effet des apports proteique et énergétique sur le métabolisme
TRANSCRIPT
GASTON RAGGIO
EFFET DES APPORTS PROTEIQUE ETÉNERGÉTIQUE SUR LE MÉTABOLISME
PROTEIQUE CHEZ LA VACHE LAITIÈRE
Thèse de doctorat en cotutelle présentéeà la Faculté des études supérieures de l'Université Laval, Québec
dans le cadre du programme de doctorat en Sciences animalespour l'obtention du grade de Philosophiae Doctor (Ph.D)
DEPARTEMENT DES SCIENCES ANIMALESFACULTÉ DES SCIENCES DE L'AGRICULTURE ET DE L'ALIMENTATION
UNIVERSITÉ LAVALQUÉBEC
Et
ÉCOLE NATIONALE SUPERIEURE AGRONOMIQUE (ENSAR)DE RENNES, FRANCE
Pour l'obtention du grade de Docteur
2006
© Gaston Raggio, 2006
II
RÉSUMÉ
Cette recherche avait pour but d'étudier le métabolisme protéique aux niveaux
splanchnique et mammaire en fonction d'apport en protéines ou en énergie. Les apports
protéiques ont été modifiés en augmentant les protéines métabolisables fournies par la
ration ou par la perfusion duodénale de caséine. Les apports d'énergie ont été modulés
par la perfusion ruminale d'acide propionique. Pour appréhender les modifications du
métabolisme protéique engendrées par l'apport de protéines et/ou d'énergie, nous avons
à la fois mesuré des débits nets des acides aminés et des débits dynamiques en utilisant
des traceurs, 13C-leucine et 2Hs-phénylalanine. Nos résultats démontrent qu'une
augmentation de l'approvisionnement en protéines métabolisables diminue l'efficacité
globale de transformation de l'apport protéique en protéines du lait. Cette diminution
d'efficacité est reliée à une augmentation du catabolisme avec des sites d'extraction se
partageant selon deux groupes d'acides aminés, soit hépatique pour l'histidine, la
méthionine, la phénylalanine et la tyrosine, soit par les tissus périphériques incluant la
glande mammaire, pour la lysine et les acides aminés à chaîne ramifiée. Quant à une
augmentation de l'apport énergétique sous forme d'acide propionique dans le rumen,
elle affecte le métabolisme protéique à deux niveaux : au niveau corporel en diminuant
l'oxydation des acides aminés indispensables et au niveau mammaire en augmentant
principalement l'extraction des acides aminés (AA) non indispensables. Ainsi, les
apports protéique et énergétique stimulent la production de protéines laitières par des
mécanismes différents, d'où l'effet additif observé sur ce paramètre. Ces résultats
confirment qu'un facteur fixe de conversion des acides aminés absorbés en sécrétion de
ces acides aminés dans le lait ne devrait pas être utilisé tel qu'il l'est actuellement dans
les différents modèles de prédiction de la production de protéines du lait. Ce facteur
varie selon le bilan nutritionnel de l'animal et selon chaque acide aminé.
III
AVANT-PROPOS
Pendant mes années de doctorat en cotutelle, j 'ai eu la chance de travailler et d'étudier
dans quatre différents endroits du monde :
- Centre de recherche d'Agriculture et Agroalimentaire Canada, Lennoxville,
Québec, Canada,
- Université Laval, Québec, Canada,
- Institut national de la recherche agronomique (INRA), dans l'unité mixte de
recherches sur la production du lait, St-Gilles, France,
- Rowett Research Institute, Aberdeen, Royaume-Uni.
Dans chaque centre, j 'ai eu la possibilité de travailler avec plusieurs personnes,
lesquelles m'ont toujours démontré leur confiance, et nous avons vraiment formé une
très bonne équipe de travail. À vous, tous les techniciens, personnes responsables
d'étable, animaliers : Merci beaucoup. Sincèrement, vous avez fait beaucoup pour moi.
Directeurs de programme d'étude:
À Doris Pellerin, mon directeur de thèse à l'Université Laval, un grand merci pour
m'avoir donné la possibilité d'être son étudiant à la maîtrise ainsi qu'au doctorat. Je
vous en remercie sincèrement.
À Jean-Louis Peyraud, mon directeur de thèse de l'INRA, merci beaucoup pour m'avoir
accepté comme son étudiant et pour votre accueil à la station de recherche de St-Gilles.
Directeurs de programme de recherche et chercheurs impliqués :
À Henri Rulquin, Jocelyne Guinard-Flament, David Pacheco Rios et Guy Allard, pour
leurs conseils et bonnes discussions, je dis merci beaucoup.
IV
À Gerald Lobley, mon superviseur et responsable au Rowett Research Institute, merci
beaucoup pour m'avoir donné l'opportunité de travailler avec vous, pour le temps que
vous m'avez dédié, pour nos nombreuses discussions et pour m'avoir communiqué
votre passion de la recherche.
À Sophie Lemosquet-Simon, je voudrais te remercier pour ton excellent accueil au
centre de recherche de même que chez ta famille. Nos discussions et les nombreuses
heures de travail que nous avons partagées, tes conseils et ta passion pour la recherche
m'ont servi énormément. Sincèrement, ton encadrement a été fondamental pour moi.
À Hélène Lapierre «LA PATRONA », sincèrement, je ne trouve pas les mots pour te
remercier pour toutes ces extraordinaires années de travail passées ensemble (6). Ta
capacité pour t'exprimer et te faire comprendre est extraordinaire, et ton amour pour la
recherche sur les acides aminés est indescriptible. Merci pour m'avoir donné le
privilège d'être ton étudiant. Aussi, PATRONA, je te remercie immensément pour me
faire sentir comme si ta famille était la mienne.
A MI FAMILIA:
Esta tesis esta dedicada a mi querida abuela FIFI, ella es una de las personas que
siempre me dio su carino incondicional en todo momento, alguien muy especial para
mi. Aunque fisicamente ya no esta con nosotros, ella siempre va a estar a mi lado.
ABUELA ESTO ES PARA VOS.
Ignacio, Caria, Justina y Tiago: Les agradezco todo corazôn lo que hicieron por mi, si
ustedes no me hubiesen recibido y ayudado durante los primeros anos en Canada este
sueno no se habria convertido en realidad.
Santiago y Maria Elena: Sinceramente le agradezco todos esos momentos que hemos
compartido juntos.
Papa y Mamâ: Ustedes me dieron la vida y la fuerza de hacer todo lo que estoy
haciendo en estos momentos. Les agradezco de corazôn todo ese carino que me han
dado y todos los valores que me inculcaron.
V
A mi mujer Anik: Amor de mi vida, te agradezco por todos estos anos juntos, toda tu
comprensiôn y el esfuerzo que hiciste para que nuestro inséparable amor siga.
Sinceramente, sos un ejemplo de perseverancia y de trabajo. Te amo para siempre.
VI
TABLE DES MATIERES
RÉSUMÉ II
AVANT-PROPOS III
TABLE DES MATIÈRES VI
LISTE DES TABLEAUX XII
LISTE DES FIGURES XV
LISTE DES ABRÉVIATIONS XVII
CHAPITRE I INTRODUCTION 1
CHAPITRE II REVUE DES TRAVAUX ANTÉRIEURS 5
2.1 Techniques de mesure du métabolisme protéique 5
2.1.1 Introduction 5
2.1.2 Mesures au niveau corporel 5
2.1.2.1 Bilan net 6
2.1.2.2 Bilan dynamique 6
2.1.2.2.1 Perfusion de traceur en bolus 8
2.1.2.2.2 Perfusion de traceur en continu 9
2.1.2.2.3 Choix du pool précurseur 11
2.1.3 Mesures au niveau tissulaire 12
2.1.3.1 Concentrations des acides aminés 12
2.1.3.2 Estimation du débit sanguin 13
2.1.3.2.1 Méthode de dilution 14
2.1.3.2.2 Sondes débitmétriques 14
2.1.3.2.3 Principe de Fick 15
2.1.3.2.4 Comparaison des méthodes 16
2.1.3.3 Débit net des acides aminés 17
VII
2.1.3.3.1 Sang entier ou plasma 17
2.1.3.4 Débit dynamique 17
2.1.3.4.1 Choix du pool précurseur 19
2.2 Métabolisme protéique chez la vache laitière 20
2.2.1 Distribution de la synthèse protéique au niveau corporel 20
2.2.2 Facteurs qui régulent la synthèse protéique 21
2.2.2.1 L'apport en protéines 22
2.2.2.2 L'apport d'énergie 23
2.3 Métabolisme mammaire des acides aminés 24
2.3.1 Introduction 24
2.3.2 Matériel et méthode 26
2.3.2.1 Formation de la base de données 26
2.3.2.2. Recalculs effectués 28
2.3.2.3 Modèle du calcul utilisé 29
2.3.3 Résultats 29
2.3.3.1 Concentrations artérielles 32
2.3.3.2 Différences artério-veineuses mammaires 34
2.3.3.3 Le prélèvement mammaire des AA 36
2.3.3.4 Le ratio prélèvement mammaire sur sécrétion en protéines du lait 38
2.3.4 Discussion 40
2.3.5 Conclusion 43
2.4 Objectifs de thèse 45
2.5 Liste des ouvrages cités 47
CHAPITRE III EFFECT OF LEVEL OF METABOLIZABLE PROTEIN ON
SPLANCHNIC FLUX OF AMINO ACIDS IN LACTATING DAIRY COWS 58
3.1 Résumé 59
3.2 Abstract 60
3.3 Introduction 61
3.4 Material and Methods 62
VIII
3.4.1 Animais and Treatments 62
3.4.2 Sampling 63
3.4.3 Laboratory Analyses 63
3.4.4 Calculations and Statistical Analyses 65
3.5 Results 66
3.5.1 Milk Production and Composition 66
3.5.2 Net Flux of Essential Amino Acids 66
3.5.3 Urea and Ammonia 67
3.5.4 N Balance 67
3.6 Discussion 68
3.6.1 Amino Acid Flux 68
3.6.2 Urea and Ammonia 71
3.6.3 Milk Production 72
3.6.4 Nitrogen Balance 73
3.7 Conclusion 74
3.8 Acknowledgements 74
3.9 Références 85
CHAPITRE IV EFFECT OF PROTEIN SUPPLY ON HEPATIC SYNTHESIS
OF CONSTITUTIVE AND PLASMA PROTEINS IN LACTATING DAIRY
COWS 91
4.1 Résumé 92
4.2Abstract 93
4.3 Introduction 94
4.4 Materials and Methods 95
4.4.1 Animais, Treatments and Sampling 95
4.4.2 Laboratory Analyses 96
4.4.3 Calculations and Statistical Analyses 97
4.5 Results 99
IX
4.6 Discussion 99
4.7 Conclusion 102
4.8 Aknowledgments 103
4.9 Références 109
CHAPITRE V EFFECT OF CASEIN AND PROPIONATE SUPPLY ON
WHOLE BODY PROTEIN METABOLISM IN LACTATING DAIRY COWS..112
5.1 Résumé 113
5.2 Abstract 114
5.3 Introduction 115
5.4 Material and Methods 116
5.4.1 Animais 116
5.4.2 Feeding and Treatments 117
5.4.3 Sampling and Laboratory Analyses 119
5.4.4 Calculations 120
5.4.5 Statistical Analyses 122
5.5Results 122
5.5.1 Milk Production 122
5.5.2 Leucine Kinetics 122
5.5.3 Amino Acid and Urea Concentrations 123
5.6 Discussion 124
5.6.1 Milk Production and Composition 124
5.6.2 Whole body leucine kinetics 125
5.6.3 Carbon Balance 127
5.7 Conclusion 128
5.8 Acknowledgments 128
5.9 Références 135
X
CHAPITRE VI EFFECT OF CASEIN AND PROPIONATE SUPPLY ON
MAMMARY PROTEIN METABOLISM IN LACTATING DAIRY COWS 141
6.1 Résumé 142
6.2 Abstract 143
6.3 Introduction 144
6.4 Materials and Methods 146
6.4.1 Animais and Treatments 146
6.4.2 Sampling and Laboratory Analyses 148
6.4.3 Calculations 149
6.4.4 Expérimental Design and Statistical Analyses 153
6.5 Results 153
6.5.1 Mammary Leucine Kinetics 153
6.5.2 Amino Acid Arterio-Venous Concentration Différences 154
6.5.3 Amino Acid Uptake and Uptake : Output Ratio 154
6.5.4 Nitrogen Balance Across the Mammary Gland 155
6.6 Discussion 156
6.6.1 Protein Metabolism 156
6.6.1.1 Leucine Net Uptake 156
6.6.1.2 Leucine Kinetics 157
6.6.1.3 Net Uptake of Amino Acids 158
6.6.1.3.1 EffectofCasein 158
6.6.1.3.2 Effect of propionate 160
6.6.1.4 Peptide Release 161
6.7 Conclusion 161
6.8 Acknowledgements 161
6.9 Références 169
CHAPITRE VII DISCUSSION ET CONCLUSION GÉNÉRALE 175
7.1 Discussion 175
XI
7.2 Conclusion 178
7.3 Liste des ouvrages cités 180
ANNEXE 1 181
1. Débit net des acides aminés (AA), mmol/h 181
2. Mouvement dynamique 181
2.1 Utilisation corporelle 181
2.2 Utilisation au niveau d'un organe 182
2.3 Glande mammaire 182
2.3.1 L'oxydation au niveau mammaire 183
2.4 Synthèse des protéines plasmatiques 184
2.5 Liste de références 185
XII
LISTE DES TABLEAUX
Tableau 2.1. Variation inter-essai, de la composition en AA des PDI dans les essais
utilisés pour constituer la base de données 30
Tableau 2.2. Caractéristiques zootechniques de la base de données 30
Table 3.1. Composition of the three dietary treatments with différent levels of
metabolizable protein (MP) 75
Table 3.2. Chemical composition of the feed ingrédients 76
Table 3.3. Effect of metabolizable protein (MP) supply on milk production and
composition in Holstein cows 77
Table 3.4. Effect of metabolizable protein (MP) supply on arterial concentration
of amino acids (uM) in Holstein cows 78
Table 3.5. Effect of metabolizable protein (MP) supply on net fluxes of amino
acids (mmol/ h) in Holstein cows 79
Table 3.6. Effect of metabolizable protein (MP) supply on net fluxes of amino
acids (mmol/ h) in Holstein cows. (continued) 80
Table 3.7. Effect of metabolizable protein (MP) supply on the ratio of amino acids
in milk protein relative to portai absorption 81
Table 3.8. Effect of metabolizable protein (MP) supply on the ratio of the hepatic
net removal of amino acids relative to portai absorption and total
influx 82
Table 3.9. Effect of metabolizable protein (MP) supply on arterial concentrations
(mM) and splanchnic flux (mmol/h) of urea and ammonia in Holstein
cows 83
Table 3.10. Effect of metabolizable protein (MP) supply on nitrogen balance (g/d)
in Holstein cows 84
XIII
Table 4.1. Effect of supply of metabolizable protein (MP) on concentrations and
protein synthesis rates of plasma albumin and total proteins 105
Table 4.2. Effect of supply of metabolizable protein (MP) on Phe concentration
and isotopic enrichment plasma samples during a 8-h infusion of 2H5-
Phe1 106
Table 4.3. Effect of supply of metabolizable protein (MP) on net transfers of Phe
(mmol/h) 107
Table 4.4. Effect of supply of metabolizable protein (MP) on Phe kinetics
(mmol/h) 108
Table 5. 1. Chemical composition of the feed ingrédients 129
Table 5.2. Effect of casein (Cas) and propionate (C3) supply on DMI, milk yield
and composition 130
Table 5.3. Effect of casein (Cas) and propionate (C3) supply on whole body (WB)
leucine kinetics 131
Table 5.4. Effect of casein (Cas) and propionate (C3) supply on 13CO2 isotopic
enrichment (IE), carbon dioxide entry rate (CER) and carbon balance 132
Table 5.5. Effect of casein (Cas) and propionate (C3) supply on plasma arterial
concentration of amino acids andurea 133
Table 6.1. Effect of casein (Cas) and propionate (C3) supply on milk yield and
composition of the right half udder in dairy cows 163
Table 6.2. Effect of casein (Cas) and propionate (C3) supply on leucine kinetics
and plasma flow of the right half udder in dairy cows 164
XIV
Table 6.3. Effect of casein (Cas) and propionate (C3) supply on plasma
arteriovenous différence of amino acids on the right half udder in dairy
cows 165
Table 6.4. Effect of casein (Cas) and propionate (C3) supply on amino acid uptake
by the right half udder in dairy cows 166
Table 6.5. Effect of casein (Cas) and propionate (C3) supply on amino acid
uptake: milk output ratio in dairy cows 166
Table 6.6. Effect of casein (Cas) and propionate (C3) supply on the nitrogen
balance across the right half udder in dairy cows 168
XV
LISTE DES FIGURES
Figure 2.1. Représentation de l'utilisation totale des AA 10
Figure 2.2. Représentation de l'oxydation de la leucine 10
Figure 2.3. Différentes possibilités de choisir le pool précurseur (PP) dans un
modèle pour l'étude de la synthèse de protéines avec trois
compartiments 20
Figure 2.4. Contribution des différents tissus à la masse protéique (Protein mass),
la synthèse protéique (protein synthesis). et à l'utilisation nette
d'oxygène (oxygen uptake) indicateur de l'énergie utilisée (tiré de
Lobley et al., 2002) 21
Figure 2.5. Augmentation de la production de protéines brutes dans le lait en
fonction des apports PDI 31
Figure 2.6. Effet des apports en PDI sur les concentrations artérielles des AA du
groupe 1 32
Figure 2.7. Effet des apports en PDI sur les concentrations artérielles des AA du
groupe 2 33
Figure 2.8. Effet des apports en PDI sur les concentrations artérielles d'AA non
indispensables 34
Figure 2.9. Effet des apports en PDI sur les différences arterio-veineuses
mammaires (DAVM) des AA du groupe 1 34
Figure 2.10. Effet des apports en PDI sur les différences arterio-veineuses
mammaires (DAVM) des AA du groupe 2 35
Figure 2.11. Effet des apports en PDI sur les différences arterio-veineuses
mammaires (DAVM) des AA non indispensables 36
Figure 2.12. Effet des apports en PDI sur le prélèvement mammaire (PM) des AA
du groupe 1 36
Figure 2.13. Effet des apports en PDI sur le prélèvement mammaire (PM) des AA
du groupe 2 37
XVI
Figure 2.14. Effet des apports en PDI sur le prélèvement mammaire (PM) des AA
non indispensables 38
Figure 2.15. Effet des apports en PDI sur le ratio prélèvement mammaire et
sécrétion en protéines du lait (PM: PL) des AA du groupe 1 38
Figure 2.16. Effet des apports en PDI sur le ratio prélèvement mammaire sur
protéines du lait (PM : PL) des AA du groupe 2 39
Figure 2.17. Effet des apports en PDI sur le ratio prélèvement mammaire et
protéines du lait (PM : PL) des AA non indispensables 40
Figure 4. 1. Average of the enrichment (MPE= molar percent excess x 1000) of
Phe in plasma total proteins and albumin during a continuous infusion
of 2H5-Phe at Low (la) or High (lb) metabolizable protein (MP)
supply 104
Figure 5.1. Effect of extra leucine intake on leucine oxidation, leucine in milk and
leucine in tissues 134
XVII
LISTE DES ABRÉVIATIONS
Françaises:
AA = acides aminés, AA-DI = AA digestible, El = enrichissement isotopique, pAH =
para-aminohippurate, DAV = différence arterio-veineuse, DAVM = DAV mammaire,
MOP =4-methyl 2-oxopentanoate, UT AA = utilisation totale d'AA, PDI = protéines
digestibles dans l'intestin, PM = prélèvement mammaire
Anglaises:
AA = amino acids, ASR = absolute synthesis rate, AV = arterio-venous, C3 =
propionate, CER = carbon dioxide entry rate, Cas = casein, Ctrl = control, Cas+C3 =
casein and propionate, BCAA = branched-chain AA, EAA = essential AA, FO =
fractional oxidation, IE = isotopic enrichment, ILR = irréversible loss rate, MG =
mammary gland, MP = metabolizable protein, MOP = 4-methyl 2-oxopentanoate,
NEAA - non-essential amino acids, PDV = portal-drained viscera, pAH = para-
aminohippurate, PS = protein synthesis; WB, whole body.
CHAPITRE I
INTRODUCTION
L'augmentation de l'efficacité alimentaire a été et sera le défi de tous les
nutritionnistes impliqués dans le secteur laitier. L'idée de produire une plus grande
quantité de protéines à un coût de production moindre exige de produire avec une
meilleure efficacité globale. Une étape essentielle pour atteindre cet objectif est la
compréhension du métabolisme protéique animal, non seulement pour améliorer la
production, mais aussi pour réduire l'impact de la production animale sur
l'environnement, relié à l'azote éliminé dans les fèces et l'urine.
L'amélioration de l'efficacité de la conversion des aliments en lait implique une
meilleure adéquation entre les nutriments fournis par la ration et ceux nécessaires au
fonctionnement optimal des tissus. Généralement, une augmentation de l'apport
protéique, soit par la perfusion de caséine (Clark et al., 1977; Guinard et Rulquin,
1994a; Hanigan et al., 1998), soit par une augmentation de protéines métabolisables par
la ration (Blouin et al., 2002) résulte en une augmentation de la production de protéines
dans le lait. Aussi, une augmentation de l'apport énergétique a un effet positif mais de
moindre envergure sur la production de protéines du lait (Clark et al., 1977; Coulon et
al., 1991; Rigout et al., 2003). Pour comprendre les mécanismes responsables de cette
augmentation, il faut comprendre comment les apports de protéines ou d'énergie vont
affecter le métabolisme des AA chez la vache laitière.
Nos connaissances sur le métabolisme des AA chez la vache laitière sont
limitées, comme le démontre notre difficulté à prédire la réponse à un accroissement
d'apports protéiques ou d'AA (Hanigan et al., 1998) ou la récupération relativement
faible d'acides aminés perfusés au niveau digestif ou vasculaire dans la lait produit
(MacRae et al., 1993). Alors que les modèles actuels prédisent une utilisation marginale
d'environ 67 % pour un supplément protéique vers la protéine du lait (NRC, 2001) si
les besoins ne sont pas rencontrés, l'efficacité marginale moyenne mesurée dans une
série de projets se situait plutôt autour de 21 % (Hanigan et al., 1998) ou aux environ de
30 % selon les données de l'INRA.
L'absorption portale nette ou apparente d'AA chez la vache laitière représentait
plus de 50 % de l'azote digéré (Blouin et al., 2002), ce qui est supérieur à ce qui est
rapporté chez le bovin en croissance où la proportion d'AA absorbés par rapport à
l'azote digéré est inférieure à 40 % (Lapierre et Lobley, 2001). L'efficacité nette
d'utilisation de ces AA absorbés pour être transformés en protéines du lait varie entre 40
et 80 % dépendant de l'AA concerné (Blouin et al., 2002). Elle diminue avec une
augmentation des apports en protéines métabolisables, ce qui indique une très grande
variation de l'utilisation des AA au niveau des tissus.
Après avoir été absorbés dans la veine porte, les AA sont dirigés vers le foie, qui
occupe une position stratégique anatomiquement, faisant l'interface entre l'apport
nutritionnel et les tissus périphériques. L'extraction hépatique des AA indispensables
varie énormément d'un AA à l'autre. En général, le foie extrait 45 % du total des AA
absorbés (Lapierre et al., 2005), mais cette extraction est très variable selon chaque AA.
Par exemple, l'extraction d'AA comme l'histidine, la méthionine, la phénylalanine, la
tyrosine et la thréonine varie de 13 à 50 % de leur absorption portale alors que d'autres
AA tels que la leucine, l'isoleucine, la valine et la lysine ne sont pas ou très peu extraits,
sur une base nette, par le foie (Blouin et al., 2002 ; Hanigan, 2005; Lapierre et al.,
2005). Ces derniers AA peuvent être catabolisés par la glande mammaire et d'autres
tissus périphériques, en accord avec la présence des différents enzymes de dégradation
de ces acides aminés dans différents tissus (Lobley et Lapierre, 2003).
De façon évidente, le foie joue un rôle de détoxication des AA lorsqu'ils sont
fournis en excès de la demande. Cependant, le foie a aussi un rôle de synthèse de
protéines d'exportation qui pourraient constituer un niveau d'extraction obligatoire
diminuant d'autant l'approvisionnement aux tissus périphériques en AA libres. En effet,
le foie fabrique des protéines plasmatiques qui vont être exportées dans la circulation
générale (Connell et al., 1997) et qui sont d'importance vitale pour les animaux.
L'importance de cette production de protéines n'est pas connue chez la vache, ni si elle
est modulée par l'apport en protéines de la ration. De plus, le foie extrait une proportion
importante d'AA non indispensables à des fins de néoglucogenèse. Jusqu'à 17 % de la
néoglucogenèse peut provenir de carbones d'AA (Reynolds et al., 1994). Ainsi, une
limitation en AA non indispensables pourrait avoir un effet négatif sur
l'approvisionnement en énergie à la glande mammaire, ou un surplus d'énergie pourrait
avoir pour effet d'épargner des AA non indispensables et se traduire par une plus grande
quantité d'AA arrivant à la glande mammaire.
L'approvisionnement post-hépatique en AA est utilisé par les tissus
périphériques, incluant la glande mammaire, un des organes les plus actifs dans
l'extraction de ces AA, pour supporter une production de protéine du lait pouvant
dépasser 1 kg par jour. La glande mammaire pourrait réguler le prélèvement des AA en
contrôlant le débit du sang ainsi qu'en augmentant l'efficacité de son extraction, tel
qu'observé par Bequette et al. (2002) chez la chèvre laitière. Une façon de mieux
comprendre ces phénomènes est par des études dynamiques avec traceurs (Bequette et
al, 1996a; Lapierre et al., 2002) ou en évaluant le ratio net du prélèvement mammaire
d'un AA sur la sécrétion de cet AA dans la protéine du lait (Clark et al., 1977; Guinard
et Rulquin, 1994a; Vanhatalo et al., 2003a). La combinaison des deux techniques
aiderait à avoir une vision plus globale.
Le ratio des AA prélevés par la glande mammaire sur la protéine sécrétée dans le
lait varie selon des groupes d'AA. Selon la classification de Mepham (1982), les AA
indispensables suivants : l'histidine, la méthionine, la phénylalanine (et la tyrosine) et la
thréonine, sont prélevés et sécrétés dans un ratio 1 : 1 . Ces AA sont ceux qui sont
principalement extraits par le foie. Dans le cas des AA indispensables qui ne sont pas
extraits par le foie (la leucine, l'isoleucine, la valine et la lysine), la glande mammaire
les extrait en excès (Mepham, 1982) par rapport aux besoins directs de synthèse de
protéines du lait : cela implique qu'ils doivent y être oxydés (Bequette et al., 1996a et b;
Lapierre et al., 2005), ou qu'il existe la possibilité d'une production intra-mammaire et
d'un largage de peptides dans le drainage mammaire (Guinard et Rulquin, 1994a:
Rulquin et al., 2004).
En conclusion, à travers différents projets de recherche, nous avons l'intention
de comprendre plus en profondeur une partie de ce complexe univers du métabolisme
des AA chez la vache laitière.
Les objectifs spécifiques de la thèse sont donc :
1- D'analyser comment différents apports de protéines métabolisables
vont affecter le métabolisme des acides aminés au niveau hépatique
et post-hépatique en se basant sur le bilan net des acides aminés au
niveau de ces tissus.
2- De déterminer si la synthèse de protéines plasmatiques au niveau
hépatique va être compromise par une restriction de l'apport de
protéines métabolisables. Un AA marqué sera perfusé pour mesurer
son incorporation dans les protéines plasmatiques.
3- D'établir si un apport variable en énergie et/ou en protéines affecte le
métabolisme corporel et mammaire des acides aminés par des
mécanismes différents. Pour cela, des bilans nets d'AA seront
effectués et un des bilans dynamiques sera mesuré en utilisant un AA
marqué.
CHAPITRE II
REVUE DES TRAVAUX ANTERIEURS
2.1 Techniques de mesure du métabolisme protéique
2.1.1 Introduction
Une meilleure compréhension du métabolisme des acides aminés (AA) dans
différents tissus et la coordination (ou compétition) entre ces tissus pour l'utilisation de
ces AA requiert de faire des mesures de façon quantitative in vivo dans un contexte
physiologique où l'on essaie de ne pas modifier les organes étudiés. Pour réaliser ce
type de travail, nous devons utiliser des méthodologies qui nous permettront de
quantifier différents phénomènes. Aux niveaux corporel et tissulaire, les bilans nets
peuvent être faits à partir des différences entre les quantités entrant et sortant du
système, soit en azote (N) total au niveau corporel ou plus spécifiquement en AA au
niveau tissulaire. Cependant, que ce soit au niveau corporel ou tissulaire, l'estimation
des voies dynamiques d'utilisation des AA pour des fins d'oxydation, de synthèse ou
provenant de dégradation protéique nécessite l'utilisation de molécules marquées dont
le devenir métabolique pourra être suivi par leur différence de masse (isotope stable) ou
par rayonnement radioactif (isotope radioactif). Cette revue a pour objectif de décrire
les techniques utilisées et le type d'information qu'elles génèrent dans la mesure du
métabolisme protéique, soit au niveau corporel ou tissulaire.
2.1.2 Mesures au niveau corporel
La masse protéique corporelle est régulée à partir d'un changement dynamique
constant entre la synthèse et la dégradation protéique appelé « turnover » protéique. La
différence entre ces deux composantes correspond au bilan net, qui représente soit la
rétention azotée pour les animaux en croissance, soit la somme de la rétention azotée et
de la production de protéines laitières pour un animal en lactation.
2.1.2.1 Bilan net
Le bilan net représente la différence entre l'ingestion totale d'azote et la somme
des excrétions sous forme urinaire ou fécale, à laquelle on ajoute la production de
protéines laitières chez la vache en lactation. On obtient le bilan azoté en mesurant sur
une période de quelques jours l'azote ingéré (alimentation servie moins refus) et l'azote
excrété (en faisant une collecte totale de fèces et d'urine et la mesure de la production
laitière). Les échantillons des différentes fractions sont analysés pour leur contenu en
azote total. La somme de l'azote retenu plus l'azote sécrété sous forme de lait représente
la différence entre la synthèse et la dégradation protéique. En production animale, c'est
ultimement le bilan net que nous voulons augmenter, mais une variation de celui-ci ne
nous renseigne pas du tout sur les mécanismes qui en sont responsables.
2.1.2.2 Bilan dynamique
Un suivi de la dynamique protéique plutôt que la mesure de son seul résultat
final peut, par une meilleure compréhension des facteurs régulant chacune de ces
fonctions, nous indiquer les meilleures stratégies à adopter pour améliorer l'efficacité de
l'utilisation de l'azote alimentaire. La synthèse corporelle de protéines représente la
somme de la synthèse protéique de tous les organes et tissus. La croissance, la nutrition,
la gestation, la lactation, les maladies et les blessures vont affecter ce processus
dynamique à différents niveaux, mais toutefois, avec flexibilité et coordination pour
maintenir leur fonction métabolique par les différents organes.
Cependant, la mesure de la dynamique protéique exige une méthodologie plus
élaborée que la simple mesure du bilan net. Elle exige l'utilisation de molécules
marquées, communément appelées « traceurs ». Pour étudier le métabolisme protéique,
les traceurs sont des AA marqués par la substitution de l'un de leurs atomes (ou
plusieurs) par un isotope de cet atome. Un isotope est un atome ayant le même nombre
de protons que l'élément original, mais avec un nombre plus élevé de neutrons, lui
conférant une masse plus grande. Il y a les isotopes stables, qui ne diffèrent que par leur
masse et les isotopes radioactifs, qui, en plus de la différence de masse, dégagent un
rayonnement radioactif qui peut être mesuré. Les techniques de mesure de turnover
protéique ont d'abord été développées avec les isotopes radioactifs (Garlick et al., 1973;
James et al., 1973; Golden et Waterlow, 1977; Reeds et al., 1978; Lobley et al., 1980;
Lobley et al., 1987) mais l'amélioration de la sensibilité des analyses en spectrométrie
de masse de même que le risque environnemental élevé à utiliser des produits
radioactifs chez des animaux de ferme (récolte et mise au rebut des produits terminaux
et des carcasses) ont grandement incité l'utilisation d'isotopes stables (Simon et al.,
1978; Liu et al., 1995; Lapierre et al., 1999) dans la recherche sur les grands animaux.
Les isotopes principalement utilisés sont le 2H, le 13C, le 15N et le 18O pour
isotopes stables et le 3H et le 14C pour les isotopes radioactifs.
Avec l'un ou l'autre des types d'isotopes, on doit prendre en considération les points
suivants (Wolfe, 1984):
- La perfusion de traceurs ne doit pas altérer la cinétique de la molécule « tracée »,
- L'animal doit être dans un état stable d'équilibre nutritionnel dans notre modèle du
métabolisme protéique,
- Le traceur est utilisé par les voies métaboliques comme la molécule tracée endogène,
- Le traceur ne doit pas être recyclé, c'est-à-dire qu'il ne doit pas revenir dans le pool
initial, pendant la période de mesure,
- Les sites de perfusion et d'échantillonnage doivent être situés de manière à refléter
l'enrichissement isotopique (El) du pool précurseur, suite à un mélange parfait
« traceur-tracé ».
- Le substrat marqué apparaît ou disparaît directement du compartiment échantillonné.
Pour les isotopes radioactifs, la mesure d'enrichissement se fait en comptant le
nombre de désintégrations par minute de la molécule étudiée, appelée radioactivité
spécifique. Pour les isotopes stables, l'EI est la mesure de la proportion de molécules
marquées dans le système étudié (habituellement sous perfusion continue ou après une
dose bolus) par rapport à l'abondance naturelle, car tous les atomes sont présents
naturellement sous forme d'isotopes stables dans des proportions variables. Ainsi, l'EI
est toujours exprimé par rapport à des taux de base mesurés avant la perfusion du
traceur. L'EI s'exprime en « atome pourcent excès (ape) ou « mole pourcent excès »
(mpe) » et est calculé des façons suivantes :
1) El (ape ou mpe) = (traceur / traceur + tracée) x 100
2) El (ape ou mpe) = (Ri - Ro) / ((1 + (Ri - Ro)) x 100
où Ri et Ro sont respectivement le ratio entre la surface sous la courbe de la masse M +
x / M + 0 (ou x représente la masse enrichie et 0 la masse sans isotope) de l'échantillon
analysé (Ri) et de l'échantillon du taux de base (Ro). Cette équation tient compte du fait
que l'AA, avant d'être séparé par chromatographie en phase gazeuse puis détecté sur le
spectromètre de masse, doit être dérivatisé, ce qui lui ajoute un nombre important
d'atomes de C, H et N ayant chacun une probabilité naturelle de contenir des isotopes
stables (Wolfe, 1984).
Afin d'estimer l'utilisation totale d'un AA, il y a différentes techniques: 1)
perfusion en bolus ou 2) perfusion en continu de traceurs.
2.1.2.2.1 Perfusion de traceur en bolus
L'utilisation d'un traceur en injection bolus permet d'évaluer sa disparition du
compartiment échantillonné, correspondant à son utilisation totale dans un état
d'équilibre (Garlik et al., 1980). Elle demande une injection en bolus, suivi de
prélèvements répétés à intervalles très courts, dus à la très courte demi-vie des AA. En
reliant le logarithme naturel de l'EI en fonction du temps, on obtient une courbe à deux
exponentiels dont les paramètres sont utilisés pour estimer l'utilisation totale (Holtrop et
al., 2004). Cette approche offre l'avantage de l'utilisation de doses beaucoup plus
faibles que la perfusion continue, mais ne permet pas de mesurer l'oxydation, paramètre
essentiel pour une estimation correcte de la synthèse protéique (c.f section 2.2.2). Chez
les ruminants, la perfusion en bolus a été relativement peu utilisée au niveau corporel
(Holtrop et al., 2004; Borucki Castro et al., 2006).
2.1.2.2.2 Perfusion de traceur en continu
Cette méthode pour mesurer la synthèse de protéines est la plus répandue chez
les animaux de ferme et a été utilisée par de nombreux scientifiques chez différentes
espèces (humain : Steffee et al., 1976; mouton : Lobley et al., 1980; porc : Reed et al.,
1981; bovin : Lapierre et al., 1999; chèvre : Bequette et al., 2002 ; vache laitière:
Bequette et al., 1996). Elle consiste en la perfusion continue, habituellement dans une
veine, du traceur durant quelques heures (4-12h), au cours desquelles l'animal en étude
doit être dans un état d'équilibre métabolique stationnaire. Celui-ci peut être atteint par
exemple en fractionnant l'alimentation (toutes les deux heures chez les ruminants). Pour
obtenir un plateau d'EI plus rapidement, il est possible d'administrer une dose
d'amorçage du traceur équivalant à une heure de perfusion (Wolfe, 1984). Dans ces
conditions, un plateau de l'EI dans le plasma est habituellement observé après deux
heures de perfusion alors qu'autrement ce plateau est atteint après 4-6 heures.
L'utilisation totale de l'AA étudié est calculée en faisant l'hypothèse qu'il y a un
équilibre entre l'entrée du traceur (perfusion) et sa sortie du compartiment
échantillonné. L'utilisation totale (UT), habituellement désignée sous le terme de
« Irréversible Loss Rate - ILR » en anglais, représente l'utilisation totale de l'AA étudié
au niveau corporel, ce qui signifie que tous les organes et tissus sont pris en compte.
L'UT est calculée avec l'équation qui suit:
3)UT= (EI inf/EIpp-l)*Inf
= synthèse protéique (SP) + oxydation (OX)
= absorption + dégradation de protéines
où El inf est l'EI de la perfusion, Inf est la dose de perfusion du traceur (mmol/h),
EIpp, l'EI de l'AA perfusé au site de prélèvement (soit artériel ou veineux en amont de
la perfusion) utilisé comme étant représentatif de l'EI du pool précurseur. Dans le cas de
la leucine, l'EI du céto-acide de la leucine, le 4-méthyl 2-oxopentanoate (MOP), formé
au niveau intracellulaire et relargué au niveau veineux est souvent utilisé comme
représentatif du pool précurseur, plutôt que la leucine elle-même (Figure 2.1)
10
Oxydation
î Synthèse
Absorption
Infusion
AA libres AAdansprotéines
Dégradation
Figure 2.1. Représentation de l'utilisation totale des AA
Cette première étape nous donne, comme dans le cas de la perfusion en bolus,
l'UT. Pour vraiment déterminer la synthèse protéique (UT - oxydation), l'oxydation de
l'AA doit aussi être mesurée : la mesure se fait à partir de l'EI du 13CO2 produit pendant
la perfusion de l'AA marqué, en combinaison avec une mesure de la production totale
de CO2 quantifiée à partir d'une perfusion de bicarbonate marqué (l3C). Un exemple
suit en utilisant la leucine comme AA traceur (Figure 2.2).
Isovaleryl CoA
13C leucine (Infusion)
13C- methyl oxopentanoate (MOP)
Oxydation
Figure 2.2. Représentation de l'oxydation de la leucine
À l'aide d'une perfusion de 13C-bicarbonate, il est possible d'estimer la
production totale de CO2 (TCO2) au niveau corporel. Normalement, cette mesure se fait
une journée avant la mesure de l'UT. La TCO2 est calculée avec l'équation qui suit :
4) T C O 2 = (El 13C-bicarbonate/ El 13CO2-bicarbonate - 1) * M l3C-bicarbonate
11
En combinant la TCO2 avec l'EI du 13CC>2 obtenu durant la perfusion de l'AA, il
est possible de calculer l'oxydation du traceur
5) Leucinetraceur oxydation = TCO2 x El 13CO2-leucine
En utilisant l'oxydation de la leucinetraceUr, la perfusion du traceur et l'EI de ce
dernier, il est possible de calculer la proportion de la leucine qui est oxydée (POx) par
l'équation :
6) POx = Leucinelraceur oxydation / (Inf 13C Leucine x El int)
On fait l'hypothèse qu'une quantité équivalente à la quantité de la leucine
perfusee comme traceur est en excès dans l'animal et sera ainsi oxydée (Lobley et al.,
2003), alors l'oxydation corporelle de la leucine se calcule comme suit :
7) Oxydation corporelle leucine (mmol/h)= POx x (UT Leucine + Inf) - Inf
À partir de ces calculs, il est possible d'estimer la synthèse protéique (SP)
comme suit :
8) SP (mmol/h) = UT (mmol/h) - Oxydation (mmol/h)
La conversion de l'utilisation de la leucine pour la synthèse protéique (mmol/h)
en kg de protéines par jour, se fait à partir de la composition moyenne de la leucine dans
le corps, soit 63 g de leucine par kg de protéines chez la vache laitière (Lobley et al.,
1980) et 98 g de leucine par kg de protéine du lait (Swaisgood, 1995).
2.1.2.2.3 Choix du pool précurseur
Le choix du pool précurseur est critique dans le calcul de l'utilisation totale des
AA, que ce soit au niveau corporel ou dans un organe déterminé. L'hypothèse est que
l'EI utilisé est représentatif de l'EI du pool précurseur réel au lieu des réactions
biochimiques. Ceci indique immédiatement une contrainte majeure pour les estimations
au niveau corporel, où de manière générale un seul El représentatif de tous les pools
précurseurs de chacun des tissus est utilisé, bien que chaque tissu ait son métabolisme
propre et en conséquence un El du pool précurseur probablement différent (ex : El
leucine au niveau artériel : 1,7 ape, portai : 1,28 ape, et hépatique : 1,22 ape : Lapierre et
12
al., 2002). L'EI au niveau artériel est choisi avec des animaux multicathétérisés où
l'accès au sang artériel est facile alors que dans les études limitées au métabolisme
corporel, un pool veineux, prélevé en amont ou de l'autre côté du site de perfusion
(Bequette et al., 1996a; Lapierre et al., 2002) est habituellement utilisé
2.1.3 Mesures au niveau tissulaire
La méthodologie décrite ci-dessus permet de déterminer l'utilisation totale, qui
est composée de l'oxydation et de la synthèse au niveau corporel, mais ne permet pas de
déterminer la contribution de chacun des tissus à ces phénomènes. Tout comme au
niveau corporel, des bilans de mouvements nets ou dynamiques peuvent être effectués
au niveau des tissus en mesurant les entrées et les sorties. Les principaux tissus étudiées
chez la vache sont : les tissus drainés par la veine porte, le tissu hépatique, la glande
mammaire et les tissus musculaires. La détermination des concentrations des AA dans
le sang artériel et veineux de même que l'estimation du débit sanguin sont deux mesures
conjointes à ces deux types de bilans et critiques pour une estimation adéquate de ceux-
ci.
2.1.3.1 Concentrations des acides aminés
La mesure des variations de concentration des AA à l'entrée et à la sortie de
l'organe étudié exige le placement de cathéters situés aux niveaux artériel et veineux.
L'implantation de cathéters artériels se fait sur une artère carotide ou une autre artère en
se basant sur l'hypothèse que la composition du sang artériel est identique, peu importe
l'artère choisie (Linzell, 1974). L'implantation du cathéter veineux se fait dans la veine
de drainage de l'organe étudié : veine porte et veine hépatique pour l'étude du
métabolisme au niveau du tissu intestinal et du foie, respectivement (Huntington et al.,
1989), et veine mammaire pour l'étude du métabolisme mammaire (Cant et al., 1993;
Guinard et Rulquin., 1994a). Il y a aussi certains auteurs comme Vanhatalo et al. (2003a
et b), Mackle et al. (2000) et Koronen et al.(2002), qui ont utilisé la technique de
ponctions veineuses mammaires et de la veine caudale pour estimer l'extraction
mammaire des AA.
13
L'utilisation de cette méthodologie est très délicate. Il faut considérer que des
problèmes concernant le positionnement du cathéter, des mouvements après le
placement et l'inflammation de la veine peuvent avoir des effets sur la mesure finale.
Par exemple, pour l'étude au niveau mammaire, des recherches ont démontré qu'il faut
considérer l'âge de l'animal et son passé reproductif parce qu'ils vont avoir une
influence sur le sang prélevé dans la veine mammaire selon la fonctionnalité des valves
de la veine mammaire et de la veine pudique externe. En général, après la deuxième
lactation, les valvules de la veine pudique externe deviennent incompétentes : ainsi du
sang veineux d'origine extra mammaire peut arriver jusqu'à la veine mammaire,
provoquant une contamination par du sang ne provenant pas du métabolisme mammaire
(Linzell, 1974; Thivierge et al., 2000). Pour prévenir ce problème, certains auteurs (ex :
Linzell, 1974; Peeters al., 1979; Mepham, 1982) recommandent l'occlusion de la veine
pudique externe ou de choisir des vaches de lere et 2e lactation (Thivierge et al., 2000).
Un autre point important dans les mesures de différences artério-veineuses
(DAV) est la précision des analyses pour déterminer la concentration des AA.
Traditionnellement, les analyses d'AA se font par colonnes échangeuses d'ions qui
séparent les AA suivies par une réaction à la ninhydrine pour quantifier les
concentrations (Guinard et al., 1994). Dans une telle analyse, habituellement, un
standard externe est ajouté et la récupération de tous les AA est ajustée à ce standard. Le
coefficient de variation (CV) de ces analyses est de 3 à 5 %, au mieux. Une nouvelle
méthode a été récemment développée afin d'améliorer la précision des analyses d'AA,
la méthode de dilution isotopique. Pour chaque analyse d'AA, le même AA marqué est
ajouté comme standard interne et les concentrations sont estimées en mesurant l'EI de
l'échantillon traité (Calder et al., 1999). Cette méthode donne des CV qui se situent
entre 0.1 et 0.3 %, beaucoup plus bas que l'analyse traditionnelle.
2.1.3.2 Estimation du débit sanguin
La mesure du débit sanguin est essentielle pour le calcul du débit des nutriments.
Il est donc important que la méthode utilisée pour mesurer le débit sanguin ne perturbe
pas ou peu l'animal. Les méthodes utilisées ont évolué au cours du temps et sont basées
14
sur différents principes physiques ou biochimiques. Les plus utilisées sont : la méthode
de dilution, les sondes débitmétriques et le principe de Fick.
2.1.3.2.1 Méthode de dilution
La méthode de dilution utilise des teintures inertes, comme le para-
aminohippurate (pAH). Elle a été décrite initialement par Katz et Bergman (1969) au
niveau splanchnique. Cette méthode consiste à infuser de façon continue du pAH au
niveau d'une (ou deux) veine mésentérique, ainsi qu'à faire des prélèvements sanguins
en amont (artère) et en aval (veine porte et veine hépatique) du site de perfusion
(Huntington 1984; Huntington et al., 1989; Ortigues et al, 1994; Lapierre et al., 1999;
Blouin et al., 2002). Lorsque la veine porte est très courte, un débit laminaire au niveau
portai peut empêcher un mélange adéquat du pAH, ce qui peut parfois fausser la mesure
de débit (Seal et Reynold, 1993).
Le calcul du débit se fait en tenant compte du débit de perfusion du pAH, et la
DVA de concentration du pAH en utilisant l'équation suivante :
9) Débit sanguin = débit de perfusion du pAH / [DVA] du pAH.
Cette méthode est aussi utilisée, mais à un degré moindre, pour estimer le débit
mammaire. Dans ce cas, le pAH est perfusé dans l'artère pudique et les prélèvements se
font en amont du site de perfusion dans l'artère pudique externe et en aval dans une
veine mammaire (Metcalf et al., 1992 ; Bequette et al., 1999).
2.1.3.2.2 Sondes débitmétriques
II existe plusieurs types de sondes débitmétriques. Les premiers modèles étaient
de type électromagnétique et demandaient un alignement perpendiculaire au vaisseau
sanguin pour donner une bonne lecture, position quasiment impossible à maintenir
stable à l'intérieur d'un animal et surtout impossible à vérifier. Cette sonde semblait
aussi très sensible aux perturbations. Ces types de sondes ont donné des estimations qui
n'avaient pas de bonne corrélation avec les débits mesurés par dilution au niveau
splanchnique (e.g. Eisemann et al., 1987) et ont finalement été peu utilisées. Par contre,
les sondes à ultrason se sont avérées beaucoup plus précises dans l'animal. Ce type de
15
sondes a été utilisé par Drost (1978) : elles contiennent deux cristaux à 4 cristaux, l'un
émet l'ultrason et l'autre les reçoit ; le temps mis par l'ultrason pour rejoindre le cristal
récepteur est fonction du débit du sang dans le vaisseau sanguin.
Les sondes à ultrason ont été utilisées pour mesurer le débit sanguin mammaire
(Guinard et al., 1994 ; Bequette et al., 1996b). Elles ont aussi été utilisées pour mesurer
le débit en veine porte chez le mouton (Rémond et al., 2000) et la vache laitière (Girard,
communication personnelle). Cette méthode présente moins de variabilité que la
méthode de dilution, les mesures étant faites continuellement (Rémond et al., 1993)
mais a cependant le désavantage, au niveau splanchnique, que la contribution artérielle
au débit hépatique n'est pas mesurée directement mais doit être considérée comme une
proportion fixe du débit sanguin portai. Par contre au niveau mammaire, Bequette et al.
(1999), Mabjeesh et al. (2000), Thivierge et al. (2002) ont comparé cette méthode avec
d'autres comme le principe de Fick ou la méthode de dilution et ont observé une sous-
estimation du débit de 15-30 % avec la sonde à ultrason. Ceci pourrait s'expliquer par
trois raisons. Premièrement, les sondes à deux cristaux (type S) sous-estiment les débits
turbulents comparativement aux sondes à 4 cristaux (Rémond, communication
personnelle). Deuxièmement, le sous-dimensionnement de la taille de la sonde par
rapport à la taille de l'artère peut aussi fausser les mesures de débit (par exemple des
sondes de 16 mm à 20 mm de diamètre sont parfois posées sur l'artère pudique de
vaches laitières). Troisièmement, la mise en place d'artéiroles collatérales après la pose
de la sonde peut entraîner une diminution du débit dans l'artère pudique sous étude.
2.1.3.2.3 Principe de Fick
À cause de sa facilité d'utilisation, il s'agit de la méthode la plus utilisée pour
estimer le débit plasmatique mammaire (Linzell, 1974). Cette méthode considère que,
pour une composante donnée, le prélèvement par le tissu (DAV x débit plasmatique) est
équivalent à la quantité sécrétée dans le lait. On considère ainsi un transfert
stœchiométrique du prélèvement mammaire dans la protéine du lait pour certains AA
tels que l'histidine, la méthionine, la phénylalanine et la tyrosine. En connaissant la
sécrétion des AA en protéines du lait (mmol/h) et la DAV mammaire des AA dans le
plasma (mmol/1), on peut donc estimer le débit plasmatique. Les AA les plus utilisés
16
sont la phénylalanine + tyrosine, la tyrosine étant tenue à compte puisqu'elle peut être
dérivée du catabolisme de la phénylalanine (Cant et al., 1993; Thivierge et al., 2002;
Korhonen et al., 2002; Vanhatalo et al., 2003a et b). Dans cette méthode, on fait une
correction du 3,5 % pour les protéines qui sont directement transportées du sang dans le
lait (Cant et al., 1993). Les quantités d'acides aminés libres dans le lait sont
négligeables et ignorées dans ce calcul.
10) Débit plasmatique = ((Phe + Tyr dans la protéine du lait) * 0,965 ) / ( DAV Phe
dans plasma + DAV Tyr dans le plasma).
2.1.3.2.4 Comparaison des méthodes
Le tableau ci-dessous décrit les caractéristiques les plus importantes et les
organes où les différentes méthodes ont été utilisées pour estimer le débit sanguin.
Sondes à ultrason
Mesure de variation dans le
temps mais semble
présenter une sous-
estimation systématique
Vie utile limitée
Coût élevé
Demande une chirurgie
Tissus splanchnique
Glande mammaire
Erreur : 20 1/h (Guinard etal , 1994)
Méthode de dilution pAH
Précision de mesure
mineure
Vie utile longue
Coût moyen
Demande l'implantation
d'un cathéter
supplémentaire - contrainte
de perfusion (pompes,
lignes de perfusion)
Tissus splanchniques
Glande mammaire
Erreur : 30 à 40 1/h
Principe de Fick
II peut y avoir
surestimation s'il y a des
peptides prélevés par la
glande mammaire
Vie utile illimitée
Coût minime
Demande une
quantification des AA du
lait (mais souvent des
valeurs de références sont
utilisées)
Glande mammaire
Erreur : 30 à 40 1/h
17
2.1.3.3 Débit net des acides aminés
Le calcul du débit net se fait par multiplication de la DAV de la concentration du
nutriment étudié et du débit plasmatique ([DAV] x DP). Ce résultat représente le
mouvement net des AA dans le tissu, qu'ils soient prélevés ou relargués. Par exemple,
dans le cas du débit net au niveau portai, il représente la quantité d'AA digérés
rejoignant la circulation après qu'aient été utilisés les AA (de source artérielle ou
provenant de la digestion) par les viscères drainés par la veine porte. Le flux net
splanchnique représente l'approvisionnement post-hépatique net d'acides aminés qui
sont disponibles pour tous les autres tissus (exemple : muscle et glande mammaire)
après l'utilisation au niveau portai et hépatique. Le flux net mammaire représente la
quantité nette d'acides aminés prélevés par la glande mammaire.
2.1.3.3.1 Sang entier ou plasma
Dans la plupart des travaux, le flux plasmatique associé à des concentrations
plasmatiques a été utilisé pour les calculs de débit des AA. Cependant, une sous-
estimation d'environ 16 % a été proposée (Heitmann and Bergman, 1980 ; Danilson et
al., 1987 ; Hanigan et al., 1991) en calculant l'extraction au niveau plasmatique vs.
sanguin par différents tissus. Pourtant, des publications récentes ont démontré au niveau
de la glande mammaire (Pacheco-Rios et al., 1999 ; Macke et al., 2000 ; Thivierge et al.,
2002) et au niveau splanchnique (Lobley et al., 1994) que le plasma est la source
principale des AA.
2.1.3.4 Débit dynamique
Le débit dynamique total représente l'utilisation totale d'un AA, qui dans le cas
le plus simple est égale à la somme de la synthèse de protéines et de l'oxydation dans
les tissus étudiés. Cette estimation ne peut se faire qu'en utilisant un AA marqué et le
calcul est basé sur l'extraction de cet AA marqué par un organe actualisé en AA non
marqué en divisant par l'EI de cet AA. Cela donne le mouvement total unidirectionnel
de la molécule (tracée + traceur) vers l'organe qui est égale à sa sortie dans un système
à l'équilibre.
L'UT-AA a été calculée a partir de l'équation suivante :
18
11) L'UT-AA au niveau d'un organe = (([AA]A x EIA) - ([AA]V
x EIV)) x DP/
Elv
où EIAA et [AA] représentent respectivement l'enrichissement isotopique (El:
mpe) de la molécule perfusée et la concentration plasmatique de l'AA étudié (uM), soit
au niveau artériel (A), soit au niveau veineux (V) et DP, le débit plasmatique de
l'organe étudié. Pour un AA ayant une voie métabolique simple, il est possible
d'estimer l'utilisation pour la synthèse protéique en soustrayant de l'utilisation totale
l'oxydation et la quantité dirigée vers d'autres métabolites. La leucine est souvent
utilisée comme un AA-traceur à cause de la simplicité de ses voies métaboliques avant
son oxydation (Figure 2.2). Nous l'utiliserons donc comme exemple.
Le MOP transformé en leucine et relargué dans la veine mammaire ou en CO2 a
été calculé comme suit:
12) MOP vers Leu ou CO2 (mmol/h) = (([MOPA] x EIMOP-A - [MOPV] x EIMOp-
v) x DPV) / (EIMOP-V)
L'oxydation est calculée en mesurant la différence entre les sites veineux et
artériel de la quantité de molécules de CO2 enrichies au 13C et en actualisant cette
quantité en CO2 total en utilisant l'EI du pool précurseur. L'oxydation est calculée de la
façon suivante :
13) Oxydation (organe ou tissu) = ((([CO2]V x El Co2-v) - ([CO2]A
x El CO2-A)) X
DS) / El MOP-V •) - correction pour l'oxydation du traceur.
Comme on fait l'hypothèse qu'une quantité équivalente à la quantité de la
leucine perfusée comme traceur est en excès dans l'animal et sera oxydée au niveau
corporel, on doit aussi soustraire de l'oxydation tissulaire une proportion de l'oxydation
corporelle que l'on estime équivalente au rapport des oxydations tissulaires sur
l'oxydation corporelle brute.
19
14) Finalement la synthèse protéique se calcule par différence entre UT-AA +
MOP vers Leu - Oxydation
La synthèse de protéines totales est utilisée pour fabriquer les protéines structurales,
fonctionnelles telles que les enzymes ainsi que les protéines d'exportation.
2.1.3.4.1 Choix du pool précurseur
Le choix du pool précurseur pour la mesure du métabolisme protéique au niveau
d'un tissu est aussi critique qu'au niveau corporel, mais demande toutefois une moins
grande généralisation et offre des options supplémentaires. Il est possible d'utiliser des
biopsies de tissus et d'y mesurer l'EI directement au niveau de l'ARN de transfert
(Airhart et al., 1974), lequel serait le précurseur le plus représentatif de la synthèse
protéique (Figure 2.3). Cette mesure étant techniquement très difficile à réaliser à cause
de très courte demi-vie de l'ARN, l'EI de l'AA au niveau intracellulaire est plutôt
utilisé lorsqu'il y a des biopsies (Connel et al., 1997). Il est aussi possible de mesurer
l'EI de l'AA étudié dans le drainage veineux (Figure 2.3) du tissu qui comporte, au
moins en partie, un apport d'AA provenant directement du tissu lui-même. Finalement
pour certains AA, il est possible de déterminer dans le drainage veineux l'EI d'un
métabolite formé intra-cellulaire qui devrait être un bon reflet de l'EI de l'AA au niveau
intra-cellulaire, tel le MOP pour la leucine (Matthews et al., 1982) ou l'homocystéine
pour la méthionine.
20
Absorption
Infusion
ArtèrePP
Intra-cellulaire
!Syntfièse
protqique
Oxydation
im
DégradationI
Protéines
Figure 2.3. Différentes possibilités de choisir le pool précurseur (PP) dans un modèlepour l'étude de la synthèse de protéines avec trois compartiments
2.2 Métabolisme protéique chez la vache laitière
L'utilisation totale des acides aminés comprend globalement la synthèse de
protéines et l'oxydation des AA. La synthèse protéique au niveau corporel est la somme
de la synthèse de protéines au niveau de tous les tissus et organes.
2.2.1 Distribution de la synthèse protéique au niveau corporel
Chez la vache laitière, la synthèse protéique totale varie de 3,5 à 5 kg/jour
(Bequette et al., 1996b ; Lapierre et al., 2002; Thivierge et al., 2002). En comparaison,
la quantité de protéines sécrétée dans le lait représente seulement 25 à 30 % de la
production corporelle totale de protéines. En supposant une rétention azotée nulle, la
différence de 70 à 75 % représente le turnover protéique, soit la synthèse de protéine et
la dégradation correspondante, indiquant l'importance du mouvement dynamique du
métabolisme protéique au niveau corporel.
La contribution du tissu gastro-intestinal et du foie à la synthèse corporelle de
protéines est d'environ 25 à 45 % chez le petit ruminant (Lobley et al., 1993; Lobley et
al., 1994), la vache laitière (Lapierre et al., 2002) et les non-ruminants (Sève et Ponter,
21
1997). Le tissu musculaire contribue seulement à 15 à 20 % de la synthèse protéique
chez un animal en croissance (Lobley et al., 1980), tandis que la synthèse protéique au
niveau mammaire représente 35 à 45 % (Bequette et al., 1996b ; Thivierge et al., 2002 )
de la synthèse protéique corporelle (Figure 2.4).
Comme nous l'avons mentionné dans cette section, la synthèse totale de
protéines par un organe, mesurée lors d'une perfusion continue, représente non
seulement les protéines constitutives mais aussi les protéines d'exportation. Dans la
glande mammaire, la proportion des protéines du lait sur la synthèse de protéines
d'exportation est d'environ 75 à 85 % (Bequette et al., 1996b ; Thivierge et al., 2002).
mammarygland
~ j % Protein mass
~1[% Pratein synthesis
':% Oxygen uptake
Figure 2.4. Contribution des différents tissus à la masse protéique (Protein mass), lasynthèse protéique (protein synthesis) et à l'utilisation nette d'oxygène (oxygen uptake)indicateur de l'énergie utilisée (tiré de Lobley et al., 2002).
2.2.2 Facteurs qui régulent la synthèse protéique
La synthèse de protéines est un processus très complexe. Il existe une série de
facteurs et de mécanismes pouvant la réguler. L'objectif de cette section est de discuter
des aspects fondamentaux impliqués dans la synthèse de protéines en relation avec les
facteurs liés à notre recherche. Chez le ruminant, une augmentation de l'ingestion
22
résulte en une augmentation de la synthèse totale de protéines (Lobley et al., 1987 ;
Lapierre et al., 1999). Mais est-ce la protéine ou l'énergie supplémentaire qui est
davantage responsable de ces modifications ?
2.2.2.1 L'apport en protéines
Une augmentation de l'ingestion de protéines entraîne habituellement une
augmentation de la synthèse de protéines au niveau corporel. Une perfusion post-
ruminale de caséine a entraîné une augmentation de la synthèse totale de protéines de 27
% chez le mouton (Liu et et al., 1995). De même, une augmentation de l'apport en
protéines métabolisables chez la vache laitière d'environ 15 % (Lapierre et al., 2002) et
de 26 % chez la truie (Reed et al., 1981), ainsi qu'une augmentation de l'ingestion de
protéines chez les non-ruminants (Garlick et al, 1991) ont entraîné une augmentation
substantielle de la synthèse protéique.
Chez la vache laitière, une mesure indirecte de la synthèse protéique peut être la
production de matières protéiques laitières. Une augmentation de l'apport en protéines
métabolisables par la ration (Blouin et al., 2002) ou la perfusion de caséine dans le
duodénum (Clark et al, 1977; Guinard et al., 1994; Vanhatalo et al., 2003a) entraîne
généralement une augmentation des protéines sécrétées dans le lait de 20 à 30 %. Tous
ces travaux nous indiquent que la synthèse protéique au niveau corporel est sensible à
l'apport protéique.
Des études in vitro ont démontré que les AA peuvent stimuler la synthèse de
protéines indépendamment des hormones (e.g insuline, IGF-1, somatotrophine et
glucagon; Campbell et al., 1999; Beugnet et al., 2003). Cet effet passerait via la
phosphorylation de la 4E-binding protéine-1 (4E-BP1) et de la 70-kDa protéine
ribosomal S6 kinase (p70 s6k) via la rapamycine mammalienne (mTOR). Ces
mécanismes sont les principaux mécanismes de régulation de la synthèse de protéines
au niveau cellulaire (Liu et al., 2006). Ces voies de signalisation sont présentes au
niveau du tissu mammaire (Choie et al, 2004; Hang et Rillema, 1997). Ce mécanisme
pourrait être celui par lequel les AA stimuleraient la synthèse et la production des
protéines chez la vache.
23
2.2.2.2 L'apport d'énergie
Qu'en est-il de l'impact de l'apport énergétique sur le métabolisme protéique ?
Les études sur le sujet sont beaucoup plus limitées que dans le cas précédent. En
utilisant la production de matières proteiques laitières comme un indice de la synthèse
protéique, il a été rapporté qu'un apport d'énergie sous forme de glucose (perfusé au
niveau du duodénum) ou d'acide propionique (perfusé au niveau ruminai) entraîne aussi
une augmentation de la production de matières proteiques du lait, mais moins prononcée
qu'avec l'accroissement de l'apport protéique (Clark et al., 1977 ; Lemosquet et al.,
1997; Hurtaud et al., 1998; Huhtanen et al., 2002; Sherperd et Combs, 1998). Le
mécanisme proposé est une meilleure utilisation des AA digérés, qui pourrait provenir
d'une diminution de l'oxydation des AA.
Au niveau corporel, Garlick et al. (1991) ont démontré une réduction de
l'utilisation totale des AA avec l'ajout d'énergie. Avec une perfusion d'acide
propionique, cette réduction était associée à une réduction de la dégradation protéique
(Abdul-Razzaq et Bickerstaffe, 1989). Reeds et al. (1981) ont aussi observé une
réduction du catabolisme de la leucine avec l'apport d'énergie.
Au niveau mammaire, les mécanismes par lesquels l'énergie pourrait affecter le
métabolisme protéique n'ont pas été étudiés directement avec des apports d'énergie.
Une perfusion intestinale de glucose à 2398 g/jour en substitution iso énergétique
(Rulquin et al., 2004) a résulté en une augmentation avec un effet quadratique de la
production des protéines du lait mais aussi une augmentation du prélèvement des AA
indispensables, ce qui suggère un autre destin de l'utilisation des ces surplus d'AA. Par
contre, une étude avec perfusion d'acide propionique ou de glucose sur une base
isoénergétique (Rigout et al., 2003) montre une augmentation des protéines dans le lait,
ce qui peut aussi suggérer des réponses similaires indépendamment de la nature du
précurseur énergétique.
Chez les monogastriques, l'insuline peut être liée à la synthèse de protéines,
principalement dans le muscle des animaux en croissance (Garlick et al., 1983). Par
contre, chez les animaux adultes, Melville et al. (1989) rapportent un effet non
significatif de l'insuline sur la synthèse protéique au niveau corporel, mais observent
24
cependant une réduction dans la dégradation des protéines. Aucun effet de l'insuline n'a
été observé sur les DAV mammaires chez la chèvre (Tesseraud et al., 1992), à l'opposé
de Mackle et al. (2000) qui ont observé une diminution des DAV pour l'histidine, la
méthionine et la phénylalanine plus la tyrosine, accompagnée d'une augmentation du
taux d'extraction et de la production de protéines du lait. Par ailleurs, Lemosquet et al.
(2004) suggèrent que l'IGF-1 pourrait être impliquée dans des mécanismes régulant
l'utilisation du glucose et indirectement les AA au niveau mammaire, l'IGF-1
intervenant dans la régulation du débit sanguin.
2.3 Métabolisme mammaire des acides aminés
2.3.1 Introduction
La production de lait et de protéines est très sensible à l'apport protéique dans
les rations (Doepel et al., 2004). Pour maximiser cette production, les producteurs
augmentent l'apport en protéines métabolisables. Cependant, les pressions sociales
exigeant de diminuer l'impact des productions animales sur l'environnement demandent
une réévaluation de cette pratique. Effectivement, lorsque l'on augmente l'apport en
protéines métabolisables, les rendements de transformation de la protéine ingérée en
protéines du lait diminuent selon une loi des rendements décroissants (INRA, 1989;
Doepel et al., 2004) et l'excès d'azote est principalement éliminé sous forme d'urée
excrétée dans l'urine, urée synthétisée lors du catabolisme que subissent les AA
absorbés. Les AA sont métabolisés au niveau de différents tissus, principalement au
niveau du foie, premier organe producteur d'urée, mais aussi, pour certains AA, au
niveau des tissus périphériques incluant la glande mammaire (Lapierre et al., 2005).
Ainsi à l'intérieur de chaque étude, l'augmentation de la production de protéines du lait
provoquée par la perfusion de caséines ou de certains AA (Clark et al., 1977; Rulquin,
1987; Rulquin, 1992; Guinard et Rulquin, 1994a; Bach et al., 2000) affectait de façon
variable les prélèvements nets mammaires des AA. Ces différences de réponses peuvent
en grande partie être liées à des différences de conditions expérimentales entre les
études. Notre objectif est de dégager des lois plus générales de réponses reliant l'apport
en protéines digestibles dans l'intestin (PDI) ou AA digestibles dans l'intestin (AA-DI ;
25
sommes des AA-DI = PDI) aux concentrations artérielles, aux DAV mammaires
(DAVM) et aux prélèvements mammaires (PM) des différents AA.
Pour raisonner sur le métabolisme intra-mammaire des AA, Mepham (1982) a
établi une première classification des AA selon les rapports entre la quantité nette
prélevée par la glande mammaire et celle sécrétée dans les protéines du lait. Cette
classification distingue 3 groupes. Le groupe 1 est composé d'AA sécrétés en quantité
quasi-équivalente à leur PM indiquant un transfert stoechiométrique, donc un ratio
prélèvement : sécrétion d'environ 1. Ce groupe comprend la méthionine, la
phénylalanine et un AA (partiellement) issu de son métabolisme la tyrosine, l'histidine
et le tryptophane. Pour ces AA, la quantité prélevée sur une base nette sert à expliquer la
quantité produite sous la forme de protéines du lait. Ce niveau de transfert rend
relativement improbable leur utilisation, sur une base nette, dans des voies métaboliques
autres que l'incorporation dans les protéines du lait.
Le groupe 2 correspond à des AA prélevés en excès par rapport à leur sécrétion,
qui peuvent donc être oxydés dans la glande mammaire : isoleucine, leucine, valine (les
AA à chaîne ramifiée), arginine, lysine et thréonine. L'excès représente une source
d'azote a-aminé et de squelettes de carbone, qui peuvent être utilisés pour la synthèse
des AA non indispensables (Lapierre et al., 2005) et possiblement d'autres métabolites
tels que l'acétyl-CoA et le succinyl-CoA et pouvant être oxydés comme source
d'énergie intra-mammaire (Clark et al., 1977; Bequette et al., 2005). Le groupe 3
correspond aux autres AA non indispensables (alanine, arginine, aspartate, cystéine,
glutamate, glutamine, proline et serine). Les AA de ce groupe sont prélevés de façon
très variable, et souvent en quantité insuffisante par rapport à leur sécrétion en protéines
du lait. Cette classification correspond-elle à une loi générale s'appliquant à une plage
de variation des apports de PDI allant d'apports ne couvrant pas les besoins en PDI à
des apports largement excédentaires aux besoins ?
Afin de décrire de manière quantitative les lois de rendements décroissants du
métabolisme des AA en réponse à l'accroissement d'apport de PDI ou en AA-DI, et de
déterminer la contribution de la glande mammaire à cette diminution de rendement,
nous avons choisi d'intégrer les résultats publiés de plusieurs études rapportant des
26
bilans nets mammaires avec une méta-analyse permettant d'éliminer l'effet lié aux
publications. Ce type d'approche quantitative a déjà été mené pour établir les lois de
réponses à l'augmentation d'apports d'AA perfusés individuellement dans la revue de
Rulquin et al. (1993 : pour lysine et méthionine). L'originalité de ce travail de méta-
analyse réside dans le fait que les lois seront établies sur un apport variable de PDI (un
ensemble d'AA-DI) et que les relations seront étudiées pour chacun des AA
individuellement mais aussi en considérant les familles décrites par Mepham (1982).
Les objectifs poursuivis ont été les suivants :
1) Déterminer la relation entre les concentrations artérielles en AA et l'apport en
AA-DI. Notre hypothèse était que la concentration artérielle reflétait en partie la
disponibilité de cet AA. Il s'agissait aussi de voir si cet accroissement de
disponibilité mesurée au niveau périphérique variait selon le groupe d'AA
considéré.
2) Déterminer si une augmentation d'AA-DI a un impact sur la DAVM qui, avec le
débit sanguin, compose le prélèvement net mammaire.
3) Déterminer comment évolue le prélèvement net mammaire des AA (individuels
ou en groupes) en fonction de l'augmentation des apports en AA-DI.
4) Vérifier si les groupes d'AA proposés par Mepham (1982) en relation avec leur
ratio prélèvement : sécrétion maintiennent le comportement suggéré lorsque
l'apport en protéines présente une large plage de variation.
2.3.2 Matériel et méthode
2.3.2.1 Formation de la base de données
Une première base de données a été constituée de 25 articles scientifiques, tous
portant sur le métabolisme mammaire des AA chez la vache laitière et publiés dans
différentes revues scientifiques internationales. À partir de cette base, notre principal
critère de sélection était que l'étude devait porter sur des traitements augmentant de
façon appréciable l'apport en PDI. Une première analyse a permis de montrer que
l'apport PDI n'était pas modifié dans les articles de Rulquin et al., 2004; Vanhatalo et
27
al., 2003b (sur le traitement de glucose), Huhntanen et al. (2002), Korhonen et al.
(2000), Metcalf et al. (1994), Lykos et Varga. (1997), Cant et al. (1993), Guinard et
Rulquin, (1994b et 1995), et Mackle et al. (2000). Dans les publications sélectionnées,
l'augmentation de l'apport en PDI a été obtenue par la perfusion post-ruminale de
caséine ou par un apport alimentaire en protéines très peu dégradées au niveau ruminai
(Prolak© et farine de poisson).
Les critères supplémentaires de sélection ont d'abord été basés sur les variables
mesurées rapportées dans les articles et sur les possibilités de recalculer les paramètres
manquants. Les articles devaient rapporter :
1. Les résultats zootechniques : ingestion des aliments et composition de la ration,
production laitière et de protéines du lait.
2. Contenir l'information pertinente sur les AA (indispensables et non
indispensables) : les concentrations artérielles, les DAVM et/ou les prélèvements
nets mammaires. Seuls les travaux de Metcalf et al. (1991 et 1994) présentaient
des données manquantes sur les concentrations. On notera que différentes
techniques de mesure et d'estimation du débit plasmatique ont été employées :
sonde débitmétrique à ultrasons (Guinard et Rulquin, 1994a), méthode de
dilution du para-aminoahippurate (Metcalf et al., 1991 et 1996) ou principe de
Fick utilisant deux « AA marqueurs » (Phe-Tyr ; Raggio et al., 2004 et 2006,
Miettinen et Huhtanen. (1997), Vanhatalo et al, 2003 a et b, Korhonen et al.,
2002).
Finalement, une vérification a été faite sur la base de données restantes pour
déterminer s'il y avait des articles dont les données étaient en dehors de l'intervalle
déterminé par la moyenne des observations ± 2 écart-types. Aucun article n'a été
enlevé sur cette base.
Compte tenu du nombre restreint de publications disponibles (8), nous avons
inclus les résultats de nos deux essais (Raggio et al, 2004 et 2006) dans cette méta-
analyse. Finalement, dix articles ont été retenus qui rencontraient nos critères de
sélection: Clark et al. (1977), Metcalf et al. (1991, 1994), Guinard et Rulquin.
(1994a), Korhonen et al., (2002), Miettinen et Huhtanen (1997), Vanhatalo et al.
(2003a, b), Raggio et al. (2004, 2006).
2.3.2.2. Recalculs effectués
Les recalculs ont permis de compléter des donnés manquantes (ex : le débit
plasmatique, les prélèvements nets d'AA). Ils ont aussi été effectués pour comparer les
résultats sur une base homogène.
a. Mesures du métabolisme mammaire
Dans les articles où le débit plasmatique n'était pas rapporté, il a été calculé avec le
principe de Fick appliqué à Phe+Tyr (Clark et al., 1977), en utilisant les DAVM et la
production des protéines rapportées. Pour cela, nous avons utilisé la composition des
AA des protéines brutes du lait proposée par Swaisgood (1995), considérant que les
protéines synthétisées par la glande mammaire représentaient 96,5 % des protéines du
lait (Cant et al., 1993).
Lorsque les prélèvements nets des AA n'ont pas été publiés ou donnés par les
auteurs (Guinard, communication personnelle), ils ont été recalculés en utilisant les
débits plasmatiques et les DAVM : Clark et al. (1977), Metcalf et al. (1991), Miettinen
et al. (1997), Vanhatalo et al. (2003a et b). Tous les prélèvements nets des AA ont été
recalculés pour une demi-mamelle, de manière à obtenir des données homogènes. Pour
pouvoir analyser de façon homogène le ratio prélèvement mammaire : sécrétion dans les
protéines du lait, ces ratios ont tous été recalculés en prenant une composition unique en
AA des protéines sécrétées dans le lait (Swaisgood, 1995). Nous avons également re-
exprimé l'ensemble des matières protéiques du lait sous forme de protéines brutes
(Guinard et Rulquin, 1994a; Raggio et al., 2006).
b. Calcul des valeurs alimentaires
Les apports en AA-DI ont été estimés à l'aide du système d'alimentation INRA, les
valeurs des tables d'alimentation INRA ainsi que les tables des AA-DI des aliments des
ruminants (Rulquin et al., 2001). Nous avons utilisé les articles rapportant la
composition biochimique des aliments à titre de référence pour réaliser le choix des
29
aliments à retenir dans les tables de référence. À défaut, toutes les valeurs relatives à la
composition des PDI, des AA-DI apportés par jour et à l'énergie apportée ont été
recalculées à partir des valeurs des tables de référence (Rulquin et al., 2001).
2.3.2.3 Modèle du calcul utilisé
Par l'intermédiaire de ces premières représentations graphiques des données
brutes, nous avons pu constater qu'il existait des relations intra-essai cohérentes entre
l'apport PDI et les paramètres du métabolisme des AA avec des pentes intra-essai
proches.
Notre objectif était de chercher une loi générale décrivant l'ensemble des
relations intra-essai observées en se départissant de différences liées au niveau de
production des vaches, à leur conduite expérimentale, aux techniques de mesure et
d'analyse de l'équipe.
C'est pourquoi à partir de ces dix articles, nous avons introduit dans le modèle
statistique d'analyse, le facteur «essai». Ce facteur est considéré comme «fixe». Dans ce
cas, chaque essai est considéré comme issu d'une population différente. L'apport PDI
était alors introduit comme une covariable :
Réponse linéaire : Yy = ESSAI + Ao + Bi X + e y
Réponse curvilinéaire: Yy = ESSAI + Ao + Bj X + Bi X2 + e y
Où X = est l'apport PDI ou l'apport AA-DI
Y = le paramètre étudié
2.3.3 Résultats
Sur cette base de données, la composition en AA exprimée en pourcentage de
PDI dans la ration a varié mais seulement entre essais (Tableau 2.1). Cette composante
ne pourra donc pas être prise en compte dans la présente analyse, vu le faible niveau de
variation à l'intérieur de chacune des études. Dans la base de données utilisée, la
production laitière (PL) variait de 20,8 à 36,2 kg/j avec une production de protéines du
lait de 686 à 1109 g/j et une augmentation de PDI de 1323 à 3090 g/j (Tableau 2.2).
Dans cette base de données obtenues chez des vaches opérées multi-cathérisées, nous
30
avons obtenu un rendement d'augmentation de la production de matières protéiques du
lait de 28 en réponse à l'apport PDI (Figure 2.5). Ce rendement est tout à fait
comparable à celui obtenu dans la littérature plus abondante sur vaches non opérées,
recevant une alimentation protéique conforme aux recommandations (Doepel et al.,
2004; Hanigan, 2005).
Acides aminés, %PDI
ArginineAspartateGlutamineGlycinePralineSerineHistidineIsoleucineLeucineLysineMéthioninePhénylalanineThréonineTyrosineValine
Minimum4,49,912,34,94,44,62,04,88,36,41,64,74,74,15,4
Maximum5,311,417,86,67,35,62,65,710,18,32,65,76,05,06,4
Variation ( %)211545346623331922286322272418
Tableau 2.1. Variation inter-essai, de la composition en AA des PDI dans les essaisutilisés pour constituer la base de données
VariableIngestion1, kg/dPL2, kg/dPDI3 g/dPB4, %
g/d
Moyenne18,828,3
1882,832,0
896,9
Minimum15,320,8
1323,924,6
686,4
Maximum24,736,2
3090,538,2
1109,0
É. Type2,44,4
415,73,24,6
CV %12,915,622,1
9,811,2
n2727272727
'Matière sècheProduction du lait2Protéines digestibles dans l'intestin estimées par le système PDI (INRA, 1989)'Protéine brute du lait ( % et g par jour)
Tableau 2.2. Caractéristiques zootechniques de la base de données
C.Pioteine biule [
1000 1500 2000 2500 3000 3500
! 27 ports / 27 données | PDI (g/j)
1300
I2IXI
11(10
1000
31»
1100
m
1)00
tProleine biule (g'i l
1000 1500127 ponts/27 durées |
2(100 2500 3000 350O
PDI W,i
Concentration en protéines brutes du lait (%) = 0.0034 X (P < O.OO1) + 26.116 (P < 0.001), r2 0.96, Syx 0.36, essai P < 0.001.Production de protéines brutes du lait (g/d) = 0.2782 X (/>< 0.001) + 358.45 (P < 0.001), r20.96, Syx 0.35, essai P< 0.001.
Figure 2.5. Augmentation de la production de protéines brutes dans le lait en fonctiondes apports PDI
Comme une des questions était de vérifier si la classification de Mepham (1982)
sur le comportement des AA se maintenait avec des apports variables en protéines, nous
avons examiné les relations étudiées à l'intérieur de chaque groupe, c'est-à-dire pour
chaque AA pris individuellement et en faisant la somme de tous les AA du même
groupe. Comme le comportement des AA à l'intérieur de chaque groupe était
effectivement relativement similaire, nous avons choisi pour simplifier la présentation,
de présenter les résultats d'un AA de chaque groupe avec la somme de chaque groupe
pour les AA indispensables. Quant aux AA non indispensables du groupe 3, les
comportements étant tellement variables d'un AA à l'autre, nous avons voulu éviter de
faire une somme de groupe mais présentons plutôt deux AA ayant des comportements
très différents : l'arginine et la glutamate. Pour le glutamate, il est important de savoir
que le glutamate DI représente la somme du glutamate et la glutamine. Il est à noter que
comme l'arginine n'est pas un AA indispensable dans le sens strict du terme (il y a une
synthèse endogène importante), et que de plus elle avait un comportement différent des
autres AA du groupe 2 dans le ratio PM : PL, nous l'avons exclue de ce groupe et
incluse dans le groupe 3.
32
2.3.3.1 Concentrations artérielles
Les AA des trois groupes répondent de façon différente à une augmentation de
la quantité d'AA-DI. A titre d'exemple, l'histidine, classée dans le groupe 1, retrouve
une réponse curvilinéaire (P < 0,01) d'augmentation des concentrations artérielles avec
un accroissement des AA-DI, avec un coefficient de détermination (r2) égal à 0,94
(Figure 2.6). Pour la somme des AA du groupe 1, pour des augmentations d'apport en
AA-DI de ce groupe allant jusqu'à 80 mmol/h, la réponse de la concentration artérielle
augmente de 2,27 uM par mmol/h d'AA DI du groupe 1 apporté. Pour des apports
supérieurs, cette réponse diminue, ce qui traduit un effet curvilinéaire (P < 0.01) avec
un coefficient de détermination élevé de 0,97.
CJliilHliwAluMl
7.0 9.5 12.0 14.5 17.0 19.5( 27 ports / 27 données ] Hitlidine DI [mmo/h)
m
225
200
1/5
150
175
100
C.Groupe 1 A |uM)
45 50 55 60 65 70 75 90 95 90 95 100127 ponts / 27 données | Groupe 1 DI (mnol/h]
|A| jiM Histidine = -0,456 X2 (P < 0,01 )+ 16,174 X (P < 0,001) - 83,412 (P < 0,01 ), r2= 0,94, Syx 5,59, essai P = 0,06.Groupe 1 = -0,002 X2 (P = 0,007) + 5,635X (P = 0,02) - 94,679 (P = 0,02), r2= 0,97, Syx 7,86, essai P < 0,01.
Figure 2.6. Effet des apports en PDI sur les concentrations artérielles des AA du groupe1
Pour la leucine et la somme des AA du groupe 2, il existe une relation linéaire
positive (P < 0,001 : r2 = 0,92 et r2 = 0,97) entre les concentrations artérielles et l'apport
d'AA-DI (Figure 2.7). Une des différences avec le groupe 1 reste cependant l'amplitude
de l'augmentation d'AA-DI du groupe 2, variant de 120 à 250 mmol/h et résultant en
une augmentation linéaire de la concentration artérielle de 4,37 uM par mmol/h d'AA
DI du groupe 2 (Figure 2.7).
33
250 r
« I
150
II»
C.LeuciiwA|uM)
35 45 55 G5
[ 27 ponls/27 donnés!]
75 85
l.eucine Dl |m»nl/h]
) ( 0 ) C_G.om.2A[uM|
875
750
625
500
250
100 125 150 175 200 225 250 275 300
127 ports/27 données] Groupe 2 Dl (maol/h]
[A] jiM Leucine = 3,5601X (P < 0,001) - 55,786 (P < 0,01), r = 0,92, Syx 15,20, essai P = 0,07.Groupe 2 = 4,3757X (P < 0,001) - 173,84 (P < 0,001), r2 = 0,97, Syx 36,93, essai P < 0,001.
Figure 2.7. Effet des apports en PDI sur les concentrations artérielles des AA du groupe2
Les concentrations artérielles des deux AA du groupe 3 que nous avons décidé
d'étudier ne répondent pas de façon similaire aux augmentations d'apports. Dans le cas
de l'arginine, nous observons aussi une relation positive en réponse à une augmentation
en arginine Dl (P < 0,001), l'augmentation de concentration étant comprise entre 45 et
105 uM en réponse à une augmentation de l'arginine Dl de 12,5 à 25 mmol/h (P <
0,001; r2 = 0,94. Figure 2.8). Par contre, la réponse de la concentration artérielle du
glutamate est une courbe curvilinéaire négative (P = 0,02), avec un coefficient de
détermination plus faible (r2 = 0,60. Figure 2.8).
105
95
m
m
65
59
C.Aig«»A(uM]1,1111
57.S
550
52.5
5(1.11
4/.!i
C_GUamatsA[uM]
12.5 15.0
124 poinls / 24 données ]
20.0 22.5 25.0
Aigimne Dl («raol/h]
55 G0 G5 70 75 80 05 90 95 100 105 110
124ports/24données ] i i lul • Dl (ramal/h)
34
[A| 11M Arginine = 4,0415 X (P = <0,001) - 5,3716 (P = 0,6), r2 = 0,938, Syx 4,49, essai P < 0,001Glutamate = 0,0066 X2 (P = 0,02) - 1,1761 X (P = 0,01) + 102,99 (P < 0,001), r2 = 0,60, Syx 2.9, essai P =
0,02
Figure 2.8. Effet des apports en PDI sur les concentrations artérielles d'AA nonindispensables
2.3.3.2 Différences artério-veineuses mammaires
Avec les AA du groupe 1 (et histidine), on observe une distribution curvilinéaire
(P — 0,10) de la DAVM en réponse à l'augmentation d'AA-DI, mais plus hétérogène,
avec un r2 =0,73, qu'avec la concentration artérielle (Figure 2.9).
C^HiilidineAVIuM)
5.0 7.5 10.0 12.5 15.0 17.5 20.0127 ports / 27 données | Hillidi» PDI [mnol/h]
80
75
71)
1,5
60
56
M
45
41)
127
C.Groupe 1 AV |uM|
45 50 55 C0 69 70 75 80 85 90 35 100
ohls / 27 données | Groupe 1 Dl (mmal/h)
|DAVM|nM Histidine =-0,091 X2 (P = 0,05) + 2,782 X (P = 0,02) - 7,046 (P = 0,02), r2 = 0,68, Syx 1.7, essaie P = 0,01Groupl = -0,0122 X2 (P = 0,10)+ 2,286 X (P = 0,04) - 34,399 (P = 0,1), r2= 0,73, Syx 7,3, essaie P = 0,07
Figure 2.9. Effet des apports en PDI sur les différences artério-veineuses mammaires(DAVM) des AA du groupe 1
Pour les AA du groupe 2, la réponse des DAVM en fonction de l'apport en AA-
DI reste linéaire (P < 0,001) avec un coefficient de détermination élevé (r2 = 0,90:
Figure 2.10). Il en résulte une augmentation linéaire de la DAVM augmentant de 1,34
uM par mmol/h des AA-DI du groupe 2 apportés. La leucine suit un schéma très
similaire, tout comme les autres AA de ce groupe (Figure 2.10).
35
100 r
CJ.eucineAV[uMl
35 40 «5 50 55[ 27 point/ 27 (tonnées ]
65 70 75 80 85
Leucine PDI (mnal/hl
300
150
C_Gioupe2 AV (uM)
100 150[ 27 ports / 27 données ]
250 300
Gii«i|>e2 PDI |mm)l/h)
IDAVM) uM Leucine = 1,195 X (P < 0,001) - 3,2402 (P = 0,3), r2l= 0,90, Syx 6,76, essaie P = 0,01.Groupe 2 = 1,3426 X (P < 0,001) - 21,675 (P = 0,5), r2 = 0.90, Syx 22,27, essaie P < 0,01.
Figure 2.10. Effet des apports en PDI sur les différences artério-veineuses mammaires(DAVM) des AA du groupe 2
Dans le cas de l'arginine, on observe que les DAVM ont augmenté de façon
linéaire (P < 0,01) de 15 à 40 uM pour une augmentation d'arginine DI d'environ 12
mmol/h, avec un r2 de 0,79 (Figure 2.11). Le glutamate montre une réponse
curvilinéaire négative (P = 0,07; Figure 2.11) entre les DAVM et les apports en
glutamate DI, mais avec un faible coefficient de détermination de (0,36).
C_ArgraeAV|uM|
12.5 15.0
12* poils H4 données |
17.5 20.0 22.5 250
AiginneDIImnolAi)
CGMamaleAVIuM]
8S 95 105
Glutamate 01 lranol/h|
|DAVM] jiM Arginine = 1,6499 X (P < 0,01 ) - 0,1912 (P = 0,8), r2 = 0,79, Syx 3,68, essaie P = 0,26Glutamate = 0,0047 X2 (P = 0,07) - 0,7915 X (P = 0,06) + 70.64 (P = 0.6), r2 = 0,35, Syx 2,27, essaie P = 0,03
36
Figure 2.11. Effet des apports en PDI sur les différences artério-veineuses mammaires(DAVM) des AA non indispensables.
2.3.3.3 Le prélèvement mammaire des AA
Pour l'ensemble des AA, les formes de courbes obtenues reliant les apports en
AA-DI et les prélèvement sont proches de celle obtenues sur les DAVM, mais les
relations avec les débits sont mieux corrélées (valeurs de P plus significatives et r2 plus
élevés) que les relations avec les DAVM seulement. Pour l'histidine et la somme de
tous les AA du groupe 1, on observe une réponse curvilinéaire (P - 0,02) du
prélèvement mammaire en fonction de l'augmentation de l'AA-DI. Toutefois, la
curvilinéarité (P = 0,02) de la réponse est moins accentuée (coefficient du X2 0,024 vs
0,003 pour le PM de l'histidine vs Groupe 1 respectivement ; Figure 2.12) quand on
tient compte de tout le groupe 1, mais toujours avec la même tendance (r2 = 0,87).
4.5
1.0
C_Hiitidine UP (nnol/hi
5.0 7.5 W.0 12.5 15.0 17.5 20.0
[ 27 pouls / 27 données | Hiilidm Dl [mnuil/h]
C_Gioi*e1UP|mnDl/h)
«5 55
|27 pouls/27 données |
65 75 85 95
Groupe 1 Dl (mnol/hl
PM mmol/h Histidine = - 0,0241 X2 (P = 0,02) + 0,749 X (P < 0,01) - 1,784(P = 0,1), r2 = 0.791, Syx 0,36, essai P <0,01.Groupe 1 PM = - 0,0029 X2 (P = 0,04) + 0,5579 X (P < 0,01) - 6,2935 (P = 0,3), r2 = 0.87, Syx 1.26, essai P < 0,01
Figure 2.12. Effet des apports en PDI sur le prélèvement mammaire (PM) des AA dugroupe 1
Par contre, pour tous les AA du groupe 2, la seule réponse sur le PM net que
nous avons observée était linéaire (P < 0,001; r2 = 0,92), ce qui indique une relation qui
ne diminue pas à des apports élevés, comme on l'avait observé pour les AA du groupe 1
(Figure 2.13).
37
C_LeucmeUP|iMiol/h| C.Gioupe 2 UP (nnol/h)
35 45
[ 27 pmnls / 27 dcnnées 1
75 86I eucine Dl (nmnf/h)
100 125 150 175 200 225 250 275
|?7 ponts/27 domées] Groupe 2 Dl (mnoWll
PM mmol/h Leucine = 0,315 X (P < 0,001 ) + 0,291 (P =0,6), r2 = 0,86, Sxy 1,98, essai P < 0,01Groupe 2 = 0,0891 X (P <0,001 ) - 0,4497 (P = 0,7), r2 = 0,92, Sxy 1,32, essai P < 0,001
Figure 2.13. Effet des apports en PDI sur le prélèvement mammaire (PM) des AA dugroupe 2
Dans le cas de l'arginine, la réponse est linéaire et positive (P = 0,05), aussi avec
un coefficient modéré de détermination (r2 = 0,58; Figure 2.14). Pour les acides aminés
non indispensables, par exemple pour le glutamate, on observe une relation globale
négative mais non significative linéaire puisque la pente négative n'est pas différente de
zéro (P = 0,12) avec un faible r2 de 0,23 (Figure 2.14).
C_Aiginine UP [nanol/hl
12.5 15.0 17.5 20.0 22.5 25.0
124 parts / 24 données ] Arginine Dl |mnnl/h)
13.50
12.75
12.10
11.26
10.M
9.B
9.00
C.GMamalsUP[mnl/h]
J
50 GO 70
) 21 points / 24 donnéES ]
90 100 110GlutamaleDI[wnoiyh|
PM mmol/h Arginine = 0,3337 X (P = 0,05) + 2,4621 (P = 0,36), rz = 0,58, Sxy 1,22, essai P = 0,02
Glutamate PM = -0,034 X (P= 0,12) + 14,002 (P = 0,02), r2 = 0,23, Sxy 1,43, essai P = 0,03
Figure 2.14. Effet des apports en PDI sur le prélèvement mammaire (PM) des AA nonindispensables
2.3.3.4 Le ratio prélèvement mammaire sur sécrétion en protéines du lait
Dans le groupe 1, on observe une réponse curvilinéaire (r2 = 0,37) entre le ratio
PM sur la sécrétion en protéines du lait (PM: PL) en fonction de l'apport en AA-DI
quand on utilise l'histidine comme représentant du groupe (Figure 2.15). Dans tout le
groupe 1, la relation entre le ratio et l'apport PDI est linéaire, avec aussi un faible
coefficient de détermination (r2 = 0,26; Figure 2.15). Les ratios sont compris entre 0,7 et
1,2 pour la moyenne de tous les AA du groupe 1, avec une pente qui a tendance à être
différente de zéro (P = 0,08), mais un très faible r2 de 0,26. Cependant, pour la majorité
des AA du groupe 1, le ratio se distribue de 0,89 à 1.1. Il y a deux observations
provenant d'une même expérience où le débit a été mesuré avec une sonde à ultrason
(Guinard et Rulquin, 1994a) qui se situait en dehors des autres : sans ces deux
observations, la pente n'est plus différente de zéro (P = 0,89).
i! il 8.5 11.0 13.5 11,11 18.5[ 31 ponts / 31 données ] Hitlidine Dl (mmol/h)
C. Groupe 1 rai»
1.2
1.1
1.0
0.3
11.»
0 7
45 55[ 27 porits / 27 données ]
85 95Groupe 1 Dl {imnl/h)
Ratio PM:PL Histidine = -0,006 X2 (P = 0,08) + 0,17 X (P = 0,05) - 0,57 (P = 0.9); r2 0,37, Sxy 0,13, essai P <0,01
Groupe 1 = -0,0027 X (P = 0,08) + 0,799 (P = 0,8); r2 = 0.26, Syx 0,08, essai P = 0,01
Figure 2.15. Effet des apports en PDI sur le ratio prélèvement mammaire et sécrétion enprotéines du lait (PM: PL) des AA du groupe 1
Pour le groupe 2, on observe une réponse linéaire positive avec un r2 = 0,76
(Figure 2.16). Cette réponse a une pente différente de zéro (P < 0,01) qui indique que le
prélèvement net mammaire est bien supérieur aux quantités sécrétées dans les protéines
du lait. Cette relation linéaire traduit un accroissement plus rapide du prélèvement net
d'AA par rapport à la quantité sécrétée dans le lait lorsque l'apport en AA-DI augmente.
1.75
1.M
1.25
1.00
0 7!,
II.M
200
1.75
1.50
1.K
100
0.75
II Ml
CJiloupe 2 tdln
35 « 45 50 55 60 E5 70 75 G0 85
127 pointe / 27 données ] Leucine DI [mniol/h]
100 125 150 175 200 225 250 275
127 ports / 27 données | Bioupe 2 Dl (mnol/h)
Ratio PM:PL Leucine = 0,0149 X (P < 0,01) + 0,4242 (P = 0,08;, r2= 0,77, Syx 0,14, essai P = 0,03Groupe 2 = 0,0048 X (P < 0,01) + 0,5336 (P= 0,02), r2 = 0,76, Syx 0,13, essai P < 0,01
Figure 2.16. Effet des apports en PDI sur le ratio prélèvement mammaire sur protéinesdu lait (PM : PL) des AA du groupe 2
Pour les AA non indispensables, les ratios PM : PL sont variables. L'arginine
présente un prélèvement supérieur à sa sécrétion dans le lait, avec des valeurs entre 1,52
à 2,78, mais avec aucune relation entre l'apport en arginine DI et le prélèvement (r2 =
0,003; Figure 2.17). Par contre, pour le glutamate, les valeurs sont inférieures à 1 (0,6 à
0,99) et négativement reliées (r2 = 0,58) à l'apport en glutamate DI (Figure 2.17).
40
3.00
2.75
2.50
2.25
2.00
1.75
1 50
CAiginine ratio
-
-
-
-
-
m126 ports/ 26 données!
l«
t i
• .
*
*|
1b
•
1 1
4 * •
i t
20 25Aiginine Dl (nmnl/h]
1.1
1.0
0.9
C_Slula<i>ate iatn
0.5
50 60124 ponts/2« données |
90 nui 110Glutamate Dl (nmolVhl
Ratio PM:PL Arginine = 0,0052 X (P = 0,91 ) + 2,1441 (P = 0,02), r2 = 0,003, Syx 0.30, essai P = 0,09Glutamate = - 0,0059 X (P< 0,01) + 1,2254 (P <0,01), r2 = 0,58, Syx 0.14, essai P =0,10
Figure 2.17. Effet des apports en PDI sur le ratio prélèvement mammaire et protéines dulait (PM : PL) des AA non indispensables
2.3.4 Discussion
La concentration artérielle en réponse à une augmentation de l'apport en PDI (ou
AA-DI) varie selon le groupe d'AA. Pour les acides aminés du groupe 1, la réponse est
curvilinéaire, avec une première partie de réponse linéaire. Les AA de ce groupe qui
sont prélevés par la glande mammaire en quantité équivalente à la sécrétion en protéines
du lait sont principalement métabolisés par le foie (Hanigan, 2005; Lapierre et al.,
2005). À des apports plus faibles, l'approvisionnement post-hépatique augmenterait
jusqu'à l'atteinte d'une certaine limite au-delà de laquelle le foie régulerait de façon
plus serrée le catabolisme de façon à éviter une hyperaminoacidémie. Avec le groupe 2,
avec les niveaux d'apport dans ces études, la réponse des concentrations artérielles à des
apports croissants de AA-DI demeure strictement une réponse linéaire. Ces AA ne sont
que peu catabolisés par le foie (Hanigan, 2005; Lapierre et al., 2005) mais plutôt par les
tissus périphériques, ce qui permettrait à une augmentation de l'apport en AA-DI
d'atteindre la circulation périphérique avant d'être catabolisée. La raison métabolique
de ces différences dans les principaux sites de catabolisme des AA indispensables n'est
pas claire, mais il a été proposé que certains AA, spécialement la leucine, agissent
comme un signal métabolique indiquant aux organes périphériques le niveau
41
d'approvisionnement en AA, agissant même en interaction avec les hormones pour
stimuler la synthèse protéique (Wray-Cahen et al., 1998). Cette démonstration a été
clairement faite chez le non ruminant, mais n'a pas été aussi clairement démontré chez
le ruminant (Wester et al., 2004). Il n'y a pas une réponse unique pour les AA du groupe
3, qui sont absorbés mais aussi produits par l'organisme. Les concentrations artérielles
des deux AA présentés (arginine et glutamate) ne reflètent plus seulement l'absorption
mais traduisent aussi le métabolisme corporel (demande et synthèse).
Les concentrations artérielles de l'arginine, bien que la synthèse de novo puisse
représenter jusqu'à 40 % de l'approvisionnement global, sont très reliées à l'arginine
DI, suggérant qu'une augmentation de l'apport digestif ne supprime pas la synthèse
endogène. Le glutamate offre une relation toute autre entre ses concentrations et
l'apport digestif. La diminution des concentrations en glutamate suite à une
augmentation de l'apport digestif en glutamate, parallèle à l'augmentation en PDI, peut
être reliée au rôle des AA non indispensables comme la glutamine à servir de «
réservoir azoté » lors de carence en azoté (Bergen, 1979; Lacey et Wilmore, 1990)
expliquant ainsi l'augmentation observée à des apports plus faibles. Cette diminution
observée avec l'augmentation en PDI pourrait aussi être associée à une demande plus
importante en glucose suite à l'augmentation de la production de lait (lactose) et aussi à
l'augmentation des besoins énergétiques pour couvrir l'accroissement de la synthèse
protéique.
Pour les AA du groupe 1, la DAVM et la quantité prélevée répondent de façon
curvilinéaire à l'apport d'AA-DI. En fait, cette relation curvilinéaire s'explique en partie
par une relation linéaire entre la concentration artérielle et la DAVM qu'il sera
intéressant d'exploiter dans le futur. La forme curvilinéaire de ces réponses reste à
expliquer par rapport à la relation linéaire entre l'apport PDI et la protéine du lait et le
ratio PM : PL s'approchant de l'unité. Les réponses curvilinéaires ne sont pas en
relation parfaite entre l'apport PDI et la protéine sécrétée dans le lait et l'analyse de la
relation PM : PL, peut-être à cause d'une accumulation des erreurs expérimentales et
aussi d'une grande variabilité de données observées. De tout manière, on n'observe pas
de relation (r2 = 0,26) entre l'augmentation de AA-DI et le ratio PM : PL, ce ratio nous
42
indiquant que le prélèvement mammaire des AA du groupe 1 s'ajuste bien avec la
quantité nécessaire pour la production lactée. Cette classification correspond à celle de
Mepham (1982) du groupe 1. Bien que la variabilité des ratios aille de 0,7 à 1,2
(Guinard et Rulquin 1994a), si on supprime 2 observations extrêmes, la variation
diminue de 0,9 à 1,03, ce qui indique un transfert prélèvement : sécrétion dans le lait
très près de l'unité, confirmée par une pente non différente de zéro (P = 0,89). Donc,
même avec des apports variables en PDI, la relation du transfert stoechiométrique est
maintenue. Cependant, pour le groupe 1, l'une des limites de notre analyse était que sur
les 10 articles, 8 étaient basés sur un débit sanguin estimé par le principe de Fick
appliqué à Phe+Tyr qui ajuste le ratio à 1 pour ceux-ci.
Pour le groupe 2, on observe une réponse linéaire entre les DAVM ou les
prélèvements et l'apport AA-DI. Cette réponse indique qu'avec des apports plus grands,
les prélèvements mammaires continuent d'augmenter de façon linéaire. Ce prélèvement
devient donc bien supérieur aux quantités sécrétées, indiquant un métabolisme intra-
mammaire majeur de ces AA. Cet excès de prélèvement peut être oxydé (Bequette et
al., 1996a) et l'azote des ces AA utilisés pour la synthèse d'AA non indispensables
(Lapierre et al., 2005). Ces résultats indiquent clairement que la glande mammaire peut
devenir un site important de catabolisme des AA du groupe 2, catabolisme qui
augmenterait en fonction de l'apport en AA-DI.
Pour les AA non indispensables, la réponse est très variable et une généralisation
de groupe ne peut être dégagée. Pour l'arginine, largement prélevée en excès par rapport
aux besoins en sécrétion de protéines du lait, l'absence de relation claire entre le ratio du
PM : PL ne permet pas vraiment de tirer de conclusion sur les besoins métaboliques de
la glande mammaire. Le ratio PM : PL du glutamate répond de façon négative, avec une
diminution des prélèvements à mesure qu'augmente l'absorption. Cette variation
pourrait être attribuée au rôle de navette azotée du glutamate, transportant au foie les
excès d'N, devant être détoxifiés en urée, prélevés sur les acides aminés du groupe 2
non-utilisés directement pour la synthèse protéique. Une augmentation du ratio PM : PL
de la lysine augmenterait de façon importante la synthèse de glutamate intra mammaire
(Lapierre et al., 2003 et 2005).
L'une des limites à cette méthode expérimentale est que tous les prélèvements
mammaires sont faits à partir des AA libres. S'il y a prélèvements de peptides par la
glande mammaire, et ce plus vraisemblablement à des apports faibles en PDI, la DAVM
sera sous-estimée. Si tel est le cas, le débit plasmatique calculé à partir du principe de
Fick sera surestimé. L'impact final sur le calcul du prélèvement mammaire sera une
surestimation s'il y a des peptides prélevés et ce biais pourrait donc varier avec l'apport
en PDI. Par contre, les prélèvements mammaires calculés avec les sondes
débitmétriques pourraient être sous-estimés si en plus des AA libres il y a
approvisionnement à la glande mammaire d'AA sous forme de peptides. Le calcul du
débit plasmatique avec le principe de Fick suppose aussi qu'à un apport élevé, il n'y a
pas de catabolisme des AA marqueurs dans la glande mammaire mamelle. Si tel n'était
pas le cas, le débit serait par ailleurs sous-estimé. Dans le travail de Bequette et al.
(1999), il a été estimé chez la chèvre que l'utilisation par la glande mammaire de
peptides contenant de la phénylalanine, de la tyrosine et de la méthionine était
dépendante du niveau circulant d'AA libres. Cependant, la différence de ratio d'EI
artériel et veineux entre le groupe 1 et 2 attribués au prélèvement de peptides pourrait
aussi provenir du prélèvement excédentaire à la sécrétion des AA du groupe 2
comparativement au groupe 1. Cependant ces conclusions restent à démontrer chez la
vache et la différence d'EI observée entres les AA pourrait ne pas impliquer de
prélèvement de peptides, à rencontre de l'hypothèse avancée par ces auteurs.
2.3.5 Conclusion
Dans ce travail de méta-analyse, on a observé que le comportement des groupes des AA
proposés par Mepham (1982) se maintenait sur une large plage d'apport PDI. Il est
intéressant d'observer le même type de réponses à l'intérieur de ces groupes d'AA. Pour
le groupe 1 : la réponse de la concentration artérielle et de prélèvement mammaire a été
curvilinéaire à des apports de PDI, ce qui se traduit par un métabolisme des ces AA au
niveau hépatique. Par contre, pour le groupe 2, la concentration artérielle et le
prélèvement mammaire, montre une réponse linéaire, mais avec un ratio PM : PL qui
augmente, indiquant une oxydation grandissante au niveau mammaire. Le ratio PM : PL
pour le groupe 1 n'est pas significativement différent de 1, ce qui démontre un transfert
44
stoechiometnque du prélèvement de ces AA dans le lait maintenu à des apports
variables d'AA-DI.
45
2.4 Objectifs de thèse
Des travaux antérieurs ont démontré qu'une augmentation des apports protéiques
entraîne généralement une augmentation de la production de matières protéiques du lait,
mais avec une efficacité marginale décroissante (Doepel et al., 2004). Ainsi, elle
entraîne une diminution de l'efficacité de la transformation des protéines digérées en
protéines de lait. Cette diminution d'efficacité doit-elle être reliée à une augmentation
du catabolisme des AA, principalement au niveau hépatique ? Par ailleurs, comment
sera affectée l'utilisation des AA au niveau mammaire ? Les objectifs de la thèse sont
donc :
Objectif 1 : Déterminer comment une augmentation de la protéine
métabolisable par la ration affecte l'extraction hépatique des AA essentiels et
l'utilisation de ces AA par la glande mammaire. Il s'agira de relier l'amplitude des
variations des extractions hépatiques avec l'efficacité «absorption-production de
protéines de lait».
Il n'y a très peu de données dans les travaux antérieurs sur la production des
protéines plasmatiques, principalement synthétisées par le foie et exportées dans le
plasma, chez les ruminants (Connell et al., 1997). Comment va être affecté le taux de
synthèse de ces protéines par différents niveaux de protéines métabolisables ingérées ?
Quelle sera la priorité métabolique de cette synthèse chez une vache laitière exigeant un
apport considérable d'AA pour maintenir sa production laitière ? Par ailleurs, quel sera
le pourcentage d'extraction hépatique d'un AA largement prélevé par le foie qui sera
utilisé pour la synthèse de ces protéines plasmatiques plutôt que d'être catabolisé ?
Nous avons donc émis l'hypothèse que le taux de synthèse des protéines d'exportation
ne varie pas en fonction des apports protéiques.
Objectif 2 : Déterminer comment varie le taux de synthèse de protéines
plasmatiques à des apports bas et élevés en protéines métabolisables et définir la
contribution de l'extraction hépatique d'un AA à cette synthèse plutôt que vers le
46
catabolisme. Il s'agit de définir la priorité métabolique de la synthèse protéique
(synthèse de protéines plasmatiques vs. de protéines du lait) à des apports restreints en
protéines et d'examiner comment le foie réagit, en terme d'extraction hépatique, pour
répondre à cette demande.
Une augmentation des apports protéiques entraîne une augmentation de la
production de protéines de lait. Aussi, il est suggéré qu'une augmentation des apports
d'énergie a un impact sur la production de protéines de lait (Hurtaud et al., 1998).
Cependant, les mécanismes d'action au niveau corporel et mammaire amenant cette
réponse ne sont pas encore bien connus. S'agit-il des mêmes mécanismes ? Existe t-il
un effet additif entre les deux nutriments ? Nous avons donc émis l'hypothèse que la
protéine et l'énergie ont des mécanismes d'action différents pour stimuler la production
de protéines du lait et auront ainsi un effet additif ou synergique sur le métabolisme
protéique chez la vache laitière.
Objectif 3 : Déterminer si des apports en caséine et en acide propionique ont des
effets additifs sur le métabolisme protéique aux niveaux corporel et mammaire. Il s'agit
également de comprendre quelle part des augmentations de production de matières
protéiques du lait s'explique par des modifications du métabolisme mammaire des AA.
En réponse à l'apport individuel ou combiné de ces deux types de nutriments,
l'utilisation totale aux niveaux corporel et mammaire de la leucine et la partition de
cette utilisation vers la synthèse de protéines ou l'oxydation nous permettra d'évaluer le
niveau d'interaction entre ces nutriments sur le métabolisme protéique.
47
2.5 Liste des ouvrages cités
Abdul-Razzaq, H. A et R. Bickerstaffe. 1989. The influence of rumen volatile fatty
acids on protein metabolism in growing lambs. Br. J. Nutr. 62:297-310.
Airhart, J., A. Vidrich et E. A. Khairallah. 1974. Compartmentation of free amino acids
for protein synthesis in rat liver. Biochem. J. 140:539-545.
Bach, A., G. B. Huntington, S. Calsamiglia et M. D. Stern. 2000. Nitrogen metabolism
of early lactation cows fed diets with two différent levels of protein and différent amino
acid profiles. J. Dairy Sci. 83:2585-95.
Bequette, B. J., F. R. Backwell, C. E. Kyle, A. G. Calder, V. Buchan, L. A. Crompton,
J. Franceet J. C. MacRae. 1999. Vascular sources of phenylalanine, tyrosine, lysine,
and methionine for casein synthesis in lactating goats. J. Dairy Sci. 82:362-377.
Bequette, B. J., F. R. Backwell, J. C. MacRae, G. E. Lobley, L. A. Crompton, J. A.
Metcalf et J. D. Sutton. 1996a. Effect of intravenous amino acid infusion on leucine
oxidation across the mammary gland of the lactating goat. J. Dairy Sci. 79:2217-24.
Bequette, B. J., C. E. Kyle, L. A. Crompton, S. E. Anderson et M. D. Hanigan. 2002.
Protein metabolism in lactating goats subjected to the insulin clamp. J. Dairy Sci.
85:1546-55.
Bequette, B. J., J. A. Metcalf, D. Wray-Cahen, F. R. Backwell, J. D. Sutton, M. A.
Lomax, J. C. MacRae, et G. E. Lobley. 1996b. Leucine and protein metabolism in the
lactating dairy cow mammary gland: responses to supplemental dietary crude protein
intake. J. Dairy Res. 63:209-222.
Bequette, B. J., S. L. Owens, N. E. Sunny, S. W. El-Kadi et L. C. Woods. 2005. Hepatic
gluconeogenesis and muscle intermediary metabolism in hybrid striped bass (HSB)
determinated by 13C-mass isotopomer distribution analysis. J. Dairy Sci. 83: (Supp 1)
293.
Bergen, W. G. 1979. Free amino acids in blood of ruminants - physiological and
nutritional régulation. J. Anim. Sci. 49:1577-1589.
Blouin, J. P., J. F. Bernier, C. K. Reynolds, G. E. Lobley, P. Dubreuil et H. Lapierre.
2002. Effect of supply of metabolizable protein on splanchnic fluxes of nutrients and
hormones in lactating dairy cows. J. Dairy Sci.. 85:2618-2630.
Borucki Castro, S. L, H. Lapierre, L. E. Phillip, P. Jardon et R. Berthiaume. 2006.
Plasma lysine irréversible loss rate to détermine the effect of treatment of soybean meal
on lysine availability in dairy cattle. J. Dairy Sci. (abstract-accepted).
Calder, A. G., K. E. Garden, S. E. Anderson et G. E. Lobley. 1999. Quantitation of
blood and plasma amino acids using isotope dilution électron impact gas
chromatography/mass spectrometry with U-(13)C amino acids as internai standards.
Rapid Commun. Mass Spectrom. 13:2080-2083.
Cant, J. P., E. J. DePeters et R. L. Baldwin. 1993. Mammary amino acid utilization in
dairy cows fed fat and its relationship to milk protein dépression. J. Dairy Sci. 76:762-
774.
Clark, J. H., H. R. Spires, R. G. Derrig et M. R. Bennink. 1977. Milk production,
nitrogen utilization and glucose synthesis in lactating cows infused postruminally with
sodium caseinate and glucose. J. Nutr. 107:631-644.
Connell, A., A. G. Calder, S. E. Anderson et G. E. Lobley. 1997. Hepatic protein
synthesis in the sheep: effect of intake as monitored by use of stable-isotope-labelled
glycine, leucine and phenylalanine. Br. J. Nutr. 77:255-271.
Coulon, J.B. et B. Remond. 1991. Variations in milk output and milk protein content in
response to the level of energy supply to the dairy cow: a review. Livestock Production
Sci. 29:31-47
Choi, K. M., I. Barash et R. E. Rhoads. 2004. Insulin and prolactin synergistically
Stimulate {beta}-casein messenger ribonucleic acid translation by cytoplasmic
polyadenylation. Mol. Endocrinol. 18:1670-1686.
Danilson, D. A., K. E. Webb, Jr. et J. H. Herbein. 1987. Transport and hindlimb
exchange of plasma and blood-cell amino acids in calves fed soy- or urea-based purified
diets. J. Anim. Sci. 64:1842-1851.
Doepel, L., D. Pacheco, J. J. Kennelly, M. D. Hanigan, I. F. Lopez et H. Lapierre. 2004.
Milk protein synthesis as a function of amino acid supply. J. Dairy Sci. 87:1279-1297.
Drost, C. J. 1978. Vessel diameter-independent volume flow measurements using
ultrasound. Proc. San Diego Biomed. Symp. 17:299-302.
Eisemann, J. H., G. B. Huntington et C. L. Ferrell. 1987. Blood flow to hindquarters of
steers measured by transit time ultrasound and indicator dilution. J. Dairy Sci. 70:1385-
1390.
Garlick, P. J., G. A. Clugston et J. C. Waterlow. 1980. Influence of low-energy diets on
whole-body protein turnover in obèse subjects. Am. J. Physiol. 238:E235-E244.
Garlick, P. J., M. Fern et V. R. Preedy. 1983. The effect of insulin infusion and food
intake on muscle protein synthesis in postabsorptive rats. Biochem. J. 210:669-676.
Garlick, P. J., M. A. McNurlan et P. E. Ballmer. 1991. Influence of dietary protein
intake on whole-body protein turnover in humans. Diabètes Care. 14:1189-1198.
Garlick, P. J., D. J. Millward et W. P. James. 1973. The diurnal response of muscle and
liver protein synthesis in vivo in meal-fed rats. Biochem. J. 136:935-945.
Golden, M. H. et J. C. Waterlow. 1977. Total protein synthesis in elderly people: a
comparison of results with [15N]glycine and [14C]leucine. Clin. Sci. Mol. Med.
53:277-288.
Guinard, J. et H. Rulquin. 1994a. Effect of graded levels of duodenal infusions of casein
on mammary uptake in lactating cows. 2. Individual amino acids. J. Dairy Sci. 77:3304-
3315.
Guinard, J. et H. Rulquin. 1994b. Effects of graded amounts of duodenal infusions of
lysine on the mammary uptake of major milk precursors in dairy cows. J. Dairy Sci.
77:3565-3576.
50
Guinard, J. et H. Rulquin. 1995. Effects of graded amounts of duodenal infusions of
methionine on the mammary uptake of major milk precursors in dairy cows. J. Dairy
Sci. 78:2196-2207.
Guinard, J., H. Rulquin et R. Vente. 1994. Effect of graded levels of duodenal infusions
of casein on mammary uptake in lactating cows. J. Dairy Sci .77:2221-2231.
Hang, J et J. A. Rillema. 1997. Effect of rapamycin on prolactin-stimulated S6 kinase
activity and milk product formation in mouse mammary explants. Biochem. Biophys.
Acta. 11:1358:209-214.
Hanigan, M. D., C.C. Calvert, E. J. DePeters, B. L. Reis et R. L. Baldwin. 1991. Whole
blood and plasma amino acid uptakes by lactating bovine mammary glands. J. Dairy
Sci. 74:2484-2490
Hanigan, M. D., J. P. Cant, D. C. Weakley et J. L. Beckett. 1998. An évaluation of
postabsorptive protein and amino acid metabolism in the lactating dairy cow. J. Dairy
Sci. 81:3385-3401.
Hanigan, M. D. 2005. Quantitative aspects of ruminant splanchnic metabolism as
related to predicting animal performance. Anim. Sci. 80:23-32.
Heitmann, R. N. et E. N. Bergman. 1980. Transport of amino acids in whole blood and
plasma of sheep. Am. J. Physiol. 239:E242-E247.
Huhtanen, P., A. Vanhatalo et T. Varvikko. 2002. Effects of abomasal infusions of
histidine, glucose, and leucine on milk production and plasma metabolites of dairy cows
fed grass silage diets. J. Dairy Sci. 85:204-16.
Huntington, G. B. 1984. Net absorption of glucose and nitrogenous compounds by
lactating Holstein cows. J. Dairy Sci. 67:1919-1927.
Huntington, G. B., C. K. Reynolds et B. H. Stroud. 1989. Techniques for measuring
blood flow in splanchnic tissues of cattle. J. Dairy Sci. 72:1583-1595.
51
Hurtaud, C, H. Rulquin et R. Vérité. 1998. Effects of level and type of energy source
(volatile fatty acids or glucose) on milk yield, composition and coagulating properties in
dairy cows. Reprod. Nutr. Dev. 38:315-330.
Institut National de la Recherche Agronomique (INRA). 1998. Ruminant Nutrition:
recommended allowances and feed tables. R. Jarrige Ed. J. Libbey Eurotext, Paris,
France.
James, W. P., P. J. Garlick, P. M. Sender et J. C. Waterlow. 1976. Studies of amino acid
and protein metabolism in normal man with L-[U-14C]tyrosine. Clin. Sci. Mol. Med.
50:525-532.
Katz, M. L. et E. N. Bergman. 1969. A method for simultaneous cannulation of the
major splanchnic blood vessels of the sheep. Am. J. Vet. Res. 30:655-661.
Korhonen, M., A. Vanhatalo et P. Huhtanen. 2002. Evaluation of isoleucine, leucine,
and valine as a second-limiting amino acid for milk production in dairy cows fed grass
silage diet. J. Dairy Sci. 85:1533-1545.
Korhonen, M., A. Vanhatalo, T. Varvikko et P. Huhtanen. 2000. Responses to graded
postruminai doses of histidine in dairy cows fed grass silage diets. J. Dairy Sci.
83:2596-2608.
Lacey, J. M. et D. W. Wilmore. 1990. Is glutamine a conditionally essential amino acid?
Nutr. Rev. 48:297-309.
Lapierre, H., J. F. Bernier, P. Dubreuil, C. K. Reynolds, C. Farmer, D. R. Ouellet et G.
E. Lobley. 1999. The effect of intake on protein metabolism across splanchnic tissues in
growing beef steers. Br. J. Nutr. 81:457-466.
Lapierre, H., R. Berthiaume, G. Raggio, L. Doepel, D. Pacheco, P. Dubreuil et G. E.
Lobley. 2005. The route of absorbed nitrogen into milk protein. Anim. Sci. 80:11-22.
Lapierre, H., J. P. Blouin, J. F. Bernier, C. K. Reynolds, P. Dubreuil et G. E. Lobley.
2002. Effect of supply of metabolizable protein on whole body and splanchnic leucine
metabolism in lactating dairy cows. J. Dairy Sci. 85:2631-2641.
52
Lapierre, H. et G. E. Lobley. 2001. Nitrogen recycling in the ruminant: a review. J.
Dairy Sci. 84, E223-E236.
Lapierre, H., E. Milne, J. Renaud et G. E. Lobley. 2003. Lysine utilisation by the
mammary gland. In Progress in research on energy and protein metabolism. EAAP
Publication No. 109. Ed. W. B. Souffrant and C. C. Metges. Pages, 777-780.
Lemosquet, S., N. Rideau, H. Rulquin, P. Faverdin, J. Simon et R. Vente. 1997. Effects
of a duodenal glucose infusion on the relationship between plasma concentrations of
glucose and insulin in dairy cows. J. Dairy Sci. 80:2854-2865.
Lemosquet, S., S. Rigout, A. Bach, H. Rulquin et J. W. Blum. 2004. Glucose
metabolism in lactating cows in response to isoenergetic infusions of propionic acid or
duodenal glucose. J. Dairy Sci. 87:1767-1777.
Linzell, J. L. 1974. Mammary blood flow and methods of identifying and measuring
precursors of milk. The mammary gland/development and maintenance Aca. Press,
London, UK. B. L. Larson et V. R. Smith: 143-220.
Liu, Z., W. Long., D. A. Fryburg., E. J. Barrett. 2006. The régulation of body and
skeletal muscle protein metabolism by hormones and amino acids. J. Nutr. 136:212S-
217S.
Liu, S. M., G. E. Lobley, N. A. Macleod, D. J. Kyle, X. B. Chen et E. R. Orskov. 1995.
Effects of long-term protein excess or deficiency on whole-body protein turnover in
sheep nourished by intragastric infusion of nutrients. Br. J. Nutr. 73:829-839.
Lobley, G. E., V. Milne, J. M. Lovie, P. J. Reeds et K. Pennie. 1980. Whole body and
tissue protein synthesis in cattle. Br. J. Nutr. 43:491-502.
Lobley, G. E., A. Connell et V. Buchan. 1987. Effect of food intake on protein and
energy metabolism in finishing beef steers. Br. J. Nutr. 57:457-465.
Lobley, G. E. 1993. Species comparisons of tissue protein metabolism: effects of âge
and hormonal action. J. Nutr: 123:337-343.
53
Lobley, G. E., A. Connell, E. Milne, A. M. Newman et T. A. Ewing. 1994. Protein
synthesis in splanchnic tissues of sheep offered two levels of intake. Br. J. Nutr. 71:3-
12.
Lobley, G. E. et H. Lapierre. 2003. Post-absorptive metabolism of amino acids. In
Progress in research on energy and protein metabolism. EAAP Publication No. 109. Ed.
W. B. Souffrant and C. C. Metges. Pages, 737-756.
Lobley, G. E., X. Shen, G. Le, D. M. Bremner, E. Milne, A. G. Calder, S. E. Anderson
et N. Dennison. 2003. Oxidation of essential amino acids by the ovine gastrointestinal
tract. Br. J. Nutr. 89:617-630.
Lykos, T. et G. A. Varga. 1997. Varying dégradation rates of total nonstructural
carbohydrates: effects on nutrient uptake and utilization by the mammary gland in high
producing Holstein cows. J. Dairy Sci. 80:3356-3367.
Mabjeesh, S. J., C. E. Kyle, J. C. MacRae et B. J. Bequette. 2000. Lysine metabolism by
the mammary gland of lactating goats at two stages of lactation. J. Dairy Sci. 83:996-
1003.
Mackle, T. R., D. A. Dwyer, K. L. Ingvartsen, P. Y. Chouinard, D. A. Ross et D. E.
Bauman. 2000. Effects of insulin and postruminai supply of protein on use of amino
acids by the mammary gland for milk protein synthesis. J. Dairy Sci. 83:93-105.
MacRae, J. C, A. Walker, D. Brown et G. E. Lobley. 1993. Acretion of total protein
and individual amino acids by organs and tissuer of growing lambs and the ability of
nitrogen balance techniquee to quantitate protein rétention. Anim. Prod. 57:237-241
Matthews, D. E., H. P. Schwarz, R. D. Yang, K. J. Motil, V. R. Young et D. M. Bier.
1982. Relationship of plasma leucine and alpha-ketoisocaproate during a L-[l-
13C]leucine infusion in man: a method for measuring human intracellular leucine tracer
enrichment. Metab. Clin. Exp. 31: 1105-1112.'
Mepham, T. B. 1982. Amino acid utilization by lactating mammary gland. J. Dairy Sci.
65:287-298.
Metcalf, J. A., J. D. Sutton, E. J. Cockburn, D. J. Napper et D. E. Beever. 1991. The
influence of Insuline and Amino Acid Supply on Amino Acid Uptake by the Lactating
Bovine Mammary Gland. J. Dairy Sci. 74: 3412-3420.
Metcalf, J. A., S. J. Roberts et J. D. Sutton. 1992. Variations in blood flow to and from
the bovine mammary gland measured using transit time ultrasound and dye dilution 2.
Res. Vet. Sci. 53:59-63
Metcalf, J. A., D. E. Beever, J. D. Sutton, D. Wray-Cahen, R. T. Evans, D. J.
Humphries, F. R. Backwell, B. J. Bequette et J. C. MacRae. 1994. The effect of
supplementary protein on in vivo metabolism of the mammary gland in lactating dairy
cows. J. Dairy Sci. 77:1816-1827.
Metcalf, J. A., L. A. Crompton, D. Wray-Cahen, M. A. Lomax, J. D. Sutton, D. E.
Beever, J. C. MacRae, B. J. Bequette, F. R. Backwell et G. E. Lobley. 1996. Responses
in milk constituents to intravascular administration of two mixtures of amino acids to
dairy cows. J. Dairy Sci. 79:1425-1429.
Miettinen, H. O et PJ Huhtanen. 1987. Effects of silage fermentation and post-ruminal
casein suplementation in lactating dairy cows: 2-energy metabolites and plasma amino
acids. J. Sci. Food Agric. 74: 459-468.
National Research Council. 2001. Nutrient Requirements of Dairy Cattle. 7th rev. éd.
Natl. Acad. Sci, Washington, DC.
Pacheco-Rios, D., J. L. Treloar, T. N. Barry et W. C. McNabb. 1999. Amino acid
utilisation by the mammary gland: whole body versus plasma free amino acid pools.
Proc. N. Z. Soc. Anim. Prod. 59: 62-65. 1999.
Peeters, G., A. Houvenaghel, E. Roets, A. M. Massart-Leen, R. Verbeke, G. Dhondt et
F. Vershcooten. 1979. Electromagnetic blood flow recording and balance of nutrients in
the udder of lactating cows. J. Anim. Sci. 48:1143-1153.
Raggio, G., S. Lemosquet, G. E. Lobley, H. Rulquin et H. Lapierre. 2006. Effect of
casein and propionate supply on mammary protein metabolism in lactating dairy cows.
J. Dairy Sci. (accepted).
55
Raggio, G., D. Pacheco, R. Berthiaume, G. E. Lobley, D. Pellerin, G. Allard, P.
Dubreuil et H. Lapierre. 2004. Effect of level of metabolizable protein on splanchnic
flux of amino acids in lactating dairy cows. J. Dairy Sci. 87:3461-3472.
Reeds, P. J., M. F. Fuller, A. Cadenhead, G. E. Lobley et J. D. McDonald. 1981. Effects
of changes in the intakes of protein and non-protein energy on whole-body protein
turnover in growing pigs. Br. J. Nutr. 45:539-546.
Reeds, P. J., M. F. Fuller, G. E. Lobley, A. Cadenhead et J. D. McDonald. 1978. Protein
synthesis and amino acid oxidation in growing pigs [proceedings]. Proc. Nutr. Soc.
37:106A.
Rémond, D., C. Poncet et J. Lefaivre. 1993. Technical note: ruminai vein catheterization
and continuous blood flow measurement in ruminai arteries of sheep. J Anim Sci.
71:1276-1280.
Reynolds, C. K., D. L. Harmon et M. J. Cecava. 1994. Absorption and delivery of
nutrients for milk protein synthesis by portal-drained viscera. J. Dairy Sci. 77:2787-
2808.
Rigout, S., C. Hurtaud, S. Lemosquet, A. Bach et H. Rulquin. 2003. Lactational effect
of propionic acid and duodenal glucose in cows. J. Dairy Sci. 86:243-253.
Rulquin, H. 1987. Détermination de certains acides aminés limitants chez la vache
laitière par la méthode des administrations post-ruminales. Reprod. Nutr. Dev. 27:299-
300.
Rulquin, H. 1992. Intérêts et limites d'un apport de méthionine et de lysine dans
l'alimentation des vaches laitières. INRA Prod. Anim. 5:29-36.
Rulquin, H., P. M. Pisulewski et R. Vérité. 1993. Milk production and composition as a
function of postruminai lysine and méthionine supply: a nutrient-response approach.
Livest. Prod. Sci. 37:69-90.
56
Rulquin, H., S. Rigout, S. Lemosquet et A. Bach. 2004. Infusion of glucose directs
circulating amino acids to the mammary gland in well-fed dairy cows. J. Dairy Sci.
87:340-349.
Rulquin, H., R. Vérité et J. Guinard-Flament. 2001. Acides aminés digestibles dans
l'intestin. Le système AADI et les recommandations d'apport pour la vache laitière.
INRA Prod. Anim. 14:3-13.
Seal, C. J. et C. K. Reynolds. 1993. Nutritional implications of gastrointestinal and liver
metabolism in ruminants. Nutr. Res. Rev. 6:185-208.
Sève, B. et A. A. Ponter. 1997. Nutrient-hormone signais regulating muscle protein
turnover in pigs. Proc. Nutr. Soc. 56:565-580.
Simon, O., R. Munchmeyer, H. Bergner, T. Zebrowska et L. Buraczewska. 1978.
Estimation of rate of protein synthesis by constant infusion of labelled amino acids in
pigs. Br. J. Nutr. 40:243-252.
Steffee, W. P., R. S. Goldsmith, P. B. Pencharz, N. S. Scrimshaw et V. R. Young. 1976.
Dietary protein intake and dynamic aspects of whole body nitrogen metabolism in adult
humans. Metabolism. 25:281-297.
Swaisgood, H. E. 1995 Protein and amino acid composition of bovine milk. Handbook
of milk composition. 1995. R. G. Jensen, éd. Académie Press, Toronto, ON. Pages 464-
468. 1995.
Tesseraud, S., J. Grizard, B. Makarski, E. Debras, G. Bayle et C. Champredon. 1992.
Effect of insulin in conjunction with glucose, amino acids and potassium on net
metabolism of glucose and amino acids in the goat mammary gland. J. Dairy Res.
59:135-149.
Thivierge, M. C, D. Petitclerc, J. F. Bernier, Y. Couture et H. Lapierre. 2000. External
pudic venous reflux into the mammary vein in lactating dairy cows. J. Dairy Sci.
83:2230-2238.
57
Thivierge, M. C, D. Petitclerc, J. F. Bernier, Y. Couture et H. Lapierre. 2002.
Variations in mammary metabolism during the natural fïlling of the udder with milk
over a 12-h period between two milkings. J. Dairy Sci. 85:1839-1854.
Vanhatalo, A., T. Varvikko et P. Huhtanen. 2003a. Effects of casein and glucose on
responses of cows fed diets based on restrictively fermented grass silage. J. Dairy Sci.
86:3260-3270.
Vanhatalo, A., T. Varvikko et P. Huhtanen. 2003b. Effects of various glucogenic
sources on production and metabolic responses of dairy cows fed grass silage-Based
diets. J. Dairy Sci. 86:3249-3259.
Wester, T. J., G. E. Lobley, L. M. Birnie, L. A. Crompton, S. Brown, V. Buchan, A. G.
Calder, E. Milne et M. A. Lomax. 2004. Effect of plasma insulin and branched-chain
amino acids on skeletal muscle protein synthesis in fasted lambs. Br. J. Nutr. 92:401-
409.
Wolfe, R. R. 1984. Tracers in metabolic research: Radioisotope and stable isotope/mass
spectrometriymethods. Alan R. Liss, Inc., New York.
Wray-Cahen, D., H. V. Nguyen, D. G. Burrin, P. R. Beckett, M. L. Fiorotto, P. J. Reeds,
T. J. Wester et T. A. Davis. 1998. Response of skeletal muscle protein synthesis to
insulin in suckling pigs decreases with development 1. Am J Physiol 275:E602-E609.
5 S
CHAPITRE III
EFFECT OF LEVEL OF METABOLIZABLE PROTEIN ON SPLANCHNIC
FLUX OF AMINO ACIDS IN LACTATING DAIRY C O W S
Ce chapitre a été publié dans la revue "Journal of Dairy Science" sous la références de
G. Raggio, D. Pacheco, R. Berthiaume, G. E. Lobley, D. Pellerin, G. Allard, P. Dubreuil
and H. Lapierre. 2004. Effect of level of metabolizable protein on splanchnic flux of
amino acids in lactating dairy cows. J. Dairy. Sci. 87:3461-3472.
3.1 Résumé
La réponse du métabolisme splanchnique des acides aminés (AA) à des apports
croissants de protéines métabolisables (PM) a été mesurée avec six vaches laitières
Holstein multicathéterisées. Trois rations, formulées pour fournir le même niveau
d'énergie et 3 différents niveaux de PM (1922, 2264 et 2517 g PM/jour) ont été testées
selon un double carré latin 3 x3 avec des périodes expérimentales de 21 jours. Aux
jours 18, 19 ou 20, des échantillons de sang ont été simultanément prélevés des veines
porte, hépatique et d'une artère. La production de lait, de protéines du lait ainsi que l'N
excrété dans l'urine ont augmenté avec l'augmentation de PM. L'absorption portale des
AA indispensables a augmenté de façon linéaire avec les apports croissants de PM.
L'extraction hépatique a augmenté seulement pour l'His, la Met, et la Phe avec
l'augmentation des PM, alors que l'extraction hépatique des acides aminés à chaîne
ramifiée et de la Lys n'a pas été affectée. L'augmentation de l'apport de PM diminue le
ratio production de protéines du lait / approvisionnement post hépatique pour la Leu,
l'Ileu, la Val, la Thr et la Lys et suggère une augmentation de l'oxydation de ces AA au
niveau des tissus périphériques. Une diminution du catabolisme des AA indispensables,
soit au niveau hépatique (His, Met, Phe, et Thr) soit au niveau des tissus périphériques
(lieu, Leu, Val, Lys et Thr) accompagne la diminution de l'efficacité du transfert des
AA absorbés en protéines du lait, à mesure qu'augmentent les apports de PM.
L'efficacité d'utilisation des AA entre l'absorption des AA et la sécrétion de protéines
dans le lait varie également entre les AA. Ces résultats confirment qu'un facteur fixe de
conversion des AA absorbés en sécrétion de ces AA dans le lait ne devrait pas être
utilisé dans les futurs modèles de prédiction des productions de protéines du lait.
Mot clés: acides aminés, tissus splanchniques, glande mammaire, vache
60
3.2 Abstract
The response of splanchnic tissue metabolism to différent levels of
metabolizable protein (MP) was measured in six catheterized multiparous lactating
Holstein cows. Three diets, balanced to provide similar energy intakes and increasing
amounts of MP (g/d), 1922 (Low), 2264 (Med) and 2517 (High), were fed according to
a replicated Latin Square with 21-d expérimental periods. On d 18, 19 or 20, six hourly
blood samples were collected simultaneously from the portai and hepatic veins plus an
artery to détermine net fluxes of nutrients across the portal-drained viscera and the liver.
Yields of milk and protein increased (P < 0.02), as did urinary N excrétion (P < 0.01)
with increasing MP. Portai absorption of essential amino acid (EAA) increased linearly
(P < 0.01) with increasing MP supply, as did liver removal of His, Met and Phe (P <
0.01). In contrast, liver removal of the branched-chain A A (BCAA) and lysine was
unaffected by diets. With increasing MP, the ratio of milk output to post-liver supply of
BCAA, Thr (P < 0.10) and Lys (P < 0.20) decreased linearly, indicating oxidation of
thèse AA in the peripheral tissues. Concomitant to a decreased catabolism of EAA in
the liver (His, Met, Phe and Thr) and/or in peripheral tissues (BCAA, Lys and Thr), the
effïciency of transfer of absorbed EAA into milk protein decreases markedly as protein
supply increases. The effïciency of transfer of absorbed AA into milk varies also greatly
between AA. Thèse two important factors should be taken into account when building
prédictive schemes for milk protein output.
Key words: amino acid, splanchnic tissue, mammary gland, cow
Abbreviation key: AA = amino acids, BCAA = branched-chain AA, EAA = essential
AA, MP = metabolizable protein, pAH = para-aminohippurate, PDV = portal-drained
viscera.
61
3.3 Introduction
Optimal conditions for milk production can be defined as either when output is
maximized or when the effïciency of transfer of dietary protein into milk protein is
optimized. Maximizing output involves increasing dietary protein supply but as the
supply approaches current recommendations, the effïciency of transfer into milk protein
decreases (Doepel et al., 2004). Highest transfer effîciencies therefore are achieved at
low dietary protein supply but are associated with reduced yields. In practical ternis, a
well-advised feeding strategy would supply protein to achieve a milk production
between extrêmes of maximal output and effîciencies. Prédiction of optimum supply
requires knowledge of how altérations in that supply affect the partition of individual
amino acids (AA) between catabolism and anabolism.
Increasing dietary protein supply usually results in higher portai absorption of
AA in lactating cows (Bach et al., 2000; Blouin et al., 2002), although elevated crude
protein intake may not lead to additional AA absorption if the energy supply is not
adéquate to support rumen microbial protein synthesis (Reynolds et al., 1992). Once
absorbed, the AA flow to the liver. Although hepatic removal of AA was first proposed
as a regulator of peripheral tissue anabolism (Reynolds et al., 1994), other évidence
suggests that hepatic AA catabolism may instead respond to peripheral tissue
requirements (Lobley and Lapierre, 2003; Ortigues-Marty et al., 2003). The question
then is: "To what extent can liver catabolism of AA be decreased, i.e. how much of the
hepatic extraction of AA is obligatory (e.g. to support plasma protein synthesis or
gluconeogenesis) and how much is non-obligate and susceptible to manipulation (e.g.
can the amount removed be reduced)?. This question can be answered by varying
protein supply below and above requirements with examination of changes in the
partition and effïciency of use of AA as supply is altered. An additional considération is
that some AA are not exclusively, or even predominantly, catabolized by the liver.
Therefore, the current study investigates the impact of increasing metabolizable
protein (MP) on hepatic removal of essential AA (EAA) and subséquent utilization of
post-splanchnic supply by the mammary gland. Given that the effïciency of conversion
of absorbed EAA into milk protein will be altered, thèse findings will lead to a better
62
understanding of how splanchnic and post-splanchnic tissues coordinate to regulate AA
metabolism.
3.4 Material and Methods
3.4.1 Animais and Treatments
Six multiparous Holstein cows, averaging 656 + 60 kg BW and 96 ± 8 DIM at
the beginning of the study, were used in a double 3 x 3 Latin Square design, balanced
for residual effects, with 21-d expérimental periods. Ail cows were surgically implanted
with vascular cathéters across the splanchnic tissues (Huntington et al., 1989). Briefly,
Teflon cathéters encased in silastic were implanted into the portai vein and one hepatic
vein while Tygon cathéters were inserted into two mesenteric veins and one in a
mesenteric artery. The right carotid artery was also raised to a subcutaneous position to
provide alternative access to arterial blood if necessary. Surgeries were performed at
least one month before the start of the experiment.
Three diets were formulated using the NRC (2001) to provide equal energy
while increasing MP, averaging 1922 (Low), 2264 (Med) or 2517 (High) g of MP/d.
Feeding différent amounts of MP was achieved by substituting Megalac™ (Church &
Dwight Co., Inc., NJ) with Prolak® (H.J. Baker & Bro. Inc., CO) (Tables 3.1 and 3.2).
The amount of feed offered provided similar energy intake (36.4 Mcal NEL/d)
throughout the study. The quantity of wet feed offered was adjusted weekly, based on
the DM content of the grass silage (determined by drying at 105 °C for 48 h). Daily
allowance was distributed in 12 equal meals every 2 h from automated feeders (Ankom,
Fairport, NY), except long hay that was offered once a day. Orts, when présent, were
recorded daily. Cows had free access to water and were housed in a tie stall barn, which
was lit from 0600 to 2200. They were milked twice daily (0600 and 1800) and
production was recorded at each milking. The expérimental protocol was approved by
the Institutional Committee for Animal Care of the Lennoxville Research Centre and
animais were cared for according to the guidelines of the Canadian Council on Animal
Care (1993).
3.4.2 Sampling
On days 13 to 18 of each period, total collection of fèces and urine were made to
détermine N balance. Urine was collected in stainless steel containers, via a Gouch tube
(BF Goodrich Co., Kitchener, ON, Canada), and acidified twice daily with 50 ml of 10
N sulfuric acid, to maintain pH below 3.0. A 2 % portion of the fèces and urine
collected was sampled daily and frozen. At the end of the period, samples were thawed,
mixed, sub-sampled and frozen for further analyses. Any feed refusais were sampled
daily. Milk samples were taken at each milking and frozen for later analysis.
On day 18, 19 or 20 (two cows sampled per day) sodium p-aminohippurate
(pAH; 100g/l) was infused continuously (14.4 g/h) into one mesenteric vein cathéter
using a syringe pump, preceded by a 2 g priming dose. After at least 40 minutes of pAH
infusion, simultaneous blood collections were taken hourly for 5 h (n=6) from the
arterial plus portai and hepatic vein cathéters. Blood samples were collected from the
mammary vein by venipuncture, every other hour (n=3). Immediately after sampling,
blood was transferred from the syringes to heparinized tubes and kept on ice prior to
centrifugation at 3000 x g for 12 min at 4 °C. The plasma was then sub-sampled for
immédiate analyses and the remainder stored at -20°C for later analyses.
3.4.3 Laboratory Analyses
Feed ingrédients, feed refusais and fèces were freeze-dried and ground to pass
through a 1 mm screen. Nitrogen content of feed ingrédients, refusais and milk samples
collected at each milking was determined by combustion (Nitrogen Determinator,
model FP-428 LECO, St-Joseph, MI). Crude protein contents of feed and milk were
calculated as N x 6.25 and N x 6.38, respectively. Urine and fèces samples were
analyzed for total N using the micro-Kjeldahl method (Tecator 1030, Hoganas Sweden;
AOAC, 1996). Milk fat was measured according to the Roese-Gottlieb method (AOAC,
1996).
Noncasein N and nonprotein N in milk were analyzed with the micro-Kjeldahl
method on pooled samples from the collection period. Noncasein N was obtained by
précipitation of the caseins at pH 4.6. Nonprotein N was obtained by précipitation with
64
trichloroacetic acid (TCA), final concentration of 12 %. Casein N is calculated by
différence between total N and noncasein N, while whey protein is estimated by
différence between noncasein nitrogen and nonprotein N. The N content was transposed
into protein using the 6.38 factor. Casein profile was determined by inverse phase
HPLC on the pellet obtained when precipitating caseins at pH 4.6 (Jaubert et al., 1992).
Packed cell volume of each blood sample was determined by the
microhematocrit method. Plasma concentrations of pAH, urea and ammonia were
measured with an automatic analyzer (Technicon Autoanalyser II, Technicon
Instruments Corporation, NY) as previously described (Huntington, 1984) on fresh
samples on the day of sampling.
For AA détermination, samples of feed were pre-digested with performic acid to
stabilize Met and Cys, treated with hydrobromic acid to destroy the performic acid and
then were acid-hydrolysed with 67V-HC1 (AOAC, 1996: method # 994-12) and AA were
quantified by ion exchange chromatography (Beckmann 6300, Palo Alto, Ca). A
separate acid hydrolysis (6iV- HCl) digestion procédure, was conducted for Phe and His,
because those AA are destroyed during the oxidation process and by reaction with
bromine.
Plasma concentrations of EAA plus tyrosine were measured by the isotope
dilution method (Calder et al., 1999). On the day of sampling, 0.2 g of an internai
standard solution was added to 1 g of plasma. The internai standard solution was
prepared with labeled AA diluted in 0.1N HCl with the following concentration (uM):
DL-His-a-15N (182), L-Ile-15N (713), L-Leu-l-13C (877), DL-Lys-2-15N-2HCl (308),
DL-Met-l-13C (98), L-Phe-l-13C (226), L-Thr-15N (45), L-Tyr-15N (469), L-Val-15N
(846). Labelled AA (95-99 atom %) were supplied by CDN isotopes (Montréal, QC,
Canada) for His, Leu, Lys, Met and Phe; Cambridge Isotope Lab (Andover, MA) for
Ile, Thr and Tyr; Isotec Inc. (Miamisburg, OH) for Val. Plasma samples with the
internai standard were kept frozen at -20°C until analyses. At the time of analysis the
plasma was deproteinized with sulfosalicylic acid (38 % w/v), the N-(tert-
butyldimethyl) AA derivative prepared (Calder and Smith, 1988) and AA concentration
determined by gas chromatography-mass spectrometry (Model HP6890, S973 mass
65
sélective detector; Hewlett Packard, CA) based on the principles described by Calder et
al. (1999).
3.4.4 Calculations and Statistical Analyses
Mean values for intake, production and composition of milk, fèces and urine
taken during the collection period were used for statistical analyses. Plasma flow across
the splanchnic tissues was calculated from downstream dilution of pAH infusion (Katz
and Bergman, 1969). If, within one sampling day, the CV of the mean plasma flow for
an animal was greater than 15 % due to only one sample, then this value was removed.
Milk protein output was estimated using milk protein yield measured during the
collection period and AA composition reported in the literature (Swaisgood, 1995).
Mammary plasma flow was estimated according to the Fick principle, using Phe and
Tyr as internai markers (Mepham, 1982), with allowance for a 3.5 % contribution from
blood-borne proteins: mammary plasma flow = ((milk Phe + Tyr) x 0.965) / (AV
différence Phe + Tyr) (Cant et al., 1993). Daily net fluxes of AA across the PDV, liver,
total splanchnic tissues and mammary gland were calculated for each cow as the
product of the average venous-arterial concentration différence and the average plasma
flow measured for that day. Blood flow was calculated as plasma flow divided by (1 -
packed cell volume). Net fluxes of ammonia were calculated using plasma
concentrations and blood flow while urea fluxes were calculated using plasma
concentrations and blood water flows. Blood water flow was calculated as blood flow
multiplied by the ratio of the dry matter of blood on plasma (Milano et al., 2000). A
négative flux indicates utilization or removal while a positive flux indicates net
production or release of the nutrient across the tissue. One cow did not hâve a patent
hepatic cathéter and results obtained for portai vein absorption strongly suggested a
misplacement of this cathéter: ail results for this cow were discarded. One cow could
not be used for her third period (treatment was High MP), as she had severe mastitis.
Therefore, n = 5 for Low and Med MP and n = 4 for High MP.
Data were analyzed statistically according to a replicated Latin Square design,
using the GLM procédure (SAS Institute Inc., 1996) with treatment, period and cow as
the main effects. Treatments were compared with linear and quadratic polynomial
66
contrasts. A linear component indicates that the différence between the fïrst and last
levels is significant, while a quadratic component will indicate the lack of fit of the
intermediary level for a linear relationship (Gill, 1978). Treatment différences were
considered significant if P < 0.10 and as a trend for 0.10 < P < 0.20.
3.5 Results
3.5.1 Milk Production and Composition
Milk production and composition data are presented in Table 3.3. Milk
production increased linearly (P = 0.02) with increasing supply of MP. Milk crude
protein yield also increased linearly (P = 0.01), as did each protein fraction yields (P <
0.03). Milk crude protein concentration increased linearly (P = 0.09), but the proportion
of true protein decreased (P < 0.01), mainly due to a decrease in the proportion of the
casein fraction (P = 0.03). The casein profile was not affected by diet (Table 3.3). Milk
fat concentration decreased linearly (P = 0.01) resulting in a decrease in fat yield at the
highest MP level (linear: P = 0.14; quadratic: P - 0.17). In this experiment the low MP
diets contained more Megalac (Table 1) and the yield and percentage of fat were higher
with this diet. Similarly, Petit (2002) reported an incrément of fat percentage in milk
with a diet supplément with Megalac, compared with other isoenergetic and
isonitrogenous diets.
3.5.2 Net Flux of Essential Amino Acids
Plasma flow across the PDV and the splanchnic tissue was not affected by MP
supply and averaged 1321, 1271 and 1236 ± 43.2 and 1676, 1608 and 1647 ± 63.0 L/h,
for Low, Med and High MP respectively. However, plasma flow across the mammary
gland increased at High MP compared with Low and Med MP (linear and quadratic
components, P = 0.04) averaging 515, 491 and 607 ± 24.1 L/h, for Low, Med and High
MP respectively. As MP intake was raised, plasma arterial concentrations (Table 3.4) of
His, Ile, Leu, Lys and Val increased linearly (P < 0.10) with a quadratic effect for Lys
(P = 0.01) and His (P = 0.14). Net fluxes of EAA and Tyr across the PDV, liver,
splanchnic tissue, mammary gland and in milk are presented in Table 3.5. With
increasing MP supply, the PDV net flux increased linearly (P < 0.01) for ail measured
67
AA with a quadratic effect for Met (P = 0.03) and Lys (P = 0.20). Hepatic removal
increased linearly (P < 0.04) for His, Met, Phe, Tyr and tended (P < 0.19) to increase
for Thr. The net flux of EAA across the splanchnic tissue increased linearly (P < 0.10)
for Ile, Leu, Lys, Thr and Val with a quadratic trend (P < 0.20) for His, Ile, Leu, Lys
and Met. Mammary gland uptake increased linearly (P < 0.10) for ail BCAA, Lys and
Phe with a tendency for a linear increase for Met (P < 0.14). Output in milk increased
linearly (P < 0.01) for ail AA, as did milk crude protein yield (Table 3.3). The ratio of
milk output / PDV net flux decreased linearly (P < 0.10) for His, Ile, Leu, Lys, Met, Phe
and Tyr and tended to decrease linearly (P < 0.20) for Thr and Val (Table 6). In
addition, the ratio of the liver flux / portai flux increased linearly (P < 0.10) with MP
supply for Met, Phe, Tyr or only with a tendency (P < 0.20) for His. Relative to total
hepatic influx, hepatic net removal Met, Phe and Tyr increased linearly (P < 0.10) while
His tended to increase (P = 0.14; Table 3.7).
3.5.3 Urea and Ammonia
Total splanchnic urea flux tended to increase (P < 0.12) with increasing MP
supply, as a resuit of a decreased return of urea into the PDV (P = 0.06) and a numerical
increase in hepatic ureagenesis (P = 0.22; Table 3.8). Accordingly, arterial urea-N
concentration increased linearly (P < 0.01) with increasing supply of MP (Table 3.5).
Arterial ammonia concentrations were unaffected by MP supply, but ammonia flux
across the PDV increased linearly and quadratically (P = 0.03) with increased MP of the
ration. Hepatic removal also increased linearly (P = 0.02) resulting in no effect of the
diet on splanchnic net flux of ammonia.
3.5.4 N Balance
The planned increase in N intake (P < 0.01) was matched by increased N
excrétion in fèces and in urine (linear contrast, P < 0.05; Table 3.9). Apparent N
digestibility also increased linearly (P < 0.02), averaging 53.2, 58.9 and 60.5 % for
Low, Med and High MP, respectively. Nitrogen in milk as a proportion of N intake
decreased (P < 0.02) with increasing N intake, averaging 27.5, 25.6, and 24.1 % of N
intake. In contrast, N rétention was not affected by treatments (Table 3.9).
68
3.6 Discussion
A récent review of studies where AA were infused postruminally in dairy cows
supports the concept that the transfer of incrémental AA supply into milk protein
diminishes as supply approaches requirements (Doepel et al., 2004). Similarly, in this
experiment although the sécrétion of N into milk was elevated by increased MP supply,
the efficiency of conversion of N intake into milk N decreased when N intake increased.
Similar results hâve also been observed by Wright et al. (1998) and Castillo et al.
(2001). The marginal recovery of incrémental MP supply in milk protein averaged 19
% between Low and High MP, which is within the range of recovery from casein
infusions (21 %) reviewed by Hanigan et al. (1998).
Therefore, although most formulation schemes focus on estimating supply of
protein at the duodénum and use fixed factors to predict transfer from absorption to
metabolic usage, metabolic events differ between cows receiving a low versus a high
supply of protein. Examination of the fate of AA between absorption, transit across the
liver and sécrétion in milk will help explain the mechanisms that resuit in higher
transfer efficiency in cows receiving a lower MP supply.
3.6.1 Amino Acid Flux
As expected, increased MP supply resulted in elevated portai absorption of
EAA, consistent with previous reports for dairy cows (Bach et al., 2000; Blouin et al.,
2002). Indeed, the measured portai absorption matched prédiction of digestible AA
(NRC, 2001), corrected for endogenous contribution to the duodenal supply. The ratios
between the NRC prédictions and measured values averaged for His (1.03), Lys (1.02),
Ile (1.14), Leu (1.24), Val (1.13) and Thr (1.27). Thèse ratios would fit with the concept
of little or no catabolism of His and Lys across the gut in the ruminant, but substantial
oxidation of the BCAA and loss of Thr through endogenous sécrétions into the post-
duodenal tract (Lobley and Lapierre, 2003). The ratios (NRC predicted/observed) for
Met (0.84) and Phe (0.79) were lower than unity, which was unexpected as limited
oxidation by the ovine gut has been reported for Met (Lobley et al., 2003). Thus, one
69
explanation would be an underestimation of the duodenal supply of thèse AA with the
NRC (2001) model.
The increased absorption with increased MP was concomitantly associated with
increased hepatic removal, both in absolute amount for His, Met, Phe and Thr, and
relative to portai absorption or total liver influx for His, Met and Phe. On the other
hand, hepatic removals of the BCAA and Lys were unaffected by diet. Thèse data are
consistent with the concept that the liver is a major site of catabolism for His, Met, Phe
and Thr and removes excess AA not used for peripheral anabolism. In contrast, for the
BCAA and Lys, catabolism occurs to a greater extent in non-splanchnic tissues. Thèse
observations agrée well with the localization of enzymes responsible for AA catabolism.
Enzymes for His, Met and Phe catabolism are almost exclusively restricted to hepatic
tissues (see review: Lobley, 2002). In contrast, enzymes responsible for BCAA
catabolism are spread across many tissues (Goodwin et al, 1987; DeSantiago et al.,
1998), and récent studies hâve identified two enzymes involved in Lys catabolism,
lysine ketoglutarate reductase and saccharopine dehydrogenase in the brain, kidney and
muscle of piglets (Pink et al., 2003).
Hepatic removal of EAA varied greatly between EAA, as previously observed in
many species, including dairy cows (Bach et al., 2000; Blouin et al., 2002). Relative to
net portai absorption, the highest fractional hepatic removal was for Phe (0.52) followed
by His (0.42), Met (0.36) and Thr (0.34). With Low MP supply, there was no net
removal of Lys, but with High MP supply, hepatic removal averaged 12 % of portai
absorption. There was no net removal and even a positive numeric flux for the BCAA
across the liver, as previously observed in dairy cows (Wray-Cahen et al., 1997; Bach et
al., 2000; Blouin et al., 2002), but hepatic-portal différences were not différent from
zéro (date not shown). Although the magnitude of hepatic removal relative to portai
absorption was lower in the current study, as would expected due to increased metabolic
demand for absorbed AA during lactation, the ranking amongst EAA agrées with
previous observations in growing cattle (Lapierre et al., 2000), non-lactating cows
(Wray-Cahen et al., 1997) and sheep (Lobley et al., 2001).
70
Only for His, and only at low MP, were mammary uptake and milk output
higher than net splanchnic flux. This means that for ail other measured AA, even at the
low MP supply, post-liver supply was suffïcient to account for mammary uptake and
milk protein output. This maintenance of lactation at a high level when supply was
reduced was achieved by a substantial réduction of AA catabolism, suffïcient to ensure
no need for contribution of body protein "reserves". Such réductions in catabolism will
hâve a fïnite limit, however, based on obligate needs for production of various
metabolites, synthesis of hepatic export proteins (Raggio et al., 2002) and basai ('non-
manipulable') ureagenesis. For His, the limit in the réduction of hepatic removal (3.3 %
of total liver influx at low MP intake) might hâve been reached, as an additional
endogenous source must hâve been necessary to cover mammary gland uptake. This
was not from mobilized body protein reserves as N rétention was not decreased at low
MP intake. One alternative are the His dipeptides, particularly carnosine, présent in mM
quantities in muscle and known to become depleted when animais are fed a diet low in
His (Amend et al., 1979). Such sources would only provide a medium-term store and
obviously such a situation could not be sustained for a whole lactation cycle. Such
constraints are a feature of the AA composition of the MP supply and the current
observation was not surprising considering that His has been found limiting when grass
silage was fed to dairy cows (Vanhatalo et al., 1999). This emphasizes the need to
consider the amount and composition of AA supplied, rather than simply MP, when
attempting to predict dietary responses and efficiencies.
Post-liver supply of the BCAA and Lys substantially exceeded both mammary
gland uptake and milk output. Earlier studies hâve shown that the mammary gland
catabolizes Leu (Bequette et al., 1996; Thivierge et al., 2002) with this oxidation
decreasing when Leu supply was low (Bequette et al., 1996). This is supported by the
current observation, as the ratio of milk output to mammary uptake for the three BCAA
decreased with increasing MP supply. Excess of EAA-N extracted by the mammary
gland relative to milk output is needed to support synthesis of non-essential AA for
which the uptake by the mammary gland is insufficient to account for milk output.
Recently, the incorporation of lysine-15N into non-essential AA within the mammary
gland has been demonstrated in dairy cows (Lapierre et al., 2003).
71
Together, thèse metabolic events across the liver and the mammary gland
explain why the ratios of transfer of absorbed AA into milk protein alter with MP
supply (Table 3.7). At low MP supply, improved efficiency is achieved through a
decreased hepatic removal of His, Met, Phe and Thr, and, probably, decreased oxidation
of the BCAA, Lys and Thr in peripheral tissues. As MP supply is decreased, the liver,
peripheral tissue and mammary gland events combine to increase the partial efficiency
of transfer of absorbed AA into milk. The design of this study does not résolve whether
the liver was responding to, or controlling use by, peripheral tissues (see Lobley and
Lapierre, 2003), but it is obvious that liver catabolism of His, Met and Phe was tightly
related with mammary requirements. Nonetheless, prédiction of milk output based on
duodenal MP or AA supply cannot rely on a fixed ratio of transfer but needs to account
for the efficiency of utilization of individual AA which decreases as supply approaches
requirements. Moreover, use of a common factor of transfer of MP into milk protein
also implies that ail AA are used with similar efficiency. In practice, efficiency of
conversion of absorbed EAA (excluding His) into milk varied widely (0.42-0.78; Table
7) at the low MP supply, where efficiency was greatest. Phenylalanine always ranked
the lowest, probably because supply was in excess of demand with thèse diets. This
variation in efficiency among AA emphasizes the uniqueness of each AA and the
importance of developing prédiction équations for milk protein output based on the
supply of individual AA rather than their total agglomération as protein.
3.6.2 Urea and Ammonia
Net portai flux of ammonia increased with increasing MP supply and increased
rumen degradable protein supply, as previously reported (Bach et al., 2000; Blouin et
al., 2002), although urea recycled to the PDV decreased at High MP. AU the extra
ammonia absorbed was removed by the liver and, combined with increased hepatic
removal of some AA, resulted in a numerically higher hepatic ureagenesis for the cows
fed the High MP diet. Lapierre and Lobley (2001) calculated the ratio between hepatic
ureagenesis and digestible N: in growing cattle, this ranged from 0.49 to 1.78 (mean
0.93) and in lactating dairy cows between 0.43 to 1.23 (mean 0.88). In the current
experiment, the ratio between ureagenesis and digestible N averaged 1.17, again
72
indicating the magnitude of the transit of N into the urea pool and the impact of
recycling. Although arterial urea concentrations were higher with the High MP diet, the
recycling of urea into the PDV decreased with increasing MP, averaging 0.46, 0.28 and
0.14 of hepatic ureagenesis for Low, Med and High MP, respectively. This suggests that
entry of urea into the digestive tract is not simply a function of plasma concentration as
suggested previously (Harmeyer and Martens, 1980). As a resuit of decreased net PDV
flux and increased hepatic flux of urea, the splanchnic net urea flux increased, which
impacted on both arterial urea concentration and élimination of N through the urine.
3.6.3 Milk Production
Increasing the MP supply, by raising the CP content of the diet from 12.7 to 16.6
%, increased both milk production and milk protein yield, in agreement with earlier
reports by Metcalf et al. (1996: 11.3 to 20.1 % CP), Wright et al. (1998: 10.2 to 24.5 %
CP) and Dinn et al. (1998: 15.3 to 18.3 % CP). In ontrast, Castillo et al. (2001) reported
no différence in milk production and composition with variation in CP content (14.0 to
18.3 % CP). On the other hand, Blouin et al. (2002) reported an increase in milk
production and milk protein yield with fixed CP, and varied ruminai undegraded protein
(RUP). Nonetheless, changes in protein degradability and RUP supply alone do not
explain variations in milk production responses (Santos et al., 1998). Together, thèse
data clearly demonstrate that CP intake or RUP supply alone should no longer be used
solely in dairy rations to evaluate the amount of protein delivered to the animal. Use of
CP and RUP has now been superseded by the more informative use of MP and when
earlier data are now re-evaluated by this approach it is clear that changes in milk
production related better to MP rather than either CP or RUP. For example, increasing
MP by 30 % raised both milk production and milk protein yield (Metcalf et al., 1996;
Wright et al, 1998; présent study). In contrast, when MP supply was increased by 18 %,
only milk production was improved (Dinn et al., 1998). In the study of Blouin et al.
(2002), despite a similar crude protein intake, the additional supply in MP (16 %)
generated by increasing RUP resulted in greater milk protein production. However, an
incrément in MP supply does not guarantee by itself an incrément in milk production. In
fact, Castillo et al. (2001), improved MP supply with increased CP or decreased protein
73
degradability did not alter milk production because energy supply would hâve limited
milk output to 22 kg (NRC, 2001).
3.6.4 Nitrogen Balance
The relatively small increase in fecal N compared with the additional N intake
resulted in greater apparent N digestibility. Using a dynamic model based on five N
balance studies with dairy cows, Kebreab et al. (2002) predicted that absolute metabolic
fecal N would increase up to intakes of 420 g N/d and remain near constant thereafter.
Therefore, the increased apparent N digestibility with increasing protein intake observed
in the current study, as in previous reports (Wright et al., 1998; Castillo et al., 2000),
would resuit from a decreased proportion of metabolic fecal N to total fecal N as N
intake increased. Thus, with higher MP more N is apparently digested and, if not used
for anabolic purposes, more should be lost via urine rather than fèces.
This is exactly what was observed, with the ratio of urinary N excrétion to N
intake increasing from 0.16, 0.22 and 0.26 for the Low MP, Med MP and High MP
diets, respectively, similar to the observations of Wright et al (1998) and Dinn et al
(1998). How much N is excreted in urine relates to a number of factors, including MP
(or CP) intake, variations in the degradability of the ration and the ability of the cow to
retain N for anabolic purposes. Indeed, when the anabolic potential is approached or
exceeded then much of the extra digested N is excreted via urine. Thus, in the current
study, although the extra digested N represented increases in portai absorption of both
AA and ammonia, much of thèse were converted to catabolic end-products (notably
urea) with urinary nitrogen excrétion increasing by more than 100 % between low to
high MP. In conséquence, the marginal return of digested N to milk was only 0.16,
compared with an apparent efficiency of 0.52 at the low MP supply. For other studies
where similar increases in digested N were achieved either by increased CP (Castillo et
al., 2001), only 0.15-0.35 of the extra digested N was secreted as milk N. Thèse various
data again reflect the declining return of milk output as maximal requirements are
approached (Doepel et al., 2004).
74
3.7 Conclusion
In dairy cows receiving diets limited in protein supply (1922 g/d MP or 12.7
%CP), increased transfer of absorbed AA into milk protein occurred concomitantly with
a réduction in hepatic removal of His, Met, Phe and Thr and a decreased catabolism /
utilization of the BCAA, Lys and Thr in peripheral tissues compared with higher MP
supply. Therefore, to improve the prédiction of protein supply and its utilization, it is
necessary to appreciate the various metabolic events that occur between absorption and
sécrétion in milk, to allow that the transfer of absorbed AA into milk is not fïxed and
will decrease as supply increases. Furthermore, it is important to take into account that
thèse transfer factors vary widely between AA, as a conséquence of both diet
composition and metabolic events.
3.8 Acknowledgements
The authors gratefùlly thank the staff of the Lennoxville dairy center for taking
care of the animais, M. Léonard and S. Provencher for their dedicated technical support,
D. St-Gelais and S. Haché for analyses of protein milk fractions as well as S. Méthot for
statistical analyses. The authors also wish to acknowledge the financial support of the
Action concertée Fonds FCAR - Novalait - MAPAQ, the National Science and
Engineering Researh Council of Canada and Agriculture and Agri-Food Canada
(Lennoxville Research Contribution Number 829) as well as H. J. Baker & Bro. Inc. for
supplying the Pro-Lak bypass protein supplément.
75
Table 3.1. Composition of the three dietary treatments with différent levels ofmetabolizable protein (MP).
Ingrédients Treatmentsg/kg DMGrass silageProlak1
Soybean hullsCorn grain, groundMegalac2
Grass hayMolasses, beet sugarMinerai & vitamin premixCorn grain, cracked
Estimation from NRC (2001)3
NEL, Mcal/dCP, %DMRDP supplied, g/dRDP balance, g/dRUP supplied, g/dRUP balance, g/dMetabolizable protein, g/d
From microbial proteinFrom undegraded feedFrom endogenous origin
1 H. J. Baker & Bro. Inc., CO2 Church & Dwight Co., Inc., NJ3 Calculated using the intake (Low MP: 23.8, Med MP 24.2, High MP 24.7 kg DM/d)and the production observed during the study, with chemical composition of the feedingrédients reported in Table 3.2
LowMP4290
243148343213217S
36.412.7
2125-162907-39619221156654112
MedMP421282431482331131775
3614
23074
1257-58
22641251899114
High MP423542341431231131575
.4 36.4
.7 16.625011821602317251712611140116
76
Table 3.2. Chemical composition of the feed ingrédients.
Analyses(DM %)CPADFNDFLigninADFIPNDFIPAshStarchLipid
Amino Acid(gAAIOOg/CP)
AlanineArginineAspartateCysteineGlutamateGlycineHistidineIsoleucineLeucineLysineMethioninePhenylalanineProlineSerineThreonineValineTotal AA
Grasssilage17.4033.1041.40
5.200.703.00
10.901.901.50
4.693.09
10.100.717.684.331.483.806.293.741.253.925.823.683.625.40
69.60
Soybeanhulls11.7046.8066.00
8.900.702.904.702.300.80
3.684.478.331.679.647.362.543.245.706.051.143.334.655.263.164.12
74.34
Grass hay
14.2033.8054.20
3.001.104.50
15.002.402.50
5.203.689.391.018.664.261.523.686.063.541.303.978.303.543.615.20
72.92
Crackedcorn8.102.009.500.500.200.701.20
71.502.90
7.784.757.392.77
19.534.093.033.69
12.803.172.374.759.505.153.965.15
99.88
Groundcorn8.302.50
10.400.600.200.801.30
76.001.30
6.634.466.272.05
16.273.732.773.25
11.202.892.054.467.594.583.374.58
86.15
Prolak®
77.609.90
19.708.708.40
12.5013.500.406.90
6.595.849.042.24
10.767.703.462.979.176.141.554.985.945.833.936.76
92.90
77
Table 3.3. Effect of metabolizable protein (MP) supply on milk production andcomposition in Holstein cows
MilkProduction, kg/dCrude protein yield, g/d
True protein yield, g/dCasein yield, g/dWhey protein yield, g/dNonprotein N yield, g/d
Crude protein, %True protein, % of CPCasein, %ofCPWhey protein, %ofCPNonprotein N, %ofCP
Casein profile, % of caseinkappaalpha-slalpha-s2gammabeta, total
Fat yield, g/dFat concentration, %
LowMP33.9
838802629173382.46
95.574.720.84.5
9.445.9
5.64.7
34.41216
3.76
TreatmentsMedMP
35.790585365919452
2.5394.372.821.5
5.7
9.445.7
5.25.1
34.71318
3.65
High MP36.2
96290268921260
2.6593.871.921.96.2
9.345.9
6.84.3
34.21199
3.29
SEM0.49
15.014.16.09.21.50.0690.160.680.680.16
0.320.420.610.300.46
39.70.104
Linear0.02
<0.01<0.01<0.01
0.03<0.01
0.09<0.01
0.030.28
<0.01
0.720.990.190.300.700.140.02
P1
Quadratic0.300.800.950.990.900.100.720.110.550.820.11
0.890.680.180.120.480.170.30
!Least squares means presented with pooled SEM (given for n = 4); n = 5 for Low andMed HP and n = 4 for High MP.2Probability corresponding to the null hypothesis with linear and quadratic contrasts.
Table 3.4. Effect of metabolizable protein (MP) supply on arterial concentration ofamino acids (uM) in Holstein cows.
HistidineIsoleucineLeucineLysineMethioninePhenylalanineThreonineTyrosineValine
TreatmentsLowMP MedMP
49.0121.7155.684.921.857.1
114.156.4
246.0
60.2129.7192.698.522.558.4
116.959.6
315.3
High MP58.7
132.3218.5
95.222.063.1
110.058.9
372.9
SEM3.062.084.111.901.062.743.962.078.45
Linear0.090.02
<0.010.010.910.200.520.45
<0.01
P2
Quadratic0.140.320.310.010.610.630.340.460.58
'Least squares means presented with pooled SEM (given for n = 4); n = 5 for Low andMed HP and n = 4 for High MP.2Probability corresponding to the null hypothesis with linear and quadratic contrasts.
Table 3.5. Effect of metabolizable protein (MP) supply on net fluxes of amino acids(mmol/ h) in Holstein cows.
Amino AcidHistidine
Isoleucine
Leucine
Lysine
Methionine
Phenylalanme
Threonine
Tissue2
PDVHEPTSPMGMILKPDVHEPTSPMGMILKPDVHEPTSPMGMILKPDVHEPTSPMGMILKPDVHEPTSPMGMILKPDVHEPTSPMGMILKPDVHEPTSPMGMILK
LowMP7.8
-3.04.8
-6.66.1
21.02.6
23.5-17.015.735.6
1.036.6
-28.326.129.1-0.129.0
-24.919.910.0-3.07.0
-6.66.6
24.4-11.812.6
-11.110.420.4-6.713.7
-11.812.6
TreatmentsMedMP
12.0-4.47.5
-6.66.6
26.64.4
30.8-19.917.043.2
6.049.1
-32.128.238.00.9
38.5-26.521.413.5-4.29.3
-6.97.0
29.8-15.114.7
-12.011.223.8-6.617.5
-13.813.6
High MP14.2-7.27.0
-6.97.0
28.81.6
30.5-22.418.049.6
2.452.0
-38.329.940.2-4.935.5
-30.122.714.3-6.67.7
-7.97.5
35.4-20.714.8
-13.111.934.5
-14.519.9
-13.214.5
SEM0.920.800.890.380.111.271.470.920.960.281.622.681.821.720.471.952.441.260.360.040.380.590.620.480.121.351.251.380.620.192.703.481.870.840.23
P*Linear<0.01
0.010.170.64
<0.01<0.01
0.69<0.01
0.01<0.01<0.01
0.75<0.01
0.01<0.01<0.01
0.240.020.01
<0.01<0.01
0.010.500.14
<0.01<0.01<0.01
0.350.09
<0.01<0.01
0.190.070.33
<0.01
Quadratic0.390.480.180.720.800.310.230.020.890.800.750.230.070.560.570.200.29
<0.010.410.800.030.440.040.620.800.960.470.570.960.800.300.360.770.260.80
80
Table 3.6. Effect of metabolizable protein (MP) supply on net fluxes of amino acids(mmol/ h) in Holstein cows. (continued)
Tyrosine
Valine
PDVHEPTSPMGMILKPDVHEPTSPMGMIL
K
18.5-8.69.8
-8.8
10.828.4
1.329.7
-20.0
19.7
21.6-11.410.1-9.9
11.734.45.3
39.5-27.0
21.2
24.1-14.010.1-9.8
12.440.5
6.246.8
-28.0
22.6
1.111.330.830.70
0.19
3.512.774.512.40
0.35
0.020.040.850.41
<0.01
0.070.290.050.07
<0.01
0.840.990.860.49
0.80
0.980.650.820.33
0.80
1 Least squares means presented with pooled SEM (given for n = 4); n = 5 for Low andMed HP and n = 4 for High MP.2 PDV: portal-drained viscera; HEP: hepatic; TSP: total splanchnic tissue; MG:mammary gland.3 Probability corresponding to the null hypothesis with linear and quadratic contrasts.
Table 3.7. Effect of metabolizable protein (MP) supply on the ratio of amino acids inmilk protein relative to portai absorption.
Amino AcidHistidineIsoleucineLeucineLysineMethioninePhenylalanineThreonineTyrosineValine
LowMP0.840.810.770.740.670.440.670.810.61
TreatmentsMedMP
0.530.620.640.560.540.360.630.590.53
High MP0.480.570.560.530.490.320.400.520.49
SEM0.070.070.060.070.060.030.100.090.05
P2
Linear0.010.080.060.080.070.030.140.060.19
Quadratic0.250.520.770.450.560.740.510.550.87
1 Least squares means presented with pooled SEM (given for n = 4); n = 5 for Low andMed HP and n = 4 for High MP.2 Probability corresponding to the null hypothesis with linear and quadratic contrasts.
Table 3.8. Effect of metabolizable protein (MP) supply on the ratio of the hepatic netremoval of amino acids relative to portai absorption and total influx.
Amino AcidHistidine
Methionine
Phenylalanine
Threonine
Tyrosine
Ratio2
PDVINFLUXPDVINFLUXPDVINFLUXPDVINFLUXPDVINFLUX
LowMP0.3460.0330.2800.0650.4660.0930.2830.0380.4450.075
TreatmentsMedMP
0.3800.0410.2960.0820.5130.1220.1930.0370.5260.099
High MP0.5060.0650.4430.1260.5810.1460.4440.0650.5760.111
SEM0.0630.0120.0400.0130.0300.0070.1200.0180.0910.009
Linear0.150.140.080.030.09
<0.010.420.380.080.05
Quadratic0.560.590.320.470.810.740.290.530.770.62
1 Least squares means presented with pooled SEM (given for n = 4); n = 5 for Low andMed HP and n = 4 for High MP; given for AA showing a net hepatic removal.2 Ratio of hepatic removal on PDV: net portai flux; or on INFLUX: total liver influx =(portai plasma flow*portal concentration) + (arterial plasma flow*arterialconcentration).3 Probability corresponding to the null hypothesis with linear and quadratic contrasts.
Table 3.9. Effect of metabolizable protein (MP) supply on arterial concentrations (raM)and splanchnic flux (mmol/h) of urea and ammonia in Holstein cows.
MetaboliteAmmonia
Urea-N
ConcentrationFlux
ConcentrationFlux
Tissue2
PDVHEPTSP
PDVHEPTSP
LowMP0.10
265-249
16
4.4-404881476
TreatmentsMedMP
0.08383
-36122
6.3-3331202869
High MP0.10
396-390
6
8.7-17412581083
SEM0.02
15.429.329.8
0.2764.3
183.0218.5
PLinear0.99
<0.010.020.84
<0.010.060.220.12
3
Quadratic0.440.030.280.76
0.430.580.560.74
1 Least squares means presented with pooled SEM (given for n = 4); n = 5 for Low andMed HP and n = 4 for High MP.2 PDV: portal-drained viscera; HEP; hepatic; TSP: total splanchnic tissue.3 Probability corresponding to the null hypothesis with linear and quadratic contrasts.
Table 3.10. Effect of metabolizable protein (MP) supply on nitrogen balance (g/d) inHolstein cows.
NitrogenIntakeFècesDigestedUrineMilkRetained in tissues
LowMP475222253
7913141
TreatmentsMedMP
55022632412214159
High MP62424637816515159
SEM6.19.31.52.72.4
11.5
pi
Linear Quadratic<0.01
0.050.02
<0.01<0.01
0.35
0.980.320.300.910.880.55
1 Least squares means presented with pooled SEM (given for n = 4); n = 5 for Low andMed HP and n = 4 for High MP.2 Probability corresponding to the null hypothesis, with linear and quadratic contrasts.
85
3.9 Références
Amend, J. F., D. H. Strumeyer, and H. Fisher. 1979. Effect of dietary histidine on tissue
concentrations of histidine-containing dipeptides in adult cockerels. J. Nutr. 109:1779-
1786.
AOAC. Association of Officiai Analytical Chemists. 1996. Officiai Methods of
Analysis 16th éd.
Bach, A., G. B. Huntington, S. Calsamiglia, and M. D. Stern. 2000. Nitrogen
metabolism of early lactation cows fed diets with two différent levels of protein and
différent amino acid profiles. J. Dairy Sci. 83:2585-2595.
Bequette, B. J., F. R. Backwell, J. C. MacRae, G. E. Lobley, L. A. Crompton, J. A.
Metcalf, and J. D. Sutton. 1996. Effect of intravenous amino acid infusion on leucine
oxidation across the mammary gland of the lactating goat. J. Dairy Sci. 79:2217-2224.
Blouin, J. P., J. F. Bernier, C. K. Reynolds, G. E. Lobley, P. Dubreuil, and H. Lapierre.
2002. Effect of supply of metabolizable protein on splanchnic fluxes of nutrients and
hormones in lactating dairy cows. J. Dairy Sci. 85:2618-2630.
Calder, A. G. and A. Smith. 1988. Stable isotope ratio analysis of leucine and
ketoisocaproic acid in blood plasma by gas chromatography/mass spectrometry. Use of
tertiary butyldimethylsilyl derivatives. Rapid Commun. Mass. Spectrom. 2:14-16.
Calder, A. G., K. E. Garden, S. E. Anderson, and G. E. Lobley. 1999. Quantitation of
blood and plasma amino acids using isotope dilution électron impact gas
chromatography/mass spectrometry with U-(13)C amino acids as internai standards.
Rapid Commun. Mass. Spectrom. 13:2080-2083.
Canadian Council on Animal Care. Guide to the care and use of expérimental animais.
Cant, J. P., E. J. DePeters, and R. L. Baldwin. 1993. Mammary amino acid utilization in
dairy cows fed fat and its relationship to milk protein dépression. J. Dairy Sci.. 76:762-
774.
Castillo, A. R., E. Kebreab, D. E. Beever, J. H. Barbi, J. D. Sutton, H. C. Kirby, and J.
France. 2001. The effect of protein supplementation on nitrogen utilization in lactating
dairy cows fed grass silage diets. J. Anim. Sci. 79:247-253.
DeSantiago, S., N. Torres, A. Suryawan, A. R. Tovar, and S. M. Hutson. 1998.
Régulation of branched-chain amino acid metabolism in lactating rat. J. Nutr. 128:
1165-1171.
Dinn, N. E., J. A. Shelford, and L. J. Fisher. 1998. Use of the Cornell net carbohydrate
and protein System and rumen-protected lysine and methionine to reduce nitrogen
excrétion from lactating dairy cows. J. Dairy Sci. 81:229-237.
Doepel, L., D. Pacheco, J. J. Kennelly, M. D. Hanigan, F. P. Lôpez, and H. Lapierre.
2004. Milk Protein Synthesis as a Function of Amino Acid Supply. J. Dairy Sci. 87:
1279-1297
Gill, J. L. 1978. Design and analysis of experiments in the animal and médical
sciences.The Iowa State Press, Iowa, USA. Vol 1. 409p.
Goodwin, G. W., W. Gibboney, R. Paxton, R. A. Harris, and J. A. Lemons. 1987.
Activities of branched-chain amino acid aminotransferase and branched-chain 2-oxo
acid dehydrogenase complex in tissues of maternai and fetal sheep. Biochem. J. 242:
305-308.
Hanigan, M. D., J. P. Cant, D. C. Weakley, and J. L. Beckett. 1998. An évaluation of
postabsorptive protein and amino acid metabolism in the lactating dairy cow. J.Dairy
Sci. 81:3385-3401.
Harmeyer, J., and H. Mertens. 1980. Aspects of urea metabolism in ruminants with
référence to the goat. J. Dairy Sci. 63:1707-1728.
Huntington, G. B. 1984. Net absorption of glucose and nitrogenous compounds by
lactating Holstein cows. J. Dairy Sci. 67:1919-1927.
Huntington, G. B., C. K. Reynolds, and B. H. Stroud. 1989. Techniques for measuring
blood flow in splanchnic tissues of cattle. J. Dairy Sci. 72:1583-1595.
«S7
Jaubert, A., and P. Martin, 1992. Reverse phase HPLC analysis of goat caseins.
Identification of asi and aS2 genetic variants. Lait 72:235-247.
Katz, M. L., and E. N. Bergman. 1969. A method for simultaneous cannulation of the
major splanchnic blood vessels of the sheep. Am. J. Vet. Res. 30:655-661.
Kebreab, E., J. France, J. A. Mills, R. Allison, and J. Dijkstra. 2002. A dynamic model
of N metabolism in the lactating dairy cow and an assessment of impact of N excrétion
on the environment. J. Anim. Sci. 80:248-259.
Lapierre, H., and G. E. Lobley. 2001. Nitrogen Recycling in the Ruminant: A Review.
J. Dairy Sci. 84:E223-E236.
Lapierre, H., C. K. Reynolds, T. H. Elsasser, P. Gaudreau, P. Brazeau, and H. F. Tyrrell.
1992. Effects of growth hormone-releasing factor and feed intake on energy metabolism
in growing beef steers: net hormone metabolism by portal-drained viscera and liver. J.
Anim. Sci. 70:742-751.
Lapierre, H., J. F. Bernier, P. Dubreuil, C.K. Reynolds, C. Farmer, D. R. Ouellet and G.
E. Lobley. 2000. The effect of feed intake level on splanchnic metabolism in growing
beef steers. J. Anim. Sci. 78:1084-1099.
Lapierre, H., E. Milne, J. Renaud, and G. E. Lobley. 2003. Lysine utilization by the
mammary gland. Pages 777-780 in Progress in research on energy and protein
metabolism. EAAP plublicationNo.109. Ed. W.B. Souffrant and C.C. Metges.
Lobley, G. E. 2002. Protein turnover - what does it mean for animal production? Pages
1-15 in: Amino acids: méat, milk and more! Ed. H. Lapierre & D.R. Ouellet, CSAS
Symposium, Québec, Canada.
Lobley, G. E., and H. Lapierre. Post-obsortive metabolism of amino acids. 2003. Pages
737-753 in Progress in research on energy and protein metabolism. Ed. W.B. Souffrant
& C.C. Metges. EAAP Pulication No. 109.
Lobley, G. E., D. M. Bremner, and D. S. Brown. 2001. Response in hepatic removal of
amino acids by the sheep to short-term infusions of varied amounts of an amino acid
NX
mixture into the mesenteric vein. Br. J. Nutr. 85:689-698.
Lobley, G. E., X. Shen, G. Le, D. M. Bremner, E. Milne, A. G. Calder, S. E. Anderson,
and N. Dennison. 2003. Oxidation of essential amino acids by the ovine gastrointestinal
tract. Br. J. Nutr. 89:617-630.
Mepham, T. B. 1982. Amino acid utilization by lactating mammary gland. J. Dairy Sci.
65:287-298.
Metcalf, J. A., D. Wray-Cahen, E. E. Chettle, J. D. Sutton, D. E. Beever, L. A.
Crompton, J. C. MacRae, B. J. Bequette, and F. R. Backwell. 1996. The effect of
dietary crude crude protein as protected soybean meal on mammary metabolism in the
lactating dairy cow. J. Dairy Sci. 79:603-611.
Milano, G. D., A. Hotston-Moore, and G. E. Lobley. 2000. Influence of hepatic
ammonia removal on ureagenesis, amino acid utilization and energy metabolism in the
ovine liver. Br. J. Nutr. 83:307-315.
National Research Council. 2001. Nutrient Requirements of Dairy Cattle. 7th rev. éd.
Natl. Acad. Sci. Washington, DC.
Ortigues-Marty, L, C. Obled, D. Darvedet and I. Savary-Auzelux. 2003. Rôle of the
liver in the régulation of energy and protein status. 2003. Pages 83-98 in Progress in
research on energy and protein metabolism. Ed. W.B. Souffrant & C.C. Metges. EAAP
Pulication No. 109.
Petit, H. V. 2002. Digestion, milk production, milk composition, and blood composition
of dairy cows fed whole flaxseed. J. Dairy Sci. 85:1482-1490.
Pink, D., R. Elango, D. Dixon, and R. O. Bail. 2003. Lysine catabolism in the neonatal
piglet during postnatal stages of growth and development. FASEB J. 17 [A702
(Abstract)].
Raggio, G., G. E. Lobley, D. Pellerin, G. Allard, R. Berthiaume, P. Dubreuil, and H.
Lapierre. 2002. Effect of protein intake on synthesis of albumin and plasma total protein
in lactating dairy cows. J. Dairy Sci. 84 (Suppl.l): 243.
89
Reynolds, C. K., D. P. Casper, D. L. Harmon, and C. T. Milton. 1992. Effect of CP and
ME intake on viscéral nutrient metabolism in beef steers. J. Anim. Sci. 70(Suppl.l):
315.
Reynolds, C. K., D. L. Harmon, and M. J. Cecava. 1994. Absorption and delivery of
nutrients for milk protein synthesis by portal-drained viscera. J. Dairy Sci. 77:2787-
2808.
Santos, F. A., J. E. Santos, C. B. Theurer, and J. T. Huber. 1998. Effects of rumen-
undegradable protein on dairy cow performance: a 12- year literature review. J. Dairy
Sci. 81:3182-3213.
SAS User's Guide: Statistics. Edition 1996. SAS Inst, Inc., Cary, NC.
Susmel, P., M. Spanghero, B. Stefanon, and C. R. Mills. 1995. Nitrogen balance and
partitioning of some nitrogen catabolites in milk and urine of lactating cows. Livestock
Prod. Sci. 44:207-219.
Swaisgood, H.E. 1995. Protein and amino acid composition of bovine milk. In:
Handbook of milk composition. R. G. Jensen., éd. Académie Press, Toronto, ON. Pages
464-468.
Thivierge, M. C, J. F. Bernier, P. Dubreuil, and H. Lapierre. 2002. The effect of jugular
or abomasal infusion of amino acids on milk yield in lactating cows fed a protein
déficient diet. Reprod. Nutr. Dev. 42:1-13.
Vanhatalo, A., P. Huhtanen, V. Toivonen, and T. Varkikko. 1999. Response of dairy
cows fed grass silage diets to abomasal infusions of histidine alone or in combination
with methionine and lysine. J. Dairy Sci. 82:2674-2685.
Wray-Cahen, D., J. A. Metcalf, F. R. Backwell, B. J. Bequette, D. S. Brown, J. D.
Sutton, and G. E. Lobley. 1997. Hepatic response to increased exogenous supply of
plasma amino acids by infusion into the mesenteric vein of Holstein-Friesian cows in
late gestation. Br. J. Nutr. 78:913-930.
Wright, T. C, S. Moscardini, P. H. Luimes, P. Susmel, and B. W. McBride. 1998.
90
Effects of rumen-undegradable protein and feed intake on nitrogen balance and milk
protein production in dairy cows. J. Dairy Sci. 81:784-793.
91
CHAPITRE IV
EFFECT OF PROTEIN SUPPLY ON HEPATIC SYNTHESIS OF
CONSTITUTIVE AND PLASMA PROTEINS IN LACTATING DAIRY COWS
Ce chapitre a été soumis dans la revue "Journal of Dairy Science" sous la référence de
Raggio, G., G. E. Lobley, R. Berthiaume, D. Pellerin, G. Allard, and H. Lapierre. 2006.
Effect of protein supply on hepatic synthesis of constitutive and plasma proteins in
lactating dairy cows. J. Dary Sci. (soumis)
4.1 Résumé
L'effet de l'apport de protéines métabolisables (PM) sur la synthèse de protéines
plasmatiques totales et de l'albumine, principalement fabriquées et exportées par le foie,
a été étudié chez la vache laitière. Six vaches multicathéterisées ont reçu deux rations
apportant deux niveaux de PM, 1922 g ou 2517 g/jour de PM, mais fournissant le même
niveau d'énergie. À la fin de chaque période expérimentale, au jour 21, les vaches ont
reçu une perfusion continue avec de 2H5[phénylalanine] (d5-Phe) dans la veine jugulaire
durant 8 h (1,3 mmol/h). Des échantillons de sang ont été prélevés au niveau artériel,
portai et hépatique entre 3 et 8 h. La phénylalanine libre plasmatique, l'albumine et la
protéine plasmatique totale ont été analysées pour déterminer leurs concentrations et
leurs enrichissements isotopiques en d5-Phe. Une diminution de la concentration de
l'albumine (P = 0,02; 32,3 versus 33,7 g/1: 0,11 SEM) a été observée au bas niveau de
PM, mais la concentration en protéines totales plasmatiques n'a pas varié (P > 0,36;
74,1 versus 75,6 g/1; 1,13 SEM). L'incorporation de d5-Phe au cours du temps dans les
protéines plasmatiques et l'albumine totale était linéaire (R2 > 0,98). Ni les taux de
synthèse absolue ou fractionnelle de l'albumine (3,4 vs. 3,4 ± 0,1 % / j ; 36,3 vs. 36,5 ±
1,11 g/j; pour les bas et haut niveaux de PM respectivement) et des protéines
plasmatiques totales (6,8 versus 6,5 ± 0,38 % / j ; 168 vs. 154 ± 19,9 g/j) n'ont été
modifiés (P > 0,20) par l'approvisionnement en PM. L'extraction hépatique nette de
phénylalanine était plus basse avec un régime faible en PM (P = 0,05; -12,3 vs. -20,2 +
1,98 mmol/h). Par conséquent, une plus grande proportion la phénylalanine extraite par
le foie a été utilisée pour la synthèse de protéines d'exportation (P = 0.07; 17,9 versus
11,1 ± 1,83 %) et la synthèse d'albumine (P = 0,09; 4,6 versus 2,9 ± 0,54 %) pour la
ration à bas niveau de PM. Ces résultats suggèrent que la synthèse hépatique de
l'albumine et des autres protéines d'exportation est maintenue chez les vaches laitières
en lactation même lorsque l'alimentation en protéines est réduite.
Mots clés : foie, albumine, synthèse protéique, vache laitière, apport en protéines
4.2 Abstract
The effects of metabolizable protein (MP) supply on synthesis of plasma total
proteins and albumin, as well as total hepatic protein synthesis, were determined in six
multicatheterized, lactating cows. Three TMR formulated to supply the same amount of
energy but différent amounts of MP, 1922 (Low), 2264 (Médium) and 2517 (High) g
MP/day, were fed every 2 h according to a double 3 x 3 Latin Square design. For the
Low and High MP treatments, the cows were continuously infused with 2H5[Phe] (d5-
Phe) into a jugular vein for 8 h (1.3 mmol/h) on d 21 of each period. Concentration and
isotopic enrichment of d5-Phe were measured for free plasma Phe, plasma total proteins
and albumin on hourly samples collected between 3 to 8 h. Low MP decreased plasma
albumin concentration (32.3 vs. 33.7 ± 0.11 g/L) but plasma total protein concentration
was unchanged (74.1 vs. 75.6 ± 1.13 g/L). Incorporation of d5-Phe over time into both
plasma total proteins and albumin was linear (R2 > 0 .98). Neither fractional nor
absolute synthesis rates of plasma total proteins (6.8 vs. 6.5 ± 0.65 %/d; 167 vs. 154 ±
19.9 g/d) or albumin (3.4 vs. 3.4 ± 0.10 %/d; 36.3 vs. 36.5 ±1.11 g/d) were affected by
MP supply. Net hepatic removal of Phe was lower with the Low MP diet (-12.3 vs. -
20.2 ± 1.98 mmol/h). As a resuit, net hepatic Phe removal used for total export protein
synthesis (17.9 vs. 11.1 ± 1.83 %) and albumin synthesis (4.6 vs. 2.9 ± 0.54 %) was
greater at Low MP. Thèse results suggest that hepatic synthesis of plasma proteins,
including albumin, is maintained in lactating dairy cows even when protein supply is
reduced.
Key words: liver, albumin, protein synthesis, dairy cow, protein intake
94
4.3 Introduction
The liver is the major site of removal of certain AA in both ruminants and non-
ruminants (Rérat et al., 1992; Lobley and Lapierre, 2003; Hanigan, 2005). In dairy
cows, for example, approximately 45 % of net portai appearance of total AA is
extracted by the liver (Lapierre et al., 2005). This composite value, however, masks the
considérable variation between AA, with fractional removals of net portai appearance of
essential A A varying between 0 (for the branched-chain A A) and 50 % (Phe). Despite
the fact that the liver is the principal site of ureagenesis in the body (Lapierre and
Lobley, 2001) thèse hepatic extractions do not encompass only catabolism. Instead, part
of the net removal can also be used for synthesis of export proteins (Jahoor et al., 1996;
Connell et al., 1997) plus any net accretion of liver protein during growth (Burrin et al.,
1990).
The différent hepatic export proteins play a variety of rôles, including
maintenance of vascular osmotic pressure (e.g. albumin), coagulation (fibrinogen),
immunity (C reactive protein) and anti-phagocytic mechanisms (ool-antitrypsin, ool-
antichymotrypsin, oo2-macroglobulin). Thèse functions are vital to the metabolic
integrity of the animal and, therefore, synthesis of export proteins is maintained even
under conditions of nutritional deprivation, including low protein intakes and fasting in
sheep and humans (Hunter et al., 1995; Connell et al., 1997; Barber et al., 2000). This is
important because the contribution of export proteins to total hepatic synthesis can be
considérable, e.g. albumin accounted for 18 % of total liver protein synthesis in
fattening lambs (Connell et al., 1997). Such synthesis may hâve implications for AA
supply beyond the liver and support of anabolic processes. For example, the unusual
AA composition of export proteins has led to the suggestion that post-hepatic supply of
free AA to support peripheral anabolism may be compromised, particularly in the case
of Phe and Cys (Reeds et al., 1992). Of course, thèse export proteins may provide an
anabolic source of AA to peripheral tissues, many of which hâve the ability to dégrade
albumin (Eskild et al., 1989; Maxwell et al., 1990). Protein sources, including those in
blood, can provide short-term réservoirs of AA and hâve the advantage of not inducing
95
regulatory mechanisms that would accompany supply of similar quantities of the free
AA.
Therefore, the objectives of this study were, in lactating dairy cows: 1) to
quantify the rate of synthesis of plasma protein synthesis; 2) to estimate the maximum
proportion of net hepatic removal Phe the liver could use to synthesize export proteins;
3) to establish the maximum proportion of hepatic protein synthesis directed towards
export proteins vs. constitutive proteins; and 4) to détermine how thèse parameters are
affected by protein supply.
4.4 Materials and Methods
4.4.1 Animais, Treatments and Sampling
Six multiparous Holstein cows, averaging 656 ± 60 kg BW and 96 ± 8 DIM at
the beginning of the study were used in a double 3 x 3 Latin square design balanced for
residual effects with 21-d expérimental periods. Ail cows were surgically implanted
with cathéters in the portai vein, one hepatic and two mesenteric veins and in a
mesenteric artery. In addition, the right carotid artery was surgically raised to a
subcutaneous position (see Raggio et al., 2004). Three diets were balanced to provide
the same amount of energy (36.4 Mcal NF^/d) but with increasing amounts of
metabolizable protein (MP), averaging 1922 (Low), 2264 (Médium), and 2517 (High)
g MP/d (NRC, 2001). Diets were as detailed previously (Raggio et al., 2004) and
offered in equal quantities every 2 h from automated feeders (Ankom, Fairport, NY),
except for 1 kg of long hay that was offered once a day. The cows had free access to
water and were housed in a tie stall barn, lit from 0600 to 2200. Cows were milked
twice daily (0600 and 1800) and yield was recorded at each milking.
On day 18, 19 or 20 of each expérimental period, measurements were made of
splanchnic plasma flows by downstream dilution of p-aminohippurate and of mammary
plasma flow using the Fick principle, as reported previously (Raggio et al., 2004). At
0600 on the last day of each period (d 21), cows on Low and High MP diets received an
8-h (1.3 mmol/h) infusion of 2H5-Phe (d5-Phe; Cambridge Isotope Laboratories,
Andover, MA, 99 atom %), preceded by a priming dose (1.3 mmol). Infusions of d5-
96
Phe were limited to the Low and High treatments due to the cost of the tracer. Blood
samples were collected simultaneously every hour from the artery plus portai and
hepatic veins starting 3 h after initiation of the infusion (n = 6). Blood samples from the
mammary vein were obtained by venepuncture every other hour (n = 3). Two blood
samples were taken from each cathéter prior to the start of infusion for détermination of
natural abundance.
The expérimental protocol was approved by the Committee for Animal Care of
the Lennoxville Research Centre and conducted according to the guidelines of the
Canadian Council on Animal Care (1993).
4.4.2 Laboratory Analyses
Plasma was obtained immediately following blood sampling by centrifugation at
1000g for 15 min at 4°C. Plasma Phe concentrations were measured by the gravimétrie
isotope dilution method (Calder et al., 1999). On the day of sampling, 0.2 g of an
internai standard (226 uM of l-13C-Phe, 99 atom %; C/D/N isotopes, Montréal, QC,
Canada) was added to 1 g of plasma and the mixture was stored at -20°C for later
analysis. At the time of analysis, thèse samples, as well as plasma taken for isotopic
enrichment analysis, were thawed, mixed, deproteinized with sulfosalicylic acid (380
g/1) and desalted on AG-50 H+ resin. The freeze-dried eluate was derivatized with N-
(tert-butyldimethysilyl)-N-methyltrifluoroacetamide (MTBSTFA): acetronile (1:1) to
form the N-(tert-butyldimethyl) AA derivative (Calder and Smith, 1988). The Phe
isotopic enrichments (IE) were quantified by gas chromatography-mass spectrometry
(GC-MS) as either m/z ions 321, 322 (to détermine concentrations on samples with the
internai standard, l-13C-Phe) or as m/z ions 336, 341 (to détermine the IE of plasma free
d5-Phe; GC-MS, Hewlett Packard, Model GC6890-MS973, Agilent Technologies,
Willmington, DE).
In addition, détermination of the enrichment of d5-Phe incorporated into plasma
total proteins and albumin was determined as described by Connell et al. (1997) as m/z
ions 336, 341 (GC-MS Voyager, Thermoquest, Wythenshawe, Lancashire, UK).
Détermination of the concentration of plasma total proteins and of albumin was
performed on a clinical analyzer (KONE Instruments Corporation, Espoo, Finland)
calibrated with appropriate standard supplied by manufacturer (Thermo Electron
Systems, Waltham, MA).
4.4.3 Calculations and Statistical Analyses
During the infusion of d5- Phe, the temporal increase in IE of d5-Phe
incorporated into plasma total proteins and albumin were plotted against time (between
3 and 8 h) to détermine linear régressions, as described previously (Connell et al.,
1997). The slope was used in conjunction with the appropriate precursor pool (taken as
the IE of free d5-Phe in the hepatic vein at plateau, i.e. from 3 to 8 h of infusion, see
Connell et al., 1997) to calculate the fractional rate of protein synthesis of both plasma
total proteins and albumin.
Daily fractional rates (FSR) of plasma total proteins (PTP) and albumin (Alb)
were calculated as follows:
FSR (%/day) = slope (IE of PTP or Alb vs. time (h)) / IEH * 24 * 100
where IE was expressed as mole percent excess (mpe).
Half-life (d) was estimated from Ln(2)/(FSR/100).
The absolute rates of PTP and Alb synthesis (ASR) were calculated as:
ASR (g/d) = FSR (PTP or Alb) * [ PTP or Alb ] * PV
where [ PTP or Alb ] refers to the plasma concentrations of plasma total proteins
or albumin (g/L), respectively, and PV to the plasma volume (L), assumed to average
49.9 mL/kg of body weight (Girard et al., 1989). To estimate the requirement of Phe for
the ASR of plasma total proteins or albumin, we assumed that plasma total proteins
contain 5 % Phe (Connell et al., 1997) and albumin contains 6 % Phe (Peters et al.,
1985)
Net fluxes of Phe across the portal-drained viscera (PDV), liver, and splanchnic
tissues (TSP = PDV plus liver) as well as mammary gland were calculated for each cow
as the product of the average of the venous-arterial concentration différence and the
plasma flow, estimated on the day prior to the d5-Phe infusion (see Raggio et al., 2004).
A négative flux means utilization or removal, while a positive flux indicates net
production or release.
Whole body (WB) Phe irréversible loss rate (ILR; mmol/h) was calculated as
follows:
WB-ILR = (IEinf / IEA - 1) x Inf
where IEjnf and IEA are the IE of the infusate and the mean IE of d5-Phe in
arterial plasma free Phe, and Inf is the infusion rate (mmol/h) of the d5-Phe.
Phenylalanine ILR (mmol/h) through the liver was calculated as follows:
Hepatic-ILR = ((([PheH] x IEphe.H - [PheP] x IEPhe.p) x PFP) + (([Pheh] x IEPhe.H -
[PheA]xIEPhe.A)xPFA)))/IEH
where the subscript indicate the site of plasma collection (A for artery, P for
portai vein, H for hepatic vein), IEPhe, [Phe] and PF represent, respectively, the IE (mpe)
and concentration (mM) of plasma free Phe and the plasma flow (L/h) in the
corresponding vessel.
Since the cows were in mid-lactation, changes in liver protein mass were
assumed to be zéro and, therefore, hepatic Phe oxidation (mmol/h) was estimated to be
the différence between net Phe liver removal minus Phe used for plasma total protein
synthesis. Then Phe used in the liver for protein synthesis was calculated as:
Hepatic protein synthesis = hepatic ILR - hepatic oxidation
Phenylalanine used for the synthesis of constitutive protein in the liver was
calculated as the différence between that used for total hepatic protein synthesis and that
used for the synthesis of plasma total proteins.
Data were analyzed statistically using the MIXED procédure of SAS (SAS
Institute Inc. 2004). One cow did not hâve a patent hepatic cathéter and her portai
cathéter was also misplaced. Therefore data for this cow were discarded from ail
analyses. In addition, one cow suffered from severe mastitis during her last period (on
High MP) and this cow period was also excluded. Therefore, n = 5 for Low MP and n =
99
4 for High MP. Treatment différences were considered significant if P < 0.05 and as a
tendency with 0.05 < P < 0.10.
4.5 Results
The increase in MP supply resulted in a 15 % improvement in milk protein yield
(848 vs. 980 g/d, P = 0.05) as reported previously (Raggio et al., 2004). For both plasma
total proteins and albumin there was a linear (R2 > 0.98) incorporation of d5-Phe
between 3 to 8 h of the d5-Phe infusion (Fig 1). Synthesis rates, both fractional and
absolute, of plasma total proteins and albumin plus their concentrations are presented in
Table 1. Albumin concentration increased with the High MP diet, by 4 % (P = 0.02).
Concentrations and 1E of plasma free Phe are presented in Table 2. The
concentration of Phe increased (P < 0.07) in ail blood vessels with the High MP diet,
whereas the 1E of Phe decreased (P < 0.03). Phenylalanine net flux across the PDV
increased (23.5 vs. 32.6; P = 0.06) with the High MP diet (Table 3). This was
accompanied by increased (P = 0.05) net hepatic Phe removal with, as a resuit, no
change (P = 0.72) in net supply beyond the total splanchnic tissues. The maximum
proportion of hepatic Phe removal used for export of plasma total protein synthesis
(17.9 vs. 11.1 ± 1.83 %; P = 0.07) and for albumin synthesis (4.6 vs. 2.9 ± 0.54 %; P =
0.09) tended to decrease at the higher MP supply. There was only a numerical increase
(P = 0.18) in mammary gland net uptake of Phe between the two diets and this uptake
matched net Phe supply beyond the liver and milk output.
Phenylalanine kinetics are presented in Table 4. The High MP diet increased (P
< 0.01) WB 1LR while Phe hepatic 1LR increased by 51 % (P = 0.04). This latter effect
was due to a combination of increased hepatic oxidation (P = 0.05) and hepatic protein
synthesis (P = 0.04). The synthesis of constitutive proteins increased with High MP (P
= 0.03), with a tendency for thèse to represent a higher proportion of total hepatic
protein synthesis (83.3 vs. 88.4 ± 1.39 %; P = 0.06).
4.6 Discussion
The objectives were to examine whether MP supply alters 1) synthesis of plasma
proteins, including albumin, mainly synthesized by the liver; 2) the proportion of net
100
hepatic Phe removal not catabolized but used to support synthesis of export proteins
(which could support milk protein production); and 3) the proportion of hepatic protein
synthesis directed towards constitutive proteins vs. export proteins. Interprétation of the
results, however, requires considération of the limitation of the techniques used.
Because multicatherized animais were used, liver biopsies were not performed and,
therefore, the IE of the hepatic precursor pool for protein synthesis was assumed similar
to the IE of plasma free d5-Phe in the hepatic vein. In sheep, this has been shown to be a
valid assumption for estimation of the synthesis of export proteins (Connell et al.,
1997), formed on the rough endoplasmic reticulum in close contact with the
extracellular pools. For constitutive protein synthesis, however, the precursor IE is
probably lower and close to that in the hepatic cytosol (Fern and Garlick, 1976; Smith
and Sun, 1995; Connell et al., 1997). As this intracellular IE could not be estimated in
the current study, the values of the constitutive protein synthesis are underestimated.
Based on data from fed sheep, this underestimate would approximate to 25 % (Connell
et al., 1997). Furthermore, plasma total proteins involve a mixture derived from liver
(including albumin, fibrinogen and 60 to 80 % of plasma globulins) and other sources
(e.g. the y globulins are synthesized in lymphoid tissues and other cells of the
reticuloendothelial System; Guyton, 1981). Therefore, assuming that plasma total
proteins are ail hepatic-derived will yield an over-estimation of liver activity.
Nonetheless, provided that the two MP supplies do not influence either the relationship
between intracellulanvascular IE or in the proportion of export proteins derived from
liver, then the relative responses to diet can be properly considered.
The absolute synthesis rates of both plasma total proteins and albumin were
unaltered by the change in MP supply. Studies in other species hâve reported albumin
synthesis to be sensitive to nutrient supply, with decreases in both 3-d starved sheep
(Connell et al., 1997) and overnight-fasted humans (Hunter et al., 1995). In addition,
low protein supply also decreases albumin synthesis in humans (Barber et al., 2000) and
pigs (Jahoor et al., 1999). Thèse reported experiments exerted moderate to severe
nutrient deprivations whereas in the current study the lowest MP supply still supported
over 30 kg of milk/d. Therefore, the lack of response in export protein synthesis in thèse
lactating cows may reflect the relatively mild nature of the nutrient différences or,
101
alternatively, may indicate the demands on the lactating cow require maintenance of
plasma protein synthesis to ensure biological integrity and function. Indeed, in
comparative terms, plasma total protein synthesis, expressed on a metabolic BW basis,
is slightly higher in the lactating dairy cow than in the non-lactating sheep (1.24 vs. 1.04
g/kg BW0'75; current study and Connell et al., 1997). A similar différence was observed
for albumin (0.28 vs. 0.21 g/kg BW075). This latter différence may reflect the need for
export albumin (and other plasma proteins) in the milk which averages 14 g of albumin
(based on 0.4 g albumin/kg milk protein), representing 38 % of albumin synthesis.
Hepatic export protein synthesis did not change even though net portai
absorption of Phe was decreased on the Low MP diet, as was the hepatic extraction.
Therefore, at the lower MP supply the proportion of the Phe removed by the liver used
for export protein synthesis increased. while Phe oxidation decreased by 33 %.
Interestingly, although export protein synthesis was not affected by protein supply, the
estimate of constitutive protein synthesis was reduced markedly (39 %, P = 0.03) at the
lower MP. This meant a substantial change occurred in the proportion of total liver
protein synthesis represented by export protein between High and Low MP supply (17
vs. 12 %). Nonetheless, thèse proportions are still much lower than in non-lactating
sheep (38-51 %; Connell et al., 1997) and reflects, for the dairy cow, a higher estimated
FSR for constitutive proteins, 0.51-0.70/d (based on 788 and 1089 g of constitutive
protein per day for Low and High MP, respectively, liver size of 1.47 % BW (Reynolds
et al., 2004) and 16.7 % of CP in the liver (Girard, personal communication)) compared
with 0.19/d in sheep (Connell et al., 1997). As the FSR of hepatic constitutive proteins
is sensitive to protein intake, the différence between thèse cows and the values observed
in sheep may reflect a combination of greater metabolic aetivity of the liver at higher
protein intakes and species différences.
As a resuit of increased MP supply, the net portai appearance of Phe increased
and this led to a 14-30 % increase in plasma Phe concentrations and conséquent
elevated net Phe hepatic removal, as observed in other studies (Bach et al., 2000; Blouin
et al., 2002). Net hepatic extraction represented 55 % and 61 % of net portai absorption
atLow and High MP intake, respectively, with Phe usually one of the AA extracted in
102
the greatest proportion relative to portai absorption in lactating cows (Hanigan, 2005;
Lapierre et al., 2005). However, the fraction of removal relative to total inflow to the
liver, i.e., total flow from the portai vein plus the hepatic artery, did not differ between
MP supply, representing 10 % vs. 13 % of total inflow for Low vs. High MP,
respectively. Hanigan (2005) and Lapierre et al. (2005) hâve previously suggested that
the fractional hepatic removal of AA relative to total inflow is relatively constant in
lactating cows. Post-splanchnic free Phe supply matched both uptake and output (milk)
by the mammary gland, as observed in a number of récent reports (see Lobley and
Lapierre, 2003; Lapierre et al., 2005). Such findings negate the need to invoke plasma
protein as a potential source of AA, at least for milk synthesis.
Based on the présent findings, the hypothesis that synthesis of hepatic export
proteins may limit free Phe supply post-liver for anabolism (Reeds et al., 1992) does not
apply in the healthy, lactating cow. Furthermore, while export proteins can provide a
source of AA to tissues, as reported for skin, muscle and liver (Eskild et al., 1989;
Maxwell et al., 1990), this does not seem to be necessary to support mammary gland
metabolism, as assessed from the close quantitative agreement between supply and
utilization for milk protein sécrétion. Whether AA from export proteins are utilized in
the dairy cow for anabolic purposes by non-hepatic tissues or enter catabolic pathways
beyond the liver (e.g. dégradation within the intestinal tract; Poppi et al., 1986) cannot
be identified from the current study. If they do provide an anabolic component, this
would supply, at maximum, an additional 18-20 % above that available as free Phe.
4.7 Conclusion
In healthy, mid-lactation dairy cows, synthesis of either plasma total proteins or
albumin is not altered by metabolizable protein supply. Their constant synthesis, across
a 30 % change in MP intake, suggests their metabolic priority is maintained even
though milk protein yield altered by 15 %. Although a larger proportion of Phe
extracted by the liver was used to support export protein synthesis at Low MP supply
this was offset by lower hepatic oxidation and so the amount of free Phe available to
support milk protein production remained unchanged.
103
4.8 Aknowledgments
The authors gratefully thank the staff of the Dairy and Swine R&D Centre for
taking care of the animais, M. Léonard, S. Provencher and J. Renaud for their dedicated
technical support, as well as S. Méthot for statistical analyses. Grateful thanks are
extended to F. Boutachkourt for isolation of the plasma albumin and total proteins and
to S. Anderson for the mass spectrometric analyses of the protein bound enrichments.
The authors also wish to acknowledge H.J. Baker & Bro. Inc. for supplying the Pro-Lak
bypass protein supplément and the financial support of the Action concertée Fonds
FCAR - Novalait - MAPAQ, the National Science and Engineering Research Council
of Canada, a grant to the Rowett Research Institute from the Scottish Executive
Environmental and Rural Affairs Department and Agriculture and Agri-Food Canada
(Contribution number xxx).
104
Figure 4. 1. Average of the enrichment (MPE= molar percent excess x 1000) of Phe inplasma total proteins and albumin during a continuous infusion of 2H5-Phe at Low (la)or High (lb) metabolizable protein (MP) supply.
Figure a. Low MP
1 5 G 1 B 9
houii al irfuiion
Albumin: R2~ 0.998, SE = 1.8 Total proteins: R = 0.998, SE= 2.2
Figure b. High MP
2 3 4 5
Albumin: R2 = 0.987, SE = 1.5 Total proteins: R2 = 0.989, SE = 2.9
105
Table 4.1. Effect of supply of metabolizable protein (MP) on concentrations and proteinsynthesis rates of plasma albumin and total proteins.
Concentration, g/1
FSR2, %
ASR3, g/d
AlbuminTotal plasma proteinAlbuminTotal plasma proteinAlbuminTotal plasma protein
TreatmentsLow MP High
32.374.1
3.46.8
36.3168
MP33.775.63.46.4
36.5154
SEM1
0.111.130.090.651.11
19.9
P0.020.360.790.680.870.64
'Least square présentée! with pooled SEM (given for n = 4) Low MP, n = 5 and High MP, n = 4.2Fractional synthesis rate3Absolute synthesis rate
106
Table 4.2. Effect of supply of metabolizable protein (MP) on Phe concentration andisotopic enrichment plasma samples during a 8-h infusion of 2Hs-Phe1.
PhenylalanineConcentration, uM
Isotopic enrichment, mpe2
SiteArterialPortaiHepaticMammaryArterialPortaiHepaticMammary
TreatmentsLowMP
59.777.766.638.3
3.52.32.33.5
High MP71.698.279.350.12.92.01.92.8
SEM1
2.612.502.072.640.090.070.060.11
P0.040.010.020.070.010.030.030.03
'Least square presented with pooled SEM (given for n = 4) Low MP; n = 5 and High MP; n = 4.2 mpe: molar percent excess
107
Table 4.3. Effect of supply of metabolizable protein (MP) on net transfers of Phe(mmol/h)
TissuePortal-drained visceraHepaticTotal splanchnic tissueMammary glandMilk
TreatmentsLowMP
23.5-12.311.2
-11.310.5
High MP32.6
-20.212.5
-12.912.1
SEM2.341.982.640.560.43
P0.060.050.720.180.05
'Least square presented with pooled SEM (given for n=4) Low MP; n = 5 and High MP; n = 4.
108
Table 4.4. Effect of supply of metabolizable protein (MP) on Phe kinetics (mmol/h)
ItemWhole body-ILR2
Hepatic-ILRHepatic OxidationHepatic total protein synthesisHepatic constitutive protein constitutiveProtein synthesis: constitutive/total
TreatmentsLowMP
35.4-23.3-10.2-13.0-10.9
0.83
High MP42.9-35.3-18.2-17.1-15.1
0.88
SEM0.872.661.890.920.810.01
P<0.01
0.040.050.040.030.06
'Least square presented with pooled SEM (given for n=4) Low MP; n = 5 and High MP; n = 4.2ILR = irréversible loss rate
109
4.9 Références
Bach, A., G. B. Huntington, S. Calsamiglia, and M. D. Stern. 2000. Nitrogenmetabolism of early lactation cows fed diets with two différent levels of protein anddifférent amino acid profiles. J. Dairy. Sci. 83:2585-2595.
Barber, M. D., K. C. Fearon, D. C. McMillan, C. Slater, J. A. Ross, and T. Preston.2000. Liver export protein synthetic rates are increased by oral meal feeding in weight-losing cancer patients. Am. J. Physiol. Endocrinol. Metab. 279:E707-E714.
Blouin, J. P., J. F. Bernier, C. K. Reynolds, G. E. Lobley, P. Dubreuil, and H. Lapierre.2002. Effect of supply of metabolizable protein on splanchnic fluxes of nutrients andhormones in lactating dairy cows. J. Dairy. Sci. 85:2618-2630.
Burrin, D. G., C. L. Ferrell, R. A. Britton, and M. Bauer. 1990. Level of nutrition andviscéral organ size and metabolic activity in sheep. Br. J. Nutr. 64:439-448.
Calder, A. G. and A. Smith. 1988. Stable isotope ratio analysis of leucine andketoisocaproic acid in blood plasma by gas chromatography/mass spectrometry. Use oftertiary butyldimethylsilyl derivatives. Rapid Commun. Mass Spectrom. 2:14-16.
Calder, A. G., K. E. Garden, S. E. Anderson, and G. E. Lobley. 1999. Quantitation ofblood and plasma amino acids using isotope dilution électron impact gaschromatography/mass spectrometry with U-(13)C amino acids as internai standards.Rapid Commun. Mass Spectrom. 13:2080-2083.
Canadian Council on Animal Care. 1993. Guide to the Care and Use of ExpérimentalAnimais. Vol. l.E.D. Offert, B.M. Cross, and A.A. McWilliam, eds. CCAC, Ottawa,Ontario
Connell, A., A. G. Calder, S. E. Anderson, and G. E. Lobley. 1997. Hepatic proteinsynthesis in the sheep: effect of intake as monitored by use of stable-isotope-labelledglycine, leucine and Phe. Br. J. Nutr. 77:255-271.
Eskild, W., G. M. Kindberg, B. Smedsrod, R. Blomhoff, K. R. Norum, and T. Berg.1989. Intracellular transport of formaldehyde-treated sérum albumin in liver endothelialcells after uptake via scavenger receptors. Biochem. J. 258:511-520.
Fern, E. B., and P. J. Garlick. 1976. Compartmentation of albumin and ferritin synthesisin rat liver in vivo. Biochem. J. 156:189-192.
Girard, C. L., J. J. Matte, and G. F. Tremblay. 1989. Sérum folates in gestating andlactating dairy cows. J. Dairy Sci. 72:3240-3246.
Guyton, A. C. 1981. Physics of Blood, Blood Flow and Pressure: Hemodynamic.Textbook of Médical Physiology. W,B. Saunders company, Philadelphia, PA.
110
Hanigan, M. D. 2005. Quantitative aspects of ruminant splanchnic metabolism asrelated to predicting animal performance. Anim. Sci .80:23-32.
Hunter, K. A., P. E. Ballmer, S. E. Anderson, J. Broom, P. J. Garlick, and M. A.McNurlan. 1995. Acute stimulation of albumin synthesis rate with oral meal feeding inhealthy subjects measured with [ring-2H5]Phe. Clin. Sci (Lond) 88:235-242.
Jahoor, F., S. Bhattiprolu, M. Del Rosario, D. Burrin, L. Wykes, and M. Frazer. 1996.Chronic protein deficiency differentially affects the kinetics of plasma proteins in youngpigs. J.Nutr. 126:1489-1495.
Jahoor, F., L. Wykes, M. Del Rosario, M. Frazer, and P. J. Reeds. 1999. Chronic proteinundernutrition and an acute inflammatory stimulus elicit différent protein kineticresponses in plasma but not in muscle of piglets. J. Nutr. 129:693-699.
Lapierre, H. and G. E. Lobley. 2001. Nitrogen Recycling in the Ruminant: J. Dairy Sci.84: (suppl) E223-E236.
Lapierre, H., R. Berthiaume, G. Raggio, L. Doepel, D. Pacheco, P. Dubreuil, and G. E.Lobley. 2005. The route of absorbed nitrogen into milk protein. Anim. Sci. 80:11-22.
Lobley, G. E. and H. Lapierre. 2003. Post-absorptive metabolism of amino acids. InProgress in research on energy and protein metabolism (ed, W. B. Souffrant and C. C.Metges). Eur. Assoc. Anim. Prod. Publ. 109:737-756.
Maxwell, J. L., L. Terracio, T. K. Borg, J. W. Baynes, and S. R. Thorpe. 1990. Afluorescent residualizing label for studies on protein uptake and catabolism in vivo andin vitro. Biochem. J. 267:155-162.
National Research Council. 2001. Nutrient Requierements of Dairy Cattle, 7th rev. ed.Natl. Acad. Sci., Washington, DC.
Peters, T. Jr. 1985. Sérum Albumin. Adv. Protein. Chem. 37: 161-245.
Poppi, D. P., J. C. MacRae, A. Brewer, and R. L. Coop. 1986. Nitrogen transactions inthe digestive tract of lambs exposed to the intestinal parasite, Trichostrongyluscolubriformis. Br. J. Nutr. 55:593-602.
Raggio, G., D. Pacheco, R. Berthiaume, G. E. Lobley, D. Pellerin, G. Allard, P.Dubreuil, and H. Lapierre. 2004. Effect of Level of Metabolizable Protein onSplanchnic Flux of Amino Acids in Lactating Dairy Cows. J. Dairy Sci. 87:3461-3472.
Reeds, P. J., D. L. Hachey, B. W. Patterson, K. J. Motil, and P. D. Klein. 1992. VLDLapolipoprotein B-100, a potential indicator of the isotopic labeling of the hepatic proteinsynthetic precursor pool in humans: studies with multiple stable isotopically labeledamino acids. J. Nutr. 122:457-466.
111
Rerat, A., C. Simoes-Nunes, F. Mendy, P. Vaissade, and P. Vaugelade. 1992.Splanchnic fluxes of amino acids after duodenal infusion of carbohydrate solutionscontaining free amino acids or oligopeptides in the non-anaesthetized pig. Br. J. Nutr.68:111-138.
Reynolds, C. K., B. Durst, B. Lupoli, D. J. Humphries, and D. E. Beever. 2004. Viscéraltissue mass and rumen volume in dairy cows during the transition from late gestation toearly lactation. J. Dairy Sci. 87:961-971.
SAS® Institute Inc. 2004. User's Guide, Version 8. Edition, SAS Inc. Cary, NC.
Smith, C. B., and Y. Sun. 1995. Influence of valine flooding on channeling of valineinto tissue pools and on protein synthesis. Am. J. Physiol. 268:E735-E744.
112
CHAPITRE V
EFFECT OF CASEIN AND PROPIONATE SUPPLY ON WHOLE BODY
PROTEIN METABOLISM IN LACTATING DAIRY COWS
Ce chapitre a été publié dans la revue "Canadian Journal of Animal Science " sous la
référence de G. Raggio, G.E. Lobley, S. Lemosquet, H. Rulquin, and H. Lapierre. 2006.
Effect of casein and propionate supply on whole body protein metabolism in lactating
dairy cows. Can. J. Anim. Sci. 86: 81-89.
113
5.1 Résumé
Les effets de perfusions de caséine (Cas) et de propionate (C3) sur le
métabolisme corporel de la leucine (Leu) ont été étudiés. Pour cela, trois vaches
Holstein multipares, fistulisées aux niveaux ruminai et duodénal ont été utilisées selon
un double carré de Youden avec des périodes expérimentales de 14 jours. Les vaches
recevaient une ration à base de balles rondes enrubannées, apportant 124,1 MJ/jour
d'ENL et 1593 g/jour de protéines digestibles dans l'intestin (PDI ; INRA 1989). Les
effets de perfusions de Cas (743 g/jour au duodénum) et de C3 (1041 g/jour au rumen)
ont été étudiés selon un plan factoriel. À chaque période, de la L[l-13C]Leu (4.3
mmol/h) a été perfusé au jour 11 et du [13C]bicarbonate de sodium (4.2 mmol/h) a été
perfusée au jour 13 dans une veine jugulaire. Des prélèvements sanguins carotidiens ont
été effectués pour mesurer l'enrichissement isotopique du 13C[4-methyl 2-
oxopentanoate] (MOP, jour 11) et du 13CO2 (jour 11 & 13). Les traitements de Cas et de
C3 ont augmenté le taux protéique du lait et la production de matières protéiques, mais
seuls les traitements de Cas ont augmenté le volume de lait (18 %). L'augmentation du
flux corporel de la Leu (26 %), de son oxydation (146 %), de son utilisation pour la
synthèse protéique (15 %) et sa sécrétion dans la protéine du lait (21 %) suggèrent une
stimulation générale du métabolisme protéique par les traitements Cas. Les traitements
C3 ont tendance à augmenter le flux corporel de Leu (5 %), la synthèse protéique (7 %)
et la Leu sécrétée dans le lait (7 %). Cependant sur l'oxydation corporelle de la Leu, il y
avait une tendance à une interaction entre Cas et C3. Ceci suggère que l'impact de
l'énergie sur le métabolisme protéique est tributaire du niveau de protéines fournies à
l'animal.
Mots-clés: vache laitière, leucine, caséine, propionate, flux corporel.
114
5.2 Abstract
The effects of Casein (Cas) and propionate (C3) on whole body (WB) leucine
(Leu) metabolism were determined in three multiparous Holstein cows, fitted with both
duodénum and rumen cannulas, used in a Youden replicated square with 14 d-periods.
Cows were fed a grass silage-based diet estimated to supply 124.1 MJ/d of NEL and
1593 g /d of protein digested in the intestine (INRA 1989). Cas (743 g/d in the
duodénum) and C3 (1041 g/d in the rumen) infusions were tested in a factorial
arrangement. For each period, on d 11, L[l-13C]leucine (4.3 mmol/h) and on d 13,
[13C]sodium bicarbonate (4.2 mmol/h) were infused into a jugular vein. Blood samples
were taken from a carotid artery to measure enrichments of 13CC>2 (d 11 & 13) and of13C[4-methyl 2-oxopentanoate] (MOP, d 11). Both Cas and C3 treatments separately
increased milk protein concentration and yield, but only Cas treatments increased milk
yield (18 %). The overall incréments in Leu irréversible loss rate (ILR) (26 %),
oxidation (146 %), protein synthesis (15 %), and output in milk protein (21 %) suggest a
gênerai response in protein turnover to Cas treatments. C3 treatments tended to increase
Leu WB ILR (5 %), protein synthesis (7 %) and in milk (7 %), with also a tendency for
a Cas x C3 interaction on WB Leu oxidation. The latter suggests that the impact of
energy on protein metabolism dépends on the level of protein supply.
Key words: Dairy cow, leucine, casein, propionate, whole body irréversible loss rate.
115
5.3 Introduction
In dairy cows, milk protein yield can be manipulated by varying either energy or
protein intakes (de Peters and Cant 1992). Thèse responses may reflect changes in
whole body (WB) protein metabolism (including synthesis and oxidation), as observed
when total food supply is altered in growing cattle (Lobley et al. 1987; Lapierre et al.
1999). Of more interest is to know whether, and how, individual macronutrients,
particularly energy or protein, might impact on protein metabolism and animal
performance. This is important because, in mechanistic terms, either protein/amino
acids (AA) or energy (glucose or precursors such as propionate) may act differently.
In the non-lactating state, there appears to be consistent responses in WB
metabolism to increased protein supply across a wide range of species. For example, in
sheep, infusion of casein into the abomasum (Liu et al. 1995) elevated WB protein
turnover, as did increased protein intake in humans (Garlick et al. 1991). In such cases,
the increased WB protein turnover was the resuit of a combined incrément in both
oxidation and protein synthesis. In contrast, for the lactating dairy cow, changes in WB
protein metabolism appear to be less consistent. For example, WB leucine kinetics can
be either unaffected by an incrément in crude protein supply (Bequette et al. 1996b),
numerically increased with a caseinate infusion (Oldham et al. 1980) or signifïcantly
improved with additional metabolisable protein supply (Lapierre et al. 2002). The
reasons for thèse différent responses are unclear and need to be understood in order to
build better prédictive models of dairy cow metabolism.
The effect of energy on WB protein metabolism is even more unclear. Compared
with the effect of additional protein supply, supplementation of pigs with non-protein
energy sources increased WB protein synthesis to a lesser extent but reduced leucine
catabolism (Reeds et al. 1981). Similarly, in humans, energy also had a smaller effect on
WB protein turnover than did protein supply (Garlick et al. 1991) while in sheep,
Abdul-Razzaq and Bickerstaffe (1989) observed a reduced protein turnover with
infusion of propionic acid, linked with decreased protein breakdown. Thèse data in
other species show that although energy supply can potentially impact protein
metabolism, the exact mechanisms may vary with expérimental conditions.
116
Although no data are available on the effect of energy supply on protein kinetics
in dairy cows, milk represents an important export protein and is probably associated
with increased rates of protein synthesis, at least for the mammary gland. Increasing
energy supply either through infusions of propionic acid into the rumen or glucose into
the duodénum can elevate milk protein output (Hurtaud et al. 1998b). Such responses
must involve more efficient use of dietary AA supply, i.e. decreased oxidation. In
ruminants, the major glucogenic source is propionate (McBride et al. 1998). Besides
acting as a precursor for glucose, in vitro studies demonstrated that this nutrient can be a
potential inhibitor of the other glucogenic pathways (Demigné et al. 1991) leading to
sparing effect on AA for a potential increase in milk protein output. This mechanism,
that should be linked to decreased ureagenesis, has not been convincingly demonstrated
in vivo in sheep (Kim et al. 1999).
In terms of interactions, none were observed between glucose and protein supply
on protein milk yield (Clark et al. 1977; Vanhatalo et al. 2003), and similarly a
combination of glucose and histidine showed additive responses (Huhtanen et al. 2002).
Surprisingly, the interaction between the main glucose precursor, propionate, and
protein has not been tested.
In an attempt to résolve some of the current uncertainties about the separate and
interactive effects of protein and propionate on metabolism in the lactating cow, the
current study was designed to examine the impact of casein and propionic acid supply
on milk production, and on mammary and WB kinetics of leucine and glucose. This
manuscript describes the findings that relate to WB leucine metabolism.
5.4 Material and Methods
5.4.1 Animais
Three multiparous Holstein cows, averaging 615 ± 24 kg BW and 65 ± 4 days in
milk at the beginning of the study, had been fitted, before calving, with both a proximal
T-shaped duodenal cannula, placed 10 to 15 cm from the pylorus, and a ruminai cannula
(Hurtaud et al. 1998a). One month before the beginning of the experiment, the cows
were further prepared with two permanent cathéters (Polyurethane cathéter tubing, i.d.
117
1.0 mm, o.d. 1.7 mm, UNO, Roestvaststaal BV, Holland), one inserted in the right
carotid artery and the other in the right subcutaneous abdominal vein as described
previously (Guinard et al. 1994). Thèse two permanent cathéters were protected in a
silicone rubber tubing (Silclear TM grade médical silicone tubing, i.d. 1.57, o.d. 3.18
mm, VWR International SAS, Briare, France). One ring of Dacron (Mersutures, TS53,
Ethicon, France) was placed around the cathéters at the point of extériorisation to
prevent infection. During the experiment, problems were encountered with one arterial
cathéter and a replacement was inserted in the left carotid artery. For the tracer
infusions, a silicone rubber cathéter (length, 21 cm; i.d., 1.02 mm; o.d., 2.16 mm;
Silclear tubing, Degania, Israël) was inserted in the left jugular vein before the
beginning of the experiment. Cathéter maintenance was performed as described
previously (Guinard et al. 1994). The expérimental procédures were reviewed and
approved by the Animal Care Committee of the French Ministry of Agriculture.
5.4.2 Feeding and Treatments
The diets were formulated with the INRA model (Institut National de la
Recherche Agronomique 1989) The same diet, 55 % grass silage and 45 % concentrate
(on DM basis: 54 % peas; 30 % dried sugar beet pulp; 1 % cane molasses; 4 % soybean
oil; 4 % NaCO3; 3 % CaCO3; 3 % CaHPO4, 1 % minerai and vitamin premix) was fed
to ail the cows (Table 5.1). The basai diet supplied 124 MJ là NEL and 1593 g/d of
protein digested in the intestine (PDI: INRA 1989) to the cows on the control treatment,
équivalent to 97 % of energy and protein requirements. When estimated using NRC
(2001), the supply on the basai diet was 119 MJ/d NEL and 1624 g/d metabolizable
protein. This diet, the same as that used by Rigout et al. (2003), was estimated to supply
little intestinal glucose. The quantity of wet feed offered was adjusted every day to
insure the same delivery of DM on each expérimental day, after détermination of DM
content of the grass silage.
Superimposed on the basai diet, four treatments were tested in a factorial
arrangement: duodenal infusion of calcium caseinate (743 + 7 g/d: 687 g/d PDI) and
estimated to provide 7.9 MJ/dof NEL (Guinard et al. 1994) and/or ruminai infusions of
propionic acid (1041 ±8 g/d), estimated to provide 15.6 MJ/d of NEL, as used in Rigout
118
et al. (2003). Therefore, treatments were: 1) control (Ctrl), 2) casein (Cas) 3) propionic
acid (C3) and 4) the casein and propionic acid combined (Cas+C3). The infusion dose
of casein was the highest dose used by Guinard et al. (1994), but within the range of
protein supply for a dairy ration. Propionate was chosen as the energy source as it is the
primary glucose precursor in dairy cow diets and should hâve the strongest effect on
sparing AA for gluconeogenesis. It is also less metabolically characterized than glucose
in dairy cows. Propionate was infused at the dose that yielded the maximal effect on
milk protein observed by Rigout et al. (2003). The original design was a 4 x 4 Latin
square but the loss of one cow led to a re-design of the experiment as two incomplète 4
x 3 Youden squares with 3 periods each for a total of 18 observations distributed as
follows: Ctrl: n = 5; Cas: n = 4; C3: n = 5; Cas+C3: n = 4. One cow had problems with
the arterial cathéter during the third period (Cas treatment), therefore, n = 3 for Cas for
the blood measurements. Each treatment period lasted 14 d, and the fïrst two days of
each period included transition for the infusions (the cows received on day 1, 0.33, and
on day 2, 0.66 of the total subséquent infusion).
The casein infusate (Armor Proteins, St Brice en Cobles, France) was prepared
daily at 30°C in 15 kg of tap water. The dose of C3 (Langlois S.A., Saint Jacques de la
Lande, France) was dissolved daily in 40 kg of tap water. A buffer solution (540 ± 4 g
of NaHCO3 plus 280 ± 2 g of KHCO3) was prepared daily in 10 kg of tap water. This
buffer was used to limit ruminai pH decrease with C3 to avoid acidosis, and the buffer
was infused in ail cows to maintain constant anion-cation balance between treatments.
Urea (average daily supply of 99 g) was added to this buffer solution as a dietary
supplément. Ail the cows in ail treatments received a similar volume in the rumen (total
of 49.5 ± 0.6 kg including 9.95 ± 0.07 kg of buffer and 39.5 ± 0.7 kg of water or
propionic acid solution) and in the duodénum (14.8 ± 0.14 kg of water or casein
solution). Thèse solutions were infused continuously by means of peristaltic pumps.
The concentrate was supplied every 3h in equal portions from automatic
feeders, starting at 0715h. Grass silage was fed three times per day: 25 % at 0715h, 25
% at 1315h, 50 % between 1715h and 1915h, except during the kinetic days (days 11
and 13) when the silage was fed four time per day: 12.5 % at 0715h, 1015h, 1315h and
119
1615h and then 50 % at 1915h. Throughout the whole study, access to the diet was
limited to lh after the concentrate distribution times. Cows had free access to water and
were housed in individual tie stalls. Every day before 0715h, any feed refusais were
weighed.
5.4.3 Sampling and Laboratory Analyses
Cows were milked twice daily (0630 and 1830 h) with yield recorded at each
milking and each sample was assayed for fat and true protein compositions by infrared
analysis (Milkoscan, Foss Electric, Hillerad, Denmark). During the second week of the
expérimental period, cows were milked from each half udder separately and milk
samples (each half udder) were assayed for fat and protein composition by infrared
analysis. Samples of the grass silage and concentrate mixture were taken every day. The
DM of the grass silage was determined on daily samples, while the DM of the
concentrate was determined every week on a pooled sample by oven drying at 80°C for
48 h. Feed samples for other analyses were stored at -20°C until analysis, where they
were pooled by period.
On day 11, L[1-13C] leucine (Cambridge Isotope Laboratories, Andover, USA.
99 ape) was infused at 4.3 mmol/h (14.20 mg/ml dissolved in stérile 0.9 % NaCl; 0.66 g
solution/min) for 7.5 h (starting 3.5h after the morning milking), preceded by a priming
dose of 4.3 mmol, équivalent to an hour of infusion (Wolfe, 1984). On day 13, a
mixture (in stérile 0.9 % NaCl) of [1-13C] bicarbonate (Cambridge Isotope Laboratories,
Andover, USA. 99 ape; 4.7 mg/ml) and [6.6-2H2] glucose (Cambridge Isotope
Laboratories, Andover, USA. 99 ape; 60 mg/ml) was infused. Within this mixture the
rate of [1-13C] bicarbonate infusion was 4.2 mmol/h (1.26 g solution/min) for 4 h
between 1200 to 1600 h, preceded by a priming dose équivalent to 3.2 mmol. Glucose
kinetics will be described in a separate paper.
On both days, cows were made to stand for at least 15 min before blood
sampling. On day 11, hourly samples were taken from the carotid artery starting 3 h
after the initiation of the infusion at lOOOh (1300, 1400, 1500, 1600 and 1700h) and
were analyzed for isotopic enrichment (IE) of leucine, 4-methyl 2-oxopentanoate
(MOP) and of CO2. On the same day, samples for AA analysis were taken at 0700,
120
0900, 1100, 1300, 1400, 1500, 1600 and 1700h and for urea at 0700, 0900, 1100, 1300,
1500 and 1700h. On day 13, samples were collected starting 1.45h after the initiation of
the infusion at 1200 (1345, 1430, 1515 and 1600h) and were analyzed to détermine the
IE of CO2. On both days 11 and 13, samples were also collected prior to the initiation of
stable isotope infusion to détermine natural abundance of leucine, MOP (2 samples:
0700 and 0900h) and CO2 (3 samples: 0800, 0900 and lOOOh).
Blood samples (7.5 ml) for AA concentrations were collected in heparinized
syringes and kept on ice. Plasma was prepared by centrifugation (2000 x g at 4°C for 15
min), 0.7g plasma was weighed, then 0.4 g of the internai standard solution was added
and the mixture was stored at -80°C. The internai standard solution was prepared at the
following concentrations: 501 nmol/g [5-I5N] Gin, 252 nmol/g [indole-15N] Trp, 260
nmol/g [1-13C] Cys, 8.8 umol/g [15N2] Urea, plus a U-13C Algal hydrolysate. This
follows the methodology for détermination of AA concentration by isotope dilution as
previously described (Calder et al. 1999). Plasma samples for urea were pooled per cow
and period and stored at -20°C until analysis on a multiparameter analyzer (KONE
Instruments Corporation, Espoo, Finland) using an enzymatic kit (KONE, urea UV
kinetic kit n°l 1703, KONE diagnostics, KONA Instruments S.A., Evry, France)
For MOP and [l-13C]leucine IE (expressed as molar percent excess, mpe), 0.7 g
plasma was weighed and 0.3g of an oxohexanoic acid solution (33.3 nmol/g) added, and
the mixture was stored at -80°C until analysis. Déterminations of MOP and leucine IE
were as described by Calder and Smith (1988). For CO2 IE, triplicate 1 ml of blood
samples were injected into evacuated vacutainers containing 1 ml of lactic acid,
immediately mixed and kept at room température until analysis was performed.
Détermination of IE (mpe) was as previously described (Lobley et al. 2003) based on
Read et al. (1984) and Reeds and Hutchens (1994).
5.4.4 Calculations
In ail équations, WB irréversible lost rate (ILR) and infusion rates are expressed
in mmol/h and IE in mpe. Tracer refers to 13C-leucine or 13C-bicarbonate and tracée is
the unlabelled leucine and bicarbonate.
121
Whole body leucine ILR or carbon dioxide entry rate (CER) were calculated as
follows:
WB-ILRtraCee or CERtracee = (IEinf / IEa - 1) x Inf
where IEjnf and IEa are the IE of the infusate and the mean enrichment of arterial
plasma MOP, leucine or 13CO2, respectively and Inf is the infusion rate of the 13C-
Leucine or 13C-bicarbonate.
Leucine tracer oxidation (mmol/h) was calculated as:
Leutracer oxidation = CER x IEa 13CC>2-Leu
Where CER is the CO2 entry rate measured during bicarbonate infusion
(described above) and IE 13CO2-Leu is the mean arterial enrichment of 13CÛ2 during
leucine infusion. For this calculation, it is assumed that the CER production during the
bicarbonate infusion day is the same as during the day of leucine infusion.
The fractional oxidation (FO) was calculated as:
FO = Leutracer oxidation / (Inf 13C-Leu x IEjnf)
For tracer kinetics, it is assumed that the animal does not distinguish between13C- and 12C-AA and it is further assumed that ail leucine tracer infused is in excess and
that a corresponding amount of leucine will be oxidized by the animal (Lobley et al.
2003). Therefore to calculate the WB leucine oxidation (WBLO)
WB-LO = FO x (ILRtracee + Inf ) - Inf
Leucine used for WB protein synthesis was calculated as the différence between
WB leucine ILR and leucine oxidation. The conversion into protein (g/d) was based on
63 g of leucine/kg of synthesized tissue protein (Lobley et al. 1980) and 98 g of
leucine/kg of CP in milk (Swaisgood 1995).
122
5.4.5 Statistical Analyses
Analyses of variance were made using the MIXED procédure of SAS (SAS
Institute Inc., 1999) according to the expérimental plan using the following statistical
model:
yijki =[i + SQUAREj + PERIODj (SQUAREj) + COWk + TREAT, + eijkl
Différences among treatments were compared using orthogonal contrasts
according to the factorial arrangement. Results are expressed as least square means with
the highest standard error of mean (SEM).
5.5 Results
5.5.1 Milk Production
Dry matter intake, milk production and composition reported are the average of
the second week of expérimental periods (Table 5.2). The DMI was not affected by
treatments. There was no interaction between Cas and C3 on milk parameters. Milk
yield increased with Cas treatments (P < 0.01). Milk fat concentration (P - 0.04) and
yield (P = 0.02) were reduced with C3 infusions, while fat yield increased with Cas
infusions (P = 0.001). True protein concentration increased with Cas (P < 0.01) and C3
(P = 0.01), as did true protein yield with Cas (P < 0.001) and C3 infusions (P = 0.02).
Similar results for true protein concentration (data not shown) and yield (see leucine in
milk, Table 5.3) were obtained on the day of WB leucine kinetics.
5.5.2 Leucine Kinetics
The WB ILR (Table 5.3), using arterial MOP IE as représentative of the
precursor pool, increased with Cas treatments (P < 0.001) and tended to increase with
C3 (P = 0.06). With arterial leucine IE selected as precursor pool to estimate WBILR,
only Cas treatments remained signifïcant (P < 0.001). Absolute ILR values were, on
average, 25 % lower with arterial leucine than MOP as precursor, due to the higher IE
of leucine. There was a tendency for an interaction (P = 0.09) between Cas and C3 on
leucine oxidation, with a slight increase when C3 was added to the Ctrl diet, but a
123
decrease by 14 % when C3 was added to the Cas infusion. There was a similar trend (P
= 0.07) for fractional oxidation.
Leucine used for WB protein synthesis increased with Cas treatments (P < 0.01),
both for milk protein (P < 0.001) and non-milk protein synthesis (tissues; P = 0.02).
Leucine used for total protein synthesis tended (P = 0.08) to increase with C3
treatments, but the effect was due to changes in leucine for milk protein (P = 0.06).
Whole body protein synthesis followed the same trend as leucine used for protein
synthesis and averaged 4.1, 4.6, 4.2 and 5.1 kg/d (SEM; 0.16) for Ctrl, Cas, C3 and Cas
+ C3, respectively.
Data for blood bicarbonate IE during the [1-13C] leucine and [NaH13CO3]
infusion and the CER are presented in Table 5.4. The blood 13CC>2 IE during the
NaH13CO3 infusion was reduced with both Cas (P = 0.02) and C3 (P = 0.02)
supplementation. In conséquence, the CER increased with Cas (P < 0.01) and C3 (P <
0.01) treatments. In contrast, the 13CC>2 enrichment measured during [1-13C] leucine
administration increased with Cas infusions (P < 0.01) with a tendency (Cas x C3
interaction, P = 0.08) for a smaller increase when Cas was added to C3 infusion.
5.5.3 Amino Acid and Urea Concentrations
Arterial AA concentrations are presented in Table 5.5. With Cas infusions
arginine, cysteine, histidine, isoleucine, leucine, lysine, methionine, phenylalanine,
proline, tryptophan, tyrosine and valine increased (P < 0.05), while glutamate and
glycine decreased (P < 0.01). During C3 infusion, glutamine and serine (P < 0.01) and
threonine, tryptophan and tyrosine (P < 0.04) increased but isoleucine, leucine and
valine (P < 0.01) decreased, with a similar tendency (P = 0.09) for lysine. There was
also a trend (P < 0.11) for an interaction between Cas and C3 on the concentrations of
the branched-chain AA with a greater increase when Cas was added to the Ctrl diet
compared with Cas+C3 responses. Arterial concentrations of urea increased (P < 0.001)
with Cas infusions but decreased (P = 0.03) with C3 (Table 5.5).
124
5.6 Discussion
In the présent experiment, C3 and Cas infusions had différent impacts on milk
yield and fat yield but both nutrients increased protein concentration and protein yield,
albeit to différent extents. The kinetic data obtained allow some of the WB interactions
on protein turnover to be considered in the light of thèse various modifications of milk
yield and composition.
5.6.1 Milk Production and Composition
The incrément in milk production with Cas treatments (18 %) is at the upper end
of numerous previous reports (Clark et al. 1977; Rulquin 1982; Guinard et al. 1994;
Choung and Chamberlain 1995; Vanhatalo 2003). As a resuit of the improved milk
yield at similar fat concentrations, milk fat yield increased with Cas infusions,
confirming earlier reports (Guinard et al. 1994; Vanhatalo et al. 2003).
Milk protein yield increased as previously observed (e.g. Guinard et al. 1994)
with Cas infusion, with approximately 25 % of the infused casein recovered as milk
protein. The observed incréments in milk and protein yields indicated that, as designed,
the control diet was not optimal to support maximum milk yield. In the présent
experiment Cas and Cas + C3 treatments provided 121 % and 112.5 % of protein
requirements, compared with 97 % for the control diet (INRA, 1989). The marginal
casein recovery (25 %) is in the range of the 30 % reported by INRA (1989) for such a
protein supply. It is also within the range from previous studies (21 % ± 15; see
Hanigan et al. 1998) with casein infusion and similar to that observed for incrément in
metabolisable protein supply from 1922 to 2517 g/d (19 %; Raggio et al. 2004).
Différences in the actual recoveries between studies probably relate to how close the
basai rations were to the requirements of the animal.
Less energy was transferred into milk lactose and fat from C3 infusions
compared with Cas treatments because C3 infusions did not change milk yield and
decreased both milk fat concentration and yield. The milk fat concentration response
was similar to other experiments when the cows received C3 infusions into the rumen,
as summarized by Rigout et al. (2003). This earlier study also reported a modest
125
improvement (0.7 kg/d) in milk yield, with dairy cows receiving 115 % of
recommended protein allowances. A similar numerical increase was observed in the
current experiment when C3 was added to Cas. This occurred despite the fact that, in
the current study the increased on milk yield was seen when the C3 was added to the
Ctrl diet, while in Rigout et al. (2003) the response occurred when C3 was substituted to
a similar amount of energy infused as volatile fatty acids for the control cows.
Protein in milk also increased (6 %) with C3 infusions. This incrément is smaller
than with Cas treatments (20 %) but is in accordance with previous experiments (see
Rigout et al. 2003). In the présent study, there were modest increases in milk protein
yield with C3 infusions in cows receiving protein supply below (97 %) or above (113
%) requirements. This suggests that part of the C3 effect on milk protein yield links to a
change in WB protein metabolism rather than major effects on intestinal AA
availability. This effect may be due to an increase of mammary blood flow favoring AA
uptake as already seen with duodenal glucose infusion (Rulquin et al. 2004). It may also
be due to less AA used as glucose precursors with, in conséquence, more available for
milk protein production.
5.6.2 Whole body leucine kinetics
The WB ILR increased with both Cas and C3 infusions. Whole body ILR is
known to be sensitive to the inflow of nutrients such as différent MP supply in dairy
cows (Lapierre et al. 2002) and at différent intake levels in both growing beef steers
(Lobley et al. 1987; Lapierre et al. 1999; Wessels et al. 1997) and sheep (Liu et al.
1995; Savary-Auzeloux et al. 2003).
In ternis of overall protein dynamics what do thèse changes in WB ILR mean?
For essential AA, such as leucine, ILR is equal to protein synthesis plus oxidation and
also to protein digested plus protein breakdown. So changes in WB ILR must reflect
altérations in at least two of thèse processes. In the current study, the increase in WB
ILR with casein infusion exceeded the additional leucine supply, similar to previous
reports in dairy cows (Lapierre et al. 2002) and growing pigs (Reeds et al. 1980),
indicative of a stimulation in tissue protein metabolism. Thus, despite increased leucine
oxidation, there was an increase in protein synthesis, linked to the improved milk
126
protein yield. Similarly, increases in both WB protein synthesis and oxidation in
response to casein supply hâve been observed in goats (Bequette et al. 1996a) and sheep
(Liu et al. 1995) with the response hypothesized as mediated via hormonal actions
involving insulin and insulin like growth factor-1 (Bequette et al. 2002). However, in a
similar dairy cow study, with the same rate of infusion (762 g/d) of casein, there was no
change in arterial plasma insulin concentrations (Guinard et al. 1994).
The impacts of C3 were weaker, with tendencies to increase WB leucine ILR
and decrease oxidation when infused with Cas. The pattern of urea concentrations
followed that of leucine oxidation, i.e. a large incrément from casein infusion, while C3
was effective in reducing oxidation only when the animais were concomitantly infused
with Cas. To our knowledge, there are no reports on the impact of C3 on WB IRL of
AA in lactating ruminants. However, in sheep, Abdul-Razzaq et al. (1989) observed a
reduced leucine oxidation with C3 infusion, but only in comparison with acétate
supplementation. In the current study, there was a tendency for WB protein synthesis to
be increased by C3, with this trend driven by the significant improvement in milk
protein output. Three mechanisms may contribute to this tendency. First, arterial
concentrations of branched-chain AA were reduced by C3 infusions, as already
observed with C3 and glucose infusions (Clark et al. 1978; Lemosquet et al. 2004). This
réduction in the branched-chain AA is often linked to an increase in insulin (Griinari et
al. 1997; Gross et al. 1990; Mackle et al. 2000) which could allow diversion of more
substrate for milk protein synthesis. Second, C3 effects on milk protein may also be due
to less AA used for glucose synthesis with, in conséquence, more available for milk
protein production. During the C3 infusions, plasma concentration of some glucogenic
AA and two essential AA (threonine and tryptophan) increased as previously observed
(Lemosquet et al. 2004). Threonine is a glucogenic AA (via serine) and may be spared
if other sources of glucose synthesis are available. Substantial quantities of tryptophan
are removed across the gut (Lobley et al. 2003), as is propionate to fuel the energy
needs of this tissue (Seal and Parker 1994). In steers, intra-ruminal infusion of
propionate (1 mol/d) led to increased tryptophan absorption and higher arterial
concentration (Seal and Parker 1996). A third possibility is increased uptake of essential
AA as observed with duodenal glucose infusion in dairy cows (Rulquin et al. 2004), a
127
process driven mainly through increased mammary plasma flow (Bequette et al. 2001;
Vanhatalo et al. 2003; Rulquin et al. 2004).
Overall, although both energy and protein independently altered WB leucine
ILR and milk protein output, there were subtle différences in the mechanisms involved.
Casein exerted a much stronger impact on WB protein synthesis, with 30-45 % of the
incrément directed towards extra milk protein. As total mammary gland protein
synthesis has been reported to average 1.3 times milk protein output (Bequette et al.
1996; Thivierge et al. 2002), not only was mammary gland protein synthesis increased
by casein supply but other tissues must also hâve responded. In contrast, C3 had a
smaller effect on protein synthesis and reduced oxidation of leucine only when added to
Cas and thus helped increased milk protein yield further. This may be through changes
in the hormone milieu, increased energy supply to support mammary gland metabolism
or from sparing of AA use for glucogenic purposes. Some of thèse issues will become
clearer from the companion measurements on glucose and mammary gland metabolism.
Effïciency of nutrient use can be considered from the leucine net balance (Figure
5.1). Any extra AA infused has three main net fates; oxidation, conversion into milk
protein and rétention in the body (mainly as tissue gain). The infusion of casein
increased leucine absorption by 20.9 mmol/h, with part recovered as extra milk protein
(5.1 mmol/h with Cas and 5.3 mmol/h with Cas + C3 infusions). Most of the remainder
was oxidized, 13.9 and 10.4 mmol/h for Cas and Cas + C3 respectively. The slight
increased transfer into milk with Cas + C3 was more than compensated by less
oxidation and, in conséquence, net balance increased from 1.9 to 5.1 mmol/h (obtained
by différence). The actual values, but not the partition, dépend on the assumption that
plasma MOP truly reflects the precursor for both oxidation and protein synthesis.
5.6.3 Carbon Balance
Casein and C3 had différent impacts on carbon effïciency (i.e. conversion of
infused C to milk C). Less of the C3 infused was converted into milk products as milk
yield did not increase and milk fat decreased. Thus, the carbon balance showed that only
5 % (0.08 mol C/h) of infused (1.74 mol C/h) C3 appeared in milk while 77 % (1.4 mol
C/h) appeared as extra CO2 production. The remaining carbon, 18 % (0.31 mol C/ h),
128
may be incorporated into body tissues other than mammary epithelial cells (e.g. liver
fat) or may represent measurement error. In contrast, for casein infusion alone (0.97 mol
C/h), 25 % (0.24 mol C/h) of the carbon intake appeared in milk with 73 % (0.70 mol
C/h) oxidized to CO2, the remaining 2 % may represent measurement error. The
incrémental ratio milk carbon / intake carbon was 0.12 for Cas + C3, therefore a better
C effïciency was obtained from Cas supply alone. Across ail treatments, most of the
infused extra carbon was recovered (90 %) either as CO2 or milk products.
5.7 Conclusion
While both Cas and C3 independently increased WB leucine ILR, protein synthesis and
leucine exported in milk, the magnitude of the responses differed, being greater with
Cas. Furthermore, Cas treatments increased oxidation while C3 decreased leucine
oxidation when infused in the présence of casein. Cas increased the arterial
concentrations of ail the essential AA, whereas C3 elevated concentrations of
glucogenic AA. Together, thèse data suggest that C3 and Cas improve milk protein
output by différent mechanisms, with an additional energy supply as C3 being also
effective when sufficient protein is supplied.
5.8 Acknowledgments
The authors gratefully thank from the UMR PL, INRA, France, Y. Lebreton for
surgeries, P. Lamberton and his team for technical support and animal care during the
experiment and C. Nahuet, L. Finot, N. Huchet, I. Jicquel, and S. Rigault for their help
in laboratory analyses and from the Rowett, A. G. Calder and S. E. Anderson for mass
spectrometry analyses.The authors also wish to acknowledge the financial support of the
INRA, the European Union for the award of Marie Curie Training Site funding to the
Rowett Research Institute, SEERAD and the National Science and Engineering
Research Council of Canada and Agriculture and Agri-Food Canada (Lennoxville
Research Centre Contribution number 874).
129
Table 5. 1. Chemical composition of the feed ingrédients.
Analysisg/kg DMCPNDFADFLigninFatOrganic MatterNDF-NZ
ADF-Ny
NEL, MJ/ kg DM
Grass silage
129536285
1931
91612.74.16.7
Concentrate
156207118
1228
88119.07.37.1
z Neutral détergent fiber insoluble nitrogenyAcid détergent fiber insoluble nitrogen
130
Table 5.2. Effect of casein (Cas) and propionate (C3) supply on DMI, milk yield andcomposition
DMI, kg/dMilk, kg/d
Fat, g/kg
g/dTrue Protein, g/kg
g/d
Ctrl
19.26.45.
1212
29792
5
93
.6
Treatments
Cas
19.731.445.0
139330.7
958
C3
19.327.1
43.01155
30.8828
Cas + C3
19.632.1
40.61287
32.81043
SEMZ
0.170.711.4029.2
0.4322.0
Cas
0.17<0.01
0.36
<0.001
<0.01O.001
C3
0.560.510.04
0.02
<0.010.02
P y
Casx C3
0.810.740.450.41
0.330.27
zLeast square means presented with pooled SEM, given for n=4; based on the secondweek from the total mammary gland (18 observations; Ctrl = 5, Cas = 4, C3 = 5, Cas +C3 = 4)yProbability corresponding to the null hypothesis with Cas, C3 and Cas x C3 contrasts.
131
Table 5.3. Effect of casein (Cas) and propionate (C3) supply on whole body (WB)leucine kinetics
Leucine, mmol/hFrom the infusion"WB ILR MOPW
WB ILR LEUW
OxidationWB Protein synthesis, PSMilkv
WB non-milk PSU
Fractional oxidation1
Ctrl
95.571.2
7.287.923.364.90.077
TreatmentCas
20.9119.995.321.298.727.971.20.177
C3 Cas
100.168.3
8.891.324.767.00.087
+ C320.9
126.196.118.2
107.930.078.3
0.140
SEMZ
2.771.841.133.500.853.710.01
Cas
O.001<0.001O.001O.001O.001
0.020O.010
Py
C3
0.060.580.560.080.060.220.33
Cas x C3
0.790.310.090.440.610.440.07
zLeast square means presented with pooled SEM, given for n=3; 17 observations: Ctrl =5, Cas = 3, C3 = 5, Cas + C3 = 4.yProbability corresponding to the null hypothesis with Cas, C3 and Cas x C3 contrasts."Leucine from the infusion calculated as 748 g * casein purity * digestibility (purity as99 %, digestibility as 90 %).WILR: irréversible loss rate, calculated with the arterial enrichment of 4-methyl 2-oxopentanoate (MOP) or leucine (LEU) as représentative of the precursor pool,vLeucine in milk was based on milk production on the half-day of leucine infusion.uWB-non-milk PS = WB protein synthesis - Leucine in milk.' Fractional oxidation : Leucine Oxidation / WB ILR MOP
132
Table 5.4. Effect of casein (Cas) and propionate (C3) supply on 13CO2 isotopicenrichment (IE), carbon dioxide entry rate (CER) and carbon balance
Treatment Py
Ctrl Cas C3 Cas+C3 SEMZ Cas C3 CasxC313CO2 IE,during leucine inf, mpe 0.0034 0.0056 0.0034 0.0044 O . 0 K 0 . 0 0 1 0.10 0.0813CO2 IE,during bicarbonate inf, mpe 0.0270 0.0261 0.0260 0.0252 <0.01 0.02 0.02 0.31
CER, mol/h 14.50 15.30 15.50 16.10 0.19 <0.01 <0.01 0.61
Extra Carbons"
Intake Carbons, mmol/h
Extra carbons CER, mmol/h
Extra carbons in milk, mmol/h
% accounted
zLeast square means presented with pooled SEM, given for n=3; 17 observations; Ctrl =5, Cas = 3, C3 = 5, Cas+C3 = 4.yProbability corresponding to the null hypothesis with Cas, C3 and Cas x C3 contrasts."Extra Carbons was calculated as Cas inf = CasCER - CtrlCER or C3inf= C3CER -CtrlCER and Cas + C3inf = (CasCER - CtrlCER) + (C3CER - CtrlCER)
97070724198
174113429082
2710204933188
133
Table 5.5. Effect of casein (Cas) and propionate (C3) supply on plasma arterialconcentration of amino acids and urea
Amino Acids, |oMEssential
HistidineIsoleucineLeucineLysineMethioninePhenylalanineThreonineTryptophanValine
Non-essentialAlanineArginineAspartic acidCysteineGlutamineGlutamic acidGlycineProlineSerineTyrosine
Urea, mM
Ctrl
20.678.890.658.028.242.8
112.245.6
139.1
187.259.82.7
81.1248.1
62.9294.0
59.691.642.9
2.9
TreatmentCas
54.1131.3169.6106.334.851.5
111.549.6
247.6
198.484.5
3.0100.8236.9
54.9241.8147.192.362.2
4.4
C3
17.761.771.252.229.241.0
127.350.1
115.9
203.057.42.9
69.9286.9
61.7317.2
59.5112.248.7
2.8
Cas + C3
54.492.8
120.896.137.148.1
134.451.8
179.5
224.183.8
3.195.7
286.054.8
269.6138.2112.669.4
3.9
SEMZ
2.716.699.124.411.302.226.611.37
12.90
11.904.360.376.769.442.1915.906.576.572.50
0.11
Cas
O.001<0.001O.001O.001O.001<0.01
0.580.05
O.001
0.15O.001
0.41<0.01
0.47<0.01<0.01O.001
0.92O.001
O.001
P y
C3
0.66<0.01O.01
0.090.220.240.020.04
<0.01
0.100.680.690.23
<0.010.740.130.520.010.03
0.03
Cas x C3
0.510.100.110.570.580.690.510.340.08
0.630.800.840.610.550.770.860.490.980.75
0.15
zLeast square means presented with pooled SEM, given for n=3; 17 observations: Ctrl:
5, Cas = 3, C3 = 5, Cas+C3 = 4.yProbability corresponding to the null hypothesis with Cas, C3 and Cas x C3 contrasts;
134
Figure 5.1. Effect of extra leucine intake on leucine oxidation, leucine in milk andleucine in tissues
Extra Leucine Intake20.9 mmol/h
Leucine Tissues1.9 mmol/h Cas5.1 mmol/h Cas+ C3
Leucine Oxidation:13.9 mmol/h Cas10.4 mmol/h Cas + C3
Leucine Milk5.1 mmol/h Cas5.3 mmol/h Cas + C3
Extra leucine intake: leucine intake in Cas and Cas + C3 infusion.Leucine-milk: was calculated as leucine in milk in Cas - Ctrl and (Cas + C3) - C3infusion.Leucine-oxidation: was calculated as leucine oxidation in Cas - Ctrl and (Cas + C3) -C3 infusion.Leucine-Tissues: was calculated as (Leucine intake) - (Leucine milk + Leucineoxidation).
135
5.9 Références
Abdul-Razzaq, H. A. and Bickerstaffe, R. 1989. The influence of rumen volatile fatty
acids on protein metabolism in growing lambs. Br. J. Nutr. 62:297-310.
Bequette, B. J., Backwell, F. R., MacRae, J. C, Lobley, G. E., Crompton, L. A.,
Metcalf, J. A. and Sutton, J. D. 1996a. Effect of intravenous amino acid infusion on
leucine oxidation across the mammary gland of the lactating goat. J. Dairy Sci. 79:2217-
2224.
Bequette, B. J., Kyle, C. E., Crompton, L. A., Anderson, S. E. and Hanigan, M. D.
2002. Protein metabolism in lactating goats subjected to the insulin clamp. J. Dairy Sci.
85:1546-1555.
Bequette, B. J., Kyle, C. E., Crompton, L. A., Buchan, V. and Hanigan, M. D. 2001.
Insulin régulâtes milk production and mammary gland and hind-leg amino acid fluxes
and blood flow in lactating goats. J. Dairy Sci. 84:241-255.
Bequette, B. J., Metcalf, J. A., Wray-Cahen, D., Backwell, F. R.C., Sutton, J. D.,
Lomax, M. A., MacRae, J. C. and Lobley, G. E. 1996b. Leucine and protein metabolism
in the lactating dairy cow mammary gland: responses to supplemental dietary crade
protein intake. J. Dairy Res. 63:209-222.
Blouin, J. P., Bernier, J. F., Reynolds, C. K., Lobley, G. E., Dubreuil, P. and Lapierre,
H. 2002. Effect of supply of metabolizable protein on splanchnic fluxes of nutrients and
hormones in lactating dairy cows. J. Dairy Sci. 85:2618-2630.
Calder, A. G., Garden, K. E., Anderson, S. E. and Lobley, G. E. 1999. Quantitation of
blood and plasma amino acids using isotope dilution électron impact gas
chromatography/mass spectrometry with U-(13)C amino acids as internai standards.
Rapid Commun. Mass Spectrom. 13:2080-2083.
Calder, A. G. and Smith, A. 1988. Stable isotope ratio analysis of leucine and
ketoisocaproic acid in blood plasma by gas chromatography/mass spectrometry. Use of
tertiary butyldimethylsilyl derivatives. Rapid Commun. Mass Spectrom. 2:14-16.
136
Choung, J. J. and Chamberlain, D. G. 1995. Effects of abomasal infusions of sodium
caseinate, a hydrolysate of casein or a corresponding mixture of free amino acids on
milk yield and composition in dairy cows. Journal Dairy Res. 62:29-37.
Clark, J. H., Spires, H. R., Derrig, R. G. and Bennink, M. R. 1977. Milk production,
nitrogen utilization and glucose synthesis in lactating cows infused postruminally with
sodium caseinate and glucose. J. Nutr. 107:631-644.
Clark, R. M., Chandler, P. T. and Park, C. S. 1978. Limiting amino acids for milk
protein synthesis by bovine mammary cells in culture. J. Dairy Sci. 61:408-413.
Demigné, C, Yacoub, C. Morand, C. and Rémésy, C. 1991. Interaction between
propionate and amino acid metabolism in isolated sheep hepatocytes. Br. J. Nutr.
61:301-317.
De Peters, E. J. and Cant, J. 1992. Nutritional factors influencing the nitrogen
composition of bovine milk: a review. J. Dairy Sci. 75: 2043-2070.
Garlick, P. J., McNurlan, M. A. and Ballmer, P. E. 1991. Influence of dietary protein
intake on whole-body protein turnover in humans. Diabètes Care 14:1189-1198.
Griinari, J. M., McGuire, M. A., Dwyer, D. A., Bauman, D. E., Barbano, D. M. and
House, W. A. 1997. The rôle of insulin in the régulation of milk protein synthesis in
dairy cows. J. Dairy Sci. 80:2361-2371.
Gross, K. L., Harmon, D. L., Minton, J. E. and Avery, T. B. 1990. Effects of
isoenergetic infusions of propionate and glucose on portal-drained viscéral nutrient flux
and concentrations of hormones in lambs maintained by total intragastric infusion. J.
Anim Sci. 68:2566-2574.
Guinard, J., Rulquin, H. and Vente, R. 1994. Effect of graded levels of duodenal
infusions of casein on mammary uptake in lactating cows. J. Dairy Sci. 77:2221-2231.
Hanigan, M. D., Cant, J. P., Weakley, D. C. and Beckett, J. L. 1998. An évaluation of
postabsorptive protein and amino acid metabolism in the lactating dairy cow. J. Dairy
Sci. 81:3385-3401.
137
Huhtanen, P., Vanhatalo, A. and Varvikko, T. 2002. Effects of abomasal infusions of
histidine, glucose, and leucine on milk production and plasma metabolites of dairy cows
fed grass silage diets. J. Dairy Sci. 85:204-216.
Hurtaud, C, Rulquin, H. and Vérité, R. 1998a. Effects of graded duodenal infusions of
glucose on yield and composition of milk from dairy cows. 1. Diets based on corn
silage. J. Dairy Sci. 81:3239-3247.
Hurtaud, C, Rulquin, H. and Vente, R. 1998b. Effects of level and type of energy
source (volatile fatty acids or glucose) on milk yield, composition and coagulating
properties in dairy cows. Reprod. Nutr. Dev. 38:315-330.
Institut National de la Recherche Agronomique. 1989. Ruminant Nutrition:
recommended allowances and feed tables. R. Jarrige Ed. J. Libbey Eurotext, Paris,
France.
Kim, H. S., Choung, J J. Chamberlain, D. and Lobley, G. E. 1999. Effect of propionate
on ovine urea kinetics. In Ed. G. E. Lobley, A. White and J. C. MacRae Proc.VIIIth
International Symposium on Protein Metabolism and Nutrition. Aberdeen, UK.. EAAP
Publication No. 96 Wageningen Press, p 57.
Lapierre, H., Bernier, J. F., Dubreuil, P., Reynolds, C. K., Farmer, C, Ouellet, D. R. and
Lobley, G. E. 1999. The effect of intake on protein metabolism across splanchnic
tissues in growing beef steers. Br. J. Nutr. 81:457-466.
Lapierre, H., Blouin, J. P., Bernier, J. F., Reynolds, C. K., Dubreuil, P. and Lobley, G.
E. 2002. Effect of supply of metabolizable protein on whole body and splanchnic
leucine metabolism in lactating dairy cows. J. Dairy Sci. 85:2631-2641.
Lemosquet, S., Rigout, S., Bach, A., Rulquin, H. and Blum, J. W. 2004. Glucose
metabolism in lactating cows in response to isoenergetic infusions of propionic acid or
duodenal glucose. J. Dairy Sci. 87:1767-1777.
Liu, S. M., Lobley, G. E., Macleod, N. A., Kyle, D. J., Chen, X. B. and Orskov, E. R.
1995. Effects of long-term protein excess or deficiency on whole-body protein turnover
in sheep nourished by intragastric infusion of nutrients. Br. J. Nutr. 73:829-839.
138
Lobley, G. E., Connell, A. and Buchan, V. 1987. Effect of food intake on protein and
energy metabolism in fïnishing beef steers. Br. J. Nutr. 57:457-465.
Lobley, G. E., Milne, V., Lovie, J. M., Reeds, P. J. and Pennie, K. 1980. Whole body
and tissue protein synthesis in cattle. Br. J. Nutr. 43:491-502.
Lobley, G. E., Shen, X., Le, G., Bremner, D. M., Milne, E., Calder, A. G., Anderson, S.
E. and Dennison, N. 2003. Oxidation of essential amino acids by the ovine
gastrointestinal tract. Br. J. Nutr. 89:617-630.
Mackle, T. R., Dwyer, D. A., Ingvartsen, K. L., Chouinard, P. Y., Ross, D. A. and
Bauman, D. E. 2000. Effects of insulin and postruminai supply of protein on use of
amino acids by the mammary gland for milk protein synthesis. J. Dairy Sci. 83:93-105.
McBride, B. W., Berthiaume, R. and Lapierre, H. 1998. Nutrient flow in the lactating
cow. C. J. Ani. Sci. 78: 91-104.
National Research Council. 2001. Nutrient Requirements of Dairy Cattle. 7th rev. éd.
Natl. Acad. Sci, Washington, DC.
Oldham, J. D., Lobley, G. E., Konig, B. A., Parker, S. D. and Smith, R. W. 1980.
Amino acid metabolism in lactating dairy cows early in lactation. Proceeding of the 3rd
EAAP Symposium on Protein Metabolism and Nutrition 2.
Raggio, G., Pacheco, D., Berthiaume, R., Lobley, G. E., Pellerin, D., Allard, G.,
Dubreuil, P. and Lapierre, H. 2004. Effect of level of metabolizable protein on
splanchnic flux of amino acids in lactating dairy cows. J. Dairy Sci. 87:3461-3472.
Read, W. W., Read, M. A. Rennie, M. J. Griggs, R. C. and Halliday, D. 1984.
Préparation of CO2 from blood and protein-bound amino acid carboxyl groups for
quantification and 13C-isotope measurements. Biomed. Mass Spectrom. 11:348-352.
Reeds, P. J., Cadenhead, A., Fuller, M. F., Lobley, G. E. and McDonald, J. D. 1980.
Protein turnover in growing pigs. Effects of âge and food intake. Br. J. Nutr. 43:445-
455.
139
Reeds, P. J., Fuller, M. F., Cadenhead, A., Lobley, G. E. and McDonald, J. D. 1981.
Effects of changes in the intakes of protein and non-protein energy on whole-body
protein turnover in growing pigs. Br. J. Nutr. 45:539-546.
Reeds, P. J., and Hutchens, T. W. 1994. Protein requirements: from nitrogen balance to
functional impact. J Nutr 124:1754S-1764S.
Rigout, S., Hurtaud, C, Lemosquet, S., Bach, A. and Rulquin, H. 2003. Lactational
effect of propionic acid and duodenal glucose in cows. J. Dairy Sci. 86:243-253.
Rulquin, H. 1982. Effects on digestion and metabolism in the dairy cow of infusions of
volatile fatty acids into the rumen and caseinate into the duodénum. I. Production and
digestion. Reprod. Nutr. Dev. 22:905-921.
Rulquin, H., Rigout, S., Lemosquet, S. and Bach, A. 2004. Infusion of glucose directs
circulating amino acids to the mammary gland in well-fed dairy cows. J. Dairy Sci.
87:340-349.
SAS® Institute Inc. User's Guide, Version 8. Edition, SAS Inst., Inc. Cary, NC, 1999.
Savary-Auzeloux, I., Hoskin, S. O. and Lobley, G. E. 2003. Effect of intake on whole
body plasma amino acid kinetics in sheep. Reprod. Nutr. Dev. 43:117-129.
Seal, C. J., and D. S. Parker. 1994. Effect of intraruminal propionic acid infusion on
metabolism of mesenteric- and portal-drained viscera in growing steers fed a forage
diet: I. Volatile fatty acids, glucose and lactate. J. Anim. Sci. 72:1325-1334.
Seal, C. J., and D. S. Parker. 1996. Effect of intraruminal propionic acid infusion on
metabolism of mesenteric- and portal-drained viscera in growing steers fed a forage
diet: H. Ammonia, urea, amino acids and peptides. J. Anim. Sci. 74:245-256.
140
Swaisgood, H. E. 1995. Protein and amino acid composition of bovine milk. Handbook
of Milk Composition. R. G. Jensen, éd. Académie Press, Toronto, ON. Pages; 464-468.
Thivierge, M. C, Bernier, J. F., Dubreuil, P. and Lapierre, H. 2002. The effect of
jugular or abomasal infusion of amino acids on milk yield in lactating cows fed a
protein déficient diet. Reprod. Nutr. Dev. 42:1-13.
Vanhatalo, A., Varvikko, T. and Huhtanen, P. 2003. Effects of casein and glucose on
responses of cows fed diets based on restnctively fermented grass silage. J. Dairy Sci.
86:3260-3270.
Wessels, R. H., Titgemeyer, E. C. and St Jean, G. 1997. Effect of amino acid
supplementation on whole-body protein turnover in Holstein steers. J. Anim. Sci.
75:3066-3073.
Wolfe, R. R. 1984 Tracers in metabolic research. Radioisotope and stable isotope/mass
spectrometry methods. Alan R. Liss, Inc., New York.
141
CHAPITRE VI
EFFECT OF CASEIN AND PROPIONATE SUPPLY ON MAMMARY
PROTEIN METABOLISM IN LACTATING DAIRY COWS
Ce Chapitre a été accepté pour publication le 07/04/06 dans la revue "Journal of Dairy
Science" sous la référence de G. Raggio, ' S. Lemosquet, G. E. Lobley, H. Rulquin, and
H. Lapierre. 2006. Effect of casein and propionate supply on mammary protein
metabolism in lactating dairy cows. J. Dairy. Sci.
142
6.1 Résumé
Les effets de la caséine (Cas) et du propionate (C3) sur le métabolisme
mammaire des acides aminés (AA) ont été étudiés sur trois vaches Holstein multipares,
possédant des canules ruminale et duodénale, selon un double carré de Youden avec des
périodes expérimentales de 14 jours. Les traitements de Cas (743 g/jour au duodénum)
et de C3 (1041 g/jour au rumen) ont été étudiés selon un plan factoriel. À chaque
période, au jour 11, de la L[l-13C]Leu (4.3 mmol/h) et au jour 13, du [13C]bicarbonate
de sodium (4.2 mmol/h) ont été perfusés dans une veine jugulaire. Des prélèvements
sanguins carotidiens et de la veine mammaire ont été effectués pour déterminer la
cinétique de la leucine (Leu) et les prélèvements mammaires nets des AA. Les
traitements de Cas et de C3 ont augmenté le taux proteique du lait et la production de
matières protéiques. Les traitements de Cas ont induit une réponse générale du
métabolisme proteique mammaire, avec une augmentation du prélèvement net
mammaire de la Leu (30 %), du ratio prélèvement net : Leu dans les protéines du lait (8
%), de la synthèse de protéines (9 %), de la production de protéines du lait (21 %) et de
l'oxydation (259 %). Avec les traitements de C3, il y a une tendance à augmenter les
protéines du lait (7 %). Lorsque le C3 est combiné avec la Cas, la Leu utilisée pour la
synthèse de protéines augmente moins que lorsque le C3 est ajouté seulement à la ration
témoin. Pour tous les traitements, la totalité des prélèvements mammaires nets de la Leu
a été utilisée pour la production de protéines du lait et l'oxydation, sans reste non
expliqué par ces deux voies métaboliques. Par conséquent il n'y a pas besoin
d'impliquer un prélèvement ou une production de peptides pour équilibrer le bilan de
prélèvement net mammaire de la Leu. Le prélèvement net des acides aminés du groupe
1 a augmenté avec les deux types de nutriments perfusés et s'équilibrait avec la quantité
d'AA sécrétés dans les protéines du lait. Pour les AA du groupe 2, le prélèvement net
mammaire était supérieur à la quantité retrouvée dans le lait, quel que soit le traitement.
Cependant ce phénomène était plus marqué avec les traitements Cas qu'avec les
traitements C3. Pour les AA du groupe 3, leur prélèvement net mammaire augmente
avec les traitements contenant du C3. Ceci suggère des mécanismes différents pour
augmenter la production de protéines du lait avec des apports protéiques (Cas) vs. des
apports énergétiques (C3).
143
Mots-clés: Caséine, propionate, glande mammaire, métabolisme de protéines.
6.2 Abstract
The effects of Casein (Cas) and propionate (C3) on mammary amino acid (AA)
metabolism were determined in three multiparous Holstein cows, fîtted with both
duodénum and rumen cannulas, used in a replicated Youden square with 14 d-periods.
Cas (743 g/din the duodénum) and C3 (1041 g/din the rumen) infusions were tested in
a factorial arrangement. For each period, L-[l-13C]leucine (d 11) and NaH[13C]C>3 (d 13)
were infused into a jugular vein and blood samples were taken from a carotid artery and
a mammary vein to détermine leucine (Leu) kinetics and net uptake of AA. Both Cas
and C3 treatments separately increased milk protein concentration and yield. With Cas
there was a gênerai response in mammary protein metabolism, involving increases in
Leu net uptake (30 %), the uptake:output ratio (8 %), protein synthesis (9 %), sécrétion
in milk protein (21 %) and oxidation (259 %). In contrast, C3 treatments only tended to
increase Leu in milk protein (7 %) and, when in combination with Cas, to reduce Leu
used for protein synthesis (6 %). Across ail treatments, most Leu uptake by the
mammary gland was accounted as Leu in milk or oxidized and therefore the Leu
balance was achieved without involvement of either net peptide utilization or
production. Mammary uptake of Group 1 AA increased to match milk output with ail
infusions. In contrast, mammary uptake of Group 2 AA exceeded output to a greater
extent with Cas compared with C3 infusion, while the incrément in uptake of Group 3
AA was higher with C3 than with Cas treatments. Overall, thèse data suggest that
différent mechanisms operate to improve milk protein production when either protein or
energy are infused.
Key words: Casein, propionate, mammary gland, protein metabolism.
144
6.3 Introduction
In dairy cows during mid-lactation, milk protein yield can be manipulated by
varying either energy or protein intakes (MacRae et al., 2000). In mechanistic terms,
however, changes in either energy or protein supply hâve been shown to exert différent
effects on whole body protein metabolism, as estimated through leucine kinetics
(Raggio et al., 2006). For example, casein infusion increased both whole body protein
synthesis and leucine exported in milk to a greater extent than did propionate (C3).
Furthermore, while leucine oxidation increased with casein treatment this oxidation was
lowered when propionate was added in combination with casein. What was unclear,
however, is whether thèse différent responses at the whole body level reflect altération
in mammary gland (MG) metabolism and the mechanisms that operate to regulate milk
protein yield.
In terms of understanding responses in mammary protein metabolism to changes
in either protein or energy supply, more is known about the impact of AA
supplementation and deficiencies than altération in either glucose or propionate
provision. For example, for dairy cows in late lactation a high protein intake increased
MG leucine irréversible loss rate (ILR) by 5 %, supporting a relatively small increase (6
%) in protein synthesis but with most of the additional leucine uptake catabolized by the
MG (Bequette et al., 1996a). In dairy goats, infusion of an AA mixture devoid of
leucine resulted in stimulation of mammary protein synthesis, with a concomitant
réduction of leucine oxidation to support the increased milk protein yield (Bequette et
al., 1996b). Such data demonstrate that, for leucine at least, the MG can adjust both
uptake and utilization mechanisms to improve milk protein output. Therefore, the
normal situation where leucine MG uptake is in excess of output in milk, usually plus or
minus 30 % (e.g. Guinard and Rulquin, 1994b; Raggio et al., 2004; Rulquin and
Pisulewski, 2000) should be seen as a fonction of availability from the diet and not a
fixed need by the MG (Bequette et al., 1996b). To date, however, no studies hâve
related the impact of energy supply alone, either in the form of glucose or C3, on
mammary protein metabolism in dairy cows. The potential impact of energy supply on
MG metabolism has been raised indirectly during studies involving hyper-insulinemic,
145
euglycemic clamps (Bequette et al., 2002). Although such experiments involved non-
physiological chronic élévation of plasma insulin concentrations, they hâve provoked
intriguing questions about the separate and combined rôles of energy and protein on
metabolism within the MG but, as yet, thèse hâve not been clearly separated and
identified.
Also, few data are available on responses in net mammary uptake of AA and
output as milk protein in response to increased energy supply and thèse hâve provided
equivocal information. For example, despite a moderate incrément in protein yield,
Huhtanen et al. (2002) were unable to detect any effect of glucose infusion (300 g/d) on
mammary net uptake of AA. Furthermore, at higher inputs of glucose (2398 g/d, as an
iso-calorie substitute for an infused VFA mixture) milk protein yield actually decreased
even though mammary uptake of EAA increased (Rulquin et al., 2004). This latter
observation suggests that thèse EAA might undergo metabolic fates within the MG
other than use for protein synthesis, e.g. they may be converted to peptides and returned
to the plasma as implied from other expérimental conditions (Guinard and Rulquin,
1994a; Rulquin et al., 2004). The form of energy supplied may also be critical in any
response as although glucose and C3 infusion both increased milk and milk protein
yield, differential effects were noted on milk fat synthesis (Rigout et al., 2003).
However, in studies with increases in either protein or energy supply there hâve
been some common responses, most notably the observation that the net uptake to
output ratio for certain EAA, notably the branched-chain AA and lysine, was also
increased (Bequette et al., 1996a; Guinard and Rulquin, 1994a; Metcalf et al., 1991;
Raggio et al., 2004; Rulquin et al., 2004; Vanhatalo et al., 2003b). This raises the
important question, what are the fates of this excess AA uptake? Some reports suggest
that thèse may be predominantly oxidized (Bequette et al., 1996a) and/or used to
provide N for synthesis of other AA (Lapierre et al., 2005). It has also been proposed
that the MG, under certain conditions, may release the excess AA taken up as peptides
into the mammary vein (Guinard and Rulquin, 1994a; Rulquin et al, 2004). Thèse
issues need to be resolved in order to better understand the various regulatory
146
mechanisms within the MG involved in the response to separate or combined altérations
in supply of nutrients.
Therefore, the objectives of this study were: 1) to differentiate how increased
milk protein sécrétion in response to change in energy and protein supply relates to
changes in MG protein metabolism and net uptake of AA and 2) quantify, for leucine,
the metabolic fate of that taken up in excess relative to milk output. Thèse objectives
were achieved with a combination of veno-arterio techniques coupled with infusions of
labelled (stable isotope) leucine and bicarbonate. Whole body protein kinetics from this
study hâve been reported previously (Raggio et al., 2006).
6.4 Materials and Methods
6.4.1 Animais and Treatments
Three multiparous Holstein cows, averaging 615 ± 24 kg BW and 65 ± 4 DIM at
the beginning of the study, were fitted with both a proximal T-shaped duodenal cannula,
placed 10 to 15 cm from the pylorus, and a ruminai cannula (Hurtaud et al., 1998)
before calving. Approximately 20 days after calving, cows were surgically prepared
with an ultrasonic flow probe (Transonic type A Systems Inc., Ithaca, NY) implanted
around the right external pudic artery as previously described (Rigout et al., 2002) and
two permanent cathéters were placed in the right carotid artery and in the right
subcutaneous abdominal vein (description of cathéters: Guinard et al., 1994, Raggio et
al., 2006). The right carotid artery was also raised to a subcutaneous position to provide
an alternative access to arterial blood if necessary. During the experiment, problems
were encountered with two cathéters: one arterial cathéter failed and a replacement
cathéter was inserted in the left carotid. In another cow the subcutaneous abdominal
vein cathéter was non functional and a temporary cathéters was inserted (for 3 days)
during the second week of each period. In addition, one ultrasonic flow probe failed
after the second period. For the tracer infusions, a cathéter was inserted in the left
jugular vein for the whole study. Cathéter maintenance was performed as described
previously (Guinard et al., 1994). The surgical préparations were reviewed and
approved by the Animal Care Committee of the French Ministry of Agriculture.
147
The same basai diet (Raggio et al., 2006) was fed during the whole study,
balanced to provide limited intestinal glucose (Rigout et al., 2003). The basai diet was
estimated to supply 119 Mj/d NEL and 1624 g/d metabolizable protein (NRC, 2001), or
1593 g/d of protein truly digested in the small intestine and 124 MJ/d NEL, (INRA,
1989). The quantity of wet feed offered was adjusted every day to ensure the same
delivery of DM on each expérimental day, based on an estimation of DM content of the
silage. The concentrate was supplied every 3h in equal portions from automatic feeders,
starting at 0715h. Grass silage was fed three times per day: 25 % at 0715h, 25 % at
1315h, 50 % between 1715h and 1915h, except during the kinetic days, when the silage
was fed five times per day: 12.5 % at 0715h, 1015h, 1315h and 1615h and then 50 %
at 1915h. Throughout the whole study, access to the diet was limited to lh after the
concentrate distribution times. Cows had free access to water and were housed in
individual tie stalls. Every day before 0715h, any feed refusais were collected, weighed
and sampled for future analyses.
Superimposed on the basai diet, infusion of two nutrients were tested in a 2 X 2
factorial arrangement: duodenal infusion of calcium caseinate (743 + 7 g/d: 687 g/d
protein truly digested in the small intestine (PDI) and estimated to provide 7.9 MJ/d of
NEL; Guinard et al., 1994) and/or ruminai infusions of propionic acid (1042 + 8 g/d,
estimated to provide 15.6 MJ/d of NEL; INRA 1989). Therefore, treatments were: 1)
control (Ctrl), 2) casein (Cas) 3) propionic acid (C3) and 4) the casein and propionic
acid combined (Cas+C3). Infusion of casein was the highest dose used by Guinard and
Rulquin (1994a), and increased the PDI balance from 97 % for the Crtl cows to an
average of 120 % for the cows receiving casein infusions. To optimize any effect, C3
was infused at the dose that yielded the maximal effect on milk protein observed by
Rigout et al. (2003). Each treatment period lasted 14 d, and the first two days of each
period included transition for the infusions (the cows received on day 1, 0.33, and on
day 2, 0.66 of the total subséquent infusion). Préparation of the infusions is detailed in
chapter V (Raggio et al., 2006).
148
6.4.2 Sampling and Laboratory Analyses
Cows were milked twice daily (0630 and 1830 h) with yield recorded at each
milking and assayed for fat and true protein composition by infrared analysis
(Milkoscan, Foss Electric, Hillerod Denmark). During the second week of the
expérimental period, cows were milked for each half udder separately and milk samples
(each half udder) were assayed for fat and true protein composition by infrared analysis
(Milkoscan, Foss Electric, Hillerad Denmark). Samples of the grass silage and
concentrate mixture were taken every day. Feed samples for analyses were stored at -
20°C until analysis, when they were pooled by period.
On day 11 of each expérimental period, leucine kinetics were determined with
an infusion of L[l-13C]leucine (Cambridge Isotope Laboratories, Andover, USA. 99
atom percent excess (ape), at 4.3 mmol/h: 14.20 mg/ml dissolved in stérile 0.9 % NaCl;
0.66 g solution/min) for 7.5 h (starting at lOOOh, 3.5h after the morning milking),
preceded by a priming dose of 4.3 mmol, équivalent to an hour of infusion (Wolfe,
1984).
Cows were made to stand for at least 15 min before blood sampling. Hourly
samples were taken from the carotid artery and mammary vein starting 3 h after
initiation of the infusion (1300, 1400, 1500, 1600 and 1700h) and were analyzed to
détermine the concentration and the isotopic enrichment (IE) of leucine, 4-methyl 2-
oxopentanoate (MOP) and CO2. In addition, samples for amino acid (AA) and urea
concentrations were taken at 0700, 0900, 1100, 1300, 1500 and 1700h.
On day 13, a mixture (in stérile 0.9 % NaCl) of [I3C]bicarbonate (Cambridge
Isotope Laboratories, Andover, USA. 99 ape; 4.7 mg/ml) and [6.6 2H2]glucose
(Cambridge Isotope Laboratories, Andover, USA. 99 ape; 60 mg/ml) was infused. The
rate of [13C]bicarbonate infusion was 4.2 mmol/h (1.26 g solution/min) for 4 h between
1200 to 1600h, preceded by a priming dose équivalent to 3.2 mmol. Glucose kinetics
will be described in a separate paper. Samples were taken from the carotid artery and
mammary vein starting 1.75h after the initiation of the infusion (1345, 1430, 1515 and
1600h) and were analyzed to détermine the concentration and the isotopic enrichment
149
(IE) of CO2. On both days, samples were also collected prior to the infusions to
détermine natural abundance of leucine and MOP (2 samples) and CO2 (3 samples).
Blood samples for AA concentrations were collected in heparinized syringes and
kept on ice. Plasma was prepared by centrifugation (2000 x g at 4°C for 15 min), 0.7 g
plasma was weighed, then 0.4 g of an internai standard solution was added and the
mixture was stored at -80°C. The internai standard solution was prepared at the
following concentrations: 501 nmol/g [5-15N]glutamine, 252 nmol/g [indole-15N]tryptophan, 260 nmol/g [l-13C]cysteine, 8.8 umol/g [15N2]urea, plus a U-13C algal
hydrolysate. This follows the methodology for détermination of AA concentration by
isotope dilution as previously described (Calder et al., 1999).
For MOP and [l-13C]leucine IE (expressed as molar percent excess, mpe), 0.7 g
plasma was weighed and 0.3 g of an oxohexanoic acid solution (33.3 nmol/g) added,
and the mixture was stored at -80°C until analysis. Déterminations of MOP and leucine
IE were as described by Calder and Smith (1988). For CO2 IE, triplicate 1 ml of blood
samples were injected into evacuated Vacutainers containing 1 ml of lactic acid,
immediately mixed and kept at room température until analysis was performed.
Détermination of IE was as previously described (Lobley et al., 2003) based on Read et
al. (1984).
6.4.3 Calculations
In addition to the blood flow probe, mammary plasma flow (PF) was estimated
according to the Fick pnnciple, using phenylalanine and tyrosine as internai markers
(Cant et al., 1993), with an allowance for a 3.56 % contribution from blood-derived
proteins. For this paper, the Fick principle was adopted to estimate plasma flow and
calculate mammary AA fluxes because the probe was not functional for one cow. For
the two functional cows, the probe measurements were 24 % lower than mammary flow
estimated with the Fick principle. Similar underestimates with ultrasound probes hâve
been observed previously when compared either with the Fick principle (Thivierge et
al., 2002) or with p-aminohippurate (Bequette et al., 1999). Nevertheless, there was no
effect of treatment on the uptake to output ratio of phenylalanine and tyrosine when the
150
uptake was estimated with the plasma flow from the probe for the two functional cows.
Sécrétions of AA into milk were calculated using published AA concentration in milk
(Swaisgood et al., 1993) except for leucine, phenylalanine and tyrosine for which milk
samples were acid-hydrolysed with 6 N phenol-HCl for 18 h at 110°C (AOAC, 2000)
and AA concentrations of the hydrolysates were determined by the isotope dilution
method (Calder et al., 1999). Concentrations leucine, phenylalanine and tyrosine
averaged 98, 49 and 56 mg of per g of crude protein, respectively.
For the following équations, infusion rates and fluxes of leucine, MOP or CO2
are in mmol/h, concentration of leucine, MOP are in uM, concentration of CO2 in mM,
IE of 13C leucine, 13C MOP and 13CO2) in mpe, with m referring to the mammary vein,
and a to the artery.
Mammary gland leucine kinetics were studied using a three compartment model
with leucine, CO2 and MOP as previously described by Bequette et al. (2002), but with
the IE of MOP from the mammary vein taken as représentative of the IE of the
precursorpool.
The MG irréversible loss rate (ILR) of leucine was calculated as follows:
MG Leu ILR = (([Leu-a] x IELeu-a - [Leu-m] x IELeu.OT) x PF ) / (IEMOp-w))
CorOx
MOP converted to Leucine was calculated as:
MG MOP to Leu = (([MOP-a] x IEMOp-a - [MOP-w] x IEMop-m) x PF ) / (IEMOp-
,«))
where [Leu] and [MOP] refer to concentrations of leucine and MOP
respectively, IE to their isotopic enrichment and PF to plasma flow. The factor CorOx
corrects for the extra oxidation in the MG due to the infusion of the tracer and is
described below.
151
Blood flow was used to calculate MG 13CO2 fluxes and was estimated as
follows:
Blood flow (BF) = PF / (1- Hematocrit-a)
The MG 13CO2 flow was calculated as follows:
MG13CO2 = ([CO2-/w] x IE13CO2-w - [CO2-fl] x IE13CO2-a ) x BF
To correct for 13CO2 séquestration (Sq) that occurred during the infusion of 13C
leucine, 13CO2 Sq was estimated during the bicarbonate infusion as the arterial inflow of13CO2 not recovered in mammary vein. During period 6, Sq was not determined due to
technical problems with the measurements of IE of 13CO2 during bicarbonate infusion.
We used an average of Sq for each cow (1.4 %, 1.2 % and 4.8 % respectively) because
no significant différences were observed between treatments.
Therefore, the corrected mammary 13CC>2 flux was calculated as follows:
MG 13CO2 corrected = (Arterial inflow of 13CO2 x Sq) + MG 13CO2
Oxidation of leucine across the MG was calculated as:
MG Leu Ox = MG 13CO2 corrected / IEMOp-m - CorOx
Tracer kinetics assumes the animal does not distinguish between 13C-AA and12C-AA. In addition, at the whole body level, it was assumed that the équivalent of
infused was in excess, and thus a corresponding amount of leucine was oxidized by the
animal (Raggio et al., 2006). Accordingly, it was assumed that mammary leucine
oxidation measured during the tracer infusion included a proportion of this extra-
oxidation, calculated to be in proportion to measured MG oxidation as a fraction of
measured whole body oxidation (before any correction was made to account for the
tracer infusion) multiplied by the infusion rate. This factor (CorOx) was subtracted from
measured leucine oxidation across the MG as well as ILR and net flux, which should
152
hâve both been increased by this amount during the tracer infusion. Fractional oxidation
was calculated as corrected oxidation divided by ILR.
Leucine used for MG protein synthesis was calculated as the différence between
MG Leu ILR, including any conversion to or from MOP, and MG oxidation. Once
leucine used for milk protein sécrétion was removed from leucine used for protein
synthesis, the remainder was converted into kg of constitutive protein applying the
factor 63 g leucine per kg of constitutive protein (Lobley et al., 1980).
Net leucine uptake was calculated using the 6 samples collected every other
hour, because milk protein yield over the 12 hours has been used to calculate the plasma
flow with the Fick principle. Measurements made during the infusion of the tracer were
corrected for CorOx.
"Residual" leucine was calculated to be the amount of leucine taken up by the
MG plus the amount of MOP converted into leucine minus oxidation and sécrétion as
milk protein, and would represent any net mammary leucine inflow having not been
accounted for in the différent outflows:
Residual = Net Leu uptake + MOP to Leu - MG Leu Ox - Leu in milk
If significantly différent from zéro this "residual" would need to be ascribed
another metabolic fate (e.g. release in peptide form into the mammary vein).
Net mammary uptake of other AA was calculated using the 6 samples collected
every other hour as follows:
net uptake = ([AA-a] - [AA-m] ) x PF
where [AA] refers to concentrations of individual AA.
153
6.4.4 Expérimental Design and Statistical Analyses
The design of the experiment was two incomplète 4 x 3 Youden squares with 3
periods each for a total of 18 observations (Ctrl: n = 5; Cas: n = 4; C3: n = 5; Cas + C3:
n = 4). During the third period, one cow had problem with her cathéter during Cas
treatment, as a conséquence n = 3 for Cas for the blood measurements.
Analyses of variance were made using the MIXED procédure of SAS (SAS
Institute Inc., 1999) according to the following statistical model:
ki =\i + SQUARE; + PERIODj (SQUAREj) + COWk + TREAT; + e iikl
Différences among treatments were compared using orthogonal contrasts
according to the factorial arrangement. Results are expressed as least square means with
the highest standard error of mean (SEM).
6.5 Results
Milk production and composition for the last week of the expérimental periods
were presented in a previous paper (Raggio et al., 2006). In the présent paper, milk
production and composition were based on the right half udder at the evening milking
(12h) of the leucine infusion day (Table 6.1). Milk yield increased with Cas treatments
(P < 0.001). True protein concentration and yield both increased with Cas (P = 0.04; P
< 0.01, respectivaly) and C3 (P = 0.03; P = 0.06 respectively) infusion. Milk fat
concentration was reduced with Cas infusion (P = 0.07), as conséquence of increased
milk yield, as fat yield was not affected by Cas infusion.
6.5.1 Mammary Leucine Kinetics
The mammary plasma flow decreased by an average of 16 % (P = 0.07) when
Cas was infused and increased by 32 % (P = 0.02) when C3 was supplied (Table 6.2).
The MG leucine kinetics are presented in Table 6.2. The net uptake of leucine and MOP
increased with Cas infusion (P < 0.01) but did not show any effect when C3 was
infused. The MG Leu ILR was increased by Cas infusion (P < 0.001) and tended to
increase with C3 (P = 0.09). In addition, there was a Cas x C3 interaction (P < 0.01),
154
because the response with combined treatments was smaller than the sum of each
nutrient supplied separately. MOP converted to leucine (MOP to Leu) and mammary
oxidation of leucme (Leu Ox) increased when Cas was supplied (P = 0.04; P < 0.001
respectively). The mammary fractional oxidation of leucine increased {P < 0.001) with
Cas supply. Leucine used for protein synthesis (Leu PS) increased with Cas (P = 0.02)
but, as with ILR, there was a tendency for Cas x C3 interaction (P = 0.09). Leucine in
milk protein (Leu Milk) increased with Cas (P < 0.001) and tended to increase with C3
(P = 0.06). In addition, différences were observed for Leu derived from protein
dégradation (estimated as Leu PS minus Leu Milk), with an increase with Cas alone
compared with Ctrl but a decrease with Cas and C3 combined compared with C3
(interaction Cas x C3, P < 0.001).
6.5.2 Amino Acid Arterio-Venous Concentration Différences
The AA arterio-venous (AV) concentration différences are presented in Table
6.3. When Cas was supplied, the AV différence increased for ail AA (P < 0.05), except
for aspartate and glutamate, where there was no effect, and for alanine, where there was
a decrease (P = 0.03). In contrast, during C3 infusions AV différences decreased for ail
AA (P < 0.06), except aspartate, cysteine, glutamate, serine and threonine which were
not affected (P > 0.10) and alanine where there was an increase (P < 0.01). There was
also a trend (P < 0.10) for an interaction with the branched-chain (leucine, isoleucine
and valine) AA, with a larger decrease with C3 and Cas were infused together compared
with Cas than when comparing C3 with Ctrl.
6.5.3 Amino Acid Uptake and Uptake : Output Ratio
The MG uptake data are presented in Table 6.4. when Cas infusion the uptake
increased (P < 0.01) for ail the essential AA as well as for glutamine, glycine, and
proline, while uptakes were reduced for alanine (P < 0.001) and glutamate (P = 0.02).
With C3 supply, uptake of alanine, glutamate and proline, as well as methionine,
phenylalanine and threonine increased (P < 0.05). There was also a Cas x C3 interaction
(P = 0.02) for alanine uptake, with a larger decrease when Cas was added to the Ctrl
diet than when Cas was added to C3.
155
The AA uptake: output ratios are présentée in Table 6.5. If the ratio exceeds
unity, then the mammary uptake was more than sufficient to account for milk protein
sécrétion; if smaller than unity then there was a need for either synthesis de novo in the
gland or uptake in peptide forai. Uptake of EAA was sufficient to account for milk
output, except for methionine when Cas was infused. Uptake of the NEAA relative to
milk output varied greatly between AA, from a large excess for arginine to ratios as low
as 0.14 for aspartate. With Cas treatments, the uptake:output ratios decreased (P <
0.001) for alanine, glutamate and glycine with a tendency (P = 0.09) for metionine,
while they increased for proline (P < 0.001), with a tendency to increase (P < 0.08) for
isoleucine and leucine. However, when C3 was supplied, the ratio for alanine increased
(P < 0.001) and tended to either increase (P = 0.07) for glutamate or decrease (P = 0.08)
for glutamine.
6.5.4 Nitrogen Balance Across the Mammary Gland
Nitrogen balances are presented in Table 6.6 This calculation is based on
transfers for the protein bound AA-N, except for asparagine, which was not measured.
Total N uptake increased (P < 0.01) when Cas was supplied (P < 0.01), with a tendency
to increase when C3 was infused (P = 0.07). Nitrogen uptake as EAA (EAA-N)
increased (P < 0.001) with Cas but C3 had no effect. Nitrogen uptake as NEAA
(NEAA-N) was increased by both Cas (P = 0.05) and C3 (P = 0.03). Total-N, EAA-N
and NEAA-N in milk were ail increased for Cas (< 0.001), and tended to increase for
C3 (P = 0.07). In tenus of N balances across the MG, the NEAA-N déficit increased (P
= 0.05) when Cas was infused.
156
6.6 Discussion
There were two objectives within this study. The first was to examine if protein
and energy, in the form of propionate, had similar mechanistic effects on protein and
AA metabolism within the MG to increase milk protein yield. The second issue was to
ascertain the fate of excess AA taken up by the MG and help résolve the issue whether
the MG can produce net sécrétion of peptides or whether oxidative and transamination
pathways are the major routes of removal. In both cases, clear answers to thèse
questions were obtained.
6.6.1 Protein Metabolism
6.6.1.1 Leucine Net Uptake
There were obvious différences in how net uptake of leucine and sécrétion as
milk were affected by supply of Cas and C3. With Cas treatments, the incrément in net
uptake of leucine was greater than the increase in milk protein yield, with only 59 % of
the incrément in net uptake transferred to milk. This supports earlier observations that
the différence between mammary net uptake of leucine and sécrétion in milk protein
becomes greater with increased protein supply (e.g. in terms of the uptake:output ratio
this increased from 1.04 to 1.87 with casein infusion, Guinard and Rulquin, 1994a; from
1.05 to 1.66 with leucine infusion, Rulquin and Pisulewski, 2000; and from 1.10 to 1.40
with a high RUP protein supplément, Raggio et al., 2004). This indicates that
metabolic pathways other than milk protein synthesis are stimulated by protein supply
and thèse can be determined using tracer infusion, as discussed below.
In contrast, when C3 was infused, the non-significant increase in leucine net
uptake (4.4 %) was less than the extra leucine secreted in milk protein. In conséquence
the apparent efficiency of utilization of the extra uptake was 130 %. Similarly,
Vanhatalo et al. (2003a) did not observe an increase in the uptake of leucine, although
milk protein yield was improved by infusion of glucose. Together, thèse data suggest
that when increased energy supply élevâtes milk protein yield, without any concomitant
incrément in protein supply, the MG uses its leucine uptake more efficiently towards
milk protein sécrétion. Energy does not always exert a positive effect, however, because
157
when a very high dose of glucose was infused, as an energy substitution, the uptake to
output ratio of leucine increased, due to a combined decrease in milk protein yield
coupled with increased leucine net uptake (Rulquin et al., 2004).
6.6.1.2 Leucine Kinetics
The current kinetic data are quantified assuming the enrichment of MOP in the
mammary venous plasma is similar to that of the precursor intracellular pool. As MOP
is formed intracellularly from leucine deamination, MOP release into the venous
drainage ideally should reflect intracellular enrichment of leucine. In practice, only part
of the venous metabolite is derived from MG intracellular outflow while the rest arises
from bypass arterial supply (Biolo et al., 1995), and so the IE of venous MOP will be
lower than MOP or Leu in the cells of the MG. Indeed, the intracellular leucine
enrichment will lie between those of MOP and leucine in the venous plasma.
Consequently, use of venous MOP enrichment as precursor will slightly over-estimate
the true rates of protein synthesis but still allows comparison between treatments.
Compared with Crtl, Cas alone increased MG ILR by 23 %, confïrming the
gênerai response observed previously by Bequette et al. (1996a) in dairy cows with
increased dietary protein supply. For the current cows the dynamic increase (5.8
mmol/h) was partitioned approximately equally between oxidation and protein
synthesis, of which 80 % went to exported proteins (milk), and the rest to constitutive
sources.
In contrast, with C3 infusions ail of the ILR incrément (0.96 mmol/h) was
diverted to protein synthesis with a minor effect on oxidation (0.04 mmol/h). This
indicates that the MG adjusts differently to increased protein versus energy supply.
While energy supply is known to impact on plasma insulin concentrations, thèse were
unaltered in the current study (unpublished data) and thus other mechanisms must
operate. As energy appears to induce différent responses than protein and thus the
possibility arises of additive, or even synergistic, effects when both nutrients are
supplied together. In practice, however, the interaction between Cas and C3 on protein
synthesis indicated that co-supply of Cas and C3 was not completely additive.
158
In this study, MG ILR accounted for 0.38 to 0.43 of whole body ILR, as
previously reported in dairy cows (37-44 %, Bequette et al., 1996a; 41-43 %: Thivierge
et al., 2002), confirming the major contribution of the MG to whole body metabolism.
The MG also played a major rôle in whole body leucine oxidation, accounting for 40-50
% across the range of treatments studied. Certainly the gland can remove close to half of
any excess supply and, therefore, in the lactating animal may be more important than
the other ruminant tissues known to be involved in leucine catabolism (Goodwin et al.,
1987). Estimated synthesis of constitutive protein by the MG averaged 980 g/d,
indicating the high turnover rate of constitutive protein in the MG.
6.6.1.3 Net Uptake of Amino Acids
The use of précise analytical techniques to measure AA concentrations in the
current study also allows examination of how Cas and C3 modified the net mammary
uptake of individual AA to support the increased milk protein yield generated.
6.6.1.3.1 Effect of Casein
Treatment effects on arterial AA concentrations hâve already been discussed in a
companion paper (Raggio et al., 2006). Across the MG the AV différences increased for
almost ail essential AA with Cas infusions, as previously reported with infusion of
casein ranging from 177 to 762 g/d (Clark et al., 1977; Guinard and Rulquin, 1994a;
Vanhatalo et al., 2003b). In thèse studies net uptake, calculated from reported AV
différences and plasma flows, increased with Cas infusions. Despite the casein-induced
decrease in plasma flow in the current experiment, Cas treatments also increased net
uptake of ail EAA.
Mepham (1982) divided AA into three groups, each having a différent
metabolism within the MG. Group 1 includes histidine, methionine, phenylalanine &
tyrosine, threonine and tryptophan, and for thèse AA their stoichiometric transfer yields
uptake:output ratios close to unity (e.g. Guinard and Rulquin, 1994a: 1.14; Miettinen et
al., 1997: 1.02; Raggio et al., 2004: 0.95). The différences between studies might be due
to différent energy and/or protein status of the animais as well as différent
méthodologies, including blood flow measurement (probe, Fick principle, p-
159
aminohippurate), AA analyses (number of samples taken, use of individual or pooled
samples, analysis technique), frequency of blood sampling relative to feeding protocols.
Due to uncertainties for Thr to belong to this group (Lapierre et al., 2005), Thr was not
included in Group 1 calculations. In line with the concept of Mepham, MG uptake of
Group 1 A A increased to mamtain at unity (0.99) the uptake :output ratio as milk protein
production changed. This supports previous reports where, with additional protein
supply, the udder extracted thèse AA in ratios equal to their sécrétion in milk protein
(Guinard and Rulquin, 1994a; Raggio et al., 2004; Vanhatalo et al., 2003a).
The second group of Mepham (1982) included EAA usually taken up in excess
by the MG relative to milk output, i.e. isoleucine, leucine and valine (the branched-
chain AA) and lysine (Clark et al., 1977; Guinard and Rulquin, 1994a; Raggio et al.,
2004). Arginine was not considered an EAA and therefore not included in Group 2 for
calculations. Casein infusions resulted in an increased uptake:output ratio for the Group
2 AA from 1.28 to 1.34 comparable to other findings e.g. Clark et al. (1977) from 1.41
to 1.53, Guinard and Rulquin (1994a) from 1.21 to 1.64, Vanhatalo et al. (2003a) from
1.57 to 1.61 and Vanhatalo et al. (2003b) from 1.33 to 1.50.
For many of the AA of the third group, which comprises the NEAA except
tyrosine, uptake was far less than output under basai conditions (Table 5), except for
glutamine and alanine (close to balance) and glycine (net release). For arginine (hère
considered as a non-essential AA) uptake was consistently in excess of output. Casein
infusions increased net uptakes of glutamine, glycine and proline, while alanine and
glutamate net uptake decreased. Vanhatalo et al. (2003a) also observed a decrease of
glutamate uptake with 350 g of Cas but no change in alanine uptake. Overall, for Group
3 AA, net uptake increased less than the output in milk in response to Cas infusions,
leading to a greater déficit in uptake relative to output (Table 6).
Therefore, the MG responded to the increased demand for NEAA for milk
protein synthesis with Cas supply by increasing the uptake of the EAA of group 2 to a
greater extent than the net uptake of the NEAA per se. Based on transfers of AA présent
in milk protein, thèse increases were enough to achieve a close overall balance for N
transactions in the MG. Why the gland increases uptake of Group 2 AA to restore
160
balance instead of extracting more NEAA from the blood during Cas supplementation
remains a mystery. Obviously this is not due to lack of NEAA as both NEAA and EAA
were supplied through infusion of casein. In addition, it has been shown, using tracer
kinetics, that for certain NEAA where the net uptake to output ratios were less than
unity across the caprine MG there was isotopic dilution between the artery and vein
(Bequette et al., 1997). This means that although the MG only extracted a proportion of
the NEAA needed to support milk protein output, there was sufficient synthesis de novo
of thèse within the MG to create an excess that was then released into the mammary
vein. This may suggest that synthesis de novo is an important process within the MG
when there is a gênerai excess of AA. Whatever the reason, the metabolism of the MG
adjusts to produce more milk protein from extra protein supply by using additional N
from those EAA extracted in excess to synthesize NEAA more than increasing, on a net
basis, extraction of NEAA.
6.6.1.3.2 Effect of propionate
When C3 was infused there were différences in how the MG achieves N
balance. Hère, plasma flow increased, regardless of whether determined by the Fick
principle (Clark et al., 1977; Huhtanen et al, 2002; Vanhatalo et al., 2003b), or
through use of flow probes (Rulquin et al., 2004). Although net uptake of Group 1 AA
increased to parallel the extra milk protein output, the largest changes were observed
with Group 3 AA. The uptake:output ratio of both alanine and glutamate increased to
considerably exceed unity. Thèse changes, coupled with increased net uptake of some of
the other NEAA, as also observed for alanine (Clark et al., 1977) and proline
(Vanhatalo et al., 2003b; Rulquin et al., 2004), were sufficient to reduce the déficit for
Group 3 AA (Table 6). Therefore, the mechanisms that operate to stimulate milk protein
output in the absence of extra protein supply, i.e. solely with energy provision, involve
increased MG uptake of some AA of Group 3. This contrasts with the response to
increased protein supply, where uptake of Group 2 AA is augmented and again this
suggests that the MG may be an important site within the body for removal of excess
Group 2 AA and the transfer of their N and C structures to form Group 3 AA may be
part of the homeostatic process. Confirmation of this concept requires more detailed
161
studies on transport kinetics and N transfers within the MG. Such approaches would
also help unravel how différent pathways respond to changes in protein and energy
supply.
6.6.1.4 Peptide Release
The second objective was to examine the fate of leucine uptake under similar
conditions to those reported where uptake of leucine was in large excess of milk output
and where there was an unaccounted AA balance, hypothesized as a release of peptides
by the MG (Guinard and Rulquin, 1994a; Rulquin et al., 2004). Within the current
study, the various metabolic fates of leucine were quantified, permitting a complète
balance for this AA. In the Ctrl cows, ail the leucine extracted by the MG was secreted
as milk or oxidized. Furthermore, the extra uptake observed with increased protein
supply was also directed towards thèse two metabolic pathways. In ail treatments, the
unaccounted leucine across the MG (the différence between uptake plus movement
from MOP minus oxidation minus milk) was small and not signifïcantly différent from
zéro. Similarly, the N balance across the MG for AA used for protein synthesis (without
asparagine) was not différent from 0. Thus, there was no need to invoke an additional
metabolic route via peptide release into the mammary vein. This is similar finding
where increased lysine-N uptake was quantitatively used for mammary synthesis of
NEAA (Lapierre et al., 2005) and where it was also unnecessary to hypothesize export
of peptides.
6.7 Conclusion
The responses in milk protein output when Cas or C3 were supplied to a basai
ration involve différent mechanisms within the MG. Both treatments increased
mammary Leu ILR, but the response was greater when Cas was infused, accompanied
by increased oxidation. During Cas infusions, N balance across the MG was mainly
achieved by increased uptake of Group 1 and 2 AA but during C3 supplementation this
was accomplished by increased uptake of AA from Group 1 and 3. No évidence was
found for release of peptides by the MG.
6.8 Acknowledgements
162
The authors gratefully thank from the UMR PL, INRA, France, Y. Lebreton for
surgeries, P. Lamberton and his team for technical support and animal care during the
experiment and C. Nahuet, L. Finot, N. Huchet, I. Jicquel, and S. Rigault for their help
in laboratory analyses and from the Rowett, A.G. Calder and S.E. Anderson for mass
spectrometry analyses. The authors also wish to acknowledge the financial support of
the INRA, the European Union for the award of Marie Curie Training Site funding to
the Rowett Research Institute, SEERAD and the National Science and Engineering
Research Council of Canada and Agriculture and Agri-Food Canada (Lennoxville
Research Centre Contribution number XXX).
163
Table 6.1. Effect of casein (Cas) and propionate (C3) supply on milk yield andcomposition of the right half udder in dairy cows.
Milk, kg/12 hFCM,kg/12hFat, g/kg
g/12hTrue Protein,
g
g/kg/12h
Ctrl6.386.7844.828229.3186
TreatmentsCas7.397.5342.330430.2222
C36.557.1146.129930.6197
Cas + C37.387.2439.028632.8240
SEM1
0.190.402.7124.310.767.12
CasO.001
0.240.070.820.04
<0.01
P2
C30.680.960.710.970.030.06
Cas x C30.600.400.360.400.350.59
'Least square means présentée! with pooled SEM, given for n=3; from the half udder in theevening (12h) of the leucine kinetic day (17 observations; Ctrl = 5, Cas = 3, C3 = 5, Cas + C3 =4).2Probability corresponding to the null hypothesis with Cas, C3 and CasxC3 contrasts.
164
Table 6.2. Effect of casein (Cas) and propionate (C3) supply on leucine kinetics andplasma flow of the right half udder in dairy cows.
mmol/hNet Leu 3
Net MOP 4
Leu ILR5
MOP to LeuLeu Ox6
Fractional OxLeu PS7
Leu Residual8
Leu Milk 9
Leu Milk / Leu PSLeu PS - Leu Milk9
MG Leu Ox / WB Leu Ox10
MGILR/WBILR11
Leu Efficiency 12
Leu PS for constitutiveprotein, g/d3
Plasma flow, 1/h
Ctrl13.480.22
18.650.131.520.08
17.250.53
11.640.754.540.490.420.85
868307
TreatmentCas
17.630.37
24.480.294.280.18
20.49-0.2113.930.774.990.430.430.78
1031241
C314.110.19
21.070.161.690.08
19.540.28
12.340.716.020.400.430.87
983386
Cas+C318.370.29
23.970.254.040.16
20.18-0.3815.010.833.620.470.380.81
1015338
SEM1
0.830.030.510.060.380.010.760.760.430.010.330.110.020.03
38.6731.50
CasO.001<0.01<0.001
0.04O.001O.001
0.020.30
O.001O.001<0.01
0.980.220.06
0.020.07
P2
C30.400.190.090.840.900.550.210.770.060.620.850.850.400.42
0.210.02
Casx C30.940.390.010.450.540.540.090.950.610.02
<0.0010.490.100.90
0.090.75
1 Least square means presented with pooled SEM, given for n=3; 17 observations; Ctrl =5, Cas = 3, C3 = 5, Cas + C3 = 4.2 Probability corresponding to the null hypothesis with Cas, C3 and Cas x C3 contrasts.3 Based on the six samples collected every other hour, corrected for increased oxidationdue to tracer infusion.4 Net 4-methyl 2-oxopentanoate (MOP) uptake during leucine kinetics measurement.5 ILR: irréversible loss rate corrected for increased oxidation due to tracer infusion.6 Leu oxidation (Ox) corrected for increased oxidation due to tracer infusion.7 Leu used for protein synthesis (PS)8 Calculated as Net Leu + Net MOP - Leu Ox - Leu Milk 12 h.9 Leu in milk protein, from milk collected from the right half udder at the eveningmilking on the day of kinetics measurements.10 Mammary gland (MG) oxidation (half udder x 2) / whole body (WB) oxidation,adjusted to use the same precursor pool.11 MG ILR (half udder x 2) / WB ILR, adjusted to use the same precursor pool.12 Leu Efficiency = Leu Milk / Net Leu Uptake corrected by Leu infusion13 Calculated as Leu PS - Leu in milk, transferred as protein using 63 mg Leu per g ofconstitutive proteins
165
Table 6.3. Effect of casein (Cas) and propionate (C3) supply on plasma arteriovenousdifférence of amino acids on the right half udder in dairy cows.
Amino AcidsHMAlaArgAspCysGinGluGlyProSerTyrHisIleLeuLysMetPheThrTrpVal
Ctrl28.524.5
1.82.4
36.239.5-3.99.36.9
14.59.5
29.945.637.79.7
15.619.63.4
37.1
TreatmentCas16.532.4
2.53.4
54.236.54.2
20.415.221.613.547.673.156.913.422.427.1
5.857.4
C333.422.3
2.32.2
26.140.6-5.48.37.4
11.78.2
25.238.432.2
8.113.817.33.2
30.5
Cas+C333.127.8
2.52.9
40.236.9
0.118.011.716.711.337.658.148.211.219.323.64.4
44.5
SEM1
2.641.340.400.304.272.611.620.793.141.780.751.602.603.000.781.070.000.321.73
Cas0.03
<0.010.270.01
<0.010.17
<0.01O.01
0.05O.001O.01O.001<0.001O.001O.001<0.01O.01<0.01<0.001
P2
C3<0.01
0.030.540.250.020.740.100.050.610.060.04
<0.01<0.01
0.040.040.040.150.01
O.001
CasxC30.390.320.630.730.600.880.370.330.490.500.490.100.120.550.650.470.710.180.07
' Least square means presented with pooled SEM, given for n=3; 17 observations; Ctrl = 5, Cas= 3, C3 = 5, Cas+C3 = 4.2 Probability corresponding to the null hypothesis with Cas, C3 and CasxC3 contrasts.
166
Table 6.4. Effect of casein (Cas) and propionate (C3) supply on amino acid uptake bythe right half udder in dairy cows.
Amino Acidsmmol/h
AlaArgAspCysGinGluGlyProSerTyrHisIle
Leu3
LysMetPheThrTrpVal
Ctrl8.87.40.50.7
11.112.1-1.32.72.04.42.99.1
13.511.42.94.75.91.2
11.2
TreatmentCas
3.68.10.60.8
13.78.91.05.13.65.43.3
11.917.614.13.35.56.71.5
14.3
C312.88.50.90.89.8
15.6-2.43.22.64.43.19.5
14.112.33.15.26.51.2
11.5
Cas+C310.99.30.80.9
13.212.2
-0.015.83.85.43.7
12.418.415.93.76.37.81.4
14.7
SEM1
0.640.830.180.090.661.290.510.230.890.250.170.570.830.840.100.200.370.090.69
Cas<0.001
0.330.990.11
O.0010.02
<0.01O.001
0.11<0.01
0.01O.001O.001<0.001O.001<0.01
0.010.01
<0.01
?z
C3O.001
0.180.130.240.210.030.070.020.670.890.120.340.400.140.050.010.050.920.59
CasxC30.020.940.710.660.520.950.990.450.760.980.690.930.940.510.260.430.520.640.94
1 Least square means presented with pooled SEM, given for n=3; 17 observations; Ctrl5, Cas = 3, C3 = 5, Cas+C3 = 4.2 Probability corresponding to thi3 Corrected for increased oxidation due to tracer infusion.
2 Probability corresponding to the null hypothesis with Cas, C3 and CasxC3 contrasts.
Table 6.5. Effect of casein (Cas) and propionate (C3) supply on amino acid uptake: milkoutput ratio in dairy cows.
Amino AcidsMmol/h
AlaArgAspCysGinGluProSerHisIleLeuLys
Ctrl1.052.440.140.801.090.930.200.211.061.291.161.29
TreatmentCas0.502.230.120.781.140.580.310.321.021.411.271.33
C32.082.640.210.830.921.140.220.261.081.281.141.30
Cas + C31.462.330.160.831.020.740.330.301.061.371.221.40
SEM1
0.090.240.040.070.070.090.010.080.050.050.050.07
CasO.001
0.230.390.860.30
O.001O.001
0.300.500.070.080.29
P'C3
O.0010.530.230.630.080.070.160.890.640.610.500.51
l
Cas x C30.060.830.610.840.670.730.930.700.910.760.760.59
167
MetThrTrpVal
1.001.051.021.28
0.950.991.071.36
0!11
.99
.09
.00
.24
0.971.060.971.29
0.020.040.070.06
0.090.270.880.23
0.940.260.430.36
0.420.720.520.78
'Least square means présentée with pooled SEM, given for n=3; milk output wasestimated from milk collected from the half udder on the evening of the leucine kineticday (17 observations; Ctrl = 5, Cas = 3, C3 = 5, Cas + C3 = 4).2Probability corresponding to the null hypothesis with Cas, C3 and Cas x C3 contrasts.3Phe and Tyr were not included because they were used to calculated plasma flow.
168
Table 6.6. Effect of casein (Cas) and propionate (C3) supply on the nitrogen balanceacross the right half udder in dairy cows.
NitrogenMmolhUptake
Total-N3
EAA-N4
NEAA-N5
Groupe l6
Groupe 27
Groupe 38
MilkTotal-NEAA-NNEAA-NGroupe 1Groupe 2Groupe 3
BalanceTotal-NEAA-NNEAA-NGroupe 1Groupe 2Groupe 3
Ctrl
163.081.381.723.056.779.6
156.168.987.122.545.384.9
6.912.3-5.40.5
11.4-5.3
TreatmentCas
189.5100.788.827.072.186.4
186.882.6
104.326.954.2
101.7
2.718.2
-15.50.1
17.9-15.3
C3
178.086.491.724.559.789.7
165.273.092.223.947.989.9
12.813.3-0.50.6
11.8-0.2
Cas + C3
212.8109.2103.629.577.4
101.1
200.988.8
112.229.058.3
109.3
11.820.3-8.60.5
19.1-8.3
SEM1
9.434.894.810.883.704.98
5.742.543.200.831.673.11
8.334.174.510.543.20
4.8
Cas
0.01<0.01
0.05<0.01<0.01
0.09
<0.01<0.01<0.01<0.01<0.01<0.01
0.720.110.050.540.040.01
?2
C3
0.070.180.030.050.260.03
0.070.070.070.070.070.07
0.360.690.200.600.790.21
Cas x C3
0.620.690.570.540.730.59
0.610.610.610.610.610.61
0.830.880.790.810.890.82
1 Least square means presented with pooled SEM, given for n=3; milk output was estimatedfrom milk collected from the half udder on the evening of the leucine kinetic day (17observations; Ctrl = 5, Cas = 3, C3 = 5, Cas + C3 = 4).2 Probability corresponding to the null hypothesis with Cas, C3 and CasxC3 contrasts.3 Total-N = N-His + N-Ileu + N-Leu + N-Lys + N-Met + N-Phe + N-Thr + N-Trp +
N- Val + N-Ala + N-Arg + N-Asp + N-Cys + N-Glu + N-Glm + N-Gly +N-Pro + N-Ser + N-Tyr.
4 EAA-N = N-His + N-Ileu + N-Leu + N-Lys + N-Met + N-Phe + N-Thr + N-Trp +N- Val.
5 NEAA-N = N-Ala + N-Arg + N-Asp + N-Cys + N-Glu + N-Glm + N-Gly + N-Pro +N-Ser + N-Tyr.
6Groupe 1 = N-His + N-Phe + N-Met + N-Tyr + N-Trp.7Groupe 2 = N-Ileu + N-Leu + N-Val + N-Lys.8Groupe 3= N-Ala + N-Arg + N-Asp + N-Cys + N-Glu + N-Glm + N-Gly + N-Pro +
N-Ser + N-Tyr.
169
6.8 Références
AOAC. Association of Officiai Analytical Chemists. 2000. Officiai Methods of
Analysis 17th éd. AOAC, Arlington, VA.
Baumann, P. Q., W. S. Stirewalt, B. D. O'Rourke, D. Howard, and K. S. Nair. 1994.
Precursor pools of protein synthesis: a stable isotope study in a swine model. Am. J.
Physiol. 267:E203-E209.
Bequette, B. J., F. R. Backwell, M. S. Dhanoa, A. Walker, A. G. Calder, D. Wray-
Cahen, J. A. Metcalf, J. D. Sutton, D. E. Beever, and G. E. Lobley. 1994. Kinetics of
blood free and milk casein-amino acid labelling in the dairy goat at two stages of
lactation. Br. J. Nutr. 72:211-220.
Bequette, B. J., J. A. Metcalf, D. Wray-Cahen, F. R. Backwell, J. D. GFtton, M. A.
Lomax, J. C. MacRae, and G. E. Lobley. 1996a. Leucine and protein metabolism in the
lactating dairy cow mammary gland: responses to supplemental dietary crude protein
intake. J. Dairy Res. 63:209-222.
Bequette, B. J., F. R. Backwell, J. C. MacRae, G. E. Lobley, L. A. Crompton, J. A.
Metcalf, and J. D. Sutton. 1996b. Effect of intravenous amino acid infusion on leucine
oxidation across the mammary gland of the lactating goat. J. Dairy Sci. 79:2217-2224.
Bequette, B. J., F. R. Backwell, C. E. Kyle, A. G. Calder, V. Buchan, L. A. Crompton,
J. France, and J. C. MacRae. 1999. Vascular sources of phenylalanine, tyrosine, lysine,
and methionine for casein synthesis in lactating goats. J. Dairy Sci. 82:362-377.
Bequette, B. J., C. E. Kyle, L. A. Crompton, S. E. Anderson, and M. D. Hanigan. 2002.
Protein metabolism in lactating goats subjcted to the insulin clamp. J. Dairy Sci.
85:1546-1555.
Beugnet, A., X. Wang, and C. G. Proud. 2003. Target of Rapamycin (TOR)-signaling
and RAIP Motifs play distinct rôles in the Mammalian TOR-dependent phosphorylation
of initiation factor 4E-binding protein. J. Biol. Chem. 278:40717-40722.
170
Biolo, G., R. Y. D. Fleming, S. P. Maggi, and R. R. Wolfe. 1995. Transmembrane
transport and intracellular kinetics of amino acids in human skeletal muscle. Am. J.
Physiol. 267:E467-E474.
Calder, A. G. and A. Smith. 1988. Stable isotope ratio analysis of leucine and
ketoisocaproic acid in blood plasma by gas chromatography/mass spectrometry. Use of
tertiary butyldimethylsilyl derivatives. Rapid Commun. Mass Spectrom. 2:14-16.
Calder, A. G., K. E. Garden, S. E. Anderson, and G. E. Lobley. 1999. Quantitation of
blood and plasma amino acids using isotope dilution électron impact gas
chromatography/mass spectrometry with U-(13)C amino acids as internai standards.
Rapid Commun. Mass Spectrom. 13:2080-2083.
Campbell, E., X. Wang, and C. G. Proud. 1999. Nutrients differentially regulate
multiple translation factors and their control by insulin. Biochem. J. 344:433-441.
Cant, J. P., E. J. DePeters, and R. L. Baldwin. 1993. Mammary amino acid utilization in
dairy cows fed fat and its relationship to milk protein dépression. J. Dairy Sci. 76:762-
774.
Clark, J. H., H. R. Spires, R. G. Derrig, and M. R. Bennink. 1977. Milk production,
nitrogen utilization and glucose synthesis in lactating cows infused postruminally with
sodium caseinate and glucose. J. Nutr. 107:631-644.
Goodwin, G. W., W. Gibboney, R. Paxton, R. A. Harris, and J. A. Lemons. 1987.
Activities of branched-chain amino acid aminotransferase and branched-chain 2-oxo
acid dehydrogenase complex in tissues of maternai and fetal sheep. Biochem. J.
242:305-308.
Greiwe, J. S., G. Kwon, M. L. McDaniel, and C. F. Semenkovich. 2001. Leucine and
insulin activate p70 S6 kinase through différent pathways in human skeletal muscle.
Am. J. Physiol. 281:E466-E471.
Guinard, J., H. Rulquin, and R. Vérité. 1994. Effect of graded levels of duodenal
infusions of casein on mammary uptake in lactating cows. 1. Major nutrients. J. Dairy
Sci. 77:2221-2231.
171
Guinard, J. and H. Rulquin. 1994a. Effect of graded levels of duodenal infusions of
casein on mammary uptake in lactating cows. 2. Individual amino acids. J. Dairy Sci.
77:3304-3315.
Guinard, J. and H. Rulquin. 1994b. Effects of graded amounts of duodenal infusions of
lysine on the mammary uptake of major milk precursors in dairy cows. J. Dairy Sci.
77:3565-3576.
Harper, A. E., R. H. Miller, and K. P. Block. 1984. Branched-chain amino acid
metabolism. Annu. Rev. Nutr. 4:409-454.
Huhtanen, P., A. Vanhatalo, and T. Varvikko. 2002. Effects of abomasal infusions of
histidine, glucose, and leucine on milk production and plasma metabolites of dairy cows
fed grass silage diets. J. Dairy Sci. 85:204-216.
Hurtaud, C, H. Rulquin, and R. Vérité. 1998. Effects of level and type of energy source
(volatile fatty acids or glucose) on milk yield, composition and coagulating properties in
dairy cows. Reprod. Nutr. Dev. 38:315-330.
Institut National de la Recherche Agronomique INRA. 1998. Ruminant Nutrition:
recommended allowances and feed tables. R. Jarrige. Ed. J. Libbey Eurotext, Paris,
France.
Kleijn, M., and C. G. Proud. 2000. Glucose and amino acids modulate translation factor
activation by growth factors in PC 12 cells. Biochem. J. 347:399-406.
Korhonen, M., A. Vanhatalo, and P. Huhtanen. 2002. Evaluation of isoleucine, leucine,
and valine as a second-limiting amino acid for milk production in dairy cows fed grass
silage diet. J. Dairy Sci. 85:1533-1545.
Lapierre, H., J. P. Blouin, J. F. Bernier, C. K. Reynolds, P. Dubreuil, and G. E. Lobley.
2002. Effect of supply of metabolizable protein on whole body and splanchnic leucine
metabolism in lactating dairy cows. J. Dairy Sci. 85:2631-2641.
Lapierre, H., L. Doepel, E. Milne, and G. E. Lobley. 2005. Effect of lysine (Lys) supply
on its utilization by the mammary gland. J. Anim. Sci. 83 / J. Dairy Sci. 88 Suppl 1: 89.
(Abstr.)
172
Liu, Z., W. Long., D. A. Fryburg., E. J. Barrett. 2006. The régulation of body and
skeletal muscle protein metabolism by hormones and amino acids. J. Nutr. 136:212S-
217S.
Lobley, G. E., E. Milne, J. M. Lovie, P. J. Reeds, and K. Pennie. 1980. Whole body and
tissue protein synthesis in cattle. Br. J. Nutr. 43:491-502.
Lobley, G. E., X. Shen, G. Le, D. M. Bremner, E. Milne, A. G. Calder, S. E. Anderson,
and N. Dennison. 2003. Oxidation of essential amino acids by the ovine gastrointestinal
tract. Br. J. Nutr. 89:617-630.
Lykos, T., and G. A. Varga. 1997. Varying dégradation rates of total nonstructural
carbohydrates: effects on nutrient uptake and utilization by the mammary gland in high
producing Holstein cows. J. Dairy Sci. 80:3356-3367.
Matthews, D. E., H. P. Schwarz, R. D. Yang, K. J. Motil, V. R. Young, and D. M. Bier.
1982. Relationship of plasma leucine and alpha-ketoisocaproate during a L-[l-13C]leucine infusion in man: a method for measuring human intracellular leucine tracer
enrichment. Metab. Clin. Exp. 31:1105-1112.
MacRae, J. C, B. J. Bequette, and L. A Crompton. 2000. Synthesis of milk protein and
opportunities for nutritional manipulation. In Milk composition. Brithish Society of
Animal Science. Occassional Pub. No. 25, pp. 179-199.
Mepham, T. B. 1982. Amino acid utilization by lactating mammary gland. J. Dairy Sci.
65:287-298.
Metcalf, J. A., J. D. Sutton, E. J. Cockburn, D. J, Napper, and D.E. Beever. 1991. The
influence of insuline and amino acid supply on amino acid uptake by the lactating
bovine mammary gland. J. Dairy Sci. 74: 3412-3420.
Miettinen, H. O., P. J. Huhtanen. 1997. Effects of silage fermentation and post-ruminal
casein suplementation in lactating dairy cows: 2-Energy metabolites and plasma amino
acids. J. Sci Food Agric. 74: 459-468.
National Research Council. 2001. Nutrient Requirements of Dairy Cattle. 7th rev. éd.
Natl. Acad. Sci., Washington, DC.
173
Raggio, G., D. Pacheco, R. Berthiaume, G. E. Lobley, D. Pellerin, G. Allard, P.
Dubreuil, and H. Lapierre. 2004. Effect of level of metabolizable protein on splanchnic
flux of amino acids in lactating dairy cows. J. Dairy Sci. 87:3461-3472.
Raggio, G., G.E. Lobley, S. Lemosquet, H. Rulquin, and H. Lapierre. 2006. Effect of
casein and propionate supply on whole body protein metabolism in lactating dairy cows.
Can. J. Anim. Sci. 86:81-89.
Read, W. W., M. A. Read, M. J. Rennie, R. C. Griggs, and D. Halliday. 1984.
Préparation of CO2 from blood and protein-bound amino acid carboxyl groups for
quantification and 13C-isotope measurements. Biomed. Mass Spectrom. 11:348-352.
Reeds, P. J. and T. W. Hutchens. 1994. Protein requirements: from nitrogen balance to
functional impact. J. Nutr. 124:1754S-1764S.
Rigout, S., S. Lemosquet, J. E. van Eys, J. W. Blum, and H. Rulquin. 2002. Duodenal
glucose increases glucose fluxes and lactose synthesis in grass silage-fed dairy cows. J.
Dairy Sci. 85:595-606.
Rigout, S., C. Hurtaud, S. Lemosquet, A. Bach, and H. Rulquin. 2003. Lactational effect
of propionic acid and duodenal glucose in cows. J. Dairy Sci. 86:243-253.
Rulquin, H. and P. M. Pisulewski. 2000. Effects of duodenal infusion of graded
amounts of Leu on mammary uptake and metablism in dairy cows. J. Dairy Sci. 83,
(Suppl.l): 164. (Abstract)
Rulquin, H., S. Rigout, S. Lemosquet, and A. Bach. 2004. Infusion of glucose directs
circulating amino acids to the mammary gland in well-fed dairy cows. J. Dairy Sci.
87:340-349.
SAS® Institute Inc. User's Guide, Version 8. Edition, SAS Ins., Inc. Cary, NC, 1999.
Swaisgood, H. E. 1993. Review and update of casein chemistry. J. Dairy Sci. 76:3054-
3061.
Thivierge, M. C, D. Petitclerc, J. F. Bernier, Y. Couture, and H. Lapierre. 2002.
Variations in mammary metabolism during the natural filling of the udder with milk
over a 12-h period between two milkings. J. Dairy Sci. 85:1839-1854.
174
Vanhatalo, A., T. Varvikko, and P. Huhtanen. 2003a. Effects of casein and glucose on
responses of cows fed diets based on restrictively fermented grass silage. J. Dairy Sci.
86:3260-3270.
Vanhatalo, A., T. Varvikko, and P. Huhtanen. 2003b. Effects of various glucogenic
sources on production and metabolic responses of dairy cows fed grass silage-based
diets. J. Dairy Sci. 86:3249-3259.
Wolfe, R. R. 1984. Tracers in metabolic research. Radioisotope and stable isotope/mass
spectrometro methods. Alan R. Liss, Inc., New York.
Zuur, G., G. E. Lobley, M. Vâzquez-Anôn, and H. Lapierre. 2002. A tracer kinetic
model for metabolism of 2-hydroxy-4-[methylthio]-butanoic acid (HMB) by the
mammary gland of lactating cows. J. Agric. Sci. (Camb) 139:106-107.
175
CHAPITRE VII
DISCUSSION ET CONCLUSION GÉNÉRALE
7.1 Discussion
Ce programme doctoral avait pour objectif de mieux comprendre comment le
métabolisme protéique aux niveaux corporel, splanchnique et mammaire était modulé
par des apports de protéine et/ou d'énergie. Pour cela, nous avons fait varier l'apport
protéique par des rations avec des niveaux différents de protéines métabolisables, ou par
la perfusion de caséine au niveau duodénal. L'apport d'énergie a été modulé par la
perfusion d'acide propionique au niveau ruminai. Ce projet a permis de mieux définir,
dans un premier temps, l'effet des apports protéiques sur les débits nets des AA aux
niveaux splanchnique et mammaire. Par ailleurs, la combinaison de perfusions de
caséine et d'acide propionique avec l'utilisation d'isotopes stables nous a permis
d'approfondir notre connaissance du métabolisme des AA aux niveaux corporel et
mammaire. Il s'agissait également de mieux comprendre les différents mécanismes
d'action par lesquels l'énergie et les protéines augmentent la production de protéines
laitières. Globalement, le projet visait l'objectif général qui est de mieux manipuler les
apports nutritionnels chez la vache laitière afin d'améliorer l'efficacité du transfert de la
protéine alimentaire en protéine laitière.
Deux expériences, lesquelles comportaient deux volets chacune, ont été réalisées
afin de pouvoir répondre aux questions de recherche posées. Dans la première
176
expérience, nous avons étudié l'effet d'apports croissants de protéines métabolisables
sur les débits nets des AA. Cette première expérience a bien démontré que l'efficacité
du transfert des AA digérés au niveau intestinal et convertis en protéines laitières
diminue à mesure que l'on augmente la quantité de protéines métabolisables dans la
ration, tel que proposé par Doepel et al., 2004. Cette diminution de l'efficacité est
directement reliée à une augmentation du prélèvement hépatique d'un groupe d'AA
indispensables (groupe 1) incluant l'histidine, la méthionine, la phénylalanine (et la
tyrosine) et la thréonine et à une augmentation du catabolisme au niveau périphérique
incluant la glande mammaire d'un autre groupe d'AA (groupe 2) composé de la leucine,
l'isoleucine, la valine et de la lysine. Ces résultats expérimentaux confirment qu'un
facteur fixe de conversion des AA absorbés en AA des protéines du lait ne devrait pas
être utilisé tel qu'il l'est actuellement dans les différents modèles de prédiction de
production de protéines du lait. Ce facteur varie selon le bilan nutritionnel de l'animal et
selon chaque AA.
Une extraction aussi importante au niveau hépatique traduit-elle directement un
catabolisme des AA ainsi prélevés ? Comme le foie est un organe où la majorité des
protéines plasmatiques sont produites, une fraction de cette extraction est effectivement
utilisée pour fabriquer des protéines qui seront exportées hors du foie, donc
potentiellement utilisées à des fins anaboliques. Chez des moutons (Connell et al.,
1997), une restriction alimentaire n'a pas réduit le taux de synthèse de ces protéines
d'exportation. Qu'en est-il d'une restriction protéique chez des vaches laitières dont la
demande mammaire pour les AA est très élevée ? Dans notre expérience, nous avons
déterminé pour la première fois chez la vache laitière en utilisant un traceur (2Hs-
phénylalanine) que le taux de synthèse des protéines plasmatiques, en majorité
exportées par le foie, est plus élevée que chez le mouton (1.24 vs. 1.04 g/kg poids vif075 par jour, notre étude vs Connell et al., 1997). Cette synthèse demeure une priorité
métabolique importante car elle n'est pas diminuée par une réduction de l'apport en
protéines, contrairement à la production de matières protéiques. Cette étude confirme le
rôle régulateur du foie qui répond à une réduction d'extraction hépatique de
phénylalanine suite à une atténuation des apports par une diminution de l'oxydation de
phénylalanine afin de maintenir constante la synthèse des protéines d'exportation.
177
Dans la deuxième partie de la thèse, nous avons modifié les apports de protéines
et d'énergie par des perfusions de caséine et/ou d'acide propionique selon des quantités
déjà utilisées dans des travaux antérieurs (Guinard et Rulquin 1994a; Rigout et al.,
2003), pour lesquels une réponse sur la production de protéines du lait avait été
observée. Les objectifs étaient de déterminer comment variaient les débits nets d'AA en
fonction d'apports protéique et/ou énergétique. Il s'agissait de plus d'estimer si les voies
métaboliques connues (synthèse protéique et oxydation) étaient suffisantes pour
expliquer l'augmentation du prélèvement mammaire de certains AA en excès par
rapport à leur sécrétion dans la protéine du lait observé avec des augmentations d'apport
en protéine. Pour répondre à ce dernier objectif, nous avons introduit des traceurs (13C-
Leucine et 13C-bicarbonate) pour mesurer le métabolisme total ainsi que l'oxydation de
la leucine aux niveaux corporel et mammaire.
Au niveau corporel, nous avons observé une augmentation de l'utilisation totale
de la leucine avec les apports des deux types de nutriments, mais à un degré moindre
avec le propionate qu'avec la caséine. Pour la répartition de l'utilisation totale de la
leucine, les deux traitements mettent en lumière les réponses suivantes : les deux
traitements entraînent une augmentation de la synthèse totale de protéines, mais la
caséine montre une réponse supérieure. La différence s'explique par l'oxydation : la
caséine l'augmente (Liu et al., 1995; Bequette et al., 1996a) alors que l'acide
propionique la diminue quand la caséine est présente. Nous en venons donc à la
conclusion qu'il y a des régulations différentes sur le métabolisme de la leucine au
niveau corporel quand la caséine et l'acide propionique sont perfusés, et que l'effet de
l'énergie sur la synthèse totale de protéines au niveau corporel dépend du niveau de
protéines. En complément, nous avons mesuré l'effet de la perfusion de caséine ou
d'acide propionique sur le métabolisme de la leucine au niveau mammaire. Nous avons
ainsi analysé l'hypothèse de largage de peptides par la glande dans le drainage veineux,
hypothèse qui semblait expliquer les prélèvements excédentaires de leucine par la
glande mammaire (Guinard et Rulquin, 1994a; Rulquin et al., 2004) à des apports élevés
de protéines. Nous avons observé une augmentation de l'utilisation totale mammaire de
la leucine pour les deux types de nutriments, reliée principalement à une grande
augmentation de l'oxydation avec la perfusion de caséine. Le suivi minutieux des
178
destins de la leucine (vers son céto-acide, vers l'oxydation et vers les protéines du lait)
nous a permis de mettre en évidence que ces voies métaboliques étaient suffisantes,
dans notre étude, pour tenir en compte les excès de prélèvement mammaire de leucine et
que le largage de peptides par la glande mammaire n'avait pas besoin d'être invoqué.
Nous avons examiné aussi les différences de prélèvements d'AA amenés par
l'apport des différents types de nutriments, mesures individuelles complétées par un
bilan mammaire azoté. Dans tous les cas, la glande mammaire maintient un bilan azoté
provenant des AA à peu près nul, ce qui signifie que dans l'ensemble, la quantité
d'azote apportée par les divers AA apporte suffisamment d'azote pour expliquer la
quantité d'azote retrouvée dans les protéines du lait. Ceci dit, sur une base globale, le
prélèvement des AA sur une base individuelle est quant à lui affecté par les deux types
de nutriments étudiés et varie selon les groupes précédemment décrits par Mepham et
al. (1982). Lors des perfusions de caséine, la glande mammaire maintient son bilan
azoté près de zéro en augmentant principalement les prélèvements des AA des groupes
1 et 2, incluant les AA indispensables, en même temps qu'en augmentant la production
de matières protéiques. Cependant, lors des perfusions d'acide propionique, le bilan
azoté est maintenu à zéro avec une augmentation moindre des productions de protéines
du lait couplées à une augmentation des prélèvements nets des AA des groupes 1 et 3.
Nous en venons donc à la conclusion que l'énergie pourrait diminuer l'utilisation d'AA
non indispensables (principalement du groupe 3) au niveau corporel; de cette façon, il y
aura plus d'AA de ce groupe disponible pour la glande mammaire. Couplée à
l'augmentation du débit sanguin mammaire, cette augmentation du prélèvement des AA
du groupe 3 pourrait permettre une amélioration de l'efficacité de l'utilisation des AA
du groupe 2 pour favoriser l'augmentation de production de protéines du lait avec la
perfusion d'acide propionique sans apport additionnel de protéines.
7.2 Conclusion
Au cours de ce travail de recherche, nous avons étudié le métabolisme protéique
aux niveaux splanchnique et mammaire en réponse à des apports de protéines ou
d'énergie. Nous avons démontré qu'une augmentation de l'apport protéique entraîne
179
une baisse de l'efficacité de transformation des AA absorbés en protéines du lait. Cela
s'explique par :
1- une augmentation de l'oxydation hépatique des acides aminés du
groupe 1 constitué de l'histidine, la méthionine, et la phénylalanine (et
la tyrosine).
2- une augmentation de l'oxydation mammaire par les tissus
périphériques des AA du groupe 2 constitué par la leucine,
l'isoleucine, la valine et la lysine.
L'énergie joue un rôle important dans le métabolisme des AA via :
1- une diminution de l'oxydation de la leucine (et probablement des
autres AA indispensables) au niveau corporel
2- une augmentation de l'extraction mammaire des AA non-
indispensables glucogéniques, possiblement par une épargne
hépatique de ces AA.
Les augmentations de matières protéiques en réponse à des apports de protéine
ou d'énergie s'expliquent par des mécanismes différents, d'où une réponse additive sur
la production de protéines du lait lors de l'ajout simultané de ces deux nutriments.
ISO
7.3 Liste des ouvrages cités
Bequette, B. J., J. A. Metcalf, D. Wray-Cahen, F. R. Backwell, J. D. Sutton, M. A.
Lomax, J. C. MacRae et G. E. Lobley. 1996. Leucine and protein metabolism in the
lactating dairy cow mammary gland: responses to supplemental dietary crude protein
intake. J. Dairy Res. 63:209-222.
Connell, A., A. G. Calder, S. E. Anderson et G. E. Lobley. 1997. Hepatic protein
synthesis in the sheep: effect of intake as monitored by use of stable-isotope-labelled
glycine, leucine and phenylalanine. Br. J. Nutr. 77:255-271.
Doepel, L., D. Pacheco, J. J. Kennelly, M. D. Hanigan, I. F. Lopez et H. Lapierre. 2004.
Milk protein synthesis as a function of amino acid supply. J. Dairy Sci. 87:1279-1297.
Guinard, J et H. Rulquin. 1994. Effect of graded levels of duodenal infusions of casein
on mammary uptake in lactating cows. 2. Individual amino acids. J. Dairy Sci. 77:3304-
3315.
Hanigan, M. D. 2005. Quantitative aspects of ruminant splanchnic metabolism as
related to predicting animal performance. Anim. Sci. 80:23-32.
Liu, S. M., G. E. Lobley, N. A. Macleod, D. J. Kyle, X. B. Chen et E. R. Orskov. 1995.
Effects of long-term protein excess or deficiency on whole-body protein turnover in
sheep nourished by intragastric infusion of nutrients. Br. J. Nutr. 73:829-839.
Mepham, T. B. 1982. Amino acid utilization by lactating mammary gland. J. Dairy Sci.
65:287-298.
Rigout, S., C. Hurtaud, S. Lemosquet, A. Bach et H. Rulquin. 2003. Lactational effect
of propionic acid and duodenal glucose in cows. J. Dairy Sci. 86:243-253.
Rulquin, H., S. Rigout, S. Lemosquet et A. Bach. 2004. Infusion of glucose directs
circulating amino acids to the mammary gland in well-fed dairy cows. J. Dairy Sci.
87:340-349.
181
ANNEXE I
1. Débit net des acides aminés (AA), mmol/h
Débit net AA = [AAV] - [AAA] x DP A OOO
Débit net portai = [AAP ] - [AAA] x DPP/1000
Débit net hépatique = ([AAH ] - [AAP] x DPP ) + ([AAH ] - [AAA] x DPAH)
/1000
Débit net splanchnique = ([AAH ] - [AAA]) x DPH/1000
Débit net mammaire = [AAM] - [AAA] x DPM/1000
où [AA] représente la concentration plasmatique des AA (uM), les vaisseaux sanguins
sont représentés comme suit : A = artère, V= veine, P = veine porte, H = veine
hépatique, M = veine mammaire, et DP indique le débit plasmatique (1/h) au niveau de
chacun des vaisseaux, DPAH étant le débit plasmatique de l'artère hépatique. Le débit
splanchnique représente la somme des débits des viscères drainés par la veine porte
(système digestif, rate, pancréas et gras mésentérique) et du foie.
2. Mouvement dynamique
2.1 Utilisation corporelle
L'utilisation corporelle totale d'un AA (UT-AA) et la production totale de
dioxyde de carbone (CER) ont été calculés ainsi :
1) UT- AA or CER (mmol/h) = (EIinf / EIA - 1) x Inf
où EIinf et EIA représentent respectivement l'enrichissement isotopique (El : mpe) du
traceur perfusé et celui du pool précurseur mesuré au niveau artériel et Inf, le taux de
perfusion du traceur.
182
L'oxydation d'un AA (Ox) au niveau corporel a été calculé ainsi;
2) Ox-Leu (mmol/h) = (CER x EIa-13CO2 /EI-Leu) - Inf
La synthèse de protéines (SP) au niveau corporel a été calcule comme suit :
3) SP (mmol/h) = UT-AA - Ox-Leu
Pour le calcul de synthèses protéines au niveau corporel, nous avons utilisé
comme référence 63 g de leucine par kg de protéines corporelles synthétisées (Lobley et
al., 1980) et 98 g de leucine par kg de protéine laitière, en accord avec nos propres
analyses et la littérature (Swaisgood, 1995).
2.2 Utilisation au niveau d'un organe
4) UT-AA (mmol/h) - (([AAA] x EIAA-A ) - ([AAV] x EIAA-V)) x DP / El AA.V
/1000
où EIAA et [AA] représentent respectivement l'enrichissement isotopique (El : mpe) du
traceur perfusé et la concentration plasmatique de l'AA étudié (uM), soit au niveau
artériel (A) ou veineux (V) et DP, le débit plasmatique de l'organe étudié.
2.3 Glande mammaire
Le métabolisme dynamique de la leucine dans la glande mammaire a été étudié
selon un modèle à trois compartiments (Bequette et al., 2002). L'utilisation totale au
niveau de la glande mammaire (GM) est calculée tel que décrit dans la section
précédente, soit :
5) GM UT-Leu (mmol/h) = ((([LeuA] xEILeu-A)- ([LeuM] xEILeu-M)) x DPM /
IEMOP-M) -CorOx
le céto-acide de la leucine, le 4-methyl 2-oxopentanoate (MOP), formé
intracellulairement et relargué au niveau du plasma veineux étant utilisé comme
représentatif du pool précurseur. Le facteur CorOx corrige pour tenir en compte
l'oxydation supplémentaire au niveau mammaire résultant de la perfusion du traceur.
183
Le MOP transformé en Leucine et relargué dans la veine mammaire (M) a été
calculé comme suit :
6) MOP vers Leu ou CO2 (mmol/h) = (([MOPA] X EI M O P-A - [MOP-M] XEIMOP-M)
x DPM) / (IEMOP-M)
où [Leu] et [MOP] représentent respectivement les concentrations de la leucine et du
MOP et El les enrichissements du MOP.
2.3.1 L'oxydation au niveau mammaire
Le débit sanguin (DS) a été utilisé pour calculer le flux de CO2
7) DS (1/h) = DP / (1- HématocriteA)
où hématocriteA représente l'hématocrite mesuré au niveau artériel (A)
Le flux de 13CO2 au niveau de la glande mammaire (GM) a été calculé :
8) GM13CO2 (mmol/h) = (([CO2M] x EI1 3C02 M j- ([CO2A] ><EI13CO2A )) x DSM
où [CO2] représente les concentrations et El les enrichissements du CO2 et 13CO2
respectivement, et A et M pour la artère et la veine mammaire.
La séquestration (Sq) de 13CO2 a été estimée par le flux artériel de 13CO2 non récupéré
dans la veine mammaire pendant la perfusion de bicarbonate. À partir de la Sq, nous
avons fait une correction sur le flux total de 13CO2 de la glande mammaire mesuré
pendant la perfusion de leucine marquée.
9) GM 13CO2 corrige pour la Sq (mmol/h) = (flux artériel de 13CO2 x Sq) + GM13CO2
L'oxydation de la leucine par la GM a été finalement calculée comme suit :
10) GM Leu Ox (mmol/h) = (GM 13CO2 corrigée pour la Sq /EIMOP-M) - CorOx
La synthèse de protéines (SP) au niveau de la glande mammaire :
184
ll)SP(mmol/h)= GM UT-Leu - MOP - GM Leu Ox
Pour transformer l'utilisation de la leucine pour la synthèse protéique (mmol/h)
en protéines totales synthétisées (g/j), nous avons utilisé comme référence 63 g de
leucine par kg de protéines synthétisées (Lobley et al., 1980), et 98 g de leucine par kg
de protéines dans le lait produit (Swaisgood, 1995).
2.4 Synthèse des protéines plasmatiques
Les taux de synthèse relative de l'albumine (Alb) et des protéines plasmatiques
totales (Prot Tôt) (FSR; %/jour d'Alb ou Prot Tôt) ont été calculés ainsi :
12) FSR ( %/jour) = pente (Alb ou Prot Tôt) / EIH * 24 * 100.
où la pente de l'EI de phénylalanine marquée (d5-Phe) dans l'albumine ou les protéines
totales plasmatiques en fonction du temps de la perfusion de d5-Phe (entre 3 et 8
heures) a été estimée pour chaque vache.
La valeur absolue de la synthèse totale des protéines plasmatiques totales ou
d'albumine (ASR; g/jour d'Alb ou Prot Tôt) a été calculée :
13) ASR (g/jour) = FSR (Alb ou Prot Tôt) * [Alb ou Prot Tôt]* VP
où VP représente le volume plasmatique, calculé comme étant 49.9 ml/kg poids
corporel (Girad al., 1989) et [Alb ou Prot Tôt] représente la concentration d'albumine
ou de protéines plasmatiques totales. Pour la conversion de ASR (g/j : Alb ou Prot Tôt)
en phénylalanine utilisée (mmol/h), nous avons utilisé 5 % de phénylalanine dans les
protéines plasmatiques totales et 6 % de phénylalanine dans l'albumine (Peters et al.,
1985).
185
La proportion maximum de phénylalanine extraite par le foie utilisée pour la synthèse
de protéines plasmatiques totales ou d'albumine a été calcule ainsi :
14) Phe Alb or Prot Tôt ( %) = Phe utilisée pour Alb or Prot Tôt / extraction
nette hépatique de Phe * 100
La phénylalanine oxydée par le foie (mmol/h) a été calculée par la différence
entre l'extraction nette hépatique de phénylalanine moins la quantité utilisée pour les
protéines d'exportation. Ainsi la phénylalanine utilisée pour la synthèse protéique
hépatique a été calculée ainsi :
15) SP Phe hépatique (mmol/h) = UT-Phe hépatique - Ox Phe
La conversion de l'utilisation de la phénylalanine pour la synthèse protéique
hépatique (mmol/h) en synthèse de protéines (g/d) a été basée sur une concentration de
36 g de phénylalanine par kg protéines constitutives du foie (Connel et al., 1997).
2.5 Liste de références
Connell, A., A. G. Calder, S. E. Anderson, and G. E. Lobley. 1997. Hepatic protein
synthesis in the sheep: effect of intake as monitored by use of stable-isotope-labelled
glycine, leucine and phénylalanine. Br. J. Nutr. 77:255-271.
Peters, T. Jr. 1985. Sérum Albumin. Adv. Protein. Chem. 37: 161-245.
Swaisgood, H. E. 1993. Review and update of casein chemistry. J. Dairy Sci. 76:3054-
3061.
Lobley, G. E., E. Milne, J.M. Lovie, J. M., P. J. Reeds, and K. Pennie. 1980. Whole
body and tissue protein synthesis in cattle. Br. J. Nutr. 43:491-502.