genetic improvement of dairy cow reproductive performance

Genetic Improvement of Dairy Cow Reproductive Performance

B Berglund

Department of Animal Breeding and Genetics, Swedish University of Agricultural Sciences and Centre for Reproductive Biology in Uppsala, Uppsala,Sweden

Contents

The welfare of cow along with profitability in production areimportant issues in sustainable animal breeding programmes.Along with an intense ⁄ intensive selection for increased milkyield, reproductive performance has declined in many coun-tries, in part due to an unfavourable genetic relationship. Thelargely unchanged genetic trend in female fertility and calvingtraits for Scandinavian Red breeds shows that it is possible toavoid deterioration in these traits if they are properly consid-ered in the breeding programme. Today’s breeding is interna-tional with a global selection and extensive use of the best bulls.The Nordic countries have traditionally recorded and per-formed genetic evaluation for a broad range of functional traitsincluding reproduction. In recent years many other countrieshave also implemented genetic evaluation for these traits. Thus,the relative emphasis of dairy cattle breeding objectives hasgradually shifted from production to functional traits such asreproduction. Improved ways of recording traits, e.g. physio-logical measures, early indicator traits, assisted reproductivetechniques and increased knowledge of genes and their regu-lation may improve the genetic selection strategies and havelarge impact on present and future genetic evaluation pro-grammes. Extensive data bases with phenotypic recordings oftraits for individuals and their pedigree are a prerequisite.Quantitative trait loci have been associated to the reproductivecomplex. Most important traits, including reproduction traitsare regulated by a multitude of genes and environmental factorsin a complex relationship, however. Genomic selection mighttherefore be important in future breeding programmes. Infor-mation on single nucleotide polymorphism has already beenintroduced in the selection programmes of some countries.

Introduction

There are strong motives for including reproduction ingenetic selection programmes. A good reproductiveperformance is crucial for economic as well as ethicalreasons. Without reproduction there will be no animalproduction. The unfavourable genetic correlation withmilk production has led to a decline in reproduction indairy cattle, at least in part due to an insufficientconsideration of this trait when selecting for a highermilk production. There are several reports that theincreasing use of Holstein genetics has caused decliningfertility as well as calving performance during the latestdecades (e.g. Berglund and Philipsson 1992; Royal et al.2000; Lucy 2001; Hansen et al. 2004). This has causeddeterioration in fertility in countries that heavily used thisbreed even if they have had reproductive performanceboth in the breeding goal and in the selection criteria. Forinstance, in Swedish Holsteins (SH), daughter fertilityhas fallen by approximately 1.2 index units per year overthe last fifteen years until now (Lindhe B, 2007: SvenskAvel, Skara, Sweden, personal communication), whereasthe Swedish Red breed (SRB), with similar milk yield

levels and genetic trend, has largely maintained fertility.A national breeding programme considering reproduc-tion traits for more than three decades may explain this.However for Holsteins the inclusion of fertility in theSwedish breeding goal has not been enough to withstandthe large imports of genetic material from countries thathave low, or no weighting on fertility in their breedingobjective.

In 2006 the World Holstein Friesian Federationinitiated a survey of the status on fertility in the Holsteinpopulation around the world. The conclusion from thissurvey was that fertility is a problem and actions need tobe taken both internationally and within each country(Sørensen et al. 2007). Many countries have imple-mented genetic evaluation for fertility traits in recentyears. More traits are gradually being evaluated andmore sophisticated evaluation methods are being imple-mented. Maybe the decline in fertility now has levelledout and reached a plateau, as was concluded from aninternational conference on fertility in dairy cows held inLiverpool, 2007 (Crowe 2007, personal communication).Still the low level of reproductive performance is aproblem. Reproduction problems are among the mostcommon reason for culling in dairy production. This isthe major cause for involuntary culling, e.g. in Swedishdairy cattle (Swedish Dairy Association 2007).

Today’s breeding is international, intensive and usesmodern reproductive and molecular genetic techniques.The welfare of cows along with profitability in produc-tion are important issues in sustainable animal breedingprogrammes. Nielsen et al. (2006) suggested that whendefining breeding goals for sustainable production,breeding organizations should predict the selectionresponse based on market economic value and addnon-market value for traits with unacceptable selectionresponses. New ways of measuring, recording andanalysing traits, new reproductive techniques andincreased knowledge of genes and their regulation mayimprove the genetic selection strategies and have a largeimpact on genetic evaluation programmes. This paperwill focus on the genetic improvement of the femalefertility and calving traits.

Difficulties in Selection for Reproduction

Many reproduction traits are difficult to handle inparameter estimation and genetic evaluation. In generalheritabilities are low, usually less than 5%, mainly due toa large influence of management and environmentaleffects. Most traits are not normally distributed and havecensored records, which complicates the analysis of thetraits. In addition to the other features of the reproductive

Reprod Dom Anim 43 (Suppl. 2), 89–95 (2008); doi: 10.1111/j.1439-0531.2008.01147.x

ISSN 0936-6768

� 2008 The Author. Journal compilation � 2008 Blackwell Verlag

traits calving performance traits are influenced by aneffect of the mother and a direct effect of the young(Steinbock et al. 2003; Hansen et al. 2004) and it may bedifficult to correctly estimate these (co)variances. To getselection response in reproduction traits a well developedrecording scheme is needed and a breeding structureallowing large daughter group sizes.

Several sub-traits

Reproduction, especially female fertility, is a complextrait composed of several sub-traits. Thus it is importantto include all important aspects of fertility to achieve agood and expected selection response. A challenge indoing this is the data collection and the quality of data.Traditionally, most fertility traits are based on calvingand insemination data and each trait has its strengthsand weaknesses. For female fertility there are measuresreflecting the ability to resume oestrous cycles aftercalving and the ability to conceive, or measurescombining these abilities like calving to last insemina-tion (CLI), often also called days open (DO). Informa-tion about treatments and culling for reproductivedisorders are recorded and used mainly in the Nordiccountries. In Sweden scores for heat symptoms arerecorded as well. Heat detection plays a considerablerole for the economy. A visible manifestation of oestrusand a high oestrus detection rate is important especiallywhen using AI and in countries where hormones foroestrus synchronization are not generally used. Dobsonet al. (2007) reported from the UK that the number ofoestrous animals standing-to-be-mounted has declinedfrom 80% to 50% over the past 30–50 years.

Antagonistic genetic correlations to other traits

The antagonistic relationship between fertility and milkproduction is well-known and was shown already byJanson and Andreasson (1981). These correlations wereconfirmed in studies by Roxstrom et al. (2001b), whereunfavourable genetic correlation for, e.g. number ofinseminations per service period, interval from calvingto first AI and treatments for reproductive disorders tomilk production was shown. The correlations rangedfrom 0.2 to 0.4, increasing with lactation number,possibly as a consequence of a higher energy demandwith increasing production level and steeper lactationcurves. Due to the antagonistic genetic correlation tomilk production, undesirable trends are expected forthe reproductive traits if they are not included in thebreeding goal. Even if included in the breeding objectivethere is a risk for deterioration of this trait due to thelow heritability if too small a breeding goal weight ordaughter groups are used. Ethic values (cow welfare,consumer preference, etc.) should be put into thebreeding goal weight and those are not easily accessible.Moreover, there are unfavourable correlations amongthe reproduction traits that are important to consider.

Negative energy balance and reproduction

A top producing cow might not cope with the energyrequirements for both milk production and maintained

reproduction and health even though the cows have acapacity to mobilize from their energy reserves. In ourstudies of Swedish Red (SRB) and Swedish Holsteins(SH) we have seen that SH has a thinner layer ofsubcutaneous fat than SRB which may be important inthis aspect (Hjerten 2006). A negative energy balance isassociated with inferior embryo quality. A review paperon metabolic changes and embryo quality was given byChagas et al. (2007). Hayhurst et al. (2007a) found asignificant genetic variation in embryo quality andestimated a heritability of 0.13 for this trait, implyinga possibility of genetically selecting cattle with inherentquality to produce high quality embryos.

Body condition score (BCS) is an internationallyaccepted, rapid, inexpensive and non-invasive method ofestimating body energy reserves in dairy cattle (Berryet al. 2007). Even though mainly used for managementpurposes, BCS may also add to the information ingenetic evaluations, especially when data on more directtraits are lacking. BCS is moderately heritable (0.09–0.45) and favourably correlated to fertility and survival.BCS is used as a predictor trait for genetic merit forfertility in some countries, e.g. the Netherlands (De Jong2005) and in Irish and UK cattle (Berry et al. 2007).

Current Status in Breeding Programmes

The general breeding goal for the reproduction of cows israther similar in different countries and may be formu-lated as follows: cows that return to normal cyclicity earlyafter calving show strong and regular heats, and whichconceive when inseminated at a correct time. Further-more, the cow should carry her pregnancy to term, have agood calving ability and give birth to viable calves.

Female fertility

Early studies by Janson (1980) showed that even thoughheritability for fertility traits was low, the additivegenetic variation was shown to be substantial. Further-more, the importance of including both interval andpregnancy measures at different ages in the geneticevaluation for fertility was underlined. These resultswere later confirmed in large field studies by Roxstromet al. (2001a,b). The Nordic countries have traditionallyrecorded and performed genetic evaluation for a broadrange of functional traits including reproduction, but inrecent years many other countries have also imple-mented genetic evaluation for these traits. Thus, therelative emphasis of dairy cattle breeding objectives hasgradually shifted from production to functional traitssuch as reproduction during the past couple of decades(Miglior et al. 2005).

Breeding values for daughter fertility were introducedin Sweden as early as 1972, and have since been used inselection. In the Nordic countries, the integration of cowdata bases (pedigree, milk recording, AI and diseasedata) has facilitated selection for reproductive traits.Three of the Nordic (Denmark, Finland and Sweden)breeding organizations have had a joint genetic evalua-tion and breeding programme since 2005 (for moreinformation see http://www.nordicebv.info). The Nordicfertility index includes the traits number of AI per service

90 B Berglund


period for heifers (h) and cows (c), interval from calvingto first AI (c), interval from first to last AI (h, c) andreproductive treatments (c). Since the mid 1990s geneticevaluation for fertility has gradually been introduced inseveral countries, e.g. the Netherlands (Hoekstra et al.1994), UK (Wall et al. 2003) and US (VanRaden et al.2004). An international genetic evaluation for fertilitytraits was introduced for Holstein populations inFebruary 2007 by Interbull and an evaluation of fertilityin the other main breeds of the Interbull membercountries is under development (Jorjani 2007) (for moreinformation see http://www.interbull.org).

Genetic trend for fertility

Lindhe and Philipsson (2001) found a clear unfavour-able genetic trend in female fertility for SwedishHolsteins and a slightly favourable genetic trend forthe Swedish Red breed. In the Norwegian Red breed notrend or a slightly favourable genetic trend for fertilitytraits was found (Chang et al. 2006). Traditional breed-ing strategies have been very successful in selecting highyielding dairy cows (Fig. 1a,b). While functional traitssuch as reproductive performance have declined in manycountries, especially for Holsteins, the largelyunchanged genetic trend in female fertility and calvingtraits for Danish, Finnish and Swedish Red (Ayrshire

type) breeds (Fig. 1a,b) shows that it is possible to avoida deterioration in these traits if they are properlyconsidered in the breeding programme. While forDanish, Swedish and Finnish Holsteins the breedingprogramme could not fully compensate for the use offoreign bull father genetic material for which functionaltraits such as reproduction were not known.

Calving performance traits

For first-parity Holstein cows, calf mortality is a greatproblem (Berglund and Philipsson 1992) and nowadaysreports are many of their high stillbirth rates, 10–13% atfirst calving. Stillbirths are approximately twice as high forHolstein first-calvers as for Swedish Red (Fig. 2). Theincreasing trend for stillbirth in Holsteins has not beenaccompanied by an increase in calving difficulty and thedivergent phenotypic trends may be an illustration of avitalityproblem.For second-calvers hardlyanydifferencesin calving difficulty and stillbirth exist between the breeds.

In a post-mortem examination of 76 calves from first-calving Holsteins, one-third of the calves were bornwithout any visible defects or injuries (Berglund et al.2003) and only half of the calves were born with signs of adifficult calving, indicating a vitality problem. The geneticcorrelations for stillbirth at first and second calving were0.45–0.48 for Swedish Holsteins (Steinbock et al. 2003),but 0.83–0.85 for the Swedish Red breed (Steinbock et al.2006). Thus there seem to be a genetic difference invitality of calves between these breeds at first calving.There are probably multifactorial reasons behindincreasing stillbirth rates. Adamec et al. (2005) recentlyshowed a consistently unfavourable effect of inbreedingon calving difficulty and stillbirth in US Holsteins, withlargest effects for first parity births. McParland et al.(2007) found that inbreeding had a deleterious effectupon most of the traits studied such as dystocia, stillbirthand calving interval in Irish Holstein-Friesians. A max-imized genetic gain should be balanced against a mini-mized inbreeding. There are now programmes to be builtinto the genetic evaluation programmes that maximizegenetic gain while minimizing inbreeding.

Because calving difficulty and stillbirth mainly is aproblem for first parity cows, this category should be themain source of information for genetic evaluation. Forcertainbreeds suchas theSwedishRedbreed the reliabilityin breeding values for calving ability could be increasedby including higher calving numbers (Steinbock et al.2006). Genetic evaluations should consider both calvingdifficulty and stillbirth as calf and maternal traits.

Calving performance has been recorded in Swedensince the 1960s. The number of countries recordingcalving traits is increasing (Mark et al. 2005), and aninternational genetic evaluation was introduced in 2005(Jacobsen and Fikse 2005). Many of the member coun-tries of Interbull genetically evaluate calving difficulty andmost of them now also evaluate bulls for stillbirth, e.g. aroutine evaluation for stillbirth in Holsteins was imple-mented in the US in 2006 (Cole et al. 2007). Difficultcalving and stillbirth reduce reproductive performance(Bicalho et al. 2007), and genes associated with difficultbirth also reduce reproductive success (Lopez de Matur-ana et al. 2007).

75

85

95

105

115

DF

S/S

WE

sca

leD

FS

/SW

E s

cale

MI FA PR SC LO CECF DOCM

75

85

95

105

115(a)

(b)

1985 1990 1995 2000Years

1985 1990 1995 2000Years

MI

FA

PR

SC

LO

CE

CF

DO

CM

Fig. 1. (a) Genetic trends for functional traits and milk productiontraits for Red (Ayrshire type) breeds. DFS ⁄ SWE, Danish, Finnish andSwedish Red breeds on a Swedish scale for breeding values (Jorjani2007, personal communication); MI, milk yield; FA, fat yield; PR,protein yield; SC, somatic cell scores; LO, longevity; CE, direct calvingease; CF, calving to first insemination; DO, days open; CM, clinicalmastitis. (b) Genetic trends for functional traits and milk productiontraits for Holsteins

Genetic Improvement of Dairy Cow Reproductive Performance 91


Genetic trend for calving traits

Mark et al. (2005) reported a slightly negative genetictrend from the Interbull genetic evaluations for maternalstillbirth in the Holstein breed since the early 1990swhile there was no clear trend in the other calving traits.For the Swedish and Norwegian Red breeds, no orslightly positive genetic trends for calving performancehave been observed (Heringstad et al. 2007; Lindhe,2007, personal communication). The genetic trends fordirect calving ease for Danish, Finnish and Swedish Redbreed and Holsteins can be seen as one of the functionaltraits in Fig. 1a,b.

Genetic defects causing reproduction loss

Mutations in single genes occur regularly and may causecongenital genetic defects. These are commonly reces-sively inherited (http://www.omia.angis.org.au). Onerecent example is the complex vertebral malformation(CVM) in theHolstein breed, first described byAgerholmet al. (2001). Complex vertebral malformation has had amajor impact on the reproductive performance inHolsteins (Agerholm 2007). In a study of SwedishHolsteins, carriers of the CVM-gene had inferior NRrate compared with non-carrier bulls reflecting a higherintra-uterine mortality (Berglund et al. 2004). It isimportant to report all kinds of malformations and tohave national control programmes for congenital defectsin order to avoid multiplication of deleterious genescausing reproductive losses and animal welfare problems.

Crossbreeding as a tool to enhance fertility

Decreasing fertility and increasing calf mortality inHolsteins have become a large problem. In some coun-tries, e.g. in the USA, Scandinavian Red and other fertilebreeds are now crossed into Holsteins to improve theHolstein reproduction. Crossbreeding may help elevatethe level of traits combining breeds with favourable traitsand by exploring heterosis effects, but the continuousgenetic improvement has to be done in the pure breeds.Heins et al. (2006a) reported that calves from Holstein

first-calf heifers sired by Scandinavian Red (NorwegianRed and Swedish Red) bulls had significantly less calvingdifficulty (5.5%) and lower stillbirth rate (7.7%) thancalves sired by Holstein bulls (16.4% difficult calving and15.1% stillbirth). Moreover, DO were significantlyshorter for all crossbred groups studied and were129 days for Scandinavian Red ⁄Holstein crossbredsand 150 days for pure Holsteins (Heins et al. 2006b).

New Tools for Genetic Improvement ofReproduction

Improved ways of recording traits, e.g. direct measuringof physiological measures may offer valuable opportu-nities to improve the genetic evaluation of fertility.Indicator traits are valuable because fertility has a lowheritability and is expressed late in life. More advancedand expensive recording technologies may be used innucleus herds and the genetic progress may be enhancedby using modern reproductive techniques such asMOET. Reproductive techniques like sexing, cloningand transfer of genetic material will impact present andfuture selection strategies in breeding programmes.

The amount of information on the molecular level israpidly increasing. The sequencing of the bovine genomewas completed in 2003 (http://www.hgsc.bcm.tmc.edu/projects/bovine). Markers are connected to the pheno-typic expression of several traits in the reproductioncomplex. Extensive data bases with phenotypic record-ings of traits for individuals and their pedigree are aprerequisite. Gene expression profiles may increase ourunderstanding of the mechanisms behind reproductivefunctions and their phenotypic expression. As genes fortraits are identified, genetic selection strategies can beimproved. Most important traits, including reproductiontraits are regulated by a multitude of genes and environ-mental factors in a complex relationship, however.

Progesterone-based measures of female fertility

Early measures of the reproductive function of cow canbe obtained from progesterone profiles. Progesterone-

0

2

4

6

8

10

12

1986 1989 1992 1995 1998 2001 2004 Years

Per

cen

t

Stillbirths SH Calving diff. SH Stillbirths SRB Calving diff. SRB

Fig. 2. Phenotypic trends in stillbirth rate and calving difficulty for first calvers of SLB ⁄ Swedish Holsteins (SH) and Swedish Red (SRB) (Swedishmilk recording statistics)

92 B Berglund


based measures of fertility have higher heritability thanthe traditional measures of fertility (Royal et al. 2002;Petersson et al. 2007). Veerkamp et al. (1998) suggestedthat selection on days to first luteal activity (CLA) basedon monthly analysis of progesterone in milk may addfurther accuracy to the genetic evaluation for fertility.Van der Lende et al. (2004) suggested to select siresbased on a measure when 50% of the daughters of a sirehad an active corpus luteum (CLA 50%) based on 3- to6-week intervals of progesterone sampling. Peterssonet al. (2007) found that direct selection on progesteronebased measures of fertility may increase the accuracy inthe genetic evaluation for an early start of cyclic ovarianactivity after calving compared to the commonly usedmeasure CFI, even with an infrequent sampling such asin the regular milk recording system. Progesteroneanalysis in the first monthly collected milk samplescould also be used as a management tool. Peterssonet al. (2008) showed that four out of five cows withdelayed cyclicity could be predicted within 60 days aftercalving enabling an earlier treatment. The cost forprogesterone analysis of milk samples could thereforebenefit from management returns as well as fromimprovements in genetic gain for fertility.

Automization of recordings

Automatic milking systems, e.g. robotic milking systemsmay allow automization of recordings. Commercialsystems for on-line recordings allowing herd-profiles offertility (e.g. based on progesterone analysis) and healthparameters are underway (Friggens and Løvendahl2007). These may offer recordings with a high accuracythat could be built into selection programmes. Løven-dahl and Chagunda (2006) used activity meters forautomatic heat controls and estimated a heritability of0.17 for this trait.

Juvenile predictors

The genes controlling fertility are present and potentiallyexpressed early in life, but it takes at least four yearsbefore a bull has milking daughters and can be progenytested for fertility. Thus early fertility predictors would bevery valuable. These could be reproductive hormones ormetabolic traits. Hayhurst et al. (2007b) suggested apossibility of using the correlation between the pre-pubertal response to gonadotrophin releasing hormone inbull calves and the fertility of their daughters as a possibleselection tool. In another study by Hayhurst et al.(2007c), it was suggested that selection for bull calveswith lower concentrations of glucose and FFA couldresult in female offspring with genetically better fertility.

Gene mapping studies

The identification of quantitative trait loci (QTL) is afirst step towards novel selection methods based on bothphenotypic and molecular information. Using QTL inselection is most beneficial for low heritability traits, sexlimited traits and traits expressed late in life such asdaughter fertility. Holmberg and Andersson-Eklund(2006) found regions with several QTLs on chromosome

9 and 11 for both reproduction and health traits. A QTLfor non-return rate was fine mapped to an interval ofless than 3 cM on chromosome 9 (Holmberg et al.2007). Marker assisted selection (MAS) can be used forpre-selection among full-sibs before progeny testing byaccounting for the Mendelian sampling and also toavoid genetic defects for which there are availablemarkers. For quantitative traits the benefit from MAS islimited by the proportion of the genetic varianceexplained by known QTL. The marker density in theQTL region can be increased by use of single nucleotidepolymorphism (SNP) markers. Single nucleotide poly-morphisms are single base-pair differences betweenindividuals within a species and where the differentvariants (most often only two allelic forms) are relativelycommon in a population.

Expression profiles of genes regulating reproduction

Gene expression profiling is a relatively recent toolcontributing to our knowledge about the various pro-cesses underlying reproduction such as the function ofgenes and their products determining the phenotype forreproduction. Beerda and Veerkamp (2006) summarizedgene expression studies related to reproduction in cattle,sheep and swine in a paper at the World Congress onGenetics Applied to Livestock Production in Brazil.They concluded that a vast amount of information hasalready been achieved in this area and as informationwill grow exponentially over the next few years theyunderlined the importance of a major effort in bringingall information together in a ‘broadly accessible ontol-ogy’. The need for compiling and analysing largeamounts of molecular data has created a new field ofscience referred to as bioinformatics. An EU integratedproject SABRE was started in 2006, which aims toprovide fundamental knowledge on the genomics andepigenetics on, e.g. reproduction traits in dairy cattle tobe used in selection for improved reproduction effi-ciency. An update on studies on expression profiles ofgenes regulating dairy cow fertility was given at theInternational conference on fertility in dairy cows inLiverpool, 2007 (Beerda and Veerkamp 2007).

Genomic selection and integration of molecular data intogenetic evaluation programmes

Genomics is the study of variation of base pairs in thenucleic acids and genomic selection means using thisinformation in genetic selection programmes. The avail-ability of large arrays of SNPs is changing the approachof predicting breeding values from molecular informa-tion. Genomic selection (GMAS) uses all markers, orrather haplotypes consisting of a pair of contiguousSNPs, spanning the genome for prediction of breedingvalues. Thus, by summing the effects of all markerhaplotypes in the genome of a bull calf, a breeding valueis obtained directly at birth whereby the generationinterval can be considerably shortened. This informationcan be used directly, hence theoretically no progenytesting is needed. Muir (2007) showed that by usingGMAS for traits of high (0.5) or low heritability (0.1) theaccuracy of selection increased between 10% and 30%.



Van Raden (2007) calculated measures of relationshipand inbreeding by use of genomic measures and demon-strated gains in reliability as compared to using tradi-tional methods, and comparatively more for lowheritability traits than high heritability traits. A disad-vantage is that the information may only be used forsome generations, maybe 7–8 generations, because theestimated effects of SNP haplotypes will change overtime. Furthermore, without progeny testing in eachgeneration the risk for late detection of undesired side-effects of selection might increase.

Along with increasing knowledge at the genomic level,effort is put on how to integrate these data in the geneticevaluation systems. Gengler and Verkenne (2007) statedthat it will be a challenge to integrate molecular andphenotypic data and that SNPs and genomic selectionwill not change this situation.

Genomic selection based on 3000 SNP markers toselect young bulls was introduced in the Netherlands inOctober 2006 (Van der Beek 2007). Denmark andNorway have also recently introduced SNP informationin their genetic selection programmes (MS Lund andS Lien, personal communication). Large scale projectsare ongoing in several livestock species to identify andvalidate several thousands of SNPs in haplotype blocksin the genome. SNP chips with 60 000 SNPs will soonbe available.

Concluding Remarks

The genetic trend for functional traits in the Red(Ayrshire) breeds shows that it is possible to maintaina good dairy cow fertility and calving performancealong with selection for increased levels of milk pro-duction, if the traits are properly considered. It ispossible to make improvements in breeding, as well as innutrition and management to keep a good reproductiveability of the cow and a good health and well-being insustainable breeding programmes. Along with increas-ing knowledge at a molecular level, large changes maybe expected in genetic evaluation programmes.

Acknowledgements

I would like to acknowledge Professor Erling Strandberg at theDepartment of Animal Breeding and Genetics, SLU, for valuablecomments on the manuscript. Associate Professor Hossein Jorjani atthe Interbull office (Jorjani 2007, communication), SLU, for kindlyproviding Interbull genetic trends and Associate Professor BengtLindhe at Svensk Avel (Lindhe 2007, communication) for kindlyproviding national genetic trends.

ReferencesAdamec V, Cassell BG, Smith EP, Pearson RE, 2005: Effectsof inbreeding in the dam on dystocia and stillbirths in USHolsteins. J Dairy Sci 89, 307–314.

Agerholm JS, 2007: Complex vertebral malformation inHolstein cattle: the story so far. Perinatal death in domesticanimals. Proc 20th NKVet Symp, Reykjavik, Iceland, p. 88.

Agerholm JS, Bendixen C, Andersen O, Arnbjerg J, 2001:Complex vertebral malformation in Holstein calves. J VetDiagn Invest 13, 283–289.

Beerda B, Veerkamp RF, 2006: Functional genomics of femalereproduction. 8th World Congress on Genetics Applied to

Livestock Production, Belo Horizonte, Brazil. Proc Com-mun 11–10 (http://www.wcgalp8.org.br).

Beerda B, Veerkamp RF, 2007: Expression profiles of genesregulating dairy cow fertility: recent findings, ongoingactivities and future possibilities. Proc Fertility in DairyCows – bridging the gaps, 30–31 August 2007, LiverpoolHope University, Liverpool, UK, p. 33.

Berglund B, Philipsson J, 1992: Increasing stillbirth rates in theSwedish Friesian population. Paper in 43rd Annual Meetingof the European Association on Animal Production,Madrid, 14–17 September 1992.

Berglund B, Steinbock L, Elvander M, 2003: Causes ofstillbirth and time of death in Swedish Holstein calvespost-mortem examined. Acta Vet Scand 44, 111–120.

Berglund B, Persson A, Stalhammar H, 2004: Effects ofcomplex vertebral malformation on fertility in SwedishHolstein cattle. Acta Vet Scand 45, 161–165.

Berry D, Roche J, Coffey M, 2007: Body condition score andfertility – more than just a feeling. Proc Fertility in DairyCows – bridging the gaps, 30–31 August 2007, LiverpoolHope University, Liverpool, UK, p. 28.

Bicalho RC, Galvao KN, Cheong SH, Gilbert RO, WarnikLD, Guard CL, 2007: Effect of stillbirths on dam survivaland reproduction performance in Holstein dairy cows.J Dairy Sci 90, 2797–2803.

Chagas LM, Bass JJ, Blache D, Burke CR, Kay JK, LindsayDR, Lucy MC, Martin GB, Meier S, Rhodes FM, RocheJR, Thatcher WW, Webb R, 2007: Invited review: Newperspectives on the roles of nutrition and metabolic prior-ities in the subfertility of high-producing dairy cows. J DairySci 90, 4022–4032.

Chang YM, Andersen-Ranberg IM, Heringstad B, Gianola D,Klemetsdal G, 2006: Bivariate analysis of number of servicesto conception and days open in Norwegian Red using acensored threshold-linear model. J Dairy Sci 89, 772–778.

Cole JB, Wiggans GR, VanRaden PM, Miller RH, 2007:Stillbirth (co)variance components for a sire-maternalgrandsire threshold model and development of a calvingability index for sire selection. J Dairy Sci 90, 2489–2496.

De Jong G, 2005: Usage of predictors for fertility in the geneticevaluation application in the Netherlands. Interbull Bull 33,69–73.

Dobson H, Walker S, Morris M, Routly J, Smith R, 2007:Why is it getting more difficult to successfully artificiallyinseminate cows? Proc Fertility in Dairy Cows – Bridgingthe Gaps, 30–31 August 2007, Liverpool Hope University,Liverpool, UK, p. 4.

Friggens N, Løvendahl P, 2007: Where did she come from?Where is she going? The potential of on-farm fertility profiles.Proc. Fertility in Dairy Cows – Bridging the Gaps, 30–31August2007,LiverpoolHopeUniversity,Liverpool,UK,p.23.

Gengler N, Verkenne C, 2007: Bayesian approach to integratemolecular data into genetic evaluations. Interbull Bull 37,37–41.

Hansen M, Lund MS, Pedersen J, Christensen LG, 2004:Genetic parameters for stillbirth in Danish Holstein cowsusing a Bayesian threshold model. Livest Prod Sci 91, 23–33.

Hayhurst C, Firth M, Christie MF, Royal MD, 2007a:Estimation of genetic variation in embryo quality: Is therepotential to genetically select cattle with the inherent abilityto produce high quality embryos? Proc Fertility in DairyCows – Bridging the Gaps, 30–31 August 2007, LiverpoolHope University, Liverpool, UK, p. 13.

Hayhurst C, Flint APF, Lovendahl P, Wolliams JA, RoyalMD, 2007b: Prepubertal predictors for fertility in dairycattle: potential use of metabolic hormones. Proc BSASAnnual Conference 2007: Southport, UK, 58.

Hayhurst C, Sorensen MK, Royal MD, Løvendahl P, 2007c:Metabolic regulation in Danish bull calves and the relation-

94 B Berglund


ship to the fertility of their female offspring. J Dairy Sci 90,3909–3916.

Heins BJ, Hansen LB, Seykora AJ, 2006a: Fertility andsurvival of pure Holsteins versus crossbreds of Holstein withNormande, Montbeliarde, and Scandinavian Red. J DairySci 89, 4944–4951.

Heins BJ, Hansen LB, Seykora AJ, 2006b: Calving difficultyand stillbirths of pure Holsteins versus crossbreds ofHolstein with Normande, Montbeliarde, and ScandinavianRed. J Dairy Sci 89, 2805–2810.

Heringstad B, Chang YM, Svendsen M, Gianola D, 2007:Genetic analysis of calving difficulty and stillbirth inNorwegian Red cows. J Dairy Sci 90, 3500–3507.

Hjerten J, 2006: Relationships between body condition score,subcutaneous fat, live weight and reproduction in SwedishHolstein and Swedish Red and White Cattle. In Swedish.Summary in English. MSc Thesis 281, SLU, Department ofAnimal Breeding and Genetics, Uppsala, Sweden.

Hoekstra J, van der Lugt AW, van der Werf JHJ, OuweltjesW, 1994: Genetic and phenotypic parameters for milkproduction and fertility traits in upgraded dairy cattle.Livest Prod Sci 40, 225–232.

Holmberg M, Andersson-Eklund L, 2006: Quantitative traitloci affecting fertility and calving traits in Swedish dairycattle. J Dairy Sci 89, 3664–3671.

Holmberg M, Sahana G, Andersson-Eklund L, 2007: Finemapping of a quantitative trait locus on chromosome 9affecting non-return rate in Swedish dairy cattle. J AnimBreed 124, 257–263.

Jacobsen JH, Fikse F, 2005: Feasibility of MACE for calvingtraits for non-Holstein breeds. Interbull Bull 33, 28–31.

Janson L, 1980: Studies on fertility traits in Swedish dairycattle. II. Genetic parameters. Acta Agr Scand 30, 427–436.

Janson L, Andreasson B, 1981: Studies on fertility traits inSwedish dairy cattle. IV. Genetic and phenotypic correlationbetween milk yield and fertility. Acta Agr Scand 31, 313–322.

Jorjani J, 2007: International genetic evaluation of femalefertility traits in five major breeds. Interbull Bull 37, 144–147.

Lindhe B, Philipsson J, 2001: Genetic trends in the twoSwedish dairy cattle breeds SRB and SLB in 1985–1999.SLU, Department of Animal Breeding and Genetics, Publ.138, Uppsala, Sweden.

Lopez de Maturana E, Legarra A, Varona L, Ugarte E, 2007:Analysis of fertility and dystocia in Holsteins using recursivemodels to handle censored and categorical data. J Dairy Sci90, 2012–2024.

Løvendahl P, Chagunda MGG, 2006: Assessment of fertility indairy cows based on electronic monitoring of their physicalactivity. 8th World Congress on Genetics Applied toLivestock Production, Belo Horizonte, Brazil. Proc Com-mun 18-04 (http://www.wcgalp8.org.br).

Lucy MC, 2001: Reproductive loss in high-producing dairycattle: where will it end? J Dairy Sci 84, 1277–1293.

Mark T, Jakobsen JH, Jorjani H, Fikse WF, Philipsson J,2005: International trends in recording and genetic evalua-tion of functional traits in dairy cattle. 56th Annual Meet ofthe European Association for Animal Production, Uppsala,Sweden, Abstract no. 536.

McParland S, Kearney JF, Rath M, Berry DP, 2007:Inbreeding effects on milk production, calving performance,fertility and conformation in Irish Holstein-Friesians.J Dairy Sci 90, 4411–4419.

Miglior F, Muir BL, Van Doormaal BJ, 2005: Selection indicesin Holstein cattle of various countries. J Dairy Sci 88, 1255–1263.

Muir WM, 2007: Genomic selection: a break through forapplication of marker assisted selection to traits of lowheritability, promise and concerns. Paper at the 58th Annual

Meet of the European Association for Animal Production,Dublin, Ireland.

Nielsen HM, Christensen LG, Ødegard J, 2006: A method todefine breeding goals for sustainable dairy cattle production.J Dairy Sci 89, 3615–3625.

Petersson K-J, Berglund B, Strandberg E, Gustafsson H, FlintAPF, Wolliams JA, Royal MD, 2007: Genetic analysis ofpostpartum measures of luteal activity in dairy cows.J Dairy Sci 90, 427–434.

Petersson K-J, Strandberg E, Gustafsson H, Royal MD,Berglund B, 2008: Detection of delayed cyclicity in dairycows based on progesterone content in monthly milksamples. Prev Vet Med (in press).

Roxstrom A, Strandberg E, Berglund B, Emanuelson U,Philipsson J, 2001a: Genetic and environmental correlationsamong female fertility traits and milk production in differentparities of Swedish Red and White dairy cattle. Acta AgricScand A Anim Sci 51, 7–14.

Roxstrom A, Strandberg E, Berglund B, Emanuelson U,Philipsson J, 2001b: Genetic and environmental correlationsamong female fertility traits and the ability to show oestrus,and milk production. Acta Agric Scand A Anim Sci 51, 192–199.

Royal MD, Mann GE, Flint APF, 2000: Strategies forreversing the trend towards subfertility in dairy cattle. VetJ 160, 53–60.

Royal MD, Flint APF, Wolliams JA, 2002: Genetic andphenotypic relationships among endocrine and traditionalfertility traits and production traits in Holstein-Friesiandairy cows. J Dairy Sci 85, 958–967.

Sørensen AC, Lawlor T, Ruiz F, 2007: A survey on fertility inthe Holstein populations of the world. Proc Fertility inDairy Cows – Bridging the Gaps, 30–31 August 2007,Liverpool Hope University, Liverpool, UK, p. 17.

Steinbock L, Nasholm A, Berglund B, Johansson K, Philips-son J, 2003: Genetic effects on stillbirth and calving difficultyin Swedish Holsteins at first and second calving. J Dairy Sci86, 2228–2235.

Steinbock L, Johansson K, Nasholm A, Berglund B, Philips-son J, 2006: Genetic effects on stillbirth and calving difficultyin Swedish Red dairy cattle at first and second calving. ActaAgric Scand A, 56, 65–72.

Van der Beek S, 2007: Effect of genomic selection on nationaland international genetic evaluations. Interbull Bull 37, 115–118.

Van der Lende T, Kaal LMTE, Roelofs RMG, Veerkamp RF,Schrooten C, Bovenhuis H, 2004: Infrequent milk proges-terone measurements in daughters enable bull selection forcow fertility. J Dairy Sci 87, 3953–3957.

VanRaden PM, 2007: Genomic measures of relationship andinbreeding. Interbull Bull 37, 33–36.

VanRaden PM, Sanders AH, Tooker ME, Miller RH,Norman HD, Kuhn MT, Wiggans GR, 2004: Developmentof a national genetic evaluation for cow fertility. J Dairy Sci87, 2285–2292.

Veerkamp RF, Oldenbroek JK, van der Lende T, 1998: The useof milk progesterone measurements for genetic improvementof fertility traits in dairy cattle. Interbull Bull 18, 62–67.

Wall E, Brotherstone S, Wolliams JA, Banos G, Coffey MP,2003: Genetic evaluation of fertility using direct andcorrelated traits. J Dairy Sci 86, 4093–4102.

Author’s address (for correspondence): B Berglund, Department ofAnimal Breeding and Genetics, Swedish University of AgriculturalSciences, Box 7023, SE-750 07, Uppsala, Sweden. E-mail: [email protected]

Conflict of interest: The author declares no conflict of interest.



genetic improvement of dairy cow reproductive performance

Documents