substrate balances across colonic carcinomas in humans

7
1995;55:1373-1378. Cancer Res Eggert Holm, Egbert Hagmüller, Ulrich Staedt, et al. Substrate Balances across Colonic Carcinomas in Humans Updated Version http://cancerres.aacrjournals.org/content/55/6/1373 Access the most recent version of this article at: Citing Articles http://cancerres.aacrjournals.org/content/55/6/1373#related-urls This article has been cited by 14 HighWire-hosted articles. Access the articles at: E-mail alerts related to this article or journal. Sign up to receive free email-alerts Subscriptions Reprints and . [email protected] Department at To order reprints of this article or to subscribe to the journal, contact the AACR Publications Permissions . [email protected] Department at To request permission to re-use all or part of this article, contact the AACR Publications American Association for Cancer Research Copyright © 1995 on February 23, 2013 cancerres.aacrjournals.org Downloaded from

Upload: doannhan

Post on 31-Dec-2016

218 views

Category:

Documents


2 download

TRANSCRIPT

Page 1: Substrate Balances across Colonic Carcinomas in Humans

  

1995;55:1373-1378.Cancer Res Eggert Holm, Egbert Hagmüller, Ulrich Staedt, et al. Substrate Balances across Colonic Carcinomas in Humans  

  

Updated Version http://cancerres.aacrjournals.org/content/55/6/1373

Access the most recent version of this article at:

  

Citing Articles http://cancerres.aacrjournals.org/content/55/6/1373#related-urls

This article has been cited by 14 HighWire-hosted articles. Access the articles at:

  

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

SubscriptionsReprints and

[email protected] atTo order reprints of this article or to subscribe to the journal, contact the AACR Publications

[email protected] at

To request permission to re-use all or part of this article, contact the AACR Publications

American Association for Cancer Research Copyright © 1995 on February 23, 2013cancerres.aacrjournals.orgDownloaded from

Page 2: Substrate Balances across Colonic Carcinomas in Humans

[CANCER RESEARCH 55, 1373-1378, March 15, 1995]

Substrate Balances across Colonie Carcinomas in Humans

Eggert Holm,1 Egbert Hagmüller,Ulrich Staedt, Gerwin Schlickeiser, Hans-Joachim Günther,Hans Leweling,

Mehmet Tokus, and Hermann B. KollmarDepartment of Pathophysiology. Medical Clinic 1 Mannheim, Theodor KuKer Ufer [E. Ho., U. S.. G. S., H. L, M. T.], and Surgical Clinic Mannheim [E. Ha., H-J. C., H. B. K./.University of Heidelberg, D-68167 Mannheim, Germany

ABSTRACT

To investigate the utilization of nutrients by malignant tumors inhumans, the balances of energy-yielding substrates and amino acids across

colonie carcinomas were assessed in 17 patients during surgery. Bloodsamples were taken from an artery and the main tumor-draining vein,

which was also used for determining tumor blood flow (direct venousoutflow technique). Additionally, the substrate exchange by peripheraltissues was studied (femoral arteriovenous differences, venous occlusionplethysmography).

Mean blood flow was greater in the carcinomas than in the leg tissues(43.2 versus 2.5 ml/100 ml/min: P < 0.001). There was a negative correlation between tumor blood flow and tumor weight (r = -0.87; P < 0.001).

Glucose net uptake and lactate release by the malignancies exceeded theperipheral exchange rates 30- and 43-fold, respectively (mean values

different at P < 0.001). The molar ratio of lactate output to glucoseconsumption was 0.78 in the tumors and 0.48 in the leg tissues (/' < 0.05).

Regarding free fatty acid and ketone body balances, no significant tumor-

periphery differences were noted. The carcinomas utilized branched chainamino acids and serine, while alanine and, in particular, ammonia werereleased in large amounts. Net glutamine retention was not consistentlyobserved.

It is concluded that the energy metabolism of human colonie carcinomas relies predominantly on glucose, with fat-derived calories making no

appreciable contribution. The impaired nutritive perfusion of malignanttumors appears to favor glycolysis and to limit both glucose oxidation andglutaminolysis. The present study has shown that the procedure chosenfor the assessment of irans-tumor substrate flux rates is a workable and

valid model for analyzing metabolic balances across human coloniecancers in vivo.

INTRODUCTION

Glucose, glutamine, and ketone bodies are widely regarded as theprimary substrates utilized by malignant tumors to meet their energyrequirements (1-4). The balances, i.e., the net uptake and releaserates, of these and other possible energy-yielding substrates across

tumors in humans have not yet been described. The same applies tothe exchange rates of most amino acids.

Around 50 years following Warburg's demonstration of aerobic

glycolysis in neoplastic tissues (5, 6), further extensive in vitro studieshave led to the conclusion that glutamine is a major energy source formalignant tumors under aerobic conditions (2, 7-10). In tumor cell

cultures, it was found that the contribution of glutamine and glutamateto total energy production amounted to 65-80% (7, 10). However, inanimals with experimental "tissue-isolated" tumors, glutamine utili

zation by various neoplasias was on a molar basis as small as 3-8%

of glucose utilization (3, 11). Even the ketone bodies were retained bythe tumors to a greater extent than was glutamine (3). Lactate could beeither produced or utilized (3, 11-14) depending on the arterial lactate

concentration but not on glucose uptake (3, 11, 14).It is only in about the past 15 years that the metabolism of human

tumors has been investigated in vivo. In studies using positron emis-

Received 7/6/94; accepted 1/17/95.The costs of publication of this article were defrayed in part by the payment of page

charges. This article must therefore be hereby marked adverlisemeni in accordance with18 U.S.C. Section 1734 solely to indicate this fact.

1To whom requests for reprints should be addressed.

sion tomography, an increased uptake of radiolabeled deoxyglucoseappeared to be a general characteristic of malignant tissues (15-18),

with an excellent correspondence between the tracer concentrationand the grade of malignancy (15, 19). Cerebral gliomas were found tometabolize glucose preferentially by aerobic glycolysis (15, 20). Direct measurement of substrate exchange by human tumors transplanted into animals revealed that glucose, in contrast to glutamine(21), was retained by all xenografts (22-25), and the extent of glucose

consumption was strongly dependent on glucose availability, i.e., ontumor blood flow (4, 22-25,). Lactate output from the carcinomas wasclosely related to glucose uptake (4, 22-25) but not to glutamine

uptake (21). Most of the cancer xenografts also utilized ketone bodies(24, 25).

It is possible that the biochemical and immunological milieu ofhuman cancer transplants in animals modifies the "normal" metabolic

processes in these tumors. Moreover, differences in vascular supplybetween xenografts and tumors in patients should also be considered.For these reasons, we have investigated colonie carcinomas duringelective operations. To obtain substrate balances across the tumors,arterial and tumor-venous substrate concentrations were measured,

and blood flow per tumor mass was quantified. Substrate flux acrossthe tumors was compared with that across the peripheral tissues.

Although the metabolic exchange rates of human malignancies can,for anatomically suitable tumors, be assessed by the procedure presented here, it is not the purpose of our paper to recommend thisprocedure as a standard method of analyzing flux rates across humantumors. Studies on cancer xenografts still need to be done as basicresearch. In the future, further advances in noninvasive techniquessuch as positron emission tomography may replace direct and invasivemetabolic measurements.

PATIENTS AND METHODS

Patient Population. Seventeen patients referred to the Surgical ClinicMannheim (Mannheim, Germany) for elective operations were recruited forthe study. All of them had colonie carcinomas which were located in theascending («= 7), transverse (n = 2), or sigmoid colon (;i = 8). As assessed

before and during surgery, tumor stages were T2 in 9 cases and T, in K cases.Cancer differentiation grade was either G2 (medium, n = 15) or G, (poor,n = 2). Five patients had métastasesin the lymph nodes hut none had distant

métastases.CEA amounted to 9.9 ±14.7 (SD) ng/ml. None of the patients hadhad any previous cancer treatment.

The following exclusion criteria were set: any kind of endocrine dysfunction; liver disease; or renal insufficiency (creatinine >1.3 mg/dl). The selectedpatients, 6 males and 11 females, were 66.5 ± 15 (SD) years old. Theydisplayed either normal or moderately increased body weight, with the percentages of optimum body weight ranging from 80 to 120% in 10 cases andexceeding 120% in 7 cases. Apart from slight anemia (hemoglobin, 11.7 ±1.9g/dl), conventional blood chemistry gave normal results (glucose, triglycérides,cholesterol, albumin, cholinesterase, transaminases, lactate dehydrogenase,bilirubin, alkaline phosphatase, 7-GT, ammonia, urea, creatinine, uric acid).

All patients gave their informed consent to participate in the study and therewere no objections from the ethical committee of our medical faculty.

Measurement of Substrate Flux across Carcinomas. Following the preparation of the tumor-bearing part of the colon, the total venous vasculature ofthe pertinent segment was examined and the main tumor-draining vessel wasidentified. After ligation of the neighboring veins the "tumor vein" was

1373

American Association for Cancer Research Copyright © 1995 on February 23, 2013cancerres.aacrjournals.orgDownloaded from

Page 3: Substrate Balances across Colonic Carcinomas in Humans

SUBSTRATE BALANCES ACROSS CARCINOMAS IN HUMANS

isolated and cannulated with a cannula having a diameter similar to that of thevein before puncture. The tumor-draining vein was used for blood flow

determination and then for blood sampling. Blood flow was estimated by the"direct venous outflow technique" (26), with the use of a 2-ml volumetric flask

and a stopwatch. The time needed to fill the volumetric flask was so short that

hyperemia of the carcinoma due to an imbalance between arterial blood supplyand venous outflow never occurred; this was assessed by a continuous obser

vation of both the visible tumor part and the surrounding bowel tissue. Bloodspecimens were taken from the radial artery and the tumor-draining vein

(approximately 15 ml). Immediately after collecting, the samples were

distributed into heparinized tubes for substrate analyses.Postoperatively the carcinomas were carefully prepared and weighed. It was

thereby possible to calculate the substrate flux rates from the arteriotumor-

venous concentration differences and blood flow per tumor mass. Histologicalclassification of the tumors was performed routinely.

Measurement of Substrate Flux across Leg Tissues. Leg blood flow wasquantified by venous occlusion plethysmography (27) with the use of a

Periquant 3500 device (Dr. Gutmann, Eurasburg, Germany); an inflatable cuffwas attached proximal to the knee joint, and a dilation-indicating tape was

attached at the maximum circumference of the calf. Because the inflated cuffblocked the venous outflow without changing the arterial inflow, the resultingincrease in tissue volume per period of time was proportional to the arterialinflow (ml/100 ml tissue/min). In each patient, the mean value of five blood

flow measurements was obtained. Thereafter, blood samples were taken fromthe femoral artery and vein (15 ml from either vessel).

For technical reasons, the substrate balances across colonie carcinomas andthe leg cannot be determined at exactly the same time. Using the upper

extremity for a roughly synchronized measurement of the flux rates during theoperation grossly interferes with the surgical procedure and may, in ouropinion, only be justified in connection with more comprehensive investigations such as tracer studies. Since a consecutive assessment of tumor and leg

metabolism unduly prolongs the duration of anesthesia, it was decided todetermine the substrate exchange by the leg 2-3 days prior to surgery in the

postabsorptive state. This study design necessitated an extra investigation ofthe possible differences between the pre- and intraoperative flux rates in the

periphery. Substrate balances across the leg were therefore measured in aseparate group of eight patients with colorectal cancers both 2 and 3 days

before and during inhalation anesthesia. Blood flow was similar prior to andduring anesthesia [2.1 ±0.5 (SEM) versus 1.9 ±0.5 ml/100 ml/min]. Inaddition, the pre- and intraoperative exchange rates of substrates did not differ

significantly from one another; this applies to the balances (nmol/100 ml/min)of glucose (+557 ±309 versus +369 ±510), láclate (-393 ±84 versus-525 ± 228), FFA2 (-542 ±444 versus -295 ± 107), ketone bodies

(+461 ±223 versus -50 ±178), essential amino acids (-40 ±52 versus-54 ±32), BCAA (+9 ±47 versus -15 ± 12), serine (-2 ±5 versus

-14 ±11), alanine (-157 ±37 versus -77 ±28), and ammonia (+8 ±2

versus + 24 ±12). Regarding the metabolic effects of inhalational anesthetic

agents, the reported results are in part inconsistent (28, 29). The findingsobtained in our patients justify a comparison being made between preoperativeperipheral flux rates and intraoperative írans-tumor balances.

Biochemical Analyses. The concentrations of glucose, pyruvate, láclate,acetoacetate and ß-hydroxybutyrate were determined in whole blood. Ice-

chilled 6% perchloric acid was added to the blood samples and enzymatictests were carried out (30). FFA were estimated in serum by the enzymaticNEFA C test (Wako Chemicals, Neuss, Germany). Before the measuring

the plasma free amino acids (including glutamate and glutamine), bloodspecimens were deproteinized with 10% sulfosalicylic acid and then deep

frozen. Analyses were done by column chromatography on a Biotronic LC5001 instrument with Durum DC 4A. The elution buffer was lithium citrate(pH 2.77, 3.10, 3.38, 4.05, and 5.30) and the internal standard was nor-leucine. Benson Type A-N-B (Biotronic) was used for calibration. For

asparagine and glutamine, an additional calibration was performed. Plasmaammonia was analyzed enzymatically (Monotest, Boehringer-Mannheim,

Mannheim, Germany).

Statistics. Results are presented as mean ±SEM. Regarding blood flowand substrate exchange, the carcinomas were compared with the peripheraltissues by means of Wilcoxon's matched pairs signed rank test.

RESULTS

Blood Flow. The average blood flow was 17 times greater in thecarcinomas than in the peripheral tissues. The respective values were43.2 + 4.9 (SEM) and 2.5 + 0.2 ml/100 g/min, the difference beingsignificant at P < 0.001. It is to be noted in this context that all thecarcinomas had a specific gravity between 0.98 and 1.02. The absolutetumor weight ranged from 6.9 to 53.3 g, with a mean value of 23.7 g.As shown in Fig. 1, there was a close and significant negativecorrelation between tumor weight and tumor blood flow.

Balances of Energy-yielding Substrates. The arterial concentrations of energy-yielding substrates obtained prior to and during the

operations are listed in Table 1. Compared with the preoperative state,anesthesia was in particular associated with an increase in the meanglucose level and a decrease in mean ketone body concentrations;these changes, however, were not statistically significant. As canfurther be seen from Table 1, several substrates showed AV concentration differences across the carcinomas deviating either quantitatively or qualitatively from those assessed across the peripheral tissues. While across the tumors a more positive AV difference ofglucose failed to become significant, a more negative difference inláclatewas observed with low error probability. Marked FFA outpulfrom ihe periphery (per liler of blood) conlrasted significantly wilhslight FFA uptake by the carcinomas. Moreover, as judged by the AVdifferences, the peripheral tissues retained acetoacelale, whereas Ihecarcinomas did not.

Due to the magnilude of lumor blood flow, the AV differencesacross the tumors fail to give an appropriate idea of substrale uplakeand release. The nel balances, i.e., Ihe exchange rales, of glucose,láclale,FFA, and ketone bodies across bolh the carcinomas and theperiphery are illustraled in Fig. 2. Glucose nel uptake by Ihe carcinomas exceeded peripheral glucose retention by a factor of 30 and wasrelated to tumor blood flow (r = 0.55; P = 0.02). Lactale oulpul from

80

r—70_c

x 60

§50

30T3Oo 20

CÛ

10

y= - 1.36x + 75.39r= 0.87p< 0.001

2 The abbreviations used are: FFA, free fatty acids; BCAA, branched-chain amino

acids; AV, arterio-venous.

O 10 20 30 40 50 60

Tumor weight [g]

Fig. 1. Tumor blood flow as a function of tumor weight.

1374

American Association for Cancer Research Copyright © 1995 on February 23, 2013cancerres.aacrjournals.orgDownloaded from

Page 4: Substrate Balances across Colonic Carcinomas in Humans

SUBSTRATE BALANCES ACROSS CARCINOMAS IN HUMANS

Table 1 Arterial concentrations and AV differences of energy-yielding substrates"

Arterial Concentration (mmol/1) Arterio-venous difference (mmol/1)

SubstratesGlucose

Py nivaleLáclateFree fatty acidsAcetoacetate3-OH-butyratePre-op.6.70

(0.22)''

0.19(0.05)0.89 (0.07)0.47 (0.04)0.15(0.02)0.26 (0.04)Intra-op.8.09

(0.39)0.13(0.01)0.86 (0.05)0.56 (0.06)0.09(0.01)0.16(0.02)Periphery+0.42

(0.07)+0.05 (0.05)-0.20 (0.02)-0.20 (0.05)+0.04(0.01)+0.02(0.01)Tumors+0.73

(0.12)-0.01 (0.01)-0.50 (0.07)''+0.01 (0.04)''±0.00 (0.01)'

-0.01 (0.01)" The arterial concentrations of the individual substrates occurring during the opera

tions (intra-op.) did not deviate significantly from those obtained prior to surgery(pre-op.). AV concentration differences were determined across both the peripheral tissues(pre-operatively; femoral artery and vein) and the colonie tumors (radial artery and maintumor-draining vein).

* Mean (SEM).e P < 0.05.

40-, "ft"ft"k

Peripheral tissues

Tumors

-30

Glucose Lactate FFA Ketonebodies

Fig. 2. Balances of energy-yielding substrates across both the peripheral tissues and thecolonie tumors. Columns, mean; bars, SEM. ***, P < 0.001.

the tumors was about 43 times greater than the peripheral release.With regard to both the glucose and láclatebalance, the carcinomasdiffered from the periphery at P < 0.001. Such a difference was notnoted for pyruvate (carcinomas, —0.38f¿mol/liter/min;periphery,

+0.13 fAmol/liter/min;mean values). The molar ratio of láclateproduction to glucose consumption amounted to 0.78 in the tumors andto 0.48 in the leg tissues (difference significant at P < 0.05). For thetumors, there was only a poor correlation between glucose and láclateexchange (r = 0.47; P < 0.05).

The exchange rates of FFA and ketone bodies measured across thecarcinomas were negligible when compared with ihose of glucose andláclate(Fig. 2). FFA and ketone body balances varied considerably,with a positive mean value for the FFA and a negative one for the

ketone bodies, but both of these mean values were not significantlydifferent from those assessed in the periphery.

Substrate uplake in relalion lo supply was calculaled as apercentage extraction:

AV difference x 100arterial concenlralion

Percenlage glucose exlraclion by the tumors was very similar lo lhalassessed in the periphery (9.1 versus 6.5%). However, in Iheperipheral lissues the percentage uptake of aceloacelale exceeded thatof glucose (21.8 versus 6.5%; P < 0.05).

Balances of Amino Acids and Ammonia. Significantdifferencesbetween arterial amino acid levels determined before and duringsurgery did not occur. There was a tendency for Ihe sum of essentialamino acids and most nonessential amino acids, such as alanine, lohave higher concentralions prior lo aneslhesia (Table 2). The BCAAshowed an opposile tendency (Table 2). Plasma ammonia was notaltered during the operations (Table 2). The AV differences showedthat Ihe essenlial amino acids were laken up by the tumors andreleased from the leg tissues, and that a slight peripheral ammoniaretention contrasted with a marked ammonia production by thecarcinomas (Table 2).

The total amino acid balance did not differ significantly betweenthe tumors and the periphery. Neither could a significant difference inthe balance be ascertained for the sum of the essential amino acids(Table 3). In contrast, the group of BCAA, as well as serine, werereleased from the leg tissues but were consumed to a considerableextent by the carcinomas (Table 3). The production of alanine andeven more that of ammonia by the tumors was excessive (Table 3).Inconsistent findings were obtained with respect to the exchange ofglutamate and, in particular, glutamine by the carcinomas. A netretention of glutamine occurred in 10 cases and a net release in 7cases.

DISCUSSION

The present results concerning tumor blood flow were obtained bymeans of the direct venous outflow technique (26), a method generallyrecognized to be superior to any indirect procedure of blood flowdetermination (4, 25). To quantify leg blood flow, venous occlusionplethysmography was used, since this procedure has an accuracy of atleast 95% (31), has been validated by comparing the measured andactual blood flow (r = 0.93, P < 0.001; Ref. 32), and is not tooinconvenient to do under clinical conditions.

Mean tumor blood flow in our patients was 43.2 ml/100 g/min. Thisis well within the range observed by Kallinowski et al. (22-25) invarious human tumors which had been transplanted into T cell-

Table2 ArterialconcentrationsandAVdifferencesofaminoacidsandammonia"

Arterial Concentration(fimol/1)VariablesEssential

AABCAASerineAlanineAmmoniaPre-op.737

(37)*

378(19)94(5)

266 (28)45(5)Intra-op.713

(34)395 (22)

91(5)214(20)42(6)Arterio-venous

difference

(funol/1)Periphery-41

(18)-15(8)

-2(2)-73 (22)

+4(3)Tumors+6(13)'

+5(7)+7(2)

-26 (15)'-159(44)''

" The arterial concentrations of the amino acids (AA) and ammonia occurring during

the operations (intra-op.) did not deviate significantly from those obtained prior to surgery(pre-op). AV concentration differences were determined across both the peripheral tissues(pre-operatively; femoral artery and vein) and the colonie tumors (radial artery and maintumor-draining vein).

* Mean (SEM).' P < 0.05.

P < 0.001.

1375

American Association for Cancer Research Copyright © 1995 on February 23, 2013cancerres.aacrjournals.orgDownloaded from

Page 5: Substrate Balances across Colonic Carcinomas in Humans

SUBSTRATE BALANCES ACROSS CARCINOMAS IN HUMANS

Table 3 Balances of amino acids and ammonia"

Balance (nmol/100 ml/min)

VariablesEssential

AABCAASerineAlanineAmmoniaPeriphery-89

(40)''

-34 (20)-4(5)

-168 (50)

+10(7)Tumors+620

(664)+473 (347)'+346(109)''-988(518)c

-8383 (2408)rf

" The balances were assessed across both the peripheral tissues (femoral artery and

vein; blood flow obtained by venous occlusion plethysmography) and the colonie tumors(radial artery and main tumor-draining vein; blood flow obtained by the "direct venousoutflow technique").

* Mean (SEM).c P < 0.05.11P< 0.001.

deficient rats as tissue-isolated preparations (0.07-0.68 ml/g/min). As

detailed by Vaupel et al. (4), malignancies can exhibit flow ratessimilar to those of either tachytrophic or bradytrophic organs. Thecolonie carcinomas presented here were "high flow tumors" and also

had a higher mean blood flow rate than the tissue of origin (43.2versus 22.7 ml/100 g/min; Ref. 33).

The close negative correlation between tumor weight and tumorblood flow, as illustrated in Fig. 1, is consistent with findingsconcerning human tumor xenografts (4, 22, 23, 25). As a malignancygrows, the vascular network becomes increasingly characterized by areduction in the number of vessels per unit tumor mass, as well as bystructural and functional abnormalities (4). Most of the deviationsfrom a normal vasculature contribute to a progressive decrease inglobal perfusion (4).

The abnormal vascularization of malignancies raises difficulties inthe interpretation of substrate exchange rates as calculated in this andother studies (3,11-14, 21-25). The blood passing through AV shunts

has been estimated to account potentially for at least 30% of the totaltumor blood flow (4). Since the shunted blood does not participate inthe metabolic exchange processes, it should lead to an overestimationof the net uptake and release rates of substrates. However, this portionof the blood does not contribute to the AV concentration differencesof the substrates, so that an underestimation of the latter and, hence,of substrate exchange should result. In connection with the presentwork, Kollmar (34) developed a mathematical model which has madeit possible to quantify the influences of the shunted blood volume, aswell as of the metabolically active and inactive tumor areas on thesubstrate balances. It was derived that it is not the shunt volume butthe proportion of metabolically inactive tumor tissue which determines the "true" uptake and release rates, leading to an underestima

tion of the substrate exchange by the metabolically active tissue of themalignancy (34).

Glucose consumption and lactate release by the carcinomas of ourpatients by far exceeded the exchange rates of FFA and ketone bodies.Also, since a minimum FFA retention was matched by a minimumketone body output, the contribution of fat-derived calories to the

energy metabolism of the tumors can be neglected. The literature onin vivo studies of tumor metabolism offers overwhelming evidencethat glucose is the primary energy source for malignancies (3, 12, 19,20-23, 25, 35, 36). The colonie tumors retained an average of 32jamol/100 g/min; for tumor-free colonie tissue, a glucose utilization

rate of 7.3 ju.mol/100 g/min has been assessed (difference significantat P < 0.01; Ref. 33). The glucose balance across neoplastic tissues isknown to be governed mainly by the supply of glucose and thus bytumor blood flow (3,4,12, 23, 25). Glucose extraction as a percentageof supply, i.e., "utilization" in a stricter sense than that commonly

used (25), amounted to only 9.1% in the colonie cancers.An abnormally high degree of lactate production is considered to be

a biochemical characteristic of malignancies (4). The glycolytic ratecan be determined as the molar ratio between lactate release andglucose uptake (4, 25). In the colonie tumors, which did not consistently utilize glutamine (the second substrate source of lactate), thevalue of this ratio (0.78) was significantly greater than that assessedfor the peripheral tissues (0.48). According to Vaupel (37) and Kal-

linowski et al. (23, 25), the metabolism of malignant tumors isdetermined mainly by the efficacy of the nutritive tumor perfusion,i.e., by oxygen and substrate availability, but not by the metabolicdemands of the cancer cells. This concept includes a consideration ofthe critical diffusion distances for oxygen and nutrients (21). Aroundthe arterial end of a microvessel in cancer tissue, only the "innercircle" (radius, about 80 /xm) is sufficiently supplied with oxygen;

therefore, processes requiring a readily available supply of oxygen(glucose oxidation, glutaminolysis) are largely restricted to this area(21). There is an "outer circle" (radius, about 200 jam), which is

reached by glucose and is exceeded by glutamine. In the area betweenthe first and second circle nonoxidative processes including glycolysispredominate (21). "Glycolysis alone can provide tumor cells withsufficient energy and metabolites for survival and growth" (1). Other

approaches to an understanding of the high glycolytic rate in neoplastic tissues refer to the altered enzyme pattern which ensures lactateformation even in the presence of oxygen (1, 38, 39) and to areduction in the number of mitochondria (40, 41).

Ketone bodies were not utilized by the colonie carcinomas of ourpatients. This is in contrast to a net ketone body uptake observed instudies on both rodent (3) and human tumors (24, 25). However, totalmolar ketone body consumption, as measured in two types of breastcancer xenografts, was 37 and 53 times less than glucose uptake (25). Asecond investigation of colonie carcinomas performed by our group hasshown that ketone body utilization occurred at very high arterial concentrations (33, 42). Likewise, the head and neck cancer patients studied byRichtsmeier et al. (36) displayed positive AV differences of the ketonebodies across their tumors and arterial ketone body levels well abovethose of our patients. Since the arterial concentration is one of the factorsdetermining the ketone body exchange by malignancies, the consumptionrate as a percentage of supply is of the utmost interest. Percentageextraction reached only 6.7 ±1.4% (SEM) in colonie tumors (33) andwas about 10% for both acetoacetate and ß-hydroxybutyratein gastric

tumors (43, 44).Regarding the exchange of glutamine by the colonie carcinomas

investigated, no consistent results were obtained. This principallyagrees with findings published by Kallinowski et al. (21). Hypoxia inneoplastic tisssues prevents glutamine from serving as a major energysource (4, 21). Most of the experimental work leading to the opinionthat glutamine is an essential substrate for the energy production inmalignancies has been done under conditions of an abundant oxygensupply (2, 3, 7, 8, 9, 10). Since glutamine is indispensable for thesynthesis of macromolecules (1), microareas of malignant tumorsshould be investigated with respect to glutamine metabolism.

The present finding that BCAA are utilized by colonie tumorsprompted us to carry out further in vivo studies on colonie and gastrictumors; these studies have confirmed that valine, leucine, and isoleu-cine are taken up (33, 34, 44-46). BCAA utilization did not occur intumor-free colonie tissue (33). In view of the consumption of BCAA at

least by gastrointestinal carcinomas, the recommendation for the use ofdiets enriched in valine, leucine, and isoleucine for oncological patients inparticular (47-49) should be reconsidered. Apart from the BCAA, serine

was consistently retained by the colonie tumors. Serine is an essentialprecursor for the synthesis of purines, pyrimidines, and phospholipids (1).

Since the total amino acid balances across the colonie carcinomasand the peripheral tissues did not differ significantly from one another,this study offers no possibility to characterize malignancies as

1376

American Association for Cancer Research Copyright © 1995 on February 23, 2013cancerres.aacrjournals.orgDownloaded from

Page 6: Substrate Balances across Colonic Carcinomas in Humans

SUBSTRATE BALANCES ACROSS CARCINOMAS IN HUMANS

"nitrogen traps" (50). The net retention of BCAA and serine was

more than matched by a marked liberation of alanine. The alanineoutput was twice as high as that from tumor-free colonie tissue(33). Although the tumors released a large quantity of nitrogen inthe form of alanine, an average of 8 times more nitrogen waseliminated as ammonia. This amount of ammonia generation couldnot be accounted for by the amino acid uptake (see Table 3). Anexcessive ammonia liberation has also been observed in studies onrodent malignancies (11, 51). Since these malignancies did notconsist of gut-derived tissue, the ammonia production by thecolonie tumors does not appear to be typical only for tumors ofintestinal origin. Protein degradation and cell necrosis may wellcontribute to the ammonia formation.

Our study has shown that the procedure chosen for the assessment of frans-tumor substrate flux rates is a workable and validmodel for analyzing metabolic exchange by human colonie cancersin vivo. Comparing the balances of energy-yielding substratesacross the carcinomas and the peripheral tissues, as determined inthe present and subsequent studies by our group, the predominantfindings can be summarized as follows: (a) the glucose and láclatebalances across the carcinomas by far exceeded the respectiveperipheral exchange rates, with fat-derived substrates making noappreciable contribution to the energy consumption of the tumors(33, 42-44); (b) glucose uptake by the peripheral tissues wasshown to be grossly impaired, while FFA and ketone bodies wereutilized to a normal degree (44, 52, 53); (c) the percentage extraction of glucose and ketone bodies, if ketone bodies are retained,were found to be similar in carcinomas but differed significantlyfrom one another in the periphery, in that the peripheral tissuesutilized, on average, less than 10% of the glucose and 22 to 34%of the acetoacetate supplied (33, 43, 44). Regarding nutrition incancer disease, these findings suggest that a diet low in carbohydrates and high in fat does not comply with the demands ofmalignant tumors but adequately meets the metabolic limitations(glucose) and potentials (fat) of the host (54).

REFERENCES

1. Eigenbrodt, E., Fister, P., and Reinacher, M. New perspectives on carbohydratemetabolism in tumor cells (review). In: R. Breitner (ed.), Regulation of CarbohydrateMetabolism, Vol. 2, pp. 141-179. Boca Raton. FL: CRC Press, 1985.

2. Matsuno, T. Bioenergetics of tumor cells: glutamine metabolism in tumor cellmitochondria. Int. J. Biochem., 19: 303-307, 1987.

3. Sauer, L. A., and Dauchy, R. T. Ketone body, glucose, lactic acid, and aminoacid utilization by tumors in vivo in fasted rats. Cancer Res., 43: 3497-3503,1983.

4. Vaupel, P., Kallinowski, F., and Okunieff, P. Blood flow, oxygen and nutrient supply,and metabolic microenvironment of human tumors: a review. Cancer Res., 49:6449-5465, 1989.

5. Warburg, O. Ãœberden Stoffwechsel der Carcinomzelle. In: O. Warburg (ed.), Ãœberden Stoffwechsel der Tumoren, pp. 187-193. Berlin: Springer-Verlag, 1926.

6. Warburg, O., Wind, F., and Negelein, E. Überden Stoffwechsel von Tumoren imKörper.In: O. Warburg (ed.). Überden Stoffwechsel der Tumoren, pp. 248-257.Berlin: Springer-Verlag, 1926.

7. Kovacevic, Z., and McGivan, J. D. Mitochondria! metabolism of glutamine andglutamate and its physiological significance. Physiol. Rev., 63: 547-605, 1983.

8. Kovacevic, Z., and Morris, H. P. The role of glutamine in the oxidative metabolismof malignant cells. Cancer Res., 32: 326-333, 1972.

9. Moreadith, R. W., and Lehninger, A. L. The pathways of glutamate and glutamineoxidation by tumor cell mitochondria. J. Biol. Chem., 259: 6215-6221, 1984.

10. Reitzer, L. J., Wice, B. M., and Kennell, D. Evidence that glutamine, not sugar, is themajor energy source for cultured HeLa cells. J. Biol. Chem., 254: 2669-2676, 1979.

11. Sauer, L. A., Stayman, J. W., and Dauchy, R. T. Amino acid, glucose, and lactic acidutilization in vivo by rat tumors. Cancer Res., 42: 4090-4097, 1982.

12. Cullino, P. M., Grantham, F. H., and Courtney, A. H. Glucose consumption bytransplanted tumors in vivo. Cancer Res., 27: 1031-1040, 1967.

13. Cullino, P. M., Grantham, F. H., Courtney, A. H., and Losonczy, I. Relationshipbetween oxygen and glucose consumption by transplanted tumors in vivo. CancerRes., 27: 1041-1052, 1967.

14. Sauer, L. A., and Dauchy, R. T. Regulation of láclateproduction and utilization in rattumors in vivo. J. Biol. Chem., 260: 7496-7501, 1985.

15. Hawkins, R. A., and Phelps, M. E. PET in clinical oncology. Cancer Metastasis Rev.,

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.38.

39.

40.

41.

42.

43.

44.

45.

1377

7: 119-142, 1988.Knopp, M. V., Bischoff. H., Oberdorfer, F., and van Kaick, G. Positronenemissions-tomographie des Thorax. Derzeitiger klinischer Stellenwert. Radiologe, 32:290-295, 1992.

Strauss, L. G. Positronenstrahler fürdie Erforschung des Tumorstoffwechsels.Radiologe, 29: 318-321, 1989.Yonekura, Y., Benua, R. S., Brill, A. B., Som, P., Yeh, S. D. J., Kemeny, N. E.,Fowler, J. S., MacGregor, R. R., Stamm. R., Christman, D. R., and Wolf, A. P.Increased accumulation of 2-deoxy-2-('8F) fluoro-D-glucose in liver métastasesfrom

colon carcinoma. J. Nucí.Med., 23: 1133-1137, 1982

Di Ghiro, G., DeLaPaz, R. L. Brooks, R. A., Sokoloff, L., Komblith. P. L., Smith, B. H.,Patrañas,N. J., Kufta, C. V, Kessler, R. M., Johnston, G. S.. Manning, R. G.. and Wolf,A. P. Glucose utilization of cerebral gliomas measured by (IHF) fluorodeoxyglucose and

positron emission tomography. Neurology, 32: 1323-1329. 1982.

Rhodes, C. G., Wise, R. J. S., Gibbs, J. M.. Frackowiak, R. S. J., Hatazawa, J., Palmer,A. J., Thomas, D. G. T., and Jones, T. In vivo disturbance of the oxidative metabolismof glucose in human cerebral gliomas. Ann. Neurol., 14: 614-626, 1983.Kallinowski, F., Runkel, S., Fortmeyer, H. P., Förster,H., and Vaupel, P. L-Gluta-mine: a major substrate for tumor cells in vivo? J. Cancer Res. Clin. Oncol., 113:209-215, 1987.

Kallinowski, F., Schienger, K. H., Kloes, M., Stohrer, M., and Vaupel. P. Tumorblood flow: The principal modulator of oxidative and glycolytic metabolism, and ofthe metabolic micromilieu of human tumor xenografts in viVo. Int. J. Cancer, 44:266-272, 1989.Kallinowski, F., Schienger, K. H., Runkel, S., Kloes, M., Stohrer, M., Okunieff, P.,and Vaupel. P. Blood flow, metabolism, cellular microenvironment, and growth rateof human tumor xenografts. Cancer Res., 49: 3759-3764, 1989.

Kallinowski, F., Schienger, K., Runkel, S., Vaupel, P., Baessler, K. H..and Fortmeyer, H. P. Glucose, láclate,and ketone body utilization by human gynecological tumors xenografted into rnu/rnu-rats. Strahlentherapie Onkol., 165:507-508,

1989.Kallinowski, F., Vaupel, P., Runkel, S., Berg, G., Fortmeyer, H. P., Baessler, K. H.,and Wagner, K. Glucose uptake, láclate release, ketone body turnover, metabolicmicromilieu, and pH distributions in human breast cancer xenografts in nude rats.Cancer Res., 48: 7264-7272, 1988.

Vaupel, P., Fortmeyer, H. P., Runkel, S., and Kallinowski, F. Blood flow, oxygenconsumption, and tissue oxygénationof human breast cancer xenografts in nude rats.Cancer Res., 47: 3496-3503, 1987.

Barbey, K., and Barbey, P. Ein neuer Plethysmograph zur Messung der Extremitätendurchblutung. Z. Kreislaufforsch., 52: 1129-1140, 1963.

Carli, F., Ramachandra, V., Gandy, J„Merrit, H., Ford, G. C., Read, M., andHalliday, D. Effect of general anaesthesia on whole body protein turnover in patientsundergoing elective surgery. Br. J. Anaesth., 65: 373-379, 1990.

Horber, F. F., Krayer, S., Miles, J., Cryer, P., Redder, K., and Haymond, M. W.Isoflurane and whole body leucine, glucose, and fatty acid metabolism in dogs.Anesthesiology, 73: 82-92, 1990.

Bergmeyer, H. U. (ed.). Methoden der enzymatischen Analyse. Vol. 2, Weinheim:Verlag Chemie, 1974.Rudofsky. G., and Gutmann, J. Venenverschluß-Plethysmografie. In: J. Gutmann

(ed.), Eurasburger Berichte zu Themen der Angiologie und Kardiologie. Eurasburg,1979.Norton, J. A., Burt, M. E., and Brennan, M. F. Capacitance plethysmograpahy:validation in an ex vivo human limb model. J. Surg. Res., 32: 127-129, 1982.

Hagmüller, E.. Der Stoffwechsel von malignen Kolontumoren beim Menschen."In vivo" Untersuchungen zum quantitativen Substrataustausch sowie zum Aminosäurenstoffwechsel mit Hilfe einer modifizierten '3C-Leuzin- Tracertechnik. Heidel

berg, Germany: Med. Habilitationsschrift, 1992.Kollmar, H. B. Stoffwechseluntersuchungen an Patienten mit kolorektalemKarzinom. Vergleich des tumoralen und peripheren Substratmetabolismussowie Analysen zur Proteinkinetik der Tumoren, der peripheren Gewebe und desGesamtkörpers mittels '3C-Leucintracertechniken. Heidelberg, Germany: Med.

Diss.. 1993.Norton, J. A., Buri, M. E., and Brennan. M. F. In vivo utilization of substrate byhuman sarcoma-bearing limbs. Cancer (Phila.), 45: 2934-2939. 1980.

Richtsmeier, W. J., Dauchy, R., and Sauer, L. A. In vivo nutrient uptake by head andneck cancers. Cancer Res., 47: 5230-5233, 1987.Vaupel, P. Hypoxia in neoplastic tissue. Microvasc. Res., 13: 399-408, 1977.Eigenbrodt, E., and Glossmann, H. Glycolysis: one of the keys to cancer? TrendsPharmacol. Sci., 240-245, 1980.Eigenbrodt, E., Reinacher, M., Scheefers-Borchel, U., Scheefers, H., and Friis, R.Double role for pyruvate kinase type M: in the expansion of phosphometabolite poolsfound in tumor cells. Crit. Rev. Oncog., 3: 91-115, 1992.Argiles, J. M., and Azcón-Bieto, J. The metabolic environment of cancer. Mol. Cell.Biochem., 81: 3-17, 1988.

Fiechter, A., and GmünderF. K. Metabolic control of glucose degradation in yeastand tumor cells. Adv. Biochem. Eng. Biotechnol., 39: 1-28, 1989.Günther,H.-J.. Hagmüller.E.. Kollmar. H., Saeger, H. D., and Holm, E. In vivoinvestigation of substrate exchange concerning energy-metabolism by human coloncarcinomas. Clin. Nutr. (Suppl.), //: 17-18, 1992.

Greschner, M., Saeger, H. D., Barth, H., Leweling, H.. and Holm, E. Balances ofenergy-yielding substrates across malignant gastric tumors in man. Clin. Nutr.

(Suppl.), 10: 12, 1991.Holm, E., Staedt, U., Leweling, H., and Tokus. M. Metabolism and artificial nutritionof patients with malignancy. S. Afr. J. Clin. Nutr., 6: 10-21, 1993.Hagmüller,E., Günther,H.-J., Kollmar, H., Saeger, H. D., and Ghoos, Y. "In vivo"

American Association for Cancer Research Copyright © 1995 on February 23, 2013cancerres.aacrjournals.orgDownloaded from

Page 7: Substrate Balances across Colonic Carcinomas in Humans

SUBSTRATE BALANCES ACROSS CARCINOMAS [N HUMANS

investigations of protein metabolism by human colon carcinomas using a modifiedtracer technique with 13C-leucine. Clin. Nutr. (Suppl.), 11: 16, 1992.

46. Hagmüller,E., Günther,H.-J., Kollmar, H., Greschner, M., Kohl, M., and Holm, E.

Transtumorale Bilanzen verzweigtkettiger Aminosäuren (VAS) und tumoralerLeuzinstoffwechsel. In vivw-Studien an Patienten mit gastrointestinalen Malignomen.Aktuel. Ernähr.-Med., 18: 59, 1993.

47. Hunter, D. C., Weintraub, M., Blackburn, G. L., and Bistrian, B. R. Branched chainamino acids as the protein component of parenteral nutrition in cancer cachexia.Br. J. Surg., 76: 149-153, 1989.

48. Tayek, J. A., Bistrian, B. R., Hehir, D., J., Martin, R., Moldawer, L. L., and BlackburnG. L. Improved protein kinetics and albumin synthesis by branched chain aminoacid-enriched total parenteral nutrition in cancer cachexia. Cancer (Phila.), 58:147-157, 1986.

49. Yamanaka, H., Kanemaki, T., Tsuji, M., Rise, Y., Hatano, T., Hioki, K., andYamamoto, M. Branched-chain amino acid-supplemented nutritional support after

gastrectomy for gastric cancer with special reference to plasma amino acidprofiles. Nutrition, 6: 241-245, 1990.

50. Torosian, M. H., and Nguyen, H. Q. Tumors: effective nitrogen traps independent ofprotein intake. J. Surg. Res., 47: 456-459, 1989.

51. Chance, W. T., Cao, L., Nelson, J. L., Foley-Nelson, T., and Fischer,J. E. Hyperammonemia in anorectic tumor-bearing rats. Life Sci., 43: 67-74,

1988.52. Holm, E. Ernährungstherapie bei Tumorkrankheiten: Wird der Tumor gefüttert?In: P.

Schauder (ed.), Ernährung und Tumorerkrankungen, pp. 454-476. Basel, Switzerland: S. Karger AG, 1991.

53. Staedt, U., Holm, E., Decken, D-W„Kühn,C., and Saeger, H-D. Peripheral exchangeof energy-yielding substrates postabsorptively and during TPN in cancer patients(CP). Clin. Nutr. (Suppl.), 8: 37, 1989.

54. Tisdale, M. J., Brennan, R. A., and Fearon K. C. Reduction of weight loss and tumoursize in a cachexia model by a high fat diet. Br. J. Cancer, 56: 39-43, 1987.

1378

American Association for Cancer Research Copyright © 1995 on February 23, 2013cancerres.aacrjournals.orgDownloaded from