Enhanced

Glutamine Stimulated

and Glucose Metabolism in Cultured Rat Splenocytes by Phorbol Myristate Acetate Plus Ionomycin Guoyao

Wu, Catherine

J. Field, and Errol B. Marliss

Metabolism of glutamine and glucose was studied in normal rat splenocytes cultured for 48 hours in the presence and absence of a mixture of the mitogens, phorbol myristate acetate (PMA) + ionomycin (lono). 3H-Thymidine uptake by splenocytes was stimulated more than 500-fold by PMA + lono. After culture, cells were incubated for 2 hours in the presence of either 2 mmol/ L [U-14C]glutamine f 5 mmol/L glucose or 5 mmol/L [LJ-‘%]glucose + 2 mmol/L glutamine in Krebs-Ringer HEPES buffer. Glutamine was metabolized mainly to ammonia, glutamate, aspartate, and CO*, and these products were all increased (P < .Ol) by twofold to 2.5.fold in stimulated cells. Glucose was metabolized mainly to lactate and, to a lesser extent, to pyruvate and COz. Lactate production from glucose was increased (P < .Ol) by 2.4.fold in stimulated cells, without changes in pyruvate or COz production. In unstimulated, cultured splenocytes, glutamine was not quantitatively as important as glucose in the provision of adenosine triphosphate (ATP), as calculated on the basis of measured metabolites. However, in stimulated cells, glutamine became a much more important energy substrate, providing similar amounts of ATP to those from glucose. The oxidation of glutamine via the Krebs cycle was the major pathway for glutamine-derived ATP production, while lactate production from glucose accounted for the major part of glucose-derived ATP in PMA + lono-stimulated splenocytes. Thus, we suggest glutamine plays a dual metabolic role in these cells, as both an important fuel and an essential source of carbon and nitrogen precursors for biosynthetic processes. Given the relative mass of cells of the immune system in the body, our findings suggest both the need for an adequate supply of glutamine for the functions of immunocytes in states of altered nutrition and stress, and the significant contribution of glucose to circulating lactate by these cells. Copyright 0 199iby W.B. Saunders Company

T

HERE HAS BEEN increasing interest in metabolism and function of amino acids in lymphocytes1~2 ever since the pioneering work of Eagle3 on amino acid metabolism in cultured mammalian cells. It is now well documented that glutamine is essential for lymphocytes to proliferate in culture.4-6 Ardawi and Newsholme4 first demonstrated that larger amounts of glutamine than glucose are used in lymphocytes, raising the possibility that, in addition to its role in macromolecule synthesis, glutamine might be a significant energy source for these cells. In both resting and mitogen-stimulated lymphocytes, glutamine is metabolized mainly to ammonia, glutamate, and, to a lesser as extent, aspartate,6-s a process termed “glutaminolysis”,9 distinct from complete oxidation to COz. The proposed role of glutaminolysis for provision of adenosine triphosphate (ATP) for lymphocytes4,6,9 appears to be uncertain.‘O Recently, O’Rourke and Rider” reported that in the presence of both glucose and glutamine, glutamine could contribute, at most, only 5% to 15% of ATP in both resting and mitogen-stimulated splenocytes. In contrast, glutamine is known to be the major energy source for rapidly dividing cells such as enterocytes, fibroblasts, reticulocytes, and certain tumor cells.‘*J3 We have also recently demonstrated that glutamine is a quantitatively important energy source for freshly isolated splenocytes and lymph node cells from normal rats, as it can contribute as much ATP as can

From the McGill Nutrition and Food Science Centre, Royal Victoria Hospital, Montreal, Quebec, Canada. Supported by the Medical Research Council of Canada. Present address (G. W): Department of Animal Science, Texas A&M University College Station, TX 77843-2471. Address reprint requests to Errol B. Marl.&, MD, McGill Nutrition and Food Science Centre, Royal Victoria Hospital, 687 Pine Ave U: Montreal, Quebec, Canada H3A IAI. Copyright 0 1992 by W.B. Saunders Company 0026049519214109-0009$03.00/0 982

glucose.13 The rates of glutamine metabolism and its contribution to ATP production are increased twofold in immunologically activated splenocytes from spontaneously diabetic BB rats,15 further supporting the capacity of glutamine to serve as an important fuel for lymphocytes. As the utilization of glutamine is increased by mitogenstimulated lymphocytes, 7.8 it could also play an important role in provision of ATP for these cells. It had previously been assumed that glucose is the major, if not the only, fuel used by lymphocytes.1h,17 Further, the conversion of glucose to lactate via glycolysis has been considered to be the predominant pathway for glucosederived ATP production in mitogen-stimulated lymphocyteslh and in proliferating cells in culture.‘” This appears to be consistent with the data suggesting that lactate production is the major source of energy from glucose metabolism in mitogen-stimulated thymocytes’ and splenocytes.ly By contrast, the recent studies by O’Rourke and Rider” indicated that the oxidation of glucose to CO: was the predominant pathway for glucose-derived ATP production in mitogen-stimulated splenocytes. Thus, it remains to be determined whether lactate production or oxidation to CO2 is the major pathway for glucose-derived ATP production in both mitogen-stimulated and resting lymphocytes. At present, little information is available regarding the quantitative importance of the pathways involved in glutaminederived ATP production in these cells. The objectives of this study were to examine the role of glutamine as an energy source for mitogen-stimulated and resting lymphocytes as compared with that of glucose, and to estimate the relative importance of pathways by which ATP could be produced from glutamine and glucose. We used a combination of two nonspecific mitogenic stimuli, phorbol myristate acetate (PMA) and ionomycin (Iono), that we have recently found to produce maximal responses in rat splenocytes.‘” Metabohsm, Vol41, No 9 (September), 1992: pp 982-988

GLUTAMINOLYSIS

AND GLYCOLYSiS

MATERIALS

IN LYMPHOCYTES

AND METHODS

Chemicals [U-14C]Glutamine (260 mCi/mmol) and [U-i4C]ghtcose (270 mCi/mmol) were obtained, respectively, from Amersham (Oakville, Ontario, Canada) and ICN Radiochemicals (Irvine, CA). Radiochemical purity of i4C-glutamine was rechecked to be more than 99% by thin-layer chromatography, and it was always purified immediately before use to minimize background values.14 3HThymidine (6.7 Ci/mmol) was alsoobtained from ICN Radiochemicals. Culture medium RPM1 1640 (10 mmol/Lglucose, glutaminefree), fetal calf serum (FCS), and other culture supplements were obtained from Flow Laboratories (Mississauga, Ontario, Canada). Iono was obtained from Calbiochem (La Jolla, CA). PMA, bovine serum albumin ([BSA] Fraction V), HEPES, and Hyamine hydroxide were obtained from Sigma Chemicals (St Louis, MO). D-Luciferin, ATP, luciferase, lactate dehydrogenase, glutamate dehydrogenase, NADH, adenosine diphosphate (ADP), and cY-ketoglutaric acid were obtained from Boehringer Mannheim (Montreal, Quebec, Canada). Animals Male Wistar-Furth rats were obtained from Harlan SpragueDawley (Indianapolis, IN). Animals were housed in a light(12-hour light-dark cycles) and climate-controlled facility and provided with ad libitum laboratory rat chow (Ralston-Purina, Woodstock, Ontario, Canada). Preparations of Splenoqtes Rats weighing 175 to 210 g were anesthetized with ether. Spleens were removed and passed through a 50-mesh grid in cold RPM1 1640 under sterile conditions. After centrifugation at 200 x g for 10 minutes at 4”C, red blood cells were lysed for 3 minutes in a solution consisting of 155 mmol/L NH&ZI, 0.1 mmol/L disodium EDTA, and 10 mmol/L KHCO3. Cells were washed three times with RPM1 1640. Cell viability was assessed by 0.2% trypan blue exclusion. Culture of Splenoqtes Isolated splenocytes were cultured (4 x lob viable cells/ml) for 48 hours in an incubator with a humidified atmosphere at 37°C in the presence of 95% Oz/5% CO2. The media were supplemented with 10% heat-inactivated FCS, penicillin (100 U/mL), streptomycin (100 p,g/mL), 2.5 kmol/L 2-mercaptoethanol, 25 mmol/L HEPES, and 2 mmol/L glutamine with or without PMA (20 ng/mL) + Ioao (0.5 pmol/L). Mitogenic response was assessed in triplicate by measurements of 3H-thymidine uptake in 96-well plates with 2.5 x lo5 cells per well. In the experiments to study the effect of glutamine on 3H-thymidine incorporation, FCS was dialyzed to remove glutamine (glutamine-free) and other small molecules (n = LO), and the culture media contained 0 to 4 mmol/L added glutamine. Cells were pulsed with 3H-thymidine (I @i/well) for 18 hours and terminated by harvesting the cells onto glass fiber filters (Skatron, Sterling, VA) using an automatic cell harvester (Skatron). After drying, filters were immersed in Universol cocktail (ICN Radiochemicals) and radioactivities were measured by a 1217 LKB Rackbeta counter (Pharmacia, Baie d’Urfe, Quebec, Canada). Metabolic Studies For studies of glutamine and glucose metabolism, splenocytes were cultured as previously described. At the end of the 48-hour culture, the cell suspensions were centrifuged for 10 minutes at 200 x g at 4°C. Cells were washed three times with Krebs-Ringer

HEPES buffer (pH 7.4) containing 0.5% BSA.lS Cell viability was again assessed by 0.2% trypan blue exclusion. Splenocytes were incubated (4 x 106viable cells/ml) for 2 hours in duplicate at 37°C in lo-mL, silicone-treated tubes, in a total volume of 1 mL Krebs-Ringer HEPES buffer (pH 7.4) containing 0.5% BSA. [U-i4C]Glutamine (2 mmol/L, 150 dpm/nmol) with or without 5 mmol/L glucose, or 5 mmol/L [U-14C]glucose (300 dpm/nmol) with or without 2 mmol/L glutamine was included in the incubation media. Tubes were gassed with 100% 02 for 20 seconds. Incubation was initiated by the addition of cells and terminated by the addition of 100 p,L 1.5 mol/L HCI04 to the incubation medium through the stopper. At the end of incubation, 0.2 mL Hyamine hydroxide was added to a 500~PL microcentrifuge tube (suspended at the top of tbe incubation tube) through the stopper. Shaking was continued for 1 hour to collect the 14COz evolved. Radioactivities in i4C02 were measured by the LKB Rackbeta counter (Pharmacia).‘4.15 The media, to which cells were added following acidification with 100 pL 1.5 mol/L HCIO4 at the end of incubation, were used as blanks. Blank values were always subtracted from the sample values. To minimize hydrolysis of glutamine to glutamate, the acidified [i4C]glutamine-containing media used for analyses of ammonia, t4C-glutamate, and 14C-aspartate were immediately neutralized with 2 mol/L KzCO3 after 14COz collection was completed. The samples were stored at -20°C and analyzed within 3 days. This resulted in very small percentages (0.79% + 0.04%, mean 2 SEM, II = 9) of added glutamine being converted to glutamate, as determined from measurement of 14C-glutamate production from i4C-glutamine in the blanks. In preliminary experiments, it was established that t4C02 production and metabolite formation from glutamine and glucose were linear during a 3-hour incubation period. Lactate and pyruvate, in acid extracts of cells plus medium, were determined by automated fluorometric enzymic methods.21 Ammonia was measured by a spectrophotometric enzymic method.22 14C-Glutamate and 14C-aspartate were separated by AG 1 x 8 (200 to 400 mesh, acetate form; BioRad Laboratories, Richmond, CA) chromatography as previously described,14,t5 and their radioactivities were measured by liquid scintillation counting using Universal cocktail (ICN). ATP in splenocytes (not acidified) was extracted by heating the samples in a boiling water bath for 5 minutes.23 After cooling on ice, the samples were centrifuged for 15 seconds in a Microcentrifuge (Fisher Scientific, Montreal, Quebec, Canada). Ten to fifty microliters supernatant was immediately used for determination of ATP using luciferase,z4 with the photons measured by liquid scintillation spectrometry.25 Calculation of Potential ATP Production Potential production of ATP from glutamine was calculated on the basis of the following assumptions, as previously described”~? 9 mol ATP per 1 mol aspartate produced and 24 mol ATP per 5 mol CO2 production via the complete oxidation of glutamine, which was equal to total CO2 production minus aspartate production, as stoichiometrically 1 mol CO1 is produced per 1 mol aspartate formed. Potential production of ATP from glucose was calculated on the following basis: 1 mol ATP per 1 mol lactate produced, and 38 mol ATP per 6 mol CO2 produced.i4 StatisticalAnalyses Results were statistically analyzed by paired t test, unpaired t test, or one-way ANOVA with the SNK test for multiple mean comparisons, using the Primer Biostatistics package (McGrawHill, Montreal, Quebec, Canada) on a Hewlett-Packard Vectra microcomputer (Sunnyvale, CA).

984

WU, FIELD, AND MARLISS

RESULTS

b

-‘H-Thymidine Uptake by Splenocytes

In splenocytes cultured in the absence of PMA + Iono (unstimulated cells), only very small amounts of “Hthymidine were taken up by cells (Table 1). The presence of PMA + Iono stimulated 3H-thymidine uptake by more than W-fold in the presence of glutamine (stimulated cells). However, in the absence of glutamine, PMA + Iono did not enhance 3H-thymidine uptake (Table 1). The 3H-thymidine uptake was maximal at 48 hours and decreased somewhat (P < .05) by 72 hours of culture (Fig 1). The glutamine concentration necessary to stimulate 3H-thymidine uptake by PMA + Iono half-maximally was 0.14 mmol/L, as determined from the double reciprocal plot (not shown), and no further increase was observed at glutamine concentrations greater than 0.5 mmol/L. Glutamine Metabolism

Glutamine was metabolized mainly to ammonia and glutamate and, to a lesser extent, to aspartate and CO* (on the basis of glutamine carbons) in both stimulated and unstimulated splenocytes (Table 2). In unstimulated cells, a small amount of pyruvate, but no lactate, was detected in the presence of glutamine alone. However, small amounts of lactate were detected in stimulated cells in the presence of glutamine alone. The rates of glutamine metabolism to ammonia, glutamate, aspartate, and COZ were all markedly increased (P < .Ol) in stimulated cells. Glucose decreased (P < .Ol) the production of 14C02 and aspartate from glutamine, but had no effect on that of glutamate and ammonia in either stimulated or unstimulated cells. Assuming that glutamine carbons are converted mainly to glutamate, aspartate, and CO2 in both mitogen-stimulated and the percentage of metabounstimulated lymphocytes, 114,6~11 lized glutamine carbons appearing in CO1 was estimated to

Table 1. WThymidine

Uptake by Splenocytes CultureCondition

Glutamine Concentration (mmol/L)

(?PMA

+ lono)

(cpm/l8 h per 2 x IO5 cells) +

0

54 2 4’

87 + 9’

0.01

93 + 5b

373 + 8b

0.05

253 2 15”

11,279 * 88OC

0.1

332 + 40”

70,213 2 1,691d

0.25

301 * 36”

161,068 lr 4,629e

0.50

299 + 14c

187,916 + 4,854‘

1.0

305 2 28c

188,122 + 4,793‘

2.0

292 + 39c

199,221 f 7,826‘

4.0

263 ? 49’

190,707 + 5,668’

NOTE. Splenocytes were cultured for 48 hours in the presence and absence of PMA + lono, and cells were pulsed with 3H-thymidine for 18 hours. initial culture medium glutamine concentrations

were 0 to 4

mmol/L.

Values are means + SEM for eight animals. Means with

different

superscript

letters are significantly different

(P < .Ol). as

analyzed by one-way ANOVA. 3H-Thymidine uptake was significantly higher (P i .Ol) in PMA + lono-stimulated cells at all glutamine concentrations,

cells than in unstimulated

as analyzed by paired t test.

Glucose (10 mmol/L) was present in the initial culture medium.

a i I

I

24

72 Culture48tirne

Fig. 1.

(h)

3H-thymidine uptake.

be 17% (unstimulated cells) and 15% (stimulated cells) in the absence of glucose. Glucose decreased (P < .Ol) the percentage of metabolized glutamine carbons appearing in COZ in unstimulated splenocytes, but had no effect in stimulated cells. Glucose Metabolism

Glucose was metabolized mainly to lactate and, to a much lesser extent, to pyruvate and CO2 (on the basis of glucose carbons) in both stimulated and unstimulated splenocytes (Table 3). The rate of glucose conversion to lactate was markedly increased (P < .Ol) in stimulated cells. However, neither pyruvate nor COZ production was altered in stimulated cells. Glutamine decreased (P < .Ol) 14COz and lactate production in both stimulated and unstimulated cells. Assuming that glucose carbons are converted mainly to lactate, pyruvate, and CO1 in both mitogen-stimulated and unstimulated lymphocytes,1,4,hJ1 the percentage of metabolized glucose carbons appearing in CO2 was estimated to be 4.9% (unstimulated cells) and 4.7% (stimulated cells) in the absence of glutamine. In both stimulated and unstimulated cells, glutamine decreased (P < .Ol) the percentage of metabolized glucose carbons appearing in COZ. Potential Production of ATP From Glutamine und Giucose

The potential production of ATP from glutamine and glucose is shown in Table 4. In unstimulated splenocytes, the oxidation of glutamine via the Krebs cycle could account for the major part (63% to 70%) of glutaminederived ATP, while the oxidation of glucose via the Krebs cycle could account for approximately 50% of glucosederived ATP. Glucose decreased (P < .Ol) potential ATP

GLUTAMINOLYSIS

985

AND GLYCOLYSIS IN LYMPHOCYTES

Table 2. Glutamine Metabolism in PMA + lono-Stimulated Unstimulated

2 mmol/L

(?PMA

‘TOZ

NO

+ lono)

_

[‘%]Aspartate

IO

6.7 + 0.3 5.1 * 0.3

Ammonia +

10.5 2 1.5*t 5.3 r 0.4 12.5 2 1.5t

IO 1.60 2 0.13

0.74 k 0.03* 1.49 + 0.11*t

IO 11.4 f. 1.4

10.8 + 1.6

7 24.6 + 4.lt

23.2 + 5.2t

IO

Lactate

Glucose

3.7 k 0.1*

7 12.5 k 1.4t 7 3.98 f 0.24t

+

carbons in CO**

Glucose

10

+

% of metabolized Gln

+5 mmol/L

n

7 13.7 k 0.9t

+ _

Pvruvate

[U-‘4C]Glutamine

(nmol products12 h per lo6 cells)

Culture Condition Product

and

Splenocytes

ND

21.6 + 2.2

+

7 1.70 + 0.30

50.5 + 5.3t

_

10 0.45 + 0.05

2.46 k 0.22*

+

7 0.46 + 0.06

2.74 2 0.31’

_

IO 17.1 -c 0.6

11.6 2 0.5*

+

7 14.9 ? 0.7

13.4 i 1.2

NOTE. Splenocytes were cultured for 48 hours in the presence or absence of PMA + lono. After culture, cells were incubated for 2 hours in the presence of 2 mmol/L

[U-‘%]glutamine

2 glucose in Krebs-

Ringer HEPES buffer. Values are means -t SEM.

ATP Concentrations in Splenocytes

ATP levels in splenocytes incubated in the presence or absence of glutamine and glucose are presented in Table 5. In unstimulated cells, glucose sustained ATP concentration over the 2-hour incubation period. However, ATP was decreased (P < .Ol) in these cells incubated with glutamine alone at the end of the 2-hour incubation. In PMA + Iono-stimulated splenocytes, either glutamine or glucose alone maintained ATP levels during the 2-hour incubation at levels found before incubation (0 hours). In the presence of both glutamine and glucose, ATP levels were increased (P < .Ol) at both 1 and 2 hours of incubation above those found before incubation. To check whether endogenous fuels are adequate to supply ATP to splenocytes, we measured ATP levels in both unstimulated and stimulated cells incubated for 1 to 2 hours in the absence of exogenous substrates. ATP concentrations (pmol/106 cells, mean t SEM, n = 6) were 128 2 22, 74 ? 6, and 35 t 7 (P < .Ol) in unstimulated cells and 430 2 22, 99 5 7.1, and 19 ? 1.2 (P < .Ol) in PMA + Iono-stimulated cells at the end of 0, 1, and 2 hours of incubation, respectively.

Abbreviation: ND, not detectable.

DISCUSSION

*P < .Ol, significantly different from control (no glucose group) as analyzed by paired 1 test. tP < .Ol, significantly different from the means for unstimulated cells as analyzed by unpaired t test. SCalculated by dividing WO,

by the sum of glutamine carbons in

%Oz, glutamate, and aspartate.

production from glutamine by 47%, whereas glutamine decreased (P < .Ol) that from glucose by 25%. In these unstimulated cells, the potential production of ATP from glutamine was approximately half that from glucose when present alone or together with glucose. In stimulated splenocytes, potential ATP production from glutamine was increased (P < .Ol) by 138% (-glucose) and 178% (+glucose). Glutamine could provide almost as much ATP as glucose when present alone or together with glucose. The oxidation of glutamine to CO* via the Krebs cycle was the major pathway for ATP production from glutamine, accounting for the major part (76%) of glutamine-derived ATP in the presence of glucose. On the other hand, the potential ATP production from glucose in stimulated cells was increased (P < .Ol), but to a lesser extent, by 55% (-glutamine) and 60% (+glutamine). This increase in glucose-derived ATP was only due to the increase in the conversion of glucose to lactate rather than to glucose oxidation via the Krebs cycle. As a result, the glycolysis of glucose to lactate could account for 72% to 74% of glucose-derived ATP. Glucose decreased (P < .Ol) potential ATP production from glutamine by 38%, whereas glutamine decreased (P < .Ol) that from glucose by 27%. It is noteworthy that the potential production of ATP per lo6 cells in the presence of both glutamine and glucose was greater (P < .Ol) than that in the presence of either substrate alone.

Requirement of Glutamine for Stimulation of 3H-Thymidine Uptake by PMA + Iono

PMA, which activates protein kinase C, and Iono (a Ca2+ ionophore), which increases the intracellular calcium concentration, act synergistically to stimulate lymphocyte proliferation, as assessed by 3H-thymidine uptake.26J7 We have also found that PMA + Iono enhanced 3H-thymidine uptake by splenocytes more than 500-fold in the presence of 0.25 to 4.00 mmol/L glutamine (Table 1). However,

Table 3. Glucose Metabolism in PMA + Iono-Stimulated

and

Unstimulated Splenocytes 5 mmol/L

Condition

h per 10scells) +2 mmol/L

Glutamine

Glutamine

‘TO2

_

IO

4.5 + 0.3

3.3 -t 0.2*

8

3.9 * 0.2

2.7 k 0.3*

Lactate

+ _

IO

27.6 2 2.4

21.2 k 2.1*

+ _

8

64.6 + 6.0t

47.7 k 5.7*t

Pyruvate

IO

2.55 k 0.20

2.38 k 0.25

8

2.41 2 0.29

2.43 k 0.37

% of metabolized

+ _

IO

4.9 + 0.4

2.0 t 0.2*

+

8

4.7 + 0.5

1.8 k 0.5*

glucose carbons

+ lono)

NO n

Product

(?PMA

IU-‘4CIGlucose

(nmol products/2

Culture

in CO$ NOTE. Splenocytes were cultured for 48 hours in the presence or absence of PMA + lono. After culture, cells were incubated in the presence of 5 mmol/L

[U-%]glucose

? glutamine in Krebs-Ringer

HEPES buffer. Values are means + SEM. *P < .Ol, significantly different from control (no glutamine group) as analyzed by paired t test. tP < .Ol, significantly different from the means for unstimulated cells as analyzed by unpaired t test. *Calculated by dividing ‘%02 by the sum of glucose carbons in 14C02. lactate, and pvruvate.

986

WU, FIELD, AND MARLISS

Table 4. Potential ATP Production From Glutamine and Glucose in PMA + lono-Stimulated

and Unstimulated Splenocytes

ATP From Glutamine (nmoli2 Medium

Substrate

Gin to CO?

Gin to Asp

ATP From Glucose

h per 106cells)

Inmoll2 Total

Glucose to Lactate

h per lo6 cells)

Glucose to CO,

Total

Unstimulated cells Glutamine (2 mmol/L)

14.4 2 1.1*

24.4 + 1.6*

38.8 t 1.5*

6.6 k 0.3

14.3 2 0.6

20.8 k 0.8

Glutamine (2 mmol/L) + glucose (5 mmol/L) Glucose (5 mmol/L)

21.2 i 2.1

20.5 -t 1.3

41.7 2 4.3

27.6 2 2.4*

28.2 2 2.1*

55.9 t 4.4*

Stimulated cells Glutamine (2 mmol/L)

35.8 k 2.2*t

46.5 + 4.4t

82.3 k 4.9*t

13.5 k 1.0t

43.4 k 6.9t

56.8 2 7.2t

Glutamine (2 mmol/L) + glucose (5 mmol/L) Glucose (5 mmol/L)

47.7 2 5.7t

17.2 k 1.8

64.7 -t 6.9t

64.6 t 6.0Xt

24.8 k l.4*

89.4 k 6.8*t

NOTE. Data were calculated based on those in Tables 2 and 3. Values are mean k SEM, with the number of animals indicated in Tables 2 and 3. *P < .Ol, significantly different from the corresponding means obtained for glutamine (2 mmol/L) plus glucose (5 mmol/L) group as analyzed by paired t test. tP < .Ol, significantly different from the corresponding means for unstimulated cells as analyzed by unpaired t test. The potential ATP production from glutamine in unstimulated cells was significantly different (P < .Ol) from that from glucose in the presence of both substrates as analyzed by paired t test.

PMA + Iono did not increase 3H-thymidine uptake in cells cultured in glutamine-free media. This would be predicted from the essential role of glutamine for the synthesis of macromolecules.‘” We observed that ‘H-thymidine uptake by stimulated splenocytes was half-maximal at 0.14 mmol/L glutamine (below normal plasma concentration), and was maximal at initial culture medium glutamine concentration, ranging from 0.5 mmol/L (close to normal plasma concentration) to 4 mmol/L (Table 1). By contrast, it was reported in cultured mesenteric lymph node lymphocytes stimulated by concanavalin A (Con A)4 or phytohemagglutininl” that 3H-thymidine uptake was maxima1 at 0.3 mmol/L glutamine and then decreased progressively as the glutamine concentration increased. This discrepancy may be due to the use of different mitogens and/or lymphocyte preparations. Metabolic Fates of Glutamine and Glucose and Their Mutual Interactions It appears that the major products of glucose are lactate, pyruvate, and CO?, while those of glutamine are ammonia, Table 5. ATP Concentrations

in PMA + lono-Stimulated

and

Unstimulated Splenocytes Unstlmulated Addition of Substrate

Incubation

to Medium

Time(h)

2 mmol/L Gln

CfZllS (pmoli106ceIls)

0

128.2 + 21.7 (6)

1

94.7 f 5.7 (5)

2

Stimulated Cells ~pmol/106cells)

430.0 2 22.1 t (9) 404.7 2 41.8t (13)

70.8 + 5.5* (10) 434.1 f 32.4t (10)

1

114.6 t 12.0 (5)

388.8 r 29.9t (13)

2

103.6 + 6.6 (10)

412.2 -t 30.3t (10)

2 mmol/L Gln +

1

136.1 r 10.3 (5)

599.2 + 36.4*t

(9)

5 mmol/L glucose

2

127.7 + 10.2 (10)

660.1 2 36.0*t

(10)

5 mmol/L glucose

NOTE. ATP in splenocytes was measured as described in the text. Values are mean r SEM with the number of animals indicated in parentheses. “P

< .Ol, significantly different from control (0 h) as analyzed by

one-way ANOVA. P < .Ol. significantly different from the means obtained for unstimulated cells as analyzed by unpaired t test.

glutamate. aspartate, and CO? in cultured splenocytes (Tables 2 and 3). We have demonstrated a reciprocal decrease in the products of the ATP-generating pathway involved in splenocyte glutamine and glucose metabolism, in the presence of either glucose or glutamine (Tables 2 and 3). Inhibition by glucose of glutamine metabolism, particularly aspartate production, has also been reported in Con A-stimulated and unstimulated thymocytes,7,8 tumor cells,‘x and cultured mammalian cells.‘9 Similarly, glutamine has also been reported to decrease glucose oxidation via the Krebs cycle in human diploid fibroblasts.?y In contrast. glutamine was claimed to markedly increase the production of lactate, pyruvate, and CO1 from glucose in both Con A-stimulated and -unstimulated splenocytes.” Similarly, Ardawi and Newsholme4 reported reciprocal stimulation of glucose and glutamine metabolism in lymph node lymphocytes. The reason for this discrepancy is not evident at present. We have suggested that glucose probably dccreases aspartate production by increasing formation of citrate from glucose-derived acetyl-CoA and glutaminederived oxaloacetate, thus decreasing the availability of the latter for transamination.‘J,‘5 Accordingly, an increase in citrate may contribute to a decrease in glycolysis, as has been suggested for lymphocytes.“’ It is noteworthy that Sri-Pathmanathan et al” have recently proposed that glucose decreases glutamine metabolism in cultured mammalian cells by decreasing intracellular phosphate concentrations, which would result in decreased activity of phoaphate-dependent glutaminase (a rate-limiting enzyme in glutamine metabolism). Whether this mechanism applies to cultured splenocytes remains to be determined. We did try to measure glutamine uptake or utilization by cultured splenocytes. However, because the amount of glutamine disappearing from the incubation medium was very small ( N 35 to 70 nmol) as compared with the amount of glutamine (2,000 nmol) present initially in the medium, we could not determine glutamine uptake or utilization with sufficient precision by this approach. Using higher cell concentrations. previous studies have shown that almost all

GLUTAMINOLYSIS

AND GLYCOLYSIS

IN LYMPHOCYTES

of the glutamine carbons used can be recovered as COZ, glutamate, and aspartate in splenocytes (eg, ref. 11). Therefore, glutamine metabolism via pathways other than the classic mitochondrial-localized enzyme, ie, via cytoplasmic glutamine transaminase or membrane-bound gamma glutamyltranspeptidase, if not impossible, must be quantitatively insignificant in splenocytes, because these alternative pathways would not generate glutamate.3Z Potential Roles of Glutamine and Glucose as Energy Substrates

The present study suggests that glutamine may not be quantitatively as important as glucose in the provision of ATP for unstimulated cultured splenocytes, on the basis of the following evidence. First, the potential production of ATP from glutamine was much lower than that from glucose when present alone or together with glucose (Table 4). Second, when present alone, glucose, but not glutamine, sustained ATP levels in unstimulated cells (Table 5). These results are consistent with the reduced rates of glutamine metabolism in unstimulated cultured rat splenocytes (Table 2) as compared with our previous results in freshly isolated cells.15 This may result, in part, from a change in cell populations in cultured splenocytes. For example, macrophages may adhere to culture plates, and therefore fewer are likely to be present in splenocytes studied after culture for 48 hours than in freshly isolated cells. In contrast to unstimulated cultured cells, glutamine could provide as much ATP as glucose in PMA + Ionostimulated splenocytes (Table 4). This is due to twofold or threefold increases in glutamine metabolism to all of its measured products (Table 2). The oxidation of glutamine via the Krebs cycle could account for the major part (76%) of glutamine-derived ATP in these cells in the presence of glucose. On the other hand, glycolysis to lactate was the major pathway for glucose-derived ATP production in PMA + Iono-stimulated splenocytes (Table 4). These results are not consistent with the suggestion of O’Rourke and Rider” that glutamine made only a minor contribution to ATP production and that oxidation of glucose to CO1 could account for the major part of glucose-derived ATP in Con A-stimulated splenocytes. The use of different mitogens (Con A stimulates mainly T cells, whereas PMA + Iono stimulates many splenocyte subsets) is unlikely to account for this discrepancy, as we have also found that glycolysis to lactate is the major pathway for glucose-derived ATP production in Con A-stimulated cultured splenocytes.‘” The most compelling reason for the discrepancy between our results and those of O’Rourke and Rider” is that splenocytes cultured with Con A for 4 and 22 hours in their study would not have reached a stage of maximal mitogenic stimulation, since this usually requires 96 hours of culture.i9 By contrast, cells used in our metabolic studies were confirmed to be maximally stimulated by PMA + Iono at 48 hours of culture (Fig 1). ATP Levels in Splenocytes

Cellular ATP levels depend on the balance between its rate of production and utilization. Along with data on calculated potential ATP production from a given sub-

987

strate, measured ATP concentration can allow for inferences about its capacity to serve as an energy substrate.14.33 On the basis of progressively decreasing ATP levels in rat chondrosarcoma cells incubated in the presence of glutamine alone, in a manner similar to that found in the absence of exogenous substrates, Spencer et a133suggested that glutamine was not a significant energy substrate for these cells. We likewise observed that in the absence of exogenous substrates, a progressive decrease in ATP levels occurred in both unstimulated and stimulated splenocytes at the end of l- and 2-hour incubations (See Results). This suggests that the supply of endogenous substrates is not adequate to provide sufficient amounts of ATP. By contrast, glutamine, like glucose, sustained ATP levels in stimulated splenocytes (but not in unstimulated cells) during the 2-hour incubation. This would also have been predicted on the basis of the potential ATP production from glutamine and glucose in stimulated cells (Table 4). Thus, our findings, taken together with data on potential ATP production, suggest that glutamine can be quantitatively as important as glucose in provision of energy for mitogen-stimulated lymphocytes, at least under the experimental conditions used here. Potential Implications of Increased Glutamine and Glucose Utilizationby Stimulated Cells of the Immune System

It appears from the present study that in PMA + Iono-stimulated splenocytes, as glutamine metabolism was markedly increased, it became a quantitatively important energy substrate. As a result, glutamine plays a dual metabolic role in these cells, serving as an important energy source via oxidation and providing carbon and nitrogen precursors for biosynthetic processes. It is not clear whether glutamine increases thymidine uptake by serving as a precursor for nucleotide synthesis or as an energy source that indirectly facilitates thymidine incorporation. Nevertheless, our findings may be of importance in understanding fuel homeostasis in pathological states associated with marked increases in the metabolic activity of cells of the immune system, such as injury, sepsis, and inflammation. The implication for greater glutamine utilization by immunocytes under these stress conditions is that if glutamine is not provided at adequate rates from endogenous (mainly skeletal muscle) or exogenous sources, functional limitations of immunocytes could hypothetically result.‘4 Indeed, glutamine has been recently suggested to be a “conditionally essential” amino acid in such settings.35 Since the body’s requirements for glutamine appear to exceed the individual’s ability to produce sufficient amounts of this amino acid during severe illness,35 the provision of supplemented glutamine, such as in the form of t_-alanyl-tglutamine or other glutamine-containing dipeptides in specialized enteral or parenteral feeding,“h might improve or maintain the body’s nutritional status and the function of immunologically active cells. Likewise, the increased metabolism of glucose by these cells, which largely return glucose carbon skeletons to the circulation as 3-carbon intermediates (especially as lactate), suggests its potential effects on in vivo glucose homeostasis, the Cori cycle, and other pathways for disposal of lactate (eg, gluconeogenesis).

WU, FIELD, AND MARLISS

988

Given the estimated 1.5 kg of lymphocytes in an adult human,37 and more in disease states such as severe infections and lymphomas, a substantial metabolic “burden” in the form of disposing of large amounts of lactate produced from glucose by immunocytes could be imposed. This suggests the importance of devising methods for quantitatively assessing the contribution by cells of the immune system to the Cori cycle and gluconeogenesis in vivo, in these stress states and certain other diseases such as

lymphomas

and leukemias

that produce elevated

blood

lactate levels.

ACKNOWLEDGMENT

The authors wish to thank M. Montambault, M. Faucher. and A. Gunasekara for their excellent technical assistance. G.W. and C.J.F. were recipients of postdoctoral fellowships from the Medical Research Council of Canada.

REFERENCES 1. Newsholme EA, Newsholme P, Curi R: The role of the citric acid cycle in cells of the immune system and its importance in sepsis, trauma and burns. Biochem Sot Symp 54:145-161, 1987 2. Wu G. Field CJ. Marliss EB: Energy metabolism in cells of the immune system and its abnormalities in the spontaneously diabetic BB rat, in Shafrir E (ed): Lessons From Animal Diabetes III. London. UK, Smith-Gordon, 1990, pp l-10 3. Eagle H: Amino acid metabolism in mammalian cell cultures. Science 130:432-437,1959 4. Ardawi MSM, Newsholme EA: Glutamine metabolism in lymphocytes of the rat. Biochem J 212:835-842.1983 5. Crawford J. Cohen HJ: The essential role of L-glutamine in lymphocyte differentiation in vitro. J Cell Physiol124:275-282. 1985 6. Ardawi MSM, Newsholme EA: Metabolism in lymphocytes and its importance in the immune response. Essays Biochem 21:1-44, 1985 7. Brand K: Glutamine and glucose metabolism during thymocyte proliferation. Biochem J 228:353-361, 1985 8. Brand K. Fekl W. von Hintzenstern J, et al: Metabolism of glutamine in lymphocytes. Metabolism 38:29-33, 1989 (suppl 1) 9. Newsholme EA, Crabtree B, Ardawi MSM: Glutamine metabolism in lymphocytes: Its biochemical, physiological and clinical importance. QJ Exp Physiol70:473-489, 1985 10. Szondy Z, Newsholme EA: The effect of glutamine concentration on the activity of carbamoyl-phosphate synthase II and on the incorporation of [3H]thymidine into DNA in rat mesenteric lymphocytes stimulated by phytohaemagglutinin. Biochem J 261: 979-983,1989 11. O’Rourke AM, Rider CC: Glucose, glutamine and ketone body utilization by resting and concanavalin A activated rat splenic lymphocytes. Biochim Biophys Acta 979:77-81, 1989 12. Kovacevic Z, McGivan JD: Mitochondrial metabolism of glutamine and glutamate and its physiological significance. Physiol Rev 63:547-605.1983 13. Krebs H: Glutamine metabolism in the animal body, in Mora J, Palacios E (eds): Glutamine: Metabolism. Enzymology, and Regulation. New York, NY, Academic, 1980, pp 319-329 14. Wu G, Field CJ, Marliss EB: Glutamine and glucose metabolism in rat splenocytes and mesenteric lymph node lymphocytes. Am J Physiol260:E141-E147, 1991 15. Wu G. Field CJ, Marliss EB: Elevated glutamine metabolism in splenocytes from spontaneously diabetic BB rats. Biochem J 274149-54, 1991 16. Hume DA, Radik JL, Ferber E, et al: Aerobic glycolysis and lymphocyte transformation. Biochem J 174:703-709, 1980 17. Hume DA, Weidemann MJ: Mitogenic Lymphocyte Transformation. New York, NY, Elsevier/North-Holland, 1980 18. Levintow L, Eagle H: Biochemistry of cultured mammalian cells. Annu Rev Biochem 30:605-640, 1961 19. Field CJ, Wu G, MCtroz-Dayer M-D. et al: Lactate production is the major metabolic fate of glucose in splenocytes and is altered in spontaneously diabetic BB rats. Biochem J 272:445-452. 1990

20. Field CJ, Chayoth R, Montambault M, et al: Enhanced glucose transport and metabolism in splenocytes from diabetic and diabetes-prone BB rats: Further evidence to support prior in vivo activation. J Biol Chem 266:3675-3681, 1991 21. Lloyd B, Burrin J. Symthe P, et al: Enzymatic fluorometric continuous flow assays for blood glucose, lactate, pyruvate, alanine. glycerol and 3-hydroxybutyrate. Clin Chem 24:1724-1729, 1978 22. Bergmeyer HU. Beutler H-O: Ammonia, in Bergmeyer HU (ed): Methods of Enzymatic Analysis, ~018. Weinheim, Germany. Verlag Chemie, 1985, pp 454-461 23. Strehler BL: Adenosine-5’.triphosphate and creatine phosphate determination with luciferase. in Bergmeyer HU (ed): Methods of Enzymatic Analysis. New York, NY. Academic. 1974. pp 2112-2126 24. Wulff K, Doppen W: ATP: Luminometric method, in Bergmeyer HU (ed): Methods of Enzymatic Analysis, ~017. Weinheim. Germany, Verlag Chemie. 1985. pp 357-364 25. Stanley PE. Williams SG: Use of the liquid scintillation spectrometer for determining adenosine triphosphate by the luciferase enzyme. Anal Biochem 29:381-392, 1969 26. Berry N. Ase K, Kikkawa U. et al: Human T-cell activation by phorbol esters and diacylglycerol analogues. J Immunol 143:14071413.1989 27. Truneh A, Albert F, Golstein P. et al: Early steps of lymphocyte activation bypassing by synergy between calcium ionophores and phorbol ester. Nature 313:318-320, 1985 2X. Kovacevic Z: The pathway of glutamine and glutamate oxidation in isolated mitochondria from mammalian cells. Biochem J 1251757-763. 1971 29. Zielke HR. Sumbilla CM, Zielke CL, et al: Glutamine metabolism by cultured mammalian cells, in Haussinger D, Sies H (eds): Glutamine Metabolism in Mammalian Tissues. Berlin, Germany, Springer-Verlag. 1984, pp 247-254 30. Ardawi MSM. Newsholme EA: Metabolism of ketone bodies. oleate and glucose in lymphocytes of the rat. Biochem J 221:255-260, 1984 31. Sri-Pathmanathan RM, Braddock P. Brindle KM: j’P-NMR studies of glucose and glutamine metabolism in cultured mammalian cells. Biochim Biophys Acta 1051:131-137. 1990 32. Meister A: Biochemistry of the Amino Acids. New York. NY. Academic, 1965, pp 621-636 33. Spencer CA, Palmer TN, Mason RM: Intermediary metabolism in the Swarm rat chondrosarcoma chondrocytes. Biochem J 265:911-914, 1990 34. Newsholme EA, Parry-Billings M: Properties of glutamine release from skeletal muscle and its importance for the immune system. JPEN 14:63S-67s. 1990 35. Lacey JM, Wilmore DW: Is glutamine a nutritionally essential amino acid? Nutr Rev 48:297-309. 1990 36. Furst P. Albers S. Stehle P: Glutamine containing dipeptides in parenteral nutrition. JPEN 14:118S-124s. 1990 37. Snyder WS, Cook MJ. Nasset ES, et al: Report of the Task Group on Reference Man. Oxford. UK, Pergamon. 1984. pp 98-99