Molecular and Cellular Biochemistry 102: 149-154, 1991. © 1991 Kluwer Academic Publishers. Printed in the Netherlands.
Respective effects of glucose and glutamine on the glutamine synthetase activity of human skin fibroblasts Thdophile Soni 1, Claire Wolfrom 2, Samia Guerroui 1, Nicole Raynaud 2, Jos6phine Poggi 1, Nicole Moatti I and Marthe Gautier z 1Facultd des Sciences pharmaceutiques et biologiques. Universitd Paris-Sud, Rue J.B. Clement, 92290 Chatenay-Malabry, France; 2 Unit~ de Recherche d'Hdpatologie Pddiatrique, INSERM U56, HOpital de BicOtre, 94270 Le Kremlin-BicOtre, France Received 18 May 1990; accepted 7 September 1990
Key words: glutamine synthetase, human fibroblasts, glucose, glutamine Summary The activity of Glutamine Synthetase (GS) was measured during the growth of human diploid skin fibroblasts cultured for three weeks in the presence or absence of either glucose or glutamine or both. In medium free of both glucose and glutamine, a single late peak in GS activity was observed concomitantly with delayed small cell protein increment. In all media containing either glucose or glutamine or both, GS activity rose sharply during rapid cell growth, displayed a plateau, and then decreased once the cells had reached confluency. The variations in extracellular aminoacid levels were also determined and were found to depend on the composition of the medium but not on the cell culture duration. These results demonstrate, for the first time as far as we know, that strong GS activity is present in rapidly growing skin fibroblasts. In contrast to many other mammalian cell types, GS activity in human skin fibroblasts appears not to be subject to regulation by extracellular glutamine. This difference may well be connected with cell differentiation.
Introduction Glutamine Synthetase (GS), i.e. (L-glutamate: ammonialigase (ADP), EC 6.3.1.2), catalyses the reversible formation of glutamine from glutamate, ammonia and ATP. This reaction plays a key role in living cell metabolism because the amide group originating from glutamine is used as the nitrogen source for the biosynthesis of various aminoacids, NAD, purine and pyrimidine nucleotides, and carbamylphosphate. Glutamine is also a maj or energy source for cultured ceils [1], and in human diploid fibroblasts this has been conclusively demonstrated by metabolic approaches [2-4]. One recent investigation showed, in particular, that the concentration of intracellular glutamine is regulated by active
uptake of extracellular glutamine, and that in glutamine-free medium containing glutamate and glucose, only one quarter of the intracellular glutamine is derived from intracellular exchangeable glutamate. Therefore compartmentation of the glutamate pool has been postulated to explain the intracellular glutamine obtained from other glutamate fractions [4]. However, it must be remembered that glutamine synthesis requires both glutamate and ammonium, which are the substrates of GS. GS activity has been demonstrated in several types of mammalian cultured ceils, mainly from rodents [5, 6], including astrocytes [7], and the embryonic human lung fibroblast strain WI-38 [8, 9]. Extracellular glutamine has been reported to repress GS activity [8, 10, 11]. We addressed the
150 question of whether GS activity is regulated by extracellular glutamine in long-term cultures of human skin fibroblasts. Another question we set out to clarify was how normal diploid human skin fibroblasts could slowly grow in hexose-free medium without glutamine, as we previously observed [12]. Accordingly, in this investigation, we measured, for the first time as far as we know, GS activity during the growth of human diploid fibroblasts, in the presence or absence of either glucose or glutamine, or both. In addition, we assessed the concomitant variations in extracellular aminoacid levels.
Materials and methods
Cel/s Human fibroblasts were obtained from a skin biopsy taken from a one month-old infant during surgery. Cultures were initiated from explants as already described [13]. Cells were initially cultured in Minimum Essential Medium (MEM 2111, Eurobio, France) containing nonessential aminoacids, and 5.5 mM D-glucose, 2mM glutamine and 10% fresh undialyzed human serum [12, 14]. At the 6th passage, cells were trypsinized and transferred into fifty 75 c m 2 replicate flasks (T75). Plating density was 4 × 105 cells per flask containing 10 ml of medium. At day 0, flasks were fed with one of the four following media: MEM containing 5.5mM glucose, either with 2 mM glutamine (Glg +) or without ( G l g - ) , and MEM containing no Hexose, either with 2mM glutamine (HOg +) or without glutamine ( H O g - ) . However, since non dialyzed serum was used, the 10% human serum in the supplemented media supplied about 0.2 mM glucose and about 0.05mM glutamine. The culture fluid was changed every 3rd or 4th day. As the cells grown in H O g - medium exhibited a very poor growth rate, this series of cultures was pursued for 27 days instead of 21, and the enzymatic assays were not, as in the other cultures, carried out every 3rd or 4th day, but at longer intervals. Observation of monolayers in phase contrast was carried out with a Wild inverted microscope.
Aminoacid determination
Before refeeding, media were sampled every 3 or 4 days, for aminoacid determination by ion exchange chromatography (Technicon T.S.M. with a Technation integrator). The percentage of variation in aminoacid levels were established from the results obtained for the corresponding reference media.
Enzymatic assay
Cells were harvested by scraping into I ml of distilled water and homogenized by sonication for 5 sec at 4° C. The pooled cell content of 1 to 4 T75 flasks was necessary for each determination. Glutamine synthetase activity was measured in duplicate by a micromethod adapted from that described by Ward and Bradford [15] without Triton X100. Aliquots of 120 ~1 of homogenate were preincubated for 5 min at 37° C before the reaction was started by adding 100 ~1 of incubation mixture. The final concentrations used were as follows: 100raM imidazole buffer (pH 7.2), 12.5mM MgC12, 20raM mercaptoethanol, 10 mM ATP, 50 mM sodium glutamate, 1 mM ouabain, 100 mM hydroxylamine (freshly neutralized to pH 7.2 with KOH or NaOH), 13 mM phosphoenol pyruvate, and 3.3 UI/ ml pyruvate kinase. After 60min incubation at 37° C, the reaction was stopped by adding 500 ~1 of a fresh mixture containing 50 mM FeCI3, 400 mM perchloric acid, and 400 mM HC1. After the tubes had been left to stand at 4° C for 20-30 minutes, the protein precipitate was sedimented by centrifugation at 4000 rpm and 4°C for 10 min. The optical density of the supernatant was measured directly at 500nm and 25°C in a Beckman spectrophotometer. Standard curves were established using glutamyl-hydroxamate (Sigma). Values are expressed as ~mol of glutamine formed per hour per mg of protein at 37° C.
Protein content
As it was considered impossible to perform on this same cell line parallel experiments for enzyme as-
151
2000 []
/ 2 c~
E100C _c
o
¢33
E
1
c
-- -- --ND-- - - NO-- - - - ' ~ [ > . . . .
I 3
I 7
] 10
[ I 14 18 Days
L 21
[ 24
L 27
0
ND"- "~
....
ND, --
i
I
I
J
I
I
I
]
3
7
10
14
18
21
24
27
Days
Fig. 1. Specific activity of glutarnine synthetase (GS) in diploid
Fig. 2. Protein increment of diploid infant human flbroblasts.
infant human flbroblasts grown in four culture media: glucose medium with glutamine (Glg+ • - - • ) or without glutamine ( G i g - IE--[]), and hexose-free medium with glutamine (HOg + O--O) or without glutamine (HOg - ~Y--~). Each point represents the mean of two determinations.
Data points represent the mean cell protein content per flask (T75), in the set of flasks used for each determination of glutamine synthetase (GS) activity, under four culture conditions: glucose medium with glutamine (Glg+ • - - • ) or without glutamine (Glg- []--i~), and hexose-free medium with glutamine (HOg + I ~ - O ) or without glutamine (HOg - 4>--~). Protein content of the homogenates was determined in duplicate.
says, and for assessing cell growth as cell numbers, due to the limited amount of cells at this early passage, the cell protein increment was used to estimate cell growth, in every set of flasks used for enzyme assays. Protein content of the homogenates was determined in duplicate according to Bradford [16] using bovine serum albumin (Sigma) as standard.
Results
and
discussion
The fluctuations in GS specific activity and the corresponding total cellular protein increment measured during cell growth (Figs 1 and 2), demonstrated that normal diploid fibroblasts are able to synthesize glutamine. In addition, we showed the related evolution of the kinetics and levels of GS activity and that of cell protein increment. W h e n cells were cultured in the presence of glucose, i.e. in conventional Gig + m e d i u m as well as in G i g - medium, GS specific activity increased
m o r e than 4-fold from days 3 to 7 of culture. A plateau was then observed for this activity for 7-10 days, and was followed by a slight decrease which coincided with the high protein content of cells in their late stationary growth phase. These results for Gag + medium do not agree with those of previous investigations showing that GS synthesis by various cell types was repressed in the presence of glutamine [10]. For glutamine-free medium ( G i g - ) , our results for normal h u m a n fibroblasts are in agreement with the findings for 3T3-L1 fibroblasts, in which glutamine deprivation had no effect on GS activity [17]. However, they diverge from those obtained for other cell lines, including the embryonic h u m a n lung WI-38 fibroblasts in which GS activity was raised by depriving cells of glutamine [8] and for a N a m a l v a cell line, in which GS activity increased when cells were adapted to continuous growth in glutamine-free m e d i u m [11]. For both G1
152 media, our results showed that the presence of extracellular glutamine, however readily the latter was utilized by the cells, had no effects on GS specific activity, and neither had its absence. A homology was recently established between mammalian and plant GS [18]; in addition, at least 3 distinct GS-encoding genes with different structures, as well as different patterns of GS expression in vivo, were reported in the pea [19]. Consequently, the observation of different patterns of GS expression in different cell types might be connected with cell differentiation. When our cells were cultured in glucose-free medium containing glutamine (HOg +), GS specific activity increased by day 7 and then remained constant, but at lower levels than in Gig + medium. During the first 14 days of culture in HOg medium cells did not multiply and some were necrosed which resulted in a low steady cell protein content, and no GS specific activity was detected; however, by day 21, a small peak in GS activity was present at the same time as cells recovered a normal appearance and started multiplying. This improvement might have resulted from the release of nutrients from damaged cells in the extracellular medium. However this hypothesis seems unlikely, since most of the cells were surviving, and the culture medium was renewed every 3rd or 4th day. The selection for a specific cell population is another possibility. Although these starved cells in HOg - medium had no major energy supply, they were able here to display transient GS activity for de n o v o synthesis of glutamine from the glutamate pool, even at very low cell density. Therefore, under our conditions of culture, cell protein increment correlated with the increase in GS specific activity, and the high cell protein content which was observed at cell confluency coincided with the decrease in this activity. A similar correlation was previously demonstrated in cultured mouse teratoma cells, [5, 6]. These parallel increases in GS specific activity and cell density were comparable to those previously observed for glutaminase activity, which was shown to be high in rapidly dividing human fibroblasts, irrespective of the glutamine concentration [20]. Figure 3 shows the corresponding variations ob-
served in the levels of extracellular nonessential aminoacids. These variations were similar to those we previously observed in human fibroblasts after 3 days of culture in MEM [21] and appeared to be closely related to the glucose and glutamine content of the medium. Thus in the presence of glucose and glutamine (Gig + medium), accumulation of alanine, glycine, proline, and glutamate was marked. After 4 days of culture, MDCK cells displayed a pattern of aminoacid variations similar to that of the present human fibroblasts, except for proline and aspartic acid, for which no change was observed [22]. Here, when glucose alone was omitted from the medium (HOg +), alanine production was greatly reduced, as its level is related to glutamine consumption [23], and also to pyruvate availability, which decreases in the absence of glucose. When glutamine was omitted from the medium (Gig - and HOg - ) , alanine and proline were utilized by the cells. Production of extracellular glutamate was observed in G l g + and H O g + media, and more surprisingly in HOg - medium. In fact, for human fibroblasts, glutamine has been clearly established as a major source of energy [1, 2, 24,25]. In conventional media, the large excess of extracellular glutamine (2 mM), which is rapidly depleted because of an efficient glutamine uptake [4], provides an ample source of energy. The glutaminolysis first step is the glutamine-toglutamate step, by glutaminase or amidotransferase, which appear to be essentially irreversible reactions. The subsequent steps of the pathway mostly involve mitochondrial enzymes [26] of the TCA cycle. A severe energy deprivation, as in our H O g - medium, might block these latter steps, which would shift the equilibrium of cytosolic aminotransferase towards the production of energy substrates such as pyruvate and oxaloacetate, and of glutamate. Conversely, a large excess in energy supply, as in our Gag + or HOg + media, might overflow these mitochondrial steps and lead to the accumulation of glutamate, which may in turn be converted to proline and to a lesser extent alanine. When only glucose is present (i.e. Glg - ) all extracellular glutamate available is taken up and additional glutamate may be supplied through the transamination of alanine and proline, whose con-
153 TGlg
+
712
.
400 200-
davs 21 100 2
~S
"l'°"I[II III[
-"
II
200
-
50n~J.
050100Ala
Gly
Pro
Glu
Gin
Asn
Ser
Tyr
Asp
Cys
t'°° Ala
% 400-
Glg-
.,,
i
III
100-
Pro
Glu
Gin
ASh
S~r
Tyr
Asp
Cys
400-
•3
200-
Gly
"&;~'~'
200
L
][L
x.I
days
27
50-
n.d
O" I It -50100Ala
Gly
Pro
Glu
Gin
Asn
Ser
Tyr
Asp
Cy5 Ala
GIy
Pro
Glu
Gln
Asn
Ser
Tyr
Asp
Cys
Fig. 3. Bar graphs represent selective aminoacid consumption (below abscissa) or production (above abcissa in the media of diploid infant human fibroblast cultures. Results are expressed in percentages of aminoacid consumption and production versus control media. Insert on the right shows cell protein content at the different days of culture. For details, see Fig. 2. Data are not related to the number of cells. In the presence of glutamine (Glg + and HOg + media) alanine, glucine, proline and glutamate were constantly produced, and the other aminoacids constantly consumed. The only marked differences concerned the levels of alanine and glutamate production. In the absence of glutamine, the difference concerned glutamate, which was entirely consumed in Glg - medium but produced in HOg medium. Consumption of all the other aminoacids was less in HOg - medium, which may reflect a low protein content per flask.
centrations in the medium were shown here to decrease. Some of the extracellular glutamine may also be chemically converted to pyrrolidone carboxylate in the medium [22]. In all 4 media, glycine accumulation was constant, probably owing to the consumption of cysteine and serine. No marked variations in the levels of essential aminoacids were noted (data not shown), except that methionine, leucine and isoleucine consumption were related to the cell growth rate. The rapid depletion of these 2 latter branched-chain aminoacids seems to be an important factor in this growth [22]. In conclusion, the role of extracellular glutamine as a source of energy is highlighted by these patterns of aminoacid metabolism in different media. When another source of energy is present, like glucose, the absence of glutamine is compensated by other aminoacids, increment in cell protein is
normal, and GS activity is unchanged. Therefore our hypothesis is that glutamine synthesis is regulated so as to fulfill the priority requirement for glutamine for protein and nucleic acid biosynthesis. This contrasts with glutaminolysis, which appears to be regulated so as to fuel oxidative metabolism concurrently with glucose. This essential anabolic role of glutamine synthesis requires a GS activity associated with cell proliferation, and independent from the provision of exogenous glutamine. The hypothesis that this newly synthesized glutamine exists as a separate pool should be tested. The recent cloning of eukaryotic GS genes [19] should help to provide more information about the signals that regulate GS activity in human fibroblasts.
154 References 1. Zielke HR, Zielke CL, Ozand PT: Glutamine: a major energy source for cultured mammalian cells. Federation Proc 43: 121-125, 1984 2. Reitzer LJ, Wice BM, Kennell D: Evidence that glutamine, not sugar, is the major energy source for cultured HeLa cells. J Biol Chem 254: 2669-2676, 1979 3. Zielke HR, Sumbilla CM, Sevdalian DA, Hawkins RL, Ozand PT: Lactate: a major product of glutamine metabolism by human diploid fibroblasts. J Cell Physiol 104: 433441, 1980 4. Darmaun D, Matthews DE, Desjeux JF, Bier DM: Glutamine and glutamate nitrogen exchangeable pools in cultured fibroblasts: a stable isotope study. J Cell Physiol 134: 143-148, 1988 5. Oppenheimer SB: Utilization of L-glutamine in intercellular adhesion: ascites tumor and embryonic cells. Exp Cell Res 77: 175-182, 1973 6. Connolly DT, Oppenheimer SB: Cell density-dependent stimulation of glutamine syntbetase activity in cultured mouse teratoma ceils. Exp Cell Res 94: 459-464, 1975. 7. Schousboe A, Svenneby G, Hertz L: Uptake and metabolism of glutamate in astrocytes cultured from dissociated mouse brain hemispheres. J Neurochem 29: 99%1005, 1977 8. Viceps D, Cristofalo VJ: Glutamine synthetase activity in WI-38 cells. J Cell Physiol 86: %14, 1975 9. Viceps-Madore D, Cristofalo VJ: Age-associated changes in glutamine synthetase activity in WI-38 cells. Mech Ageing Dev 8: 43-50, 1978 10. StamatiadouMN: Effects of glutamine on the two catalytic activities of glutamine synthetase in cultured mouse cells strain L. Biochem Biophys Res Commun 47: 485-490, 1972 11. Hosoi S, Mioh H, Anzai C, Sato S, Fujiyoshi N: Establishment of Namalva cell lines which grow continuously in glutamine-free medium. Cytotechnology 11: 151-158, 1988 12. Wolfrom C, Kadhom N, Polini G, Poggi J, Moatti N, Gautier M: Glutamine dependency of human skin fibroblasts: Modulation by hexoses. Exp Cell Res 183: 303-318, 1989 13. Lemonnier F, Gautier M, Wolfrom C, Lemonnier A: Some metabolic differences between human skin and aponeurosis fibroblasts in culture. J Cell Physiol 104: 415--423, 1980 14. Wolfrom C, Polini G, Delhotal B, Lemonnier F, Gautier M: Comparative effects of glucose and fructose on the growth and morphology of cultured human skin fibroblasts. Exp Cell Res 149: 535-546, 1983
15. Ward HK, Bradford HF: Relative activities of glutamine synthetase and glutaminase in mammalian synaptosomes. J Neurochem 33: 339-342, 1979 16. Bradford MM: A rapid and sensitive method for the quantiration of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72: 248-254, 1976 17. Miller RE, Carrino DA: An association between glutamine synthetase activity and adipocyte differentiation in cultured 3T3-L1 cells. Arch Biochem Biophys 209: 486--503, 1981 18. Hayward B, Hussain A, Wilson RH, Lyons A, Woodcock V, Mclntosh B, Harris TJR: The cloning and nucleotide sequence of cDNA for an amplified glutamine synthetase gene from the Chinese hamster. Nucleic Acids Res 14: 999-1008, 1986 19. Tingey SV, Walker EL, Coruzzi GM: Glutamine synthetase genes of pea encode distinct polypeptides which are differentially expressed in leaves, roots and nodules. EMBO J 6: 1-9, 1987 20. Sevdalian DA, Ozand PT, Zielke HR: Increase in glutaminase activity during the growth cycle of cultured human diploid fibroblasts. Enzymes 25: 142-144, 1980 21. Lemonnier F, Gautier M, Moatti N, Lemonnier A: Comparative study of extraceUular aminoacids in culture of human liver and fibroblastic cells. In Vitro 12: 450-466, 1976 22. Butler M, Thilly WG: MDCK microcarrier cultures: seeding density effects and aminoacid utilization. In Vitro 18: 213-219, 1982 23. Wagner R, Ryll T, Krafft H, Lehman N: Variation of aminoacid concentrations in the medium of HU ~-IFN and HU IL-2 producing cell lines. Cytotechnology 1: 145-150, 1988 24. Donnely M, 8cheffler IE: Energy metabolism in respiration deficient and wild Chinese hamster fibroblasts. J Cell Physiol 89: 3%52, 1976 25. Sumbilla CM, Zielke CL, Reich WD, Ozand PT, Zielke HR: Comparison of the oxidation of glutamine, glucose, ketone bodies and fatty acids by human diploid fibroblasts. Biochem Biophys Acta 675: 301-304, 1981 26. McKeehan WL: Glutaminolysis in animal cells. In: MG Morgan (ed.) Carbohydrate metabolism in cultured cells. Plenum Publishing Corporation, 1986, pp 111-150.
Address for offprints: M. Gautier, Inserm U56, H6pital de Bic~tre, 78, rue du G6n6ral Leclerc, 94270 Le Kremlin-Bic6tre, France
Respective effects of glucose and glutamine on the glutamine synthetase activity of human skin fibroblasts Thdophile Soni 1, Claire Wolfrom 2, Samia Guerroui 1, Nicole Raynaud 2, Jos6phine Poggi 1, Nicole Moatti I and Marthe Gautier z 1Facultd des Sciences pharmaceutiques et biologiques. Universitd Paris-Sud, Rue J.B. Clement, 92290 Chatenay-Malabry, France; 2 Unit~ de Recherche d'Hdpatologie Pddiatrique, INSERM U56, HOpital de BicOtre, 94270 Le Kremlin-BicOtre, France Received 18 May 1990; accepted 7 September 1990
Key words: glutamine synthetase, human fibroblasts, glucose, glutamine Summary The activity of Glutamine Synthetase (GS) was measured during the growth of human diploid skin fibroblasts cultured for three weeks in the presence or absence of either glucose or glutamine or both. In medium free of both glucose and glutamine, a single late peak in GS activity was observed concomitantly with delayed small cell protein increment. In all media containing either glucose or glutamine or both, GS activity rose sharply during rapid cell growth, displayed a plateau, and then decreased once the cells had reached confluency. The variations in extracellular aminoacid levels were also determined and were found to depend on the composition of the medium but not on the cell culture duration. These results demonstrate, for the first time as far as we know, that strong GS activity is present in rapidly growing skin fibroblasts. In contrast to many other mammalian cell types, GS activity in human skin fibroblasts appears not to be subject to regulation by extracellular glutamine. This difference may well be connected with cell differentiation.
Introduction Glutamine Synthetase (GS), i.e. (L-glutamate: ammonialigase (ADP), EC 6.3.1.2), catalyses the reversible formation of glutamine from glutamate, ammonia and ATP. This reaction plays a key role in living cell metabolism because the amide group originating from glutamine is used as the nitrogen source for the biosynthesis of various aminoacids, NAD, purine and pyrimidine nucleotides, and carbamylphosphate. Glutamine is also a maj or energy source for cultured ceils [1], and in human diploid fibroblasts this has been conclusively demonstrated by metabolic approaches [2-4]. One recent investigation showed, in particular, that the concentration of intracellular glutamine is regulated by active
uptake of extracellular glutamine, and that in glutamine-free medium containing glutamate and glucose, only one quarter of the intracellular glutamine is derived from intracellular exchangeable glutamate. Therefore compartmentation of the glutamate pool has been postulated to explain the intracellular glutamine obtained from other glutamate fractions [4]. However, it must be remembered that glutamine synthesis requires both glutamate and ammonium, which are the substrates of GS. GS activity has been demonstrated in several types of mammalian cultured ceils, mainly from rodents [5, 6], including astrocytes [7], and the embryonic human lung fibroblast strain WI-38 [8, 9]. Extracellular glutamine has been reported to repress GS activity [8, 10, 11]. We addressed the
150 question of whether GS activity is regulated by extracellular glutamine in long-term cultures of human skin fibroblasts. Another question we set out to clarify was how normal diploid human skin fibroblasts could slowly grow in hexose-free medium without glutamine, as we previously observed [12]. Accordingly, in this investigation, we measured, for the first time as far as we know, GS activity during the growth of human diploid fibroblasts, in the presence or absence of either glucose or glutamine, or both. In addition, we assessed the concomitant variations in extracellular aminoacid levels.
Materials and methods
Cel/s Human fibroblasts were obtained from a skin biopsy taken from a one month-old infant during surgery. Cultures were initiated from explants as already described [13]. Cells were initially cultured in Minimum Essential Medium (MEM 2111, Eurobio, France) containing nonessential aminoacids, and 5.5 mM D-glucose, 2mM glutamine and 10% fresh undialyzed human serum [12, 14]. At the 6th passage, cells were trypsinized and transferred into fifty 75 c m 2 replicate flasks (T75). Plating density was 4 × 105 cells per flask containing 10 ml of medium. At day 0, flasks were fed with one of the four following media: MEM containing 5.5mM glucose, either with 2 mM glutamine (Glg +) or without ( G l g - ) , and MEM containing no Hexose, either with 2mM glutamine (HOg +) or without glutamine ( H O g - ) . However, since non dialyzed serum was used, the 10% human serum in the supplemented media supplied about 0.2 mM glucose and about 0.05mM glutamine. The culture fluid was changed every 3rd or 4th day. As the cells grown in H O g - medium exhibited a very poor growth rate, this series of cultures was pursued for 27 days instead of 21, and the enzymatic assays were not, as in the other cultures, carried out every 3rd or 4th day, but at longer intervals. Observation of monolayers in phase contrast was carried out with a Wild inverted microscope.
Aminoacid determination
Before refeeding, media were sampled every 3 or 4 days, for aminoacid determination by ion exchange chromatography (Technicon T.S.M. with a Technation integrator). The percentage of variation in aminoacid levels were established from the results obtained for the corresponding reference media.
Enzymatic assay
Cells were harvested by scraping into I ml of distilled water and homogenized by sonication for 5 sec at 4° C. The pooled cell content of 1 to 4 T75 flasks was necessary for each determination. Glutamine synthetase activity was measured in duplicate by a micromethod adapted from that described by Ward and Bradford [15] without Triton X100. Aliquots of 120 ~1 of homogenate were preincubated for 5 min at 37° C before the reaction was started by adding 100 ~1 of incubation mixture. The final concentrations used were as follows: 100raM imidazole buffer (pH 7.2), 12.5mM MgC12, 20raM mercaptoethanol, 10 mM ATP, 50 mM sodium glutamate, 1 mM ouabain, 100 mM hydroxylamine (freshly neutralized to pH 7.2 with KOH or NaOH), 13 mM phosphoenol pyruvate, and 3.3 UI/ ml pyruvate kinase. After 60min incubation at 37° C, the reaction was stopped by adding 500 ~1 of a fresh mixture containing 50 mM FeCI3, 400 mM perchloric acid, and 400 mM HC1. After the tubes had been left to stand at 4° C for 20-30 minutes, the protein precipitate was sedimented by centrifugation at 4000 rpm and 4°C for 10 min. The optical density of the supernatant was measured directly at 500nm and 25°C in a Beckman spectrophotometer. Standard curves were established using glutamyl-hydroxamate (Sigma). Values are expressed as ~mol of glutamine formed per hour per mg of protein at 37° C.
Protein content
As it was considered impossible to perform on this same cell line parallel experiments for enzyme as-
151
2000 []
/ 2 c~
E100C _c
o
¢33
E
1
c
-- -- --ND-- - - NO-- - - - ' ~ [ > . . . .
I 3
I 7
] 10
[ I 14 18 Days
L 21
[ 24
L 27
0
ND"- "~
....
ND, --
i
I
I
J
I
I
I
]
3
7
10
14
18
21
24
27
Days
Fig. 1. Specific activity of glutarnine synthetase (GS) in diploid
Fig. 2. Protein increment of diploid infant human flbroblasts.
infant human flbroblasts grown in four culture media: glucose medium with glutamine (Glg+ • - - • ) or without glutamine ( G i g - IE--[]), and hexose-free medium with glutamine (HOg + O--O) or without glutamine (HOg - ~Y--~). Each point represents the mean of two determinations.
Data points represent the mean cell protein content per flask (T75), in the set of flasks used for each determination of glutamine synthetase (GS) activity, under four culture conditions: glucose medium with glutamine (Glg+ • - - • ) or without glutamine (Glg- []--i~), and hexose-free medium with glutamine (HOg + I ~ - O ) or without glutamine (HOg - 4>--~). Protein content of the homogenates was determined in duplicate.
says, and for assessing cell growth as cell numbers, due to the limited amount of cells at this early passage, the cell protein increment was used to estimate cell growth, in every set of flasks used for enzyme assays. Protein content of the homogenates was determined in duplicate according to Bradford [16] using bovine serum albumin (Sigma) as standard.
Results
and
discussion
The fluctuations in GS specific activity and the corresponding total cellular protein increment measured during cell growth (Figs 1 and 2), demonstrated that normal diploid fibroblasts are able to synthesize glutamine. In addition, we showed the related evolution of the kinetics and levels of GS activity and that of cell protein increment. W h e n cells were cultured in the presence of glucose, i.e. in conventional Gig + m e d i u m as well as in G i g - medium, GS specific activity increased
m o r e than 4-fold from days 3 to 7 of culture. A plateau was then observed for this activity for 7-10 days, and was followed by a slight decrease which coincided with the high protein content of cells in their late stationary growth phase. These results for Gag + medium do not agree with those of previous investigations showing that GS synthesis by various cell types was repressed in the presence of glutamine [10]. For glutamine-free medium ( G i g - ) , our results for normal h u m a n fibroblasts are in agreement with the findings for 3T3-L1 fibroblasts, in which glutamine deprivation had no effect on GS activity [17]. However, they diverge from those obtained for other cell lines, including the embryonic h u m a n lung WI-38 fibroblasts in which GS activity was raised by depriving cells of glutamine [8] and for a N a m a l v a cell line, in which GS activity increased when cells were adapted to continuous growth in glutamine-free m e d i u m [11]. For both G1
152 media, our results showed that the presence of extracellular glutamine, however readily the latter was utilized by the cells, had no effects on GS specific activity, and neither had its absence. A homology was recently established between mammalian and plant GS [18]; in addition, at least 3 distinct GS-encoding genes with different structures, as well as different patterns of GS expression in vivo, were reported in the pea [19]. Consequently, the observation of different patterns of GS expression in different cell types might be connected with cell differentiation. When our cells were cultured in glucose-free medium containing glutamine (HOg +), GS specific activity increased by day 7 and then remained constant, but at lower levels than in Gig + medium. During the first 14 days of culture in HOg medium cells did not multiply and some were necrosed which resulted in a low steady cell protein content, and no GS specific activity was detected; however, by day 21, a small peak in GS activity was present at the same time as cells recovered a normal appearance and started multiplying. This improvement might have resulted from the release of nutrients from damaged cells in the extracellular medium. However this hypothesis seems unlikely, since most of the cells were surviving, and the culture medium was renewed every 3rd or 4th day. The selection for a specific cell population is another possibility. Although these starved cells in HOg - medium had no major energy supply, they were able here to display transient GS activity for de n o v o synthesis of glutamine from the glutamate pool, even at very low cell density. Therefore, under our conditions of culture, cell protein increment correlated with the increase in GS specific activity, and the high cell protein content which was observed at cell confluency coincided with the decrease in this activity. A similar correlation was previously demonstrated in cultured mouse teratoma cells, [5, 6]. These parallel increases in GS specific activity and cell density were comparable to those previously observed for glutaminase activity, which was shown to be high in rapidly dividing human fibroblasts, irrespective of the glutamine concentration [20]. Figure 3 shows the corresponding variations ob-
served in the levels of extracellular nonessential aminoacids. These variations were similar to those we previously observed in human fibroblasts after 3 days of culture in MEM [21] and appeared to be closely related to the glucose and glutamine content of the medium. Thus in the presence of glucose and glutamine (Gig + medium), accumulation of alanine, glycine, proline, and glutamate was marked. After 4 days of culture, MDCK cells displayed a pattern of aminoacid variations similar to that of the present human fibroblasts, except for proline and aspartic acid, for which no change was observed [22]. Here, when glucose alone was omitted from the medium (HOg +), alanine production was greatly reduced, as its level is related to glutamine consumption [23], and also to pyruvate availability, which decreases in the absence of glucose. When glutamine was omitted from the medium (Gig - and HOg - ) , alanine and proline were utilized by the cells. Production of extracellular glutamate was observed in G l g + and H O g + media, and more surprisingly in HOg - medium. In fact, for human fibroblasts, glutamine has been clearly established as a major source of energy [1, 2, 24,25]. In conventional media, the large excess of extracellular glutamine (2 mM), which is rapidly depleted because of an efficient glutamine uptake [4], provides an ample source of energy. The glutaminolysis first step is the glutamine-toglutamate step, by glutaminase or amidotransferase, which appear to be essentially irreversible reactions. The subsequent steps of the pathway mostly involve mitochondrial enzymes [26] of the TCA cycle. A severe energy deprivation, as in our H O g - medium, might block these latter steps, which would shift the equilibrium of cytosolic aminotransferase towards the production of energy substrates such as pyruvate and oxaloacetate, and of glutamate. Conversely, a large excess in energy supply, as in our Gag + or HOg + media, might overflow these mitochondrial steps and lead to the accumulation of glutamate, which may in turn be converted to proline and to a lesser extent alanine. When only glucose is present (i.e. Glg - ) all extracellular glutamate available is taken up and additional glutamate may be supplied through the transamination of alanine and proline, whose con-
153 TGlg
+
712
.
400 200-
davs 21 100 2
~S
"l'°"I[II III[
-"
II
200
-
50n~J.
050100Ala
Gly
Pro
Glu
Gin
Asn
Ser
Tyr
Asp
Cys
t'°° Ala
% 400-
Glg-
.,,
i
III
100-
Pro
Glu
Gin
ASh
S~r
Tyr
Asp
Cys
400-
•3
200-
Gly
"&;~'~'
200
L
][L
x.I
days
27
50-
n.d
O" I It -50100Ala
Gly
Pro
Glu
Gin
Asn
Ser
Tyr
Asp
Cy5 Ala
GIy
Pro
Glu
Gln
Asn
Ser
Tyr
Asp
Cys
Fig. 3. Bar graphs represent selective aminoacid consumption (below abscissa) or production (above abcissa in the media of diploid infant human fibroblast cultures. Results are expressed in percentages of aminoacid consumption and production versus control media. Insert on the right shows cell protein content at the different days of culture. For details, see Fig. 2. Data are not related to the number of cells. In the presence of glutamine (Glg + and HOg + media) alanine, glucine, proline and glutamate were constantly produced, and the other aminoacids constantly consumed. The only marked differences concerned the levels of alanine and glutamate production. In the absence of glutamine, the difference concerned glutamate, which was entirely consumed in Glg - medium but produced in HOg medium. Consumption of all the other aminoacids was less in HOg - medium, which may reflect a low protein content per flask.
centrations in the medium were shown here to decrease. Some of the extracellular glutamine may also be chemically converted to pyrrolidone carboxylate in the medium [22]. In all 4 media, glycine accumulation was constant, probably owing to the consumption of cysteine and serine. No marked variations in the levels of essential aminoacids were noted (data not shown), except that methionine, leucine and isoleucine consumption were related to the cell growth rate. The rapid depletion of these 2 latter branched-chain aminoacids seems to be an important factor in this growth [22]. In conclusion, the role of extracellular glutamine as a source of energy is highlighted by these patterns of aminoacid metabolism in different media. When another source of energy is present, like glucose, the absence of glutamine is compensated by other aminoacids, increment in cell protein is
normal, and GS activity is unchanged. Therefore our hypothesis is that glutamine synthesis is regulated so as to fulfill the priority requirement for glutamine for protein and nucleic acid biosynthesis. This contrasts with glutaminolysis, which appears to be regulated so as to fuel oxidative metabolism concurrently with glucose. This essential anabolic role of glutamine synthesis requires a GS activity associated with cell proliferation, and independent from the provision of exogenous glutamine. The hypothesis that this newly synthesized glutamine exists as a separate pool should be tested. The recent cloning of eukaryotic GS genes [19] should help to provide more information about the signals that regulate GS activity in human fibroblasts.
154 References 1. Zielke HR, Zielke CL, Ozand PT: Glutamine: a major energy source for cultured mammalian cells. Federation Proc 43: 121-125, 1984 2. Reitzer LJ, Wice BM, Kennell D: Evidence that glutamine, not sugar, is the major energy source for cultured HeLa cells. J Biol Chem 254: 2669-2676, 1979 3. Zielke HR, Sumbilla CM, Sevdalian DA, Hawkins RL, Ozand PT: Lactate: a major product of glutamine metabolism by human diploid fibroblasts. J Cell Physiol 104: 433441, 1980 4. Darmaun D, Matthews DE, Desjeux JF, Bier DM: Glutamine and glutamate nitrogen exchangeable pools in cultured fibroblasts: a stable isotope study. J Cell Physiol 134: 143-148, 1988 5. Oppenheimer SB: Utilization of L-glutamine in intercellular adhesion: ascites tumor and embryonic cells. Exp Cell Res 77: 175-182, 1973 6. Connolly DT, Oppenheimer SB: Cell density-dependent stimulation of glutamine syntbetase activity in cultured mouse teratoma ceils. Exp Cell Res 94: 459-464, 1975. 7. Schousboe A, Svenneby G, Hertz L: Uptake and metabolism of glutamate in astrocytes cultured from dissociated mouse brain hemispheres. J Neurochem 29: 99%1005, 1977 8. Viceps D, Cristofalo VJ: Glutamine synthetase activity in WI-38 cells. J Cell Physiol 86: %14, 1975 9. Viceps-Madore D, Cristofalo VJ: Age-associated changes in glutamine synthetase activity in WI-38 cells. Mech Ageing Dev 8: 43-50, 1978 10. StamatiadouMN: Effects of glutamine on the two catalytic activities of glutamine synthetase in cultured mouse cells strain L. Biochem Biophys Res Commun 47: 485-490, 1972 11. Hosoi S, Mioh H, Anzai C, Sato S, Fujiyoshi N: Establishment of Namalva cell lines which grow continuously in glutamine-free medium. Cytotechnology 11: 151-158, 1988 12. Wolfrom C, Kadhom N, Polini G, Poggi J, Moatti N, Gautier M: Glutamine dependency of human skin fibroblasts: Modulation by hexoses. Exp Cell Res 183: 303-318, 1989 13. Lemonnier F, Gautier M, Wolfrom C, Lemonnier A: Some metabolic differences between human skin and aponeurosis fibroblasts in culture. J Cell Physiol 104: 415--423, 1980 14. Wolfrom C, Polini G, Delhotal B, Lemonnier F, Gautier M: Comparative effects of glucose and fructose on the growth and morphology of cultured human skin fibroblasts. Exp Cell Res 149: 535-546, 1983
15. Ward HK, Bradford HF: Relative activities of glutamine synthetase and glutaminase in mammalian synaptosomes. J Neurochem 33: 339-342, 1979 16. Bradford MM: A rapid and sensitive method for the quantiration of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72: 248-254, 1976 17. Miller RE, Carrino DA: An association between glutamine synthetase activity and adipocyte differentiation in cultured 3T3-L1 cells. Arch Biochem Biophys 209: 486--503, 1981 18. Hayward B, Hussain A, Wilson RH, Lyons A, Woodcock V, Mclntosh B, Harris TJR: The cloning and nucleotide sequence of cDNA for an amplified glutamine synthetase gene from the Chinese hamster. Nucleic Acids Res 14: 999-1008, 1986 19. Tingey SV, Walker EL, Coruzzi GM: Glutamine synthetase genes of pea encode distinct polypeptides which are differentially expressed in leaves, roots and nodules. EMBO J 6: 1-9, 1987 20. Sevdalian DA, Ozand PT, Zielke HR: Increase in glutaminase activity during the growth cycle of cultured human diploid fibroblasts. Enzymes 25: 142-144, 1980 21. Lemonnier F, Gautier M, Moatti N, Lemonnier A: Comparative study of extraceUular aminoacids in culture of human liver and fibroblastic cells. In Vitro 12: 450-466, 1976 22. Butler M, Thilly WG: MDCK microcarrier cultures: seeding density effects and aminoacid utilization. In Vitro 18: 213-219, 1982 23. Wagner R, Ryll T, Krafft H, Lehman N: Variation of aminoacid concentrations in the medium of HU ~-IFN and HU IL-2 producing cell lines. Cytotechnology 1: 145-150, 1988 24. Donnely M, 8cheffler IE: Energy metabolism in respiration deficient and wild Chinese hamster fibroblasts. J Cell Physiol 89: 3%52, 1976 25. Sumbilla CM, Zielke CL, Reich WD, Ozand PT, Zielke HR: Comparison of the oxidation of glutamine, glucose, ketone bodies and fatty acids by human diploid fibroblasts. Biochem Biophys Acta 675: 301-304, 1981 26. McKeehan WL: Glutaminolysis in animal cells. In: MG Morgan (ed.) Carbohydrate metabolism in cultured cells. Plenum Publishing Corporation, 1986, pp 111-150.
Address for offprints: M. Gautier, Inserm U56, H6pital de Bic~tre, 78, rue du G6n6ral Leclerc, 94270 Le Kremlin-Bic6tre, France
