Glutamine synthetase is essential for proliferation of fetal skin fibroblasts

Archives of Biochemistry and Biophysics 478 (2008) 96–102

Contents lists available at ScienceDirect

Archives of Biochemistry and Biophysics journal homepage: www.elsevier.com/locate/yabbi

Glutamine synthetase is essential for proliferation of fetal skin fibroblasts T. Vermeulen a, B. Görg b, T. Vogl c, M. Wolf c, G. Varga c,d, A. Toutain e, R. Paul f, F. Schliess b, D. Häussinger b, J. Häberle a,* a

Universitätsklinikum Münster, Klinik und Poliklinik für Kinder- und Jugendmedizin, Albert-Schweitzer-Straße 33, 48129 Münster, Germany Heinrich-Heine Universität Düsseldorf, Klinik für Gastroenterologie, Hepatologie und Infektiologie, Moorenstraße 5, 40225 Düsseldorf, Germany c Universitätsklinikum Münster, Institut für Immunologie, Röntgenstraße 21, 48149 Münster, Germany d Universitätsklinikum Münster, Klinik und Poliklinik für Hautkrankheiten, Von-Esmarch-Straße 58, 48149 Münster, Germany e Service de Génétique, Centre Hospitalier Universitaire Bretonneau, 2 boulevard Tonnellé, 37044 Tours Cedex 9, France f Universität Münster, Institut für Zoophysiologie, Hindenburgplatz 55, 48143 Münster, Germany b

a r t i c l e

i n f o

Article history: Received 28 May 2008 and in revised form 2 July 2008 Available online 17 July 2008 Keywords: Cell proliferation Skin fibroblasts Glutamine synthetase Fetal period

a b s t r a c t Background. Glutamine synthetase (GS) is ubiquitously expressed in the human and plays a major role for many metabolic pathways. However, little is known about its role during the fetal period. Methods. Cultured skin fibroblasts derived from an aborted fetus deficient in GS activity due to a R324C exchange as well as fetal and mature controls were used to determine the level of GS-expression, apoptosis, and proliferation in presence or absence of exogenous glutamine. Results. Glutamine synthetase can be found at early gestational stages. Loss of GS activity either inherited or induced through L-methionine sulfoximine leads to an upregulation of the GS protein but not of the GS mRNA and results in a significant drop in the proliferation rate but has no effect on apoptosis. Exogenous glutamine does not influence the rate of apoptosis but increases proliferation rates of the fetal but not the mature fibroblasts. Conclusion. GS can be found during early human fetal stages when it displays a significant effect on cell proliferation. Ó 2008 Elsevier Inc. All rights reserved.

Glutamine synthetase (GS1; glutamate ammonia ligase, EC 6.3.1.2) catalyzes the conversion of glutamate and ammonia to glutamine by an ATP-dependent reaction. GS is encoded by a highly conserved gene and can be considered as a key enzyme in metabolism of pro- and eukaryotes [1]. In human plasma, glutamine is the most abundant amino acid and an important source for the biosynthesis of several amino acids, purines, and pyrimidines [2]. GS plays a decisive role in ammonia and glutamate detoxification, pH homeostasis, interorgan nitrogen flux, acid–base regulation, and cell signaling [3]. GS is localized in the cytoplasm and is ubiquitously expressed in the human organism with high concentrations in liver, brain, and skeletal muscle [4]. In the liver, GS expression is confined to a small cell population at the hepatic venous outflow of the liver acinus [5–7]. In astrocytes, GS operates as a neuroprotective enzyme by removing excess of ammonia and glutamate [8]. Recently, an inherited defect of GS activity in two * Corresponding author. Fax: +49 251 8347809. E-mail address: [email protected] (J. Häberle). 1 Abbreviations used: GS, glutamine synthetase; SNAT 1, System N amino acid transporter 1; GS, fetal GS activity-deficient fibroblasts; GS+, fetal GS wild type fibroblasts; GS control, GS wild type fibroblasts from a 2-year-old child; MEM, Minimal Essential Medium; FCS, fetal calf serum; MSO, methionine sulfoximine; SDS, sodium dodecyl sulphate-polyacrylamide; GAPDH, glyceraldehyde phosphate dehydrogenase; TBST, Tris buffer saline tween; RPL, ribosomal protein L13a; cpm, counts per minute; CFDA, carboxyfluorescein diacetate; MFI, mean fluorescence intensity. 0003-9861/$ - see front matter Ó 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.abb.2008.07.009

children of consanguineous Turkish background who both died from multi-organ failure was reported and ascribed to the exchange of R324C and R341C, respectively, in the active site [9]. Systemic glutamine deficiency was the leading biochemical finding. As the most striking clinical sign, severe brain malformation with abnormal gyration and marked white-matter lesions was observed [9]. There are only few studies on GS expression and activity as well as glutamine metabolism during fetal development. In rat liver, an atypical GS distribution pattern developing in the late fetal period was described. Before that stage, megakaryocytes showed intense staining on immunohistochemistry but liver parenchyma was only faintly stained [10]. Other than in the liver, rat astrocytes showed the appearance of GS mRNA and protein as early as on embryonic day 14 [11,12]. In a porcine model, glutamine exhibited the highest fetal:maternal plasma ratio among all amino acids, suggesting active synthesis and release of glutamine by the placenta [13]. In an ovine model, about two-thirds of glutamine transported in the umbilical vein was found to be derived from uterine uptake with one-third derived from placental production [14]. Transport of glutamine via the placental barrier is provided by System N amino acid transporter 1 (SNAT 1) that showed an increasing rate of expression in the rat placenta during the final third of gestation [15]. In line with a role of GS in fetal development, complete knockout of the mouse GS resulted in early

T. Vermeulen et al. / Archives of Biochemistry and Biophysics 478 (2008) 96–102

embryonic lethality [16]. In the same study, GS deficient embryonic stem cells exhibited a reduced survival if transplanted into wild type blastocysts and failed to survive in aggregates with tetraploid wild type blastomeres [16]. Confirming the importance of glutamine for fetal development, exogenous glutamine was shown to improve mouse embryonic development when added to the culture medium [17]. In this paper, cultured human skin fibroblasts derived from a GS activity-deficient fetus of 17 weeks of gestation were investigated under different culture conditions as a model to study the role of GS on basal growth properties in early fetal stages. Hereby, an important role of GS for the proliferation of skin fibroblasts at early fetal stages was elucidated.

97

Roche, Mannheim, Germany). The protein content was determined using the Coomassie Plus Assay Reagent (Bio-Rad, München, Germany). Proteins were analyzed by Western blotting with use of 10% SDS (sodium dodecyl sulphate-polyacrylamide) gels and a semidry transfer apparatus (Amersham Pharmacia, England). Blots were probed overnight at 4 °C with antisera against GS (1:5000 solution, Sigma–Aldrich, München, Germany) and against GAPDH (glyceraldehyde phosphate dehydrogenase, 1:10,000 solution, Biodesign International Saco, USA), respectively. All antibodies were diluted in Tris buffer saline tween (TBST, Sigma, Taufkirchen, Germany). After washing with TBST and incubation with horseradish-peroxidase coupled antirabbit IgG or antimouse IgG antibody (1:2500 solution, Dako, Cambridgeshire, England) at 20 °C for 1 h, blots were developed with the use of Western-Lightning Chemiluminescence Reagent Plus (Perkin-Elmer, MA, USA).

Materials and methods Quantitative real-time PCR Skin fibroblasts Cultured skin fibroblasts were obtained from an aborted fetus after prenatal diagnosis by molecular genetic means, which had revealed a homozygous mutation (R324C) of the GS gene. The index patient of this family suffered from congenital GS deficiency and has been reported earlier [9]. As a result of the prenatal diagnosis, the gestation was terminated at 16+4 weeks of gestational age and a fetal skin biopsy was taken. The skin biopsy was initially cultured in standard medium containing a surplus of 10 mM glutamine in order to prevent growth failure as reported before [9]. As control cells, skin fibroblasts from a fetus aged 16+6 weeks of gestation who was aborted because of a severe feto-fetal transfusion syndrome as well as skin fibroblasts from a 2-year-old child taken for other diagnostic purposes were used. For this study, cells are named GS (fetal GS activity-deficient fibroblasts), GS+ (fetal GS wild type fibroblasts), and GS control (GS wild type fibroblasts from a 2-year-old child). All cell lines were investigated at the GS locus by DNA-sequencing of all coding exons including the flanking intronic regions as described [9]. The parents agreed to the use of the skin fibroblasts of their children for scientific purposes by written informed consent. The study was approved by the Ethics Committee of the University of Münster. Cell culture Cells were cultured in 10 ml supplemented Minimal Essential Medium (MEM, PAA, Pasching, Austria) containing 10% fetal calf serum (FCS, Biowest, Nuaillé, France), 1% streptomycin, 1% penicillin (Biochrome, Berlin, Germany), and 2 mM glutamine and routinely passaged twice weekly by trypsinizing and seeding at 1:4 dilution in 75 cm2 flasks (Greiner Bio One, Frickenhausen, Germany). For all experiments, GS, GS+, and GS control cells were washed by PBS, trypsinized and counted using the Neubauer–Zählkammer. Cells were then plated in 96-well or 6-well plates. Culture conditions varied with respect to the concentrations of extracellular glutamine (0–10 mM) or the presence of L-methionine sulfoximine (MSO, Sigma, Taufkirchen, Germany), an irreversible inhibitor of GS [18,19]. Cerebro-cortical astrocytes from Wistar rats were cultured in DMEM as described before and used for control purposes [20]. Western blot analysis Cells were harvested in lysis buffer containing 50 mM Tris hydrochloride (pH 7.4), 1% Triton X-100, 150 mM sodium chloride, 1 mM EDTA, and a protease inhibitor cocktail (1 tablet per 15 ml;

To determine the level of the GS transcript in addition to the Western blot analyses cultured GS, GS+, and GS control skin fibroblasts were taken for preparation of RNA using a standard kit (RNeasy Mini Prep, Qiagen, Hilden, Germany). This was performed twice using independent cultures of each cell line. Total RNA (1 lg) was used for reverse transcription utilizing the first strand cDNA synthesis kit and oligo-p(dT)15 primers (Boehringer, Mannheim, Germany). The quantitative cDNAs were used to determine the level of gene expression by real-time reverse transcription-polymerase chain reaction (RT-PCR) as described previously [21]. As endogenous housekeeping genes, fragments of glyceraldehyde phosphate dehydrogenase (GAPDH) and ribosomal protein L13a (RPL) were amplified. Primers used for PCR analysis were as follows: for GS forward 50 -GCTGGTGTAGCCAATCGTAGC-30 and reverse 50 -GGCTTCTGTCACCGAAAAGG-30 (fragment size 122 bp); for GAPDH forward 50 -TGCACCACCAACTGCTTAGC-30 and reverse 50 GGCATGGACTGTGGTCATGAG-30 (fragment size 86 bp); for RPL forward 50 -AGGTATGCTGCCCCACAAAAC-30 and reverse 50 -TGTAGG CTTCAGACGCACGAC-30 (fragment size 141 bp). The relative expression was calculated as 2DC t GS =2DC t Housekeeping gene . All measurements were repeated two times. Determination of cell viability GS, GS+, and GS control cells were cultured as described above in the presence of 0, 2, 4, 6, 8, and 10 mM glutamine for 6 days in T25 flasks at 37 °C and 5% CO2 incubation. Then, cells were washed twice with PBS, trypsinized and centrifuged for 10 min at 600g. Cell pellets were solved in 400 ll standard MEM and quantified by staining with 5 ll Annexin V-FITC (PharMingen, San Diego, CA) following a 15 min incubation in darkness before analysis with FACScan flow cytometry (Becton–Dickinson, San Jose, CA). As a positive control of Annexin-V-FITC, cells were cultured for 12 h in standard medium with 1 lM staurosporine (Axxora, Lörrach, Germany). Measurement of cell proliferation by [3H]-thymidine incorporation All cell lines (2.5  103/well) were plated in 96-well round-bottom microtiter plates (Nunc, Rosklide, Denmark) in triplicate. Cells were cultured for 3 days at 37 °C and 5% CO2 under the experimental conditions described above. [3H]-thymidine incorporation was performed in a final volume of 200 ll with 1 lCi/well [3H]-thymidine (Hartman Analytics, Braunschweig, Germany) that was present for the last 12 h of the experiment. Plates were frozen at 20 °C for 2 days in order to remove the adherent cells. After thawing at 37 °C, cells were transferred to a filter pad by filter mate Harvester (Perkin-Elmer, MA, USA). Filter

98

T. Vermeulen et al. / Archives of Biochemistry and Biophysics 478 (2008) 96–102

pads were dried at 60 °C for 2 h and then put into a paper bag with 5 ml scintillation liquid and finally in the casket. [3H]-thymidine incorporation was measured by liquid scintillation counting that resulted in the determination of radioactive counts per minute (cpm) allowing to conclude on the rate of proliferation (Wallac, Turku, Finland). The different experimental conditions were as follows: GS, GS+, and GS control cells were plated at 2.5  103/well and cultured for 3 days in supplemented MEM with or without 2 mM glutamine. Cells were plated at 96-well plates and cultured for 3 days in supplemented MEM containing 2 mM glutamine and 10 mM MSO. Cell proliferation was estimated by liquid scintillation counting after 12 h [3H]-thymidine incorporation.

For analysis of the results of the RT-PCR the calculated relative expression of GS cells was set to 1. When GAPDH was used for control purposes, GS+ and GS control cells showed an n-fold expression of the GS transcript of 0.81 and 0.80, respectively. With RPL being the reference, GS+ and GS control cells yielded an n-fold expression of 1.05 and 0.90, respectively. Taken together, the analysis of the GS-expression by quantitative RT-PCR revealed no significant differences between the GS, GS+, and GS control cells. To compare the level of expression to a cell type with known abundant GS, rat astrocytes were used. Under identical culture conditions, the GS protein is much higher expressed in astrocytes than in skin fibroblasts (Fig. 1b). Influence of MSO and exogenous glutamine on GS-expression

Measurement of cell proliferation by flow cytometry Cells (6  105) of each cell line were incubated for 5 min in PBS containing 0.5 lM carboxyfluorescein diacetate (CFDA, Molecular Probes, Leiden, The Netherlands) at 37 °C in the dark, washed in PBS/1% FCS, and employed in proliferation assays. CFDA-labelled cells/well (2  105) were seeded into 6-well plates and cultured for 3 days. Cells were then trypsinized and added in 15 ml Falcon tubes (BD, Heidelberg, Germany) and centrifuged for 7 min at 600g. Cells were washed in PBS/1% FCS and the proliferation analyzed by FACScalibur (BD Biosciences, Heidelberg, Germany).

When the fibroblasts were cultured for 3 days in standard medium containing MSO, Western blot analysis displayed a strong upregulation of the GS protein in the GS+ and GS control cells. In GS cells, however, MSO did not further increase GS expression levels (Fig. 1c). Addition of up to 10 mM glutamine had no effect on the MSO-induced upregulation of GS expression (Fig. 1d). To examine the influence of varying exogenous glutamine supply in the absence of MSO, all cells were cultured in the presence of 0, 2, and 10 mM glutamine. Western blot analysis revealed that addition of glutamine had no effect on GS expression levels in either fibroblast type under investigation (Fig. 1e).

Determination of intracellular amino acid concentrations

Glutamine and genotype effects on cell viability

Skin fibroblasts were cultured in standard medium containing 2 mM glutamine. After trypsinizing and centrifugation for 10 min at 1100g, cells were sonicated and resolved in 100 ll 0.9% NaCl. Five microliters of each cell probe were used for determination of the protein content by Coomassie Plus Assay Reagent. Fifty microliters of the cell suspension were treated with 200 ll 3% sulfosalicylacid and centrifuged for 10 min at 1100 rpm. One hundred microliters of the supernatant were added to 100 ll sample dilution buffer. Quantitative analysis of the amino acid concentration was performed by standard cation exchange chromatography using a LC 3001 amino acid analyzer (Eppendorf-Biotronic, Germany). All measurements were done in triplicates and were repeated in three independent experiments.

After culturing of GS, GS+, and GS control cells in standard medium (MEM) and following addition of variable amounts of glutamine for 6 days, apoptotic cells were counted by flow cytometry upon staining with Annexin-V-FITC. Irrespective of the amount of glutamine added and independent of the fibroblast type investigated the percentage of apoptotic cells ranged from 6% to 8% (Fig. 2). This result indicates that GS deficiency does not increase apoptosis in the skin fibroblasts.

Statistical analysis For comparison of the results we used Student’s t-test. A difference was considered statistically significant at a level of p < 0.05(), p < 0.01(), and p < 0.001().

Results Mutation analysis Sequencing of the coding exons of the GS gene confirmed the homozygous mutation c.970C>T (R324C) in the fetal GS skin fibroblasts. Fetal and mature control cells displayed wild type sequences in the GS gene. Level of GS-expression in cells cultured in standard medium When fibroblasts were cultured in standard medium containing 2 mM glutamine, Western blot analysis revealed a strong upregulation of GS in the fetal GS fibroblasts, compared to the GS+ and GS control cells (Fig. 1a).

Influence of GS on cell proliferation Proliferation of skin fibroblasts in standard medium [3H]-thymidine incorporation displayed 7590 cpm ± 337 in GS cells, 25.919 cpm ± 1844 in GS+ cells, and 4521 cpm ± 324 in GS control cells. The proliferation rate of the GS cells was significantly different from that of both, GS+ (p < 0.01) and GS control cells (p < 0.001) as was proliferation of GS+ and GS control cells (p < 0.01) (Fig. 3a). This indicates a major role for GS on the proliferation of fetal skin fibroblasts. Effects of MSO and exogenous glutamine on fibroblast proliferation [3H]-thymidine incorporation in GS cells decreased by about 23% to 5865 cpm ± 16 when cells were cultured in a medium without additional glutamine (p < 0.01). In the presence of 10 mM MSO, [3H]-thymidine incorporation of GS cells further decreased by 85% to 1147 cpm ± 60 (p < 0.01). MSO plus (10 mM) the absence of exogenous glutamine decreased the [3H]-thymidine incorporation by about 97% to 218 cpm ± 42 (p < 0.001) (Fig. 3b). Similar effects on [3H]-thymidine incorporation were observed with GS+ cells (Fig. 3c). Proliferation of mature control cells, however, resulted in a median of 4521 cpm ± 324 upon culture in standard medium. Interestingly, [3H]-thymidine incorporation (6417 cpm ± 272) was not decreased in a medium lacking glutamine and thus differed fundamentally from the fetal cells. 10 mM MSO strongly decreased [3H]-thymidine incorporation by about 96% both in the presence or absence of additional glutamine (p < 0.001) (Fig. 3d).

T. Vermeulen et al. / Archives of Biochemistry and Biophysics 478 (2008) 96–102

99

Fig. 1. Western blot analysis of GS expression of GS, GS+, and GS control cells after culture for 3 days under variable media conditions. Equal amounts (120 lg) of total protein were separated by electrophoresis. GAPDH was used as reference protein. Q: glutamine. (a) GS expression in GS, GS+, and GS control cells. (b) GS expression in GS and GS+ compared to that found in cultured rat astrocytes. (c) Effect of MSO on GS expression. (d) Effect of MSO and addition of glutamine on GS expression. (e) Effect of varying concentrations of glutamine on GS expression.

eration rate (p < 0.001). GS cells showed a median of 13.3 MFI ± 0.10 after culture in standard medium and this decreased to 8.48 MFI ± 0.20 (p < 0.001) when glutamine was lacking. GS+ cells indicated a median of 28.9 MFI ± 0.35 after culture in standard medium and a reduction to 20 MFI ± 0.20 in a glutamine-free medium (p < 0.001). Thus, rates of proliferation of fetal fibroblasts were in accordance both when determined by [3H]-thymidine incorporation and by flow cytometry, respectively. GS control cells presented a median of 9 MFI ± 0.22 after culture in standard medium and, other than in the [3H]-thymidine incorporation experiment, an unchanged proliferation rate of 9.26 MFI ± 0.17 in the absence of glutamine (Fig. 4). Intracellular concentration of amino acids in fibroblasts Fig. 2. Cell viability of GS, GS+, and GS control cells was addressed by Annexin VFITC staining. None of the differences did reach statistical significance (n = 2).

Cell proliferation determined by CFDA As in the above experiment, GS+ cells exhibited the highest proliferation rate as determined by mean fluorescence intensity (MFI) measurement while GS cells showed a significantly lower prolif-

The level of intracellular amino acid concentrations was determined after growth in standard medium and neither did vary between GS and GS+ cells nor between fetal and mature fibroblasts. Also with respect to the intracellular glutamine concentration, no differences were observed, as detailed in Table 1.

100

T. Vermeulen et al. / Archives of Biochemistry and Biophysics 478 (2008) 96–102

Fig. 3. [3H]-thymidine incorporation of GS, GS+, and GS control cells after culture for 3 days (n = 3). Bars denote SD. The asterisks indicate significant differences. Q: glutamine. (a) [3H]-thymidine incorporation of cells cultured in medium containing 2 mM Q. (b–d) Influence of MSO and addition of Q on [3H]-thymidine incorporation in (b) GS cells, (c) GS+ cells, and (d) mature control cells.

Table 1 Measurements of intracellular amino acids by cation exchange chromatography in GS, GS+, and GS control skin fibroblasts (n = 6)

Fig. 4. Effect of exogenous glutamine (2 mM) on proliferation of skin fibroblasts measured by CFDA dilution using flow cytometry. Q: glutamine.

Discussion There is only scarce knowledge on the role of GS during fetal development. We took advantage on the availability of cultured homozygous GS mutant fetal skin fibroblasts and determined their proliferation and viability both in the presence and absence of the GS inhibitor MSO and under the influence of different concentrations of extracellular glutamine. GS cells exhibited a significant upregulation of the GS protein, which is in accordance with the findings in immortalized lymphocytes from other members of the same family [9]. Fetal control skin fibroblasts evinced only a weak level of GS expression in this early stage of development which was even lower than in the mature control cells. However, the findings indicate that GS is already expressed at the beginning of the second trimester of gestation, and that a defective enzyme results in upregulation of GS expression during this early fetal stage similar to that found in inherited GS deficiency [9]. GS upregulation may reflect a cellular attempt to compensate for the loss of GS activity.

Aspartate Threonine Serine Asparagine Glutamate Glutamine Glycine Alanine Citrulline Valine Cysteine Methionine Isoleucine Leucine Tyrosine Phenylalanine Histidine Ornithine Lysine Arginine Proline

GS

GS+

GS control

112.1 ± 52.8 152.8 ± 64.8 141.2 ± 60.3 49.1 ± 23.0 766.7 ± 317.6 13.9 ± 6.5 241.9 ± 115.3 248.3 ± 109.4 1.3 ± 1.9 118.6 ± 49.6 13.4 ± 6.2 58.4 ± 26.6 94.7 ± 41.7 193.9 ± 84.0 98.3 ± 43.4 101.8 ± 45.2 58.3 ± 27.2 9.2 ± 4.0 197.3 ± 91.0 210.8 ± 94.2 225.1 ± 90.5

125.8 ± 33.9 162.4 ± 45.7 72.0 ± 29.5 56.1 ± 17.0 645.2 ± 208.8 10.8 ± 5.4 368.5 ± 95.1 242.5 ± 82.2 0 ± 0.22 140.3 ± 35.1 10.0 ± 3.4 62.2 ± 18.8 99.5 ± 27.6 209.5 ± 59.9 104.8 ± 29.6 110.4 ± 31.1 64.4 ± 17.3 74.7 ± 97.4 205.0 ± 57.8 190.3 ± 51.5 277.8 ± 60.4

183.2 ± 70.3 212.0 ± 33.5 117.3 ± 31.4 57.4 ± 9.5 1527.6 ± 297.0 20.9 ± 5.4 345.3 ± 35.4 347.6 ± 56.2 0.66 ± 1.1 127.1 ± 19.4 12.0 ± 2.9 55.2 ± 10.6 110.4 ± 19.3 199.1 ± 35.8 99.2 ± 16.1 102.6 ± 16.9 66.4 ± 9.7 20.95 ± 2.5 202.6 ± 36.0 199.3 ± 29.8 363.6 ± 41.6

Concentrations are given in lmol/g total protein.

To confirm the immediate role of GS inactivation for the GS upregulation in GS cells, we treated all cell lines with MSO, which is an effective and irreversible inhibitor of GS activity [19]. Previously, MSO was shown to inhibit GS activity by at least 97% in rat intestinal epithelial cells [18,22]. When treated with MSO, a change in the level of GS expression was revealed in the GS competent fetal and mature cells, confirming that the loss of GS activity in the homozygously affected fetal fibroblasts led to the observed increase in the expression level. Glutamine has been shown to be critical for growth and function of the gastrointestinal epithelium [23] and to have a proliferative effect throughout the rat intestinal epithelium cells [24].

T. Vermeulen et al. / Archives of Biochemistry and Biophysics 478 (2008) 96–102

Osmotic cell swelling due to intracellular accumulation of glutamine was suggested to be a critical signal mediating progression through the cell cycle [25,26]. A role of GS for cell growth seemed to be further underlined, when skin biopsies taken from patients suffering from inherited GS deficiency failed to grow when put into a standard culture medium containing 2 mM glutamine [9]. To prevent a growth failure of the skin biopsy taken from an aborted GS deficient fetus, the growth medium was supplemented with 10 and 20 mM glutamine, respectively. These conditions yielded morphologically normal fetal skin fibroblasts in the cultures. However, since the skin biopsy was not grown in parallel in a standard medium containing the usual 2 mM glutamine, there is no proof that the cells investigated would have failed again under these conditions. Expression of GS is known to be upregulated as a response to glucocorticoid action [27]. Another mechanism has been found in L-asparaginase-resistent leukemia cells. As a result of L-asparaginase treatment, these cells were deprived of asparagine and glutamine and consequently increased their rate of GS expression and activity but not of GS mRNA pointing towards a posttranscriptional regulation of GS expression [28–30]. The investigations of the GS transcript in this study also point towards a posttranscriptional mechanism: to determine whether the Western blot findings would be the result of either a transcript upregulation or a decreased degradation of the GS protein, we performed quantitative RT-PCR experiments in all three cell lines. Hereby, transcript upregulation as the underlying mechanism was excluded. Instead, a decreased rate of degradation of the mutated GS protein is likely to be the mechanism chosen by the cell to adjust the GS deficiency but further studies are needed to clarify the details of this. Cultured skin fibroblasts were shown to depend on an external supply of glutamine [23]. We therefore speculated that an increase of the extracellular glutamine concentration might be able to influence the level of GS expression, and that it could lead to a diminished upregulation in GS cells. However, an increased supplementation of glutamine up to 10 mM in the medium did not antagonize GS expression neither in GS nor in MSO-treated GS+ cells. Likewise, when cultured in the absence of glutamine fetal and mature control cells did not show an increase of GS expression. Thus, at the level of GS expression we did not observe the product feedback inhibition as demonstrated in L2-cells [27] but the difference might be explained by the fact that we examined another cell system with an overall lower level of GS expression. Further, different pools of intracellular glutamine may exist that differentially affect cell proliferation [31]. Glutamine deprivation was shown to increase apoptosis as noted by an increased DNA fragmentation and caspase-3 activity [32]. Along this line, glutamine deprivation-mediated cell shrinkage promoted a ligand-independent activation of the CD95-mediated apoptotic pathway in leukemia-derived CEM and HL-60 cells [33]. Thus, we expected an increase in the rate of apoptosis of GS fibroblasts in comparison to wild type cells. However, variable extracellular glutamine concentrations ranging from 0 to 10 mM yielded no influence on the rate of apoptosis in either of the three cell lines tested. This is in contrast to other studies [23,32,34] but again, the difference might be explained by a reduced functional importance of GS for skin fibroblasts in comparison to other cell types as demonstrated by the relatively low level of its expression. High concentrations of extracellular glutamine resulted in an increased rate of cell proliferation as measured by the uptake of [3H]-thymidine into DNA of preconfluent rat intestinal epithelia cells [18]. We therefore investigated the influence of glutamine on skin fibroblast proliferation and could confirm the expected decreased rate of proliferation of fetal fibroblasts in the absence of glutamine. Fetal GS+ cells showed a significant higher proliferation

101

rate in comparison to mature cells, a finding that might reflect the overall lower need of a postnatal organism for rapid cell proliferation. These findings were confirmed in another experiment when the rate of cell proliferation was investigated by the use of CFDA. Hereby, a similar pattern was elucidated with respect to the developmental stage of the cell lines and the influence of extracellular glutamine. Regarding the intracellular concentration of glutamine, a decreased level in the GS cells was hypothesized but could not be substantiated. While glutamate was the most abundant intracellular amino acid, glutamine was found in comparatively low concentrations in all cell lines. This may reside on the considerable overexpression of GS in the GS cells and could explain why the GS genotype was largely without effect on cell viability. Alternatively, distinct intracellular glutamine pools might exist which would not have been unveiled by the applied method [31]. Also, changes in the transport of glutamine and glutamate into the mutated cells might confound differences in the intracellular concentrations of these amino acids. In conclusion, the study corroborates the function of GS at early fetal stages and supports its importance for fetal cell proliferation. This adds another important role of GS for the human organism.

Acknowledgments The authors are grateful to the technical help of I. Neumann, M. Grüneberg, M. Steinert, T. Janssen, and J. Beckstedde. The work has been supported in part by a grant from the Deutsche Forschungsgemeinschaft to J.H. (HA 4376/1-2) and by the Sonderforschungsbereich 575 ‘‘Experimental Hepatology”, Düsseldorf.

References [1] Y. Kumada, D.R. Benson, D. Hillemann, T.J. Hosted, D.A. Rochefort, C.J. Thompson, W. Wohlleben, Y. Tateno, Proc. Natl. Acad. Sci. USA 90 (1993) 3009–3013. [2] J. Häberle, B. Görg, A. Toutain, F. Rutsch, J.F. Benoist, A. Gelot, A.L. Suc, H.G. Koch, F. Schliess, D. Häussinger, J. Inherit. Metab. Dis. 29 (2006) 352–358. [3] D. Häussinger, Adv. Enzymol. Relat. Areas Mol. Biol. 72 (1998) 43–86. [4] D. Häussinger, H. Sies, Biochem. J. 221 (1984) 651–658. [5] F. Gaunitz, D. Deichsel, K. Heise, M. Werth, U. Anderegg, R. Gebhardt, Hepatology 41 (2005) 1225–1232. [6] D. Häussinger, Biochem. J. 267 (1990) 281–290. [7] D. Häussinger, J. Parenter. Enteral. Nutr. 14 (1990) 56–62. [8] I. Suarez, G. Bodega, B. Fernandez, Neurochem. Int. 41 (2002) 123–142. [9] J. Häberle, B. Görg, F. Rutsch, E. Schmidt, A. Toutain, J.F. Benoist, A. Gelot, A.L. Suc, W. Höhne, F. Schliess, D. Häussinger, H.G. Koch, N. Engl. J. Med. 353 (2005) 1926–1933. [10] J.W. Gaasbeek Janzen, R. Gebhardt, G.H. ten Voorde, W.H. Lamers, R. Charles, A.F. Moorman, J. Histochem. Cytochem. 35 (1987) 49–54. [11] K.M. Mearow, J.F. Mill, L. Vitkovic, Brain Res. Mol. Brain Res. 6 (1989) 223–232. [12] J. Akimoto, H. Itoh, T. Miwa, K. Ikeda, Brain Res. Dev. Brain Res. 72 (1993) 9–14. [13] J.T. Self, T.E. Spencer, G.A. Johnson, J. Hu, F.W. Bazer, G. Wu, Biol. Reprod. 70 (2004) 1444–1451. [14] F.C. Battaglia, T.R. Regnault, Placenta 22 (2001) 145–161. [15] D. Novak, M. Lehman, H. Bernstein, M. Beveridge, S. Cramer, Placenta 27 (2006) 510–516. [16] Y. He, T.B. Hakvoort, J.L. Vermeulen, W.H. Lamers, M.A. Van Roon, Dev. Dyn. 236 (2007) 1865–1875. [17] Y. Rezk, C. Huff, B. Rizk, Am. J. Obstet. Gynecol. 190 (2004) 1450–1454. [18] V. Demarco, K. Dyess, D. Strauss, C.M. West, J. Neu, J. Nutr. 129 (1999) 57–62. [19] D. Häussinger, W. Gerok, Eur. J. Biochem. 136 (1983) 421–425. [20] F. Schliess, R. Sinning, R. Fischer, C. Schmalenbach, D. Häussinger, Biochem. J. 320 (1996) 167–171. [21] W. Liu, D.A. Saint, Anal. Biochem. 302 (2002) 52–59. [22] R.A. Ronzio, W.B. Rowe, A. Meister, Biochemistry 8 (1969) 1066–1075. [23] H.G. Windmueller, Adv. Enzymol. Relat. Areas Mol. Biol. 53 (1982) 201–237. [24] D.O. Jacobs, D.A. Evans, K. Mealy, S.T. O’Dwyer, R.J. Smith, D.W. Wilmore, Surgery 104 (1988) 358–364. [25] D. Häussinger, F. Schliess, Front. Biosci. 12 (2007) 371–391. [26] F. Lang, G.L. Busch, M. Ritter, H. Völkl, S. Waldegger, E. Gulbins, D. Häussinger, Physiol. Rev. 78 (1998) 247–306. [27] B.I. Labow, W.W. Souba, S.F. Abcouwer, J. Nutr. 131 (2001) 2467–2474.

102

T. Vermeulen et al. / Archives of Biochemistry and Biophysics 478 (2008) 96–102

[28] B.M. Rotoli, J. Uggeri, V. Dall’Asta, R. Visigalli, A. Barilli, R. Gatti, G. Orlandini, G.C. Gazzola, O. Bussolati, Cell Physiol. Biochem. 15 (2005) 281–292. [29] A.M. Aslanian, M.S. Kilberg, Biochem. J. 358 (2001) 59–67. [30] A.M. Aslanian, B.S. Fletcher, M.S. Kilberg, Biochem. J. 357 (2001) 321–328. [31] E. Ziemin´ska, W. Hilgier, H.S. Waagepetersen, L. Hertz, U. Sonnewald, A. Schousboe, J. Albrecht, Neurochem. Res. 29 (2004) 2121–2123.

[32] S.D. Larson, J. Li, D.H. Chung, B.M. Evers, Am. J. Physiol. Gastrointest. Liver Physiol. 293 (2007) G1262–G1271. [33] C. Fumarola, A. Zerbini, G.G. Guidotti, Cell Death. Differ. 8 (2001) 1004–1013. [34] H.T. Papaconstantinou, K.O. Hwang, S. Rajaraman, M.R. Hellmich, C.M. Townsend Jr., T.C. Ko, Surgery 124 (1998) 152–159.