- Email: [email protected]
Mycophenolic acid does not inhibit protein glycosylation in T lymphocytes
Transplant Immunology 8 Ž2000. 169᎐175
Mycophenolic acid does not inhibit protein glycosylation in T lymphocytes Scott JepsonU , Iain J. Brogan, R.W. Stoddart, Ian V. Hutchinson School of Biological Sciences, The Uni¨ ersity of Manchester, Oxford Road, Manchester M13 9PT, UK Received 6 June 2000; accepted 10 July 2000
Abstract Background: Mycophenolic acid inhibits guanosine nucleotide synthesis and has been shown to be a potent inhibitor of lymphocyte proliferation as well as being effective at decreasing the incidence of graft rejection. Guanosine nucleosides are essential for protein glycosylation and many cell surface proteins including adhesion molecules, which are important for graft infiltration and rejection, are glycoproteins. There have been conflicting reports concerning the ability of MPA to interfere with glycosylation in lymphoid cells. Therefore, the purpose of this study was to investigate the effects of MPA on cell surface protein glycosylation in lymphoid cells. Methods: Cells were cultured in the presence of increasing concentrations of MPA for different lengths of time and stained with fluorescent-labelled lectins specific for either mannose or fucose residues on glycoproteins. Analysis was then performed by flow cytometry. Results: MPA treatment had no effect on the binding of either fucose or mannose-specific lectins to Con A stimulated human PBLs and rat lymph node lymphocytes or to a CEMC7a T cell line. Conclusion: The results show that, contrary to previous reports, MPA does not affect cell surface glycosylation in T cells using T cells from different sources of both human and non-human origin. 䊚 2000 Elsevier Science B.V. All rights reserved. Keywords: Mycophenolic acid; T lymphocytes; Glycosylation
1. Introduction The immunosuppressive drug mycophenolic acid ŽMPA. is a potent inhibitor of nucleic acid synthesis that suppresses proliferative responses of human peripheral blood B and T lymphocytes to mitogens and allogeneic lymphocytes w1x. MPA is administered as mycophenolate mofetil ŽMMF. the morpholinoethylester prodrug of MPA, which has an increased bioavailability compared with MPA due to enhanced absorption from the gastrointestinal tract ŽGIT. w2x. In normal mammalian cells, guanine and adenine U
Corresponding author. Department of Cell Biology and Anatomy, The University of Miami School of Medicine, 1550 NW 10th Ave, Miami FL 33136, USA. Tel.: q1-305-243-5812; fax: q1-305243-4431. E-mail address: [email protected] ŽS. Jepson..
nucleotides are manufactured from small precursor molecules through the de novo pathway or by recycling purine bases through the salvage pathway w3x. Resting lymphocytes meet their metabolic requirements by salvage pathways of purine nucleotide synthesis, but upon stimulation with mitogen Že.g. PHA. there is an increase in salvage pathway activity and more importantly there is also activation of de novo purine biosynthesis w4x. This higher activity of de novo synthesis in proliferating lymphocytes compared to resting cells is thought to be due to failure of the salvage pathway to provide sufficient nucleotides for DNA synthesis w5x. MPA is a potent, non-competitive, reversible inhibitor of inosine monophosphate dehydrogenase ŽIMPDH. the rate-limiting enzyme in the de novo synthesis of guanine nucleotides. IMPDH catalyses the NADqdependent oxidation of inosine monophosphate to xanthosine monophosphate at the branch point in the
0966-3274r00r$ - see front matter 䊚 2000 Elsevier Science B.V. All rights reserved. PII: S 0 9 6 6 - 3 2 7 4 Ž 0 0 . 0 0 0 2 3 - X
170
S. Jepson et al. r Transplant Immunology 8 (2000) 169᎐175
purine metabolic pathway. IMPDH activity, mediated by two isoenzymes denoted IMPDH type I and type II is, therefore, essential for providing the necessary precursors for DNA and RNA synthesis. Post-translational modification of proteins by glycosylation is an important process that modifies both their structure and function. Glycoproteins contain oligosaccharide chains covalently attached to specific amino acid residues within the polypeptide chain. Glycoprotein synthesis involves the transfer of sugar residues from carrier molecules to the growing oligosaccharide chain. Importantly, the carrier molecule for the monosaccharides mannose and fucose is GDP. MPA was postulated to interfere with glycoprotein synthesis by inhibiting the transfer of mannose and fucose to the oligosaccharide chain because of its ability to deplete guanosine nucleotides. However, there have been few convincing reports demonstrating the effect of MPA on lymphocyte glycoprotein synthesis in vitro, despite it being postulated as a mechanism of action of the immunosuppressive effect of MPA. Experiments to look at the in vivo effects of MMF showed that grafts in MMF treated allograft recipients survived longer than controls and had better preserved graft structure w6x. Cellular infiltration was also decreased by MMF, and there was decreased binding of lymphocytes to the graft, suggesting that MPA may be acting to change glycoproteins structures on the recipients lymphocytes or on the donor tissue. We have attempted to further define the role of MPA in protein glycosylation by investigating the effect of MPA on cell surface protein glycosylation in T cells of different origins, including a T cell line. The cells chosen for this study were human PBLs, rat lymph node lymphocytes and CEM-C7a T cells. Cells were cultured for varying lengths of time in the presence of different concentrations of MPA then stained with fluorescently labelled lectins, specific for mannose or fucose residues, and analysed by flow cytometry. In contrast to previous reports, we were unable to detect any effect of MPA on the binding of fucose or mannose-specific lectins to any of the cells used in the study. By contrast treatment with tunicamycin, a known inhibitor of glycosylation, was able to decrease glycosylation. We, therefore, conclude that the immunosuppressive effect of MPA is not due to an effect on lymphocyte glycosylation.
2. Materials and methods 2.1. Materials Mycophenolic acid was purchased from Sigma ŽPoole, UK. and reconstituted in DMSO at 0.1 M then diluted in medium to the required concentration. Tunicamycin
was also purchased from Sigma. All lectins were purchased from Sigma and reconstituted at 1 mgrml in DAB-2-azide for FACs staining. The major specificities of the lectins used were: Lotus tetragonolobus ŽLTA.Lfuc ␣ 1,2Gal1,4 Ž Lfuc ␣ 1,3-GlcNAc; Vicia faba ŽVFA. ᎐branched mannose residues 2.2. Cell culture CEM-C7a cells were kindly provided by Dr Ged Brady, School of Biological Sciences, University of Manchester. CEM-C7 cells were cultured in suspension in Optimem ŽGibco BRL. supplemented with 5% fetal calf serum, 100 units penicillin, 0.1 mgrml streptomycin and 2 mM glutamine. Rat lymph node cells were dissected from mesenteric, cervical and axilliary lymph nodes then washed under sterile conditions and resuspended in RPMI Žsupplemented with 10% fetal calf serum, 2 mM glutamine, 1 = non-essential amino acids and 100 units penicillinr0.1 mgrml streptomycin, 1 mM sodium pyruvate and 10 mM hepes solution.. Human peripheral blood mononuclear cells ŽHuPBMC. were purified from the blood of normal healthy volunteers in the laboratory. The blood was centrifuged on histopaque and the layer containing the mononuclear cells removed and the cells cultured in RPMI Žsupplemented as for rat LNCs.. 2.3. Measuring cellular proliferation by 3[H]thymidine incorporation Cells were cultured at 1 = 10 6 cellsrml in 96-well flat bottom plates for 72 h in a humidified incubator Ž37⬚C, 5% CO 2 .. They were pulsed for the final 24 h by adding 1 Ci of 3 wHxthymidine ŽAmersham. into each well. The cells were then harvested ŽDynatech Multimash 2000. onto filter paper and dried in scintillation vials ŽSarstedt. in an oven Ž37⬚C. for 1᎐2 h. Scintillation fluid Ž2 ml. ŽOptiphase Hi-safe, fisons chemicals. was added to each vial and thymidine incorporation measured using a scintillation counter ŽBeckman LS1801. as described by the manufacturers. 2.4. Flow cytometry Cells Ž1 = 10 6rml. were cultured for 24, 48 or 72 h at 37⬚C, 5% CO 2 in a humidified incubator in 24 well plates. Prior to staining, cells were counted by trypan blue exclusion using a haemocytometer and 1 = 10 5 cells centrifuged in FACS tubes ŽAlpha Laboratories. at high speed for 1 min and the cell pellet washed twice with DAB-2-azide ŽDAB containing 2% FCS Žvrv., 20 mM sodium azide.. After the final wash, cells were incubated, on ice, with FITC-conjugated lectin for 30 min. The cells were again washed with DAB-2-azide
S. Jepson et al. r Transplant Immunology 8 (2000) 169᎐175
Žthree times. and after the final wash fixed by resuspending in 1% formaldehyde Žvrv. in PBS before analysis by flow cytometry ŽBeckton Dickinson Facscan..
3. Results Initial experiments were to determine the effect of MPA on proliferation in a number of different cell types. As can be seen in Fig. 1, MPA is a potent inhibitor of proliferation in all the cells with similar sensitivity in all cases. The effect of MPA on cell surface protein glycosylation was assessed by flow cytometry using fluorescently labelled lectins. Different lectins specific for either mannose or fucose residues attached to glycoproteins were used to identify alterations in their incorporation into cell surface glycoproteins. Rat LNCs stimulated with Con A in the presence or absence of MPA were stained with either the fucose-specific lectin Lotus tetragonolobus ŽLTA. or the mannose-specific lectins Vicia faba ŽVFA.. However, we did not detect any differences in lectin staining between MPA treated cells and controls, using either of the fucose or mannose-specific lectins ŽFig. 2.. Subsequently, human peripheral blood mononuclear cells ŽHuPBMC. isolated from normal healthy human
171
volunteers were used to determine whether MPA may affect glycosylation in human cells if not in rat cells. Fig. 3 shows the effect of MPA on the binding of fluorescently labelled lectins to Con A stimulated HuPBMCs. Again, there were no differences in the binding of the fucose and mannose-specific lectins LTA and VFA to Con A stimulated HuPBMCs following MPA treatment. Other lectins specific for fucose and mannose were also used and they too did not reveal any differences between control and MPA treated cells Ždata not shown.. The experiments with rat LNCs and HuPBMCs described above are performed with heterogeneous populations of cells. Therefore, the use of CEM-C7a cells, a human T cell line, enabled us to look at the specific effect of MPA on an isolated T cell population. CEM-C7a cells were cultured with MPA for 24, 48 and 72 h and then stained with either Lotus tetragonolobus ŽLTA., or Vicia faba ŽVFA.. MPA treated cells showed no difference in lectin binding when compared to control cells at 24, 48, and 72 h ŽFigs. 4 and 5.. However, when a known inhibitor of protein glycosylation was included, tunicamycin, fluorescence intensity was decreased significantly showing that it was possible to detect changes in lectin binding and fluorescence ŽFig. 4, insert.. These results, therefore, suggest that
Fig. 1. The effect of MPA on proliferation in HUPBMCs Ža., Rat LNCs Žb. and CEM-C7a T cells Žc.. Proliferation was measured by Hxthymidine incorporation after 72 h culture in the presence of increasing concentration of MPA. The results shown are the means of four individual experiments except where indicated and are plotted as the percentage of the cpm from stimulated cells alone. The symbols U and UU denote P- 0.05 and - 0.01, respectively, one-way ANOVA followed by Dunnetts multiple comparisons test. 3w
172
S. Jepson et al. r Transplant Immunology 8 (2000) 169᎐175
Fig. 2. Lack of effect of MPA on lectin binding to Con A stimulated rat lymph node cells. Rat lymph nodes were removed and cells extracted then cultured for 72 h in the presence of Con A and MPA Ž1 = 10y6 M. before removing for FACs analysis with different lectins. The figures show the effect of MPA Žblack line. on lectin binding compared with Con A stimulated cells alone Žfilled trace.. The figures shown are representative of three individual experiments.
Fig. 3. Lack of effect of MPA on lectin binding to Con A stimulated human peripheral blood mononuclear cells. HuPBMCs were cultured for 72 h in the presence of Con A and MPA Ž1 = 10y6 M. before removing for FACs analysis with different lectins. The figures show the effect of MPA Žblack line. on lectin binding compared with Con A stimulated cells alone Žfilled trace.. The figures shown are representative of three individual experiments.
MPA has no effect on expression of cell surface fucose in CEM-C7a cells. There was also no difference in fluorescent intensities between MPA treated cells and controls when cells were stained with the mannose-specific lectin VFA, except again for tunicamycin treated cells. Tunicamycin
treatment caused a dramatic decrease in fluorescence intensity ŽFig. 5, insert.. We were, therefore, unable to show any effect of MPA on the cell-surface expression of mannose or fucose residues in CEM-C7a T cells.
Fig. 4. Lack of effect of MPA on the binding of LTA to fucose residues on CEM-C7a T cells. Cells were cultured with MPA for: Ža. 24 h; Žb. 48 h; and Žc. 72 h. The figures show the effect of MPA 1 = 10y6 M Žsolid black line. compared with untreated cells Žfilled trace.. The figure inset shows the effect of tunicamycin Ž10 grml, solid black line. a known inhibitor of glycosylation.
S. Jepson et al. r Transplant Immunology 8 (2000) 169᎐175
173
Fig. 5. Lack of effect of MPA on the binding of VFA to mannose residues CEM-C7a T cells. Cells were cultured with MPA for: Ža. 24 h; Žb. 48 h; and Žc. 72 h. The figures show the effect of MPA 1 = 10y6 M Žsolid black line. compared with untreated cells Žfilled trace.. The figure inset shows the effect of tunicamycin Ž10 grml, solid black line. a known inhibitor of glycosylation.
4. Discussion We have demonstrated that MPA treatment has no effect on the cell surface expression of mannose and fucose on rat LNCs, human PBLs or CEM-C7a T cells. These data are in contrast to a previous report in which experiments by Allison et al. w7x suggested that MPA inhibited the expression of mannose containing oligosaccharides on Con A stimulated human peripheral blood lymphocytes using the mannose-specific lectin VFA. Experiments in vivo w6x showed mycophenolate mofetil ŽMMF. treated rats to have a decreased cellular infiltrate in kidney allografts, resulting in better graft preservation. The authors attributed this to an effect of MMF on lymphocyte adhesion molecules and possible interactions with macrophages. In addition, MPA was shown to cause a decrease in mannosylation of membrane glycoproteins in human monocytes and decrease the adhesiveness of these cells to human umbilical vein endothelial cells ŽHUVECs. and laminin w8x. We did not look at the effect of MPA on monocytes and, therefore, cannot rule out the possibility of there being different effects on cells from different lineages. We have used primary cells from heterogeneous lymphocyte populations of both human and rat origin and also a homogeneous human T cell line. However, we were unable to show any difference between control and MPA-treated groups with respect to lectin binding in any of the cells tested. In addition, the effective concentration of MPA for inhibition of cellular prolif-
eration was very similar between the different cell types suggesting they show similar sensitivity toward MPA. It may be argued that our methodology for detecting lectin binding was not sensitive enough to detect changes in the binding of the different lectins to fucose and mannose residues. However, we were able to detect a significant decrease in lectin binding to cells cultured in the presence of tunicamycin, a known inhibitor of glycosylation. Tunicamycin inhibits the first reaction in the lipid-saccharide pathway, the formation of Dol-PP-N-acetylglucosamine from UDP-Nacetylglucosamine and dolicholphosphate w9᎐11x. Therefore, it was possible to detect changes in cell surface expression of mannose and fucose with the use of tunicamycin. However, when we cultured cells with MPA we were unable to detect any differences in lectin binding to CEM-C7a cells. Following mitogen stimulation of HuPBLs, there is a 10-fold increase in the rate on N-glycosylation w12x. The effect on O-glycosylation is less clear, because following mitogen stimulation of murine T cells there is an increase in O-glycosylation of nuclear proteins but a decrease in glycosylation of cytosolic proteins w13x. We have used activated cells in all our experiments and, therefore, should expect them to have an increased rate of glycosylation. In addition, the differences detected following tunicamycin treatment suggest that the glycoproteins are turning over and it is possible to detect differences in protein glycosylation using our methodology.
174
S. Jepson et al. r Transplant Immunology 8 (2000) 169᎐175
An explanation for the effect of MPA on lymphocyte proliferation but not glycosylation may be that intracellular nucleotides are not homogeneously distributed within a cell and that compartmentalisation of intracellular nucleotide pools does occur w14x. In lymphocytes it has been suggested that there are two pools of dTTPs with one of them being fed by de novo synthesis and used preferentially for DNA synthesis. From experiments using MOLT-4 cells, an acute lymphoblastic leukaemia cell line, it was suggested that a discrete compartment of thymine nucleotides exists serving exclusively as precursors for DNA synthesis w15x. This suggests that another pool of dTTP exists which is not available for DNA synthesis. Similar evidence has been found for CTP pools, with one pool being preferentially available for DNA synthesis w16x. The effect of compartmentalisation of guanine ribonucleotides has also been studied w17x. It was shown in S-49 lymphoma cells that reduction of DNA synthesis caused by inhibition of IMPDH is associated with depletion of guanine ribonucleotide pools rather than dGTP pools ᎏ the latter were found not to be available for DNA synthesis. This indicated that a discrete guanine ribonucleotide precursor pool might play a crucial role in limiting substrate supply for DNA replication under conditions of purine starvation. These data showed that the immediate and dramatic inhibition of DNA synthesis and, therefore, proliferation, following MPA treatment of cells occurs before guanine nucleotides other than GMP are exhausted. It is, therefore, possible that following treatment of lymphocytes with MPA, GMP pools utilised for DNA synthesis and cellular proliferation are depleted leading to inhibition of these processes, but that sufficient levels of GDP remain within a separate cellular pool to allow protein glycosylation to continue normally. We have conducted preliminary experiments using HUVECs treated with MPA, which suggest that MPA may be able to selectively interfere with glycosylation in endothelial cells. This stands in contrast to our findings in lymphocytes, however, similar effects on endothelial cells have been published elsewhere w18᎐20x. This effect on the endothelium would contribute to the immunosuppressive effects of MPA because glycosylation of cell᎐cell adhesion molecules is important for their function. If glycosylation in endothelial cells is inhibited, leucocyte adhesion to the graft endothelium would be disrupted and thus immune cell extravasation into the tissue would consequently be reduced. These data, therefore, provide a potential mechanism to explain the decreased graft infiltration following MPA treatment and suggest that the immunosuppressive effects of MPA are not a result of effects directly on lymphocyte glycosylation per se.
Acknowledgements This work was supported by a grant from the Biotechnology and Biological Sciences Research Council ŽBBSRC.. References w1x Eugui EM, Almquist SJ, Muller CD, Allison AC. Lymphocyteselective cytostatic and immunosuppressive effects of mycophenolic acid in vitro- role of deoxyguanosine nucleotide depletion. Scand J Immunol 1991;33:161᎐173. w2x Sievers TM, Rossi SJ, Ghobrial RM et al. Mycophenolate mofetil. Pharmacotherapy 1997;17:1178᎐1197. w3x Allison AC, Eugui EM. Immunosuppressive and other effects of mycophenolic acid and an ester prodrug, mycophenolate mofetil. Immunol Rev 1993;136:5᎐28. w4x Barankiewicz J, Cohen A. Purine nucleotide metabolism in phytohemagglutinin-induced human T lymphocytes. Arch Biochem Biophys 1987;258:167᎐175. w5x Marijnen YMT, de Korte D, Haverkort WA, den Breejen EJS, van Gennip AH, Roos D. Studies on the incorporation of precursors into purine and pyrimidine nucleotides via ‘de novo’ and ‘salvage’ pathways in normal lymphocytes and lymphoblastic cell-line cells. Biochim Biophys Acta 1989;1012:148᎐155. w6x Heeman U, Azuma H, Hamar P, Schmid C, Tilney N, Philipp T. Mycophenolate mofetil inhibits lymphocyte binding and the upregulation of adhesion molecules in acute rejection of rat kidney allografts. Transplant Immunol 1996;4:64᎐67. w7x Allison AC, Kowalski WJ, Muller CD, Waters RV, Eugui EM. Mycophenolic acid and brequinar, inhibitors of purine and pyrimidine synthesis, block the glycosylation of adhesion molecules. Transplant Proc 1993;25Žsuppl 2.:67᎐70. w8x Laurent AF, Dumont S, Poindron P, Muller CD. Mycophenolic acid suppresses protein N-linked glycosylation in human monocytes and their adhesion to endothelial cells and to some substrates. Exp Haematol 1996;24:59᎐67. w9x Speake BK, Hemming FW, White DA. The effects of tunicamycin on protein glycosylation in mammalian and fungal systems. Biochem Soc Trans 1980;8:166᎐168. w10x Schwarz RT, Datema R. Inhibitors of protein glycosylation. Trends Biochem Sci 1980;5:65᎐68. w11x Schwarz RT, Datema R. The lipid pathway of protein glycosylation and its inhibitors: the biological significance of proteinbound carbohydrates. Adv Carbohydrate Chem Biochem 1982;40:287᎐379. w12x Kumar V, Heinemann FS, Ozols J. Interleukin-2 induces N-glycosylation in T cells: characterization of human lymphocyte oligosaccharyltransferase. Biochem Biophys Res Commun 1998;247:524᎐529. w13x Kearse KP, Hart GW. Lymphocyte activation induces rapid changes in nuclear and cytoplasmic glycoproteins. Proc Natl Acad Sci USA 1991;88:1701᎐1705. w14x Moyer JD, Henderson JF. Compartmentation of intracellular nucleotides in mammalian cells. CRC Crit Rev Biochem 1985;19:45᎐61. w15x Taheri MR, Wickremasinghe RG, Hoffbrand AV. Functional compartmentation of DNA precursors in human leukaemoblastoid cell lines. Br J Haematol 1982;52:401᎐409. w16x Reichard P. Regulation of deoxyribotide synthesis. Biochemistry 1987;26:3245᎐3248. w17x Nguyen BT, Sadee W. Compartmentation of guanine nucleotide precursors for DNA synthesis. Biochem J 1986; 234:263᎐269.
S. Jepson et al. r Transplant Immunology 8 (2000) 169᎐175 w18x Hauser IA, Johnson DR, Thevenod F, Goppelt-Strube M. Effect of mycophenolic acid on TNF␣-induced expression of cell adhesion molecules in human venous endothelial cells in vitro. Br J Pharmacol 1997;122:1315᎐1322. w19x Bertalanffy P, Dubsky P, Seebacher G, Griesmacher A, Weigel G, Wolner E. Mycophenolic acid influences the expression of
175
ICAM-1 on human endothelial cells. Adv Exp Med Biol 1998;43:565᎐568. w20x Bertalanffy P, Dubsky P, Wolne E, Weige G. Alterations of endothelial nucleotide levels by mycohenolic acid result in changes of membrane glycosylation and E-selectin expression. Clin Chem Lab Med 1999;37:259.
Mycophenolic acid does not inhibit protein glycosylation in T lymphocytes Scott JepsonU , Iain J. Brogan, R.W. Stoddart, Ian V. Hutchinson School of Biological Sciences, The Uni¨ ersity of Manchester, Oxford Road, Manchester M13 9PT, UK Received 6 June 2000; accepted 10 July 2000
Abstract Background: Mycophenolic acid inhibits guanosine nucleotide synthesis and has been shown to be a potent inhibitor of lymphocyte proliferation as well as being effective at decreasing the incidence of graft rejection. Guanosine nucleosides are essential for protein glycosylation and many cell surface proteins including adhesion molecules, which are important for graft infiltration and rejection, are glycoproteins. There have been conflicting reports concerning the ability of MPA to interfere with glycosylation in lymphoid cells. Therefore, the purpose of this study was to investigate the effects of MPA on cell surface protein glycosylation in lymphoid cells. Methods: Cells were cultured in the presence of increasing concentrations of MPA for different lengths of time and stained with fluorescent-labelled lectins specific for either mannose or fucose residues on glycoproteins. Analysis was then performed by flow cytometry. Results: MPA treatment had no effect on the binding of either fucose or mannose-specific lectins to Con A stimulated human PBLs and rat lymph node lymphocytes or to a CEMC7a T cell line. Conclusion: The results show that, contrary to previous reports, MPA does not affect cell surface glycosylation in T cells using T cells from different sources of both human and non-human origin. 䊚 2000 Elsevier Science B.V. All rights reserved. Keywords: Mycophenolic acid; T lymphocytes; Glycosylation
1. Introduction The immunosuppressive drug mycophenolic acid ŽMPA. is a potent inhibitor of nucleic acid synthesis that suppresses proliferative responses of human peripheral blood B and T lymphocytes to mitogens and allogeneic lymphocytes w1x. MPA is administered as mycophenolate mofetil ŽMMF. the morpholinoethylester prodrug of MPA, which has an increased bioavailability compared with MPA due to enhanced absorption from the gastrointestinal tract ŽGIT. w2x. In normal mammalian cells, guanine and adenine U
Corresponding author. Department of Cell Biology and Anatomy, The University of Miami School of Medicine, 1550 NW 10th Ave, Miami FL 33136, USA. Tel.: q1-305-243-5812; fax: q1-305243-4431. E-mail address: [email protected] ŽS. Jepson..
nucleotides are manufactured from small precursor molecules through the de novo pathway or by recycling purine bases through the salvage pathway w3x. Resting lymphocytes meet their metabolic requirements by salvage pathways of purine nucleotide synthesis, but upon stimulation with mitogen Že.g. PHA. there is an increase in salvage pathway activity and more importantly there is also activation of de novo purine biosynthesis w4x. This higher activity of de novo synthesis in proliferating lymphocytes compared to resting cells is thought to be due to failure of the salvage pathway to provide sufficient nucleotides for DNA synthesis w5x. MPA is a potent, non-competitive, reversible inhibitor of inosine monophosphate dehydrogenase ŽIMPDH. the rate-limiting enzyme in the de novo synthesis of guanine nucleotides. IMPDH catalyses the NADqdependent oxidation of inosine monophosphate to xanthosine monophosphate at the branch point in the
0966-3274r00r$ - see front matter 䊚 2000 Elsevier Science B.V. All rights reserved. PII: S 0 9 6 6 - 3 2 7 4 Ž 0 0 . 0 0 0 2 3 - X
170
S. Jepson et al. r Transplant Immunology 8 (2000) 169᎐175
purine metabolic pathway. IMPDH activity, mediated by two isoenzymes denoted IMPDH type I and type II is, therefore, essential for providing the necessary precursors for DNA and RNA synthesis. Post-translational modification of proteins by glycosylation is an important process that modifies both their structure and function. Glycoproteins contain oligosaccharide chains covalently attached to specific amino acid residues within the polypeptide chain. Glycoprotein synthesis involves the transfer of sugar residues from carrier molecules to the growing oligosaccharide chain. Importantly, the carrier molecule for the monosaccharides mannose and fucose is GDP. MPA was postulated to interfere with glycoprotein synthesis by inhibiting the transfer of mannose and fucose to the oligosaccharide chain because of its ability to deplete guanosine nucleotides. However, there have been few convincing reports demonstrating the effect of MPA on lymphocyte glycoprotein synthesis in vitro, despite it being postulated as a mechanism of action of the immunosuppressive effect of MPA. Experiments to look at the in vivo effects of MMF showed that grafts in MMF treated allograft recipients survived longer than controls and had better preserved graft structure w6x. Cellular infiltration was also decreased by MMF, and there was decreased binding of lymphocytes to the graft, suggesting that MPA may be acting to change glycoproteins structures on the recipients lymphocytes or on the donor tissue. We have attempted to further define the role of MPA in protein glycosylation by investigating the effect of MPA on cell surface protein glycosylation in T cells of different origins, including a T cell line. The cells chosen for this study were human PBLs, rat lymph node lymphocytes and CEM-C7a T cells. Cells were cultured for varying lengths of time in the presence of different concentrations of MPA then stained with fluorescently labelled lectins, specific for mannose or fucose residues, and analysed by flow cytometry. In contrast to previous reports, we were unable to detect any effect of MPA on the binding of fucose or mannose-specific lectins to any of the cells used in the study. By contrast treatment with tunicamycin, a known inhibitor of glycosylation, was able to decrease glycosylation. We, therefore, conclude that the immunosuppressive effect of MPA is not due to an effect on lymphocyte glycosylation.
2. Materials and methods 2.1. Materials Mycophenolic acid was purchased from Sigma ŽPoole, UK. and reconstituted in DMSO at 0.1 M then diluted in medium to the required concentration. Tunicamycin
was also purchased from Sigma. All lectins were purchased from Sigma and reconstituted at 1 mgrml in DAB-2-azide for FACs staining. The major specificities of the lectins used were: Lotus tetragonolobus ŽLTA.Lfuc ␣ 1,2Gal1,4 Ž Lfuc ␣ 1,3-GlcNAc; Vicia faba ŽVFA. ᎐branched mannose residues 2.2. Cell culture CEM-C7a cells were kindly provided by Dr Ged Brady, School of Biological Sciences, University of Manchester. CEM-C7 cells were cultured in suspension in Optimem ŽGibco BRL. supplemented with 5% fetal calf serum, 100 units penicillin, 0.1 mgrml streptomycin and 2 mM glutamine. Rat lymph node cells were dissected from mesenteric, cervical and axilliary lymph nodes then washed under sterile conditions and resuspended in RPMI Žsupplemented with 10% fetal calf serum, 2 mM glutamine, 1 = non-essential amino acids and 100 units penicillinr0.1 mgrml streptomycin, 1 mM sodium pyruvate and 10 mM hepes solution.. Human peripheral blood mononuclear cells ŽHuPBMC. were purified from the blood of normal healthy volunteers in the laboratory. The blood was centrifuged on histopaque and the layer containing the mononuclear cells removed and the cells cultured in RPMI Žsupplemented as for rat LNCs.. 2.3. Measuring cellular proliferation by 3[H]thymidine incorporation Cells were cultured at 1 = 10 6 cellsrml in 96-well flat bottom plates for 72 h in a humidified incubator Ž37⬚C, 5% CO 2 .. They were pulsed for the final 24 h by adding 1 Ci of 3 wHxthymidine ŽAmersham. into each well. The cells were then harvested ŽDynatech Multimash 2000. onto filter paper and dried in scintillation vials ŽSarstedt. in an oven Ž37⬚C. for 1᎐2 h. Scintillation fluid Ž2 ml. ŽOptiphase Hi-safe, fisons chemicals. was added to each vial and thymidine incorporation measured using a scintillation counter ŽBeckman LS1801. as described by the manufacturers. 2.4. Flow cytometry Cells Ž1 = 10 6rml. were cultured for 24, 48 or 72 h at 37⬚C, 5% CO 2 in a humidified incubator in 24 well plates. Prior to staining, cells were counted by trypan blue exclusion using a haemocytometer and 1 = 10 5 cells centrifuged in FACS tubes ŽAlpha Laboratories. at high speed for 1 min and the cell pellet washed twice with DAB-2-azide ŽDAB containing 2% FCS Žvrv., 20 mM sodium azide.. After the final wash, cells were incubated, on ice, with FITC-conjugated lectin for 30 min. The cells were again washed with DAB-2-azide
S. Jepson et al. r Transplant Immunology 8 (2000) 169᎐175
Žthree times. and after the final wash fixed by resuspending in 1% formaldehyde Žvrv. in PBS before analysis by flow cytometry ŽBeckton Dickinson Facscan..
3. Results Initial experiments were to determine the effect of MPA on proliferation in a number of different cell types. As can be seen in Fig. 1, MPA is a potent inhibitor of proliferation in all the cells with similar sensitivity in all cases. The effect of MPA on cell surface protein glycosylation was assessed by flow cytometry using fluorescently labelled lectins. Different lectins specific for either mannose or fucose residues attached to glycoproteins were used to identify alterations in their incorporation into cell surface glycoproteins. Rat LNCs stimulated with Con A in the presence or absence of MPA were stained with either the fucose-specific lectin Lotus tetragonolobus ŽLTA. or the mannose-specific lectins Vicia faba ŽVFA.. However, we did not detect any differences in lectin staining between MPA treated cells and controls, using either of the fucose or mannose-specific lectins ŽFig. 2.. Subsequently, human peripheral blood mononuclear cells ŽHuPBMC. isolated from normal healthy human
171
volunteers were used to determine whether MPA may affect glycosylation in human cells if not in rat cells. Fig. 3 shows the effect of MPA on the binding of fluorescently labelled lectins to Con A stimulated HuPBMCs. Again, there were no differences in the binding of the fucose and mannose-specific lectins LTA and VFA to Con A stimulated HuPBMCs following MPA treatment. Other lectins specific for fucose and mannose were also used and they too did not reveal any differences between control and MPA treated cells Ždata not shown.. The experiments with rat LNCs and HuPBMCs described above are performed with heterogeneous populations of cells. Therefore, the use of CEM-C7a cells, a human T cell line, enabled us to look at the specific effect of MPA on an isolated T cell population. CEM-C7a cells were cultured with MPA for 24, 48 and 72 h and then stained with either Lotus tetragonolobus ŽLTA., or Vicia faba ŽVFA.. MPA treated cells showed no difference in lectin binding when compared to control cells at 24, 48, and 72 h ŽFigs. 4 and 5.. However, when a known inhibitor of protein glycosylation was included, tunicamycin, fluorescence intensity was decreased significantly showing that it was possible to detect changes in lectin binding and fluorescence ŽFig. 4, insert.. These results, therefore, suggest that
Fig. 1. The effect of MPA on proliferation in HUPBMCs Ža., Rat LNCs Žb. and CEM-C7a T cells Žc.. Proliferation was measured by Hxthymidine incorporation after 72 h culture in the presence of increasing concentration of MPA. The results shown are the means of four individual experiments except where indicated and are plotted as the percentage of the cpm from stimulated cells alone. The symbols U and UU denote P- 0.05 and - 0.01, respectively, one-way ANOVA followed by Dunnetts multiple comparisons test. 3w
172
S. Jepson et al. r Transplant Immunology 8 (2000) 169᎐175
Fig. 2. Lack of effect of MPA on lectin binding to Con A stimulated rat lymph node cells. Rat lymph nodes were removed and cells extracted then cultured for 72 h in the presence of Con A and MPA Ž1 = 10y6 M. before removing for FACs analysis with different lectins. The figures show the effect of MPA Žblack line. on lectin binding compared with Con A stimulated cells alone Žfilled trace.. The figures shown are representative of three individual experiments.
Fig. 3. Lack of effect of MPA on lectin binding to Con A stimulated human peripheral blood mononuclear cells. HuPBMCs were cultured for 72 h in the presence of Con A and MPA Ž1 = 10y6 M. before removing for FACs analysis with different lectins. The figures show the effect of MPA Žblack line. on lectin binding compared with Con A stimulated cells alone Žfilled trace.. The figures shown are representative of three individual experiments.
MPA has no effect on expression of cell surface fucose in CEM-C7a cells. There was also no difference in fluorescent intensities between MPA treated cells and controls when cells were stained with the mannose-specific lectin VFA, except again for tunicamycin treated cells. Tunicamycin
treatment caused a dramatic decrease in fluorescence intensity ŽFig. 5, insert.. We were, therefore, unable to show any effect of MPA on the cell-surface expression of mannose or fucose residues in CEM-C7a T cells.
Fig. 4. Lack of effect of MPA on the binding of LTA to fucose residues on CEM-C7a T cells. Cells were cultured with MPA for: Ža. 24 h; Žb. 48 h; and Žc. 72 h. The figures show the effect of MPA 1 = 10y6 M Žsolid black line. compared with untreated cells Žfilled trace.. The figure inset shows the effect of tunicamycin Ž10 grml, solid black line. a known inhibitor of glycosylation.
S. Jepson et al. r Transplant Immunology 8 (2000) 169᎐175
173
Fig. 5. Lack of effect of MPA on the binding of VFA to mannose residues CEM-C7a T cells. Cells were cultured with MPA for: Ža. 24 h; Žb. 48 h; and Žc. 72 h. The figures show the effect of MPA 1 = 10y6 M Žsolid black line. compared with untreated cells Žfilled trace.. The figure inset shows the effect of tunicamycin Ž10 grml, solid black line. a known inhibitor of glycosylation.
4. Discussion We have demonstrated that MPA treatment has no effect on the cell surface expression of mannose and fucose on rat LNCs, human PBLs or CEM-C7a T cells. These data are in contrast to a previous report in which experiments by Allison et al. w7x suggested that MPA inhibited the expression of mannose containing oligosaccharides on Con A stimulated human peripheral blood lymphocytes using the mannose-specific lectin VFA. Experiments in vivo w6x showed mycophenolate mofetil ŽMMF. treated rats to have a decreased cellular infiltrate in kidney allografts, resulting in better graft preservation. The authors attributed this to an effect of MMF on lymphocyte adhesion molecules and possible interactions with macrophages. In addition, MPA was shown to cause a decrease in mannosylation of membrane glycoproteins in human monocytes and decrease the adhesiveness of these cells to human umbilical vein endothelial cells ŽHUVECs. and laminin w8x. We did not look at the effect of MPA on monocytes and, therefore, cannot rule out the possibility of there being different effects on cells from different lineages. We have used primary cells from heterogeneous lymphocyte populations of both human and rat origin and also a homogeneous human T cell line. However, we were unable to show any difference between control and MPA-treated groups with respect to lectin binding in any of the cells tested. In addition, the effective concentration of MPA for inhibition of cellular prolif-
eration was very similar between the different cell types suggesting they show similar sensitivity toward MPA. It may be argued that our methodology for detecting lectin binding was not sensitive enough to detect changes in the binding of the different lectins to fucose and mannose residues. However, we were able to detect a significant decrease in lectin binding to cells cultured in the presence of tunicamycin, a known inhibitor of glycosylation. Tunicamycin inhibits the first reaction in the lipid-saccharide pathway, the formation of Dol-PP-N-acetylglucosamine from UDP-Nacetylglucosamine and dolicholphosphate w9᎐11x. Therefore, it was possible to detect changes in cell surface expression of mannose and fucose with the use of tunicamycin. However, when we cultured cells with MPA we were unable to detect any differences in lectin binding to CEM-C7a cells. Following mitogen stimulation of HuPBLs, there is a 10-fold increase in the rate on N-glycosylation w12x. The effect on O-glycosylation is less clear, because following mitogen stimulation of murine T cells there is an increase in O-glycosylation of nuclear proteins but a decrease in glycosylation of cytosolic proteins w13x. We have used activated cells in all our experiments and, therefore, should expect them to have an increased rate of glycosylation. In addition, the differences detected following tunicamycin treatment suggest that the glycoproteins are turning over and it is possible to detect differences in protein glycosylation using our methodology.
174
S. Jepson et al. r Transplant Immunology 8 (2000) 169᎐175
An explanation for the effect of MPA on lymphocyte proliferation but not glycosylation may be that intracellular nucleotides are not homogeneously distributed within a cell and that compartmentalisation of intracellular nucleotide pools does occur w14x. In lymphocytes it has been suggested that there are two pools of dTTPs with one of them being fed by de novo synthesis and used preferentially for DNA synthesis. From experiments using MOLT-4 cells, an acute lymphoblastic leukaemia cell line, it was suggested that a discrete compartment of thymine nucleotides exists serving exclusively as precursors for DNA synthesis w15x. This suggests that another pool of dTTP exists which is not available for DNA synthesis. Similar evidence has been found for CTP pools, with one pool being preferentially available for DNA synthesis w16x. The effect of compartmentalisation of guanine ribonucleotides has also been studied w17x. It was shown in S-49 lymphoma cells that reduction of DNA synthesis caused by inhibition of IMPDH is associated with depletion of guanine ribonucleotide pools rather than dGTP pools ᎏ the latter were found not to be available for DNA synthesis. This indicated that a discrete guanine ribonucleotide precursor pool might play a crucial role in limiting substrate supply for DNA replication under conditions of purine starvation. These data showed that the immediate and dramatic inhibition of DNA synthesis and, therefore, proliferation, following MPA treatment of cells occurs before guanine nucleotides other than GMP are exhausted. It is, therefore, possible that following treatment of lymphocytes with MPA, GMP pools utilised for DNA synthesis and cellular proliferation are depleted leading to inhibition of these processes, but that sufficient levels of GDP remain within a separate cellular pool to allow protein glycosylation to continue normally. We have conducted preliminary experiments using HUVECs treated with MPA, which suggest that MPA may be able to selectively interfere with glycosylation in endothelial cells. This stands in contrast to our findings in lymphocytes, however, similar effects on endothelial cells have been published elsewhere w18᎐20x. This effect on the endothelium would contribute to the immunosuppressive effects of MPA because glycosylation of cell᎐cell adhesion molecules is important for their function. If glycosylation in endothelial cells is inhibited, leucocyte adhesion to the graft endothelium would be disrupted and thus immune cell extravasation into the tissue would consequently be reduced. These data, therefore, provide a potential mechanism to explain the decreased graft infiltration following MPA treatment and suggest that the immunosuppressive effects of MPA are not a result of effects directly on lymphocyte glycosylation per se.
Acknowledgements This work was supported by a grant from the Biotechnology and Biological Sciences Research Council ŽBBSRC.. References w1x Eugui EM, Almquist SJ, Muller CD, Allison AC. Lymphocyteselective cytostatic and immunosuppressive effects of mycophenolic acid in vitro- role of deoxyguanosine nucleotide depletion. Scand J Immunol 1991;33:161᎐173. w2x Sievers TM, Rossi SJ, Ghobrial RM et al. Mycophenolate mofetil. Pharmacotherapy 1997;17:1178᎐1197. w3x Allison AC, Eugui EM. Immunosuppressive and other effects of mycophenolic acid and an ester prodrug, mycophenolate mofetil. Immunol Rev 1993;136:5᎐28. w4x Barankiewicz J, Cohen A. Purine nucleotide metabolism in phytohemagglutinin-induced human T lymphocytes. Arch Biochem Biophys 1987;258:167᎐175. w5x Marijnen YMT, de Korte D, Haverkort WA, den Breejen EJS, van Gennip AH, Roos D. Studies on the incorporation of precursors into purine and pyrimidine nucleotides via ‘de novo’ and ‘salvage’ pathways in normal lymphocytes and lymphoblastic cell-line cells. Biochim Biophys Acta 1989;1012:148᎐155. w6x Heeman U, Azuma H, Hamar P, Schmid C, Tilney N, Philipp T. Mycophenolate mofetil inhibits lymphocyte binding and the upregulation of adhesion molecules in acute rejection of rat kidney allografts. Transplant Immunol 1996;4:64᎐67. w7x Allison AC, Kowalski WJ, Muller CD, Waters RV, Eugui EM. Mycophenolic acid and brequinar, inhibitors of purine and pyrimidine synthesis, block the glycosylation of adhesion molecules. Transplant Proc 1993;25Žsuppl 2.:67᎐70. w8x Laurent AF, Dumont S, Poindron P, Muller CD. Mycophenolic acid suppresses protein N-linked glycosylation in human monocytes and their adhesion to endothelial cells and to some substrates. Exp Haematol 1996;24:59᎐67. w9x Speake BK, Hemming FW, White DA. The effects of tunicamycin on protein glycosylation in mammalian and fungal systems. Biochem Soc Trans 1980;8:166᎐168. w10x Schwarz RT, Datema R. Inhibitors of protein glycosylation. Trends Biochem Sci 1980;5:65᎐68. w11x Schwarz RT, Datema R. The lipid pathway of protein glycosylation and its inhibitors: the biological significance of proteinbound carbohydrates. Adv Carbohydrate Chem Biochem 1982;40:287᎐379. w12x Kumar V, Heinemann FS, Ozols J. Interleukin-2 induces N-glycosylation in T cells: characterization of human lymphocyte oligosaccharyltransferase. Biochem Biophys Res Commun 1998;247:524᎐529. w13x Kearse KP, Hart GW. Lymphocyte activation induces rapid changes in nuclear and cytoplasmic glycoproteins. Proc Natl Acad Sci USA 1991;88:1701᎐1705. w14x Moyer JD, Henderson JF. Compartmentation of intracellular nucleotides in mammalian cells. CRC Crit Rev Biochem 1985;19:45᎐61. w15x Taheri MR, Wickremasinghe RG, Hoffbrand AV. Functional compartmentation of DNA precursors in human leukaemoblastoid cell lines. Br J Haematol 1982;52:401᎐409. w16x Reichard P. Regulation of deoxyribotide synthesis. Biochemistry 1987;26:3245᎐3248. w17x Nguyen BT, Sadee W. Compartmentation of guanine nucleotide precursors for DNA synthesis. Biochem J 1986; 234:263᎐269.
S. Jepson et al. r Transplant Immunology 8 (2000) 169᎐175 w18x Hauser IA, Johnson DR, Thevenod F, Goppelt-Strube M. Effect of mycophenolic acid on TNF␣-induced expression of cell adhesion molecules in human venous endothelial cells in vitro. Br J Pharmacol 1997;122:1315᎐1322. w19x Bertalanffy P, Dubsky P, Seebacher G, Griesmacher A, Weigel G, Wolner E. Mycophenolic acid influences the expression of
175
ICAM-1 on human endothelial cells. Adv Exp Med Biol 1998;43:565᎐568. w20x Bertalanffy P, Dubsky P, Wolne E, Weige G. Alterations of endothelial nucleotide levels by mycohenolic acid result in changes of membrane glycosylation and E-selectin expression. Clin Chem Lab Med 1999;37:259.