Nephrol Dial Transplant (2002) 17: 580–586
Original Article
Sphingosine 1-phosphate stimulates rat mesangial cell proliferation from outside the cells Norio Hanafusa, Yutaka Yatomi1, Koei Yamada, Yuichi Hori, Masaomi Nangaku, Toshihiro Okuda, Toshiro Fujita, Kiyoshi Kurokawa2 and Masafumi Fukagawa3 Division of Nephrology and Endocrinology, University of Tokyo School of Medicine, Tokyo, 1Department of Laboratory Medicine, Yamanashi Medical University, Yamanashi, 2Division of Nephrology, Tokai University School of Medicine, Kanagawa, and 3Division of Nephrology and Dialysis Center, Kobe University School of Medicine, Kobe, Japan
Abstract Background. Proliferation of mesangial cells (MCs) is the initial step in glomerulonephritis, and plateletderived mediators have been shown to play a significant role in this proliferation. Sphingosine 1-phosphate (S1P), one of the sphingolipids, is abundantly stored in platelets and is released upon stimulation. We examined the effects of S1P and related sphingolipids on the cell fate of cultured MCs in order to elucidate potential roles of these lipid mediators in glomerulonephritis. Methods. Cell proliferation was evaluated by bromodeoxy uridine (BrdU) incorporation together with MTS assay. Apoptosis of MCs was evaluated by examining annexin V staining and typical morphological changes in nuclei. We also examined the metabolism of w3Hxsphingosine in MCs in either the presence or absence of platelet-derived growth factor (PDGF). The expression of endothelial differentiation genes (edg), which are the cell surface receptors for S1P in MCs, was examined by RT-PCR. Results. S1P, but not the other sphingolipids, stimulated MC proliferation. In contrast, dimethylsphingosine (DMS) induced apoptosis in the MCs. The amount of sphingosine (Sph) converted into S1P was small and was not affected by PDGF. This observation suggested that Sph kinase activity producing S1P from Sph was low in the MCs. Furthermore, expression of edg-1, -2 and -5 in MCs was confirmed by RT-PCR. Conclusions. Our observations suggest that S1P stimulates MC proliferation from outside the cells, and not as a second messenger for PDGF. The modulation of MC fate with sphingolipids may provide possible strategies for the treatment of glomerulonephritis.
Correspondence and offprint requests to: Masafumi Fukagawa, Associate Professor and Chief, Division of Nephrology and Dialysis Center, Kobe University School of Medicine, 7-5-2 Kusunoki-cho, Chuo-ku, Kobe 650-0017, Japan. Email:
[email protected] #
Keywords: apoptosis; dimethylsphingosine; endothelial differentiation genes; mesangial cell; proliferation; sphingosine 1-phosphate
Introduction Sphingolipids have recently emerged as a class of lipids involved in signal transduction w1,2x. The sphingolipids have been distinguished by their apparent ability to participate in pro- or anti-proliferative pathways w1,2x. For example, it has been shown that ceramide (Cer) and dimethylsphingosine (DMS) induce apoptosis w2x, whereas sphingosine 1-phosphate (S1P) stimulates mitosis w1,3x and works as a second messenger in the proliferation induced by platelet-derived growth factor (PDGF) and by serum w3x. It has also been proposed that the balance between intracellular levels of Cer and S1P, as well as regulatory effects of these sphingolipids on members of mitogen-activated protein kinases, may determine cell fate w4x. Platelets demonstrate a unique metabolism of S1P. These anucleate cells exhibit high sphingosine (Sph) kinase activity, which phosphorylates Sph into S1P; however, they lack lyase activity, which degrades S1P into ethanolamine phosphate and fatty aldehyde w5x. As a result, platelets abundantly accumulate S1P w6x and release it upon stimulation w5x. Furthermore, S1P itself has been shown to stimulate platelets w5x as well as to protect endothelial cells from apoptosis w7x. These observations suggest that S1P acts as an autocrine or paracrine mediator released from activated platelets. Proliferation of mesangial cells (MCs) is an important and initial step in the pathogenesis of glomerulonephritis. This early step is usually followed by the accumulation of extracellular matrix within the glomeruli w8x. Platelets have been shown to play a significant role in mediating MC proliferation in vivo. For example, platelets were found together with neutrophils in damaged or inflamed glomeruli w9x.
2002 European Renal Association–European Dialysis and Transplant Association
Sphingolipids on mesangial cell fate
In anti-Thy-1 nephritis, an experimental model of mesangioproliferative glomerulonephritis, platelet depletion by anti-platelet antibody decreased the extent of MC proliferation w10x. These actions have been mainly attributed to factors released by platelets, such as PDGF. To elucidate the role of S1P in MC proliferation, we investigated the effects of S1P and related sphingolipids on the fate of MCs. We found that S1P, but not other Sph derivatives, stimulates MC proliferation, whereas DMS induced MC apoptosis. These findings suggest that S1P stimulates proliferation of MCs from outside the cells.
Materials and methods
581
absorption at 450 nm was measured to detect BrdU using DIGISCAN (ASYS Hitech GmbH, Eugendorf, Austria), according to the supplied protocols. The results were expressed in relative data. The mean values of control wells without mitogen were defined as zero and wells containing 17% FCS were defined as 100. CellTiter 96 AQueous One Solution Cell Proliferation Assay (Promega, Madison, WI, USA) was employed for MTS assays. We followed the manufacturer’s protocol. During the final 15 min of incubation with various stimulants, 20 ml of MTS solution was added to each well. At the end of the incubation period, optic absorbance at 550 nm was measured. Various numbers of MCs, counted using a counting chamber, were seeded and subjected to MTS assays on the calibration plate. The results were expressed as relative values. The mean number of MCs in the wells without mitogen was defined as 100. Both BrdU assays and MTS assays were performed with an n value of 8 and were duplicated.
Preparation of rat MCs Primary culture of rat MCs was prepared by a standard sieving method w11x. The cells were maintained in Dulbecco’s modified Eagle’s Medium (DMEM) (Nissui Seiyaku, Tokyo, Japan) containing 17% foetal bovine serum (JRH Biosciences, Lenexa, KS, USA) at 378C under a humidified atmosphere of 5% CO2u95% air. In order to avoid contamination of epithelial cells or endothelial cells, we used cells between the 5th and 15th passage number. The cells were confirmed as MCs by positive staining for vimentin, alfa-smooth muscle actin and Thy-1 antigen (data not shown).
MC proliferation MC proliferation was examined by two methods. The first method measured the amount of bromodeoxy uridine (BrdU) incorporated into chromosomal DNA during synthesis. The second method used a colorimetric assay of MTS, one of the tetrazolium compounds (3-(4,5-dimethylthiazol-2-yl)5-(3-carboxymethoxyphenyl)-2-(4-sulphophenyl)-2H-tetrazolium) that changes colour due to the intracellular reductive products in proportion to the number of cells. MCs were seeded in 96-well plates (Becton Dickinson, Franklin Lakes, NJ, USA). When cells reached ;60% confluency, the medium was changed to DMEM without serum. After a 48 h serum starvation period, specific concentrations of S1P or 10 mM of other sphingolipids were added to the wells and the cells were incubated for an additional 24 h. S1P and DMS were obtained from BIOMOL (Plymouth Meeting, PA, USA). Sph and Cer (C2-Cer, cell membrane permeable type) were purchased from Sigma (St Louis, MO, USA). S1P was resolved in 50% (vuv) ethanol (EtOH) in phosphatebuffered saline (PBS). The other sphingolipids were resolved in 100% EtOH. Concentrations of EtOH in the media, including samples without sphingolipids, were adjusted to 0.5%, a concentration that did not have effects on MC proliferation or apoptosis. For the BrdU incorporation assay, performed 6 h prior to the assays, 10 mM BrdU was added to the medium. The amount of incorporated BrdU was then determined by Cell Proliferation ELISA system, version 2 (Amersham LIFE SCIENCE, Amersham Place, Buckinghamshire, UK). This non-RI system detects the amount of BrdU within chromosomal DNA by peroxidase-labelled anti-BrdU antibody-coloured 3,39,5,59-tetramethylbenzidine. The optic
Detection of apoptosis MCs were cultured on 6 cm plates (Corning, Corining, NY, USA) with up to 60% confluency. After 48 h of serum starvation, the cells were challenged with 10 mM of S1P or DMS for 24 h. The cells were then trypsinized (TrypsinEDTA, Life Technologies, Rockville, MD, USA) and scraped from the plates. After centrifugation, the cells were stained with 1 mM bisbenzimide hydrochloride (Hoechst 33258) (Sigma, St Louis, MO, USA) in PBS for 2 min. Morphological changes to the nuclei were studied under a fluorescent microscope with FITCuDAPIuTex Red filter (Olympus, Tokyo, Japan), and photographs were taken using Kodak Gold 100 film (Eastman Kodak, Rochester, NY, USA). In addition to the morphological studies, the percentages of apoptotic cells were determined by flow cytometry. Cells were stained with both fluorescein (FITC)-labelled annexin V and propidium iodide (PI). The cultured cells were submitted to the same procedures described in the morphological study. They underwent apoptosis with each of the 10 mm sphingolipids on the 6 cm plates. Cells were removed from the plates, trypsinized and subjected to the assays described below. For annexin V binding assays, we used Annexin V-FITC Apoptosis Detection Kit (Medical & Biological Laboratories, Nagoya, Japan). Cells were incubated at room temperature in binding buffer for 10 min with annexin V-FITC and PI. The cells were then analysed by FACScan and LYSYS II software for analysis (both from Becton Dickinson, Franklin Lakes, NJ, USA). The fraction of cells that were annexin positive and PI negative were defined as apoptotic.
Metabolism of [3H]Sph in MCs MCs, cultured in 12-well plates (Becton Dickinson) to 90% confluency, were incubated with 1 mM Sph containing w3HxSph (0.1 mCi) (NEN Life Science Products, Boston, MA, USA) in the presence or absence of rat recombinant PDGF-BB (5 nguml) (Sigma, St Louis, MO, USA). At the indicated time points, the reaction was terminated by removing the medium and by adding ice-cold methanol to the cells. Lipids were extracted separately from the cells and medium by the method of Bligh and Dyer w12x, and were then analysed using tritium-labelled sphingolipids as described
582
previously w5x. Portions of lipids obtained from the chloroform phase were applied to silica gel high-performance thin layer chromatography (TLC) plates (Merck, Darmstadt, Germany), and the plates were then developed in butanolu acetic aciduwater (3 : 1 : 1). The TLC plates were scanned with a BAS-2000 (Fujifilm, Tokyo, Japan).
N. Hanafusa et al.
and 59-ACCACACGTCGGTTGCTCATTC-39 (624–645 bp) (antisense) for edg-2 w14x; and 59-CATGTACCTGTTCCTCGGCAAC-39 (201–222 bp) (sense) and 59-GCGAAGGCAAAGAAATAATGGG-39 (809–830 bp) (antisense) for h218uagr16, the rat homologue of edg-5 w15x.
Statistical analyses Detection of edg expression The expression of the edg family of receptors was detected by RT-PCR. Total RNA was extracted by a standard acid guanidinium thiocyanate-phenol-chloroform (AGPC) method from 90% confluent MCs cultured on 10-cm plates (Becton Dickinson). Complementary DNA was synthesized from 2 mg of total RNA with SuperScript II (Life Technologies, Rockville, MD, USA) according to the manufacturer’s instructions. PCR reactions were performed with synthetic primers (Nippon Flour Mills, Tokyo, Japan), ExTaq and the appropriate buffer (TaKaRa, Tokyo, Japan). Amplification was conducted during 30 cycles of 30 s at 948C, 30 s at 548C, and 30 s at 728C. We also performed PCR on nucleic acid samples without reverse transcription in order to exclude the possibility of contamination by chromosomal DNA. The following oligonucleotide primer pairs were used: 59-CCTCCTTGCTATCGCCATTGAG-39 (667–688 bp) (sense) and 59-TGAGTTCAGCACAGCCAGAACC-39 (1156–1177 bp) (antisense) for edg-1 w13x; 59ACGAGTTGCTTCTTGTGCCACC-39 (84–105 bp) (sense)
Stat View Version 5.0, (SAS Institute, Cary, NC, USA) was used for statistical analysis. BonferroniuDunn tests were used for comparisons between three or more groups. P values of -0.05 were defined as statistically significant. The data were expressed as means"SD (n s 8 and duplicated for BrdU incorporation and MTS assays; n s 3 for annexin V binding assay).
Results DNA synthesis in MCs stimulated by S1P The incorporation of BrdU into chromosomal DNA revealed that S1P stimulates DNA synthesis in MCs (Figure 1A). A mitogenic effect was observed at 0.1–3 mM. Regression analysis confirmed a dose relationship at doses between 0.1 and 1 mM (r s0.494, P s 0.0004). For unknown reasons, the
Fig. 1. Stimulation of MC proliferation by S1P. (A) BrdU incorporation into MCs treated with the indicated concentrations of S1P was measured as described in Materials and methods. BrdU incorporation was significantly increased by S1P at concentrations between 0.1 and 3 mM compared with control cells. *P-0.05. Dose dependency was also observed between 0.1 and 1 mM (r s 0.494, P-0.05). (B) The number of MCs was also increased by S1P. The maximum effect was observed at 3 mM. At that concentration, the increase in cell number was statistically significant (P-0.05). (C) DMS, another sphingolipid, decreased the number of MCs and totally abolished the reaction of MTS at 10 mM (#). Sph at 10 mM also decreased MC numbers compared with control (P-0.05). Open bars represent 1 mM, shaded bars are for 3 mM, and filled bars are for 10 mM of sphingolipids.
Sphingolipids on mesangial cell fate
stimulatory effect of S1P became less evident at 10 mM. The MTS assays also revealed that 3 mM of S1P significantly increased the number of MCs (Figure 1B). In contrast, DMS at 10 mM decreased MC numbers and totally abolished MC viability (Figure 1C). We additionally counted actual numbers of MCs per 1 mm2 on 96-well plates using inverted microscopy (Olympus, Tokyo, Japan) and found that MC numbers were increased to values similar to the values in the MTS assay (data not shown).
Effects of sphingolipids on MC apoptosis Sphingolipid mediators, such as Cer, Sph and DMS, have been shown to induce apoptosis in other cell types w2x. In the present study, nuclei from either serumdeprived or S1P-added cells stained with Hoechst 33258 showed normal appearances (Figure 2A and B). However, a significant number of cells treated with DMS manifested characteristic appearances of apoptosis, including nuclear and cytoplasmic condensation and fragmentation (Figure 2C).
583
Quantitative analyses of the proportion of apoptotic cells measured by annexin V binding revealed that DMS significantly increased the number of apoptotic MCs (Figure 3A and C). Conversely, MCs challenged with S1P did not show a significant difference compared with the vehicle-treated group (Figure 3A). These results indicated that DMS, unlike S1P, induces apoptosis in MCs. An extracellular action of S1P in MC proliferation It has been reported that Sph kinase activity is elevated and that S1P levels are increased by PDGF in Swiss 3T3 fibroblasts w3x. We examined the amount of S1P converted from Sph in order to determine the activity of Sph kinase in MCs. In these cells, incorporated w3HxSph was mainly converted to w3HxCer and then further to w3Hxsphingomyelin in a time-dependent manner, whereas only a small amount of w3HxS1P was transiently produced from w3HxSph (Figure 4A). Moreover, the conversion of Sph into S1P was not affected by PDGF (Figure 4B). This observation argues against a role of S1P as a second messenger for PDGF in MCs.
Fig. 2. Detection of apoptosis in MCs treated with various sphingolipids. MCs were challenged with 10 mM S1P (A), serum starvation only (B), or with 10 mM DMS (C). Morphological changes of MCs were detected using Hoechst 33258 staining and examined under a fluorescence microscope. DMS markedly induced apoptotic morphological changes in nuclei, such as apoptotic bodies (C, arrowheads). These changes were also induced by serum starvation in a limited number of cells (not shown).
584
Fig. 3. MC apoptosis induced by various sphingolipids determined by annexin V binding to the cells. (A) Percentage of apoptotic MCs determined by annexin V binding. Annexin V-positive and PI-negative cells were detected by flow cytometry. Ten micromoles of DMS significantly increased the population of apoptotic MCs compared with control. *P-0.05. FITC-labelled annexin V (horizontal axis) and PI (vertical axis) fluorescence profiles were plotted on a logarithmic scale (B and C). The results of cytograms show the population of apoptotic cells (lower right quadrant) induced by either serum starvation (B) or DMS (C).
w3HxS1P was not detected in the medium and this did not change with PDGF stimulation. This finding indicates that MCs do not release S1P into the extracellular environment. In order to examine the possibility that S1P acts from outside the MCs, we measured the expression of S1P receptors, edgs w16x by RT-PCR. Both edg-1 and -5 were identified as receptors for S1P, and edg-2 was identified for both S1P and lysophosphatidic acid (LPA). RNA from MCs was prepared for reverse transcription, followed by PCR amplification of specific transcripts. As shown in Figure 5, MCs expressed mRNA for edg-1, -2 and -5.
Discussion In agreement with mitogenic properties of S1P in several cell lines w1,3,4,7x, S1P stimulated MC proliferation and activated DNA synthesis. Similar findings
N. Hanafusa et al.
Fig. 4. Metabolism of w3HxSph in MCs. MCs incubated with w3HxSph (0.1 mCi) were challenged with (B) or without (A) 5 nguml PDGF for various durations. Lipids were then extracted from the cells or the media and analysed for tritium-labelled sphingolipids by TLC autoradiography. Locations of standard lipids are indicated on the right. SM, sphingomyelin; Ori, origin.
were reported in a recent short communication w17x. Although S1P has been proposed as an intracellular second messenger in certain cell types w1,3,4x, this possibility is unlikely in MCs. This is because Sph, the precursor of S1P, was not mitogenic for MCs. In addition, the level of S1P converted from Sph was not affected by PDGF even though Sph is mitogenic only when it is converted into S1P intracellularly. It is more likely that edgs, cell surface receptors for lysophospholipids including S1P w18x, mediate the extracellular actions of S1P. The observation that S1P receptors were expressed in MCs supports an extracellular action of S1P on MCs. Recently, it has been shown that intracellular Ca2q signalling was induced by S1P in MCs w18x. This novel sphingolipid mediator is probably an important agonist for MCs. Furthermore, modulation of MCs by sphingolipids may have pathophysiological importance since S1P is released from platelets w5x, and platelets are found in damaged or inflamed glomeruli
Sphingolipids on mesangial cell fate
585
may be associated with MC activation and lead to mesangial matrix expansion w8x and glomerulosclerosis. Alternatively, apoptosis of MCs may be the major mechanism for resolution of glomerular hypercellularity in mesangioproliferative glomerulonephritis w20x. Accordingly, mechanisms that initiate, maintain and limit the proliferative response of MCs in glomerular diseases may provide useful insights into future therapeutic strategies for these diseases. In this context, modulation of MC fate by sphingolipids may prove to have important implications. Treatment of glomerulonephritis is currently limited to supportive therapy with or without non-specific immunosuppressive drugs. Further knowledge about the involvement of platelets or related sphingolipids in MC fate may identify new possibilities for treatment of kidney diseases in the future w20x.
Fig. 5. Detection of mRNA expression by RT-PCR for S1P receptors in the MCs. Amplified products for edg-1, -2 and -5 were electrophoresed in 2% agarose gel. The size of each product was consistent with the predicted values from the published sequences: 511, 562 and 630 bp, respectively.
w9x. S1P may be supplied as a survival factor for MCs from activated platelets in vivo. Supporting this, results of w3HxSph metabolism indicated that S1P was not released from PDGF-activated MCs. Of the various sphingolipids examined, DMS most potently induced apoptosis in the MCs. Although Cer had been reported to induce apoptosis in MCs w19x, its effect in the present study was much smaller than DMS and not significant. DMS is a potent inhibitor of Sph kinase. However, the activity of Sph kinase appeared to be low in MCs because only a limited amount of S1P was produced from Sph. Thus, the action of DMS to inhibit Sph kinase may not be responsible for the apoptotic effect of DMS. Although it has been suggested that DMS induces apoptosis by inhibiting protein kinase C, recent reports have revealed that DMS induces apoptosis through complex mechanisms involving more than protein kinase C inhibition w2x. Further studies will elucidate the mechanisms of DMS-induced apoptosis in MCs. Even though DMS exerts a variety of cellular actions (including induction of apoptosis) and Sph N-methyltransferase, the enzyme that synthesizes DMS, exhibits activity in certain tissues or cells w2x, DMS was formed in neither resting nor PDGFstimulated w3HxSph-labelled MCs (Figure 4), nor in the media. Thus, the physiological and pathophysiological implications of apoptosis induced by DMS remain to be solved. Nevertheless, the effects of DMS may be therapeutically important, as discussed below. Various glomerular diseases, including glomerulonephritis, are characterized by MC proliferation. This
Acknowledgements. The authors are grateful to Dr Libo Yang for her technical assistance. This work was supported in part by a Program Project Grant from the Ministry of Health (to M.F.), a Special Grant for Medical Research from Ministry of Post and Telecommunications (to M.F.), a grant from Ministry of Science and Education (to Y.Y.), grants in Aid for Scientific Research from the Ministry of Education, Science and Culture (Nos 11671030 and 13671100 to M.N.), and a grant from the Ministry of Health, Labour and Welfare (No. H13-21st century(Seikatu)-17 to M.N.).
References 1. Spiegel S and Merrill AH Jr. Sphingolipid metabolism and cell growth regulation. FASEB J 1996; 10: 1388–1397 2. Igarashi Y. Functional roles of sphingosine, sphingosine 1-phosphate, and methylsphingosines: in regard to membrane sphingolipid signaling pathways. J Biochem 1997; 122: 1080–1087 3. Olivera A, Spiegel S. Sphingosine-1-phosphate as second messenger in cell proliferation induced by PDGF and FCS mitogens. Nature 1993; 365: 557–560 4. Cuvillier O, Pirianov G, Kleuser B et al. Suppression of ceramide-mediated programmed cell death by sphingosine1-phosphate. Nature 1996; 381: 800–803 5. Yatomi Y, Yamamura S, Ruan F, Igarashi Y. Sphingosine 1-phosphate induces platelet activation through an extracellular action and shares a platelet surface receptor with lysophosphatidic acid. J Biol Chem 1997; 272: 5291–5297 6. Yatomi Y, Igarashi Y, Yang L et al. Sphingosine 1-phosphate, a bioactive sphingolipid abundantly stored in platelets, is a normal constituent of human plasma and serum. J Biochem 1997; 121: 969–973 7. Hisano N, Yatomi Y, Satoh K et al. Induction and suppression of endothelial cell apoptosis by sphingolipids: a possible in vitro model for cell–cell interactions between platelets and endothelial cells. Blood 1999; 93: 4293–4299 8. Couser WG. Pathogenesis of glomerulonephritis. Kidney Int Suppl 1993; 42: S19–S26 9. Johnson RJ, Alpers CE, Pritzl P et al. Platelets mediate neutrophil-dependent immune complex nephritis in the rat. J Clin Invest 1988; 82: 1225–1235 10. Johnson RJ, Garcia RL, Pritzl P, Alpers CE. Platelets mediate glomerular cell proliferation in immune complex nephritis induced by anti-mesangial cell antibodies in the rat. Am J Pathol 1990; 136: 369–374 11. Okuda T, Yamashita N, Kurokawa K. Angiotensin II and vasopressin stimulate calcium-activated chloride conductance in rat mesangial cells. J Clin Invest 1986; 78: 1443–1448 12. Bligh EG, Dyer WJ. A rapid method of total lipid extraction and purification. Can J Biochem Physiol 1959; 37: 911–917
586 13. Lado DC, Browe CS, Gaskin AA, Borden JM, MacLennan AJ. Cloning of the rat edg-1 immediate-early gene: expression pattern suggests diverse functions. Gene 1994; 149: 331–336 14. Allard J, Barron S, Diaz J et al. A rat G protein-coupled receptor selectively expressed in myelin-forming cells. Eur J Neurosci 1998; 10: 1045–1053 15. Okazaki H, Ishizaka N, Sakurai T et al. Molecular cloning of a novel putative G protein-coupled receptor expressed in the cardiovascular system. Biochem Biophys Res Commun 1993; 190: 1104–1109 16. Goetzl EJ, An S. Diversity of cellular receptors and functions for the lysophospholipid growth factors lysophosphatidic acid and sphingosine 1-phosphate. FASEB J 1998; 12: 1589–1598
N. Hanafusa et al. 17. Gennero I, Simon MF, Gaits F et al. Effect of sphingosine1-phosphate and analogues of lysophosphatidic acid on mesangial cell proliferation. Ann NY Acad Sci 2000; 905: 340–343 18. Chen PF, Chin TY, Chueh SH. Ca2q signaling induced by sphingosylphosphorylcholine and sphingosine 1-phosphate via distinct mechanisms in rat glomerular mesangial cells. Kidney Int 1998; 54: 1470–1483 19. Baker AJ, Mooney A, Hughes J, Lombardi D, Johnson RJ, Savill J. Mesangial cell apoptosis: the major mechanism for resolution of glomerular hypercellularity in experimental mesangial proliferative nephritis. J Clin Invest 1994; 94: 2105–2116 20. Shayman JA. Sphingolipids. Kidney Int 2000; 58: 11–26 Received for publication: 20.2.01 Accepted in revised form: 9.10.01
