NEFROLOGÍA. Vol. XXV. Número 2. 2005 ORIGINALS All-trans retinoic acid induces apoptosis in human mesangial cells: involvement of stress active p38 kinase J. C. Sepúlveda*, V. Moreno Manzano*, M. Alique*, P. Reyes*, M. Calvino**, J. Pérez de Hornedo**, T. Parra** and F. J. Lucio* *Departamento de Fisiología, Facultad de Medicina, Universidad de Alcalá. Alcalá de Henares, Madrid. **Unidad de Investigación, Hospital Universitario de Guadalajara. Guadalajara. SUMMARY All-trans retinoic acid (AR-t) is used for treating acute promyelocytic leukemia and renal cell carcinoma and it also has therapeutic value in several animal models of renal disease. Among its renal targets, mesangial cells have been widely studied: they have both retinoic acid receptors (RAR) and retinoid X receptors (RXR) and the cell growth is inhibited when human mesangial cells are incubated with 1-10 µM AR-t. Although his effect has been related with the antiproliferative action of AR-t, there are no studies on the involvement of apoptosis in AR-t induced cell growth when higher concentrations of retinoid are used. Our studies show that 25 µM AR-t triggers mesangial cell apoptosis assessed by light and fluorescence microscopy (Giemsa stain and acridine orange stain, respectively), DNA electrophoresis, flow cytometry (annexin-V) and immunocytochemistry (TUNEL). AR-t induced apoptosis was not inhibited by preincubation with the RXR pan-antagonist HX531 nor with the RAR pan-antagonist AGN 193109, this suggesting RAR and RXIR are not involved in AR-t induced cell death. Previous results of our group showed that ERK (extracellular regulated kinase) and JNK (c-Jun kinase), t wo members of the MAP (mitogen activated protein) kinase family, are involved in non apoptotic effects of AR-t on mesangial cells. Therefore we focussed on the stress activated p38 kinase, the third member of the MAPK family, to investigate its involvement in AR-t induced apoptosis. The results confirmed a role of p38 since: 1) preincubation with SB203589, a p38 inhibitor, inhibited ARA induced apoptosis; 2) incubation with AR-t induced p38 phosphorilation after few minutes and p38 remained phosphorilated for at least 8 hours and 3) AR-t induced p38 phosphorilation was inhibited by SB203589. These data suggest that AR-t migth have toxic side effects on the kidney but also suggest that AR-t could be an useful inhibitor of pathological mesangial cell expansion. Key words: All-trans retinoic acid. Mesangial cell. P38 kinase. Apoptosis. Retinoic acid receptor. Retinoid X receptor. Correspondence: Dr. Francisco Javier de Lucio Cazaña Departamento de Fisiología. Facultad de Medicina Universidad de Alcalá Alcalá de Henares (Madrid) E-mail: javier.lucio@uah.es 131 J. C. SEPÚLVEDA y cols. EL ÁCIDO RETINOICO TODO-TRANS INDUCE APOPTOSIS EN CÉLULAS MESANGIALES HUMANAS: IMPLICACIÓN DE LA QUINASA ACTIVADA POR ESTRES P38 RESUMEN El ácido retinoico todo-trans (AR-t) se utiliza en clínica en el tratamiento de la leucemia promielocítica aguda y el cáncer renal. También presenta efecto terapéutico en diversas formas de enfermedad renal experimental. Las células mesangiales son una de las dianas farmacológicas de AR-t mejor estudiadas: presentan receptores de ácido retinoico (RAR) y receptores X de retinoides (RXR) y el AR-t, a concentraciones entre 1 y 10 µM, inhibe su crecimiento. Este efecto se ha relacionado con la acción antiproliferativa del AR-t, aunque no se ha estudiado la participación de mecanismos apoptóticos cuando se utilizan mayores concentraciones de AR-t. El presente trabajo demuestra que AR-t 25 µM induce apoptosis de células mesangiales humanas en cultivo, caracterizada por estudios de microscopía óptica y de fluorescencia (tinciones de Giemsa y naranja de acridina, respectivamente), electroforesis del ADN fragmentado, citometría de flujo (anexina-V/ioduro de propidio) e inmunocitoquímica (TUNEL). Ni HX531 (pan-antagonista RXR), ni AGN193109 (pan-antagonista RAR) redujeron el grado de muerte celular inducido por el AR-t, lo que sugiere un mecanismo independiente de receptores. Resultados previos de nuestro grupo indican que dos de los tres miembros de las quinasas activadas por mitógenos (MAP), ERK (quinasa regulada por estímulos extracelulares) y JNK (quinasa de c-Jun), están implicados en efectos no apoptóticos del AR-t en células mesangiales. Nos centramos, pues, en el potencial pro-apoptótico del tercer miembro, la quinasa activada por estrés p38. Confirmamos su implicación en la apoptosis inducida por el AR-t porque: 1) su inhibidor farmacológico, SB203580, previno dicha apoptosis 2) El AR-t indujo en pocos minutos la fosforilación de p38, manteniéndose fosforilada durante las 8 horas posteriores; y 3) dicha fosforilación se inhibió por preincubación con SB203580. Estos datos sugieren una posible toxicidad renal del AR-t, pero también su utilidad para controlar la proliferación patológica de células mesangiales. Palabras clave: Ácido retinoico todo-trans. Célula mesangial. Quinasa p38. Apoptosis. Receptores de ácido retinoico. Receptores X de retinoides. INTRODUCTION All-trans retinoic acid (RA-t) has demonstrated its usefulness in the treatment of several forms of experimental renal desease1-3. Glomerular mesangial cells are one of the best studied RA-t pharmacological targets: these cells have , and, retinoic acid receptors (RAR) and , , and X receptors for retinoids (RXR), and treatment with RA-t exerts on these cells the well known suppressor effect on cellular growth of this vitamin A derivative4. Furthermore, RA-t protects mesangial cells from hydrogen peroxide (H2O2)-induced apoptosis, it inhibits mesangial expression of molecules implicated in monocyte adhesion as well as monocyte adhesion to mesangial cells cultures4, and it has demonstrated its efficacy in prevention and treatment of antiThy1.1-induced nephritis3, a condition where me132 sangial damage and proliferation are particularly prominent. Knowing more about the mechanisms implicated in the number of mesangial cells control by RA-t is important for two reasons: 1) because mesangial proliferation and apoptosis are implicated in phenomena of glomerular damage and repair, being such that modulation of these phenomena has an obvious therapeutic interest5; 2) because RA-t is already being used in clinical practice to treat acute promyelocytic leukemia6 and it could directly affect renal function by reducing the number of glomerular cells. In deed, many retinoids exert a negative effect on cellular growth since they inhibit proliferation and/or induce apoptosis7. In the case of mesangial cells, we know that the anti-proliferative effect acts on cellular growth inhibition4 when RA-t concentrations lower than 10 µM APOPTOSIS IN MESANGIAL CELLS are used. However, it is unknown whether apoptotic type mechanisms might participate at higher concentrations. In the present study, we analyze the RA-t action at high concentrations on mesangial cells, confirming that it induces apoptosis. Next, we assess the role of RAR and RXR receptors in this phenomenon. On the other hand, our previous studies indicate that two of the three mitogen-activated kinases (MAPK), ERK (extracellular stimuli-regulated kinase) and JNK (c-Jun kinase), are activated by RA-t concentrations that do not induce death in mesangial cells8,9, and indeed, JNK induction is implicated in the anti-apoptotic effects of RA-t on these cells9. In this sense, it was of interest to establish the implication of the third type of MAPIK, p38 kinase, which is activated by cellular stress and may have a clear role in apoptosis induction by several chemical substances, among which there are synthetic molecules related to retinoids10,11. MATERIAL AND METHODS Experimental design Confluent (4 to 8 passes) and quiescent cells were used by means of incubation for 48 hours in 0.5% fetal calf serum-enriched media. Then, cells were cultivated with RA-t 25-100 µM (10 µM are an innocuous concentration and, in fact, it protects mesangial cells against hydrogen peroxide9 in preliminary experiments that indicated that incubation with RA-t 25 µM for 16 hours is sufficient to cause mesangial cell death, identified by internalization of tripan blue into the cell, which reveals the loss of selective permeability of the cellular membrane). Then, we performed the following studies: (1) Apoptosis identification by presence of pyknotic nuclei, severe chromatin concentration and nuclear fragmentation into spherical structures in Giemsa- or acridine orange-stained cells; by confirmation that nuclear fragments stained by means of TUNEL (in situ labeling of the DNA 3-OH terminals); by electrophoretic demonstration of DNA fragments with an internucleosomic size; and by flow-cytometry of phosphatidyl serine translocation from the inner side to the outer side of plasma membrane (anexin-V). (2) Role of retinoids receptors and of p38 pathway in RA-t-induced apoptosis. In these experiments, before adding RA-t, we pre-incubated with the RXR pan-antagonist (HX531)8 and RAR pan-antagonist (AGN193109)9 in order to identify a possible intervention of retinoids receptors in apoptosis. In the same way, to evaluate stress-activated p38 kinase we pre-incu- bated with its inhibitor SB203580 before adding RA-t. Since apoptosis was prevented in this later step, we proceeded to confirm that RA-t promotes p38 activation through the appropriate kinase assay. Reactants Rochefarma S.A. (Spain) provided RA-t free of charge. If not indicated other way, all chemical reactants were from Sigma Chemical Co. Materials, media and sera were obtained from Gibco. Human glomerular mesangial cells culture These cells were obtained from cortical regions of adult kidney, as previously described4, establishing cellular identity through morphological, immunohystochemical, and functional studies4. The culture medium used was RPMI 1640 enriched with 10% fetal calf serum, L-glutamine (200 mM), sodium penicillin G (100 µg/ml), and streptomycin sulfate (100 µg/ml), buffered with HEPES (N-2 hydroxi-ethyl-piperazine-N-2-ethanolsulphonic acid) 20 mM and sodium bicarbonate (NaHCO3) 24 mM. Apoptosis and cellular viability assays (1) Microscopy. Cells for light microscopy were fixed with 4% formaldehyde in phosphate-buffered saline, and were stained with Giemsa for 1 hour; for fluorescence microscopy, cells were stained with acridine-orange for 10 min (4 µg/ml). (2) In situ detection of apoptosis. After cell fixation for 20 min with 4% formaldehyde in phosphate-buffered saline, TUNEL assay was performed using the in situ detection kit "cell death kit" (Boehringer Mannheim) according to the manufacturer's instructions. (3) Analysis of tripan blue exclusion. Cells were trypsinized and stained with tripan blue, and viable and non-viable cells count was done in a Neubauer chamber. (4) Anexin V/propidium iodide stain. Trypsinized cells were dark-incubated in a buffer made up of HEPES 10 mM, NaCl 50 mM, KCl 5 mM, MgCL2 1 mM, CaCl2 1.8 mM, pH 7.4, which contains anexin-V conjugated with phyco-erithrin or propidium iodide, for 30 min at 37º C. After that time, cells were washed with buffer and flow-cytometry analysis was performed. (5) DNA electrophoresis. Cells were lysed with lysis buffer (EDTA 10 mM, Triton X-100, 0.5% 133 J. C. SEPÚLVEDA y cols. Stress-activated p38 kinase assay A) 100 50 0 RA-t ­ + Cells were lysed with the sample buffer (2% SDS, 5% glycerol, 0.003% bromophenol blue, and 1% -mercaptoethanol in Tris-HCl 125 mM, pH 6.8), boiled for 5 min and passed through 23-G needles for several times. After spinning, the supernatants were placed on 10% polyacrilamide gels and were transferred to nitrocellulose membranes. Analyses of phosphorylated (active) p38 and total p38 (phosphorylated and non-phosphorylated) were done using the PhosphoPlus p38 MAP Kinase antibody kit (New England Biolabs, UK), according to the manufacturer's instructions. Statistical analysis Cellular viability (%) B) CONTROL RA-t All the results are expressed as mean ± standard deviation. Data were compared using the MannWhitney U-test for non-parametric variables. A p value < 0.05 was considered statistically significant. All experiments were done at least 4 times by triplicate. RESULTS As a result of the morphological study of exposure of mesangial cells to RA-t 25 µM, a highly significant number presented the typical apoptosis morphology (see Experimental design) (Fig. 1B). Quantification of dead cells after incubation with RA-t for 48 h is presented in Figure 1 A. The TUNEL technique (Fig. 2 A) confirmed that nuclear morphological changes were associated to DNA fragmentation (Fig. 2 B). Finally, flow-cytometry results demonstrated that RA-t produced an increase in cellular labeling with anexin V (Fig. 2 C). In order to find out the mechanism of action of RA-t, we used antagonists for each retinoids receptor type: the pan-RXR antagonist (HX531) and the panRAR antagonist (AGN193109)8. Our results (Fig. 3) indicate that none of them significantly reduced the degree of RA-t-induced cellular death, which suggest that the later occurs following an RAR and RXR receptors-independent mechanism. One should bear in mind that antagonists per se had some toxicity, which may conceal a possible protective effect against retinoids-induced apoptosis. However, the results reflect the experiments done with 48 h of incubation. Shorter 12 h RA-t incubations produced a milder degree of apoptosis but none toxicity from the antagonists, and in this case, none RA-t-induced apoptosis inhibition was found out either (results not shown). Fig. 1.--Apoptosis induction by RA-t. A) The histogram shows tripan blue incorporation after incubation with RA-t 25 µM/48 h (*p < 0.01 vs control). B) The photographs show the apoptotic morphology both with Giemsa (superior panel) and acridine orange (inferior panel) staining. Tris-HCl 10 mM, pH 7.4). After spinning, the supernatant was incubated with proteinase K (300 µg/ml) for 1 hour, and DNA was precipitated with isopropanol, proceeding to a new centrifugation. The precipitate was dissolved in 10 µl of Tris-EDTA (pH 7.6) that contained RNAase (300 µg/ml). The presence of internucleosomic DNA fragmentation was determined through 1.5% agarose gel electrophoresis. 134 APOPTOSIS IN MESANGIAL CELLS A) CONTROL RA-t C) Propidium iodide (fluorescence relative units) CONTROL Anexin-V-PE (fluorescence relative units) B) C RA-t M RA-t Propidium iodide (fluorescence relative units) Anexin-V-PE (fluorescence relative units) Fig. 2.--Apoptosis induction by RA-t (treatment with Rat- 25 µM/24 h). A) TUNEL staining. B) DNA electrophoresis; C, control, RAt; all-trans retinoic acid; M, pattern of different molecular weight DNAs. C) Flow-cytometry: scatter diagram of phyco-erithrin (PE)-labeled anexin-V and propidium iodide uptake. Dead cells (%) We observed the p38 implication in RA-t-induced apoptosis because 1) its pharmacologic inhibitor, SB203580, prevented apoptosis (Figs. 4 A and B), 2) RA-t induced in few minutes p38 phosphorylation, remaining phosphorylated for the following 8 h (Fig. 4 C) and 3) the afore mentioned phosphorylation was inhibited with SB203580 preincubation (Fig. 4 C). DISCUSSION RA-t is the most important active metabolite of vitamin A since, with the exception of some aspects of vision and reproduction, it reproduces virtually all the biological effects of vitamin A12. However, the apoptosis that it induces in mesangial cells (Figs. 1 and 2) appears at a concentration well above its physiologic plasma levels, which are in the nanomolar range13. Thus, it is not surprising 100 50 0 RA-t AGN193109 HX531 ­ ­ ­ + ­ ­ ­ + ­ + + ­ ­ ­ + + ­ + Fig. 3.--RAR and RXR receptors pan-antagonists (AGN193109 and HX531, respectively) do not inhibit RA-t-induced cellular death. Viable cells quantification (tripan blue exclusion) was done after pre-incubating with antagonists (1 µM) for 30 min and incubating with RA-t (25 µM/48 h; *p < 0.01 vs control). 135 J. C. SEPÚLVEDA y cols. A) SB SB + RA-t Anexin-V-PE Propidium iodide (fluorescence relative units) Propidium iodide Anexin-V-PE B) 100 C) 0 Dead cells (%) RA-t 25 µM 50 0.1 0.3 1 4 8 (h) Phospho-p38 p38 total Phospho-p38 p38 total RA-t 25 µM + SB 203580 0 Control SB RA-t SB + RA-t Fig. 4.--p38 kinase participates in RA-t-induced apoptosis. A) Flow-cytometry: after pre-incubating with a p38 inhibitor (SB203580 25 µM/1 h), cells were incubated with RA-t 25 µM/24 h (SB + RA-t) or with RA-t vehicle (SB). PE: phyco-erithrin-labeled anexin-V. B) Viable cells quantification (tripan blue exclusion). Cells were pre-incubated with SB203580 25 µM/1 h (SB and SB + RA-t) and incubated for 48 h with RA-t 25 µM (AR-t and SB + RA-t) or with its vehicle (Control and SB); *p < 0.01 vs control, SB and SB + RAt; **p < 0.01 vs control, SB and RA-t, C) p 38 activation. Cells pre-incubated as before with SB203580 were incubated with RA-t 25 µM (0-8 h). Phospho-p38: phosphorylated p38 (active); p38 total: phosphorylated and non-phosphorylated p38. that RAR receptors antagonists, which are saturated in the nanomolar range7, do not block the RA-tinduced apoptosis (Fig. 3). RXR receptors are not activated by RA-t but by its isomer: the 9-cys-retinoic-acid7. But when RA-t is at high concentrations (micromolar or higher), a proportion may spontaneously isomerize to 9-cys-retinoic acid14. For that reason, it would be possible that at a 25 µM RA-t concentration mesangial apoptosis would depend on activation of RXR receptors by 9-cysretinoic acid (produced from RA-t isomerization). 136 However, preincubation of mesangial cells with an RXR antagonist din not modify RA-t-induced apoptosis. Altogether, the results obtained suggest that apoptosis occurs through an RAR and RXR receptors-independent pathway. MAPK family includes three varieties, ERK, JNK, and p38, which are activated in response to a great variety of stimuli and they mediate important signaling pathways in the generation of biological responses15, including mesangial responses to RA-t8,9. J. C. SEPÚLVEDA y cols. In the present study, we demonstrated p38 implication in RA-t-induced apoptosis because 1) its pharmacologic inhibitor, SB203580, prevented apoptosis (Figs. 4 A and B); 2) RA-t induced within few minutes p38 phosphorylation, remaining phosphorylated for the following 8 hours (Fig. 4 C); and 3) that phosphorylation was inhibited by preincubation with SB203580 (Fig. 4 C). Participation of p38 in mesangial cells apoptosis has been previously described16 and it is not unusual, either, that there exist RA-t effects independent of its receptors but kinase-dependent: it has been described that RA-t concentrations higher than 10 µM may bind with high affinity to protein kinase C in in vitro systems17 and modulate that kinase activity. More recently, is has also been demonstrated that p38 is activated by RA-t18, which is very interesting since this implies effects such as transcription factors activation19, cytokines production19, or induction of erithroid cells differentation20, among others. Treatment of acute promyelocytic leukemia with ATRA induces very high remission rates --because of induction of promyelocytes differentiation-- but rarely confirmed at a molecular level. Treatments with ATRA encapsulated in liposomes appear to overcome this problem since they achieve stable ATRA levels with a plasma peak around 11 µM21 versus the 1-4 µM routinely used22,23. Although 11 µM are somehow lower than the 25 µM concentration required to induce mesangial apoptosis, in view on the tendency to increase ATRA plasma levels in order to improve its therapeutic effect, it is likely that in the short term sufficiently high ATRA plasma concentrations will be achieved to induce mesangial cells apoptosis. In turn, lower RA-t concentrations may inhibit both apoptosis and mesangial proliferation4. These data suggest that RA-t has the capacity to modulate the number of mesangial cells within the glomerulus. The clinical relevance of the regulatory capacity on mesangial cells numbers is based in two assumptions: on the one hand, mesangial cells hyperplasia is a common feature of many types of glomerulonephritis, so that apoptosis constitutes an important mechanism for resolving glomerular hypercellularity in several forms of human and experimental glomerulonephritis. On the other hand, apoptosis participation in hypocellularity of chronic glomerulonephritis has also been documented in animal models and in the human being24,25. However, it will be necessary to perform further studies in animal models of glomerulonephritis to clarify whether RA-t and other retinoids could be incorporated to the nephrologist therapeutic armamentarium. 138 REFERENCES 1. Moreno-Manzano V, Mampaso F, Sepúlveda-Muñoz JC, Alique M, Chen S, Ziyadeh FN, lglesias-de la Cruz MC, Rodríguez J, Nieto E, Orellana JM, Reyes P, Arribas I, Xu Q, Kitamura M, Lucio Cazana FJ: Retinoids as a potential treatment for experimental puromycin-induced nephrosis. Br J Pharmacol 139: 823-831, 2003. 2. Oseto S, Moriyama T, Kawada N, Nagatoya K, Takeji M, Ando A, Yamamoto T, Imai E, Hori M: Therapeutic effect of all-trans retinoic acid on rats with anti-GBM antibody glomerulonephritis. Kidney Int 64: 1241-1252, 2003. 3. Dechow C, Morath C, Peters J, Lehrke I, Waldherr R, Haxsen V, Ritz E, Wagner J: Effects of all-trans retinoic acid on renin-angiotensin system in rats with experimental nephritis. Am J Physiol Renal Physiol 281: F909-F919, 2001. 4. Moreno-Manzano M Muñoz X, Jiménez JR, Puyol MR, Puyol DR, Kitamura M, Cazana FJ: Human renal mesangial cells are a target for the anti-inflammatory action of 9-cis retinoic acid. Br J Pharmacol 131: 1673-1683, 2000. 5. Ortiz A, Justo P, Catalán MP, Sanz AB, Lorz C, Egido J: Apoptotic cell death in renal injury: the rationale for intervention. Curr Drug Targ 2: 181-192, 2002. 6. Schwartz EL, Hallam S, Gallagher RE, Wiernik PH: Inhibition of all-trans-retinoic acid metabolism by fluconazole in vitro and in patients with acute promyelocytic leukernia. Biochem Pharmacol 50: 923-928, 1995. 7. Thacher SM, Vasudevan J, Chandraratna RAS: Therapeutic applications for ligands of retinoid receptors. Curr Pharm Des 6: 25-58, 2000. 8. Xu Q, Konta T, Furusu A, Nakayama K, Lucio-Cazana J, Fine LG, Kitamura M: Transcriptional induction of mitogen-activated protein kinase phosphatase 1 by retinoids. Selective roles of nuclear receptors and contribution to the antiapoptotic effect. J Biol Chem 277: 41693-41700, 2000. 9. Moreno-Manzano V, Ishikawa Y, Lucio-Cazana J, Kitamura M: Suppression of apoptosis by all-trans-retinoic acid. Dual intervention in the c-Jun N-terminal kinase-AP-1 pathway. J Biol Chem 274: 20251-20258, 1999. 10. Xia Z, Dickens M, Raingeaud J, Davis RJ, Greenberg ME: Opposing effects of ERK and JNK-p38 MAP kinases on apoptosis. Science 270: 1326-1331, 1996. 11. Ono K, Han J: The p38 signal transduction pathway: activation and function. Cell Signal 12: 1-13, 2000. 12. Ross SA, McCaffery PJ, Drager UC, De Luca LM: Retinoids in embryonal development. Physiol Rev 80: 1021-1054, 2000. 13. Philips WEJ, Murray TK, Campbell JS: Serum vitamin A and carotenoids of Canadians. Can Med Assoc J 102: 1085-1086, 1970. 14. Mangelsdorf DJ, Borgmeyer U, Heyman RA, Zhou JY, Ong ES, Oro AE, Kakizuka A, Evans RM: Characterization of three RXR genes that mediate the action of 9-cis retinoic acid. Genes Dev 6: 329-344, 1992. 15. Schaeffer HJ, Weber MJ: Mitogen-activated protein kinases: specific messages from ubiquitous messengers. Mol Cell Biol 19: 2435-2444, 1999. 16. Parameswaran N, Spielman WS, Brooks DP, Nambi P: SB203580 reverses adenomedullin's effect on proliferation and apoptosis in cultured mesangial cells. Eur J Pharmacol 371: 75-82, 1999. 17. Radominska-Pandya A, Chen G, Czernik PJ, Little JM, Samokyszyn VM, Carter CA, Nowak G: Direct interaction of all-trans-retinoic acid with protein kinase C. Implications for PKC signaling and cancer therapy. J Biol Chem 275: 2232422330, 2000. 18. Alsayed Y, Uddin S, Mahmud N, Lekmine F, Kalvakolanu DV, Minucci S, Bokch G, Platanias LC: Activation of Rac1 and the J. C. SEPÚLVEDA y cols. p38 mitogen-activated protein kinase pathway in response to all-trans-retinoic acid. J Biol Chem 276: 4012-9, 2001. 19. Raingeaud J, Whitmarsh AJ, Barrett T. Derijard B, Davis RJ: MKK3- and MKK6- regulated gene expression is mediated by the p38 mitogen-activated protein kinase signal transduction pathway. Mol Cell Biol 16: 1247-1255, 1996. 20. Nagata Y, Takahashi N, Davis RJ, Todokoro: Activation of p38 MAP kinase and JNK but not ERK is required for erythropoietin-induced erythroid differentiation. Blood 15, 92: 18591869, 1998. 21. Estey E, Thall PF, Mehta K, Rosenblum M, Brewer T Jr, Simmons V, Cabanillas F, Kurzrock R, López-Berges G: Alterations in tretinoin pharmacokinetics following administration of liposomal all-trans retinoic acid. Blood 87: 3650-3654, 1996. 22. Muindi JR, Frankel SR, Huselton C, DeGrazia F, Garland Wk Young CW, Warrell RP Jr: Clinical pharmacology of oral alltrans retinoic acid in patients with acute promyelocytic leukemia. Cancer Res 52: 2138-2142, 1992. 23. Takinati K, Tamai H, Morinobu T, Kawamura N, Miyake M, Fujimoto T, Mino M: Pharmacokinetics of all-trans retinoic acid in pediatric patients with leukemic. Jpn J Cancer Res 86: 400-5, 1995. 24. Ortiz A: Proapoptotic regulatory proteins in renal injury. Kidney Int 58:485-487, 2000. 25. Sugiyama H, Kashihara N, Makino H, Yamasaki Y, Ota A: Apoptosis in glomerular sclerosis. Kidney Int 49: 103-111, 1996. 140