- Email: [email protected]
Protection of oligodendrocyte precursor cells by low doses of HSP90 inhibitors in cell culture
Experimental Neurology 225 (2010) 29–33
Contents lists available at ScienceDirect
Experimental Neurology j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / y e x n r
Short Communication
Protection of oligodendrocyte precursor cells by low doses of HSP90 inhibitors in cell culture Cristina Cid a, Alberto Alcazar b,⁎ a b
Center for Astrobiology, CSIC-INTA, Torrejon de Ardoz, Spain Department of Investigation, Hospital Ramón y Cajal, Madrid, Spain
a r t i c l e
i n f o
Article history: Received 30 July 2009 Revised 4 November 2009 Accepted 21 November 2009 Available online 4 December 2009 Keywords: 17-allylamino-17-demethoxygeldanamycin Heat shock protein 90 HSP90 antibodies HSP90 beta HSP90 inhibitors O-2A cells Cell death Oligodendrocyte precursor cells Radicicol
a b s t r a c t Oligodendrocyte precursor cells (OPCs) become myelin-forming after their differentiation into post-mitotic oligodendrocytes. OPCs are extremely efficient at myelin repair and contribute to remyelination. However, remyelination fails in multiple sclerosis (MS), which suggest that the OPCs are ineffective in this disorder. We have studied previously the expression of heat shock protein 90 (HSP90) in OPCs and have reported autoantibodies against HSP90 in MS patients, which recognize the antigen on the OPC surface. The present study investigated a protective effect of HSP90 inhibitors observed in cultured OPCs. Radicicol and 17allylamino-17-demethoxygeldanamycin (17-AAG) at non-cytotoxic doses targeted cell-surface HSP90 in OPCs. Thus, 0.01 nM 17-AAG or 10 nM radicicol competed with the anti-HSP90 antibodies for binding to cellsurface HSP90. These low doses of HSP90 inhibitors prevented HSP90-antibody-induced OPC death and protected the oligodendrocyte population against antibody attack. Adult oligodendrocytes were protected by these low doses of HSP90 inhibitors in a similar fashion to perinatal cells. The present results show that, despite OPCs being very sensitive to HSP90 inhibitors, low and non-cytotoxic doses of 17-AAG and radicicol protect oligodendrocytes from anti-HSP90 antibody attack. They may have therapeutic potential for MS patients that have anti-HSP90 autoantibodies and provide a novel strategy for therapeutic intervention with HSP90 inhibitors. © 2009 Elsevier Inc. All rights reserved.
Oligodendrocyte precursor cells (OPCs) retain characteristics of multipotent central nervous system (CNS) stem cells (Kondo and Raff, 2000). They are also known as O-2A progenitor cells, are able to proliferate, and most differentiate into post-mitotic oligodendrocytes (Temple and Raff, 1986). These cells have been identified in rodent cell cultures and in the adult rodent and human CNS (Scolding et al., 1999; Roy et al., 1999; Raff et al., 1983; Shi et al., 1998). In humans, these cells are similar immunophenotypically to their rodent counterparts (Scolding et al., 1999; Armstrong et al., 1992). OPCs are extremely efficient at myelin repair and contribute to remyelination, which assigns them a role in myelination (Scolding et al., 1999; Blakemore and Keirstead, 1999). However, failure to remyelinate is a pathological characteristic of the human demyelinating disease multiple sclerosis (MS), which suggests that these cells are ineffective in this disorder (Franklin, 2002). The reason for this remains unclear (Franklin, 2002), but it is known that oligodendrocytes that survive in an area of demyelination do not contribute to remyelination, which is undertaken by the OPCs (Blakemore and Keirstead, 1999). We have
Abbreviations: 17-AAG, 17-allylamino-17-demethoxygeldanamycin; HSP90, heat shock protein 90; OPCs, oligodendrocyte precursor cells. ⁎ Corresponding author. Department of Investigation, Hospital Ramón y Cajal, Ctra. Colmenar km 9.1, 28034 Madrid, Spain. Fax: +34 91 336 9016. E-mail address: [email protected] (A. Alcazar). 0014-4886/$ – see front matter © 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.expneurol.2009.11.017
studied previously the expression of heat shock protein 90 (HSP90) in OPCs. HSP90 is a conserved molecular chaperone (rat HSP90β only differs in 10 amino acids out of 723 from the human form of the protein) and is part of a multichaperone complex whose association is required for the stability and function of multiple signaling proteins that promote cell growth, differentiation and apoptosis. We have found that HSP90β is expressed at the surface of OPCs, and it is not expressed when OPCs differentiate into pre-oligodendrocytes (Cid et al., 2004, 2005). These results, in addition to those that describe expression of HSP90 at the surface of cerebellar and Schwann cells (Sidera et al., 2004), suggest that cell-surface HSP90 has a role in differentiation, cell migration and development of certain types of cells in the nervous system. In addition, we have reported autoantibodies against HSP90β in MS patients that recognize the antigen on the OPC surface (Cid et al., 2004, 2007b). Rabbit anti-HSP90 antibodies against the N-terminal region of HSP90β recognize this protein at the cell surface of OPCs in a similar fashion to anti-HSP90 autoantibodies from MS patients (Cid et al., 2004, 2005) and indicate the extracellular localization of this region. Human and rabbit antiHsp90β antibodies attack OPCs and decrease the oligodendrocyte population in cell cultures (Cid et al., 2005). Development of new strategies to protect the new oligodendrocyte population is important for promoting remyelination in MS. In this regard, we have studied the effect of HSP90 inhibitors geldanamycin, its derivate 17-
30
C. Cid, A. Alcazar / Experimental Neurology 225 (2010) 29–33
allylamino-17-demethoxygeldanamycin (17-AAG) and the macrolactone radicicol on cultured OPCs. These HSP90 inhibitors interact with the ATP-binding site located at the N-terminal region of HSP90 (Chaudhury et al., 2006). We have found that OPCs are extremely sensitive to HSP90 inhibitors compared to other CNS cells (Alcázar and Cid, 2009). Thus, the IC50 of geldanamycin, 17-AAG and radicicol for OPCs was 7.1, 10.7 and 137 nM, respectively, compared to 1–2 μM for pre-oligodendrocytes, astrocytes or neurons. However, ≤1 nM geldanamycin or 17-AAG and ≤10 nM radicicol were non-cytotoxic for OPCs (Alcázar and Cid, 2009). The present work investigated the use of non-cytotoxic doses of HSP90 inhibitors as a new potential therapeutic strategy. We studied the possible binding of HSP90 inhibitors to cell-surface HSP90 in OPCs, and a novel effect on oligodendrocyte lineage cells, cultured OPCs and pre-oligodendrocytes, and found a protective effect at low doses. Primary cell cultures were prepared from cerebral hemispheres of 17- to 19-day-old Sprague–Dawley rat embryos, as described previously (Cid et al., 2004, 2005). All animal procedures were in accordance with The Declaration of Helsinki and Spanish legislation for the use of animals in biomedical experimentation. OPCs were identified phenotypically by their characteristic bipolar morphology and A2B5+ immunoreactivity labeling with mouse monoclonal A2B5 IgM (Millipore) (Raff et al., 1983). The oligodendrocytes were identified by their multipolar morphology and O4+ immunoreactivity labeling with mouse monoclonal O4 IgM (Millipore). O4 is a marker of pre-oligodendrocytes, which are cells that have differentiated into the oligodendrocyte lineage (Armstrong et al., 1992). In 6day-old cell cultures, A2B5+ OPCs comprised 12 ± 1% of the cells, and very few O4+ pre-oligodendrocytes were present (≤1 ± 0.2%); from days 6 to 12 of culture, a gradual reduction in OPCs was observed, while the number of pre-oligodendrocytes correspondingly increased to 10–12% of all cells (Cid et al., 2004). The number of GFAP+ astrocytes and βIII-tubulin+ neurons remained constant over time and were negative for A2B5 (Cid et al., 2004; Alcázar and Cid, 2009). Competition assays between HSP90 inhibitors at non-toxic concentrations and anti-HSP90β antibodies were performed in live cells. Geldanamycin shares biological activities with 17-AAG and has an effect on OPCs at similar doses to 17-AAG (Alcázar and Cid, 2009); therefore, these experiments were done only with 17-AAG and the chemically unrelated HSP90 inhibitor radicicol. Cells cultured for 6 days were untreated (control) or treated with 0.001–1.0 nM 17-AAG or with 0.1–10 nM radicicol (both from Sigma) for 1 h. Live cells were then labeled with rabbit polyclonal anti-HSP90β (HSP84) antibodies (Millipore) to target cell-surface HSP90 as described previously (Cid
et al., 2004, 2005). Living cells were incubated with anti-HSP90β antibodies (1:10 dilution) for 30 min at 25°C and labeled with lissamine rhodamine-conjugated goat anti-rabbit IgG (Millipore) (1:100 dilution) for 1 h at 4 °C. Cells were fixed and exposed to A2B5 antibody to label OPCs (Figs. 1A–D). Radicicol at 10 and 0.01 nM 17-AAG decreased significantly the labeling with anti-HSP90β antibodies on the surface of OPCs (Figs. 1A–D). This decrease in labeling with anti-HSP90β antibodies was quantified by measuring their fluorescence intensity in labeled OPCs. Thus, the fluorescence intensity of anti-HSP90β antibodies was 230±14 in control cells compared to 26±8 in 10 nM radicicol-treated cells (p = 0.0003) (Fig. 1C). Concentrations of 0.01 nM and even 0.001 nM 17-AAG also decreased the fluorescence intensity of the label for HSP90β in OPCs (Fig. 1D). The results demonstrated that these HSP90 inhibitors interacted with cell-surface HSP90β and competed with the antiHSP90β antibodies in OPCs. Competitive assays were performed with western blotting using samples of cell membrane fractions, which were obtained from 6-day-old cultured cells as described previously (Cid et al., 2004, 2007a). Membrane fraction samples were subjected to SDS–PAGE with 5–12% acrylamide (2.6% cross-linker) discontinuous gels and blotted in polyvinylidene-fluoride membranes. Samples for western blotting were cut into strips and incubated individually with different concentrations of HSP90 inhibitors for 1 h. They were exposed to anti-HSP90β antibodies for 1 h and incubated with peroxidase-conjugated goat anti-rabbit IgG (Bio-Rad) and developed with ECL reagent (GE Healthcare) (Fig. 1E). The HSP90 inhibitors decreased the anti-HSP90β antibody reaction in immunoblots in a dose-dependent manner (Fig. 1E), which demonstrated that these inhibitors competed with anti-HSP90β antibodies for binding to HSP90. We have reported previously that anti-HSP90β antibodies induce OPC death by a mechanism mediated by complement activation, which leads to a decrease in the pre-oligodendrocyte population (Cid et al., 2005). Besides, as describe above, HSP90 inhibitors at noncytotoxic doses inhibited the recognition of cell-surface HSP90 by anti-HSP90β antibodies in OPCs. We studied the possible protective effect of the inhibitors against anti-HSP90β antibody attack in preoligodendrocytes. Six-day cell cultures were untreated or treated with HSP90 inhibitors and exposed to anti-HSP90β antibodies (2.5 μg/ml; a concentration in the range determined in CSF from MS patients; Cid et al., 2005) and rabbit complement (Calbiochem). Cells were washed and placed in fresh medium for another 6 days. These 12-day cultured cells were fixed and exposed to O4 antibody to label pre-oligodendrocytes (Figs. 1F−I). Additional experiments were performed
Fig. 1. (A–D) HSP90 inhibitors interacted with cell-surface HSP90 in OPCs and competed with anti-HSP90 antibodies. Cells untreated (control) or treated with 17-AAG or radicicol were incubated with anti-HSP90β antibodies and the secondary antibody. Cells were then fixed with 3.7% paraformaldehyde for 10 min and incubated for 1 h at room temperature or overnight at 4 °C with A2B5 antibody (1:100 dilution) to label OPCs. Cells were washed and incubated with fluorescein-conjugated anti-mouse IgM (1:40 dilution) for 1 h at room temperature, mounted with anti-fade solution and visualized by confocal microscopy. The label for cell-surface HSP90 (red) and labeled OPCs (green) was detected by laser scanning confocal microscopy (MRC-1024, Bio-Rad) in 568- and 488-nm laser lines, respectively. Experiments omitting the primary antibodies were performed to test the background. No cross-talk between laser lines was observed. The immunofluorescence images were 11 optical sections (1 μm each) and were merged to form one image. (A and B) Representative images of untreated control cells (A) or cells treated with 10 nM radicicol (B). Red label was only detected in the top section and the yellow mark was caused by superimposition over the green label. Scale bar in μm. (C and D) Quantification of fluorescence intensity of anti-HSP90β antibodies in labeled OPCs in the absence (control) or presence of HSP90 inhibitors radicicol (C) or 17-AAG (D). Fluorescence intensity was quantified in each laser line separately using LaserSharp (Bio-Rad) software. Data are expressed as mean ± SEM pixel intensity, in gray scale (0–255 values) corrected for the background, determined in three independent experiments. Twenty to 50 cells per sample were averaged. ⁎p b 0.05, ⁎⁎⁎p b 0.001, compared to control cells (Student's t test). (E) Competition of HSP90 inhibitors against anti-HSP90β antibodies in western blotting. Membrane fraction samples subjected to western blotting were incubated individually without (control) or with radicicol or 17-AAG and exposed to anti-HSP90β antibodies (1.0 μg/ml) and peroxidaseconjugated secondary antibody (1:3,000 dilution). Data in parentheses express the quantification of the anti-HSP90 antibody reaction with reference to the control value (100%) and represent the average of three to four independent experiments. ⁎p b 0.05, ⁎⁎p b 0.01, compared to 100% control value (one-sample t test). (F–I) HSP90 inhibitors protected preoligodendrocytes from HSP90-antibody-induced cell death. Cells cultured for 6 days were untreated (control, white bars) or treated with 17-AAG (gray bars) or radicicol (black bars) for 30 min, followed by anti-HSP90β antibodies (2.5 μg/ml) and complement (1:10 dilution) for 1 h. Cells were washed and after 6 days in culture, fixed, incubated with O4 antibody (1:100 dilution) and rhodamine-red-X-conjugated anti-mouse IgM (1:400 dilution), mounted and visualized by fluorescence microscopy. (F) Cell protection was evaluated comparing the number of immunolabeled cells in the presence of HSP90 inhibitors with the number of cells treated with antibody and complement. The cells counted were all intact, stained and showed a full cell body. Two independent observers counted the labeled cells in a double blind fashion. At least 100–200 cells from nine (3 × 3) or more fields per coverslip were counted. Data are the mean ± SEM of three or four independent experiments run in duplicate using different batches of cell cultures. The number of O4+ preoligodendrocytes is represented as a percentage with reference to their number in control cultures, 36 ± 3 cells/mm2, defined as 100%. ###p b 0.001 versus control; ⁎p b 0.05, ⁎⁎p b 0.01 and ⁎⁎⁎p b 0.001, compared to cells treated with antibody and complement. Fluorescence images of labeled cells show a representative result in: (G) untreated cells (control); (H) cells treated with anti-HSP90β antibodies and complement showing pre-oligodendrocytes as “footprints”; and (I) cells treated with anti-HSP90β antibodies and complement in the presence of 10 nM radicicol. Scale bar in μm.
C. Cid, A. Alcazar / Experimental Neurology 225 (2010) 29–33
labeling the oligodendrocytes with both O1 and anti-galactocerebroside (GalC) monoclonal antibodies (both from Millipore) (Bansal et al., 1986) and similar results were found (data not shown). The reduction in the pre-oligodendrocyte population caused by antiHSP90β antibody and complement was prevented by lower concentrations of HSP90 inhibitors in a dose-dependent fashion (Fig. 1F). The concentration of HSP90 inhibitor needed to prevent the antibody
31
attack was dependent on the inhibitor used. Radicicol at a concentration of 10 nM prevented antibody attack, while 0.001 and 0.01 nM 17-AAG inhibited the antibody attack, but this was ineffective at higher concentrations (0.1 and 1.0 nM 17-AAG) (Figs. 1F−I). The concentrations of 17-AAG and radicicol that were effective at preventing antibody attack were similar to those required to compete with the anti-HSP90β antibodies in live cell experiments (Figs. 1C and D). When
32
C. Cid, A. Alcazar / Experimental Neurology 225 (2010) 29–33
OPCs in culture were treated with 10 nM 17-AAG or 100 nM radicicol, the antibody-induced cell death and the reported cytotoxicity of the inhibitors (Alcázar and Cid, 2009) were additive. It was of interest to confirm the potential biological activity of low doses of these drugs in perinatal and adult cells. OPCs from the optic nerve of 12- and N60-day-old rats were prepared according to the procedure of Raff et al. (1983), to obtain perinatal and adult oligodendrocyte lineage cells, respectively. Two-day-old optic nerve cultures consist largely of OPCs and maintaining the cells in lowserum medium yields cultures enriched in oligodendrocytes. In 7day-old optic nerve cultures, multipolar oligodendrocytes were evident under phase contrast microscopy and represented N95% of the living cells. The oligodendrocyte population is also sensitive to complement (Cid et al., 2005); therefore, control experiments were always performed by incubating the cells with complement alone to distinguish between inhibitor-dependent effects and complementmediated lysis. Two-day-old OPC cultures from the perinatal and adult optic nerve were untreated (control) or treated with HSP90 inhibitors in the presence of complement (1:20 dilution) and then exposed to anti-HSP90β antibodies (2.5 μg/ml). They were washed and placed in fresh culture medium for another 5 days. These cultures were fixed and labeled with O4 antibody to identify preoligodendrocytes by immunofluorescence (Supplementary material). The significant reduction in the number of pre-oligodendrocytes induced by antibody treatment of OPCs (1052 ± 60 and 632 ± 57 cells/cm2 in perinatal control and antibody-treated cells, respectively, p b 0.01; and 189 ± 39 and 53 ± 23 cells/cm2 in adult control and antibody-treated cells, p b 0.01, Student's t test) was prevented completely by 10 nM radicicol in perinatal and adult cultures (1017 ± 35 and 168 ± 25 cells/cm2, respectively, p b 0.01, compared to antibody-treated cells). Similar results were obtained with 0.01 nM 17-AAG (979 ± 95 and 152 ± 18 cells/cm2, in perinatal and adult cells, respectively, p b 0.05, compared to antibody-treated cells) (Supplementary material). The present work investigated a novel effect of HSP90 inhibitors at non-cytotoxic doses in OPCs. 17-AAG was tested and compared to radicicol, another chemically unrelated HSP90 inhibitor, to establish whether the observed effects were caused by binding of the inhibitor to HSP90 and to avoid any unknown effects that were not related to HSP90. The results for live cells demonstrated that HSP90 inhibitors targeted cell-surface HSP90 in OPCs and competed with anti-HSP90 antibodies for binding to HSP90. These inhibitors were able to compete with the antibodies in western blotting experiments and inhibited the antibody reaction. The effects of radicicol in both competitive assays, live immunofluorescence and immunoblotting experiments, were dose dependent. When 17-AAG was used in the competition assays, concentrations of 0.01 nM and even 0.001 nM inhibited anti-HSP90β antibody binding to live cells, whereas concentrations of 0.1–1.0 nM did not decrease antibody binding. This effect also occurred in western blotting experiments, in which 10 nM 17-AAG was more effective than 1 μM for competition against anti-HSP90β antibodies. These results were consistent with the fact that 0.01 and 0.001 nM 17-AAG protected significantly against antiHSP90β antibody attack, whereas 0.1 and 1.0 nM did not protect the pre-oligodendrocytes. OPCs are myelin-forming after their differentiation into postmitotic oligodendrocytes (Levine et al., 2001). Investigation of whether HSP90 inhibitors are potentially protective for OPCs that develop into oligodendrocytes is important for considering this effect as a new protective strategy for the prevention of anti-HSP90 antibody attack. Radicicol at 10 and 0.01 nM 17-AAG prevented HSP90β-antibody-induced OPC death and preserved the pre-oligodendrocytes. Furthermore, we demonstrated these results in adult OPCs to confirm the therapeutic potential of low doses of HSP90 inhibitors in MS patients that have HSP90 autoantibodies. Adult OPCs express many of the same phenotypic marker antigens as their
perinatal counterparts, although they differ in terms of their cell cycle duration and time course of differentiation, all of which are slower (Levine et al., 2001). HSP90 inhibitors also prevented antibodyinduced cell death in adult OPCs, protected adult oligodendrocytes and demonstrated the potential biological activity of low doses of these drugs. We report that, compared to other cells, OPCs are very sensitive to HSP90 inhibitors; these inhibitors induce significant OPC death without affecting other types of CNS cells (Alcázar and Cid, 2009). The present work found a dual effect of these HSP90 inhibitors at low and non-cytotoxic doses, which interacted with the HSP90 exposed on the OPC surface. In another study, geldanamycin and radicicol at low doses of 2 and 20 nM, respectively, have been described as protective against neurocytotoxicity induced by anticancer drugs, but higher doses of both inhibitors are cytotoxic (Sano, 2001), in accordance with the results described here. Cell-surface HSP90 expressed in OPCs may be a target for the immune response in MS. Although a physiological role for anti-HSP90 autoantibodies in MS remains to be established, it has been demonstrated that these antibodies can reduce the new preoligodendrocyte population in culture (Cid et al., 2005). MS is an immune-mediated disease, and it is recognized that antibodydependent immune mechanisms are involved in the formation of its characteristic lesions (Noseworthy et al., 2000). Moreover, it is known that HSP90 is highly immunogenic at the N-terminal region (Kishimoto et al., 2005). This protein has been described as an autoantigen in several auto-immune diseases such as rheumatoid arthritis (Hayem et al., 1999), systemic lupus erythematosus (Ripley et al., 2001), type 1 diabetes (Qin et al., 2003) and ovarian autoimmunity (Pires and Khole, 2009). Finally, the N-terminal domain also binds peptides and participates in direct and indirect antigen presentation (Callahan et al., 2008). HSP90 inhibitors that bind to this region, i.e. radicicol and 17-AAG, at non-cytotoxic doses can interact with HSP90 and compete with anti-HSP90 autoantibodies and can prevent their immune attack, which contributes to the protection of new oligodendrocytes and remyelination. They may have therapeutic potential for MS patients who have anti-HSP90 autoantibodies. Our results report three novel findings: (i) a new effect for HSP90 inhibitors, by binding to cell-surface HSP90 and competing with antiHSP90 antibodies; (ii) this effect was induced by low and noncytotoxic doses of the drugs and protected oligodendrocytes against these antibodies; and (iii) a novel strategy for the therapeutic use of HSP90 inhibitors. We are indebted to Ms. M. Gómez-Calcerrada for her technical assistance. This work was supported by ISCIII grants 08/0761 and RETICS-RD06/0026/0008, and grant CTM2008-00304 from Spanish Ministry of Science and Innovation. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.expneurol.2009.11.017. References Alcázar, A., Cid, C., 2009. High cytotoxic sensitivity of the oligodendrocyte precursor cells to HSP90 inhibitors in cell cultures. Exp. Neurol 216, 511–514. Armstrong, R.C., Dorn, H.H., Kufta, C.V., Friedman, E., Dubois-Dalcq, M.E., 1992. Preoligodendrocytes from adult human CNS. J. Neurosci. 12, 1538–1547. Bansal, R., Warrington, A.E., Gard, A.L., Ranscht, B., Pfeiffer, S.E., 1986. Multiple and novel specificities of monoclonal antibodies O1, O4, and R-mAb used in the analysis of oligodendrocyte development. J. Neurosci. Res. 24, 548–557. Blakemore, W.F., Keirstead, H.S., 1999. The origin of remyelinating cells in the central nervous system. J. Neuroimmunol. 98, 69–76. Callahan, M.K., Garg, M., Srivastava, P.K., 2008. Heat-shock protein 90 associates with N-terminal extended peptides and is required for direct and indirect antigen presentation. Proc. Natl. Acad. Sci. USA 105, 1662–1667. Chaudhury, S., Welch, T.R., Blagg, B.S.J., 2006. Hsp90 as a target for drug development. Chem. Med. Chem. 1, 1331–1340.
C. Cid, A. Alcazar / Experimental Neurology 225 (2010) 29–33 Cid, C., Álvarez-Cermeño, J.C., Camafeita, E., Salinas, M., Alcázar, A., 2004. Antibodies reactive to heat shock protein 90 induce oligodendrocyte precursor cell death in culture. Implications for demyelination in multiple sclerosis. FASEB J. 18, 409–411. Cid, C., Álvarez-Cermeño, J.C., Salinas, M., Alcázar, A., 2005. Anti-heat shock protein 90 β antibodies decrease pre-oligodendrocyte population in perinatal and adult cell cultures. Implications for remyelination in multiple sclerosis. J. Neurochem. 95, 349–360. Cid, C., García-Villanueva, M., Salinas, M., Alcázar, A., 2007a. Detection of anti-heat shock protein 90 β (Hsp90β) antibodies in cerebrospinal fluid. J. Immunol. Methods 318, 153–157. Cid, C., Regidor, I., Alcázar, A., 2007b. Anti-heat shock protein 90b antibodies are detected in patients with multiple sclerosis during remission. J. Neuroimmunol. 184, 223–226. Franklin, R.J., 2002. Why does remyelination fail in multiple sclerosis? Nat. Rev. Neurosci. 3, 705–714. Hayem, G., De Brand, M., Palazzo, E., Rouxd, S., Combe, B., Eliaou, J.F., Sany, J., Kahn, M.F., Meyer, O., 1999. Anti-heat shock protein 70 kDa and 90 kDa antibodies in serum of patients with rheumatoid arthritis. Ann. Rheum. Dis. 58, 291–296. Kishimoto, J., Fukuma, Y., Mizuno, A., Nemoto, T.K., 2005. Identification of the pentapeptide constituting a dominant epitope common to all eukaryotic heat shock protein 90 molecular chaperones. Cell Stress Chaperones 10, 296–311. Kondo, T., Raff, M., 2000. Oligodendrocyte precursor cells reprogrammed to become multipotential CNS stem cells. Science 289, 1754–1757. Levine, J.M., Reynolds, R., Fawcett, J.W., 2001. The oligodendrocyte precursor cell in health and disease. Trends Neurosci 24, 39–47. Noseworthy, J.H., Lucchinetti, C., Rodriguez, M., Weinshenker, B.G., 2000. Multiple sclerosis. N. Engl. J. Med. 343, 938–952.
33
Pires E.S., Khole V.V., 2009. A block in the road to fertility: autoantibodies to heat-shock protein 90-β in human ovarian autoimmunity. Fertil. Steril. 92, 1395–1409. Qin, H.Y., Mahon, J.L., Atkinson, M.A., Chaturvedi, P., Lee-Chan, E., Singh, B., 2003. Type 1 diabetes alters anti-hsp90 autoantibody isotype. J. Autoimmun. 20, 237–245. Raff, M.C., Miller, R.H., Noble, M., 1983. A glial progenitor cell that develops in vitro into an astrocyte or an oligodendrocyte depending on culture medium. Nature 303, 390–396. Ripley, B.J., Isenberg, D.A., Latchman, D.S., 2001. Elevated levels of the 90 kDa heat shock protein (hsp90) in SLE correlate with levels of IL-6 and autoantibodies to hsp90. J. Autoimmun. 17, 341–346. Roy, N.S., Wang, S., Harrison-Restelli, C., Benraiss, A., Fraser, R.A.R., Gravel, M., Braun, P.E., Goldman, S.A., 1999. Identification, isolation, and promoter-defined separation of mitotic oligodendrocyte progenitor cells from the adult human subcortical white matter. J. Neurosci. 19, 9986–9995. Sano, M., 2001. Radicicol and geldanamycin prevent neurotoxic effects of anticancer drugs on cultured embryonic sensory neurons. Neuropharmacology 40, 947–953. Scolding, N.J., Rayner, P.J., Compston, D.A.S., 1999. Identification of A2B5-positive putative oligodendrocyte progenitor cells and A2B5-positive astrocytes in adult human white matter. Neuroscience 89, 1–4. Shi, J., Marinovich, A., Barres, B.A., 1998. Purification and characterization of adult oligodendrocyte precursor cells from the rat optic nerve. J. Neurosci. 15, 4627–4636. Sidera, K., Samiotaki, M., Yfanti, E., Panayotou, G., Patsavoudi, E., 2004. Involvement of cell surface HSP90 in cell migration reveals a novel role in the developing nervous system. J. Biol. Chem. 279, 45379–45388. Temple, S., Raff, M.C., 1986. Clonal analysis of oligodendrocyte development in culture: evidence for a developmental clock that counts cell divisions. Cell 44, 773–779.
Contents lists available at ScienceDirect
Experimental Neurology j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / y e x n r
Short Communication
Protection of oligodendrocyte precursor cells by low doses of HSP90 inhibitors in cell culture Cristina Cid a, Alberto Alcazar b,⁎ a b
Center for Astrobiology, CSIC-INTA, Torrejon de Ardoz, Spain Department of Investigation, Hospital Ramón y Cajal, Madrid, Spain
a r t i c l e
i n f o
Article history: Received 30 July 2009 Revised 4 November 2009 Accepted 21 November 2009 Available online 4 December 2009 Keywords: 17-allylamino-17-demethoxygeldanamycin Heat shock protein 90 HSP90 antibodies HSP90 beta HSP90 inhibitors O-2A cells Cell death Oligodendrocyte precursor cells Radicicol
a b s t r a c t Oligodendrocyte precursor cells (OPCs) become myelin-forming after their differentiation into post-mitotic oligodendrocytes. OPCs are extremely efficient at myelin repair and contribute to remyelination. However, remyelination fails in multiple sclerosis (MS), which suggest that the OPCs are ineffective in this disorder. We have studied previously the expression of heat shock protein 90 (HSP90) in OPCs and have reported autoantibodies against HSP90 in MS patients, which recognize the antigen on the OPC surface. The present study investigated a protective effect of HSP90 inhibitors observed in cultured OPCs. Radicicol and 17allylamino-17-demethoxygeldanamycin (17-AAG) at non-cytotoxic doses targeted cell-surface HSP90 in OPCs. Thus, 0.01 nM 17-AAG or 10 nM radicicol competed with the anti-HSP90 antibodies for binding to cellsurface HSP90. These low doses of HSP90 inhibitors prevented HSP90-antibody-induced OPC death and protected the oligodendrocyte population against antibody attack. Adult oligodendrocytes were protected by these low doses of HSP90 inhibitors in a similar fashion to perinatal cells. The present results show that, despite OPCs being very sensitive to HSP90 inhibitors, low and non-cytotoxic doses of 17-AAG and radicicol protect oligodendrocytes from anti-HSP90 antibody attack. They may have therapeutic potential for MS patients that have anti-HSP90 autoantibodies and provide a novel strategy for therapeutic intervention with HSP90 inhibitors. © 2009 Elsevier Inc. All rights reserved.
Oligodendrocyte precursor cells (OPCs) retain characteristics of multipotent central nervous system (CNS) stem cells (Kondo and Raff, 2000). They are also known as O-2A progenitor cells, are able to proliferate, and most differentiate into post-mitotic oligodendrocytes (Temple and Raff, 1986). These cells have been identified in rodent cell cultures and in the adult rodent and human CNS (Scolding et al., 1999; Roy et al., 1999; Raff et al., 1983; Shi et al., 1998). In humans, these cells are similar immunophenotypically to their rodent counterparts (Scolding et al., 1999; Armstrong et al., 1992). OPCs are extremely efficient at myelin repair and contribute to remyelination, which assigns them a role in myelination (Scolding et al., 1999; Blakemore and Keirstead, 1999). However, failure to remyelinate is a pathological characteristic of the human demyelinating disease multiple sclerosis (MS), which suggests that these cells are ineffective in this disorder (Franklin, 2002). The reason for this remains unclear (Franklin, 2002), but it is known that oligodendrocytes that survive in an area of demyelination do not contribute to remyelination, which is undertaken by the OPCs (Blakemore and Keirstead, 1999). We have
Abbreviations: 17-AAG, 17-allylamino-17-demethoxygeldanamycin; HSP90, heat shock protein 90; OPCs, oligodendrocyte precursor cells. ⁎ Corresponding author. Department of Investigation, Hospital Ramón y Cajal, Ctra. Colmenar km 9.1, 28034 Madrid, Spain. Fax: +34 91 336 9016. E-mail address: [email protected] (A. Alcazar). 0014-4886/$ – see front matter © 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.expneurol.2009.11.017
studied previously the expression of heat shock protein 90 (HSP90) in OPCs. HSP90 is a conserved molecular chaperone (rat HSP90β only differs in 10 amino acids out of 723 from the human form of the protein) and is part of a multichaperone complex whose association is required for the stability and function of multiple signaling proteins that promote cell growth, differentiation and apoptosis. We have found that HSP90β is expressed at the surface of OPCs, and it is not expressed when OPCs differentiate into pre-oligodendrocytes (Cid et al., 2004, 2005). These results, in addition to those that describe expression of HSP90 at the surface of cerebellar and Schwann cells (Sidera et al., 2004), suggest that cell-surface HSP90 has a role in differentiation, cell migration and development of certain types of cells in the nervous system. In addition, we have reported autoantibodies against HSP90β in MS patients that recognize the antigen on the OPC surface (Cid et al., 2004, 2007b). Rabbit anti-HSP90 antibodies against the N-terminal region of HSP90β recognize this protein at the cell surface of OPCs in a similar fashion to anti-HSP90 autoantibodies from MS patients (Cid et al., 2004, 2005) and indicate the extracellular localization of this region. Human and rabbit antiHsp90β antibodies attack OPCs and decrease the oligodendrocyte population in cell cultures (Cid et al., 2005). Development of new strategies to protect the new oligodendrocyte population is important for promoting remyelination in MS. In this regard, we have studied the effect of HSP90 inhibitors geldanamycin, its derivate 17-
30
C. Cid, A. Alcazar / Experimental Neurology 225 (2010) 29–33
allylamino-17-demethoxygeldanamycin (17-AAG) and the macrolactone radicicol on cultured OPCs. These HSP90 inhibitors interact with the ATP-binding site located at the N-terminal region of HSP90 (Chaudhury et al., 2006). We have found that OPCs are extremely sensitive to HSP90 inhibitors compared to other CNS cells (Alcázar and Cid, 2009). Thus, the IC50 of geldanamycin, 17-AAG and radicicol for OPCs was 7.1, 10.7 and 137 nM, respectively, compared to 1–2 μM for pre-oligodendrocytes, astrocytes or neurons. However, ≤1 nM geldanamycin or 17-AAG and ≤10 nM radicicol were non-cytotoxic for OPCs (Alcázar and Cid, 2009). The present work investigated the use of non-cytotoxic doses of HSP90 inhibitors as a new potential therapeutic strategy. We studied the possible binding of HSP90 inhibitors to cell-surface HSP90 in OPCs, and a novel effect on oligodendrocyte lineage cells, cultured OPCs and pre-oligodendrocytes, and found a protective effect at low doses. Primary cell cultures were prepared from cerebral hemispheres of 17- to 19-day-old Sprague–Dawley rat embryos, as described previously (Cid et al., 2004, 2005). All animal procedures were in accordance with The Declaration of Helsinki and Spanish legislation for the use of animals in biomedical experimentation. OPCs were identified phenotypically by their characteristic bipolar morphology and A2B5+ immunoreactivity labeling with mouse monoclonal A2B5 IgM (Millipore) (Raff et al., 1983). The oligodendrocytes were identified by their multipolar morphology and O4+ immunoreactivity labeling with mouse monoclonal O4 IgM (Millipore). O4 is a marker of pre-oligodendrocytes, which are cells that have differentiated into the oligodendrocyte lineage (Armstrong et al., 1992). In 6day-old cell cultures, A2B5+ OPCs comprised 12 ± 1% of the cells, and very few O4+ pre-oligodendrocytes were present (≤1 ± 0.2%); from days 6 to 12 of culture, a gradual reduction in OPCs was observed, while the number of pre-oligodendrocytes correspondingly increased to 10–12% of all cells (Cid et al., 2004). The number of GFAP+ astrocytes and βIII-tubulin+ neurons remained constant over time and were negative for A2B5 (Cid et al., 2004; Alcázar and Cid, 2009). Competition assays between HSP90 inhibitors at non-toxic concentrations and anti-HSP90β antibodies were performed in live cells. Geldanamycin shares biological activities with 17-AAG and has an effect on OPCs at similar doses to 17-AAG (Alcázar and Cid, 2009); therefore, these experiments were done only with 17-AAG and the chemically unrelated HSP90 inhibitor radicicol. Cells cultured for 6 days were untreated (control) or treated with 0.001–1.0 nM 17-AAG or with 0.1–10 nM radicicol (both from Sigma) for 1 h. Live cells were then labeled with rabbit polyclonal anti-HSP90β (HSP84) antibodies (Millipore) to target cell-surface HSP90 as described previously (Cid
et al., 2004, 2005). Living cells were incubated with anti-HSP90β antibodies (1:10 dilution) for 30 min at 25°C and labeled with lissamine rhodamine-conjugated goat anti-rabbit IgG (Millipore) (1:100 dilution) for 1 h at 4 °C. Cells were fixed and exposed to A2B5 antibody to label OPCs (Figs. 1A–D). Radicicol at 10 and 0.01 nM 17-AAG decreased significantly the labeling with anti-HSP90β antibodies on the surface of OPCs (Figs. 1A–D). This decrease in labeling with anti-HSP90β antibodies was quantified by measuring their fluorescence intensity in labeled OPCs. Thus, the fluorescence intensity of anti-HSP90β antibodies was 230±14 in control cells compared to 26±8 in 10 nM radicicol-treated cells (p = 0.0003) (Fig. 1C). Concentrations of 0.01 nM and even 0.001 nM 17-AAG also decreased the fluorescence intensity of the label for HSP90β in OPCs (Fig. 1D). The results demonstrated that these HSP90 inhibitors interacted with cell-surface HSP90β and competed with the antiHSP90β antibodies in OPCs. Competitive assays were performed with western blotting using samples of cell membrane fractions, which were obtained from 6-day-old cultured cells as described previously (Cid et al., 2004, 2007a). Membrane fraction samples were subjected to SDS–PAGE with 5–12% acrylamide (2.6% cross-linker) discontinuous gels and blotted in polyvinylidene-fluoride membranes. Samples for western blotting were cut into strips and incubated individually with different concentrations of HSP90 inhibitors for 1 h. They were exposed to anti-HSP90β antibodies for 1 h and incubated with peroxidase-conjugated goat anti-rabbit IgG (Bio-Rad) and developed with ECL reagent (GE Healthcare) (Fig. 1E). The HSP90 inhibitors decreased the anti-HSP90β antibody reaction in immunoblots in a dose-dependent manner (Fig. 1E), which demonstrated that these inhibitors competed with anti-HSP90β antibodies for binding to HSP90. We have reported previously that anti-HSP90β antibodies induce OPC death by a mechanism mediated by complement activation, which leads to a decrease in the pre-oligodendrocyte population (Cid et al., 2005). Besides, as describe above, HSP90 inhibitors at noncytotoxic doses inhibited the recognition of cell-surface HSP90 by anti-HSP90β antibodies in OPCs. We studied the possible protective effect of the inhibitors against anti-HSP90β antibody attack in preoligodendrocytes. Six-day cell cultures were untreated or treated with HSP90 inhibitors and exposed to anti-HSP90β antibodies (2.5 μg/ml; a concentration in the range determined in CSF from MS patients; Cid et al., 2005) and rabbit complement (Calbiochem). Cells were washed and placed in fresh medium for another 6 days. These 12-day cultured cells were fixed and exposed to O4 antibody to label pre-oligodendrocytes (Figs. 1F−I). Additional experiments were performed
Fig. 1. (A–D) HSP90 inhibitors interacted with cell-surface HSP90 in OPCs and competed with anti-HSP90 antibodies. Cells untreated (control) or treated with 17-AAG or radicicol were incubated with anti-HSP90β antibodies and the secondary antibody. Cells were then fixed with 3.7% paraformaldehyde for 10 min and incubated for 1 h at room temperature or overnight at 4 °C with A2B5 antibody (1:100 dilution) to label OPCs. Cells were washed and incubated with fluorescein-conjugated anti-mouse IgM (1:40 dilution) for 1 h at room temperature, mounted with anti-fade solution and visualized by confocal microscopy. The label for cell-surface HSP90 (red) and labeled OPCs (green) was detected by laser scanning confocal microscopy (MRC-1024, Bio-Rad) in 568- and 488-nm laser lines, respectively. Experiments omitting the primary antibodies were performed to test the background. No cross-talk between laser lines was observed. The immunofluorescence images were 11 optical sections (1 μm each) and were merged to form one image. (A and B) Representative images of untreated control cells (A) or cells treated with 10 nM radicicol (B). Red label was only detected in the top section and the yellow mark was caused by superimposition over the green label. Scale bar in μm. (C and D) Quantification of fluorescence intensity of anti-HSP90β antibodies in labeled OPCs in the absence (control) or presence of HSP90 inhibitors radicicol (C) or 17-AAG (D). Fluorescence intensity was quantified in each laser line separately using LaserSharp (Bio-Rad) software. Data are expressed as mean ± SEM pixel intensity, in gray scale (0–255 values) corrected for the background, determined in three independent experiments. Twenty to 50 cells per sample were averaged. ⁎p b 0.05, ⁎⁎⁎p b 0.001, compared to control cells (Student's t test). (E) Competition of HSP90 inhibitors against anti-HSP90β antibodies in western blotting. Membrane fraction samples subjected to western blotting were incubated individually without (control) or with radicicol or 17-AAG and exposed to anti-HSP90β antibodies (1.0 μg/ml) and peroxidaseconjugated secondary antibody (1:3,000 dilution). Data in parentheses express the quantification of the anti-HSP90 antibody reaction with reference to the control value (100%) and represent the average of three to four independent experiments. ⁎p b 0.05, ⁎⁎p b 0.01, compared to 100% control value (one-sample t test). (F–I) HSP90 inhibitors protected preoligodendrocytes from HSP90-antibody-induced cell death. Cells cultured for 6 days were untreated (control, white bars) or treated with 17-AAG (gray bars) or radicicol (black bars) for 30 min, followed by anti-HSP90β antibodies (2.5 μg/ml) and complement (1:10 dilution) for 1 h. Cells were washed and after 6 days in culture, fixed, incubated with O4 antibody (1:100 dilution) and rhodamine-red-X-conjugated anti-mouse IgM (1:400 dilution), mounted and visualized by fluorescence microscopy. (F) Cell protection was evaluated comparing the number of immunolabeled cells in the presence of HSP90 inhibitors with the number of cells treated with antibody and complement. The cells counted were all intact, stained and showed a full cell body. Two independent observers counted the labeled cells in a double blind fashion. At least 100–200 cells from nine (3 × 3) or more fields per coverslip were counted. Data are the mean ± SEM of three or four independent experiments run in duplicate using different batches of cell cultures. The number of O4+ preoligodendrocytes is represented as a percentage with reference to their number in control cultures, 36 ± 3 cells/mm2, defined as 100%. ###p b 0.001 versus control; ⁎p b 0.05, ⁎⁎p b 0.01 and ⁎⁎⁎p b 0.001, compared to cells treated with antibody and complement. Fluorescence images of labeled cells show a representative result in: (G) untreated cells (control); (H) cells treated with anti-HSP90β antibodies and complement showing pre-oligodendrocytes as “footprints”; and (I) cells treated with anti-HSP90β antibodies and complement in the presence of 10 nM radicicol. Scale bar in μm.
C. Cid, A. Alcazar / Experimental Neurology 225 (2010) 29–33
labeling the oligodendrocytes with both O1 and anti-galactocerebroside (GalC) monoclonal antibodies (both from Millipore) (Bansal et al., 1986) and similar results were found (data not shown). The reduction in the pre-oligodendrocyte population caused by antiHSP90β antibody and complement was prevented by lower concentrations of HSP90 inhibitors in a dose-dependent fashion (Fig. 1F). The concentration of HSP90 inhibitor needed to prevent the antibody
31
attack was dependent on the inhibitor used. Radicicol at a concentration of 10 nM prevented antibody attack, while 0.001 and 0.01 nM 17-AAG inhibited the antibody attack, but this was ineffective at higher concentrations (0.1 and 1.0 nM 17-AAG) (Figs. 1F−I). The concentrations of 17-AAG and radicicol that were effective at preventing antibody attack were similar to those required to compete with the anti-HSP90β antibodies in live cell experiments (Figs. 1C and D). When
32
C. Cid, A. Alcazar / Experimental Neurology 225 (2010) 29–33
OPCs in culture were treated with 10 nM 17-AAG or 100 nM radicicol, the antibody-induced cell death and the reported cytotoxicity of the inhibitors (Alcázar and Cid, 2009) were additive. It was of interest to confirm the potential biological activity of low doses of these drugs in perinatal and adult cells. OPCs from the optic nerve of 12- and N60-day-old rats were prepared according to the procedure of Raff et al. (1983), to obtain perinatal and adult oligodendrocyte lineage cells, respectively. Two-day-old optic nerve cultures consist largely of OPCs and maintaining the cells in lowserum medium yields cultures enriched in oligodendrocytes. In 7day-old optic nerve cultures, multipolar oligodendrocytes were evident under phase contrast microscopy and represented N95% of the living cells. The oligodendrocyte population is also sensitive to complement (Cid et al., 2005); therefore, control experiments were always performed by incubating the cells with complement alone to distinguish between inhibitor-dependent effects and complementmediated lysis. Two-day-old OPC cultures from the perinatal and adult optic nerve were untreated (control) or treated with HSP90 inhibitors in the presence of complement (1:20 dilution) and then exposed to anti-HSP90β antibodies (2.5 μg/ml). They were washed and placed in fresh culture medium for another 5 days. These cultures were fixed and labeled with O4 antibody to identify preoligodendrocytes by immunofluorescence (Supplementary material). The significant reduction in the number of pre-oligodendrocytes induced by antibody treatment of OPCs (1052 ± 60 and 632 ± 57 cells/cm2 in perinatal control and antibody-treated cells, respectively, p b 0.01; and 189 ± 39 and 53 ± 23 cells/cm2 in adult control and antibody-treated cells, p b 0.01, Student's t test) was prevented completely by 10 nM radicicol in perinatal and adult cultures (1017 ± 35 and 168 ± 25 cells/cm2, respectively, p b 0.01, compared to antibody-treated cells). Similar results were obtained with 0.01 nM 17-AAG (979 ± 95 and 152 ± 18 cells/cm2, in perinatal and adult cells, respectively, p b 0.05, compared to antibody-treated cells) (Supplementary material). The present work investigated a novel effect of HSP90 inhibitors at non-cytotoxic doses in OPCs. 17-AAG was tested and compared to radicicol, another chemically unrelated HSP90 inhibitor, to establish whether the observed effects were caused by binding of the inhibitor to HSP90 and to avoid any unknown effects that were not related to HSP90. The results for live cells demonstrated that HSP90 inhibitors targeted cell-surface HSP90 in OPCs and competed with anti-HSP90 antibodies for binding to HSP90. These inhibitors were able to compete with the antibodies in western blotting experiments and inhibited the antibody reaction. The effects of radicicol in both competitive assays, live immunofluorescence and immunoblotting experiments, were dose dependent. When 17-AAG was used in the competition assays, concentrations of 0.01 nM and even 0.001 nM inhibited anti-HSP90β antibody binding to live cells, whereas concentrations of 0.1–1.0 nM did not decrease antibody binding. This effect also occurred in western blotting experiments, in which 10 nM 17-AAG was more effective than 1 μM for competition against anti-HSP90β antibodies. These results were consistent with the fact that 0.01 and 0.001 nM 17-AAG protected significantly against antiHSP90β antibody attack, whereas 0.1 and 1.0 nM did not protect the pre-oligodendrocytes. OPCs are myelin-forming after their differentiation into postmitotic oligodendrocytes (Levine et al., 2001). Investigation of whether HSP90 inhibitors are potentially protective for OPCs that develop into oligodendrocytes is important for considering this effect as a new protective strategy for the prevention of anti-HSP90 antibody attack. Radicicol at 10 and 0.01 nM 17-AAG prevented HSP90β-antibody-induced OPC death and preserved the pre-oligodendrocytes. Furthermore, we demonstrated these results in adult OPCs to confirm the therapeutic potential of low doses of HSP90 inhibitors in MS patients that have HSP90 autoantibodies. Adult OPCs express many of the same phenotypic marker antigens as their
perinatal counterparts, although they differ in terms of their cell cycle duration and time course of differentiation, all of which are slower (Levine et al., 2001). HSP90 inhibitors also prevented antibodyinduced cell death in adult OPCs, protected adult oligodendrocytes and demonstrated the potential biological activity of low doses of these drugs. We report that, compared to other cells, OPCs are very sensitive to HSP90 inhibitors; these inhibitors induce significant OPC death without affecting other types of CNS cells (Alcázar and Cid, 2009). The present work found a dual effect of these HSP90 inhibitors at low and non-cytotoxic doses, which interacted with the HSP90 exposed on the OPC surface. In another study, geldanamycin and radicicol at low doses of 2 and 20 nM, respectively, have been described as protective against neurocytotoxicity induced by anticancer drugs, but higher doses of both inhibitors are cytotoxic (Sano, 2001), in accordance with the results described here. Cell-surface HSP90 expressed in OPCs may be a target for the immune response in MS. Although a physiological role for anti-HSP90 autoantibodies in MS remains to be established, it has been demonstrated that these antibodies can reduce the new preoligodendrocyte population in culture (Cid et al., 2005). MS is an immune-mediated disease, and it is recognized that antibodydependent immune mechanisms are involved in the formation of its characteristic lesions (Noseworthy et al., 2000). Moreover, it is known that HSP90 is highly immunogenic at the N-terminal region (Kishimoto et al., 2005). This protein has been described as an autoantigen in several auto-immune diseases such as rheumatoid arthritis (Hayem et al., 1999), systemic lupus erythematosus (Ripley et al., 2001), type 1 diabetes (Qin et al., 2003) and ovarian autoimmunity (Pires and Khole, 2009). Finally, the N-terminal domain also binds peptides and participates in direct and indirect antigen presentation (Callahan et al., 2008). HSP90 inhibitors that bind to this region, i.e. radicicol and 17-AAG, at non-cytotoxic doses can interact with HSP90 and compete with anti-HSP90 autoantibodies and can prevent their immune attack, which contributes to the protection of new oligodendrocytes and remyelination. They may have therapeutic potential for MS patients who have anti-HSP90 autoantibodies. Our results report three novel findings: (i) a new effect for HSP90 inhibitors, by binding to cell-surface HSP90 and competing with antiHSP90 antibodies; (ii) this effect was induced by low and noncytotoxic doses of the drugs and protected oligodendrocytes against these antibodies; and (iii) a novel strategy for the therapeutic use of HSP90 inhibitors. We are indebted to Ms. M. Gómez-Calcerrada for her technical assistance. This work was supported by ISCIII grants 08/0761 and RETICS-RD06/0026/0008, and grant CTM2008-00304 from Spanish Ministry of Science and Innovation. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.expneurol.2009.11.017. References Alcázar, A., Cid, C., 2009. High cytotoxic sensitivity of the oligodendrocyte precursor cells to HSP90 inhibitors in cell cultures. Exp. Neurol 216, 511–514. Armstrong, R.C., Dorn, H.H., Kufta, C.V., Friedman, E., Dubois-Dalcq, M.E., 1992. Preoligodendrocytes from adult human CNS. J. Neurosci. 12, 1538–1547. Bansal, R., Warrington, A.E., Gard, A.L., Ranscht, B., Pfeiffer, S.E., 1986. Multiple and novel specificities of monoclonal antibodies O1, O4, and R-mAb used in the analysis of oligodendrocyte development. J. Neurosci. Res. 24, 548–557. Blakemore, W.F., Keirstead, H.S., 1999. The origin of remyelinating cells in the central nervous system. J. Neuroimmunol. 98, 69–76. Callahan, M.K., Garg, M., Srivastava, P.K., 2008. Heat-shock protein 90 associates with N-terminal extended peptides and is required for direct and indirect antigen presentation. Proc. Natl. Acad. Sci. USA 105, 1662–1667. Chaudhury, S., Welch, T.R., Blagg, B.S.J., 2006. Hsp90 as a target for drug development. Chem. Med. Chem. 1, 1331–1340.
C. Cid, A. Alcazar / Experimental Neurology 225 (2010) 29–33 Cid, C., Álvarez-Cermeño, J.C., Camafeita, E., Salinas, M., Alcázar, A., 2004. Antibodies reactive to heat shock protein 90 induce oligodendrocyte precursor cell death in culture. Implications for demyelination in multiple sclerosis. FASEB J. 18, 409–411. Cid, C., Álvarez-Cermeño, J.C., Salinas, M., Alcázar, A., 2005. Anti-heat shock protein 90 β antibodies decrease pre-oligodendrocyte population in perinatal and adult cell cultures. Implications for remyelination in multiple sclerosis. J. Neurochem. 95, 349–360. Cid, C., García-Villanueva, M., Salinas, M., Alcázar, A., 2007a. Detection of anti-heat shock protein 90 β (Hsp90β) antibodies in cerebrospinal fluid. J. Immunol. Methods 318, 153–157. Cid, C., Regidor, I., Alcázar, A., 2007b. Anti-heat shock protein 90b antibodies are detected in patients with multiple sclerosis during remission. J. Neuroimmunol. 184, 223–226. Franklin, R.J., 2002. Why does remyelination fail in multiple sclerosis? Nat. Rev. Neurosci. 3, 705–714. Hayem, G., De Brand, M., Palazzo, E., Rouxd, S., Combe, B., Eliaou, J.F., Sany, J., Kahn, M.F., Meyer, O., 1999. Anti-heat shock protein 70 kDa and 90 kDa antibodies in serum of patients with rheumatoid arthritis. Ann. Rheum. Dis. 58, 291–296. Kishimoto, J., Fukuma, Y., Mizuno, A., Nemoto, T.K., 2005. Identification of the pentapeptide constituting a dominant epitope common to all eukaryotic heat shock protein 90 molecular chaperones. Cell Stress Chaperones 10, 296–311. Kondo, T., Raff, M., 2000. Oligodendrocyte precursor cells reprogrammed to become multipotential CNS stem cells. Science 289, 1754–1757. Levine, J.M., Reynolds, R., Fawcett, J.W., 2001. The oligodendrocyte precursor cell in health and disease. Trends Neurosci 24, 39–47. Noseworthy, J.H., Lucchinetti, C., Rodriguez, M., Weinshenker, B.G., 2000. Multiple sclerosis. N. Engl. J. Med. 343, 938–952.
33
Pires E.S., Khole V.V., 2009. A block in the road to fertility: autoantibodies to heat-shock protein 90-β in human ovarian autoimmunity. Fertil. Steril. 92, 1395–1409. Qin, H.Y., Mahon, J.L., Atkinson, M.A., Chaturvedi, P., Lee-Chan, E., Singh, B., 2003. Type 1 diabetes alters anti-hsp90 autoantibody isotype. J. Autoimmun. 20, 237–245. Raff, M.C., Miller, R.H., Noble, M., 1983. A glial progenitor cell that develops in vitro into an astrocyte or an oligodendrocyte depending on culture medium. Nature 303, 390–396. Ripley, B.J., Isenberg, D.A., Latchman, D.S., 2001. Elevated levels of the 90 kDa heat shock protein (hsp90) in SLE correlate with levels of IL-6 and autoantibodies to hsp90. J. Autoimmun. 17, 341–346. Roy, N.S., Wang, S., Harrison-Restelli, C., Benraiss, A., Fraser, R.A.R., Gravel, M., Braun, P.E., Goldman, S.A., 1999. Identification, isolation, and promoter-defined separation of mitotic oligodendrocyte progenitor cells from the adult human subcortical white matter. J. Neurosci. 19, 9986–9995. Sano, M., 2001. Radicicol and geldanamycin prevent neurotoxic effects of anticancer drugs on cultured embryonic sensory neurons. Neuropharmacology 40, 947–953. Scolding, N.J., Rayner, P.J., Compston, D.A.S., 1999. Identification of A2B5-positive putative oligodendrocyte progenitor cells and A2B5-positive astrocytes in adult human white matter. Neuroscience 89, 1–4. Shi, J., Marinovich, A., Barres, B.A., 1998. Purification and characterization of adult oligodendrocyte precursor cells from the rat optic nerve. J. Neurosci. 15, 4627–4636. Sidera, K., Samiotaki, M., Yfanti, E., Panayotou, G., Patsavoudi, E., 2004. Involvement of cell surface HSP90 in cell migration reveals a novel role in the developing nervous system. J. Biol. Chem. 279, 45379–45388. Temple, S., Raff, M.C., 1986. Clonal analysis of oligodendrocyte development in culture: evidence for a developmental clock that counts cell divisions. Cell 44, 773–779.