FKBP36 Is an Inherent Multifunctional Glyceraldehyde-3-phosphate ...

THE JOURNAL OF BIOLOGICAL CHEMISTRY VOL. 284, NO. 2, pp. 766 –773, January 9, 2009 © 2009 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in the U.S.A.

FKBP36 Is an Inherent Multifunctional Glyceraldehyde-3-phosphate Dehydrogenase Inhibitor*□ S

Received for publication, November 29, 2007, and in revised form, October 28, 2008 Published, JBC Papers in Press, November 10, 2008, DOI 10.1074/jbc.M709779200

Franziska Jarczowski, Gu¨nther Jahreis, Frank Erdmann, Angelika Schierhorn, Gunter Fischer1, and Frank Edlich2 From the Max Planck Research Unit for Enzymology of Protein Folding, Weinbergweg 22, D-06120 Halle/Saale, Germany

FKBP36 (FK506-binding protein 36; FKBP6) belongs to the human immunophilins and is predominantly found in testes. FKBP6 was initially identified as one of the genes on chromosome 7 whose deletion is linked to Williams-Beuren syndrome (1). Williams-Beuren syndrome is a dominant autosomal disorder, which may include supravalvular aortic stenosis, multiple peripheral pulmonary arterial stenoses, elfin face, mental and statural deficiency, characteristic dental malformation, and infantile hypercalcemia (2). FKBP36 is predominantly expressed in testes and plays an important role in spermatogenesis. The loss of FKBP36 causes a significant reduction of the size of testes and the infertility of male mice. The testis-specific FKBP was identified as a component of the synaptonemal complex, which forms between two homologous chromosomes during meiosis (3). The synaptonemal complex participates in chromosome pairing, synapsis, and recombination (4). FKBP36 contains an N-terminal peptidyl-prolyl cis/trans isomerase domain and three tetratricopeptide repeat (TPR)3

* This work was supported by Sonderforschungsbereich 604 of the Deutsche Forschungsgemeinschaft. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. □ S The on-line version of this article (available at http://www.jbc.org) contains supplemental Figs. 1–3. 1 To whom correspondence may be addressed. Tel.: 345-55-22800; Fax: 34555-11972; E-mail: [email protected]. 2 To whom correspondence may be addressed: Surgical Neurology Branch, NINDS, National Institutes of Health, Porter Neuroscience Research Center, 35 Convent Dr., MSC 3704, Bethesda, MD 20892-3704. Tel.: 301-451-1717; Fax: 301-402-0380; E-mail: [email protected]. 3 The abbreviations used are: TPR, tetratricopeptide repeat; GA-3-P, glyceraldehyde 3-phosphate; GST, glutathione S-transferase; FKBP, FK506-binding

766 JOURNAL OF BIOLOGICAL CHEMISTRY

motifs in its C-terminal segment. Furthermore, FKBP36 displays a high sequence homology to the regulated FKBP38, and the absence of FKBP36 activity might suggest a similar activation scenario (5). The TPR domains of FKBP38, FKBP51, and FKBP52 have been shown to interact with the heat shock protein Hsp90 (5, 6). The interactions between the different FKBPs and Hsp90 are mediated by the Hsp90 C90 domain. Complexes of FKBP51 and FKBP52 with Hsp90 have been reported to interact with client molecules, e.g. steroid hormone receptors, to modify the activity, stability, or subcellular localization of the client proteins (7). Within these complexes, the FKBPs interact with both the receptor and Hsp90 (8). In the present study, we show that FKBP36 interacts with Hsp90 via the TPR domain. A second binding site in the TPR domain of FKBP36 mediates the interaction with GAPDH, leading to GAPDH-FKBP36-Hsp90 complexes, which are comparable to complexes formed by other TPR-containing immunophilins. Furthermore, our experiments showed that FKBP36 controls GAPDH activity either by direct interference with binding of the coenzyme NAD⫹ or down-regulation of GAPDH levels. Hence, FKBP36 represents a novel non-metabolic and multifunctional inhibitor of GAPDH.

EXPERIMENTAL PROCEDURES Sources of enzymes used in the experiments, recombinant human FKBP36 and recombinant human Hsp90 C90 (Hsp90627–731), were recombinantly expressed by using a pET28a vector in Escherichia coli RosettaTM cells. Glutathione S-transferase (GST) fusion FKBP36 variants were expressed by using a pGEX4T1 vector in E. coli RosettaTM cells. The C-terminal Hsp90 peptides Hsp90C20 (Hsp90712–731, PDEIPPLEGDEDASRMEEVD), Hsp90C7 (Hsp90725–731, SRMEEVD), and Hsp90C5 (Hsp90727–731, MEEVD) were produced by solidphase synthesis using the Fmoc (N-(9-fluorenyl)methoxycarbonyl) strategy. Rabbit muscle GAPDH was purchased from Sigma. The FKBP36 antibody was a section 4 polyclonal antibody from rabbit against purified FKBP36. Additional antibodies used were polyclonal rabbit anti-actin (Sigma), monoclonal mouse antiHis5 (Qiagen), monoclonal mouse anti-GST (Sigma), monoclonal rat anti-Hsp90 (Stressgen), polyclonal rabbit antiGAPDH (Sigma), and polyclonal rabbit anti-H3 (Cell Signaling Technology, Danvers, MA). MG115, NAD⫹, and glyceraldehyde 3-phosphate (GA-3-P) were obtained from Sigma. protein; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; MALDITOF, matrix-assisted laser desorption ionization time-of-flight.

VOLUME 284 • NUMBER 2 • JANUARY 9, 2009

Downloaded from http://www.jbc.org/ by guest on January 2, 2016

FKBP36 has been previously shown to be a crucial factor in spermatogenesis because of its interplay with the synaptonemal complex protein SCPI. Here we show that beyond this function, FKBP36 forms complexes with glyceraldehyde-3-phosphate dehydrogenase (GAPDH; EC 1.2.1.12) and Hsp90. Both proteins bind independently to different sites of the FKBP36 tetratricopeptide repeat domain. The interaction between FKBP36 and GAPDH directly inhibits the catalytic activity of GAPDH. In addition, FKBP36 expression causes a significant reduction of the GAPDH level and activity in COS-7 cells. Particularly in the cytosolic fraction, GAPDH was depleted by FKBP36 expression. Thus, FKBP36 diminishes GAPDH activity by direct interaction and down-regulation of GAPDH, which represents a previously unknown mechanism of GAPDH regulation and a novel function of FKBP36 in testis-specific signaling.

FKBP36 Is a GAPDH Inhibitor

JANUARY 9, 2009 • VOLUME 284 • NUMBER 2

FIGURE 1. FKBP36 interacts with GAPDH and Hsp90 from rat testis lysate. A, FKBP36-interacting proteins from rat testis cells were identified by incubation of rat testis cell lysate with GST-FKBP36- or GST-pre-loaded GSH-Sepharose. Samples (15 ␮g) of bound proteins were separated and visualized by Coomassie-stained SDS-PAGE (12.5%). Endogenous rat proteins identified by MALDI-TOF analysis are labeled in the right. Molecular mass according to the migration of the protein marker is shown on the left. B, shown are the results from Western blot analysis of proteins from rat-testes lysate bound to a Streptag II-FKBP36-loaded StrepTactin matrix using anti-Hsp90 and anti-GAPDH antibodies. The StrepTactin matrix alone served as a control.

brane fractions. The preparation of rat testis cytoplasm was described previously (10).

RESULTS FKBP36 Interacts with GAPDH and Hsp90—The TPR domains of several immunophilins have been shown to mediate interactions with Hsp90 (7). Hence, we tested whether FKBP36 is able to interact with Hsp90 as well. We performed a GST pulldown experiment loading GSTFKBP36 on GSH-Sepharose and applying rat testis lysate to the matrix. Several endogenous proteins from rat testis cells bound to the FKBP36 affinity matrix, Hsp90, Hsp70, tubulin, actin, GAPDH, and GST (Fig. 1A), as identified by MALDI-TOF mass spectrometry (data not shown). However, Hsp70, tubulin, and actin were also found in the control experiment using only a GST matrix. Thus, only Hsp90 and GAPDH can be considered as specific binders of FKBP36. To confirm these results, a second interaction assay was performed using FKBP36 with Strep-Tag II fusion immobilized on a StrepTactin matrix. The FKBP36-affinity matrix was incubated with rat-testis lysate in parallel to the previous experiment, and bound proteins were analyzed by Western blotting with anti-GAPDH and anti-Hsp90 antibodies (Fig. 1B). GAPDH and Hsp90 only interacted with the FKBP36 affinity matrix, whereas no binding was observed to the matrix alone. JOURNAL OF BIOLOGICAL CHEMISTRY

767


GAPDH Activity Assay—GAPDH activity was measured with an HP 8452A spectrophotometer by following the reduction of NAD⫹ at 340 nm. The reaction mixture in a volume of 1.5 ml contained 10 mM sodium arsenate, 2.5 mM EDTA, 2.5 mM dithiothreitol, 150 mM NaCl (pH 8.5). Unless mentioned otherwise, 25 ␮l of COS-7 crude cell extract, 1 mM NAD⫹, and 1 mM GA-3-P were used in all assays. One unit of enzyme activity is expressed as the amount of enzyme that forms 1 ␮mol of NADH/min at 25 °C. Protein Interaction Assays—For GSH-Sepharose-binding assay, 30-␮l volume of GSH-Sepharose was pre-equilibrated in 25 mM Tris-HCl buffer (pH 7.5) 150 mM NaCl, 1 mM dithiothreitol and saturated with GST or GST fusion proteins. The beads were then washed twice with buffer B (25 mM Tris-HCl, (pH 7.5), 150 mM NaCl, 1 mM dithiothreitol) and subsequently incubated for 2 h with 100 ␮l of rat testis cytoplasm. In some experiments, 50 ␮M Hsp90 C90 or 1 mM NAD⫹, NADH ⫹ H⫹, GA-3-P, or arsenate was added. Beads were centrifuged, washed three times, and mixed with an equal volume of Laemmli sample buffer. Samples were then subjected to 12.5% SDS-PAGE and analyzed by Western blotting using monoclonal mouse anti-His5 (Pharmingen), monoclonal mouse antiGST, monoclonal rat anti-Hsp90 (Stresgen), and polyclonal rabbit anti-GAPDH antibody. Co-immunoprecipitation was performed with 5 ml of rat testis cytosol (pre-cleaned with 200 ␮l of Protein G-SepharoseTM 4 Fast Flow for 1 h at 4 °C). The samples (3 ⫻ 1.5 ml) were subjected to 5 ␮g of polyclonal goat anti-Hsp90 (Santa Cruz Biotechnology) and 5 ␮g of polyclonal goat anti-GAPDH antibody (Santa Cruz Biotechnology) or buffer, respectively. After a 3-h pre-incubation at 4 °C, Protein G SepharoseTM 4 Fast Flow (50 ␮l) (GE Healthcare) was added and incubated for 2 h at 4 °C. Subsequently, samples were centrifuged at 12,000 ⫻ g for 1 min, washed five times with 200 ␮l of buffer B, boiled with 50 ␮l of sample buffer, subjected to SDS-PAGE, and analyzed by Western blotting. Cell Culture—For FKBP36 expression, transformed African green monkey kidney fibroblast cells (COS-7 line) (Deutsche Sammlung von Mikroorganismen und Zellkulturen, Braunschweig, Germany) were used. The cells were cultured in Dulbecco’s modified Eagle’s medium (Biochrom, Berlin, Germany) supplemented with 2 mM L-glutamine and 10% (v/v) heat-inactivated fetal calf serum in a humidified incubator at 37 °C in 10% (v/v) CO2. COS-7 cells were transfected with pcDNA3.1 or pcDNA-FKBP36 and seeded at 5 ⫻ 106 cells/ml in 6-well plates. To analyze GAPDH proteasomal degradation, COS-7 cells were transfected with either pcDNA3.1 or pcDNA-FKBP36 constructs. After 48 h, cells were treated with 0, 25, and 50 ␮M MG115 for 5 h. Subsequently, cells were harvested and analyzed by Western blot. To prepare COS-7 crude cell extract, cells were harvested and centrifuged for 10 min at 2,000 ⫻ g, and cell pellets were resuspended in the same volume of hypotonic lysis buffer (50 mM Tris-HCl, (pH 7.4), 20 mM EDTA, 1 mM dithiothreitol, and 1% Triton X-100), followed by sonication. The crude extract was centrifuged for 10 min at 800 ⫻ g. The pellet was used for the preparation of nuclei (9). The supernatant was centrifuged again for 45 min at 100,000 ⫻ g to prepare cytosol and mem-




The FKBP36 TPR Domain Binds GAPDH and Hsp90—To test whether FKBP36 binds directly to GAPDH, we made use of purified GAPDH from rabbit muscle, which was previously utilized in several similar studies (11–13). Isolated GAPDH was incubated with the GST-FKBP36 affinity matrix. Fig. 2A shows that GAPDH bound specifically to FKBP36, demonstrating direct interaction between both proteins. Next, several GST-FKBP36 variants were applied to GSHSepharose and incubated with rat testis lysate to investigate which segment of FKBP36 mediates the interaction with GAPDH. Western blot analysis using anti-GAPDH antibody showed that GAPDH bound only to the full-length FKBP36 and to the FKBP36145–327 variant containing the three TPR motifs (Fig. 2B) and a C-terminal segment. A similar interaction pattern was observed for Hsp90 using the different FKBP36 variant affinity matrices. Thus, the C-terminal FKBP36 segment, which contains the three TPR motifs, interacts with both Hsp90 and GAPDH. Further experiments using a FKBP36 variant lacking the C-terminal extension were carried out to investigate the contribution of the FKBP36 C terminus to the interaction with either GAPDH or Hsp90. Fig. 2C shows that both GAPDH and Hsp90 bound similarly to the FKBP36 variants independent of the presence of the C-terminal FKBP36 extension. Hence, only the FKBP36 segment that contains the TPR motifs mediates the interactions with GAPDH and Hsp90. The Hsp90 MEEVD Segment and GAPDH Interact with Different Binding Sites in the FKBP36 TPR Domain—Given the interaction between the FKBP36 TPR domain and Hsp90, we tested if this interaction is mediated by the C90 domain of Hsp90 in parallel to other immunophilin-Hsp90 complexes. To this end, we incubated FKBP36 with either a C90-GST affinity matrix or a GST matrix (Fig. 3A). Indeed, the C90 domain sufficed to interact with FKBP36. Hence, the interaction between FKBP36 and Hsp90 is mediated by the TPR domain and the C90 domain similar to other FKBP-Hsp90 complexes. Furthermore, we investigated using a FKBP36 affinity matrix which C-terminal segments of Hsp90 can compete with endogenous rat Hsp90 and thus mediate the interaction with FKBP36. The results in Fig. 3B show that the different segments of the C90 domain including the very C-terminal MEEVD motif competed with endogenous Hsp90 from testis lysate for FKBP36 binding. Thus, the interaction between Hsp90 and FKBP36 is mediated by the C-terminal MEEVD motif of Hsp90, which is similar to previously described interactions between Hsp90 and human FKBPs. Far-UV CD spectroscopic measurements show changes in the secondary structure elements of either FKBP36 or the C90 domain because of complex formation (Fig. 3C). Interestingly, these changes of the protein ellipticity in the far-UV CD spectra were not observed for interactions between FKBP36 and the MEEVD peptide, suggesting either structural changes in the C90 domain upon FKBP36 binding or additional binding sites between Hsp90 and FKBP36. Furthermore, we analyzed whether Hsp90 and GAPDH compete for the same interaction site in the FKBP36 TPR domain or perhaps can be found together in a complex with FKBP36. Thereto, we applied purified proteins to a FKBP36

FIGURE 2. The FKBP36 TPR domain binds GAPDH and Hsp90. A, direct binding between FKBP36 and GAPDH was investigated by incubation of purified GAPDH from rabbit muscle with GST-FKBP36- or GST-pre-loaded GSHSepharose. The samples were analyzed by Coomassie-stained SDS-PAGE (12.5%). Samples without GAPDH served as a control. B, binding of GAPDH and Hsp90 to different domains of FKBP36 was tested by incubation of rat testis lysate with GSH-Sepharose pre-loaded with different GST-FKBP36 variants or GST alone. Bound proteins were analyzed by SDS-PAGE and Western blotting using anti-Hsp90 and anti-GAPDH antibodies. Eluted GST and GSTFKBP36 variants were analyzed by Western blotting using anti-GST antibodies (lower panel). C, endogenous GAPDH and Hsp90 from rat testis lysate bound to GSH-Sepharose loaded with GST-FKBP36 variants was analyzed by SDS-PAGE and Western blotting using anti-Hsp90 and anti-GAPDH antibodies. The Western blot analysis of eluted GST-FKBP36 variants using anti-GST antibodies is displayed in the lower panel.



Downloaded from http://www.jbc.org/ by guest on January 2, 2016 FIGURE 3. GAPDH and Hsp90 bind to different interaction sites in the FKBP36 TPR domain. A, binding of recombinant FKBP36 and FKBP52 to GSTHsp90C90- or GST-pre-loaded GSH-Sepharose was analyzed by Coomassie-stained SDS-PAGE (12.5%) of supernatant (S) and pellet (P). A sample without FKBP served as a control. B, interaction between endogenous Hsp90 from rat testis lysate and FKBP36-pre-loaded GSH-Sepharose in the presence of different segments derived from the Hsp90 C terminus was analyzed by Western blotting using anti-Hsp90 antibodies. Loaded GST-FKBP36 was analyzed by Western blotting using anti-GST antibodies (lower panel). C, far-UV CD spectrum of FKBP36 and the Hsp90 C90 domain either separated (E) or mixed (F) in 10 mM Tris-HCl buffer (pH 7.5). Changes in the minimum at 208.5 nm are displayed in the inlay. mdpg; millidegrees. D, the influence of GAPDH and Hsp90 on the interactions between FKBP36 and both proteins was examined by incubating rat-testes cell lysate with GST-FKBP36-loaded GSH-Sepharose in the absence or presence of Hsp90 C90 or purified GAPDH from rabbit muscle cells. Input and eluate were subjected to SDS-PAGE and subsequently analyzed by Western blotting using anti-Hsp90 and anti-GAPDH antibodies. The presence of Hsp90 C90 was analyzed using anti-His5 antibodies. E, the presence of GAPDH-FKBP36Hsp90 complexes was studied by co-immunoprecipitation of endogenous proteins from rat testis cytosol using goat anti-Hsp90 or goat anti-GAPDH antibodies. Antibody-protein complexes were bound to Protein G Sepharose. After washing precipitates, input and samples were subjected to SDS-PAGE and analyzed by Western blotting using rat anti-Hsp90 antibody, rabbit anti-GAPDH antibody, and rabbit anti-FKBP36 antibody. As a control, cell lysate pre-incubated with Protein G-Sepharose was used. IP, immunoprecipitate.

affinity matrix that was incubated with testis lysate (Fig. 3D). The presence of additional C90 domain had no effect on the binding of GAPDH to the FKBP36 affinity matrix. Further addition of purified GAPDH (5 ␮M) did not influence the FKBP36/ Hsp90 interaction either. The lack of competition between Hsp90 and GAPDH for the TPR domain binding implies different interaction sites for both proteins in FKBP36. JANUARY 9, 2009 • VOLUME 284 • NUMBER 2

To analyze the presence of protein complexes containing FKBP36, GAPDH, and Hsp90 in spermatocytes, we performed a co-immunoprecipitation using cytosol from rat testis cells (Fig. 3E). Endogenous FKBP36 bound either Hsp90 or GAPDH, and in addition, Hsp90 was detected in the fraction of GAPDHinteracting proteins and vice versa, suggesting the presence of GAPDH-FKBP36-Hsp90 complexes in testis cells. FurtherJOURNAL OF BIOLOGICAL CHEMISTRY

769




more, we incubated cell lysate of COS-7 cells either expressing FKBP36 or lacking the human FKBP with a GST-Hsp90 C90 affinity matrix (supplemental Fig. 1). GAPDH only interacted with the C90 affinity matrix in the presence of FKBP36, demonstrating the requirement of FKBP36 for the formation of the trimeric complex. FKBP36 Inhibits GAPDH Activity—Next, GAPDH activity measurements were performed in the presence of FKBP36 to investigate the role of FKBP36 in complexes with GAPDH. FKBP36 had only a slight effect on GAPDH activity at 1 mM NAD⫹, reducing the residual activity to about 90% compared with the absence of FKBP36 (Fig. 4A). FKBP36-mediated GAPDH inhibition was not influenced by GA-3-P concentrations in the range of 0.01–1 mM. However, the inhibitory effect of FKBP36 on GAPDH activity was increased significantly at lower NAD⫹ concentrations of 0.01 mM. Under these conditions, FKBP36 reduced the GAPDH activity in a concentrationdependent manner to less than 40% residual activity at 2.5 ␮M FKBP36 with a Ki of 1.5 ⫾ 0.45 ␮M. Thus, we analyzed whether FKBP36 competes with NAD⫹ for GAPDH binding. The EadieHofstee plot (Fig. 4B) shows deviations from the linear function, which are caused by negatively cooperative NAD⫹ binding at four binding sites in a GAPDH tetramer (14 –16). The FKBP36-mediated GAPDH inhibition differs from the pattern of a competitive inhibitor by indicating different Vmax, and thus indicates a mixed form of inhibition that is not solely caused by competition with NAD⫹. Interactions between FKBP36 and NAD⫹ can be excluded because the fluorescence spectra of both FKBP36 and NAD⫹ remain unaltered in the presence of each other (supplemental Figs. 2 and 3). Furthermore, the interference of GAPDH substrates and effectors with the interaction between GAPDH and FKBP36 was tested using the FKBP36 affinity matrix (Fig. 4C). In the presence of NAD⫹ or NADH ⫹ H⫹, the amount of FKBP36-bound GAPDH was significantly reduced, which corresponds to the kinetic measurements. However, neither GA-3-P nor arsenate influenced the FKBP36/ GAPDH interaction, demonstrating interference only between FKBP36/GAPDH interactions and NAD⫹ binding. FKBP36 Reduces GAPDH Levels and Activity in Cells—Next, we studied the influence of FKBP36 on endogenous GAPDH activity to get inside the physiological effect of this interaction. Thus, the testis-derived cell lines NTERA-2, F9, TM3, and TM4 were analyzed for FKBP36 and GAPDH expression levels. However, the FKBP36 levels in the different cell lines were significantly lower than in testis lysate. In the absence of a cell system reflecting the physiological FKBP36 concentration, COS-7 cells were, thus, transfected with a pcDNA-FKBP36 construct, and the GAPDH activity in transfected cells was compared with that in normal cells. Fig. 5A shows that FKBP36 reduced the GAPDH activity by about 70 units/g of total cell protein, which is 32% of the GAPDH activity in these cells. To test whether the reduction of GAPDH activity is caused by an altered GAPDH concentration or by direct inhibition because of FKBP36 binding, the GAPDH protein levels were analyzed. Interestingly, the presence of FKBP36 reduced the GAPDH protein level by about 33% in COS-7 cells based on the densitometric analysis, which is likely to be the major cause of the reduction of GAPDH activity (Fig. 5, B and C). Furthermore,

FIGURE 4. GAPDH is inhibited by FKBP36. A, activity measurements of GAPDH (60 nM) at various FKBP36 concentrations (0.25–2.5 ␮M). GAPDH activity is displayed in percent residual activity at 1 mM NAD⫹ and 1 mM GA-3-P (E), 1 mM NAD⫹ and 0.01 mM GA-3-P (F), and 0.01 mM NAD⫹ and 1 mM GA-3-P (Œ). B, Eddie-Hofstee plot of residual GAPDH (1.2 ␮g/ml) activity and NAD⫹ concentration in the presence of 0 ␮M (F), 0.7 ␮M (E), 1.4 ␮M (), and 2.8 ␮M (ƒ) FKBP36. The measurements were carried out in the presence of 1 mM GA-3-P. Linear regression for measurements at 10, 25, 50, and 100 ␮M NAD⫹ is displayed as lines. C, binding of endogenous GAPDH from rat testis cell lysate to GST-FKBP36 loaded onto GSH-Sepharose in the presence of 1 mM NAD⫹, NADH ⫹ H⫹, GA-3-P, or arsenate. The GAPDH binding was analyzed by Western blotting using anti-GAPDH antibodies. The amount of FKBP36 loaded onto the matrix was analyzed by anti-GST antibodies.

COS-7 cells expressing FKBP36 were fractionized, and the GAPDH localization was compared with the pattern observed in wild-type cells (Fig. 5D). GAPDH was found in the cytosol and the membrane fraction of COS-7 cells both in presence and absence of FKBP36. However, FKBP36 expression significantly reduced the GAPDH levels in both cell fractions. In fact, the VOLUME 284 • NUMBER 2 • JANUARY 9, 2009


Downloaded from http://www.jbc.org/ by guest on January 2, 2016 FIGURE 5. FKBP36 expression reduces GAPDH protein levels in COS-7 cells. A, GAPDH activity measurements using cell lysate of either untransfected COS-7 cells or COS-7 cells transfected with pcDNA3.1 or pcDNA-FKBP36 constructs, respectively. Data were statistically analyzed by a non-paired Student’s t test and are represented as mean ⫾ S.D. from six replicates. *, p ⬍ 0.05. B, Western blot analysis of proteins present in COS-7 cells either not transfected or transfected with pcDNA3.1 or pcDNA-FKBP36 constructs, using anti-GAPDH, anti-actin, and anti-FKBP36 antibodies. C, densitometric analysis of the GAPDH levels of the Western blot analysis depicted in B. GAPDH levels are displayed as percent of endogenous protein found in untransfected COS-7 cells. Data are represented as mean ⫾ S.D. from four replicates. *, p ⬍ 0.05. D, Western blot analysis of a cell fractionation of COS-7 cells were either transfected with pcDNA3.1 or pcDNA3.1 FKBP36 construct using anti-FKBP36, anti-GAPDH, and anti-histone antibodies. E, influence of FKBP36 on proteasomal GAPDH degradation analyzed by treatment of pcDNA3.1 or pcDNA-FKBP36-transfected COS-7 cells with 0, 25, or 50 ␮M MG115 for 5 h and subsequent analysis of crude cell extracts by anti-GAPDH, anti-FKBP36, and anti-␤ actin antibodies.


JOURNAL OF BIOLOGICAL CHEMISTRY

771

FKBP36 Is a GAPDH Inhibitor cytosolic GAPDH concentration seems to be affected in particular by FKBP36 expression. GAPDH was previously found to be degraded by the proteasome (17). Thus, we analyzed if the presence of FKBP36 increases GAPDH degradation using the proteasome inhibitor MG115. Fig. 5E shows a significant reduction of GAPDH levels in the presence of FKBP36. However, this reduction was not prevented by 50 ␮M MG115. On the other hand, FKBP36 was found at slightly higher levels in the presence of the proteasome inhibitor. Thus, FKBP36 does not increase the GAPDH degradation rate, but is likely reduces GAPDH expression.


Acknowledgments—We are grateful to Melanie Kirchner, Martina Heidler, and Beate Rappsilber for excellent technical assistance and Corinna Brandsch and Reiner Klinger for the support of animal material.



DISCUSSION FKBP36 is a multifunctional immunophilin, which has been reported to interact with components of the synaptonemal complex. Here, we show interactions of this testis-specific FKBP with Hsp90 and GAPDH, suggesting additional cellular functions. The interaction between FKBP36 and the C-terminal domain of Hsp90 is mediated by its TPR domain (Figs. 2B and 3A) and is thus similar to interactions between Hsp90 and other immunophilins (e.g. FKBP52) (7). The FKBP52 residues Lys282, Lys354, and Arg358 were reported to assemble a dicarboxylate clamp, which binds the Hsp90 C-terminal MEEVD motif to form the FKBP52-Hsp90 complex (18). FKBP36 features identical residues in the corresponding positions, and the MEEVD pentapeptide was sufficient for the interaction with FKBP36 (Fig. 3C), implying a similar binding mechanism. Changes in the far-UV CD spectrum (Fig. 3C) suggest either additional binding sites between FKBP36 and Hsp90 or structural changes in the C90 domain upon FKBP36 binding. However, the competition between the MEEVD peptide suggests that the latter possibility is more likely. GAPDH also binds to the TPR domain of FKBP36, but GAPDH and Hsp90 interact with different sites in FKBP36 because both proteins do not compete for FKBP36 binding (Fig. 3D). In fact, the results of binding studies with rat-testis proteins and COS-7 cell lysate containing or lacking FKBP36 suggest the presence of the GAPDH-FKBP36-Hsp90 complexes in the cytosol, where all three proteins are present. These interactions occurred with recombinant human proteins, purified rabbit GAPDH, or endogenous rat proteins, suggesting similar interactions in different mammalian organisms perhaps based on the high similarity of the interaction partners in different mammals. Such GAPDH-FKBP36-Hsp90 complexes might be similar to steroid hormone receptor complexes. Thus, the FKBP36-mediated GAPDH control is perhaps comparable with the reduction of steroid hormone receptor activity induced by high cellular FKBP51 concentrations (19), but presents a unique mechanism, because GAPDH, in contrast to receptor molecules, is not a client protein of Hsp90 and direct interactions between both proteins were not observed. Nevertheless, Hsp90 might interact with GAPDH in complex with FKBP36 or recruit further interaction partners that could participate in the regulation of GAPDH activity. Endothelial nitric-oxide synthase , for instance, is activated by Hsp90 and S-nitrosylation at Cys149 and is known to inhibit GAPDH activity (20). FKBP36 also participates in the control of GAPDH in an Hsp90-independent manner via direct interaction with the

NAD⫹-binding site of GAPDH. Even though only minor FKBP36-mediated effects on GAPDH activity were observed at NAD⫹ saturation, significant GAPDH inhibition was found at lower NAD⫹ concentrations. Similar effects are caused by the interaction between GAPDH and adenine nucleotides or RNA (21). Based on our results, FKBP36-mediated GAPDH inhibition might play an important role in spermatocytes because the NAD⫹ levels in spermatocytes are significantly lower than in COS-7 cells and comparable with low NAD⫹ concentrations in the kinetic measurements (Fig. 4) because of lactate-based nutrition (22). FKBP36 expression furthermore reduced the GAPDH concentration in COS-7 cells (Fig. 5), demonstrating a control of GAPDH protein levels. GAPDH is likely diminished by FKBP36-mediated reduction of GAPDH expression because the use of MG115 showed no involvement of proteasomal degradation in reduced GAPDH levels. Similar down-regulation of GAPDH levels have been reported previously. However, formerly identified regulators of GAPDH expression are nutrients (e.g. amino acids and ethanol) or hormones such as insulin (21, 23, 24). FKBP36, by contrast, is the first described interaction partner that affects the GAPDH expression level. Due to the FKBP36-mediated reduction of GAPDH expression, the GAPDH activity in COS-7 cell lysate was diminished. Moreover, the FKBP36 expression changed the subcellular GAPDH distribution by reducing the cytosolic GAPDH population in particular, which might potentiate the FKBP36-mediated GAPDH inhibition in the cytosol of spermatocytes. Hence, FKBP36 controls GAPDH activity in a combination of direct inhibition and reduction of the GAPDH concentration, particularly in the cytosol. Previous reports also describe a testis-specific GAPDH isoform, which is expressed in late spermatid stages (25, 26). However, glycerinaldehyde-3-phosphate dehydrogenase, spermatogenic (GAPDS) isoform is not regulated by FKBP36 because the FKBP is only present in early stages of spermatocytes, not in spermatids (3). GAPDH participates in glycolysis by catalyzing the oxidative phosphorylation of GA-3-P into 1,3-diphosphoglycerate. GAPDH is expressed ubiquitously, but the cellular concentrations vary significantly from cell type to cell type (27). Interestingly, particularly low GAPDH concentrations were identified in testes (28). Thus, only in spermatogenic cells, GAPDH is limiting for the glycolysis (29). The interaction between GAPDH and the testis-specific FKBP36 may participate in glycolysis and several other processes because GAPDH was found to promote membrane-membrane fusion processes, modulate the cytoskeleton, and be involved in apoptosis control, DNA repair, DNA replication, and tRNA export (30 –33). On the basis of the bifunctional reduction of GAPDH activity by FKBP36 and the expression pattern of both proteins, an influence of FKBP36 on GAPDH-dependent processes in testes is likely and might be involved in the regulation of physiological processes that depend on GAPDH activity.

FKBP36 Is a GAPDH Inhibitor REFERENCES


17. Buchczyk, D. P., Grune, T., Sies, H., and Klotz, L. O. (2003) Biol. Chem. 384, 237–241 18. Wu, B., Li, P., Liu, Y., Lou, Z., Ding, Y., Shu, C., Ye, S., Bartlam, M., Shen, B., and Rao, Z. (2004) Proc. Natl. Acad. Sci. U. S. A. 101, 8348 – 8353 19. Scammell, J. G., Denny, W. B., Valentine, D. L., and Smith, D. F. (2001) Gen. Comp. Endocrinol. 124, 152–165 20. Mohr, S., Stamler, J. S., and Brune, B. (1996) J. Biol. Chem. 271, 4209 – 4214 21. Banerjee, K., Mohr, L., Wands, J. R., and de la Monte, S. M. (1998) Alcohol. Clin. Exp. Res. 22, 2093–2101 22. Nakamura, M., Okinaga, S., and Arai, K. (1984) Biol. Reprod. 30, 1187–1197 23. Alexander, M. C., Lomanto, M., Nasrin, N., and Ramaika, C. (1988) Proc. Natl. Acad. Sci. U. S. A. 85, 5092–5096 24. Claeyssens, S., Gangneux, C., Brasse-Lagnel, C., Ruminy, P., Aki, T., Lavoinne, A., and Salier, J. P. (2003) Am. J. Physiol. 285, G840 –G849 25. Welch, J. E., Brown, P. L., O’Brien, D. A., Magyar, P. L., Bunch, D. O., Mori, C., and Eddy, E. M. (2000) J. Androl. 21, 328 –338 26. Welch, J. E., Barbee, R. R., Magyar, P. L., Bunch, D. O., and O’Brien, D. A. (2006) Mol. Reprod. Dev. 73, 1052–1060 27. Barber, R. D., Harmer, D. W., Coleman, R. A., and Clark, B. J. (2005) Physiol. Genomics 21, 389 –395 28. Piechaczyk, M., Blanchard, J. M., Marty, L., Dani, C., Panabieres, F., El Sabouty, S., Fort, P., and Jeanteur, P. (1984) Nucleic Acids Res. 12, 6951– 6963 29. Nakamura, M., Okinaga, S., and Arai, K. (1984) Andrologia 16, 446 – 450 30. Glaser, P. E., and Gross, R. W. (1995) Biochemistry 34, 12193–12203 31. Meyer-Siegler, K., Mauro, D. J., Seal, G., Wurzer, J., deRiel, J. K., and Sirover, M. A. (1991) Proc. Natl. Acad. Sci. U. S. A. 88, 8460 – 8464 32. Singh, R., and Green, M. R. (1993) Science 259, 365–368 33. Zheng, L., Roeder, R. G., and Luo, Y. (2003) Cell 114, 255–266

JOURNAL OF BIOLOGICAL CHEMISTRY

773


1. Meng, X., Lu, X., Morris, C. A., and Keating, M. T. (1998) Genomics 52, 130 –137 2. Tassabehji, M. (2003) Hum. Mol. Genet. 12, R229 –R237 3. Crackower, M. A., Kolas, N. K., Noguchi, J., Sarao, R., Kikuchi, K., Kaneko, H., Kobayashi, E., Kawai, Y., Kozieradzki, I., Landers, R., Mo, R., Hui, C. C., Nieves, E., Cohen, P. E., Osborne, L. R., Wada, T., Kunieda, T., Moens, P. B., and Penninger, J. M. (2003) Science 300, 1291–1295 4. Page, S. L., and Hawley, R. S. (2003) Science 301, 785–789 5. Edlich, F., Erdmann, F., Jarczowski, F., Moutty, M. C., Weiwad, M., and Fischer, G. (2007) J. Biol. Chem. 282, 15341–15348 6. Pirkl, F., and Buchner, J. (2001) J. Mol. Biol. 308, 795– 806 7. Pratt, W. B., Galigniana, M. D., Harrell, J. M., and DeFranco, D. B. (2004) Cell. Signal. 16, 857– 872 8. Smith, D. F., Baggenstoss, B. A., Marion, T. N., and Rimerman, R. A. (1993) J. Biol. Chem. 268, 18365–18371 9. Blobel, G., and Potter, V. R. (1966) Science 154, 1662–1665 10. Zakeri, Z. F., and Wolgemuth, D. J. (1987) Mol. Cell. Biol. 7, 1791–1796 11. Kragten, E., Lalande, I., Zimmermann, K., Roggo, S., Schindler, P., Muller, D., van Oostrum, J., Waldmeier, P., and Furst, P. (1998) J. Biol. Chem. 273, 5821–5828 12. Olah, J., Tokesi, N., Vincze, O., Horvath, I., Lehotzky, A., Erdei, A., Szajli, E., Medzihradszky, K. F., Orosz, F., Kovacs, G. G., and Ovadi, J. (2006) FEBS Lett. 580, 5807–5814 13. Laschet, J. J., Minier, F., Kurcewicz, I., Bureau, M. H., Trottier, S., Jeanneteau, F., Griffon, N., Samyn, B., Van Beeumen, J., Louvel, J., Sokoloff, P., and Pumain, R. (2004) J. Neurosci. 24, 7614 –7622 14. Nagradova, N. K. (2001) Biochemistry (Mosc.) 66, 1067–1076 15. Meunier, J. C., and Dalziel, K. (1978) Eur. J. Biochem. 82, 483– 492 16. Henis, Y. I., and Levitzki, A. (1980) Proc. Natl. Acad. Sci. U. S. A. 77, 5055–5059

pcDNAFKBP36

pcDNA3.1

Supplementary Figure 1

input

GAPDH

elution input

FKBP36

elution

anti-GST

The formation of FKBP36/GAPDH/Hsp90C90 complexes in cell lysate was analysed by incubating Cos7 cell lysate either expressing or lacking FKBP36 with a GST-Hsp90C90 preloaded GSH-sepharose. Input and eluate were subjected to SDS-PAGE and subsequently analysed by Western blot with anti-FKBP36 and anti-GAPDH antibodies. The loaded Hsp90 C90 was analyzed using anti-GST antibodies.

Supplementary Figure 2 2,5e+6

-1

flourecence (counts x s )

2,0e+6

1,5e+6

1,0e+6

5,0e+5

0,0 300

320

340

360

380

400

420

440

wavelength (nm)

Fluorescence measurements of 1 µM FKBP36 (---) and 100 µM NAD+ (○○○) separated and mixed (__) were performed at an excitation wavelength of 280 nm. In addition, the calculated spectrum (...) represents the sum of the separated components spectra.

Supplementary Figure 3 0,10

0,08

AU

0,06

0,04

0,02

0,00 300

320

340

360

wavelength (nm)

380

400

+

UV/Vis measurements of 14.8 µM FKBP36 (●) and 50 µM NAD (▲) separated and mixed (○) in the range between 300 and 400 nm.

Enzyme Catalysis and Regulation: FKBP36 Is an Inherent Multifunctional Glyceraldehyde-3-phosphate Dehydrogenase Inhibitor Franziska Jarczowski, Günther Jahreis, Frank Erdmann, Angelika Schierhorn, Gunter Fischer and Frank Edlich J. Biol. Chem. 2009, 284:766-773. doi: 10.1074/jbc.M709779200 originally published online November 10, 2008

Find articles, minireviews, Reflections and Classics on similar topics on the JBC Affinity Sites. Alerts: • When this article is cited • When a correction for this article is posted Click here to choose from all of JBC's e-mail alerts

Supplemental material: http://www.jbc.org/content/suppl/2008/11/11/M709779200.DC1.html This article cites 33 references, 18 of which can be accessed free at http://www.jbc.org/content/284/2/766.full.html#ref-list-1


Access the most updated version of this article at doi: 10.1074/jbc.M709779200