NovelInteractionofClassIIbHistoneDeacetylase6(HDAC6 ... · 2015-05-22 · may29,2015•volume290•number22 journalofbiologicalchemistry 14045 cell movement via acetylation of -tubulin

Novel Interaction of Class IIb Histone Deacetylase 6 (HDAC6)with Class IIa HDAC9 Controls Gonadotropin ReleasingHormone (GnRH) Neuronal Cell Survival and Movement*

Received for publication, January 26, 2015, and in revised form, April 8, 2015 Published, JBC Papers in Press, April 14, 2015, DOI 10.1074/jbc.M115.640482

Smita Salian-Mehta‡, Mei Xu‡, Timothy A. McKinsey§, Stuart Tobet¶, and Margaret E. Wierman‡�1

From the ‡Division of Endocrinology, Metabolism, and Diabetes and §Division of Cardiology, University of Colorado School ofMedicine, Aurora, Colorado 80045, �Research Service Veterans Affairs Medical Center, Denver, Colorado 80220, and ¶Departmentof Biomedical Sciences, Colorado State University, Fort Collins, Colorado 80523

Background: HDACs expression is increased in differentiated GnRH neuronal cells.Results: HDAC9/6 promotes GnRH neuronal cell survival and halts neuron movement.Conclusion: Class II HDACs can modulate GnRH neuronal functions.Significance: This study identifies a novel interaction of HDAC9 with HDAC6 in GnRH neuronal cells.

The impact of histone deacetylases (HDACs) in the controlof gonadotropin releasing hormone (GnRH) neuronal devel-opment is unknown. We identified an increase in manyHDACs in GT1-7 (differentiated) compared with NLT (undif-ferentiated) GnRH neuronal cell lines. Increased HDAC9mRNA and protein and specific deacetylase activity in GT1-7cells suggested a functional role. Introduction of HDAC9 inNLT cells protected from serum withdrawal induced apopto-sis and impaired basal neuronal cell movement. Conversely,silencing of endogenous HDAC9 in GT1-7 cells increased apo-ptosis and cell movement. Comparison of WT and mutantHDAC9 constructs demonstrated that the HDAC9 pro-sur-vival effects required combined cytoplasmic and nuclearlocalization, whereas the effects on cell movement required acytoplasmic site of action. Co-immunoprecipitation demon-strated a novel interaction of HDAC9 selectively with theClass IIb HDAC6. HDAC6 was also up-regulated at themRNA and protein levels, and HDAC6 catalytic activity wassignificantly increased in GT1-7 compared with NLT cells.HDAC9 interacted with HDAC6 through its second catalyticdomain. Silencing of HDAC6, HDAC9, or both, in GT1-7 cellsaugmented apoptosis compared with controls. HDAC6 and -9had additive effects to promote cell survival via modulatingthe BAX/BCL2 pathway. Silencing of HDAC6 resulted in anactivation of movement of GT1-7 cells with induction inacetylation of �-tubulin. Inhibition of HDAC6 and HDAC9together resulted in an additive effect to increase cell move-ment but did not alter the acetylation of �tubulin. Together,these studies identify a novel interaction of Class IIa HDAC9with Class IIb HDAC6 to modulate cell movement and sur-vival in GnRH neurons.

During embryogenesis gonadotropin releasing hormone(GnRH)2 neurons migrate from the olfactory placode to theforebrain. A myriad of factors can influence the process ofGnRH neuronal development, and any failure in survival ormigration of these neurons can lead to incomplete or absentsexual maturation (1–3).

A growing number of studies suggest that neuronal develop-ment may be controlled epigenetically (4, 5). Epigeneticchanges are governed by a variety of enzymatic modificationsincluding DNA methylation and histone modification. Histonedeacetylases (HDACs) are a family of proteins initially de-scribed for their role in deacetylating histone tails acting asco-repressors to gene transcription. Whereas classes I HDACsare ubiquitous nuclear modulators of gene expression, theother HDAC classes (II-IV) have diverse, compartment-spe-cific roles in normal physiology and pathophysiologic states(6, 7).

Class II HDACs are subdivided into IIa (HDAC 4, 5, 7, and 9)and IIb (HDAC6 and -10) (8, 9). The N-terminal regulatorydomain mediates protein-protein interactions in the nucleuswith tissue-specific partners, and the C-terminal domain con-tains the deacetylase activity (10). Class IIa HDACs display tis-sue-specific expression and contribute to differentiation anddevelopment (11). HDAC9 is highly expressed in striated mus-cle, where it acts as a negative regulator of differentiation andgrowth (12–14), and in brain where it is linked to neuronalmorphogenesis (15). Based upon its sequence, HDAC9 is pre-dicted to have weak HDAC activity, often requiring a partner tomediate its effects as a transcriptional modulator (16). The abil-ity of Class IIa HDACs to shuttle from the nucleus to the cyto-plasm also suggests they may have additional roles beyond theirfunction as transcriptional co-repressors. Class IIb HDACsinclude a unique member, HDAC6, that has two-deacetylasedomains. HDAC6 is a predominant cytoplasmically localizedprotein that has been shown to control rates of cell death viatrafficking misfolded ubiquitinated proteins as well as foster

* This work was supported, in whole or in part, by National Institutes ofHealth Grant HD31191 (to M. E. W.), Grants HL116848 and AG043822 (toT. A. M.), and by the American Heart Association (Grant-in-Aid,14510001).

1 To whom correspondence should be addressed: Endocrinology MS8106,University of Colorado Denver,12801 East 17th Ave., RC1 South, Aurora, CO80045. Tel.: 303-724-3952; Fax: 303-724-3920; E-mail: [email protected].

2 The abbreviations used are: GnRH, gonadotropin releasing hormone; HDAC,histone deacetylase; IP, immunoprecipitate; qPCR, quantitative PCR; DC,double catalytic; HDRP, HDAC-related protein.

THE JOURNAL OF BIOLOGICAL CHEMISTRY VOL. 290, NO. 22, pp. 14045–14056, May 29, 2015Published in the U.S.A.

MAY 29, 2015 • VOLUME 290 • NUMBER 22 JOURNAL OF BIOLOGICAL CHEMISTRY 14045

by guest on Novem

ber 13, 2020http://w

ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

cell movement via acetylation of �-tubulin and cortactin infibroblasts and lymphocytes (17).

Because HDACs have been suggested to be negative regula-tors of differentiation (12–14), we initially hypothesized thatmost HDACs would be higher in the less differentiated NLTGnRH cells and may be switched off with differentiation asexemplified by the GT1-7 GnRH neuronal cells. However, anopposite molecular signature in microarrays was observed withconsistently higher levels of all and in particular Class II HDACmembers in the differentiated (GT1-7) compared with less dif-ferentiated (NLT) cell lines derived from immortalized GnRHneurons. To understand the GnRH neuron-specific role ofClass II HDACs, we analyzed their functional role in the GnRHcell lines and asked if the effects are mediated through thenuclear or cytoplasmic compartments. Our data demonstrate anovel interaction between HDAC9 and HDAC6 to modulateGnRH neuronal cell survival and movement, thus supportingtheir potential developmental role for modulating reproductivefunction.

Experimental Procedures

Plasmids—FLAG-pcDNA3-HDAC6 was a gift from EricVerdin (Addgene plasmid #13823). Enhanced GFP (EGFP)-tagged mouse HDAC9 overexpressing plasmid (EGFP-HDAC9)and its translocating deficient mutant (EGFP-S218A/S448A-Hdac9) were provided by Noriyuki Sugo (Osaka University,Japan). pcMV 3Tag 6 vectors inserted with FLAG-tagged full-length HDAC9 cDNA construct (amino acids 1–1069), trun-cated FLAG-tagged N-terminal HDAC9 construct (aminoacids 1– 636), and FLAG-tagged C-terminal HDAC9 construct(amino acids 637–1069) were a generous gift provided by TapanChatterjee (Georgia Reagents University). pcDNA3.1(�) full-length FLAG-HDAC6 (1–1215), truncated FLAG-HDAC6(1–503 and 1– 840), FLAG-HDAC9-HA, and HA-HDAC6 con-structs were provided by Xiaohong Zhang (University of SouthFlorida, FL) and Edward Seto (Moffitt Cancer Center).pCMV-3 tag6-HDAC9 and pcDNA3.1(�)-HA(3)-HDAC9-FLAG plasmids were excised at the XhoI and BamHI sitesto obtain HDAC9 and pcDNA3.1(�)-HA(3), respectively.HDAC9 and pcDNA3.1(�)-HA(3) were ligated to obtain theplasmid pcDNA3.1(�)-HA(3)-HDAC9. pcDNA-HDAC6.DC-FLAG was a gift from Tso-Pang Yao (Addgene plasmid#30483). Sequences for all plasmids were confirmed using DNAsequencing (University of Colorado Denver, Cancer CenterSequencing Core Facility).

Antibodies and Reagents—HDAC6 (CO226) antibody waspurchased from Assay Biotech (Sunnyvale, CA). HDAC9(ab18970) antibody was purchased from Abcam (Cambridge,MA). For immunofluorescence experiments, rabbit anti-HDAC9 antibody (AP1109a) was obtained from Abgent (SanDiego, CA). Donkey-anti-rabbit-FITC was purchased fromJackson ImmunoResearch (103161). GAPDH and �-tubulinantibodies were purchased from Millipore (Billerica, MA) andAbcam, respectively. Anti-�-tubulin (2144) and anti-caspase 3antibody (9662) were purchased from Cell Signaling Technol-ogy (Beverly, MA). Acetylated �-tubulin (6-11B-1), BCL2(7382), protein A/G Plus-agarose immunoprecipitation re-agent, normal mouse, and rabbit immunoglobulin G (IgG) were

purchased from Santa Cruz Biotechnology (Santa Cruz, CA).Horseradish peroxidase-conjugated secondary antibodies (goatanti-rabbit IgG and goat anti-mouse IgG) were purchased fromBio-Rad. Anti-FLAG mouse antibody for immunoblotting(F3165) and immunofluorescence (F4049) and anti-HA anti-body (H6908) were purchased from Sigma. BAX antibody(BMS-163) was obtained from Bender MedSysyems (Austria).Lamin was purchased from Cell Signaling. Tubastatin A wasprovided by T. McKinsey.

Cell Culture and Transfections—NLT (18) and GT1-7 GnRHcell lines (19) were cultured as previously described (20). Over-expression of vector and HDAC9 constructs in NLT cells wereperformed using Lipofectamine Plus reagent (Invitrogen) as perthe manufacturer’s instructions. For silencing experiments,scrambled siRNAs and stealth siRNA pool for HDAC9 (Invit-rogen) and HDAC6 siRNA (Ambion, NY) targeting HDAC9and HDAC6, respectively, were transfected in GT1-7 cellsusing Lipofectamine 2000 (Invitrogen, CA). Cells were evalu-ated in functional assays 24 – 48 h post-transfection.

Adenoviral Transductions—Adenoviruses containing FLAG-tagged WT-HDAC6 and HDAC6 catalytic domain mutants(H216A/DD1 mutant, H611A/DD2 mutant, and H216A/H611A/DD1�DD2 mutant) were used. HDAC6 catalyticmutants H216A(DD1) and H611A(DD2) have single mutationsat His-216 to Ala and His-611 to Ala. Double catalytic HDAC6mutants (DD1�DD2) contain two mutations, His-216 to Alaand His-611 to Ala. For immunoprecipitation (IP) experiments,NLT cells were first transfected with FLAG-tagged vector andfull-length HDAC9 constructs for 17 h followed by adenoviraltransductions with FLAG tagged WT-HDAC6 and HDAC6catalytic domain mutants (DD1, DD2, and DD1�DD2) at anmultiplicity of infection of 25 pfu/cell. Cells were harvested 24 hpost transduction and used for IP experiments.

Microarray—Total RNA was isolated from three separate ali-quots of NLT and GT1-7 cells, and DNA microarray was per-formed. Bioinformatics analysis of the data were performed aspreviously described (21). Data were analyzed using GeneSpring software (Agilent Technologies, Santa Clara, CA) andIngenuity Pathway Analysis (Redwood City, CA).

RT-PCR—RNA from hypothalamic, pituitary, and ovariantissues of female adult mice were extracted using TRIzol re-agent as previously described earlier (22). Total RNA from NLTand GT1-7 cells was extracted using RNeasy kit (Qiagen, Valen-cia, CA). RT was performed using iScript (Bio-Rad), and qPCRwas performed using Power SYBR Green PCR master mix(Applied Biosystems, Foster city, CA) in a real time PCR systemas described earlier (23). The primer sequences used to amplifyHdac9 were 5�-CCATTGCCACGTGAACAACC-3� and 5�-TTCACGTCCACTGAGCTGAT-3�, Hdac6 were 5�-TGGCG-GACTAGAAAGAGCCT-3� and 5�-GAAGGGGTGACTGG-GGATTG-3�, and glyceraldehyde-3-phosphate dehydrogenase(Gapdh) were 5�-CGACCCCTTCATTGACCTCA-3� and 5�-CCACGACATACTCAGCACCG-3�. The cycle thresholdvalues (Ct) obtained for Hdac9 and Hdac6 genes were norma-lized against Gapdh to calculate ��Ct values from triplicateexperiments.

Immunoblot and IP—Immunoblotting of neuronal cell lysateswere performed as previously described (23). Densitometry anal-

Novel Interaction of HDAC6 and HDAC9 in GnRH Neurons

14046 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 290 • NUMBER 22 • MAY 29, 2015

by guest on Novem


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

ysis using GAPDH as the internal loading control from three sep-arate experiments was performed with the Bio-Rad Fluor-S multi-imager and NIH Image J software. The IP experiment wasperformed as previously described (21). For Tubastatin A effects,GT1-7 cells were treated with Tubastatin A (1 �M) for 24 h fol-lowed by harvesting and immunoprecipitation with HDAC6.

HDAC Activity Assay—Neuronal cell lysates were treatedwith acetylated HDAC substrates (I-1985 for Class IIa HDACs,I-1875 for Class IIb HDACs), and HDAC activity was deter-mined as described (24). For HDAC activity assay, IP complexeswere washed five times with HDAC assay buffer. The raw fluo-rescence signal was corrected for background, and data fromthree separate experiments were analyzed for significance.

Migration Assay—24 h after transfection, transfected NLTand GT1-7 cells were starved in serum-free DMEM for 5 h, andmigration assay was performed as described earlier (19). Basalmigration after 16 –18 h in serum-free medium was determinedby counting four fields on each membrane in three separateexperiments.

Apoptosis Assays—To assess rates of apoptosis, cleavedcaspase 3 assays were performed. 24 h after transfection, trans-fected NLT and GT1-7 cells were serum-starved for 48 h and16 h, respectively. Cells were harvested and used for immuno-blotting with cleaved caspase 3. For Hoechst staining trans-fected NLT and GT1-7 neuronal cells were plated on coverslipsin serum-free medium for 16 h, then fixed and stained withHoechst stain (33258) for 30 min (13). Apoptotic cells (withcondensed or fragmented chromatin) from 8 randomly chosenfields were counted in 1000 cells from duplicate coverslips in 3separate experiments using a fluorescent microscope (Axiovert200 Zeiss microscope, Carl Zeiss, Oberkochen, Germany).

Immunofluorescence—For endogenous detection and over-expression experiments (24 h post-transfection), GnRH cellswere plated (15,000/well) on coverslips, and immunofluores-cence experiments with FITC-FLAG (1:200) and HDAC6(1:200) were carried out as described (25). Immunofluores-cence for HDAC9 (1:200) was performed as described earlier(15). Coverslips were mounted with prolonged gold containingDAPI (Invitrogen) and observed under confocal microscope(Olympus FV1000 FCS/RCIS, Tokyo, Japan).

Nuclear and Cytoplasmic Fractionation—A NE-PER nuclearand cytoplasmic extraction kit (Thermo Fisher Scientific) wasused for nuclear and cytoplasmic extraction from NLT andGT1-7 cells with a modified protocol. For overexpressionexperiments, cells were harvested 48-h post-transfection andused for fractionation studies.

Statistical Analysis—Statistical analyses were performedusing GraphPad Software (La Jolla, CA). Data are representedas the mean � S.E., and statistical differences was analyzedusing Student’s t test for two groups and one way analysis ofvariance with the Bonferroni post hoc test among multiplegroups with p � 0.05 considered significant.

Results

Differential Expression of Hdac9 in GnRH Neuronal CellLines—DNA microarray of gene expression profiles in NLTcompared with GT1-7 cells were used to determine differencesin the transcript levels of different classes of HDACs using

bioinformatics tools including Gene Spring software and Inge-nuity Pathway Analysis (21). Analysis revealed a consistentincrease in the transcript levels for all classes of HDACs (exceptHDAC7a) in the GT1-7 cells compared with the less differen-tiated NLT neuronal cells (Fig. 1A). Higher levels of expressionfor the Class II HDACs, especially Hdac9 (6.7-fold), and wereobserved.

Increased HDAC9 Expression in GT1-7 Cells—Differentialexpression of Hdac9 in GnRH cell lines at the mRNA level wasconfirmed using qPCR. Increased expression of Hdac9 (4.1-fold, p � 0.0005) (Fig. 1B) along with variable expression ofother Class IIa Hdac members including Hdac4 (1.5-fold, p �0.74) and Hdac5 (1.3-fold, p � 0.08) (data not shown) was seenin GT1-7 cells compared with NLT cells. Hdac7 mRNA levelswere not detected in the GnRH neuronal cell lines (data notshown). Increased HDAC9 protein levels were confirmed byimmunoblot (4.5-fold, p � 0.03) (Fig. 1C). Immunofluores-cence demonstrated that endogenous HDAC9 was predomi-nately localized in the nucleus with a slight localization in thecytoplasmic compartment in the differentiated GT1-7 cells(Fig. 1D). Endogenous localization of HDAC9 was further con-firmed using fractionation studies. Endogenous HDAC9 washighly expressed in the nuclear extracts compared with thecytoplasmic extracts in both GT1-7 and NLT cells. Whole cellextracts reveal the difference in HDAC9 expression betweenboth these cell lines. Lamin and �-tubulin were used as nuclearand cytoplasmic markers, respectively (Fig. 1E). BecauseHDACs are thought to have their major function as deacety-lases, we next examined the neuronal-specific HDAC activitylevels.

HDAC9 Activity Is Up-regulated in GT1-7 Cells—Class-spe-cific HDAC activity was measured in NLT and GT1-7 cellsusing acetylated class-specific substrates. Acetylated class-spe-cific HDAC substrates, once deacetylated by their respectiveHDACs, become susceptible to cleavage by trypsin, whichreleases the fluorophore 7-amino-4-methylcoumarin. Consis-tent with the observed differences in gene expression, Class IHDAC activity was increased by 1.7-fold in GT1-7 cells (p �0.05; data not shown). The overall activity in an assay that mea-sures Class IIa HDACs (4, 5, 7, and 9) was not different betweenthe GnRH neuronal cells (Fig. 1F). Because there are noHDAC9-specific substrates available, GT1-7 and NLT celllysates were immunoprecipitated with a HDAC9-specific anti-body (Fig. 1G) and the Class IIa HDAC activity assay wasrepeated. Fig. 1H shows the increased activity of lysatesenriched for HDAC9 (1.5-fold, p � 0.05) in the differentiatedGT1-7 cells compared with NLT cells, reflective of the differ-ential expression at the mRNA and protein levels. Togetherthese data suggest that HDAC9 expression and activity increasein models of GnRH neuronal development.

Pro-survival Role of HDAC9 in GnRH Neuronal Cells—Toexplore potential functional roles, HDAC9 was introduced inthe less differentiated NLT cells to create a model of GnRHneuronal development (top panel in Fig. 2A), and converselyendogenous HDAC9 was silenced in the more differentiatedGT1-7 cells (Top panel in Fig. 2C). Each transfectant was testedin assays of cell survival. Vector control and NLT-HDAC9 cellswere exposed to serum deprivation as a model of growth factor



by guest on Novem


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

withdrawal to trigger cell death. NLT-HDAC9 cells showeddecreased Hoechst staining of condensed nuclei (0.5-fold, p �0.04, Fig. 2A) and reduced caspase-3 cleavage (0.6 versus 1.2-fold) (Fig. 2B) in response to serum deprivation compared withvector controls. These data suggested that HDAC9 drives sur-vival of GnRH cell lines. In support of this hypothesis, silencingof endogenous HDAC9 in GT1-7 cells (89% silencing by qPCR;

data not shown) resulted in a 4.1-fold (p � 0.01) increase inapoptotic cells as assessed by Hoechst staining compared withscramble controls (Fig. 2C). Similarly, silencing of HDAC9 inGT1-7 neurons reversed the protective effects of HDAC9 asassessed by caspase-3 cleavage (2.4-fold, p � 0.05) (Fig. 2D).

To determine if silencing of HDAC9 altered the levels ofother Class I (Hdac 1, 2, 3, and 8) or Class IIa Hdacs (-4, -5 and

FIGURE 1. HDAC9 RNA, protein, and activity are up-regulated in GT1-7 compared with NLT GnRH neuronal cells. Panel A, a table showing changes in classI Hdacs (-1, -2, -3, -8), IIa (-4, -5, -7, -9), and IV (-11) in GnRH neuronal cell lines. Panel B, increased mRNA expression of Hdac9 in GT1-7 neuronal cells comparedwith NLT cells using Gapdh as a housekeeping control. *, p � 0.05, n � 3. Panel C, bar graphs depicting increased HDAC9 protein levels (shown in the inset) inGT1-7 cells as compared with NLT cells with GAPDH as the loading control. *, p � 0.05, n � 3. Panel D, immunocytochemical localization of HDAC9 in GT1-7 cellsshowed higher nuclear than cytoplasmic localization. Nuclei stained with DAPI appear in blue (60�). Panel E, HDAC9 expression higher in nuclear extracts (NE)and slight in cytoplasmic extracts (CE) compared with whole cell extracts (WC) in both GT1-7 and NLT GnRH neuronal cells. Lamin and �-tubulin were used asnuclear and cytoplasmic markers. Panel F, overall Class IIa HDAC activity was similar NLT and GT1-7 neuronal cells. Panel G, IP of lysates with a HDAC9-specificantibody in NLT and GT1-7 cells. Panel H, immunoprecipitated HDAC9 lysates confirmed increased HDAC9 activity in GT1-7 neuronal cells. *, p � 0.05, n � 3.



by guest on Novem


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

-7), qPCRs were performed. Similar levels of Hdac1, Hdac2,Hdac3, Hdac8, Hdac4, and Hdac5 were observed after silencingof Hdac9, showing selective inhibition of Hdac9 (data notshown). Hdac7 was not detected in the cell lines. Togetherthese data suggest that HDAC9 may play a role in promotingGnRH neuron survival.

HDAC9 Acts as a Brake on GnRH Neuronal Cell Movement—To assess the role of HDAC9 in cell movement, migrationassays were performed in modified Boyden chamber (23).Overexpression of HDAC9 in NLT neuronal cells resulted in adramatic inhibition of neuronal cell movement toward serum(6.3-fold, p � 0.04) compared with control cells (Fig. 2E). Con-versely, silencing of endogenous HDAC9 (86%) in GT1-7 cellsresulted in the induction of movement of GT1-7 neuronal cells,which are usually not motile (26) (1.7-fold, p � 0.007; Fig. 2F).These results support the hypothesis that induction of HDAC9during development may play a role in cessation of GnRH neu-ronal migration.

Migration and Not Survival Effects of HDAC9 Are Dependenton Nuclear and Cytoplasmic Localizations—To ask if the effectsof HDAC9 to ensure cell survival or halt movement was depen-dent on a nuclear or cytoplasmic location of the protein anddetermine which domain of the protein mediated these effects,wild type (WT) and truncated HDAC9 (N terminus 1– 636, Cterminus 637–1069) constructs were tested in the functional

assays. Immunohistochemistry revealed that the full-lengthprotein when expressed in NLT cells was localized predomi-nantly in the nuclear compartment with some cytoplasmicexpression, similar to endogenous levels in GT1-7 neurons (seeFig. 3A). The HDAC-N truncated mutant (consisting of theregulatory domain and nuclear localization sequences) was spe-cifically targeted to the nucleus, and the HDAC9-C mutant(consisting of the HDAC catalytic domain with a nuclear exclu-sion sequence) was expressed exclusively in the cytoplasm (Fig.3A). WT and N- and C-HDAC9 truncated constructs wereexpressed at equivalent levels in the NLT cells by immunoblot(data not shown). Wild type and truncated HDAC9 proteinswere then tested for their effects on cell survival in response togrowth factor restriction by Hoechst staining of nuclear con-densation (Fig. 3B) and cell migration (Fig. 3C). In response toserum starvation-induced apoptosis, both N- and C-terminaltruncated HDAC9 constructs, similar to WT HDAC9, showeddecreased rates of apoptosis as compared with vector control(73 and 71%, respectively Fig. 3B), suggesting that nuclear aswell as cytoplasmic localization may be critical for mediatingthe pro-survival roles of Hdac9. NLT cells expressing WTHDAC9 (48%, p � 0.005 from control) and C-terminal HDAC9truncated constructs (45%, p � 0.006 from control) retainedtheir ability to inhibit basal GnRH neuronal movement (Fig.3C). However, this function was lost with nuclear-targeted

FIGURE 2. HDAC9 mediates cell survival and acts as a brake to neuronal cell movement. Panel A, the top panel shows representative immunoblots with aFLAG antibody with tubulin as a loading control. The bar graph depicts reduced % of apoptotic cells in NLT cells overexpressing HDAC9 as assessed by Hoechststaining. *, p � 0.05, n � 3. Panel B, bar graphs depict the effect of HDAC9 overexpression in NLT cells on the mean relative expression of FLAG-HDAC9, cleavedcaspase 3 to that of �-tubulin, and total caspase 3, respectively, in vector and HDAC9-overexpressed lysates in 10 and 0% serum containing DMEM (the toppanel depicts a representative immunoblot). Panel C, the top panel shows representative immunoblots for silencing endogenous HDAC9 with tubulin as aloading control. The bar graph depicts increased % of apoptotic cells in HDAC9 silenced in GT1-7 cells as assessed by Hoechst staining. *, p � 0.05, n � 3. PanelD, bar graphs depict the effect of HDAC9 silencing in GT1-7 cells on the mean relative expression of endogenous HDAC9, cleaved caspase 3 to that of �-tubulin,and total caspase 3, respectively, in scramble and HDAC9-silenced lysates in serum-free media (the top panel depicts a representative immunoblot). Panel E, bargraph depicting impaired basal movement in HDAC9-overexpressing NLT cells as assessed by a modified Boyden chambers assay. *, p � 0.05, n � 3. Panel F, bargraph depicting induced basal migration in GT1-7 neuronal cells after silencing of endogenous HDAC9 as assessed by a modified Boyden chambers assay. *,p � 0.05, n � 3.



by guest on Novem


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

HDAC9 mutant (24%, p � not significant), suggesting that acytoplasmic localization is critical for the effects of HDAC9 toinhibit cell movement. Overexpression of N-HDAC9 andC-HDAC9 confirmed their expression in the nuclear extractsand cytoplasmic extracts, respectively, in NLT cells. LAMINand �-tubulin were used as nuclear and cytoplasmic markersrespectively (Fig. 3D).

Nuclear Localization of HDAC9 Leads to Loss of Its Effect onCell Migration—Additional mutants were then analyzed to askif the effects of HDAC9 to halt cell movement were dependenton a nuclear to cytoplasmic shuttling function (inset in Fig. 3E).Immunohistochemistry confirmed that the HDAC9 mutantcontaining a two-amino acid mutation that blocks cytoplasmictranslocation (S218A/S448A) showed exclusive nuclear local-ization (Fig. 3E). In contrast to WT HDAC9, nuclear-localizedHDAC9 did not inhibit cell movement (1.4-fold, p � 0.0005increase compared with controls) in comparison to the inhibi-tory effects on movement exhibited by the WT mHDAC9 (0.5-fold, p � 0.002) (Fig. 3F). Together these data confirm thatcytoplasmic HDA9 contributes to repression of GnRH neuro-nal cell movement.

Class IIb HDAC6 Is a Novel Interacting Partner of HDAC9 —Little is known about the cytoplasmic partners of HDAC9.Because the Class IIb HDAC6 expression was also increased inGT1-7 cells in the microarray and is a predominantly cytoplas-mic protein, experiments determined whether it was a partnerof HDAC9. Anti-HDAC6 and anti-HDAC9 antibodies wereused to immunoprecipitate HDAC9 or HDAC6 protein. GT1-7lysates were used as an input control. HDAC9 and HDAC6were detected at 140 and 132 kDa in immunoprecipitatedHDAC6 (Fig. 4A) and HDAC9 (Fig. 4B) lysates, respectively.The interaction of HDAC9 and HDAC6 was specific, as non-immune rabbit IgG precipitated neither HDAC6 nor HDAC9.Immunoprecipitation of HDAC9 revealed no interaction withother members of Class IIa HDACs (HDAC4, HDAC5, andHDAC7) (data not shown). Hdac10, another member of ClassIIb, showed higher mRNA levels in NLT rather than GT1-7neuronal cells, similar to those seen at the transcript levels (datanot shown), and did not interact with HDAC9 (data notshown). Together, these data indicate a class member-specificinteraction of Class IIa, HDAC9, with Class IIb HDAC6 inGnRH neuronal cells.

FIGURE 3. HDAC9 promotes survival via nuclear and cytoplasmic location but impairs movement of NLT cells dependent on a cytoplasmic localization.Panel A, immunocytochemical localization of HDAC9 WT and N-terminal and C-terminal-truncated mutants with schematic illustration of truncations. Panel B,compared with vector controls, the bar graph depicts the reduced % of apoptotic cells in both N-terminal- and C-terminal-overexpressing HDAC9 constructssimilar to WT HDAC9-overexpressing NLT cells as assessed by Hoechst staining. *, p � 0.05, n � 3. Panel C, the bar graph depicts reduced basal movement in onlyC-terminal and WT-overexpressing HDAC9 constructs, whereas N-terminal HDAC9-overexpressing NLT cells were similar to vector controls as assessed by amodified Boyden chamber assay. *, p � 0.05, n � 3. Panel D, overexpression of N-HDAC9 and C-HDAC9 (as shown by FLAG) is exclusively expressed in thenuclear extracts (NE) and cytoplasmic extracts (CE), respectively, in NLT GnRH neuronal cells. LAMIN and �-tubulin were used as nuclear and cytoplasmicmarkers. Panel E, immunocytochemical localization of enhanced GFP-tagged HDAC9 WT and nucleocytoplasmic shuttling-deficient mutants in NLT cells withan inset of its schematic illustration of mutations. Panel F, the bar graph depicts reduced basal migration in only WT overexpressing HDAC9 constructs, whereasnuclear-localized nucleocytoplasmic-deficient mutant-overexpressing NLT cells were similar to vector controls as assessed by modified Boyden chamberassay. *, p � 0.05, n � 3.



by guest on Novem


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

Cytoplasmic-localized HDAC6 Is Up-regulated in GT1-7Cells—Transcript levels of Hdac6 were up-regulated in GT1-7cells (6.4-fold) compared with NLT cells (Fig. 4C). Differentialtranscript expression of Hdac6 at the mRNA level was con-firmed by semiquantitative qPCR (4.3-fold, p � 0.005; Fig. 4D).Hdac6 protein (4.2-fold, p � 0.05) was also higher in GT1-7compared with NLT cells as assessed by immunoblotting (Fig.4E). Immunocytochemical localization of endogenous HDAC6confirmed its cytoplasmic localization (Fig. 4F). Endogenouslocalization of HDAC6 was further confirmed using fraction-ation studies. HDAC6 was exclusively localized in the cyto-plasmic extracts and absent in the nuclear extracts in bothNLT and GT1-7 cells. Whole cell extracts reveal the differencein HDAC6 expression between both these cell lines (Fig. 4G).Lamin and �-tubulin were used as nuclear and cytoplasmicmarkers (Fig. 1E). Importantly, HDAC6 activity was also signif-icantly increased (5.4-fold, p � 0.005) in GT1-7 compared withNLT neuronal cells, thus supporting the functional importanceof cytoplasmic HDAC6 (Fig. 4H).

HDAC6 Interacts with HDAC9 via Its Second CatalyticDomain—NLT cells co-expressing FLAG-HDAC9 and HA-HDAC6 were immunoprecipitated with FLAG. HA-taggedHDAC6 was detected in the IP FLAG-HDAC9 lanes, suggesting

a specific HDAC6 and HDAC9 interaction in the cell lines (Fig.4I). IgG control was used as a negative control. To ask whichdomains of HDAC6 interact with HDAC9, co-immunoprecipi-tation experiments were performed in NLT cells. FLAG-taggedfull-length HDAC6 and truncated HDAC6 constructs contain-ing only the first catalytic domain (1–503) or a construct withboth catalytic domains (1– 840) were co-expressed with HA-HDAC9 in NLT cells. Lysates were immune-precipitated with aFLAG-specific antibody and immunoblotted with an anti-HAand anti-FLAG antibody (Fig. 5A). HA-HDAC9 was co-precip-itated in lysates expressing the FLAG-WT-HDAC6 and trun-cated FLAG-HDAC6 (1– 840) construct but not with the trun-cated HDAC6 construct containing a single catalytic domain(1–503). To demonstrate that FLAG antibody has no affinity forHA, NLT cells co-expressing FLAG-vector and FLAG-HDAC9with HA-HDAC6 were immunoprecipitated using anti-FLAGantibody and probed with HA. HA and FLAG was detected onlyin the HA-HDAC9 lane and not in the vector lane (data notshown). These findings suggest that HDAC6 interacts directlywith HDAC9 via its second catalytic domain.

To further elucidate the exact role of the second catalyticdomain, WT-HDAC6 and HDAC6 single and double catalyticdomain mutants were generated with point mutations in either

FIGURE 4. Class IIb HDAC6 is a novel interacting partner of HDAC9 and has up-regulated RNA, protein, and activity in GT1-7 neuronal cell lines. PanelsA and B, GT1-7 lysate (Input, control), lysates immunoprecipitated (IP) with anti-HDAC6 and anti-HDAC9, and rabbit IgG (negative control) were immunoblotted(IB) with anti-HDAC9 and anti HDAC6, n � 3. Panel C, the table shows specific changes in class IIb Hdacs (6 and 10) in cell lines. Panel D, mRNA expression ofHdac6 in GT1-7 neuronal cells compared with NLT cells using GAPDH as a housekeeping control. *, p � 0.05, n � 3. Panel E, bar graphs depicting increasedHDAC6 protein levels (shown in the inset) in GT1-7 cells as compared with NLT cells with GAPDH as loading control. *, p � 0.05, n � 3. Panel F, immunocyto-chemical localization of endogenous HDAC6 in GT1-7 cells showed cytoplasmic localization (60�). Panel G, HDAC6 expression exclusively seen in cytoplasmicextracts (CE) and in whole cell extract (WE) but absent in nuclear extracts (NE) in both GT1-7 and NLT GnRH neuronal cells. Panel H, overall Class IIb HDAC6activity is higher in GT1-7 compared with NLT cells. Panel I, HA-HDAC6 was detected in NLT cells co-expressing FLAG-HDAC9 and HA-HDAC6 and co-immunoprecipitated using anti-FLAG antibodies. IgG was used as a negative control.



by guest on Novem


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

or both of the HDAC domains. In NLT cells, low amounts ofendogenous HDAC6 were detected across all input and IPlanes (data not shown). Exogenous HDAC6 in both the inputand IP lanes was present only in the vector/WT HDAC6- andWT HDAC6-overexpressing lanes and was not in only vectorand WT HDAC9-overexpressing lysates, thus confirming over-expression of exogenous HDAC6 and its immunoprecipitationin NLT cells. HDAC9 was detected in all the HDAC6 IP lanes,confirming our earlier observation of interaction betweenHDAC6 and HDAC9 (Fig. 5B). Overexpression of WT HDAC6and catalytic domain mutants (in variable amounts) wasdetected in all the input and IP HDAC6 lanes (Fig. 5B). HDAC9was identified in immunoprecipitated WT HDAC6/WTHDAC9 and IP mtHDAC6 (DD1)/WT HDAC9; however, thisinteraction was disrupted in the immunoprecipitation ofmtHDAC6 (DD2)/WT HDAC9 lanes (Fig. 5B). To confirm thespecificity of the interaction of HDAC6 and HDAC9 via the sec-

ond catalytic domain of HDAC6, a double catalytic (DC) HDAC6mutant (H216A/H611A) was co-expressed with WT HDAC9 inNLT cells. HDAC9 was detected using IP of HDAC6 in NLTcells co-expressing WT HDAC6 and WT HDAC9 (Fig. 5C) butlost when WT HDAC9 was co-expressed with the DC-HDAC6mutant (Fig. 5C). Similarly, HA-HDAC9 was immunoprecipi-tated with FLAG-HDAC6 but not with the FLAG-DC-HDAC6mutant using FLAG immunoprecipitation (data not shown).Additional studies demonstrated that treatment of GT1-7 cellswith Tubastatin A (1 �M), a known specific catalytic inhibitor ofHDAC6 that targets its second catalytic domain, abolished theHDAC9 and HDAC6 interaction (Fig. 5D). Tubulin acetylationwas increased after treatment with Tubastatin, indicating theeffectiveness of HDAC6 inhibition (data not shown). Together,these data confirm that the interaction of Class IIa HDAC9 withClass IIb HDAC6 occurs via the second catalytic domain ofHDAC6.

FIGURE 5. HDAC6 and HDAC9 interaction is dependent on activity of second catalytic domain of HDAC6. Panel A, NLT cells co-expressing FLAG-HDAC9and HDAC6 truncated plasmids (schematic, not drawn to scale) were co-immunoprecipitated using anti-FLAG and immunoblotted (IB) with anti-HA (top panel)and anti-FLAG (middle panel). IgG was used as a negative control. Input lysates (bottom panel) were analyzed by immunoblotting using anti-HA antibodies.Panel B, overexpressed WT-HDAC6/WT-HDAC9 and HDAC6 domain mutants (the inset shows schematic illustration of mutations) in NLT (Input, control) wereimmunoprecipitated with anti-HDAC6 and rabbit IgG (negative control) and immunoblotted with anti-HDAC6 and anti-HDAC9. The ratio of IP HDAC9 withinput HDAC6 controls is indicated as -fold change (FC). Panel C, IP of HDAC6 failed to show HDAC9 in NLT cells co-expressing WT-HDAC9 and double catalyticHDAC6 (H216A/H611A) compared with cells co-expressing WT-HDAC9 and WT-HDAC6. Rabbit IgG was used as negative control. Panel D, GT1–7 cells treatedwith Tubastatin-A (1 �M) lysates immunoprecipitated with anti-HDAC6 failed to show interaction when immunoblotted with HDAC9. IgG was used as anegative control.



by guest on Novem


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

Interaction of HDAC6 and HDAC9 Controls Cell Survival viathe BAX/BCL2 Pathway—The functional role of HDAC6 in cellsurvival was then tested by silencing it alone or in combinationwith HDAC9 in GT1-7 cells. Hdac6 and Hdac9 and combinedsilencing showed significant knockdown of Hdac6 and Hdac9in GT1-7 cells as seen by real-time PCR (Fig. 6, A and B). Proteinlevels of knockdown of HDAC6, HDAC9, and HDAC6 and -9together are shown in Fig. 6C. Cleaved Caspase-3 levels wereassessed as an index of cell death. Silencing of HDAC6,HDAC9, or both augmented caspase-3 cleavage (3.4-, 4.4-, and7.1-fold, respectively, p � 0.02, Fig. 6D), suggesting thatHDAC6 and -9 have additive effects to promote GnRH neu-ronal cell survival. To further elucidate the downstreameffectors of HDAC9 and -6 that can promote cell survival,expression of several of the pro and anti-apoptotic pathwaymolecules were assessed. Silencing of either HDAC6 orHDAC9 alone slightly increased the expression of BAX, apro-apoptotic marker by (1.6-fold, p � 0.48; 1.3-fold, p �0.65). However, this induction was significant only whenboth were silenced (3.9-fold, p � 0.02). Conversely, silencingof HDAC6 inhibited the anti-apoptotic marker BCL2 by(3.7-fold, p � 0.06) and showed a significant decrease only insilencing HDAC9 (9.3-fold, p � 0.04) alone and in combina-tion 35.6-fold (p � 0.04) compared with scramble controls(Fig. 6E). Thus HDAC9/6 interaction regulates cell survivalvia modulating BAX and BCL2, the pro- and anti-apoptoticproteins in the death pathway.

Interaction of HDAC6 and HDAC9 Affects Migration inGnRH Neuronal Cells—To evaluate the combinatorial effects ofHDAC6 and HDAC9 on the control of cell movement, modi-fied Boyden chamber assays were performed. Silencing of eitherHDAC6 or HDAC9 (Fig. 7A) induced GT1-7 cell movementcompared with controls (2.6- and 2.3-fold, respectively, p �0.05) and together resulted in an additive effect to promoteGT1-7 GnRH neuronal movement (5-fold, p � 0.05, Fig. 7A).Together, these in vitro data raise the novel hypothesis thatClass IIa and Class IIb HDACs may play a role in the timing ofcessation of GnRH neuronal migration in vivo.

To further elucidate the mechanism of HDAC6 and HDAC9interaction to alter cell movement, levels of acetylated tubulinwere measured as an index of cell motility. In NLT cells, over-expression of HDAC6 reduced acetylated �-tubulin levels com-pared with vector control (56.6%). However, overexpression ofHDAC6 and HDAC9 together failed to reduce this acetylationstatus, suggesting that HDAC6 interaction with HDAC9 abro-gates the acetylation of �-tubulin, thus leading to impairmentin migration (Fig. 7B). Conversely, in GT1-7 cells, silencing ofHDAC9 resulted in reduced acetylated tubulin (59.4%). In con-trast, silencing of HDAC6 resulted in higher levels of acetylatedtubulin. However, silencing of both HDAC9 and HDAC6blocked the induction of acetylation that was seen with silenc-ing HDAC6 alone (Fig. 7B). These data suggest that the abilityof HDAC6 to deacetylate tubulin to induce neuronal migrationis lost due to its interaction with HDAC9 in GT1-7 cells, result-

FIGURE 6. Pro-survival role of HDAC9 and -6 in GT1-7 cells. Panels A and B, semiquantitative real time PCR for Hdac6 and Hdac9 in scramble controlsand Hdac6 and Hdac9 silenced GT1-7 cell using Gapdh as loading control. *, p � 0.05, n � 3. Panel C, bar graphs depict effect of silencing HDAC6 andHDAC9 in GT1-7 cells compared with scramble controls. Tubulin was used as a loading control (lower panel) (the inset depicts a representativeimmunoblot). Panel D, bar graphs, combined HDAC6 and HDAC9 silencing induces cleaved caspase-3 expression in GT1-7 cells compared with scramblecontrols (the inset depicts a representative immunoblot). Panel E, bar graphs, combined HDAC6 and HDAC9 silencing induces BAX (upper panel) andreduces BCL-2 (middle panel) expression in GT1-7 cells. �-Tubulin was used as a loading control (lower panel) (the inset depicts a representativeimmunoblot). *, p � 0.05, n � 3.



by guest on Novem


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

ing in the cessation of movement, a process that is critical forthe appropriate targeting of GnRH neurons to the forebrainduring development.

Discussion

Appropriate development and targeting of GnRH neurons isa complex process that involves precise controls on their sur-vival and migration that can affect ultimately sexual maturationand reproductive competency (1, 22, 27). Cell adhesion mole-cules, neurotransmitters, growth factors, G-protein receptors,and transcription factors contribute to GnRH neuronal devel-opment (1). Additional mechanisms that govern the success ofneuronal targeting remain to be elucidated. The processwherein GnRH neurons stop their migration to appropriatelylocalize in the hypothalamus has largely remained unexplored.Recently, Kurian and Terasawa (4) demonstrated that in vitroneuronal maturation of GnRH gene expression coincides with

developmental changes in DNA demethylation in post-mitoticneurons, drawing attention to the relatively unexplored field ofepigenetic regulation in GnRH neurons.

The roles of HDACs in reproductive and neuronal develop-ment are under active investigation. Using HDAC inhibitors asa tool to modulate HDAC function, investigators have showntheir role as nuclear transcriptional repressors critical for sper-matogenesis (29) to repress FSH� and LH� gene expression inimmature pituitary gonadotropes (30) and to influence GnRHgene expression (31), thus emphasizing their critical roles in thehypothalamic-pituitary-gonadal axis. Our studies identify ClassII HDACs as additional migratory and survival cues in GnRHneuron cells and suggest potential roles for these HDACs inGnRH neurons in vivo.

Our observation of a global activation of all classes of HDACsin the cell models of GnRH neuronal differentiation is in con-trast to other systems where many HDACs act only as tran-scriptional “brakes” and are turned off with development.Undifferentiated myocytes, adipocytes, and cholinergic neu-rons demonstrated high expression of HDAC9, whereas rela-tively lower levels were detected in the differentiated state (12,32, 33). For example, Mitr/Hdrp, an Hdac9 splice variant,represses MEF2 targets by recruiting Class I HDACs but then isdown-regulated with muscle denervation to allow up-regula-tion of the acetylcholine receptor subunits (33). In cholinergicneurons, HDAC9 is also a negative regulator of differentiationand is inhibited to allow activation of the choline acetyltrans-ferase gene during neuronal development (32). However, in thecurrent experiments, most HDACs and Class II HDACs, in par-ticular, were up-regulated in the more differentiated cell line,suggesting a cell-specific developmental program. HDAC9 washighly expressed in the dissected hypothalamus from adultmice, suggesting it could influence the hypothalamic locus ofthe reproductive axis in vivo.

Prior work has shown that Class IIa HDACs (HDAC4, -5, and-7) compensate for each other when one is silenced to promoteresponses to glucagon and insulin signaling in the liver (14).HDAC9 can interact with nuclear co-repressor (N-CoR) as wellas HDAC family members (HDAC1, HDAC3, and HDAC4)involved in hematological malignancies (16). In the currentstudy, HDAC9 did not co-immunoprecipitate with any otherClass IIa HDAC family members. Importantly, HDAC9 showeda novel ability to interact with HDAC6, a cytoplasmically local-ized class IIb HDAC. HDAC6, a major inducer of cell migrationin lymphocytes and mouse embryonic fibroblasts cells, hasbeen reported to have a number of non-histone interactingpartners including �-tubulin, HSP90, cortactin, and others (17,35) that play a significant role in cell movement, ubiquitination,management of misfolded proteins, apoptosis, transcription,and cell proliferation (36 –38). Earlier reports indicated thatHDAC6 could interact with HDAC11, a class IV HDAC mem-ber, in both the cytoplasmic and nuclear compartments and actas a transcriptional activator for IL-10 (39). HDAC6 has alsobeen reported to interact with SIRT2, a Class III HDAC mem-ber (40). Interaction of HDAC6 with Class I or II HDAC familymembers had not been previously reported. Our study is thefirst to demonstrate this novel interaction between a Class IIaand IIb HDAC family member.

FIGURE 7. HDAC9 and HDAC6 modulates migration. Panel A, the bar graphdepicts amplified basal movement in both HDAC6 and HDAC9 silencedGT-1–7 cells with augmented effects on basal cell movement after silencingboth HDAC6 and HDAC9 as assessed by modified Boyden chamber assay. *,p � 0.05, n � 3. Panel B, overexpression (top panel) or silencing (bottom panel)of HDAC6 and HDAC9 together impedes acetylation of �-tubulin in NLT andGT1-7 cells, respectively, with total �-tubulin as a loading control. Panel C,schematic illustration of HDAC9 and HDAC6 interaction and their role in mod-ulating survival and movement of GT1-7 neuronal cells.



by guest on Novem


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

Isotype effects of HDACs have been previously shown tohave additive, independent effects on T cell function. HDAC6and HDAC9 can each modulate Foxp3 expression and hyper-acetylate Hsp90, thus controlling regulatory T cell function;however, no analysis was performed of any interaction betweenHDACs (41). HDAC6 and SIRT2 can positively regulate acety-lation of K-ras, an oncogene in cancer cells, thus suggesting apossible combined therapeutic target utilizing interactionsbetween different classes of HDACs (40). The current studiessuggest a paradigm-shifting model where Class IIa HDAC9 andClass IIb HDAC6 directly interact to modulate cell survival andhalt cell movement through unique cytoplasmic and nuclearactions (Fig. 7C). In other systems calcium/calmodulin andprotein phosphatases modulate shuttling of Class IIa HDACs.The specific phosphatases reported to regulate Class IIaHDACs include PP1B and MYPT1 as HDAC7 partners in theimmune system and PP2A with HDAC7 and -4 in muscle andbone cells (11). Also, recent library screens suggest that Class IIHDACs may interact with other cell-specific kinases includingAMPK, MASK2, and Mirk/SIK1, -2, and -3 and checkpointkinases (34). Further studies are needed to determine whetherthere are GnRH neuronal cell-specific mechanisms that mod-ulate nuclear-cytoplasmic shuttling of HDAC9.

Interaction with HDAC9 in GnRH neuronal cell lines wasdependent on a direct interaction with the second catalyticdomain of HDAC6. Among the HDACs, the first and secondcatalytic domains of HDAC6 show a higher (50 and 55%) simi-larity to the catalytic domain of HDAC9 (42). HDAC9 did notinteract with the other Class IIb HDAC, HDAC10, confirmingthe specificity of the HDAC9 and -6 interaction. BecauseHDAC10 lacks a functional second catalytic domain comparedwith HDAC6, the lack of interaction between HDAC9 andHDAC10 supports the hypothesis that interaction of HDAC9and HDAC6 is dependent on the second catalytic domain.

In GnRH neuronal cells, HDAC9 and HDAC6 control cellsurvival by modulating the BAX/BCL2 pathway. Prior litera-ture suggests that general HDAC inhibitors (HDACi) can act aspro-apoptotic inducers in numerous cancer cell types (43).With regard to the effect of HDACs on cell survival, variableeffects have been reported. In colon cancer cells, silencing ofHDAC1 and HDAC3 trigger cell death (44). In cerebellar neu-rons, activation of HDAC1 and HDAC3 triggers cell death,whereas in cortical neurons, activation of HDAC1 and HDRPpromotes survival (45). In the fibroblast cell line AT5BIVA,HDAC9 had no effect on apoptosis induced by irradiation (46).Nuclear HDRP, an alternative splice variant of HDAC9 thatlacks an intrinsic catalytic domain, maintained survival of cer-ebellar granular neurons under serum deprivation with non-depolarizing levels of potassium (47). HDRP was not detected atthe transcript level in our GnRH cell lines. HDAC6 overexpres-sion has been suggested to increase survival in multiplemyeloma cells (28). We speculate that interaction of HDAC6with HDAC9 may alter the levels of HDAC6 and HDAC9,which leads to augmentation of BAX and abrogation of BCL2levels in GT1-7 cells. The current data strongly suggest a pro-survival role of HDACs in GnRH neuronal cells.

In contrast to other systems, in GnRH neuronal cells, thepro-migratory action of HDAC6 is reversed by its interaction

with increasing expression of HDAC9. The effects of HDAC9on cell movement had not been previously studied; however,overexpression of HDAC6 in other systems induces motility byreducing acetylation of its substrates �-tubulin and cortactinand increasing interaction with F-actin. The second catalyticdomain of HDAC6 is mainly required to mediate this tubulindependent deacetylase mechanism of migration (17). In GT1-7cells, HDAC6 interaction with HDAC9 via this second catalyticdomain could potentially impede this process of tubulindeacetylation, thus blocking migration. Interestingly, silencingof HDAC6 in GT1-7 cells promoted migration rather thandecreasing it, suggesting that cell migration mediated byHDAC6 could be influenced by cell-specific partners that maybe independent from its deacetylase-dependent mechanisms.

In summary, the current studies identified a unique molecu-lar footprint of differential expression of HDAC members incell lines derived from immortalized GnRH neurons, increasingrather than decreasing across models of differentiation.Detailed experiments demonstrate a novel interaction of ClassIIa HDAC9 with Class IIb HDAC6 to mediate the cessation ofGnRH neuronal cell movement via cytoplasmic actions and theability to promote cell survival via both cytoplasmic and nuclearactions. Evaluation of GnRH neuron development and carefulreproductive phenotyping in Hdac9 and Hdac6 GnRH-specificknock-out models will provide validation of these potentiallyimportant actions of Class II HDACs on reproductive develop-ment. Examination for mutations in these proteins and theirpathways in patients with hypogonadotropic hypogonadismmay be warranted to ask if alterations in histone deacetylasesimpact sexual maturation in humans.

Acknowledgments—We thank Todd Horn for performing the HDACcatalytic activity assays. Imaging experiments were performed in theUniversity of Colorado Anschutz Medical Campus Advance LightMicroscopy Core supported in part by National Center for AdvancingTranslational Sciences, National Institutes of Health Colorado Clin-ical and Translational Sciences Institute Grant UL1 TR001082.

References1. Wierman, M. E., Kiseljak-Vassiliades, K., and Tobet, S. (2011) Gonadotro-

pin-releasing hormone (GnRH) neuron migration: initiation, mainte-nance and cessation as critical steps to ensure normal reproductive func-tion. Front. Neuroendocrinol. 32, 43–52

2. Schwarting, G. A., Wierman, M. E., and Tobet, S. A. (2007) Gonadotropin-releasing hormone neuronal migration. Semin. Reprod. Med. 25, 305–312

3. Tobet, S. A., and Schwarting, G. A. (2006) Minireview: recent progress ingonadotropin-releasing hormone neuronal migration. Endocrinology 147,1159 –1165

4. Kurian, J. R., and Terasawa, E. (2013) Epigenetic control of gonadotropinreleasing hormone neurons. Front. Endocrinol. (Lausanne) 4, 61

5. Terasawa, E., Kurian, J. R., Guerriero, K. A., Kenealy, B. P., Hutz, E. D., andKeen, K. L. (2010) Recent discoveries on the control of gonadotrophin-releasing hormone neurones in nonhuman primates. J. Neuroendocrinol.22, 630 – 638

6. Witt, O., Deubzer, H. E., Milde, T., and Oehme, I. (2009) HDAC family:what are the cancer relevant targets? Cancer Lett. 277, 8 –21

7. Martin, M., Kettmann, R., and Dequiedt, F. (2007) Class IIa histonedeacetylases: regulating the regulators. Oncogene 26, 5450 –5467

8. de Ruijter, A. J., van Gennip, A. H., Caron, H. N., Kemp, S., and vanKuilenburg, A. B. (2003) Histone deacetylases (HDACs): characterizationof the classical HDAC family. Biochem. J. 370, 737–749



by guest on Novem


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

9. Haberland, M., Montgomery, R. L., and Olson, E. N. (2009) The manyroles of histone deacetylases in development and physiology: implicationsfor disease and therapy. Nat. Rev. Genet. 10, 32– 42

10. Zhou, X., Marks, P. A., Rifkind, R. A., and Richon, V. M. (2001) Cloningand characterization of a histone deacetylase, HDAC9. Proc. Natl. Acad.Sci. U.S.A. 98, 10572–10577

11. Bertos, N. R., Wang, A. H., and Yang, X. J. (2001) Class II histone deacety-lases: structure, function, and regulation. Biochem. Cell Biol. 79, 243–252

12. Chatterjee, T. K., Idelman, G., Blanco, V., Blomkalns, A. L., Piegore, M. G.,Jr., Weintraub, D. S., Kumar, S., Rajsheker, S., Manka, D., Rudich, S. M.,Tang, Y., Hui, D. Y., Bassel-Duby, R., Olson, E. N., Lingrel, J. B., Ho, S. M.,and Weintraub, N. L. (2011) Histone deacetylase 9 is a negative regulatorof adipogenic differentiation. J. Biol. Chem. 286, 27836 –27847

13. Zhang, C. L., McKinsey, T. A., Chang, S., Antos, C. L., Hill, J. A., and Olson,E. N. (2002) Class II histone deacetylases act as signal-responsive repres-sors of cardiac hypertrophy. Cell 110, 479 – 488

14. Chang, S., McKinsey, T. A., Zhang, C. L., Richardson, J. A., Hill, J. A., andOlson, E. N. (2004) Histone deacetylases 5 and 9 govern responsiveness ofthe heart to a subset of stress signals and play redundant roles in heartdevelopment. Mol. Cell. Biol. 24, 8467– 8476

15. Sugo, N., Oshiro, H., Takemura, M., Kobayashi, T., Kohno, Y., Uesaka, N.,Song, W. J., and Yamamoto, N. (2010) Nucleocytoplasmic translocation ofHDAC9 regulates gene expression and dendritic growth in developingcortical neurons. Eur. J. Neurosci. 31, 1521–1532

16. Petrie, K., Guidez, F., Howell, L., Healy, L., Waxman, S., Greaves, M., andZelent, A. (2003) The histone deacetylase 9 gene encodes multiple proteinisoforms. J. Biol. Chem. 278, 16059 –16072

17. Valenzuela-Fernández, A., Cabrero, J. R., Serrador, J. M., and Sánchez-Madrid, F. (2008) HDAC6: a key regulator of cytoskeleton, cell migration,and cell-cell interactions. Trends Cell Biol. 18, 291–297

18. Radovick, S., Wray, S., Lee, E., Nicols, D. K., Nakayama, Y., Weintraub,B. D., Westphal, H., Cutler, G. B., Jr., and Wondisford, F. E. (1991) Migra-tory arrest of gonadotropin-releasing hormone neurons in transgenicmice. Proc. Natl. Acad. Sci. U.S.A. 88, 3402–3406

19. Mellon, P. L., Windle, J. J., Goldsmith, P. C., Padula, C. A., Roberts, J. L.,and Weiner, R. I. (1990) Immortalization of hypothalamic GnRH neuronsby genetically targeted tumorigenesis. Neuron 5, 1–10

20. Salian-Mehta, S., Xu, M., Knox, A. J., Plummer, L., Slavov, D., Taylor, M.,Bevers, S., Hodges, R. S., Crowley, W. F., Jr., and Wierman, M. E. (2014)Functional consequences of AXL sequence variants in hypogonadotropichypogonadism. J. Clin. Endocrinol. Metab. 99, 1452–1460

21. Pierce, A., Bliesner, B., Xu, M., Nielsen-Preiss, S., Lemke, G., Tobet, S., andWierman, M. E. (2008) Axl and Tyro3 modulate female reproduction byinfluencing gonadotropin-releasing hormone neuron survival and migra-tion. Mol. Endocrinol. 22, 2481–2495

22. Pierce, A., Xu, M., Bliesner, B., Liu, Z., Richards, J., Tobet, S., and Wier-man, M. E. (2011) Hypothalamic but not pituitary or ovarian defects un-derlie the reproductive abnormalities in Axl/Tyro3 null mice. Mol. Cell.Endocrinol. 339, 151–158

23. Salian-Mehta, S., Xu, M., and Wierman, M. E. (2013) AXL and METcrosstalk to promote gonadotropin releasing hormone (GnRH) neuronalcell migration and survival. Mol. Cell. Endocrinol. 374, 92–100

24. Lemon, D. D., Horn, T. R., Cavasin, M. A., Jeong, M. Y., Haubold, K. W.,Long, C. S., Irwin, D. C., McCune, S. A., Chung, E., Leinwand, L. A., andMcKinsey, T. A. (2011) Cardiac HDAC6 catalytic activity is induced inresponse to chronic hypertension. J. Mol. Cell Cardiol 51, 41–50

25. Allen, M. P., Linseman, D. A., Udo, H., Xu, M., Schaack, J. B., Varnum, B.,Kandel, E. R., Heidenreich, K. A., and Wierman, M. E. (2002) Novel mech-anism for gonadotropin-releasing hormone neuronal migration involvingGas6/Ark signaling to p38 mitogen-activated protein kinase. Mol. Cell.Biol. 22, 599 – 613

26. Maggi, R., Pimpinelli, F., Molteni, L., Milani, M., Martini, L., and Piva, F.(2000) Immortalized luteinizing hormone-releasing hormone neuronsshow a different migratory activity in vitro. Endocrinology 141, 2105–2112

27. Herbison, A. E., Porteous, R., Pape, J. R., Mora, J. M., and Hurst, P. R. (2008)Gonadotropin-releasing hormone neuron requirements for puberty, ovu-lation, and fertility. Endocrinology 149, 597– 604

28. Hideshima, T., Bradner, J. E., Wong, J., Chauhan, D., Richardson, P.,Schreiber, S. L., and Anderson, K. C. (2005) Small-molecule inhibition ofproteasome and aggresome function induces synergistic antitumor activ-ity in multiple myeloma. Proc. Natl. Acad. Sci. U.S.A. 102, 8567– 8572

29. Fenic, I., Sonnack, V., Failing, K., Bergmann, M., and Steger, K. (2004) Invivo effects of histone-deacetylase inhibitor trichostatin-A on murinespermatogenesis. J. Androl. 25, 811– 818

30. Lim, S., Luo, M., Koh, M., Yang, M., bin Abdul Kadir, M. N., Tan, J. H., Ye,Z., Wang, W., and Melamed, P. (2007) Distinct mechanisms involvingdiverse histone deacetylases repress expression of the two gonadotropin�-subunit genes in immature gonadotropes, and their actions are over-come by gonadotropin-releasing hormone. Mol. Cell. Biol. 27, 4105– 4120

31. Gan, L., Ni, P. Y., Ge, Y., Xiao, Y. F., Sun, C. Y., Deng, L., Zhang, W., Wu,S. S., Liu, Y., Jiang, W., and Xin, H. B. (2012) Histone deacetylases regulategonadotropin-releasing hormone I gene expression via modulating Otx2-driven transcriptional activity. PLoS ONE 7, e39770

32. Aizawa, S., Teramoto, K., and Yamamuro, Y. (2012) Histone deacetylase 9as a negative regulator for choline acetyltransferase gene in NG108 –15neuronal cells. Neuroscience 205, 63–72

33. Méjat, A., Ramond, F., Bassel-Duby, R., Khochbin, S., Olson, E. N., andSchaeffer, L. (2005) Histone deacetylase 9 couples neuronal activity tomuscle chromatin acetylation and gene expression. Nat. Neurosci. 8,313–321

34. Walkinshaw, D. R., Weist, R., Kim, G. W., You, L., Xiao, L., Nie, J., Li, C. S.,Zhao, S., Xu, M., and Yang, X. J. (2013) The tumor suppressor kinase LKB1activates the downstream kinases SIK2 and SIK3 to stimulate nuclearexport of class IIa histone deacetylases. J. Biol. Chem. 288, 9345–9362

35. Gao, Y. S., Hubbert, C. C., Lu, J., Lee, Y. S., Lee, J. Y., and Yao, T. P. (2007)Histone deacetylase 6 regulates growth factor-induced actin remodelingand endocytosis. Mol. Cell. Biol. 27, 8637– 8647

36. Li, Y., Shin, D., and Kwon, S. H. (2013) Histone deacetylase 6 plays a role asa distinct regulator of diverse cellular processes. FEBS J 280, 775–793

37. Zhang, Y., Li, N., Caron, C., Matthias, G., Hess, D., Khochbin, S., andMatthias, P. (2003) HDAC-6 interacts with and deacetylates tubulin andmicrotubules in vivo. EMBO J. 22, 1168 –1179

38. Zhang, X., Yuan, Z., Zhang, Y., Yong, S., Salas-Burgos, A., Koomen, J.,Olashaw, N., Parsons, J. T., Yang, X. J., Dent, S. R., Yao, T. P., Lane, W. S.,and Seto, E. (2007) HDAC6 modulates cell motility by altering the acety-lation level of cortactin. Mol. Cell 27, 197–213

39. Gao, L., Cueto, M. A., Asselbergs, F., and Atadja, P. (2002) Cloning andfunctional characterization of HDAC11, a novel member of the humanhistone deacetylase family. J. Biol. Chem. 277, 25748 –25755

40. Yang, M. H., Laurent, G., Bause, A. S., Spang, R., German, N., Haigis, M. C.,and Haigis, K. M. (2013) HDAC6 and SIRT2 regulate the acetylation stateand oncogenic activity of mutant K-RAS. Mol. Cancer Res. 11, 1072–1077

41. Beier, U. H., Wang, L., Han, R., Akimova, T., Liu, Y., and Hancock, W. W.(2012) Histone deacetylases 6 and 9 and sirtuin-1 control Foxp3� regu-latory T cell function through shared and isoform-specific mechanisms.Sci. Signal. 5, ra45

42. Bhatia, U., Cai, R. L., Cohen, D., and Fischer, D. D. (April 18, 2002) Histonedeacetylase-related gene and protein. U. S. Patent WO2002030970A2

43. Insinga, A., Monestiroli, S., Ronzoni, S., Gelmetti, V., Marchesi, F., Viale,A., Altucci, L., Nervi, C., Minucci, S., and Pelicci, P. G. (2005) Inhibitors ofhistone deacetylases induce tumor-selective apoptosis through activationof the death receptor pathway. Nat. Med. 11, 71–76

44. Thangaraju, M., Carswell, K. N., Prasad, P. D., and Ganapathy, V. (2009)Colon cancer cells maintain low levels of pyruvate to avoid cell deathcaused by inhibition of HDAC1/HDAC3. Biochem. J. 417, 379 –389

45. Bardai, F. H., Price, V., Zaayman, M., Wang, L., and D’Mello, S. R. (2012)Histone deacetylase-1 (HDAC1) is a molecular switch between neuronalsurvival and death. J. Biol. Chem. 287, 35444 –35453

46. Yuan, Z., Peng, L., Radhakrishnan, R., and Seto, E. (2010) Histone deacety-lase 9 (HDAC9) regulates the functions of the ATDC (TRIM29) protein.J. Biol. Chem. 285, 39329 –39338

47. Morrison, B. E., Majdzadeh, N., Zhang, X., Lyles, A., Bassel-Duby, R.,Olson, E. N., and D’Mello, S. R. (2006) Neuroprotection by histonedeacetylase-related protein. Mol. Cell. Biol. 26, 3550 –3564



by guest on Novem


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

WiermanSmita Salian-Mehta, Mei Xu, Timothy A. McKinsey, Stuart Tobet and Margaret E.

Survival and MovementHDAC9 Controls Gonadotropin Releasing Hormone (GnRH) Neuronal Cell Novel Interaction of Class IIb Histone Deacetylase 6 (HDAC6) with Class IIa

doi: 10.1074/jbc.M115.640482 originally published online April 14, 20152015, 290:14045-14056.J. Biol. Chem.

10.1074/jbc.M115.640482Access the most updated version of this article at doi:

Alerts:

When a correction for this article is posted•

When this article is cited•

to choose from all of JBC's e-mail alertsClick here

http://www.jbc.org/content/290/22/14045.full.html#ref-list-1

This article cites 46 references, 18 of which can be accessed free at

by guest on Novem


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/lookup/doi/10.1074/jbc.M115.640482

http://www.jbc.org/cgi/alerts?alertType=citedby&addAlert=cited_by&cited_by_criteria_resid=jbc;290/22/14045&saveAlert=no&return-type=article&return_url=http://www.jbc.org/content/290/22/14045

http://www.jbc.org/cgi/alerts?alertType=correction&addAlert=correction&correction_criteria_value=290/22/14045&saveAlert=no&return-type=article&return_url=http://www.jbc.org/content/290/22/14045

http://www.jbc.org/cgi/alerts/etoc

http://www.jbc.org/content/290/22/14045.full.html#ref-list-1

http://www.jbc.org/

Documents

NovelInteractionofClassIIbHistoneDeacetylase6(HDAC6 ... · 2015-05-22 · may29,2015•volume290•number22 journalofbiologicalchemistry 14045 cell movement via acetylation of -tubulin