Overexpression of Hdac6 extends reproductive lifespan in mice

Histone deacetylase 6 (Hdac6) was discovered as a deacetylase of α-tubulin and functions in cell migration, immunity and resistance to virus infection in vitro (Hubbert et al., 2002; Valenzuela-Fernandez et al., 2008). Overex-pression of Hdac6 enhances resistance to virus infection in embryonic stem (ES) cells and in mice (Wang et al., 2015). Hdac6 also can function to deacetylate protein and is involved in protein ubiquitination and degradation (Seigneurin-Berny et al., 2001; Zhang et al., 2014), and self-clearance of misfolded proteins, promoting autophagy and preventing neurodegeneration (Lee et al., 2010; Pandey et al., 2007). HDAC6 also is implicated in DNA damage response and depletion or inhibition of HDAC6 induces DNA damage and apoptosis (Namdar et al., 2010; Zhang et al., 2014), suggesting that HDAC6 could be important for DNA repair and integrity maintenance. DNA damage accumulates with somatic aging and reproductive aging (Titus et al., 2013), and enhanced DNA damage repair can reverse aging (Maynard et al., 2015). Reproductive aging in females mainly results from ovarian senescence, as shown by reduced number and quality of follicles and germ cells. Interestingly, Hdac6 is highly expressed in germ cell tissues, testis and spermatogenic cells, relative to other somatic tissues, such as liver, heart, muscle, spleen and kidney (Seigneurin-Berny et al., 2001; Zhang et al., 2008), and yet Hdac6-deficient mice are viable and fertile, and show normal development and function of testis (Zhang et al., 2008). Initially, we tested whether DNA damage is reduced in Hdac6 overexpression embryonic stem (ES) cells, which expressed higher Hdac6 protein levels and lower levels of α-acetylated tubulin than did WT ES cells (Wang et al., 2015). Treatment of mouse embryonic fibroblasts (MEFs) with mitomycin C can generate mitotically inactive feeder cells and these cells showed high levels of DNA damage as shown by many γH2AX foci served as positive control (Fig. S1A and 1B). Fewer γH2AX foci were seen in WT ES cells, and comparatively, even fewer γH2AX foci in Hdac6 overexpression ES cells than in WT ES cells (Fig. S1A and 1B). Also, Hdac6 overexpression ES cells expressed lower γH2AX protein levels than did WT ES cells (Fig. S1C). These data show that high levels of Hdac6 can reduce DNA damage or induce more robust DNA repair. Further, we compared telomere lengths of Hdac6 over-expression ES cells with those of WT ES cells. Telomeres were longer in Hdac6 overexpression ES cells than in WT ES cells shown …


Mice and breeding
Hdac6 overexpression transgenic mice used in this study were generated previously (Wang et al., 2015), and age-matched WT mice served as aging controls (12-21 months old). Young male ICR mice at the age of 2-3 month old were used for breeding to test fertility of females. Late in the afternoon, one female mouse was placed in the cage of one male for mating. On the following morning, successfully mated female mice with plug were returned to their cages, and those without mating plugs were mated with a young male for another night. Females with no plugs for three successive days were not used for breeding experiments. The number of offspring within one week after delivery was recorded. Hdac6 knockout (KO) mice were in 129/C57BL6 mixed genetic background, and were obtained from Tso-Pang Yao (Liu et al., 2015). All mouse experiments were carried out in accordance with the guidelines and regulations and approved by the Institutional Animal Care and Use Committee of Nankai University.

Serial sections of ovary or testis and follicle count
Ovaries or testis were collected and fixed by immersion in 4% paraformaldehyde for 8-24 h, and tissues were embedded with paraffin wax. Based on previous methods (Liu et al., 2013), serial sections (5 µm) from each ovary were aligned in order on glass microscope slides, stained with hematoxylin and eosin Y, and analyzed for the number of follicles in three or four different developmental stages in every fifth section with random start in the first five sections. The total number of follicles per ovary was calculated by cumulating the count of every fifth section throughout the whole ovaries.
The follicles were categorized into primordial and primary, secondary, antral and preovulatory, accordingly (Myers et al., 2004). Follicles were classified as primordial and primary if they contained an oocyte surrounded by a single layer of squamous or cuboidal granulosa cells. Follicles at an intermediate-stage also were scored in this group. Secondary follicles were identified as having more than one layer of granulosa cells with no visible antrum. Antral follicles possessed one or two small areas of follicular fluid (antrum) or a single large antral space. Preovulatory follicles had a rim of cumulus cells surrounding the oocyte.
Total RNA was isolated from tissues using RNeasy mini kit (Qiagen), and subject to cDNA synthesis using Moloney Murine Leukemia Virus Reverse Transcriptase (Invitrogen). PCR reactions were set up in duplicates using the FastStart Universal SYBR Green Master (Roche) and run on the Mastercycler® RealPlex2 real time PCR detection system (Eppendorf). At least two parallel samples were run for analysis of each gene. Primers were designed using the IDT DNA website and listed in Table S1. The final PCR reaction volume in 20µl contained 10µl SYBR Green PCR Master Mix, 1µl cDNA template, 2µl primer mixture and 7µl water. Thermal cycling was carried out with a 10 min denaturation step at 95 °C, followed by two-step cycles, 15s at 95°C and 1min at 60°C.

Immunocytochemistry and fluorescence microscopy
Briefly, after deparaffinizing, rehydrating and washing in 0.01 M PBS (pH 7.2-7.4), sections were incubated with 3% H2O2 for 10 min at room temperature to block endogenous peroxidase, subjected to high pressure antigen recovery sequentially in 0.01% citrate buffer (pH 6.0) for 3 min, incubated with blocking solution (5% goat serum and 1% BSA in PBS) for 2 h at room temperature, and then incubated with the diluted primary antibodies overnight at 4°C. Blocking solution without the primary antibody served as negative control. After washing with PBS, sections were incubated with appropriate secondary antibodies (Alexa Fluor® 568, 488 or 594, Invitrogen). The sections were then stained with 1 µg/ml Hoechst 33342 for 10 min to

Immunohistochemistry
Immunocytochemistry was performed based on the method described previously (Yuan et al., 2013).

TUNEL assay
For detection of apoptosis cells, the In situ cell death detection kit was applied (Roche). Briefly, after deparaffinizing, rehydrating and washing in 0.01 M PBS (pH 7.2-7.4), sections were incubated in 0.1% TritonX-100 for 8-10 min. Then TUNEL reaction mixture was applied to the sections. After being stained with 1 µg/ml Hoechst33342 and mounted in Vectashield mounting medium (Vector Laboratories, CA, USA), sections were imaged using a Zeiss Axio Imager and apoptotic nuclei counted.

Telomere measurement
Telomere lengths were measured by qPCR (Callicott and Womack, 2006), and telomere Q-FISH following the protocol described (Dan et al., 2014). For qPCR of telomere length, the telomere signal was normalized to the signal from the single-copy gene (36B4) to generate a T/S ratio indicative of relative telomere length of the given sample. The TRF analysis by southern blot was performed using a commercial kit (TeloTAGGG Telomere Length Assay, catalog no. 12209136001, Roche Diagnostics), based on the method described previously (Blasco et al., 1997), with slight modifications. Cells were isolated and embedded in agarose plugs (Pulsed Field Certified Agarose, 162-0137, Bio-Rad) to let plugs containing 5 × 10 5 cells and treated with Proteinase K (PCR Grade, 03115879001, Roche Life Science). The plug was digested with MboI (R0147L, NEB) for 15 h and underwent electrophoresis through a 1% agarose gel in 1XTAE at 14°C for 16 h at 6 V/cm at an initial pulse for 1 s and ending in 12 s using the Bio-Rad CHEF DR-III pulse-field system.

ChIP-qPCR
ChIP-qPCR analysis of H3K9ac and H3K9me3 was performed as previously described (Dan et al., 2014). Briefly, ~5 × 10 7 cells were fixed with 1% paraformaldehyde, lysed, and sonicated to achieve the majority of DNA fragments with 100-1000 bp. DNA fragments were then enriched by immunoprecipitation using 7 µg H3K9ac antibody (04-1003, Millipore) or 5 µg H3K9me3 antibody (ab8898, Abcam). The eluted protein:DNA complex was reverse-crosslinked at 65 °C overnight. DNA was recovered after RNase A and proteinase treatment. ChIP enriched DNA was analyzed by real-time qPCR and β-actin locus served controls. Primers are listed in Table S2.

Genome-wide gene expression by microarray analysis
ES cells were maintained on mitomycin-C treated mouse embryonic fibroblasts (feeders) in ES cell medium. Trizol extraction of total RNA was performed according to the manufacturer's instructions. Microarray was performed by CapitalBio Corporation (Beijing, China) using SurePrint G3 Mouse GE 8x60K Microarray (Agilent Technologies). The analysis was carried out according to the manufacturer's protocol, described in details on the website of CapitalBio (http://www.capitalbio.com). Only probe sets showing more than 1.5-fold change were retained in the final list. We then performed hierarchical clustering with the differentially expressed genes using cluster software (version 3), and the data has been deposited (GSE81710).

Statistical analysis
Statistical analyses were performed by ANOVA and means compared by Fisher's protected least-significant difference (PLSD) using StatView software from SAS Institute Inc. (Cary, NC). P-value <0.05 or lower was considered statistically significant.