Phosphorylation of the Anaphase Promoting Complex activator CDH1/FZR regulates the transition from Meiosis I to Meiosis II in mouse male germ cell

CDH1/FZR is an activator of Anaphase promoting complex/Cyclosome (APC/C), best known for its role as E3 ubiquitin ligase that drives the cell cycle. APC/C activity is regulated by CDK-mediated phosphorylation of CDH1 during mitotic cell cycle. Although the critical role of CDH1 phosphorylation has been shown mainly in yeast and in vitro cell culture studies, its biological significance in mammalian tissues in vivo remained elusive. Here, we examined the in vivo role of CDH1 phosphorylation using a mouse model, in which non-phosphorylatable substitutions were introduced in the putative CDK-phosphorylation sites of CDH1. Although ablation of CDH1 phosphorylation did not show substantial consequences in mouse somatic tissues, it led to severe testicular defects resulting in male infertility. In the absence of CDH1 phosphorylation, male juvenile germ cells entered meiosis normally but skipped meiosis II producing diploid spermatid-like cells. In aged testis, male germ cells were overall abolished, showing Sertoli cell-only phenotype. The present study demonstrated that phosphorylation of CDH1 is required for temporal regulation of APC/C activity at the transition from meiosis I to meiosis II, and for spermatoginial stem cell maintenance, which raised an insight into the sexual dimorphism of CDH1-regulation in germ cells.


Introduction
Anaphase promoting complex/Cyclosome (APC/C) controls timely transitions of mitotic cell cycle phases by promoting ubiquitylation and degradation of many key cell cycle regulators (Acquaviva and Pines, 2006). APC/C activity is regulated by either of two co-activators CDC20 and CDH1/FZR, which determine the substrate specificity of ubiquitylation for each cell cycle phase (Peters, 2006). APC/C CDC20 activity appears in metaphase-to-anaphase transition, when it has an essential function in promoting chromosome segregation by mediating cyclin B1 and securin degradation. APC CDH1 is thought to regulate a wide range of cell cycle events, in which more than 30 proteins have been identified as APC/C CDH1 substrates (Ramanujan and Tiwari, 2016).
While CDC20 plays a role in APC/C activity at metaphase, when the cyclin-dependent kinase 1 (CDK1) activity is high, CDH1 contributes to APC/C activity when CDK1 activity is sustained at a low level (Peters, 2006). APC/C CDH1 activity is negatively regulated by CDK-mediated phosphorylation of CDH1 during mitotic cell cycle. At mitotic exit, reduction of CDK1 activity leads to phosphatase-mediated dephosphorylation of CDH1 which then binds and activates APC/C until late G1 phase (Wurzenberger and Gerlich, 2011). In yeast mitotic cell cycle, CDH1 is phosphorylated by increased level of CDK activity after late G1/S onward, which subsequently leads to dissociation of CDH1 from APC/C and its inactivation (Zachariae et al., 1998) (Jaspersen et al., 1999) (Blanco et al., 2000) (Robbins and Cross, 2010) (Ondracka et al., 2016). In mammals, HeLa cells transiently transfected with mutant CDH1 that lacked potential CDK1 phosphorylation sites, resulted in premature reduction of cyclin A and cyclin B with decrease in G2/M phase-cells (Kramer et al., 2000). Thus, CDK-mediated phosphorylation of CDH1 plays a crucial role in temporal regulation of APC/C activity during mitotic cell cycle. However, the physiological significance of CDH1 phosphorylation is yet to be examined in vertebrate mitotic tissues in vivo.
The meiotic cell cycle consists of one round of DNA replication followed by two rounds of chromosome segregation, producing haploid gametes from diploid cells.
Although study in budding yeast elucidated that the role of APC/C in meiosis relies on the meiosis-specific co-activator AMA1, its homolog or counterpart does not exist in mammals. Instead, CDC20 and CDH1 contribute to the regulation of APC/C activity in mammalian meiosis. In mice, it was demonstrated that loss of CDH1 led to abnormal spermatogonial proliferation and defects in progression of meiotic prophase I (Holt et al., 2014).This indicates CDH1 is required for mammalian meiosis, but it remains elusive how APC/C activity regulated by phosphorylation of CDH1 is involved in meiotic cell cycle.
Here, we examined whether phosphorylation of CDH1 is required for the regulation of APC/C activity during mitotic and meiotic cell cycle in vivo, using knockin mice that carries non-phosphorylatable mutations of CDH1. Our study demonstrates that phosphorylation of CDH1 is required for spermatoginial stem cell maintenance for a long period of time. Furthermore, we show that CDH1 phosphorylation has a crucial role in the regulation of APC/C activity during meiosis I-meiosis II transition in juvenile male but not in female. Sexual dimorphism in the requirement of CDH1 phosphorylation raised an insight into different modes of meiosis I-to-meiosis II progression in spermatocytes and oocytes.

Generation of Cdh1 9A/9A knockin mice
It has been shown that human CDH1 is phosphorylated during mitotic cell cycle.
Ectopic expression of non-phosphorylatable CDH1 mutant in HeLa cells resulted in formation of constitutively active APC/C CDH1 and a reduction of G2/M phase (Kramer et al., 2000). To analyze the physiological role of CDH1-phosphorylation, we generated a mouse model, Cdh1 9A/9A knockin (Cdh1 9A/9A KI) mice, in which the nine putative CDK-phosphorylation sites of Ser/Thr residues in CDH1 were substituted with alanine amino acids ( Fig.1A 1E). We noticed that the level of CDC20 incorporated in APC/C in Cdh1 9A/9A testis was less than that in natural WT Cdh1 +/+ and Cdh1 Gt wt/Gt wt KI testes. This implies that persistent inclusion of CDH1-9A in APC/C impedes the association of CDC20 to APC/C.  were eliminated at least in part by apoptosis (Fig. 3F).

Phosphorylation of CDH1 is required for the entry into meiosis II in male meiosis
Given that Cdh1 9A/9A KI spermatocytes undergo meiosis I but do not produce haploid cells, the round spermatid-like cells may be a consequence of the failure of meiosis II.
Homologous chromosomes are segregated in meiosis I, whereas sister chromatids are separated in meiosis II. In order to examine the chromosome composition in Cdh1 9A/9A KI round spermatid-like cells, we performed immuno-FISH assays using a probe that detects a specific DNA sequence in the mid-arm region of chromosomes 3. In both WT and Cdh1 9A/9A KI mice, interkinesis spermatocytes showed a pair of FISH signals, indicating that homologous chromosomes were disjoined after meiosis I (Fig. 4A). In contrast, while WT round spermatids showed a single FISH signal, most of Cdh1 9A/9A KI round spermatid-like cells showed a pair of FISH signals (Fig. 4B), suggesting that in

Cdh1
9A/9A KI interkinesis spermatocytes failed to enter or complete meiosis II, and consequently led to non-disjunction of sister chromatids. Thus, these results suggested that the progression from interkinesis to meiosis II was regulated by phosphorylation of CDH1 in male meiosis. In mitotic cell cycle, constitutively active APC/C Cdh1 reduces cyclin A and cyclin B levels prematurely with decrease in G2/M phase cells (Kramer et al., 2000). Our observations in Cdh1 9A/9A KI male implies that without phosphorylation of CDH1, APC/C Cdh1 substrates might undergo premature degradation, which impairs male meiosis I -meiosis II transition.

Phosphorylation of CDH1 is required for maintaining spermatogonial stem cell population
We noticed that the loss of germ cells was more severe in aged Cdh1 9A/9A KI seminiferous tubules, with residual spermatogonia, spermatocytes and Sertoli cells remaining along the basement compartment of the tubules (Fig.5A), which was accompanied by complete absence of spermatozoa in the epididymis (Fig.5B).
Immunostaining of the aged Cdh1 9A/9A KI seminiferous tubules revealed that TRA98 positive cells were lost while SOX9 positive cells remained in a subpopulation of the tubules, suggesting that germ cell populations were depleted (Fig.5C). This trend was further augmented at older age in Cdh1 9A/9A KI seminiferous tubules (Fig.5D). The aged Cdh1 9A/9A KI seminiferous tubules showed a severe loss of male germ cells, which was similar to Sertoli cell-only phenotype. This implied spermatogonial stem cell population was depleted over a long period of time in Cdh1 9A/9A KI testis. Therefore, it is plausible that phosphorylation of CDH1 is required also for maintaining the spermatogonial stem cell population.

Discussion
Contrary to previous observation in human cell culture study (Kramer et al., 2000), Crucially, we have shown that introduction of non-phosphorylatable mutations of CDH1 results in spermatogenesis defect and subsequent male infertility (Fig 2), which cannot be compensated as in mitotic cell cycle. The defects in the Cdh1 9A/9A KI testes primarily arose from a process during meiosis I-meiosis II transition producing diploid round spermatid-like cells (Figs 3 and 4). This phenotype is reminiscent of the  (Jaspersen et al., 1999) (Blanco et al., 2000) (Kramer et al., 2000), it is possible that constitutively active APC/C CDH1-9A compromise the levels of its critical substrates, such as cyclin A and cyclin B, which in turn prevent the CDK activity required for nuclear breakdown at interkinesis and entry into meiosis II. Alternatively, constitutive binding of CDH1-9A may compete with another activator CDC20 for the association with APC/C before meiosis II, which in turn raises a situation that APC/C CDC20 specific substrates fail to undergo destruction in meiosis II.
We have also shown that the Cdh1 9A/9A KI testes show Sertoli-cell only phenotype at older age (Fig 5). Depletion of germ cells in Cdh1 9A/9A KI testes may imply that constitutive APC/C Cdh1-9A activity enforces persistent transition of spermatogonial stem cell population into G1/S, preventing their entry into G0 quiescence state. Thus, CDK-mediated phosphorylation of CDH1 is required for sustaining spermatogonial stem cells over a long period of time.
Overall, temporal regulation of CDH1 by phosphorylation is essential for the maintenance of spermatogonial stem cell population and meiosis II transition in male germ cells.

Declaration of interests:
The authors declare no competing interests. supervised experiments and conducted the study. K.I. and N.T. wrote the manuscript.
Whenever possible, each knockin animal was compared to littermates or age-matched non-littermates from the same colony, unless otherwise described. Since Cdh1 +/+ ,
To produce ES cells in which the β-geo gene cassette of Cdh1 +/GT cells was replaced with cDNA encoding mouse wild type Cdh1 (Cdh1 for 1 day to isolate cell lines that had undergone Cre-mediated recombination. Puromycin selection was performed twice at a 2-day interval. To detect the expression from the Cdh1 wt and Cdh1 9A knock-in (KI) alleles in the recombinant ES cell lines, we performed reverse transcription-PCR (RT-PCR) analysis using the primers 5NC-s2 (5'-TCGAACAGGCGCGGCGTGTT-3') and mFzr as2 (5'-ATAGTCCTGGTCCATGGTG GAG-3'). The PCR product was cloned into the pGEM-T easy vector (Promega) and sequences were verified. Chimera mice were generated by morula injection (host ICR) of recombinant ES cells. Chimeric males were mated to C57BL/6N females and the progenies were genotyped by PCR using the following primers.  and Gas7 (5'-CTCCAAGGCCTTTGTGAGGC-3') for the wild-type allele (0.4kb).  for the knock-in allele (0.7kb).

Preparation of testis extracts and immunoprecipitation
Testis extracts were prepared as described previously (Ishiguro et al., 2011

Antibodies
The following antibodies were used for immunoblot (

Histological Analysis
For, hematoxylin and eosin staining, testes, epididymis and ovaries were fixed in 10%

Immunostaining of spermatocytes
Spread nuclei from spermatocytes were prepared as described (Ishiguro et al., 2014).
Briefly testicular cells were suspended in PBS, then dropped onto a slide glass together with an equal volume of 2% PFA, 0.2% (v/v) Triton X-100 in PBS, and incubated at room temperature in humidified chamber. The sides were then air-dried and washed with PBS containing 0.1% Triton-X100 or frozen for longer storage at -80ºC. The serial sections of frozen testes were fixed in 4% PFA for 5 min at room temperature and permeabilized in 0.1% TritonX100 in PBS for 10 min. The slides were blocked in 3% BSA/PBS, and incubated at room temperature with the primary antibodies in a blocking solution. After three washes in PBS, the sections were incubated for 1 h at room temperature with Alexa-dye-conjugated secondary antibodies (1:1000; Invitrogen) in a blocking solution. DNA was counterstained with Vectashield mounting medium containing DAPI (Vector Laboratory).

Fluorescence in situ hybridization (FISH) on immunostained nuclei.
For immuno-FISH, structurally preserved nuclei (SPN) from spermatocytes were prepared as described (Ishiguro et al., 2014) with modification. Briefly testicular cells were collected in PBS by mincing seminiferous tubules into small pieces with fine-tipped tweezers and then pipetting. After removal of tissue pieces, the cell suspension was filtered through a Cell strainer (BD Falcon) to remove debris. The cell suspension (~5 μl) was dropped onto a MAS-coated slide glass (Matsunami) and fixed with 10 μl of 2% Paraformaldehyde (PFA)/100 mM sucrose in PBS for 10 min followed by the addition of 1.5 μl of 1.25 M Glycine/PBS, and then air-dried at room temperature.
Immediately before they were completely air-dried, the slide glasses were washed with PBS containing 0.1% Triton-X100 or frozen for longer storage at -80ºC.
Immuno-stained samples of SPN were fixed in 4% paraformaldehyde for 5 min, washed with PBS, and subjected to sequential dehydration through 70%, 80%, 90%, 100% ethanol. Immuno-stained SPNs were denatured in 50% formamide, 2x SSC at 72°C for 10 min. Hybridization was conducted with a fluorescence-labeled point probe in buffer containing 50% formamide, 2x SSC, 20% dextran sulfate at 37°C for 12-16 h. The slides were washed sequentially at room temperature in 2x SSC for 1 min, 0.4x SSC/0.3% Tween20 solution for 2 min, and 2xSSC at room temperature for 1 min. The mouse point probe derived from BAC clone RP23-6I6 detects the mid region of chromosome 3.

Imaging
Immunostaining images were captured with DeltaVision (GE Healthcare). The projection of the images were processed with the SoftWorx software program (GE Healthcare). For Fig S2B, images were captured with Zeiss LSM-710 confocal microscope and processed with ZEN software. All images shown were Z-stacked.
Bright field images were captured with OLYMPUS BX53 fluorescence microscope and processed with CellSens standard program. For counting seminiferous tubules, immunostaining images were captured with BIOREVO BZ-X710(KEYENCE), and processed with BZ-H3A program.