Deletion of inositol polyphosphate 4-phosphatase type-II B affects spermatogenesis in mice.

Inositol polyphosphate-4-phosphatase type II (INPP4B) is a dual-specificity phosphatase that acts as a tumor suppressor in multiple cancers. INPP4B dephosphorylates phospholipids at the 4th position of the inositol ring and inhibits AKT and PKC signaling by hydrolyzing of PI(3,4)P2 and PI(4,5)P2, respectively. INPP4B protein phosphatase targets include phospho-tyrosines on Akt and phospho-serine and phospho-threonine on PTEN. INPP4B is highly expressed in testes, suggesting its role in testes development and physiology. The objective of this study was to determine whether Inpp4b deletion impacts testicular function in mice. In testis, Inpp4b expression was the highest in postmeiotic germ cells in both mice and men. The testes of Inpp4b knockout male mice were significantly smaller compared to the testes of wild-type (WT) males. Inpp4b-/- males produced fewer mature sperm cells compared to WT, and this difference increased with age and high fat diet (HFD). Reduction in early steroidogenic enzymes and luteinizing hormone (LH) receptor gene expression was detected, although androgen receptor (AR) protein level was similar in WT and Inpp4b-/- testes. Germ cell apoptosis was significantly increased in the knockout mice, while expression of meiotic marker γH2A.X was decreased. Our data demonstrate that INPP4B plays a role in maintenance of male germ cell differentiation and protects testis functions against deleterious effects of aging and high fat diet.


Animal studies
Mice were maintained at the AAALAC accredited animal facility at Florida International University and all experimental protocols were performed in accordance with the regulations of the Institutional Animal Care and Use Committees at FIU and the National Academy of Science Guide for Care and Use of Laboratory Animals. The Institutional Animal Care and Use Committees at Florida International University approved this research, protocol AN18-055.
Generation of conventional knockout Inpp4b -/- [22,23] and cryptorchid Rxfp2 -/- [24] mouse models were described previously. Mice with the Inpp4b knockout allele were backcrossed to FVB/N inbred strain for 4 generations and then intercrossed to obtain Inpp4b -/homozygotes in order to decrease genomic background variability. Mice were fed with low fat diet (LFD) with 12.9% fat, 63.8% carbohydrate, and 23.2% protein (total 13.6 kJ/g) (LabDiet 5V75, St. Louis, MO) or high fat diet (HFD) with 59.4% fat, 25.7% carbohydrate, and 14.9% protein (total 22.8 kJ/g) content (TestDiet 58R3, St. Louis, MO) [25]. For the HFD group, the females were on HFD for a month prior, during the pregnancy, and after delivery until weaning and the pups were on HFD from weaning until the euthanasia [25]. It was shown that under this protocol, male pups displayed more drastic changes in the urogenital system. Agematched WT and Inpp4b -/male mice were used in our studies at 2, 3, 4 and 6 months of age. Mice were euthanized by isoflurane (Patterson Veterinary, Greeley, CO) overexposure and testes and seminal vesicles were dissected and weighed.
Sperm count was performed as described by Huang et al [24], with minor modification. Briefly, sperm was released from the cauda epididymis into PBS, then diluted at a 1:5 ratio with distilled water, loaded into a hemocytometer and counted manually under the microscope by two investigators.

Hormone measurement
Testosterone (T) and LH levels were determined in male blood serum. The mice were euthanized, and blood collected from 3-month old mice, with 6 males per group. The serum was collected after centrifugation at 3000 × g for 15 min and frozen until hormonal analysis. Testosterone and LH levels were determined in the University of Virginia Center for Research at the Reproduction Ligand Assay and Analysis Core (University of Virginia, Charlottesville, VA) using mouse/rat testosterone ELISA (IBL America, Minneapolis, MN) and RIA (in house protocol) assays respectively.

Gene expression analysis
mRNA samples were isolated from mouse testes using Tri-Reagent (Molecular Research Center, Cincinnati, OH) and reverse transcribed using the Verso cDNA synthesis Kit (Thermo Fisher Scientific, Waltham, MA). For quantitative PCR, primers were designed using online software (Roche, Basel, Switzerland) and probes were purchased from Universal ProbeLibrary (Roche). Roche 480 LightCycler (Roche) was used for probe-based real-time PCR. GoTaq qPCR master mix (Promega, Madison, WI) was used for experiments using BRYT dye on the Mastercycler RealPlex 2 system (Eppendorf, Westbury, NY). The relative fold change in gene mRNA level was calculated by the comparative cycle threshold (2 -ΔΔCt ) method using 18S rRNA for normalization of the expression data. Primer sequences are shown in Table 1. The number of animals used in each group is shown in figure legends.

Western blotting
Testes from WT, Inpp4b -/and cryptorchid Rxfp2 -/mice were homogenized with a glass tissue grinder in ice-cold RIPA buffer supplemented with protease and phosphatase inhibitors [23]. The lysates were diluted tenfold and 20-30 μg of protein were resolved on SDS-PAGE and transferred to PVDF membrane. For immunoblotting, rabbit polyclonal primary antibody against AR (1:1000 dilution, catalog number 06-680, Millipore, Carlsbad, CA) and mouse monoclonal β-tubulin (1:5000, #05-661, Millipore) were used. Signal was visualized using ImageQuant LAS 500 imaging system (GE Healthcare Life Sciences, Marlborough, MA) and quantitative analysis was performed with ImageQuant TL software.

Propidium iodide staining and flow cytometry
Testis digestion and flow cytometry experiments were performed as previously described [26]. Six-month old WT and Inpp4b -/mice were used (n = 4 for each group). Seminiferous tubules were washed with 1X HBSS, digested with collagenase and trypsin, and filtered. Next, cells were washed with 0.1% BSA in PBS, centrifuged at 400 g for 2 minutes, counted, and resuspended at 2x10 6 cells/ml. Cells were fixed in 70% ethanol and stained for 30 minutes with 25 μg/ml of propidium iodide dissolved in 0.1% BSA in PBS. Analysis was performed using Accuri C6 flow cytometer (Becton-Dickinson, Franklin Lakes, NJ).

Histology and immunohistochemistry (IHC)
Testis samples were fixed in 4% PFA overnight and embedded in paraffin. The embedded tissue was sectioned at 4.5 μm. H&E staining and IHC were performed as previously described [27]. Rabbit polyclonal antibodies to INPP4B (1:150 dilution, #8450, Cell Signaling, Danvers, MA) and γH2A.X (1:700, #2577, Cell Signaling, Danvers, MA) were used as primary antibodies and sections were counterstained with hematoxylin (Millipore). For TUNEL assay, Apop-Tag Plus Peroxidase In Situ Apoptosis Detection Kit (Millipore) was used following manufacturer protocol. Three mice were analyzed for each group and minimum ten circular tubules were counted per mouse. The images were captured using a Carl Zeiss Axio A1 microscope with an AxioCam MRc5 CCD camera (Carl Zeiss, New York, NY).

Statistical analysis
Student's t-test for two groups, one-way ANOVA, and two-way ANOVA for more than two groups were used to assess the significance of differences using Prism 7.0 software (GraphPad Software, La Jolla, CA). All data are presented as mean ± SEM. p values less than 0.05 (p<0.05) were accepted as significant. The number of samples analyzed is shown in figure legends.

INPP4B expression is highest in post-meiotic germ cells
To understand the function of INPP4B in the testes, we analyzed the expression pattern of INPP4B in various testicular cell populations. Using a previously generated transcriptional cell atlas [19] derived from human testicular single-cell RNA sequencing, we detected the highest level of INPP4B expression in the round and elongating spermatids and in differentiating sperm ( Fig 1A). A lower level of expression was also present in Leydig and Sertoli cells. Interestingly, PTEN was expressed ubiquitously in stromal and germ cells of the testis ( Fig 1A).
Clustering gene analysis and gene ontology analysis showed that the top 222 genes upregulated with correlation coefficient cut-off value 0.4 in INPP4B-positive cells are involved in fertilization, reproductive development, cellular signaling pathways and various stages of spermatogenesis ( Fig 1B). Thus, INPP4B-positive cells represent germ rather than somatic cell populations in testis. We next used the microarray data from Chalmel et al. [28], and analyzed INPP4B expression in the testes of men with cryptorchidism or infertility and compared it to INPP4B expression in healthy adult individuals. We divided the infertile population into two groups, those who had spermatids present in the seminiferous tubules and those who did not. Among all groups, INPP4B expression was highest in healthy controls and it was significantly decreased in cryptorchid testes. Within the infertile population, INPP4B expression was significantly lower in the group lacking spermatids (Fig 2A).
The first cycle of spermatogenesis in mice is synchronized over seminiferous tubules. This timely process allows the comparison of the gene expression within distinct stages of germ cell differentiation. We compared Inpp4b and Pten expression in the testes at three important time-points during germ cell development: day 8, when the testes contain pre-meiotic diploid differentiating germ cells up to the type B spermatogonia stage; day 20, when the germ cells are differentiated into round spermatids; and after day 35, when testes contain all stages of germ cells [29]. We also analyzed adult testes from 2-month old Rxfp2 -/males with high intraabdominal cryptorchidism, which lack all stages of spermatogenesis past the spermatogonial cells, with only a few early spermatocytes present [26]. When compared to adult WT mice, Inpp4b expression in these mutants was significantly lower in the testes of 8-and 20-day old mice, and cryptorchid adult mice ( Fig 2B). Cryptorchid testes show a somewhat lower level of expression compared to 20-day old WT testis ( Fig 2B), although this difference does not reach statistical significance (p = 0.868). Thus, as in human, in mouse germ cells Inpp4b expression was highest during the late stages of spermatogenesis. The variation of PTEN expression was significantly less pronounced in the same groups ( Fig 2C). IHC staining indicated robust expression of INPP4B was present in elongated spermatids and lower expression was observed in round spermatids, Leydig and Sertoli cells in WT mouse testes ( Fig 2D).

Testis weight and sperm counts are decreased in Inpp4b -/males
To characterize the effect of INPP4B loss, we measured body, testes, and seminal vesicle weight and the epididymal sperm count of 2-, 3-, 4-and 6-month old WT and Inpp4b -/mice. Body weight remained comparable among age groups until 6-months of age, when Inpp4b -/mice weighed slightly less (Fig 3A). In Inpp4b -/males, testes weight was consistently smaller when compared to WT controls, and this difference increased with age ( Fig 3B). There were no statistically significant differences in seminal vesicle weight between WT and Inpp4b -/mice at any age (Fig 3C) or visible changes in mutant epididymides. Epididymal sperm counts were higher in the WT groups. The difference was not statistically significant between 2-month old

PLOS ONE
and 3-month old males, but it became significant in 4-and 6-month old groups. (Fig 3D). Additionally, there were no differences in the diameter of the seminiferous tubules between WT and Inpp4b -/testes at any age group (S1 Fig).

INPP4B deficiency causes a shift from haploid to diploid cell population in testis
Since Inpp4b -/males had significantly decreased sperm counts in aged groups, we used flow cytometry to evaluate changes in the ploidy of germ cells in 6-month old WT and mutant testes [30,31]. We compared the percentages of testicular haploid, diploid, and tetraploid populations. In the haploid populations, we examined round (1C) and hypostained (HC) elongated spermatids separately to determine whether the reduction in sperm counts was due to a reduction in meiosis (round spermatids) or spermiogenesis (elongated spermatids) stages (Fig 4A  and 4B) [32]. The percentage of elongated spermatids was significantly decreased in Inpp4b -/males (WT-21.8%, Inpp4b -/--8.7%, p = 0.0032), whereas the percentage of diploid cells in

High fat diet exacerbates the effects of INPP4B deficiency on testis functions
Since INPP4B is a key regulator of PIP substrates that participate in metabolic signaling pathways, we examined if a high fat diet (HFD) can exacerbate the effects of INPP4B loss on spermatogenesis. Body weight and seminal vesicle weights showed no difference when comparing 3-month old Inpp4b -/and WT mice fed either diet (Fig 5A and 5B). The testes of Inpp4b -/mice weighed significantly less than testes of the WT controls ( Fig 5C) and the consumption of a HFD significantly decreased sperm count in Inpp4b -/mice when compared to WT ( Fig  5D). The LFD group is the same as the 3-month old group used in Fig 3. Since the hypothalamic-pituitary-gonadal (HPG) hormonal axis has an important role in the regulation of spermatogenesis [33,34], we compared the levels of testosterone and luteinizing hormone (LH) in the serum of mutant and WT males on a LFD or a HFD. While there was a trend suggesting that Inpp4b -/mice produce less testosterone than WT controls, this difference was not statistically significant (Fig 5E). In testis, LH stimulates the production of testosterone in Leydig cells. Notably, in the HFD group, the LH levels were significantly lower in the Inpp4b -/group than in WT mice (Fig 5F).
Steroidogenic hormone synthesis in testis is controlled by a series of enzymatic steps ( Fig  6A). We compared the gene expression of the key steroidogenic enzymes Cyp11a1, Cyp17a1, Hsd3b6, Hsd17b3, Cyp19 and Srd5a1, steroidogenic factor-1 Nr5a1, the cholesterol transporting protein Star, and the LH receptor, Lhcgr, in WT and mutant testes in LFD and HFD groups. Cyp11a1 expression was significantly lower in the Inpp4b -/-LFD group when compared to WT mice, but this decrease was not significant in HFD groups (Fig 6B). Cyp17a1 and Lhcgr expression was significantly lower in the LFD mutant group and in mice on HFD when

PLOS ONE
compared with the WT LFD group (Fig 6B). Srd5a1 expression was not changed between LFD groups, but it was significantly decreased in Inpp4b -/-HFD mice when compared with the WT HFD group. The expression of Nr5a1, Star, Hsd3b6, Hsd17b3, and Cyp19 were not affected by the loss of INPP4B or the diet (Fig 6B).
Additionally, there was no difference in testicular androgen receptor (AR) protein levels between WT and Inpp4b -/groups (Fig 7). There was a significantly higher level of AR expression in the cryptorchid testes, likely due to the lack of a germ cell population beyond the spermatogonia stage. As the germ cells do not express AR, cryptorchid testes have a higher proportion of AR-positive stromal cells compared to WT [35].

Loss of INPP4B increases apoptosis rate in early spermatogenesis
A decrease in the haploid testicular cell population and an increase in diploid cells indicated an abnormality in meiosis or spermiogenesis. We analyzed the rate of apoptosis in testes from 3-month old WT and Inpp4b -/testes in LFD and in age-matched HFD groups (Fig 8). The apoptosis rate was significantly higher in the mutant group compared to WT testes, independent of the diet (Fig 8B). Apoptosis in testis was mainly observed in spermatogonia cells and in primary spermatocytes (Fig 8C).
We next analyzed the expression of a meiosis marker, phosphorylated form of histone 2A.X (γH2A.X), which marks the double strand breaks in preleptotene through zygotene spermatocytes and sex bodies in pachytene spermatocytes. It is highly expressed in B type spermatogonia, primary spermatocytes from preleptotene to pachytene stages and elongated spermatids [36][37][38][39] (Fig 9A). Based on their position within the seminiferous tubules and cellular morphology at stage VIII-XI, the γH2A.X-positive cells were primary spermatocytes from preleptotene to pachytene spermatocytes, a population of actively dividing cells that determines efficiency of sperm production, and elongated spermatids (Fig 9C). The number of γH2A.X-positive cells were significantly lower in the LFD Inpp4b -/group compared to LFD WT controls (p = 0.0178). Consistent with reduced sperm count in HFD Inpp4b -/males, there was a significantly lower number of γH2A.X-positive cells in the HFD Inpp4b -/group when compared to HFD WT (p = 0.0348) or LFD mutant testes (p = 0.0258) (Fig 9B). Among the prophase I primary spermatocytes, the γH2A.X positive pachytene spermatocyte count was significantly higher in LFD WT males (37.63 cells per tubule) than in the LFD Inpp4b -/-(31.82 cells/per tubule, p = 0.0135) and HFD Inpp4b -/-(26.30 cells per tubule, p = 0.0002) groups.

Expression of cytokines is altered in INPP4B deficient testis
Cytokines like interleukins (ILs) and tumor necrosis factors (TNFs) have been shown to play an important role in testicular homeostasis and spermatogenesis [40]. In testis, IL1β, IL6 and TNFα are produced by Sertoli, Leydig, and germ cells in a cyclical manner, stimulating Sertoli cell function, spermatogenesis, and steroidogenesis in Leydig cells [41]. We compared the expression of Il1b, Il6 and Tnfa in 4-month old WT and Inpp4b -/testis on LFD (Fig 10A) and in 3-month old WT and Inpp4b -/testes in mice on LFD and HFD (Fig 10B). At three months no significant difference was observed in Il1b and Il6 expression, however, Il1b and Il6 expression was significantly higher in the WT than in Inpp4b -/testes in 4-month old males. The expression of Tnfa showed no significant difference between any groups (Fig 10).

Discussion
Recent data demonstrated that PIP signaling plays an important role in spermatogenesis and germ cell maintenance [9]. The involvement of members of this pathway in testicular functions was revealed through analysis of loss-of-function mutants. It was shown the PTEN/ PI3K/Akt pathway is important in controlling the proliferation and division of spermatogonial stem cells in mouse testis. Disruption of this signaling through knockout of the key genes in this pathway leads to the loss of spermatogonial cells and infertility in males [9,42,43]. Studies performed on Drosophila indicate that PI(4,5)P2 and PI(4)P are central regulators in germ cell meiosis and spermiogenesis, however the role of this pathway in mammals is less clear.
In this study we analyzed the expression and function of INPP4B in testes. INPP4B antagonizes the PI3K-AKT/PKB signaling pathway by dephosphorylating phosphoinositides and thereby modulating cell cycle progression and cell survival. We have shown that during spermatogenesis, in mice and men, INPP4B expression significantly increases at the round spermatid stage and continues into the mature sperm stage. Infertile men with no spermatids and cryptorchid male mice with spermatogenesis arrest at the spermatogonial-spermatocyte stage both had significantly lower levels of testicular INPP4B expression. Analysis of testicular phenotype in Inpp4b -/males showed reduced testis weight, lower sperm count, increased apoptosis rate, and lower LH concentrations. The expression of certain enzymes mediating early stages of androgen synthesis was also reduced in mutant testes. Importantly, some of these abnormalities were exacerbated in animals maintained on a high fat diet, suggesting that INPP4B plays a role in male germ cell differentiation. In both mice and men, high fat diet correlates with impaired intratesticular signaling and spermatogenesis [44,45]. PKC pathway dysregulation caused by diet with high fat content leads to disruption in several pathways such as lipid metabolism and reactive oxygen species (ROS) formation, which are important for normal testis functions [46,47].The synergistic stimulatory effect of Inpp4b loss and HFD on the PKC pathway might have a negative impact on testosterone metabolism and spermatogenesis. Expression of steroidogenic enzyme Cyp17a1 and luteinizing hormone receptor Lhcgr were decreased in the HFD groups which, in combination with the decrease of Srd5a1, led to a reduction in sperm count in Inpp4b -/mice when compared to the LFD group. Increased germ cell apoptosis and the reduction of expression of meiotic marker γH2A.X in the HFD mutant group support this conclusion. It should be pointed out that the relatively modest effects of the HFD in younger males might be due to the resistance of the FVB mouse strain to HFD induced obesity [48].
Analysis of previously reported gene expression datasets revealed a high level of expression of INPP4B in human and mouse testes [15,17]. In cryptorchid mice, in which the testes are devoid of germ cells beyond the early spermatocyte stage, there was a low level of Inpp4b gene expression. The same was true when we analyzed previously published data on gene expression in infertile men with no detectable spermatids [26]. The advance of single cell RNA sequencing allowed us to map the highest INPP4B expression to postmeiotic germ cells beginning from the round spermatids; the conclusion was also supported by IHC analysis in mouse testis. Consistently, the genes positively correlated with expression of INPP4B were associated with pathways activated during spermatogenesis, such as spermatid differentiation, flagellum formation, and fertilization. The question arises as to whether INPP4B plays any functional role in these processes or whether it is just a marker for specific stages of spermatogenesis. Our data suggest that the former might be true. INPP4B mainly catalyzes the hydrolysis of the phosphate located in the 4 th position of inositol ring of PI(3,4)P2, PI(4,5)P2, and inositol 1, 3, 4-trisphosphate. Experiments in Drosophila clearly show that PI4P cell signaling in germ cells has the same pattern: it is most prominent in spermatocyte through spermatid stages of differentiation, showing modest effects on the premeiotic population. Germline stem cells are affected by the PI3K pathway, mediated by insulin receptor and FOXO transcription factors. All these pathways are nutritionally regulated [9]. Activation of the PI3K pathway (especially isoform PI3KDN) in somatic cyst cells improves the transition from germline stem cells to spermatocytes [49]. PI4P and PI(4,5)P2 are regulators of meiotic cytokinesis in Drosophila spermatocytes [9]. Deletion of four wheel drive (fwd) encoding Drosophila PI 4-kinase IIIβ (PI4KIIIβ) [50], Class I PI transfer protein (PITP) [51,52], trafficking factors such as GOLPH3, a Golgi PI4P-binding protein [52], and other members of PIP pathway all lead to abnormal cytokinesis. Involvement of the homologous genes in spermatogenesis in mice is less clear. Here we showed that INPP4B is highly expressed in postmeiotic germ cells during spermatogenesis in mice and men, and the deletion of this gene in mice leads to a decrease in mature sperm. However, the mild effect of Inpp4b deficiency is likely due to compensation by other members of the pathway, by modifier genes, or the relatively young age of the analyzed mice.
The deletion of another member of the INPP family, INPP5B, causes male infertility in mice due to a reduced sperm count, motility, and fertilization defects [53]. Interestingly, that backcrossing of the mutant allele on FVB/N inbred background, the same genetic background UH2A.X-positive cells (brown staining) at prophase I in stage VIII-XI tubules were counted under 40X objective in at least 10 tubules that appear circular on the slide per animal. � p<0.05; n = 3/group. Data shown as mean ± SEM. Statistical analysis was performed using 2-way ANOVA. C) Magnified sections from WT males on LFD and Inpp4b -/males on HFD, pre-leptone to zygotene spermatocytes, pachytene spermatocytes and elongated spermatids showed by black arrowheads, arrows and white arrowheads respectively. Scale bar is 20 μm. as in our mice, partially rescued mutant male infertility, suggesting the existence of genetic modifier gene(s) in this mouse line. Conditional deletion of Inpp5b in germ cells resulted in normal male fertility [53], indicating that the abnormalities in mutant sperm function and maturation were not due to a deletion of the gene in germ cells. Further analysis of Inpp5b -/mutant Sertoli cells revealed the appearance of abnormal vacuoles affecting germ cell adhesion [54]. A similar effect of Inpp4b deletion in somatic testicular cells may account for the observed increase of apoptosis in spermatogonial and spermatocyte cells detected in Inpp4b -/mutants. One possible explanation of this phenomenon is that it might be related to a decrease in testosterone production in mutants along with the reduced expression of several steroidogenesis genes in mutant testis. The role of INPP4B targets, such as PI3K/Akt, in the survival of Leydig cells and steroidogenesis has been previously demonstrated [55]. Additionally, decreased expression of cytokines such as Il1b and Il6 in mutant Sertoli cells might have caused reduced testosterone production, resulting in impaired spermatogenesis [4]. Ablation of AR signaling in Leydig, Sertoli, or peritubular myoid cells all leads to deficient spermatogenesis [56]. This is also consistent with our findings of prostate hyperplasia in one year old Inpp4b mutant males [23]. Thus, it is possible that in addition to a direct effect of Inpp4b deletion in germ cells, indirect effects of reduced testosterone and LH signaling in testicular somatic cells may be responsible for the observed phenotype. Further analysis of conditional Inpp4b deletion in various testicular cells might define the role of this gene in spermatogenesis. objective and analyzed with 2-way ANOVA. Data shown as mean ± SEM. n = 3/group. (TIF) S1 Raw data. (PDF)