SPAM1/HYAL5 double deficiency in male mice leads to severe male subfertility caused by a cumulus‐oocyte complex penetration defect

Abstract

The glycosylphosphatidylinositol‐anchored sperm hyaluronidases (Hyals), sperm adhesion molecule 1 (SPAM1) and HYAL5, have long been believed to assist in sperm penetration through the cumulus‐oocyte complex (COC), but their role in mammalian fertilization remains unclear. Previously, we have shown that mouse sperm devoid of either Spam1 or Hyal5 are still capable of penetrating the COC and that the loss of either Spam1 or Hyal5 alone does not cause male infertility in mice. In the present study, we found that Spam1/Hyal5 double knockout (dKO) mice produced significantly fewer offspring compared with wild‐type (WT) mice, and this was due to defective COC dispersal. A comparative analysis between WT and Spam1/Hyal5 dKO epididymal sperm revealed that the absence of these 2 sperm Hyals resulted in a marked accumulation of sperm on the outside of the COC. This impaired sperm activity is likely due to the deficiency in the sperm Hyals, even though other somatic Hyals are expressed normally in the dKO mice. The fertilization ability of the Spam1/Hyal5 dKO sperm was restored by adding purified human sperm Hyal to the in vitro fertilization medium. Our results suggest that Hyal deficiency in sperm may be a significant risk factor for male sterility.—Park, S., Kim, Y.‐H., Jeong, P.‐S., Park, C., Lee, J.‐W., Kim, J.‐S., Wee, G., Song, B.‐S., Park, B.‐J., Kim, S.‐H., Sim, B.‐W., Kim, S.‐U., Triggs‐Raine, B., Baba, T., Lee, S.‐R., Kim, E. SPAM1/HYAL5 double deficiency in male mice leads to severe male subfertility caused by a cumulus‐oocyte complex penetration defect. FASEB J. 33, 14440‐14449 (2019). www.fasebj.org

ABBREVIATIONS

Cas9

CRISPR‐associated protein 9

COC

cumulusoocyte complex

CRISPR

clustered regularly interspaced short palindromic repeat

dKO

double knockout

G418

Geneticin

HA

hyaluronic acid

Hyal

hyaluronidase

IVF

in vitro fertilization

KO

knockout

SPAM1

sperm adhesion molecule 1

TYH

Toyoda‐Yokoyama‐Hoshi

WT

wild type

To ensure successful fertilization, mammalian sperm must penetrate the cumulus‐oocyte complex (COC) in the ampulla region of the oviduct and fuse with the egg plasma membrane (1, 2). Although the precise mechanisms underlying these events remain unclear, evidence has suggested that sperm hyaluronidases (Hyals) play a critical role in the fertilization process (3–6). Hyaluronic acid (HA), a polysaccharide composed of alternating repeats of N‐acetylglucosamine and D‐glucuronic acid disaccharide units, is a key component of the matrix of the oocyte COC and of cumulus cells and is essential for successful ovulation and in vivo fertilization (7).

Hyals are a family of enzymes that catalyze the degradation of HA and were first placed into 2 distinct groups in 1971 by Meyer (6). In mice, the 7 Hyal‐like genes Hyal1, Hyal2, Hyal3, Hyal4, Hyal5, Hyal6, and sperm adhesion molecule 1 (Spam1) are clustered as 2 tightly linked multiplets on chromosomes 9F1‐9F2 and 6A2 (8, 9). In mice, the 2 sperm‐specific Hyal isoforms, namely SPAM1 and HYAL5, are glycosylphosphatidylinositol‐anchored enzymes that are attached to the plasma membrane and the acrosomal membrane. In mice, Spam1 and Hyal5 are both single‐copy genes and are localized side‐by‐side on chromosome 6 (9–12). They are primarily expressed in male germ cells (13, 14), although low levels of SPAM1 expression have also been detected in the male vas deferens and accessory duct (15) and the female reproductive tract (16). The function of SPAM1, also referred to as PH‐20, has been investigated using Spam1 knockout (KO) mice. Spam1^‐/‐ male mice are fertile, although Spam1^‐/‐ epididymal sperm exhibited a delay in COC dispersal compared with wild‐type (WT) mice (17). Furthermore, there were no significant differences in the zona pellucida‐binding ability between Spam1^2/2 and WT sperm. HYAL5 is a newly identified sperm‐specific Hyal composed of a single 55‐kDa polypeptide chain. SPAM1, in contrast, is composed of 2 polypeptide chains connected by a covalent bond (8). In 2009, Kimura et al. (18) demonstrated that the fertilization ability of Hyal5‐/‐sperm was similar to that of WT sperm. It has been suggested that the absence of fertilization defects in Spam1^2/2 or Hyal5^‐/‐ mice might be attributed to compensation by the other remaining Hyals. Interestingly, with the exception of rodents and bovine, the genome of most mammals encodes only a single gene for a sperm‐specific Hyal.

Although sperm Hyals are critical for in vitro fertilization (IVF) in mammals (19), their in vivo function still remains unclear (20, 21). In order to assess the proposed compensatory activity played by sperm Hyals in mouse fertilization, it is necessary to generate a double knockout (dKO) mouse lacking both Hyal5 and Spam1. To this end, we generated Spam1 and Hyal5 dKO mice using the clustered regularly interspaced short palindromic repeat (CRISPR)/CRISPR‐associated protein 9 (Cas9) system. The resulting male KO mice lacking both Spam1 and Hyal5 exhibited fertility defects that were characterized by a marked reduction in the number of offspring and impaired COC dispersal. These results indicate that sperm Hyal activity is required in vivo for the fertilization process. In addition, the impaired fertility observed in the dKO mice could be largely attributed to a severe defect in COC dispersal. By generating male KO mice lacking SPAM1 and HYAL5 activity, we have therefore demonstrated that sperm Hyal plays a critical role in promoting COC dispersal and that a lack of Hyal enzyme activity causes severe male subfertility in mice. These data suggest that Hyal activity is also required for human sperm to promote COC dispersal. In this regard, male infertility could also be associated with impaired sperm Hyal activity.

MATERIALS AND METHODS

RT‐PCR analysis

Total RNA was extracted from mice testes using Isogen (Nippon Gene, Toyama, Japan), as previously described by Kim et al. (8). Five micrograms of total RNA was reverse‐transcribed to cDNA using the SuperScript III First‐Strand Synthesis System (Thermo Fisher Scientific, Waltham, MA, USA). PCR amplification was carried out using Ex Taq DNA polymerase (Takara Bio, Kusatsu, Japan) according to the manufacturer's instructions. A 10th of the first‐strand cDNA reaction mixture was used as the PCR template. The oligonucleotide sequences of all primers used in this study were as follows: Spam1:5′‐AAGGAAGTTTATGGAAGGAACT‐3′ (forward), 5′‐GAGCAACAATTTCACCAATTGT‐3′ (reverse), Hyal5:5‐TTTCACAGAAGCTGTCAAGTGG‐3 (forward), 5‐GTAATCATGTAAGTCCAGACAA‐3 (reverse), Hyal4:5′‐TGGAGGAAACTATCAAATTGGG‐3 (forward), 5′‐GCTCTGGTCACATTGATTATGTA‐3 (reverse), Hyal6:5′‐TAGGAATTGGGGCAGTAGAATG‐3 (forward), 5′‐ACTTTCGCCAATTGTGTGCACT‐3 (reverse), Hyall:5′‐ATTCGTCCGTGATCAGAACGAC‐3 (forward), 5′‐CAATGGAGAAACTGGCAGGGTT‐3 (reverse), Hyal2:5′‐AGACCCCAGCACCTGTGGGGCT‐3 (forward), 5′‐TACTGGGTGGCCCAGGACACGT‐3 (reverse), Hyal3:5′‐ACA‐GATATCCAGCCTGTGGCAA‐3 (forward), 5′‐TCGACCATGG‐CCATGACATCGC‐3′ (reverse), Gapdh: 5′‐AGATTGTCAGCAAT‐CATCCTG‐3 (forward), and 5′‐TGCTTCACCACCTTCTTGAT‐GT‐3 (reverse).

Preparation of protein extracts

Testicular germ cells and epididymal sperm were prepared from WT and dKO mice, as previously described by Kim et al. (8). Cells were suspended in a lysis buffer consisting of 20 mM Tris‐HCl (pH 7.4), 1% Triton X‐100, 150 mM NaCl, and a 10% protease inhibitor cocktail (MilliporeSigma, Burlington, MA, USA) and kept on ice for 20 min. After centrifugation at 10,000 g for 10 min at 4°C, proteins in the supernatant fraction were analyzed by Western blotting. Protein concentration was determined by the Bradford method (22).

Fertility testing

Sexually mature male Spam1^+/‐/Hyal5^+/‐ and Spam1^‐/‐ /Hyal5^‐/‐ mice were housed with female F₁ progeny (>2 mo old) from a C57BL/6N × DBA/2 cross (also referred to as B6D2F1 mice) for 4 mo, and the number of pups in each cage was determined within a week of their birth. Copulation was confirmed by the presence of vaginal plugs, which were checked every morning.

Western blot analysis

Proteins were separated using SDS‐PAGE, and the separated proteins were then electrophoretically transferred to Immobilon‐P PVDF membranes (MilliporeSigma). The membranes were blocked with 2% skimmed milk in 50 mM Tris/HCl (pH 7.5) supplemented with 0.3 M NaCl and 0.1% Tween‐20. The blots were probed with the primary antibodies and subsequently incubated with the appropriate horseradish peroxidase‐conjugated secondary antibody. Immunoreactive protein bands were visualized using an ECL Western Blotting Detection Kit (Elpis Biotech, Daejeon, South Korea) according to the manufacturer's instructions.

Zymography

Proteins with Hyal activity were visualized following separation by SDS‐PAGE in the presence of 0.01% human umbilical cord hyaluronan (MilliporeSigma) under nonreducing conditions. To remove residual SDS, the gels were washed 3 times with 50 mM sodium acetate buffer (pH 6.0) supplemented with 0.15 M NaCl and 3% Triton X‐100 at room temperature for 1.5 h. Following this, the gels were incubated with 50 mM sodium acetate buffer (pH 6.0) supplemented with 0.15 M NaCl at 37°C overnight. The hyaluronan‐hydrolyzing proteins were detected as transparent bands against a blue background by staining the gels with 0.5% Alcian Blue 8Gx and Coomassie Brilliant Blue R250 (MilliporeSigma).

Agarose gel electrophoresis

Hyaluronan (1%, 50 μl) (MilliporeSigma) was digested with proteins extracted from mouse tissues for 12 h in a 37°C incubator. In addition, in order to confirm that tissues and sperm Hyal activity patterns vary according to pH, we carried out the hyaluronan digestion assay over a range of pH values from 1 to 10. Following this, the digested hyaluronan samples were separated by electrophoresis on 0.8% agarose gels, after which the gels were stained with 0.5% Alcian Blue 8Gx to visualize the HA bands.

IVF assay

C57BL/6N mice (~6‐7 wk old) were superovulated using intraperitoneal injections of 7.5 IU pregnant mare serum gonadotropin and human chorionic gonadotropin at 46‐h intervals. The COC was collected from the ampulla of the oviduct, and denuded oocytes were prepared by treating the COC with 0.1% Hyal. The oocytes were cultured in M16 medium (MilliporeSigma) prior to insemination. Cauda epididymides were dissected from WT and dKO mice and gently squeezed to collect the spermatozoa in human tubal fluid medium (Origio, Måløv, Denmark). The spermatozoa were capacitated at 37°C for 1 h prior to insemination and subsequently mixed with the oocytes at a final level of 1.5 χ 10⁶ sperm/ml. The oocytes and spermatozoa were coincubated for 4 h at 37°C in a 5% CO₂ atmosphere. Cumulus‐intact or cumulus‐free eggs were inseminated by the capacitated sperm (1.5 × 10⁵ sperm/ml) in a 0.2‐ml drop of Toyoda‐Yokoyama‐Hoshi (TYH) medium (i.e., a modified Krebs‐Ringer bicarbonate solution supplemented with glucose, sodium pyruvate, bovine serum albumin, and antibiotics). The inseminated eggs were incubated for 6 h at 37°C in a 5% CO₂ environment. Following this, the cumulus cells were removed by incubating the eggs with bovine Hyal (3 U/ml) for 15 min, and the eggs were subsequently washed. The female and male pronuclei of the inseminated eggs were stained with DAPI (10 μg/ml) for 30 min and then viewed under a fluorescence microscope (Leica Microsystems, Buffalo Grove, IL, USA).

Dispersal of COC in vitro

Eggs surrounded by a dense layer of cumulus cells were collected from the oviductal ampulla of superovulated mice 16 h after the injection of human chorionic gonadotropin. The eggs were placed in a 0.2‐ml drop of TYH medium covered with mineral oil. Fresh cauda epididymal sperm from 3‐mo‐old dKO mice were capacitated by incubation in a 0.2‐ml drop of TYH medium for 2 h at 37°C in an atmosphere of 5% CO₂. Human sperm extracts (4 μg protein/μl) were added to the eggs in TYH medium in the presence or absence of 300 μM apigenin and further incubated for 1‐3 h at 37°C in a 5% CO₂ atmosphere. The morphology of the COC was evaluated using an Eclipse Ti‐S microscope equipped with a D5‐Qi1Mc digital camera (Nikon, Tokyo, Japan). The sperm motility assays were conducted in the presence of 0.4% HA in TYH medium. Briefly, dKO sperm were incubated in the medium at 37°C for 30 min and subsequently observed using a Nikon microscope.

Recombinant Hyals

DNA fragments encoding the entire protein‐coding regions of Hyal1, Hyal2, and Hyal3 were introduced into a pCXN2 vector. Human embryonic kidney 293 (HEK293) cells were transfected with the plasmid constructs (20 μg) by the lipofection method and cultured in DMEM/10% heat‐inactivated fetal bovine serum in the presence of 0.5 mg/ml Geneticin (G418) at 37°C in 5% CO2/95% air. To confirm Hyal activity, cell lysates prepared from the G418‐resistant clones were subjected to an agarose gel analysis.

Statistical analysis

The data are presented as the means ± sd of $3 independent experiments. Statistical comparisons were made using Student's t test in Excel (Microsoft, Redmond, WA, USA) after the distribution of the data had been examined.

RESULTS

Generation of the double (Spam1‐/‐/Hyal5‐/‐) KO mice

To elucidate the functional roles of sperm Hyals in fertilization, we knocked out both Spam1 and Hyal5 in the mouse embryo using CRISPR/Cas9‐mediated genome editing (Fig. 1A) (23). The progeny of the cross of heterozygous mice that generated the dKO mice exhibited the expected Mendelian ratios (11:23:10 from 4 breeding pairs). The genotypes of WT (Spam1^+/+/Hyal5^+/+), heterozygous (Spam^+/‐ /Hyal5^+/‐), and homozygous (Spam1‐/‐/Hyal5‐/‐⁾ mice were determined using PCR analysis of genomic DNA (Fig. 1B). RT‐PCR analysis of testicular cDNA demonstrated that both Spam1 and Hyal5 mRNAs were absent in the dKO mice (Fig. 3). Moreover, the immunoreactive 52‐ and 55‐kDa protein bands, corresponding to SPAM1 and HYAL5, respectively, were absent from Western blots of protein extracts derived from cauda epididymal sperm isolated from the dKO mice (Fig. 5A). The fertilin, heterodimeric protein complex composed of ADAM1b and ADAM2 subunits on the sperm surface (24), which play a role during fertilization, were expressed at normal levels in the sperm of the dKO mice (Fig. 5A). These data indicate that the sperm of the dKO mice completely lacked the 2 sperm‐specific Hyals (Fig. 5). In addition, zymographic assays measuring Hyal enzyme activity confirmed the complete absence of Hyal activity in the dKO sperm (Figs. 3A and 5B), suggesting that no other Hyals remained to compensate the loss of the SPAM1 and HYAL5 proteins at the pH at which the assay was performed. Moreover, an RT‐PCR analysis of total cellular RNA from the dKO testes demonstrated that Hyal1, Hyal2, Hyal3, and Hyal6 were expressed at levels similar to the levels in WT mice. HYAL4 was not examined because it encodes a chondroitinase that lacks Hyal activity (25, 26). These data clearly demonstrate that sperm Hyal activity was completely abolished in the dKO mouse.

Figure 1

A) The generation of Spam1/Hyal5 dKO mice. A schematic representation of chromosome 6 in mice. A genomic region encompassing the coding regions was deleted using the CRISPR/Cas9 system. B) PCR analysis of the target mutation in genomic DNA from WT (^+/+), heterozygous (+/ ‐), and homozygous (‐/‐) mice. Primers for SPAM1 (P1): 5′‐TTGCGCAAAGCTAAGACTGA‐3′, and (P2): 5′‐CATTCCCAACCTTGGACACT‐3′, Primers for HYAL5 (P3): 5′‐AGTGATCGTCTGGGTCTCTATCCT‐3′, (P4): 5′‐TATATCAGGGCATTCTCCTTTGTA‐3′.

Figure 3

Characterization of dKO sperm and testis. Zymography assay at pH 6.0 of protein extracts from sperm derived from WT, HT, and dKO mice (A). RT‐PCR analysis of SPAM1, Hyal5, Hyal6, and GPDH using cDNA from WT and dKO tissues (B). GPDH, glycerol‐3‐phosphate dehydrogenase; HT, heterozygous type.

Figure 5

Expression of sperm proteins from epididymal sperm from WT and dKO mice. A) Proteins from the sperm were separated by SDS‐PAGE under reducing conditions, and Western blot analysis was performed with antibodies against SPAM1, HYAL5, ADAM1b, and ADAM2. Hyaluronan degradation assays were performed at pH 7.4. For each lane, the hyaluronan was incubated with or without a source of Hyal and then separated by 1.0% agarose gel electrophoresis, which was followed by staining with Alcian Blue 8Gx. B) No Hyal active at the pH 7.4 is present in the No treatment lane and the dKO sperm extract lane. The white arrow head indicates degraded hyaluronan, and the asterisk indicates intact hyaluronan.

Impairment of fertility in Spam1‐/‐/Hyal5‐/‐male mice

The dKO mice were normal with respect to general health, body size, and behavior. The morphology of spermatogenic cells appeared normal in the seminiferous tubules in the testes of the dKO mice (Figs. 1 and 2A and Supplemental Table S1). Similarly, cauda epididymal sperm isolated from the dKO mice were indistinguishable from those isolated from WT mice with respect to shape, motility, and the percentage of acrosome‐reacted sperm (Supplemental Fig. S1). In addition, the mean number of sperm (1.19 ± 1.15 × 10⁷ sperm, 5 mice) in the cauda epididymis and the sperm count (1.18 ± 1.75 × 10⁶ sperm, 5 mice) in the uterus from the ejaculated semen of dKO mice were similar to those of WT mice (1.24 ± 2.09 × 10⁷ sperm and 1.2 ± 1.39 × 10⁶ sperm, respectively). These findings indicate that there were no apparent abnormalities in spermatogenesis, sperm maturation, or ejaculation in male dKO mice. Despite the normal formation of copulation plugs in mated females, the fertilization ability of sperm in the dKO mice was severely impaired. Only a few litters were generated from crosses between a total of 10 males lacking the 2 sperm‐specific Hyals and 30 females over a period >2 mo (mean of 1.8 offspring). Compared with the dKO mice, however, the litter sizes of WT and heterozygous mice were normal (mean 10.3 and 9.7 offspring, respectively) (Fig. 2B) from crosses between male (n = 10) and female (n = 30) mice, respectively.

Figure 2

A) The loss of sperm‐specific Hyals does not affect spermatogenesis or the rate of fertilization. Hematoxylin and eosin staining of the Spam1/Hyal5 dKO mouse testis. B) Mean litter size from Spam1/Hyal5 dKO male mice. HT, heterozygous type.

Defective COC dispersal conferred by the sperm of dKO mice

To assess whether the loss of sperm Hyals affects the ability of sperm to penetrate the COC, we evaluated capacitated cauda epididymal sperm using fertilization assays. Compared with heterozygous sperm, Spam1^—/—/ Hyal5‐^/— sperm were defective in COC dispersal, suggesting that this defect might be directly linked to impaired male fertility in vivo (Figs. 4A and 6A). The fertilization ratio of intact eggs inseminated with Hyal‐deficient sperm markedly decreased, but such a decrease in the fertilization ratio was not observed in the COC‐free zona pellucida‐intact eggs (Fig. 4B).

Figure 4

IVF assay of dKO sperm. A) The fertilization ratio of intact eggs inseminated with capacitated cauda epididymal sperm from dKO (open column) and Hetero type (shaded column) male mice. B) Cumulus‐free eggs were inseminated with capacitated cauda epididymal sperm derived from Hetero type (shaded column) and dKO (open column) mice, and fertilization ratio was determined. Hetero, heterozygous.

Figure 6

COC dispersion assay. A, B) The COCs were incubated for 30 min to 3 h with WT or dKO mouse sperm as well as with sperm extracts, and COC dispersion was evaluated. C) Purified human sperm (hSperm) Hyal was assessed in the mouse COC dispersion assay in the presence or absence of 300 μM of apigenin at 37°C for 3 h.

Sperm Hyal plays a key role in mammalian fertilization

To ascertain if the loss of SPAM1 and HYAL5 proteins on the surface of mouse sperm affects their motility during the process of COC dispersion, we performed an in vitro sperm motility assay with the addition of 0.4% HA diluted in a drop of TYH medium. WT sperm moved forward progressively and dispersed the HA spontaneously. However, the Hyal‐deficient sperm moved slowly during the course of the assay (data not shown) and were unable to penetrate the COC (Fig. 6). The COC treated with the Hyal‐deficient sperm or sperm extracts remained densely packed during the entire 3‐h incubation period, similar to what was observed in control COCs incubated in the absence of sperm or sperm extracts (Fig. 6A, B). In the presence of WT sperm, however, the cumulus cells were readily dispersed from the COC, and they were completely eliminated 3 h after insemination (Fig. 6A). These results indicate that the defective COC dispersal caused by Hyal‐deficient sperm was due to the loss of the sperm Hyal activity responsible for hydrolyzing HAs in the extracellular matrix of the cumulus cells. These findings indicate that sperm‐specific Hyal activity is required for the ability of sperm to penetrate cumulus cells in the COC. To parallel these findings in mouse sperm, a COC dispersion assay was performed using purified human sperm Hyal to determine if this enzyme was also essential for human sperm to penetrate the COC. As shown in Fig. 6C, COC dispersion was strongly inhibited in the presence of a Hyal inhibitor (300 μM apigenin), although the human sperm Hyal was still capable of promoting spontaneous COC dispersal (Fig. 6C). In addition to IVF, we also investigated the effect of Hyal on mouse fertilization using purified human sperm Hyal. The low IVF level obtained with the Hyal‐deficient mouse sperm was restored by the addition of purified human sperm Hyal to the IVF medium (Fig. 7). Taken together, these data indicate that sperm Hyal activity plays a key role in mammalian fertilization.

Figure 7

Effect of human sperm Hyal during IVF using dKO sperm. The ration of IVF was determined using cauda epididymal sperm from WT male mice (shaded column), dKO male mice (black column), WT (shaded column), and purified human sperm Hyal (open column). The COCs were incubated with capacitated epididymal dKO sperm in IVF medium, following the addition of purified human sperm Hyal. Error bars represent the means ± sd (n = 3) performed in triplicate.

Role of somatic Hyals in mammalian fertilization

The presence of somatic Hyals in mouse sperm led us to postulate that a unique somatic Hyal (HYAL1, HYAL2, or HYAL3) may have a role in the sperm dispersal of the COC complex. To determine whether these Hyals play a role in the fertilization process, an RT‐PCR analysis was firstly conducted in various tissues from the dKO fertilization ratio male mice using specific primers against Hyal1, Hyal2, and Hyal3. The data showed that the somatic Hyal genes were ubiquitously expressed in all mouse tissues tested as well as expressed at similar levels in both WT and dKO mice (Fig. 8A). To assess if the catalytic activity of these somatic Hyals was present in the dKO sperm, an HA dispersal assay was performed using dKO tissues extracts. Despite the presence of somatic Hyal molecules in dKO sperm, no signals were detected in this assay (Fig. 8B). These data suggest that somatic Hyal do not contribute to the digestion of HA. In order to further explore possible conditions that could detect the hyaluronic activity of HYAL1, HYAL2, and HYAL3, we generated expression vectors encoding the mouse hyal1, hyal2, and hyal3 genes (Fig. 8D). As shown in Fig. 8E, protein extracts from HEK293 cells transfected with the different somatic Hyals did not show evidence for Hyal activity. Because the pH conditions in the female reproductive tract are nearly neutral during natural fertilization, Hyals present on the sperm head must be active at a neutral pH. When the Hyal activities in the WT and dKO mice were assayed over a range of different pH values, there were no differences in the activities between the WT and dKO mice samples (Fig. 8B). Nonetheless, it is possible that in vivo activities not detectable using these in vitro assays were responsible for the weak fertility of the dKO mice.

Figure 8

Expression of mouse HYAL1, HYAL2, and HYAL3 in HEK 293 cells. A) The somatic mouse Hyals HYAL1, HYAL2, and HYAL3 are conserved in mouse germ cells, but they do not have sperm‐associated extracellular activity. RT‐PCR shows the presence of a 500 bp PCR band in both dKO and WT mice organs tested. A hyaluronan digestion assay showed that WT sperm had activity at pH 4, 7.0, and 9.0, but not at pH 2.0. B) This was indicated by a low molecular mass band (white arrow head) present in the lane representing the WT sperm, but there was no evidence of any digested bands in lanes representing any of the dKO mice tissues. C) Based on the IVF rate at both pH 4.0 and 7.4, there were no surviving sperm or eggs and no fertilized eggs at pH 4.0. HEK293 cells were transfected with plasmids containing Hyal1, Hyal2, and Hyal3. After selection of G418 positive clones, the cells were harvested, and a hyaluronan digestion assay was performed using the extracts. D) Cells transfected with the control plasmid did not show expression of Hyal1, Hyal2, or Hyal3. E) No enzyme activity was found at pH 4 or 7.4 when protein extracts from the HYAL1, HYAL2, and HYAL3 expressing cells were mixed with 1% hyaluronan. GDH, glycerophosphate dehydrogenase; GPDH, glycerol‐3‐phosphate dehydrogenase; mHyal, mouse Hyal.

DISCUSSION

Based on the lack of fertility phenotype seen in mice deficient in either Spam1 or Hyal5, we hypothesized that SPAM1 and HYAL5 Hyals could compensate for each other, and, as such, we set out to generate mice with a dKO of these glycosylphosphatidylinositol‐anchored Hyals. By generating Spam1/Hyal5 dKO mice and providing both genetic and biochemical evidence, we now show that sperm deficient in both of these Hyals are severely impaired in their ability to fertilize cumulus‐intact eggs because of a defect in COC dispersal, thereby confirming that sperm Hyal activity is a key factor required for fertilization. To our knowledge, this is the first report that clearly demonstrates the prime importance that sperm Hyal activity plays in the COC dispersal event.

In contrast to the previous single KO mice lacking only one of these 2 enzymes, male fertility was severely impaired in the Spam1/Hyal5 dKO mice (Figs. 2B and 4). The fertility defects were characterized by a reduction in offspring number and an impairment of the ability of sperm to penetrate the COC (Fig. 2B). Despite the importance of the cumulus matrix in fertilization, the molecular mechanisms by which ejaculated sperm obtain access to cumulus‐intact eggs by penetrating the COC remain unclear. Previously, only ambiguous observations in male infertility were made using individual Spam1‐ and Hyal5‐deficient mice because double gene KO mice were unavailable. Although it was clearly important to generate double gene KO mice lacking both SPAM1 and HYAL5 in order to clarify the exact functions of these 2 mouse sperm Hyals in fertilization, to date this has been unsuccessful because of the difficulty of generating gene KO mice using homologous recombination procedures due to these genes being localized on the same chromosome. With the advent of more efficient genome‐editing tools, we could finally generate these double gene KO mice using the CRISPR/C as 9 system (23).

It appears that a key factor affecting the severe defects in male fertility in the dKO mice is the loss of the sperm's ability to promote COC dispersal and penetration (Fig. 6). Hyal‐deficient sperm from dKO mice showed a severe reduction in fertility compared with WT sperm because of their impaired penetration into the COC even after adhering to the COC surface. Previous reports have shown that SPAM1 is expressed on the entire surface of the sperm head and on the outer acrosomal membrane, whereas HYAL5 is expressed on both the outer and inner acrosomal membranes (8). Many researchers believe that digestion of the outer COC begins when SPAM1 first contacts the COC. Following this, the activity of HYAL5 further facilitates penetration of the sperm into the COC space, thereby allowing the sperm to gain access to the egg.

Despite these hypotheses, it remains unclear why the Hyal‐deficient sperm from the dKO mice were not completely infertile. One possibility is that the packing of the COC becomes loose as the ovulated COC travels through the oviductal fluid. In this case, the physical force of the sperm might at times be sufficient to penetrate the egg in the absence of Hyal activity. In support of this, sometimes no offspring were produced in the mating experiment, whereas in other cases there were litter sizes ranging from 1 to 4 mice. These variable litter sizes might be due to the influence of a variety of factors, such as the ovulation status in the COC in the oviduct and the age of the female mouse used for mating. The evidence for this is that while no offspring were produced from the initial 3 mating experiments between the dKO male mice and 6‐wk‐old WT female mice, later matings with older female mice resulted in birth. For this reason, the mean litter size of 1.8 obtained from mating with the dKO male mouse might be dependent on the ovulated COC status.

Interestingly, the IVF rate using Hyal‐deficient sperm was also not 0% (Fig. 4A). The most likely reason for this is that fertilization succeeded because the egg is located outside of the superovulated COC, making it so that the sperm could more easily come into contact with the superovulated egg in the IVF assay. The COC of 1 mouse contains multiple oocytes; among these, an oocyte located outside the COC would easily be able to make contact with a Hyal‐deficient sperm. In the case of humans with Hyal‐deficient sperm, the probability of male infertility would be very high because humans have only 1 egg in the COC. Interestingly, we found that the fertilization rate was restored by adding purified Hyal from human sperm during the IVF assay using the dKO sperm and the mouse COC (Fig. 7).

Another potential explanation for the observation that fertility was not entirely abolished in the dKO mice is the existence of other Hyals; HYAL1, HYAL2, and HYAL3 are all expressed in the testis, and HYAL2 and HYAL3 have been reported to have acid Hyal activity in mouse and human sperm (27, 28). Although we were unable to detect activity associated with these enzymes in our expression studies, this could be a bias of our assay, and it is possible that these enzymes have activities that we did not detect. These Hyal‐like genes are clustered as 2 tightly linked multiplets on 2 chromosomes; Hyal1, Hyal2, and Hyal3 on mouse chromosome 9F1‐F2 and Hyal4, Hyal5, Hyalp1, and Spam1 on mouse chromosome 6A2.

Despite the fact that Hyal1‐, Hyal2‐, and Hyal3‐deficient male mice have normal fertility (29–31), there is some controversy as to why these proteins are present in the sperm of normal mice. This study reveals that in the IVF treatment there is a failure of the sperm‐egg interaction because of the death of both gametes in the low‐medium pH conditions (Fig. 8C). On the other hand, although we expressed the mouse HYAL1, HYAL2, and HYAL3 proteins in HEK293 cells, their enzymatic activity was not found at neutral pH nor even active at low pH. Thus, although somatic Hyals may be present on the sperm surface, they are not thought to be involved in fertilization.

One of the most puzzling questions is whether the sperm Hyal plays a role in COC dispersal in human fertilization. Although the sperm Hyal activity from infertile human males has not been assessed, a defect in sperm Hyal function would be a potential risk for causing infertility in the monotocous human species in parallel with the mutant mouse case shown in this study. Consistent with our hypothesis, the ability of human sperm to disperse and penetrate the COC was inhibited in the presence of a Hyal inhibitor (Fig. 6C). Interestingly, the IVF assay using purified human Hyal showed that the fertility rate was ~70% higher than that without purified human Hyal (Fig. 7). Because the human genome encodes only 1 sperm neutral‐active Hyal, abnormalities in the activity of this sperm Hyal might lead to more severe fertility defects in humans compared with mice. The mechanisms underlying sperm entry into the COC and sperm passage through the cumulus matrix can therefore provide key insights into male infertility, which could contribute to the development of therapeutic approaches to treat male infertility as well as novel contraceptive agents.

ACKNOWLEDGMENTS

The authors thank Jinyoung Kim (Korea University, Seoul, South Korea) for figure assistance. This research was partly supported by Korea Research Institute of Bioscience and Biotechnology (KRIBB) Research Initiative Program (KGM5281921 and KGM4251913) and a National Research Foundation of Korea Grant funded by the Korean Government (NRF‐2017R1D1A1B03031420). The authors declare no conflicts of interest.

AUTHOR CONTRIBUTIONS

S.‐U. Kim, S.‐R. Lee, and E. Kim conceived of the study; S. Park, Y.‐H. Kim, P.‐S. Jeong, C. Park, J.‐W. Lee, G. Wee, B.‐S. Song, and B.‐W. Sim performed experiments; S.‐U. Kim, B.‐J. Park, S.‐H. Kim, and T. Baba analyzed data; S.‐R. Lee and E. Kim wrote the manuscript; B. Triggs‐Raine commented on and revised the manuscript; and all authors were involved in the discussion of results.