bioRxiv preprint doi: https://doi.org/10.1101/2021.05.02.442348; this version posted May 2, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

The sperm SPACA4 is required for efficient fertilization in mice

Sarah Herberg1#, Yoshitaka Fujihara2,3#, Andreas Blaha1,4, Karin Panser1, Kiyonari Kobayashi2, Tamara Larasati2, Maria Novatchkova1, H. Christian Theußl1, Olga Olszanska1, Masahito Ikawa2,5,6,7*, Andrea Pauli1*

1 Research Institute of Molecular Pathology (IMP), Vienna BioCenter (VBC), Campus-Vienna-Biocenter 1, 1030 Vienna, Austria. 2 Research Institute for Microbial Diseases, Osaka University, Suita, Osaka 565-0871, Japan. 3 Department of Bioscience and Genetics, National Cerebral and Cardiovascular Center, Suita, Osaka 564-8565, Japan. 4 Vienna BioCenter PhD Program, Doctoral School of the University at Vienna and Medical University of Vienna, Vienna, Austria. 5 Immunology Frontier Research Center, Osaka University, 3-1 Yamadaoka, Suita, Osaka 565-0871, Japan. 6 Graduate School of Pharmaceutical Sciences, Osaka University, 1-6 Yamadaoka, Suita, Osaka 565-0871, Japan. 7 The Institute of Medical Science, The University of Tokyo, 4-6-1 Shirokanedai, Minato-ku, Tokyo 108-8639, Japan.

# equal contribution * corresponding authors

Contact details for corresponding authors: [email protected], [email protected]

SUMMARY

Fertilization is the fundamental process that initiates the development of a new individual in all sexually reproducing species. Despite its importance, our understanding of the molecular players that govern mammalian sperm-egg interaction is incomplete, partly because many of the essential factors found in non-mammalian species do not have obvious mammalian homologs. We have recently identified the Ly6/uPAR protein Bouncer as a new, essential fertilization factor in zebrafish (Herberg et al., 2018). Here, we show that Bouncer’s homolog in mammals, SPACA4, is also required for efficient fertilization in mice. In contrast to fish, where Bouncer is expressed specifically in the egg, SPACA4 is expressed exclusively in the testis. Male knockout mice are severely sub-fertile, and sperm lacking SPACA4 fail to fertilize wild-type eggs in vitro. Interestingly, removal of the zona pellucida rescues the fertilization defect of Spaca4-deficient sperm in vitro, indicating that SPACA4 is not required for the interaction of sperm and the oolemma but rather of sperm and zona pellucida. Our work identifies SPACA4 as an important sperm protein necessary for zona pellucida penetration during mammalian fertilization.

Keywords: fertilization, sperm-egg interaction, zona pellucida, mouse

1 bioRxiv preprint doi: https://doi.org/10.1101/2021.05.02.442348; this version posted May 2, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

INTRODUCTION fertilization factors JUNO, TMEM95, FIMP and SOF1 lack clear non-mammalian homologs (S1 Fertilization is the fundamental process by Fig) (Grayson, 2015; Pausch et al., 2014). In line which two gametes, the sperm and the egg, fuse to with this observation, rapid protein evolution and form a single cell which gives rise to a new divergence has been noted as a general hallmark of organism. Despite being essential for all sexually involved in reproduction (Swanson and reproducing organisms, the molecular mechanisms Vacquier, 2002). that mediate sperm-egg interaction remain poorly This evolutionary divergence limits the direct understood. One important step towards gaining transfer of knowledge gained from studies in mechanistic insights into fertilization is the invertebrates (Evans and Sherman, 2013; Kosman identification of molecules that can mediate sperm- and Levitan, 2014; Levitan, 2017; Palumbi, 2009) egg interaction. Several proteins that are and plants (reviewed in (Dresselhaus et al., 2016)) specifically required for gamete interaction in to fertilization in vertebrates. For example, the mammals have been identified with the help of mechanistically best-understood fertilization genetic mouse models (reviewed in (Evans, 2012; proteins are lysin and Verl from the marine Krauchunas et al., 2016; Okabe, 2018a)). Required mollusk abalone (reviewed in (Gert and Pauli, proteins on the sperm are the transmembrane 2019)). Lysin is a sperm-expressed and highly proteins IZUMO1 (Inoue et al., 2005; Satouh et al., abundant secreted protein, whereas Verl is an egg- 2012), SPACA6 (Lorenzetti et al., 2014; Noda et coat protein that shows structural homology to al., 2020), TMEM95 (Fernandez-Fuertes et al., mammalian ZP2 (Raj et al., 2017). Species-specific 2017; Noda et al., 2020; Pausch et al., 2014), FIMP binding of lysin to Verl causes non-enzymatic (Fujihara et al., 2020), DCST1/2 (Inoue et al., disruption of the abalone egg coat, thereby 2021; Noda et al., 2021) and the secreted protein allowing conspecific sperm to fertilize the egg SOF1 (Noda et al., 2020). The factors required on (Lewis et al., 1982; Lyon and Vacquier, 1999; Raj the egg are the tetraspanin protein CD9 (Kaji et al., et al., 2017; Swanson and Vacquier, 1997). 2000; Miyado et al., 2000; Le Naour et al., 2000) However, lysin has no known homolog in as well as the GPI-anchored protein JUNO vertebrates, leaving it open whether a similar (Bianchi et al., 2014). The only known interacting mechanism might contribute to mammalian sperm protein pair is IZUMO1 and JUNO (Aydin et al., passage through the ZP. 2016; Bianchi et al., 2014; Han et al., 2016; Kato Apart from the lack of clear homologs, et al., 2016; Ohto et al., 2016), which are known to identification of further factors required for sperm- mediate adhesion between sperm and egg (Bianchi egg interaction in vertebrates has also been et al., 2014; Inoue et al., 2005; Kato et al., 2016; hampered by the almost exclusive focus on Satouh et al., 2012). How the other known mammalian fertilization. Mammalian fertilization mammalian fertilization factors enable sperm-egg occurs internally, which poses additional binding and/or fusion remains unclear and is experimental challenges due to the low number of subject to active research. eggs and inaccessibility of gametes. Moreover, In addition to sperm/egg surface proteins, possible functional redundancies among genetic analyses also revealed the importance of fertilization factors have limited the informative proteins of the mammalian egg coat, called zona value of single- knockout studies in mice. pellucida (ZP), in fertilization. ZP proteins were Accordingly, many that had been implicated shown to be required for the initial step of sperm as potential fertilization factors based on in vitro binding to the ZP as well as the subsequent block studies were later shown to be dispensable for to polyspermy, which is induced by the first sperm fertilization in vivo (Ikawa et al., 2010; Okabe, entering the egg (Avella et al., 2013, 2014; Gahlay 2018b; Park et al., 2020). Other gamete-specific et al., 2010; Rankin et al., 1996, 1998). While proteins that might play a role in fertilization have IZUMO1, CD9, SPACA6, DCST1/2 and ZP not yet been analyzed for their function in vivo, proteins have homologs in non-mammalian leaving their roles during mammalian fertilization vertebrates, phylogenetic analyses by us and others unclear. suggest that the other known essential mammalian 2 bioRxiv preprint doi: https://doi.org/10.1101/2021.05.02.442348; this version posted May 2, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

One of these gamete-specific proteins is S2B), consistent with their origin by gene SPACA4 (Sperm Acrosome Associated 4; also duplication (Loughner et al., 2016), and are called SAMP14 (Sperm Acrosomal Membrane expressed in diverse tissues in mice and Associated 14)), which was initially identified by (Fig 1B; Fig S2C). While mammalian Spaca4 is mass spectrometry in a screen for membrane- not the only Ly6/uPAR-type gene expressed bound sperm proteins (Shetty et al., 2003). specifically in the male germline (testis) (Fig 1B; SPACA4 is particularly interesting for three Fig S2C), it stands out for having homologs in fish reasons: First, its fish homolog Bouncer was (bouncer) and amphibians (Spaca4) that are also recently shown to be essential for sperm-egg germline-specifically expressed yet in the opposite membrane interaction in zebrafish (Herberg et al., sex (ovary) (Fig 1A, Fig S2A). 2018). Secondly, while Bouncer is expressed To confirm that Spaca4 was indeed expressed exclusively in the egg in fish and frogs, its closest in male but not female gametes in mice, we mammalian homolog SPACA4 is expressed analyzed the expression level of Spaca4 mRNA in exclusively in the testis (Herberg et al., 2018; different tissues using RT-qPCR. Murine Spaca4 Shetty et al., 2003). Thirdly, the incubation of was detected specifically in testis and enriched human sperm with SPACA4-specific antibodies 100- to 1000-fold compared to other tissues (Fig was shown to decrease the binding and fusion of 1C; Fig S2D), which agrees with published RNA- sperm with zona pellucida-free hamster eggs in Seq data from mice (Fig 1A,B) and with the vitro (Shetty et al., 2003). Taken together, these reported testis-specific expression in humans observations point towards an important function (Shetty et al., 2003). Analysis of published single- of SPACA4 for mammalian fertilization. cell RNA-Seq data from murine spermatogenesis Here, we investigate the functional relevance revealed a peak of Spaca4 mRNA expression in of SPACA4 in mammals by analyzing the round spermatids, which resembles the expression phenotypic consequence of genetic loss of of Izumo1 mRNA in timing and magnitude (Fig SPACA4 in mice. S3) (Ernst et al., 2019). Using anti-mouse SPACA4 antibodies, SPACA4 was found to localize to the sperm head (Fig 1D) and was readily detected on RESULTS acrosome-reacted but not on acrosome-intact live sperm (Fig 1E). This expression pattern is Murine SPACA4 is expressed in testis and consistent with the reported acrosomal membrane localizes to the inner sperm membrane localization of human SPACA4 (Shetty et al., Bouncer was recently discovered as an 2003). essential fertilization factor in zebrafish that is attached to the egg surface via a glycosyl- Male mice lacking SPACA4 are sub-fertile phosphatidylinositol (GPI)-anchor and enables To investigate the function of mammalian sperm binding to the egg membrane (Herberg et SPACA4 in vivo, we generated Spaca4 knockout al., 2018). Evolutionary analysis revealed that mice by CRISPR/Cas9-mediated targeted Bouncer has a mammalian homolog, SPACA4 (Fig mutagenesis. We recovered two mutant alleles in 1A, Fig S2A) (Herberg et al., 2018), which raised the C57BL/6J background. The first one, in the the immediate question whether SPACA4 might following called Spaca477del, contains a 77-nt also be important for mammalian reproduction. deletion leading to a frameshift after 42 Bouncer and SPACA4 belong to the large (Fig 2A-B; Fig S4A). The second allele, in the lymphocyte -6 (Ly6)/urokinase-type following called Spaca4117del, contains a 117-nt in- plasminogen activator receptor (uPAR) protein frame deletion, which removes half (39 amino family, which is characterized by a conserved acids) of the mature Spaca4 protein and is thus also 60-80 amino acid protein domain containing 8-10 predicted to result in a full knockout mutation (Fig cysteines that adopt a characteristic three-finger 2A-B; Fig S4A). An independent Spaca4 knockout fold (Loughner et al., 2016). Most Ly6/uPAR-type mouse (Spaca4tm1Osb) was generated in the BDF1 genes occur in clusters in the mouse genome (Fig background by replacing the whole of the

3 bioRxiv preprint doi: https://doi.org/10.1101/2021.05.02.442348; this version posted May 2, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

Figure 1 A B C

ovary testis spleen muscle liver heart brain z-score ovary testis stomach spleen muscle kidney heart brain z-score Cyprinus carpio XP 018955710.1 52-339 Ly6i:15_74979534 Salmo salar XP 013981440.1 52-576 2.5 Ly6g5b:17_35113948 Cd177:7_24743983 2 Oryzias latipes AM153096.1 52-444 2 Ly6d:15_74762056 Cynoglossus semilaevis XP 008335986.1 52-516 1.5 Ly6g6f:17_35080480 Ly6g:15_75155240 1 Takifugu rubripes XP 011605858.1 52-387 1 Ly6e:15_74955051 Bouncer Oreochromis niloticus XP 013126046.1 52-420 0.5 Ly6c2:15_75108158 0 Ly6a2:15_75131377 Haplochromis burtoni XP 014191263.1 49-426 0 Ly6a:15_74994877 Lepisosteus oculatus XP 015207687.1 52-342 Lypd2:15_74732247 −1 -0.5 Danio rerio XP 005173770.1 46-297 Ly6c1:15_75044018 -1 Gpihbp1:15_75596628 Cyprinus carpio XP 018955736.1 52-360 Slurp1:15_74724318 −2 -1.5 Plaur:7_24462484

Hippocampus comes XP 019712504.1 49-312 Hrpt -2 Ly6g6d:17_35071443 other Cynoglossus semilaevis XP 008310642.1 49-411 Cd59a:2_104095801 Takifugu rubripes XP 011605859.1 52-354 -2.5 Pate7:9_35776524 Ly6/uPAR Ly6g6g:15_74770908 Oreochromis niloticus XP 019217266.1 52-336 Ly6g2:15_75216376 proteins Larimichthys crocea KKF27993.1 46-303 Ly6f:15_75268421 Oryzias latipes AM154129.1 52-327 Gm5916:9_36119934 Ly6g5c:17_35108279 Salmo salar XP 013981439.1 49-318 Pinlyp:7_24541658 Oncorhynchus mykiss XP 021433692.1 52-336 Ly6g6c:17_35065388 Lypd5:7_24349196 Xenopus tropicalis OCA29924.1 37-306 Ly6m:15_74876987 Xenopus laevis OCT73178.1 37-309 Pate4:9_35607093 Pate3:9_35645113 SPACA4 Myotis brandtii XP 005884576.1 40-303 Pate9:9_36532397 Myotis brandtii XP 005871655.1 40-303 Pate14:9_36635754 Cd59b:2_104069849 Myotis brandtii XP 005878225.1 40-309 1810065E05Rik:11_58421111 Myotis brandtii EPQ10600.1 34-375 Pate1:9_35685079 Canis lupus XP 852068.1 40-306 Pate2:9_35570284 transcript level relative to Spaca4:7_45725107 Felis catus XP 006941098.1 40-309 Gm27235:9_35776524 Homo sapiens NP 598005.1 40-306 Ly6k:15_74796874 Gm3443:19_21552204 Camelus ferus XP 006194501.1 40-306 Pate6:9_35788050 Sus scrofa NP 001171400.1 40-306 Gml:15_74813452 Spaca4 Gml2:15_74819071 Otolemur garnettii XP 012667956.1 40-306 Lypd4:7_24864620 Fukomys damarensis XP 010639093.1 46-321 Lypd9:11_58446149 Lypd10:7_24709242 Mus musculus BAB24224.1 40-306 liver heart testis oocyte Pate5:9_35838329 Cricetulus griseus XP 003508724.1 40-306 Pate13:9_35908423 Loxodonta africana XP 010585179.1 40-303 Lypd11:7_24715023 Tex101:7_24668007 Ornithorhynchus anatinus XP 007657128.1 52-306 Acrv1:9_36693220 Sarcophilus harrisii XP 012408976.1 52-321 Lynx1:15_74747852 Lypd1:1_125867622 Sarcophilus harrisii XP 012409037.1 46-390 Ly6h:15_75551268 Sarcophilus harrisii XP 003775118.1 52-321 Lypd3:7_24636550 2210407C18Rik:11_58608204 Monodelphis domestica XP 007507575.1 52-321 Psca:15_74714839 Alligator mississippiensis XP 019356419.1 52-318 Lypd8:11_58379043 Gekko japonicus XP 015269869.1 52-300 Slurp2:15_74742764 Lypd6:2_50066429 Python bivittatus XP 007426908.1 52-300 Lypd6b:2_49787688 $#Anol%&'()(*+is carolinensis XP 008115456.1, 52-300'+%&0-+1.##2324 # Ly6g6e:17_35076902 $#%&'()(*+$#%&'()(*+,'+%,&'+-%+&.##-+.# D ###)/+# '*# # 012324E # SPACA4!"#$%Acr&###::EGFP'()/+)####'*#)/+# '#*# #BF SPACA4 IZUMO1 BF $#%&'()(*+ ,'+%&-+.## 20 µm $#%&'()(*+$#%&'()(*+,'+%&-+.# $# $# a $#* $# * $# * 012324 # acrosome ###)/+# '*# # 012324012#324!"## $%&###'()/+)# '*#####)/+# '*# intact * b * * 3 µm * $# $# $# $# $# * *$#$# * $#$*# $# $# acrosome reacted 5 µm a b a b a b $# 3 µm $# $# $# $# $# $# $# * * * $# * $*# $# $# $$##acrosome$ reacted# (* $ ) # $# $$## $ Fig 1. SPACA4 is expressed in murine sperm. (A) Mammalian SPACA4 is the homolog for fish Bouncer and is expressed specifically in testis. Part of a maximum-likelihood phylogenetic tree based on Ly6/uPAR protein sequence alignments across vertebrates showing SPACA4 and Bouncer (see Fig S2A for the full tree; adapted from (Herberg et al., 2018)). Branches supported by ultrafast bootstrap values (>=95%) are marked with a blue dot. Z-scores of averaged values across adult tissues are shown on the right for Danio rerio (Noda et al., 2021), Xenopus laevis (Session et al., 2016), Mus musculus (Li et al., 2017) and Homo sapiens (www.gtexportal.org). While Bouncer/ Spaca4 mRNAs are expressed in oocytes in fish and frog, mammalian Spaca4 mRNAs are expressed in testis. (B) Mouse Ly6/uPAR genes show diverse expression patterns across adult tissues. The heatmap is color-coded based on z-scores of the normalized gene expression values (average of the square-root) of RNA-Seq data from murine adult tissues (Li et al., 2017). The clustering and dendrogram (left) are based on expression scores. Spaca4 is highlighted in red. Numbers behind gene names indicate location in mice. (C) Mouse Spaca4 RNA is expressed in the male germline. RT-qPCR from murine cDNA from different adult tissues reveals enrichment of Spaca4 specifically in mouse testis. Primers amplifying Hypoxanthineguanine phosphoribosyl transferase (Hrpt) were used as a control. (D) Mouse SPACA4 protein localizes to the sperm head. Immunostaining of SPACA4 (red) and IZUMO1 (green) under permeabilizing conditions detects SPACA4 and IZUMO1 in the sperm head. In contrast to IZUMO1 that relocalizes after the acrosome reaction, SPACA4 does not change its localization. BF, brightfield image. Scale bar, 3 µm. (E) Murine SPACA4 localizes to the inner sperm membrane. Immunofluorescence staining of SPACA4 (red) in mouse spermatozoa under non-permeabilizing conditions. Spermatozoa were derived from transgenic mice expressing EGFP under the control of the Acrosin promoter (Nakanishi et al., 1999), labeling sperm with intact acrosomes (green). Acrosome-reacted spermatozoa are highlighted by an asterisk. Boxed areas are shown below at higher magnification (a – acrosome-reacted sperm; b – spermatozoa with intact acrosome). SPACA4 was detected in acrosome-reacted spermatozoa (a) but not in spermatozoa with intact acrosomes (b). BF, brightfield image. Scale bar, 20 µm (top) and 5 µm (bottom).

4 bioRxiv preprint doi: https://doi.org/10.1101/2021.05.02.442348; this version posted May 2, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

Fig 2. SPACA4 is required for efficient fertilization in male mice. (A and B) Overview of the C57BL/6J-Spaca4 knockout alleles generated by CRISPR/Cas9-mediated targeted mutagenesis. One allele contains a 117-nt in-frame deletion after amino acid 42. The other allele contains a 77-nt out-of-frame deletion after amino acid 47. (A) Schematic of the wild-type and knockout alleles. Yellow dashed lines indicate the site of the deletions. Predicted bridges are indicated in orange. SP, signal peptide; TM, transmembrane region. (B) Genotyping of Spaca4 knockout mice. Spaca4 PCR products are separated based on their sizes on an Agarose-gel.

5 bioRxiv preprint doi: https://doi.org/10.1101/2021.05.02.442348; this version posted May 2, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

Fig 2 (continued). SPACA4 is required for efficient fertilization in male mice. (C) Spaca4 knockout male mice are sub-fertile. Litter sizes of C57BL/6J-Spaca4 wild-type (+/+), heterozygous (+/-) and transheterozygous (-/-) males caged with B6129F1 wild-type females, or B6129F1 wild-type males caged with transheterozygous (-/-) females. Successful mating was confirmed by plug checks. Data are means ± SD. ***p < 0.0001 (Kruskal-Wallis test with Dunn multiple-comparisons test); n.s., not significant. n = number of litters; m = number of male mice tested. (D) Sperm morphology is normal in the absence of SPACA4 protein. Immunostaining of sperm detects SPACA4 protein (red) under permeabilizing conditions in the head of sperm from wild-type (+/+) mice but not in sperm from Spaca4 knockout (-/-) mice. Sperm morphology is normal. DAPI (cyan) staining labels the sperm nucleus. DIC, differential interference contrast image. Scale bar, 10 µm. (E) Sperm from Spaca4 knockout mice has a severely reduced ability to fertilize wild-type oocytes. In vitro fertilization of oocytes from C57BL/6J wild-type females using sperm from C57BL/6J-Spaca4 wild-type (+/+), heterozygous (+/-) or transheterozygous (-/-) males. Plotted is the percentage of 2-cell stage embryos as a measure of successful fertilization. Data are means ± SD. *p = 0.014 (Kruskal-Wallis test with Dunn multiple- comparisons test); n.s., not significant. n = total number of oocytes; m = number of males tested. (F) SPACA4 protein is absent in sperm and testis from Spaca4 knockout mice. Western blot analysis showing the absence of SPACA4 protein in testes and spermatozoa of Spaca4 knockout mice. Testes and spermatozoa of wild-type mice as well as BASIGIN protein, which undergoes proteolytic cleavage during sperm maturation (Noda et al., 2020), are used as control. (G) IZUMO1 localization is normal in Spaca4 knockout mice. Immunostaining of sperm shows normal localization of IZUMO1 (green) in wild-type and SPACA4-deficient sperm. BF, brightfield image. Scale bar, 10 µm.

Spaca4 gene with a neomycin resistance cassette average fertilization rates using spermatozoa from using homologous recombination (Fig S5A-C). wild-type or heterozygous (Spaca4117del/+ or Spaca4 knockout mice appeared indistinguishable Spaca477del/+) mice were at 59.4% or 33.4%, from wild-type and heterozygous littermates, respectively (Fig 2E). Similar defects in IVF were revealing that SPACA4 is dispensable for somatic observed in Spaca4tm1Osb-/- sperm (Fig S5E). The development in mice. inability of SPACA4-deficient sperm to fertilize In line with SPACA4’s sperm-specific wild-type oocytes could be due to loss of expression in mammals, we found that SPACA4 is expression and/or improper localization of necessary for male fertility. The litter size of IZUMO1. However, Western blotting and immuno- transheterozygous (Spaca4117del/77del) as well as fluorescence staining revealed that sperm of homozygous mutant male mice was significantly Spaca4 knockout mice showed normal expression lower than the litter sizes of wild-type or of IZUMO1 (Fig 2F-G). Overall, we conclude that heterozygous male mice (p < 0.0001) (Fig 2C; Fig SPACA4 is an important albeit not absolutely S4B and S5D). In contrast, female fertility was not essential protein for mammalian gamete affected by the Spaca4 mutation (Fig 2C). The interaction. observed defect in male fertility was not due to an inability of Spaca4 knockout males to mate as SPACA4 is required for efficient penetration of verified by the presence of vaginal plugs. the zona pellucida The severely reduced fertility of males lacking SPACA4 could have multiple reasons, including The marked in vitro fertilization defect of -/- reduced sperm count, immotility, or a defect in sperm derived from Spaca4 males showed that gamete interaction. Sperm morphology, numbers SPACA4 is necessary for sperm to efficiently and overall sperm motility were similar in wild- interact with the egg. In mammals, this interaction type and knockout mice (Fig 2D, Fig S4C-E), occurs in two steps: sperm first needs to bind and suggesting that SPACA4 is required during sperm- penetrate the zona pellucida (ZP) before binding to egg interaction. To test this hypothesis, we the egg membrane (oolemma), which enables performed in vitro fertilization (IVF) experiments, sperm-egg fusion. To determine at which stage of which revealed that Spaca4 mutant sperm was fertilization SPACA4-deficient sperm is impaired, severely compromised in its ability to fertilize we quantified the number of spermatozoa bound to wild-type oocytes in vitro: Spermatozoa from the ZP 30 minutes after insemination. We found -/- transheterozygous (Spaca4117del/77del) male mice that in both Spaca4 mutant strains, fewer mutant resulted in a severely reduced average fertilization spermatozoa bound the ZP of wild-type oocytes rate of 2.9% (2-cell stage embryos), whereas the compared to control wild-type sperm (23.5 ± 9.5 in 6 bioRxiv preprint doi: https://doi.org/10.1101/2021.05.02.442348; this version posted May 2, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

Fig 3. SPACA4 enables sperm to bind to and penetrate the zona pellucida. (A) SPACA4 is required for efficient binding of sperm to the ZP. Sperm of the indicated genotypes was incubated for 30 minutes with THY-treated oocytes (with intact ZP) of matching genetic backgrounds. Plotted is the number of sperm bound to ZP-containing oocytes. Data are means ± SD. P-values are of Student’s t test. n = total number of oocytes; N = number of replicates. (B) Schematic of sperm bound to the cumulus-oocyte-complex (COC), and experimental treatments used to remove the cumulus cells (by treatment of COCs with hyaluronidase) and the zona pellucida (ZP) (by treatment of cumulus-free oocytes with acidified Tyrode’s solution) from the COCs. (C) SPACA4 is required for ZP penetration but not for oolemma binding and fusion. IVF performed with COCs from super-ovulated C57BL/6J females, cumulus cell-free oocytes (oocyte with ZP), and ZP-free oocytes with sperm from either wild-type C57BL/6J males or age-matched Spaca4117nt-del/77nt-del males. Plotted is the percentage of 2-cell stage embryos as a measure of successful fertilization. Data are means ± SD. P-values are of a Kruskal-Wallis test with Dunn multiple-comparisons; n.s., not significant. n = total number of oocytes; N = number of replicates. (D) Model of SPACA4’s function in murine fertilization. SPACA4 is required for efficient ZP penetration in mice. 7 bioRxiv preprint doi: https://doi.org/10.1101/2021.05.02.442348; this version posted May 2, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

the case of Spaca4117/77 vs. 30.7 ± 7.0 for wild type mutations in humans and male sub-fertility in (BJ6) (p < 0.001); 34.3 ± 11.1 in the case of patients presenting with normal sperm count, Spaca4tm1Osb vs. 59.7 ± 10.7 for wild type (BDF1) morphology and motility. (p < 0.001)) (Fig 3A). The reduced binding of SPACA4 and Bouncer are not the only Ly6/ SPACA4-deficient sperm to the ZP, combined with uPAR proteins that have been linked to vertebrate a reduced motility of mutant sperm during reproduction. Several other members of this large incubation with cumulus-oocyte-complexes gene family show a testis-restricted expression (COCs) (Fig S6), prompted us to ask whether pattern in mammals (Fig 1B, Fig S2C), some of SPACA4 might be involved in the penetration of which (Tex101, Lypd4, Pate4 and the Pate gene the protective layers that surround the oocyte, cluster) have been confirmed in genetic knockout namely the cumulus layer and the ZP. To test studies to be required for male reproduction in whether removal of the cumulus cells and/or the mice (Fujihara et al., 2014, 2019; Noda et al., ZP can rescue the fertility defect of SPACA4- 2019). In light of these known male-restricted deficient sperm in IVF, wild-type oocytes were requirements for Ly6/uPAR proteins in mammals, treated with hyaluronidase and acidified Tyrode’s SPACA4 and Bouncer present an interesting to remove the cumulus layer and ZP, respectively example of homologous proteins that diverged (Richard Behringer, Marina Gertsenstein, Kristina both in terms of gene expression pattern and Nagy, 2014) (Fig 3B). Fertilization by wild-type function. Our phylogenetic sequence analysis sperm was not significantly affected by the shows that Bouncer and SPACA4 are the closest different treatments (Fig 3C). Similarly, removal of homologs among all other Ly6/uPAR family the cumulus cells did not rescue the low members (Fig S2A), yet they have opposing fertilization rate of SPACA4-deficient sperm germline-specific expression patterns that broadly (18.8% (COCs) vs. 4.7% (cumulus cells removed) correlate with external (fish, amphibians; (p = 0.884); Fig 3C). However, removal of the ZP expressed in the egg) versus internal (mammals, enabled SPACA4-deficient sperm to fertilize zona- reptiles; expressed in the sperm) fertilization (Fig free oocytes at a similarly high rate (95.8%) as 1A, Fig S2A). We currently do not know how this wild-type sperm (97.1%) (p-value for mutant different gene expression pattern arose, nor sperm fertilizing oocytes with ZP vs. ZP-free = whether it evolved as a consequence of the 0.033) (Fig 3C). We therefore conclude that different fertilization modes. One possibility is that murine SPACA4 is required for the sperm’s ability an ancestral Ly6/uPAR protein might have been to efficiently traverse the ZP (Fig 3D). expressed in both male and female gonads, and that sex-specific loss of expression occurred either by DISCUSSION chance in different lineages or in response to Here we reveal that SPACA4, the mammalian functional requirements that made restricted homolog of fish Bouncer (Herberg et al., 2018), is expression of SPACA4/Bouncer in either the male required for normal male fertility in mice (Fig 2). or in the female germline beneficial. Acquisition of We find that SPACA4 is expressed in the sperm a restricted expression domain from an initially head and gets exposed upon the acrosome reaction broader expression pattern has been proposed for (Fig. 1D-E), which enables sperm to efficiently other members of the Ly6/uPAR protein family, bind to and penetrate through the zona pellucida namely for snake toxins that evolved a venom (Fig 3). Thus, our work provides genetic evidence gland-restricted expression pattern (Hargreaves et for an important function for SPACA4 in male al., 2014). reproduction in mice. Together with previous The difference in expression pattern also SPACA4-antibody blocking experiments with extends to a difference in functionality at least in human sperm and the conserved expression pattern the case of the two example model organisms of human SPACA4 in sperm (Shetty et al., 2003), zebrafish (Herberg et al., 2018) and mice (this our results in mice could have direct relevance for work): While Bouncer in fish is required in the egg male fertility in humans. Future work will be for sperm binding to the oolemma (Herberg et al., needed to explore a possible link between Spaca4 2018), the results presented here reveal that in mice SPACA4 is dispensable for sperm binding to the 8 bioRxiv preprint doi: https://doi.org/10.1101/2021.05.02.442348; this version posted May 2, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

oolemma and instead required for the preceding conspecific sperm reaches the egg (Levitan, 2017; process of zona pellucida penetration/binding. In Levitan and Ferrell, 2006), and whose oocyte- this regard it is interesting to note that mammalian specific expression of Bouncer could contribute to and fish gametes differ in key aspects: Mammalian post-copulation female mate choice (also called sperm has an acrosome, a specialized vesicle that cryptic female mate choice) (Firman et al., 2017). needs to first undergo exocytosis in order for Thus, sperm-expressed proteins like SPACA4 important membrane-localized fertility factors to could promote the efficiency of fertilization while oocyte-expressed proteins like Bouncer could become exposed on the surface of the sperm head support the selection of conspecific sperm. Future (Hirohashi and Yanagimachi, 2018; Satouh et al., experiments will be required to elucidate how 2012). This step is important for successful zona SPACA4 promotes fertilization in mammals and to pellucida penetration and fertilization. Moreover, which extent the mechanism differs between mammalian sperm needs to first bind to and SPACA4 and Bouncer. penetrate the outer coat before gaining access to Overall, our study on SPACA4/Bouncer the egg membrane. Fish sperm, on the other hand, highlights an interesting example of a vertebrate- lacks an acrosome and has direct access to the specific sperm-egg interaction protein that evolved oolemma through a pre-formed funnel in the outer a different function and gene expression pattern in protective layer of the fish egg. One can therefore fish versus mammals. speculate that acquisition of a sperm-specific expression of SPACA4 in mammals was beneficial to promote sperm traversal through the additional protective barrier that the zona pellucida ACKNOWLEDGMENTS comprises. In zebrafish, current data suggests that We thank the animal facility personnel at IMP/ Bouncer acts in a unilateral manner by interacting IMBA, in particular Jasmin Ecker and Lukas with a still unknown factor on the opposing gamete Laister, and Naoko Nagasawa at Osaka University/ since (1) expression in sperm does not rescue National Cerebral and Cardiovascular Center for infertility of bouncer mutant females, and (2) excellent care of animals and for help with mouse successful sperm entry requires compatibility husbandry; J.R. Wojciechowski, L. Haas from the between a species’ Bouncer and the sperm Obenauf lab, A. Fedl from the Busslinger lab, D. (Herberg et al., 2018). Whether a similar mode of Keays and I. Baptista from the Keays lab, and E. action, e.g. a SPACA4 interacting protein Chatzidaki and J. Gassler from the Tachibana lab expressed in the oocyte and a possible involvement for technical assistance and advice with mouse in species-specificity of sperm-egg interaction, also work; K. Aumayer and her team of the biooptics applies to mammals is currently unclear. Given the facility at the ViennaBiocenter for support with divergence in function of Ly6/uPAR proteins microscopy; the entire Pauli lab, particularly Krista (Bychkov et al., 2017; Loughner et al., 2016), it is Gert and Victoria Deneke, for valuable discussions possible that SPACA4 acquired a different mode of and comments on the project and manuscript. action from Bouncer, e.g. by acting in cis through interacting with other sperm-expressed membrane proteins, or by interacting with oviductal proteins AUTHOR CONTRIBUTIONS as suggested for its interaction with plasminogen Y.F., A.P and M.I. designed the study. S.H., (Ferrer et al., 2016). Moreover, one can speculate Y.F., A.B., K.P., K.K. and T.L. performed that for species with internal fertilization, a experiments with the help of K.K., T.L., H.C.T. and selection step determining gamete compatibility at O.O. S.H., Y.F., A.B., K.P., K.K., T.L., H.C.T., M.I. the stage of sperm-egg interaction might be less and A.P. analyzed the data. M.N. performed important since mating partner selection alone can evolutionary and gene expression analyses. A.P. ensure that only the selected partner’s sperm will wrote the manuscript with help of S.H., Y.F., A.B. be available for fertilization. This is not the case and M.I. and input from all authors. for species performing external fertilization who cannot guarantee by pre-mating choice that only 9 bioRxiv preprint doi: https://doi.org/10.1101/2021.05.02.442348; this version posted May 2, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

FUNDING Aydin, H., Sultana, A., Li, S., Thavalingam, A., and This work was supported on the Austrian side Lee, J.E. (2016). Molecular architecture of the (Pauli lab) by the Research Institute of Molecular human sperm IZUMO1 and egg JUNO fertilization Pathology (IMP), Boehringer Ingelheim, the complex. Nature 534, 562–565. Austrian Academy of Sciences, FFG (Headquarter Bianchi, E., Doe, B., Goulding, D., and Wright, grant FFG-852936), the FWF START program (Y G.J. (2014). Juno is the egg Izumo receptor and is essential for mammalian fertilization. Nature 508, 1031-B28) to A.P., the HFSP Career Development 483–487. Award (CDA00066/2015) and the HFSP Young Investigator Award to A.P., EMBO-YIP funds to Bychkov, M., Shenkarev, Z.O., and Lyukmanova, E. (2017). Three Finger Proteins from the Ly6 / A.P., a DOC Fellowship from the Austrian uPAR Family: Functional diversity within one Academy of Sciences to S.H., and a Boehringer structural motif. Biochemistry. Ingelheim Fonds (BIF) PhD fellowship to A.B. Cabrera-Quio, L.E., Schleiffer, A., Mechtler, K., This work was supported on the Japanese side and Pauli, A. (2021). Zebrafish Ski7 tunes RNA (Ikawa lab, Fujihara lab) by the Ministry of levels during the oocyte-to-embryo transition. Education, Culture, Sports, Science and PlosGenetics 17(2): e10, 1–30. Technology/Japan Society for the Promotion of Dresselhaus, T., Sprunck, S., and Wessel, G.M. Science KAKENHI Grants JP15H05573, (2016). Fertilization mechanisms in flowering JP16KK0180, and JP20KK0155 to Y.F., and plants. Curr. Biol. 26, R125–R139. JP25112007, JP17H01394, and JP19H05750 to Ernst, C., Eling, N., Martinez-Jimenez, C.P., M.I.; Japan Agency for Medical Research and Marioni, J.C., and Odom, D.T. (2019). Staged Development Grant JP19gm5010001 to M.I.; developmental mapping and X chromosome Takeda Science Foundation grants to Y.F. and M.I.; transcriptional dynamics during mouse spermatogenesis. Nat. Commun. 10. Mochida Memorial Foundation for Medical and Pharmaceutical Research grant to Y.F.; The Evans, J.P. (2012). Sperm-egg interaction. Annu. Sumitomo Foundation Grant for Basic Science Rev. Physiol. 74, 477–502. Research Projects to Y.F.; Senri Life Science Evans, J.P., and Sherman, C.D.H. (2013). Sexual Foundation grant to Y.F.; Intramural Research Fund Selection and the Evolution of Egg-Sperm Interactions in Broadcast-Spawning Invertebrates. (grants 30-2-5 and 31-6-3) for Cardiovascular Biol. Bull. 224, 166–183. Diseases of National Cerebral and Cardiovascular Center to Y.F.; Eunice Kennedy Shriver National Fernandez-Fuertes, B., Laguna-Barraza, R., Fernandez-Gonzalez, R., Gutierrez-Adan, A., Institute of Child Health and Human Development Blanco-Fernandez, A., O’Doherty, A.M., Di Fenza, Grants R01HD088412 and P01HD087157 to M.I.; M., Kelly, A.K., Kölle, S., and Lonergan, P. (2017). and the Bill & Melinda Gates Foundation (grant Subfertility in bulls carrying a nonsense mutation INV-001902 to M.I.). The funders had no role in in transmembrane protein 95 is due to failure to study design, data collection and analysis, decision interact with the oocyte vestments. Biol. Reprod. 97, 50–60. to publish, or preparation of the manuscript. Ferrer, M.J.S., Xu, W., Shetty, J., Herr, J., and Oko, R. (2016). Plasminogen Improves Mouse IVF by DECLARATION OF INTERESTS Interactions with Inner Acrosomal Membrane- The authors declare no competing interests. Bound MMP2 and SAMP141. Biol. Reprod. 94, 1– 11. Firman, R.C., Gasparini, C., Manier, M.K., and REFERENCES Pizzari, T. (2017). Postmating Female Control: 20 Years of Cryptic Female Choice. Trends Ecol. Avella, M.A., Xiong, B., and Dean, J. (2013). The Evol. 32, 368–382. molecular basis of gamete recognition in mice and humans. Mol. Hum. Reprod. 19, 279–289. Fujihara, Y., Kaseda, K., Inoue, N., Ikawa, M., and Okabe, M. (2013a). Production of mouse pups Avella, M.A., Baibakov, B., and Dean, J. (2014). A from germline transmission-failed knockout single domain of the ZP2 zona pellucida protein chimeras. Transgenic Res. 22, 195–200. mediates gamete recognition in mice and humans. J. Cell Biol. 205, 801–809. Fujihara, Y., Tokuhiro, K., Muro, Y., Kondoh, G., Araki, Y., Ikawa, M., and Okabe, M. (2013b).

10 bioRxiv preprint doi: https://doi.org/10.1101/2021.05.02.442348; this version posted May 2, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

Expression of TEX101, regulated by ACE, is Herberg, S., Gert, K.R., Schleiffer, A., and Pauli, essential for the production of fertile mouse A. (2018). The Ly6/uPAR protein Bouncer is spermatozoa. Proc. Natl. Acad. Sci. 110, 8111– necessary and sufficient for species-specific 8116. fertilization. Science (80-. ). 361, 1029–1033. Fujihara, Y., Okabe, M., and Ikawa, M. (2014). Hirohashi, N., and Yanagimachi, R. (2018). Sperm GPI-Anchored Protein Complex, LY6K/TEX101, acrosome reaction: Its site and role in fertilization. Is Required for Sperm Migration into the Oviduct Biol. Reprod. 99, 127–133. and Male Fertility in Mice1. Biol. Reprod. 90, 1–6. Hu, Y., Flockhart, I., Vinayagam, A., Bergwitz, C., Fujihara, Y., Noda, T., Kobayashi, K., Oji, A., Berger, B., Perrimon, N., and Mohr, S.E. (2011). Kobayashi, S., Matsumura, T., Larasati, T., Oura, An integrative approach to ortholog prediction for S., Kojima-Kita, K., Yu, Z., et al. (2019). disease-focused and other functional studies. BMC Identification of multiple male reproductive Bioinformatics 12. tractspecific proteins that regulate sperm migration Ikawa, M., Benham, A.M., Okabe, M., Ikawa, M., through the oviduct in mice. Proc. Natl. Acad. Sci. Inoue, N., Benham, A.M., and Okabe, M. (2010). U. S. A. 116, 18498–18506. Fertilization : a sperm ’ s journey to and interaction Fujihara, Y., Lu, Y., Noda, T., Oji, A., Larasati, T., with the oocyte Find the latest version : Review Kojima-kita, K., Yu, Z., Matzuk, R.M., Matzuk, series Fertilization : a sperm ’ s journey to and M.M., and Ikawa, M. (2020). Spermatozoa lacking interaction with the oocyte. 120, 984–994. Fertilization Influencing Membrane Protein Ikawa, M., Tokuhiro, K., Yamaguchi, R., Benham, (FIMP) fail to fuse with oocytes in mice. PNAS 1– A.M., Tamura, T., Wada, I., Satouh, Y., Inoue, N., 8. and Okabe, M. (2011). Calsperin is a testis-specific Gagnon, J.A., Valen, E., Thyme, S.B., Huang, P., chaperone required for sperm fertility. J. Biol. Ahkmetova, L., Pauli, A., Montague, T.G., Chem. 286, 5639–5646. Zimmerman, S., Richter, C., and Schier, A.F. Inoue, N., Ikawa, M., Isotani, A., and Okabe, M. (2014). Efficient mutagenesis by Cas9 protein- (2005). The immunoglobulin superfamily protein mediated oligonucleotide insertion and large-scale Izumo is required for sperm to fuse with eggs. assessment of single-guide RNAs. PLoS One 9, 5– Nature 434, 234–238. 12. Inoue, N., Hagihara, Y., and Wada, I. (2021). Gahlay, G., Gauthier, L., Baibakov, B., Epifano, Evolutionarily conserved sperm factors , DCST1 O., and Dean, J. (2010). Gamete Recognition in and DCST2 , are required for gamete fusion. Elife Mice Depends on the Cleavage Status of an Egg’s 49, 1–12. Zona Pellucida Protein. Science (80-. ). 329, 216– 219. Kaji, K., Oda, S., Shikano, T., Ohnuki, T., Uematsu, Y., Sakagami, J., Tada, N., Miyazaki, S., Gert, K.R., and Pauli, A. (2019). Species-specific and Kudo, a (2000). The gamete fusion process is mechanisms during fertilization (Elsevier Inc.). defective in eggs of Cd9-deficient mice. Nat. Goodson, S.G., Zhang, Z., Tsuruta, J.K., Wang, W., Genet. 24, 279–282. and O’Brien, D.A. (2011). Classification of mouse Kato, K., Satouh, Y., Nishimasu, H., Kurabayashi, sperm motility patterns using an automated A., Morita, J., Fujihara, Y., Oji, A., Ishitani, R., multiclass support vector machines model. Biol. Ikawa, M., and Nureki, O. (2016). Structural and Reprod. 84, 1207–1215. functional insights into IZUMO1 recognition by Grayson, P. (2015). Izumo1 and Juno : the JUNO in mammalian fertilization. Nat. Commun. evolutionary origins and coevolution of essential 7, 1–9. sperm – egg binding partners. R. Soc. Open Sci. 2. Kosman, E.T., and Levitan, D.R. (2014). Sperm Han, L., Nishimura, K., Sadat Al Hosseini, H., competition and the evolution of gametic Bianchi, E., Wright, G.J., and Jovine, L. (2016). compatibility in externally fertilizing taxa. Mol. Divergent evolution of vitamin b9 binding Hum. Reprod. 20, 1190–1197. underlies juno-mediated adhesion of mammalian Krauchunas, A.R., Marcello, M.R., and Singson, A. gametes. Curr. Biol. 26, R100–R101. (2016). The molecular complexity of fertilization: Hargreaves, A.D., Swain, M.T., Hegarty, M.J., Introducing the concept of a fertilization synapse. Logan, D.W., and Mulley, J.F. (2014). Restriction Mol. Reprod. Dev. 83, 376–386. and recruitment-gene duplication and the origin Levitan, D.R. (2017). Do Sperm Really Compete and evolution of snake venom toxins. Genome and Do Eggs Ever Have a Choice? Adult Biol. Evol. 6, 2088–2095. Distribution and Gamete Mixing Influence Sexual Selection, Sexual Conflict, and the Evolution of 11 bioRxiv preprint doi: https://doi.org/10.1101/2021.05.02.442348; this version posted May 2, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

Gamete Recognition Proteins in the Sea. Am. Nat. Noda, T., Blaha, A., Fujihara, Y., Gert, K.R., 191, 000–000. Emori, C., Deneke, V.E., Oura, S., Berent, S., Kodani, M., Panser, K., et al. (2021). Sperm Levitan, D.R., and Ferrell, D.L. (2006). Selection membrane proteins DCST1 and DCST2 are on Gamete Recognition proteins depends on sex, required for the sperm-egg fusion process in mice density, and genotype frequency. Science (80-. ). and fish. BioRxiv. 312, 267–270. Ohto, U., Ishida, H., Krayukhina, E., Uchiyama, S., Lewis, C.A., Talbot, C.F., and Vacquier, V.D. Inoue, N., and Shimizu, T. (2016). Structure of (1982). A protein from abalone sperm dissolves the IZUMO1-JUNO reveals sperm-oocyte recognition egg vitelline layer by a nonenzymatic mechanism. during mammalian fertilization. Nature 534, 566– Dev. Biol. 92, 227–239. 569. Li, B., Qing, T., Zhu, J., Wen, Z., Yu, Y., Okabe, M. (2018a). Sperm-egg interaction and Fukumura, R., Zheng, Y., Gondo, Y., and Shi, L. fertilization: Past, present, and future. Biol. (2017). A Comprehensive Mouse Transcriptomic Reprod. 99, 134–146. BodyMap across 17 Tissues by RNA-seq. Sci. Rep. 7, 1–10. Okabe, M. (2018b). Beware of memes in the interpretation of your results – lessons from gene- Lorenzetti, D., Poirier, C., Zhao, M., Overbeek, disrupted mice in fertilization research. FEBS Lett. P.A., Harrison, W., and Bishop, C.E. (2014). A 592, 2673–2679. transgenic insertion on mouse chromosome 17 inactivates a novel immunoglobulin superfamily Palumbi, S.R. (2009). Speciation and the evolution gene potentially involved in sperm-egg fusion. of gamete recognition genes: Pattern and process. Mamm. Genome 25, 141–148. Heredity (Edinb). 102, 66–76. Loughner, C.L., Bruford, E.A., McAndrews, M.S., Park, S., Shimada, K., Fujihara, Y., Xu, Z., Delp, E.E., Swamynathan, S., and Swamynathan, Shimada, K., Larasati, T., Pratiwi, P., Matzuk, S.K. (2016). Organization, evolution and functions R.M., Devlin, D.J., Yu, Z., et al. (2020). CRISPR/ of the human and mouse Ly6/uPAR family genes. Cas9-mediated genome-edited mice reveal 10 Hum. Genomics 10, 10. testis-enriched genes are dispensable for male fecundity. Biol. Reprod. 103, 195–204. Lyon, J.D., and Vacquier, V.D. (1999). Regions Mediating Species-Specific Recognition of the P a u s c h , H . , K ö l l e , S . , Wu r m s e r, C . , Abalone Egg Vitelline Envelope. 159, 151–159. Schwarzenbacher, H., Emmerling, R., Jansen, S., Trottmann, M., Fuerst, C., Götz, K.U., and Fries, Miyado, K., Yamada, G., Yamada, S., Hasuwa, H., R. (2014). A Nonsense Mutation in TMEM95 Nakamura, Y., Ryu, F., Suzuki, K., Kosai, K., Encoding a Nondescript Transmembrane Protein Inoue, K., Ogura, A., et al. (2000). Requirement of Causes Idiopathic Male Subfertility in Cattle. PLoS CD9 on the egg plasma membrane for fertilization. Genet. 10. Science (80-. ). 287, 321–324. Raj, I., Sadat, H., Hosseini, A., Dioguardi, E., Nakanishi, T., Ikawa, M., Yamada, S., Parvinen, Villa, A., Sanctis, D. De, Jovine, L., Raj, I., Sadat, M., Baba, T., Nishimune, Y., and Okabe, M. H., Hosseini, A., et al. (2017). Structural Basis of (1999). Real-time observation of acrosomal Egg Coat-Sperm Recognition at Article Structural dispersal from mouse sperm using GFP as a marker Basis of Egg Coat-Sperm Recognition at protein. FEBS Lett. 449, 277–283. Fertilization. Cell 169, 1315-1326.e17. Le Naour, F., Rubinstein, E., Jasmin, C., Prenant, Rankin, T., Familari, M., Lee, E., Ginsberg, A., M., and Boucheix, C. (2000). Severely Reduced Dwyer, N., Blanchette-Mackie, J., Drago, J., Female Fertility in CD9-Deficient Mice. Science Westphal, H., and Dean, J. (1996). Mice (80-. ). 287, 319–321. homozygous for an insertional mutation in the Zp3 Noda, T., Fujihara, Y., Matsumura, T., Oura, S., gene lack a zona pellucida and are infertile. Kobayashi, S., and Ikawa, M. (2019). Seminal Development 122, 2903–2910. vesicle secretory protein 7, PATE4, is not required Rankin, T.L., Tong, Z.B., Castle, P.E., Lee, E., for sperm function but for copulatory plug Gore-Langton, R., Nelson, L.M., and Dean, J. formation to ensure fecundity. Biol. Reprod. 100, (1998). Human ZP3 restores fertility in Zp3 null 1035–1045. mice without affecting order-specific sperm Noda, T., Lu, Y., Fujihara, Y., Oura, S., Koyano, T., binding. Development 125, 2415–2424. Kobayashi, S., Matzuk, M.M., and Ikawa, M. Richard Behringer, Marina Gertsenstein, Kristina (2020). Sperm proteins SOF1, TMEM95, and Nagy, A.N. (2014). Manipulating the Mouse SPACA6 are required for sperm−oocyte fusion in Embryo (Cold Spring Harbor Laboratory Press.). mice. PNAS 1–10. 12 bioRxiv preprint doi: https://doi.org/10.1101/2021.05.02.442348; this version posted May 2, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

Satouh, Y., Inoue, N., Ikawa, M., and Okabe, M. mice, C57BL/6N-Spaca4tm1Osb, were deposited to (2012). Visualization of the moment of mouse the RIKEN BioResource Center (http:// sperm-egg fusion and dynamic localization of mus.brc.riken.jp/en/) and are available to the IZUMO1. J. Cell Sci. 125, 4985–4990. scientific community. Session, A.M., Uno, Y., Kwon, T., Chapman, J.A., Generation of Spaca4 knockout mice Toyoda, A., Takahashi, S., Fukui, A., Hikosaka, A., The mouse Spaca4 gene consists of a single Suzuki, A., Kondo, M., et al. (2016). Genome exon and maps to chromosome 7. Two strategies evolution in the allotetraploid frog Xenopus laevis. were used to generate Spaca4 knockout mice: 1) Nature 538, 336–343. CRISPR-Cas9-based gene targeting (IMP, Vienna, Shetty, J., Wolkowicz, M.J., Digilio, L.C., Klotz, Austria); and 2) replacement of the Spaca4- K.L., Jayes, F.L., Diekman, A.B., Westbrook, V.A., encoding exon with a neomycin selection cassette Farris, E.M., Hao, Z., Coonrod, S.A., et al. (2003). via gene targeting of ES-cells (Osaka University, SAMP14, a novel, acrosomal membrane- Japan). associated, glycosylphosphatidylinositol-anchored CRISPR-Cas9-based gene targeting (resultant member of the Ly-6/urokinase-type plasminogen Spaca4 alleles: C57BL/6J-Spaca4117del and C57BL/ activator receptor superfamily with a role in sperm- 6J-Spaca477del): The Spaca4 knockout mice were egg interaction. J. Biol. Chem. 278, 30506–30515. generated using CRISPR-Cas9-based gene targeting. Two guide RNAs (gRNAs) targeting the Swanson, W.J., and Vacquier, V.D. (1997). The coding region of Spaca4 were generated according abalone egg vitelline envelope receptor for sperm to published protocols (Gagnon et al., 2014) by lysin is a giant multivalent molecule. Proc. Natl. oligo annealing followed by T7 polymerase-driven Acad. Sci. U. S. A. 94, 6724–6729. in vitro transcription (gene-specific targeting oligo: Swanson, W.J., and Vacquier, V.D. (2002). The SPACA4_gRNA1 and 2; common tracer oligo). rapid evolution of reproductive proteins. Nat. Rev. For gRNA injections, zygotes were isolated from Genet. 3, 137–144. super-ovulated donor female (C57BL/6J) mice on the day of the coagulation plug (= E0.5). To Yamaguchi, R., Yamagata, K., Ikawa, M., Moss, remove cumulus cells, zygotes were incubated in S.B., and Okabe, M. (2006). Aberrant Distribution hyaluronidase solution (~0.3 mg/ml). The injection of ADAM3 in Sperm from Both Angiotensin- mix (50 ng/µl Cas9 mRNA (Sigma-Aldrich), 50 Converting Enzyme (Ace)- and Calmegin (Clgn)- ng/µl SPACA4_gRNA1 and 2) was injected into Deficient Mice1. Biol. Reprod. 75, 760–766. the cytoplasm of the zygotes. Injected zygotes Yamaguchi, R., Yamagata, K., Hasuwa, H., Inano, were incubated for at least 15 minutes at 37°C and E., Ikawa, M., and Okabe, M. (2008). Cd52, 5% CO2. Surviving zygotes were transferred into known as a major maturation-associated sperm the oviducts of pseudo-pregnant recipient females. membrane antigen secreted from the epididymis, is The resulting pubs were genotyped using primers not required for fertilization in the mouse. Genes to SPACA4_gt_F and SPACA4_gt_R. Cells 13, 851–861. Gene targeting in ES cells (resultant Spaca4 allele: C57BL/6N-Spaca4tm1Osb): The targeting vector was constructed using pNT1.1 (https:// www.ncbi.nlm.nih.gov/nuccore/JN935771). A 1.9 kb NotI-XhoI short arm fragment and a 5.1 kb PacI-MfeI long arm fragment were obtained by MATERIALS AND METHODS PCR amplification using BAC DNA (RP24-343L2) as a template. The primers used were: Mouse lines and husbandry Spaca4_targeting-s_F, Spaca4_targeting-s_R, for All mouse experiments were conducted the short arm; Spaca4_targeting-l_F, according to Austrian and European guidelines for Spaca4_targeting-l_R for the long arm. The animal research and approved by local Austrian targeting construct was linearized with ClaI and authorities or by the Animal Care and Use electroporated into EGR-G101 [C57BL/6N- Committee of the Research Institute for Microbial Tg(CAG/Acr-EGFP)] embryonic stem (ES) cells Diseases, Osaka University, Japan (#Biken-AP- (Fujihara et al., 2013a). Potentially targeted ES cell H30-01). Mice were maintained under a 10/14-hr clones were separated by positive/negative light/dark cycle (IMP, Vienna, Austria) or under a selection with G418 and ganciclovir. Correct 12-hr light/dark cycle (Osaka University, Japan). targeting of the Spaca4 allele in ES cell clones and Wild-type mice were purchased from CLEA Japan germline transmission were determined by PCR. (Tokyo, Japan) and Japan SLC (Shizuoka, Japan) S c r e e n i n g p r i m e r s u s e d w e r e : (Osaka University, Japan). The mouse strain S p a c a 4 _ s c r e e n i n g + g t # 7 8 1 a n d Tg(Acr-EGFP)1Osb has been reported before Spaca4_screening+gt#5081 for the short arm; (Nakanishi et al., 1999). The Spaca4 knockout 13 bioRxiv preprint doi: https://doi.org/10.1101/2021.05.02.442348; this version posted May 2, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

S p a c a 4 _ s c r e e n i n g + g t # 5 1 7 3 a n d cloning vectors provided by the StrataClone PCR Spaca4_screening+gt#678 for the long arm. The cloning kit (Stratagene). The cloning was mutant ES clones were injected into 8-cell stage performed according to the manufacturer’s ICR embryos, and chimeric blastocysts were protocol. Of the resulting bacterial colonies 96 transferred into the uterine horns of pseudo- colonies were picked and sequenced using primer pregnant ICR females the next day. To confirm SPACA4_gt_F. Two mice, one with a 77-nt and the germline transmission, chimeric males were mated other one with a 117-nt deletion in the coding with B6D2F1 females. Offspring from sequence of Spaca4 were selected (see below) and heterozygous intercrosses were genotyped by PCR. used for further in- and out-crossing. Both The genotyping primers used were: deletions were also confirmed by sequencing the Spaca4_gt#5269, Spaca4_gt#5298, and PCR products from the genotyping of the next Spaca4_screening+gt#781. Two bands, a 0.3-kb generation (F2). band as the wild-type allele and a 0.5-kb band as the knockout allele, were amplified by PCR. Wild-type Spaca4 ORF: Atggtccttggctggccactgcttctggtgttggttctttgcccaggtgtgacag Genotyping: Extraction of gDNA from mouse gcatcaaggactgcgtcttctgtgagctgactgactctgctcggtgccctggca ear clips and genotyping PCR cacacatgcgctgtggggatgacgaagattgcttcacaggccacggagtag Mice were genotyped at weaning age (around cccagggtgtggggcccatcatcaacaaaggctgcgtgcactccaccag ctgtggccgcgaggaacccatcagctacatgggcctcacatacagtctcacc 3 weeks) using ear clips. gDNA was extracted from accacctgctgttctggccacctttgcaataagggcactggcctttccacaggg the ear clips using one of the following three gctaccagcctgtcactgggtctgcagctgctcctgggcctgttgctgctgcttc alternative protocols. According to one protocol, aatactggctgtga ear clips were lysed by incubation in 25 µl of Bold sequence is deleted in the mutant Spaca477del QuickExtract Solution (QE09050, Lucigen) at 65 Underlined sequence is deleted in the mutant °C and shaking at 600 rpm for 30 min. According Spaca4117del to a second protocol, ear clips were lysed in 100 µl lysis buffer (0.1 M NaCl, 0.5% SDS, 10 mM TRIS Wild-type SPACA4 protein: pH 8.0, 0.25 mM EDTA, 2µg/µl proteinase K) at MVLGWPLLLVLVLCPGVTGIKDCVFCELTDSARC 55 °C for 3 hours. To precipitate the cellular debris PGTHMRCGDDEDCFTGHGVAQGVGPIINKGCVHST 60 µl NaCl were added and the sample was SCGREEPISYMGLTYSLTTTCCSGHLCNKGTGLSTG ATSLSLGLQLLLGLLLLLQYWL* centrifuged at (21000 g) for 10 min at 4 °C. The Underlined sequence: signal peptide supernatant was centrifuged a second time (21000 Bold sequence is deleted in the mutant Spaca477del g, 10 min at 4 °C). To precipitate the DNA, 160 µl Underlined italics sequence is deleted in the mutant of cold 100% ethanol was added, and the sample Spaca4117del was centrifuged again (21000 g, 10 min at 4 °C). The pellet was washed a second time using 250 µl Spaca477del ORF: of 75% ethanol (centrifugation at 21000 g, 5 min at Atggtccttggctggccactgcttctggtgttggttctttgcccaggtgtgacag 4 °C). The cleaned DNA pellet was dried at 37 °C gcatcaaggactgcgtcttctgtgagctgactgactctgctcggtgccctggc acacacatgcgctgtggggatgacgaagattgcttaggaacccatcagcta for 5-15 min and solubilized in 50 µl ddH2O. According to a third protocol, a commercial lysis catgggcctcacatacagtctcaccaccacctgctgttctggccacctttgca buffer was used to extract the gDNA from ear ataa clippings (DirectPCR Lysis Reagent (Mouse Tail), SPACA477del protein: ViagenBiotech). 200 µl of the commercial lysis MVLGWPLLLVLVLCPGVTGIKDCVFCELTDSARC buffer and 1 µl of 100 mg/ml proteinase K were PGTHMRCGDDEDCLGTHQLHGPHIQSHHHLLF added to each ear clip and incubated at 55 °C under WPPLQ* shaking (600 rpm) overnight. For heat inactivation (bold sequence does not exist in the wild type as it is a the samples were heated to 85 °C for 45 min under consequence of the out-of-frame deletion.) vigorous shaking (800 rpm). To amplify the Spaca4 coding region, standard Spaca4117del ORF: Taq Polyermase or Q5 Hot Start polymerase (New Atggtccttggctggccactgcttctggtgttggttctttgcccaggtgtgacag England Biolabs) was used with primers gcatcaaggactgcgtcttctgtgagctgactgactctgctcggtgccctggc SPACA4_gt_F and SPACA4_gt_R according to acacacatgcgctgtggcctcacatacagtctcaccaccacctgctgttctg gccacctttgcaataagggcactggcctttccacaggggctaccagcctgtc the manufacturer’s protocol. The size of the PCR actgggtctgcagctgctcctgggcctgttgctgctgcttcaatactggctgtg product was analyzed on a 2% agarose gel. a

Identification of the nature of the SPACA477del SPACA4117del protein: and SPACA4117del mutations MVLGWPLLLVLVLCPGVTGIKDCVFCELTDSARC To analyze the nature of the Spaca4 mutations, PGTHMRCGLTYSLTTTCCSGHLCNKGTGLSTGAT the PCR products from the genotyping of the first SLSLGLQLLLGLLLLLQYWL* outcross (generation F1) were cloned into the

14 bioRxiv preprint doi: https://doi.org/10.1101/2021.05.02.442348; this version posted May 2, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

RT-PCR and RT-qPCR of Spaca4 hour. Oocytes from superovulated female mice Mouse cDNA was prepared from multiple (C57BL/6J) were introduced into a drop of CARD adult tissues of ICR mice and from testes of MEDIUM. To prepare cumulus- or zona-free Spaca4 knockout mice. Total RNA was reverse- oocytes (Fig 3C), cumulus-oocyte-complexes transcribed into cDNA using a SuperScript III (COCs) were collected in M2 medium (MR-015P-5D, Merck) and treated with 300 µg/ First-Strand Synthesis System for RT-PCR mL hyaluronidase (H3884, Sigma-Aldrich) until (Invitrogen). The amplification conditions for the cumulus cells were removed, and washed in PCR were 2 min at 50°C, 30 sec at 95°C, followed M2. For ZP-removal, the cumulus cell-free oocytes by 39 cycles of 95°C for 15 sec and 60°C for 1 min were moved to a droplet of acidified Tyrode's (+ plate read) (RT-qPCR Fig 1C) or 1 min at 94°C, solution (T1788, Sigma-Aldrich) for a few followed by 30 cycles of 94°C for 30 sec, 65°C for seconds, then washed with M2 and finally 30 sec, and 72°C for 30 sec, with a final 7 min transferred into CARD MEDIUM. Afterwards, the extension at 72°C (RT-PCR Fig S2D), using preincubated sperm was added to the differently primers targeting Spaca4 (Spaca4_qPCR_F2 and treated oocytes for fertilization. Sperm and eggs Spaca4_qPCR_R2) and Hprt (HPRT_qPCR_F and were incubated at 37°C and 5% CO2 and washed 3 H P RT _ q P C R _ R ) ( F i g 1 C ) o r Spaca4 hours after incubation. Fertilization rates were (Spaca4_qPCR_F1 and Spaca4_qPCR_R1) and recorded by counting the number of two-cell stage Gapdh (Gapdh_RT_F and Gapdh_RT_R) (S2D embryos on the next day. Fig). Sperm number and motility analyses In vivo fertility assays The number (Fig S4D) and overall motility Fertility assays in mice were performed (motile, progressive motile; Fig S4E) of sperm according to two alternative methods: In the case were measured using the computer-assisted sperm of CRISPR-generated C57BL/6J-Spaca4 alleles, analysis system CEROS II animal (Hamilton C57BL/6J-Spaca4 wild-type, heterozygous, Thorne) according to the manufacturer’s protocol. transheterozygous (Spaca4117del/77del) or In brief, 1µl of sperm was diluted 1:200 in DPBS homozygous (Spaca4117del/117del or Spaca477del/77del) (MR-006C, Merck) and then overall motility mutant male or female mice were caged with 2-4- assessed on a CEROS II. month-old B6129F1 wild-type mice in the evening. To quantify sperm motility under IVF Females were checked for plugs every morning conditions (Fig S6), cauda epididymal and separated from the males as soon as a plug spermatozoa were squeezed out, and then dispersed could be observed. The number of pups for each in TYH (for sperm motility and IVF). After female was counted within a week of birth. In case incubation of 10 and 120 minutes in TYH, sperm of the mutant male mice, this procedure was motility patterns were examined using the CEROS repeated at least once before the mutant mice were II sperm analysis system (Goodson et al., 2011; kept caged with a B6129F1 female for 3-10 weeks Noda et al., 2019). after the initial plug. In the case of C57BL/6N-Spaca4tm1Osb Assessment of sperm binding to the zona mutants, sexually mature wild-type, heterozygous pellucida or homozygous mutant male mice were caged with Sperm ZP-binding assay was performed as 2-month-old B6D2F1 for several months, and the described previously (Yamaguchi et al., 2006). number of pups in each cage was counted within a Briefly, 30 min after mixing with 2-hour-incubated week of birth. Average litter sizes are presented as spermatozoa, cumulus-free eggs were fixed with the number of total pups born divided by the 0.25% glutaraldehyde. The bound spermatozoa number of litters for each genotype. were observed with an Olympus IX73 microscope.

In vitro fertilization assays Immunostaining of mouse sperm Before in vitro fertilization, female mice were Immunostaining of mouse sperm was superovulated by injection of CARD HyperOva performed as described previously (Yamaguchi et (KYD-010-EX, Cosmo Bio Co) approx. 63 hours al., 2008). Briefly, all samples were mounted on before and hCG (Chorulon) 14-16 hours before glass slides and dried. After washing with PBS, harvesting the oocytes. In vitro fertilization was slides were blocked with 10% Normal Goat Serum/ performed using CARD MEDIUM (KYD-003-EX, PBS (B6J) or 10% Newborn Calf Serum (NBCS)/ C o s m o B i o C o ) a n d C A R D F E RT I U P PBS (BDF1) for 1 hour and incubated with Preincubation Medium (KYD-002-EX, Cosmo Bio primary antibodies (rat monoclonal anti-mouse Co) according to the manufacturer’s protocol. SPACA4 (KS139-281) (Fujihara et al., 2013b); rat Sperm was prepared from the cauda epididymides monoclonal anti-mouse IZUMO1 (KS64-125) and capacitated in CARD FERTIUP medium for 1 (Fujihara et al., 2013b; Ikawa et al., 2011); rabbit

15 bioRxiv preprint doi: https://doi.org/10.1101/2021.05.02.442348; this version posted May 2, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

anti-mouse IZUMO1 (Inoue et al., 2005)) in Protein sequence alignments and percent identity blocking buffer at 4°C overnight. After washing matrices were generated with clustal omega with PBS-T (B6J) or 10% NBCS/PBS-T (BDF1), (https://www.ebi.ac.uk/Tools/msa/clustalo/) using the slides were incubated with secondary default parameters and plotted with JalView antibodies (goat anti-rat IgG Alexa Fluor 488 (https://www.jalview.org/). (A11006); goat anti-rat IgG Alexa Fluor 546 (A11081); goat anti-rabbit IgG Alexa Fluor 488 Bioinformatic analyses of Ly6/uPAR genes (A11008) (all purchased from Thermo Fisher Our analysis of the mouse Ly6/uPAR family Scientific)) in blocking buffer for 1 hour. After considered 66 mouse genes annotated in the washing with PBS containing 0.05% Tween-20, the Ensembl release 79 of GRCm38. The Ensembl slides were observed under a fluorescence release was chosen to match the mouse expression microscope (B6J: Axio Imager.Z2, Zeiss; BDF1: data (Li et al., 2017). The set of genes was selected IX70, Olympus). The mouse strain Tg(Acr- based on sequence similarity to the mouse Ly6/ EGFP)1Osb was used to identify spermatozoa with uPAR family published in (Loughner et al., 2016) unreacted acrosomes (Nakanishi et al., 1999). and domain similarity searches with Ly6/uPAR domain HMMs against the Ensembl proteome Western blotting (Herberg et al., 2018). For defining the set of Immunoblot analysis from mouse sperm was human orthologs of the mouse Ly6/uPAR genes, performed as described previously (Yamaguchi et mouse gene symbols were used to query against al., 2006). Briefly, sperm samples were collected Diopt v8.0 (Hu et al., 2011), and hits were only from the cauda epididymis and vas deferens. These considered with Best.Score, Best.Score.Reverse samples were homogenized in lysis buffer and Rank "high". Mouse expression data were containing 1% Triton X-100 and 1% protease obtained from (Li et al., 2017). Human expression inhibitors (Nacalai Tesque), centrifuged, and the d a t a w e r e o b t a i n e d f r o m G T E x v 8 supernatants were collected. Protein lysates were (www.gtexportal.org; GTEx_Analysis_ resolved by SDS/PAGE under reducing condition 2017-06-05_v8_RNASeQCv1.1.9_gene_media and transferred to PVDF membranes (Merck n_ Millipore). After blocking, blots were incubated tpm). Xenopus laevis expression data were derived with primary antibodies (rat monoclonal anti- from the NCBI GEO entry GSE73419 (Session et mouse SPACA4 (KS139-281) (Fujihara et al., al., 2016). Danio rerio expression data were 2013b); goat polyclonal anti-BASIGIN sc-9757 derived from GEO entries GSE111882 (testis, (Santa Cruz Biotechnology)) overnight at 4°C, and ovary, mature oocytes) (Herberg et al., 2018), then incubated with secondary antibodies GSE147112 (oogenesis, mature oocytes) (Cabrera- conjugated with horseradish-peroxidase (HRP) Quio et al., 2021) and GSE171906 (adult tissues) ( HRP-conjugated goat anti-rat (Noda et al., 2021). Heatmaps were used to immunoglobulins (IgGs) (112-035-167) and illustrate the relative expression levels of genes across tissues. To this end, published library size HRP-conjugated goat anti-mouse IgGs and length corrected expression values (TPM/ (115-036-062) (Jackson ImmunoResearch FPKM) were obtained; after square-root-based Laboratories)). The detection was performed variance stabilizing transformation the obtained using an ECL plus Western blotting detection kit values were centered and scaled per gene (z-score). (GE Healthcare). Statistical analysis Protein sequence alignments of fertility factors Statistical analysis was performed with the Protein sequences of human CD9 (P21926), GraphPad Prism 7 software. Statistical tests are IZUMO1 (Q8IYV9), SPACA6 (W5XKT8), JUNO/ detailed in each Figure legend. Differences were IZUMO1R (A6ND01), TMEM95 (Q3KNT9), considered significant at p < 0.05 (*) (p < 0.01 FIMP (Q96LL3) and mouse LLCFC/SOF1 (**); p < 0.001 (***); n.s., not significant). Error (Q9D9P8) were downloaded from (https:// bars represent standard deviation. Figure legends www.uniprot.org/), and searched via NCBI protein indicate the number of n values for each analysis. BLAST blastp (https://blast.ncbi.nlm.nih.gov/ Blast.cgi?PAGE=Proteins) for homologous protein sequences in Mus musculus (taxid:10090) or Homo sapiens (taxid:9606), reptiles (taxid:8459), Data availability statement Xenopus (taxid:8353), fugu (taxid:31032), Danio All data are available in the manuscript or the rerio (taxid:7955), platypus (taxid:9258) and supplementary material and are available from the armadillo (taxid:9359). If no homologous protein corresponding authors upon request. sequence was found, the search was extended to translated nucleotide sequences using tblastn.

16 bioRxiv preprint doi: https://doi.org/10.1101/2021.05.02.442348; this version posted May 2, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

List of primers

name of oligo sequence comments Spaca4_qPCR_F1 AGGACTGCGTCTTCTGTGAGC qRT-PCR #1 primer for spaca4 Spaca4_qPCR_R1 GCCCTTATTGCAAAGGTGGCC A SPACA4_qPCR_F2 TGGTGTTGGTTCTTTGCCCA qRT-PCR #2 primer for spaca4 SPACA4_qPCR_R2 GTAGCTGATGGGTTCCTCGC Gapdh_RT_F AGTGGAGATTGTTGCCATCAA RT-PCR primer for gapdh CGAC Gapdh_RT_R GGGAGTTGCTGTTGAAGTCGC AGGA HPRT_qPCR_F GAACCAGGTTATGACCTAGAT qRT-PCR primer for HRPT TTGTT HPRT_qPCR_R CAAGTCTTTCAGTCCTGTCCA TAAT SPACA4_gRNA1 taatacgactcactataGGTGACGAAGA T7 promoter sequence - gRNA TTGCTTCACgttttagagctagaaatagca TARGETING SEQUENCE – ag tracerOligo annealing sequence SPACA4_gRNA2 taatacgactcactataGGTAGCTGATG T7 promoter sequence - gRNA GGTTCCTCGgttttagagctagaaatagc TARGETING SEQUENCE – aag tracerOligo annealing sequence tracer_oligo AAAAGCACCGACTCGGTGCCA Common oligo used to generate CTTTTTCAAGTTGATAACGGA guideRNAs by annealing and in CTAGCCTTATTTTAACTTGCTA vitro transcription (Gagnon et TTTCTAGCTCTAAAAC al., 2014) SPACA4_gt_F CACTACCAGCAGAACACACCT SPACA4 genotyping

SPACA4_gt_R AGCTCACTGTCTCTGACCGC

Spaca4_targeting-s_F CTCGAGACGCACATCTTTCCA Spaca4 short arm targeting CATTGACG vector Spaca4_targeting-s_R GCGGCCGCTAGATGCAGCTGA AGCTCACT Spaca4_targeting-l_F CAATTGATCCTTCCGCCATGT Spaca4 long arm targeting GGTTTC vector Spaca4_targeting-l_R TTAATTAAGCTCTTCAGTCCTC GGGTTG Spaca4_screening+gt#78 GCTTGCCGAATATCATGGTGG Screening PCR for ES cell 1 AAAATGGCC targeting (short arm); #781 is Spaca4_screening#5081 AGGCTGTACACGTGCTCCTCT also used for genotyping the GGT neomycin-containing Spaca4 knockout mouse Spaca4_screening#5173 TGGTGCTTGCCTCTCAAACCA Screening PCR for ES cell GCTAGACTC targeting (long arm) Spaca4_screening#678 TCTGTTGTGCCCAGTCATAGC CGAATAGCC Spaca4_gt#5269 GCATGTCACGAACTCTCAGGT Genotyping PCR for neomycin- GACACCAA containing Spaca4 knockout Spaca4_gt#5298 CTGCTTCAATACTGGCTGTGA mouse (Osaka, Japan) CCCACAAG

17 bioRxiv preprint doi: https://doi.org/10.1101/2021.05.02.442348; this version posted May 2, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

SUPPLEMENTARY FIGURES

Supplementary Figure 1 A CD9

B IZUMO1

C SPACA6

D JUNO/IZUMO1R A not WR

not H/R-K/T F

E TMEM95

F FIMP

G LLCFC1/SOF1

Fig S1. Analysis of the conservation of mammalian fertility factors across vertebrates. (A-G) Protein sequence alignments (top) and percent identity matrices (bottom) of mammalian fertility factors known to be essential for sperm-egg interaction in mice. (A) CD9, (B) IZUMO1, (C) SPACA6, (D) JUNO/IZUMO1R, (E) TMEM95, (F) FIMP, (G) LLCFC1/SOF1. To D: Although folate receptor homologs (JUNO/IZUMO1R belong to the folate receptor family) are found outside of mammals, amino acids defining the JUNO/IZUMO1R (highlighted in red) are different outside of mammals (Grayson, 2015). Based on this definition, JUNO/IZUMO1R homologs are only present in mammals (species highlighted in the red box in the percent identity matrix). 18 bioRxiv preprint doi: https://doi.org/10.1101/2021.05.02.442348; this version posted May 2, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

Supplementary Figure 2

Anolis carolinensis X Pelodiscus sinensis X

Gekko japonicus X Python bivittatus X Pelodiscus sinensis X Chelonia mydas X M

Monodelphis domestica X 1 Gekko japonicus X y .

o Xenopus tropicalis X 0

Chelonia mydas X t

5

Cynoglossus semilaevis X 1 i

s . A 6

4

3 Sarcophilus harrisii X b 008309600.1

Xenopus laevis X Gallus gallus X 0 008329743.1 9 r P

a 9 010963001.1 3 P

n 1

4 P

d 0

1 Oncorhynchus mykissT X 003497485.1 005457336.1

t 010739991.1

0

etraodon nigroviridis CAF93928.1 Q i P P i 005947804.1 W68192.1 P 17969.1

P

X P P A

E 1607742.1 11

P X 533156.1

P P

P i 1

i P i

P P 008 i

015283491.1 t 007433280.1 0 999335.1 Salmo salar X t 0 00 007068022.1

006127369.1 d P 0 d P P Oryzias latipes X P P

n 015266351.1 5 P n 006127368.1 022523923.1

a P P 8 Oryzias latipes X 007068021.1 1 a Larimichthys crocea X r P r

6 P 004913413.1 421075.4 15595.1 b 008323628.1 b

4 Oryzias latipes X P 018 P

s P 4 014030639.1 s 0074973 i

i

t

6 T P 012408016.1 t P

o 013863997.1 6

etraodon nigroviridis CAF93927.1 o 016896885.1 1 y P Oncorhynchus mykiss X . 12862.1 y P 1

M 005812177.1

M 014032462.1 Homo sapiens E Canis lupus X P Camelus bactrianus X P Sus scrofa N Astyanax mexicanus X Cricetulus griseus X 001313314.1 Salmo salar X 021427362.1 Cynoglossus semilaevis X 004086091.1 P Cynoglossus semilaevis X P 1 P 0 Larimichthys crocea X 001298266.1 P 1.1 Oreochromis niloticus X P 004073817.11 Haplochromisakifugu burtoni rubripes X X Oncorhynchus mykiss X P 1479031.1 T etraodon nigroviridis CAG10812.1 004073820.1 T 005735060.1 P Salmo salar X P 010733659.1 010747866.1 Oreochromis niloticusSalmo salarX X CD59 Oncorhynchus mykiss N P Astyanax mexicanus X 019716351.1 P Cyprinus carpio KTG31434.1 P 014032463.1 Danio rerio N Pundamilia nyererei X P 021427364.1 Cynoglossus semilaevis X 003966760.1 P Oryzias latipes X P 022527661.1 Larimichthys crocea KKF25049.1P Austrofundulus limnaeus X 019729560.1 013997487.1 Xiphophorus maculatus X P P 006642765.1 021416322.1 Oreochromis niloticus N P Cyprinus carpio KTG47768.1P Pundamilia nyererei X 17894.1 003447201.1 Larimichthys crocea X 11 P 005748749.1 Hippocampus comes X P 00 Takifugu rubripes X 013979476.1 Danio rerio X y-6 P 007235220.1 L Tetraodon nigroviridis CAG01322.1P Hippocampus comes X Lepisosteus oculatus X P 706427.1 fish Salmo salar X 009296055.1 P P 018932540.1 Oncorhynchus mykiss N 003967388.1 Astyanax mexicanus X P P 001290270.1 Danio rerio X P 005937001.1 Anolis carolinensis X mammals/reptiles/ Cyprinus carpio X Takifugu rubripes X P 003458330.1 Python bivittatus X Larimichthys crocea N P 008 Gekko japonicus X 115456.1 amphibians/fish Haplochromis burtoni X Alligator mississippiensis X P 007426908.1 Oreochromis niloticus X P 004070197.1 005795873.1 P 015269869.1 Tetraodon nigroviridis CAF89761.1P Monodelphis domestica X P 019356419.1 Oryzias latipes X P 013882247.1

reptiles/amphibians Xiphophorus maculatus X Sarcophilus harrisii XP 007507575.1 005994334.1 Austrofundulus limnaeusP X P 003775 Sarcophilus harrisii X 1 18.1 coelocanth Latimeria chalumnae X

P 012409037.1 ‘sharks’, Sarcophilus harrisii X Squalus acanthias ERR1530319.17641298P 007910730.1 P 012408976.1 mammals/ Ornithorhynchus anatinus X Callorhinchus milii X P 007910729.1 P 007657128.1 Callorhinchus milii X Loxodonta africana XP 010585179.1 S Callorhinchus milii XP 007910731.1 Cricetulus griseus XP 003508724.1 P Astyanax mexicanus XP 007239783.2 ACA4 Takifugu rubripes XP 003976061.1 Mus musculus BAB24224.1 Tetraodon nigroviridis CAF93926.1 P 010639093.1 Fukomys damarensis X Austrofundulus limnaeus X P 012667956.1 Oreochromis niloticus X P 013868546.1 Otolemur garnettii X P 001171400.1 Hippocampus comes X P 019206432.1 Sus scrofa N P 006194501.1 Oryzias latipes X P 019726365.1 Camelus ferus X P 598005.1 Xiphophorus maculatusP 004073814.1 X Larimichthys crocea X Homo sapiensP N 006941098.1 852068.1 Cynoglossus semilaevis X P 014329583.1 Felis catus X P Cyprinus carpio KTG47773.1P 010733660.1 Canis lupus X fish Danio rerio X P 008329424.1 Myotis brandtiiP EPQ10600.1 005878225.1 Cyprinus carpio KTG31369.1 Oncorhynchus mykissP 002665541.1 X P 005871655.1 Salmo salar X Myotis brandtii X 005884576.1 P Spaca4lSalmo salar N Myotis brandtii X Oncorhynchus mykiss X P 014032465.1 Myotis brandtii X Astyanax mexicanus X P 021427368.1 fish P Xenopus laevis OCT73178.1P 021433692.1 Cyprinus carpio X 00 Bouncer Cyprinus carpio X 1134891.1 P 013981439.1 Xenopus tropicalis OCA29924.1 AM154129.1 Oncorhynchus mykiss X Salmo salar X P 021416321.1 Larimichthys crocea KKF32626.1P P Salmo salar X 019217266.1 Hippocampus comes X 018947454.1 007250506.1 Oncorhynchus mykiss X P 1605859.1 Oryzias latipes 1 Cynoglossus semilaevis X P P 0 018974377.1 Oryzias latipes X P 008310642.1 T 014057583.1 P etraodon nigroviridis CAG 019712504.1 Xenopus laevis X Larimichthys crocea KKF27993.1P P 021444203.1 018955736.1 Xenopus tropicalis X P mammals/reptiles/ Xenopus laevis OCT66209.1 005173770.1 Xenopus laevis OCT67908.1 P OreochromisTakifugu niloticus rubripes X X 015207687.1 Xenopus tropicalis OCA25513.1 P L Latimeria chalumnae X P P 014191263.1 ypc amphibians/fish Latimeria chalumnae X 004077925.1 019747621.1 P 1605858.1 Latimeria chalumnae X 013126046.11 P Gekko japonicus X 0 Anolis carolinensis X P P P Ophiophagus hannah ETE61453.1 018085649.1 016893477.1 Cynoglossus semilaevisCyprinusDanio carpio X rerio X X 008335986.1 Python bivittatus X Hippocampus comes X P AM153096.1 Alligator mississippiensis X Pelodiscus sinensis X Chelonia mydas X Ophiophagus hannah ETE61452.1 P 013981440.1 LY6G6e/d 1 Anolis carolinensis X 002939497.1 1914.1 Gekko japonicus X P Canis lupus X Felis catus X Camelus bactrianus X 018955710.1 Mus musculus N S P 010733656.2 1 1

u P . .

008329428.1 s

3 1

Lepisosteus oculatus X P 014329986.1

3 6 s P 005754472.1 c 19928.1 7 1 018934481.1 P 1 r 8 P Haplochromis burtoni X 3 akifugu rubripesOryzias X latipes P 014032458.1 o 0 006007251.1 0 P T P 003200124.1 258439.1 f P

a Oreochromis niloticus X 006013400.1 6 015276067.1 Q 006013542.1 P 015276065.1 P P

Salmo salar X P 008 0 X P P P 006013543.1 1 P P P P 003228198.1 003508529.1 E P 010608159.1 0 007435798.1 020953812.1 019685930.1

i P 022281559.1 i P P 0 P t P

2 007070490.1 Cyprinus carpio X P d

Cynoglossus semilaevis X X P

0 P

n 081642.1

s 006120941.1 9 015276068.1 P a i

r

5

s 008

b P 4

n P Larimichthys crocea X Salmo salar X 1 019346841.1

Danio rerio X s e 010959804.1

i

r Gallus gallus BAJ53004.1 1 t Cyprinus carpio X 1

a 3 o 19924.1

.

y

m Cynoglossus semilaevisPundamilia X nyererei X Mus musculus N 1

Sus scrofa X a Xiphophorus maculatus X M

d

Gekko japonicus X

s

y Anolis carolinensis X Latimeria chalumnae X

m Cricetulus griseus X Tree scale: 1 o

k

u

Fukomys damarensis X

F

B ypd1 C L mouse human

chr1 y y t t y y e e r r r r a a ain ain uscle uscle r r v v ypd6b ypd6 hea o testis stomach spleen lung m kidn hea b o testis stomach spleen lung m kidn b L L Cd59b Cd59a Ly6i:15_74979534 Ly6g5b:17_35113948 LY6G5B chr2 Cd177:7_24743983 CD177 Ly6d:15_74762056 LY6D Ly6g6f:17_35080480 LY6G6F x101 ypd5 ypd3 ypd10 ypd11 ypd4

e Ly6g:15_75155240 L Plaur Pinlyp L T L L Cd177 L Spaca4 Ly6e:15_74955051 LY6E Ly6c2:15_75108158 Ly6a2:15_75131377 chr7 Ly6a:15_74994877 Lypd2:15_74732247 LYPD2 v1

r Ly6c1:15_75044018 ate6 ate12 ate8 ate4 ate3 ate1 ate10 ate7 ate11 ate9 ate14 Gm27235 P P P Ac P P P P P P P P Gpihbp1:15_75596628 GPIHBP1 Slurp1:15_74724318 SLURP1 Plaur:7_24462484 PLAUR chr9 Ly6g6d:17_35071443 LY6G6D Cd59a:2_104095801 Pate7:9_35776524

ypd8 ypd9 Ly6g6g:15_74770908 2210407C18Rik L L Ly6g2:15_75216376 Ly6f:15_75268421 Gm5916:9_36119934 chr11 Ly6g5c:17_35108279 LY6G5C Pinlyp:7_24541658 PINLYP p1 p2

r r Mm Ly6g6c:17_35065388 LY6G6C y6i y6c1 y6c2 ypd2 ynx1 y6d y6g6g y6k y6m y6e y6a y6g y6g2 y6f y6l y6h 3 L L L Psca Slu L Slu L L L L Gml Gml2 L L L L L L L L Gpihbp1 Lypd5:7_24349196 LYPD5 Ly6m:15_74876987 Pate4:9_35607093 PATE4 0 chr15 Pate3:9_35645113 PATE3 Pate9:9_36532397 −3 Pate14:9_36635754

y6g6c y6g6d y6g6e y6g5c y6g5b Cd59b:2_104069849 CD59 Hs L L L L L 1810065E05Rik:11_58421111 3 Pate1:9_35685079 PATE1 chr17 Pate2:9_35570284 PATE2 0 Spaca4:7_45725107 SPACA4 Gm27235:9_35776524 Ly6k:15_74796874 LY6K −3 Gm3443 Gm3443:19_21552204 Pate6:9_35788050 Gml:15_74813452 GML chr19 Gml2:15_74819071 Lypd4:7_24864620 LYPD4 Lypd9:11_58446149 Lypd10:7_24709242 Pate5:9_35838329 Pate13:9_35908423 Lypd11:7_24715023 D Tex101:7_24668007 TEX101 Acrv1:9_36693220 ACRV1 Lynx1:15_74747852 LYNX1 Lypd1:1_125867622 LYPD1 brain lung heart liver spleen kidney testis ovary uterus Ly6h:15_75551268 LY6H Lypd3:7_24636550 LYPD3 2210407C18Rik:11_58608204 Spaca4 Psca:15_74714839 PSCA Lypd8:11_58379043 LYPD8 Slurp2:15_74742764 Lypd6:2_50066429 LYPD6 Gapdh Lypd6b:2_49787688 LYPD6B Ly6g6e:17_35076902 LY6G6E

19 bioRxiv preprint doi: https://doi.org/10.1101/2021.05.02.442348; this version posted May 2, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

Fig S2. Expression analysis of mammalian Ly6/uPAR proteins. (A) Phylogenetic tree of selected groups of Ly6/uPAR proteins across vertebrates. Maximum-likelihood phylogenetic tree based on Ly6/uPAR protein sequence alignments across vertebrates (adapted from (Herberg et al., 2018)). Branches supported by ultrafast bootstrap values (>=95%) are marked with a blue dot. SPACA4 (red) is present in mammals, reptiles and amphibians; Bouncer (yellow) is present in fish. The sperm and egg symbols next to selected species’ genes indicate expression in testis (blue) or ovaries (green). (B) Genomic localization of Ly6/uPAR genes in mice. Ly6/uPAR genes are clustered in the genome. Only containing at least one Ly6/uPAR gene are shown. Testis-expressed genes are highlighted with a blue box; Spaca4 is highlighted in red. (C) Mouse and human Ly6/uPAR proteins are expressed in various tissues, including in gametes (extension of Fig 1B including human orthologs). Heatmaps of expression levels of homologous mouse (Mm; left) and human (Hs; right) Ly6/uPAR proteins (homologous proteins are shown in one row) across various adult tissues. Heatmaps are color-coded based on z-scores of the normalized gene expression values (average of the square- root) of RNA-Seq data from murine and human adult tissues (Li et al., 2017) (www.gtexportal.org). Adult tissues are indicated at the bottom. Gene names are given on the left (mouse) and right (human) of the corresponding heat-maps. Numbers behind gene names indicate chromosome locations in mice or humans. Spaca4 is highlighted in red. (D) Murine Spaca4 is expressed in the male but not the female germline. RT-PCR for Spaca4 from cDNA of the indicated mouse tissues relative to the housekeeping gene Gapdh. The template cDNA was generated from a mixture of the tissue from 3-4 different mice.

20 bioRxiv preprint doi: https://doi.org/10.1101/2021.05.02.442348; this version posted May 2, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

Supplementary Figure 3 A

Spaca4

B

Izumo1

C

Spaca6

D

Tmem95

Fig S3. Analysis of the expression of selected mammalian fertility factors during murine spermatogenesis. (A-D) Expression values for selected fertility factors during murine spermatogenesis were derived from published single-cell RNA-Seq data (Ernst et al., 2019). The web-browser version https://marionilab.cruk.cam.ac.uk/ SpermatoShiny/ was used to generate the plots for (A) Spaca4, (B) Izumo1, (C) Spaca6, (D) Tmem95. The Spaca4 temporal expression pattern resembles Izumo1’s pattern.

21 bioRxiv preprint doi: https://doi.org/10.1101/2021.05.02.442348; this version posted May 2, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

Supplementary Figure 4 A MVLGWPLLLVLVLCPGVTGIKDCVFCELTDSARCPGTHMRCG Spaca4 SP SPACA4 TM DDEDCFTGHGVAQGVGPIINKGCVHSTSCGREEPISYMGLTY SLTTTCCSGHLCNKGTGLSTGATSLSLGLQLLLGLLLLLQYWL*

MVLGWPLLLVLVLCPGVTGIKDCVFCELTDSARCPGTHMRCG Spaca4 117del SP TM ------LTY SLTTTCCSGHLCNKGTGLSTGATSLSLGLQLLLGLLLLLQYWL*

77del MVLGWPLLLVLVLCPGVTGIKDCVFCELTDSARCPGTHMRCG Spaca4 SP DDEDCLGTHQLHGPHIQSHHHLLFWPPLQ*

n.s. B 14 ** n.s. *** ** 12 10 8 6

litter size 4 2 0 male female genotype of male +/+ +/- 117/77 77/77 117/117 117/77 # of litters n=24 n=16 n=13 n=6 n=11 n=5 # of males m=6 m=6 m=4 m=3 m=4 m=3

C +/+ Spaca4 77/77del +/+ w/o primary Ab

5 μm DAPI SPACA4 DIC DAPI SPACA4 DIC DAPI SPACA4 DIC D E n.s. n.s. motile n.s. n.s. 100 progressive 1200

1000 80 m 800 r 60 spe f o

m number 600 r %

pe 40 s 400

200 20

0 0 genotype of male +/+ +/- 117/77 genotype of male +/+ +/- 117/77 # of males m=3 m=3 m=6 # of males m=2 m=2 m=5

22 bioRxiv preprint doi: https://doi.org/10.1101/2021.05.02.442348; this version posted May 2, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

Fig S4. Generation and phenotypic analysis of Spaca4 knockout mice. (A) Overview of the C57BL/6J-Spaca4 knockout alleles generated by CRISPR/Cas9-mediated targeted mutagenesis. (Left) Schematics of the wild-type and knockout alleles. Yellow dashed lines indicate the site of the deletions. Predicted disulfide bridges are indicated in orange. SP, signal peptide; TM, transmembrane region. (Right) Amino acid sequences of the protein products that are predicted to be produced from the different alleles. One mutant allele contains a 117-nt in-frame deletion after amino acid 42. The other allele contains a 77-nt out- of-frame deletion after amino acid 47. Cysteines predicted to form disulfide bridges are indicated in orange; the SP sequence is shown in grey; the out-of-frame additional protein sequence in the 77del allele is shown in red. (B) Spaca4 knockout male mice are sub-fertile. Litter sizes of C57BL/6J-Spaca4 wild-type (+/+), heterozygous (+/-), transheterozygous (117/77) and homozygous (77/77 or 117/117) males of the indicated allele combinations caged with B6129F1 wild-type females, or B6129F1 wild-type males caged with transheterozygous (-/-) females. Successful mating was confirmed by plug checks. Data are means ± SD. ***p < 0.0001, **p < 0.001 (Kruskal- Wallis test with Dunn multiple-comparisons test); n.s., not significant. n = number of litters; m = number of male mice tested. (C) SPACA4 protein is absent in morphologically normal sperm of Spaca4 knockout mice. Immunostaining of sperm detects SPACA4 protein (magenta) under permeabilizing conditions in the head of sperm derived from wild-type (+/+) but not Spaca4 knockout (77/77del) mice. DAPI (cyan) staining labels the sperm nucleus. A control immunostaining in which the primary antibody was omitted is shown on the right. DIC, differential interference contrast image. Scale bar, 25 µm. (D) Sperm number is normal in Spaca4 knockout males. Sperm number of Spaca4 wild-type (+/+), heterozygous (+/-) and transheterozygous (-/-) males. Data are means ± SD. n.s., not significant (one-way ANOVA with Dunnett’s multiple-comparisons test). m = number of male mice tested. (E) Overall sperm motility is not affected in Spaca4 knockout males. Sperm motility of Spaca4 wild-type (+/ +), heterozygous (+/-) and transheterozygous (-/-) males. Shown is the percentage of sperm that was motile (open circles, comparison +/+ versus -/-: p = 0.65) and progressively motile (closed circles, comparison +/+ versus -/-: p = 0.41). Data are means ± SD. n.s., not significant (One-way ANOVA with Dunnett’s multiple-comparisons test). m = number of male mice tested.

23 bioRxiv preprint doi: https://doi.org/10.1101/2021.05.02.442348; this version posted May 2, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

Supplementary Figure 5 A Fam83e Spaca4 5269 wild-type allele

X 5298 X neo neo

targeting vector r

1.9 kb 5.1 kb

781 neo neo

knockout allele r Spaca4 tm1Osb 5081 678

B C

+/+ +/- -/- +/+ -/- knockout allele Spaca4 Spaca4 tm1Osb wild-type allele Gapdh

D E n.s. ** 15 100 n.s. * 80 10 60

litter size 40 5 20 % 2-cell stage embryos 0 0 genotype of male +/+ +/- -/- genotype of male +/+ +/- -/- (Spaca4 tm1Osb) (Spaca4 tm1Osb) # of litters n=11 n=11 n=8 # of males m=7 m=3 m=7 # of males m=2 m=2 m=2

Fig S5. Male Spaca4 knockout mice are sub-fertile (A) Targeted disruption of the murine Spaca4 gene to generate Spaca4tm1Osb. To disrupt the Spaca4 gene, the single exon was replaced with a neomycin resistance cassette (neo), and a thymidine kinase cassette (tk) was used for negative selection. Labeled arrows indicate primer binding sites. (B) Genotyping of Spaca4tm1Osb knockout mice. Both the wild-type allele (a 0.3-kb band) and the knockout allele (a 0.5-kb band) were amplified by PCR, using primers #5269 and #5298 for the wild-type allele and primers #781 and #5269 for the knockout allele. (C) RT-PCR analysis of testis in wild-type and Spaca4tm1Osb knockout mice. The Spaca4-specific 236-nt band was amplified from wild-type but not from Spaca4tm1Osb knockout testis. Gapdh was used as a control. (D-E) Spaca4tm1Osb knockout mice are sub-fertile. (D) Spaca4tm1Osb knockout males (-/-) copulated normally but were sub-fertile compared to wild-type (+/+) and heterozygous Spaca4tm1Osb (+/-) mutant males (comparison +/+ versus -/-: p = 0.0333, unpaired t-test). Data are means ± SD. n.s., not significant. n = number of litters; m = number of male mice tested. (E) In vitro fertilization of oocytes from wild-type females using sperm from wild- type (+/+), heterozygous Spaca4tm1Osb (+/-) or homozygous Spaca4tm1Osb (-/-) males. Plotted is the percentage of 2-cell stage embryos as a measure of successful fertilization. Data are means ± SD. p = 0.003 (Kruskal-Wallis test with Dunn multiple-comparisons test); n.s., not significant. m = number of males tested.

24 bioRxiv preprint doi: https://doi.org/10.1101/2021.05.02.442348; this version posted May 2, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

Supplementary Figure 6

300 +/+ (B6J)

) -/- (B6J Spaca4 117del/77del) 200 +/+ (BDF1) m/sec tm1Osb ( -/- (BDF1 Spaca4 )

100 velocity

0

10 min 10 min 10 min P 120 min L 120 min L 120 min A P L L V A VC VS V VC VS

Fig S6. Spaca4 knockout sperm has reduced motility under IVF conditions Sperm motility was assessed under IVF conditions using the computer-assisted sperm analysis system CEROS II. Sperm motility of transheterozygous B6J (117del/77del; -/-) males was compared to wild-type B6J (+/+) males; similarly, sperm motility of BDF1 Spaca4tm1Osb knockout (-/-) males was compared to wild- type BDF1 (+/+) males to control for possible differences in the genetic backgrounds of both mutants. There are significant differences of all three parameters (VAP, VCL, and VSL) between wild-type and Spaca4 mutant (transheterozygous B6J and BDF1 knockout) spermatozoa after 10 min incubation (p < 0.05). After 120 min incubation, however, there are no significant differences between B6J backgrounds although there are significant differences between BDF1 backgrounds. Plotted is the velocity in µm/sec. VAP: average path velocity, VSL: straight line velocity, VCL: curvilinear velocity. Data are means ± SD. n.s., not significant (Student’s t test). m = 3 male mice tested for each genotype.

25