Bulls have a greater impact on the genetic gain of a herd than any single cow because of the large number of cows artificially inseminated (AI) or serviced by an individual bull compared with a single offspring generated by a cow each year. Therefore, it is important to study the impact of male fertility on cattle operations. Thus, the identification of seminal traits that can be predictors of a bull’s fertility is of great importance. After differentiation, sperm lose the ability to grow, divide, repair, and synthesize proteins (Hammerstedt, 1993). Although the metabolic function of sperm may be limited, it is heavily regulated by proteins (Talluri et al., 2022) and its environment (Zoca et al., 2022a). After spermiation, sperm travel through the testis tubules into the epididymis (where further maturation occurs) and are stored in the epididymis tail in a quiescent state until ejaculation (Acott and Carr, 1984; Carr and Acott, 1984; Barth and Oko, 1989). On ejaculation, epididymal sperm are diluted with seminal plasma from accessory sex glands and motility is initiated (Acott and Carr, 1984; Carr and Acott, 1984). Following natural service (Hunter and Wilmut, 1984; Wilmut and Hunter, 1984; Lefebvre et al., 1995) or AI (Mitchell et al., 1985. Suarez et al., 1997), sperm with fertilizing ability reach the oviduct approximately 6–12 h after insemination, populate the isthmus portion of the oviduct, and form the sperm reservoir. Sperm that bind to oviductal cells in vitro have prolonged motility and fertilization ability (≈ 30 h) compared with sperm free in the media (Pollard et al., 1991).

Cell-to-cell interactions (i.e., sperm to oviduct and sperm to oocyte) are mediated through proteins; therefore, these interactions are important for successful fertilization. The sperm’s apical surface binds to the oviductal isthmus and ampullary ciliated cells (Pollard et al., 1991; Lefebvre et al., 1995) and binder of sperm proteins (BSP) has been reported to be involved in sperm reservoir formation (Ignotz et al., 2001; Gwathmey et al., 2003; Gwathmey et al., 2006). There are few proteins known to be required for fertilization, and these include CD9 (Kaji et al., 2000; Le Naour et al., 2000; Miyado et al., 2000) and JUNO (Bianchi et al., 2014) on the egg, and IZUMO1 in the sperm (Inoue et al., 2005). Other proteins have been identified as being associated with mammalian fertility but are not required (reviewed by Sutovsky, 2009).

Both DAG1 and SERPINA5 were present and loosely attached to the ejaculated sperm of bulls, but they were not present on the epididymal sperm, although SERPINA5 was present in both epididymal fluid and seminal plasma (increased abundance in seminal plasma) and DAG1 was present in seminal plasma (Zoca et al., 2022a; Zoca et al., 2022b). Furthermore, the function of both proteins is associated with cell-to-cell interactions. More specifically, the DAG1 gene encodes the dystroglycan precursor that generates two proteins, alpha- and beta-dystroglycan, through posttranslational modification. Alpha-dystroglycan is an extracellular/surface protein and beta-dystroglycan is a transmembrane protein (Ibraghimov-Beskrovnaya et al., 1992). In humans, the presence of DAG1 has been reported in seminal plasma but not on sperm (Jodar et al., 2016). Beta-dystroglycan has been reported to be localized to the tail middle piece of guinea-pig sperm (Hernández-González et al., 2001) and the post-acrosomal region and middle piece of mouse sperm (Hernández-González et al., 2005). The gene SERPINA5 encodes the plasma serine protease inhibitor. This protein is also known as serpin family A member 5, protein C inhibitor, and others. The presence of the SERPINA5 protein has been reported in many body fluids, including plasma (blood), seminal plasma, follicular fluid, amniotic fluid, and milk (Laurell et al., 1992). In knockout mice for SERPINA5, females were fertile and males were infertile in both in vitro (0.5% pregnancy) and in vivo (0% pregnancy) experiments. Also, sperm motility (12.5% motility) and the percentage of morphologically normal sperm (5% normal morphology) were decreased in knockout mice (Uhrin et al., 2000). Similarly, SERPINA5 concentrations were decreased in normozoospermic infertile men compared with normozoospermic fertile men (Panner Selvam et al., 2019). Nevertheless, in men, SERPINA5 has been localized to the sperm head (Zheng et al., 1994; Elisen et al., 1998). In bovines, the localization of SERPINA5 and DAG1 is still uncertain, and the function of DAG1 on sperm is not well understood. Furthermore, the impact of the abundance of these proteins on bull fertility has not been evaluated. It was hypothesized that both proteins would be associated with bull fertility and that bulls with greater concentrations would have both greater field and in vitro fertility. Thus, the first objective of this study was to characterize DAG1 and SERPINA5 immunolocalization on bovine sperm and their potential as fertility markers by evaluating variability within and among bulls. The second objective was to investigate the relationship of DAG1 and SERPINA5 with field fertility [i.e., sire conception rate (SCR)], in vitro fertility [in vitro embryo production (IVP)], and sperm parameters.

Immunolocalization demonstrated that the SERPINA5 protein was present on both the sperm head ( Figure 1 ) and the tail ( Figure 1 ). On the sperm head, the most consistent pattern of SERPINA5 covered the proximal region over the acrosomal cap ( Figure 1 ). There were 34.2% ± 12.7% [mean ± standard deviation (SD)] of sperm tails that were positive for SERPINA5 (ranging from 6.0% to 57.4%; Figure 2 ). The relative concentration of SERPINA5 on the sperm head ranged from 38.9 ± 1.1 au (arbitrary unit) to 68.4 ± 1.1 au ( Figure 3 ) with an average of 53.2 ± 6.6 au (mean ± SD). The abundance of SERPINA5 had a greater variance among bulls than within bulls (P < 0.0001; variance 325.8 vs. 285.4, respectively). Furthermore, immunolocalization determined that DAG1 was present on the sperm head in the proximal apical region, over the acrosomal cap ( Figure 4 ). The percentage of sperm with DAG1 present on the sperm head did not differ between treatments (P = 0.74; 32.1% ± 4.1%, 38.1% ± 3.5%, 31.5% ± 4.7%, 34.6% ± 3.4% for high-SCR/good, high-SCR/poor, low-SCR/good, and low-SCR/poor, respectively). Abundance (fluorescence insensitivity) ranged from 29.6 ± 0.5 to 45.7 ± 0.5 au ( Figure 5 ) and averaged 36.0 ± 4.6 au (mean ± SD). There was greater variation in the abundance of DAG1 among bulls than within bulls (P < 0.0001; variance was 69.4 vs. 49.1, respectively).

There were positive correlations between TMOT and PROG (P < 0.01; Table 2 ), PMI and percentage of sperm tail labeled for SERPINA5 (P = 0.05; Table 2 ); all other correlations were not significant (P < 0.12; Table 2 ). There was no SCR classification by IVP classification interaction (P < 0.12) for SCR, PROG, PMI, CL, BL, BL/CL, SERPINA5 and DAG1 concentrations, and the percentage of sperm tail labeled for SERPINA5 ( Table 3 ); nevertheless, the interaction was significant (P = 0.02) for TMOT. Bulls classified as high-SCR and good IVP had greater TMOT than bulls classified as high-SCR and poor IVP and tended (P = 0.08) to have greater TMOT than low-SCR and good IVP; however, they were not different than low-SCR and poor IVP. In addition, low-SCR and poor IVP tended (P = 0.07) to have greater TMOT than high-SCR and poor IVP ( Table 3 ).

As designed, when evaluating the SCR classification, high-SCR bulls had greater SCRs values than low-SCR bulls (P < 0.0001; Table 4 ). Interestingly, low-SCR bulls had greater (P = 0.03) BL production and tended (P = 0.10) to have greater BL/CL ratios than high-SCR bulls ( Table 4 ). No other parameters evaluated were associated with SCR classification (P < 0.15; Table 4 ). Similarly, as designed, bulls classified as having good IVP had greater (P < 0.0001) BL/CL ratios than poor IVP bulls ( Table 5 ). Also, good IVP bulls had greater BL production (P < 0.0001) than poor IVP bulls, but there was no difference (P = 0.96) in CL. Interestingly, good IVP bulls tended (P = 0.07) to have greater SCRs than poor IVP bulls ( Table 5 ). No other parameters evaluated were associated with IVP classification (P < 0.34; Table 5 ).

The rate of genetic improvement in a herd is far more efficient through bull selection than female selection owing to the larger number of offspring generated by one single bull compared with one single female. This is especially true in dairy cattle, in which 90% of females are artificially inseminated (Starbuck et al., 2004; Valour et al., 2015; García-Ruiz et al., 2016; Wiggans et al., 2017; USDA, 2018). Bull fertility, as evaluated in the AI industry, has been evaluated heavily through semen quality, which relies predominantly on sperm motility and morphology, and more recently sperm PMI (Barth and Oko, 1989; Koziol and Armstrong, 2018; DeJarnette et al., 2021). Sire conception rate is one of the most common methods of evaluation for bull field fertility. The SCR value given to a bull is generated based on field reports of pregnancies: SCR values represent the bull’s deviation in pregnancy rates at day 70 of gestation compared with the mean pregnancy rates from all other bulls that could have been used (Kuhn et al., 2006; Norman et al., 2011). It was observed that some low-SCR bulls had good BL production; interestingly, overall, low-SCR bulls had greater BL production than high-SCR bulls and tended to have greater BL/CL ratios. Furthermore, in this group of bulls, animals that were classified as good had slightly greater SCRs than those ranked as poor. Ortega et al. (2018) reported similar findings in which one (out of three) low-SCR bull had BL production rates similar to high-SCR bulls; however, overall, in their group of bulls, high-SCR bulls had greater BL production than low-SCR bulls. Sperm must endure far fewer challenges to fertilize an embryo in vitro than in vivo. In vivo, sperm must navigate the female reproductive tract, survive uterine contraction, overcome the utero–tubular junction, form the sperm reservoir, capacitate, and “find” the ovulated oocyte to then fertilize that single oocyte (Suarez, 2015; Suarez, 2016). In addition, AI may happen in different moments during estrus/pro-estrus resulting in the requirement for sperm to survive for prolonged periods of time or capacitate “quicker”, both affecting fertilization rate and embryo quality (Saacke et al., 2000; Dalton et al., 2001; Richardson et al., 2017). In contrast, in vitro, sperm must tolerate manipulation insults (Baldi et al., 2020); however, barriers in the female reproductive tract (except those from the oocyte) are eliminated. Thus, it is possible that bulls with low-SCRs but with good BL production rates have sperm transport problems or are more susceptible to the timing of insemination (i.e., sperm longevity) or the uterine/oviduct environment than low-SCR bulls with lower BL production rates, where the problem may be related to fertilization itself rather than sperm transport. This hypothesis is partially explained by the “compensable” and “uncompensable” characteristics of sperm previously reported (Saacke et al., 1994; Saacke, 2008; Amann et al., 2018). It was also observed that some high-SCR bulls had poor IVP, and a few hypotheses may be drawn from these observations: (1) these bulls could be more sensitive to laboratory manipulation procedures for IVP; (2) the female reproductive tract in vivo may be acting as a “filter” and some of the sperm with uncompensable characteristics may not reach the site of fertilization, whereas in vitro they are not being removed; (3) the ejaculates from these bulls used for IVP were not accurate representations of their field fertility (i.e., high variability between ejaculates); or (4) these bulls (with SCRs of 1.1 and 2.8) were misclassified as high-SCR.

The objective of the bovine AI industry is to provide semen of high quality to cattle producers; semen that passes quality control and is commercially available has met specific thresholds (Harstine et al., 2018; DeJarnette et al., 2021). With that, sperm motility, morphology, and PMI of commercially available semen are not expected to correlate with field fertility, especially in large samples (DeJarnette et al., 2021). The present results corroborated the industry efforts and did not identify any relationship or correlation between field fertility (SCR) or IVP classification and TMOT, PROG, or PMI. Nevertheless, an SCR by IVP classification interaction on TMOT was observed in the present study; however, differences in motility did not conclusively explain fertility differences, especially because high-SCR/good had similar motility to low-SCR/poor.

Sperm interact with the utero–tubular junction (UTJ), oviduct (formation of sperm reservoir and movement through the oviduct), and oocyte through proteins (Lefebvre et al., 1995; Gwathmey et al., 2003; Gwathmey et al., 2006; Ignotz et al., 2007; Sutovsky, 2009; Suarez, 2015; Suarez, 2016). The formation of the sperm reservoir in bovines involves BSP (Ignotz et al., 2001; Gwathmey et al., 2003; Gwathmey et al., 2006). These groups of proteins are attached to the sperm during ejaculation when epididymal sperm come into contact with seminal plasma (Desnoyers and Manjunath, 1992; Müller et al., 1998; Nauc and Manjunath, 2000); similarly, as observed by Zoca et al (2022a; 2022b), DAG1 and SERPINA5 are attached to the sperm during ejaculation, as DAG1 and SERPINA5 were not detected in epididymal sperm samples. Liquid chromatography with tandem mass spectrometry analysis (LCMS/MS) results demonstrated that DAG1 was not very abundant (spectra count 1.1), whereas SERPINA5 was highly abundant (spectra count 37.3) on the sperm (Zoca et al., 2022b). The present results ( Figures 1 , 3 , 4 , 5 ) agree with LCMS/MS findings that SERPINA5 is present in greater abundance on the sperm than DAG1; interestingly, DAG1 and SERPINA5 concentrations were correlated ( Table 2 ). The localization of both DAG1 and SERPINA5 on the sperm head is interesting and consistent with the region of the sperm that attaches to oviductal epithelial cells to form the sperm reservoir (Lefebvre et al., 1995). The function of DAG1 on the sperm is not fully understood, especially alpha-dystroglycan, which is more likely than beta-dystroglycan to have been measured owing to the fact that alpha-dystroglycan is an extracellular/surface protein (Ibraghimov-Beskrovnaya et al., 1992). Beta-dystroglycan, a transmembrane protein (Ibraghimov-Beskrovnaya et al., 1992), was reported on the tail middle piece of guinea-pig sperm (Hernández-González et al., 2001) and the middle piece and acrosomal region of mice sperm (Hernández-González et al., 2005). Hernández-González et al. (2005) demonstrated that mice sperm with beta-dystroglycan deficiencies had increased morphological abnormalities in the sperm tail, and the number of sperm capable of fertilization was lower (≈ 50% less) than in sperm from wild-type mice. In the present study, DAG1 was not associated with field fertility or field and in vitro embryo fertility. Furthermore, DAG1 was not correlated with SCR, CL, or BL. Thus, it may be hypothesized that DAG1 may function to stabilize the acrosomal region as a decapacitating factor, preventing premature acrosomal reaction or formation of the sperm reservoir owing to its localization on the sperm ( Figure 4 ); however, this study failed to identify any relationship between DAG1 concentration and bull fertility, either in vivo or in vitro.

The abundance of SERPINA5 in the seminal plasma and loosely attached to the sperm ranked 13th and 11th based on spectra count, respectively (Zoca et al., 2022b), which agrees with previous reports for SERPINA5 in seminal plasma (Druart et al., 2013; Kasimanickam et al., 2019). Reference for immunolocalization of SERPINA5 in bovine or other livestock species could not be found, but within human sperm, SERPINA5 was localized to the acrosomal region of epididymal and ejaculated sperm (Zheng et al., 1994; Elisen et al., 1998). There was no difference in SERPINA5 localization between capacitated and non-capacitated sperm; however, when an acrosome reaction was induced, SERPINA5 was limited to the equatorial region (Zheng et al., 1994; Elisen et al., 1998). The immunolocalization of SERPINA5 on the bovine sperm head ( Figure 1 ) was similar to that of human sperm (Zheng et al., 1994; Elisen et al., 1998); however, bovine sperm also had SERPINA5 on the sperm tail, differing from human sperm. The protease inhibitory activity of SERPINA5 has been found in multiple body tissues and fluids (España et al., 1989; Ecke et al., 1992; Christensson and Lilja, 1994; Hermans et al., 1994; Zheng et al., 1994; Elisen et al., 1998). The activity of SERPINA5 can be modulated by heparin and other glycosaminoglycans (Kuhn et al., 1990; Pratt and Church, 1992; Ecke et al., 1997). Heparin and glycosaminoglycans are present in the oviduct from oviductal fluid and follicular fluid that have been shown to induce sperm capacitation (Parrish et al., 1985; Parrish et al., 1988; Mahmoud and Parrish, 1996; Bergqvist et al., 2007). When the SERPINA5 gene was disrupted in mice, male mice were infertile both in vitro and in vivo because of morphologically abnormal sperm, lower motility, and lack of sperm–egg binding (Uhrin et al., 2000). In addition, normozoospermic men with an unknown reason for infertility had a lower concentration of SERPINA5 than their fertile counterparts (Panner Selvam et al., 2019). The ability of human sperm to bind to the human zona pellucida was evaluated in the presence of different concentrations of anti-SERPINA5 or SERPINA5 in the media (Elisen et al., 1998). Interestingly, a lower concentration of anti-SERPINA5 increased the ability of sperm to bind to the zona pellucida; however, the greater the concentration of SERPINA5 in the media the lower the ability of sperm to bind to the zona pellucida (Elisen et al., 1998). Another member of the serine protease inhibitor (SERPIN) family, called glia-derived nexin or protease nexin-1 (SERPINE2), has been reported to be a decapacitating factor in mice (Lu et al., 2011). Controversially, there was no association of SERPINA5 concentration or percentage of tail labeled for SERPINA5 with bull fertility (SCR and IVP). The present results indicate that neither SERPINA5 nor DAG1 is a good indicator of in vivo or in vitro bull fertility.

In conclusion, by using immunofluorescent microscopy, DAG1 and SERPINA5 proteins were localized on the bovine sperm head, and SERPINA5 was also localized on the sperm tail. For a protein to be considered a marker of fertility, it is necessary to have animal variation. Also, any new test must not be correlated with current evaluations of semen quality or must provide a simpler method of evaluation over current analyses (DeJarnette, 2005; Harstine et al., 2018; DeJarnette et al., 2021). A greater variation among bulls than within bulls was observed for both DAG1 and SERPINA5, fulfilling the first characteristics for a potential fertility marker. Furthermore, when evaluated by immunofluorescent microscopy, DAG1 and SERPINA5 were not correlated with TMOT, PROG, or PMI, fulfilling the second characteristic of a potential fertility marker; however, the percentage of tails labeled for SERPINA5 was correlated with PMI. In addition, the relative concentrations of both proteins in the sperm were correlated with each other. Although fulfilling the requirements for a new fertility marker, the proteins were not associated with bull fertility. Thus, using immunofluorescent microscopy to determine SERPINA5 and DAG1 may not be a good marker of bull fertility; however, as SERPINA5 and DAG1 do have variation among animals (first criteria), increased sensitivity by evaluating a greater proportion of an ejaculate by immunofluorescent flow cytometry may impact its ability to predict bull fertility.

The original contributions presented in the study are included in the article. Further inquiries can be directed to the corresponding author.

SZ, RC, and GP: experimental design and conceptualization. SZ, AK, TA, JR, KE, JD, MO, and GP: data collection. SZ: data management. SM, RC, and GP: statistical analyses. JW and GP: funding acquisition. SM wrote the original draft of the manuscript. All authors contributed to the article and approved the submitted version.