Novel proteomic profiling of epididymal extracellular vesicles in the domestic cat reveals proteins related to sequential sperm maturation with differences observed between normospermic and teratospermic individuals.

Extracellular vesicles (EVs) secreted by the epididymal epithelium transfer to spermatozoa key proteins that are essential in promoting motility and subsequent fertilization success. Using the domestic cat model, the objectives were to (1) characterize and compare protein content of EVs between segments of the epididymis, and (2) compare EV protein compositions between normo- and teratospermic individuals (producing >60% of abnormal spermatozoa). Epididymal EVs from adult cats were isolated and assessed via liquid chromatography tandem mass spectrometry. Both male types shared 3,008 proteins in total, with 98 and 20 EV proteins unique to normospermic and teratospermic males, respectively. Expression levels of several proteins changed between epididymal segments in both male types. Several proteins in both groups were related to sperm motility (hexokinase 1, adenylate kinase isoenzyme) and zona pellucida or oolemma binding (disintegrin and metalloproteinase domain proteins, zona binding proteins 1 and 2). Interestingly, seven cauda-derived EV proteins trended downward in teratospermic compared to normospermic males, which may relate to poor sperm quality. Collective results revealed, for the first time, EV proteins related to sequential sperm maturation with differences observed between normospermic and teratospermic individuals.

characterized; elucidating the function of these key factors will contribute to our understanding of sperm maturation processes in mammals.
Discerning key proteins contained within EVs is critical to developing interventions for fertilization failure and early embryo loss. This is especially vital for individuals that exhibit teratospermia; a condition in which more than 60% of sperm cells exhibit morphological abnormalities (27,28). This condition has been observed in several species, including humans and in wild species (29). Research suggests this condition may result from detrimental membrane modifications during the sperm maturation process, which subsequently hinders spermatozoa structural integrity (30,31). Supporting evidence for this hypothesis are in examples of these modifications observed in humans including acrosome and nuclear abnormalities (32), resulting in diminished ability of the sperm cell to penetrate the zona pellucida (33). This reduced sperm functionality has also been observed in the teratospermic domestic cat (29,34,35). The vast majority of research in cats focused on abnormalities which arise during spermatogenesis, resulting in primary abnormalities, adversely impacting sperm functionality (e.g. micro-, macro-, or polycephalic heads, acrosomal defects, mitochondrial sheath defects, biflagellate or tightly coiled flagellum; 29,[36][37][38]. Studies on secondary abnormalities are scarcer, rendering it difficult to understand the underlying etiology of these defects (e.g. detached head or flagellum, retained cytoplasmic droplets, bent midpiece or flagellum; 29,38). While not as severe, these secondary abnormalities still greatly diminish sperm quality and overall fertility. Comparing EV protein content of teratospermic males to the normospermic baseline will further elucidate the underlying basis of teratospermia, which is rampant in the 38 species of felids that are listed as threatened or endangered as reported by the International Union for Conservation of Nature (39).
The domestic cat serves as an ideal model because it is a close relative of wild felids. Furthermore, studies in the domestic cat have improved our basic understanding of human reproductive physiology, including potential physiological sources of teratospermia (40).
The objectives of the study were to (1) characterize and compare protein content of EVs between segments of the cat epididymis, and (2) compare EV protein compositions between normo-and teratospermic adults.

Materials and Methods
All chemicals and reagents were purchased from Sigma Chemical Company (St. Louis, MO, USA), unless otherwise stated.

Tissue collection
The study did not require the approval of the Animal Care and Use Committee of the Smithsonian Conservation Biology Institute because cat testes were collected at local veterinary clinics as byproducts from owner-requested routine orchiectomies. Adult (>1 year) domestic cat testis samples were supplied by local veterinary clinics following routine orchiectomy (n = 20 male tracts total). Tracts were transported and stored in phosphate buffered saline (PBS) at 4°C until processing within a 24-hour time period. Epididymal tissues were then removed from the rest of the testis in PBS using a scalpel blade until further processing.

Classifying normospermic vs teratospermic samples
A small sample of spermatozoa was isolated by making one 3 mm incision into the distal end of the cauda segment (42 (34).
To first assess overall abnormalities in this sample type, sperm cells were isolated from the cauda segment of individuals until a sample size of n = 5 normospermic males, and n = 7 teratospermic males were attained, and the different morphological abnormalities recorded. It should be noted that the prevalence of teratospermia in the samples used in this study does not represent the overall prevalence of the general domestic cat population. A separate set of samples were then assessed for the collection of EVs (n = 4 normospermic males, and n = 4 teratospermic males).
EV samples were subsequently collected within hour to reduce sample loss and protein degradation.

EV collections
Following identification of whether a sample was normospermic or teratospermic, entire epididymides were separated into the different, consecutive, segments (caput, corpus, cauda; 42) and further minced using a scalpel blade in PBS (n = 4 normospermic males, and n = 4 teratospermic males). Luminal fluid was allowed to seep for 5-10 minutes before collected into microcentrifuge tubes.
Cell debris was discarded from the supernatant by a series of centrifugations at 700 x g for 10 min and 3,000 x g for 10 min at room temperature, with the supernatant transferred to a new microcentrifuge tube following each centrifugation. The EV fraction was isolated from the remaining luminal fluid by ultracentrifugation at 100,000 x g for 2 h at 4°C and re-suspended in fresh PBS (Beckman Coulter Optima L-90K, SW 55 Ti rotor, 3.5 mL polycarbonate tubes catalogue number: 349622, filled to 3 mL each, with full dynamic braking to 0 rpm, Kadj= 88). Aliquots of EV samples were then stored at -20°C. Samples were processed, and analyzed via mass spectrometry (described below) within one month's time. Successful isolation of EVs was previously confirmed via observations performed using a transmission electron microscope (Zeiss 10 CA Transmission Electron Microscope) at the University of Maryland Laboratory for Biological Ultrastructure, USA (Rowlison et al., 2018). All relevant data of collection may be found in the EV-TRACK knowledgebase (EV-TRACK ID: EV200074, https://evtrack.org/). It should be noted that this study is using the operative term, "Extracellular Vesicle" (EV), in accordance to the guidelines set forth by the International Society for Extracellular Vesicles (17). ISEV endorses the use of this term when defining an isolated sample that has not been further analyzed for EV subtype (e.g. exosome, microvesicle, apoptotic vesicle, etc.).

Mass spectrometry analysis
EV samples were isolated from each consecutive segment of the epididymis (caput, corpus, and cauda), as described above, and the mass spectrometry analyses carried out by MS Bioworks (Ann Arbor, Michigan, USA) using nano liquid chromatography-tandem mass spectrometry.
Sample processing by MS Bioworks: The volume of each submitted sample was adjusted to 50μL with 50mM Tris.HCl, pH 8.0 and transferred to a 1.5mL microcentrifuge tube. 50μL of 2X modified RIPA buffer was added to each sample and sonicated briefly. Samples were incubated at 60°C for 15 minutes prior to clarification by centrifugation. The protein concentration of each sample was determined by Qubit fluorometry. 10μg of protein from each sample was processed by SDS-PAGE using a 10% Bis-Tris NuPAGE gel (Invitrogen) with the MES buffer system, the gel was run approximately 2cm and the mobility region was excised into by guest on October 3, 2020 https://www.mcponline.org Downloaded from 10 equally sized bands. In-gel digestion with trypsin was performed on each band using a ProGest robot (DigiLab) protocol: gels were first washed with 25mM ammonium bicarbonate followed by acetonitrile, reduced with 10mM dithiothreitol at 60°C, followed by alkylation with 50mM iodoacetamide at room temperature. Reduced gels were digested with trypsin (Promega) at 37°C for 4h, and quenched with formic acid. The supernatant was analyzed directly without further processing.
Analyses were performed with a Waters NanoAcquity HPLC system interfaced to a ThermoFisher QExactive HF. Peptides were loaded on a trapping column and eluted over a 75μm analytical column at 350nL/min; both columns were packed with Luna C18 resin (Phenomenex). The mass spectrometer was operated in data-dependent mode, with the Orbitrap operating at 60,000 FWHM and 17,500 FWHM for MS and MS/MS respectively. The fifteen most abundant ions were selected for MS/MS.

Proteomic searching
All RAW cat epididymosome files were converted to mzML files using Proteowizard product mass tolerance = ±20 ppm; report PSM ambiguity = True. PSMs, peptides, and proteins were filtered at 1% false discovery rate (FDR) based on a slided decoys.
To confirm the successful isolation of EVs and assess the quality of isolates, proteins known to be associated with exosomes, as well as commonly found contaminants, were assessed as according to the International Society for Extracellular Vesicles (17).

Experimental design and statistical analysis
Spermatozoa were isolated from the cauda epididymal segment and assessed for morphological abnormalities (n = 5 normospermic males, and n = 7 teratospermic males). by guest on  Statistical analyses were conducted using Graphpad Prism (version 6). Percent of sperm cells displaying each abnormality was then compared between treatment groups via Student's T-test.
Subsequently, the resultant MetaMorpheus data was analyzed by the FlashLFQ Bayesian foldchange tool (43). The following parameters were used: 10 ppm tolerance, normalize intensities, match between runs, Log2 Fold-change cutoff 0.5. For searches that required two runs (i.e., Normal/Teratospermic caput vs corpus vs cauda), the same random seed was utilized to allow similarity in MCMC outcomes. Significant differences were restricted to 1% FDR or <10% posterior probability, whichever was smaller. Fold changes were evaluated by groups in Table 1.

Comparative sperm morphological abnormalities between normospermic and teratospermic males
Morphological analyses revealed both primary and secondary abnormalities in mature, cauda-derived sperm samples. As expected, samples isolated from teratospermic individuals (as evidenced by ≥ 60% of spermatozoa with one or more morphological abnormalities) had significantly (p < 0.05) more abnormal cells compared to samples isolated from normospermic individuals (p = 0.0034; Table 2). Of the abnormalities observed, three were classified as primary (i.e. micro-or macrocephalic heads, polycephalic, as well as biflagellate) and five were secondary (i.e. missing acrosome, abnormal head-tail junction, bent midpiece, bent flagellum, and retained cytoplasmic droplet). Teratospermic samples had a significantly higher percentage of sperm cells with bent flagellum as compared to normospermic samples (p = 0.0085). While by guest on October 3, 2020 https://www.mcponline.org Downloaded from not significant, teratospermic samples also trended toward having higher percentages of missing acrosomes as compared to normospermic (p = 0.0522) as well as abnormal head-tail junctions (p = 0.0773).

Qualitative analysis of EV content
Proteomic analyses of normospermic males identified a total of 3,029 proteins including 115, 10, and 28 proteins unique to caput, corpus, and cauda-derived EVs, respectively, and 2,687 proteins present in all three segments ( Figure 1A). Analyses of teratospermic males identified a total of 3,028 proteins including 101, 15, and 48 proteins unique to caput, corpus, and caudaderived EVs, respectively, and 2,685 proteins present in all three segments ( Figure 1B). Several known proteins associated with exosomes were detected including 30 transmembrane or GPIanchored proteins, and 29 cytosolic proteins (Table 3) as according to the International Society for Extracellular Vesicles (17). Additionally, three proteins known to be common contaminants of exosome isolates were also detected ( Table 3; 17).
Normospermic and teratospermic males shared a total of 3,008 proteins, with 98 and 20 proteins unique to normospermic and teratospermic males, respectively ( Figure 1C). Further analysis of each epididymal segment identified differences in protein composition when comparing samples isolated from normospermic versus teratospermic males (Table 4). Several identified EV proteins in both sample groups had functions related to sperm motility (e.g. cysteine-rich secretory protein 1-CRISP1, hexokinase 1, adenylate kinase isoenzyme 1, sorbitol dehydrogenase), as well as zona and oolemma binding (e.g. acrosin, zona binding protein 1,2-  (Table 5). Proteins unique to normospermic or teratospermic males were also classified based on their known or potential biological processes ( Figure 2 A-B, Table 6, Supplementary File 2), with teratospermic males lacking numerous proteins with key biological functions including cellular structure, stimulus response, signaling, multicellular organismal processes, developmental processes, adhesion, locomotion, and immune system support.

Discussion
This study is the first to characterize the proteomic profile of epididymal EVs in the cat model, showing conserved molecular functions similar to other organisms including men, bull and mouse (16,22,24). Morphological analyses of teratospermic sperm samples showed a significant increase in the percentage displaying bent flagellum, a secondary abnormality that occurs from suboptimal maturation during epididymal transit (29,38). Main findings from this study suggest that proteins brought by EVs can be specific to caput, corpus, or cauda segments.
Several proteins were associated with reproductive processes (i.e. sperm motility, zona and oolemma binding) in specific segments. Furthermore, this was the first study to compare epididymal EV protein content between normospermic and teratospermic individuals of any species. Normospermic and teratospermic males shared many similar proteins (3,008 in total); however, 98, and 20 proteins were unique to normospermic and teratospermic males, respectively, with additional variances observed when comparing the consecutive segments between each group.
In total, 3,029 and 3,028 proteins were identified in EVs isolated from normospermic and teratospermic individuals, respectively. Comparative studies in the mouse and human resulted in much fewer identified proteins (1,640, and 1,022, respectively); however, this may be a result of intrinsic interspecies variation, or the use of different methods for collecting and processing of tissue samples, as well as protein identification, especially as the study in humans was published years prior when methods were not as advanced (16,22). The reference proteomes used may also have included fewer annotated proteins at the time of analysis. Regardless, there were many similar GO-classified molecular functions observed in all species including catalytic activity, binding, and transportation. Numerous proteins known to be associated with exosomes were detected including several transmembrane or GPI-anchored proteins as according to the International Society for Extracellular Vesicles (17), confirming the successful isolation of these epididymal exosomes.
Previous analysis of the EVs using transmission electron microscopy also confirmed successful isolation (26). Additionally, three proteins known to be common contaminants of exosome isolates were also detected likely as a result of the collection method (17). Mincing of the epididymal tissue proved to be the most feasible option, as compared to profusion of the luminal fluid; however, this also resulted in the release and collection of extraneous lipoproteins and albumin. While effort was made to reduce the amount of these contaminants, it should be noted that these were still detected in the overall collection.
Multiple candidate proteins brought by EVs during sperm maturation have a direct effect on the fertilization and subsequent quality of the preimplantation embryo in the domestic cat.
There were also several key proteins which aid in reproductive processes identified in this study that were also observed within the EVs of the other species. These included sorbitol dehydrogenase which was previously reported in the bovine and mouse (16,24), interleukin enhancer binding factor (also reported in the mouse; 16), cysteine rich secretory protein and zona pellucida binding proteins (mouse and human; 16,22), as well as disintegrin and metalloproteinase domain-containing proteins which were identified in all three models (16,22,24). Comparisons between these species illustrate the evolutionary conservation of certain proteins previously seen in other studies which contribute to the structure and function of the sperm cell (44)(45)(46). In addition to the conservation of certain proteins, it is also of note that the mechanism itself in which these vesicles are secreted and acquired by the maturing sperm population appear to be very similar between the studied animal models (3, 6, 14-17, 23, 47).  EVs identified several candidate proteins contributing to the zona   binding process including CRISP1, ZPBP1, ZPBP2, ADAM2, ADAM7, ADAM28, ADAM32 and acrosin. Specifically, ZPBP1, and ZPBP2 function in the binding of the sperm cell to the zona pellucida, while acrosin acts as a protease to aid in the penetration (1,4,5). Proteins CRISP1, and ADAM2, ADAM 7, ADAM 28, ADAM 32 then subsequently aid in the sperm penetration of the oolemma. Other key proteins identified from these analyses have also been previously reported to aid in various sperm functions in the human including motility (sorbitol dehydrogenase, interleukin enhancer binding factors 2 and 3, and hexokinase 1; 48-50) as well as maintenance of DNA integrity (DNA repair protein-RAD50; 51). Interestingly, the epididymal sperm binding protein (ELSPBP1) was also identified like in bovine EVs, which specifically targets dead sperm cells to presumably initiate degradation and removal from the maturing population (53). Together, acquiring these proteins can aid in the achievement of key functions while simultaneously eliminating poor-quality sperm cells to improve the condition of the overall population. As an example, Figure 6 illustrates predicted interactions between a subset of identified EV proteins in this study, and domestic cat sperm cell proteins that have been previously identified to aid in the fertilization process (STRING CONSORTIUM 2020, version 11.0). These interactions highlight the potential benefits that the content of EVs may provide to the developing sperm cell. Additionally, this figure highlights the potential interactions that EV proteins may have with each other, possibly altering their downstream effects and adding another layer of complexity to the study of these vesicles that to date has not been assessed in any species.
Comparing the EV content of normospermic individuals to those exhibiting teratospermia also provides novel information which will aid in better understanding the underlying physiology by guest on  of this condition. Previous research has implicated certain gene alterations in teratospermic individuals which may contribute to diminished sperm quality, but the downstream effects on protein expression have yet to be determined (29). Investigating the potential influence of epididymal EVs is also of great value as the etiology of secondary abnormalities, which occur during epididymal transit, are poorly understood. For example, sperm samples isolated from teratospermic individuals in this study were observed to have a significantly increased percentage of cells with bent flagellum, a secondary abnormality. As this deformity arises during the sperm epididymal transit, it is possible that the spermatozoa were unable to acquire the appropriate milieu of proteins in the proper sequential order as a result of altered EV content.
Abnormalities in the head-tail junction region of the sperm cell also tended to increase in teratospermic males. Contained within this region is the sperm centrosome which is essential in coordinating the first mitotic division of the zygote, diminished functionality of this cellular structure may inhibit subsequent embryonic development (53,54). Previous studies in our laboratory have observed significant improvement in the sperm centrosome maturation when exposing immature spermatozoa to EVs (25,26). Therefore, it is also possible that modifying the content of these EVs in teratospermic individuals may contribute to the development of malformed centrosomes.
Certain factors, such as hormone imbalance, altered luminal composition, and pH have been found as contributing factors to secondary abnormalities, but how these interplay with the miRNAs and proteins supplied via EVs remains unknown (38). Additionally, variance of any of these factors may also influence protein expression at the cellular level, potentially altering which proteins are even secreted from the epididymal epithelium. For example, many of the identified proteins unique to normospermic males, and not present in teratospermic EVs, were by guest on October 3, 2020 https://www.mcponline.org Downloaded from GO-classified in a number of vital biological functions including cellular processes (29.7%), intracellular transportation (13.5%), metabolism (13.5%), cellular structure (10.8%), stimulus response (8.1%), signaling (4.1%), multicellular organismal processes (4.1%), developmental processes (2.7%), adhesion (1.4%), locomotion (1.4%), and immune system support (1.4%).
While it is possible that these proteins may be provided via another mechanism, their absence in these EVs may still hinder sperm cell maturation. This is further evident as many proteins supplied by EVs have been identified as lacking any signal peptide sequence (3,6,(14)(15)(16)(17). While these proteins may somehow be secreted into the luminal fluid, possibly through the classical merocrine pathway, successful binding with the sperm cell is still unlikely as there is no signal sequence available to achieve this. One protein of interest for example, zona pellucida binding protein1 (ZPBP1), was observed to be trending toward a significant decrease in expression in the cauda-derived EVs of teratospermic males as compared to normospermic males. As this protein enables the sperm cell to penetrate the oocyte zona pellucida, a decrease in its acquisition may greatly hinder the success of fertilization.
Our collective results elucidate the type and quantity of proteins involved in the sperm cell maturation during epididymal transit. Proteomic analyses of the epididymal EV revealed vital information regarding mechanisms of protein transport and identified key paternal factors which may contribute to improved sperm quality and early embryonic development in the domestic cat model. Future studies will investigate roles of candidate proteins to assess the downstream effects on sperm maturation and developmental potential. Additional analyses will also be performed to further assess the subtype of extracellular vesicles secreted by the epididymal epithelium. Information gained from this study and future analyses will contribute to     Zona pellucida sperm-binding protein 2 ZPBP2 Zona pellucida binding protein 2 Figure 6. Example of interactions observed between a subset of proteins identified in exosomes of normospermic males, and sperm cell proteins that are involved in the fertilization process (STRING 11.0). Table 1. Groups used in fold-change analysis. Table 3. EV characterization based on known associated proteins, as well as contaminants commonly found in EV isolates (17).