Towards the Identification of Reliable Sperm Biomarkers for Male Infertility: a Sperm Proteomic Approach

Accepted: 1 September 2017

DOI: 10.1111/and.12919

ORIGINAL ARTICLE

Towards the identification of reliable sperm biomarkers for male infertility: A sperm proteomic approach

P. Intasqui1,2 | A. Agarwal1 | R. Sharma1 | L. Samanta3 | R. P. Bertolla2

1American Center for Reproductive Medicine, Cleveland Clinic, Cleveland, OH, Summary USA Male infertility evaluation is mainly based on semen analysis. Thus, identification of 2 Department of Surgery, Division of additional diagnostic methods is valuable. The aim of this study was to analyse the Urology, Human Reproduction Section, Sao Paulo Federal University – Sao Paulo Hospital, sperm proteome of infertile men to identify the underlying mechanisms and reliable Sao Paulo, Brazil diagnostic biomarkers. This cross-sectional study consisted of 16 infertile men and 3Department of Zoology, Ravenshaw seven proven fertile men. An LC-MS/MS approach was performed in five pooled sam- University, Cuttack, India ples of each group (proven fertile men, primary infertility and secondary infertility). Correspondence Differentially expressed proteins were used for functional enrichment analyses, and Ashok Agarwal, American Center for Reproductive Medicine, Cleveland Clinic, the most central proteins involved in altered functions in both infertile groups and the Cleveland, OH, USA. testis-specific proteins were validated using Western blotting and immunocytochem- Email: [email protected] istry. In total, 1,305 sperm proteins were identified, of which 102 were underex- Funding information This work was conducted with financial pressed and 15 were overexpressed proteins in both infertile groups. Underexpressed support from the American Center for proteins were mostly related to protein post-translational modification and folding, Reproductive Medicine, Cleveland Clinic, USA, and the São Paulo Research Foundation, Brazil especially BAG6, HSPA2 and SPA17. Validation analysis revealed an underexpression (FAPESP, grant# 2015/18867-6) of BAG6 in infertile men, whereas HSPA2 and SPA17 expressions did not differ between the groups. No differences were observed in the sperm localisation of these proteins. An overexpression of HIST1H2BA—a testis-specific protein—was observed in both proteomic approaches. Therefore, BAG6 and HIST1H2BA are potential candi- dates for male infertility biomarkers.

KEYWORDS biomarkers, infertility, male, proteomics, spermatozoa, validation studies

1 | INTRODUCTION any major alterations in semen variables (Abu, Franken, Hoffman, & Henkel, 2012; Shamsi et al., 2010). Infertility affects approximately 15% of reproductive age couples, and Due to these limitations, additional methods are required to eval- male factors are responsible in approximately 50% of cases (Brugh uate male fertility. Thus, efforts have been made towards understand- & Lipshultz, 2004). Male infertility can be attributed to a myriad of ing and identifying altered molecules that might be utilised during causes ranging from varicocele (1992), obesity (Jensen et al., 2004), infertility management (Kovac, Pastuszak, & Lamb, 2013). In this genetic factors (Shamsi, Kumar, & Dada, 2011), lifestyle factors (i.e., context, the study of the sperm proteome is of great interest because smoking; Hassan & Killick, 2004) and spinal cord injury (Monga, Bernie, sperm proteins play a key role in the maintenance of sperm morphol- & Rajasekaran, 1999). Currently, clinical evaluation of male infertility ogy, motility patterns, acrosome formation and reaction, capacitation, is mainly based on semen analysis, which, although being important to and fertilisation (Ashrafzadeh, Karsani, & Nathan, 2013). Hence, al- assess the male fertility potential, does not provide a direct prediction tered sperm protein expression in male infertility may be further used of reproductive success. Therefore, infertile men can present without as noninvasive diagnostic, prognostic and/or therapeutic tools of male

Andrologia. 2017;e12919. wileyonlinelibrary.com/journal/and © 2017 Blackwell Verlag GmbH | 1 of 11 https://doi.org/10.1111/and.12919 2 of 11 | INTASQUI et al. infertility (Kovac et al., 2013). Moreover, it may provide important in- Center. Infertile men were divided into primary and secondary infer- sights into sperm physiology and its underlying mechanisms (Rahman, tility groups for the shotgun proteomic study. For the proven fertile Lee, Kwon, & Pang, 2013). group, only normozoospermic men who had fathered a child in the It has been hypothesised that the sperm protein profile in infer- past year were included. Infertile men were aged 20–50 years, pre- tile men is altered, and some researchers have assessed these pro- sented with infertility of at least one-year duration, irrespective of files in several male infertility conditions including varicocele (Agarwal, the cause and with a round cell concentration lower than 1 × 106/ Sharma, Samanta, Durairajanayagam, & Sabanegh, 2016; Agarwal, ml. Exclusion factors were azoospermia, history of cancer and cancer Sharma, Durairajanayagam, Ayaz, et al., 2015; Agarwal, Sharma, treatment, previous surgery of the male reproductive tract, erectile Durairajanayagam, et al., 2016; Hosseinifar et al., 2013), obesity dysfunction, cryptorchidism or use of exogenous testosterone. Men (Kriegel et al., 2009; Liu et al., 2015), smoking (Chen et al., 2015), failed were also excluded if their female partner had been diagnosed with fertilisation after assisted reproduction techniques (ART; Frapsauce a condition associated with female infertility, such as endometriosis, et al., 2014; Legare et al., 2014; Pixton et al., 2004), asthenozoosper- advanced age or irregular menses. mia (Amaral et al., 2014; Hashemitabar, Sabbagh, Orazizadeh, Ghadiri, No differences between the groups were observed regarding age, & Bahmanzadeh, 2015; Liu et al., 2015; Martinez-Heredia, de Mateo, body mass index (BMI), semen volume, sperm concentration and mo- Vidal-Taboada, Ballesca, & Oliva, 2008; Parte et al., 2012; Shen, Wang, tility, and round cells concentration. Infertility duration, described as Liang, & He, 2013; Siva et al., 2010; Zhao et al., 2007) and idiopathic median; inter-quartile range (1st quartile–3rd quartile), was 15; 27 (9– infertility (Legare et al., 2014; McReynolds et al., 2014; Xu et al., 2012). 36) months for the primary infertility group, 18; 12 (12–24) months for Sperm protein profiles have also been evaluated in men with semen the secondary infertility group (p > .05) and 12; 12 (12–24), if merging oxidative stress (Ayaz et al., 2015; Sharma et al., 2013) and sperm both groups. In the primary infertility group, 22.2% of men were smok- DNA fragmentation (Intasqui et al., 2013; de Mateo et al., 2007), ers, 11.1% had varicocele, 44.4% were obese, and 77.8% had alter- which are common causes of male infertility. Additionally, the sperm ation in their semen analysis. In the secondary infertility group, 42.9% proteome was also compared in samples from infertile men with those were smokers, 57.1% had obesity, and 71.4% presented with altered of men with proven fertility, although infertility was described only as semen analysis. oligoasthenozoospermia (Thacker et al., 2011). Semen was collected at the Andrology Laboratory by mastur- To the best of our knowledge, no study has analysed the sperm bation following 2–6 days of ejaculatory abstinence (World Health proteome of men with primary or secondary infertility, regardless of Organization, 2010). On average, samples in the fertile, primary infer- the cause of infertility, and compared the results with those of proven tility and secondary infertility groups had an ejaculate volume of 4.06, fertile men. Such an analysis could help us better understand the un- 4.95 and 3.21 ml respectively, which was then used for the shotgun derlying molecular mechanisms of male infertility that are common to proteomic analysis and the validation studies. different causes of infertility as well as provide sperm biomarkers of After liquefaction, a semen aliquot was utilised for conventional infertility that would be applicable to most infertile men. semen analysis, which was performed according to the WHO (2010) Therefore, the aim of this study was to compare the sperm pro- guidelines (World Health Organization, 2010). Another aliquot (200 μl) teome between men with primary or secondary infertility, irrespective was centrifuged for 10 min at 500 g; the pellet was fixed in 4% (v/v) of the cause, with that of proven fertile men to identify altered molec- paraformaldehyde in phosphate-buffered saline (PBS, Irvine Scientific, ular pathways and potential sperm biomarkers of male infertility. For CA, USA) for 20 min at room temperature, washed in PBS (3 × 5 min, this, we utilised a shotgun proteomic approach to identify the differ- 500 g), resuspended in 1 ml of PBS and kept at 4°C for protein lo- entially expressed sperm proteins, and we validated the results using calisation analysis by immunocytochemistry. The remaining semen Western blotting and immunocytochemistry in order to assess protein volume was frozen without cryoprotectants and kept at −80°C for expression and localisation respectively. shotgun proteomic analysis and Western blotting. All reagents used in this study were purchased from Thermo Fisher Scientific (MA, USA), Sigma-Aldrich (MO, USA) or Bio-Rad Laboratories (CA, USA), unless 2 | MATERIAL AND METHODS otherwise described. Five pooled sperm samples from each group (control, primary 2.1 | Ethical approval infertility and secondary infertility) were subjected to shotgun pro- This study was approved by the Institutional Review Board from the teomic analysis using one-dimensional electrophoresis followed by Cleveland Clinic. All included subjects provided written informed consent. liquid chromatography coupled to tandem mass spectrometry (1D- LC-MS/MS). Results were used for bioinformatic analyses to identify the protein–protein interaction networks and the altered biological 2.2 | Study design pathways. From the comparative proteomic analysis, key proteins of A cross-sectional observational study was conducted using infer- the top molecular pathway in both primary and secondary infertility tile patients, who were referred to the Andrology Laboratory of the groups were chosen for validation by Western blotting and immuno- American Center for Reproductive Medicine (ACRM), Cleveland cytochemistry, together with proteins responsible for altered sperm Clinic, USA, and men with proven fertility, who were enrolled by the function. Because in a clinical setting, the identification of a general INTASQUI et al. | 3 of 11 biomarker of infertility—without the distinction of primary or second- Proteome Discoverer (version 1.4.1.288, Thermo Fisher Scientific). ary infertility—would be more valuable, and for increased sensitivity, Charge state deconvolution and de-isotoping were not performed. the infertility groups were merged into one infertility group for the For protein identification, all MS/MS data were analysed using validation study. Mascot (version 2.3.02, Matrix Science, London, UK), SEQUEST (version 1.4.0.288, Thermo Fisher Scientific) and X! Tandem (version CYCLONE, The GPM) search programs. Mascot, Sequest and X! 2.3 | Shotgun proteomic analysis (1D-LC-MS/MS) Tandem were set up to search the human reference (33,292 entries) At the time of the analyses, whole ejaculate samples were thawed at assuming trypsin as the digestion enzyme. These searches were per- 4°C and centrifuged at 500 g for 10 min at 4°C. After the supernatant formed with a fragment ion mass tolerance of 0.8 Da and a parent seminal plasma was completely removed, the pellet was washed with ion tolerance of 10 parts per million (PPM). Carbamidomethylation of PBS (3 × 10 min, 500 g, 4°C). For sperm protein extraction, radioim- cysteine was specified as a fixed modification, and oxidation of methi- munoprecipitation assay (RIPA) buffer (Sigma-Aldrich) supplemented onine was specified as a variable modification. with protease inhibitors (cOmplete, Mini Protease Inhibitor Cocktail For validation of MS/MS-based peptide and protein identifications, Tablets, EDTA-free, Roche Applied Science, Mannheim, Penzberg, these search results were then uploaded into the Scaffold 4.0.6.1 pro- Germany) was then added to the washed pellet according to the gram (Proteome Software Inc., OR, USA). Only a >99% probability to sperm concentration. Samples were thoroughly homogenised and achieve a false detection rate (FDR) < 1.0% was accepted as well as kept overnight at 4°C. After centrifugation at 13,000 g for 30 min at at least two identified peptides. Proteins were annotated with gene 4°C, an aliquot of the supernatant was utilised for protein estimation, ontology (GO) terms from the National Center for Biotechnology and the remaining volume was stored at −20°C until proteomic and Information (NCBI, downloaded 21 October 2013). Western blotting analyses. Protein quantification was performed by comparing the nor- Protein estimation was performed using the bicinchoninic acid malised spectral counts for these samples. Appropriate filters were (BCA) protein assay kit (Thermo Fisher Scientific), using bovine serum used to identify the differentially expressed proteins (DEPs) between albumin (BSA) as standard. Samples quantified with coefficients of each comparison (primary infertility × proven fertile; secondary infer- variation over 5% were quantified again in another run. tility × proven fertile; primary infertility × secondary infertility). DEPs For shotgun proteomic analysis, samples were pooled sepa- were dependent on the overall abundance of the proteins among the rately for each group (five samples per group), and each pool was three replicate runs (Agarwal, Sharma, Samanta, et al., 2016; Gokce, run in triplicate (technical replicates) following the same protocol Shuford, Franck, Dean, & Muddiman, 2011). described in depth in (Agarwal, Sharma, Durairajanayagam, et al., 2016; Agarwal, Sharma, Samanta, et al., 2016). Both spermatozoa 2.4 | Bioinformatic analysis and selection of proteins and protein concentrations were normalised in each group prior for validation to analysis, that is, equal quantity of protein was obtained by an equal number of spermatozoa. Briefly, each pool was mixed with Protein–protein interaction networks, functional annotation and sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS- functional enrichment analysis of DEPs were performed using publicly PAGE) buffer, boiled for protein denaturation, and separated on available bioinformatic annotation tools and databases such as GO one-dimensional (1D) 12.5% Tris–HCl gels (15 μg protein/lane) Term Finder; GO Term Mapper; UniProt; Software for Researching with a constant voltage of 150 V for 35 min. Prior to in-gel diges- Annotations of Proteins (STRAP); Database for Annotation, tion, protein bands were reduced with dithiothreitol (DTT) and al- Visualization, and Integrated Discovery (DAVID); and proprietary soft- kylated with iodoacetamide. Each gel lane was cut into six pieces ware packages such as Ingenuity Pathway Analysis (IPA) and Search and washed with milli-Q water, then dehydrated in acetonitrile. Tool for the Retrieval of Interacting Genes/Proteins (STRING). Complete protein digestion was obtained with 5 μl of 10 ng/μl tryp- Thereafter, the major enriched functions related to the underex- sin in 50 mm ammonium bicarbonate overnight at room tempera- pressed and overexpressed proteins in the primary and secondary ture. Formed peptides were extracted from the gel and analysed infertility groups were identified. The proteins related to these func- by LC-MS/MS in a LTQ-Orbitrap Elite Hybrid Mass Spectrometer tions that were differentially expressed in both infertility groups and system (Thermo Fisher Scientific). The HPLC system was a Dionex with the highest number of interactions in the protein–protein net- with an Acclaim Pepmap C18 2 μm and 100 Å reversed-phase cap- work were then detected and, through an in-depth literature search illary chromatography column. Tandem MS data were obtained in a of these proteins, the proteins more commonly related to infertility data-dependent method. Briefly, a full-profile MS scan at 60,000 were selected for validation: large proline-rich protein BAG6 (BAG6), resolution (at 400 m/z) between 300 and 2,000 m/z was performed heat shock-related 70-kDa protein 2 (HSPA2) and sperm surface pro- on the Orbitrap FT instrument using a precursor isolation window tein Sp17 (SPA17). Moreover, testis-specific proteins differentially ex- of 3 m/z at 35% relative collision energy. The charge state screen- pressed in all infertility groups were also validated, given that these ing was not enabled, and the dynamic exclusion option was enabled proteins may reflect sperm maturation: SPA17 and histone H2B type after three repeated acquisitions within a 20-s duration with each 1-A (HIST1H2BA). These selected proteins may play the most import- exclusion lasting 90 s. Mass spectra were then extracted by the ant role in the observed infertility. 4 of 11 | INTASQUI et al.

(sc-2005, Santa Cruz Biotechnology, TX, USA); (ii) HSPA2—1:5,000 2.5 | Protein validation (v/v) anti-HSPA2 (ab154374, Abcam) and 1:10,000 (v/v) anti-rabbit To identify potential biomarkers of male infertility that could be ap- (ab97051, Abcam); (iii) 1:1,000 (v/v) anti-SPA17 (ab172626, Abcam) plied to most of the infertile men, irrespective of the cause or type of and 1:20,000 (v/v) anti-rabbit; and (iv) HIST1H2BA—1:1,000 (v/v) the infertility, primary and secondary infertility groups were merged anti-HIST1H2BA (ab178426, Abcam) and 1:20,000 (v/v) anti-rabbit. into one infertility group for the validation study. Decision to merge Secondary antibodies were washed three times for 40 min with TBST both primary and secondary infertility groups as one group was due at room temperature, and peroxidase activity was revealed using to (i) a relatively small variation in protein expression between both enhanced chemiluminescence (Pierce ECL Western blotting sub- infertility groups, (ii) the possibility of increasing the number of in- strate, Thermo Fisher Scientific) as a substrate. Chemiluminescence dividuals, thus decreasing the likelihood of incurring in type II error was detected in the ChemiDoc Touch Imaging System (Bio-Rad). and (iii) the common basis of the clinical phenotype—both groups Housekeeping protein expression was detected after membrane were infertile (with diverse underlying causes for infertility between stripping (Restore PLUS Western blot stripping buffer, Thermo Fisher different patients in the same group). Although we understand this Scientific, 15 min at room temperature and 15 min at 37°C, followed increases the chance of an alpha error, we maintained stringency at by one wash with TBST for 10 min), blocking and re-incubation with the % cut-off for the alpha error while decreasing the odds of incur- 1:1,000 (v/v) anti-α-Tubulin primary antibody (ab4074, Abcam) and ring in a beta error. Therefore, for protein expression and localisation 1:10,000 (v/v) anti-rabbit secondary antibody. analyses, proven fertile men were compared to infertile men. With Densitometric analysis and quantification of the bands were this approach, sensitivity is increased, albeit at the cost of reducing done using the Image Lab 5.2.1 Software (Bio-Rad), and each band specificity. intensity was normalised by the housekeeping protein (α-Tubulin), in order to facilitate a better comparison between different membranes and decrease intra-assay variability. Data are shown as the 2.5.1 | Protein expression analysis densitometric value of the protein divided by the value of α-Tubulin Western blotting was performed in individual and pooled samples (arbitrary units). depending on the protein concentration (samples with low-protein concentration were pooled). For the proven fertile group, seven in- 2.5.2 | Immunocytochemistry analysis dividual samples (n = 7) were utilised, whereas for the infertile group, seven individual samples were utilised and nine samples were pooled For the immunocytochemistry, a new and independent cohort of sam- into four different pools (final n = 11), assuring biological variability ples was collected. A concentration of 2 × 106 fixed spermatozoa/ml between the pools (pool 1 with two different samples, pool 2 with from three proven fertile men and two infertile men was placed on the three samples and pools 3 and 4 with two samples each). coverslips (Deckglaser cover glasses, 12 mm, Marienfeld, Germany)

Western blotting was carried out as previously described by previously coated with 0.01% (v/v) poly-l-lysine (Sigma-Aldrich) and Agarwal, Sharma, Samanta et al. (2016). Briefly, 20 μg of proteins of allowed to dry at room temperature overnight. Cells were permeabi- each pool or individual sample was added to a final volume of 30 μl of lised with 0.2% (v/v in PBS) Triton X-100 (Bio-Rad) for 20 min, washed RIPA buffer and Laemmli buffer (4× Laemmli sample buffer—Bio-Rad four times with 0.05 m glycine in PBS and blocked with 1% (w/v) BSA with 10% [v/v] 2-mercaptoethanol—Sigma-Aldrich). Samples were in PBS for 1 hr, both at room temperature. Samples were then incu- boiled for 5 min and loaded into 10%–20% Tris–Tricine polyacryl- bated with primary antibodies at 4°C overnight, in the following dilu- amide gels (Mini-PROTEAN, Bio-Rad). SDS–PAGE was performed tions: (i) BAG6—10 μg/ml; (ii) HSPA2 (ab89130, Abcam)—20 μg/ml; in the Bio-Rad system in a running buffer (Bio-Rad) consisting of and (iii) SPA17—1:100 (v/v). Coverslips were again washed four times

25 mm Tris, 192 mm glycine and 0.1% (w/v) SDS, pH 8.3. Proteins and incubated with Alexa Fluor 488-conjugated secondary antibodies were then transferred to a polyvinylidene fluoride (PVDF) membrane for 1.5 hr at 37°C. For BAG6 and HSPA2, an anti-mouse antibody was (Thermo Fisher Scientific) in a semi-dry system (Bio-Rad), using a used (A-11001, Thermo Fisher Scientific), whereas for SPA17, an anti- transfer buffer with 25 mm Tris–HCl (Sigma-Aldrich), 192 mm glycine rabbit antibody (ab150077, Abcam) was used, both in a 1:1,000 dilu- (Sigma-Aldrich) and 20% (v/v) methanol (Sigma-Aldrich). Blocking tion. Coverslips were then washed again and incubated with 1 μg/ml was performed in SuperBlock Blocking Buffer in PBS (Thermo Fisher 7-aminoactinomycin D (7-AAD, Thermo Fisher Scientific) in DMSO/ Scientific), for 1 hr at room temperature. Primary antibodies were in- PBS for 30 min at 37°C. After washing three times, coverslips were cubated overnight at 4°C, diluted in 1% (w/v) BSA (Sigma-Aldrich) in added to clean slides and mounted with mounting medium contain- tris-buffered saline Tween-20 (TBST) buffer (25 mm Tris, 0.15 m NaCl, ing 0.01 μg/ml Hoechst 33258 (Sigma-Aldrich) in water and equal 0.05% Tween-20, Thermo Fisher Scientific). After washing three volumes of PBS and glycerol (Sigma-Aldrich). Cells were visualised by times with TBST (3 × 10 min), the membranes were incubated with confocal microscopy using a Leica DM6000CFS upright microscope horseradish peroxidase-conjugated secondary antibody diluted in 1% and LAS AF software v2.7.3 (Leica Microsystems, GmbH, Wetzlar, (w/v) BSA in TBST, for 1 hr at room temperature. Antibody dilutions Germany). Approximately 150 cells per group were analysed, and the for each protein were as follows: (i) BAG6—0.5 μg/ml anti-BAG6 percentage of cells showing protein localisation in the head or tail was (ab88292, Abcam, Cambridge, UK) and 1:2,000 (v/v) anti-mouse counted for each protein. INTASQUI et al. | 5 of 11

infertility), as well as the protein–protein interaction network of the 2.6 | Statistical analyses DEPs related to altered biological functions, are shown in Figure 1. A Statistical analyses were performed in the PASW (SPSS) 18.0 soft- total of 117 DEPs were differentially expressed in the infertile men ware for Windows. For the quantitative data from shotgun proteomic (both primary and secondary infertility taken together) in comparison analysis, the numerical values used in the quantitation correspond to with fertile counterparts. Furthermore, 356 proteins were differen- the normalised spectral counts (nSC, SC/ΣSC). In order for the pro- tially expressed when comparing primary infertility to secondary in- teins to be considered significant, the nSC ratio should be greater than fertility (data not shown), whereas 1,188 were common to all three 2.0 or less than 0.5 with a p-value < .05 using t test. Each set of data groups. DEPs in each comparison are further described in Table S1. was tested for normality (Shapiro–Wilk test) and homogeneity of vari- Of all the identified networks by Ingenuity Pathway Analysis (IPA), ance (Levene’s test). Data were compared between the groups using a the one involved in protein folding and post-translational modifications Student’s t test to identify the DEPs. (PTM) was the most prominent one as all the 35 focus proteins were Semen analysis and Western blotting (protein/α-tubulin) data were underexpressed in the infertile groups of our data set (Figure 1). From compared between controls and infertile men by the nonparametric this network, the testis-enriched chaperone HSPA2 and its cochaper- Mann–Whitney test. In this case, data are presented as median; inter- one BAG6, important hubs in the protein–protein interaction network, quartile range (Q1–Q3). Furthermore, proteins presenting a statisti- were chosen for validation based on previous reports on their involve- cally significant difference in the Western blotting analysis were used ment in sperm–oocyte recognition (Bromfield, Aitken, & Nixon, 2015). in a logistic regression analysis. These results were used to construct Furthermore, important proteins implicated in sperm dysfunction, that a receiver operating characteristics (ROC) curve, in order to assess is, SPA17 and HIST1H2BA, were also validated, the latter being over- specificity, sensitivity and the area under the curve. expressed in these groups in the shotgun proteomic analysis. For immunocytochemistry data, a Mann–Whitney test was also Protein validation data (expression and localisation) demonstrated performed to compare the percentages of each protein localisation that, as expected, BAG6 was significantly underexpressed in sperm between the groups, and the results are presented as a boxplot. In all samples from infertile men (3.9-fold decrease, compared to proven cases, a p value < .05 was considered statistically significant. fertile men, p = .003, Figure 2a), whereas HIST1H2BA was overexpressed in these men (1.8-fold increase, p = .001, Figure 3c). SPA17 was identified with two different molecular masses—17 3 | RESULTS and 20 kDa (Lea, Richardson, Widgren, & O’Rand, 1996; Figure 4a). However, the expression of both SPA17 forms and also of HSPA2 Using a shotgun proteomic approach, we were able to identify a total did not differ between the groups (p > .05, Figures 3a and 4a). BAG6 of 1,305 sperm proteins among the primary and secondary infertility and HSPA2 were expressed throughout the spermatozoon, both in and the proven fertile groups. Venn diagrams with the DEPs in three the head and in tail (Figures 2b and 3b). SPA17, although also ex- different comparisons (proven fertile × primary infertility; proven fer- pressed along the spermatozoa, had its highest expression observed tile × secondary infertility; proven fertile × primary and secondary in the sperm tail, particularly in the tail principal piece (Figure 4b).

FIGURE 1 Shotgun proteomic results. Top panel (a-c): Venn diagrams representing all identified proteins. In green, underexpressed or absent proteins in the infertile groups. In red, overexpressed or exclusively expressed proteins in the infertility groups. The intersection represents the conserved proteins (observed in equal quantities in both groups) of each comparison. Bottom panel (a-c): Biological networks of the functions enriched by the differentially expressed proteins in each comparison. The proteins selected for validation are highlighted in red 6 of 11 | INTASQUI et al.

FIGURE 2 Validation study of BAG6 protein. (a) Protein expression analysis by Western blotting. a.u., arbitrary units; F, proven fertility group; I, infertility group. (b) Protein localisation analysis in spermatozoa by immunocytochemistry using confocal microscopy. For all cases, groups were compared by the Mann–Whitney test

FIGURE 3 Validation study of HSPA2 and HIST1H2BA proteins. (a) HSPA2 expression analysis by Western blotting. (b) HSPA2 localisation analysis in spermatozoa by immunocytochemistry using confocal microscopy. (c) HIST1H2BA expression analysis by Western blotting. a.u., arbitrary units; F, proven fertility group; I, infertility group; *, outlier. For all cases, groups were compared by the Mann–Whitney test

No differences were seen in the sperm localisation of these proteins a negative predictive value of 85.7% and a total predictive value of (p > .05, Figures 2–4). 93.8%. For HIST1H2BA, a significant model was achieved (p = .0003), Finally, two logistic regression models were constructed, one with a positive predictive value (for detecting the “Infertile” group) for BAG6 and one for HIST1H2BA. For BAG6, a significant model of 90.9%, a negative predictive value of 100% and a total predictive was achieved (p = .0002), with a positive predictive value of 100%, value of 94.4%. A ROC curve was constructed for each model, with an INTASQUI et al. | 7 of 11

FIGURE 4 Validation study of SPA17 protein. (a) Protein expression analysis by Western blotting. a.u., arbitrary units; F, proven fertility group; I, infertility group. (b) Protein localisation analysis in spermatozoa by immunocytochemistry using confocal microscopy. For all cases, groups were compared by the Mann–Whitney test area under the curve value of 92.1% (p = .005) for BAG6 and of 93.5% (p = .004) for HIST1H2BA (Figure 5). Sensitivity and specificity values were, respectively, 100% and 85.7% for BAG6 and 88.9% and 100% for HIST1H2BA.

4 | DISCUSSION

Despite the high prevalence of infertility, and, particularly, of male infertility, little is known about its underlying molecular mechanisms. Furthermore, male infertility evaluation is based upon semen analysis, which is unable to provide a reliable diagnosis of infertility. Therefore, the aim of this study was to compare the sperm proteome between proven fertile men and infertile men (both with primary or secondary infertility), focusing on detecting altered post-genomic pathways and identifying potential biomarkers that could be used during male infertility manage- FIGURE 5 ROC curve generated by the logistic regression model ment. Using a shotgun proteomic approach, we were able to identify a including BAG6 and HIST1H2BA as independent variables and group total of 1,305 sperm proteins. Of these, 117 proteins were differentially (proven fertility or infertility) as dependent variable expressed in both infertile groups, which were mostly involved with protein PTM and folding functions, according to our bioinformatic analyses. Post-translational modifications are especially important in mature 1995). Similarly, a correct folding of the tertiary protein structure is spermatozoa, in which the nucleus is transcriptionally and transla- essential for protein activity and to avoid protein degradation and the tionally inert. Therefore, protein PTM or the acquisition of new pro- accumulation of misfolded proteins, which are involved with several teins during epididymal transit is essential for regulation of sperm diseases (Valastyan & Lindquist, 2014). Thus, the overall underexpres- maturation and function (Baker, 2016; Baker et al., 2012). The most sion of sperm proteins related to PTM and folding in men with primary studied sperm PTMs so far are protein phosphorylation, glycosylation, and secondary infertility may indicate an alteration in these processes methylation, acetylation and ubiquitination (Baker, 2016). Protein in male infertility, but more studies are needed to confirm this find- phosphorylation, especially tyrosine phosphorylation, for instance, ing. Similar results were also observed in infertile men with varico- is important to the process of sperm capacitation (Bajpai & Doncel, cele (Agarwal, Sharma, Durairajanayagam, Ayaz, et al., 2015; Agarwal, 2003; Battistone et al., 2014; Ficarro et al., 2003; Visconti et al., Sharma, Durairajanayagam, Cui, et al., 2015), asthenozoospermia 8 of 11 | INTASQUI et al.

(Amaral et al., 2014; Siva et al., 2010) and sperm DNA fragmentation in male germ cells and probably acts as a nucleosome dissociating fac- (Intasqui et al., 2013), corroborating our data. Of all the proteins re- tor. Therefore, it appears to be important for chromatin condensation lated to protein PTM and folding, HSPA2, BAG6 and SPA17 were se- during spermatogenesis. In mice, HIST1H2BA expression is highly in- lected for validation due to their central role in these pathways. creased in 25-days post-partum testis and remains highly expressed HSPA2 is a testis-enriched member of the HSP70 chaperone in the adult testis (Sun & Qi, 2014). In rat spermatozoa, its expression family, and it is capable of binding to misfolded or denatured pro- was 1.8 higher than that of the control spermatozoa after a 9-week teins, inhibiting their aggregation and enabling their correct folding coadministration of bleomycin, etoposide and cis-platinum—the most (Beckmann, Mizzen, & Welch, 1990). It is also involved in protein commonly used drugs for testicular cancer treatment—followed by transport through cell membrane (Chirico, Waters, & Blobel, 1988) a 9-week recovery period (Maselli, Hales, Chan, & Robaire, 2012). and DNA repair (Dix et al., 1997). In humans, it is expressed during This alteration is inherited in the progeny, with an overexpression of spermatogenesis, where it plays a key role in the formation of the HIST1H2BA in spermatozoa from the F1 generation (Maselli, Hales, & synaptonemal complex (Huszar, Stone, Dix, & Vigue, 2000), in sperm Robaire, 2013). These animals also had had lower sperm protamination nuclear DNA protamination (Govin et al., 2006) and in removal of ex- levels, suggesting that this protein overexpression may be involved with cess cytoplasm after spermiogenesis (Huszar et al., 2000). It is also chromatin alterations leading to male infertility (Maselli et al., 2012). expressed during sperm maturation, and it has been considered a On the other hand, SPA17 expression (both of 17 and 20 kDa) and marker of mature spermatozoa (Huszar et al., 2000). HSPA2 is an im- localisation were not statistically different between the groups. portant protein in mature spermatozoa for binding to the oocyte zona SPA17 is a low molecular weight, mannose-binding protein that pellucida (Redgrove et al., 2012). Several studies have shown that binds to zona pellucida carbohydrates and is therefore an important HSPA2 is underexpressed in infertile men and is involved with failure protein for a number of processes including sperm capacitation, acro- in sperm–oocyte recognition (Cedenho et al., 2006; Redgrove et al., some reaction, recognition and binding to the oocyte zona pellucida, 2012, 2013). In azoospermic men, a complete absence of HSPA2 was and fertilisation (Chiriva-Internati et al., 2009; O’Rand, Widgren, & observed, which was associated with cell arrest as primary spermato- Fisher, 1988; Richardson, Yamasaki, & O’Rand, 1994; Wen, Richardson, cyte (Dix et al., 1996). Furthermore, a reduction in HSPA2 expression & O’Rand, 1999). In the human testis, a weak expression of this protein was observed in oligozoospermic men (Cedenho et al., 2006; Lima was observed in the cytoplasm of spermatocytes, whereas a strong ex- et al., 2006) and men with pregnancy failure following in vitro fertili- pression was detected in early and late spermatids, which suggests that sation (Huszar et al., 2003) or varicocele (Lima et al., 2006). Stability SPA17 plays a role in the differentiation process. Furthermore, 95% of of HSPA2 depends upon its interaction with BAG6 protein (Kawahara, ejaculated spermatozoa also expressed SPA17, which was observed Minami, & Yokota, 2013; Sasaki et al., 2008), and deletion of Bag6 using light microscopy in the tail principal piece (Chiriva-Internati et al., gene leads to degradation of HSPA2 and, consequently, to male in- 2009; Grizzi et al., 2003). Under electron microscopy, it was demon- fertility (Sasaki et al., 2008). In humans, HSPA2 and BAG6 have been strated that this expression was restricted to the sperm fibrous sheath shown to be expressed and colocalised both in testis, epididymis and (Chiriva-Internati et al., 2009). Another study, however, detected mature spermatozoa, with equal expression between noncapacitated SPA17 expression in the sperm head and tail, mostly in the principal and capacitated spermatozoa, but in different localisations (equatorial piece (Lea, Widgren, & O’Rand, 2004). In agreement with this finding, region and anterior region of the head respectively). Studying the sper- we also observed the presence of SPA17 in the principal piece of the matozoa from men with zona pellucida binding failure demonstrated a sperm tail and in the head of about 5% of spermatozoa. SPA17 localisa- concomitant lack of HSPA2 and BAG6 (Bromfield et al., 2015). tion in the sperm principal piece appears to be important for cell signal- Although our validation results showed the underexpression of ling and motility, as this protein may regulate protein kinase A anchoring BAG6 in spermatozoa from infertile men, no differences were ob- protein (AKAP) activity (Carr et al., 2001; Chiriva-Internati et al., 2009; served in HSPA2 expression. Moreover, both BAG6 and HSPA2 were Colledge & Scott, 1999). Therefore, we can suggest that infertile men, expressed throughout the spermatozoon, both in the head and in tail, both with primary or with secondary infertility, do not present with an and this localisation did not change in infertile men (Figures 2b and 3b alteration in SPA17 expression and localisation in spermatozoa. respectively). Therefore, it is possible to conclude that regardless of Using BAG6 and HIST1H2BA, the proteins presenting a statisti- cellular localisation, the overall underexpression of BAG6 is an import- cal difference between infertile and fertile groups, in separate logis- ant alteration related to male infertility. Given the interaction of this tic regression models demonstrated that these proteins present high protein with HSPA2, we also suggest that this underexpression may specificities and sensitivities as sperm biomarkers of male infertility. affect HSPA2 activity, but not its expression or localisation, but more Therefore, these proteins should be further studied to evaluate their studies are needed to support this hypothesis. role in male infertility, as well as to assess their real potential as diag- HIST1H2BA and SPA17, two testis-specific proteins, were also dif- nostic methods using targeted proteomic approaches and confirma- ferentially expressed in both infertile groups in the shotgun proteomic tory studies. If confirmed, the evaluation of these proteins expression analysis. Validation analyses revealed that HIST1H2BA is significantly could be added to semen analysis as an additional diagnostic method overexpressed in infertile men. HIST1H2BA is a testis-specific histone of male infertility. expressed in meiotic and post-meiotic cell types (Sun & Qi, 2014). It is The most important limitation of this study was the small sam- required specifically during the replacement of histones by protamines ple size and the pooling of samples. In the Western blotting analysis, INTASQUI et al. | 9 of 11 because sperm concentration in some samples was low and, thus, Agarwal, A., Sharma, R., Durairajanayagam, D., Ayaz, A., Cui, Z., Willard, samples had to be pooled, we could not detect all inter-individual dif- B., … Sabanegh, E. (2015). Major protein alterations in spermatozoa from infertile men with unilateral varicocele. Reproductive Biology and ferences within the groups. To address this issue, we ran individual Endocrinology, 13, 8. samples as well. Additionally, we only had a limited number of samples Agarwal, A., Sharma, R., Durairajanayagam, D., Cui, Z., Ayaz, A., Gupta, S., from infertile men with varicocele. In the immunocytochemistry anal- … Sabanegh, E. (2015). Differential proteomic profiling of spermatozoal ysis, the small sample size may have impaired our ability to identify proteins of infertile men with unilateral or bilateral varicocele. Urology, 85, 580–588. minor differences. Finally, in the shotgun proteomic analysis, several Agarwal, A., Sharma, R., Durairajanayagam, D., Cui, Z., Ayaz, A., Gupta, S., … proteins were differentially expressed between the groups, and each Sabanegh, E. (2016). Spermatozoa protein alterations in infertile men one of them could have been utilised for validation. We chose the pro- with bilateral varicocele. Asian Journal of Andrology, 18, 43–53. posed proteins based on their involvement with protein turnover and Agarwal, A., Sharma, R., Samanta, L., Durairajanayagam, D., & Sabanegh, E. (2016). Proteomic signatures of infertile men with clinical varicocele sperm function, which suggests that these proteins could be more im- and their validation studies reveal mitochondrial dysfunction leading to portant for infertility than the others. However, it is highly likely that infertility. Asian Journal of Andrology, 18, 282–291. we may still have missed some other important proteins that could Amaral, A., Paiva, C., Attardo Parrinello, C., Estanyol, J. M., Ballesca, J. L., be validated, such as proteins involved in proteasomal and lysosomal Ramalho-Santos, J., & Oliva, R. (2014). Identification of proteins in- pathways. The validated proteins, although with altered expression in volved in human sperm motility using high-throughput differential proteomics. Journal of Proteome Research, 13, 5670–5684. the infertile groups, cannot be related to a specific defect because of Ashrafzadeh, A., Karsani, S. A., & Nathan, S. (2013). Mammalian sperm the heterogeneity of the infertility groups. fertility related proteins. International Journal of Medical Sciences, 10, 1649–1657. Ayaz, A., Agarwal, A., Sharma, R., Arafa, M., Elbardisi, H., & Cui, Z. (2015). Impact of precise modulation of reactive oxygen species levels on sper- 5 | CONCLUSION matozoa proteins in infertile men. Clinical Proteomics, 12, 4. Bajpai, M., & Doncel, G. F. (2003). Involvement of tyrosine kinase and In conclusion, the sperm proteome of infertile men with primary or cAMP-dependent kinase cross-talk in the regulation of human sperm secondary infertility differs from that of proven fertile men. PTMs and motility. Reproduction, 126, 183–195. Baker, M. A. (2016). Proteomics of post-translational modifications of protein folding are enriched by the underexpressed proteins in the pri- mammalian spermatozoa. Cell and Tissue Research, 363, 279–287. mary and secondary infertility groups, suggesting that they may play Baker, M. A., Hetherington, L., Weinberg, A., Naumovski, N., Velkov, a role in male infertility. After confirmatory studies, we suggest that T., Pelzing, M., … Aitken, R. J. (2012). Analysis of phosphopeptide BAG6 and HIST1H2BA proteins may be potential biomarkers of male changes as spermatozoa acquire functional competence in the epididymis demonstrates changes in the post-translational modification of infertility, irrespective of its cause. Izumo1. Journal of Proteome Research, 11, 5252–5264. Battistone, M. A., Alvau, A., Salicioni, A. M., Visconti, P. E., Da Ros, V. G., & Cuasnicu, P. S. (2014). Evidence for the involvement of proline-rich ACKNOWLEDGEMENTS tyrosine kinase 2 in tyrosine phosphorylation downstream of protein The authors acknowledge Ms. Banu Gopalan and Vathsalya Senapathi kinase A activation during human sperm capacitation. Molecular Human Reproduction, 20, 1054–1066. for the bioinformatic annotation analysis on this dataset and Mr. John Beckmann, R. P., Mizzen, L. E., & Welch, W. J. (1990). Interaction of Hsp 70 Peterson and Ms. Gauravi Deshpande for the confocal analysis. with newly synthesized proteins: Implications for protein folding and assembly. Science, 248, 850–854. Bromfield, E., Aitken, R. J., & Nixon, B. (2015). Novel characterization of CONFLICT OF INTEREST the HSPA2-stabilizing protein BAG6 in human spermatozoa. Molecular Human Reproduction, 21, 755–769. None. Brugh, V. M. 3rd, & Lipshultz, L. I. (2004). Male factor infertility: Evaluation and management. The Medical Clinics of North America, 88, 367–385. Carr, D. W., Fujita, A., Stentz, C. L., Liberty, G. A., Olson, G. E., & Narumiya, ORCID S. (2001). Identification of sperm-specific proteins that interact with A-kinase anchoring proteins in a manner similar to the type II reg- P. Intasqui http://orcid.org/0000-0002-8494-6009 ulatory subunit of PKA. The Journal of Biological Chemistry, 276, A. Agarwal http://orcid.org/0000-0003-0585-1026 17332–17338. Cedenho, A. P., Lima, S. B., Cenedeze, M. A., Spaine, D. M., Ortiz, V., & R. Sharma http://orcid.org/0000-0002-9664-6978 Oehninger, S. (2006). Oligozoospermia and heat-shock protein expres- L. Samanta http://orcid.org/0000-0002-2969-0071 sion in ejaculated spermatozoa. Human Reproduction, 21, 1791–1794. Chen, X., Xu, W., Miao, M., Zhu, Z., Dai, J., Chen, Z., … Shi, H. (2015). R. P. Bertolla http://orcid.org/0000-0001-6892-9244 Alteration of sperm protein profile induced by cigarette smoking. Acta Biochimica et Biophysica Sinica, 47, 504–515. Chirico, W. J., Waters, M. G., & Blobel, G. (1988). 70K heat shock related REFERENCES proteins stimulate protein translocation into microsomes. Nature, 332, 805–810. Abu, D. A., Franken, D. R., Hoffman, B., & Henkel, R. (2012). Sequential Chiriva-Internati, M., Gagliano, N., Donetti, E., Costa, F., Grizzi, F., analysis of sperm functional aspects involved in fertilisation: A pilot Franceschini, B., … Kast, W. M. (2009). Sperm protein 17 is expressed study. Andrologia, 44(Suppl 1), 175–181. in the sperm fibrous sheath. Journal of Translational Medicine, 7, 61. 10 of 11 | INTASQUI et al.

Colledge, M., & Scott, J. D. (1999). AKAPs: From structure to function. obesity-associated proteomic changes in human spermatozoa by differ- Trends in Cell Biology, 9, 216–221. ence gel electrophoresis. Reproductive Biomedicine Online, 19, 660–670. Dix, D. J., Allen, J. W., Collins, B. W., Mori, C., Nakamura, N., Poorman-Allen, P., Lea, I. A., Richardson, R. T., Widgren, E. E., & O’Rand, M. G. (1996). Cloning … Eddy, E. M. (1996). Targeted gene disruption of Hsp70-2 results in failed and sequencing of cDNAs encoding the human sperm protein, Sp17. meiosis, germ cell apoptosis, and male infertility. Proceedings of the National Biochimica et Biophysica Acta, 1307, 263–266. Academy of Sciences of the United States of America, 93, 3264–3268. Lea, I. A., Widgren, E. E., & O’Rand, M. G. (2004). Association of sperm Dix, D. J., Allen, J. W., Collins, B. W., Poorman-Allen, P., Mori, C., Blizard, D. protein 17 with A-kinase anchoring protein 3 in flagella. Reproductive R., … Eddy, E. M. (1997). HSP70-2 is required for desynapsis of syn- Biology and Endocrinology, 2, 57. aptonemal complexes during meiotic prophase in juvenile and adult Legare, C., Droit, A., Fournier, F., Bourassa, S., Force, A., Cloutier, F., … mouse spermatocytes. Development, 124, 4595–4603. Sullivan, R. (2014). Investigation of male infertility using quantitative Ficarro, S., Chertihin, O., Westbrook, V. A., White, F., Jayes, F., Kalab, P., … comparative proteomics. Journal of Proteome Research, 13, 5403–5414. Visconti, P. E. (2003). Phosphoproteome analysis of capacitated human Lima, S. B., Cenedeze, M. A., Bertolla, R. P., Filho, P. A., Oehninger, S., & sperm. Evidence of tyrosine phosphorylation of a kinase-anchoring Cedenho, A. P. (2006). Expression of the HSPA2 gene in ejaculated protein 3 and valosin-containing protein/p97 during capacitation. The spermatozoa from adolescents with and without varicocele. Fertility Journal of Biological Chemistry, 278, 11579–11589. and Sterility, 86, 1659–1663. Frapsauce, C., Pionneau, C., Bouley, J., Delarouziere, V., Berthaut, I., Ravel, Liu, Y., Guo, Y., Song, N., Fan, Y., Li, K., Teng, X., … Ding, Z. (2015). Proteomic C., … Mandelbaum, J. (2014). Proteomic identification of target proteins pattern changes associated with obesity-induced asthenozoospermia. in normal but nonfertilizing sperm. Fertility and Sterility, 102, 372–380. Andrology, 3, 247–259. Gokce, E., Shuford, C. M., Franck, W. L., Dean, R. A., & Muddiman, D. C. Martinez-Heredia, J., de Mateo, S., Vidal-Taboada, J. M., Ballesca, J. L., & (2011). Evaluation of normalization methods on GeLC-MS/MS label- Oliva, R. (2008). Identification of proteomic differences in asthenozo- free spectral counting data to correct for variation during proteomic ospermic sperm samples. Human Reproduction, 23, 783–791. workflows. Journal of the American Society for Mass Spectrometry, 22, Maselli, J., Hales, B. F., Chan, P., & Robaire, B. (2012). Exposure to bleomy- 2199–2208. cin, etoposide, and cis-platinum alters rat sperm chromatin integrity and Govin, J., Caron, C., Escoffier, E., Ferro, M., Kuhn, L., Rousseaux, S., … sperm head protein profile. Biology of Reproduction, 86, 166, 161-110. Khochbin, S. (2006). Post-meiotic shifts in HSPA2/HSP70.2 chaper- Maselli, J., Hales, B. F., & Robaire, B. (2013). The effects of chemother- one activity during mouse spermatogenesis. The Journal of Biological apy with bleomycin, etoposide, and cis-platinum (BEP) on rat sperm Chemistry, 281, 37888–37892. chromatin remodeling, fecundity and testicular gene expression in the Grizzi, F., Chiriva-Internati, M., Franceschini, B., Hermonat, P. L., Soda, progeny. Biology of Reproduction, 89, 85. G., Lim, S. H., & Dioguardi, N. (2003). Immunolocalization of sperm de Mateo, S., Martinez-Heredia, J., Estanyol, J. M., Dominguez-Fandos, D., protein 17 in human testis and ejaculated spermatozoa. Journal of Vidal-Taboada, J. M., Ballesca, J. L., & Oliva, R. (2007). Marked correla- Histochemistry and Cytochemistry, 51, 1245–1248. tions in protein expression identified by proteomic analysis of human Hashemitabar, M., Sabbagh, S., Orazizadeh, M., Ghadiri, A., & Bahmanzadeh, spermatozoa. Proteomics, 7, 4264–4277. M. (2015). A proteomic analysis on human sperm tail: Comparison be- McReynolds, S., Dzieciatkowska, M., Stevens, J., Hansen, K. C., Schoolcraft, tween normozoospermia and asthenozoospermia. Journal of Assisted W. B., & Katz-Jaffe, M. G. (2014). Toward the identification of a subset Reproduction and Genetics, 32, 853–863. of unexplained infertility: A sperm proteomic approach. Fertility and Hassan, M. A., & Killick, S. R. (2004). Negative lifestyle is associated with Sterility, 102, 692–699. a significant reduction in fecundity. Fertility and Sterility, 81, 384–392. Monga, M., Bernie, J., & Rajasekaran, M. (1999). Male infertility and erectile Hosseinifar, H., Gourabi, H., Salekdeh, G. H., Alikhani, M., Mirshahvaladi, S., dysfunction in spinal cord injury: A review. Archives of Physical Medicine Sabbaghian, M., … Gilani, M. A. (2013). Study of sperm protein profile and Rehabilitation, 80, 1331–1339. in men with and without varicocele using two-dimensional gel electro- O’Rand, M. G., Widgren, E. E., & Fisher, S. J. (1988). Characterization of the phoresis. Urology, 81, 293–300. rabbit sperm membrane autoantigen, RSA, as a lectin-like zona binding Huszar, G., Ozenci, C. C., Cayli, S., Zavaczki, Z., Hansch, E., & Vigue, L. protein. Developmental Biology, 129, 231–240. (2003). Hyaluronic acid binding by human sperm indicates cellular ma- Parte, P. P., Rao, P., Redij, S., Lobo, V., D’Souza, S. J., Gajbhiye, R., & Kulkarni, turity, viability, and unreacted acrosomal status. Fertility and Sterility, V. (2012). Sperm phosphoproteome profiling by ultra performance liq- 79(Suppl 3), 1616–1624. uid chromatography followed by data independent analysis (LC-MS(E)) Huszar, G., Stone, K., Dix, D., & Vigue, L. (2000). Putative creatine kinase M- reveals altered proteomic signatures in asthenozoospermia. Journal of isoform in human sperm is identified as the 70-kilodalton heat shock Proteomics, 75, 5861–5871. protein HspA2. Biology of Reproduction, 63, 925–932. Pixton, K. L., Deeks, E. D., Flesch, F. M., Moseley, F. L., Bjorndahl, L., Ashton, Intasqui, P., Camargo, M., Del Giudice, P. T., Spaine, D. M., Carvalho, V. P. R., … Brewis, I. A. (2004). Sperm proteome mapping of a patient who M., Cardozo, K. H., … Bertolla, R. P. (2013). Unraveling the sperm pro- experienced failed fertilization at IVF reveals altered expression of at teome and post-genomic pathways associated with sperm nuclear least 20 proteins compared with fertile donors: Case report. Human DNA fragmentation. Journal of Assisted Reproduction and Genetics, 30, Reproduction, 19, 1438–1447. 1187–1202. Rahman, M. S., Lee, J. S., Kwon, W. S., & Pang, M. G. (2013). Sperm pro- Jensen, T. K., Andersson, A. M., Jorgensen, N., Andersen, A. G., Carlsen, E., teomics: Road to male fertility and contraception. International Journal Petersen, J. H., & Skakkebaek, N. E. (2004). Body mass index in rela- of Endocrinology, 2013, 360986. tion to semen quality and reproductive hormones among 1,558 Danish Redgrove, K. A., Anderson, A. L., McLaughlin, E. A., O’Bryan, M. K., Aitken, men. Fertility and Sterility, 82, 863–870. R. J., & Nixon, B. (2013). Investigation of the mechanisms by which the Kawahara, H., Minami, R., & Yokota, N. (2013). BAG6/BAT3: Emerging roles molecular chaperone HSPA2 regulates the expression of sperm sur- in quality control for nascent polypeptides. Journal of Biochemistry, 153, face receptors involved in human sperm-oocyte recognition. Molecular 147–160. Human Reproduction, 19, 120–135. Kovac, J. R., Pastuszak, A. W., & Lamb, D. J. (2013). The use of genomics, Redgrove, K. A., Nixon, B., Baker, M. A., Hetherington, L., Baker, G., Liu, proteomics, and metabolomics in identifying biomarkers of male infer- D. Y., & Aitken, R. J. (2012). The molecular chaperone HSPA2 plays a tility. Fertility and Sterility, 99, 998–1007. key role in regulating the expression of sperm surface receptors that Kriegel, T. M., Heidenreich, F., Kettner, K., Pursche, T., Hoflack, B., mediate sperm-egg recognition. PLoS ONE, 7, http://journals.plos.org/ Grunewald, S., … Paasch, U. (2009). Identification of diabetes- and plosone/article?id=10.1371/journal.pone.0050851. INTASQUI et al. | 11 of 11

Richardson, R. T., Yamasaki, N., & O’Rand, M. G. (1994). Sequence of a rab- Valastyan, J. S., & Lindquist, S. (2014). Mechanisms of protein-folding dis- bit sperm zona pellucida binding protein and localization during the eases at a glance. Disease Models & Mechanisms, 7, 9–14. acrosome reaction. Developmental Biology, 165, 688–701. Visconti, P. E., Bailey, J. L., Moore, G. D., Pan, D., Olds-Clarke, P., & Kopf, Sasaki, T., Marcon, E., McQuire, T., Arai, Y., Moens, P. B., & Okada, H. G. S. (1995). Capacitation of mouse spermatozoa. I. Correlation be- (2008). Bat3 deficiency accelerates the degradation of Hsp70-2/ tween the capacitation state and protein tyrosine phosphorylation. HspA2 during spermatogenesis. The Journal of Cell Biology, 182, Development, 121, 1129–1137. 449–458. Wen, Y., Richardson, R. T., & O’Rand, M. G. (1999). Processing of the sperm Shamsi, M. B., Kumar, K., & Dada, R. (2011). Genetic and epigenetic fac- protein Sp17 during the acrosome reaction and characterization as a tors: Role in male infertility. Indian Journal of Urology: IJU: Journal of the calmodulin binding protein. Developmental Biology, 206, 113–122. Urological Society of India, 27, 110–120. World Health Organization. (2010). WHO laboratory manual for the exam- Shamsi, M. B., Venkatesh, S., Tanwar, M., Singh, G., Mukherjee, S., Malhotra, ination and processing of human semen. Geneva, Switzerland: World N., … Dada, R. (2010). Comet assay: A prognostic tool for DNA integrity Health Organization. assessment in infertile men opting for assisted reproduction. The Indian Xu, W., Hu, H., Wang, Z., Chen, X., Yang, F., Zhu, Z., … Qiao, Z. (2012). Journal of Medical Research, 131, 675–681. Proteomic characteristics of spermatozoa in normozoospermic pa- Sharma, R., Agarwal, A., Mohanty, G., Hamada, A. J., Gopalan, B., Willard, tients with infertility. Journal of Proteomics, 75, 5426–5436. B., … du Plessis, S. (2013). Proteomic analysis of human spermatozoa Zhao, C., Huo, R., Wang, F. Q., Lin, M., Zhou, Z. M., & Sha, J. H. (2007). proteins with oxidative stress. Reproductive Biology and Endocrinology, Identification of several proteins involved in regulation of sperm motil- 11, 48. ity by proteomic analysis. Fertility and Sterility, 87, 436–438. Shen, S., Wang, J., Liang, J., & He, D. (2013). Comparative proteomic study between human normal motility sperm and idiopathic asthenozoospermia. World Journal of Urology, 31, 1395–1401. SUPPORTING INFORMATION Siva, A. B., Kameshwari, D. B., Singh, V., Pavani, K., Sundaram, C. S., Rangaraj, N., … Shivaji, S. (2010). Proteomics-based study on asthe- Additional Supporting Information may be found online in the nozoospermia: Differential expression of proteasome alpha complex. supporting information tab for this article. Molecular Human Reproduction, 16, 452–462. Sun, R., & Qi, H. (2014). Dynamic expression of combinatorial replication- dependent histone variant genes during mouse spermatogenesis. Gene How to cite this article: Intasqui P, Agarwal A, Sharma R, Expression Patterns, 14, 30–41. Thacker, S., Yadav, S. P., Sharma, R. K., Kashou, A., Willard, B., Zhang, D., Samanta L, Bertolla RP. Towards the identification of reliable & Agarwal, A. (2011). Evaluation of sperm proteins in infertile men: A sperm biomarkers for male infertility: A sperm proteomic proteomic approach. Fertility and Sterility, 95, 2745–2748. approach. Andrologia. 2017;e12919. https://doi.org/10.1111/ UniProt AC Q96A08. (1992). The influence of varicocele on parameters of and.12919 fertility in a large group of men presenting to infertility clinics. World Health Organization. Fertility and Sterility, 57, 1289–1293.