Isolation and identification of proteins from swine sperm chromatin and nuclear matrix

The aim of this study was to perform a proteomic analysis to isolate and identify proteins from the swine sperm nuclear matrix to contribute to a database of swine sperm nuclear proteins. We used prechilled diluted semen from seven boars (19 to 24 weekold) from the commercial line Landrace x Large White x Pietran. The semen was processed to separate the sperm heads and extract the chromatin and nuclear matrix for protein quantification and analysis by mass spectrometry, by LTQ Orbitrap ELITE mass spectrometer (Thermo-Finnigan) coupled to a nanoflow chromatography system (LC-MS/MS). We identified 222 different proteins in the sample; a total of 159 (71.6%) were previously described as present in the somatic or sperm nuclei of other species, 41 (18.5%) did not have a previously reported nuclear presence and 22 (9.9%) had not been characterized. The most abundant family of proteins corresponded to ribosomal (13.1%), followed by cytoskeleton (12.2%), uncharacterized (9.9%), histones (5.4%), proteasome subunits (3.6%) and heat shock (1.8%). The other proteins clustered in other families accounted for 54% of the total proteins. The protein isolation of the nuclear matrix of the swine spermatozoa was satisfactory, thus demonstrating that the protocol used was efficient. Several proteins were identified and described. However, it was not possible to identify some protein structures. Therefore, this study helps to establish a starting point for future proteomic studies comparing fertile and sub-fertile animals.


Introduction
It has long been thought that the only function of sperm cells is to transmit the paternal genomic DNA to the next generation.This idea was challenged by the discovery of the imprinting of sex-specific genes mediated by DNA methylation differences (set during gametogenesis) that were epigenetically transmitted to the next generation (Oliva and Ballescá, 2012).
DNA condensation by sperm protamines leaves only a small fraction of the sperm genome accessible for DNA binding proteins, which are necessary to enable DNA replication and genes transcription.These sites may be the most important sites for the initiation of paternal genome functions in the early embryo (Yamauchi et al., 2011).According to the same authors, these active sperm chromatin sites in protamine toroids may contain important epigenetic information for the developing embryo.
The isolated use of genomic and transcriptomic information may be insufficient to fully understand a complex organism because proteomics and transcriptomics can be discordant and DNA-RNA relationships cannot be fully correlated.Thus, measurements of other metabolic levels should also be obtained, such as the study of proteins (Wright et al., 2012).According to these same authors, large-scale protein research in organisms (i.e., the proteome-protein complement expressed by a genome) is equally important because it provides information about the real factors (i.e., enzymes) involved in the metabolic process.However, unlike other areas (i.e., genomics and transcriptomics), proteomics and its present techniques and strategies are still under development.
Proteomics projects related to studies of nuclear proteins in sperm have enabled the creation of catalogs.However, DeMateo et al. (2011) related that only small subsets of the identified proteins are nuclear proteins.The aim of this study was to perform a proteomic analysis to isolate and identify proteins from the swine sperm nuclear matrix to contribute to a database of swine sperm nuclear proteins.

Semen and sperm processing
Prediluted (diluent BTS) and cooled boar semen was used in this study, with concentration of 2.5 x 10 9 (80 ml/dose).The doses were stored in thermal chamber between 15 and 18°C.The semen was provided by an artificial insemination center located in Uberlandia, Minas Gerais, Brazil.We used semen from seven boars (19 to 24 week-old) from the commercial line Landrace x Large White x Pietran.The period between the collection and analysis ranges from 24 to 48 h.The breeders chosen were normally used by insemination center, where the resulting litters, borned to the date of collection, were within zootechnical levels proposed for the line in question.
Based on routine testing using the CASA system (computerized analysis of sperm), the boars presented semen with 84.63 to 93.54% of motility and 58.78 to 84.11% of progressive motility.We use the standard method for evaluation of sperm morphology, which consisted of 200 cell count at 1000X magnification in phase contrast microscopy with immersion oil.It was observed defects of head, tail and curled tail, acrosome, midpiece, colon and proximal and distal citoplasmic drop.The rate of morphological defects (apart from the distal cytoplasmic drop) was 2.5 to 10.5%.

Analysis of chromatin alterations using semen smears stained with toluidine blue
All semen was evaluated by the toluidine blue technique (Beletti et al., 2005) to determine whether the boars in question had normal sperm with respect to condensation of chromatin and the morphology of the sperm head.
The samples of boar semen were fixed using formalin citrate (2 drops of semen to 1 ml of formaldehyde citrate).A drop of sample was fixed in two smears and subsequently air dried at room temperature.These smears were subjected to acid hydrolysis in 4N hydrochloric acid for 20 min and washed in distilled water.After drying, the smears were stained with a drop of 0.025% toluidine blue (pH 4.0) in phosphate-citric acid buffer (McIlvaine buffer) on the slide, followed by placement of the coverslip.After three minutes, 50 digital images in grayscale were captured from each slide using a Leica DM500 optical microscope coupled with a Leica ICC 50 camera (Wetzlar, Germany) with an oil immersion lens at 100X magnification.Digital images were used to segment by thresholding 100 sperm heads from each slide.
The samples were analyzed using routines developed in the SCILAB environment to obtain the mean and standard deviation of the pixel values within the head of each image.To obtain a reference for the normal color of the sperm head, six sperm heads that were the most homogeneous and had a lighter color (i.e., the sperm was more homogeneous and intensely compacted) were automatically selected in each smear.The average pixel values of these heads were used as the reference value for the normal staining of the sperm (standard head).Then, the differences between the average values of the standard heads and the average values of each head examined were determined for each image.This difference was transformed into a percentage (% unpacking) based on the average value of the standard heads.The coefficient of variation (heterogeneity %) of the gray levels was also calculated (Beletti et al., 2005).

Sperm head segregation
The methodology used to segregate the sperm heads was modified from a previous study (Morandi-Filho, 2013).Each semen sample (4 ml) cooled to 2 to 8 degrees Celsius (C) was placed in 15 ml conical bottom tubes containing 8 ml of buffer (50 mM Tris-HCl, pH 7.5, and 1 mM EDTA).The flask was homogenized and centrifuged at 750 x g for 15 min at 4°C, followed by removal of the supernatant.The pellet was resuspended in 8 ml of the same buffer, homogenized and centrifuged again.This procedure was repeated three times.
After the third centrifugation, the pellet was resuspended in 1.5 ml of the same buffer.The material was sonicated on ice for 10 min with 30 s pulses and intervals of 5 s.Subsequently, the material was centrifuged at 1000 x g for 15 min at 4°C, the supernatant was removed, and 2 ml of buffer (50 mM Tris-HCl and 1.1 M saccharose, pH 7.5) was added.
Part of this material was diluted 1:100 in distilled water for counting in a Neubauer chamber to measure the concentration of the heads in the sample.Then, the concentration was adjusted to 1 x 10 7 head/ml using 50 mM Tris-HCl buffer with 1.1 M saccharose (pH 7.5).
The heads were isolated from the tails using ultracentrifugation at 75,600 x g for 45 min at 4°C in a gradient consisting of 2 ml of cesium chloride (2.82 M cesium, 25 mM Tris-EDTA, 5 mM MgCl2 and 0.5% Triton X-100) at the bottom of a 12 ml ultracentrifuge tube that was overlaid with 4 ml of 2.2 M saccharose and covered with 2 ml of the sample in 50 mM Tris-HCl and 1.1 M saccharose (pH 7.5).After centrifugation, the supernatant containing the tails was carefully removed by pipetting.The bottom sediment was resuspended in 25 mM Tris buffer and washed three times by centrifugation at 1000 x g for 30 min at 4°C in 25 mM Tris buffer to remove the excess cesium chloride.After this process, a smear was made of a drop of the sample.The smear was dried in an oven for 15 min, stained with xylidine for 15 min and washed with distilled water to evaluate the purity of the sample relative to the absence of tails.The purity was approximately 95% according to visual evaluation by light microscopy based on counting 100 cells in a field.

Extraction of chromatin and the nuclear matrix
The extraction of chromatin and the nuclear matrix followed the methodology adapted from Codrington et al. (2007).The isolated heads were resuspended in 500 μl of a solution containing 1% Triton X-100, 50 mM Tris-HCl (pH 7.5), 1 mM EDTA, and 5 μl of protease inhibitor cocktail (Sigma Aldrich P8340/011M4000) and vortexed for 10 min at room temperature.This treatment removed the acrosome and all membranes while leaving the nucleus condensed and connected to the nuclear envelope and some vestige of the perinuclear material.
The samples were washed three times by centrifugation at 1100 x g for 30 min with 1.5 ml of 50 mM Tris-HCl (pH 7.5).After the last wash, the material was resuspended in 500 µl of decondensation buffer consisting of 40 mM 1,4-dithiothreitol (DTT), 0.25 M (NH 4 ) 2 S 4 , 25 mM Tris-HCl (pH 7.5), and 5 μl of a protease inhibitor cocktail and incubated for 40 min at room temperature.Then, 4000U of RNAse-free deoxyribonuclease I was added and the samples were homogenized for 60 min under a vortex at room temperature.Finally, the samples was frozen, lyophilized and stored in a freezer prior to processing for mass spectrometry.Anim.Reprod., v.14, n.2, p.418-428, Apr./Jun.2017

Protein quantification
At this point the lyophilized samples were mixed, and from here they were processed as single sample.Initially, the lyophilized sample was resuspended in 100 µl of 0.1 M Tris-HCl buffer (pH 8.8) containing 8 M urea.The Bradford method (Bradford, 1976) with the Protein Assay Dye Concentrate Reagent (Bio-Rad, Hercules, California, USA) was used for protein quantification.The standard curve was performed using different dilutions of bovine serum albumin prepared from a commercially acquired stock (200 mg/ml protein standard, Sigma, St. Louis, Missouri, USA).The sample was distributed in triplicate in microplates.The absorbance at 595 nm was read in a spectrophotometer (Molecular Devices, SpectraMax Plus 384).The quantification of protein by the Bradford method indicated concentration of 3.4 mg/ml.

Sample preparation
The sample preparation for mass spectrometry consisted of three main steps: i) reduction and alkylation of proteins, ii) enzymatic digestion of the proteins with trypsin and iii) clean up/desalting of the samples.We used 38 µl of each sample (50 µg).Briefly, the sample was subjected to the reduction of disulfide bonds of the protein by the addition of DTT (dithiothreitol) in a proportion of 1 mg DTT/mg protein and incubated for 2 h at room temperature.Then, alkylating IA (iodoacetamide) was added in a proportion of 3 mg IA/mg of protein and incubated for 1 h at room temperature in the dark.The volume of the sample was diluted 5-fold in a 0.1 M solution of ammonium bicarbonate (pH ≥ 8,0) to obtain a final volume of 500 µl.The sample was incubated with 1 µg of trypsin (Promega, Madison, Wiscousin, USA) at 37°C overnight.Prior to application of the sample into the mass spectrometer, clean-up/desalting of the sample was performed using the OASIS HLB cartridge 1 cc column according to the manufacturer's instructions.The column was equilibrated with a 5% acetonitrile solution containing 0.1% formic acid, and elution of the material of interest was performed with 80% acetonitrile.The sample was dried in a speed vac and applied to a mass spectrometer.

Mass spectrometry analysis
The digested sample was dried and analyzed in the LTQ Orbitrap ELITE mass spectrometer (Thermo-Finnigan) coupled to a nanoflow chromatography system (LC-MS/MS).The acquired data were automatically processed by the Computational Proteomics Analysis System (CPAS; Rauch, 2006).The identified peptides were grouped into proteins using the algorithm Protein Prophet, and a list of identifications with error rates less than 2.0% was created.A general database of all species was used (Uniprot, 2016).

Statistical analysis
Descriptive statistical analyses were performed on the presented data.

Toluidine blue method
The toluidine blue method was used to analyze the chromatin in 195 sperm heads of each boar.The averages were: unpacking chromatin (%) 2.46 ± 1.73 and heterogeneity of chromatin (%) 4.49 ± 0.94.

Proteins found
In the mass spectrometry analysis, 222 different proteins were identified in the sample (Table 1); a total of 159 of these (71.6%) were previously described as being present in the somatic or sperm nuclei of other species (Uniprot, 2016), 41 (18.5 %) had no previously described nuclear presence and 22 (9.9%) were uncharacterized.

Discussion
The results of toluidine blue analyses suggest that the animals in question presented normal sperm with respect to chromatin compaction (Beletti et al., 2005).
According to DeMateo et al. (2011) the proteomic analyses of human sperm nuclei revealed that the most abundant proteins were the histone family (9.7%), followed by cytoskeletal proteins (cytokeratins, tubulin and tektinas, 8.6%), ribosomal proteins (6.7%), proteasome subunits (6.2%), uncharacterized proteins (6.2%), spanx proteins (1.7%), and heat shock proteins (1.2%).Notably DeMateo et al. (2011) studied the proteins of human sperm nuclei, whereas this study focused on the proteins of the swine sperm nuclear matrix.Similarities between the identified families can be observed, although there is little similarity between the described proportions.
Notably LOC100626209 protein had the highest number of peptides (218) in this study, followed by FAM71B (140 peptides).The third most detected (105 peptides) was protamine 2. The LOC100626209 protein was classified in this study as an uncharacterized protein and had not previously been described in any species, and thus we cannot speculate as to its possible actions.Lemos (2013) described uncharacterized proteins with high molecular weights in the bovine nuclear annulus, which may indicate that the LOC100626209 protein is present in this structure.The second protein FAM71B was identified in the human sperm nucleus and might be involved in RNA biogenesis (Van Koningsbruggen et al., 2007).This protein was identified for the first time in swine sperm nuclear matrix.
Interestingly, the third protein in terms of the number of peptides found was protamine 2. Protamine 2 was described as being absent in the swine sperm nucleus, whereas protamine 1 was identified in various mammalian species, including swine (Pirhonen et al., 1994;Andrabi, 2007).Although some studies did not identify protamine 2, this protein was reported to be transcribed and translated in swine at low levels.However, an 8 amino acid deletion occurred at the amino acid terminus of the molecule and probably had functional relevance (Maier et al., 1990).
The absence of protamine 1 in the sample of swine sperm chromatin also occurred in the evaluation of the human sperm nucleus (DeMateo et al., 2011).The same authors reported that protamine 2 was present among the basic nuclear proteins, as was demonstrated in this study.These authors also suggested a possible explanation for the lack of protamine 1 based on its particular amino acid composition.Protamine is a basic protein that is highly enriched in lysine and arginine (more than 50% in the mature form).Trypsin cleaves peptide chains primarily on the carboxyl side of lysine or arginine, which results in very small peptide fragments that cannot be detected under the conditions used in mass spectrometry.This limitation is particularly important for protamine 1 because it is more enriched in arginine than protamine 2.
Genes for two protamines (PRM1 and PRM2) and two transition proteins (TNP1 and TNP2) have been characterized in several mammalian species (Engel et al., 1992).According to the same authors, the human, swine and bull genes for PRM1, PRM2 and TNP2 are closely connected along a specific stretch of DNA, whereas the gene for TNP1 in all of the species studied is located on another chromosome.In this study we detected just genes for PRM2 and TNP2, but we also detected a peptide of the CHTOP protein, which is responsible for the chromatin marking of protamine 1.This protein is also involved in transcriptional regulation.It contains a region rich in glycine and arginine that interacts with the RNA or DNA directly or Anim.Reprod., v.14, n.2, p.418-428, Apr./Jun.2017 in combination with other nucleotide binding proteins (Takai et al., 2014).
Proteins related to the ribonucleosome were identified in the sample.The 40S ribosomal proteins RPS25, RPS3A and RPS6 were also reported in the human sperm nuclear proteome by DeMateo et al. (2011).Some 60S ribosomal proteins were also found in the human sperm nucleus, as in this study.The presence of RPL9 protein in human sperm nuclei was confirmed using an immunofluorescence technique (DeMateo et al., 2011).
The proteomic analysis of isolated sperm nuclei indicated the presence of only the cytoplasmic ribosomal proteins (80S (60S + 40S)) and not the mitochondrial ribosomal proteins (55S).Thus, the detection of cytoplasmic ribosomal proteins in this study was consistent with the cytoplasmic translation proposed by some authors (Lambard et al., 2004;Galeraud-Denis et al., 2007;DeMateo et al., 2011The ribosomal proteins are recognized as cytoplasmic proteins.Thus, their detection in swine sperm nuclei is an important finding and may clarify possible paternal epigenetic actions in swine. The family of cytoskeletal proteins (i.e., keratin, tubulin and actin) that was identified in this study has also been identified in the human sperm nucleus (DeMateo et al., 2011).The same authors reported that some cytoskeletal molecules were demonstrated to participate in the formation of the sperm tail and sperm nucleus format.Specifically, tubulin was also detected in the heads of sperm, suggesting a possible role related to the acrosome reaction.Cytokeratin and actin were also associated with the sperm nuclear matrix in guinea pigs (Ocampo et al., 2005), and now in swine sperm nuclear matrix.
This study identified several histone variants in mature and ejaculated sperm.In this respect, we demonstrated the transport and incorporation of the sperm histones to the zygote, indicating another potential effect of parental epigenetic reprogramming on the zygote after fertilization that is independent of the imprinting state (Miller et al., 2010).Histones were only a small part of the proteins identified in this sperm chromatin sample, indicating that many other proteins can transmit epigenetic information to the zygote.
Epigenetic control of gene expression exists for the activation of DNA methylation because methylation, acetylation and phosphorylation of histones and histone entry into the oocyte leaves a space for DNA and epigenetic signaling based on histones, which can be important for subsequent embryonic development (Miller et al., 2010).The same authors described results from analyses of the composition of soluble (linked to histones) and insoluble (linked to protamines) domains in human and murine sperm and indicated that chromatin was actually the most significant contributor to an epigenetic signal in these cells.
Concurrent with visible changes in the organization of the sperm chromatin, the histones would be removed from the DNA of initial spermatids and spermatocytes and replaced by transition proteins.Subsequently, the transition proteins are replaced by protamines that are responsible for the final condensation and stabilization of the sperm chromatin (D 'Occhio et al., 2007).The same authors reported that histones could persist in ejaculated mature sperm from humans and other mammals, as demonstrated in this study for several types of histones.
The fact that sperm nucleus histones are not necessarily characterized as an error in chromatin compaction has led to a concept change.Beletti (2013) reported that there are regions interspersed between the toroidal structures of sperm chromatin that contain nucleosome sequences, which often contain hypomethylated DNA.These regions may be involved in important functions related to early embryonic development and paternal epigenetic inheritance.
A total of 25 different proteasome subunits were reported in the human sperm nucleus (DeMateo et al., 2011); PSMA4, 5, 6 and 8 were also identified in this study.These proteins are the protein machine for ubiquitin-mediated proteolytic degradation, which has been implicated in many cellular processes including cell cycle progression, transcriptional regulation, signal transduction, and determining the fate of the cell (Zhong and Belote, 2007).
Is interesting to note the identification of proteins involved in the regulation of gene expression, such as PHTF1 protein (Table 1).It was thought that the sperm is an inert transcriptionally cell, with the sole objective is to provide its DNA (packed by protamines) to the oocyte (DeMateo et al., 2011).Thus, the sperm nuclear proteins identified here may be only a remnant of the sperm cell differentiation process, or may be relevant to fertilization or success of embryo development.
For example, the PPP1CC protein, that according to MacLeod et al. (2014) is an essential phosphatase protein in spermatogenesis, was found in the sample, which concurs with a previous result (DeMateo et al., 2011).An important nuclear protein (WBP2NL) found in this study, was also identified in rats by Chen et al. (2014) and may play a role in meiotic resumption and male pronucleus formation, which are related to early embryonic development.The protein glyceraldehyde 3-phosphate dehydrogenase (GAPDH) found in this sample performed functions in glycolysis and participated in nuclear events, including transcription, RNA transport, DNA replication and apoptosis (Applequist et al., 1995).
The heterogeneous nuclear ribonucleoprotein (HNRNPs) is involved in many biological processes, such as cell signaling, DNA repair, and regulation of gene expression and protein (Almeida et al., 2014).The HNRNPC presented 8 peptides in the sample and was previously described in the human sperm nucleus (DeMateo et al., 2011).The heterogeneous nuclear ribonucleoprotein K (HNRNPK) was also found in this sample, and according to Almeida et al. (2014) this protein is predominantly located in the nucleus, where it is involved in multiple steps in gene expression such as transcription, RNA splicing and translation.
Proteins of the nuclear matrix may be involved in different genetic markers that can contribute after fertilization to establish the order of paternal gene reactivation (Oliva, 2006) and may serve as an indication of protein epigenetic function.Therefore, based on the results obtained in this study and the functions of some described proteins, the nuclear matrix, which is composed in part of chromatin proteins, transmits essential mechanisms for the growth and differentiation of the oocyte.Thus, the proteins become not only structural components of the chromatin/matrix architecture but also essential components of successful reproduction.
In this survey, 9.9% of the proteins were classified as uncharacterized (i.e., there are no reports in the literature regarding the description, identification and isolation of these proteins for any species).Thus, future studies can be designed in an attempt to characterize these proteins and elucidate their functions to clarify possible epigenetic paternal inheritance that is beyond the current scientific knowledge.
The set of proteins present in swine sperm chromatin shows that the nuclear matrix plays important roles in the development and maturation of sperm cells.Therefore, it would be important to determine how these protein structures are linked to the establishment of epigenetic functions and how they can affect the embryo development.
In conclusion the protein isolation from the swine sperm nuclear matrix was satisfactory and demonstrated that the protocol was efficient.Some protein families were identified and described.However, it was not possible to identify some protein structures, such as protamine 1.
Therefore, this study contributes to a catalog of protein structures that may be useful in future proteomics studies.The comparison between fertile and sub-fertile animals can contribute to the search for proteomic variations.These studies can improve reproductive technologies and animal breeding of several species by focusing breeders' choices on epigenetic potential as well as high genetic potential.