Head morphometry and chromatin instability in normal boar spermatozoa and in spermatozoa with cytoplasmic droplets

In boar studs, morphological analyses are used to evaluate sperm quality and thereby categorize ejaculates as either approved or rejected. Normally, morphological characteristics correlate with chromatin disorders, but studies to date have only considered the average of abnormalities; cells were not segregated as normal or abnormal. The aim of this study was to assess whether the presence of cytoplasmic droplets was associated with morphometric characteristics and chromatin instability of spermatozoa heads. Morphological analyses were performed on semen from 11 boars using phase contrast microscopy (200 cells per sample). Normal cells were differentiated from those with cytoplasmic droplets and both types were evaluated separately. Photomicrographs were acquired of normal spermatozoa (Group NOR, N = 1,207) as well as spermatozoa with proximal and distal cytoplasmic droplets (Group DROP, N = 725). Sperm-head morphometry and chromatin structure were evaluated using the toluidine blue technique. Spermatozoa heads in the DROP group were longer (8.37 ± 0.60 × 8.31 ± 0.53; P = 0.025), narrower (4.16 ± 0.21 × 4.19 ± 0.19; P = 0.03), and more symmetric on the sides (0.973 ± 0.012 × 0.971 ± 0.011; P = 0.007) than were spermatozoa heads of the NOR group. The DROP group also had a greater average ellipticity (0.335 ± 0.034 × 0.329 ± 0.031; P = 0.0004), a greater percentage of decondensed chromatin (2.71 ± 3.87 × 2.28 ± 1.38; P < 0.0008), and a greater chromatin heterogeneity (4.66 ± 1.40 × 4.40 ± 1.42; P < 0.0001). A greater frequency of semen collection results in a shorter period of cell maturation and this probably affected the degree of chromatin condensation and the cytoplasmic droplet migration, with concomitant effect on the head morphometry measurements. In conclusion, compared with normal spermatozoa, those with cytoplasmic droplets show altered morphometric characteristics, such as longer and narrower spermatozoa heads. They likewise have greater chromatin instability, resulting in a higher percentage of decondensed chromatin and more heterogeneous chromatin.


Introduction
During sperm maturation in the epididymis, cytoplasmic droplets migrate from the head to the tail, and they are released when spermatozoa mature (Cooper and Yeung, 2003;Briz et al., 1995).Generally, in systems with intensive semen collection, immature sperm can be released in the ejaculate because the time for droplet migration is reduced.An important characteristic of such immature cells is the presence of proximal (PCD) and distal (DCD) cytoplasmic droplets in the sperm tail (Flowers, 2004).
The spermatozoa head, the mid-piece, and the tail develop concurrently (Gil et al., 2009); therefore, it has been hypothesized that the presence of cytoplasmic droplets is associated with abnormal head measurements and chromatin instability.Spermatozoa head morphometry has shown that some of these measurements correlate with female reproductive performance (Hirai et al., 2001), with chromatin destabilization (Hingst et al., 1995;Karabinus et al., 1997), and with the presence of abnormalities in the ejaculate (Gaggini et al., 2015).However, few data support this hypothesis, and spermatozoa head morphometry is not routinely performed in the field during semen evaluation.
In order to prevent ejaculates bearing a high percentage of immature cells from being incorporated into artificial insemination doses, morphological tests are performed monthly in ejaculates of all boars or every 50-60 days in boar studs.Ejaculates with more than 10% of cells showing PCD are deemed unsuitable (Feitsma, 2009).A high percentage of DCD is not a cause for rejection, but it also indicates immaturity of sperm (Saravia et al., 2007).Thus, rejection of ejaculates with more than 10% PCD and DCD is a sound approach, because the presence of cytoplasmic droplets is negatively correlated with pregnancy rate and litter size (Waberski et al., 1994).
It is not possible to identify, through morphological tests alone, certain sperm alterations responsible for a reduction in fertilization rate or increased embryonic mortality.Altered chromatin compaction has been studied and is known to be one factor responsible for low male fertility or subfertility in humans (Liu and Backer, 1992;Spano et al., 2000;Bungum et al., 2004) as well as in boars (Boe-Hansen et al., 2008).In bulls (Januskauskas et al., 2003;Khalil, 2004), altered chromatin compaction reduces fertility even when sperm motility and morphology results are considered acceptable (Beletti and Mello, 1996).
While several studies have used samples from bulls and humans (Januskauskas et al., 2003;Bungum et al., 2004;Lucio et al., 2016), few have correlated sperm morphology and chromatin instability in swine (Saravia et al., 2007;Tsakmakidis et al., 2010).Further, many Anim.Reprod., v.14, (Suppl.1),p. 1253-1258. 2017 studies that correlated morphological data with chromatin instability did not distinguish normal cells from those with cytoplasmic droplets, therefore, it is not possible to determine whether the presence of cytoplasmic droplets is related to chromatin instability.Thus, the aim of the study was to assess whether cytoplasmic droplets are associated with altered morphometric characteristics and chromatin instability in boar spermatozoa.

Materials and Methods
Boars from the same genetic line (n = 11, hybrid line -Pietrain x Large White x Landrace -from the same boar stud) were sampled and one ejaculate was obtained from each animal.The average age of the animals on the day of sample collection was 585.95 ± 108.25 days and the average collection interval (defined as the average number of days between the last collection and the day of sample collection) was 5.59 ± 1.80 days.All animals were given ad libitum access to water and fed 2.5 kg/d cornsoybean diet containing 0.55% digestible lysine and 3300 kcal metabolizable energy.Environmental temperature of the boar stud was controlled to a maximum of 22°C.
Semen from the boars was collected using the glove-hand technique (Hancock and Hovel, 1959) in a pre-warmed (36°C) plastic bag equipped with a filter to remove the gel fraction and collect only the rich fraction.Ejaculates were evaluated macroscopically for color, aspect, temperature, volume, and odor, and microscopically by subjective evaluation for sperm motility by the same technician.Only semen samples with a minimum of 85% motile cells were approved.
Semen samples approved in motility assessment were then prepared in Eppendorf tubes using an electronic mixing pipette such that 0.5 mL of semen was diluted with 2 mL of warm (36°C) buffered formalin.These samples were used for morphological examination, morphometric measurements, and chromatin instability evaluation.All samples were analyzed no later than 72 h after semen collection.
The morphological examination was performed in a phase contrast microscope at 100× magnification using an oil immersion lens.Two hundred spermatozoa from each sample were classified according to sperm morphology as normal, or with acrosome defect, abnormal head, neck defect, midpiece defect, folded tail, coiled tail, PCD, or DCD (Pursel et al., 1972).All samples used in the study presented at least one cytoplasmic droplet count.
The morphometric and chromatin instability examinations were performed following the protocol of Beletti et al. (2005).Briefly, one droplet of each sample was smeared onto a glass slide, dried, hydrolyzed with 4N HCl for 15 min, washed three times with distilled water, dried, stained with one droplet of blue toluidine (0.025%, pH 4, in McIlvaine buffer), and covered with a cover slip after one minute.Pictures of spermatozoa were acquired on a microscope (Leica DM500 with Leica ICC50 digital camera).Stained slides were photographed at 100× magnification with oil immersion.Normal cells (NOR group) and cells with cytoplasmic droplets (DROP group) were photographed separately; the pictures of each sample (separate by animals) and each group (NOR and DROP) were saved in different folders such that during analyses it was possible to see each cell from each boar and assess whether the cells were normal or had cytoplasmic droplets.The number of cells used per boar in NOR and DROP groups were defined by the number of cells that had no overlap with other cells or with dirt.
Head segmentation was assessed by a semiautomatic method and was performed on 1,207 heads of the NOR group (average of 109.72 ± 38.74 per boar with minimum value of 61 and maximum value of 162), and 725 heads of the DROP group (average of 68.63 ± 56.99 per boar with minimum value of 7 and maximum value of 161).Heads were evaluated using algorithms developed in Scilab (Beletti et al., 2005).Parameters evaluated were area, perimeter, width, length, ratio of width to length, ellipticity (e), and shape factor (SF). Ellipticity is defined as elongation of the head contour: (length − width) / (length + width), normalized to -1 < e < 1 (Beletti and Costa, 2003).Shape factor is defined as degree of deviation of the head contour from a smooth ellipse: (1−e) ✕ perimeter 2 /4π ✕ area (Beletti and Costa, 2003).Additional parameters were side symmetry (identifies asymmetries along the principal spermatozoa axis) and anteriorposterior symmetry (identifies asymmetry along the secondary spermatozoa axis).These were calculated by flipping the heads along their major and minor axis, filling the voids obtained by the orthogonal lattice representation of the head, and then identifying the intersection between the original and flipped areas (Beletti et al., 2005).
The algorithm also evaluated pixels of each head and selected 20 heads with minor standard deviation from each sample.From these 20 selected heads, the algorithm then selected six with the maximum average of pixel values (faintly stained heads) because those six heads theoretically represented the cells with the most compacted chromatin.The average values of these cells were considered as the default value for all heads of that sample.Next, the algorithm calculated the difference between the default value and the value of each photographed head of the same sample, then converted this difference to percent thereby yielding the percentage of decondensed chromatin.The coefficient of variation for the gray scale intensity value of each head was also calculated, which represents the heterogeneity of chromatin condensation (Beletti et al., 2005).
All analyses were performed with Statistical Analysis System, version 9. 1.3 (SAS Institute, 2005).Head spermatozoa were the experimental unit for the analyses.All parameters were analyzed using the general linear model (GLM procedure) considering individual boar effect and age at collection as covariates.Correlation test (CORR procedure) was used to evaluate relationships between percentage of cytoplasmic droplets in ejaculate (obtained in morphological exam), morphometric values, and 1255 chromatin characteristics.Differences were considered as significant at P < 0.05.

Results
Spermatozoa heads from the NOR group were wider than those in the DROP group (P = 0.030), while DROP spermatozoa heads were longer than NOR heads (P = 0.025).The higher width-to-length ratio of NOR spermatozoa heads (P = 0.005) further confirmed the observation that NOR heads were wider and shorter than DROP heads.Similarly, ellipticity results confirmed that DROP heads were more elongated than NOR heads (P = 0.0004), and shape factor analysis likewise indicated that NOR heads were more similar to an ellipse than were DROP heads (P = 0.018; Table 1).Compared to NOR heads, DROP heads were more symmetric on the sides (P = 0.007), but no difference was seen in anterior-posterior symmetry (P = 0.430).Chromatin heterogeneity and percent decondensed chromatin were higher in the DROP group than in the NOR group (P < 0.001 for both).Significant correlations were found among percent decondensed chromatin, chromatin heterogeneity, and all morphometric measurements except ante posterior symmetry and chromatin heterogeneity (Table 2).Chromatin heterogeneity was also correlated with percent decondensed chromatin, and both parameters were significantly correlated with percent cytoplasmic droplets.

Discussion
Studies in other species have shown that variation in spermatozoa morphology is a sensitive marker of chromatin abnormality and animal fertility (Hingst et al., 1995;Karabinus et al., 1997;Beletti et al., 2005).Thus, in the present study, normal spermatozoa and spermatozoa with cytoplasmic droplets were evaluated separately.To the best of our knowledge, this is the first report of such an evaluation; Anim.Reprod., v.14, (Suppl.1),p. 1253-1258. 2017 no comparable results exist in the literature.Nonetheless, studies that have compared rejected and approved ejaculates based on sperm morphological examination have shown results similar to those reported here.A positive correlation (r = 0.44; P = 0.01) between decondensed chromatin and percent cytoplasmic droplets has been reported (Martínez, 2005) based on the sperm chromatin structure assay (SCSA): a modified fluorescence microscopy assay.Fischer et al. (2003) also showed a positive correlation (r = 0.59, P < 0.001) between DNA denaturation and the presence of cytoplasmic droplets in human sperm using flow cytometry analysis of acridine-orange-treated spermatozoa.Further, Volker (2004) reported that when the occurrence of cytoplasmic droplets is higher than 50% in boars, the presence of decondensed chromatin is greater than 5%.However, we report a comparatively smaller but significant difference in percent decondensed chromatin.No defined limits for percent decondensed chromatin exist, but previous reports have used 5% (Wabersky et al., 2002;Volker, 2004;Martínez, 2005).We showed that sperm from NOR and DROP groups both were within this limit, and it is important to consider the differences in methodologies, genetic lines and species used in our study compared to those in literature.
Although most correlations were statistically significant, correlation coefficient (r) values were considered low and, for this reason, only the highest values are considered in this discussion.Moreover, it is not possible to include individual boar effect and age as covariates in the correlation test; thus, some correlation values could be influenced by these characteristics.Chromatin heterogeneity was correlated with the percent decondensed chromatin, which indicates that cells with abnormal chromatin have greater heterogeneity in chromatin compaction within the spermatozoa head.In agreement with the results reported by Beletti et al. (2005) for bulls, correlations among chromatin heterogeneity, decondensed chromatin, and head morphometric measurements reported here indicate that chromatin condensation can be associated with sperm head morphometry even in the absence of morphological abnormalities.The positive correlation between percent cytoplasmic droplets and chromatin heterogeneity can possibly be explained by the immaturity of cells when they were ejaculated.More frequent semen collection results in less time for cell maturation, and this probably influences the completion of chromatin condensation (Yoseffi et al., 1994;Hingst et al., 1995;Golan et al., 1996) as well as cytoplasmic droplet migration (Flowers, 2004).Also important is that retention of droplets, primarily proximal droplets, correlates strongly with production of reactive oxygen species (ROS) in semen (Gomez et al., 1996).ROS has been shown to affect sperm function in different ways, which include impaired fertilization capacity and DNA integrity (Irvine et al., 2000;Aitken and De Iuliis, 2010;Aitken et al., 2013).Hirai et al. (2001) studied sperm of Pietrain boars using automated sperm morphometry analysis (ASMA); Saravia et al. (2007) used ASMA and the ISAS® morphometric module to evaluate sperm of Duroc, Large White, Landrace, and hybrid lines; and Gil et al. (2009) used ISAS® to evaluate sperm from Iberian boars.All these studies reported lower average head area (28.45-36.20 µm 2 ) than that in the present study, but Saravia et al. (2007) showed higher values for average head perimeter (26.00 µm), width (4.50 µm), length (9.00 µm), and ellipticity (2.00).Similarly, Gil et al. (2009) reported greater values for head perimeter (22.35 µm) and ellipticity (1.99), but lower values for head width (4.07 µm) and length (8.11 µm) than those reported here.Importantly, it has been shown that differences in measurements arise dependent on the boar lines (Saravia et al., 2007;Kondracki et al., 2012) and evaluation technique (Boersma et al., 1999).Therefore, we propose that morphometric data should be compared between studies only when the analysis method and the boar genetic line are identical.
We found that ellipticity and shape factor are in agreement with width and length, which show that the DROP group sperm heads were more elongated than those of the NOR group.Symmetry measurements relate to hydrodynamic properties of the cell and can therefore be used for identification of specific alterations in sperm heads, such as the pyriform head (Beletti and Costa, 2003).Even though cells in the DROP group did not demonstrate head abnormalities, the average for side head symmetry in this group was higher than for the NOR group.This implies that the presence of droplets could be associated with head shape.
Very few studies have analyzed the effects of sperm head morphometric measurements on the reproductive performance of sows.Hirai et al. (2001) showed that sows inseminated with semen containing longer sperm heads had a reduced farrowing rate, and the cutoff value in their analysis was 86%.Even though the present study did not evaluate the effect of cytoplasmic droplet presence on the reproductive performance of sows, the results of the morphometric analyses were similar in both studies, i.e., longer sperm produced poorer results.This similarity can be explained by the relationship between sperm head morphology and fertility potential.Accordingly, we showed that cells with cytoplasmic droplets have longer heads and, when present during morphological exam, this abnormality is generally negatively correlated with reproductive performance (Waberski et al., 1994;Benchaib et al., 2003;Feitsma et al., 2008;Feitsma, 2009;McPherson et al., 2014).
In conclusion, in abnormal sperm, the presence of cytoplasmic droplets is associated not only with altered morphometric characteristics (in particular, longer and narrower sperm heads) but also with chromatin instability (higher percent decondensed chromatin and more heterogeneous chromatin).

Table 1 .
Morphometric measurements, percentage of decondensed chromatin and chromatin heterogeneity of the boar sperm heads of normal sperm and sperm with proximal or distal cytoplasmic droplets (Mean ± Standard Deviation).

Table 2 .
Correlation among percentage of decondensed chromatin, morphometric measurements and chromatin heterogeneity of boar sperm.