- Top of page
- Materials and methods
In this study, the effect of the chamber used for the automated analysis of sperm motility by a computer-assisted semen analysis system on sperm kinematics was evaluated, and the cause of this effect was also verified. Twenty-three bull semen batches were thawed, and semen was diluted, aliquoted and analysed with six different chambers, (three capillary-loaded chambers and three droplet (DR)-loaded chambers). For each chamber type, each sample was analysed in quadruplicate, and the reliability of the analysis was tested using an intraclass correlation coefficient (ICC). Furthermore, sperm membrane integrity (MI) was evaluated, for each sample and chamber, in 12 randomly selected central and 12 edge fields. The ICC analysis showed that some parameters could have a significant variability related to the chamber. High stability of results was detected in Leja 4-chamber slide. Furthermore, as previously reported in other studies, capillary-loaded chambers seemed to affect the total and progressive motility and sperm velocities (average path velocity, straight line velocity and curvilinear velocity). These findings were corroborated by the evaluation of sperm MI that was significantly higher in the DR-loaded chambers. This study confirms that the chamber used for the objective kinetic evaluation of bull-thawed spermatozoa significantly affects the result. These differences could be present also in other species, even if the specific effect on the sperm kinematics should be verified. The Makler chamber seemed to give reliable results with negligible effects on sperm kinematics.
- Top of page
- Materials and methods
Sperm motility is considered as one of the most important parameters for the evaluation of the animal sperm quality worldwide. Furthermore, in human seminology specific thresholds were reported for classifying ejaculates with normal total (lower reference limit of 40%) and progressive (lower reference limit of 32%) motility (WHO, 2010). The development of computer-assisted semen analysis (CASA), using a software that analyses and records every sperm track characteristic, allows the precise and comparable estimation of sperm kinetic. The availability of data recorded by CASA facilitates the comparison of results and makes it possible to find subtle differences between male subjects or treatments (Verstegen et al., 2002). Furthermore, CASA systems appear to have high accuracy and repeatability (Davis et al., 1992; Farrell et al., 1995). On the other hand, CASA systems require a strict setting to achieve reliable and comparable results (Mortimer et al., 1995; Contri et al., 2010). A strict setting is also required to differentiate motile and immotile spermatozoa from debris and other cells and particles.
In a previous study, the effect of some settings was found to be relevant for the motility evaluation results in the bull (Contri et al., 2010). One of the most important variables is the chamber used for the analysis. Different types of chambers could be used on the CASA systems, and they differ in terms of volume, depth, shape and loading modality. The effect of different chambers used for CASA analysis was marginally studied for the assessment of sperm concentration in human (Mahmoud et al., 1997; Sokol et al., 2000), but few data are reported in animals (Prathalingam et al., 2006). Most of these data showed that the concentration of spermatozoa changed dependently to the device used for the analysis, and most researchers agreed with the assumption that a different distribution of the latex beads and sperm suspension was recorded in different chambers. This phenomenon could be the results of the Segre–Silberberg effect (Douglas-Hamilton et al., 2005; Kuster, 2005). The preliminary evaluation of the chamber effect by Contri et al. (2010) showed that, similar to the concentration, sperm motility also could be different in capillary-loaded chamber (Leja 4 chamber) compared with data recorded in DR-loaded chamber (Makler). These results were afterwards confirmed by other studies in bovine and equine semen (Lenz et al., 2011; Hoogewijs et al., 2012).
Little is known about the effect of different chambers used on sperm kinetic evaluation in the bull. Furthermore, no data are available on the variability in different chambers on sperm kinetic results; thus, the experimental design proposed in this study was developed with the aim of evaluating if and how the chamber used for the sperm motility analysis affect kinetic parameter of the sample. Furthermore, we tried to provide an explanation of these differences by the study of the distribution of live and dead spermatozoa within these chambers.
- Top of page
- Materials and methods
In this study, cryopreserved semen from 23 different bulls was evaluated using six different support devices by CASA system. As recorded in the preliminary trial, sperm motility parameters showed similar values in stained and non-stained samples in both L2 and MK at any time (data not shown). Thus in this study, samples were stained before sperm motility analysis, to allow a rapid and contextual evaluation of MI of the same cells in each device.
The amount of spermatozoa counted in each slide was comparable between chambers. The ICC was higher for L4, L2 and CVD, whereas it was lower for CVC and SL and MK (Table 2).
Table 2. Intraclass correlation coefficient (ICC) values of each sperm parameter measured in bull-thawed spermatozoa evaluated using computer-assisted semen analysis system through different chambers
|Rapid cells (%)||0.752**||0.925**||0.823**||0.889**||0.853**||0.906**|
|Medium cells (%)||0.834**||0.871**||0.705*||0.958**||0.932**||0.902**|
|Slow cells (%)||0.86**||0.904**||0.796*||0.932**||0.294||0.933**|
|Static cells (%)||0.859**||0.95**||0.711*||0.757*||0.979**||0.855**|
The TM was very stable within the same chamber, but it was very different between chambers. The reliability, by the ICC, was higher for L2, CVD, MK and SL, whereas it was significant, but less repeatable in L2 and CVC. On the other hand, the percentage of motile spermatozoa was significantly higher in all chambers loaded by DR-loaded than the CP-loaded chambers (Table 3). This finding was different for PM. For this parameter, the ICC evaluation showed lower repeatability for SL (Table 2). The values found in SL and CVD were significantly higher than the values recorded for capillary-loaded chambers and for MK (Table 3).
Table 3. Mean (±SD) of sperm kinetic parameters in bovine post-thawing semen analysed by computer-assisted semen analysis using different chambers: Leja 2 chamber (L2), Leja 4 chamber (L4), Cell-Vu with fix coverslip (CVC), Cell-Vu sperm counting chamber (CVD), slide-coverslip (SL), and Makler chamber (MK)
|Mean ± SD||Mean ± SD||Mean ± SD||Mean ± SD||Mean ± SD||Mean ± SD|
|Cell counted||2845.8 ± 611.8||2414.3 ± 362.7||2764.2 ± 695.6||2220.9 ± 955.9||2165.5 ± 1059.5||2899 ± 820.9|
|Total motility (%)||47.6 ± 8a||51.2 ± 8.2a||44.2 ± 9.8a||74.4 ± 12.8b||67.3 ± 10.8b||71.3 ± 10.6b|
|Progressive motility (%)||30.2 ± 5.4a||34.5 ± 7.2a||28.8 ± 6.4a||53 ± 9.8b||47 ± 8.8b||51.4 ± 5.1b|
|VAP (μm/s)||101.1 ± 6.5a||103.3 ± 9.7a||102 ± 8.2a||111 ± 16.5b||118.8 ± 11.5c||113.7 ± 7.9bc|
|VSL (μm/s)||82.4 ± 5.4a||85.7 ± 7.7a||83.9 ± 6.5a||92.4 ± 15.7b||99.1 ± 10.9b||89 ± 5.8b|
|VCL (μm/s)||169.1 ± 11.9a||171.4 ± 19.2a||170.6 ± 15.9a||172.7 ± 22.1a||186.4 ± 18b||191.9 ± 13.8b|
|ALH (μm)||7 ± 0.5a||6.8 ± 0.8a||6.8 ± 0.7a||6.7 ± 0.8a||6.9 ± 0.7a||7.8 ± 0.4b|
|BCF (Hz)||34.4 ± 1.7a||35.5 ± 2.6a||35.1 ± 1.9a||34.1 ± 2.5a||33.4 ± 1.8a||30.2 ± 1.8b|
|STR (%)||79 ± 1.5a||80.6 ± 2a||79.8 ± 2.2a||81.8 ± 4.2a||82.5 ± 2.4a||76.9 ± 1.9a|
|LIN (%)||49.5 ± 1.7a||50.9 ± 3.1a||50.5 ± 2.8a||55.2 ± 6.7a||54.9 ± 3.6a||47.7 ± 2.1a|
|Rapid cells (%)||39.9 ± 7.3a||43.6 ± 8.2a||37.6 ± 8.6a||67.5 ± 12.6b||61.2 ± 11.6b||58 ± 10.4b|
|Medium cells (%)||7.8 ± 1.1a||7.6 ± 1.4a||6.6 ± 1.7a||7 ± 3.8a||6.2 ± 4.5a||13.4 ± 2.6b|
|Slow cells (%)||41.2 ± 5.8a||43.7 ± 6.9a||37.5 ± 6.2a||17.3 ± 10.3b||25.1 ± 5.2bc||26.2 ± 8.2bc|
|Static cells (%)||11.1 ± 6.5||5.1 ± 4.2||18.2 ± 10.7||8.2 ± 8.8||7.2 ± 8.3||2.5 ± 2.9|
|Membrane integrity (%)||51.3 ± 16.9a||59.8 ± 5.1a||56.2 ± 6.9a||79.5 ± 9.7b||75.1 ± 15.7b||75 ± 9.9b|
The ICC revealed a high stability for values of all velocities in all chambers; however, the value detected using the CASA system seemed significantly affected by the chamber. The values of SL, MK and CVD were higher than the velocities recorded with the other chambers (Table 3).
The lateral head displacement was significantly higher in the MK chamber, whereas the value is similar for the other devices. This parameter was very variable in CVD and SL, in which the ICC was low and not significant. Similarly, the BCF was comparable in all chambers, but was significantly variable in CVD, SL and MK (Table 2). The straightness and the LIN were similar for all chambers, and ICC showed a good reliability of these parameters in all chambers.
Finally, for the sperm velocity subclasses, only the slow sperm category seemed to have a low ICC value in the SL chamber. The percentage of rapid spermatozoa was significantly higher in DR-loaded chambers compared with CP-loaded chambers. Conversely, a significantly higher proportion of spermatozoa were judged slow by the parameters used for the classification of this category in all L2, L4 and CVD chambers.
The distribution of spermatozoa in the chamber seemed different in different devices, as suggested by the number of spermatozoa counted in the MI evaluation. A wide discrepancy in sperm count and MI between the centre and the periphery was found in all chambers, but higher values were found for L4 and CVC. Furthermore, in the Leja 4 chamber an inhomogeneous number of spermatozoa was detected in some points of the perimeter of the chamber (Fig. 1). The percentage of spermatozoa with MI was higher in SL, and especially in the MK chamber, although was comparable in the other chambers (Table 4).
Figure 1. Leja 4 chamber and representative images of the number of spermatozoa in the field near the posterior angle (on the left) compared with the central field (on the right).
Download figure to PowerPoint
Table 4. Mean (±SD) of sperm per field and sperm membrane integrity in central fields (CFs) and in edge fields (EFs) in 23 bull semen samples analysed using different chambers
|Chamber||Loading method||Sperm/field||Membrane integrity|
|Mean ± SD||Mean ± SD||Mean ± SD||Mean ± SD|
|L2||Capillarity||15.8 ± 3.7||14.2 ± 4||51.3 ± 16.9a||37.6 ± 7.5b|
|L4||Capillarity||15.4 ± 3.1a||35.2 ± 21.1b||59.8 ± 5.1a||38.5 ± 8.4b|
|CVC||Capillarity||14.6 ± 2.4||13.5 ± 3.2||56.2 ± 6.9a||33.9 ± 9.4b|
|CVD||Droplet||17.1 ± 3.3||14.6 ± 3||79.5 ± 9.7a||36.5 ± 9b|
|SL||Droplet||16.4 ± 4.9||12.1 ± 3.2||75.1 ± 15.7a||42.5 ± 9.3b|
|MK||Droplet||18.3 ± 4||14.4 ± 3.3||75 ± 9.9a||61 ± 14.3a|
The agreement between MI estimated in the CF compared with the peripheral fields was higher in the L2 chamber (0.732; p ≤ 0.01), SL (0.852; p ≤ 0.01) and MK (0.934; p ≤ 0.01), whereas in L4 (0.427), CVC (0.319) and CVD (0.233) was not significant (Fig. 2).
Figure 2. Passing and Bablok plots for the agreement between sperm membrane integrity detected in the central (centre) and peripheral (edge) fields in the chambers studied in bull frozen-thawed samples. The dotted line represents the identity line (x = y), the solid line represents the regression line and the dashed lines represent the confidence interval for the regression line.
Download figure to PowerPoint
No significant correlations were found among TM and MI (L2 = 0.527; L4 = 0.542; CVC = 0.484; CVD=0.516; SL = 0.468; MK=0.486), such as among PM and MI in all chambers (L2 = 0.476; L4 = 0.528; CVC = 0.436; CVD = 0.381; SL = 0.412; MK = 0.489). In all chambers, the correlations between kinetic parameters were similar: VAP, VSL and VCL were highly correlated between them (p ≤ 0.01). Lateral head displacement was positively correlated with sperm velocities, but it was negatively correlated with BCF, STR and LIN in all chambers. A negative correlation between VCL and BCF was detected in the MK.
- Top of page
- Materials and methods
Motility is considered as an important parameter for the evaluation of frozen-thawed semen quality in bovines and other domestic species. Motility estimation is an essential step in evaluating male fertility and can be considered as a functional test because of its relationship with the energy status of the mammalian spermatozoa (Roldan, 1998; Quintero-Moreno et al., 2004).
Although motility and kinetic parameters cannot be considered a reliable marker for the fertilizing ability of a given ejaculate (Holt et al., 1985, 1997; Marshburn et al., 1992; Barratt et al., 1993; Krause, 1995; MacLeod & Irvine, 1995), spermatozoa with low or altered movement are unlikely to reach the oviduct, and it is reasonable to presume that the more the spermatozoa with PM, the higher the chance that one of them will reach the ampulla of the oviduct (Muino et al., 2008). The evaluation of thousands of sperm tracks using CASA systems and the quantification of numerous kinetic parameters at cell level allowed the evaluation of specific characteristics of groups of spermatozoa with similar kinetic characteristics (Martínez-Pastor et al., 2011). CASA systems are useful tools for objective and detailed evaluation of kinetic characteristics of an ejaculate; however, they are not ready-to-use devices. In our previous study, we reported an effect of some settings and sample preparation variables on sperm kinetic results in the bull (Contri et al., 2010). Similar results were reported in other species such as the dog (Rijsselaere et al., 2003). These studies showed the need of a standardization of the procedures developed in andrology (ESHRE Andrology Special Interest Group, 1998).
A source of variability seemed related to the chamber used in the CASA system, as suggested in the previous study (Contri et al., 2010). Thus, the aim of this study was the evaluation of the effect of the chamber on sperm kinetic results in the bull. Different chambers are commercialized, with variable characteristics. In this study, we tested three capillary-loaded chambers (Leja 2 chamber, Leja 4 chamber and CVC) and three DR-loading chambers (CVD, SL and MK). This study clearly showed that the type of chamber significantly affect sperm motility characteristics. The amount of cells detected was variable in different chambers. The number of spermatozoa was highly repeatable in L2 and L4 chambers, CVD and MK, whereas it was less stable in CVC and SL. This inhomogeneous value could be the result of a non-uniform distribution in the surface of the chamber, and this could affect the succeeding results of motility. This could be supported by the high discrepancy between the number of cells counted in the MI evaluation between the centre of these chambers and the edge (data not shown). The uniform distribution of particles in the different chambers is a relevant issue in sperm objective analysis by CASA. In different studies, the hydrodynamic movement of the fluid within the closed chambers, termed Segre–Silderberg effect, was suggested to be responsible for a different distribution of suspended particles in capillary-loaded chambers (Douglas-Hamilton et al., 2005; Kuster, 2005). In those studies, the authors hypothesized that the spermatozoa loaded in a chamber by CP were influenced, in their distribution, by the Poiseuille flow. This effect is not able to explain why the proportion of motile and progressive motile spermatozoa, such as their mean velocity, was higher in DR-loading chambers. No visible passive movement of spermatozoa was detected in any chamber before the analysis. Furthermore, a random distribution of spermatozoa within chambers loaded by DR and coverslip could reduce the repeatability of the analysis. These considerations were not verified in this study: a high repeatability was found especially in MK for most parameters, suggesting that the random distribution should not be considered causative.
Chamber used for the evaluation by CASA affects also the kinetic characteristics of spermatozoa, with higher values in DR-loaded chambers. On the other hand, the repeatability was significantly higher in capillary-loaded chambers. Total and PM were highly repeatable in all chambers, with the exception of the SL, but were significantly affected by the device used, as previously reported (Contri et al., 2010; Lenz et al., 2011; Hoogewijs et al., 2012). Sperm velocities were highly repeatable in all chambers, suggesting that these parameters were very stable in all conditions. However, lower values were found in L2, L4 and CVC, compared with DR-loading chambers. The significant low values in these parameters, in agreement with a previous study (Contri et al., 2010), but in contrast with others (Lenz et al., 2011; Hoogewijs et al., 2012), underline a possible negative effect specific of the fix coverslip. As for the amount of sperm counted in different parts of the chamber, the loading of a sample in a capillary-loaded vs. DR-loaded chamber seemed to affect the kinetic results via a different hydrodynamic behaviour of the fluid. A different explanation could be related to the possible toxicity of the glue used for the fixing of the coverslip or the paint used for the serigraphy of chambers. It could be hypothesized that the adhesive used for the application of the coverslip could be toxic for spermatozoa. This hypothesis could be corroborated by the reduced MI recorded in the chambers with fix coverslip and by the high proportion of spermatozoa with membrane damage detected in the edge of these chambers. In addition, the toxic effect could alter sperm kinetic parameters before the reduction in sperm MI, explaining the lower values for VAP, VSL and VCL recorded in this study. However, this hypothesis was not tested in our study, thus specific studies on the effect of the adhesive are needed to exclude definitively this possibility.
In this study, the high values for sperm total and PM, such as sperm velocities, were detected in SL system. This modality for sperm motility evaluation is extensively used during the subjective motility estimation. In our study, we used a 10-μL DR under a 22 × 22-mm coverslip, obtaining a chamber of about 20 μm in depth (WHO, 2010). However, several factors have an effect on the distribution of spermatozoa in a flattened DR, affecting sperm kinetic characteristics (Nöthling & dos Santos, 2012). This finding was confirmed by our data, in which a great variability in kinetic results was obtained after repeated analysis of the same sample. Thus, the data presented here suggested that sperm motility kinetics were similar to other DR-loaded chambers, but a lower repeatability was detected for some parameters, such as PM and BCF. A lower proportion of spermatozoa with MI was detected in the SL chamber in the peripheral fields compared with CF. This finding was in agreement with previous results and could be explained by the possible effect of the surface tension on the perimeter of the coverslip (Lenz et al., 2011).
The ALH was significantly higher in the MK compared with those in the other devices (Table 2). These values were lower than those reported previously in the bull (Contri et al., 2010), but showed the same differences compared with the values recorded using the Leja chambers. This wide amplitude of the sperm movement could be related to the depth of the chamber because the Makler was the only chamber with a 10-μm depth. The flagellar beat is almost a mysterious phenomenon, but there is a general agreement on the rotational movement of spermatozoa (Ishijima et al., 1992), as a result of the flagellar dynamic that requires an adequate free space around each cell to move freely. On this basis, the ALH could be considered the 2D simplification of a 3D real movement of spermatozoa. A reduction in the depth of the chamber could limit the amplitude of the flagellar excursion in one of the dimensions, resulting in a more wide 2D movement. In a previous study (Makler, 1978), the motility of spermatozoa was prevented when the depth of the chamber was 2 μm or lower, but increased progressively as the depth increases between 2 and 10 μm. The author concluded that the optimal depth of the chamber for the evaluation of human spermatozoa was 8–10 μm. However, this study did not consider that the amplitude of the flagellar beat could be different in different species. Furthermore, the study considered only the subjective motility to corroborate this effect, because the objective quantification of sperm developed later. Now, we have devices that geometrically quantify the kinetic parameters of spermatozoa, thus it is possible that a depth of 10 μm could affect some characteristics of the sperm movement, such as the ALH or the BCF, without affecting the TM. In our study, the motility and PM, such as sperm velocities, were significantly higher in the MK.
The WHO laboratory manual (WHO, 2010) mentioned the CASA system as a reliable method for the evaluation of sperm kinematics in human. Our data showed that the chamber used for this analysis could affect the results, thus guidelines for the specimen analysis using CASA should consider this variability.
In conclusion, this study showed that the kinetic of spermatozoa in the bull measured using CASA systems was significantly affected by the chamber used for the analysis. TM, PM and all velocities (VAP, VSL and VCL) were significantly higher in DR-loaded chambers compared with capillary-loaded chambers. This effect could be related in part to the Segre–Silderberg effect, as previously hypothesized, but could also be also the result of a possible detrimental effect of one of the components of these types of chamber (glue used to fix the coverslip, paint used for the serigraphy) even if specific studies are required to verify specifically this second possibility. The different distribution of spermatozoa in the different chambers was corroborated by the high discrepancy between the number of spermatozoa evaluated in the centre of the field compared with the number detected at the edge.
The repeatability of the values recorded for kinetic parameters by the CASA systems could be affected by the chamber used. Although the use of the Leja 4 chamber resulted in significant lower values for total and PM and sperm velocities, the repeated analysis of the same sample resulted in a high stability of the measures for all kinetic variables. On the other hand, the SL allowed the higher values of several sperm characteristics, but the repeatability was low in several cases. The MK seemed to ensure a high repeatability and higher values for TM, PM, VAP, VSL and VCL, but showed a significantly higher ALH compared with the other chambers. It must be remarked that a high stability of the results recorded with this device should not be considered as a high truthfulness of the analysis, since we do not have a real value from a golden standard analysis which compare it.
Data reported in this study suggested that the results recorded by the CASA systems should be evaluated on the basis of the chamber used for the analysis.