Propidium Iodide Labeling
For each experiment on unfixed cells, the percentage of dead cells (i.e., cells with high permeability to PI) was assessed. Contrary to U937, where 25 µM VP-16 apoptosis induction did not increase PI permeability, cryopreservation induced statistically significant sperm cell death (P 28% ± 12.7% of the ejaculate cells showed high permeability to PI, whereas after cryopreservation, this proportion increased to 50.8% ± 14.2%. In spite of various precautions (specific medium for freezing, slow cooling), the number of dead cells was multiplied by 2.1% ± 1.1% after cryopreservation.
Blebbing membranes are morphological modifications characteristic of apoptosis. Using light microscopy, it was noted that VP-16 induced changes in U937 shape from round (healthy) to blebbing cells (apoptotic) (Fig. 1). In contrast, cryopreservation did not induce any change in spermatozoon shape (data not shown). The study then focused on other events in the apoptotic cascade.
Determination of Mitochondrial Membrane Potential after Cryopreservation in Living Sperm
The accumulation of the cationic lipophilic fluorochromes DiOC6(3) in the inner membrane of mitochondria enables detection of m variations . Using fluorescence microscopy, we found that DiOC6(3) fluorescence was mainly detected in the intermediate piece of the spermatozoa, where sperm mitochondria are located (Fig. 2D).
Using DiOC6(3)/PI, three cell patterns were detected: i) necrotic cells were labeled with PI, ii) living cells with normal m showed normal mean green fluorescence intensity, and iii) living cells with low m, characteristic of apoptotic phenomena, showed low mean fluorescence intensity. Typical cytograms of living (PI-negative) U937 and bovine sperm cells labeled with DiOC6(3)/PI are shown in Figure 2, A and B, respectively.
The m decrease induced by FCCP in U937 and in sperm cells confirmed the efficiency of DiOC6(3) labeling (Fig. 2, A and B, respectively). Furthermore, after apoptosis induction in the U937 cell line, the observed increase in the proportion of living cells with low m validated the experiment (Fig. 2A).
Cryopreservation induced a statistically significant (P 0.0001) increase in the proportion of bovine sperm cells with low m. Before cryopreservation, 11.3% ± 10.6% cells of the ejaculate showed mitochondria with low m, whereas after cryopreservation, this proportion dramatically increased to 44.9% ± 17%. Furthermore, the high m population nearly disappeared (4.9% ± 3.8%) (Table 1 and Fig. 3C).
Detection of Active Caspases in Sperm
During apoptosis, the decrease in m results from the opening of membrane pores located in the mitochondrial membrane. The consequence is the translocation and activation of the various proapoptotic factors. Our study focused on the activation of the caspase family of proteins, one of these main components.
Typical cytograms of U937 and bovine sperm cells labeled with FITC-VAD-FMK/PI are shown in Figure 3, A and B, respectively. Using FITC-VAD-FMK/PI, three cell patterns were detected: i) necrotic cells, labeled with PI, were found in the top quadrant; ii) living cells without active caspase were found in the lower left quadrant; and iii) apoptotic cells (i.e., living cells containing active caspases) were found in the lower right quadrant.
Induction of apoptosis in the U937 cells resulted in the appearance of living cells containing active caspases (Fig. 3A).
Cryopreservation of bovine spermatozoa induced a statistically significant (P with active caspases. Of the fresh cells of the ejaculate, 2.2% ± 1% showed active caspases, whereas after cryopreservation, this proportion reached 12% ± 6.3% (Table 1 and Fig. 3C). This proportion was not significantly (P > 0.05) modified after 4 h of incubation in PBS at 37°C (7.3% ± 0.7%, n = 3).
Active caspases were found to be mainly detected in the intermediate piece of spermatozoa (Fig. 3D). Most stained spermatozoa had normal morphology without any cytoplasmic droplets (immature sperm cell morphology) or typical characteristics of apoptotic cell (blebbing membranes) (Fig. 3D).
Assessment of Membrane Modifications in Living Sperm Cells
The protease activity of apoptotic factors contributes to the degradation phase of apoptosis when PS is exposed to the external leaflet of the cytoplasmic membrane. PS externalization was studied using the annexin V probe. Typical cytograms of cells labeled with annexin V-FITC/PI are shown for U937, human sperm, and bovine sperm cells in Figure 4. Three patterns of cell were detected: i) necrotic cells, labeled with PI, were found in the top quadrant; ii) living cells without PS exposure were found in the lower left quadrant; and iii) apoptotic cells with PS exposure were found in the lower right quadrant. This experiment gave clearly different populations in U937 (Fig. 4A) and in human sperm (Fig. 4B) cells, but the green fluorescence intensity of bovine sperm cells was too low to discriminate apoptotic from living cells (Fig. 4C).
During the degradation phase, the cytoplasmic membrane becomes slightly permeable. Apoptotic cells are permeable to Yo-Pro-1 green fluorochrome and impermeable to PI. Thus, use of combined Yo-Pro-1 and PI dyes provides a sensitive indicator for apoptosis . Typical cytograms of U937 and bovine sperm cells labeled with Yo-Pro-1/PI are shown in Figure 5, A and B, respectively. Using Yo-Pro-1/PI, three patterns of cell were clearly detected: i) necrotic cells, labeled with PI were found in the top quadrant; ii) living cells, with low permeability membranes, were found in the lower left quadrant; and iii) apoptotic cells (i.e., living cells with modified membranes) were found in the lower right quadrant.
Induction of apoptosis in the U937 cell line increased the number of living cells exhibiting a high permeability to Yo-Pro-1 (Fig. 5A).
Cryopreservation of bovine spermatozoa induced a statistically significant (P exhibiting high permeability to Yo-pro-1: only 1.4% ± 1.1% of the fresh cells of the ejaculate showed permeability to Yo-Pro-1, whereas after cryopreservation, this proportion reached 10.8% ± 5.8% (Table 1 and Fig. 5C).
DNA Fragmentation and Nucleus Condensation
Two late manifestations of the degradation phase of apoptosis are DNA fragmentation and nucleus condensation. Typical cytograms of TUNEL assay-labeled bovine sperm cells are shown in Figure 6A. Cells with DNA fragmentation were gated as shown in Figure 6A.
Induction of apoptosis in U937 cell line led to DNA fragmentation (data not shown). Moreover, the TUNEL assay was validated using DNase-treated spermatozoa as positive controls (Fig. 6A).
Cryopreservation did not induce a statistically significant (P > 0.05) increase in the proportion of DNA-fragmented cells (Table 1 and Fig. 6B). Of the fresh bovine cells, 9.1% ± 7.7% of the ejaculate showed DNA fragmentation and, after cryopreservation, this proportion was 11.1% ± 5.7%. Only 5.9% ± 8.4% of cells survived after 30 h of incubation in PBS at 37°C. This massive cell death was not associated with a significant (P > 0.05) increase of the fragmented DNA cells percentage (11.3% ± 3.6%; Fig. 7).
Accessibility of PI to the spermatozoon nucleus enabled two populations with differently condensed nuclei to be quantified . Figure 6A shows these cell populations with normally condensed and decondensed chromatin.
Of the fresh cells of the bovine ejaculate, 46% ± 12.7% exhibited a normal nucleus condensation, and after cryopreservation, this proportion was 43.8% ± 13.1%. Cryopreservation/thawing or cryopreservation/thawing followed by 30 h of incubation in PBS at 37°C did not induce a statistically significant (P > 0.05) decrease in the proportion of cells with a normally condensed nucleus (Table 1, Figs. 6B and 7).
We also compared the percentage DNA fragmentation of cells with normal and with decondensed nuclei. A positive correlation (r = 0.92) was found for DNA fragmentation in cells with a normally condensed nucleus and in cells with a decondensed nucleus (Fig. 6C). Furthermore, the DNA fragmentation in cells with a normally condensed nucleus was significantly (P in cells with a decondensed nucleus (17.2% ± 10% vs. 8.3% ± 5.5%) (Fig. 6B).