Search

We previously isolated epithelial-like cells (ELC) from sheep (Ovis aries) and eland (Taurotragus oryx) ejaculates (Nel-Themaat et al. 2004 Reprod. Fertil. Dev. 16, 152). Success rates were low, and the presence of live sperm during initial culture may have altered medium properties, physically prevented contact between cells and the substrate, and/or damaged somatic cell membranes. Using the complete semen sample requires the sacrifice of valuable live spermatozoa without the certainty of obtaining a somatic cell population. Therefore, the purpose of the present study was to develop a method for isolating somatic cells from semen before culture. Separation by discontinuous Percoll gradient centrifugation was performed. Ejaculates obtained by electro-ejaculation from two Louisiana Native rams and rectal massage of one eland bull were washed, layered on columns of 2.5 mL each of 20% (P-20), 50% (P-50), and 90% (P-90) Percoll in 15-mL conical tubes, and centrifuged for 20 min at 400g. Bands were obtained at the three interfaces, as well as a pellet in the P-90 fraction. Each Percoll layer plus the interface band immediately below it was collected by aspiration. Cells from each fraction were re-suspended in serum-supplemented Minimum Essential Medium Alpha containing 500 IU/mL of penicillin, 0.5 mg/mL of streptomycin and 250 g/mL of gentamycin, and were plated on collagen-1 coated dishes. Trypan blue staining was used to determine cell viability in each fraction. After 24 h, dishes were washed and the culture continued until cells attached and proliferated. For both species, the surface fraction yielded primarily dead ELC that were morphologically flat and angular in shape. In the P-20 band, more viable cells were isolated compared with other fractions, although dead cells were also present. In the P-50 and P-90 fractions, few ELC were observed compared with the surface and P-20 fractions. A negligible number of sperm was observed in the surface and P-20 fractions. In contrast, mostly dead sperm were found at the P-50 and P-90 interface band, whereas live spermatozoa were detected in the P-90 pellet. Epithelial-like cells isolated from semen of both species attached (A, initial 24 h of culture), divided (D, within 3 days after plating), and proliferated (P, after 1 week of culture) (see Table). Ram ELC derived from the surface and from the P-20 and P-50 layers attached, but proliferation was detected only with cells collected in the P-20 fraction. In contrast, eland cells collected from the surface and P-20 layers readily attached and proliferated. In summary, a technique has been developed for the isolation of somatic cells from semen using a three-layer Percoll gradient. This method also allows the isolation of motile spermatozoa, hich are available for further use.

We previously isolated epithelial-like cells (ELC) from sheep (Ovis aries) and eland (Taurotragus oryx) ejaculates (Nel-Themaat et al. 2004 Reprod. Fertil. Dev. 16, 152). Success rates were low, and the presence of live sperm during initial culture may have altered medium properties, physically prevented contact between cells and the substrate, and/or damaged somatic cell membranes. Using the complete semen sample requires the sacrifice of valuable live spermatozoa without the certainty of obtaining a somatic cell population. Therefore, the purpose of the present study was to develop a method for isolating somatic cells from semen before culture. Separation by discontinuous Percoll gradient centrifugation was performed. Ejaculates obtained by electro-ejaculation from two Louisiana Native rams and rectal massage of one eland bull were washed, layered on columns of 2.5 mL each of 20% (P-20), 50% (P-50), and 90% (P-90) Percoll in 15-mL conical tubes, and centrifuged for 20 min at 400g. Bands were obtained at the three interfaces, as well as a pellet in the P-90 fraction. Each Percoll layer plus the interface band immediately below it was collected by aspiration. Cells from each fraction were re-suspended in serum-supplemented Minimum Essential Medium Alpha containing 500 IU/mL of penicillin, 0.5 mg/mL of streptomycin and 250 g/mL of gentamycin, and were plated on collagen-1 coated dishes. Trypan blue staining was used to determine cell viability in each fraction. After 24 h, dishes were washed and the culture continued until cells attached and proliferated. For both species, the surface fraction yielded primarily dead ELC that were morphologically flat and angular in shape. In the P-20 band, more viable cells were isolated compared with other fractions, although dead cells were also present. In the P-50 and P-90 fractions, few ELC were observed compared with the surface and P-20 fractions. A negligible number of sperm was observed in the surface and P-20 fractions. In contrast, mostly dead sperm were found at the P-50 and P-90 interface band, whereas live spermatozoa were detected in the P-90 pellet. Epithelial-like cells isolated from semen of both species attached (A, initial 24 h of culture), divided (D, within 3 days after plating), and proliferated (P, after 1 week of culture) (see Table). Ram ELC derived from the surface and from the P-20 and P-50 layers attached, but proliferation was detected only with cells collected in the P-20 fraction. In contrast, eland cells collected from the surface and P-20 layers readily attached and proliferated. In summary, a technique has been developed for the isolation of somatic cells from semen using a three-layer Percoll gradient. This method also allows the isolation of motile spermatozoa, hich are available for further use.

We previously isolated epithelial-like cells (ELC) from sheep (Ovis aries) and eland (Taurotragus oryx) semen (Nel-Themaat et al. 2004 Reprod. Fertil. Dev. 16, 152) and subsequently developed a system to separate ELC before plating (Nel-Themaat et al. 2005 Reprod. Fertil. Dev. 17, 314). Cells attached and proliferated in only 50% and 31% of the attempts, respectively. Therefore, the purposes of the present study were to improve ELC proliferation by co-culture with inactivated 3T3 mouse fibroblasts and to characterize the obtained cells. Semen fractions from two mature Gulf Coast Native rams (n = 20 ejaculates) and one common eland bull (n = 2 ejaculates) were plated on feeder layers (A), on collagen with feeder cell inserts (B), or on collagen alone (C). For B and C, cell attachment and division were assessed; proliferation and passage 1 (P1) confluence were evaluated for A, B, and C. No difference in attachment rates between B (80%) and C (70%) were found for ovine cells, but (P < 0.05) cells divided more times in B (80%) than in C (35%). All colonies in A (60%) and B (70%) reached P1 confluence and no difference was detected between A and B, but less proliferation (10%) and P1 confluence (5%) were observed in C. Therefore, contact between epithelial cells and feeders was not necessary for growth stimulation by the feeder cells. No difference among A, B, and C was detected for eland ELC proliferation (100%), but no P1 confluence was observed in C. Ram cells were subsequently characterized by immunohistochemical detection of keratin and vimentin, as well as morphology. The 3T3 cells cross-reacted with keratin, and characterization was thus performed mainly on morphology and vimentin expression. Distribution of vimentin microfilaments differed between different epithelial morphologies and fibroblasts. Expression in epithelial cells was faint and patchy in confluent colonies and located around cytoplasmic extremities in semi-confluent colonies. In 3T3 cells, expression was very prominent throughout the cytoplasm and around the nucleus. For treatments A and B, 63 and 57%, respectively, were characterized as only epithelial cells; 25 and 36%, respectively, appeared to contain a mixture of epithelial and fibroblast cells; and 13 and 7%, respectively, contained only fibroblast cells. Only one sample was evaluated from treatment C and only keratin was detected in the epithelial-like colony. We conclude that culture of semen-derived ELC is markedly improved by 3T3 fibroblast co-culture. Further research on conditioned media may simplify the system and reduce chances of 3T3 cell contamination.