The objective of the present study was to describe the basic characteristics of brown bear spermatozoa collected by electroejaculation and from the cauda epididymis after necropsy. Semen was collected during the mating season of 2004 and 2005 from five mature males, aged from 3.5 to 22 years. Animals were anesthetized with 7 to 10 mg/kg body weight of Zoletil^® (tiletamine + zolazepam; Virbac, Carros, France). Bears were placed on a special stretcher in ventral recumbency. A rectal probe with three longitudinal electrodes and a diameter of 2 cm (Electrojac IV; Minitübe, Tiefenbach, Germany) was placed into the rectum. Two series of stimuli were administered with a 2–3 min rest between series. Each stimulus lasted 2 s and was administered at 2 s intervals increasing intensity of stimulation from 0 to 15V. Of 23 electroejaculation attempts, 17 ejaculates were collected and two were eliminated because of poor semen quality. Epididymal spermatozoa were collected from three animals after euthanasia by flushing cauda epididymis with the freezing medium. Two animals showed unilateral testicular hypoplasia and only one testicle was processed from each of them. Volumes were recorded and concentration was evaluated in the laboratory. Semen quality of epididymal and ejaculated samples was assessed by: progressive sperm motility (%IM); quality of movement on a scale of 0 to 5 (Q); normal acrosome status (%NAS) and normal sperm morphology (%NOR), assessed by phase contrast microscopy of glutaraldehyde fixed samples; membrane response to hypo-osmotic test (%HOST); and sperm viability estimated using eosin-nigrosin vital staining (%V). Results were evaluated by Student t-test. Mean volume of ejaculates was 0.79 ± 0.73 mL. The average concentration of samples collected by electroejaculation was 519 ± 278 × 10⁶ sperm/mL. The average number of spermatozoa collected per ejaculate was lower than the average number of spermatozoa collected from the epididymis, 476 ± 352 × 10⁶ vs. 640 ± 164 × 10⁶, respectively (P > 0.05). Sperm quality was not different between epididymal and electroejaculated samples (Table 1), except that the %NORwas higher in electroejaculated samples than in epididymal samples (62 ± 16 vs. 3 ± 15; P < 0.01). Ahigh percentage of spermatozoa with abaxial midpiece attachment (78.6 ± 10.4%) was observed in all samples. All electroejaculates contained leukocytes and 53% of the ejaculates had agglutinated spermatozoa. All 15 ejaculate and all epididymal samples were frozen in TES-Tris extender containing 4% glycerol using a standard protocol for bull semen. Our results indicate that electroejaculated and epididymal brown bear spermatozoa may be of sufficient quality for use in AI programs to increase gene flow between isolated captive populations. Whether cryopreserved samples can be used for artificial breeding remains to be determined.

This work was supported by Zoological Society of San Diego.