MOLECULAR REPRODUCTION A N D DEVELOPMENT 2 5 6 1 - 6 6 (1990)
Collection and Quality of Rhesus Monkey Semen S.E. LANZENDORF,' P.M. GLIESSMAN? A.E. ARCHIBONG,l M. ALEXANDER,' AND D.P. WOLF' 'Division of Reproductive Biology and Behavior, 'Biomedical Engineering, Oregon Regional Primate Research Center, Beauerton, Oregon ABSTRACT Electroejaculation is an accepted method of semen collection from nonhuman primates. Although both penile and rectal probe stimulation techniques have been used, there has been a general lack of consistency and detail regarding their applicotion. This report describes the collection, processing, and evaluation of rhesus monkey semen contrasting two methods of penile electroejaculation: 1 ) a constant-voltage method where stimulus current is a variable and 2) a constant-current method where stimulus current is operator-controlled. The constant-current method was the more efficient procedure, requiring a lower stimulus current for successful electroejaculation. The influence on semen quality of potentially toxic agents used in the procedure, surgical glove powder and electrolyte cream, was tested; both were detrimental as measured by motility loss. No correlation was found between coagula volume and sperm numbers. The intra- and interanimal variability in semen samples from six monkeys was also evaluated. Penile electroejaculation, combined with control of stimulus current, provides a consistent, successful, and humane method for the collection of semen in the rhesus monkey. Key Words: Spermatozoa, Penile probes, Electroejaculation
INTRODUCTION Electroejaculation is the accepted method of semen collection from nonhuman primates and typically is performed using direct penile (Mastroianni and Manson, 1963; Valerio et al., 1969; Settlage and Hendrickx, 1974; Matsubayashi, 1982; Binor e t al., 1988; Gould and Mann, 1988) or rectal (Weisbroth and Young, 1965; Bennett, 1967; Roussel and Austin, 1967a; Matsubayashi, 1982; Gould and Mann, 1988) probe electrostimulation. Although both techniques are widely used and are associated with specific advantages, there has been a general lack of consistency in efficacy and in the level of detail provided in the published literature for either approach. This report describes the routine collection and processing of semen from rhesus monkeys and evaluation of the resulting sperm preparation. During the evolution of this penile electroejaculation method, from the
O 1990 WILEY-LISS, INC.
original report (Mastroianni and Manson, 1963) a s modified by Hoskins and Patterson (19671, efforts were made to standardize parameters critical to achieving consistent, successful, and humane electroejaculation. The control of stimulus current makes the present approach a highly successful and unique adaptation of the original technique.
MATERIALS AND METHODS Animal Selection Healthy male rhesus monkeys (6-9 kg) were selected from the Oregon Regional Primate Research Center's breeding colony. Before assignment to the project, each animal received a physical examination from the veterinary staff and was sedated with Vetalar (ketamine HC1; 5 mg/kg intramuscularly; ParkeDavis, Morris Plains, NJ) and fitted with a special neck chain to facilitate restraint. All monkeys had their canine teeth capped before use. Animals were individually caged in a temperature-controlled (22"C), lightregulated (12 hr:12 hr, 1ight:dark) room. Diet consisted of Purina monkey chow (Ralston Purina Company, Richmond, IN) supplemented with fresh fruit three times weekly and water ad libitum. Animals assigned to the electroejaculation program were evaluated on the basis of ease of restraint, number of attempts required to obtain a sample, and a qualitative judgment regarding the animal's tolerance of the procedure. If a n animal proved uncooperative or a sample was not obtained after attempts on three separate occasions, it was rejected from the program. Remaining candidates were then accepted or rejected on the basis of a qualitative analysis of their sperm samples. Only monkeys with sperm numbers greater than 100 million per ejaculate and with more than 70% motile cells with normal morphology were acceptable. Apparatus Animals were transferred from their home cage using a n aluminum transfedrestraint cage (model LC-
Received June 13, 1989; accepted August 6, 1989. Address reprint requests to Don P. Wolf, PhD, Oregon Regional Primate Research Center, 505 N.W. 185th Avenue, Beaverton, OR 97006.
62
S.E. LANZENDORF ET AL.
1410; Research Equipment Company, Bryan, TX). voltage setting until ejaculation occurred or until the Electroejaculation was performed in a modified Hazel- maximum output of 80 V was reached. Peak current ton model NSS-3234 primate cage (now Lab Products levels reached a t the 80 V stimulus were approxiInc., Maywood, NJ). The cage front, door, and internal mately 50 mA. Total stimulus time varied from 30 to perches were removed and reinforced anchor points or 90 sec, during which the stimulus voltage was allowed tie-downs were installed in the floor, sides, and back of to reach the maximum (80 V) only for 5-10 sec. Curthe cage. Animals were guided and controlled in this rent was not operator-controlled and varied significage utilizing a n integral guide-tube restraint chain cantly during stimulation, with impedance changes a t that was attached to the neck collar with a trigger-snap the electrode/tissue interface. To determine peak stimulus current using this method, peak stimulus voltage swivel eye. Stimulation was performed using a Grass S-5 (suc- was measured across a 10 ohm precision resistor in ceeded by S-9) stimulator (Grass Instruments, Quincy, series with the animal with a calibrated oscilloscope MA) adjusted to produce bipolar monophasic square isolated from ground. Using Ohm’s Law, measured wave pulses a t 20 Hz with a duration of 20 msec. A voltage divided by a known resistance equals the stimGrass model CCU-1 constant current unit was used to ulating current (Greenberg, 1975). The constant-current method required converting convert the stimulus pulse from constant voltage to constant current. Disposable electrodes were made by the constant-voltage output of the stimulator (set to folding a 6 x 7 cm piece of ordinary light-weight alu- deliver 80 V) to a constant current using a Grass CCUminum foil six times length-wise to produce a strip 1active feedback converter. The peak current output of measuring 1 x 6.0 cm. Disposable rubber surgical the converter (limited to 50 mA) was then controlled by gloves (Surgikos, Inc., Arlington, TX) were worn to the operator independently of impedance changes a t prevent tissue contact and to insulate the operator the tissue/electrode interface and was monitored difrom stimulus current. Gloves were prewashed to re- rectly on the CCU-1 meter. The current output control was adjusted to approximately 1 mA and was left a t move powder. this level for 10-15 sec to act as a “priming stimulus” Electroejaculation to the animal. Current was then slowly and steadily Before each procedure, the bottom of the animal‘s increased until a slight erection, engorgement of the home cage was visually checked for the presence of a glans, or elevation of the testicles into the inguinal white coagulum, evidence of recent masturbation. An- region was observed. Or, if data were available from imals masturbate frequently during the day; therefore, prior stimulation, current was adjusted to a level that semen collections were best performed in the morn- had been previously successful for that particular animal. Stimulus current was left a t this level for a maxings. After the animal was transferred to the electroejac- imum of 30-40 sec. Electrodes used with these animals had 300 ( 2 2 0 ) ulation cage and restrained, electrolyte cream (Redux Creme; Hewlet Packard, Waltham, MA) was applied to mm2 of “contact” surface area. Current densities were the entire shaft of the penis with the exception of the calculated by dividing total measured peak current by glans. One electrode was wrapped around the base of the surface area of one (active) electrode. the penis and the excess length folded to create a tab to Evaluation of the Semen Sample which the negative stimulator lead was connected. The Each sample was allowed to liquefy a t room tempersecond electrode was positioned immediately behind the glans and connected to the positive stimulator lead. ature for approximately 15 min, at which time the liqWith the electrodes attached, the penis was grasped uid portion was aspirated from the coagulum. The volgently between the index and second finger, extended ume was determined, and the semen was transferred to slightly, and positioned over a collection beaker. With a 15 ml centrifuge tube. The coagulum was removed the free hand, the stimulator was activated and the from the container and weighed. The sample was twice stimulus was adjusted and maintained for 30-60 sec. If washed in 3.0 ml Talp-Hepes medium containing 0.3% no ejaculate was obtained, the stimulation was re- bovine serum albumin (Bavister et al., 1983) by cenpeated within 1-3 min. No more than three consecu- trifugation (7 min, 350g). Analyses of motility and tive stimulus attempts were made per animal on a sperm number per ejaculate were performed on the given day. Each animal was given a rest period of a t washed sample. A visual assessment of sperm cell morphology was performed. The washed sample was resusleast 48 h r between sample collections. Electroejaculation was achieved using either a con- pended in Talp-Hepes medium to a concentration of 20 stant-voltage or a constant-current method. Significant x lo6 celldm1 and stored a t room temperature. electrical parameters for both procedures were voltage, Toxicity Testing current, and impedance at the stimulus site. The constant-voltage method involved simply setting Ejaculates collected from three different monkeys the stimulator output to produce appropriate pulses, were tested for the effect of glove powder and electroconnecting the electrodes, and slowly increasing the lyte cream contamination on sperm motility. Each
COLLECTION OF MONKEY SEMEN
63
T A B L E 1. A Comparison of the Electrical Stimulus U s e d With the Constant-Voltage and C o n s t a n t - C u r r e n t Techniques in the Electroejaculation of Five Rhesus Monkeys? Animal 1
Trial 1 2 1 2 1 2 1 2 1 2
2 3 4 5
mean
5
SEM
Constant voltage cp* CD* 50 0.167 38 0.127 32 0.107 41 0.137 15 0.050 20 0.067 35 0.117 50 0.167 30 0.100 39 0.130 35 2 3.61 0.117 t 0.01
S No Yes Yes No Yes Yes Yes No Yes Yes
Constant current CP* CD* 9 0.030 9 0.030 13 0.043 14 0.047 11 0.037 9 0.030 11 0.037 11 0.037 13 0.043 12 0.040 11.2 2 0.57 0.037 2 0.002
S Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
?Abbreviations: Cp, measured peak current (mA); CD, calculated current density (mA/mm2); S, sample obtained (yes or no). *Significantly different between two techniques, P < 0.001.
sample was washed twice in Talp-Hepes medium and resuspended in 6.0 ml fresh medium. Glove powder contamination was evaluated as follows. While the glove was worn, 2.0 ml of medium was poured over the fingers. An aliquot of each sperm sample was added to contaminated media and allowed to incubate 15 min at room temperature. In testing the effect of electrolyte cream contamination, approximately 200 p1 of cream was smeared on the bottom of three 10 ml glass beakers. An aliquot of each sperm sample was poured into the beakers and incubated for 15 min a t room temperature. Following contamination, samples and controls were washed twice and resuspended in fresh medium. All samples were evaluated for motility prior to contamination and each hour for 3 hr following contamination.
Statistical Analysis Data on measured peak current and calculated current density within and between the constant voltage technique and the constant current technique of semen collection were compared by paired t test, as were semen volume, sperm concentration, and sperm motility, between the two techniques. Data on motility were analyzed by split-plot analysis of variance, and differences among means were tested for significance by the use of orthogonal contrast (Steel and Torrie, 1980). Correlation coefficients of semen coagulum, volume, and sperm concentration were determined to assess possible negative relationships.
RESULTS Of the 16 animals assigned to this program over the past 18 months, 11 were released: four due to uncooperative behavior, one because of poor semen quality, and one for postejaculatory urination. An additional five animals were released, for unrelated medical reasons, after being successfully used for various time periods. The remaining five animals are utilized for semen collection approximately two or three timesiweek.
A comparison of the electrical stimulus used with the constant voltage and constant current techniques is summarized in Table 1. Using the constant-current technique, the data demonstrate a n average 68% reduction in stimulus current densities applied to animals accompanied by a 30% increase in successful collection of semen samples. The overall efficiency of the constant-voltage technique was 70% (7/10), compared with 100% (10/10) with the constant-current technique. The failed collection attempts using the constantvoltage technique are unexplained. Since adopting the constant-current technique, we have failed to obtain a semen sample in only two of 30 trials: once due to equipment failure and once due to unrelated animal illness. Additionally, with this technique, animal intratrial variability in stimulus current has been reduced from a n average of 33% to 7%. Comparisons between the two collection techniques in semen volume, sperm number per ejaculate, and motility were performed in three animals (Table 2 ) . All three parameters were similar with the exception of sperm number per ejaculate, which was significantly lower (P < 0.05) in animal 2 with the constant-current technique, and sperm motility, which was significantly lower (P < 0.02) in animal 4 with the constant-current technique. Using either procedure, full penile erection usually did not occur and was not required for ejaculation. Elevation of the testicles into the inguinal region routinely occurred just prior to ejaculation. There were no obvious indications of animal discomfort during these procedures. Urination into the semen sample when it occurred following collection was always noted, since contamination resulted in decreased sperm viability and, therefore, a n inadequate sample. Unlike human semen, monkey semen rarely contains morphologically abnormal sperm cells (Dukelow, 1971). However, morphological sperm variation was observed in these studies. For example, one monkey repeatedly produced sperm cells with a bent midpiece
64
S.E. LANZENDORF ET AL.
TABLE 2. A Comparison of Semen Parameters From Three Rhesus Monkeys Undergoing Constant-Voltageor Constant-CurrentTechniques of Electroejaculationt
Constant voltage Volume No. spermisample
No. of Animal samnles (ml) 0.38 ? 0.05 2 8 0.26 ? 0.06 3 7 0.52 t 0.14 4 6 ~
x 109 384.4 t 74.4* 408.0 107.85 364.7 & 105.9
*
Constant current Motility (%)
No. of
samdes
82.8 k 1.62 80.3 t 1.52 86.8 2 1.35*’$
8 7 6
Volume No. spermisample (ml) ( x 106) 185.4 t 30.6* 0.30 t 0.07 263.0 2 50.8 0.20 2 0.06 304.0 Ifr 24.9 0.23 t 0.06
Motility (%)
78.8 t 3.63 82.4 t 1.59 78.7 2 2.23*”
tValues shown are mean 2 SEM. “Number of spermisample significantly different ( P < 0.05) between two techniques. **Percentageof motile sperm significantly different (P < 0.02) between two techniques.
and coiled tail defect. In addition, electroejaculation in decreases of 40% and 46%, respectively, compared to the afternoon often resulted in samples containing control samples (Fig. 2). high percentages of sperm with cytoplasmic droplets, DISCUSSION suggesting the collection of immature cells. This was associated with monkeys that had masturbated prior to In 1963, Mastroianni and Manson first described the electroejaculation. technique of penile stimulation for semen collection in The intra- and interanimal variability in semen the rhesus monkey. As described, Mastroianni and samples of six monkeys collected using either method Manson’s method utilized a constant-voltage stimulus are presented in Table 3. Both semen volume and total source, with current as the variable. However, current sperm numbers per ejaculate varied within the same (specifically current density a t the electrode) is the critanimal as well as between different animals. Among ical parameter in the stimulation of sympathetic and the six animals, mean volumes ranged from 0.22 to parasympathetic pathways (Dalziel and Phillips, 1948; 0.56 ml, and mean sperm numbers ranged from 287.0 Warner et al., 1974). With a constant voltage source, to 374.0 million. The percent of motile cells ranged stimulus current varies with parameters that effect imfrom a mean of 76.8% to 86.8%. Figure 1 presents the pedance such as skin resistivity, amount and hydration total number of motile cells in each of 10 consecutive of electrolyte cream a t the electrode interface, and ejaculates of two monkeys and demonstrates the vari- quality of electrode contact with skin. Therefore, the ation encountered within a n individual animal over same voltage setting may produce different current time. Mean motilities for both animals were greater levels on different occasions, even with the same anithan 70%. mal, and the success of the technique is significantly To determine if the amount of coagulum produced in dependent on the skill and finesse of the operator. With the sample had a negative effect on the number of so many uncontrolled variables, it is also difficult to sperm per ejaculate, 10 semen samples collected by the replicate stimulus patterns from day to day or between constant current technique from five monkeys were operators or laboratories. evaluated. The weights of the coagula (after removal of In a n effort to overcome these limitations, to stanliquefied semen) ranged from 0.10 g to 1.44 g, with a dardize the procedure, and to improve the percentage of mean value of 0.75 g. Total sperm numbers ranged successful semen collections, a constant current stimfrom 65.0 to 525.0 million, with a mean sperm number ulus method was evaluated. With this methodology, a n per ejaculate of 231.0 million. Statistical analysis effective current range was defined such that, for new showed no correlation between the amount of coagu- animals, a narrow current range known to produce lum present in the sample and the sperm number (r = ejaculation successfully could be selected. If prior stim-0.35). In addition, no correlation was observed be- ulus data were known, previously successful parametween the volume of semen and the sperm number (r = ters could be duplicated to obtain a sample. In addition, -0.053). since current output was limited to a maximum of 50 Toxicity testing was performed on three ejaculates to mA, there was little danger of tissue damage from exdetermine the effect of glove powder and electrolyte cessive stimulus current. cream on monkey sperm motility. Initial motilities perAlthough absolute current levels are controlled with formed on the three samples before contamination the constant-current technique, current density is still were not significantly different (P < 0.01). In all three directly proportional to the surface area of the elecejaculates, sample contamination resulted in a de- trode. Therefore, electrodes of standard dimensions crease in sperm cell motility. This decrease was signif- should be used. Little interanimal variation in penile icantly greater (P < 0.01) than that seen in control size was found, and a 1 x 3 cm active surface area samples taken over the 3 h r time period. After the 3 hr, produced optimal results. glove and electrolyte cream exposure resulted in mean A comparison of semen parameters following collec-
COLLECTION OF MONKEY SEMEN TABLE 3. Intra- and Interanimal Variability in Semen Parameters for Six Rhesus Monkeys Electroejaculated With the Constant-Currentor Constant-Voltage Technique* Number of Animal samples 2" 27 3" 20 4 13 5" 5 6 10 7 5
Volume (ml) 0.25 0.02 0.22 0.03 0.38 ? 0.08 0.48 ? 0.11 0.24 2 0.02 0.56 2 0.12
*
*
Number spermisample ( x lo6) 310.5 2 34.3 330.6 2 46.3 374.0 t 62.6 287.0 2 75.5 350.0 2 55.0 299.1 2 63.8
ACocmd 0
Motility
*Values shown are mean 2 SEM. "Animals proved fertile in IVF program.
oAnirmlNo.2 A A n i n l No. e
Ekctrolyte Cream
O G b v e Powder
(% motile)
84.1 2 1.3 82.6 ? 1.0 82.5 +- 1.6 82.0 f 3.3 76.8 2 3.8 86.8 f 1.2
65
n
dp
Y
75
t
*st 0
1
2
3
Time (hours) Fig. 2. Mean motilities of three ejaculates over a 3 hr period following contamination with electrolyte cream and surgical glove powder.
--a 0
I-
6 1
S
10
Ejaculate Number
Fig. 1. Total number of motile spermatozoa in 10 consecutive ejaculates of two rhesus monkeys.
tion using the two techniques (Table 2 ) revealed a significantly lower (P < 0.05) mean number of sperm per ejaculate in one animal and a significantly lower (P < 0.02) mean motility in another animal with the constant-current technique for electroejaculation. However, semen parameter values were still within acceptable limits for use in research and in vitro fertilization (IVF), and these lower values may reflect intraanimal variation over time. The selection of a tolerant animal was critical in obtaining quality samples, with tolerance evaluated after the first three attempts a t electroejaculation. The skill of the animal handler collecting the sample was also important. The animals should be familiar with the handler, and the handler should be acquainted with the equipment.
With any semen collection technique, sample quality is of major importance. Potential sources of toxic contamination tested here and found detrimental include the powder present on handlers' surgical gloves and the electrolyte cream. Residual powder should be washed off the outside of surgical gloves before collecting the sample, and the glans penis and the inside of the collecting vessel should be kept free of electrolyte cream. Urine contamination of semen is also detrimental to sample quality (pH effect), and animals showing a tendency towards postejaculatory urination may have to be rejected. The semen of most monkeys undergoes rapid coagulation after ejaculation due to a reaction between the secretions of the cranial lobe of the prostate gland and seminal vesicles (van Wagenen, 1936; Dukelow, 1971). The resulting semen sample thus consists of two fractions: a solid, rubbery coagulum, which, over time, undergoes partial liquefaction; and the liquid portion containing sperm. In the rhesus monkey, only a small portion of the semen will liquefy, and the bulk of the sample remains as the solid coagulum (Dukelow, 1971) despite enzymatic treatments (Weisbroth and Young, 1965; Roussel and Austin, 196713, 1968) or sample collection directly into enzyme solution (Hoskins and Patterson, 1967). In the present study, sample liquefaction occurred at room temperature within 15 min of collection, and no correlation was established between coagulum volume and recovered sperm density or quality. The initial description of direct penile electroejaculation in the rhesus monkey (Mastroianni and Manson, 1963) reported sperm concentrations ranging from 93
66
S.E. LANZENDORF ET AL.
to 807 million/ml and motilities ranging from 30% to 99%. Roussel and Austin (1967a) also reported a wide range of values for motility ( 1 0 4 5 % ) and sperm numbers (100-3,600 million/ejaculate). The present study also demonstrates variations in semen characteristics within and between animals but with values that consistently fall within acceptable limits for use in research and IVF. The described technique for semen collection and preparation provides quality spermatozoa for IVF and the development of cryopreservation protocols. Sperm preparation for IVF is similar to that previously described by Bavister et al. (1983), with washed sperm “activated” by exposure to dibutyryl cyclic adenosine monophosphate and caffeine. Oocytes are routinely inseminated with 50,000 motile spermatozoa in 1.0 ml culture medium (Wolf e t al., 1989). Recently, four embryo-transfer procedures were performed with embryos fertilized with sperm collected by the constant-current electrostimulation method. Two embryos were transferred to each of four animals; three pregnancies are currently ongoing from this trial.
ACKNOWLEDGMENTS The authors thank the animal technicians, Rick Jones, Kevin Grund, and Randy Shoemake for their skilled handling of the animals and Patsy Kimzey for assisting in the preparation of the manuscript. This research (Publication No. 1683 of the Oregon Regional Primate Research Center) was supported by NIH grants RR00163 and HD23786.
REFERENCES Bavister BD, Boatman DE, Leibfried L, Loose M, Vernon MW (1983): Fertilization and cleavage of rhesus monkey oocytes in vitro. Biol Reprod 28:983-999. Bennett JP (1967): Semen collection in the squirrel monkey. J Reprod Fertil 13:353-355.
Binor Z, Rawlins RG, Van der Ven H, Dmowski WP (1988): Rhesus monkey sperm penetration into zona-free hamster ova: Comparison of preparation and culture conditions. Gamete Res 19:91-99. Dalziel CF, Phillips CL (1948):Electric ejaculation. Determination of optimum electric shock to produce ejaculation in chinchillas and guinea pigs. Am J Vet Res 9:225-232. Dukelow WR (1971):Semen and artificial insemination. In ESE Hafez (ed): “Comparative Reproduction of Nonhuman Primates.” Springfield, IL: Charles C. Thomas, pp 115-127. Gould KG, Mann DR (1988): Comparison of electrostimulation methods for semen recovery in the rhesus monkey (Macaca mulattaj. J Med Primatol 17:95-103. Greenberg LH (1975): “Physics for Biology and Pre-Med Students.” Philadelphia: W.B. Saunders Company, pp 470-472. Hoskins DD, Patterson DL (1967): Prevention of coagulum formation with recovery of motile spermatozoa from rhesus monkey semen. J Reprod Fertil 13:337-340. Mastroianni L J r , Manson WA Jr (1963):Collection of monkey semen by electroejaculation. Proc SOC Exp Biol Med 112:1025-1027. Matsubayashi K (1982): Comparison of the two methods of electroejaculation in the Japanese monkey (Mucaca fuscatal. Exp Anim (Tokyo) 31:l-6. Roussel JD, Austin CR (1967a): Preservation of primate spermatozoa by freezing. J Reprod Fertil 13:333-335. Roussel JD, Austin CR (196713): Enzymatic liquefaction of primate semen. Int J Fertil 12:288-290. Roussel JD, Austin CR (1968): Improved electro-ejaculation of primates. J Inst Anim Techno1 19:22-32. Settlage DSF, Hendrickx AG (1974): Electroejaculation technique in Macaca mulatta (rhesus monkeys). Fertil Steril 25:157-159. Steel RGD, Torrie J H (1980): “Principles and Procedures of Statistics, A Biometrical Approach.” New York: McGraw-Hill Book Company, pp 383-387. Valerio DA, Ellis EB, Clark ML, Thompson GE (1969): Collection of semen from macaques by electroejaculation. Lab Anim Care 1 9 250-252. Van Wagenen G (1936): The coagulating function of the cranial lobe of the prostate gland in the monkey. Anat Rec 66:411-421. Warner H, Martin DE, Keeling ME (1974). Electroejaculation of the great apes. Ann Biomed Eng 2:419-432. Weisbroth S, Young FA (1965): The collection of primate semen by electro-ejaculation. Fertil Steril 16229-235, Wolf DP, VandeVoort CA, Meyer-Haas GR, Zelinski-Wooten MB, Hess DL, Baughman WL, Stouffer RL (1989):In vitro fertilization and embryo transfer in the rhesus monkey. Biol Reprod 41:335346.
Collection and Quality of Rhesus Monkey Semen S.E. LANZENDORF,' P.M. GLIESSMAN? A.E. ARCHIBONG,l M. ALEXANDER,' AND D.P. WOLF' 'Division of Reproductive Biology and Behavior, 'Biomedical Engineering, Oregon Regional Primate Research Center, Beauerton, Oregon ABSTRACT Electroejaculation is an accepted method of semen collection from nonhuman primates. Although both penile and rectal probe stimulation techniques have been used, there has been a general lack of consistency and detail regarding their applicotion. This report describes the collection, processing, and evaluation of rhesus monkey semen contrasting two methods of penile electroejaculation: 1 ) a constant-voltage method where stimulus current is a variable and 2) a constant-current method where stimulus current is operator-controlled. The constant-current method was the more efficient procedure, requiring a lower stimulus current for successful electroejaculation. The influence on semen quality of potentially toxic agents used in the procedure, surgical glove powder and electrolyte cream, was tested; both were detrimental as measured by motility loss. No correlation was found between coagula volume and sperm numbers. The intra- and interanimal variability in semen samples from six monkeys was also evaluated. Penile electroejaculation, combined with control of stimulus current, provides a consistent, successful, and humane method for the collection of semen in the rhesus monkey. Key Words: Spermatozoa, Penile probes, Electroejaculation
INTRODUCTION Electroejaculation is the accepted method of semen collection from nonhuman primates and typically is performed using direct penile (Mastroianni and Manson, 1963; Valerio et al., 1969; Settlage and Hendrickx, 1974; Matsubayashi, 1982; Binor e t al., 1988; Gould and Mann, 1988) or rectal (Weisbroth and Young, 1965; Bennett, 1967; Roussel and Austin, 1967a; Matsubayashi, 1982; Gould and Mann, 1988) probe electrostimulation. Although both techniques are widely used and are associated with specific advantages, there has been a general lack of consistency in efficacy and in the level of detail provided in the published literature for either approach. This report describes the routine collection and processing of semen from rhesus monkeys and evaluation of the resulting sperm preparation. During the evolution of this penile electroejaculation method, from the
O 1990 WILEY-LISS, INC.
original report (Mastroianni and Manson, 1963) a s modified by Hoskins and Patterson (19671, efforts were made to standardize parameters critical to achieving consistent, successful, and humane electroejaculation. The control of stimulus current makes the present approach a highly successful and unique adaptation of the original technique.
MATERIALS AND METHODS Animal Selection Healthy male rhesus monkeys (6-9 kg) were selected from the Oregon Regional Primate Research Center's breeding colony. Before assignment to the project, each animal received a physical examination from the veterinary staff and was sedated with Vetalar (ketamine HC1; 5 mg/kg intramuscularly; ParkeDavis, Morris Plains, NJ) and fitted with a special neck chain to facilitate restraint. All monkeys had their canine teeth capped before use. Animals were individually caged in a temperature-controlled (22"C), lightregulated (12 hr:12 hr, 1ight:dark) room. Diet consisted of Purina monkey chow (Ralston Purina Company, Richmond, IN) supplemented with fresh fruit three times weekly and water ad libitum. Animals assigned to the electroejaculation program were evaluated on the basis of ease of restraint, number of attempts required to obtain a sample, and a qualitative judgment regarding the animal's tolerance of the procedure. If a n animal proved uncooperative or a sample was not obtained after attempts on three separate occasions, it was rejected from the program. Remaining candidates were then accepted or rejected on the basis of a qualitative analysis of their sperm samples. Only monkeys with sperm numbers greater than 100 million per ejaculate and with more than 70% motile cells with normal morphology were acceptable. Apparatus Animals were transferred from their home cage using a n aluminum transfedrestraint cage (model LC-
Received June 13, 1989; accepted August 6, 1989. Address reprint requests to Don P. Wolf, PhD, Oregon Regional Primate Research Center, 505 N.W. 185th Avenue, Beaverton, OR 97006.
62
S.E. LANZENDORF ET AL.
1410; Research Equipment Company, Bryan, TX). voltage setting until ejaculation occurred or until the Electroejaculation was performed in a modified Hazel- maximum output of 80 V was reached. Peak current ton model NSS-3234 primate cage (now Lab Products levels reached a t the 80 V stimulus were approxiInc., Maywood, NJ). The cage front, door, and internal mately 50 mA. Total stimulus time varied from 30 to perches were removed and reinforced anchor points or 90 sec, during which the stimulus voltage was allowed tie-downs were installed in the floor, sides, and back of to reach the maximum (80 V) only for 5-10 sec. Curthe cage. Animals were guided and controlled in this rent was not operator-controlled and varied significage utilizing a n integral guide-tube restraint chain cantly during stimulation, with impedance changes a t that was attached to the neck collar with a trigger-snap the electrode/tissue interface. To determine peak stimulus current using this method, peak stimulus voltage swivel eye. Stimulation was performed using a Grass S-5 (suc- was measured across a 10 ohm precision resistor in ceeded by S-9) stimulator (Grass Instruments, Quincy, series with the animal with a calibrated oscilloscope MA) adjusted to produce bipolar monophasic square isolated from ground. Using Ohm’s Law, measured wave pulses a t 20 Hz with a duration of 20 msec. A voltage divided by a known resistance equals the stimGrass model CCU-1 constant current unit was used to ulating current (Greenberg, 1975). The constant-current method required converting convert the stimulus pulse from constant voltage to constant current. Disposable electrodes were made by the constant-voltage output of the stimulator (set to folding a 6 x 7 cm piece of ordinary light-weight alu- deliver 80 V) to a constant current using a Grass CCUminum foil six times length-wise to produce a strip 1active feedback converter. The peak current output of measuring 1 x 6.0 cm. Disposable rubber surgical the converter (limited to 50 mA) was then controlled by gloves (Surgikos, Inc., Arlington, TX) were worn to the operator independently of impedance changes a t prevent tissue contact and to insulate the operator the tissue/electrode interface and was monitored difrom stimulus current. Gloves were prewashed to re- rectly on the CCU-1 meter. The current output control was adjusted to approximately 1 mA and was left a t move powder. this level for 10-15 sec to act as a “priming stimulus” Electroejaculation to the animal. Current was then slowly and steadily Before each procedure, the bottom of the animal‘s increased until a slight erection, engorgement of the home cage was visually checked for the presence of a glans, or elevation of the testicles into the inguinal white coagulum, evidence of recent masturbation. An- region was observed. Or, if data were available from imals masturbate frequently during the day; therefore, prior stimulation, current was adjusted to a level that semen collections were best performed in the morn- had been previously successful for that particular animal. Stimulus current was left a t this level for a maxings. After the animal was transferred to the electroejac- imum of 30-40 sec. Electrodes used with these animals had 300 ( 2 2 0 ) ulation cage and restrained, electrolyte cream (Redux Creme; Hewlet Packard, Waltham, MA) was applied to mm2 of “contact” surface area. Current densities were the entire shaft of the penis with the exception of the calculated by dividing total measured peak current by glans. One electrode was wrapped around the base of the surface area of one (active) electrode. the penis and the excess length folded to create a tab to Evaluation of the Semen Sample which the negative stimulator lead was connected. The Each sample was allowed to liquefy a t room tempersecond electrode was positioned immediately behind the glans and connected to the positive stimulator lead. ature for approximately 15 min, at which time the liqWith the electrodes attached, the penis was grasped uid portion was aspirated from the coagulum. The volgently between the index and second finger, extended ume was determined, and the semen was transferred to slightly, and positioned over a collection beaker. With a 15 ml centrifuge tube. The coagulum was removed the free hand, the stimulator was activated and the from the container and weighed. The sample was twice stimulus was adjusted and maintained for 30-60 sec. If washed in 3.0 ml Talp-Hepes medium containing 0.3% no ejaculate was obtained, the stimulation was re- bovine serum albumin (Bavister et al., 1983) by cenpeated within 1-3 min. No more than three consecu- trifugation (7 min, 350g). Analyses of motility and tive stimulus attempts were made per animal on a sperm number per ejaculate were performed on the given day. Each animal was given a rest period of a t washed sample. A visual assessment of sperm cell morphology was performed. The washed sample was resusleast 48 h r between sample collections. Electroejaculation was achieved using either a con- pended in Talp-Hepes medium to a concentration of 20 stant-voltage or a constant-current method. Significant x lo6 celldm1 and stored a t room temperature. electrical parameters for both procedures were voltage, Toxicity Testing current, and impedance at the stimulus site. The constant-voltage method involved simply setting Ejaculates collected from three different monkeys the stimulator output to produce appropriate pulses, were tested for the effect of glove powder and electroconnecting the electrodes, and slowly increasing the lyte cream contamination on sperm motility. Each
COLLECTION OF MONKEY SEMEN
63
T A B L E 1. A Comparison of the Electrical Stimulus U s e d With the Constant-Voltage and C o n s t a n t - C u r r e n t Techniques in the Electroejaculation of Five Rhesus Monkeys? Animal 1
Trial 1 2 1 2 1 2 1 2 1 2
2 3 4 5
mean
5
SEM
Constant voltage cp* CD* 50 0.167 38 0.127 32 0.107 41 0.137 15 0.050 20 0.067 35 0.117 50 0.167 30 0.100 39 0.130 35 2 3.61 0.117 t 0.01
S No Yes Yes No Yes Yes Yes No Yes Yes
Constant current CP* CD* 9 0.030 9 0.030 13 0.043 14 0.047 11 0.037 9 0.030 11 0.037 11 0.037 13 0.043 12 0.040 11.2 2 0.57 0.037 2 0.002
S Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
?Abbreviations: Cp, measured peak current (mA); CD, calculated current density (mA/mm2); S, sample obtained (yes or no). *Significantly different between two techniques, P < 0.001.
sample was washed twice in Talp-Hepes medium and resuspended in 6.0 ml fresh medium. Glove powder contamination was evaluated as follows. While the glove was worn, 2.0 ml of medium was poured over the fingers. An aliquot of each sperm sample was added to contaminated media and allowed to incubate 15 min at room temperature. In testing the effect of electrolyte cream contamination, approximately 200 p1 of cream was smeared on the bottom of three 10 ml glass beakers. An aliquot of each sperm sample was poured into the beakers and incubated for 15 min a t room temperature. Following contamination, samples and controls were washed twice and resuspended in fresh medium. All samples were evaluated for motility prior to contamination and each hour for 3 hr following contamination.
Statistical Analysis Data on measured peak current and calculated current density within and between the constant voltage technique and the constant current technique of semen collection were compared by paired t test, as were semen volume, sperm concentration, and sperm motility, between the two techniques. Data on motility were analyzed by split-plot analysis of variance, and differences among means were tested for significance by the use of orthogonal contrast (Steel and Torrie, 1980). Correlation coefficients of semen coagulum, volume, and sperm concentration were determined to assess possible negative relationships.
RESULTS Of the 16 animals assigned to this program over the past 18 months, 11 were released: four due to uncooperative behavior, one because of poor semen quality, and one for postejaculatory urination. An additional five animals were released, for unrelated medical reasons, after being successfully used for various time periods. The remaining five animals are utilized for semen collection approximately two or three timesiweek.
A comparison of the electrical stimulus used with the constant voltage and constant current techniques is summarized in Table 1. Using the constant-current technique, the data demonstrate a n average 68% reduction in stimulus current densities applied to animals accompanied by a 30% increase in successful collection of semen samples. The overall efficiency of the constant-voltage technique was 70% (7/10), compared with 100% (10/10) with the constant-current technique. The failed collection attempts using the constantvoltage technique are unexplained. Since adopting the constant-current technique, we have failed to obtain a semen sample in only two of 30 trials: once due to equipment failure and once due to unrelated animal illness. Additionally, with this technique, animal intratrial variability in stimulus current has been reduced from a n average of 33% to 7%. Comparisons between the two collection techniques in semen volume, sperm number per ejaculate, and motility were performed in three animals (Table 2 ) . All three parameters were similar with the exception of sperm number per ejaculate, which was significantly lower (P < 0.05) in animal 2 with the constant-current technique, and sperm motility, which was significantly lower (P < 0.02) in animal 4 with the constant-current technique. Using either procedure, full penile erection usually did not occur and was not required for ejaculation. Elevation of the testicles into the inguinal region routinely occurred just prior to ejaculation. There were no obvious indications of animal discomfort during these procedures. Urination into the semen sample when it occurred following collection was always noted, since contamination resulted in decreased sperm viability and, therefore, a n inadequate sample. Unlike human semen, monkey semen rarely contains morphologically abnormal sperm cells (Dukelow, 1971). However, morphological sperm variation was observed in these studies. For example, one monkey repeatedly produced sperm cells with a bent midpiece
64
S.E. LANZENDORF ET AL.
TABLE 2. A Comparison of Semen Parameters From Three Rhesus Monkeys Undergoing Constant-Voltageor Constant-CurrentTechniques of Electroejaculationt
Constant voltage Volume No. spermisample
No. of Animal samnles (ml) 0.38 ? 0.05 2 8 0.26 ? 0.06 3 7 0.52 t 0.14 4 6 ~
x 109 384.4 t 74.4* 408.0 107.85 364.7 & 105.9
*
Constant current Motility (%)
No. of
samdes
82.8 k 1.62 80.3 t 1.52 86.8 2 1.35*’$
8 7 6
Volume No. spermisample (ml) ( x 106) 185.4 t 30.6* 0.30 t 0.07 263.0 2 50.8 0.20 2 0.06 304.0 Ifr 24.9 0.23 t 0.06
Motility (%)
78.8 t 3.63 82.4 t 1.59 78.7 2 2.23*”
tValues shown are mean 2 SEM. “Number of spermisample significantly different ( P < 0.05) between two techniques. **Percentageof motile sperm significantly different (P < 0.02) between two techniques.
and coiled tail defect. In addition, electroejaculation in decreases of 40% and 46%, respectively, compared to the afternoon often resulted in samples containing control samples (Fig. 2). high percentages of sperm with cytoplasmic droplets, DISCUSSION suggesting the collection of immature cells. This was associated with monkeys that had masturbated prior to In 1963, Mastroianni and Manson first described the electroejaculation. technique of penile stimulation for semen collection in The intra- and interanimal variability in semen the rhesus monkey. As described, Mastroianni and samples of six monkeys collected using either method Manson’s method utilized a constant-voltage stimulus are presented in Table 3. Both semen volume and total source, with current as the variable. However, current sperm numbers per ejaculate varied within the same (specifically current density a t the electrode) is the critanimal as well as between different animals. Among ical parameter in the stimulation of sympathetic and the six animals, mean volumes ranged from 0.22 to parasympathetic pathways (Dalziel and Phillips, 1948; 0.56 ml, and mean sperm numbers ranged from 287.0 Warner et al., 1974). With a constant voltage source, to 374.0 million. The percent of motile cells ranged stimulus current varies with parameters that effect imfrom a mean of 76.8% to 86.8%. Figure 1 presents the pedance such as skin resistivity, amount and hydration total number of motile cells in each of 10 consecutive of electrolyte cream a t the electrode interface, and ejaculates of two monkeys and demonstrates the vari- quality of electrode contact with skin. Therefore, the ation encountered within a n individual animal over same voltage setting may produce different current time. Mean motilities for both animals were greater levels on different occasions, even with the same anithan 70%. mal, and the success of the technique is significantly To determine if the amount of coagulum produced in dependent on the skill and finesse of the operator. With the sample had a negative effect on the number of so many uncontrolled variables, it is also difficult to sperm per ejaculate, 10 semen samples collected by the replicate stimulus patterns from day to day or between constant current technique from five monkeys were operators or laboratories. evaluated. The weights of the coagula (after removal of In a n effort to overcome these limitations, to stanliquefied semen) ranged from 0.10 g to 1.44 g, with a dardize the procedure, and to improve the percentage of mean value of 0.75 g. Total sperm numbers ranged successful semen collections, a constant current stimfrom 65.0 to 525.0 million, with a mean sperm number ulus method was evaluated. With this methodology, a n per ejaculate of 231.0 million. Statistical analysis effective current range was defined such that, for new showed no correlation between the amount of coagu- animals, a narrow current range known to produce lum present in the sample and the sperm number (r = ejaculation successfully could be selected. If prior stim-0.35). In addition, no correlation was observed be- ulus data were known, previously successful parametween the volume of semen and the sperm number (r = ters could be duplicated to obtain a sample. In addition, -0.053). since current output was limited to a maximum of 50 Toxicity testing was performed on three ejaculates to mA, there was little danger of tissue damage from exdetermine the effect of glove powder and electrolyte cessive stimulus current. cream on monkey sperm motility. Initial motilities perAlthough absolute current levels are controlled with formed on the three samples before contamination the constant-current technique, current density is still were not significantly different (P < 0.01). In all three directly proportional to the surface area of the elecejaculates, sample contamination resulted in a de- trode. Therefore, electrodes of standard dimensions crease in sperm cell motility. This decrease was signif- should be used. Little interanimal variation in penile icantly greater (P < 0.01) than that seen in control size was found, and a 1 x 3 cm active surface area samples taken over the 3 h r time period. After the 3 hr, produced optimal results. glove and electrolyte cream exposure resulted in mean A comparison of semen parameters following collec-
COLLECTION OF MONKEY SEMEN TABLE 3. Intra- and Interanimal Variability in Semen Parameters for Six Rhesus Monkeys Electroejaculated With the Constant-Currentor Constant-Voltage Technique* Number of Animal samples 2" 27 3" 20 4 13 5" 5 6 10 7 5
Volume (ml) 0.25 0.02 0.22 0.03 0.38 ? 0.08 0.48 ? 0.11 0.24 2 0.02 0.56 2 0.12
*
*
Number spermisample ( x lo6) 310.5 2 34.3 330.6 2 46.3 374.0 t 62.6 287.0 2 75.5 350.0 2 55.0 299.1 2 63.8
ACocmd 0
Motility
*Values shown are mean 2 SEM. "Animals proved fertile in IVF program.
oAnirmlNo.2 A A n i n l No. e
Ekctrolyte Cream
O G b v e Powder
(% motile)
84.1 2 1.3 82.6 ? 1.0 82.5 +- 1.6 82.0 f 3.3 76.8 2 3.8 86.8 f 1.2
65
n
dp
Y
75
t
*st 0
1
2
3
Time (hours) Fig. 2. Mean motilities of three ejaculates over a 3 hr period following contamination with electrolyte cream and surgical glove powder.
--a 0
I-
6 1
S
10
Ejaculate Number
Fig. 1. Total number of motile spermatozoa in 10 consecutive ejaculates of two rhesus monkeys.
tion using the two techniques (Table 2 ) revealed a significantly lower (P < 0.05) mean number of sperm per ejaculate in one animal and a significantly lower (P < 0.02) mean motility in another animal with the constant-current technique for electroejaculation. However, semen parameter values were still within acceptable limits for use in research and in vitro fertilization (IVF), and these lower values may reflect intraanimal variation over time. The selection of a tolerant animal was critical in obtaining quality samples, with tolerance evaluated after the first three attempts a t electroejaculation. The skill of the animal handler collecting the sample was also important. The animals should be familiar with the handler, and the handler should be acquainted with the equipment.
With any semen collection technique, sample quality is of major importance. Potential sources of toxic contamination tested here and found detrimental include the powder present on handlers' surgical gloves and the electrolyte cream. Residual powder should be washed off the outside of surgical gloves before collecting the sample, and the glans penis and the inside of the collecting vessel should be kept free of electrolyte cream. Urine contamination of semen is also detrimental to sample quality (pH effect), and animals showing a tendency towards postejaculatory urination may have to be rejected. The semen of most monkeys undergoes rapid coagulation after ejaculation due to a reaction between the secretions of the cranial lobe of the prostate gland and seminal vesicles (van Wagenen, 1936; Dukelow, 1971). The resulting semen sample thus consists of two fractions: a solid, rubbery coagulum, which, over time, undergoes partial liquefaction; and the liquid portion containing sperm. In the rhesus monkey, only a small portion of the semen will liquefy, and the bulk of the sample remains as the solid coagulum (Dukelow, 1971) despite enzymatic treatments (Weisbroth and Young, 1965; Roussel and Austin, 196713, 1968) or sample collection directly into enzyme solution (Hoskins and Patterson, 1967). In the present study, sample liquefaction occurred at room temperature within 15 min of collection, and no correlation was established between coagulum volume and recovered sperm density or quality. The initial description of direct penile electroejaculation in the rhesus monkey (Mastroianni and Manson, 1963) reported sperm concentrations ranging from 93
66
S.E. LANZENDORF ET AL.
to 807 million/ml and motilities ranging from 30% to 99%. Roussel and Austin (1967a) also reported a wide range of values for motility ( 1 0 4 5 % ) and sperm numbers (100-3,600 million/ejaculate). The present study also demonstrates variations in semen characteristics within and between animals but with values that consistently fall within acceptable limits for use in research and IVF. The described technique for semen collection and preparation provides quality spermatozoa for IVF and the development of cryopreservation protocols. Sperm preparation for IVF is similar to that previously described by Bavister et al. (1983), with washed sperm “activated” by exposure to dibutyryl cyclic adenosine monophosphate and caffeine. Oocytes are routinely inseminated with 50,000 motile spermatozoa in 1.0 ml culture medium (Wolf e t al., 1989). Recently, four embryo-transfer procedures were performed with embryos fertilized with sperm collected by the constant-current electrostimulation method. Two embryos were transferred to each of four animals; three pregnancies are currently ongoing from this trial.
ACKNOWLEDGMENTS The authors thank the animal technicians, Rick Jones, Kevin Grund, and Randy Shoemake for their skilled handling of the animals and Patsy Kimzey for assisting in the preparation of the manuscript. This research (Publication No. 1683 of the Oregon Regional Primate Research Center) was supported by NIH grants RR00163 and HD23786.
REFERENCES Bavister BD, Boatman DE, Leibfried L, Loose M, Vernon MW (1983): Fertilization and cleavage of rhesus monkey oocytes in vitro. Biol Reprod 28:983-999. Bennett JP (1967): Semen collection in the squirrel monkey. J Reprod Fertil 13:353-355.
Binor Z, Rawlins RG, Van der Ven H, Dmowski WP (1988): Rhesus monkey sperm penetration into zona-free hamster ova: Comparison of preparation and culture conditions. Gamete Res 19:91-99. Dalziel CF, Phillips CL (1948):Electric ejaculation. Determination of optimum electric shock to produce ejaculation in chinchillas and guinea pigs. Am J Vet Res 9:225-232. Dukelow WR (1971):Semen and artificial insemination. In ESE Hafez (ed): “Comparative Reproduction of Nonhuman Primates.” Springfield, IL: Charles C. Thomas, pp 115-127. Gould KG, Mann DR (1988): Comparison of electrostimulation methods for semen recovery in the rhesus monkey (Macaca mulattaj. J Med Primatol 17:95-103. Greenberg LH (1975): “Physics for Biology and Pre-Med Students.” Philadelphia: W.B. Saunders Company, pp 470-472. Hoskins DD, Patterson DL (1967): Prevention of coagulum formation with recovery of motile spermatozoa from rhesus monkey semen. J Reprod Fertil 13:337-340. Mastroianni L J r , Manson WA Jr (1963):Collection of monkey semen by electroejaculation. Proc SOC Exp Biol Med 112:1025-1027. Matsubayashi K (1982): Comparison of the two methods of electroejaculation in the Japanese monkey (Mucaca fuscatal. Exp Anim (Tokyo) 31:l-6. Roussel JD, Austin CR (1967a): Preservation of primate spermatozoa by freezing. J Reprod Fertil 13:333-335. Roussel JD, Austin CR (196713): Enzymatic liquefaction of primate semen. Int J Fertil 12:288-290. Roussel JD, Austin CR (1968): Improved electro-ejaculation of primates. J Inst Anim Techno1 19:22-32. Settlage DSF, Hendrickx AG (1974): Electroejaculation technique in Macaca mulatta (rhesus monkeys). Fertil Steril 25:157-159. Steel RGD, Torrie J H (1980): “Principles and Procedures of Statistics, A Biometrical Approach.” New York: McGraw-Hill Book Company, pp 383-387. Valerio DA, Ellis EB, Clark ML, Thompson GE (1969): Collection of semen from macaques by electroejaculation. Lab Anim Care 1 9 250-252. Van Wagenen G (1936): The coagulating function of the cranial lobe of the prostate gland in the monkey. Anat Rec 66:411-421. Warner H, Martin DE, Keeling ME (1974). Electroejaculation of the great apes. Ann Biomed Eng 2:419-432. Weisbroth S, Young FA (1965): The collection of primate semen by electro-ejaculation. Fertil Steril 16229-235, Wolf DP, VandeVoort CA, Meyer-Haas GR, Zelinski-Wooten MB, Hess DL, Baughman WL, Stouffer RL (1989):In vitro fertilization and embryo transfer in the rhesus monkey. Biol Reprod 41:335346.
