THE JOURNAL OF EXPERIMENTAL ZOOLOGY 260382-390 (1991)
Genetic Analysis of Androgenetic Rainbow Trout PAUL D. SCHEERER, GARY H. THORGAARD, FRED W. ALLENDORF Department of Zoology and Program in Genetics and Cell Biology, Washington State University, Pullman, Washington 991 64-4236 (P.D.S., G.H.T.); Division of Biological Sciences, University of Montana, Missoula, Montana 59812 (F.W.A.) AND
We analyzed a number of genetic characteristics in androgenetic rainbow trout ABSTRACT (Oncorhynchus mykiss) and their progeny. The androgenetic progeny of individual androgenetic males appeared genetically identical to each other based on eight enzyme loci. Their viability was no higher than that of androgenetic progeny of outbred males. Homozygous androgenetic female rainbow trout produced very poor quality eggs. When common eggs and sperm from outbred individuals were used to produce androgenetic and gynogenetic progeny, the yield of gynogenetic progeny was higher but some were heterozygous at protein loci, while no androgenetic progeny were heterozygous. Some androgenetic diploid rainbow trout were successfully produced from cryopreserved sperm. The progeny of some androgenetic males crossed to normal females were virtually all males, while the progeny of other males were virtually all females. This suggests that both XX and YY androgenetic individuals may develop as males. Androgenesis is likely to be useful for generating homozygous clones for research and for recovering strains from cryopreserved sperm.
Isogenic lines of animals and plants are valu- androgenesis (all-paternal inheritance) (Scheerer able research tools. Inbred lines and hybrids be- et al., '86; May et al., '88). Androgenesis involves tween lines are useful in genetic, developmental inactivating the maternal genetic material by usbiology, immunology, toxicology, and cancer re- ing gamma radiation (Romashow and Belyaeva, search (Allard, '60; Festing, '79; Green, '81; Ab- '64; Purdom, '69; Arai et al., '79; Parsons and planalp, '86). However, the common method of Thorgaard, '841, fertilizing these treated eggs generating inbred strains involves 10-20 genera- with normal sperm, and restoring diploidy by suptions of sib-mating, requiring substantial com- pressing the first cleavage by using hydrostatic pressure (Onozato, '82, in Yamazaki, '83; Parsons mitments of both time and money. Animal researchers have recently used alter- and Thorgaard, '85). Production of homozygous nate approaches for rapidly producing inbred rainbow trout by androgenesis has been generally lines in several species. These approaches involve successful (Parsons and Thorgaard, '85; Scheerer inactivating the entire chromosome set from one et al., '86). Spontaneous heterozygous individuals parent followed by duplicating the remaining set. have not been observed, and both the XX and YY Homozygous frogs have been produced by block- offspring appear to be viable (Parsons and Thoring the first cleavage of parthenogenetically acti- gaard, '85). The gametes from these androgenetic vated haploid eggs (Reinschmidt et al., '79). progeny can, in turn, be used t o found isogenic Clonal lines of two aquarium fish species have lines. Androgenesis could also facilitate the genbeen produced gynogenetically by blocking the eration of YY males which could be useful in first cleavage of eggs fertilized with UV-inacti- aquaculture programs where production of allvated sperm, followed by producing gynogenetic male populations can be desirable (Scott et al., progeny from the homozygous individuals (Strei- '$9). singer et al., '81; Naruse et al., '85). This approach has been unsuccessful in mammals, where both the maternal and paternal pronuclei must apparReceived January 11, 1991; revision accepted February 15, 1991. ently be active in the developing embryo (McAddress reprint requests to Gary H. Thorgaard, Department of Zoology and Program in Genetics and Cell Biology, Washington State Grath and Solter, '84). Pullman, WA 99164-4236. A promising approach for generating inbred University, Paul D. Scheerer'spresent address is 24886 Grange Hall Rd., Philolines in fish and amphibians involves inducing math, OR 97370. 0 1991 WILEY-LISS,
INC.
ANDROGENESIS IN RAINBOW TROUT
This study used several approaches to genetically analyze androgenetic rainbow trout (Oncorhynchus mykiss) and their progeny. Androgenetic clones of rainbow trout were generated by using sperm from homozygous androgenetic males and their survival was compared to that of androgenetic trout produced from sperm from outbred males. Gynogenetic offspring were also generated by using eggs from homozygous androgenetic females. The yields of homozygous diploid androgenetic and gynogenetic progeny produced by using the same eggs, sperm, and pressure treatments were compared. Androgenetic offspring were produced from cryopreserved trout semen, demonstrating the potential application of this approach to gene banking. Finally, we examined the sex ratios in the progeny of androgenetic males crossed t o normal females.
MATERIALS AND METHODS Production of homozygous clones by androgenesis Rainbow trout sperm was obtained in December 1986 from six homozygous androgenetic males and six outbred males. The outbred strains included the Arlee strain from the Jocko River Trout Hatchery (Montana Department of Fish, Wildlife and Parks), the Spokane Hatchery strain (Washington Department of Wildlife), an albino heterozygote (Albino from the Egan, Utah State Fish Hatchery x Spokane rainbow), and a palomino rainbow (a co-dominant golden from The Pennsylvania State University x Spokane rainbow) reared at Washington State University. The homozygous androgenetic males were produced from the Arlee, Spokane, and Hot Creek (California Department of Fish and Game) strains (see Scheerer et al., '86). Eggs were obtained from six females from the Spokane hatchery. Gametes were transported on ice to Washington State University. Androgenetic progeny were produced from both outbred and androgenetic male parents. Eggs from the six rainbow trout females were used separately. The majority of the eggs from each female were exposed to 60cobaltgamma radiation at the Washington State University Nuclear Radiation Center to a final dose of 3.6 x lo4 R (Parsons and Thorgaard, '84); a portion of each egg lot was left unirradiated. The irradiated eggs were divided into two treatment groups and fertilized separately with sperm from one outbred and one androgenetic male. The unirradiated eggs were
383
also divided into two groups and fertilized with sperm from the same males, serving as diploid controls. After fertilization, all egg lots were placed in an incubator for further development at 7-9°C. A portion of each irradiated egg lot was separated off to serve as androgenetic haploid controls; the remainder were divided into two groups. Both groups were treated with hydrostatic pressure at 9,000 psi for a 3 minute duration to block the first cleavage division and restore diploidy (Parsons and Thorgaard, '85). One group was pressure treated at 320 minutes post-fertilization and the other at 340 minutes post-fertilization. The two times of application of hydrostatic pressure were used t o increase the chance of blocking the first cleavage since the interval between fertilization and the first cleavage division varies between strains and between females within strains (unpublished data). Relative survival rates of the androgenetic diploid lots were calculated as a percentage survival compared to the control survival (survival androgenetic lot/survival control lot). Survival was monitored at the eyed-egg stage (20 days), at hatching (35 days), and at the initiation of feeding (62 days). Tissues from a sample of androgenetic fish were removed and stored at - 70°C. Tissue extracts were later prepared and analyzed by horizontal starch-gel electrophoresis as described by Allendorf et al. ('86) t o test homozygosity and the clonal nature of the offspring. We attempted to score the androgenetic progeny for genetic variation at the following enzyme loci: AAT-3; HEX; IDH-2; IDH3; LDH-4; Mdh-B; PGM-2; SOD-1. Not all progeny could be scored at all loci because some fish were sampled at a very young age and some were dead for a period of time before being frozen. Five embryos from two separate lots of both outbred androgenetic diploids and androgenetic clones were sacrificed at day 17 and chromosome preparations were made (Thorgaard et al., '81). Production of homozygous ggnogenetic clones by using eggs from homozggous androgenetic females Eggs were obtained in the 1986-87 and the 1987-88 spawning seasons from androgenetic female trout produced in December 1984 (see Scheerer et al., '86), held on ice, and fertilized within 1 hour of spawning. Sperm was collected from males reared at Washington State University. When available, males that possessed the
384
P.D. SCHEERER ET AL.
dominant golden color marker were used, enabling us to monitor the effectiveness of the ultraviolet radiation treatment used to inactivate the sperm chromosomes (see below). Eggs obtained from 11 androgenetic females were each divided into two groups. One group was fertilized with unirradiated sperm to serve as a diploid control. The other group was fertilized with ultraviolet-irradiated sperm (Chourrout, '82; Thorgaard et al., '83). A portion of this second group was separated off to serve as haploid gynogenetic controls. The remainder were placed in a 29°C recirculating water bath for a 10 minute treatment applied 10 minutes post-fertilization t o induce the retention of the second polar body, thus creating diploid gynogenetic fish (Thorgaard et al., '83).
Production of androgenetic and ggnogenetic rainbow trout from common parents Rainbow trout eggs and sperm were obtained in November 1986 from four females and three males from the Arlee strain of the Jocko River Trout Hatchery (Montana Department of Fish, Wildlife, and Parks). These fish were a subsample from a larger group of trout which were sacrificed after gamete collection and analyzed by starchgel electrophoresis t o allow planned crosses with the gametes. The gametes were then transported on ice to Washington State University. Four separate pair-matings were designed t o compare the survival and yield of homozygous offspring from androgenetic and gynogenetic crosses derived from the same eggs and sperm. A portion of the eggs from each of the four female rainbow trout were fertilized with sperm from a separate male rainbow trout to serve as diploid controls. The remaining eggs from each female were divided into two groups. The eggs in group one were exposed t o 60cobaltradiation to a final dose of 3.6 x lo4 (Parsons and Thorgaard, '84). These eggs were fertilized with unirradiated sperm from separate males. A portion of these fertilized eggs were separated off t o serve as haploid androgenetic controls. The remainder of the eggs from each female in group one were subdivided into two lots and treated with hydrostatic pressure (9000 psi) for a 3 minute duration at 320 and 340 minutes post-fertilization, respectively. The eggs in group 2 were not irradiated, but instead were fertilized with sperm which was inactivated with ultraviolet radiation (Thorgaard et al., '83). A portion of the fertilized eggs from each lot was separated off to serve as haploid gyno-
genetic controls. The remainder were pressure treated simultaneously with the corresponding androgenetic lots (from common parents), one subgroup exposed at 320 minutes and the other at 340 minutes post-fertilization. The experimental lots were incubated at 11°C and survival was monitored at the eyed-egg stage (18 days), at hatching (31 days), and at the initiation of feeding (59 days). Tissues from a sample of fish from the androgenetic and gynogenetic diploid lots were removed and stored at - 70°C. At a later date, these tissue samples were thawed and analyzed by starch-gel electrophoresis to test the homozygosity and uniparental nature of the offspring. Females were selected which were heterozygous for the enzymes malate dehydrogenase (Mdh-B) and superoxide dismutase (SOD-1). Males were selected which were homozygous for either isocitrate dehydrogenase (IDH-21, glycerol-3-phosphate dehydrogenase (G3P), or hexoseaminidase (HEX) alleles not present in the females. These mating combinations allowed us t o determine if the gynogenetic offspring were truly homozygous, or if they were partially heterozygous due to spontaneous polar body retention. Also, the effectiveness of the ultraviolet treatment in sperm inactivation could be monitored in individuals showing expression of the appropriate enzyme loci.
Production of androgenetic trout from crgopreserved semen Albino rainbow trout semen was obtained in March 1987 from the University of Washington hatchery in Seattle. Eggs from two female steelhead (sea-run rainbow) trout were obtained from the Dworshak National Fish Hatchery, Ahsahka, Idaho. Albino rainbow trout semen was divided into two equal portions. One portion was stored at 4°C in an antibiotic extender solution (see Disney et al., '87). The remainder was cryopreserved in a solution containing 10% ( v h ) dimethyl sulfoxide in 0.3 M glucose, which was mixed in a 9 : l ratio with fresh chicken egg yolk (Alderson and MacNeil, '84; Scheerer and Thorgaard, '89). One milliliter of semen was mixed with 1.5 ml of cryoprotectant solution and allowed to equilibrate for 5 minutes before adding another 1.5 ml of the cryoprotectant solution, resulting in a final ratio of 3 : l cryoprotectant to semen. After another 5 minute equilibration, extended semen was drawn into 0.25 ml straws and frozen immediately on dry ice. After all the semen from one male was
ANDROGENESIS IN RAINBOW TROUT
drawn into straws and frozen on dry ice, the straws were transferred to liquid nitrogen for storage for 48 hours. A pooled sample of steelhead trout eggs was divided into two groups. Group 1was divided into two lots, one of which was fertilized with unfrozen albino rainbow semen. The other lot was fertilized with cryopreserved semen which was thawed by immersion in 10°C dechlorinated tap water. Group 2 was also divided into two lots which were exposed to 60cobalt irradiation as described previously. One lot was fertilized with unfrozen semen and the other was fertilized with cryopreserved semen. Fertilized eggs were placed in an 1l.C incubator for further development. A portion of each of the androgenetic lots was separated off to serve as androgenetic haploid controls. The remainder were pressure treated at 9,000 psi for a 3 minute duration at 340 minutes post-fertilization to restore diploidy. Survival was monitored at the eyed-egg stage (18 days), at hatching (31 days), and at the initiation of feeding (59 days). Sex ratios in the progeny of androgenetic males Androgenetic male rainbow trout previously produced from the Arlee, Spokane and Hot Creek (California Department of Fish and Game) strains (Scheerer et al., '86) were crossed to eggs from normal females and the progeny from nine
385
androgenetic males were sacrificed and sexed at 9- 11 months of age.
RESULTS Production of homozygous clones by androgenesis When androgenetic diploid rainbow trout were generated from the same eggs by using sperm from outbred and androgenetic (homozygous) sperm sources, survival was similar and significantly below that of the diploid control lots (Table 1). Five families of androgenetic progeny from four different outbred male parents (a total of 43 individuals) were examined for genetic variation by using protein electrophoresis. All 43 individuals were homozygous at all scorable loci, and differences were detected among homozygous individuals in all five families, reflecting heterozygosity in the outbred male parent. Four families of androgenetic progeny from four different androgenetic male parents (a total of 34 individuals) were also examined. All 34 individuals were homozygous and identical within families at all scorable loci. In addition one androgenetic male parent was shown to be identical to all 12 of his progeny at all scorable loci. Chromosome spreads from five individuals each from two outbred androgenetic diploid lots and two androgenetic clone lots were analyzed and all were found to be diploid with no chromosome fragments.
TABLE 1. Mean percent survival (f S.D.) of androgenetic rainbow trout produced from outbred and androgenetic sperm sources (ranges are provided i n parentheses) Mean survival
No. of Cross' Outbred sperm source Diploid controls
Eyed-egg
Lots
Eggs/lot
stage (20 days)
6
339 t 43 (318-355) 74 f 21 (48-125) 593 f 115 (437-701) 596 ? 117 (462- 724)
88.3 4.9 (83.0-93.4) 25.2 2 18.6 (4.3-47.9) 7.9 t 6.8 (0.2-18.7) 7.9 t 6.5 (0.0-14.5)
327 f 47 (251-376) 80 t 25 (51- 116) 597 t 110 (416-728) 606 f 92 (451-716)
87.3 6.7 (74.5-92.2) 27.3 f 15.3 (8.6-47.4) 8.1 8.1 (0.0-22.1) 8.3 f 6.6 (0.4-15.4)
~~
Androgenetic haploids
6
Androgenetic diploids, 320 Androgenetic diploids, 340 Androgenetic sperm source Diploid controls
6
6
Androgenetic haploids
6
Androgenetic diploids, 320 Androgenetic diploids, 340
6
6
6
*
*
*
Hatching (35 days)
Initiation of feeding (62 days)
87.5 t 4.5 (82.1-92.8) 0.0 t 0.0 (0.0-0.0) 3.1 t 2.8 (0.2-7.1) 2.7 t 2.5 (0.0-6.3)
85.8 f 4.4 (80.7-90.7)
84.7 f 6.3 (72.6-90.1) 0.0 t 0.0 (0.0-0.0) 1.8 t 1.9 (0.0-5.2) 1.9 t 1.9 (0.0-4.8)
82.3 & 5.9 (71.3-88.9) -
'Times refer to the time (minutes) post-fertilization when pressure treatment was applied.
-
2.1 t 2.2 (0.0-5.5) 2.2 t 2.0 (0.0-5.1)
*
1.4 1.6 (0.0-4.5) 1.5 1.4 (0.0-3.4)
*
P.D. SCHEERER ET AL.
386
Production of homozygous gynogenetic clones by using eggs from homozygous androgenetic females During the winter of 1986, eggs were obtained from four androgenetic females. Three of the females were originally produced from the Hot Creek strain and one was produced from the Arlee strain. The egg quality was poor and none of the eggs, even those fertilized with normal, unirradiated sperm, developed to the eyed-egg stage. In 1987, similar results were obtained from crosses involving eggs from three Hot Creek androgenetic females and two Arlee androgenetic females. However, viable yet marginal quality eggs were obtained from one Hot Creek androgenetic female and one Spokane androgenetic female. Control survival to the initiation of feeding was 15.8% and 16.3% in these groups, respectively. Survival of the gynogenetic diploid clones produced from the homozygous androgenetic females was low, yet the clone survival relative to the control survival (0.2/15.8 = 1.3%; 0.9/16.3 = 5.5% from the Hot Creek and Spokane female, respectively) was similar to the survival of the androgenetic clones relative to their controls (1.5/82.3 = 1.8%). No offspring expressing the dominant golden color marker were observed in the gynogenetic lots, indicating that the ultraviolet radiation treatment of the sperm was effective.
Production of androgenetic and gynogenetic rainbow trout from common parents When homozygous androgenetic and homozygous gynogenetic trout were generated by using the same eggs, sperm, and pressure treatments, the survival was higher in the gynogenetic lots. At the initiation of feeding, gynogenetic survival averaged 4.0% in the lots exposed to hydrostatic pressure 320 minutes post-fertilization, and 0.9% in the lots exposed at 340 minutes post-fertilization, compared to 0.7% and 0.1% in the respective androgenetic lots (Table 2). However, some of the gynogenetic offspring, when analyzed by starchgel electrophoresis, were found to be heterozygous at some enzyme loci. Among 23 gynogenetic progeny from females heterozygous at the SOD-1 locus, all 16 individuals in two families were homozygous at all loci, including SOD-1. However, among seven gynogenetic progeny in a third family, five were heterozygous at SOD-1 and two of the five were also heterozygous at the Mdh-B locus. This indicates that a substantial portion of the gynogenetic progeny in this family resulted from spontaneous retention of the second polar body; both the SOD-1 and Mdh-B loci normally show a n extremely high proportion of heterozygotes among gynogenetic progeny after retention of the second polar body (Thorgaard et al., '83; Guyomard, '84; Allendorf et al., '86). As expected
TABLE 2. Mean percent survival (tS.D.) of androgenetic and gynogenetic rainbow trout produced from common gametes (ranges are provided in parentheses) Mean survival
Lots
Eggdot
Eyed-egg stage (18 days)
Diploid controls
4
Androgenetic haploids
4
Androgenetic diploids, 3202 Androgenetic diploids, 3402 Gynogenetic haploids3
4
110 ? 19 (91-135) 60 t 16 (49-83) 539 t 135 (450-741) 608 ? 69 (549-693) 73 & 18 (59-98) 530 t 69 (457-607) 532 t 58 (450-580)
64.8 ? 38.1 (8.0-89.6) 11.8 t 9.8 (0.0-20.4) 6.4 t 11.9 (0.0-24.2) 3.0 r+ 4.2 (0.0-9.2) 69.3 ? 14.7 (56.1-86.4) 28.4 t 13.6 (9.3-39.7) 23.2 t 13.5 (8.3-40.7)
No. of Cross'
Gynogenetic diploids, 3202 Gynogenetic diploids, 3402
4 4 4 4
Hatching (35 days)
lnitiation of feeding (59 days)
57.9 ? 39.8 (0.0-87.4) 0.0 t 0.0
56.8 2 39.3 (0.0-86.7)
1.3 t 2.7 (0.0-5.3) 0.9 ? 1.1 (0.0-1.9) 2.2 t 2.1 (0.0-5.0) 5.9 & 5.6 (1.1-13.6) 2.8 t 2.5 (0.5-6.2)
0.7 t- 1.3 (0.0-2.6) 0.1 t- 0.2 (0.0-0.4) 1.5 2 1.4 (0.0-3.3) 4.0 t- 5.2 (0.6-1 1.7) 0.9 t- 1.0 (0.0-2.3)
-
'All crosses were made with gametes from the Arlee, Montana, hatchery rainbow trout strain. 'Times refer to the time (minutes) post-fertilization when pressure treatment was applied. 3Percent survival beyond the eyed stage represents survivors which were presumably diploids resulting from spontaneous polar body retention.
ANDROGENESIS IN RAINBOW TROUT
in the androgenetic progeny of an outbred male all six of the androgenetic progeny from one family sampled were homozygous at all loci examined but all were not identical t o each other.
Production of androgenetic trout from cryopreserved semen Androgenetic offspring were produced by using cryopreserved albino rainbow trout semen. Survival of the androgenetic offspring produced from frozen semen was lower than that from unfrozen semen, yet the survival of the androgenetic diploids relative t o the survival of the respective control lots was nearly identical (Table 3). Albinism is a recessive color marker in trout and was used to confirm all-paternal inheritance in these crosses. All androgenetic offspring were albino and all diploid control offspring were normally pigmented.
387
Control survival using cryopreserved semen, relative t o that of the controls produced from non-frozen semen, was low and averaged 39.4% at the eyed-egg stage and 35.4% at the initiation of feeding (Table 3). Sex ratios in the progeny of androgenetic males The sex ratios from the eight lots of progeny of androgenetic males with sufficient numbers to be informative all deviated significantly from 1:1 (Table 4). Four lots had all males and one had one female and 15 males. However, one lot had all females and two had large excesses of females. These results suggest that both XX and YY individuals were represented among the androgenetic males that were used as parents. The rare females or males in some lots might reflect autosomal genes or environmental influences affecting sex determination.
TABLE 3. Percent survival
of androgenetic rainbow trout produced from cryopreserved semen‘
Mean survival
Cross Non-frozen semen Diploid controls Androgenetic haploids Androgenetic diploids Cryopreserved semen Diploid controls Androgenetic haploids Androgenetic diploids
No. of eggs per lot
Eyed-egg stage (18 days)
Hatching (31 days)
Initiation of feeding (59 days)
245 310 73 1
84.1 17.4 (20.7) 19.8 (23.5)
75.9 0.0 (0.0) 3.8 (5.0)
68.6 -
181 26 1 767
33.1 5.4 (16.3) 8.4 (25.4)
27.6 0.0 (0.0) 1.3 (4.7)
24.3 0.9 (3.7)
-
_
2.5 (3.6)
‘Dworshak steelhead eggs (n = 2 females, pooled) were crossed with semen from a single albino rainbow trout from the University of Washington. Numbers in parentheses represent survival of androgenetic individuals relative to the respective diploid controls.
TABLE 4 . Sex ratios in the woeeny o f androgenetic male rainbow trout’ Female parent
Androgenetic male parent
Females
Males
P (1:l ratio)
Spokane pool Spokane pool Spokane pool Spokane pool Spokane pool Spokane pool Utah albino Spokane No. 4 Spokane No. 3
Hot Cr. #57 Arlee #37 Arlee #51 Hot Cr. #59 Spokane #1 Hot Cr. #60 Arlee clone #11 Arlee #410 Arlee #12
19 0 0 1 12 20 0 0 0
0 1 15 15 4 2 27 14 40
Genetic Analysis of Androgenetic Rainbow Trout PAUL D. SCHEERER, GARY H. THORGAARD, FRED W. ALLENDORF Department of Zoology and Program in Genetics and Cell Biology, Washington State University, Pullman, Washington 991 64-4236 (P.D.S., G.H.T.); Division of Biological Sciences, University of Montana, Missoula, Montana 59812 (F.W.A.) AND
We analyzed a number of genetic characteristics in androgenetic rainbow trout ABSTRACT (Oncorhynchus mykiss) and their progeny. The androgenetic progeny of individual androgenetic males appeared genetically identical to each other based on eight enzyme loci. Their viability was no higher than that of androgenetic progeny of outbred males. Homozygous androgenetic female rainbow trout produced very poor quality eggs. When common eggs and sperm from outbred individuals were used to produce androgenetic and gynogenetic progeny, the yield of gynogenetic progeny was higher but some were heterozygous at protein loci, while no androgenetic progeny were heterozygous. Some androgenetic diploid rainbow trout were successfully produced from cryopreserved sperm. The progeny of some androgenetic males crossed to normal females were virtually all males, while the progeny of other males were virtually all females. This suggests that both XX and YY androgenetic individuals may develop as males. Androgenesis is likely to be useful for generating homozygous clones for research and for recovering strains from cryopreserved sperm.
Isogenic lines of animals and plants are valu- androgenesis (all-paternal inheritance) (Scheerer able research tools. Inbred lines and hybrids be- et al., '86; May et al., '88). Androgenesis involves tween lines are useful in genetic, developmental inactivating the maternal genetic material by usbiology, immunology, toxicology, and cancer re- ing gamma radiation (Romashow and Belyaeva, search (Allard, '60; Festing, '79; Green, '81; Ab- '64; Purdom, '69; Arai et al., '79; Parsons and planalp, '86). However, the common method of Thorgaard, '841, fertilizing these treated eggs generating inbred strains involves 10-20 genera- with normal sperm, and restoring diploidy by suptions of sib-mating, requiring substantial com- pressing the first cleavage by using hydrostatic pressure (Onozato, '82, in Yamazaki, '83; Parsons mitments of both time and money. Animal researchers have recently used alter- and Thorgaard, '85). Production of homozygous nate approaches for rapidly producing inbred rainbow trout by androgenesis has been generally lines in several species. These approaches involve successful (Parsons and Thorgaard, '85; Scheerer inactivating the entire chromosome set from one et al., '86). Spontaneous heterozygous individuals parent followed by duplicating the remaining set. have not been observed, and both the XX and YY Homozygous frogs have been produced by block- offspring appear to be viable (Parsons and Thoring the first cleavage of parthenogenetically acti- gaard, '85). The gametes from these androgenetic vated haploid eggs (Reinschmidt et al., '79). progeny can, in turn, be used t o found isogenic Clonal lines of two aquarium fish species have lines. Androgenesis could also facilitate the genbeen produced gynogenetically by blocking the eration of YY males which could be useful in first cleavage of eggs fertilized with UV-inacti- aquaculture programs where production of allvated sperm, followed by producing gynogenetic male populations can be desirable (Scott et al., progeny from the homozygous individuals (Strei- '$9). singer et al., '81; Naruse et al., '85). This approach has been unsuccessful in mammals, where both the maternal and paternal pronuclei must apparReceived January 11, 1991; revision accepted February 15, 1991. ently be active in the developing embryo (McAddress reprint requests to Gary H. Thorgaard, Department of Zoology and Program in Genetics and Cell Biology, Washington State Grath and Solter, '84). Pullman, WA 99164-4236. A promising approach for generating inbred University, Paul D. Scheerer'spresent address is 24886 Grange Hall Rd., Philolines in fish and amphibians involves inducing math, OR 97370. 0 1991 WILEY-LISS,
INC.
ANDROGENESIS IN RAINBOW TROUT
This study used several approaches to genetically analyze androgenetic rainbow trout (Oncorhynchus mykiss) and their progeny. Androgenetic clones of rainbow trout were generated by using sperm from homozygous androgenetic males and their survival was compared to that of androgenetic trout produced from sperm from outbred males. Gynogenetic offspring were also generated by using eggs from homozygous androgenetic females. The yields of homozygous diploid androgenetic and gynogenetic progeny produced by using the same eggs, sperm, and pressure treatments were compared. Androgenetic offspring were produced from cryopreserved trout semen, demonstrating the potential application of this approach to gene banking. Finally, we examined the sex ratios in the progeny of androgenetic males crossed t o normal females.
MATERIALS AND METHODS Production of homozygous clones by androgenesis Rainbow trout sperm was obtained in December 1986 from six homozygous androgenetic males and six outbred males. The outbred strains included the Arlee strain from the Jocko River Trout Hatchery (Montana Department of Fish, Wildlife and Parks), the Spokane Hatchery strain (Washington Department of Wildlife), an albino heterozygote (Albino from the Egan, Utah State Fish Hatchery x Spokane rainbow), and a palomino rainbow (a co-dominant golden from The Pennsylvania State University x Spokane rainbow) reared at Washington State University. The homozygous androgenetic males were produced from the Arlee, Spokane, and Hot Creek (California Department of Fish and Game) strains (see Scheerer et al., '86). Eggs were obtained from six females from the Spokane hatchery. Gametes were transported on ice to Washington State University. Androgenetic progeny were produced from both outbred and androgenetic male parents. Eggs from the six rainbow trout females were used separately. The majority of the eggs from each female were exposed to 60cobaltgamma radiation at the Washington State University Nuclear Radiation Center to a final dose of 3.6 x lo4 R (Parsons and Thorgaard, '84); a portion of each egg lot was left unirradiated. The irradiated eggs were divided into two treatment groups and fertilized separately with sperm from one outbred and one androgenetic male. The unirradiated eggs were
383
also divided into two groups and fertilized with sperm from the same males, serving as diploid controls. After fertilization, all egg lots were placed in an incubator for further development at 7-9°C. A portion of each irradiated egg lot was separated off to serve as androgenetic haploid controls; the remainder were divided into two groups. Both groups were treated with hydrostatic pressure at 9,000 psi for a 3 minute duration to block the first cleavage division and restore diploidy (Parsons and Thorgaard, '85). One group was pressure treated at 320 minutes post-fertilization and the other at 340 minutes post-fertilization. The two times of application of hydrostatic pressure were used t o increase the chance of blocking the first cleavage since the interval between fertilization and the first cleavage division varies between strains and between females within strains (unpublished data). Relative survival rates of the androgenetic diploid lots were calculated as a percentage survival compared to the control survival (survival androgenetic lot/survival control lot). Survival was monitored at the eyed-egg stage (20 days), at hatching (35 days), and at the initiation of feeding (62 days). Tissues from a sample of androgenetic fish were removed and stored at - 70°C. Tissue extracts were later prepared and analyzed by horizontal starch-gel electrophoresis as described by Allendorf et al. ('86) t o test homozygosity and the clonal nature of the offspring. We attempted to score the androgenetic progeny for genetic variation at the following enzyme loci: AAT-3; HEX; IDH-2; IDH3; LDH-4; Mdh-B; PGM-2; SOD-1. Not all progeny could be scored at all loci because some fish were sampled at a very young age and some were dead for a period of time before being frozen. Five embryos from two separate lots of both outbred androgenetic diploids and androgenetic clones were sacrificed at day 17 and chromosome preparations were made (Thorgaard et al., '81). Production of homozygous ggnogenetic clones by using eggs from homozggous androgenetic females Eggs were obtained in the 1986-87 and the 1987-88 spawning seasons from androgenetic female trout produced in December 1984 (see Scheerer et al., '86), held on ice, and fertilized within 1 hour of spawning. Sperm was collected from males reared at Washington State University. When available, males that possessed the
384
P.D. SCHEERER ET AL.
dominant golden color marker were used, enabling us to monitor the effectiveness of the ultraviolet radiation treatment used to inactivate the sperm chromosomes (see below). Eggs obtained from 11 androgenetic females were each divided into two groups. One group was fertilized with unirradiated sperm to serve as a diploid control. The other group was fertilized with ultraviolet-irradiated sperm (Chourrout, '82; Thorgaard et al., '83). A portion of this second group was separated off to serve as haploid gynogenetic controls. The remainder were placed in a 29°C recirculating water bath for a 10 minute treatment applied 10 minutes post-fertilization t o induce the retention of the second polar body, thus creating diploid gynogenetic fish (Thorgaard et al., '83).
Production of androgenetic and ggnogenetic rainbow trout from common parents Rainbow trout eggs and sperm were obtained in November 1986 from four females and three males from the Arlee strain of the Jocko River Trout Hatchery (Montana Department of Fish, Wildlife, and Parks). These fish were a subsample from a larger group of trout which were sacrificed after gamete collection and analyzed by starchgel electrophoresis t o allow planned crosses with the gametes. The gametes were then transported on ice to Washington State University. Four separate pair-matings were designed t o compare the survival and yield of homozygous offspring from androgenetic and gynogenetic crosses derived from the same eggs and sperm. A portion of the eggs from each of the four female rainbow trout were fertilized with sperm from a separate male rainbow trout to serve as diploid controls. The remaining eggs from each female were divided into two groups. The eggs in group one were exposed t o 60cobaltradiation to a final dose of 3.6 x lo4 (Parsons and Thorgaard, '84). These eggs were fertilized with unirradiated sperm from separate males. A portion of these fertilized eggs were separated off t o serve as haploid androgenetic controls. The remainder of the eggs from each female in group one were subdivided into two lots and treated with hydrostatic pressure (9000 psi) for a 3 minute duration at 320 and 340 minutes post-fertilization, respectively. The eggs in group 2 were not irradiated, but instead were fertilized with sperm which was inactivated with ultraviolet radiation (Thorgaard et al., '83). A portion of the fertilized eggs from each lot was separated off to serve as haploid gyno-
genetic controls. The remainder were pressure treated simultaneously with the corresponding androgenetic lots (from common parents), one subgroup exposed at 320 minutes and the other at 340 minutes post-fertilization. The experimental lots were incubated at 11°C and survival was monitored at the eyed-egg stage (18 days), at hatching (31 days), and at the initiation of feeding (59 days). Tissues from a sample of fish from the androgenetic and gynogenetic diploid lots were removed and stored at - 70°C. At a later date, these tissue samples were thawed and analyzed by starch-gel electrophoresis to test the homozygosity and uniparental nature of the offspring. Females were selected which were heterozygous for the enzymes malate dehydrogenase (Mdh-B) and superoxide dismutase (SOD-1). Males were selected which were homozygous for either isocitrate dehydrogenase (IDH-21, glycerol-3-phosphate dehydrogenase (G3P), or hexoseaminidase (HEX) alleles not present in the females. These mating combinations allowed us t o determine if the gynogenetic offspring were truly homozygous, or if they were partially heterozygous due to spontaneous polar body retention. Also, the effectiveness of the ultraviolet treatment in sperm inactivation could be monitored in individuals showing expression of the appropriate enzyme loci.
Production of androgenetic trout from crgopreserved semen Albino rainbow trout semen was obtained in March 1987 from the University of Washington hatchery in Seattle. Eggs from two female steelhead (sea-run rainbow) trout were obtained from the Dworshak National Fish Hatchery, Ahsahka, Idaho. Albino rainbow trout semen was divided into two equal portions. One portion was stored at 4°C in an antibiotic extender solution (see Disney et al., '87). The remainder was cryopreserved in a solution containing 10% ( v h ) dimethyl sulfoxide in 0.3 M glucose, which was mixed in a 9 : l ratio with fresh chicken egg yolk (Alderson and MacNeil, '84; Scheerer and Thorgaard, '89). One milliliter of semen was mixed with 1.5 ml of cryoprotectant solution and allowed to equilibrate for 5 minutes before adding another 1.5 ml of the cryoprotectant solution, resulting in a final ratio of 3 : l cryoprotectant to semen. After another 5 minute equilibration, extended semen was drawn into 0.25 ml straws and frozen immediately on dry ice. After all the semen from one male was
ANDROGENESIS IN RAINBOW TROUT
drawn into straws and frozen on dry ice, the straws were transferred to liquid nitrogen for storage for 48 hours. A pooled sample of steelhead trout eggs was divided into two groups. Group 1was divided into two lots, one of which was fertilized with unfrozen albino rainbow semen. The other lot was fertilized with cryopreserved semen which was thawed by immersion in 10°C dechlorinated tap water. Group 2 was also divided into two lots which were exposed to 60cobalt irradiation as described previously. One lot was fertilized with unfrozen semen and the other was fertilized with cryopreserved semen. Fertilized eggs were placed in an 1l.C incubator for further development. A portion of each of the androgenetic lots was separated off to serve as androgenetic haploid controls. The remainder were pressure treated at 9,000 psi for a 3 minute duration at 340 minutes post-fertilization to restore diploidy. Survival was monitored at the eyed-egg stage (18 days), at hatching (31 days), and at the initiation of feeding (59 days). Sex ratios in the progeny of androgenetic males Androgenetic male rainbow trout previously produced from the Arlee, Spokane and Hot Creek (California Department of Fish and Game) strains (Scheerer et al., '86) were crossed to eggs from normal females and the progeny from nine
385
androgenetic males were sacrificed and sexed at 9- 11 months of age.
RESULTS Production of homozygous clones by androgenesis When androgenetic diploid rainbow trout were generated from the same eggs by using sperm from outbred and androgenetic (homozygous) sperm sources, survival was similar and significantly below that of the diploid control lots (Table 1). Five families of androgenetic progeny from four different outbred male parents (a total of 43 individuals) were examined for genetic variation by using protein electrophoresis. All 43 individuals were homozygous at all scorable loci, and differences were detected among homozygous individuals in all five families, reflecting heterozygosity in the outbred male parent. Four families of androgenetic progeny from four different androgenetic male parents (a total of 34 individuals) were also examined. All 34 individuals were homozygous and identical within families at all scorable loci. In addition one androgenetic male parent was shown to be identical to all 12 of his progeny at all scorable loci. Chromosome spreads from five individuals each from two outbred androgenetic diploid lots and two androgenetic clone lots were analyzed and all were found to be diploid with no chromosome fragments.
TABLE 1. Mean percent survival (f S.D.) of androgenetic rainbow trout produced from outbred and androgenetic sperm sources (ranges are provided i n parentheses) Mean survival
No. of Cross' Outbred sperm source Diploid controls
Eyed-egg
Lots
Eggs/lot
stage (20 days)
6
339 t 43 (318-355) 74 f 21 (48-125) 593 f 115 (437-701) 596 ? 117 (462- 724)
88.3 4.9 (83.0-93.4) 25.2 2 18.6 (4.3-47.9) 7.9 t 6.8 (0.2-18.7) 7.9 t 6.5 (0.0-14.5)
327 f 47 (251-376) 80 t 25 (51- 116) 597 t 110 (416-728) 606 f 92 (451-716)
87.3 6.7 (74.5-92.2) 27.3 f 15.3 (8.6-47.4) 8.1 8.1 (0.0-22.1) 8.3 f 6.6 (0.4-15.4)
~~
Androgenetic haploids
6
Androgenetic diploids, 320 Androgenetic diploids, 340 Androgenetic sperm source Diploid controls
6
6
Androgenetic haploids
6
Androgenetic diploids, 320 Androgenetic diploids, 340
6
6
6
*
*
*
Hatching (35 days)
Initiation of feeding (62 days)
87.5 t 4.5 (82.1-92.8) 0.0 t 0.0 (0.0-0.0) 3.1 t 2.8 (0.2-7.1) 2.7 t 2.5 (0.0-6.3)
85.8 f 4.4 (80.7-90.7)
84.7 f 6.3 (72.6-90.1) 0.0 t 0.0 (0.0-0.0) 1.8 t 1.9 (0.0-5.2) 1.9 t 1.9 (0.0-4.8)
82.3 & 5.9 (71.3-88.9) -
'Times refer to the time (minutes) post-fertilization when pressure treatment was applied.
-
2.1 t 2.2 (0.0-5.5) 2.2 t 2.0 (0.0-5.1)
*
1.4 1.6 (0.0-4.5) 1.5 1.4 (0.0-3.4)
*
P.D. SCHEERER ET AL.
386
Production of homozygous gynogenetic clones by using eggs from homozygous androgenetic females During the winter of 1986, eggs were obtained from four androgenetic females. Three of the females were originally produced from the Hot Creek strain and one was produced from the Arlee strain. The egg quality was poor and none of the eggs, even those fertilized with normal, unirradiated sperm, developed to the eyed-egg stage. In 1987, similar results were obtained from crosses involving eggs from three Hot Creek androgenetic females and two Arlee androgenetic females. However, viable yet marginal quality eggs were obtained from one Hot Creek androgenetic female and one Spokane androgenetic female. Control survival to the initiation of feeding was 15.8% and 16.3% in these groups, respectively. Survival of the gynogenetic diploid clones produced from the homozygous androgenetic females was low, yet the clone survival relative to the control survival (0.2/15.8 = 1.3%; 0.9/16.3 = 5.5% from the Hot Creek and Spokane female, respectively) was similar to the survival of the androgenetic clones relative to their controls (1.5/82.3 = 1.8%). No offspring expressing the dominant golden color marker were observed in the gynogenetic lots, indicating that the ultraviolet radiation treatment of the sperm was effective.
Production of androgenetic and gynogenetic rainbow trout from common parents When homozygous androgenetic and homozygous gynogenetic trout were generated by using the same eggs, sperm, and pressure treatments, the survival was higher in the gynogenetic lots. At the initiation of feeding, gynogenetic survival averaged 4.0% in the lots exposed to hydrostatic pressure 320 minutes post-fertilization, and 0.9% in the lots exposed at 340 minutes post-fertilization, compared to 0.7% and 0.1% in the respective androgenetic lots (Table 2). However, some of the gynogenetic offspring, when analyzed by starchgel electrophoresis, were found to be heterozygous at some enzyme loci. Among 23 gynogenetic progeny from females heterozygous at the SOD-1 locus, all 16 individuals in two families were homozygous at all loci, including SOD-1. However, among seven gynogenetic progeny in a third family, five were heterozygous at SOD-1 and two of the five were also heterozygous at the Mdh-B locus. This indicates that a substantial portion of the gynogenetic progeny in this family resulted from spontaneous retention of the second polar body; both the SOD-1 and Mdh-B loci normally show a n extremely high proportion of heterozygotes among gynogenetic progeny after retention of the second polar body (Thorgaard et al., '83; Guyomard, '84; Allendorf et al., '86). As expected
TABLE 2. Mean percent survival (tS.D.) of androgenetic and gynogenetic rainbow trout produced from common gametes (ranges are provided in parentheses) Mean survival
Lots
Eggdot
Eyed-egg stage (18 days)
Diploid controls
4
Androgenetic haploids
4
Androgenetic diploids, 3202 Androgenetic diploids, 3402 Gynogenetic haploids3
4
110 ? 19 (91-135) 60 t 16 (49-83) 539 t 135 (450-741) 608 ? 69 (549-693) 73 & 18 (59-98) 530 t 69 (457-607) 532 t 58 (450-580)
64.8 ? 38.1 (8.0-89.6) 11.8 t 9.8 (0.0-20.4) 6.4 t 11.9 (0.0-24.2) 3.0 r+ 4.2 (0.0-9.2) 69.3 ? 14.7 (56.1-86.4) 28.4 t 13.6 (9.3-39.7) 23.2 t 13.5 (8.3-40.7)
No. of Cross'
Gynogenetic diploids, 3202 Gynogenetic diploids, 3402
4 4 4 4
Hatching (35 days)
lnitiation of feeding (59 days)
57.9 ? 39.8 (0.0-87.4) 0.0 t 0.0
56.8 2 39.3 (0.0-86.7)
1.3 t 2.7 (0.0-5.3) 0.9 ? 1.1 (0.0-1.9) 2.2 t 2.1 (0.0-5.0) 5.9 & 5.6 (1.1-13.6) 2.8 t 2.5 (0.5-6.2)
0.7 t- 1.3 (0.0-2.6) 0.1 t- 0.2 (0.0-0.4) 1.5 2 1.4 (0.0-3.3) 4.0 t- 5.2 (0.6-1 1.7) 0.9 t- 1.0 (0.0-2.3)
-
'All crosses were made with gametes from the Arlee, Montana, hatchery rainbow trout strain. 'Times refer to the time (minutes) post-fertilization when pressure treatment was applied. 3Percent survival beyond the eyed stage represents survivors which were presumably diploids resulting from spontaneous polar body retention.
ANDROGENESIS IN RAINBOW TROUT
in the androgenetic progeny of an outbred male all six of the androgenetic progeny from one family sampled were homozygous at all loci examined but all were not identical t o each other.
Production of androgenetic trout from cryopreserved semen Androgenetic offspring were produced by using cryopreserved albino rainbow trout semen. Survival of the androgenetic offspring produced from frozen semen was lower than that from unfrozen semen, yet the survival of the androgenetic diploids relative t o the survival of the respective control lots was nearly identical (Table 3). Albinism is a recessive color marker in trout and was used to confirm all-paternal inheritance in these crosses. All androgenetic offspring were albino and all diploid control offspring were normally pigmented.
387
Control survival using cryopreserved semen, relative t o that of the controls produced from non-frozen semen, was low and averaged 39.4% at the eyed-egg stage and 35.4% at the initiation of feeding (Table 3). Sex ratios in the progeny of androgenetic males The sex ratios from the eight lots of progeny of androgenetic males with sufficient numbers to be informative all deviated significantly from 1:1 (Table 4). Four lots had all males and one had one female and 15 males. However, one lot had all females and two had large excesses of females. These results suggest that both XX and YY individuals were represented among the androgenetic males that were used as parents. The rare females or males in some lots might reflect autosomal genes or environmental influences affecting sex determination.
TABLE 3. Percent survival
of androgenetic rainbow trout produced from cryopreserved semen‘
Mean survival
Cross Non-frozen semen Diploid controls Androgenetic haploids Androgenetic diploids Cryopreserved semen Diploid controls Androgenetic haploids Androgenetic diploids
No. of eggs per lot
Eyed-egg stage (18 days)
Hatching (31 days)
Initiation of feeding (59 days)
245 310 73 1
84.1 17.4 (20.7) 19.8 (23.5)
75.9 0.0 (0.0) 3.8 (5.0)
68.6 -
181 26 1 767
33.1 5.4 (16.3) 8.4 (25.4)
27.6 0.0 (0.0) 1.3 (4.7)
24.3 0.9 (3.7)
-
_
2.5 (3.6)
‘Dworshak steelhead eggs (n = 2 females, pooled) were crossed with semen from a single albino rainbow trout from the University of Washington. Numbers in parentheses represent survival of androgenetic individuals relative to the respective diploid controls.
TABLE 4 . Sex ratios in the woeeny o f androgenetic male rainbow trout’ Female parent
Androgenetic male parent
Females
Males
P (1:l ratio)
Spokane pool Spokane pool Spokane pool Spokane pool Spokane pool Spokane pool Utah albino Spokane No. 4 Spokane No. 3
Hot Cr. #57 Arlee #37 Arlee #51 Hot Cr. #59 Spokane #1 Hot Cr. #60 Arlee clone #11 Arlee #410 Arlee #12
19 0 0 1 12 20 0 0 0
0 1 15 15 4 2 27 14 40
