39 Electrophoretic analysis of the diploid progenies from triploid × diploid crosses in the loach M

368K. ARAI AND M. MUKAINO THE JOURNAL OF EXPERIMENTAL ZOOLOGY 280:368–374 (1998)

? 1998 WILEY-LISS, INC.

Electrophoretic Analysis of the Diploid Progenies

From Triploid × Diploid Crosses in the Loach Misgurnus anguillicaudatus (Pisces: Cobitidae)

KATSUTOSHI ARAI* AND MIKIO MUKAINO

Faculty of Applied Biological Science, Hiroshima University,Higashi-Hiroshima 739-8528, Japan

ABSTRACT In the loach Misgurnus anguillicaudatus (Pisces: Cobitidae), triploid females de-rived from the hybridization of normal diploid females and natural tetraploid males generate both triploid and haploid eggs simultaneously. Diploid progenies were produced by fertilizing the hap-loid eggs of triploids with haploid spermatozoa of normal diploids. Ldh-1 and Mdh-2 allozyme loci were electrophoretically diagnostic, because diploid and tetraploid loach exhibited the diverged genotype aa and bbbb at each locus, respectively. Diploids from triploid (abb ) × diploid (aa ) crosses frequently exhibited a heterozygous ab genotype at each locus but sometimes showed genotypic frequencies that fit a random pairing model (aa:ab = 1:2) of two of three chromosomes in each set of homologues. These results show that the entire haploid genome of one parent is not completely excluded in the oogenesis of triploid loach, and the haploid eggs are not formed in a typical hybridogenetic manner. However, the two homologous chromosomes derived from the tetraploid parent are likely to pair with each other preferentially during the meiosis of triploid loach. J. Exp.Zool. 280:368 374, 1998.? 1998 Wiley-Liss, Inc.

Contract Grant Sponsor: Fisheries Agency of Japan for Research on Breeding of Aquatic Organisms.

*Correspondence to: Katsutoshi Arai, Faculty of Applied Biological Science, Hiroshima University, 4-4, 1-Chome, Kagamiyama, Higashi-Hiroshima, 739-8528, Japan. E-mail: araikt@ipc.hiroshima-u.ac.jp Received 3 June 1997; Accepted 11 November 1997

The chromosome number for the loach Misgur-nus anguillicaudatus in Japan is 2n = 50, with a few specimens showing polyploidy (Ojima and Takai, ’79; Arai et al., ’91). Natural tetraploids have been detected in specimens obtained from fish deal-ers, though their exact origin is still unknown. Lo-ach with 100 chromosomes are not considered diploidized tetraploids, but evolutionary recent tetraploids with four sets of homologous chromo-somes. Viable gynogens were produced from eggs fertilized with UV-irradiated spermatozoa without treatment for chromosome doubling (Arai et al., ’93).Normally , gynogens induced from diploid loach are inviable because of a haploid syndrome (Suzuki et al., ’85). An androgenetic experiment in which UV-irradiated eggs of diploid loach were fertilized with spermatozoa of tetraploids generated viable diploid offspring and gave similar results (Arai et al., ’95).Triploid loach produced by crossing normal dip-loids with natural tetraploids were viable, with the females laying triploid eggs of large size to-gether with haploid eggs of small size, and the males sterile (Matsubara et al., ’95). When trip-loid gynogens were induced by fertilizing the large triploid eggs with UV-irradiated spermatozoa of heterospecific species, they demonstrated DNA fin-gerprints identical or highly similar to the somatic cells of the mother fish (Arai and Mukaino, ’97).

This suggested that the large egg of the triploid loach is formed by unreduced oogenesis.

In contrast, we know little about the mechanism by which triploid females generate small-sized hap-loid eggs. In this breeding system, it is apparent that chromosome segregation does not follow the normal pattern, and only a haploid genome can be transmitted to mature eggs. Such an atypical mode of reproduction may be a derivative of hybrido-genesis, which was first described in the unisexual hybrid fish Poeciliopsis (Schultz, ’63, ’66, ’69). These hybrid-origin fish are all females, and the pater-nally derived genome is premeiotically excluded be-cause it remains in the cytoplasm of the oogonial cell and is not transmitted to mature ova (Cimino,’72). Analogous examples have been reported in trip-loid teleosts and anurans. The allotriploid minnow fish Phoxinus eos-neogaeus-eos, which had two ge-nomes of P. eos and one genome of P. neogaeus, did lay haploid eggs comprising only P. eos chromosomes (Goddard and Schultz, ’93). Günther et al. (’79) con-cluded that the triploid frog Rana esculenta pro-

ALLOZYMES IN DIPLOID PROGENIES OF TRIPLOID LOACH369

duced gametes whose chromosome set derived from a parental species giving two genomes to the trip-loidy. Such a hybridogenetic gametogenesis was re-ported in allotriploids comprising two genomes of Japanese frogs and one genome of European frogs, genus Rana (Nishioka and Ohtani, ’84; Ohtani, ’93) and those comprising tetraploid and diploid species of treefrogs, genus Hyla (Mable and Bogart, ’95). We were able to obtain diploid progenies derived from fertilization of small eggs laid by triploid (diploid × tetraploid) females with spermatozoa of normal diploid males. In these progeny, we elec-trophoretically analyzed selected allozyme loci of-fering allelic differences between diploid and tetraploid loach. We then determined whether a hybridogenetic system for haploid egg formation is present in the triploid loach by examining allozyme genotypes at diagnostic loci.

We demonstrate here genetic evidence for pref-erential pairing between homologous chromo-somes derived from tetraploids in the oogenesis of the triploid loach and report an unusual mecha-nism for gametogenesis, distinguishable from the typical mode of hybridogenesis.

MATERIALS AND METHODS

Fish

To examine diagnostic allozyme loci of diploid and tetraploid loach, we used tetraploid (n = 67) from four different crosses (3026, 4026, 4028, and 4041) using females and males developed from the first generation of the tetraploid × tetraploid cross (024) (Arai et al., ’93; Zhang and Arai, ’96) and dip-loid fish (n = 37) from a mixed population of six different crosses (3005, 3010, and the other four crosses in 1992) using wild fish caught in a rice field in Sera Town, Sera County, Hiroshima Prefec-ture, Japan (Zhang and Arai, ’96). We also used 23 wild loach collected in Sera in 1994. Triploid loach (n = 27) in three different crosses (025, 026, 027) were also sampled to confirm heterozygosity at the putative diagnostic loci. These triploids were the progeny of hybridizations between normal diploid females and a tetraploid male obtained from a com-mercial dealer (Arai et al., ’91, ’93).

For breeding, mature triploid females in the 027 group were selected based on external appearance and abdomen size. Wild diploid males were caught in Sera and used for cross-fertilization.

Crosses

The procedure for artificial ovulation and fer-tilization has been described by Arai et al. (’93). The diploid progeny were produced from hybrid-izations between triploid females (2025, 4016, 5010, and 5016) and diploid males caught in a rice field in Sera in 1992, 1994, and 1995. Fertil-ized eggs of each cross were placed in an enamel pan half-filled with freshwater and then incubated at room temperature. Small eggs were visually sorted from each batch by removing large eggs; triploid (diploid × tetraploid) loach lay both trip-loid eggs of large size and haploid eggs of small size at the same time (Matsubara et al., ’95). The number of small eggs used and normal 5-day-old fry at the beginning of feeding were recorded. Af-ter absorption of yolk, the fry were fed rotifers and brine shrimp and later artificial carp food. The fry from each cross were reared in a plastic container (40 liters) half-filled with freshwater.

Flow cytometry

Ploidy status of diploid, triploid, and tetraploid loach for surveying the diagnostic allozyme loci was determined by flow cytometry for DNA con-tents of somatic cells when loach were killed to take tissue specimens for electrophoresis. In ad-dition, the triploidy of females used for breeding was confirmed by flow cytometry before crossing. Ploidy status of individual fish from each triploid × diploid cross was also determined by flow cytometry. Sample preparation and measurements were as described by Zhang and Arai (’96). Electrophoresis and staining of allozymes The skeletal muscle and liver were separated in diploid (n = 60), triploid (n = 27), and tetrap-loid loach (n = 67) and then kept in a deep freezer (–80°C) until electrophoretic analysis. Tissue specimens were not taken from females used for the cross, as the fish were needed for further breeding, while the muscle of the diploid males used for cross was taken and deep frozen for elec-trophoresis. Diploid progenies from triploid (small eggs) × diploid crosses were also killed to take muscle specimens for electrophoresis.

Lactate dehydrogenase (LDH, E.C.1.1.1.27), malate dehydrogenase (MDH, E.C.1.1.1.37), isoci-trate dehydrogenase (IDHP, E.C.1.1.1.42), phos-phoglucomutase (PGM, E.C.5.4.2.2), and aspartate aminotransferase (AAT, E.C.2.4.1.1) were ana-lyzed by horizontal starch-gel electrophoresis, us-ing a buffer system of citrate-N(-3-aminopropyl)-morpholine (pH. 6.1) as described by Clayton and Tretiak (’72). A drip of defrosted tissue specimen was applied to the gel as a sample using a paper wick. Gels (200 mm width × 150 mm length × 5 mm depth) were made using 12–14% starch, and

370K. ARAI AND M. MUKAINO

samples were electrophoresed for 4–6 h at con-stant voltage (150 V) in a 4°C refrigerator. After electrophoresis, the gel slices (1 mm thick) were histochemically stained as described by Allendorf et al. (’77) and Aebersold et al. (’87). Genetic loci were serially nominated from anode to cathode and are described with italic letters and numbers.Alleles are described with small italic letters.

RESULTS

Electrophoretic comparison of diploid and

tetraploid loach For lactate dehydrogenase (LDH) there are at least two loci (Ldh-1 and -2) expressed in muscle tissue of the loach (Fig. 1). A total of five bands generated by tetrameric combination of gene prod-ucts (subunit polypeptides) of the two loci were observed both in diploid and tetraploid loach, but the homotetramer migrated more anodally in dip-

loids than in tetraploids (Fig. 1). Alleles a and b of Ldh-1 were unique to diploid and tetraploid lo-ach, respectively (Table 1). Triploid loach produced by diploid × tetraploid hybridization showed com-plicated expression of multiple numbers of tet-rameric products formed by a single a and two b alleles at Ldh-1 and three common alleles at the Ldh-2 locus (Fig. 1). Electrophoretic phenotypes and ploidy status indicated aa, bbbb, and abb genotypes of diploid, tetraploid, and triploid lo-ach at the Ldh-1 locus, respectively.

For malate dehydrogenase (MDH), four pre-sumptive loci, Mdh-1, -2, -3, and -4, are expressed in muscle tissue (Fig. 2). Mdh-1 and -2 were pre-dominantly expressed in liver tissue, but Mdh-3and -4 were not (figure not shown). Gene prod-ucts of the Mdh-1 locus were less active than those of other loci in muscle. Mdh-2 was actively ex-pressed and diagnostic: diploid loach expressed the a allele and tetraploid loach exhibited the b al-lele. Common alleles both for diploid and tetrap-loid loach were detected in Mdh-3 and -4 loci.Triploid loach (n = 27) gave heterozygous geno-type abb comprising a single allele from the dip-loid (genotype aa ) and two alleles from the tetraploid (genotype bbbb ) loach (Fig. 2; Table 1).In aspartate aminotransferase (AAT) and isoci-trate dehydrogenase (IDHP), tetraploids gave polymorphisms, but the most frequent allele was that common to monomorphic diploids. Diploid and tetraploid loach showed the same phospho-glucomutase (PGM) phenotype because there was no allelic difference between the two. Thus, the loci encoding the above three enzymes were not considered diagnostic.

Survival and ploidy of progenies from

triploid × diploid crosses Normal fry developed from the small eggs of triploids fertilized with spermatozoa of normal diploids. The percentage of small eggs laid by trip-loid females (4016, 5010, and 5016) was 83.67 ±12.35% (mean ± S.D.). The mean percentage (± S.D.) of normal 5-day-old fry from the three females was 7.73 ± 4.49%, relative to the total number of small eggs used. Percentages of small eggs and normal 5-day-old fry in cross 2025 were described in Matsubara et al. (’95). Surviving fish from each cross were used for electrophoresis in 1995 and 1996. Most of the fish that developed from the small eggs after normal fertilization with diploid males were determined as diploidy by flow cytometry, but 5 of 24 fish in cross 4016 and 2 of

13 fish in cross 5016 were triploid.

Fig. 1.Photograph (A ) and schematic representation (B )of electrophoretic patterns for lactate dehydrogenase (LDH)of diploid, tetraploid, and triploid (diploid × tetraploid) loach.Presumptive genetic loci are given to the right of the gel. A:Lanes 1, 2: tetraploid (Ldh-1: bbbb ), Lanes 3, 4: diploid (Ldh -1: aa ), Lanes 5, 6: triploid (Ldh-1: ab b ). Arrows indicate homotetrameric products of diagnostic alleles. B: Solid sym-bols represent homotetramers a 4 in diploids and b 4 in tetraploids and homotetramers (a 4, b 4) and resultant hetero-tetramers (a 3b 1, a 2b 2, a 1b 3) in triploids. Numbers in paren-theses indicate theoretical intensities of the tetrameric products of Ldh-1 locus in triploids. Open symbols represent homo- and heterotetramic products from different loci.

ALLOZYMES IN DIPLOID PROGENIES OF TRIPLOID LOACH

371

Segregation in diploid progenies of

triploid × diploid crosses Based on the examination of the diagnostic loci,each triploid female was expected to have an abb genotype for each locus. Diploid males used for the

crosses were electrophoretically confirmed to be homozygous aa for Ldh-1 and Mdh-2 loci. In dip-loid progenies of triploid × diploid crosses, both ho-mozygous (aa ) and heterozygous (ab ) genotypes for Ldh-1 and Mdh-2 were detected (Fig. 3). Genotypes of Ldh-1 were not jointly segregated with those of Mdh-2: a homozygote for Ldh-1 was not always ho-mozygous for Mdh-2. These observations did not show whether small haploid eggs were generated hybridogenetically in triploid oogenesis from the two genomes of the tetraploid grandparent.

Genotype frequencies for the Ldh-1 and Mdh-2 loci are summarized in Table 2. Assuming a random pairing between two of three homologous chromosomes in triploid loach, a 1:2 ratio for homozygotes (aa ) and heterozygotes (ab ) was expected at each diagnostic locus. In Ldh-1,however, 4016 produced significantly more prog-eny with the heterozygous ab genotype. Because the male genotype for the crosses was homozy-gous (aa ), the excessive occurrence of ab indicates a selective formation of eggs carrying the b al-lele derived from the tetraploid grandparent in

TABLE 1.Ldh-1 and Mdh-2 allozyme genotypes of diploid, tetraploid, and triploid (diploid × tetraploid)

loach Misgurnus anguillicaudatus

No. of fish Ldh-1

Mdh-2Ploidy Origin/cross used

aa bbbb abb aa bbbb abb Diploid Sera 1

2323002300Mixed population 137********Sum 6060006000Tetraploid

30262902900290402616016001604028110110011040411101100110Sum 6706700670Triploid

025140014001402650050050278008008Sum

27

27

27

1

Wild population collected in Sera Town, Hiroshima Prefecture, 1994.

2

Six different crosses (3005–3010 from 1993 and four crosses from 1992) using wild loach caught in Sera Town, Hiroshima Prefecture.

Fig. 2.Photograph (A ) and schematic representation (B )of electrophoretic patterns for malate dehydrogenase (MDH)of diploid, tetraploid, and triploid (diploid × tetraploid) loach.presumptive genetic loci are given to the right of the gel. A:Lanes 1, 2: tetraploid (Mdh-2: bbbb ), Lanes 3, 4: diploid (Mdh -2: aa ), Lanes 5, 6: triploid (Mdh-2: ab b ). Arrows indicate homodimeric products of diagnostic alleles. B: Solid symbols represent homodimers a 2 in diploids and b 2 in tetraploids and homodimers (a 2, b 2) and the resulting heterodimer (ab) in triploids. Numbers in parentheses indicate theoretical inten-sities of the dimeric products of Mdh-2 locus in triploids. Open symbols represent homo- and heterodimeric products from different loci.

372K. ARAI AND M. MUKAINO

the oogenesis of triploidy. Only the cross 5016 fit a model of random segregation between two of three homologues (χ2 = 0.124).

An excess of heterozygotes (ab) for Mdh-2 was also observed in the three crosses, 2025, 5010, and 5016, and only one homozygote (aa) was detected in cross 5016. However, cross 4016 gave 4 homozy-gotes (aa) and 15 heterozygotes (ab) and fit the model for segregation after random pairing of three homologues (χ2 = 1.256).

DISCUSSION

In wild Japanese populations, no tetraploid lo-ach have been found, but specimens have been obtained from fish dealers (Arai et al., ’91). Al-though the exact origin of tetraploid loach is still

Fig. 3.Photographs and schemata of selected electro-phoretic patterns for lactate dehydrogenase (LDH: A, B) and malate dehydrogenase (MDH: C, D) in the diploid progeny of triploid × diploid loach (cross 4016) and the male parent (lane 1). Genotypes of diploid loach for Ldh-1 (A) and Mdh-2 (C) are given under each gel. In schemata (B, D), solid symbols show homomeric products (a4, b4 of LDH; a2, b2 of MDH) of diagnostic loci and resultant heteromeric products (a3b1, a2b2, a1b3 of LDH; ab of MDH) in diploid progeny. Diploid geno-types (aa, ab, b b) are shown in each schema, but theoreti-cally no homozygous bb has occurred in the progeny. Numbers in parentheses indicate theoretical intensities of the polymeric products of Ldh-1 (B) and Mdh-2 (D) loci. Open symbols rep-resent dimeric or tetrameric products from different loci.

TABLE 2.Observed (obs.) and expected (exp.) segregation of Ldh-1 and Mdh-2 allozyme genotypes in the diploid progeny from triploid (genotype:abb) × diploid (genotype:aa) crosses

Cross No. of fish Ldh-1Mdh-2

7c58dc471ed9ad51f01df295ed aa abχ2aa abχ2 20259obs.18—09—exp. (3.0)(6.0)(3.0)(6.0)

401619obs.217 4.390415 1.256 exp.(6.3)(12.7)(6.3)(12.7)

501111obs.110—011—exp.(3.7)(7.3)(3.7)(7.3)

501636obs.13230.12413515.125 exp.(12.0)(24.0)(12.0)(24.0)

Total75obs.1758 3.84057024.000 exp.(25.0)(50.0)(25.0)(50.0)

ALLOZYMES IN DIPLOID PROGENIES OF TRIPLOID LOACH373

unknown, the present electrophoretic studies re-

vealed that natural tetraploid loach had different

alleles from those of normal diploids, in Hiro-

shima, Japan, at least two diagnostic enzyme loci, Ldh-1 and Mdh-2. Allelic substitution observed may suggest the possibility of inter-subspecific or

inter-specific differentiation between the two

forms of loach. Triploid (normal diploid × natural

tetraploid) loach were reported to show male

sterility and unusual formation of triploid and hap-

loid eggs (Matsubara et al., ’95). These reproduc-

tive characteristics are similar to those reported in

interspecific hybrids of teleosts (Ojima et al., ’75;

Johnson and Wright, ’86; Hamaguchi and Sakai-

zumi, ’92; Sakaizumi et al., ’92, ’93; Cherfas et al.,

’94) and other lower vertebrates (for review, see

Dawley, ’89). Such an analogy between the triploid

loach and other hybrids supports the hypothesis

that the diploid and tetraploid loach may have di-

verged to different subspecies or species. Because

of the absence of population genetic studies, how-

ever, it is premature to conclude that we know the

origin and taxonomy of this species.

In the present breeding study, all the triploid

loach laid two types of eggs, large and small, and

most of the progeny developed from the artificial

fertilization of small eggs with spermatozoa of

normal diploid were diploid as described by

Matsubara et al. (’95) and Zhang and Arai (’96).

Thus, the present study shows that triploid lo-

ach produce small haploid eggs. Such unusual

modes of reproduction have been suggested to be

hybridogenesis (Günther et al., ’79; Tunner and

Heppich, ’81; Heppich et al., ’82; Nishioka and

Ohtani, ’84; Ohtani, ’93; Mable and Bogart, ’95).

However, studies of diagnostic Ldh-1 and Mdh-2

allozymes in diploid progeny of triploid (genotype abb) × diploid (genotype aa) crosses have shown that the haploid genome of one parent is not always entirely excluded in the oogenesis of trip-loid loach, and that haploid eggs are not gener-ated via a typical hybridogenetic mechanism, because a low frequency of homozygous aa geno-types was observed at each diagnostic locus. The present results have also revealed the frequent occurrence of the heterozygous ab genotype in the diploid progeny of triploid loach. This excess of heterozygosity may be a consequence of prefer-ential pairing between the two homologous chro-mosomes derived from the tetraploid parent. As in the hybridogenesis of allotriploid frogs (Günther et al., ’79; Nishioka and Ohtani, ’84; Ohtani, ’93; Mable and Bogart, ’95), homologous chromosomes transmitted from tetraploids are likely to pair because of their affinity in the first meiosis and resulting bivalents then segregate in a quasi-normal meiotic manner. Exotic non-ho-mologous chromosomes derived from one species were premeiotically excluded in the gametogonial cells of hybrid Poeciliopsis fish (Cimino, ’72) and allotriploid frogs (Tunner and Heppich, ’81; Heppich et al., ’82; Nishioka and Ohtani, ’84; Ohtani, ’93). The cytogenetic mechanisms for preferential pair-ing and elimination of chromosomes have not yet been elucidated in triploid loach developed from hy-bridization of diploid and tetraploid individuals. In hybridogenetic anurans, incompatibility of the cen-tromere structure between different species was suggested as a cause of the unusual behavior of chromosomes (Tunner and Heppich, ’81; Heppich et al., ’82). The preferential but non-hybridogenetic mode of oogenesis in triploid loach suggests less cy-togenetic divergence between diploid and tetraploid loach than between interspecific pairs that induce typical hybridogenesis in the hybrids.

Although triploid (diploid × tetraploid) loach regularly lay both triploid and haploid eggs (Matsubara et al., ’95; Zhang and Arai, ’96), high frequencies of unexpected triploids (15–21%) were recorded in the offspring of small eggs fertilized with haploid spermatozoa in crosses 4016 and 5016. These observations show that triploid lo-ach sometimes produce small diploid eggs. Such an elevation of egg ploidy status has been ex-plained by spontaneous inhibition of the second polar body released after fertilization, but this is a relatively rare event with a frequency that nor-mally does not exceed 1% to 1.5% (Cherfas, ’81). Triploidy was not detected in a total of 20 fish in the other two crosses examined and was not re-ported in 81 fish developed from small eggs of 9 triploid females fertilized with haploid spermato-zoa (Matsubara et al., ’95). Therefore, in the near future, we should examine possible mechanisms by which triploid loach produce diploid eggs.

ACKNOWLEDGMENTS

This work was supported in part by a Grant-in-Aid from the Fisheries Agency of Japan for Research on Breeding of Aquatic Organisms to K.A. We thank Dr. Quanqi Zhang, Laboratory of Aquaculture, Graduate School of Biosphere Sci-ence, Hiroshima University for discussing the ideas in this manuscript with us.

LITERATURE CITED

Aebersold, P.B., G.A. Winans, D.J. Teel, G.B. Milner, and F.M. Utter (1987) Manual for starch gel electrophoresis: A method for the detection of genetic variation. NOAA Tech. Rep. NMFS, 61:1–19.

374K. ARAI AND M. MUKAINO

Allendorf, F.W., N. Mitchell, N. Ryman, and G. Stahl (1977) Isozyme loci in brown trout (Salmo trutta L.): detection and interpretation for population data. Hereditas, 86:179–190. Arai, K., M. Ikeno, and R. Suzuki (1995) Production of an-drogenetic diploid loach Misgurnus anguillicaudatus us-ing spermatozoa of natural tetraploids. Aquaculture, 137;131–138.

Arai, K., K. Matsubara, and R. Suzuki (1991) Karyotype and erythrocyte size of spentaneous tetraploid and triploid in the loach Misgurnus anguillicaudatus. Nippon Suisan Gakkaishi, 57:2167–2172.

Arai, K., K. Matsubara, and R. Suzuki (1993) Production of polyploids and viable gynogens using spontaneously occur-ring tetraploid loach Misgurnus anguillicaudatus. Aquac-ulture, 117:227–235.

Arai, K., and M. Mukaino (1997) Clonal nature of gynogenetically induced progeny of triploid (diploid × tetraploid) loach, Misgurnus anguillicaudatus (Pisces: Cobitididae). J. Exp. Zool., 278:412–421.

Cherfas, N.B. (1981) Gynogenesis in fish. In: Genetic Bases of Fish Selection. V.S. Kirpichnikov, ed. Springer-Verlag, Berlin, pp. 199–214.

Cherfas, N.B., B.I. Gomelsky, O.V. Emelyanova, and A.V. Recoubratsky (1994) Induced diploid gynogenesis and poly-ploidy in crucian carp, Carassius auratus gibelio (Bloch), ×common carp, Cyprinus carpio L., hybrids. Aquacult. Fish. Manage. 25:943–954.

Cimino, M.C. (1972) Egg-production, polyploidization, and evo-lution in a diploid all-female fish of the genus Poeciliopsis. Evolution, 26:294–306.

Clayton, J.W., and D.N. Tretiak (1972) Amine-citrate buffer for pH control of starch gel electrophoresis. J. Fish. Res. Bd. Can., 29:1169–1172.

Dawley, R.M. (1989) An introduction to unisexual vertebrates. In: Evolution and Ecology of Unisexual Vertebrates. R.M. Dawley and J.P. Bogart, eds. Bull. 466, New York State Mu-seum, Albany, NY, pp. 1–18.

Goddard, K.A., and R.J. Schultz (1993) Aclonal reproduction by polyploid member of the clonal hybrid species Phoxinus eosneogaeus (Cyprinidae). Copeia, 1993:650–660.

Günther, R., T. Uzzell, and L. Berger (1979) Inheritance pat-tern in triploid Rana esculenta (Amphibia: Salientia). Mitt. Zool. Mus. Berlin, 55:35–57.

Hamaguchi, S., and M. Sakaizumi (1992) Sexually differen-tiated mechanisms of sterility in interspecific hybrids be-tween Oryzias latipes and O. curvinotus. J. Exp. Zool., 263:323–329.

Heppich, S., H.G. Tunner, and J. Greilhugen (1982) Premei-otic chromosome doubling after genome elimination during spermatogenesis of the species hybrid Rana esculenta. Theor. Appl. Genet., 61:101–104.Johnson K.R., and J.E. Wright (1986) Female brown trout ×Atlantic salmon hybrids produce gynogens and triploids when backcrossed to male Atlantic salmon. Aquaculture, 57:345–358.

Mable, B.K., and J.P. Bogart (1995) Hybridization between tetraploid and diploid species of treefrogs (genus Hyla). J. Hered., 86:432–440.

Matsubara, K., K. Arai, and R. Suzuki (1995) Survival po-tential and chromosomes of progeny of triploid and pentap-loid females in the loach, Misgurnus anguillicaudatus. Aquaculture, 131:37–48.

Nishioka, M., and H. Ohtani (1984) Hybridogenetic repro-duction of allotriploids between Japanese and European pond frog. Zool. Sci., 1:291–316.

Ohtani, H. (1993) Mechanism of chromosome elimination in the hybridogenetic spermatogenesis of allotriploid males between Japanese and European water frogs. Chromosoma, 102:158–162.

Ojima, Y., M. Hayashi, and K. Ueno (1975) Triploidy appeared in the back-cross offspring from funa-carp crossing. Proc. Japan Acad., 51B:702–706.

Ojima, Y., and A. Takai (1979) The occurrence of spontane-ous polyploid in the Japanese common loach Misgurnus anguillicaudatus. Proc. Japan Acad., 55B:487–491. Sakaizumi, M., Y. Shimizu, and S. Hamaguchi (1992) Elec-trophoretic studies of meiotic segregation in inter- and in-traspecific hybrids among East Asian species of the genus Oryzias (Pisces: Oryziatidae). J. Exp. Zool., 264:85–92. Sakaizumi, M., U. Shimizu, T. Matsuzaki, and S. Hamaguchi (1993) Unreduced diploid eggs produced by interspecific hy-brids between Oryzias latipes and O. curvinotus. J. Exp. Zool., 266:312–318.

Schultz, R.J. (1963) Reproductive mechanisms of unisexual and bisexual strains of the viviparous fish Poeciliopsis. Evo-lution, 15:302–325.

Schultz, R.J. (1966) Hybridization experiments with an all-female fish of the genus Poeciliopsis. Biol. Bull. Woods Hole, 130:415–429.

Schultz, R.J. (1969) Hybridization, unisexuality and polyp-loidy in the teleost Poeciliopsis (Poeciliidae) and other ver-tebrates. Am. Nat., 103:605–619.

Suzuki, R., T. Oshiro, and T. Nakanishi (1985) Survival, growth and fertility of gynogenetic diploids induced in the cyprinid loach, Misgurnus anguillicaudatus. Aquaculture, 48:45–55.

Tunner, H.G., and S. Heppich (1981) Premeiotic genome ex-clusion during oogenesis in the common edible frog, Rana esculenta. Naturwissenschaften, 68:207–208.

Zhang, Q., and K. Arai (1996) Flow cytometry for DNA con-tents of somatic cells and spermatozoa in the progeny of natural tetraploid loach. Fish. Sci., 62:870–877.

本文来源：https://www.bwwdw.com/article/fsde.html