1. The Origin of Reproductive Isolation
5
tail (X, helleri) have a uniform grey body colouration due to small black pigment
cells (micromelanophores). Some geographical races of platifish, however, exhibit
spot patterns of melanophores that are much larger than normal (macromelanophores). These patterns are determined by a sex-linked locus which is closely
linked to an oncogene, Tumor (Tu) (Weis and Schartl 1998). Tu-bearing platifish,
by nature, do not suffer from melanoma, because they are assumed to have an autosomal dominant suppressor gene, Regulator (R). The platifish with spots are genotypically described as Tu/Tu; R/R. On the contrary, the swordtails have neither of
these genes: in this respect, described as -/-; -/-. F1 hybrids are Tu/-; R/-, and show
enhanced expression of the pigmentation pattern (but still benign melanoma). When
the F1 hybrids are backcrossed to swordtails, four genotypic classes are produced. A
quarter of the Be 1 hybrids are Tu/-; -/-, and suffer from malignant melanoma.
The Tu gene was cloned by Wittbrodt et al. (1989) and was found to be a novel
putative receptor tyrosine kinase, now named Xiphophorus melanoma receptor kinase (Xmrk). The sequence analysis of the oncogene (ONC-Xmrk) and the protooncogene (INV-Xmrk) (Adam et al. 1993) indicated that Tu was an Xmrk duplication produced by a non-homologous recombination. The second copy has a 5' region derived from an anonymous locus, designated D, rather than the normal 5' end
of the original copy. The consequence of this event is that INV-Xmrk and ONCXmrk are subject to different transcriptional regulation. Specific overexpression of
ONC-Xmrk seems to be responsible for melanoma formation. It is assumed that the
R-locus-dependent transcriptional control of the oncogene promoter allows high
levels of expression only in pigment cells of certain hybrid genotypes. Then, what
is the product of the R locus'? Is it a transcription factor? The gene has been located
to linkage group V and a candidate, CDKN2-like, is already cloned (Nairn et al.
1996). Anyway, the control of melanomas by the R locus can thus be viewed as "an
accidental side effect of the regulation R exerts on D" (Friend 1993).
2.2 Sexual Reversion
Sexual reversion of hybrids is well-known in several crosses. A classical example is
inter-racial hybrids of the gypsy moth (Lyman tria dispar) [for a review see
Goldschmidt (1940)]. This phenomenon was attributed to the geographical variation of the "strength" of sex determinants. In crosses where the strength of the sex
determinants in the parental races differ, intersexual hybrids are produced. When
the difference is slight, intersexes are so much like normal females or males that
they are fertile. Where the difference is greater, however, the intersexes are sterile.
When the difference is much greater, the intersexes transform into individuals of
the sex opposite to the chromosomal sex. Similar cases are known in several lepidopteran interspecific crosses [cited in Haldane (1922)]. Although more detailed
genetic bases have not been elucidated, it may be explicable by more elaborate
terms of sex determination. In Drosophila, for example, sex is determined by the
X-chromosome/autosome (X/A) ratio. The X chromosome has numerator genes
5
tail (X, helleri) have a uniform grey body colouration due to small black pigment
cells (micromelanophores). Some geographical races of platifish, however, exhibit
spot patterns of melanophores that are much larger than normal (macromelanophores). These patterns are determined by a sex-linked locus which is closely
linked to an oncogene, Tumor (Tu) (Weis and Schartl 1998). Tu-bearing platifish,
by nature, do not suffer from melanoma, because they are assumed to have an autosomal dominant suppressor gene, Regulator (R). The platifish with spots are genotypically described as Tu/Tu; R/R. On the contrary, the swordtails have neither of
these genes: in this respect, described as -/-; -/-. F1 hybrids are Tu/-; R/-, and show
enhanced expression of the pigmentation pattern (but still benign melanoma). When
the F1 hybrids are backcrossed to swordtails, four genotypic classes are produced. A
quarter of the Be 1 hybrids are Tu/-; -/-, and suffer from malignant melanoma.
The Tu gene was cloned by Wittbrodt et al. (1989) and was found to be a novel
putative receptor tyrosine kinase, now named Xiphophorus melanoma receptor kinase (Xmrk). The sequence analysis of the oncogene (ONC-Xmrk) and the protooncogene (INV-Xmrk) (Adam et al. 1993) indicated that Tu was an Xmrk duplication produced by a non-homologous recombination. The second copy has a 5' region derived from an anonymous locus, designated D, rather than the normal 5' end
of the original copy. The consequence of this event is that INV-Xmrk and ONCXmrk are subject to different transcriptional regulation. Specific overexpression of
ONC-Xmrk seems to be responsible for melanoma formation. It is assumed that the
R-locus-dependent transcriptional control of the oncogene promoter allows high
levels of expression only in pigment cells of certain hybrid genotypes. Then, what
is the product of the R locus'? Is it a transcription factor? The gene has been located
to linkage group V and a candidate, CDKN2-like, is already cloned (Nairn et al.
1996). Anyway, the control of melanomas by the R locus can thus be viewed as "an
accidental side effect of the regulation R exerts on D" (Friend 1993).
2.2 Sexual Reversion
Sexual reversion of hybrids is well-known in several crosses. A classical example is
inter-racial hybrids of the gypsy moth (Lyman tria dispar) [for a review see
Goldschmidt (1940)]. This phenomenon was attributed to the geographical variation of the "strength" of sex determinants. In crosses where the strength of the sex
determinants in the parental races differ, intersexual hybrids are produced. When
the difference is slight, intersexes are so much like normal females or males that
they are fertile. Where the difference is greater, however, the intersexes are sterile.
When the difference is much greater, the intersexes transform into individuals of
the sex opposite to the chromosomal sex. Similar cases are known in several lepidopteran interspecific crosses [cited in Haldane (1922)]. Although more detailed
genetic bases have not been elucidated, it may be explicable by more elaborate
terms of sex determination. In Drosophila, for example, sex is determined by the
X-chromosome/autosome (X/A) ratio. The X chromosome has numerator genes