1. The Origin of Reproductive Isolation
11
X-linked factor involved (Sawamura et al. 1993b; K. Sawamura, unpublished observation).
The mutation, zhr, was found to be a deficiency of a specific region of the X
heterochromatin, and has been precisely mapped by using deficiencies and duplications of the X heterochromatin (Sawamura and Yamamoto 1993). The viability/
inviability of hybrids is determined by the absence/existence of the locus, so the
incompatible component of D. melanogaster is neomorphic. The region where zhr
was localized is rich in a 359 bp repetitive sequence of satellite DNA whose buoyant density is 1.688 g/cm3 (Lohe et al. 1993). Further, it is assumed that the gene
consists of a kind of repetitive sequence based on its quantitative effect (Sawamura
et al. 1995; Sawamura and Yamamoto 1997). The 1.688 satellite DNA itself or
some repetitive sequences embodied in it (e.g., transposable elements) seem to be
the factor responsible for the embryonic hybrid inviability. If the sequences have
some extent of homology to the recognition sites of a DNA binding protein [in fact,
1.688 satellite DNA-related sequences exist on several locations of the X euchromotin
(di Bartolomeis et al. 1992)], they will titrate out the maternally-supplied binding
factor(s) necessary to regulate some essential genes. Coevolution of the transcription factor and its binding sites including the titration region will cause this incompatibility between species. Besides transcriptional regulation, the DNA binding
protein may play an important role in chromosome condensation and/or chromosome segregation. It is interesting in this context that some heterochromatic satellite DNAs bound by well-known DNA-binding proteins (e.g., GAGA factor, Proliferation Disruptor protein) are species-specific (Platero et al. 1998).
3 Conclusion
It is apparent that postzygotic reproductive isolation is the manifestation of interspecific genetic incompatibility. A gene that functions normally in a pure species
may cause an anomaly in the background of different species. Most of the cases arc
not the consequences of allelic interaction (underdominance) but of gene interaction (epistasis): genetic incompatibility usually involves at least a pair of loci
(Dobzhansky 1937; Muller 1940; Hutter et al. 1990; Sawamura et al. 1993c; Wu
and Davis 1993; Wu and Palopoli 1994; Orr 1997; Coyne and Orr 1998). This is
because popUlations cannot cross the adaptive valley if a single locus is responsible
for the unadapted genotype (Fig. 1). Assume one locus, A, whose allelic status is
indicated by subscripts (1 and 2). The missing link betwecn A r41 and Az-42 species
should be A r42' an inferior genotype. On the other hand, if two loci, A and B, are
involved, Ar4~lBl proto-species can evolve to Az-4~~l and Ar4~~2 species
throughAr4~lBl and Ar4!l!l2 genotypes, respectively, avoiding the inferior genotype equivalent to the interspecific hybrids, Ar4~~2. Theoretically, any genetic
interactions in pure species can cause genetic incompatibility in interspecific hybrids, and evolutionary geneticists haven't seriously cared about its biological mechanisms so far. Recent advances in developmental biology have shown that gene regu-
11
X-linked factor involved (Sawamura et al. 1993b; K. Sawamura, unpublished observation).
The mutation, zhr, was found to be a deficiency of a specific region of the X
heterochromatin, and has been precisely mapped by using deficiencies and duplications of the X heterochromatin (Sawamura and Yamamoto 1993). The viability/
inviability of hybrids is determined by the absence/existence of the locus, so the
incompatible component of D. melanogaster is neomorphic. The region where zhr
was localized is rich in a 359 bp repetitive sequence of satellite DNA whose buoyant density is 1.688 g/cm3 (Lohe et al. 1993). Further, it is assumed that the gene
consists of a kind of repetitive sequence based on its quantitative effect (Sawamura
et al. 1995; Sawamura and Yamamoto 1997). The 1.688 satellite DNA itself or
some repetitive sequences embodied in it (e.g., transposable elements) seem to be
the factor responsible for the embryonic hybrid inviability. If the sequences have
some extent of homology to the recognition sites of a DNA binding protein [in fact,
1.688 satellite DNA-related sequences exist on several locations of the X euchromotin
(di Bartolomeis et al. 1992)], they will titrate out the maternally-supplied binding
factor(s) necessary to regulate some essential genes. Coevolution of the transcription factor and its binding sites including the titration region will cause this incompatibility between species. Besides transcriptional regulation, the DNA binding
protein may play an important role in chromosome condensation and/or chromosome segregation. It is interesting in this context that some heterochromatic satellite DNAs bound by well-known DNA-binding proteins (e.g., GAGA factor, Proliferation Disruptor protein) are species-specific (Platero et al. 1998).
3 Conclusion
It is apparent that postzygotic reproductive isolation is the manifestation of interspecific genetic incompatibility. A gene that functions normally in a pure species
may cause an anomaly in the background of different species. Most of the cases arc
not the consequences of allelic interaction (underdominance) but of gene interaction (epistasis): genetic incompatibility usually involves at least a pair of loci
(Dobzhansky 1937; Muller 1940; Hutter et al. 1990; Sawamura et al. 1993c; Wu
and Davis 1993; Wu and Palopoli 1994; Orr 1997; Coyne and Orr 1998). This is
because popUlations cannot cross the adaptive valley if a single locus is responsible
for the unadapted genotype (Fig. 1). Assume one locus, A, whose allelic status is
indicated by subscripts (1 and 2). The missing link betwecn A r41 and Az-42 species
should be A r42' an inferior genotype. On the other hand, if two loci, A and B, are
involved, Ar4~lBl proto-species can evolve to Az-4~~l and Ar4~~2 species
throughAr4~lBl and Ar4!l!l2 genotypes, respectively, avoiding the inferior genotype equivalent to the interspecific hybrids, Ar4~~2. Theoretically, any genetic
interactions in pure species can cause genetic incompatibility in interspecific hybrids, and evolutionary geneticists haven't seriously cared about its biological mechanisms so far. Recent advances in developmental biology have shown that gene regu-