Hidenori Hayashida

Close structural resemblance between putative polymerase of a Drosophila transposable genetic element 17.6 and pol gene product of Moloney murine leukaemia virus.

We have made a computer‐assisted search for homology among polymerases or putative polymerases of various viruses and a transposable element, the Drosophila copia‐like element 17.6. The search revealed that the putative polymerase (second open reading frame) of the copia‐like element 17.6 bears close resemblance in overall structural organization to the pol gene product of Moloney murine leukaemia virus (M‐MuLV): they show significant homology to each other at both the N‐ and C‐terminal portions, suggesting that the 17.6 putative polymerase carries two enzymatic activities, related to reverse transcriptase and DNA endonuclease. The putative polymerase of cauliflower mosaic virus (CaMV) shows striking homology with the putative polymerase of 17.6 over almost its entire length, but it lacks the DNA endonuclease‐related sequence. Furthermore, it was shown that the N‐terminal ends of the M‐MuLV pol product and the CaMV and 17.6 putative polymerases exhibit strong sequence homology with the gag‐specific protease (p15) of Rous sarcoma virus (RSV) as well as the amino acid sequence predicted from the gag/pol spacer sequence of human adult T‐cell leukaemia virus (HTLV). These p15‐related sequences contain a highly conserved stretch of amino acids which show a close similarity with sequences around the active site amino acids Asp‐Thr‐Gly of the acid protease family, suggesting that they have an activity similar to acid protease. On the basis of the alignment of reverse transcriptase‐related sequences, a dendrogram representing phylogenetic relationships among all the viruses compared together with 17.6 was constructed and its evolutionary implication is discussed.

Molecular clock of silent substitution: At least six-fold preponderance of silent changes in mitochondrial genes over those in nuclear genes

SummaryFor each of eleven different types of nuclear genes, comparisons of the protein coding sequences were made between human, mouse and rat pairwisely, and the evolutionary rate of silent substitution, vSnucl., was estimated. It is shown that the vSnucl. is not only very high (=5.37×10−9/site/yr), but also approximately uniform for different genes regardless of the types, which confirms our previous results (Miyata et al. 1980b). This is in sharp contrast to the rate of protein evolution which differes greatly from protein to protein. Furthermore the vSnucl. is shown to be approximately constant with respect to different divergence times, at least within a short time period (≤75 Myr). Based on these observations, we propose a new molecular clock which has several advantages over a protein clock. Using this clock, we show that the rate of amino acid replacement in the immunoglobulin Ck gene of b4 rabbit is unexpectedly high, almost comparable to the rate of silent changes. This rate may be the highest one for protein evolution that we know so far. We further examine the rate of silent substitutions in mitochondrial genes comparing mouse and rat. Surprisingly the rate is extremely high (≥35×10−9/site/yr), at least 6-times as high as the corresponding rate of nuclear genes. Based on the estimate, we discuss a possible origin of the rapid rate found in mitochondrial DNA.

The nucleotide sequence of the mouse immunoglobulin epsilon gene: comparison with the human epsilon gene sequence.

We have determined the nucleotide sequence of the immunoglobulin epsilon gene cloned from newborn mouse DNA. The epsilon gene sequence allows prediction of the amino acid sequence of the constant region of the epsilon chain and comparison of it with sequences of the human epsilon and other mouse immunoglobulin genes. The epsilon gene was shown to be under the weakest selection pressure at the protein level among the immunoglobulin genes although the divergence at the synonymous position is similar. Our results suggest that the epsilon gene may be dispensable, which is in accord with the fact that IgE has only obscure roles in the immune defense system but has an undesirable role as a mediator of hypersensitivity. The sequence data suggest that the human and murine epsilon genes were derived from different ancestors duplicated a long time ago. The amino acid sequence of the epsilon chain is more homologous to those of the gamma chains than the other mouse heavy chains. Two membrane exons, separated by an 80‐base intron, were identified 1.7 kb 3′ to the CH4 domain of the epsilon gene and shown to conserve a hydrophobic portion similar to those of other heavy chain genes. RNA blot hybridization showed that the epsilon membrane exons are transcribed into two species of mRNA in an IgE hybridoma.

Nucleotide sequences of immunoglobulin-epsilon pseudogenes in man and apes and their phylogenetic relationships☆

To understand the phylogenetic relationships between hominoids, the nucleotide sequences of immunoglobulin-epsilon processed pseudogenes from chimpanzee, gorilla and orangutan were determined. The basic structures of these processed pseudogenes agreed with their human counterpart. Although the degrees of nucleotide differences between man and the African apes had no statistical significance, all the analytical data examined supported the theory that chimpanzee is the closest relative of man. This result was consistent with that deduced by our recent qualitative study. Studies on the nucleotide sequences of globin genes have suggested that the molecular clock runs more slowly in hominoids than in non-hominoid primates. According to the present data, however, further retardation of the evolutionary rate was not observed in the human lineage. Assuming that orangutan diverged 14 million years ago and that the evolutionary rate between the orangutan lineage and the lineage leading to the other three species is constant, the divergence dates of chimpanzee and gorilla were estimated to be 4.9(+/- 0.9) and 5.9(+/- 0.9) million years ago, respectively.

Concerted evolution of the mouse immunoglobulin gamma chain genes.

The nucleotide sequences of the immunoglobulin heavy‐chain constant region genes of mouse, C gamma 3, C gamma 1, C gamma 2b and C gamma 2a, together with that of a human equivalent C gamma 4 were compared. All the six pairs of genes within the mouse C gamma gene family contain DNA segments that exhibit marked homology, whereas no such segmental homology was found in interspecies comparisons. This result indicates that the four C gamma genes of the mouse evolved concertedly by exchanging parts of their genetic information with each other either by gene conversion or by double unequal crossing‐over. Another example of such concerted evolution was found in gene regions encoding membrane domains of the mouse C gamma chains. We also searched for such segmental homologies in other mammalian C gamma gene families and found at least two more examples in man and guinea‐pig. In the mouse C gamma gene family, the silent positions of an exon encoding the third domain of C gamma chains show much greater divergence in sequence than other regions, indicating that the genetic information encoded by this gene region was least scrambled during recent evolution. A phylogenetic tree constructed from the nucleotide differences of this exon demonstrates that at least two C gamma genes had already existed before mammalian radiation. Based on these results, evolution of mammalian C gamma gene families is discussed.

Interchromosomal gene conversion as a possible mechanism for explaining divergence patterns of ZFY-related genes

SummaryThe divergence pattern of mammalian ZFY-related genes from human (ZFY and ZFX) and mouse (Zfy-1 and Zfx) was reexamined on the basis of nucleotide substitutions at the synonymous codon-alternating positions. It is possible to explain the unusual divergence pattern of the mammalian Y-linked ZF genes by interchromosomal gene conversion by X-linked ZF genes. Furthermore, the rates of evolution of mammalian X- and Y-linked ZF genes were shown to agree well with those expected from our model.

Sequence similarity between epidermal growth factor precursor and atrial natriuretic factor precursor

Computer‐assisted analysis for homology of the EOF precursor revealed the presence of two large duplication units, each comprising 5 non‐EGF‐like homologous segments each of about 40 residues length and 3 or 4 EGF‐like segments. The amino acid sequences of the non‐EGF‐like repeats were subjected to search for homology with 2600 known protein sequences compiled in our database. An unexpected but statistically significant homology has been found, when compared with the atrial natriuretic factor precursor. The functional and evolutionary implications of the homology observed between the two different precursors are also discussed.