With advances in human genome project, the structural analyses of human genome will be completed by the beginning of the 21st century. However, nucleotide sequences do not give sufficient information about the biological functions of genome. Thus, we need another approaches to analyze the biological functions. One of the promising approaches is the gene trap mutagenesis. Gene trapping in embryonic stem (ES) cells has been developed as a powerful tool for identifying novel, developmentally regulated genes in mouse embryos. When the gene trap vector inserts into cellular genes, a fusion transcript between the endogenous gene and the reporter gene is produced, so that the expression pattern of the tagged gene can be traced by simple histochemical staining of the reporter gene. Furthermore, in many cases, the en-dogeneous gene is disrupted by the insertion, which can lead to loss of function. Trapped clones can be analyzed for significant mutations that give rise to mutant phenotypes after germline transmission. In addition, the reporter gene insertion facilitates cloning cDNAs of exon (s) upstream of the insertion site by rapid amplification of cDNA 5'-end (5'RACE) . Thus, the gene trap approach has a number of advantages for functional analysis of the genome. However, it has one shortcoming that is its relative inability to induce subtle mutations. Recently, we have demonstrated knock-in system in ES cells using Cre-mutant lox system, which was first reported by Albert, H. et al. (1995) . In this system, LE mutant lox, lox71, which has 5 base pair changed in the left end and RE mutant lox, lox66, which has 5 base pair changed in the right end are used. In the case of recombination between loxP sites, the integrated DNA flanked by two loxP sites is easily removed again if Cre recombinase is present. Consequently, the insertion product is unstable during transient Cre expression. On the other hand, recombination between a lox71 site and a lox66 site produces a double (LE+RE) mutant site and a wild-type loxP site. Since the LE+RE mutant site exhibits reduced binding affinity for Cre recombinase, the inserted DNA is stable. By applying the Cre-mutant lox system to gene trapping, we have developed “exchangeable gene trapping system”. In this system, we can carry out random insertional mutagenesis as the first, and then we can introduce another DNA fragment that we want to express under the control of the trapped gene through Cre-mediated integration in the second step to induce any type of mutation.

View full abstract

Mouse hair mutants are useful animal models to study molecular mechanism of hair development and human hair disease. A spontaneous mouse hair mutation, Hague (Japanese word for bald) on Chromosome 15 was recovered in C3H/HeN mice colony. In this mutant colony, the semidominant allele is unstable and easily turn to be recessive one. The purpose of this study is to clarify histopathological characteristics of the mutant hair follicle and to examine the expression of a novel type II cytokeratin, keratin2-6g in the mutant and control mice. We compared immunohistologically and by in situ hybridization the dorsal skin hair follicles of wild type (+/+), semidominant heterozygotes (Hag/+), semidominant homozygotes (Hag/Hag) and recessive homozygotes (hag/hag) . The results indicated that the morphology of hair follicles and hair shafts of each genotype clearly differed to each other. In Hag/Hag mice the inner root sheath (IRS) cells were degenerated and disrupted by large vacuolation. In Hag/+ mice vacuolation also detected in the IRS. But the number of vacuoles were less than those detected in Hag/Hag mice. In hag/hag mice vacuolation was not observed, but instead, tricohyaline granules in the IRS cells were found larger than those of wild type mice. In situ hybridization and immunohistochemistry revealed that the expression of keratin2-6g gene was only small amount and the keratin protein was not detectable in the mutant. These results indicated that keratin2-6g in the IRS was essential for hair follicle development, and Hag and hag alleles had different affects on the IRS cells.

View full abstract

Waardenburg syndrome (WS) is an auditory-pigmentary syndrome associated with deafness and iris/hair/skin pigmentation anomaly. Depending on various other symptoms, WS is classified into four types: WS type 1 (WS1), WS2, WS3 and WS4. While WS2 is not associated with any additional symptoms, WS1, WS2 and WS4 are associated with, respectively, distopia canthorum (wide distance between inner canthi of eyes), upper limb deformity, and Hirschsprung disease (aganglionic megacolon) . Thanks to the existence of mouse models, many WS genes have been identified: Genomic analysis of splotch (sp) mice, for example, resulted in identification of the Pax-3 gene, whose human homolog was later found mutated in WS1 and WS3 individuals. Similarly, analysis of VGA-9 mice, which have an insertional mutation at microphthalmia locus, identified the Mitf (microphthalmia-associated transcription factor) gene, and later mutation of its human homolog was found in WS2 individuals. Cochleas of VGA-9 mice showed cochleo-saccular degeneration due to lack of melanocytes (intermediate cells) in stria vascularis. Since MITF mutations have been found in only -15% of WS2 individuals, other genes probably are also play a part: One possibility is that mutations of genes responsible for other types of WS also cause WS2 owning to, for example, differences in genetic backgrounds. This hypothesis should be testable using mouse models of WS.
We recently reported mutant mice with white coat color that die around one month after birth due to aganglionic megacolon. Breeding analyses of these mice revealed that these mice were allelic with piebald-lethal and JF1 mice but not with lethal spotting mice, indicating that they are mutated in Ednrb (the endothelin receptor gene B), whose human homolog is mutated in WS4 individuals. Genomic analysis did, in fact, revealed that exons 2 and 3 of Ednrb are deleted. These mice did not respond to sound, and their cochleas showed changes similar to VGA-9 mice: Stria vascularis lacked melanocytes and hair cells of the organ of Corti showed severe degeneration. We therefore proposed these mice as mouse model of WS4 and named them WS4 mice. Lethal-spotting (ls) mice are mutated in the endothelin-3 gene (Edn3), and Dominant megacolon (Dom) mice are mutated in Sox10: Both showed white spotting and aganglionic megacolon, and thus are good candidate for WS4 models. Disorder of cochleas of these mice, however, remains to be confirmed.

View full abstract

Ischemic heart disease is important in developed countries. However, there are few suitable animal models for myocardial infarction. To develop myocardial infarction (MI) -prone WHHL rabbits, we have carried out selective breeding by using coronary atherosclerosis-prone WHHL rabbits in our colony since 1994. We used following indices for the selective breeding; 1) vulnerable coronary plaque, 2) severe coronary stenosis greater than the lumen area stenosis with the mean value plus the standard error value of the parents' generation, 3) plasma cholesterol levels above 800 mg/dl at 12 months old, and 4) occurrence of myocardial infarction. During the selective breeding, housing condition was maintained constant and rabbits were fed standard rabbit chaw. We did not treat rabbits especially except the selective breeding. At 1999, after the selective breeding, myocardial infarction was observed in 94% of rabbits deceased up to 35 months old. Age of 35 months old in rabbits corresponds to approximately 44 years old in human by calculation of their life span (75 years old in humans and 60 months old in experimen-tal rabbits) . In histological examination, many rabbits after the selective breeding showed findings of acute myocardial infarction alone or in combination with findings of old myocardial infarction. Myocardial lesions of MI-prone WHHL rabbits were observed mainly at the subendocardial region circumf erentially, and about 60% rabbits showed transmural infarction in combination with subendocardial infarction. In some MI-prone WHHL rabbits, the profile of the electrocardiogram showed typical findings of myocardial infarction. The culprit coronary arteries had severe lumen area stenosis greater than 90% at about 20 serial sections prepared at 500μm interval. These findings of MI-prone WHHL rabbits resembled those of human multivessel disease. Atheromatous plaques of the coronary arteries of WHHL rabbits were changed from fibromuscular type to macrophage/extracellular lipid-rich type by the selective breeding. Many coronary plaques showed occlusion by macrophage accumulation at the superficial area of the plaque. In this case, denudation of the endothelial cells was observed. Some others showed thin fibromuscular cap with large lipid core. In this case, macrophages were detected at the border between the attenuated fibromuscular cap and the lipid core. This finding was considered as vulnerable plaque, which probably causes plaque rupture and consequent occlusive thrombus formation. However, we did not detect any plaque rupture or luminal thrombus in the coronary arteries. This suggests that we can examine which factors are important as trigger of plaque rupture by using MI-prone WHHL rabbits. We consider that MI-prone WHHL rabbits will contribute to studies of myocardial infarction.

View full abstract