Search

Statistical analysis of parental age data from 225 sporadic cases of bilateral retinoblastoma, plus ten sporadic cases of chromosome deletion or translocation involving 13q14 that was identified as of paternal origin, revealed no evidence of paternal or maternal age effect. Parental exposure to ionizing radiation or chemical mutagens, the effect of which is accumulated with advancing age, does not seem to play a major role in the production of germinal mutations at the responsible (RB) locus. Furthermore, analysis of variation in the month of birth of 753 children with sporadic unilateral retinoblastoma did not show any significant deviation from the controls or a cyclic trend. The occurrence of nonheritable retinoblastoma is not likely to be associated with certain viruses such as human adenovirus 12 whose activity varies markedly with season. These results, together with the fairly uniform pattern in the incidence of this tumor among different populations, suggest that most, if not all, cases of sporadic retinoblastoma are caused by some intrinsic biological mechanisms, and not by environmental mutagens that may vary with respect to time and place.

We analyzed mitochondrial DNA purified from the livers of 10 Japanese monkeys, Macaca fuscata, derived from four local populations using the cleavage patterns for 18 restriction enzymes. Of the 18 enzymes used, polymorphism of the cleavage patterns was observed in 12 enzyme digestions. By combining the cleavage patterns for each enzyme, the 10 samples were classified into four restriction types. Each of the four types was found exclusively in the respective population from which the samples were obtained. Each type had 43 to 49 sites for the 18 enzymes. The average number of nucleotide substitution per site was estimated to be 0.0132. This value is several times larger than that observed in human population but at approximately the same level compared with those observed in great apes and rodents. The phylogenetic tree constructed for the four types indicate that mtDNAs of the 10 Japanese monkeys examined were divided into two clusters and these two clusters diverged approximately one million years ago. This work has proved that the restriction enzyme analysis is useful for elucidating genetic variations in Japanese monkeys.

The size of C-positive qh regions of chromosomes 1, 9 and 16 was determined using a Nikon-Vickers M85 microdensitometer. Results (area) were expressed based on how different they were from an average of each 1qh, 9qh or 16qh region in terms of standard deviations (SD). It was proposed that a variant within ±1 SD of an average be assigned to the intermediate class (class 3) in the Paris Conference system (Suppl. 1975), one within ±2 SD to either small or large class (class 2 or 4) and all others to either very small or very large (class 1 or 5). The present system compares favorably with any other methods so far tested as to the detection of a qh variant.

It is generally accepted that the overall process of carcinogenesis, with its long latent period, usually comprises at least two steps, initiation and promotion, each of which is probably controlled or modified by genetic and environmental factors. Retinoblastoma is a malignant neoplasm arising from the transformation of an embryonic retinal cell that may involve the minimal number of carcinogenic steps. For this and several other reasons, it is a cancer most suitable for studying the mechanisms of carcinogenesis in humans. This article reviews recent studies on the genetics and genesis of retinoblastoma with special reference to the host resistance theory proposed by the author (Matsunaga, 1978; 1979). Main conclusions are as follows: 1) Among all retinoblastomas, the ratio of unilateral to bilateral cases is approximately 2∶1. The vast majority occur sporadically, and familial cases represent less than 5% (Table 1). Follow-up studies of survivors of sporadic cases showed that all bilateral and about 10% of the unilateral cases are of the hereditary form, the rest being nonhereditary and presumably due to somatic mutation. Thus, of all sporadic cases that are heritable, the ratio of unilateral to bilateral disease is estimated at 0.17∶0.83. Although sporadic cases may contain a few inherited from an unaffected carrier parent, this ratio may be regarded approximately as referred to the general population who developed retinoblastoma as a result of fresh germinal mutation in the gonads of a healthy parent. 2) There are two types in the hereditary form; one is attributable to an autosomal dominant gene with variable penetrance and expressivity, and the other, referred to a minority of cases usually accompanied by developmental retardation, is associated with chromosomal deletion of 13q14. Fifty-seven cases of the deletion type, collected from the literature, included 25 unilateral and 32 bilateral cases (Table 2). Most of them were sporadic in origin, while some were familial, transmitted from a parent carrying an insertional translocation (Stronget al., 1981); in the latter instance, the non-expression of retinoblastoma in the parent was due to the balanced type of translocation and not to any host resistance factor. Therefore, with respect to host factors, the 57 cases are comparable to those sporadic cases that are heritable owing to the mutant gene. Comparison of the distribution of unilateral and bilateral cases between the two groups reveals a significant excess of unilateral disease in the cases associated with 13q-, indicating somewhat reduced carcinogenic potential of deletion of 13q14 as compared with the mutant gene. Since the gene has been assigned to 13q14 by linkage to Esterase D (Sparkeset al., 1982), this finding suggests that the mutant gene is probably not of null type. 3) Analysis of two-generation data from published pedigrees with familial cases of retinoblastoma showed that not only penetrance of the gene in children but also expressivity, as measured by the proportion of bilateral disease or age at diagnosis of the affected children, vary consistently with the degree of expressivity in the carrier parents (Tables 3–5). Maternal effect can be excluded because the variation in expressivity in the affected children did not change with the sex of the parents who transmitted the gene. The possibility of multiple alleles with different penetrance and expressivity can also be ruled out because all the three phenotypes occurred frequently among members of the same family who share the identical mutant gene. The distribution of the three phenotypes among the gene carriers is consistent with a multifactorial model with two thresholds (Fig. 1), but not with the two-mutation model (Knudson, 1971). These findings indicate that inherited host resistance plays a significant role in the malignant transformation of an already initiated or mutated cell. 4) Although the transmission pattern of retinoblastoma in certain kindreds showing affected collateral relatives who were connectedvia unaffected persons may be explained by a chromosomal mechanism as well as by successive selection of host resistance genes, the possibility of delayed mutation cannot be excluded. However, examination of these pedigrees revealed no sign of cytoplasmic inheritance associated with a vertically transmitting virus. 5) In most bilateral cases of retinoblastoma, the disease is already bilateral when the diagnosis was first made in either eye (Table 6), although the tumor in the first eye with the diagnosed disease had usually more progressed in terms of the Reese group than had the tumor in the second eye. Based on the data for sequentially bilateral cases (Table 7), the time required for a growing tumor to progress by one grade in the Reese group was estimated at less than 2 months and probably about 1 month. After removal of the bias due to the fact that the diagnosis for each eye is never independent, the correlation between ages of the patients at diagnosis in the right and left eyes was estimated at 0.82 (p

A new approach is proposed for the evaluation of chromosome variants, which uses a scanning microdensitometer in the determination of the area of a variant. Results are assigned into five classes based on the difference from an average in terms of standard deviation. In the first two papers of the present series, results obtained in C variants of 1,9, and 16 and LBA variants in 12 pairs lacking an established variable site (e.g., nos. 2, 5, 6, etc.) were described. In the present communication, results obtained in pairs with a known Q-variable site are described. When a variable region outside of +/- 1 SD of the average is defined as a variant, 9, 11, 7, 10, and 10 variants are detected in pairs 13, 14, 15, 21, and 22, respectively, from 12 individuals by means of LBA preparations, in addition to Q variants, which can be detected by the standard QFQ technique.

Data on karyotype and maternal age of 1954 cases of Down's syndrome were analyzed to see if the rate of chromosome mutations leading to this abnormality has been enhanced during the last 20 years. Comparison of the data for patients born in 1947–1960 with those in 1961–1975 revealed little change with time in the proportions of cases due to different karyotypes, the overwhelming majority being of 21 trisomy type in both periods. However, there has been a remarkable decline in the mean maternal age from 33.1 years to 29.7 years as well as in the variance from 50.5 to 29.4. While the rate of decline in the variance was almost the same as that for all births occurring in the same periods, the decline in the mean maternal age was much greater for the patients than for all births, suggesting that the rate of nondisjunction might have increased in younger rather than in older mothers. However, when the risk of brearing a child with Down's syndrome for mothers aged 40–44 is taken as unity, no evidence was found for an increase with time in the relative risk for younger mothers. Moreover, results of surveys made in 1960 and thereafter in different parts of Japan indicate that the crude incidence rate of Down's syndrome at birth has been around 0.10%, giving no indication of an upward trend. These findings are discussed with reference to the serious environmental pollution, including possible genetic hazards, with which Japan has been faced since the 1960s.

Chromosome variants were evaluated on the basis of their DNA-replication pattern (LBA). The size of late-replicating centromeric heterochromatin of chromosomes 2, 5, 6, 7, 8, 10, 11, 12, 17, 18, 19, and 20, i.e., pairs without Q or C (qh) variants, was measured by means of a microdensitometer. The results were expressed in area, related to that of a euchromatic segment of a given chromosome, and were assigned into five classes based on the difference in standard deviation from an average relative size. LBA variants in each of 12 pairs were found in 29%-42% of the chromosomes.

Morphological data from 3,614 well-preserved human embryos derived from artificial termination of pregnancy were used to determine whether ectopic implantation or enlarged myomas could enhance the prevalence of localized malformations of the embryo. The frequency of malformed embryos was 11.6% among 43 recovered from ectopic pregnancies, 6.2% among 97 from myomatous pregnancies, and 3.3% among 3,474 from normally implanted pregnancies not complicated by myomas. Unilateral amelia in the ectopic cases and caudal dysplasia in the myoma cases were significantly increased. Both malformations were quite unusual in the control group. It is argued that spatial restrictions could be a teratogenic agent in human embryos. Mothers of both ectopic and myoma cases were on average much older than mothers of the control specimens, and those with myomas had a higher frequency of previous pregnancy wastages. Ectopic pregnancy was found to be associated with lowered parity, previous ectopic pregnancy, and maternal smoking and drinking. These associations are discussed and interpreted in relation to etiology of each of the two conditions.

Analyses of family data published in Japan concerning 29 kindred with 2 or more cases of retinoblastoma revealed that, in the children who received the gene from a carrier parent, both penetrance and expressivity increase with increasing degree of expressivity in the parent. The estimate for the average degree of penetrance varies from 0.7 to 0.9, depending upon the method employed. The value will be increased with increasing number of survivors of hereditary retinoblastoma. Data were presented to indicate that in the hereditary form of retinoblastoma onset tends to be later in unilateral than in bilateral affection, though in the non-hereditary form onset will be further delayed. No effect of birth order was detected. It was suggested that suppressor genes at other loci play a significant role in manifestation of the major dominant gene for retinoblastoma. Implications of these findings were discussed.

A molecular species of sialic acid was isolated in a free form from cerumen of the wet type, but not of the dry type, by an ion-exchange column chromatography and preparative high-performance liquid chromatography. Structural analysis of this sialic acid was performed by gas-liquid chromatography/mass spectrometry with chemical ionization (CI) and electron ionization (EI). In the CI mass spectra, the protonated molecular ion of the trimethylsilyl derivative was observed at m/z 580. and that of the methyl ester-trimethylsilyl derivative was at m/z 522. In the EI mass spectrum, the methyl ester-trimethylsilyl derivative gave characteristic ions at m/z 506, 462, 418, 416, 328, 316, 238, 228, 205, 186, and 173. This mass spectrum was identical with that of 2,7-anhydro-N-acetylneuraminic acid, which was reported by Lifely and Cottee (Carbohydr. Res. 107, 187-197, 1982) as the mass spectrum of a by-product prepared from N-acetylneuraminic acid by methanolysis. These results indicate that the compound in the wet cerumen is 2,7-anhydro-N-acetylneuraminic acid. Since this sialic acid species could not be detected in cerumens of the dry type, its formation in the wet type may be controlled by an autosomal dominant gene.

Data for 27 cases of retinoblastoma that developed in patients with 13q-were collected from the literature and analyzed. The distribution of unilateral and bilateral cases of retinoblastoma differed significantly from the expectation that the degree of expressivity does not differ between the retinoblastoma gene and deletion of 13q. The excess of unilateral cases among the patients with 13q-, which could not be accounted for by ascertainment bias, was attributed to somewhat lowered carcinogenic potential of deletion of 13q14 as compared with the retinoblastoma gene. It was argued that the retinoblastoma gene is probably not located on 13q, and perhaps 20% or more of the individuals with a deletion of 13q14 would not develop retinoblastoma. The normal allele at the retinoblastoma locus, the haplicon in the segment of 13q14, and the suppressor genes as defined by the host resistance model, may be all concerned, in their function additively and without dominance, with normal differentiation of the embryonic retinal cells.

One hundred and twenty-nine embryos with polydactyly were found among 36,380 human conceptuses obtained through induced abortion during the period from 1962 to 1974, giving an overall incidence of 0.35%; this must be taken as an underestimate because the denominator includes a large number of damaged embryos. A significant increase was noted in 1971 and 1972, and some exogenous factors, most probably ascertainment bias, were suspected to be responsible. About three quarters of the present cases had preaxial polydactyly in contrast to the predominance of postaxial type in Negro and Caucasian populations. The ratio unilateral to bilateral affection was approximately 2∶1.

With increasing success in medical treatment for many common disorders with complex inheritance, it is expected that the incidence of patients requiring treatment will gradually rise from generation to generation. While this is theoretically true for a long range of future, the pattern of possible increase in the frequency of a disorder resulting from complete cure has been examined with special reference to the next few generations. It was shown that the maximum possible increment will take place in the first generation, with additional increment rapidly diminishing with subsequent generations. For congenital heart disease, for example, the restoration of full fertility to all of the patients would incur an increase in the frequency at birth of the disease by about 10% after 5 to 10 generations.

SummaryIn order to test the possible effect of maternal inbreeding upon non-disjunction, ancestors of 104 children with Down's syndrome born to younger mothers were identified on the basis of official records. Among them, 6 had related parents, while 10 of the fathers and 12 of the mothers had also such parents. The relationship comprises first-cousins, first-cousins once removed, and second cousins. It was concluded that recessive genes are unlikely to be involved in the aetiology of non-disjunction of the 21st autosome, at least in younger mothers.

There is evidence from diverse sources to indicate that ABO blood groups are subject to natural selection at various developmental stages. This paper reviews the findings on associations between ABO blood groups and diseases and on ABO incompatibility as a major cause of prenatal and perinatal selection; provides evidence for prezygotic selection in ABO blood groups; and discusses selection for heterozygotes in ABO blood groups and in MN blood groups. From the evidence examined, the author concludes that the principal mechanism which maintains the ABO blood groups acts during prenatal and perinatal periods in addition to prezygotic stages; that the effective result of associations between ABO blood groups and adult diseases seems to be small; and that the MN polymorphism is probably maintained by the advantage of heterozygotes, apparently limited to MN mothers.

It is well known that there is a very marked racial difference in the nature of cerumen or ear-wax. Caucasian and Negro peoples have generally brown, sticky and wet cerumen, while this type of cerumen is less common among Japanese where the majority has gray, brittle and dry cerumen. The distinction between these two types, wet and dry, is usually sharp and manifested soon after birth, so that diagnosis can easily be made by inspection of external ear canal if no pathologic complication is involved. It has been further noticed, even by layman, that there is an association between wet cerumen and axillary odor, the presence of which is deemed to be pathologic in Japan and is often considered to require medical treatment. Probably for this reason, there are a considerable number of Japanese literature which deal with the variation in cerumen, mostly in connection with axillary odor. Based on his own observations besides the results collected by other authors, the present author proposes to review the subject on the apparent polymorphism in ear-wax types from the view-point of genetics with special ref er ence to the anthropological significance of the finding on this trait.

Summary Various aspects of the dimorphism in human cerumen have been described on the basis of the data collected mostly by Japanese investigators. It was shown that the ear-wax types are apparently determined by a simple genetic mechanism, and can be used as a polymorphic system for both genetics and anthropology. Frequencies of the alleles for the ear-wax types have been estimated for different ethnic groups. It was pointed out that the marked difference in the distribution of ear-wax types among the major races is probably a consequence of natural selection having acted in a long course of evolution, and there may be some differences in the selective value between individuals with wet cerumen and those with dry. In view of the apparent association with axillary odour, it was suggested that the alleles for ear-wax types may control certain metabolic processes, through which individuals with different types of oerumen may display different reactions to environmental changes. The need for future studies along this line is stressed.

Red cell ESD activities have been determined in 78 individuals with ESD-1 phenotype, in 94 with ESD2-1 and in 28 with ESD-2 phenotype. The mean activities of these three groups were 276.7, 216.6 and 171.5 expressed as 10-7 M methyl-umbelliferone produced/h/g Hb, respectively. The activities associated with a single ESD * 1 gene are estimated to be 60% higher than ESD * 2.

Data for 26 cases of nonradiogenic osteosarcomas that developed in patients with retinoblastoma and in their close relatives were collected from the literature and analyzed. The distribution of unaffected and unilaterally and bilaterally affected cases of retinoblastoma agreed with the expectation that risk to osteosarcoma did not differ by the phenotypes of the gene carriers. This finding indicated not only that increased risk to second primary tumors was a pleiotropic effect of the retinoblastoma gene but also that host resistance to the gene effects was tissue-specific. The risk of gene carriers to nonradiogenic osteosarcoma was estimated at 1.2%, and the relative risk was estimated at 230. Liability of the gene carriers to osteosarcoma could be regarded as a multifactorial threshold character, and heritability was estimated at 71%. Results of analyses of ages at diagnosis suggested that nonradiogenic osteosarcoma developed slightly earlier in carriers of the retinoblastoma gene than in children not carrying the gene. These findings are discussed in the light of the host resistance model in which genes involved in tissue resistance must be suppressors to an error in differentiation (presumably a crucial event in malignant transformation) and radiogenic osteosarcoma in gene carriers is ascribed to the accumulation of induced mutations in that suppressor system.

Carlson and Desnick's [1979] proposals of multiple allelism and mutational mosaicism to account for the observed variability in penetrance and expressivity in retinoblastoma do not seem to give a clear-cut explanation for the facts that 1) a bilaterally affected patient could have an unaffected carrier child who in turn would produce bilaterally affected offspring, and 2) carriers of the retinoblastoma gene are prone to nonradiogenic osteosarcoma irrespective of their phenotypes. However, these observations are readily explained by assuming that inherited host resistance plays an important role in the manifestation of a single major gene for retinoblastoma. Based on their model, Carlson and Desnick offer recurrence risk figures for genetic counseling, but their figures concerning the transmissible types of retinoblastoma do not seem more accurate, as assumed by them, than those previously estimated from the family data. We present evidence that host resistance may be regarded as a multifactorial threshold character with high heritability, and we propose that tissue-specific genes, which we presume to be involved in normal differentiation of retinoblasts and osteoblasts, play a crucial role in malignant transformation. Two lines of suggestions are made for future studies: one, to search for polymorphic genes associated with retinoblastoma, and the other, to see whether the principle of synchronous appearance of multiple tumors in persons at high risk applies to other cancers as well.

Causal relations between maternal genital bleeding, supportive hormone therapy and external malformations of the embryos were investigated with special reference to the critical period of organogenesis. This was done using morphological and obstetrical data obtained by Nishimura and his associates from 667 undamaged embryos derived from induced abortions whose mothers had genital bleeding in early pregnancy. In addition, data from 90 embryos with polydactyly and 38 with limb reductions in the Nishimura collection were used for case history studies. Evidence was presented to demonstrate that, for major malformations such as CNS anomalies, cleft lip, polydactyly and limb reductions, maternal genital bleeding was not a cause but a consequence of the conception of an abnormal embryo. No indication was revealed that exogenous female hormones currently used in Japan for preventing miscarraiges could produce major malformations recognizable at the embryonic stage, including limb reductions, nor salvage the severely malformed embryos. This does not however mean to exclude the possible relationship of progestogens/estrogens intake during early pregnancy with an increased incidence at birth of certain internal and/or external malformations. It was suggested that most, if not all, of the minor anomalies observed at certain embryonic stages are kinds of normal variants without any functional impairment of embryonic development.This paper explored, using induced abortuses both alive and dead in utero in a Nishimura consecutive series, the causal relationships among maternal genital bleeding, supportive hormone therapy, and congenital anomalies, with special reference to the critical organogenesis period. Data from 667 undamaged embryos from induced abortions whose mothers had had early genital bleeding and from 90 embryos with polydactyly and 38 with limb reductions were studied. The following conclusions were drawn from examining this series and the damaged abortuses' case histories: 1) Major malformations such as central nervous system anomalies, cleft lip, polydactyle, and limb reductions were not caused by maternal genital bleeding but that the bleeding was a consequence of conceiving an abnormal embryo. 2) No indication was found that exogenous hormones, especially those used in Japan for preventing miscarriages, induced major malformations at the embryonic stage, including limb reductions, nor that they could salvage embryos already severely malformed. 3) Most of the minor anomalies seen at embryonic stages are suggested to be normal variants without any functional impairment of embryonic development. 4) Though this retrospective investigation could not link estrogens used to prevent miscarriages with major malformations, the authors do not rule out the possibility that progestogen/estrogen intake during early pregnancy may be related to increased incidence at birth of internal anomalies such as cardiac anomalies and esophageal atresia, or external malformations manifested at the fetal or neonatal stage.

One hundred fifty embryos with holoprosencephaly were found among the total of 36,380 conceptuses obtained through induced abortion in the period from 1962 to 1974, giving an overall incidence of 0.4 percent. The occurrence was period from 1962 to 1974, giving an overall incidence of 0.4 percent. The occurrence was largely at random through time, and no "epidemic" was noted in particular years or months, but there appeared more cases derived from conceptions in winter than in summer months. The mean maternal age did not differ significantly from that of the general embryonic population, indicating that, although none of our cases were karyotyped, chromosome aberrations such as trisomies 13 and 18 that are closely associated with maternal age may not constitute a major part as causes of holoprosencephaly in human embryos. Materal age did not differ by the presence or absence of associated external anomalies. No association was found with paternal age, parental consanguinity nor with maternal medical history, including irregularity of menstrual cycles, and smoking and drinking habits. There was an indication that the mothers were prone to have repeated miscarriages, supporting the view that some kind of maternal predisposition is responsible for the causation of holoprosencephaly. Argument was made that, apart from various chromosome aberrations well documented as causes of this malformation, polygenic mechanism probably accounts for the majority of the cases with normal karyotype.

We attempted to determine whether all cases of AWTA (anirida-Wilms tumor association) or any of the following groups of patients show 11p deletion: cases of Wilms tumor with congenital abnormalities other than aniridia, those without any congenital abnormalities, tumor itself in cases of Wilms tumor without constitutional 11p deletion and cases of aniridia or hemihypertrophy without Wilms tumor. We studied a total of 29 index patients including five cases of AWTA, four cases of Wilms tumor with various congenital abnormalities, 16 cases of Wilms tumor without other abnormalities, three cases of aniridia in one of which Wilms tumor developed later and a case of hemihypertrophy. In all five cases of AWTA and in a case of aniridia who later developed Wilms tumor, 11p deletion involving the p13 band was detected. The mother of the latter also showed an identical 11p deletion. The common segment of deletion was the middle part of the p13. Two possible hypotheses on the mechanism through which Wilms tumor might develop were evaluated, based on the distribution of break points. All other cases, including five with tumor culture, showed a normal karyotype.

The analyses of published data for Wilms' tumor show the following results: 1) The mode of inheritance of familial cases, which constitute less than 1% of all Wilms' tumors, is autosomal dominant with variable penetrance and expressivity. The familial pattern reveals no evidence that this tumor is associated with a vertically transmitting tumor virus. 2) There is a significant correlation between the first-degree relatives affected with Wilms' tumor in regard to expressivity as measured by the proportion of bilaterally affected or by the age of the patients at diagnosis. In particular, expressivity in the children affected varies consistently with expressivity in the carrier parents, indicating that inherited host resistance plays an important role in the manifestation of the major gene. The average risk of developing Wilms' tumor for offspring of unaffected carriers, who may be regarded as inherently resistant to tumor formation, is estimated at about 0.30. Although the data are limited, the risk for offspring of survivors of hereditary unilateral or bilateral cases may be as high as 0.40 or more. The data from seven pairs of monozygotic twins, including the discordant one associated with del(11 p) and aniridia, are consistent with the host resistance model. 3) In sporadic cases, the proportion of bilateral involvement is about 3%, in contrast with 20% in the familial cases. For bilateral cases, however, the distribution of ages of the patients at diagnosis is virtually the same whether sporadic or familial, and in two-thirds of the cases the disease is already bilateral when the first tumor is noted. These findings suggest that bilateral cases are probably always hereditary. Ages of the patients at diagnosis of sporadic unilateral cases tend to be much higher than in the patients with familial unilateral cases, indicating that most sporadic unilateral cases are nonhereditary. The overall risk for survivors of sporadic unilateral Wilms' tumor to transmit the disease to children may be as low 2∼4%. 4) Thus, a number of parallels are found between Wilms' tumor and retinoblastoma. The rate of germinal mutation per locus per generation is estimated at 3∼8×10-6 for Wilms' tumor, which does not seem to differ from 8∼14×10-6 for retinoblastoma. However, the incidence of nonhereditary Wilms' tumor per newborn is 7∼11×10-5, which is about twice that (3∼5×10-5) of nonhereditary retinoblastoma. This difference is possibly due to the greater number of cell divisions, resulting in an increased frequency of somatic mutations, of the embryonic cells involved in morphogenesis of the kidney than in the case of retinal development. 5) With increasing success in therapeutic management, it is anticipated that more and more cases of Wilms' tumor inherited from the survivors are emerging. However, even if complete cure of the disease and full recovery of fertility could be achieved for all the patients, about 10 generations may be required to double the present incidence of Wilms' tumor in the population, provided that other conditions remained the same. 6) Some guidelines for genetic counseling in families affected by Wilms' tumor are given.

A huge ‘satellite’ (22s+) was observed in a mentally retarded and malformed girl. It consisted of constitutive heterochromatin based on the C-band technique (CBG). In neither of the parents was the variant observed. Genetic marker studies revealed that the father was, indeed, the father of the proposita (probability, 99.8%). It is emphasized that caution must be taken in the use of chromosome variants for paternity dispute cases.

Analysis of two-generation data from published pedigrees with familial cases of retinoblastoma showed that, while the proportion of bilateral cases among affected children varied consistently with expressivity of the parents who transmitted the gene, the proportion did not differ with the sex of the parents. These findings support the view that inherited host resistance to the development of hereditary retinoblastoma is determined by modifying genes, and not influenced by transplacental maternal factors, and that the resistance is mainly directed to the induction (promotion) phase. There was a significant deficit of bilaterally affected mothers relative to bilaterally affected fathers, which could not be accounted for by ascertainment bias or by differential survival or fertility between sexes. It is suggested that female patients who survived bilateral retinoblastoma had much less chance of marriage as compared to male survivors. Examination of pedigrees with multiple cases of retinoblastoma, transmitted via unaffected carriers, revealed no sign of cytoplasmic inheritance associated with a vertically transmitting virus.

A method is described (LBA method) which used DNA replication pattern in the detection of chromosome variants in man. In Part I, results on chromosomes with known sites of Q-variants, i.e., 5 pairs of acrocentrics, as well as 3 and 4, were presented. Fourty-one variants were detected in a total of 40 acrocentrics. Twenty-eight of them were detected only by the LBA technique; 11 of them being in short arms and 17 of them in centromeres. Seven variants, including 4 of those in satellites, were detected only in QFQ-stained metaphases. Six short-arm variants were observed by both methods. It appears that a sequential QFQ-LBA technique is very useful in the detection of variants in D- or G-group chromosomes.

Using data from 3,474 human embryos derived form induced abortion, various maternal factors ascertained prior to morphologic examination of the specimens were analyzed in association with the frequency of malformed embryos. The overall frequency of malformed embryos was 3.3%; polydactyly, holoprosencephaly, exencephaly with or without myeloschisis, and cleft lip accounted for 88% of the total malformations. Mothers of low parity showed an increased frequency of malformed embryos. There were nine embryos form mothers with thyroid disorders, of which one was affected with exencephaly and myeloschisis and another with holoprosencephaly and polydactyly; this finding may support the relation of maternal hyperthyroidism with anencephaly suggested by previoius case reports. The frequency of malformed embryos from mothers who drank alcohol was significantly lower than from those who did not. Several possibilities were suggested to interpret this unexpected finding. No association was verified with maternal age, irregular menstrual cycles, irradiation, contraceptive methods, maternal occupation, and a smoking habit, or with parental consanguinity. The limitation of this study is discussed in relation to both negative and positive results.

Mitochondrial DNA (mtDNA) from 116 Japanese was analyzed with nine restriction enzymes that recognize a four or five base pair sequence. The sizes of the mtDNA fragments produced by digestion by each enzyme were compared after gel electrophoresis. Double digestion experiments indicated that, in the coding region from URF2 (unidentified reading frame) to tRNAAsn (bp 5274-5691), there is an insertion of about 60 base pairs (bp) compared with the published mtDNA sequence, which is common to all individuals in the present sample. A total of 95 different morphs were detected with the nine enzymes, 60 of which have not been documented previously. Based on a comparison of the cleavage maps of all individuals, 62 different combinations of restriction types were observed. By pairwise comparison of each restriction type, the average number of nucleotide substitutions per nucleotide site (delta) was estimated to be 0.0026. Phylogenetic analysis of the present data indicates that at least two distinct lineages exist in the Japanese population.

In Part I of this communication, a technique (LBA) was described which used DNA replication in the evaluation of chromosome variants in man. It was shown that the method was very useful in the detection of variants in D- and G-group chromosomes. Results on pars 3 and 4 were also presented. In Part II, the rest of chromosomes were examined. In the evaluation of qh variants in 1, 9 and 16, the LBA technique proved itself to be a very effective implement. It was practically free of technical variables coherent with C-band technique and, therefore, it was possible to use the size of an euchromatic segment of a chromosome as a reference standard. LBA variants were observed in about 50% of the members of the remaining 12 pairs of chromosomes, i.e., 2, 5, 6, 7, 8, 10, 11, 12, 17, 18, 19, and 20.

The mitochondrial DNA (mtDNA) from 120 Japanese was analysed with 15 restriction enzymes that recognize six base pairs, of which 11 enzymes showed at least one atypical cleavage pattern. Digestion patterns with Hinc II and Hae II were highly polymorphic. The observed restriction enzyme morphs were classified into 22 types of distinct cleavage patterns. By pairwise comparison of each restriction type, the average number of nucleotide substitutions per nucleotide site (delta) was estimated at 0.00417, which agreed with the values obtained from other human populations in previous studies. There were 11 site gains, of which seven were transitions and four were transversions. Phylogenetic analysis of the present data suggested that the Japanese population conceals a considerably high degree of mtDNA diversity.

Paternal age distribution for 1279 cases of Down's syndrome born in 1952--1968 was compared with the corresponding distribution for the general population, corrected for the maternal age as well as for the year of birth of the patients. Although there was no difference in the mean paternal age, the two distributions differed significantly, largely due to the excess of fathers aged 55 years and over and to the deficit of those aged 40--44 years in the patients born to mothers aged 30 years and over. The overall pattern of the relative incidence of Down's syndrome with advancing paternal age, with maternal age controlled, seems consistent with the hypothesis proposed by Stene et al. (1977). It increased from 0.8 for fathers aged 20--24 years slowly up to 1.2 for those aged 45--49 years, though with an intermediate drop to 0.8 at the age of 40--44 years, and then sharply to 2.4 for those aged 55 years and over. This rising pattern of the relative incidence with paternal age was essentially the same for the patients born in 1952--1960 and for those born in 1961--1968, although the slope was less steep in the latter than in the former group.

Age-specific incidence data from 244 bilateral and 31 familial unlateral cases of retinoblastoma were analyzed. In the bilateral cases, a high intraclass correlation between ages of the patients at diagnosis in the right and left eyes was noted, even after removal of the bias due to the fact that the diagnosis for each eye is never independent. This finding indicated strongly that the age at onset was largely determined by host factors common to both eyes. The mean ages of the patients at diagnosis of the inherited cases among the bilateral or unilateral group varied consistently with parental phenotype. This finding implied that inherited host resistance played an important role in the latency period for the gene carriers. The fraction of cases not yet diagnosed fit well to a negative exponential distribution, and in the less susceptible group with hereditary unilateral cases, tumor formation may have been suppressed about twice that in the most susceptible group with bilateral cases. It was argued that the presumed second hit initiating tumor formation in the gene carriers was not a mutational process, but probably an error in the process of differentiation that could be suppressed completely in the most resistant group who could remain unaffected, but that could not be suppressed completely in the less resistant group.

A statistical method was devised to test whether prezygotic selection was operating in ABO blood groups, and it was demonstrated, with data from Japanese families, that heterozygous AO and BO fathers transmitted more than 50 percent O-bearing sperm (approximately 55 percent) to their children. Neither sperm incompatibility nor reproductive compensation could account for the results.

DNA sequence analysis and electrophoresis in denaturing gel revealed that a 60 base pair insertion which had been previously postulated on the basis of native polyacrylamide gel electrophoresis of mitochondrial DNA from Japanese (Horai and Matsunaga 1986) did not exist at all. Unusual behavior of certain restriction fragments in native polyacrylamide gels apparently resulted in what appeared to be an insertion. Further study revealed that this behavior is most likely due to secondary structures of the fragments. The results of the present study suggest that adequate care should be taken when assessing molecular weights of restriction fragments by native polyacrylamide gel electrophoresis.