Biomedical subjects
P H Kay
Publications and source records attributed to P H Kay.
A novel MspI PCR-RFLP in the human cytosine 5-methyltransferase gene: lack of relevance for malignant lymphoproliferative disease and breast cancer.
Two alternate allelic forms of human cytosine 5-methyltransferase, 5-MT I and 5-MT II, which differ by the absence or presence of an intronic MspI recognition sequence, have been recognised. The polymorphic region was localised using a series of subprobes prepared upon MspI digestion of the 2.5-kb cDNA probe (hmt-2.5). A PCR-based method was then developed for rapid 5-MT genotyping. The gene and phenotype frequencies of 5-MT I and 5-MT II were not significantly different in genomic DNA samples from a series of non-Hodgkin's lymphomas and breast cancer cases compared with DNA from normal subjects. Allelism of 5-MT allows new approaches to the assessment of variation in gene copy number of 5-MT in different types of neoplasia.
Allele-specific variation in the gene copy number of human cytosine 5-methyltransferase.
Previously, we have identified two alternate allelic forms of cytosine 5-methyltransferase, 5-MT I and 5-MT II, specified by polymorphic fragments of 1.5 and 1.1 kb, respectively. In the presence study, a 0.8-kb genomic probe was prepared which was confirmed to be included within the polymorphic fragments. The 0. 8-kb probe hybridised with greater intensity to the 1.1-kb fragment than the 1.5-kb fragment. Densitometric analysis indicated that there is 1 copy of 5-MT associated with 5-MT I, whereas there may be 1-4 copies of the gene associated with the 5-MT II allele. Segregation studies demonstrated that the multiple copies of 5-MT II are inherited in a Mendelian fashion. These results allow novel approaches to investigating the underlying mechanisms of cytosine methylation and gene duplication.
TaqI polymorphism of the epidermal growth factor receptor gene in Caucasoids and Japanese.
Genetic polymorphism of the epidermal growth factor (EGF) receptor gene following Taq I digestion was compared between samples of genomic DNA from glioma-derived cell lines and Caucasoid and Japanese subjects. The same three allelic forms of the EGF receptor gene, marked by variant fragments of approximately 12.8, 11.6 and 10.8 kb in size were common to both ethnic groups and the 12.8- and 11.6-kb fragments were found in the glioma-derived cell line DNA. A further variant fragment of approximately 13.8 kb in size has been shown to be thus far restricted to the Japanese. These data suggest that most allelic forms of the EGF receptor gene recognized by Taq I restriction fragment length polymorphism have a long evolutionary history and probably do not predispose to development of malignant glioma.
Hypomethylation of cytosine 5-methyltransferase in human neoplasms.
Cytosine methylation is an epigenetic modification of DNA involved in control of gene expression. Neoplastic cells exhibit various alterations both in DNA methylation and activity of the enzyme responsible for this modification, 5-methyltransferase (5-MeTase). As there is little requirement for 5-methyltransferase expression in normal cells except during mitosis, we argued that the gene would be hypermethylated in normal cells. Southern analysis revealed almost complete methylation of the gene in genomic DNA from the peripheral blood leukocytes of healthy subjects and a primary fibroblast derived cell line. In contrast, in DNA from a range of tumour tissues and tumour derived cell lines, 5-MeTase exhibited marked hypomethylation. The results of this study indicate that dysregulation of the DNA methylating machinery, especially with respect to the methylation status of 5-MeTase, is a feature of a wide range of neoplasms.
Bf S gene frequency in Nigerians with HIV infection.
Complement immunogenetic susceptibility to human immunodeficiency virus (HIV) infection was examined in 40 Nigerians with serological and/or clinical evidence of the infection. A mild increase in C4A null alleles (C4AQO) frequency was observed in the patient group compared to a group of healthy subjects (25 per cent vs 17 per cent) but overall the HIV infected and the reference groups did not differ significantly in the frequency of alleles of C4A or C4B. In contrast, properdin factor B (Bf) S gene frequency was significantly higher in the patients with HIV infection (p less than 0.025). There was a concomitant decrease in Bf F allele and gene frequencies (p less than 0.01, and p less than 0.05), respectively. Furthermore, blank Bf allotypes due to excessive complement consumption were detected in two asymptomatic patients. These findings suggest that Major Histocompatibility Complex (MHC) located complement genes may be important HIV infection. In particular Bf S gene or even C4AQO alleles may be permissive or influence outcome of infection with HIV.
Ten-year survival following aortic valve replacement: A multivariate analysis of coronary bypass as a risk factor.
The additional risk of coronary bypass surgery was analysed in 664 patients over 40 years of age undergoing aortic valve replacement between 1969 and 1981. Four hundred sixty-seven patients underwent aortic valve replacement alone, while 197 patients with coronary artery disease underwent combined aortic valve replacement and coronary bypass surgery. There were no significant differences in the preoperative hemodynamic characteristics of the two groups of patients. There were 41 (9%) operative deaths following aortic valve replacement alone and 20 (10%) following aortic valve replacement with coronary bypass surgery. Since 1976, operative mortality has fallen to 5% and perioperative myocardial infarction to 2% following the combined procedure. Ten-year actuarial survival (standard error) was 56 (3%) following aortic valve replacement and 49 (6%) following aortic valve replacement and coronary bypass surgery. A multivariate analysis including both groups of patients revealed that age, functional class and year of operation significantly affected ten-year survival (p less than 0.05). The same analysis showed that coronary artery disease requiring coronary bypass surgery also decreased ten year survival in patients undergoing aortic valve replacement (p = 0.06).
A current appraisal of the Waterston shunt.
Recent reports have suggested that the classical Blalock-Taussig shunt and its prosthetic modification have a low patency rate in neonates. This prompted a review of 58 neonates undergoing a Waterston shunt for cyanotic congenital heart disease. The 4 mm anastomosis was constructed using a calibrated clamp. There were 20 (35%) hospital deaths, of which 3 were directly related to shunt failure. Operative mortality was associated with increasing complexity of the cardiac lesion, emergency operation, weight less than 3 kg and age less than 48 hours. There were 5 late deaths resulting in an actuarial survival of 52% at ten years. There was one case of late occlusion corresponding to a patency of 92% at 2 years. Anastomotic kinking with preferential flow to one lung occurred in 2 (7%) of 28 patients undergoing repeat angiography at 3 and 5 years respectively. Nineteen patients underwent corrective surgery a mean of 27 months after shunt construction. Six (32%) required an angioplastic repair of the pulmonary artery. Six (16%) of the 38 operative survivors required diuretic therapy for excessive pulmonary flow. The high patency ensures that the Waterston shunt continues to play an important role in the management of low weight neonates who require an emergency systemic-pulmonary shunt. The use of a calibrated clamp reduces the incidence of excessive pulmonary flow. Early corrective surgery and shunt dismantling may reduce the necessity for angioplastic repair.
Aortic valvotomy in the treatment of congenital aortic stenosis.
Forty patients aged 1.5-23 (mean 11) years underwent open aortic valvotomy for congenital aortic stenosis. Mean preoperative peak aortic gradient for the group was 88 (SD +/- 24) mmHg. Follow-up data was available on all patients for a total of 269 patient-years (means 6.6 year). There was no early mortality, and only one (2.5%) late death. During the follow-up period 3 patients (7.5%) required aortic valve replacement. Actuarial survival curves predict a 7 year survival of 95.7 (SE +/- 8.4%) for all patients, and a freedom from reoperation rate at 6 years of 8.7 (SE +/- 9.3)%. Symptomatic results in the survivors were excellent (97.5% NYHA Class I). Data from this series compares favourably with the literature which is reviewed for a total of 1136 patients undergoing open aortic valvotomy.
The results of surgery for active endocarditis of the native aortic valve.
Between 1968 and 1981, 40 patients with active endocarditis of the native aortic valve were treated by aortic valve replacement (A.V.R.). There were 8 postoperative deaths (hospital mortality 20%). This included 5 patients who had developed cardiogenic shock prior to surgery. Antibiotic treatment for less than one week and positive cultures on the excised valve had poor prognostic implications but tended to be associated with irreversible haemodynamic failure. Twenty five patients underwent A.V.R. following the onset of severe pulmonary oedema. The hospital mortality in this group was 28% and the 5 year actuarial survival 56% (+/- 11%). Fifteen patients who had developed premature closure of the mitral valve (P.C.M.V.) on M-mode echocardiography but who had no overt signs of cardiac failure underwent A.V.R. with a single death (7% hospital mortality). The 5 year actuarial survival in this group was 87% (+/- 9%). P.C.M.V. is a useful prognostic sign identifying those patients with endocarditis on the native aortic valve likely to benefit from early surgery.
The management of prosthetic valve endocarditis.
During the period 1970 to 1980 2,945 patients underwent valve replacement at the Brompton Hospital. Thirty-one (1%) patients subsequently developed prosthetic valve endocarditis (P.V.E.). Twenty-one (0.7%) patients developed endocarditis within two months of valve replacement. Ten patients were treated medically, with seven deaths, whilst four of the eleven surgical cases died. There were ten cases of late P.V.E. occurring between two months and eight years after initial valve replacement (0.12% per annum). Eight patients were treated surgically, with two deaths, whilst both medical cases died. Thus 67% patients were effectively treated by immediate valve replacement, whilst only 25% medical cases survived. Actuarial survival curves predict a 41% five-year survival following surgery for P.V.E. We stress the importance of early diagnosis of this condition and recommend immediate valve replacement in all patients with P.V.E. developing signs of haemodynamic failure.
Aortic valve replacement in the over seventy age group.
Between 1970 and 1979 50 patients over the age of seventy years underwent aortic valve replacement. The overall hospital mortality was 14% but has been reduced to 5% following the introduction of cold cardioplegia for myocardial protection. Actuarial survival curves predict a 60% five year survival. Whilst all patients considered for surgery were severely limited in terms of effort tolerance, only 5% of the survivors remain in New York Heart Association functional class III or IV. In conclusion aortic valve replacement may be safely performed in septagenarians and results in increased longevity and quality of life.
Hypermethylation of the Myf-3 gene in human colorectal cancer.
Abnormal methylation of DNA is one of the earliest detectable changes in human tumors. We investigated hypermethylation of the Myf-3 gene, the human homolog of the mouse myogenic gene Myo-D1, in colorectal adenomas and adenocarcinomas. Histologically normal gut mucosa from colorectal cancer patients showed moderately higher levels of methylation than in other normal tissues. Varying degrees of Myf-3 hypermethylation were found in all five adenomas and eight adenocarcinomas examined. Carcinomas arising from within adenomas generally showed a decrease in both the heterogeneity and intensity of Myf-3 methylation. Regional hypermethylation of this gene therefore appears to be an early event in human colorectal neoplasia.
Evidence that DNA methylation imbalance is not involved in the development of malignant mesothelioma.
Methylation dysregulation has been a consistent finding in various malignancies, particularly those where the pathogenetic mechanisms are unclear. In order to test the hypothesis that methylation imbalance may not be a feature of cancers where the aetiologic agent or process is known, we studied the methylation status of the myogenic genes Myf-3 and Myf-4 by Southern blotting in malignant mesothelioma, a cancer strongly associated with asbestos exposure. DNA samples obtained from controls and mesothelioma patients did not exhibit hypermethylation of Myf-3 and hypomethylation of Myf-4, as noted in malignant lymphomas. The methylation status of Myf-3 and Myf-4 in malignant mesothelioma was similar to that of non-malignant cells indicating that dysregulation of the DNA methylating machinery may not be involved in mesothelioma development. The present findings do not support the view that methylation imbalance is a consequence of neoplastic transformation, but indicate that it may be one of the early molecular events involved in the genesis of some cancers.
Hypermethylation of HpaII recognition sequences within the 5' coding region of the estrogen receptor gene is not associated with estrogen receptor negativity in primary breast tumours.
Estrogen receptor negative tumours are unable to respond to antiestrogen therapy. The underlying molecular mechanisms of estrogen receptor negativity are poorly understood. Cytosine mechylation is one of the mechanisms of gene control and previous studies, particularly on breast tumour derived cell lines have suggested that hypermethylation of HpaII recognition sequences within the 5' coding region of estrogen receptor may be responsible for gene inactivity. This study has examined the methylation status of HpaII recognition sequences in estrogen receptor positive and negative breast tumours taking into account a polymorphic HpaII site in the 5' coding region of the estrogen receptor gene. It is concluded that hypermethylation of one or more of the 4 or 5 HpaII recognition sequences in the 5' coding region of estrogen receptor is not associated with ER negativity in primary breast tumours.