A hypertensive father, but not hypertensive mother, determines blood pressure in normotensive male offspring through body mass index

Journal of Human Hypertension (1998) 12, 441 445 1998 Stockton Press. All rights reserved 0950-9240/98 $12.00 http://www.stockton-press.co.uk/jhh ORIGINAL ARTICLE A hypertensive father, but not hypertensive mother, determines blood pressure in normotensive male offspring through body mass index Y Uehara 1,2, WS Shin 1, T Watanabe 1, T Osanai 1, M Miyazaki 1, H Kanase 1, R Taguchi 1, K Sugano 1 and T Toyo-Oka 1 1 Health Service Center and 2 Second Department of Medicine, University of Tokyo, Tokyo, Japan This investigation was to assess the role of genetic loading of hypertensive parents in the determination of blood pressure (BP) in their normotensive offspring. The medical check-up data from 7279 Japanese university students aged 19.22 ± 0.01 years were analysed of which 641 students had only one hypertensive parent with or without hypertensive grandparents, and from this number 609 cases were available for the present analysis. The BP in the students having only one hypertensive parent were in the normotensive range, but was significantly higher than in those students without hypertensive relatives. Analyses of the data from the students having only one hypertensive parent revealed that systolic BP (SBP) and body mass index (BMI) were higher in the male than in the female students. In addition, there were no differences in BP and BMI between the male students with a hypertensive father and the male students having a normotensive father. However, multivariate analyses revealed that BMI was an independent predictor of SBP solely in the male students having a hypertensive father, but not in the male students having a normotensive father. Such a relationship between BMI and BP determination was not observed in the female students with one hypertensive parent. It is suggested that there are different mechanisms for the determination of BP in normotensive offspring of hypertensive parents, and genetic loading of a hypertensive father plays a critical role in the determination of BP through BMI. Keywords: family history of hypertension; body mass index; obesity; Y chromosome; autosomal chromosome Introduction A family history of essential hypertension is known to be a strong predictor of development of high blood pressure (BP). 1 In fact, normotensive offspring of parents with hypertension have slightly, but significantly higher BP levels than offspring of normotensive parents. 2 4 We have recently demonstrated that normotensive freshman students having both hypertensive parents have higher BP levels than those having normotensive parents. 5 The body mass index (BMI) of these students with higher genetic loading is a strong predictor of their BP levels. More interestingly, the offspring of both hypertensive parents have higher plasma insulin concentrations than those of normotensive parents. These data, being consistent with previous studies, 6 8 strongly suggest that there is a genetic mechanism between abnormalities of glucose metabolism and BP determination in the offpring of hypertensive parents. In this context, recent studies on genetic loading Correspondence: Dr Yoshio Uehara, Second Department of Medicine, University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-0033, Japan Received 12 August 1997; revised 29 December 1997; accepted 9 January 1998 of hypertension, using genetic strain of spontaneous hypertensive rats (SHR) and backcrossing techniques, have provided evidence for the role of Y chromosome, as well as autosomal chromosomes, in the pathogenesis of hypertension. Ely et al 9 11 have reported that Y chromosome independently elevates BP in male SHR probably through increased sympathetic nervous activity. The enhanced adrenergic nerve activity influences both BP values and glucose metabolism. These data seem plausible to explain the relationship between BP regulation and abnormalities of glucose metabolism. To our knowledge, however, there have been few data available addressing the role of Y chromosome in the BP regulation in humans. At the University of Tokyo, more than 3000 freshmen aged 18 to 20 years are enrolled every year. As part of the enrolment process, these students are required to have a medical check-up sponsored by the university. This programme has existed for more than 10 years, using almost the same protocol throughout that period. The programme has provided information on changes in health-related parameters over time. 5 In the present study, we analysed the database from students enrolled in 1993 and 1994, to investigate the role of a hypertensive father

442 in the determination of BP in normotensive freshman students. Subjects and methods Design of study The medical check-up programme for the freshmen students is operated by the university s Health Service Centre and performed by medical doctors and registered nurses designated by the university. These medical examinations are performed between late April and early May of each year. We analysed the data from 7279 students enrolled in the 2 years of 1993 and 1994. At first, the students were classified into four groups: (1) students having both parents who were diagnosed definitely by doctors with essential hypertension or have been treated with anti-hypertensive drugs. Their grandparents might be hypertensive or normotensive (FH-3); (2) students with one parent diagnosed definitely by doctors with essential hypertension and with or without hypertensive grandparents (FH-2); (3) students with at least one hypertensive grandparent without hypertensive parents (FH-1); and (4) students with neither hypertensive parents nor hypertensive grandparents or those who did not definitely know their family history of hypertension (FH-0). Records of 32 students in FH-2 were not available because they had dropped out until the year 1996 when this analysis was performed. To investigate the role of a hypertensive father in the determination of BP, we analysed the records from 609 students in group FH-2. The students were divided into four groups according to the sex of parents: (1) male students with a hypertensive father (father male student); (2) female students with a hypertensive father (father female student); (3) male students with a hypertensive mother (mother male student); and (4) female students with a hypertensive mother (mother female student). This programme includes a medical interview, physical examination, a chest roentgenogram, electrocardiogram and several biochemical tests. The interviews were conducted by registered nurses, including the students family history of various illnesses and their general health status. BP measurement and heart rate determination were performed by registered nurses who had reviewed the procedure for BP determination prior to the check-up. Medical doctors then performed the physical examinations. Whole blood was collected from the students after they had fasted for at least 12 h in order to determine total cholesterol level and glutamyloxaloacetic transaminase (GOT) activity in plasma. Blood pressure determination After the interview, the student s BP was measured by the cuff-sphygmomanometer method in a quiet, comfortable room. The first and fourth measuring points were used to indicate the systolic and diastolic BP, respectively. If the fourth measuring point could not be read definitely, the fifth point was used to indicate diastolic BP (DBP). BP was measured four times at 5-min intervals with the student in a relaxed upright position. The last three measurements were averaged. The student s heart rate was measured simultaneously. Determination of variables Height and body weight were measured using an automatic electric device (Model TBF-1, Tanita Co, Ltd, Tokyo, Japan). In biochemical testing, total plasma cholesterol and GOT levels were determined using an automatic analyser (Model Hitachi 736, Hitachi Co, Ltd, Tokyo, Japan). Statistical analysis Values were expressed as means ± s.e. Differences were evaluated by one-way analysis of variance (ANOVA), Student s t-test, multivariate and multiple regression analysis using STATISTICA programme (StatSoft, Tulsa, OK, USA) and Gateway 2000 P4D-100 computer system (Gateway 2000 Inc, N Sioux City, SD, USA) running on Windows 95 operating system. P values less than 0.05 were accepted as statistically significant. Results Annual medical check-up programme All 7279 freshmen students enrolled at the University of Tokyo participated in this medical check-up programme. Table 1 summarised their demographic data and family history of hypertension. There were 6175 males and 1104 females, age range 17 to 28 years; mean 19.22 years. More than 99% students were 17 to 21 years old. Approximately 8.8% of the students had one hypertensive parent (FH-2); for 0.3% of the students, both parents were hypertensive (FH-3). For all groups of students, systolic BP (SBP) and DBP were within the normotensive ranges. BP was comparable in the students without hypertensive relatives (FH-0) and students with only hypertensive grandparents (FH-1). However, those who had hypertensive parents (FH-3 and FH-2) showed a significant increase in BP values as compared with those with no family history of hypertension (FH-0). Interestingly, the BP of students where both parents were hypertensive (FH-3) appeared to be higher than in those with one hypertensive parent (FH-2) although the differences were not significant. Heart rates did not differ significantly among the four student groups. There were no significant differences in body weight (BW), height and body mass index (BMI) among the four groups of students. Total cholesterol levels tended to be higher in those with hypertensive parents (FH-2 and FH-3). GOT levels did not differ among the four groups. Analyses of FH-2 group We investigated the role of a hypertensive father in determination of BP level using data from 609 students in group FH-2. The various parameters in each

Table 1 Various parameters in students enrolled in 1993 and 1994 443 Parameters FH-0 FH-1 P values FH-2 P values FH-3 P values No. of Students 5122 1494 641 22 Height (cm) 169.6 ± 0.1 169.3 ± 0.3 0.040 169.0 ± 0.3 0.038 170.9 ± 1.6 0.379 Body weight (kg) 61.5 ± 0.2 61.2 ± 0.3 0.092 62.5 ± 0.4 0.066 62.3 ± 2.8 0.790 BMI (kg/m 2 ) 21.3 ± 0.1 21.3 ± 0.1 0.291 21.8 ± 0.1 0.003 21.3 ± 0.8 0.929 SBP (mm Hg) 122.8 ± 0.2 121.5 ± 0.4 0.703 125.7 ± 0.5 2.0E-8 128.7 ± 2.1 0.022 DBP (mm Hg) 71.0 ± 0.1 70.6 ± 0.3 0.199 73.1 ± 0.4 2.9E-7 76.5 ± 2.3 0.009 HR (beat/min) 76.9 ± 0.2 77.3 ± 0.4 0.265 77.4 ± 0.4 0.430 78.7 ± 1.6 0.559 GOT (U/L) 19.5 ± 0.1 19.7 ± 0.3 0.525 20.2 ± 0.5 0.093 22.2 ± 3.6 0.226 TC (mg/dl) 169.4 ± 0.4 173.0 ± 1.1 0.670 173.9 ± 1.3 0.658 172.8 ± 6.5 0.594 Summarised data for the students examined during the 2-year study period. FH-0, students with normotensive relatives; FH-1, students with at least one hypertensive grandparent and no hypertensive parents; FH-2, students with one parent with hypertension and with or without hypertensive grandparents; FH-3, students with hypertensive parents and with or without hypertensive grandparents; BMI, body mass index; SBP, systolic blood pressure; DBP, diastolic blood pressure; HR, heart rate; GOT, plasma glutamyl-oxaloacetic transaminase activity; TC, plasma total cholesterol level; 2.0E-8, 2.0 10 8 ; and 2.9E-7, 2.9 10 7. The difference was compared with the respective values in FH-0 group using one-way analysis of variance and Student s t-test. P values less than 0.05 were considered statistically significant. subgroup of FH-2 were shown in Table 2. BMI, SBP, DBP and plasma GOT levels were higher in the male students than in the female students of FH-2 group. More accurately, the male students having a hypertensive father (father male student) exhibited significantly higher BMI, BP levels and plasma GOT activity than the female students having a hypertensive father (father female student). More interestingly, there were no differences in BMI, SBP, DBP or plasma GOT levels between the male students having a hypertensive father (father male student) and the male students having a normotensive father (mother male student). Multivariate analyses To assess separately the independent predictors of BP in the male students of FH-2 group, raw determinants were analysed using multiple regression analyses. In the male students having one hypertensive parent, BMI, heart rate and DBP, in order of strength, were independent predictors of SBP; however, in the female students having one hypertensive parent, BMI did not influence the determination of SBP (Table 3). With regard to DBP, BMI was not related to the values either in the male or in the female students. To investigate these points further, we re-analysed the effects of the sex of parents and their offspring on BP determination using the data from four subgroups of FH-2 (Table 4). BMI and heart rate were major independent predictors of SBP solely in the male students having a hypertensive father (father male student), but not in the male students having a normotensive father (mother male student). This relationship was observed in neither the female students having a hypertensive father (father female student) nor the female students having a hypertensive mother (mother female student). With regard to DBP, we could not extract its independent predictors in any of the four subgroups. Discussion An increasing number of studies have demonstrated that normotensive persons with hypertensive parents exhibit higher levels of BP than the subjects having normotensive parents do. 1 5,12,13 The Table 2 Sex differences in various parameters of students having one hypertensive parent Group (n) BMI P SBP P DBP P HR P GOT P TC P (kg/m 2 ) values (mm Hg) values (mm Hg) values (beat/min) values (unit/l) values values Total of male 22.1±0.1 1.5E-8 126.9±0.5 5.2E-9 73.7±0.5 0.0024 77.4±0.5 0.9926 20.9±0.6 0.0022 173.1±1.4 0.1328 students (510) Total of female 20.2±0.2 119.2±1.2 70.2±1.0 77.4±0.8 16.7±0.4 178.2±3.2 students (99) Father male 22.2±0.2 3.7E-7 127.1±0.6 1.2E-5 74.2±0.5 0.0106 77.9±0.6 0.5138 21.4±0.8 0.0101 172.0±1.5 0.1744 student (368) Father female 20.1±0.3 120.0±1.5 70.7±1.3 77.1±1.0 16.4±0.5 177.4±3.9 student (70) Mother male 21.8±0.3 0.0403 126.3±1.0 0.0002 72.8±1.0 0.1769 75.9±0.9 0.4119 19.6±0.5 0.0372 173.7±2.9 0.2314 student (142) Mother female 20.6±0.4 117.1±2.1 69.6±1.7 77.6±1.4 17.1±0.9 182.3±6.4 student (29) Abbreviations as in Table 1. 1.5E-8, 1.5 10 8 ; 3.7E-7, 3.7 10 7 ; 5.2E-9, 5.2 10 9 ; and 1.2E-5, 1.2 10 5. n represents number of students. Differences between total of male students and female students, father male student and father female student or mother male student and mother female student were assessed by one-way analysis of variance and Student s t-test. P values less than 0.05 were considered statistically significant.

444 Table 3 Predictors of blood pressure in male and female students with one hypertensive parent Parameter Male P values Female P values student student Systolic blood pressure: BMI 0.2771 0.0000 0.1984 0.0607 DBP 0.1692 0.0001 0.2352 0.0193 HR 0.1579 0.0000 0.1098 0.2780 GOT 0.0357 0.3899 0.0644 0.5322 TC 0.0326 0.4347 0.0075 0.9440 Diastolic blood pressure: BMI 0.0296 0.5289 0.1719 0.1024 SBP 0.1881 0.0001 0.2314 0.0193 HR 0.0650 0.1309 0.0793 0.4304 GOT 0.0181 0.6801 0.2018 0.0466 TC 0.0823 0.0613 0.0505 0.6368 Abbreviations as in Table 1. Raw values of 510 male students and 99 female were analysed by multiple regression analysis. The independent variables were BMI, GOT, TC and HR, and the dependent variables various BP values. The values in the table represent standardised beta values. P value less than 0.05 were considered statistically significant. difference in BP level between the offspring of normotensive and hypertensive parents is observed in childhood as well as in adulthood. More interestingly, higher levels of BP in the offspring of hypertensive parents are often associated with an increase in plasma insulin concentrations or a decrease in insulin sensitivity and increased response of adrenergic nervous system to mental stress. 9 11,14,15 In fact, Scherrer et al 16 have demonstrated the relationship between BMI and the rate of sympathetic nerve discharge to skeletal muscle vasculature in normotensive subjects. In this context, we have demonstrated in our previous study that BP levels are well correlated to BMI in non-obese, normotensive freshman students having both hypertensive parents, and that such offspring have relatively higher plasma insulin concentrations. 5 We demonstrated in the present study that in FH- 2 group, BP levels were significantly higher in the male students than in the female students. The male students also had significantly higher BMI and GOT levels. Indeed, multivariate analyses revealed that in the male students, BMI was an independent predictor of BP, but not in the female students. When we analysed the sex-differences in the determination of BP in the normotensive offspring, it was revealed that BMI was an independent predictor of SBP solely in the male students having a hypertensive father (father male student), but not in the male students having a normotensive father (mother male student). These data clearly suggest that there are differences in mechanisms for BP determination between the male students having a hypertensive father and those having from a normotensive father. Genetic factor(s) from a hypertensive father are probably responsible in part for this linkage between abnormalities of glucose metabolism and BP determination. At this time, it is not clear how the genetic loadings from a hypertensive father are related to abnormalities of glucose metabolism. However, recent studies using hypertensive rat models have provided an insight into the understanding of the mechanism of genetic loading of hypertension. These have emphasised the role of autosomal chromosomes and Y chromosome in determination of BP in SHR. 9,10,17,18 Considering these data from backcrossed SHR, Y chromosome from a male rat with hypertension is responsible in part for elevation of BP probably through enhanced adrenergic nerve system or androgen receptor mediated events. 11,19 If hypertensive Y chromosome influences adrenergic nerve activity in human as well, 12 15 this theory is able to explain the linkage between the genetic loading from a hypertensive father and BMI observed in the present study. In this context, we demonstrated in the present study that heart rate is one of the predictors of SBP in the male students having a hypertensive father. This suggests that adrenergic nervous system plays a role in producing the relationship between BP determination and glucose metabolism. To address these questions, however, we have to expand our study and investigate Table 4 Sex effects of hypertensive parent and offspring on prediction of blood pressure Parameter Father male P values Father female P values Mother male P values Mother female P values student (368) student (70) student (142) student (29) Systolic blood pressure: BMI 0.3136 0.0000 0.1573 0.2268 0.1331 0.1643 0.2347 0.2298 DBP 0.1589 0.0010 0.2956 0.0186 0.1889 0.0242 0.1573 0.4212 HR 0.1715 0.0005 0.1653 0.1786 0.0958 0.2428 0.2967 0.1373 GOT 0.0533 0.2781 0.0262 0.8382 0.1255 0.1756 0.0883 0.6353 TC 0.0321 0.5185 0.0282 0.8310 0.0438 0.6164 0.3792 0.0841 Diastolic blood pressure: BMI 0.0105 0.8539 0.2356 0.0618 0.1062 0.2749 0.0433 0.8383 SBP 0.1838 0.0010 0.2821 0.0186 0.1945 0.0242 0.1800 0.4212 HR 0.1002 0.0606 0.0677 0.5752 0.0063 0.9397 0.2895 0.1772 GOT 0.0426 0.4205 0.1730 0.1653 0.1010 0.2835 0.2372 0.2273 TC 0.0541 0.3115 0.0330 0.7978 0.1399 0.1135 0.4432 0.0574 Abbreviations as in Table 1. Raw values of 510 male students and 99 female having one hypertensive parent were divided into four subgroups as described in text. Number in parentheses represents number of students. The values were analysed by multiple regression analysis. The independent variables were BMI, GOT, TC and HR, and the dependent variables various BP values. The values in the table represent standardised beta values. P values less than 0.05 were considered statistically significant.

more precisely the alterations in adrenergic nerve system in the student group having only one hypertensive parent (FH-2). In the present study, we interviewed the students to investigate their family history of hypertension. We reconfirmed the family history from their parents whenever it was not clear. These procedures were of use to increase the reliability of history taking. Despite such efforts, ultimately, we could not determine the family history in approximately 1.2% of the students. These unknown cases were included in FH-0 group. There is a difficulty in assessing how much is genetic loading of hypertension in humans because a number of hypertensive parents are presumed to be heterogeneous in hypertensive genes. Strictly speaking, a hypertensive father may include hypertensive and normotensive chromosomes. However, considering the fact that the parents in the present study were in their forties or fifties and develop hypertension at relatively younger ages, it seemed possible that the genetic loading of hypertension is greater, and that hypertensive genes occur in the Y chromosome as well as autosomes. In the present study, we demonstrated that the relationship between BMI and BP determination was observed in the male students having a hypertensive father, but not in the male students having a hypertensive mother and a normotensive father. The male students having a hypertensive mother and a normotensive father must possess hypertensive genes in their 22 autosomes. This suggests that autosomes from a hypertensive parent are not responsible for the relationship between BP determination and BMI. In addition, we also demonstrated that there were no differences in BP determination between the female students having a hypertensive mother and those having a hypertensive father. These students also have hypertensive autosomes from a hypertensive mother or the autosomes from a hypertensive father, suggesting that the hypertensive autosomes do not produce the relationship between BP determination and BMI. These data favours the hypothesis that genetic factor(s) from a hypertensive father alone are involved in producing the linkage between BP determination and BMI. Finally, we demonstrated in the present study that genetic loading of normotensive and hypertensive fathers increases BP, as compared with female students. In addition, genetic factor(s) from a hypertensive father are required to establish the relationship between BP determination and BMI in very young normotensive offspring. At this particular time, however, we are unable to address definitely the mechanism of this relationship. Population analyses using genetic techniques would provide much information to understand the pathogenesis of hypertension in humans. Acknowledgements The authors thank Yukari Kawabata for data analyses. This study was supported in part by a Grantin-Aid from the Ministry of Education, Science and Culture of Japan (No. 07670762). References 1 Alleman Y, Weidmann P. Cardiovascular, metabolic and hormonal dysregulations in normotensive offspring of essential hypertensive parents. J Hypertens 1995; 13: 163 173. 2 Himmelmann A, Svensson A, Hansson L. Five-year follow-up of blood pressure and left ventricular mass in children with different maternal histories of hypertension: the Hypertension in Pregnancy Offspring Study. J Hypertens 1994; 12: 89 95. 3 Munger RC, Prineas RJ, Comez-Marin O. Persistent elevation of blood pressure among children with a family history of hypertension: The Minneapolis Children s Blood Pressure Study. J Hypertens 1988; 6: 647 653. 4 van Hooft IMS, Hofman A, Grobbee DE, Valkenburg HA. Change in blood pressure in offspring of parents with high or low blood pressure: The Dutch Hypertension and Offspring Study. J Hypertens 1988; 6 (Suppl 4): 5594 5596. 5 Uehara Y et al. Body mass index is a determinant of blood pressure in young adults with essential hypertensive parents. J Hum Hypertens 1996; 10: 601 606. 6 Facchini F et al. Insulin resistance, hyperinsulinemia, and dyslipidemia in nonobese individuals with a family history of hypertension. Am J Hypertens 1992; 5: 694 699. 7 Ferrannini E et al. Insulin resistance in essential hypertension. N Engl J Med 1987; 317: 350 357. 8 Ohno Y et al. Impaired insulin sensitivity in young, lean normotensive offspring of essential hypertensives: possible role of disturbed calcium metabolism. J Hypertens 1993; 11: 421 426. 9 Ely DL, Turner ME. Hypertension in the spontaneously hypertensive rat is linked to the Y chromosome. Hypertension 1990; 16: 277 281. 10 Ely DL et al. The hypertensive Y chromosome elevates blood pressure in F 11 normotensive rats. Hypertension 1993; 21 (Part 2): 1071 1075. 11 Ely DL et al. Spontaneously hypertensive rat Y chromosome increases indexes of sympathetic nervous system activity. Hypertension 1997; 29: 613 618. 12 Doyle AE, Fraser JRE. Essential hypertension and inheritance of vascular reactivity. Lancet 1961; ii: 509 511. 13 Bianchetti MG, Beretta-Picolli C, Weidmann P, Ferrier C. Blood pressure control in normotensive members of hypertensive families. Kidney Int 1986; 29: 882 888. 14 Turner ST, Reilly SL. Renal sodium excretion in sons of hypertensive parents. Hypertension 1993; 22: 323 330. 15 Perni C et al. Effects of psychological and physical covariates on plasma catecholamines in borderline hypertensives and offspring of hypertensive parents. Clin Exp Hypertens (A) 1990; 12: 137 150. 16 Scherrer U et al. Body fat and sympathetic nerve activity in healthy subjects. Circulation 1994; 89: 2634 2640. 17 Turner ME, Johnson ML, Ely DL. Separate sex-influenced and genetic components in spontaneously hypertensive rat hypertension. Hypertension 1991; 7 (Part 2): 1097 1103. 18 Davidson AO et al. Blood pressure in genetically hypertensive rats. Influence of the Y chromosome. Hypertension 1995; 26: 452 459. 19 Johnson ML, Ely DL, Turner ME. Steroid sulfatase and the Y chromosome hypertensive locus of the spontaneously hypertensive rat. Steroids 1995; 60: 681 685. 445