Tobacco Documents Online

U(

96-274 INCREASED SERUM MITOGENIC ACTIVITY FOR ARTERIAL SMOOTH MUSCLE CELLS ASSOCIATED WITH RELAXATION AND LOW EDUCATIONAL LEVEL IN HUMAN SUBJECTS WITH HIGH BUT NOT LOW HOSTILITY TRAITS: IMPLICATIONS FOR ATHEROGENESIS William H. Gutstein, M.D.; Jeanne A. Teresi, Ed.D., Ph.D.; Joseph M. Wu, Ph.D.; Fereshteh Salimian, M.S.; Yanning Cui, M.D., Inge Paul, B.S., Sana Jabr, B.S. Running Title: Hostility and Atherosclerosis Department of Pathology, New York Medical College (W.H.G., J.M.W., F.S., Y.C., I.P., S.J.); Hebrew Home for the Aged at Riverdale and Columbia University Center for Geriatrics and New York State Office of Mental Health (J.T.) For correspondence contact: Dr. Gutstein at the Department of Pathology, Basic Science Building, New York Medical College, Valhalla, New York 10595. Telephone: (914)-993-4154 Fax (914)-993-4679. 9/24/96 s

underlying trait of hostility in humans. Electrical stimulation of the diencephalon also produces lesions in arteries (aorta) of rodents (15) and non-human primates (16) which bear the essential characteristics of atherosclerotic plaques - i.e., proliferation of arterial smooth muscle cells, collagen and elastin synthesis and, in squirrel monkeys, lipid deposition even when diets are not high in fat content. When rats experiencing hypothalamic stimulation were investigated in an in vitro system for a substance in their sera with growth-promoting properties for cells of the arterial wall which participate in the formation of the atherosclerotic plaque, i.e., arterial smooth muscle cells, such a substance(s) was often present (17-19). This finding suggested a possible link between aggressive behavior - emanating from the subcortical regions of the brain - and a cellular response in arteries contributing to atherogenesis via a serum factor which was mitogenic for these cells, although the mechanisms involved have not yet been elucidated. Reasoning, however, that a similar relationship might exist in human populations, the present study was undertaken among 225 normal subjects. METHODS Study Desian Two hundred and twenty five healthy volunteers, aged 21-65, 3

INTRODUCTION Atherosclerosis is a disease in which the end-stage lesion - the fibrous plaque - is characterized by an abnormal focal proliferation of cells in the intima of the arterial wall, which have migrated there from the media and undergone active replication. Although this replication - or proliferation - is known to be influenced in vitro and in vivo by a number of blood- borne substances (1), designated as either "mitogens" or "growth factors," in humans the stimulus for this phenomenon is unknown. In recent years, studies have shown that various psychological factors may play an important, independent, role in the development of this pathology (for review see reference 2). More recent evidence suggests that anger and "hostility" (Ho) are some of the main factors not only in the coronary circulation (3,4), but.in connection with cerebrovascular disease (5,6) and peripheral arterial disease, as well (7). However, the means by which these characteristics translate into pathology at the level of the arterial wall are not understood. In both animals (8-10) and humans (11-14), electrical stimulation of the diencephalon (hypothalamus) and structures closely connected with it, such as the limbic system and tegmentum of the midbrain, may elicit "aggression" and aggressive feelings - forms of behavior which are closely linked to the 2

at least 5 years at sometime within the past 15 years. These were admitted into the study, since the great majority of subjects consisted of individuals who were not current smokers. Of the 225 subjects, 125 were male and 100 were female. Most (97.3%) were white and 2.7% were black. Marital status was reported as 28.3% single, 58.4% married and 13.3% separated, divorced or widowed. The educational level consisted of 12-14 years in 25.6%, 15-17 years in 51.8% and 18-20 years in 22.6%. The average age was 40.85 (s.d.=10.77, median=41); 18% of the sample was between the ages of 21 and 30, 31% between 31 and 40; 31% between 41 and 50 and 20% between 51 and 65. Physical exercise (moderate) was reported as < 3 x 1 week by 49.3-t and > 3 x 1 week by 50.7%. Psychological instruments The 50 item Cooke-Medley Hostility subscale of the Minnesota Multiphasic Personality Inventory (MMPI) (20)was administered prior to the intervention. The Hostility subscale is composed of fifty dichotomous (yes/no) items. The scale alpha was .82. For some analyses, the hostility measure was dichotomized at the median with scores of 17 and over representing high hostility. Cardiovascular Measures Blood pressure (BP), pulse rate (PR), serum total and high density lipoprotein cholesterol concentrations (T/HDL-C) were measured immediately before and after the induced emotional

in an "exasperating" manner. Most of the subjects later reported some degree of "irritation" or "anger" at these remarks. Group 3: Relaxed (n=40) The test session here differed radically from those of the previous 2 groups. During the 30 minute session, the subjects did not play the electronic game and were not exposed to any critical remarks, but were comfortably seated in the reclining position of a sofa-chair and asked to close their eyes and sleep if they wished. A relaxation tape with musical background at a low level was played during the first ten minutes. The interviewer left the room far the entire 30 minute period. 'Almost all of the individuals reported that they closed their eyes and many said that they ^dozed," but reported falling fully asleep. Serum biochemical assays PDGF was measured on whole blood serum by a sensitive none radioimmunoassay method (Amersham International Pic, Amersham, LTK) . Total and HDL-cholesterol concentrations were also measured in whole blood serum using the Abbot Vision System (Abbot Laboratories, Abbot Park,IL 60064) with total cholesterol assayed by the coupled cholesterol oxidase-peroxidase method (22), and HDL cholesterol, after initial precipitation of LDL and VLDL by magnesium and dextran sulphate, based on the method of Artiss and Zak (23). 8

were exposed to three induced emotional states: a) frustration (group 1), b) provoked anger (group 2) and c) relaxation (group 3). The third state, "relaxation," was included, because one of the objectives was to examine the influence of suppressed, but not elicited, hostility on the possible increase of a serum substance similar to that found in the animal experiments. Characteristics of the sample Volunteers were recruited from the general public via the news media (local newspapers and radio). They were screened to rule out recent history of acute or chronic illness, surgery, known psychologic illness, current use of any medication, and occurrence of a major life event (separation/divorce/death of a loved one) within the past 6 months. Of the 250 individuals initially recruited, 25 had to be dropped for various reasons, such as not meeting the above criteria, inability to keep necessary appointments, etc. _ All remaining participants were required to pass a complete physical examination, routine clinical laboratory examinations and an exercise stress test for cardiovascular performance. Individuals whose basal systolic blood pressure was greater than 140 mm Hg or whose diastolic pressure exceeded 90 mm Hg and those whose fasting serum cholesterol concentration was higher than 220 mg/dL were excluded. However, there were 50 individuals (22.0%) who reported smoking at least 20 cigarettes / day for a period of N~f1 4 ~ N . O 0D O

states. Serum values of platelet derived growth factor (PDGF) and thymidine (Thy) incorporation were also measured as indices of mitogenic activity. All measurements were expressed as changes in value as a result of the intervention. Procedures Prior to acceptance into the study, informed written consent to participate in each phase was obtained from all subjects. The study design was approved by the Institutional Review Board of New York Medical College. Upon arrival, subjects completed the MMPI questionnaire, and were allowed to rest for a period of 15 minutes. An IV catheter was then placed in the antecubital vein of the left forearm. A BP cuff was placed around the right arm. Catheter and cuff were arranged so as not to interfere with hand movements during the test session. Following another 15 minute period of adjustment, BP was recorded 3x from an automatic monitor at 2 minute intervals and the results (systolic/diastolic) averaged. Pulse rate was simultaneously measured. Blood samples were obtained from the indwelling catheter and assayed as below. All samples were obtained at least 3 hours after a light breakfast consisting of 1 slice of dry toast and an unsweetened beverage of tea or coffee. The test session described was then administered for 30 minutes. During this period all subjects were video filmed. At the end of the session BP and PR were recorded as before N 01 0 U1 6 ~ N f~1

and a second blood sample obtained. All subjects were then debriefed and notified of the "purpose" of the study, which was to determine if an atherogenic substance had been increased in their blood under the conditions of the test session to which they were exposed. Test Session: Subjects were assigned to one of three groups. Group 1: Frustrated (n=40) Individuals were asked to play a hand held electronic game (Mario Brothers) in the presence of one of the members of the research team within the time period of the session and to achieve a score, known by the team to be extremely difficult or impossible to obtain. None of the participants was familiar with the game, so did not question this objective. During the session, the interviewer periodically encouraged the subject to improve his/her score, but scrupulously avoided criticism of any kind. All subjects later reported some degree of "frustration" with the task. Group 2: Harassed (n=145) The same strategy as used for group 1 was employed except that the interviewer (the same individual as in group 1), rather than encouraging the subject, uttered "subtly" disparaging and critical remarks concerning the accomplishments of the participants during the session. Comments such as "I expected you to do better than that," or "you're not even performing at the level of a ten year old child," were said 7

(intervention effects), comparing those with low and high hostility. Selection of statistical models and tests of model assumptions Indices of serum mitogenic activity reflected by changes concentration of PDGF in vivo or of thymidine in vitro (i.e., in without PDGF) were statistically analyzed using a general linear model. Both ANCOVA and hierarchical multiple regression analyses were used to examine the study hypothesis; a two-tailed test was used. Preparatory to analyses, extensive examination of variable distributions were undertaken to ensure adequate ranges or cell sizes. The dependent variable for the final analysis was percent change in thymidine incorporation (proliferation assay) treated as a continuous variable. Because we were dealing with a dependent variable where the units are proportions, an arcsine transformation of the thymidine change variable was first performed in order to obtain a distribution whose units were more linearly related to other variables. (Parenthetically, this correction, while increasing the zero-order associations slightly with most variables, did not appreciably affect the multivariate results.) Colinearity diagnostics were performed examining correlations among input variables, variance inflation factors and tolerances. In instances where colinearity was observed, the 11