Journd of Psychosomatic Research, Vol. 34, No. I, pp. 29-33, 1990. Pnnted in Great Britain.
0
0022-3999/W $3.00 + .Xl 1990 Pergamon Press plc
BETA-ADRENERGIC RECEPTOR SENSITIVITY IN SUBJECTS PRACTICING TRANSCENDENTAL MEDITATION PAUL J. MILLS,* ROBERT H. SCHNEIDER, DAVID HILL, KENNETH G. WALTON,** and R. KEITH WALLACE (Received 26 April 1989; accepted in revised form 13 June 1989) Abstract-Several studies suggest that behavioral techniques such as meditation and relaxation may be associated with reduced end organ adrenergic receptor sensitivity. Thus far the evidence supporting this hypothesis has been indirect. We present preliminary findings showing reduced beta-adrenergic receptor sensitivity in a group of subjects practicing Transcendental Meditation. The meditation group (N = lo), compared to controls (N = lo), had a lower percentage of functional lymphocyte beta-adrenergic receptors (p = 0.009), but showed no difference in total receptor number or plasma catecholamines. There were no differences between the groups in Type A behavior, the Type A components, exercise, or family history of hypertension. The results provide some support for studies postulating that meditation is associated with reduced sympathetic adrenergic receptor sensitivity, and provide encouragement for the efficacy of receptor measurement in psychophysiology research.
INTRODUCTION BEHAVIORAL techniques continue to receive attention as potential methods for modifying reactivity to stressors [l]. For example, studies using Transcendental Meditation have reported changes in blood pressure [2] and heart rate [3] reactivity to stressors. In addition to changes in reactivity, studies suggest that behavioral techniques modify end organ receptor sensitivity. Hoffman and associates [4] reported that, compared to controls, subjects taught relaxation had disproportionately higher norepinephrine responses to a stressor despite nearly identical blood pressure and heart rate responses. The authors interpreted this pattern as an adrenergic end organ desensitization associated with the behavioral treatment. A later study of nearly identical design also found differences in cardiovascular and adrenergic responsiveness to stressors in the relaxation group as compared to controls, and suggested the need for further research on potential underlying mechanisms, such as adrenergic receptors [5]. Recent studies showing a relationship between beta-adrenergic receptor function and blood pressure and blood pressure reactivity further support the potential usefulness of this approach [6, 7, 81. This study addressed this issue by directly measuring beta-adrenergic receptor sensitivity in subjects practicing the Transcendental Meditation (TM) technique. TM is a stress management technique which requires no special belief systems or lifestyle and which has been associated with reduced cardiovascular disease [9]. In contrast to previous studies, we attempted to control for possible differences in behavior and *Present address: Dr Paul Mills, Department of Pschiatry (T-004), University of California, San Diego, La Jolla, California 92093, U.S.A. Department of Physiological and Biological Sciences, Maharishi International University, Fairfield, Iowa, U.S.A. **Department of Chemistry, Maharishi International University, Fairfield, Iowa, U.S.A. Address reprint requests to: Dr Robert Schneider, Department of Physiological and Biological Sciences, Maharishi International University, Fairfield, 52556, U.S.A. 29
30
PAUL J. MILLS et al.
lifestyle patterns thought to influence sympathetic nervous system functioning. Another difference from previous studies, the majority of which used short-term practitioners [l], is that we used relatively long-term TM practicing subjects. Beta-adrenergic receptors were measured in lymphocytes. Lymphocytes are used as a relatively noninvasive model of human beta-adrenergic receptor function because lymphocyte beta2-adrenergic receptors parallel beta-adrenergic receptors in the heart and vasculature [ 10, 11, 121. METHODS The subjects were 20 non-obese male volunteers. Subjects were recruited through an advertisement placed both in the local newspaper and at employment sites where permission was obtained. For all subjects, inclusion criteria included being male and in good health. For the TM group, a minimum of 2 to 3 three years of regular TM practice was required. For the controls, it was required that they were not practicing any type of behavioral technique. All were employed, none were taking medication, and all abstained from coffee, tea, or cocoa for at least 15 hr before the study. The experimental group consisted of 10 subjects practicing TM, mean age 32 (SD = 5) years, mean length of TM practice 9 (SD = 5) years. The control group consisted of 10 age- and occupation-matched subjects not practicing TM or any behavioral technique, mean age 30 (SD = 7) years. Subjects were evaluated for Type A behavior, the Type A behavior components, exercise habits, and family history of hypertension. Type A behavior and the component scores were derived from the Structured Interview (SI). The SI is considered the most reliable method of assessing Type A behavior 1131. Behavior was scored blind to meditator/control status on a scale of 1 to 6, with extreme Type A and extreme Type B corresponding to number 1 and to number 6 respectively. The components were scored on a scale of 1 to 5 according the method of Dembroski et al. [14]. Number 1 and number 5 corresponded to a strong and weak presence respectively of that component. The seven components were loud, explosive, rapid, response latency, potential for hostility, angerout, and verbal competition. Exercise and family history were scored as either a 1 (Yes) or 2 (No) to indicate regular exercise or a lack thereof, and the presence or absence of high blood pressure in one or both parents. Subjects sat in a comfortable chair and a 21-gauge cannula was inserted into a superficial vein of the right forearm; the catheter was kept patent with a heparinized saline infusion. Following a 30 min rest, a blood sample was drawn to assess lymphocyte beta-adrenergic receptors and plasma catecholamines. Lymphocytes were isolated from fresh heparinized blood [15] and membrane lysates were prepared according to the method of Aarons et al. [16]. Radioligand binding experiments were conducted to determine the density of beta2-adgenergic receptors (B,,,), the dissociation constant (Kd), and the percentage of high affinity state receptors (Rh). The Kd is a measure of the affinity or ‘binding strength’ of the radioligand for the receptor. The high affinity state is considered the functional configuration of the beta receptor where agonist binding translates into an end organ response. binding isotherms were performed according to the methods of Feldman et al. [17] For I%,,, , saturation using ‘251-iodocyanopindolol (ICYP) at six concentrations from 5 to 30 picoM. Both B,,, and the Kd for ICYP were determined using Scatchard analysis [18]. The Rh was derived from competition binding of ICYP with 14 concentrations of isoprotereno: from 100 microM to 1 nanoM. These data were analyzed
TABLE I.-SAMPLE
CHAKACTERISTICS(MEAN,* SD)
TM Type A behavior Loud Explosive Rapid Response latency Hostility Angerout Verbal competition Exercise Family history of hypertension *See text for scoring
2.6 3.4 3.5 3.0 3.5 3.2 3.5 3.5 1.4
(I .4) (0.91) (0.92) (0.53) (0.92) (1.1) (0.92) (0.91) (0.52)
1.5 (0.51) method.
Control 2.4 3.8 3.8 3.1 3.4 3.5 3.4 3.2 1.3
P
(0.88) (1.0) (0.92) (0.33) (0.72) (0.52) (0.72) (0.66) (0.51)
0.75 0.30 0.40 0.61 0.89 0.48 0.89 0.48 0.27
1.3 (0.48)
0.26
Beta-adrenergic
receptors
and meditation
31
3s 30
PercentRh
E, 20 15 10 5 0 M
FIG. I.-The percent of beta-adrenergic receptors in the high affinity state (Rh). The TM group had a lower Rh than the controls indicating a reduced percent of functional receptors (p = 0.009). by a nonlinear curve-fitting procedure using a generalized model for complex ligand receptor systems (The Ligand Program, NIH). The quality of the fit of the data to the one- vs the two-affinity state model was compared by F-test and the fit to the two-affinity state model was deemed signiticanly better when p < 0.05. For the two-affinity state model, the computer estimated the proportion of receptors in the high (Rh) and low (Rl) affinity state. Plasma samples were analyzed for norepinephrine and epinephrine by high performance liquid chromatography with electrochemical detection, with an interassay coefficient of < 10% [19]. Separate one-way ANOVAs using BMDP statistical software were used to compare groups for the dependent (receptors and catecholamines) and independent variables (Type A behavior, the components, exercise, and family history). RESULTS
There were no significant differences between groups for Type A behavior, the components, family history, or exercise patterns, see Table I. Therefore, these variables were not controlled for in the receptor analyses. Compared to controls, the TM group had a lower Rh (F( 1,18) = 8.3, p = 0.009), indicating a reduced percentage of receptors in the functional high affinity state, see Figure 1. There were no differences in total receptor number, the dissociation constant, Kd, or plasma catecholamines between the groups, see Table II. DISCUSSION
Receptor number, affinity, and sensitivity are traditionally evaluated in receptor research. By directly measuring lymphocyte beta-adrenergic receptors, we have found some support for the hypothesis that TM is associated with reduced end organ receptor sensitivity. A variety of diverse mechanisms mediate adrenergic receptor sensitivity and may account for this finding [20,21]. For example, both human and animal studies indicate that corticosteroids are important in facilitating the formation of high affinity state beta-adrenergic receptors [22,23,24]. Several studies have TABLE
II.-BETA-ADRENERGIC
RECEPTOR AND (MEAN,SD)
TM B,,, (fmol/mg protein) Kd (pmol) Norepinephrine (pg/ml) Epinephrine (pg/ml)
186.7 (116) 3.46 (2) 316 (124) 47 (25)
~ATEC~IOLAMINE DATA
Control 128.5 (51) 3.23 (1) 372 (300) 71 (104)
P
0.16 0.74 0.55 0.54
PAUL J. MILLS ef al.
32
found lower cortisol in subjects practicing TM [25,26]. Although we do not have cortisol data on this particular group of TM subjects, perhaps lower cortisol levels in the TM group contributed to the lower percentage of high affinity state receptors. Changes in adrenergic receptor function may also help explain reported findings of lower blood pressure in TM subjects [27,28]. We found no difference in receptor number of plasma catecholamine levels between the TM and controls subjects. Beta-adrenergic receptor number is strongly influenced by plasma catecholamine levels [21]. Higher levels are associated with a reduction or ‘down regulation’ of receptor number and lower levels are associated with an increase or ‘up-regulation’ of receptor number. The similarity in plasma catecholamines between the groups may help explain the similarity in receptor number, and suggests that factors other than adrenergic activation are responsible for the change in receptor affinity. The dissociation constant for the radioligand, Kd, was nearly identical between the groups. From a pharmacological perspective this is what one would expect because this binding characteristic is quite stable. This study employed a cross-sectional design using long-term TM subjects. The question then arises as to the cause of the receptor differences. Is it self-selection bias, the meditation technique itself, possible changes in lifestyle, or a combination of these? Given the limitations of a cross-sectional study, it is difficult to adequately evaluate these questions with the present data. We attempted to address some of these issues by matching for sex and age and by measuring a number of pertinent independent variables. We found no differences between the two groups on these independent variables. Whether a more thorough evaluation of other personality or lifestyle variables would uncover other factors influencing the receptors remains to be determined. These questions can only be fully answered using a prospective design. Considering the results of both this and other studies [4, S] which have examined the hypothesis that behavioral techniques influence receptor physiology, a prospective study seems warranted to further delineate the intriguing effects of these techniques on receptor function. The results also point to the potentially useful contribution of receptor measurement in psychophysiology research [29]. .4cknoM/~dg~menr.s-The authors would like to thank Dr Gordon Moss and the Behavioral Medicine Group at Eastern Michigan University for scoring the Structured Interviews. We would also like to thank Dr Marcia M. Ward for helpful advice, Dr Ivan Mefford for performing the catecholamine analyses, and Mary Chirico, RN for technical assistance. This study was supported by a grant from the Davis Foundation.
REFERENCES JA(.ORRG, CHESNEYMA. Psychological
In Hundhook
and behavioral
methods
to reduce cardiovascular reactivity. MATTHEWSKA, WEISS SM, SB, WILLIAMSRB), pp. 417457. Wiley, New York.
of Stress, Reacririty, und Cardiot~ascular Disease. (Edited by
DETKET, DEMBKOSKI TM. FALKNERB. MANXK
1986. LAY
0
0022-3999/W $3.00 + .Xl 1990 Pergamon Press plc
BETA-ADRENERGIC RECEPTOR SENSITIVITY IN SUBJECTS PRACTICING TRANSCENDENTAL MEDITATION PAUL J. MILLS,* ROBERT H. SCHNEIDER, DAVID HILL, KENNETH G. WALTON,** and R. KEITH WALLACE (Received 26 April 1989; accepted in revised form 13 June 1989) Abstract-Several studies suggest that behavioral techniques such as meditation and relaxation may be associated with reduced end organ adrenergic receptor sensitivity. Thus far the evidence supporting this hypothesis has been indirect. We present preliminary findings showing reduced beta-adrenergic receptor sensitivity in a group of subjects practicing Transcendental Meditation. The meditation group (N = lo), compared to controls (N = lo), had a lower percentage of functional lymphocyte beta-adrenergic receptors (p = 0.009), but showed no difference in total receptor number or plasma catecholamines. There were no differences between the groups in Type A behavior, the Type A components, exercise, or family history of hypertension. The results provide some support for studies postulating that meditation is associated with reduced sympathetic adrenergic receptor sensitivity, and provide encouragement for the efficacy of receptor measurement in psychophysiology research.
INTRODUCTION BEHAVIORAL techniques continue to receive attention as potential methods for modifying reactivity to stressors [l]. For example, studies using Transcendental Meditation have reported changes in blood pressure [2] and heart rate [3] reactivity to stressors. In addition to changes in reactivity, studies suggest that behavioral techniques modify end organ receptor sensitivity. Hoffman and associates [4] reported that, compared to controls, subjects taught relaxation had disproportionately higher norepinephrine responses to a stressor despite nearly identical blood pressure and heart rate responses. The authors interpreted this pattern as an adrenergic end organ desensitization associated with the behavioral treatment. A later study of nearly identical design also found differences in cardiovascular and adrenergic responsiveness to stressors in the relaxation group as compared to controls, and suggested the need for further research on potential underlying mechanisms, such as adrenergic receptors [5]. Recent studies showing a relationship between beta-adrenergic receptor function and blood pressure and blood pressure reactivity further support the potential usefulness of this approach [6, 7, 81. This study addressed this issue by directly measuring beta-adrenergic receptor sensitivity in subjects practicing the Transcendental Meditation (TM) technique. TM is a stress management technique which requires no special belief systems or lifestyle and which has been associated with reduced cardiovascular disease [9]. In contrast to previous studies, we attempted to control for possible differences in behavior and *Present address: Dr Paul Mills, Department of Pschiatry (T-004), University of California, San Diego, La Jolla, California 92093, U.S.A. Department of Physiological and Biological Sciences, Maharishi International University, Fairfield, Iowa, U.S.A. **Department of Chemistry, Maharishi International University, Fairfield, Iowa, U.S.A. Address reprint requests to: Dr Robert Schneider, Department of Physiological and Biological Sciences, Maharishi International University, Fairfield, 52556, U.S.A. 29
30
PAUL J. MILLS et al.
lifestyle patterns thought to influence sympathetic nervous system functioning. Another difference from previous studies, the majority of which used short-term practitioners [l], is that we used relatively long-term TM practicing subjects. Beta-adrenergic receptors were measured in lymphocytes. Lymphocytes are used as a relatively noninvasive model of human beta-adrenergic receptor function because lymphocyte beta2-adrenergic receptors parallel beta-adrenergic receptors in the heart and vasculature [ 10, 11, 121. METHODS The subjects were 20 non-obese male volunteers. Subjects were recruited through an advertisement placed both in the local newspaper and at employment sites where permission was obtained. For all subjects, inclusion criteria included being male and in good health. For the TM group, a minimum of 2 to 3 three years of regular TM practice was required. For the controls, it was required that they were not practicing any type of behavioral technique. All were employed, none were taking medication, and all abstained from coffee, tea, or cocoa for at least 15 hr before the study. The experimental group consisted of 10 subjects practicing TM, mean age 32 (SD = 5) years, mean length of TM practice 9 (SD = 5) years. The control group consisted of 10 age- and occupation-matched subjects not practicing TM or any behavioral technique, mean age 30 (SD = 7) years. Subjects were evaluated for Type A behavior, the Type A behavior components, exercise habits, and family history of hypertension. Type A behavior and the component scores were derived from the Structured Interview (SI). The SI is considered the most reliable method of assessing Type A behavior 1131. Behavior was scored blind to meditator/control status on a scale of 1 to 6, with extreme Type A and extreme Type B corresponding to number 1 and to number 6 respectively. The components were scored on a scale of 1 to 5 according the method of Dembroski et al. [14]. Number 1 and number 5 corresponded to a strong and weak presence respectively of that component. The seven components were loud, explosive, rapid, response latency, potential for hostility, angerout, and verbal competition. Exercise and family history were scored as either a 1 (Yes) or 2 (No) to indicate regular exercise or a lack thereof, and the presence or absence of high blood pressure in one or both parents. Subjects sat in a comfortable chair and a 21-gauge cannula was inserted into a superficial vein of the right forearm; the catheter was kept patent with a heparinized saline infusion. Following a 30 min rest, a blood sample was drawn to assess lymphocyte beta-adrenergic receptors and plasma catecholamines. Lymphocytes were isolated from fresh heparinized blood [15] and membrane lysates were prepared according to the method of Aarons et al. [16]. Radioligand binding experiments were conducted to determine the density of beta2-adgenergic receptors (B,,,), the dissociation constant (Kd), and the percentage of high affinity state receptors (Rh). The Kd is a measure of the affinity or ‘binding strength’ of the radioligand for the receptor. The high affinity state is considered the functional configuration of the beta receptor where agonist binding translates into an end organ response. binding isotherms were performed according to the methods of Feldman et al. [17] For I%,,, , saturation using ‘251-iodocyanopindolol (ICYP) at six concentrations from 5 to 30 picoM. Both B,,, and the Kd for ICYP were determined using Scatchard analysis [18]. The Rh was derived from competition binding of ICYP with 14 concentrations of isoprotereno: from 100 microM to 1 nanoM. These data were analyzed
TABLE I.-SAMPLE
CHAKACTERISTICS(MEAN,* SD)
TM Type A behavior Loud Explosive Rapid Response latency Hostility Angerout Verbal competition Exercise Family history of hypertension *See text for scoring
2.6 3.4 3.5 3.0 3.5 3.2 3.5 3.5 1.4
(I .4) (0.91) (0.92) (0.53) (0.92) (1.1) (0.92) (0.91) (0.52)
1.5 (0.51) method.
Control 2.4 3.8 3.8 3.1 3.4 3.5 3.4 3.2 1.3
P
(0.88) (1.0) (0.92) (0.33) (0.72) (0.52) (0.72) (0.66) (0.51)
0.75 0.30 0.40 0.61 0.89 0.48 0.89 0.48 0.27
1.3 (0.48)
0.26
Beta-adrenergic
receptors
and meditation
31
3s 30
PercentRh
E, 20 15 10 5 0 M
FIG. I.-The percent of beta-adrenergic receptors in the high affinity state (Rh). The TM group had a lower Rh than the controls indicating a reduced percent of functional receptors (p = 0.009). by a nonlinear curve-fitting procedure using a generalized model for complex ligand receptor systems (The Ligand Program, NIH). The quality of the fit of the data to the one- vs the two-affinity state model was compared by F-test and the fit to the two-affinity state model was deemed signiticanly better when p < 0.05. For the two-affinity state model, the computer estimated the proportion of receptors in the high (Rh) and low (Rl) affinity state. Plasma samples were analyzed for norepinephrine and epinephrine by high performance liquid chromatography with electrochemical detection, with an interassay coefficient of < 10% [19]. Separate one-way ANOVAs using BMDP statistical software were used to compare groups for the dependent (receptors and catecholamines) and independent variables (Type A behavior, the components, exercise, and family history). RESULTS
There were no significant differences between groups for Type A behavior, the components, family history, or exercise patterns, see Table I. Therefore, these variables were not controlled for in the receptor analyses. Compared to controls, the TM group had a lower Rh (F( 1,18) = 8.3, p = 0.009), indicating a reduced percentage of receptors in the functional high affinity state, see Figure 1. There were no differences in total receptor number, the dissociation constant, Kd, or plasma catecholamines between the groups, see Table II. DISCUSSION
Receptor number, affinity, and sensitivity are traditionally evaluated in receptor research. By directly measuring lymphocyte beta-adrenergic receptors, we have found some support for the hypothesis that TM is associated with reduced end organ receptor sensitivity. A variety of diverse mechanisms mediate adrenergic receptor sensitivity and may account for this finding [20,21]. For example, both human and animal studies indicate that corticosteroids are important in facilitating the formation of high affinity state beta-adrenergic receptors [22,23,24]. Several studies have TABLE
II.-BETA-ADRENERGIC
RECEPTOR AND (MEAN,SD)
TM B,,, (fmol/mg protein) Kd (pmol) Norepinephrine (pg/ml) Epinephrine (pg/ml)
186.7 (116) 3.46 (2) 316 (124) 47 (25)
~ATEC~IOLAMINE DATA
Control 128.5 (51) 3.23 (1) 372 (300) 71 (104)
P
0.16 0.74 0.55 0.54
PAUL J. MILLS ef al.
32
found lower cortisol in subjects practicing TM [25,26]. Although we do not have cortisol data on this particular group of TM subjects, perhaps lower cortisol levels in the TM group contributed to the lower percentage of high affinity state receptors. Changes in adrenergic receptor function may also help explain reported findings of lower blood pressure in TM subjects [27,28]. We found no difference in receptor number of plasma catecholamine levels between the TM and controls subjects. Beta-adrenergic receptor number is strongly influenced by plasma catecholamine levels [21]. Higher levels are associated with a reduction or ‘down regulation’ of receptor number and lower levels are associated with an increase or ‘up-regulation’ of receptor number. The similarity in plasma catecholamines between the groups may help explain the similarity in receptor number, and suggests that factors other than adrenergic activation are responsible for the change in receptor affinity. The dissociation constant for the radioligand, Kd, was nearly identical between the groups. From a pharmacological perspective this is what one would expect because this binding characteristic is quite stable. This study employed a cross-sectional design using long-term TM subjects. The question then arises as to the cause of the receptor differences. Is it self-selection bias, the meditation technique itself, possible changes in lifestyle, or a combination of these? Given the limitations of a cross-sectional study, it is difficult to adequately evaluate these questions with the present data. We attempted to address some of these issues by matching for sex and age and by measuring a number of pertinent independent variables. We found no differences between the two groups on these independent variables. Whether a more thorough evaluation of other personality or lifestyle variables would uncover other factors influencing the receptors remains to be determined. These questions can only be fully answered using a prospective design. Considering the results of both this and other studies [4, S] which have examined the hypothesis that behavioral techniques influence receptor physiology, a prospective study seems warranted to further delineate the intriguing effects of these techniques on receptor function. The results also point to the potentially useful contribution of receptor measurement in psychophysiology research [29]. .4cknoM/~dg~menr.s-The authors would like to thank Dr Gordon Moss and the Behavioral Medicine Group at Eastern Michigan University for scoring the Structured Interviews. We would also like to thank Dr Marcia M. Ward for helpful advice, Dr Ivan Mefford for performing the catecholamine analyses, and Mary Chirico, RN for technical assistance. This study was supported by a grant from the Davis Foundation.
REFERENCES JA(.ORRG, CHESNEYMA. Psychological
In Hundhook
and behavioral
methods
to reduce cardiovascular reactivity. MATTHEWSKA, WEISS SM, SB, WILLIAMSRB), pp. 417457. Wiley, New York.
of Stress, Reacririty, und Cardiot~ascular Disease. (Edited by
DETKET, DEMBKOSKI TM. FALKNERB. MANXK
1986. LAY
