Sleep, 18(1):39-42 © 1995 American Sleep Disorders Association and Sleep Research Society

Sleep Disordered Breathing

132-Adrenergic Receptor Characteristics in Sleep Apnea Patients

*Paul J. Mills, *Joel E. Dimsdale, *Timothy V. Coy,

*tSonia Ancoli-Israel, tJack L. Clausen and *Richard A. Nelesen Downloaded from https://academic.oup.com/sleep/article/18/1/39/2749623 by guest on 01 October 2021

*Department of Psychiatry, University of California, San Diego, La Jolla, California, U.S.A.; tVeterans Affairs Medical Center, San Diego, California, U.S.A.; and tDepartment of Medicine, University of California, San Diego, La Jolla, California, U.S.A.

Summary: This study examined the relationship between sleep apnea and i3,-adrenergic receptor characteristics. Using standard polysomnography, individuals were classified as either apneic (n = 15) or mild to nonapneic (n = 15) according to their respiratory disturbance index (RDI). Subjects were similar in terms of sodium excretion and blood pressure. Apneic subjects showed a decrease in i3,-adrenergic receptor sensitivity (p = 0.01) [as determined by isoproterenol-stimulated cyclic adenosine 5' -monophosphate (AMP) production in lymphocytes] and an increased binding affinity to the i3 receptor antagonist ['25I]iodopindolol (p < 0.001). i3 receptor density was also diminished in apneics, but not significantly (p = 0.08). Forskolin-stimulated cyclic AMP was not significantly different between the groups, indicating a similarity in postreceptor G,-adenylate cyclase activation. Across all subjects, RDI was negatively correlated with i3 receptor sensitivity (r = -0.35, p = 0.05) and Kd (r = -0.54, p < 0.01) and positively correlated with systolic blood pressure (r = 0.37, p < 0.05). The findings indicate that sleep apnea is associated with a diminished i3,-adrenergic receptor function but no change in postreceptor components and suggest a mech­ anism for the high comorbidity between sleep apnea and hypertension. Key Words: Sleep apnea-i3-Adrenergic receptor-RDI - Lymphocytes.

Acute episodes of sleep apnea are associated with amined the effect of sleep apnea on lymphocyte f3r sympathetic nervous system activation as assessed by adrenergic receptors. The lymphocyte serves as a gen­ firing of sympathetic nerves and plasma catechol­ eral model for human f32 receptors because lymphocyte amines (1,2). Even when awake, patients with sleep f32 receptor density and sensitivity are similar to f3 re­ apnea frequently show an elevation in plasma norep­ ceptors on other tissue such as heart and lung (9-11). inephrine (3-5), although this is not universally seen The measurement oflymphocyte f32-adrenergic recep­ (6). Few if any studies have examined the effects of tors has benefited several areas of clinical research, sleep apnea on end-organ receptors. including psychiatric (12,13), pulmonary (14) and car­ Norepinephrine exerts its biologic effects through diovascular (11). Such measurement has been useful activation of adrenergic receptors. These recognition both in terms of the understanding of mechanisms as sites are in turn regulated by exposure to adrenergic well as the evaluation of treatment effects (11,15). agonists such as norepinephrine; short-term exposure to high concentrations leads to a temporary desensi­ tization, an uncoupling of the receptor from the postre­ METHODS ceptor guanine nucleotide coupling protein (Gs) and Subjects adenylate cyclase; long-term exposure can lead to a downregulation (loss of receptor number) (7,8). Subjects were recruited through advertisement and Given the growing literature demonstrating an effect referral from sleep-disorder clinics. Thirty males were of apnea on sympathetic physiology, this study ex- studied after obtaining written informed consent [mean age 48 years (SD = 7)]. Twenty were normotensive and Accepted for publication August 1994. 10 were classified as mildly hypertensive (BP > 140 Address correspondence and reprint requests to Paul J. Mills, Ph.D., UCSD Medical Center, 200 West Arbor Dr., San Diego, CA mm Hg systolic or >90 mm Hg diastolic). Three of 92103-0804, U.S.A. the 10 hypertensives had been on antihypertensive 39 ...... ------

40 P. J. MILLS ET AL.

therapy at the time of recruitment (one on a (3 blocker, TABLE 1. one on a diuretic and one on a calcium channel blocker) Nonapneics Apneics and were tapered off antihypertensive medication and (n = 15) (n = 15) were medication free for at least 2 weeks prior to study. No. hypertensives 4 6 Given the effects of weight on sympathetic physiology, Age 48 (7) 50 (7) only subjects whose weight was between 1.0 and 1.5 Ideal weight 1.2 (0.18) 1.2 (0.13) Systolic BP (mm Hg) 131 (12) 140 (17) times ideal body weight (16) were studied. The pro­ Diastolic BP (mm Hg) 86 (10) 88 (9) tocol was approved by the University of California, 24-hour Na+ excretion San Diego (UCSD), Institutional Review Board. All (mEq) 168 (34) 165 (36) RDI** 9.5 (5.5) 59 (23) subjects were studied at the UCSD Clinical Research (range 2-19) (range 27-103) Center. {3 receptor sensitivitya* 4.8 (1.6) 3.6 (0.9) Forskolin stimulation" 5.1 (1.5) 4.9 (1.9) Downloaded from https://academic.oup.com/sleep/article/18/1/39/2749623 by guest on 01 October 2021 (3 receptor Bmaxc 226 (150) 132 (102) Procedure {3 receptor Kd (picoM)** 74 (16) 55 (19) Sleep polysomnography included central and occip­ Values expressed as means with SD in parentheses. ital electroencephalogram, bilateral electrooculogram, a Ratio of isoproterenol-stimulated cyclic AMP to basal, nonstim­ ulated cyclic AMP. submental and tibialis electromyogram and electro­ " Ratio offorsko1in-stimulated cyclic AMP to basal, nonstimulated cardiogram. Respiration was assessed with nasal/oral cyclic AMP. airflow, abdominal and thoracic respiratory effort, and c fmoles/mg protein. * p < 0.02. oximetry. Sleep was scored according to standard cri­ ** p < 0.001. teria (17). The respiratory disturbance index (RDI) was quantified as the average number of hypopneas plus ceptors; a low Kd means a high binding affinity to the apneas per hour of sleep. {3 receptor and vice versa. Immediately following the night's sleep, before the Both analysis of variance and correlation analyses subject was ambulatory, 60 ml of blood was drawn for were used to examine the relationship between the {3 receptor studies via an indwelling heplock placed the receptors and RDI. previous evening. The lymphocyte isolation and {32- adrenergic receptor assays were conducted according to previously published methods (18). Lymphocytes RESULTS were isolated from whole blood according to the meth­ Across all subjects, RDI ranged from 2 to 103, with od of Boyum (19), using a density gradient medium a mean of33 (SO = 30). Individuals with an RDI ~20 (Ficoll-Paque, Pharmacia, Piscataway, NJ, U.S.A.). were classified as apneic; those with an RDI <20 were Contaminating leukocytes and platelets were removed classified as mild to nonapneic. by centrifugation. Lymphocyte {32-adrenergic receptor Subjects were similar in age, weight, sodium excre­ sensitivity was determined in whole cells by quanti­ tion and diastolic blood pressure; apneics showed a fying cyclic adenosine 5' -mono phosphate (AMP) ac­ non significantly higher systolic blood pressure (p = cumulation following incubation with 10 /LM isopro­ 0.07) (Table 1). High apneic subjects showed a de­ terenol. Cyclic AMP accumulation was assessed in creased {3 receptor sensitivity (p = 0.0 18) and decreased lymphocytes incubated with forskolin (20 ILM) in order Kd (p < 0.001). High apneic subjects also showed a to evaluate postreceptor Gs-adenylate cyclase activa­ non significantly lower {3 receptor density (p = 0.080). tion. Basal, nonstimulated, intracellular cyclic AMP Covariate analyses indicated that these receptor find­ was also determined. {3 receptor sensitivity was defined ings were independent of the marginal difference in as the ratio of stimulated to basal cyclic AMP (maximal blood pressure. Forskolin-stimulated cyclic AMP was isoproterenol stimulation typically causes a three- to not significantly different between the groups (Table five-fold increase in intracellular cyclic AMP levels in 1). lymphocytes). Lymphocyte {32 -adrenergic receptor Across all subjects, RDI was negatively correlated density was determined using radioligand binding with with {3 receptor sensitivity (r = -0.35, p = 0.05) and [I25I]iodopindolol at six concentrations from 10 to 320 Kd (r = -0.54, p < 0.01) and positively correlated picoM (18). Specific binding was determined by using with systolic blood pressure (r = 0.37, p < 0.05). 10-6 M propranolol. {3-adrenergic receptor density (Bmax) and the dissociation constant (Kd) were cal­ DISCUSSION culated using a nonlinear regression receptor binding program (GraphPad Software, San Diego, CA, U.S.A.). This study examined the relationship between sleep Kd is the concentration of the {3 antagonist apnea and IJ-adrenergic receptor characteristics. IJ re­ (125I]iodopindolol needed to occupy one-half of the re- ceptor function was evaluated from blood drawn im- Sleep. Vol. 18. No. I. 1995 {3 RECEPTORS IN SLEEP APNEA 41 mediately following the night's sleep upon which the apnea could lead to a reduction in {32-mediated vaso­ sleep apnea classifications were made. This provided dilation. Similarly, there are instances where {3- and the opportunity to determine the acute effects of apneic a-adrenergic receptors are regulated in opposite direc­ episodes on {3 receptors. The findings suggest that acute tion by factors such as temperature and hormonal ma­ apneic events are associated with a {3 receptor desen­ nipulations such that {3 receptors decrease and a re­ sitization, an approximate 25% reduction in sensitiv­ ceptors increase in responsiveness (and vice versa, ity. Forskolin-stimulated cyclic AMP was similar be­ depending on the regulating factors and organ systems tween the groups, suggesting that the isoproterenol­ under question) (25,26). If apnea is associated with a stimulated deficit in cyclic AMP is indeed in the {3 preferential loss of {32 receptor function, but not a cor­ receptor itself and not in postreceptor Gs and adenylate responding change in a receptor function, this could cyclase activation. Future studies should examine lead to an imbalance in terms of an unopposed pe­

whether {3 receptors regain their sensitivity at a point ripheral a-adrenergic-mediated vasoconstriction. Thus, Downloaded from https://academic.oup.com/sleep/article/18/1/39/2749623 by guest on 01 October 2021 in time farther from the acute apnea events or whether a selective loss in {32-adrenergic receptors may be a this is indeed a trait associated with sleep apnea. potential mechanism for the high comorbidity seen This study attempted to control for possible con­ between sleep apnea and hypertension (27). founding factors, such as sodium and antihypertensive medication, both of which can affect adrenergic recep­ Acknowledgements: This work was supported by grants tor physiology (20-22). Twenty-four-hour sodium ex­ MOlRR-00827, HL3600S, HL40l02, AG027ll and cretionwas similar between the two groups. Given that HL47074 from the National Institutes of Health. antihypertensive medication can alter (3-adrenergic re­ ceptors (10,18,22), hypertensive subjects were off an­ REFERENCES tihypertensive medication for at least 2 weeks. Al­ though it is possible that there were still residual effects 1. Eisenberg E, Zimlichman R, Lavie P. Plasma norepinephrine levels in patients with sleep apnea syndrome. New Engl J Med of the medications on the {3 receptors, a repeat of the 1990;332:932-3. analyses excluding the three subjects previously on an­ 2. Watanabe T, Mano T, Iwase S, et al. Enhanced muscle sym­ tihypertensive medication yielded essentially identical pathetic nerve activity during sleep apnea in the elderly. J Auto Nerv System 1992;37:223-6. results. 3. Marrone 0, Ricobono L, Salvaggio A, Mirabella A, Bonanno Physiologically, these data are compatible with pre­ A, Bonsignore MR. Catecholamines and blood pressure in ob­ vious studies showing sympathetic nervous system ac­ structive sleep apnea syndrome. Chest 1993;103:722-7. 4. Fletcher EC, Miller J, Schaaf JW, Fletcher JG. Urinary cate­ tivation accompanying apneic events (1-5). Acute ap­ cholamines before and after tracheostomy in patients with ob­ neic-associated increases in norepinephrine, for structive sleep apnea and hypertension. Sleep 1987;10:35-44. example, could lead to an immediate receptor un­ 5. Carlson J, Hedner J, Elam M, Ejnell H, Wallin G. Augmented sympathetic activity in awake patients with obstructive sleep coupling and, in the long term, an eventual receptor apnea. Sleep Res 1991 ;20A:297. downregulation. In the high apneic subjects, {3 recep­ 6. Levinson P, Randall C, Goldstein D, Carlisle C, Losiphidis A, tors were desensitized; {3 density was lower, but non­ Millman R. Catecholamine and hormone levels in obstructive sleep apnea. Am J Hyper 1990;3(pt. 2):26A. significantly (this may be due in part to the greater 7. Insel P. Adrenergic receptor in man. New York: Marcell Dekker, variance seen in this particular measure). The high 1987. apneic subjects also showed an increased binding af­ 8. Mills PJ, Dimsdale JE. The promise of adrenergic receptor stud­ ies in psychophysiologic research II: applications, limitations, finity (lower Kd) to the {3 receptor antagonist iodopin­ and progress. Psychosom Med 1993;55:448-57. dolol. It is tempting to speculate that the increased 9. Brodde 0, Kretsch R, Ikezono K, Zerkowski H, Reidmeister binding affinity may reflect an attempt to compensate Je. Human j3-adrenoceptors: relation of myocardial and lym­ phocyte j3-adrenoceptor density. Science 1986;231: 1584-5. for the reduced receptor sensitivity and/or density. 10. Aarons R, Nies A, Gerber J, MolinoffP. Decreased j3 adrenergic Studies demonstrate that receptor affinity for antago­ receptor density on human lymphocytes after chronic treatment nist (as with agonist) is highly variable and fluid, being with agonists. J Pharmacol Exper They 1983;224: 1-6. II. Hom E, Corwin S, Steinberg S, et al. Reduced lymphocyte stim­ determined by factors such as the receptors' micro­ ulatory guanine nucleotide regulatory protein and j3-adrenergic environment and intracellular processes (23, for re­ receptors in congestive heart failure and reversal with angioten­ view). sin converting enzyme inhibitor therapy. Circulation 1988;78: 1373-9. Respiratory disturbance index was also related to 12. Mann J, Brown R, Halper J, et al. Reduced sensitivity oflym­ the subject's systolic blood pressure. It would be of phocytc j3-adrenergic receptors in patients with endogenous de­ interest to examine the effects of sleep apnea on {31 and pression and psychomotor agitation. New Engl J Med 1985;313: 715-20. on a-adrenergic receptors. {31- and {32-adrenergic re­ 13. Pandey G, Janicak P, Davis J. Decreased j3-adrenergic receptors ceptors are regulated differently; {32 receptors are more in leukocytes ofdepressed patients. Psychiatry Res 1987;22:265- susceptible to desensitization and downregulation (24). 73. 14. lizuka K, Yoshie Y, Nakazawa T. Hormone-sensitive adenylate Given that the predominant {3 receptor subtype in the cyclase system in lymphocytes from asthmatic patients: possible vasculature is {32, a selective loss of {32 receptors in sleep defects at the postreceptor sites. Ann Allergy 1991;66:167-72.

Sleep. Vol. 18. No.1. 1995 42 P. J. MILLS ET AL.

15. Mann J, Mahler J, Wilner P, et al. Normalization of blunted relationship of leukocyte /'I receptor density to the cardiac sen­ lymphocyte /'I-adrenergic responsivity in melancholic inpatients sitivity to isoproterenol. J Clin Invest 1981 ;67: 1777-84. by a course of electroconvulsive therapy. Arch Gen Psychiatry 22. Brodde OE, Wang XL, O'Hara N, Daul A, Schiess W. Effect of 1990;47:461-4. propranolol, alprenolol, pindolol, and bopindolol on /'I2-adreno­ 16. Metropolitan Life Foundation. 1983 Metropolitan height and ceptor density in human lymphocytes. J Cardiovasc Pharmacol weight tables. Stat Bull 1983;64: 1. 1986;8:S70-3. 17. Rechtschatten A, Kales A. A manual 0/ standardized terminol­ 23. Bevan J, Bevan R, Shreeve SM. Variable receptor affinity hy­ ogy and techniques and scoring system/or sleep stages o/human pothesis. FASEB J 1989;3:1696-704. subjects. NIH publication #204. Washington, DC: U.S. Gov­ 24. Muntz KH, Zhao M, Miller Je. Downregulation of myocardial ernment Printing Office, 1968. /'I-adrenergic receptors: receptor subtype selectivity. Circ Res 18. Mills PJ, Dimsdale JE, Ziegler MG, Hauger R, Nelesen RA, 1994;74:360-75. Brown M. Sympathetic alterations following sodium restriction 25. Kunos G, Kunos I, Hirata F, Ishac E. Adrenergic receptors: and short-term captopril. J Am Coil CardioI1993;21: 177-81. possible mechanism of inverse regulation of a- and /'I-receptors. 19. Boyum A. Isolation of mononuclear cells and granulocytes from J Allergy Clin lmmunol 1985;76:346-51. blood. Scand J Clin Invest 1968;21(SuppI97):77-89. 26. Michel M, Mindermann G, Dau1 A, Brodde O. Effects of an­

20. Dimsdale J, Ziegler M, Graham R. The effect of hypertension, tihypertensive therapy on human a- and (J-adrenoceptors. J Hy­ Downloaded from https://academic.oup.com/sleep/article/18/1/39/2749623 by guest on 01 October 2021 sodium, and race on isoproterenol sensitivity. Clin Exper Hypert per 1991 ;9:60 1-6. I 988;AIO(5):747-56. 27. Jeong D, Dimsdale J. Sleep apnea and essential hypertension: 21. Fraser J, Nadeau J, Robertson D, Wood AJJ. Regulation of a critical review of the epidemiological evidence for co-mor­ human leukocyte (J receptors by endogenous catecholamines: bidity. Clin Exper Hyper 1989;A11:1301-23.

Sleep. Vol. 18. No.1. 1995