Modulation of Blood Pressure by Central Melanocortinergic Pathways

Transcription

1 The new england journal of medicine original article Modulation of Blood Pressure by Central Melanocortinergic Pathways Jerry R. Greenfield, M.D., Ph.D., Jeffrey W. Miller, M.D., Julia M. Keogh, B.Sc., Elana Henning, B.Soc.Sc., Julie H. Satterwhite, Ph.D., Gregory S. Cameron, Ph.D., Beatrice Astruc, M.D., John P. Mayer, Ph.D., Soren Brage, Ph.D., Teik Choon See, M.D., David J. Lomas, M.D., Stephen O Rahilly, M.D., and I. Sadaf Farooqi, M.D., Ph.D. Abstract From the University of Cambridge Metabolic Research Laboratories (J.R.G., J.M.K., E.H., S.O., I.S.F.) and Medical Research Council Epidemiology Unit (S.B.), the Institute of Metabolic Science and the Department of Radiology (T.C.S., D.J.L.), Addenbrooke s Hospital, Cambridge, United Kingdom; Eli Lilly, Lilly Corporate Center, Indianapolis (J.W.M., J.H.S., G.S.C., J.P.M.); and Biotrial, Rennes, France (B.A.). Address reprint requests to Dr. Farooqi at the Institute of Metabolic Science, Metabolic Research Laboratories, Addenbrooke s Hospital, Hills Rd., Box 289, Cambridge CB2 QQ, United Kingdom, or at isf2@cam.ac.uk. This article (1.156/NEJMoa8385) was published at NEJM.org on December 17, 28. N Engl J Med 29;36: Copyright 29 Massachusetts Medical Society. Background Weight gain and weight loss are associated with changes in blood pressure through unknown mechanisms. Central melanocortinergic signaling is implicated in the control of energy balance and blood pressure in rodents, but there is no information regarding such an association with blood pressure in humans. Methods We assessed blood pressure, heart rate, and urinary catecholamines in overweight or obese subjects with a loss-of-function mutation in MC4R, the gene encoding the melanocortin 4 receptor, and in equally overweight control subjects. We also examined the effects of an MC4R agonist administered for 7 days in 28 overweight or obese volunteers. Results The prevalence of hypertension was markedly lower in the MC4R-deficient subjects than in the control subjects (24% vs. 53%, P =.9). After the exclusion of subjects taking antihypertensive medications, blood-pressure levels were significantly lower in MC4R-deficient subjects than in control subjects, with mean (±SE) systolic blood pressures of 123±14 mm Hg and 131±12 mm Hg, respectively (P =.2), and mean diastolic blood pressures of 73±1 mm Hg and 79±7 mm Hg, respectively (P =.3). As compared with control subjects, MC4R-deficient subjects had a lower increase in heart rate on waking (P =.7), a lower heart rate during euglycemic hyperinsulinemia (P<.1), and lower 24-hour urinary norepinephrine excretion (P =.4). The maximum tolerated daily dose of 1. mg of the MC4R agonist led to significant increases of 9.3±1.9 mm Hg in systolic blood pressure and of 6.6±1.1 mm Hg in diastolic blood pressure (P<.1 for both comparisons) at 24 hours, as compared with placebo. Differences in blood pressure were not explained by changes in insulin levels; there were no significant adverse events. Conclusions Results of our genetic and pharmacologic studies implicate melanocortinergic signaling in the control of human blood pressure through an insulin-independent mechanism. 44 n engl j med 36;1 nejm.org january 1, 29 Downloaded from nejm.org on April 11, 218. For personal use only. No other uses without permission. Copyright 29 Massachusetts Medical Society. All rights reserved.

2 Epidemiologic and physiological studies have consistently demonstrated that obesity is a major cause of hypertension. 1,2 Studies of rodents with diet-induced obesity suggest that increased activation of the sympathetic nervous system is an important mediator of obesity-induced hypertension: alpha- and beta-adrenergic receptor antagonists and renal denervation significantly blunt the rise in arterial pressure associated with weight gain. 3-5 However, the physiological mechanisms linking the development of obesity and hypertension are unclear. The hypothalamic leptin melanocortin pathway is critically involved in the control of energy balance, and genetic disruption of molecules in this pathway leads to severe obesity in rodents and humans. 6 A key component of this pathway is the melanocortin system, which includes leptinresponsive neurons expressing neuropeptide Y and agouti-related-protein and those expressing proopiomelanocortin, which is cleaved to produce the melanocyte-stimulating hormones (MSH) (alpha-, beta-, and gamma-msh). 7 Downstream targets of these neurons express the melanocortin 3 receptor (MC3R) and the melanocortin 4 receptor (MC4R). Melanocortins are agonists of MC3R and MC4R, whereas agouti-related-protein is a high-affinity antagonist. 8 Studies in rodents suggest that the melanocortin system is important in cardiovascular regulation. 9 Acute central administration of alpha-msh increases mean arterial pressure and heart rate. 1 Pharmacologic studies indicate that this effect is attributable to activation of the sympathetic nervous system. 11 Chronic pharmacologic blockade of MC3R and MC4R causes weight gain and a reduction in heart rate but no increase in arterial pressure. 12 These pressor and depressor effects are mediated predominantly by signaling through MC4R-expressing neuronal pathways, because Mc4rnull mice maintain a normal blood pressure despite marked obesity and are unresponsive to the pressor effects of central alpha-msh administration. 1 MC4R deficiency is the most common form of inherited human obesity, with a mutation prevalence of approximately 6% in subjects with severe early-onset obesity, % in unselected obese adults, 14,15 and.1% in a population-based study of unselected subjects. 16 We have previously described the clinical characteristics of human MC4R deficiency. 13 Many of these phenotypes are closely replicated in the Mc4r-deficient mouse, suggesting that it is an excellent model for the human disease. 17 To examine the effects of loss of function of the melanocortin pathway on blood pressure in humans, we studied adults who were heterozygous for complete loss-of-function mutations in MC4R (i.e., with haploinsufficiency) and equally overweight or obese control subjects who did not have MC4R mutations. To examine the effects of increased signaling through melanocortin receptors, we studied the short-term effects of a melanocortin receptor agonist in overweight or obese volunteers. Methods Physiologic Studies in MC4R-Deficient Subjects We analyzed blood pressure and metabolic data for 46 MC4R-deficient adults from 32 nuclear families identified as part of the Genetics of Obesity Study. 13 All subjects were heterozygous for complete loss-of-function mutations in MC4R (for details, see Table 1 in the Supplementary Appendix, available with the full text of this article at NEJM. org). We recruited 3 adult volunteers whom we determined to be overweight, which was defined as having a body-mass index (BMI, the weight in kilograms divided by the square of the height in meters) of more than 25, or obese (BMI, >3) and who were shown to have a normal MC4R genetic sequence. All subjects provided written informed consent. Studies were conducted at the Wellcome Trust Clinical Research Facility at Addenbrooke s Hospital in accordance with the principles of the Declaration of Helsinki after approval by the ethics committee at Addenbrooke s Hospital. Eight MC4Rdeficient adults of European descent and eight control subjects who were matched in age, ancestral background, and BMI agreed to undergo more detailed studies (Table 2 in the Supplementary Appendix). European descent was determined by the reported country of birth of at least three generations of family members. Details of measurement of blood pressure, autonomic nervous system activation, body composition, measurement of abdominal and liver fat by magnetic resonance imaging, hyperinsulinemic euglycemic and hyperglycemic clamp studies, along with analytical methods, definitions, and calculations, are pro- n engl j med 36;1 nejm.org january 1, Downloaded from nejm.org on April 11, 218. For personal use only. No other uses without permission. Copyright 29 Massachusetts Medical Society. All rights reserved.

3 The new england journal of medicine Table 1. Characteristics of the Subjects.* Characteristic MC4R-Deficient Subjects (N = 46) Control Subjects (N = 3) Sex (M/F) 17/29 25/5 Age (yr) 34.4± ±7.9 Body-mass index 38.9± ±5.1 Fasting plasma glucose (mmol/liter) 5.±.5 5.±.7 Fasting plasma insulin (pmol/liter) Geometric mean SE range Fasting total cholesterol (mmol/liter) 4.9±.8 5.±1 Fasting triglycerides (mmol/liter) Geometric mean SE range * Plus minus values are means ±SD. Measures of fasting plasma glucose and insulin were excluded for eight subjects with MC4R deficiency and six control subjects who had diabetes. Measures of fasting total cholesterol and triglycerides were excluded for two subjects with MC4R deficiency and one control subject who were receiving lipid-lowering therapy. The body-mass index is the weight in kilograms divided by the square of the height in meters. To convert the values for glucose to milligrams per deciliter, divide by To convert the values for insulin to microunits per milliliter, divide by 6.. To convert the values for cholesterol to milligrams per deciliter, divide by To convert the values for triglycerides to milligrams per deciliter, divide by vided in the Methods section in the Supplementary Appendix. Pharmacologic Study in Overweight or Obese Volunteers We conducted a double-blind, dose-escalating, crossover trial of an MC4R agonist, LY (Eli Lilly), as compared with placebo, in 28 healthy men and women who were overweight or obese and whose weight had been stable during the previous 6 months (Table 3 in the Supplementary Appendix). The MC4R agonist was administered in two doses (.45 mg and 1. mg) as a subcutaneous infusion over a 24-hour period. After each 24-hour infusion of either the MC4R agonist or placebo, there was a washout interval of 3 to 6 days before the next round of infusions was started in the same subjects. Seven-day infusions of the MC4R agonist at one dose or infusions of placebo were also administered with a washout interval of 7 days in a crossover design. Medications that were known to alter appetite, blood pressure, and autonomic function were prohibited. Studies were conducted in the Clinical Research Unit at Biotrial in Rennes, France, under Good Clinical Practice guidelines as part of a program for the development of drugs targeting obesity. The study protocol was approved by the ethics review board (Comité de Protection des Personnes de Brest, France). Details regarding the study design, methods for measurement of blood pressure and biochemical values, statistical analysis, and a synopsis of the protocol are provided in the Methods section in the Supplementary Appendix. Results Blood Pressure and MC4R Deficiency We studied the blood pressure and metabolic phenotype in 46 adults with MC4R haploinsufficiency and in 3 control subjects with a normal MC4R genotype (Table 1). The prevalence of hypertension was significantly lower in the MC4R-deficient group than in the control group (Fig. 1A). After the exclusion of subjects receiving antihypertensive medications, the mean systolic and diastolic bloodpressure levels were lower in MC4R-deficient adults than in control subjects (Fig. 1B and 1C). The prevalence of diabetes was similar in the MC4R-deficient group and the control group (18% and 2%, respectively). After the exclusion of subjects with diabetes, fasting plasma glucose levels were similar in the two groups (Table 1). Consistent with our previous observation that the relative hyperinsulinemia associated with childhood MC4R deficiency becomes less marked with age, 13 there was no significant difference in fasting plasma insulin levels between the two groups (Table 1). Autonomic Nervous System Activity To examine autonomic nervous system function, we measured variability in heart rate in MC4R-deficient subjects and in control subjects (Table 2 in the Supplementary Appendix). There was no difference in sleeping heart rate between the two study groups (Fig. 2A). On waking, the heart rate increased in the two groups; however, the magnitude of this increase was significantly attenuated in MC4R-deficient subjects (P =.7) (Fig. 2A). Infusion of insulin increased the heart rate in the two groups. Although the magnitude of the increase was similar, MC4R-deficient subjects maintained significantly lower heart rates throughout the clamp procedure, with a difference of 7 bpm between the two groups during the steady-state period (P<.1) (Fig. 2A). 46 n engl j med 36;1 nejm.org january 1, 29 Downloaded from nejm.org on April 11, 218. For personal use only. No other uses without permission. Copyright 29 Massachusetts Medical Society. All rights reserved.

4 During sleep, there was no significant difference in the high-frequency component of heart rate variability (predominantly a measure of parasympathetic activation) between the two study groups (Fig. 2B). On waking, the high-frequency component decreased in control subjects, a drop that was attenuated in MC4R-deficient subjects (Fig. 2B). After the infusion of insulin, the highfrequency component decreased in the two groups but remained higher in MC4R-deficient subjects throughout the clamp procedure (P =.2), consistent with greater parasympathetic activity and relatively lower sympathetic activity than in control subjects (Fig. 2B). The low-frequency component of heart rate variability, a composite measure of sympathetic and parasympathetic activity, was similar in the two groups throughout the study (data not shown). The root mean square of successive differences (RMSSD) between adjacent normal RR intervals, an index of parasympathetic activity, was similar in MC4R-deficient subjects and control subjects during sleep (Fig. 2C). The RMSSD decreased in response to waking in control subjects, but there was no change in MC4R-deficient subjects. Infusion of insulin led to a reduction in the RMSSD values in both groups, although the values were higher in MC4R-deficient subjects throughout the hyperinsulinemic euglycemic clamp procedure (P =.2) (Fig. 2C). Urinary Catecholamine Excretion We measured 24-hour urinary catecholamine excretion in 1 adults with haploinsufficiency for MC4R with a mean (±SD) age of 33.8±13.6 years and a mean BMI of 43±7.5 and in 19 control subjects with a mean age of 39.8±8.5 years and a mean BMI of 37.2±3.8. MC4R-deficient subjects had lower 24-hour norepinephrine excretion than did the control subjects (199±23 nmol and 291±28 nmol, respectively; P<.5) (Fig. 3A). Although 24-hour urinary epinephrine excretion also tended to be lower in MC4R-deficient subjects, the difference was not significant (Fig. 3B). There was no significant difference in 24-hour dopamine excretion between the two groups (Fig. 3C). Nonparametric tests gave the same results for these comparisons. A Hypertension Prevalence (%) B Systolic Blood Pressure Blood Pressure (mm Hg) C Diastolic Blood Pressure Blood Pressure (mm Hg) P=.9 P<.5 P<.5 Figure 1. Prevalence of Hypertension and Blood- Pressure Measures. The graphs show the prevalence of hypertension (Panel A) and measures of systolic blood pressure (Panel B) and diastolic blood pressure (Panel C) in 46 subjects with MC4R deficiency, as compared with 3 overweight or obese control subjects. Four subjects with MC4R deficiency and seven control subjects were excluded from the analyses in Panels B and C because they were taking antihypertensive medications. The I bars represent the standard error. Other Phenotypic Signatures We measured body composition, fat distribution, insulin secretion, and insulin sensitivity to ascertain whether there might be other phenotypic signatures associated with MC4R deficiency. MC4Rdef icient subjects and control subjects had similar percentages of total body fat and fat-free mass, intraabdominal adipose tissue area, subcutaneous n engl j med 36;1 nejm.org january 1, Downloaded from nejm.org on April 11, 218. For personal use only. No other uses without permission. Copyright 29 Massachusetts Medical Society. All rights reserved.

5 The new england journal of medicine A Heart Rate Heart Rate (beats/min) Asleep Baseline Awake MC4R deficient B High-Frequency Component High-Frequency Component (log 1 msec 2 ) C RMSSD RMSSD (log 1 msec) P=.7 MC4R deficient Asleep Baseline Awake MC4R deficient Asleep Baseline Awake P< >3 45 >45 6 >6 75 >75 9 >9 15 Minutes since Start of Clamp Procedure P= >3 45 >45 6 >6 75 >75 9 >9 15 Minutes since Start of Clamp Procedure P= >3 45 >45 6 >6 75 >75 9 >9 15 Minutes since Start of Clamp Procedure Figure 2. Heart Rate and Its High-Frequency Component and RMSSD. The graphs show the heart rate (Panel A) and its high-frequency component (predominantly a measure of parasympathetic activation) (Panel B) and the root mean square of successive differences (RMSSD) between adjacent normal RR intervals (Panel C) during sleep, in the awake state, and during the hyperinsulinemic euglycemic clamp procedure in eight subjects with MC4R deficiency and eight control subjects. The I bars represent the standard error. abdominal adipose tissue area, and liver fat (Table 4 in the Supplementary Appendix). During the hyperinsulinemic euglycemic clamp procedure in MC4R-deficient subjects and control subjects, there were similar mean (±SE) steadystate levels of blood glucose (5.± mmol per liter in the two groups) and geometric mean steadystate levels of plasma insulin (138 pmol per liter [1-SE range, 1231 to 139] and 1299 pmol per liter [1-SE range, 1217 to 1386], respectively). During steady state, there was no significant difference between the two groups in insulin-mediated glucose uptake, with similar rates of insulin-stimulated oxidative and nonoxidative glucose disposal (Fig. 4A). Posthepatic insulin clearance rates were also similar (data not shown). Reduced activity in the sympathetic nervous system would be expected to increase insulin secretion in the MC4R-deficient subjects. Although there was a trend in this direction, there was no significant difference between MC4R-deficient subjects and control subjects in geometric mean levels of first-phase insulin secretion (614 pmol per liter [1-SE range, 471 to 81] and 474 pmol per liter [1-SE range, 368 to 612], respectively) or second-phase insulin secretion (45 pmol per liter [1-SE range, 332 to 494] and 332 pmol per liter [1-SE range, 272 to 45], respectively) during the intravenous glucose tolerance test (Fig. 4B). Results were similar when insulin responses were compared with the use of areaunder-the-curve analysis and repeated-measures analysis of variance (data not shown). Geometric mean levels of second-phase insulin secretion that was measured during the hyperglycemic clamp procedure were also similar in the two groups (944 pmol per liter [1-SE range, 76 to 1172] and 729 pmol per liter [1-SE range, 575 to 924], respectively). There was no significant difference in the disposition index (the product of insulin sensitivity and first-phase insulin secretion) between the two groups. Nonparametric tests provided the same results for these comparisons. Effects of MC4R Agonist Of the 28 healthy overweight or obese subjects who were enrolled, 26 completed the study (Table 3 in the Supplementary Appendix). Steady-state plasma levels of drugs were achieved within 24 hours and were proportional to the dose. Infusion of the MC4R agonist LY resulted in symptoms 48 n engl j med 36;1 nejm.org january 1, 29 Downloaded from nejm.org on April 11, 218. For personal use only. No other uses without permission. Copyright 29 Massachusetts Medical Society. All rights reserved.

6 known to be mediated by melanocortin pathways, such as yawning, a sensation of muscular stiffness, and increased penile erection. The need to stretch or yawn was significantly increased at all doses during the 7-day infusion. A significant increase in male erectile function was observed at the.45-mg and 1.-mg daily doses only. Adverse effects, including headache, asthenia, nausea, vomiting, and diarrhea, are presented in Table 5 in the Supplementary Appendix. There was a dose-dependent increase in blood pressure after infusion of LY in overweight or obese adults during a 24-hour period (Fig. 5). Three hours after the start of infusion, the average systolic blood pressure was 4. to 8.5 mm Hg higher than placebo-controlled values (Fig. 5A). Maximum increases from the placebo values were generally observed at 24 hours. The average difference in systolic blood pressure at 24 hours was 9.3±1.9 mm Hg for the maximum tolerated dose of 1. mg per day (P<.1). Diastolic blood pressure 1 hour after the start of infusion was 2.5 to 8.2 mm Hg higher than placebo-controlled values for the dose range of.15 to 2. mg (Fig. 5B). The average difference in diastolic blood pressure at 24 hours was 6.6±1.1 mm Hg for the maximum tolerated dose of 1. mg per day (P<.1). Blood-pressure changes were sustained for the 7 days of LY infusion (Table 6 in the Supplementary Appendix). A decrease in mean heart rate of approximately 5 to 6 bpm was observed 1 hour after infusion in all dose groups, with the exception of the.45-mg dose. However, the average heart rate increased during the 7-day infusions, with an average difference of 3 bpm, as compared with placebo (Table 6 in the Supplementary Appendix). There were no significant differences in levels of plasma insulin and glucose and urinary norepinephrine after LY administration (Table 7 in the Supplementary Appendix). Discussion A Norepinephrine Norepinephrine (nmol/24 hr) B Epinephrine Epinephrine (nmol/24 hr) C Dopamine Dopamine (nmol/24 hr) P<.5 Figure 3. Twenty-four-Hour Levels of Urinary Norepinephrine, Epinephrine, and Dopamine. The graphs show levels of urinary norepinephrine (Panel A), epinephrine (Panel B), and dopamine (Panel C) in 1 subjects with MC4R deficiency and 19 control subjects. The I bars represent the standard error. Obesity is a major risk factor for hypertension, 1 and weight loss reduces blood pressure. 18 Studies in animals have indicated that induced weight gain results in increased blood pressure. 19 Numerous mechanisms have been suggested, 2 many of which involve the assumption that insulin resistance or hyperinsulinemia that accompanies weight gain is involved through effects on renal sodium retention, vascular structure and function, ion transport, or activity of the sympathetic nervous system. 21 Melanocortinergic signaling is critical to the control of weight in humans 6 and in animals. 8 Studies in rodents have clearly implicated cen- n engl j med 36;1 nejm.org january 1, Downloaded from nejm.org on April 11, 218. For personal use only. No other uses without permission. Copyright 29 Massachusetts Medical Society. All rights reserved.

7 The new england journal of medicine A Insulin Sensitivity 7 B Plasma Insulin 12 Insulin Sensitivity (µmol/kg FFM/min) Plasma Insulin (pmol/liter) MC4R deficient Minutes since Glucose Bolus Figure 4. Insulin Sensitivity with Oxidative and Nonoxidative Glucose Disposal during Hyperinsulinemic Euglycemic Clamp Studies and Insulin Levels during an Intravenous Glucose Tolerance Test. Panel A shows steady-state insulin-mediated glucose uptake, with oxidative glucose disposal (light blue) and nonoxidative glucose disposal (dark blue) during hyperinsulinemic euglycemic clamp studies in eight subjects with MC4R deficiency and eight control subjects. FFM denotes fat-free mass. Panel B shows plasma insulin levels during the intravenous glucose tolerance test in the same subjects. The I bars represent the standard error (SE) in Panel A and the 1-SE range in Panel B. trally expressed MC4R in the control of blood pressure. 9,1 In this study, we combined observations from subjects with reduced MC4R signaling and pharmacologic studies using a centrally active melanocortinergic agonist to examine the relationship between melanocortin signaling and blood pressure in humans. We conclude that both increases and decreases in central melanocortin signaling influence blood pressure in humans and that the effects are not explained by changes in circulating insulin levels or insulin sensitivity. In the case of long-standing decreases in melanocortinergic tone, these changes are associated with reduced activity in the sympathetic nervous system. We studied a clearly defined group of subjects with MC4R haploinsufficiency. Because some of these subjects were related to each other, they may share genetic factors other than mutated MC4R that contribute to blood-pressure regulation. Careful matching of control subjects to MC4R-deficient subjects with respect to BMI, age, and ancestry revealed striking differences both in the prevalence of hypertension and in mean levels of systolic and diastolic blood pressure. The prevalence of hypertension and mean blood pressures seen in the control group were very similar to those of unselected overweight or obese subjects, 2 a finding that supported our assertion that the MC4Rdeficient subjects were relatively protected from the expected relationship between overweight and raised blood pressure. We studied a synthetic peptide agonist that is highly selective for MC4R. The drug, MC4R agonist LY , had an increased affinity for MC4R over MC1R of more than a factor of 3 and over MC3R of more than a factor of 1. Administration of the MC4R agonist to overweight or obese volunteers produced the classic symptoms of yawning, stretching, and penile erection associated with activation of central MC4R and a marked short-term increase in blood pressure that was sustained for the 7 days of the drug s administration. The average increase in systolic and diastolic blood pressure in these studies was quantitatively similar to the difference in blood pressure between MC4R-deficient subjects and control subjects. A deficiency in MC4R (in both children 13 and mice 22 ) appears to be associated with hyperinsulinemia at a rate that is disproportionate to the degree of obesity, and several studies have implicated hyperinsulinemia as a link between obesity and hypertension. 21 Our finding that MC4R deficiency is associated with reduced, and not increased, blood pressure does not support such a link. In addition, we could find no difference in measures of insulin sensitivity and intraabdominal 5 n engl j med 36;1 nejm.org january 1, 29 Downloaded from nejm.org on April 11, 218. For personal use only. No other uses without permission. Copyright 29 Massachusetts Medical Society. All rights reserved.

8 fat between MC4R-deficient subjects and control subjects. Finally, the administration of the MC4R agonist for 7 days did not significantly alter fasting insulin levels, despite having a marked effect on blood pressure. Central control of the sympathetic nervous system can clearly influence blood pressure, 23 and the administration of an MC4R agonist in mice is associated with increased activity of the sympathetic nervous system. 24,25 The rarity of finding subjects with MC4R deficiency without other coexisting illnesses and the complexity of our protocol made it impossible to incorporate direct electrophysiological studies of activity in the sympathetic nervous system. However, surrogate measures support such an effect. Although the sleeping heart rate (under strong vagal influence) was similar in MC4R-deficient subjects and control subjects, the increases in heart rate on waking and during a hyperinsulinemic clamp procedure were significantly lower in the MC4R-deficient subjects than in the control subjects. Also, urinary norepinephrine levels were significantly reduced in the MC4Rdeficient subjects. The urinary norepinephrine level is usually thought to be a relatively crude measure of activity in the sympathetic nervous system and is useful only in large populations, but we believe that it is all the more striking that we observed significant differences in relatively small numbers of subjects, presumably because the effect of MC4R haploinsufficiency in this carefully defined group was both substantial and consistent. This pattern has been observed in persons with recent weight loss, in whom activation of the autonomic nervous system has been more directly quantified. 26 Finally, direct measures of sympathetic nerve activity are unlikely to have provided additional information, because studies involving animals have shown that sympathetic nerve output to the renal area may be most relevant to the control of blood pressure. 27,28 Despite a significant increase in heart rate, we did not observe a significant increase in levels of urinary norepinephrine in the pharmacologic study, although we saw a nonsignificant trend for an increase in this measure. The explanation for this lack of symmetry with the genetic studies is unclear but may imply that discernible effects on this measure require alterations in signaling through MC4R of a longer duration or that the effect of LY on other melanocortin receptors might modify the effects that we observed. A Systolic Blood Pressure Difference from Placebo (mm Hg) Difference from Placebo (mm Hg) LY : B Diastolic Blood Pressure.5 mg (N=8).15 mg (N=18).45 mg (N=17) Hours Hours In conclusion, we have demonstrated that central melanocortinergic tone significantly influences blood pressure in humans. The therapeutic use of a melanocortin agonist in common human obesity may be complicated by the increase in blood pressure, as seen in our short-term studies. However, the effects of administration of a melano- 1. mg (N=24) 2. mg (N=8) Figure 5. Differences in Mean Blood Pressure in Subjects Receiving Melanocortin Agonist LY or Placebo at 24 Hours, According to Dose. The graphs show the differences from placebo in mean systolic blood pressure (Panel A) and diastolic blood pressure (Panel B) in 28 subjects receiving a melanocortin agonist, according to the dose of the melanocortin agonist. The I bars represent 9% confidence intervals. n engl j med 36;1 nejm.org january 1, Downloaded from nejm.org on April 11, 218. For personal use only. No other uses without permission. Copyright 29 Massachusetts Medical Society. All rights reserved.

9 cortin agonist to patients with MC4R haploinsufficiency would be of great interest. Given the effects of MC4R tone on blood pressure, we conclude that the well-established effects of weight loss and obesity on human blood pressure may involve alterations in central signaling at MC4R, mediated through changes in sympathetic neural activity. Supported by grants from the National Health and Medical Research Council of Australia, the Royal Australasian College of Physicians, and St. Vincent s Clinic Foundation, Sydney (to Dr. Greenfield); from the Wellcome Trust and the Cambridge Biomedical Research Centre of the National Institute for Health Research (to Drs. O Rahilly and Farooqi); and from the Medical Research Council (to Drs. O Rahilly, Brage, and Farooqi); and by Eli Lilly for the pharmacologic studies. Drs. Miller, Satterwhite, Cameron, and Mayer report being employed by Eli Lilly and having an equity interest in the company; and Dr. O Rahilly, receiving consulting fees from Eli Lilly. No other potential conflict of interest relevant to this article was reported. We thank Adrian Dixon, Sri Aitken, and the radiography staff at Addenbrooke s Hospital for their assistance with imaging studies; Fiona Tulloch and Keith Burling, who performed the biochemical assays, and David Moller for helpful discussions; the patients and volunteers for their participation; and the physicians involved in the Genetics of Obesity Study. References 1. Dustan HP. Research contributions toward prevention of cardiovascular disease: research related to the underlying mechanisms in hypertension. Circulation 1979;6: Garrison RJ, Kannel WB, Stokes J III, Castelli WP. Incidence and precursors of hypertension in young adults: the Framingham Offspring Study. Prev Med 1987;16: Hall JE. Mechanisms of abnormal renal sodium handling in obesity hypertension. Am J Hypertens 1997;1:49S-55S. 4. Kassab S, Kato T, Wilkins FC, Chen R, Hall JE, Granger JP. Renal denervation attenuates the sodium retention and hypertension associated with obesity. Hypertension 1995;25: Hall JE, Brands MW, Hildebrandt DA, Kuo J, Fitzgerald S. Role of sympathetic nervous system and neuropeptides in obesity hypertension. Braz J Med Biol Res 2;33: Farooqi IS, O Rahilly S. Monogenic obesity in humans. Annu Rev Med 25; 56: Schwartz MW, Woods SC, Porte D Jr, Seeley RJ, Baskin DG. Central nervous system control of food intake. Nature 2; 44: Cone RD. Anatomy and regulation of the central melanocortin system. Nat Neurosci 25;8: Tallam LS, Stec DE, Willis MA, da Silva AA, Hall JE. Melanocortin-4 receptor-deficient mice are not hypertensive or salt-sensitive despite obesity, hyperinsulinemia, and hyperleptinemia. Hypertension 25;46: Ni XP, Butler AA, Cone RD, Humphreys MH. Central receptors mediating the cardiovascular actions of melanocyte stimulating hormones. J Hypertens 26;24: Kuo JJ, da Silva AA, Tallam LS, Hall JE. Role of adrenergic activity in pressor responses to chronic melanocortin receptor activation. Hypertension 24;43: Kuo JJ, Silva AA, Hall JE. Hypothalamic melanocortin receptors and chronic regulation of arterial pressure and renal function. Hypertension 23;41: Farooqi IS, Keogh JM, Yeo GSH, Lank EJ, Cheetham T, O Rahilly S. Clinical spectrum of obesity and mutations in the melanocortin 4 receptor gene. N Engl J Med 23;348: Larsen LH, Echwald SM, Sørensen TI, Andersen T, Wulff BS, Pedersen O. Prevalence of mutations and functional analyses of melanocortin 4 receptor variants identified among 75 men with juvenile-onset obesity. J Clin Endocrinol Metab 25;9: Stutzmann F, Tan K, Vatin V, et al. Prevalence of melanocortin-4 receptor deficiency in Europeans and their age-dependant penetrance in multigenerational pedigrees. Diabetes 28;57: Alharbi KK, Spanakis E, Tan K, et al. Prevalence and functionality of paucimorphic and private MC4R mutations in a large, unselected European British population, scanned by meltmadge. Hum Mutat 27;28: Huszar D, Lynch CA, Fairchild-Huntress V, et al. Targeted disruption of the melanocortin-4 receptor results in obesity in mice. Cell 1997;88: Mertens IL, Van Gaal LF. Overweight, obesity, and blood pressure: the effects of modest weight reduction. Obes Res 2;8: Rocchini AP, Moorehead CP, DeRemer S, Bondie D. Pathogenesis of weight-related changes in blood pressure in dogs. Hypertension 1989;13: Hall JE. Pathophysiology of obesity hypertension. Curr Hypertens Rep 2;2: Hall JE, Summers RL, Brands MW, Keen H, Alonso-Galicia M. Resistance to metabolic actions of insulin and its role in hypertension. Am J Hypertens 1994;7: Fan W, Dinulescu DM, Butler AA, Zhou J, Marks DL, Cone RD. The central melanocortin system can directly regulate serum insulin levels. Endocrinology 2; 141: Vaz M, Jennings G, Turner A, Cox H, Lambert G, Esler M. Regional sympathetic nervous activity and oxygen consumption in obese normotensive human subjects. Circulation 1997;96: Brito MN, Brito NA, Baro DJ, Song CK, Bartness TJ. Differential activation of the sympathetic innervation of adipose tissues by melanocortin receptor stimulation. Endocrinology 27;148: Song CK, Jackson RM, Harris RB, Richard D, Bartness TJ. Melanocortin-4 receptor mrna is expressed in sympathetic nervous system outflow neurons to white adipose tissue. Am J Physiol Regul Integr Comp Physiol 25;289: R1467-R Rosenbaum M, Goldsmith R, Bloomfield D, et al. Low-dose leptin reverses skeletal muscle, autonomic, and neuroendocrine adaptations to maintenance of reduced weight. J Clin Invest 25;115: Haynes WG, Sivitz WI, Morgan DA, Walsh SA, Mark AL. Sympathetic and cardiorenal actions of leptin. Hypertension 1997;3: Haynes WG, Morgan DA, Walsh SA, Mark AL, Sivitz WI. Receptor-mediated regional sympathetic nerve activation by leptin. J Clin Invest 1997;1:27-8. Copyright 29 Massachusetts Medical Society. 52 n engl j med 36;1 nejm.org january 1, 29 Downloaded from nejm.org on April 11, 218. For personal use only. No other uses without permission. Copyright 29 Massachusetts Medical Society. All rights reserved.