Reviews/Commentaries/ADA Statements PERSPECTIVES ON THE NEWS

Obesity: Mediators and Treatment Approaches

ZACHARY T. BLOOMGARDEN, MD with the latter being G protein–coupled receptors and with abnormalities of the ␤-2 associated with obesity (4). his is the fifth of a series of six arti- particularly with a high-fat diet and envi- A number of other genetic polymor- cles based on presentations at the ronmental stress. Interestingly, ob/ob phisms appear to regulate adipose tissue T American Diabetes Association Sci- mice have a sevenfold increase in NPY ex- mass (5). Using a genome-wide scan, a entific Sessions held 6–10 June 2008 in pression. NPY has -like effects on receptor involved in pain San Francisco, California. release of adipokines and . In vivo, perception, GPR-74 (the G protein– NPY administration by a slow-release pel- coupled receptor for the morphine- Obesity hormones let stimulates fat formation in lean mice modulating A18 famide Zofia Zukowska (Washington, DC) dis- and monkeys. The effect of stress with a [NPAF] and NPFF), is highly expressed in cussed the peripheral actions of neu- high-fat diet is blocked with a y2 receptor adipose tissue. NPFF decreases lipolysis ropeptide Y (NPY), an important antagonist or by deletion of the y2 recep- in human fat cells in a dose-dependent mediator by which diet and lack of phys- tor using local injection of a viral vector, fashion and may be as potent as insulin, ical activity lead to weight gain. NPY is both of which increase apoptosis and de- appearing to act by inhibiting the signal synthesized as a 97 amino acid precursor, crease proliferation of endothelial cells from the ␤-adrenergic receptor to hor- which is processed and amidated to NPY and adipocytes, decreasing obesity and mone-sensitive lipase. There are several 1–36. NPY 1–36 is cleaved by dipeptidyl improving insulin sensitivity and glucose GPR-74 polymorphisms in man, with the peptidase (DPP)-4 to NPY 3–36, which tolerance. Stress-induced increased sym- ATAG haplotype associated with 45% re- has other biologic activities, leading to the pathetic tone, then, would lead to weight duction in likelihood of obesity (6) and possibility that inhibitors of DPP-4 may loss, but desensitization of that pathway, with increased lipolysis and increased potentiate NPY action with potentially ad- accompanied by upregulation of the y2 adipocyte norepinephrine sensitivity, verse (or beneficial) effects. NPY is con- pathway, as well as increased local glu- suggesting that individuals with the pro- tained in all sympathetic nerves together cocorticoids, leads to fat accumulation tective polymorphism have a greater lipo- with norepinephrine, although their (3). Interestingly, peptide YY 3–36 has ef- lytic response to norepinephrine and modes of release are somewhat dissimilar. fects similar to those of NPY in fat, stim- decreased NPFF/NPAF response. An- This is, then, a major sympathetic ner- ulating angiogenesis and adipogenesis, other question is whether there are ge- vous system effector, elevated in stress, although in the brain PYY has the oppos- netic variations in the adipocyte hypertension, cardiac and renal failure, ing effect of inducing satiety and promot- generation involved in obesity. Growth of and malignancies, which, Zukowska ing weight loss. human adipose tissue has been thought to stated, is the most abundant peptide in Peter Arner (Karolinska, Stockholm, be regulated by adipogenesis in youth, the brain and heart. There are multiple G Sweden) discussed human obesity genes, with subsequent fat expansion only by protein–coupled receptors for NPY in- including those influencing fatty acid growth of preexisting adipocytes, al- cluding y1, which is vasoconstrictive, has combustion in muscle, adipocyte lipid though above a certain obesity level adi- atherogenic effects, and acts in the brain metabolism, adipogenesis, and brain con- pogenesis also may occur later in life. to reduce anxiety and increase appetite trol of appetite and energy expenditure. Arner reviewed his studies with measure- (1), and y2 and y5, which are also in- Human fat cells increase lipid content un- ment of 14C (derived from atmospheric volved in the peptide’s vascular effects der circumstances of increased insulin or nuclear weapon testing) in adipose tissue, (2). She cited evidence that although decreased catecholamine levels, which in- suggesting that adipocyte turnover is stress with increased sympathetic tone crease fat uptake and decrease its release. ϳ10% per year and that lean individuals causes lipolysis by increasing catechol- Genetic variants in lipolysis have been doc- produce new fat cells at lower rates than amine levels, sympathetic activation may umented, with a group of 311 obese and obese individuals (7). Genetic factors also increase NPY, leading to angiogenesis 223 nonobese women and 114 and 82 such linked to greater or lesser degrees of adi- and adipogenesis causing fat cell growth. men, showing that obesity was associated pogenesis may, then, play a role in human Cold stress with a fat- and sugar-enriched with increased insulin-induced inhibition obesity. diet increases body weight. An experi- of lipolysis and decreased norepinephrine- Several studies presented at the meet- mental model of daily 10-min encounters induced lipolysis. Resistance to catechol- ing examined genetic associations of obe- with an aggressive mouse was also associ- amine action would, Arner suggested, sity. Lindgren et al. (abstract 1751) ated with increased NPY as well as in- promote obesity. There are four catechol- analyzed genome wide association data creased Y2 receptor expression, seen amine receptors, ␣-2A and ␤-1, -2, and -3, pertaining to obesity from Ͼ90,000 indi- ●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●● viduals, finding that variants in the fused Zachary T. Bloomgarden, MD, is a practicing endocrinologist in New York, New York, and is affiliated with toe (FTO) gene, which have already been the Division of Endocrinology, Mount Sinai School of Medicine, New York, New York. described and appear to relate to hypo- DOI: 10.2337/dc09-zb05 © 2009 by the American Diabetes Association. Readers may use this article as long as the work is properly thalamic and pancreatic islet signaling, cited, the use is educational and not for profit, and the work is not altered. See http://creativecommons. and in the melanocortin four receptor org/licenses/by-nc-nd/3.0/ for details. gene, involved in another hypothalamic e48 DIABETES CARE, VOLUME 32, NUMBER 5, MAY 2009 Bloomgarden satiety receptor pathway, appear to show and an acute nausea response to very high that Bloom founded (and of which he is the strongest signals. Traurig et al. (ab- gut signal levels. Chief Scientific Officer) focused on pep- stract 1,752) reported that gene variants Peptide YY (PYY) 36 was identified in tides regulating appetite. Onyntomodulin among 3,501 Pima Indians associated 1982 as a peptide produced by gut endo- appears to act at the central GLP-1 recep- with obesity may alter transcription of crine cells and released into the circula- tor but also has action at the arcuate nu- SIM1, the human homolog of the Dro- tion (10). Bloom’s group subsequently cleus, leading to greater weight loss and sophila Single Minded gene, required for found the peptide to be present in the less nausea than GLP-1. Its effect is addi- differentiation of hypothalamic neuroen- distal gut, with malabsorption increasing tive to that of rimonabant. docrine cells, and further that four of 96 delivery of nutrients to this site, which Mary Dallman (San Francisco, CA) morbidly obese Pima Indians had abnor- particularly increases PYY levels (11). PYY discussed effects of glucocorticoids on malities in the SIM1 gene sequence. Scott levels increase fivefold after major small feeding. Insulin and glucocorticoids have et al. (abstract 1,763) reported that T al- bowel resection (12). After bariatric sur- in many ways opposite effects, inhibiting lele carriers of the 3111T/C polymor- gery with bypass, there is elevation in versus promoting gluconeogenesis, with phism of the clock gene had greater waist PYY. Infusion of PYY 3–36 similarly re- anabolic versus catabolic actions, and re- circumference among 207 individuals duces food intake and hunger scores and ducing versus promoting food intake. To- with histologically confirmed non alco- inhibits the response to ghrelin, and gether, the two synergize or antagonize in holic fatty liver disease, suggesting a role Bloom suggests that ghrelin is not stimu- a fashion ultimately increasing caloric in- of the molecular clock in visceral obesity, lated by the presence of food in the stom- take. Adrenalectomized rats given in- although there was no association with ach but rather by changes in the release of creasing corticosterone doses lose weight other insulin resistance characteristics. gut hormones such as PYY; obese individ- as a result of its catabolic effect but with (Abstract numbers refer to the ADA Sci- uals have lower PYY release, potentially mesenteric fat accumulation, thus shifting entific Sessions, Diabetes 57 [Suppl. 2], explaining weight differences between caloric stores from peripheral to central 2008). obese and lean individuals. PYY3–36 is sites and from muscle to fat. Fat intake is Stephen Bloom (London, U.K.) dis- rapidly cleared, rapid peaks lead to nau- not increased in adrenalectomized dia- cussed gut hormones regulating appetite. sea, and appetite rebound occurs when betic animals, however, unless insulin is Bloom termed bariatric surgery the only levels fall. -like peptide (GLP)-1 replaced, and fat intake is proportional to currently available successful treatment of reduces food intake by central effects, the dose of insulin administered, and cor- obesity, with the Swedish obese subjects while its DPP-4 degradation product and ticosterone does not increase mesenteric study of 4,000 individuals showing that exendin 9–39 block the effect, increasing fat in the absence of insulin, with an in- those who had the procedure maintained food intake. Interestingly, DPP-4 inhibi- sulin dose-related increase in mesenteric 25% weight loss after 10 years, with a tion decreases formation of PYY3–36 fat. Effects of corticosterone are mediated 29% increase in mortality in matched (but from PYY1–36, and administration of the in part by increasing dopamine levels over not randomized) controls (8) and with DPP-4 inhibitor vildagliptin has been re- the shell of the nucleus accumbens (14). similar suggestion of benefit in a retro- ported to decrease PYY concentrations Behavioral actions of insulin depend on spective matched control study of nearly (13). Human de- hepatic vagal afferents, with insulin in- 10,000 individuals undergoing such pro- creases food intake in a fashion similar to creasing fat intake to a greater extent cedures (9). There are, however, 30–40% that associated with PYY3–36. when infused into the splanchnic rather rehospitalization rates and 1–2% surgical is a 37–amino acid than the systemic circulation, while va- mortality, leading Bloom to conclude, gut peptide derived from the preproglu- gotomy increases fat intake, suggesting “It’s expensive, it’s dangerous, but it cagon molecule, which does not bind to that hepatic vagal afferents to the brain are works.” Bariatric surgery appears to act in the . Its initial discovery involved in inhibition of fat intake (15). It part by increasing satiety, perhaps by in- was based on reduction in gastric acid appears that fat intake alters c-Fos expres- duction of gut hormones. Ghrelin is pro- production. Oxyntomodulin is released sion in brain sites such as the nucleus duced in the stomach and stimulates in proportion to meal size, and levels are tractus solitaries and nucleus accumbens appetite, with treatment useful in individ- high in bowel disorders associated with known to participate in pleasurable be- uals with impaired appetite from cancer decreased appetite. Chronic oxynto- haviors (16). or renal failure, but levels typically are low modulin treatment decreases weight to a The function of this dual hormonal in obesity, so blocking ghrelin may not be greater extent than seen in pair-fed animals, system may be related to the effect of helpful, just as obese individuals typically suggesting an effect in increasing energy ex- stress and high glucocorticoids on behav- have high , with administration of penditure. In a 4-week study of oxynto- ior and autonomic outflow. High glu- this agent not effective when given as a modulin treatment in obese individuals, 2.5 cocorticoid levels stimulate insulin and single agent. Reduction in hunger after kg weight loss was seen over 4 weeks, com- food ingestion—a metabolic signal that in meals is not related to gastric distension pared with a 0.5-kg decrease in placebo- turn reduces the brain chronic stress– or to nutrients in the circulation and ap- treated patients. No evidence of escape from response network, which constitutes a bi- pears to be caused by neural or hormonal the effect was seen at the end of the study. ological explanation for the stressed signals derived from the gastrointestinal Measurement of metabolic rate showed no individual feeling and, perhaps, function- tract. Gut hormones do not exhibit down- change in basal levels, but activity-related ing better with increasing food intake, regulation of response with prolonged ex- energy expenditure was 9% greater, with thus leading to a learned behavioral pat- posure. Three separate gut hormonal a 26% increase in physical activity. An tern (17). Chronic corticosterone admin- responses may be involved in weight reg- oxyntomodulin analog not susceptible to istration increases nonhypothalamic ulation; short-lived reduction in appetite; degradation by DPP-4, TKS1225 is being corticotropin-releasing factor (CRF) lev- chronic elevation, seen with gut disease; developed by thiakis, a biotech company els, while insulin increases fat stores, re-

DIABETES CARE, VOLUME 32, NUMBER 5, MAY 2009 e49 Perspectives on the News ducing CRF expression. As another months, at 2 years neither group had sus- tor ␥ coactivator (PGC)-1␤ is a transcrip- feedback loop in the stress system, su- tained weight loss or change in blood tional coactivator for sterol response crose and/or fat reduce the CRF, ACTH, pressure or lipids, or, among the diabetic element binding protein-1c, the tran- and corticosterone responses to single or patients, in glycemia. Buscemi et al. (ab- scriptional regulator of hepatic lipogene- repeated acute stress. Sucrose ingestion stract 1746) randomized 20 overweight sis. They reduced PGC-1␤ expression inhibits paraventriclar nucleus CRF, so or obese women to two different hypoca- with targeted antisense oligonucleotides mesenteric fat mass may produce a meta- loric diets: a low-CHO “Atkins” diet vs. a in the liver and adipose tissue of rats fed bolic feedback signal. Mediterranean diet, with 8 vs. 5 kg weight high-fructose diets for 4 weeks and found loss at 60 days but 57% decreased vs. 41% 20 and 70% reductions in plasma and Studies of relationships between increased brachial artery flow-mediated liver triglyceride, respectively, and lower diabetes and obesity dilation at 5 days. The low-CHO diet in- fasting glucose and insulin levels with in- Despres et al. (abstract 949) analyzed the creased interleukin-6 and 8-iso- creased hepatic and peripheral insulin relationship between waist circumference prostaglandin F2␣ at 5 days, suggesting sensitivity, suggesting that PGC-1␤ may and BMI among 19,605 diabetic and that it may acutely cause inflammation be a therapeutic target for treatment of 147,567 nondiabetic patients examined and oxidative stress. Kodama et al. (ab- insulin-resistant states including nonal- by almost 6,400 primary care physicians stract 1703) presented a meta-analysis of coholic fatty liver disease, diabetes, and in 63 countries. Waist and BMI were 24 studies of type 2 diabetic patients in hypertriglyceridemia. Le et al. (abstract strongly correlated, but at every BMI level whom the dietary CHO-to-fat ratio was 35), noting that consumption of sugar- waist circumference was greater in dia- measured, showing no effect of the ratio sweetened beverages containing fructose betic than in nondiabetic patients: by ϳ4 on fasting glucose or A1C but an associa- increased by ϳ135% from 1977 to 2001, and 10 cm among men and women, re- tion of a high CHO-to-fat ratio with studied 11 male offspring of type 2 dia- spectively. The regression equations were higher fasting insulin and higher post- betic patients versus 8 control subjects af- as follows: for diabetic men, waist circum- meal glucose and 2-h insulin levels, sug- ter standard and high-fructose diets. ference (cm) ϭ 33.3 ϩ 2.1 ϫ BMI (kg/ gesting reduced insulin sensitivity. Fasting triglyceride increased 81 vs. 32% m2) ϩ 0.12 ϫ age (years); for nondiabetic Menchikova et al. (abstract 66) com- and insulin 17 vs. 30%, respectively, with men, waist circumference ϭ 29.3 ϩ pared seven older overweight to obese in- similarly increased urate and decreased 2.2 ϫ BMI ϩ 0.14 ϫ age; for diabetic dividuals undergoing caloric restriction free fatty acids and ketones in both women, waist circumference ϭ 39.3 ϩ with 8 kg weight loss with eight individ- groups. Hepatic lipid content increased 1.8 ϫ BMI ϩ 0.09 ϫ age; and for nondi- uals beginning an exercise program and 103 vs. 73%. Hepatic insulin sensitivity abetic women, waist circumference ϭ showed a similar (22 vs. 26%) improve- decreased 13%, and alanine aminotrans- 29.4 ϩ 2.0 ϫ BMI ϩ 0.12 ϫ age. Mea- ment in insulin sensitivity using a hyper- ferase increased 88% in the offspring surement of waist circumference, then, insulinemic-euglycemic clamp. Vastus while not changing in controls. Stanhope adds information for characterization of lateralis muscle biopsies showed similar et al. (abstract 352) randomized 23 over- patients’ diabetes risk in all BMI catego- mitochondrial content (based on cardio- weight or obese individuals to fructose- or ries. Iacobellis and Gerstein (abstract lipin content), but NADH oxidase, a mea- glucose-sweetened beverage consump- 1735) analyzed echocardiograms of 246 sure of activity of the electron transport tion for 10 weeks. Abdominal computer- individuals and found that among the chain, increased only with exercise, sug- ized tomography scanning showed an 58% with metabolic syndrome, epicardial gesting that the apparent mitochondrial increase in fat area with fructose, and fast- fat thickness was 9.5 mm and 7.5 mm in dysfunction of aging (and of insulin- ing glucose and insulin increased 5 and men and women, respectively, which ex- resistant states) is a function of physical 12%, with reduction in insulin sensitivity ceeds that in individuals without the syn- inactivity, while weight loss but not exer- index, measured from deuterated glucose drome. These data suggest a useful cise increased citrate synthase activity, disposal and net insulin output during a measure for risk stratification and, per- which is the regulatory step of the Krebs glucose tolerance test; these parameters haps, for therapeutic interventions. Say- cycle. Paniagua et al. (abstract 1687) were unchanged with glucose-sweetened dah and Eberhardt (abstract 993) studied 11 abdominally obese offspring of beverages. Yeung et al. (abstract 353) analyzed data from 1999–2004 National obese type 2 diabetic parents on three 28- noted that the distinctive photosynthetic Health and Nutrition Examination Sur- day diets (one high in saturated fat, one high mechanism of corn and sugar cane selec- veys and found a sedentary lifestyle in in monounsaturated fat, and one high in tively concentrates the 13-C isotope of 35% of known diabetic subjects but 26% CHO) and found lower postprandial free carbon, measuring the C13-to-C12 ratio of pre-diabetic subjects, 24% of those fatty acids and higher glucose with the high using mass spectroscopy in 131 individ- with undiagnosed diabetes, and 24% of CHO diet, whereas glucose was lower and uals. Comparing those consuming Ͼ2–4 those without diabetes; 7, 4, 6, and 2% of fatty acid levels were higher with the satu- servings/week vs. Յ1–3 servings/month the groups, respectively, reported an in- rated fat diet. Lipid oxidation was lower of sodas or sweetened juices, C13 in- ability to vigorously exercise, while 25, with the high-CHO diet, which was associ- creased with increasing consumption of 17, 20, and 14% reported watching tele- ated with higher adipose tissue uncoupling these beverages and was significantly as- vision at least 5 h daily. protein-2 expression. Postprandial oxi- sociated with waist circumference and Dyson et al. (abstract 1703) com- dized LDL levels were lower with the high BMI. pared a low-carbohydrate (CHO) diet monounsaturated fat diet. Gillum et al. (abstract 1552) noted with a standard diet in 13 type 2 diabetic Dietary fructose appears to be an im- that the endocannabinoid precursor N- and 13 nondiabetic obese individuals portant mediator of insulin resistance. acylphosphatidyl-ethanolamines are syn- with BMI 35 kg/m2. Although weight loss Nagai et al. (abstract 1514) noted that thesized in the small intestine in response was greater with the low-CHO diet at 3 peroxisome proliferator–activated recep- to ingested fat and that plasma and lymph e50 DIABETES CARE, VOLUME 32, NUMBER 5, MAY 2009 Bloomgarden levels increase 50–100% following food cating a weight-independent mechanism viduals. There was greater gastric empty- or intraduodenal lipid infusion, with sys- of improved glycemia. ing suppression with GLP-1 than with temic administration reducing food con- Several studies addressed aspects of but similar reduction in food in- sumption 90% and suppressing bariatric surgery. Bose et al. (abstract take in 12 nondiabetic individualsl; the hypothalamic transcription of NPY 64% 1440) compared nine type 2 diabetic GLP-1 resistance in type 1 diabetes per- in a rodent model and with intracerebro- women 1 month after Roux-en-Y gastric haps indicates action in part via stimula- ventricular administration reducing food bypass surgery with 10 having diet- tion of amylin secretion. had intake 56%, suggesting potential as a induced weight loss, both by ϳ10 kg, and effects similar to those of human amylin. therapeutic target. Scheen et al. (abstract found similar improvement in fasting glu- King et al. (abstract 549) reported use of 101) presented 2-year results of 599 indi- cose and insulin but greater reduction in continuous glucose monitoring to titrate viduals receiving 20 mg daily of the can- post–glucose load glycemia and greater prandial insulin dosing in nine pump- nabinoid 1 receptor antagonist increase in GLP-1, which may therefore treated type 1 diabetic patients before and rimonabant along with those of 305 re- be a consequence of the surgery rather after addition of 60 ␮g pramlintide before ceiving placebo, finding 5.5 vs. 1.2 kg than of the weight loss. Schernthaner et each meal. Insulin dosing was adjusted to weight loss and 6 vs. 2 cm reduction in al. (abstract 257) studied 95 patients with have fewer than 20% of glucose readings waist circumference, with improvement a mean 39 kg weight loss after bariatric Ͼ170 mg/dl, fewer than 10% Ͻ70 mg/dl, in fasting glucose and insulin and in glu- surgery, finding an increase in circulating and 4-h postmeal glucose within 20% of cose tolerance. Nausea occurred in 13 vs. endothelial precursor cells, potentially re- premeal glucose. At 5 weeks, A1C de- 4%, anxiety in 5 vs. 3%, and depression in ducing atherogenesis. Tarnoff et al. (ab- creased from 7.4 to 7.1%, weight de- 4 vs. 2%, with most of these side effects stract 105) randomized 11 type 2 diabetic creased 1 kg, and basal insulin decreased observed during the first year of treat- patients, 7 receiving metformin alone and 4 11%, while the carbohydrate–to– ment. Hollander et al. (abstract 330) ran- also receiving a sulfonylurea, to endoscopic prandial insulin dose ratio increased 8%; domized 368 type 2 diabetic patients insertion of a 61-cm duodenal-jejunal by- they suggest that in this setting prandial treated with insulin monotherapy to 20 pass impermeable fluoropolymer sleeve, insulin bolus doses should not be re- mg daily of rimonabant versus placebo for fastened with a barbed metal anchor at the duced. Huffman and McLean (abstract 48 weeks, with baseline a A1C of 9.1% duodenal entrance—versus 5 having sham 199) treated 11 type 1 diabetic individu- and BMI 35 kg/m2, and found that A1C endoscopy. At 1 week, fasting glucose de- als with continuous subcutaneous pram- decreased 0.9 vs. 0.2%, weight decreased creased 52 vs. increasing 17 mg/dl, respec- lintide infusion using an insulin pump at 2.5 kg vs. an increase of 0.1 kg, HDL cho- tively, with mean 7-point daily glucose nine ␮g/h basally, titrating to 60 ␮g three lesterol increased 3% vs. a decrease of 7%, profile decreasing 55 vs. increasing 1 mg/ times daily before meals over 3 weeks. and triglyceride decreased 4% vs. an in- dl. At 12 weeks, one versus three contin- Baseline vs. 12-week fasting glucose was crease of 8%. Nausea occurred in 11 vs. ued to require hypoglycemic medication, 187 vs. 182 mg/dl, but there was a post- 2%, anxiety in 14 vs. 4%, and depression with weight loss 8.5 vs. 7.8 kg. At 31 prandial glucose increment of 24 mg/dl in 10 vs. 4%. weeks, eight patients receiving the sleeve vs. a decrement of 17 mg/dl, with a 26% Garvey et al. (abstract 390) random- had a decrease in A1C of 2.9% from 8.9% reduction in the bolus insulin–to– ized 206 diabetic patients to a combina- at baseline, while three sham-treated pa- carbohydrate ratio. A1C decreased from tion of 15 mg phentermine in the tients had a 0.8% fall. The sleeve came 8.2 to 7.7%, and there was mean 1.4 kg morning (typical dose 15–37.5 mg daily), loose in three patients, requiring retrieval weight loss, suggesting that this could be and 100 mg topiramate in the evening through endoscopy; eight had abdominal a potentially useful approach to pramlint- (typical anticonvulsant dose 100–200 mg pain; three had diarrhea; three had vom- ide administration. daily) vs. placebo, with a 16-week A1C iting; two had hypoglycemia; and one had In a study of type 2 diabetic patients, decrease 1.1 vs. 0.6% and weight loss 6 nausea. Bruno et al. (abstract 1741) found Riddle (abstract 524) randomized 61 in- vs. 1%; no treatment-related serious ad- that two serum markers of bone turnover dividuals receiving ti- verse events were reported. Karnieli et al. increased in 20 patients 18 months fol- trated to target fasting glucose 70–100 (abstract 430) implanted a gastric stimu- lowing Roux-en-Y gastric bypass surgery: mg/dl to 120 ␮g pramlintide three times lator in 10 obese type 2 diabetic patients, bone-specific alkaline phosphatase from daily versus placebo for 16 weeks, finding applying gastric contractility modulation 18 to 22 ng/ml and N-telopeptide cross- the insulin dose to increase from 51 to 65 signals to the antrum of the stomach dur- linked type 1 from 11 to 17 vs. from 48 to 57 units daily. Fasting glu- ing meals. At 24 weeks, A1C decreased nmol/l; however, 25-hydroxy vitamin D cose decreased from 144 to 81 vs. from from 8.1 to 7.3%, fasting glucose from increased from from 17 to 26 ng/ml (pre- 122 to 85 mg/dl, the postprandial glucose 174 to 143 mg/dl, homeostasis model as- sumably reflecting supplementation). increment decreased from 55 to 30 vs. sessment of insulin resistance from 10.3 Both of these studies suggest that there from 61 to 54 mg/dl, and A1C was re- to 6.2, systolic blood pressure from 134 to may be harm as well as benefit from bari- duced from 8.7 by 1.2% vs. from 8.2 by 126 mmHg, weight from 103 to 98 kg, atric surgery, which is in keeping with 0.5%. Body weight decreased 1 kg versus and waist circumference from 121 to 115 Bloom’s observations. increasing 1 kg, hypoglycemia occurred cm, with HDL cholesterol increasing in 57 vs. 55% of subjects, and nausea oc- ϳ9%. Kipnes et al. (abstract 573) used Amylin and pramlintide curred in 39 vs. 18%. Given the small the same device in 12 type 2 diabetic pa- Asmar et al. (abstract 1446) compared the numbers and somewhat dissimilar base- tients for 12 weeks, with a decrease in effects of human amylin, pramlintide, and line glycemia, the study can only be re- A1C from 8.5 to 7.6% and in weight from GLP-1 on gastric emptying, appetite, and garded as suggestive of benefit. 106 to 103 kg. The weight change did not food intake, finding similar reduction in Amylin may be particularly important correlate with that in A1C, perhaps indi- all parameters in 11 type 1 diabetic indi- as a component of multiple drug regi-

DIABETES CARE, VOLUME 32, NUMBER 5, MAY 2009 e51 Perspectives on the News mens, with interesting evidence that it Lilly, Amylin, Daichi Sankyo, and Glaxo- obese subjects. N Engl J Med 2007 Aug may restore leptin responsiveness in obe- SmithKline; has served on advisory panels for 23;357:741–752 sity (18); a number of studies presented at Medtronics, Takeda, Merck, AtherGenics, CV 9. Adams TD, Gress RE, Smith SC, Halver- the ADA assessed amylin-related treat- Therapeutics, Daichi Sankyo, BMS, and Astra- son RC, Simper SC, Rosamond WD, Lam- ments as approaches to obesity manage- Zeneca; holds stock in Abbott, Bard, onte MJ, Stroup AM, Hunt SC: Long-term Medtronic, Merck, Millipore, Novartis, and mortality after gastric bypass surgery. ment. Nicandro et al. (abstract 1543) Roche; and has served as a consultant for No- N Engl J Med 2007;357:753–761 described phase one studies with vartis, Dainippon Sumitomo Pharma America, 10. Tatemoto K: Isolation and characteriza- AC2307, an amylin mimetic with in- Forest Laboratories, and Nastech. No other tion of peptide YY (PYY), a candidate gut creased duration of action in obese non- potential conflicts of interest relevant to this hormone that inhibits pancreatic exocrine diabetic individuals, showing reduction article were reported. secretion. Proc Natl Acad Sci U S A 1982; in food intake. Trevaskis et al. (abstract 79:2514–2518 192) administered amylin, leptin, and 11. Adrian TE, Ferri GL, Bacarese-Hamilton protein YY (3–36) to diet-induced obese References AJ, Fuessl HS, Polak JM, Bloom SR: Hu- 1. Eva C, Serra M, Mele P, Panzica G, Oberto man distribution and release of a putative rats for 4 weeks. As single agents, amylin A: Physiology and gene regulation of the and leptin led to 8 and 4% weight loss, new gut hormone, peptide YY. Gastroen- brain NPY Y1 receptor. Front Neuroendo- terology 1985;89:1070–1077 respectively, while there was no weight crinol 2006;27:308–339 loss with PYY. Addition of PYY to amylin 12. Adrian TE, Savage AP, Fuessl HS, Wolfe 2. Abe K, Tilan JU, Zukowska Z: NPY and K, Besterman HS, Bloom SR: Release of and leptin led to 10 and 9% weight loss, NPY receptors in vascular remodeling. peptide YY (PYY) after resection of small respectively. Amylin plus leptin led to Curr Top Med Chem 2007;7:1704–1709 bowel, colon, or pancreas in man. Surgery 20% weight loss, and the triple combina- 3. Kuo LE, Kitlinska JB, Tilan JU, Li L, Baker 1987;101:715–719 tion led to 24% weight loss, with the latter SB, Johnson MD, Lee EW, Burnett MS, 13. Vella A, Bock G, Giesler PD, Burton DB, two regimens reducing hepatic expres- Fricke ST, Kvetnansky R, Herzog H, Serra DB, Saylan ML, Deacon CF, Foley Zukowska Z: acts directly sion of the lipogenic genes stearoyl-CoA JE, Rizza RA, Camilleri M: The effect of in the periphery on fat tissue and mediates dipeptidyl peptidase-4 inhibition on gas- desaturase-1 and fatty acid synthase, as stress-induced obesity and metabolic syn- tric volume, satiation and enteroendo- well as reducing fat mass to undetectable drome. Nat Med 2007;13:803–811 crine secretion in type 2 diabetes: a levels without reducing lean mass. Weyer 4. Large V, Hellstro¨m L, Reynisdottir S, Lo¨n- double-blind, placebo-controlled cross- et al. (abstract 1738) treated 177 over- nqvist F, Eriksson P, Lannfelt L, Arner P: over study. Clin Endocrinol (Oxf) 2008; weight or obese individuals on a 40% cal- Human beta-2 adrenoceptor gene poly- 69:737–744 orie-deficit diet with 180 ␮g pramlintide morphisms are highly frequent in obesity ␮ and associate with altered adipocyte be- 14. Barrot M, Abrous DN, Marinelli M, twice daily for 2 weeks and then 360 g Rouge´-Pont F, Le Moal M, Piazza PV: In- twice daily (twice the current maximal ta-2 adrenoceptor function. J Clin Invest 1997;100:3005–3013 fluence of glucocorticoids on dopaminer- daily dose) for 4 weeks. A total of 139 gic transmission in the rat dorsolateral subjects completed this period and 5. Dahlman I, Arner P: Obesity and poly- morphisms in genes regulating human striatum. Eur J Neurosci 2001;13: achieved 2–8% weight loss and were ran- adipose tissue. Int J Obes (Lond) 2007; 812–818 domized to add 5 mg twice 31:1629–1641 15. Warne JP, Foster MT, Horneman HF, Pec- daily, continue pramlintide alone, or 6. Dahlman I, Dicker A, Jiao H, Kere J, oraro NC, Ginsberg AB, Akana SF, Dall- change to 5 mg metreleptin twice daily Blomqvist L, van Harmelen V, Hoffstedt J, man MF: Hepatic branch vagotomy, like alone. Of 93 evaluable participants at 20 Borch-Johnsen K, Jorgensen T, Hansen T, insulin replacement, promotes voluntary weeks, weight loss from baseline was 13, Pedersen O, Laakso M, Arner P: A com- lard intake in streptozotocin-diabetic rats. 8, and 8%, with weight loss stabilizing mon haplotype in the G-protein-coupled Endocrinology 2007;148:3288–3298 16. Warne JP, Horneman HF, Ginsberg AB, after 12 weeks in the monotherapy receptor gene GPR74 is associated with leanness and increased lipolysis. Am J Pecoraro NC, Foster MT, Akana SF, Dall- groups but continuing at an average rate man MF: Mapping brain c-Fos immuno- of 0.2 kg/week in the combination group. Hum Genet 2007;80:1115–1124 7. Spalding KL, Arner E, Westermark PO, reactivity after insulin-induced voluntary Aronne et al. (abstract 99) randomized Bernard S, Buchholz BA, Bergmann O, lard intake: insulin- and lard-associated 244 overweight or obese individuals to Blomqvist L, Hoffstedt J, Na¨slund E, Brit- patterns. J Neuroendocrinol 2007;19: placebo or to 120 ␮g pramlintide three ton T, Concha H, Hassan M, Ryde´n M, 794–808 times daily alone or in combination with Frise´n J, Arner P: Dynamics of fat cell 17. Dallman MF, Warne JP, Foster MT, Pec- 10 mg sibutramine daily or 37.5 mg turnover in humans. Nature 2008; oraro NC: Glucocorticoids and insulin phentermine daily for 24 weeks, finding 453:783–787 both modulate caloric intake through ac- weight loss in evaluable participants of 8. Sjo¨stro¨m L, Narbro K, Sjo¨stro¨m CD, Kara- tions on the brain. J Physiol 2007;583: 2.1, 3.6, 11.3, and 11.3 kg, with 28, 36, son K, Larsson B, Wedel H, Lystig T, Sul- 431–436 78, and 82% of subjects, respectively, livan M, Bouchard C, Carlsson B, 18. Roth JD, Roland BL, Cole RL, Trevaskis JL, Weyer C, Koda JE, Anderson CM, Parkes achieving Ն5% weight loss. Bengtsson C, Dahlgren S, Gummesson A, Jacobson P, Karlsson J, Lindroos AK, Lo¨n- DG, Baron AD: Leptin responsiveness re- roth H, Na¨slund I, Olbers T, Stenlo¨f K, stored by amylin agonism in diet-induced Torgerson J, Agren G, Carlsson LM: obesity: evidence from nonclinical and clin- Acknowledgments— Z.T.B. has served on Swedish Obese Subjects Study. Effects of ical studies. Proc Natl Acad SciUSA2008; speaker’s bureausof Merck, Novo Nordisk, bariatric surgery on mortality in Swedish 105:7257–7262

e52 DIABETES CARE, VOLUME 32, NUMBER 5, MAY 2009