Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

A Marker of Endotoxemia Is Associated With Obesity and Related Metabolic Disorders in Apparently Healthy Chinese

A Marker of Endotoxemia Is Associated With Obesity and Related Metabolic Disorders in Apparently... Cardiovascular and Metabolic Risk ORIGINAL ARTICLE A Marker of Endotoxemia Is Associated With Obesity and Related Metabolic Disorders in Apparently Healthy Chinese 1 1 LIANG SUN, MSC HONGYU WU, MSC idemic of obesity and associated meta- 1,7 1 ZHIJIE YU, MD, PHD YAN CHEN, MD, PHD bolic disorders. However, increasing 2 4 XINGWANG YE, PHD JOEL DORE, PHD evidence supports the fact that chronic 3 5 SHURONG ZOU, MD, MSC KARINE CLEMENT ´ , MD, PHD low-grade inflammation is one of the key 1 6 HUAIXING LI, PHD FRANK B. HU, MD, PHD mechanisms underlying the pathogenesis 1 1 DANXIA YU, MSC XU LIN, MD, PHD of obesity-related metabolic disorders (2). However, the agents responsible for initi- ating and sustaining this low-grade in- OBJECTIVE — Elevated lipopolysaccharide-binding protein (LBP), a marker of subclinical en- flammatory signal have yet to be dotoxemia, may be involved in the pathogenesis of obesity and metabolic risk. We aimed to inves- identified. tigate the association between plasma LBP and metabolic disorders in apparently healthy Chinese. A link between chronic infection and atherosclerosis has been postulated, in RESEARCH DESIGN AND METHODS — A population-based study including 559 2 2 which lipopolysaccharide (LPS) derived overweight/obese (BMI 24.0 kg/m ) and 500 normal-weight (18.0  BMI 24.0 kg/m ) from various Gram-negative bacteria subjects aged 35–54 years was conducted in Shanghai, China. Fasting plasma glucose, lipid profile, LBP, high-sensitivity C-reactive protein, interleukin-6, high-molecular-weight (HMW) might play a pivotal role (3). Animal stud- adiponectin, leptin, hepatic enzymes, and body composition were measured. Metabolic syn- ies (4,5) and human evidence (6) sug- drome was defined by the updated National Cholesterol Education Program Adult Treatment gested that subclinical endotoxemia, Panel III criterion for Asian Americans. indicated by low to moderately elevated LPS, may be involved in the pathogenesis RESULTS — LBP levels were significantly higher in overweight/obese individuals than in normal- of metabolic disorders. Lowering LPS weight individuals (geometric mean 27.6 [95% CI 25.2–30.3] vs. 10.0 [9.1–11.1]g/ml; P 0.001). concentrations through antibiotic (4) or After multiple adjustments including BMI, the odds ratios were 3.54 (95% CI 2.05– 6.09) and 5.53 rosiglitazone therapy (6) could improve (95% CI 2.64 –11.59) for metabolic syndrome and type 2 diabetes, respectively, comparing the metabolic outcomes. LPS is a potential highest with the lowest LBP quartile. Further adjustments for inflammatory markers almost abolished factor in triggering the innate immune re- the significant association of LBP with metabolic syndrome but not that with type 2 diabetes, and controlling for adipokines and hepatic enzymes did not substantially alter the results. sponse and modulating the inflammatory cascade via stimulation of the nuclear fac- CONCLUSIONS — Elevated circulating LBP was associated with obesity, metabolic syn- tor-B pathway and transcription of drome, and type 2 diabetes in apparently healthy Chinese. These findings suggested a role of proinflammatory genes (7). However, the lipopolysaccharide via initiation of innate immune mechanism(s) in metabolic disorders. Pro- short half-life of LPS (8) and the difficulty spective studies are needed to confirm these results. of removing interference in blood (9) limit the utility of LPS testing in clinical or Diabetes Care 33:1925–1932, 2010 research settings. Lipopolysaccharide-binding protein besity, a major risk factor for meta- 2005 (1). An unhealthy diet and lifestyle- (LBP), an acute-phase protein synthe- bolic disorders, is reaching epi- associated obesogenic environment, sized in liver, initiates recognition and demic proportions worldwide with along with genetic predisposition, are the monomerization of LPS and amplifies 1.6 billion overweight/obese adults in main recognized drivers of the global ep- host responses to LPS (10). Having a rel- atively long half-life, LBP delivers LPS to ●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●● membrane and soluble forms of CD14 From the Key Laboratory of Nutrition and Metabolism, Institute for Nutritional Sciences, Shanghai Institutes for and consequently interacts with Toll-like Biological Sciences, Chinese Academy of Sciences and Graduate School of the Chinese Academy of Sciences, receptor 4, triggering a downstream sig- Shanghai, China; the Department of Health Sciences, Northeastern University, Boston, Massachusetts; the 3 4 Shanghai Municipal Center for Disease Control and Prevention, Shanghai, China; the Institut National de la naling cascade that leads to the upregula- Recherche Agronomique, Micalis, Unite ´ Mixte de Recherche 1319, Jouy en Josas, France; the Institut National tion of proinflammatory cytokines (7,10). de la Sante ´ et de la Recherche Me ´ dicale, Nutriomique, U872 team 7, Paris, France, the Universite ´ Pierre et Marie Therefore, the presence of LBP could re- Curie-Paris 6, Centre de Recherche des Cordeliers, Paris, France, and the Assistance Publique-Ho ˆ pitaux de flect an “effective” LPS level and innate Paris, Pitie ´ -Salpe ˆ trie ` re Hospital, Nutrition and Endocrinology Department, Paris, France; Departments of Nutrition and Epidemiology, Harvard School of Public Health, Boston, Massachusetts; and the SIBS-Novo immune response triggered by LPS Nordisk Translational Research Centre for PreDiabetes, Shanghai Institutes for Biological Sciences, Chinese (11,12). Academy of Sciences, Shanghai, China. Limited data have suggested that ele- Corresponding author: Xu Lin, [email protected]. vated LBP levels were observed in patients Received 20 February 2010 and accepted 29 May 2010. Published ahead of print at http://care. diabetesjournals.org on 8 June 2010. DOI: 10.2337/dc10-0340. with nonalcoholic fatty liver disease (11) © 2010 by the American Diabetes Association. Readers may use this article as long as the work is properly and were associated with unfavorable ef- cited, the use is educational and not for profit, and the work is not altered. See http://creativecommons. fects on metabolic traits in glucose- org/licenses/by-nc-nd/3.0/ for details. intolerant men (13) or with the increased The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. prevalence of coronary artery disease care.diabetesjournals.org DIABETES CARE, VOLUME 33, NUMBER 9, SEPTEMBER 2010 1925 Marker of endotoxemia and metabolic disorders (12). However, the role of LBP on obesity- tained from all participants. A total of Whole-body densitometry was con- related metabolic disorders has not been 1,059 eligible participants (559 over- ducted with the participant in light cloth- explored. Therefore, we aimed to investi- weight/obese and 500 normal-weight) ing and without carrying any metal gate whether plasma LBP concentrations were successfully recruited. objects by using a Hologic DXA (QDR- were associated with metabolic disorders Of these individuals, 960 (90.7%) 4500; Hologic, Waltham, MA). such as metabolic syndrome, insulin re- completed a dual-energy X-ray absorpti- sistance, and type 2 diabetes and also to ometry (DEXA) scan. Baseline character- Laboratory methods what extent the associations were ex- istics (BMI, waist circumference, family Fasting peripheral venous EDTA blood plained by adiposity, adipokines, hepatic history of chronic diseases, smoking, al- samples were collected and centrifuged at enzymes, and inflammation in apparently cohol drinking, and educational attain- 4°C and 3,000 rpm for 15 min. After be- healthy Chinese. ment) were similar between those with ing frozen, the samples were shipped in and without DEXA scans. However, indi- dry ice to the Institute for Nutritional Sci- RESEARCH DESIGN AND viduals who did not participate in the ences and stored at 80°C until analyses. METHODS — The Gut Microbiota DEXA scan were younger (43.0 vs. 46.2 Total cholesterol, HDL cholesterol, LDL and Obesity Study was a population- years; P  0.001) and tended to be male cholesterol, triglycerides, glucose, aspar- based case-control study among noninsti- (57.6 vs. 36.7%, P  0.001) compared tate aminotransferase (AST), alanine tutionalized residents aged 35–54 years with those who had the scan. aminotransferase (ALT), and -glutamyl- in Shanghai, China. This study investi- transferase (GGT) were measured enzy- gated the effects of gut microbiota and en- Data collection matically on an automatic analyzer vironmental factors on obesity and Home interviews were conducted by (Hitachi 7080, Tokyo, Japan) with re- related metabolic disorders. Two urban trained physicians or public health work- agents purchased from Wako Pure Chem- districts (Luwan and Zhabei) were chosen ers from the local Centers for Disease ical Industries (Osaka, Japan). Plasma to represent people with high and low so- Control and Prevention and community high-sensitivity (hs) C-reactive protein cioeconomic status in urban Shanghai. clinics. Information on demographic vari- (CRP) was measured by a particle- Participants were enrolled through their ables, health status, and behavior was enhanced immunoturbidimetric assay response to an advertisement. The field- collected using a standardized question- (Roche Diagnostics, Indianapolis, IN). work was conducted simultaneously in naire. Smoking was defined as never, cur- A1C was quantified from resolved eryth- both districts from November 2007 rent, and former. Alcohol drinking was rocytes with an automated immunoassay through January 2008. Five hundred defined as “yes” or “no.” Physical activity (Roche Diagnostics). Insulin was mea- pairs of age- and sex-matched subjects data were collected by using the Interna- sured with a completely homologous ra- (overweight/obesity) and control subjects tional Physical Activity Questionnaire dioimmunoassay (Linco Research, St. (normal-weight) were planned to be re- (short last 7-day format, http://www. Charles, MO), which had 0.2% cross- cruited. This sample size presumably ipaq.ki.se/scoring.pdf), and the level for reactivity with proinsulin. The insulin re- could provide 80% power to detect a 25% each individual was calculated as a sum of sistance index (homeostasis model difference in gut microbiota composition MET-minute/week score and then classi- assessment of insulin resistance [HOMA- between groups (14). Overweight/obesity fied as low and high by sex-specific total IR]) was calculated using updated ho- and normal weight were defined as BMI MET median. Educational attainment was meostasis model assessment methods 24.0 kg/m and 18.0  BMI 24.0 kg/ categorized into three groups (0 to 9, 10 (http://www.dtu.ox.ac.uk/). m , respectively, which was modified to 12, and 13 years of education). Fam- Plasma LBP levels were determined from the recommendation by the Work- ily history of chronic diseases was positive by a sandwich ELISA (USCN Life Science ing Group on Obesity in China (15). if a parent or sibling had coronary heart & Technology, Missouri City, TX). Eligible candidates were adult resi- disease, stroke, or type 2 diabetes. Plasma samples were diluted at least 200 dents who have lived in Shanghai for at After a home interview, all partici- times and assayed according to the man- least 10 years. Exclusion criteria included pants had a physical examination after ufacturer’s instructions. The assay has a 1) diarrhea for 3 consecutive days within overnight fasting. Body weight and height sensitivity of 0.2 ng/ml and a measurable the previous 3 months, 2) heavy alcohol were measured to the nearest 0.1 kg and concentration range of 0.78 –50 ng/ml. consumption (40 g/day ethanol for men 0.1 cm, respectively, with the participants The intra-assay and interassay coefficients and 20 g/day for women), 3) diagnosed in light indoor clothing without shoes. of variation were 5 and 10%, respec- diabetes, taking oral antidiabetic agents or BMI was calculated as weight in kilograms tively. Enzyme immunoassays were used insulin, cancer, coronary heart disease, divided by the square of height in meters. to measure plasma interleukin (IL)-6, lep- myocardial infarction, stroke, or severe Waist circumference was obtained at the tin (R&D Systems, Minneapolis, MN), kidney or liver diseases, 4) infectious dis- midpoint between the lowest rib and the and high-molecular-weight (HMW)- eases including tuberculosis, AIDS, and iliac crest to the nearest 0.1 cm, after in- adiponectin (Millipore, St. Charles, MO). hepatitis, 5) severe psychological disor- halation and exhalation. Blood pressure The interassay coefficients of variation ders or physical disabilities, 6) antibiotics was measured by using an electronic were 9.6 and 6.5% for IL-6 at 0.49 and used for 3 consecutive days within the blood pressure monitor (Omron HEM- 5.65 pg/ml, 5.4 and 3.5% for leptin at previous 3 months, 7) gastrointestinal 705CP; OMRON Healthcare, Vernon 65.7 and 581 pg/ml, and 8.1 and 3.8% for surgery within 1 year, or 8) pregnancy or Hills, IL) on the right arm of the partici- HMW-adiponectin at 21.23 and 61.50 lactation in women. The protocol was ap- pant in a comfortable sitting position after ng/ml, respectively. Hepatitis B surface proved by the Institutional Review Board at least a 5-min rest. Three measurements antigen (HBsAg) was detected by a Murex of the Institute for Nutritional Sciences were taken and the mean of the last two HBsAg ELISA kit (Murex Biotech, Dart- and written informed consent was ob- measurements was used for the analyses. ford, Kent, U.K.). 1926 DIABETES CARE, VOLUME 33, NUMBER 9, SEPTEMBER 2010 care.diabetesjournals.org Sun and Associates Definition of diseases Stata 9.2 (StataCorp, College Station, TX). Plasma LBP was also highly correlated Metabolic syndrome was defined accord- Statistical tests were two-sided and P  with inflammatory markers (hs-CRP r ing to the updated National Cholesterol 0.05 was considered statistically 0.93 and IL-6 r  0.50, both P  0.001) Education Program Adult Treatment significant. after adjustment for age and sex. These Panel III criteria for Asian-Americans correlation coefficients were attenuated (16), including at least three of the follow- RESULTS but remained statistically significant after ing components: 1) waist circumferences further adjustment for BMI (supplemen- 90 cm in men or 80 cm in women, 2) General characteristics tary Table A1). When the analyses were triglycerides 1.7 mmol/l, 3) HDL cho- The prevalence of metabolic syndrome stratified by obesity status, significant cor- lesterol 1.03 mmol/l in men or 1.30 and newly defined type 2 diabetes was relations with the metabolic traits were mmol/l in women, 4) blood pressure 42.0% (n  445) and 13.8% (n  146), more pronounced among overweight/ 130/85 mmHg or current use of antihy- respectively. The mean BMI values were obese subjects than among normal- pertensive medications, and 5) fasting 28.0  2.7 kg/m in overweight/obese weight individuals. In addition, LBP was plasma glucose 5.6 mmol/l. Type 2 di- subjects and 21.0  1.4 kg/m in normal- strongly correlated with total body fat abetes was defined as fasting plasma glu- weight subjects (P  0.001) (Table 1). mass/percentage and trunk fat mass in cose7.0 mmol/l or 2-h postload plasma Compared with normal-weight sub- sex-stratified correlation models (all P glucose11.1 mmol/l during an oral glu- jects, overweight/obese individuals were 0.001) (data not show). cose tolerance test. The selection proce- more likely to have lower educational at- dure for the oral glucose tolerance test is tainment and higher prevalence of meta- Associations of LBP concentrations described in supplementary Fig. A1 bolic syndrome and type 2 diabetes (all with metabolic syndrome, type 2 (available in an online appendix at http:// P  0.01) (Table 1). They also had higher diabetes, and insulin resistance care.diabetesjournals.org/cgi/content/ values for waist circumference, blood The risk for metabolic syndrome in the full/dc10-0340/DC1). Elevated hepatic pressure, glucose, A1C, insulin, HOMA- whole study sample increased progres- enzymes were defined as the highest IR, total cholesterol, LDL cholesterol, and sively across the LBP quartiles (P trend quartiles of one or more hepatic enzymes: triglycerides and lower HDL cholesterol 0.001) (Table 2) and those in the highest AST 30 IU/l, ALT 33 IU/l, or GGT concentration (all P  0.05). Meanwhile, LBP quartile had an OR of 3.54 (95% CI 42 IU/l. they exhibited higher levels of total body 2.05– 6.09) compared with those in the fat mass/percentage and trunk fat mass lowest quartile (model 2), after adjust- Statistical analyses (all P  0.001), hepatic enzymes (AST, ment for age, sex, lifestyle factors, family Log transformations were performed for ALT, and GGT), inflammatory markers history of chronic diseases, and BMI. Sim- LBP, triglycerides, insulin, HOMA-IR, in- (hs-CRP and IL-6) and leptin, accompa- ilar trends were also observed for the met- flammatory makers, adipokines, and he- nied by lower concentrations of HMW- abolic syndrome components. Further patic enzymes to approximate normality. adiponectin (all P  0.001). Importantly, adjustment for inflammatory markers ANCOVA for continuous variables and plasma LBP levels were significantly (model 3) abolished the significant asso- logistic regression models for categorical higher in overweight/obese participants ciations for metabolic syndrome and most variables were applied for the comparison than in normal-weight participants (geo- of its components except for hypertriglyc- across obesity status. Multiple compari- metric mean 27.6 [95% CI 25.2–30.3] vs. eridemia and low HDL cholesterol. son corrections were performed using the 10.0 [9.1–11.1] g/ml; P  0.001). In For those with newly defined type 2 Benjamini and Hochberg procedure (17). addition, LBP levels were higher in sub- diabetes, the OR in the highest quartile Partial Spearman correlation coefficients jects with impaired fasting glucose than in was 5.53 (95% CI 2.64 –11.59) compared between LBP and various parameters those with normal fasting glucose (16.7 with that in the lowest LBP quartile (P trend were calculated by adjustment for age, [15.0 –18.6] vs. 13.8 [12.2–15.6] g/ml;  0.001) in the multivariable model (Ta- sex, and BMI. Individuals with presum- P  0.041) and higher in subjects with ble 2, model 2). A positive association be- ably acute inflammation (hs-CRP 10 impaired glucose tolerance than in those tween LBP and insulin resistance, mg/l) were excluded (n 18, 1.7%), leav- with normal glucose tolerance (31.2 represented by the highest quartile of ing 1,041 subjects in the analysis. Of [24.3– 40.1] vs. 14.4 [12.8 –16.3] g/ml; HOMA-IR, was also observed in nondia- these, 942 subjects completed a DEXA P  0.001), after adjustment for age and betic participants (OR 1.90 [95% CI scan. sex. 1.10 –3.28]). The ORs for diabetes and Participants were classified into four insulin resistance were slightly attenuated groups according to their LBP quartiles in Correlation between LBP but remained statistically significant after the whole sample. Multivariate logistic re- concentrations and metabolic further adjustment for hs-CRP and IL-6 gression models were used to estimate the parameters (model 3). odds ratios (ORs) for metabolic syn- LBP was positively correlated with BMI, The associations between LBP and drome, type 2 diabetes, and insulin resis- waist circumference (all P  0.001) (sup- metabolic disorders were not attenuated tance. Potential confounding variables plementary Table A1, available in an on- by additionally controlling for HMW- include age, sex, lifestyle factors, educa- line appendix), blood pressure, total adiponectin, leptin, and elevated hepatic tion level, family history of chronic dis- cholesterol, LDL cholesterol, triglycer- enzymes (as categorical or continuous eases, and BMI (or total body fat). In ides, glucose, insulin, HOMA-IR, hepatic variables) in model 2 (data not shown). addition, we adjusted for inflammatory enzymes, and leptin and negatively corre- Replacing BMI with total body fat mass in markers, adipokines, and elevated he- lated with HDL cholesterol and HMW- the multiple regression models also did patic enzymes. Data management and sta- adiponectin (all P  0.05, data not not materially change the magnitude of tistical analyses were performed using shown), after adjustment for age and sex. the associations. The results remained care.diabetesjournals.org DIABETES CARE, VOLUME 33, NUMBER 9, SEPTEMBER 2010 1927 Marker of endotoxemia and metabolic disorders Table 1—Characteristics of participants according to obese status significant interactions were observed be- tween LBP and these factors (P  0.05 for all interaction tests). Normal weight Overweight/obesity 2 2 (18.0  BMI 24.0 kg/m ) (BMI 24.0 kg/m ) P value* CONCLUSIONS — Our data showed n 500 559 significant associations between elevated Age (years)† 45.8  5.5 46.0  5.4 0.604 LBP concentrations and the risk for met- BMI (kg/m ) 21.0  1.4 28.0  2.7 0.001‡ abolic syndrome, insulin resistance, and Men† 175 (35.0) 234 (41.9) 0.022‡ type 2 diabetes independent of conven- Physical inactivity 249 (49.8) 281 (50.3) 0.869 tional cardiovascular risk factors in an ap- Education levels 0.002‡ parently healthy Chinese population. 0–9 years 114 (22.8) 175 (31.3) Further adjustment for inflammatory fac- 10–12 years 273 (54.6) 281 (50.3) tors, but not for adipokines and elevated 12 years 113 (22.6) 103 (18.4) hepatic enzymes, substantially attenuated Current smoker (yes) 117 (23.4) 158 (28.3) 0.767 the associations for metabolic syndrome Alcohol drinker (yes) 183 (36.6) 204 (36.5) 0.245 and most of its components, suggesting Family history of chronic that chronic inflammation may mediate diseases 200 (40.0) 222 (39.7) 0.939 the effects of innate immune response in- Newly defined type 2 diabetes 36 (7.2) 110 (19.7) 0.001‡ duced by LPS-LBP. Metabolic syndrome 51 (10.2) 394 (70.5) 0.001‡ In a previous human study, Ghanim Waist circumference (cm) 75.9  6.1 93.2  8.2 0.001‡ Systolic blood pressure et al. (18) reported that increased plasma (mmHg) 118.4  15.3 130.9  17.6 0.001‡ LPS and LBP and also expression of Toll- Diastolic blood pressure like receptors and suppressor of cytokine (mmHg) 74.7  9.8 84.0  11.6 0.001‡ signaling-3 in mononuclear cells (MNCs) Glucose (mmol/l) 5.81  1.13 6.30  1.54 0.001‡ could be induced by consuming a meal A1C (%) 5.58  0.62 5.79  0.78 0.001‡ with a high-fat, high-carbohydrate con- Insulin (U/ml) 7.3 (7.07.7) 11.3 (10.911.8) 0.001‡ tent, but not with isocaloric fruit and HOMA-IR 0.85 (0.820.89) 1.33 (1.271.38) 0.001‡ fiber, implying a potential role of the Total cholesterol (mmol/l) 5.16  1.14 5.33  1.18 0.014‡ LPS-LBP pathway in postprandial inflam- LDL cholesterol (mmol/l) 3.14  0.93 3.41  0.99 0.001‡ mation and related metabolic disorders. HDL cholesterol (mmol/l) 1.53  0.44 1.23  0.34 0.001‡ In addition, positive correlations between Triglycerides (mmol/l) 1.01 (0.961.06) 1.61 (1.531.69) 0.001‡ LBP and metabolic traits such as BMI, di- DEXA scan 456 (91.2) 504 (90.2) 0.697 astolic blood pressure, fasting glucose, in- Total body fat mass (kg) sulin, and triglycerides were observed in (n  960) 60 men with glucose intolerance (13). Men 11.8  3.3 20.7  4.3 0.001‡ Moreover, a higher LBP level was associ- Women 15.7  2.8 25.7  5.1 0.001‡ ated with increased prevalence of coro- Total body fat (%) (n  960) nary artery disease independent of Men 19.0  4.3 25.6  3.4 0.001‡ established cardiovascular risk factors in Women 29.4  3.8 36.1  3.6 0.001‡ 247 male patients (12). With a relatively Trunk fat mass (kg) (n  960) large sample size of apparently healthy Men 6.3  2.1 12.1  2.7 0.001‡ men and women, our study provides Women 7.7  1.8 13.7  3.0 0.001‡ more convincing evidence about the rela- HBsAg carriage 44 (8.8) 50 (8.9) 0.916 tionship between LBP and metabolic Elevated hepatic enzymes 141 (28.2) 294 (52.6) 0.001‡ abnormalities. AST (IU/l) 23.6 (22.9–24.3) 26.3 (25.4–27.2) 0.001‡ In recent years, the effects of micro- ALT (IU/l) 19.1 (18.2–20.0) 27.8 (26.5–29.2) 0.001‡ biota on health have attracted increasing GGT (IU/l) 22.5 (21.3–23.8) 33.1 (31.2–35.0) 0.001‡ attention, and low-grade endotoxemia or hs-CRP (mg/l) 0.61 (0.57–0.66) 1.38 (1.28–1.49) 0.001‡ LPS was found to link to various meta- IL-6 (pg/ml) 1.19 (1.13–1.26) 1.67 (1.59–1.76) 0.001‡ bolic consequences. However, most stud- HMW-adiponectin (g/ml) 3.16 (2.93–3.41) 1.88 (1.73–2.04) 0.001‡ ies have been performed in mice and few Leptin (ng/ml) 3.99 (3.70–4.31) 8.94 (8.40–9.51) 0.001‡ in human populations. Studies in mice LBP (g/ml) 10.0 (9.1–11.1) 27.6 (25.2–30.3) 0.001‡ demonstrated that two- to threefold in- Data are arithmetic mean  SD, n (%), or geometric mean (95% CI). n  1,059. Percentages may not sum to 100 because of rounding. *P value was calculated after adjustment for age and sex; P value for body fat creased circulating LPS induced by a comparison was not adjusted for sex. †Data not adjusted for itself. ‡Benjamini and Hochberg– corrected high-fat diet or LPS infusion led to in- statistical significance (17). creased levels of fasting glucose and insu- lin and body weight gain (4,5). The similar after exclusion of HBsAg-positive tion analyses to examine whether obesity, occurrences of a metabolic response subjects (n  91, 8.7%) or subjects with trunk fat mass, HMW-adiponectin, and could be counteracted by a CD14 mutant 18.0 BMI 18.5 kg/m (n  14, 1.3%) elevated hepatic enzymes modified the as- (5) or improved by changing gut micro- (data not shown). sociations of LBP with metabolic syn- biota (4). In humans, a high LPS concen- We further conducted joint classifica- drome and type 2 diabetes (Fig. 1). No tration was found in individuals with type 1928 DIABETES CARE, VOLUME 33, NUMBER 9, SEPTEMBER 2010 care.diabetesjournals.org Sun and Associates Table 2—ORs (95% CI) for metabolic syndrome, type 2 diabetes, and insulin resistance according to quartiles of LBP Quartile of LBP Q1 (LBP 6.5 Q2 (6.5  LBP Q3 (15.8  LBP Q4 (LBP 42.0 g/ml) 15.8 g/ml) 42.0 g/ml) g/ml) P trend Metabolic syndrome (n  1,041) 35/260 91/260 132/261 176/260 Model 1 1 3.45 (2.22–5.37) 6.25 (4.05–9.65) 12.90 (8.28–20.10) 0.001 Model 2 1 2.78 (1.64–4.72) 3.04 (1.81–5.12) 3.54 (2.05–6.09) 0.001 Model 3 1 2.56 (1.50–4.38) 2.38 (1.36–4.15) 1.80 (0.84–3.88) 0.053 Central obesity 59/260 110/260 164/261 204/260 Model 1 1 2.50 (1.71–3.66) 5.92 (4.01–8.72) 12.80 (8.42–19.47) 0.001 Model 2 1 1.64 (0.85–3.16) 2.54 (1.30–4.97) 2.53 (1.18–5.43) 0.005 Model 3 1 1.49 (0.77–2.90) 1.88 (0.90–3.92) 1.09 (0.35–3.35) 0.304 Elevated blood pressure 66/260 89/260 119/261 157/260 Model 1 1 1.51 (1.02–2.23) 2.24 (1.53–3.28) 4.19 (2.85–6.15) 0.001 Model 2 1 1.16 (0.76–1.76) 1.32 (0.87–2.00) 1.73 (1.11–2.69) 0.013 Model 3 1 1.09 (0.71–1.67) 1.08 (0.69–1.68) 0.94 (0.50–1.75) 0.982 Hypertriglyceridemia 25/260 70/260 102/261 130/260 Model 1 1 3.61 (2.18–5.99) 5.82 (3.56–9.52) 9.34 (5.72–15.25) 0.001 Model 2 1 3.11 (1.85–5.24) 4.05 (2.43–6.76) 4.98 (2.94–8.43) 0.001 Model 3 1 3.04 (1.80–5.12) 3.77 (2.23–6.38) 4.10 (2.14–7.85) 0.001 Low HDL cholesterol 65/260 82/260 116/261 118/260 Model 1 1 1.43 (0.97–2.11) 2.64 (1.81–3.86) 2.77 (1.89–4.04) 0.001 Model 2 1 1.17 (0.78–1.75) 1.72 (1.15–2.58) 1.39 (0.90–2.15) 0.056 Model 3 1 1.18 (0.78–1.76) 1.76 (1.15–2.68) 1.54 (0.87–2.72) 0.027 Hyperglycemia 140/260 157/260 174/261 188/260 Model 1 1 1.28 (0.90–1.82) 1.64 (1.14–2.35) 2.14 (1.48–3.09) 0.001 Model 2 1 1.18 (0.82–1.69) 1.34 (0.92–1.97) 1.49 (0.99–2.27) 0.047 Model 3 1 1.08 (0.75–1.56) 1.00 (0.66–1.52) 0.60 (0.32–1.13) 0.382 Type 2 diabetes (n  1,041) 10/260 22/260 43/261 67/260 Model 1 1 2.29 (1.06–4.95) 4.71 (2.30–9.61) 8.30 (4.15–16.59) 0.001 Model 2 1 1.99 (0.91–4.33) 3.53 (1.69–7.37) 5.53 (2.64–11.59) 0.001 Model 3 1 1.93 (0.88–4.21) 3.06 (1.45–6.46) 3.23 (1.38–7.58) 0.003 Q1 (LBP 5.9 Q2 (5.9  LBP Q3 (12.7  LBP Q4 (LBP 37.8 g/ml) 12.7 g/ml) 37.8 g/ml) g/ml) P trend Insulin resistance (n  899) 27/224 38/225 72/225 87/225 Model 1 1 1.49 (0.87–2.54) 3.48 (2.13–5.70) 4.64 (2.85–7.55) 0.001 Model 2 1 1.12 (0.64–1.98) 1.95 (1.14–3.32) 1.90 (1.10–3.28) 0.005 Model 3 1 1.14 (0.64–2.01) 1.97 (1.14–3.41) 1.95 (0.97–3.92) 0.014 Model 1: adjusted for age and sex. Model 2: further adjusted for smoking (current, yes or no), alcohol drinking (yes or no), physical activity (low or high by the sex-specific MET/week median), education (0 –9, 10 –12, and 13 years), family history of chronic diseases (yes or no), and BMI. Model 3: further adjusted for hsCRP and IL-6. 2 diabetes (6). The administration of ros- We observed a stronger correlation plasma HMW-adiponectin contributed iglitazone, a hypoglycemic agent with a between LBP and inflammatory markers additional risk for metabolic syndrome potential anti-inflammatory effect (19), than that between LBP and adipokines af- and type 2 diabetes under the high LBP could reduce LPS levels (6). However, the ter adjustment for BMI. Moreover, adjust- condition (Fig. 1E and F). In fact, existing short half-life of LPS (8) and the disadvan- ing for hs-CRP and IL-6 almost eliminated data showed that LPS increased plasma tages associated with its assay (9) have the associations of LBP with metabolic leptin levels and downregulated adi- limited its potential applications in rou- syndrome and most of its traits (Table 2, ponectin receptor mRNA in healthy vol- tine clinical settings and large-scale stud- model 3), whereas controlling for HMW- unteers (20). ies. Having a relatively long half-life and adiponectin and leptin had little impact We also examined whether adjust- reliable measurements, LBP might serve on the associations. A potential mecha- ment for hepatic enzymes altered the as- as a marker reflecting an “effective” LPS nism is that LPS-LBP–triggered immune sociations, because LBP is synthesized in level and innate immune response trig- response activates the nuclear factor-B liver and LPS could cause hepatocyte gered by LPS (11,12). Therefore, circulat- pathway, stimulates formation of IL-6, damage by inducing upregulation of tu- ing LBP levels can be considered a IL-1, and tumor necrosis factor- (7), and mor necrosis factor- expression in promising early biomarker and interven- upregulates CRP synthesis in hepato- Kupffer cells (7). Previous prospective tion target for inflammatory conditions. cytes. However, it is noteworthy that low studies also suggested that elevated he- care.diabetesjournals.org DIABETES CARE, VOLUME 33, NUMBER 9, SEPTEMBER 2010 1929 Marker of endotoxemia and metabolic disorders Figure 1—ORs for metabolic disorders according to joint classification of LBP and obesity status (A and B), trunk fat (sex and obesity-stratified tertile [T], C and D), HMW-adiponectin (E and F), and hepatic enzymes (G and H). A–D: Modified metabolic syndrome was defined as having two or more components of metabolic syndrome without central obesity. Adjusted for age, sex, smoking, alcohol drinking, physical activity, education, and family history of chronic diseases. E–H: Adjusted for age, sex, smoking, alcohol drinking, physical activity, education, family history of chronic diseases, and BMI. 1930 DIABETES CARE, VOLUME 33, NUMBER 9, SEPTEMBER 2010 care.diabetesjournals.org Sun and Associates who.int/mediacentre/factsheets/fs311/en/ patic enzymes predicted risk of metabolic by applying strict exclusion criteria and print.html/. Accessed 26 October 2007 syndrome and type 2 diabetes (21). How- also included a relatively large sample size 2. Dandona P, Aljada A, Chaudhuri A, Mo- ever, no material change was observed with both sexes. Certainly, these results hanty P, Garg R. Metabolic syndrome: a when we controlled for elevated hepatic should be examined prospectively and in comprehensive perspective based on in- enzymes or excluded individuals having different populations to establish the teractions between obesity, diabetes, hepatitis B infection. Collectively, our causal relationship between LBP, inflam- and inflammation. Circulation 2005;111: findings indicate that the inflammatory mation, and metabolic outcomes. 1448 –1454 cascade initiated by LPS-LBP triggered an Our study indicates that LBP is signif- 3. Wiedermann CJ, Kiechl S, Dunzendorfer innate immune response, which is a ma- icantly associated with obesity-related S, Schratzberger P, Egger G, Oberhollen- jor mechanism linking LBP to the patho- metabolic disorders in apparently healthy zer F, Willeit J. Association of endotox- emia with carotid atherosclerosis and genesis of metabolic disorders. Chinese. These findings indicate the role cardiovascular disease: prospective re- Interestingly, unlike the case for met- of LPS-initiated innate immune mecha- sults from the Bruneck Study. J Am Coll abolic syndrome risk, controlling for in- nisms in metabolic diseases. Future pro- Cardiol 1999;34:1975–1981 flammatory markers in current study did spective studies are needed to clarify 4. Cani PD, Bibiloni R, Knauf C, Waget A, not alter the significant associations of whether LBP predicts future risk of meta- Neyrinck AM, Delzenne NM, Burcelin R. LBP with dyslipidemia, insulin resistance, bolic disorders. Changes in gut microbiota control meta- and type 2 diabetes (Table 2). The minor bolic endotoxemia-induced inflammation effect of inflammation on the association in high-fat diet-induced obesity and dia- Acknowledgments — This study was sup- between LBP and dyslipidemia might be betes in mice. Diabetes 2008;57:1470 – ported by research grants from the Chinese explained by a sequence identity shared Academy of Sciences (KSCX2-YW-R-116), between LBP and lipid transfer proteins 5. Cani PD, Amar J, Iglesias MA, Poggi M, the Ministry of Science and Technology of Knauf C, Bastelica D, Neyrinck AM, Fava that could modify lipid homeostasis dur- China (2008DFA31960, 2008AA02Z315, F, Tuohy KM, Chabo C, Waget A, Delme ´e ing subclinical endotoxemia (22). Previ- 2009AA022704, and 2007AA027332), the E, Cousin B, Sulpice T, Chamontin B, Fer- ous studies demonstrated that Science and Technology Commission of rie ` res J, Tanti JF, Gibson GR, Casteilla L, hyperinsulinemia might affect immune Shanghai Municipality (075407001), the Delzenne NM, Alessi MC, Burcelin R. competence, including functions of neu- Shanghai Institutes for Biological Sciences, the Metabolic endotoxemia initiates obesity trophils (13) and hepatic Kupffer cells Chinese Academy of Sciences (SIBS2008006), and insulin resistance. Diabetes 2007;56: (23), which subsequently influenced LPS the National Natural Science Foundation of 1761–1772 China (30930081), and the Novo Nordisk clearance and LBP synthesis. Neverthe- 6. Creely SJ, McTernan PG, Kusminski CM, A/S. J.D. and K.C. also acknowledge the less, endotoxemia might induce insulin Fisher M, Da Silva NF, Khanolkar M, French National Agency for Research (ANR resistance, whereas CD14 mutant mice Evans M, Harte AL, Kumar S. Lipopoly- MicroObes program). saccharide activates an innate immune showed hypersensitivity to insulin (5). The sponsors were not involved in the system response in human adipose tissue MNCs were also suggested to be a target study design, data collection, analysis, or in obesity and type 2 diabetes. Am J of insulin action and involved in the in- interpretation. Physiol Endocrinol Metab 2007;292: teraction between inflammation and in- No other potential conflicts of interest rele- E740 –E747 sulin resistance (24). The increased vant to this article were reported. 7. Siebler J, Galle PR, Weber MM. The gut- suppressor of cytokine signaling-3 ex- L.S. contributed to the study design, re- liver-axis: endotoxemia, inflammation, pression in MNCs, accompanied by in- searched data, contributed to discussion, and insulin resistance and NASH. J Hepatol wrote the manuscript. Z.Y. contributed to the creased LPS and LBP concentrations after 2008;48:1032–1034 study design, researched data, contributed to a high-fat, high-carbohydrate meal, might 8. Munford RS. Detoxifying endotoxin: discussion, and reviewed/edited the manu- also interfere with insulin signaling and time, place and person. J Endotoxin Res script. X.Y., S.Z., and H.L. contributed to the play a role in insulin resistance (18,24). 2005;11:69 – 84 study design, researched data, and reviewed/ 9. Novitsky TJ. Limitations of the Limulus Our data suggested that LBP might exert edited the manuscript. D.Y. and H.W. re- amebocyte lysate test in demonstrating its influence directly on insulin action in searched data and reviewed/edited the circulating lipopolysaccharides. Ann NY addition to stimulating a downstream manuscript. Y.C. contributed to the study de- Acad Sci 1998;851:416 – 421 pathway generating inflammatory cyto- sign and reviewed/edited the manuscript. J.D. 10. Weiss J. Bactericidal/permeability-in- kines. Moreover, the LBP gene was re- reviewed/edited the manuscript. K.C. wrote creasing protein (BPI) and lipopolysac- cently found to be genetically susceptible the manuscript. F.B.H. and X.L. contributed charide-binding protein (LBP): structure, to the conception and design of the study, to type 2 diabetes in Japanese (25), which function and regulation in host defence contributed to discussion, reviewed/edited may provide one possible reason that the against Gram-negative bacteria. Biochem the manuscript. conventionally inflammatory cytokine Soc Trans 2003;31:785–790 We are grateful to An Pan, Qibin Qi, Ling showed little impact on the association 11. Ruiz AG, Casafont F, Crespo J, Cayo ´n A, Lu, Chen Liu, Geng Zhang, Geng Zong, Sha- Mayorga M, Estebanez A, Fernadez-Es- between LBP and type 2 diabetes. ojie Ma, He Zheng, and the local Center for calante JC, Pons-Romero F. Lipopolysac- To our knowledge, this is the first Disease Control and Prevention staff of Shang- charide-binding protein plasma levels study that systematically investigated the hai for their kind help at various stages of this and liver TNF- gene expression in obese associations of LBP concentrations with study. patients: evidence for the potential role of the risk of obesity-related metabolic dis- endotoxin in the pathogenesis of non-al- orders. Admittedly, the cross-sectional coholic steatohepatitis. Obes Surg 2007; nature of this study does not allow for a References 17:1374 –1380 causal inference. Nonetheless, we tried to 1. World Health Organization. Obesity and 12. Lepper PM, Schumann C, Triantafilou K, eliminate potential effects of acute inflam- overweight: Fact Sheet No. 311 [article on- Rasche FM, Schuster T, Frank H, Schneider mation and other potential confounders line], 2006. Available from http://www. EM, Triantafilou M, von Eynatten M. Asso- care.diabetesjournals.org DIABETES CARE, VOLUME 33, NUMBER 9, SEPTEMBER 2010 1931 Marker of endotoxemia and metabolic disorders ciation of lipopolysaccharide-binding pro- Golani I. Controlling the false discovery the Mexico City Diabetes study. Diabetes tein and coronary artery disease in men. rate in behavior genetics research. Behav Care 2005;28:1757–1762 J Am Coll Cardiol 2007;50:25–31 Brain Res 2001;125:279 –284 22. Hudgins LC, Parker TS, Levine DM, Gor- 13. Gubern C, Lo ´ pez-Bermejo A, Biarne ´s J, 18. Ghanim H, Abuaysheh S, Sia CL, Korze- don BR, Saal SD, Jiang XC, Seidman CE, Vendrell J, Ricart W, Ferna ´ ndez-Real JM. niewski K, Chaudhuri A, Fernandez-Real Tremaroli JD, Lai J, Rubin AL. A single Natural antibiotics and insulin sensitiv- JM, Dandona P. Increase in plasma endo- intravenous dose of endotoxin rapidly al- ity: the role of bactericidal/permeabili- toxin concentrations and the expression ters serum lipoproteins and lipid transfer ty-increasing protein. Diabetes 2006; of Toll-like receptors and suppressor of proteins in normal volunteers. J Lipid Res 55:216 –224 cytokine signaling-3 in mononuclear cells 2003;44:1489 –1498 14. Li M, Wang B, Zhang M, Rantalainen M, after a high-fat, high-carbohydrate meal: 23. Cornell RP, McClellan CC. Modulation Wang S, Zhou H, Zhang Y, Shen J, Pang X, implications for insulin resistance. Diabe- of hepatic reticuloendothelial system Zhang M, Wei H, Chen Y, Lu H, Zuo J, Su tes Care 2009;32:2281–2287 phagocytosis by pancreatic hormones. M, Qiu Y, Jia W, Xiao C, Smith LM, Yang S, 19. Mohanty P, Aljada A, Ghanim H, J Reticuloendothel Soc 1982;32:397– Holmes E, Tang H, Zhao G, Nicholson JK, Hofmeyer D, Tripathy D, Syed T, Al- 407 Li L, Zhao L. Symbiotic gut microbes mod- Haddad W, Dhindsa S, Dandona P. Ev- 24. Ghanim H, Aljada A, Daoud N, Deo- ulate human metabolic phenotypes. Proc idence for a potent antiinflammatory purkar R, Chaudhuri A, Dandona P. Role Natl Acad Sci USA 2008;105:2117–2122 effect of rosiglitazone. J Clin Endocrinol of inflammatory mediators in the sup- 15. Wu Y. Overweight and obesity in China. Metab 2004;89:2728 –2735 pression of insulin receptor phosphoryla- BMJ 2006;333:362–363 20. Anderson PD, Mehta NN, Wolfe ML, tion in circulating mononuclear cells of 16. Grundy SM, Cleeman JI, Daniels SR, Do- Hinkle CC, Pruscino L, Comiskey LL, obese subjects. Diabetologia 2007;50: nato KA, Eckel RH, Franklin BA, Gordon Tabita-Martinez J, Sellers KF, Rickels MR, 278 –285 DJ, Krauss RM, Savage PJ, Smith SC Jr, Ahima RS, Reilly MP. Innate immunity 25. Takeuchi F, Yanai K, Inomata H, Kuzuya Spertus JA, Costa F. Diagnosis and manage- modulates adipokines in humans. J Clin N, Kajio H, Honjo S, Takeda N, Kaburagi ment of the metabolic syndrome: an Amer- Endocrinol Metab 2007;92:2272–2279 Y, Yasuda K, Shirasawa S, Sasazuki T, ican Heart Association/National Heart, 21. Nannipieri M, Gonzales C, Baldi S, Posa- Kato N. Search of type 2 diabetes suscep- Lung, and Blood Institute scientific state- das R, Williams K, Haffner SM, Stern MP, tibility gene on chromosome 20q. Bio- ment. Circulation 2005;112:e285– e290 Ferrannini E. Liver enzymes, the meta- chem Biophys Res Commun 2007;357: 17. Benjamini Y, Drai D, Elmer G, Kafkafi N, bolic syndrome, and incident diabetes: 1100 –1106 1932 DIABETES CARE, VOLUME 33, NUMBER 9, SEPTEMBER 2010 care.diabetesjournals.org http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Diabetes Care Pubmed Central

A Marker of Endotoxemia Is Associated With Obesity and Related Metabolic Disorders in Apparently Healthy Chinese

Download PDF
8 pages

Loading next page...
 
/lp/pubmed-central/a-marker-of-endotoxemia-is-associated-with-obesity-and-related-u9vhKrpSUu

References (58)

Publisher
Pubmed Central
Copyright
© 2010 by the American Diabetes Association.
ISSN
0149-5992
eISSN
1935-5548
DOI
10.2337/dc10-0340
Publisher site
See Article on Publisher Site

Abstract

Cardiovascular and Metabolic Risk ORIGINAL ARTICLE A Marker of Endotoxemia Is Associated With Obesity and Related Metabolic Disorders in Apparently Healthy Chinese 1 1 LIANG SUN, MSC HONGYU WU, MSC idemic of obesity and associated meta- 1,7 1 ZHIJIE YU, MD, PHD YAN CHEN, MD, PHD bolic disorders. However, increasing 2 4 XINGWANG YE, PHD JOEL DORE, PHD evidence supports the fact that chronic 3 5 SHURONG ZOU, MD, MSC KARINE CLEMENT ´ , MD, PHD low-grade inflammation is one of the key 1 6 HUAIXING LI, PHD FRANK B. HU, MD, PHD mechanisms underlying the pathogenesis 1 1 DANXIA YU, MSC XU LIN, MD, PHD of obesity-related metabolic disorders (2). However, the agents responsible for initi- ating and sustaining this low-grade in- OBJECTIVE — Elevated lipopolysaccharide-binding protein (LBP), a marker of subclinical en- flammatory signal have yet to be dotoxemia, may be involved in the pathogenesis of obesity and metabolic risk. We aimed to inves- identified. tigate the association between plasma LBP and metabolic disorders in apparently healthy Chinese. A link between chronic infection and atherosclerosis has been postulated, in RESEARCH DESIGN AND METHODS — A population-based study including 559 2 2 which lipopolysaccharide (LPS) derived overweight/obese (BMI 24.0 kg/m ) and 500 normal-weight (18.0  BMI 24.0 kg/m ) from various Gram-negative bacteria subjects aged 35–54 years was conducted in Shanghai, China. Fasting plasma glucose, lipid profile, LBP, high-sensitivity C-reactive protein, interleukin-6, high-molecular-weight (HMW) might play a pivotal role (3). Animal stud- adiponectin, leptin, hepatic enzymes, and body composition were measured. Metabolic syn- ies (4,5) and human evidence (6) sug- drome was defined by the updated National Cholesterol Education Program Adult Treatment gested that subclinical endotoxemia, Panel III criterion for Asian Americans. indicated by low to moderately elevated LPS, may be involved in the pathogenesis RESULTS — LBP levels were significantly higher in overweight/obese individuals than in normal- of metabolic disorders. Lowering LPS weight individuals (geometric mean 27.6 [95% CI 25.2–30.3] vs. 10.0 [9.1–11.1]g/ml; P 0.001). concentrations through antibiotic (4) or After multiple adjustments including BMI, the odds ratios were 3.54 (95% CI 2.05– 6.09) and 5.53 rosiglitazone therapy (6) could improve (95% CI 2.64 –11.59) for metabolic syndrome and type 2 diabetes, respectively, comparing the metabolic outcomes. LPS is a potential highest with the lowest LBP quartile. Further adjustments for inflammatory markers almost abolished factor in triggering the innate immune re- the significant association of LBP with metabolic syndrome but not that with type 2 diabetes, and controlling for adipokines and hepatic enzymes did not substantially alter the results. sponse and modulating the inflammatory cascade via stimulation of the nuclear fac- CONCLUSIONS — Elevated circulating LBP was associated with obesity, metabolic syn- tor-B pathway and transcription of drome, and type 2 diabetes in apparently healthy Chinese. These findings suggested a role of proinflammatory genes (7). However, the lipopolysaccharide via initiation of innate immune mechanism(s) in metabolic disorders. Pro- short half-life of LPS (8) and the difficulty spective studies are needed to confirm these results. of removing interference in blood (9) limit the utility of LPS testing in clinical or Diabetes Care 33:1925–1932, 2010 research settings. Lipopolysaccharide-binding protein besity, a major risk factor for meta- 2005 (1). An unhealthy diet and lifestyle- (LBP), an acute-phase protein synthe- bolic disorders, is reaching epi- associated obesogenic environment, sized in liver, initiates recognition and demic proportions worldwide with along with genetic predisposition, are the monomerization of LPS and amplifies 1.6 billion overweight/obese adults in main recognized drivers of the global ep- host responses to LPS (10). Having a rel- atively long half-life, LBP delivers LPS to ●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●● membrane and soluble forms of CD14 From the Key Laboratory of Nutrition and Metabolism, Institute for Nutritional Sciences, Shanghai Institutes for and consequently interacts with Toll-like Biological Sciences, Chinese Academy of Sciences and Graduate School of the Chinese Academy of Sciences, receptor 4, triggering a downstream sig- Shanghai, China; the Department of Health Sciences, Northeastern University, Boston, Massachusetts; the 3 4 Shanghai Municipal Center for Disease Control and Prevention, Shanghai, China; the Institut National de la naling cascade that leads to the upregula- Recherche Agronomique, Micalis, Unite ´ Mixte de Recherche 1319, Jouy en Josas, France; the Institut National tion of proinflammatory cytokines (7,10). de la Sante ´ et de la Recherche Me ´ dicale, Nutriomique, U872 team 7, Paris, France, the Universite ´ Pierre et Marie Therefore, the presence of LBP could re- Curie-Paris 6, Centre de Recherche des Cordeliers, Paris, France, and the Assistance Publique-Ho ˆ pitaux de flect an “effective” LPS level and innate Paris, Pitie ´ -Salpe ˆ trie ` re Hospital, Nutrition and Endocrinology Department, Paris, France; Departments of Nutrition and Epidemiology, Harvard School of Public Health, Boston, Massachusetts; and the SIBS-Novo immune response triggered by LPS Nordisk Translational Research Centre for PreDiabetes, Shanghai Institutes for Biological Sciences, Chinese (11,12). Academy of Sciences, Shanghai, China. Limited data have suggested that ele- Corresponding author: Xu Lin, [email protected]. vated LBP levels were observed in patients Received 20 February 2010 and accepted 29 May 2010. Published ahead of print at http://care. diabetesjournals.org on 8 June 2010. DOI: 10.2337/dc10-0340. with nonalcoholic fatty liver disease (11) © 2010 by the American Diabetes Association. Readers may use this article as long as the work is properly and were associated with unfavorable ef- cited, the use is educational and not for profit, and the work is not altered. See http://creativecommons. fects on metabolic traits in glucose- org/licenses/by-nc-nd/3.0/ for details. intolerant men (13) or with the increased The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. prevalence of coronary artery disease care.diabetesjournals.org DIABETES CARE, VOLUME 33, NUMBER 9, SEPTEMBER 2010 1925 Marker of endotoxemia and metabolic disorders (12). However, the role of LBP on obesity- tained from all participants. A total of Whole-body densitometry was con- related metabolic disorders has not been 1,059 eligible participants (559 over- ducted with the participant in light cloth- explored. Therefore, we aimed to investi- weight/obese and 500 normal-weight) ing and without carrying any metal gate whether plasma LBP concentrations were successfully recruited. objects by using a Hologic DXA (QDR- were associated with metabolic disorders Of these individuals, 960 (90.7%) 4500; Hologic, Waltham, MA). such as metabolic syndrome, insulin re- completed a dual-energy X-ray absorpti- sistance, and type 2 diabetes and also to ometry (DEXA) scan. Baseline character- Laboratory methods what extent the associations were ex- istics (BMI, waist circumference, family Fasting peripheral venous EDTA blood plained by adiposity, adipokines, hepatic history of chronic diseases, smoking, al- samples were collected and centrifuged at enzymes, and inflammation in apparently cohol drinking, and educational attain- 4°C and 3,000 rpm for 15 min. After be- healthy Chinese. ment) were similar between those with ing frozen, the samples were shipped in and without DEXA scans. However, indi- dry ice to the Institute for Nutritional Sci- RESEARCH DESIGN AND viduals who did not participate in the ences and stored at 80°C until analyses. METHODS — The Gut Microbiota DEXA scan were younger (43.0 vs. 46.2 Total cholesterol, HDL cholesterol, LDL and Obesity Study was a population- years; P  0.001) and tended to be male cholesterol, triglycerides, glucose, aspar- based case-control study among noninsti- (57.6 vs. 36.7%, P  0.001) compared tate aminotransferase (AST), alanine tutionalized residents aged 35–54 years with those who had the scan. aminotransferase (ALT), and -glutamyl- in Shanghai, China. This study investi- transferase (GGT) were measured enzy- gated the effects of gut microbiota and en- Data collection matically on an automatic analyzer vironmental factors on obesity and Home interviews were conducted by (Hitachi 7080, Tokyo, Japan) with re- related metabolic disorders. Two urban trained physicians or public health work- agents purchased from Wako Pure Chem- districts (Luwan and Zhabei) were chosen ers from the local Centers for Disease ical Industries (Osaka, Japan). Plasma to represent people with high and low so- Control and Prevention and community high-sensitivity (hs) C-reactive protein cioeconomic status in urban Shanghai. clinics. Information on demographic vari- (CRP) was measured by a particle- Participants were enrolled through their ables, health status, and behavior was enhanced immunoturbidimetric assay response to an advertisement. The field- collected using a standardized question- (Roche Diagnostics, Indianapolis, IN). work was conducted simultaneously in naire. Smoking was defined as never, cur- A1C was quantified from resolved eryth- both districts from November 2007 rent, and former. Alcohol drinking was rocytes with an automated immunoassay through January 2008. Five hundred defined as “yes” or “no.” Physical activity (Roche Diagnostics). Insulin was mea- pairs of age- and sex-matched subjects data were collected by using the Interna- sured with a completely homologous ra- (overweight/obesity) and control subjects tional Physical Activity Questionnaire dioimmunoassay (Linco Research, St. (normal-weight) were planned to be re- (short last 7-day format, http://www. Charles, MO), which had 0.2% cross- cruited. This sample size presumably ipaq.ki.se/scoring.pdf), and the level for reactivity with proinsulin. The insulin re- could provide 80% power to detect a 25% each individual was calculated as a sum of sistance index (homeostasis model difference in gut microbiota composition MET-minute/week score and then classi- assessment of insulin resistance [HOMA- between groups (14). Overweight/obesity fied as low and high by sex-specific total IR]) was calculated using updated ho- and normal weight were defined as BMI MET median. Educational attainment was meostasis model assessment methods 24.0 kg/m and 18.0  BMI 24.0 kg/ categorized into three groups (0 to 9, 10 (http://www.dtu.ox.ac.uk/). m , respectively, which was modified to 12, and 13 years of education). Fam- Plasma LBP levels were determined from the recommendation by the Work- ily history of chronic diseases was positive by a sandwich ELISA (USCN Life Science ing Group on Obesity in China (15). if a parent or sibling had coronary heart & Technology, Missouri City, TX). Eligible candidates were adult resi- disease, stroke, or type 2 diabetes. Plasma samples were diluted at least 200 dents who have lived in Shanghai for at After a home interview, all partici- times and assayed according to the man- least 10 years. Exclusion criteria included pants had a physical examination after ufacturer’s instructions. The assay has a 1) diarrhea for 3 consecutive days within overnight fasting. Body weight and height sensitivity of 0.2 ng/ml and a measurable the previous 3 months, 2) heavy alcohol were measured to the nearest 0.1 kg and concentration range of 0.78 –50 ng/ml. consumption (40 g/day ethanol for men 0.1 cm, respectively, with the participants The intra-assay and interassay coefficients and 20 g/day for women), 3) diagnosed in light indoor clothing without shoes. of variation were 5 and 10%, respec- diabetes, taking oral antidiabetic agents or BMI was calculated as weight in kilograms tively. Enzyme immunoassays were used insulin, cancer, coronary heart disease, divided by the square of height in meters. to measure plasma interleukin (IL)-6, lep- myocardial infarction, stroke, or severe Waist circumference was obtained at the tin (R&D Systems, Minneapolis, MN), kidney or liver diseases, 4) infectious dis- midpoint between the lowest rib and the and high-molecular-weight (HMW)- eases including tuberculosis, AIDS, and iliac crest to the nearest 0.1 cm, after in- adiponectin (Millipore, St. Charles, MO). hepatitis, 5) severe psychological disor- halation and exhalation. Blood pressure The interassay coefficients of variation ders or physical disabilities, 6) antibiotics was measured by using an electronic were 9.6 and 6.5% for IL-6 at 0.49 and used for 3 consecutive days within the blood pressure monitor (Omron HEM- 5.65 pg/ml, 5.4 and 3.5% for leptin at previous 3 months, 7) gastrointestinal 705CP; OMRON Healthcare, Vernon 65.7 and 581 pg/ml, and 8.1 and 3.8% for surgery within 1 year, or 8) pregnancy or Hills, IL) on the right arm of the partici- HMW-adiponectin at 21.23 and 61.50 lactation in women. The protocol was ap- pant in a comfortable sitting position after ng/ml, respectively. Hepatitis B surface proved by the Institutional Review Board at least a 5-min rest. Three measurements antigen (HBsAg) was detected by a Murex of the Institute for Nutritional Sciences were taken and the mean of the last two HBsAg ELISA kit (Murex Biotech, Dart- and written informed consent was ob- measurements was used for the analyses. ford, Kent, U.K.). 1926 DIABETES CARE, VOLUME 33, NUMBER 9, SEPTEMBER 2010 care.diabetesjournals.org Sun and Associates Definition of diseases Stata 9.2 (StataCorp, College Station, TX). Plasma LBP was also highly correlated Metabolic syndrome was defined accord- Statistical tests were two-sided and P  with inflammatory markers (hs-CRP r ing to the updated National Cholesterol 0.05 was considered statistically 0.93 and IL-6 r  0.50, both P  0.001) Education Program Adult Treatment significant. after adjustment for age and sex. These Panel III criteria for Asian-Americans correlation coefficients were attenuated (16), including at least three of the follow- RESULTS but remained statistically significant after ing components: 1) waist circumferences further adjustment for BMI (supplemen- 90 cm in men or 80 cm in women, 2) General characteristics tary Table A1). When the analyses were triglycerides 1.7 mmol/l, 3) HDL cho- The prevalence of metabolic syndrome stratified by obesity status, significant cor- lesterol 1.03 mmol/l in men or 1.30 and newly defined type 2 diabetes was relations with the metabolic traits were mmol/l in women, 4) blood pressure 42.0% (n  445) and 13.8% (n  146), more pronounced among overweight/ 130/85 mmHg or current use of antihy- respectively. The mean BMI values were obese subjects than among normal- pertensive medications, and 5) fasting 28.0  2.7 kg/m in overweight/obese weight individuals. In addition, LBP was plasma glucose 5.6 mmol/l. Type 2 di- subjects and 21.0  1.4 kg/m in normal- strongly correlated with total body fat abetes was defined as fasting plasma glu- weight subjects (P  0.001) (Table 1). mass/percentage and trunk fat mass in cose7.0 mmol/l or 2-h postload plasma Compared with normal-weight sub- sex-stratified correlation models (all P glucose11.1 mmol/l during an oral glu- jects, overweight/obese individuals were 0.001) (data not show). cose tolerance test. The selection proce- more likely to have lower educational at- dure for the oral glucose tolerance test is tainment and higher prevalence of meta- Associations of LBP concentrations described in supplementary Fig. A1 bolic syndrome and type 2 diabetes (all with metabolic syndrome, type 2 (available in an online appendix at http:// P  0.01) (Table 1). They also had higher diabetes, and insulin resistance care.diabetesjournals.org/cgi/content/ values for waist circumference, blood The risk for metabolic syndrome in the full/dc10-0340/DC1). Elevated hepatic pressure, glucose, A1C, insulin, HOMA- whole study sample increased progres- enzymes were defined as the highest IR, total cholesterol, LDL cholesterol, and sively across the LBP quartiles (P trend quartiles of one or more hepatic enzymes: triglycerides and lower HDL cholesterol 0.001) (Table 2) and those in the highest AST 30 IU/l, ALT 33 IU/l, or GGT concentration (all P  0.05). Meanwhile, LBP quartile had an OR of 3.54 (95% CI 42 IU/l. they exhibited higher levels of total body 2.05– 6.09) compared with those in the fat mass/percentage and trunk fat mass lowest quartile (model 2), after adjust- Statistical analyses (all P  0.001), hepatic enzymes (AST, ment for age, sex, lifestyle factors, family Log transformations were performed for ALT, and GGT), inflammatory markers history of chronic diseases, and BMI. Sim- LBP, triglycerides, insulin, HOMA-IR, in- (hs-CRP and IL-6) and leptin, accompa- ilar trends were also observed for the met- flammatory makers, adipokines, and he- nied by lower concentrations of HMW- abolic syndrome components. Further patic enzymes to approximate normality. adiponectin (all P  0.001). Importantly, adjustment for inflammatory markers ANCOVA for continuous variables and plasma LBP levels were significantly (model 3) abolished the significant asso- logistic regression models for categorical higher in overweight/obese participants ciations for metabolic syndrome and most variables were applied for the comparison than in normal-weight participants (geo- of its components except for hypertriglyc- across obesity status. Multiple compari- metric mean 27.6 [95% CI 25.2–30.3] vs. eridemia and low HDL cholesterol. son corrections were performed using the 10.0 [9.1–11.1] g/ml; P  0.001). In For those with newly defined type 2 Benjamini and Hochberg procedure (17). addition, LBP levels were higher in sub- diabetes, the OR in the highest quartile Partial Spearman correlation coefficients jects with impaired fasting glucose than in was 5.53 (95% CI 2.64 –11.59) compared between LBP and various parameters those with normal fasting glucose (16.7 with that in the lowest LBP quartile (P trend were calculated by adjustment for age, [15.0 –18.6] vs. 13.8 [12.2–15.6] g/ml;  0.001) in the multivariable model (Ta- sex, and BMI. Individuals with presum- P  0.041) and higher in subjects with ble 2, model 2). A positive association be- ably acute inflammation (hs-CRP 10 impaired glucose tolerance than in those tween LBP and insulin resistance, mg/l) were excluded (n 18, 1.7%), leav- with normal glucose tolerance (31.2 represented by the highest quartile of ing 1,041 subjects in the analysis. Of [24.3– 40.1] vs. 14.4 [12.8 –16.3] g/ml; HOMA-IR, was also observed in nondia- these, 942 subjects completed a DEXA P  0.001), after adjustment for age and betic participants (OR 1.90 [95% CI scan. sex. 1.10 –3.28]). The ORs for diabetes and Participants were classified into four insulin resistance were slightly attenuated groups according to their LBP quartiles in Correlation between LBP but remained statistically significant after the whole sample. Multivariate logistic re- concentrations and metabolic further adjustment for hs-CRP and IL-6 gression models were used to estimate the parameters (model 3). odds ratios (ORs) for metabolic syn- LBP was positively correlated with BMI, The associations between LBP and drome, type 2 diabetes, and insulin resis- waist circumference (all P  0.001) (sup- metabolic disorders were not attenuated tance. Potential confounding variables plementary Table A1, available in an on- by additionally controlling for HMW- include age, sex, lifestyle factors, educa- line appendix), blood pressure, total adiponectin, leptin, and elevated hepatic tion level, family history of chronic dis- cholesterol, LDL cholesterol, triglycer- enzymes (as categorical or continuous eases, and BMI (or total body fat). In ides, glucose, insulin, HOMA-IR, hepatic variables) in model 2 (data not shown). addition, we adjusted for inflammatory enzymes, and leptin and negatively corre- Replacing BMI with total body fat mass in markers, adipokines, and elevated he- lated with HDL cholesterol and HMW- the multiple regression models also did patic enzymes. Data management and sta- adiponectin (all P  0.05, data not not materially change the magnitude of tistical analyses were performed using shown), after adjustment for age and sex. the associations. The results remained care.diabetesjournals.org DIABETES CARE, VOLUME 33, NUMBER 9, SEPTEMBER 2010 1927 Marker of endotoxemia and metabolic disorders Table 1—Characteristics of participants according to obese status significant interactions were observed be- tween LBP and these factors (P  0.05 for all interaction tests). Normal weight Overweight/obesity 2 2 (18.0  BMI 24.0 kg/m ) (BMI 24.0 kg/m ) P value* CONCLUSIONS — Our data showed n 500 559 significant associations between elevated Age (years)† 45.8  5.5 46.0  5.4 0.604 LBP concentrations and the risk for met- BMI (kg/m ) 21.0  1.4 28.0  2.7 0.001‡ abolic syndrome, insulin resistance, and Men† 175 (35.0) 234 (41.9) 0.022‡ type 2 diabetes independent of conven- Physical inactivity 249 (49.8) 281 (50.3) 0.869 tional cardiovascular risk factors in an ap- Education levels 0.002‡ parently healthy Chinese population. 0–9 years 114 (22.8) 175 (31.3) Further adjustment for inflammatory fac- 10–12 years 273 (54.6) 281 (50.3) tors, but not for adipokines and elevated 12 years 113 (22.6) 103 (18.4) hepatic enzymes, substantially attenuated Current smoker (yes) 117 (23.4) 158 (28.3) 0.767 the associations for metabolic syndrome Alcohol drinker (yes) 183 (36.6) 204 (36.5) 0.245 and most of its components, suggesting Family history of chronic that chronic inflammation may mediate diseases 200 (40.0) 222 (39.7) 0.939 the effects of innate immune response in- Newly defined type 2 diabetes 36 (7.2) 110 (19.7) 0.001‡ duced by LPS-LBP. Metabolic syndrome 51 (10.2) 394 (70.5) 0.001‡ In a previous human study, Ghanim Waist circumference (cm) 75.9  6.1 93.2  8.2 0.001‡ Systolic blood pressure et al. (18) reported that increased plasma (mmHg) 118.4  15.3 130.9  17.6 0.001‡ LPS and LBP and also expression of Toll- Diastolic blood pressure like receptors and suppressor of cytokine (mmHg) 74.7  9.8 84.0  11.6 0.001‡ signaling-3 in mononuclear cells (MNCs) Glucose (mmol/l) 5.81  1.13 6.30  1.54 0.001‡ could be induced by consuming a meal A1C (%) 5.58  0.62 5.79  0.78 0.001‡ with a high-fat, high-carbohydrate con- Insulin (U/ml) 7.3 (7.07.7) 11.3 (10.911.8) 0.001‡ tent, but not with isocaloric fruit and HOMA-IR 0.85 (0.820.89) 1.33 (1.271.38) 0.001‡ fiber, implying a potential role of the Total cholesterol (mmol/l) 5.16  1.14 5.33  1.18 0.014‡ LPS-LBP pathway in postprandial inflam- LDL cholesterol (mmol/l) 3.14  0.93 3.41  0.99 0.001‡ mation and related metabolic disorders. HDL cholesterol (mmol/l) 1.53  0.44 1.23  0.34 0.001‡ In addition, positive correlations between Triglycerides (mmol/l) 1.01 (0.961.06) 1.61 (1.531.69) 0.001‡ LBP and metabolic traits such as BMI, di- DEXA scan 456 (91.2) 504 (90.2) 0.697 astolic blood pressure, fasting glucose, in- Total body fat mass (kg) sulin, and triglycerides were observed in (n  960) 60 men with glucose intolerance (13). Men 11.8  3.3 20.7  4.3 0.001‡ Moreover, a higher LBP level was associ- Women 15.7  2.8 25.7  5.1 0.001‡ ated with increased prevalence of coro- Total body fat (%) (n  960) nary artery disease independent of Men 19.0  4.3 25.6  3.4 0.001‡ established cardiovascular risk factors in Women 29.4  3.8 36.1  3.6 0.001‡ 247 male patients (12). With a relatively Trunk fat mass (kg) (n  960) large sample size of apparently healthy Men 6.3  2.1 12.1  2.7 0.001‡ men and women, our study provides Women 7.7  1.8 13.7  3.0 0.001‡ more convincing evidence about the rela- HBsAg carriage 44 (8.8) 50 (8.9) 0.916 tionship between LBP and metabolic Elevated hepatic enzymes 141 (28.2) 294 (52.6) 0.001‡ abnormalities. AST (IU/l) 23.6 (22.9–24.3) 26.3 (25.4–27.2) 0.001‡ In recent years, the effects of micro- ALT (IU/l) 19.1 (18.2–20.0) 27.8 (26.5–29.2) 0.001‡ biota on health have attracted increasing GGT (IU/l) 22.5 (21.3–23.8) 33.1 (31.2–35.0) 0.001‡ attention, and low-grade endotoxemia or hs-CRP (mg/l) 0.61 (0.57–0.66) 1.38 (1.28–1.49) 0.001‡ LPS was found to link to various meta- IL-6 (pg/ml) 1.19 (1.13–1.26) 1.67 (1.59–1.76) 0.001‡ bolic consequences. However, most stud- HMW-adiponectin (g/ml) 3.16 (2.93–3.41) 1.88 (1.73–2.04) 0.001‡ ies have been performed in mice and few Leptin (ng/ml) 3.99 (3.70–4.31) 8.94 (8.40–9.51) 0.001‡ in human populations. Studies in mice LBP (g/ml) 10.0 (9.1–11.1) 27.6 (25.2–30.3) 0.001‡ demonstrated that two- to threefold in- Data are arithmetic mean  SD, n (%), or geometric mean (95% CI). n  1,059. Percentages may not sum to 100 because of rounding. *P value was calculated after adjustment for age and sex; P value for body fat creased circulating LPS induced by a comparison was not adjusted for sex. †Data not adjusted for itself. ‡Benjamini and Hochberg– corrected high-fat diet or LPS infusion led to in- statistical significance (17). creased levels of fasting glucose and insu- lin and body weight gain (4,5). The similar after exclusion of HBsAg-positive tion analyses to examine whether obesity, occurrences of a metabolic response subjects (n  91, 8.7%) or subjects with trunk fat mass, HMW-adiponectin, and could be counteracted by a CD14 mutant 18.0 BMI 18.5 kg/m (n  14, 1.3%) elevated hepatic enzymes modified the as- (5) or improved by changing gut micro- (data not shown). sociations of LBP with metabolic syn- biota (4). In humans, a high LPS concen- We further conducted joint classifica- drome and type 2 diabetes (Fig. 1). No tration was found in individuals with type 1928 DIABETES CARE, VOLUME 33, NUMBER 9, SEPTEMBER 2010 care.diabetesjournals.org Sun and Associates Table 2—ORs (95% CI) for metabolic syndrome, type 2 diabetes, and insulin resistance according to quartiles of LBP Quartile of LBP Q1 (LBP 6.5 Q2 (6.5  LBP Q3 (15.8  LBP Q4 (LBP 42.0 g/ml) 15.8 g/ml) 42.0 g/ml) g/ml) P trend Metabolic syndrome (n  1,041) 35/260 91/260 132/261 176/260 Model 1 1 3.45 (2.22–5.37) 6.25 (4.05–9.65) 12.90 (8.28–20.10) 0.001 Model 2 1 2.78 (1.64–4.72) 3.04 (1.81–5.12) 3.54 (2.05–6.09) 0.001 Model 3 1 2.56 (1.50–4.38) 2.38 (1.36–4.15) 1.80 (0.84–3.88) 0.053 Central obesity 59/260 110/260 164/261 204/260 Model 1 1 2.50 (1.71–3.66) 5.92 (4.01–8.72) 12.80 (8.42–19.47) 0.001 Model 2 1 1.64 (0.85–3.16) 2.54 (1.30–4.97) 2.53 (1.18–5.43) 0.005 Model 3 1 1.49 (0.77–2.90) 1.88 (0.90–3.92) 1.09 (0.35–3.35) 0.304 Elevated blood pressure 66/260 89/260 119/261 157/260 Model 1 1 1.51 (1.02–2.23) 2.24 (1.53–3.28) 4.19 (2.85–6.15) 0.001 Model 2 1 1.16 (0.76–1.76) 1.32 (0.87–2.00) 1.73 (1.11–2.69) 0.013 Model 3 1 1.09 (0.71–1.67) 1.08 (0.69–1.68) 0.94 (0.50–1.75) 0.982 Hypertriglyceridemia 25/260 70/260 102/261 130/260 Model 1 1 3.61 (2.18–5.99) 5.82 (3.56–9.52) 9.34 (5.72–15.25) 0.001 Model 2 1 3.11 (1.85–5.24) 4.05 (2.43–6.76) 4.98 (2.94–8.43) 0.001 Model 3 1 3.04 (1.80–5.12) 3.77 (2.23–6.38) 4.10 (2.14–7.85) 0.001 Low HDL cholesterol 65/260 82/260 116/261 118/260 Model 1 1 1.43 (0.97–2.11) 2.64 (1.81–3.86) 2.77 (1.89–4.04) 0.001 Model 2 1 1.17 (0.78–1.75) 1.72 (1.15–2.58) 1.39 (0.90–2.15) 0.056 Model 3 1 1.18 (0.78–1.76) 1.76 (1.15–2.68) 1.54 (0.87–2.72) 0.027 Hyperglycemia 140/260 157/260 174/261 188/260 Model 1 1 1.28 (0.90–1.82) 1.64 (1.14–2.35) 2.14 (1.48–3.09) 0.001 Model 2 1 1.18 (0.82–1.69) 1.34 (0.92–1.97) 1.49 (0.99–2.27) 0.047 Model 3 1 1.08 (0.75–1.56) 1.00 (0.66–1.52) 0.60 (0.32–1.13) 0.382 Type 2 diabetes (n  1,041) 10/260 22/260 43/261 67/260 Model 1 1 2.29 (1.06–4.95) 4.71 (2.30–9.61) 8.30 (4.15–16.59) 0.001 Model 2 1 1.99 (0.91–4.33) 3.53 (1.69–7.37) 5.53 (2.64–11.59) 0.001 Model 3 1 1.93 (0.88–4.21) 3.06 (1.45–6.46) 3.23 (1.38–7.58) 0.003 Q1 (LBP 5.9 Q2 (5.9  LBP Q3 (12.7  LBP Q4 (LBP 37.8 g/ml) 12.7 g/ml) 37.8 g/ml) g/ml) P trend Insulin resistance (n  899) 27/224 38/225 72/225 87/225 Model 1 1 1.49 (0.87–2.54) 3.48 (2.13–5.70) 4.64 (2.85–7.55) 0.001 Model 2 1 1.12 (0.64–1.98) 1.95 (1.14–3.32) 1.90 (1.10–3.28) 0.005 Model 3 1 1.14 (0.64–2.01) 1.97 (1.14–3.41) 1.95 (0.97–3.92) 0.014 Model 1: adjusted for age and sex. Model 2: further adjusted for smoking (current, yes or no), alcohol drinking (yes or no), physical activity (low or high by the sex-specific MET/week median), education (0 –9, 10 –12, and 13 years), family history of chronic diseases (yes or no), and BMI. Model 3: further adjusted for hsCRP and IL-6. 2 diabetes (6). The administration of ros- We observed a stronger correlation plasma HMW-adiponectin contributed iglitazone, a hypoglycemic agent with a between LBP and inflammatory markers additional risk for metabolic syndrome potential anti-inflammatory effect (19), than that between LBP and adipokines af- and type 2 diabetes under the high LBP could reduce LPS levels (6). However, the ter adjustment for BMI. Moreover, adjust- condition (Fig. 1E and F). In fact, existing short half-life of LPS (8) and the disadvan- ing for hs-CRP and IL-6 almost eliminated data showed that LPS increased plasma tages associated with its assay (9) have the associations of LBP with metabolic leptin levels and downregulated adi- limited its potential applications in rou- syndrome and most of its traits (Table 2, ponectin receptor mRNA in healthy vol- tine clinical settings and large-scale stud- model 3), whereas controlling for HMW- unteers (20). ies. Having a relatively long half-life and adiponectin and leptin had little impact We also examined whether adjust- reliable measurements, LBP might serve on the associations. A potential mecha- ment for hepatic enzymes altered the as- as a marker reflecting an “effective” LPS nism is that LPS-LBP–triggered immune sociations, because LBP is synthesized in level and innate immune response trig- response activates the nuclear factor-B liver and LPS could cause hepatocyte gered by LPS (11,12). Therefore, circulat- pathway, stimulates formation of IL-6, damage by inducing upregulation of tu- ing LBP levels can be considered a IL-1, and tumor necrosis factor- (7), and mor necrosis factor- expression in promising early biomarker and interven- upregulates CRP synthesis in hepato- Kupffer cells (7). Previous prospective tion target for inflammatory conditions. cytes. However, it is noteworthy that low studies also suggested that elevated he- care.diabetesjournals.org DIABETES CARE, VOLUME 33, NUMBER 9, SEPTEMBER 2010 1929 Marker of endotoxemia and metabolic disorders Figure 1—ORs for metabolic disorders according to joint classification of LBP and obesity status (A and B), trunk fat (sex and obesity-stratified tertile [T], C and D), HMW-adiponectin (E and F), and hepatic enzymes (G and H). A–D: Modified metabolic syndrome was defined as having two or more components of metabolic syndrome without central obesity. Adjusted for age, sex, smoking, alcohol drinking, physical activity, education, and family history of chronic diseases. E–H: Adjusted for age, sex, smoking, alcohol drinking, physical activity, education, family history of chronic diseases, and BMI. 1930 DIABETES CARE, VOLUME 33, NUMBER 9, SEPTEMBER 2010 care.diabetesjournals.org Sun and Associates who.int/mediacentre/factsheets/fs311/en/ patic enzymes predicted risk of metabolic by applying strict exclusion criteria and print.html/. Accessed 26 October 2007 syndrome and type 2 diabetes (21). How- also included a relatively large sample size 2. Dandona P, Aljada A, Chaudhuri A, Mo- ever, no material change was observed with both sexes. Certainly, these results hanty P, Garg R. Metabolic syndrome: a when we controlled for elevated hepatic should be examined prospectively and in comprehensive perspective based on in- enzymes or excluded individuals having different populations to establish the teractions between obesity, diabetes, hepatitis B infection. Collectively, our causal relationship between LBP, inflam- and inflammation. Circulation 2005;111: findings indicate that the inflammatory mation, and metabolic outcomes. 1448 –1454 cascade initiated by LPS-LBP triggered an Our study indicates that LBP is signif- 3. Wiedermann CJ, Kiechl S, Dunzendorfer innate immune response, which is a ma- icantly associated with obesity-related S, Schratzberger P, Egger G, Oberhollen- jor mechanism linking LBP to the patho- metabolic disorders in apparently healthy zer F, Willeit J. Association of endotox- emia with carotid atherosclerosis and genesis of metabolic disorders. Chinese. These findings indicate the role cardiovascular disease: prospective re- Interestingly, unlike the case for met- of LPS-initiated innate immune mecha- sults from the Bruneck Study. J Am Coll abolic syndrome risk, controlling for in- nisms in metabolic diseases. Future pro- Cardiol 1999;34:1975–1981 flammatory markers in current study did spective studies are needed to clarify 4. Cani PD, Bibiloni R, Knauf C, Waget A, not alter the significant associations of whether LBP predicts future risk of meta- Neyrinck AM, Delzenne NM, Burcelin R. LBP with dyslipidemia, insulin resistance, bolic disorders. Changes in gut microbiota control meta- and type 2 diabetes (Table 2). The minor bolic endotoxemia-induced inflammation effect of inflammation on the association in high-fat diet-induced obesity and dia- Acknowledgments — This study was sup- between LBP and dyslipidemia might be betes in mice. Diabetes 2008;57:1470 – ported by research grants from the Chinese explained by a sequence identity shared Academy of Sciences (KSCX2-YW-R-116), between LBP and lipid transfer proteins 5. Cani PD, Amar J, Iglesias MA, Poggi M, the Ministry of Science and Technology of Knauf C, Bastelica D, Neyrinck AM, Fava that could modify lipid homeostasis dur- China (2008DFA31960, 2008AA02Z315, F, Tuohy KM, Chabo C, Waget A, Delme ´e ing subclinical endotoxemia (22). Previ- 2009AA022704, and 2007AA027332), the E, Cousin B, Sulpice T, Chamontin B, Fer- ous studies demonstrated that Science and Technology Commission of rie ` res J, Tanti JF, Gibson GR, Casteilla L, hyperinsulinemia might affect immune Shanghai Municipality (075407001), the Delzenne NM, Alessi MC, Burcelin R. competence, including functions of neu- Shanghai Institutes for Biological Sciences, the Metabolic endotoxemia initiates obesity trophils (13) and hepatic Kupffer cells Chinese Academy of Sciences (SIBS2008006), and insulin resistance. Diabetes 2007;56: (23), which subsequently influenced LPS the National Natural Science Foundation of 1761–1772 China (30930081), and the Novo Nordisk clearance and LBP synthesis. Neverthe- 6. Creely SJ, McTernan PG, Kusminski CM, A/S. J.D. and K.C. also acknowledge the less, endotoxemia might induce insulin Fisher M, Da Silva NF, Khanolkar M, French National Agency for Research (ANR resistance, whereas CD14 mutant mice Evans M, Harte AL, Kumar S. Lipopoly- MicroObes program). saccharide activates an innate immune showed hypersensitivity to insulin (5). The sponsors were not involved in the system response in human adipose tissue MNCs were also suggested to be a target study design, data collection, analysis, or in obesity and type 2 diabetes. Am J of insulin action and involved in the in- interpretation. Physiol Endocrinol Metab 2007;292: teraction between inflammation and in- No other potential conflicts of interest rele- E740 –E747 sulin resistance (24). The increased vant to this article were reported. 7. Siebler J, Galle PR, Weber MM. The gut- suppressor of cytokine signaling-3 ex- L.S. contributed to the study design, re- liver-axis: endotoxemia, inflammation, pression in MNCs, accompanied by in- searched data, contributed to discussion, and insulin resistance and NASH. J Hepatol wrote the manuscript. Z.Y. contributed to the creased LPS and LBP concentrations after 2008;48:1032–1034 study design, researched data, contributed to a high-fat, high-carbohydrate meal, might 8. Munford RS. Detoxifying endotoxin: discussion, and reviewed/edited the manu- also interfere with insulin signaling and time, place and person. J Endotoxin Res script. X.Y., S.Z., and H.L. contributed to the play a role in insulin resistance (18,24). 2005;11:69 – 84 study design, researched data, and reviewed/ 9. Novitsky TJ. Limitations of the Limulus Our data suggested that LBP might exert edited the manuscript. D.Y. and H.W. re- amebocyte lysate test in demonstrating its influence directly on insulin action in searched data and reviewed/edited the circulating lipopolysaccharides. Ann NY addition to stimulating a downstream manuscript. Y.C. contributed to the study de- Acad Sci 1998;851:416 – 421 pathway generating inflammatory cyto- sign and reviewed/edited the manuscript. J.D. 10. Weiss J. Bactericidal/permeability-in- kines. Moreover, the LBP gene was re- reviewed/edited the manuscript. K.C. wrote creasing protein (BPI) and lipopolysac- cently found to be genetically susceptible the manuscript. F.B.H. and X.L. contributed charide-binding protein (LBP): structure, to the conception and design of the study, to type 2 diabetes in Japanese (25), which function and regulation in host defence contributed to discussion, reviewed/edited may provide one possible reason that the against Gram-negative bacteria. Biochem the manuscript. conventionally inflammatory cytokine Soc Trans 2003;31:785–790 We are grateful to An Pan, Qibin Qi, Ling showed little impact on the association 11. Ruiz AG, Casafont F, Crespo J, Cayo ´n A, Lu, Chen Liu, Geng Zhang, Geng Zong, Sha- Mayorga M, Estebanez A, Fernadez-Es- between LBP and type 2 diabetes. ojie Ma, He Zheng, and the local Center for calante JC, Pons-Romero F. Lipopolysac- To our knowledge, this is the first Disease Control and Prevention staff of Shang- charide-binding protein plasma levels study that systematically investigated the hai for their kind help at various stages of this and liver TNF- gene expression in obese associations of LBP concentrations with study. patients: evidence for the potential role of the risk of obesity-related metabolic dis- endotoxin in the pathogenesis of non-al- orders. Admittedly, the cross-sectional coholic steatohepatitis. Obes Surg 2007; nature of this study does not allow for a References 17:1374 –1380 causal inference. Nonetheless, we tried to 1. World Health Organization. Obesity and 12. Lepper PM, Schumann C, Triantafilou K, eliminate potential effects of acute inflam- overweight: Fact Sheet No. 311 [article on- Rasche FM, Schuster T, Frank H, Schneider mation and other potential confounders line], 2006. Available from http://www. EM, Triantafilou M, von Eynatten M. Asso- care.diabetesjournals.org DIABETES CARE, VOLUME 33, NUMBER 9, SEPTEMBER 2010 1931 Marker of endotoxemia and metabolic disorders ciation of lipopolysaccharide-binding pro- Golani I. Controlling the false discovery the Mexico City Diabetes study. Diabetes tein and coronary artery disease in men. rate in behavior genetics research. Behav Care 2005;28:1757–1762 J Am Coll Cardiol 2007;50:25–31 Brain Res 2001;125:279 –284 22. Hudgins LC, Parker TS, Levine DM, Gor- 13. Gubern C, Lo ´ pez-Bermejo A, Biarne ´s J, 18. Ghanim H, Abuaysheh S, Sia CL, Korze- don BR, Saal SD, Jiang XC, Seidman CE, Vendrell J, Ricart W, Ferna ´ ndez-Real JM. niewski K, Chaudhuri A, Fernandez-Real Tremaroli JD, Lai J, Rubin AL. A single Natural antibiotics and insulin sensitiv- JM, Dandona P. Increase in plasma endo- intravenous dose of endotoxin rapidly al- ity: the role of bactericidal/permeabili- toxin concentrations and the expression ters serum lipoproteins and lipid transfer ty-increasing protein. Diabetes 2006; of Toll-like receptors and suppressor of proteins in normal volunteers. J Lipid Res 55:216 –224 cytokine signaling-3 in mononuclear cells 2003;44:1489 –1498 14. Li M, Wang B, Zhang M, Rantalainen M, after a high-fat, high-carbohydrate meal: 23. Cornell RP, McClellan CC. Modulation Wang S, Zhou H, Zhang Y, Shen J, Pang X, implications for insulin resistance. Diabe- of hepatic reticuloendothelial system Zhang M, Wei H, Chen Y, Lu H, Zuo J, Su tes Care 2009;32:2281–2287 phagocytosis by pancreatic hormones. M, Qiu Y, Jia W, Xiao C, Smith LM, Yang S, 19. Mohanty P, Aljada A, Ghanim H, J Reticuloendothel Soc 1982;32:397– Holmes E, Tang H, Zhao G, Nicholson JK, Hofmeyer D, Tripathy D, Syed T, Al- 407 Li L, Zhao L. Symbiotic gut microbes mod- Haddad W, Dhindsa S, Dandona P. Ev- 24. Ghanim H, Aljada A, Daoud N, Deo- ulate human metabolic phenotypes. Proc idence for a potent antiinflammatory purkar R, Chaudhuri A, Dandona P. Role Natl Acad Sci USA 2008;105:2117–2122 effect of rosiglitazone. J Clin Endocrinol of inflammatory mediators in the sup- 15. Wu Y. Overweight and obesity in China. Metab 2004;89:2728 –2735 pression of insulin receptor phosphoryla- BMJ 2006;333:362–363 20. Anderson PD, Mehta NN, Wolfe ML, tion in circulating mononuclear cells of 16. Grundy SM, Cleeman JI, Daniels SR, Do- Hinkle CC, Pruscino L, Comiskey LL, obese subjects. Diabetologia 2007;50: nato KA, Eckel RH, Franklin BA, Gordon Tabita-Martinez J, Sellers KF, Rickels MR, 278 –285 DJ, Krauss RM, Savage PJ, Smith SC Jr, Ahima RS, Reilly MP. Innate immunity 25. Takeuchi F, Yanai K, Inomata H, Kuzuya Spertus JA, Costa F. Diagnosis and manage- modulates adipokines in humans. J Clin N, Kajio H, Honjo S, Takeda N, Kaburagi ment of the metabolic syndrome: an Amer- Endocrinol Metab 2007;92:2272–2279 Y, Yasuda K, Shirasawa S, Sasazuki T, ican Heart Association/National Heart, 21. Nannipieri M, Gonzales C, Baldi S, Posa- Kato N. Search of type 2 diabetes suscep- Lung, and Blood Institute scientific state- das R, Williams K, Haffner SM, Stern MP, tibility gene on chromosome 20q. Bio- ment. Circulation 2005;112:e285– e290 Ferrannini E. Liver enzymes, the meta- chem Biophys Res Commun 2007;357: 17. Benjamini Y, Drai D, Elmer G, Kafkafi N, bolic syndrome, and incident diabetes: 1100 –1106 1932 DIABETES CARE, VOLUME 33, NUMBER 9, SEPTEMBER 2010 care.diabetesjournals.org

Journal

Diabetes Care – Pubmed Central

Published: Jun 8, 2010

There are no references for this article.