Ramon-Krauel et al. Molecular and Cellular Pediatrics (2021) 8:3 https://doi.org/10.1186/s40348-021-00113-z Molecular and Cellular Pediatrics RESEARCH Open Access Relationship between adiponectin, TNFα, and SHBG in prepubertal children with obesity Marta Ramon-Krauel1, María Jesús Leal-Witt1, Óscar Osorio-Conles1, Montse Amat-Bou1, Carles Lerin1* and David M. Selva2* Abstract Background: Sex hormone-binding globulin (SHBG) levels are low in adult subjects with obesity when compared to normal-weight individuals. Obesity is associated with higher tumor necrosis factor alpha (TNFα) plasma levels and lower adiponectin levels. Moreover, we have recently elucidated the molecular mechanisms by which TNFα and adiponectin regulate hepatic SHBG production. Aim: The main objective of this study was to assess if the adult associations between TNFα, adiponectin, and SHBG are present in prepubertal children. Methods: We determined several morphometric and biochemical parameters in normal-weight (n=15) and obese prepubertal (n=51) children, as well as quantified plasma SHBG, TNFα receptor 1 (TNFα-R1), and adiponectin levels. Results: Our results showed that prepubertal children with obesity had decreased plasma SHBG levels compared to normal-weight controls (67 nmol/L vs 172 nmol/L). Importantly, SHBG plasma levels correlated significantly (P < 0.05) with TNFα (negatively, ßstd= − 0.31) and adiponectin (positively, ßstd= 0.58) suggesting an important role of these two cytokines in determining plasma SHBG levels in prepubertal children. Conclusions: Our results suggest that plasma adiponectin levels may play a more important role than TNFα in influencing plasma SHBG levels in our prepubertal population with obesity. Keywords: Prepubertal children, Obesity, Sex hormone-binding globulin, Cytokines Background including cardiovascular disease, type 2 diabetes, meta- Childhood obesity prevalence has increased at an alarm- bolic syndrome, and several types of cancer [3–8]. ing rate globally, and the International Association for It is well-known that adult individuals with obesity the Study of Obesity estimates that up to 200 million show lower plasma sex hormone-binding globulin school-aged children are either overweight or obese [1, (SHBG) levels than normal-weight subjects [9, 10]. 2]. This is important because children and adolescents SHBG is a protein produced and secreted into the circu- with obesity are at higher risk of obesity during adult- lation by the liver that binds androgens and estrogens hood [3, 4], which is often linked to serious conditions with high affinity. In blood, SHBG acts as a carrier of these sex steroids and regulates their bioavailability and access to target tissues and cells [11]. Interestingly, we * Correspondence: clerin@fsjd.org; david.martinez.selva@vhir.org have previously described that low-grade inflammation 1 Endocrinology Department, Institut de Recerca Sant Joan de Déu, Hospital occurring in obesity is an important regulator of plasma Sant Joan de Déu, 08950 Barcelona, Spain 2 Diabetes and Metabolism Research Unit, Vall Hebron Institut de Recerca SHBG levels [12]. Obesity is associated with higher (VHIR), Universitat Autònoma de Barcelona, Pg Vall d’Hebron 119-129, 08035 tumor necrosis factor alpha (TNFα) plasma levels and Barcelona, Spain © The Author(s). 2021 Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. Ramon-Krauel et al. Molecular and Cellular Pediatrics (2021) 8:3 Page 2 of 6 lower adiponectin levels in adults [13, 14]. Moreover, we cholesterol, HDL-cholesterol, triglycerides) were mea- have recently elucidated the molecular mechanisms by sured using standard protocols at the Hospital’s clinical which TNFα and adiponectin regulate hepatic SHBG laboratory. An oral glucose tolerance test (1.75 g glu- production [15–17]. Low plasma SHBG levels have been cose/kg with a maximum of 75 g per subject) was per- also found in children with obesity; however, the reasons formed in a subset of children with obesity (n=47); for such reduction need to be yet determined. glucose and insulin levels were measured at 0, 30, 60, 90, In the present work, we aimed to study whether these and 120 min post-glucose load. The insulinogenic index regulatory mechanisms were already present in pre- (IGI) and the whole-body insulin sensitivity index pubertal children. For this purpose, we determined sev- (WBISI) were calculated from the tolerance test data as eral morphometric and biochemical parameters in previously described [20]. Plasma levels for SHBG, adi- normal-weight and prepubertal children with obesity, as ponectin, and TNFα-R1 were measured by ELISA (R&D well as quantified plasma SHBG, TNFα receptor 1 Systems, Madrid, Spain). (TNFα-R1), and adiponectin levels. Statistical analysis Methods Data are shown as mean (SD) for normally distributed Study participants data, and median [IQR] for non-normally distributed The study was approved by the Hospital’s Ethic Com- data; groups were compared using two-paired t-test or mittee (PIC-21-12). Subsets of data from this cohort Mann-Whitney U test, respectively. Normality was have been reported elsewhere [18, 19]. Children with tested with the Shapiro-Wilk W test. SHBG levels were obesity attending the Hospital Sant Joan de Déu, Barce- log-transformed to achieve normality before testing cor- lona (Spain) were recruited between the months of Janu- relations. Associations between variables were deter- ary 2013 and December 2014. Inclusion criteria were mined by standard least-square regression, with sex, age, children (a) age from 7 to 10 years old; (b) prepubertal, and BMI-SDS as covariates where indicated. Data ana- defined as Tanner stage I breast development for girls lysis and calculations were performed in JMP v14.3 (SAS and testicular volume less than 4 ml in boys; (c) with Institute Inc., Cary, NC). A p value below 0.05 was con- BMI-SDS for a given age and sex using the World sidered significant. Health Organization (WHO) standards greater than 2 standard deviations for the obesity group, and between Results −1 and 1 standard deviations for the normal-weight Circulating SHBG levels are decreased in prepubertal group. Exclusion criteria included any form of endogen children with obesity obesity, major congenital or chronic disease, drug- The main clinical and biochemical features of normal- induced obesity, use of drugs for weight loss, and in- weight (n=15) and children with obesity (n=51) are dis- volvement in another weight-loss program. All parents played in Table 1. All morphometric parameters were signed the informed consent document. Blood samples significantly elevated in subjects with obesity compared were obtained from the hospital’s biobank and were with lean subjects, including weight, BMI-SDS, and waist available from 15 normal-weight controls and from 51 circumference (Table 1). Regarding the glycemic param- of the 53 subjects with obesity initially included in the eters, no differences were found between lean and obese study. subjects in glucose and HbA1c levels; however, insulin concentration and HOMA-IR values were significantly Physiologic and biochemical measurements increased in children with obesity (Table 1). Regarding Blood samples were obtained in the morning after an the lipid profile, while total cholesterol was not altered, overnight fast (8–10 h of fasting). We measured weight we observed increased LDL-cholesterol and decreased (kg) and height (m) with light clothing in a calibrated HDL-cholesterol, as well as higher triacylglyceride scale and rigid stadiometer. Body mass index (BMI) was (TAG) levels in the group with obesity (Table 1). More- calculated, and BMI-SDS was obtained by using the over, subjects with obesity had significantly higher C- “Anthro Plus” software (WHO). Blood pressure was reactive protein (CRP) and TNFα-R1 and reduced SHBG measured in the right arm using an automated system and adiponectin plasma levels compared to the normal- with the appropriate sleeve size for the arm diameter. weight controls (Table 1). Waist circumference was determined as middle point between the last rib and iliac crest. Blood samples were SHBG correlations with anthropometric and biochemical taken in tubes containing EDTA, and plasma was imme- parameters in prepubertal children diately separated, aliquoted, and stored at −80 °C until After analyzing all the measured parameters in our study further use. Glucose, insulin, glycated hemoglobin population, our results showed that SHBG levels were (HbA1c), and the lipid profile (total cholesterol, LDL- negatively correlated to BMI-SDS, waist circumference, Ramon-Krauel et al. Molecular and Cellular Pediatrics (2021) 8:3 Page 3 of 6 Table 1 Demographic and physiologic characteristics of study subjects. Normally distributed variables are shown by mean (SD) and groups compared by two-tail t-test. aNon-normally distributed data are shown by median [IQR] and groups compared by Mann- Whitney U test. Bold font indicates p<0.0001. Min, minimum value; Max, maximum value; BMI-SDS, BMI standard deviation score; SP, systolic pressure; DP, diastolic pressure; ADIPOQ, adiponectin; TAG, triacylglycerides; LDL-Cho, low-density lipoprotein cholesterol; HDL- Cho, high-density lipoprotein cholesterol; n.a., not applicable Normal weight (n=15) Obesity (n=51) Mean ± SD Min-max Mean ± SD Min-max p value Sex (F/M) 7/8 . 28/23 n.a. Age (years) 8.4±0.8 (7.3, 10.1) 9.0±1.1 (7.0, 11.0) 0.048 Weight (kg) 26.8±2.8 (22.8, 34.8) 53.5±12.6 (32.2, 96.3) < 0.001 BMI-SDS − 0.01±0.89 (− 1.71, 1.3) 3.41±0.76 (2.04, 5.76) < 0.001 Waist Circ. (cm) 56±3 (50, 60) 83±9 (65, 104) < 0.001 SP (mmHg) 104±7 (94, 116) 113±9 (92, 143) 0.001 DP (mmHg) 64±6 (52, 73) 70±7 (54, 83) 0.003 a SHBG (nmol/L) 172±24 (115, 218) 67±68 (12, 195) < 0.001 TNFα-R1 (pg/mL) 854±230 (578, 1435) 1084±205 (638, 1573) 0.001 CRP (mg/L) a 0.6±2.1 (0.2, 27.4) 3.0±3.2 (0.5, 52.2) 0.005 ADIPOQ (pg/mL) a 12380±4739 (6431, 20138) 7057±3778 (1214, 17748) < 0.001 Glycemia (mg/dL) 86±6 (77, 94) 85±8 (70, 113) 0.765 Insulin (mU/L) 3.2±1.6 (0.5, 5.7) 13.6±6.6 (2.1, 33.3) <0.001 HbA1c (%) 5.3±0.2 (5, 5.7) 5.3±0.2 (4.9, 5.8) 0.894 Cholesterol (mg/dL) 160±21 (132, 200) 165±24 (115, 214) 0.554 a TAG (mg/dL) 42±17 (29, 70) 70±41 (39, 183) <0.001 LDL-Cho (mg/dL) 89±19 (67, 132) 104±23 (63, 150) 0.036 HDL-Cho (mg/dL) 62±11 (44, 82) 44±10 (29, 82) <0.001 IGI a 1.27±1.00 (− 0.01, 5.69) n.a. a WBISI 3.06±2.93 (1.23, 11.74) n.a. HOMA-IR 0.70±0.35 (0.10, 1.25) 2.87±1.47 (0.44, 6.66) <0.001 insulin, HOMA-IR, TAG, TNFα-R1, and CRP plasma while they correlated positively with adiponectin (Table levels adjusting for sex and age. In addition, plasma 3). No correlation was found between SHBG and BMI- SHBG levels correlated significantly and positively with SDS, % fat mass, TNFα-R1, CRP glucose, total choles- adiponectin and HDL plasma levels (Table 2). No correl- terol, LDL, or HDL plasma levels (Table 3). Further ad- ation was found between SHBG and glucose, HbA1c, justment by BMI-SDS did not substantially modify these total cholesterol, or LDL levels (Table 2). Interestingly, correlations (Table 3). while further adjustment for BMI-SDS did not substan- tially modify SHBG correlations with insulin, HOMA-IR, Discussion and adiponectin, its association with inflammation Childhood obesity has become a major health concern markers TNFα-R1 and CRP plasma levels was com- in the last decades and has increased its prevalence pletely lost (Table 2). worldwide, in both developed and developing countries [21–23]. It has been already described that childhood SHBG plasma level correlations in prepubertal children and adolescent obesity is associated with an increase in with obesity BMI and excess of adiposity [5, 24] which was also evi- Given that obesity is intimately associated with SHBG dent in our study since prepubertal children with obesity levels, insulin resistance, and inflammation markers, we showed higher BMI-SDS and waist circumference when analyzed SHBG correlations with all measured parame- compared with lean children. This adiposity increase oc- ters only in prepubertal children with obesity to avoid curring in obesity is also associated with insulin resist- this confounding factor. Our results showed that SHBG ance, hypertension, and dyslipidemia, which in turn plasma levels correlated significantly and negatively with increase risk of developing atherosclerotic vascular waist circumference, insulin, and TAG concentration, disease and type 2 diabetes [5, 24]. Our results Ramon-Krauel et al. Molecular and Cellular Pediatrics (2021) 8:3 Page 4 of 6 Table 2 Correlations between SHBG levels and anthropometric Table 3 Correlations between SHBG levels and anthropometric and physiologic parameters in all subjects. Multivariate linear and physiologic parameters in subjects with obesity. regression was applied to determine correlations between SHBG Multivariate linear regression was applied to determine levels and other variables in all subjects. Bold font indicates P < correlations between SHBG levels and other variables only in 0.05. Model 1 adjusted for sex and age. Model 2 adjusted for subjects with obesity. Bold font indicates P < 0.05. Model 1 sex, age, and BMI-SDS. ßstd, standardized beta coefficient from adjusted for sex and age. Model 2 adjusted for sex, age, and the regression model; BM-SDS, BMI standard deviation score; BMI-SDS. ßstd, standardized beta coefficient from the regression ADIPOQ, adiponectin; TAG, triacylglycerides; LDL-Cho, low-density model; BMI-SDS, BMI standard deviation score; ADIPOQ, lipoprotein cholesterol; HDL-Cho, high-density lipoprotein adiponectin; TAG, triacylglycerides; LDL-Cho, low-density cholesterol; n.a., not applicable lipoprotein cholesterol; HDL-Cho, high-density lipoprotein Model 1 Model 2 cholesterol; n.a., not applicable ßstd p value ßstd p value Model 1 Model 2 BMI-SDS − 0.65 < 0.001 - - ßstd p value ßstd p value Waist circ. − 0.61 < 0.001 − 0.14 0.013 BMI-SDS − 0.34 0.008 n.a. n.a. TNFα-R1 − 0.31 0.012 − 0.02 0.914 Waist circ − 0.36 0.001 − 0.16 0.053 log CRP − 0.31 0.022 − 0.11 0.478 TNFα-R1 − 0.14 0.313 − 0.01 0.926 log ADIPOQ 0.58 < 0.001 0.57 < 0.001 log CRP − 0.14 0.324 0.03 0.856 Glucose − 0.12 0.355 -0.23 0.173 log ADIPOQ 0.58 < 0.001 0.57 < 0.001 Insulin − 0.73 < 0.001 -0.63 < 0.001 Glucose − 0.19 0.198 − 0.27 0.081 HbA1c − 0.01 0.923 − 0.02 0.910 Insulin − 0.64 < 0.001 − 0.62 < 0.001 HOMA-IR − 0.72 < 0.001 − 0.63 < 0.001 HbA1c − 0.03 0.823 0.03 0.840 TAG − 0.48 < 0.001 − 0.37 0.018 HOMA-IR − 0.63 < 0.001 − 0.62 < 0.001 Cholesterol 0.18 0.201 0.18 0.301 log IGI − 0.58 < 0.001 − 0.53 0.001 LDL-Cho 0.09 0.516 0.21 0.225 log WBISI 0.59 < 0.001 0.57 < 0.001 HDL-Cho 0.43 0.001 0.16 0.269 TAG − 0.33 0.028 − 0.36 0.025 Cholesterol 0.26 0.090 0.12 0.427 LDL-Cho 0.27 0.076 0.14 0.363 corroborated the latter since obese prepubertal children had higher insulin plasma levels together with an in- HDL-Cho 0.24 0.114 0.22 0.177 crease in HOMA-IR and higher LDL and TAG plasma levels when compared with lean prepubertal children. insulin levels has been previously described [28, 29], An excess of adiposity may also influence various as- which we also found in our study. However, insulin as a pects of pubertal development, such as timing of puber- major factor downregulating SHBG production does not tal initiation and hormonal parameters during puberty. come along with the fact that obese subjects are charac- How obesity may perturb several hormonal aspects of terized by insulin resistance [30, 31], and, to the best of puberty is currently under debate; however, it may in- our knowledge, the molecular mechanism by which in- volve sex hormone-binding globulin (SHBG), a liver- sulin reduces SHBG production has not yet been de- produced glycoprotein that binds sex steroids with high scribed. In this regard, we have demonstrated that affinity in blood regulating their bioavailability [11]. insulin does not regulate hepatic SHBG production [15, Interestingly, SHBG plasma levels change through hu- 32]. Nevertheless, we found strong correlations between man life-span, remaining high during infancy and de- SHBG and insulin plasma levels when analyzing our creasing when puberty approaches [25]. It is important study population or children with obesity alone. Further- to mention that low plasma SHBG levels are found in more, SHBG was strongly associated with both insulin children with obesity; this is in agreement with our re- sensitivity (HOMA-IR and WBISI) and insulin secretory sults showing that prepubertal children with obesity had (IGI) indexes in children with obesity. half the plasma SHBG levels present in normal-weight Accumulated evidence over the last decade points to prepubertal children. inflammation as one of the critical processes associated Low plasma SHBG levels are also present in adult sub- with the development of obesity, insulin resistance, and jects with obesity [9, 10]. The reason why this may occur type 2 diabetes [33, 34]. In fact, obesity is considered a has been investigated during the last three decades; ini- state of chronic low-grade inflammation [35, 36]. This is tially, it was suggested that hyperinsulinemia was the of importance since we have reported that alterations in main factor reducing plasma SHBG levels [26, 27]. In proinflammatory and anti-inflammatory cytokines oc- fact, a negative correlation between plasma SHBG and curring in obesity [37, 38] play an important role in Ramon-Krauel et al. Molecular and Cellular Pediatrics (2021) 8:3 Page 5 of 6 regulating hepatic SHBG production [12, 15–17, 39]. Acknowledgements Specifically, using in vitro and in vivo models as well as We are indebted to the families and children who participated in the study. We thank the “Biobanc de l’Hospital Infantil Sant Joan de Déu per a la human samples, we have shown that hepatic SHBG pro- Investigació” integrated in the Spanish Biobank Network of Instituto de Salud duction was reduced by TNFα and increased by adipo- Carlos III for sample procurement. nectin [15–17]. Mechanistically, we have determined Authors’ contributions that TNFα decreases SHBG production by reducing the MRK: researched data, contributed to discussion, reviewed/edited protein levels of hepatocyte nuclear receptor 4 alpha manuscript. MJLW: contributed to discussion, reviewed/edited manuscript. (HNF-4α), the main activator of SHBG gene expression OOC: contributed to discussion, reviewed/edited manuscript. MAB: contributed to discussion, reviewed/edited manuscript. CL: researched data, [15, 16]. On the other hand, adiponectin increases SHBG contributed to discussion, wrote manuscript, reviewed/edited manuscript. production via HNF-4α levels by inhibiting hepatic lipo- DMS: researched data, contributed to discussion, wrote manuscript, genesis and stimulating hepatic β-oxidation [17]. Inter- reviewed/edited manuscript. The authors read and approved the final manuscript. estingly, we showed that plasma TNFα-R1 levels correlated inversely with plasma SHBG levels in adult Funding obese subjects [15]. Remarkably, prepubertal children This work was supported by grants from the Instituto de Salud Carlos III with obesity from our study had higher TNFα-R1 plasma (DMS) PI12/01357, CIBERDEM (CIBER de Diabetes y Enfermedades Metabólicas Asociadas) an initiative of Instituto de Salud Carlos III (DMS), levels than lean children, as described previously [40]. European Regional Development Funds (ERDF), and Marie Curie FP7-PEOPLE- Importantly, SHBG plasma levels correlated significantly 2011-CIG (CL). and negatively with TNFα-R1 plasma levels. However, this correlation was lost after adjusting for BMI-SDS. Availability of data and materials The datasets used and analyzed during the current study are available from This is of importance since TNFα levels could be an im- the corresponding author on reasonable request. portant factor downregulating SHBG production during early stages of obesity development, but once children Declarations reach a certain point of inflammation, TNFα plasma Ethics approval and consent to participate levels may not influence SHBG production. All parents signed the informed consent document. The study was approved Plasma adiponectin levels are lower in subjects with by the Hospital’s Ethic Committee (PIC-21-12). obesity, both adult [41] and children, compared to Consent for publication normal-weight controls [42]. These findings were cor- Not applicable. roborated in our study, since prepubertal children with obesity had lower levels of adiponectin when compared Competing interests with lean subjects. In addition, it has been previously re- The authors declare that they have no competing interests. ported that adiponectin levels were inversely associated Received: 25 September 2020 Accepted: 1 March 2021 with SHBG [43, 44], and this correlation was also found in our prepubertal population. Moreover, this correl- ation between SHBG and adiponectin was maintained References 1. Wang Y, Lim H (2012) The global childhood obesity epidemic and the after adjusting for BMI-SDS. More importantly, this association between socio-economic status and childhood obesity. Int Rev SHBG correlation with adiponectin was also found when Psychiatry 24(3):176–188 analyzing only subjects with obesity. 2. Fruhstorfer BH, Mousoulis C, Uthman OA et al (2016) Socio-economic status and overweight or obesity among school-age children in sub-Saharan In conclusion, the results obtained in this study show Africa - a systematic review. Clin Obes 6(1):19–32 that prepubertal children with obesity had decreased 3. The NS, Suchindran C, North KE et al (2010) Association of adolescent plasma SHBG levels compared to normal-weight con- obesity with risk of severe obesity in adulthood. JAMA 304(18):2042–2047 4. Pi-Sunyer X (2009) The medical risks of obesity. Postgrad Med 121(6):21–33 trols. Importantly, SHBG plasma levels correlated signifi- 5. Biro FM, Wien M (2010) Childhood obesity and adult morbidities. Am J Clin cantly with TNFα (negatively) and adiponectin Nutr 91(5):1499S–1505S (positively), suggesting that adiponectin and TNFα could 6. Harrell JS, Jessup A, Greene N (2006) Changing our future: obesity and the metabolic syndrome in children and adolescents. J Cardiovasc Nurs 21(4): play a role in determining plasma SHBG levels in pre- 322–330 pubertal children. In addition, our results suggest that 7. Bjerregaard LG, Jensen BW, Ängquist L et al (2018) Change in overweight plasma adiponectin levels may play a more important from childhood to early adulthood and risk of type 2 diabetes. New England Journal of Medicine 378:1302–1312 role than TNFα in influencing plasma SHBG levels in 8. Juonala M, Magnussen CG, Berenson GS et al (2011) Childhood adiposity, our prepubertal population with obesity. adult adiposity, and cardiovascular risk factors. N Engl J Med 365:1876–1885 9. De Moor P, Joossens JV (1970) An inverse relation between body weight and the activity of the steroid binding-globulin in human plasma. Abbreviations Steroidologia 1(3):129–136 BMI-SDS: Body mass index standard deviation score; 10. Kopelman PG, Pilkington TR, White N et al (1980) Abnormal sex steroid EDTA: Ethylenediaminetetraacetic acid; ELISA: Enzyme-linked immunosorbent secretion and binding in massively obese women. Clin Endocrinol (Oxf) assay; HDL: High-density lipoprotein; HOMA-IR: Homeostasis model 12(4):363–369 assessment of insulin resistance; LDL: Low-density lipoprotein 11. Siiteri PK, Murai JT, Hammond GL et al (1982) The serum transport of steroid hormones. Recent Prog Horm Res 38:457–510 Ramon-Krauel et al. Molecular and Cellular Pediatrics (2021) 8:3 Page 6 of 6 12. Simó R, Sáez-López C, Barbosa-Desongles A et al (2015) Novel insights in 38. Pirola L, Ferraz JC (2017) Role of pro- and anti-inflammatory phenomena in SHBG regulation and clinical implications. Trends Endocrinol Metab 26(7): the physiopathology of type 2 diabetes and obesity. World J Biol Chem 26; 376–383 8(2):120-128. 13. Katsuki A, Sumida Y, Murashima S et al (1998) Serum levels of tumor 39. Simó R, Barbosa-Desongles A, Hernandez C et al (2012) IL1β down- necrosis factor-alpha are increased in obese patients with noninsulin- regulation of sex hormone-binding globulin production by decreasing HNF- dependent diabetes mellitus. J Clin Endocrinol Metab 83(3):859–862 4α via MEK-1/2 and JNK MAPK pathways. Mol Endocrinol 26(11):1917–1927 14. Weyer C, Funahashi T, Tanaka S et al (2001) Hypoadiponectinemia in obesity 40. Dzienis-Straczkowska S, Straczkowski M, Szelachowska M et al (2003) Soluble and type 2 diabetes: close association with insulin resistance and tumor necrosis factor-alpha receptors in young obese subjects with normal hyperinsulinemia. J Clin Endocrinol Metab 86(5):1930–1935 and impaired glucose tolerance. Diabetes Care 26(3):875–880 15. Simo R, Barbosa-Desongles A, Lecube A et al (2012) Potential role of tumor 41. Oh KW, Lee WY, Rhee EJ et al (2005) The relationship between serum necrosis factor-alpha in downregulating sex hormone-binding globulin. resistin, leptin, adiponectin, ghrelin levels and bone mineral density in Diabetes. 61:372–382 middle-aged men. Clin Endocrinol (Oxf) 63:131–138 16. Simo R, Barbosa-Desongles A, Sáez-Lopez C et al (2012) Molecular 42. Weiss R, Dufour S, Groszmann A et al (2003) Low adiponectin levels in mechanism of TNFalpha-induced down-regulation of SHBG expression. Mol adolescent obesity: a marker of increased intramyocellular lipid Endocrinol 26:438–446 accumulation. J Clin Endocrinol Metab 88(5):2014–2018 17. Simó R, Saez-Lopez C, Lecube A et al (2014) Adiponectin upregulates SHBG 43. Vanbillemont G, Lapauw B, De Naeyer H et al (2012) Sex hormone-binding production: molecular mechanisms and potential implications. globulin at the crossroad of body composition, somatotropic axis and Endocrinology 155(8):2820–2830 insulin/glucose homeostasis in young healthy men. Clin Endocrinol (Oxf) 76: 18. Leal-Witt MJ, Ramon-Krauel M, Samino S et al (2018) Untargeted 111–118 metabolomics identifies a plasma sphingolipid-related signature associated 44. Yasui T, Tomita J, Miyatani Y, et al (2007) Associations of adiponectin with with lifestyle intervention in prepubertal children with obesity. Int J Obes sex hormone-binding globulin levels in aging male and female populations. 42:72–78 Clin Chim Acta 386:69–75. 19. Leal-Witt MJ, Llobet M, Samino S et al (2018) Lifestyle intervention decreases urine trimethylamine N-oxide levels in prepubertal children with obesity. Obesity (Silver Spring) 26:1603–1610 Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in 20. Matsuda M, DeFronzo RA (1999) Insulin sensitivity indices obtained from published maps and institutional affiliations. oral glucose tolerance testing: comparison with the euglycemic insulin clamp. Diabetes Care 22:1462–1470 21. Institute of Medicine (2012) Accelerating progress in obesity prevention: solving the weight of the nation. National Academies Press, Washington, DC 22. US Department of Health and Human Services (2010) The surgeon general’s vision for a healthy and fit nation. US Dept., Health and Human Services, Rockville, MD 23. Must A, Hollander SA, Economos CD (2006) Childhood obesity: a growing public health concern. Expert Rev Endocrinol Metab 1(2):233–254 24. Adair LS (2008) Child and adolescent obesity: epidemiology and developmental perspectives. Physiol Behav. 22;94(1):8-16. 25. Hammond GL (2011) Diverse roles for sex hormone-binding globulin in reproduction. Biol Reprod 85:431–441 26. Plymate SR, Matej LA, Jones RE et al (1988) Inhibition of sex hormone- binding globulin production in the human hepatoma (Hep G2) cell line by insulin and prolactin. J. Clin. Endocrinol. Metab 67:460–464 27. Loukovaara M, Carson M, Adlercreutz H (1995) Regulation of production and secretion of sex hormone binding globulin in HepG2 cell cultures by hormones and growth factors. J. Clin. Endocrinol. Metab 80:160–164 28. Haffner SM, Katz MS, Stern MP et al (1988) The relationship of sex hormones to hyperinsulinemia and hyperglycemia. Metabolism 37:683–688 29. Pugeat M, Cousin P, Baret C et al (2000) Sex hormone-binding globulin during puberty in normal and hyperandrogenic girls. J. Pediatr. Endocrinol. Metab 13:1277–1279 30. Bonadonna RC, Groop L, Kraemer N et al (1990) Obesity and insulin resistance in humans: a dose-response study. Metabolism 39(5):452–459 31. Dunaif A, Segal KR, Futterweit W et al (1989) Profound peripheral insulin resistance, independent of obesity, in polycystic ovary syndrome. Diabetes 38(9):1165–1174 32. Selva DM, Hogeveen KN, Innis SM et al (2007) Monosaccharide-induced lipogenesis regulates the human hepatic sex hormone-binding globulin gene. J Clin Invest 117:3979–3987 33. Wellen KE, Hotamisligil GS (2005) Inflammation, stress, and diabetes. J Clin Invest 115(5):1111–1119 34. Olefsky JM, Glass CK (2010) Macrophages, inflammation, and insulin resistance. Annu Rev Physiol 72:219–246 35. Mathieu P, Lemieux I, Després JP (2010) Obesity, inflammation, and cardiovascular risk. Clin Pharmacol Ther 87(4):407–416 36. Hotamisligil GS (2003) Inflammatory pathways and insulin action. Int J Obes 27:S53–S55 37. Yudkin JS (2003) Adipose tissue, insulin action and vascular disease: inflammatory signals. Int J Obes 27(3):S25–S28