Fig 2: Relation of CC16 serum concentrations and selected influencing factors. We assessed data from 497 human subjects for an initial identification of non-modifiable and common modifiable metabolic risk factors on CC16 serum concentrations. Here, we were able to display significant contribution of (A) the variation at the genetic loci chr11:3741240 whereby both homozygous and heterozygous mutations were associated to decreased CC16, (B) the participants’ biological sex with decreased CC16 in women prior to menopause as well as (C) a strong positive correlation with the participants’ age (D). Lower CC16 was also seen with active smoking, (E) decreased gut microbial species richness (F). In return, the intake of uricosuric medication, which is a treatment of hyperuricemia, significantly increased serum CC16. Statistical significance was tested using univariate pairwise Mann-Whitney-U tests for categorical and Spearman’s rank correlation test for continuous variables. Based on the displayed strongly significant influences (p < 0.01), these factors were defined as confounders for the following analyses.

Fig 3: Effect of weight loss on serum CC16 concentrations. We assessed the effect of weight loss in the time course of 26 week achieved by three different intervention types: non-surgical conservative obesity complex therapy (Dietary Intervention, n = 48), surgical sleeve gastrectomy (n = 22) and surgical gastric bypass (n = 28). Data indicate that the patients’ BMI drastically decreased (A-C) and metabolic profiles improved (data not shown) due to either dietary or surgical intervention. After surgical intervention, alterations are overall accompanied by a mild decrease in serum CC16 (E + F), while there were no differences in concentrations at the end of the dietary intervention program (D). Statistical significance was tested using univariate paired Wilcoxon-tests.

Fig 4: Effect of metabolic and cardiovascular implications on CC16 serum concentrations. We assessed data from 497 human subjects on the presence and, where appropriate, treatment of metabolic and cardiovascular diseases in relation to CC16 serum levels using general linear regression models adjusted for biological age, sex, smoking habit, the presence of CC16 A38G polymorphism (rs3741240) and uricosuric intake. These analyses indicate an inverse relation of severe obesity, and especially abdominal body fat distribution reflected by higher waist-to-hip ratio, and CC16 serum concentrations. Moreover, untreated arterial hypertension is associated with decreased circulating CC16 which can be counteracted by antihypertensive medication treatment. Atherosclerotic cardiovascular diseases (myocardial infarction, coronary artery diseases and chronic heart failure) display increased CC16 serum levels that, in case of chronic heart failure, are even robust against confounding factors and remain statistically significant. Conforming with these findings, mild associations with small effect sizes were seen for cardio-metabolic measures, like blood pressure, triglyceride levels, as well as HOMA index indicative for insulin resistance. For all disease analyses, the absence was set as reference group. Results are displayed as estimate (95% confidence interval) and corresponding p-values. Abbreviations: CRP = C-reactive Protein, HOMA-IR = Homeostatic Model Assessment of Insulin Resistance, CI = confidence interval.

Fig 5: Cohort set-up and study workflow. This study uses data from two Northern German cohorts constituted for the extensive study of obesity and associated cardio-metabolic implications. From the cross-sectional FoCus cohort, n = 497 participants aged 18–75 years were included. From the longitudinal Kiel Intervention Cohort (KIK), n = 99 severely obese participants (BMI ≥ 40 kg/m2) who either participated in a conservative dietary weight loss program or underwent bariatric surgery with baseline and 6-month weight loss data available for analysis were included. The purpose of the study is a comprehensive metabolism-centered assessment of the pulmonary anti-inflammatory and immunomodulatory marker CC16. This comprises the validation of known non-modifiable CC16 determinants (e.g. biological age, gender and hormonal status), identification of further influencing factors on lifestyle level, the evaluation of cardio-metabolic disease presence, severity and intervention as well as the estimation of variable importance and intercorrelations both with CC16 and among determinants. For this, CC16 has been quantified in blood serum samples using ELISA and assessed in relation to several cardio-metabolic variables using correlation tests, multiparameter linear regression models and random forest-based machine learning as validation.