Distinct whole-blood transcriptome profile of children with metabolic healthy overweight/obesity compared to metabolic unhealthy overweight/obesity

Integrated analysis of methylome and transcriptome responses to exercise training in children with overweight/obesity

We examined the effects of a 20-wk exercise intervention on whole blood genome-wide DNA methylation signature and its association with the exercise-induced changes in gene expression profiles in boys and girls with overweight/obesity (OW/OB). Twenty-three children (10.05 ± 1.39 yr, 56% girls) with OW/OB were randomized to either a 20-wk exercise intervention [exercise group (EG); n = 10; 4 boys/6 girls] or to usual lifestyle [control group (CG); n = 13; 6 boys/7 girls]. Whole blood genome-wide methylome (CpG sites) analysis using Infinium Methylation EPIC array and transcriptome analysis using RNA-seq (STRT2 protocol) were performed. Exercise-induced modifications in DNA methylation at 485 and 386 CpGs sites in boys and girls, respectively. These CpG sites are mapped to loci enriched in distinct gene pathways related to metabolic diseases, fatty acid metabolism, and immune function. In boys, changes in the DNA methylation of 87 CpG sites (18% of the 485 CpGs sites altered by exercise) were associated with changes in the gene expression levels of 51 genes also regulated by exercise. Among girls, changes in DNA methylation at 46 CpG sites (12% of the initial 386 significant CpGs) were associated with changes in the expression levels of 30 exercise-affected genes. Genes affected by exercise that were associated with DNA methylation are related to obesity, metabolic syndrome, and inflammation. Multiomics analysis of whole blood samples from children with OW/OB suggests that gene expression response to exercise may be modulated by DNA methylation and involve gene pathways related to metabolism and immune functions.NEW & NOTEWORTHY This study pioneers the exploration into the effects of exercise on whole blood genome-wide DNA methylation patterns and its association with changes in transcriptome profiles in children with overweight/obesity. Exercise potentially impacts molecular pathways involved in metabolism and immune functions in children with overweight/obesity (sex-specific responses) through the modification of epigenetic and transcriptomic profiles. Our preliminary results provide initial steps to understand better the molecular mechanisms underlying cardiometabolic benefits of exercise in children with overweight/obesity.

The declining insulinogenic index correlates with inflammation and metabolic dysregulation in non-obese individuals assessed by blood gene expression

AIMS

Diabetes onset is difficult to predict. Since decreased insulinogenic index (IGI) is observed in prediabetes, and blood gene expression correlates with insulin secretion, candidate biomarkers can be identified.

METHODS

We collected blood from 96 participants (54 males, 42 females) in 2008 (age: 52.5 years) and 2016 for clinical and gene expression analyses. IGI was derived from values of insulin and glucose at fasting and at 30 min post-OGTT. Two subgroups were identified based on IGI variation: "Minor change in IGI" group with absolute value variation between -0.05 and +0.05, and "Decrease in IGI" group with a variation between -20 and -0.05.

RESULTS

Following the comparison of "Minor change in IGI" and "Decrease in IGI" groups at time 0 (2008), we identified 77 genes correlating with declining IGI, related to response to lipid, carbohydrate, and hormone metabolism, response to stress and DNA metabolic processes. Over the eight years, genes correlating to declining IGI were related to inflammation, metabolic and hormonal dysregulation. Individuals with minor change in IGI, instead, featured homeostatic and regenerative responses.

CONCLUSIONS

By blood gene expression analysis of non-obese individuals, we identified potential gene biomarkers correlating to declining IGI, associated to a pathophysiology of inflammation and metabolic dysregulation.

Donor white blood cell differential is the single largest determinant of whole blood gene expression patterns

It has become widely accepted that sample cellular composition is a significant determinant of the gene expression patterns observed in any transcriptomic experiment performed with bulk tissue. Despite this, many investigations currently performed with whole blood do not experimentally account for possible inter-specimen differences in cellularity, and often assume that any observed gene expression differences are a result of true differences in nuclear transcription. In order to determine how confounding of an assumption this may be, in this study, we recruited a large cohort of human donors (n = 138) and used a combination of next generation sequencing and flow cytometry to quantify and compare the underlying contributions of variance in leukocyte counts versus variance in other biological factors to overall variance in whole blood transcript levels. Our results suggest that the combination of donor neutrophil and lymphocyte counts alone are the primary determinants of whole blood transcript levels for up to 75% of the protein-coding genes expressed in peripheral circulation, whereas the other factors such as age, sex, race, ethnicity, and common disease states have comparatively minimal influence. Broadly, this infers that a majority of gene expression differences observed in experiments performed with whole blood are driven by latent differences in leukocyte counts, and that cell count heterogeneity must be accounted for to meaningfully biologically interpret the results.

Comparative analyses of surrogates of metabolic syndrome in children and adolescents with metabolically healthy obesity vs. metabolically unhealthy obesity according to Damanhoury's criteria

OBJECTIVES

Metabolically healthy obesity (MHO) has been reported with varying frequencies in children. The reasons of metabolically healthy phenotype in some obese subjects are unclear. Our aim was to identify the frequency of MHO in obese subjects, to assess the potential associations of demographic characteristics, serum uric acid, alanine transaminase (ALT), pediatric nonalcoholic fatty liver disease fibsosis score probability (PNFS p) with MHO status and to evaluate the differences between MHO and metabolically unhealthy obesity (MUO) with regard to metabolic syndrome surrogates.

METHODS

251 consecutive obese subjects (125 females) aged 7-18 years were included. Subjects were classified as having MHO according to Damanhoury's criteria. Several metabolic variables were measured, PNFS p was calculated by using the formula: z=1.1+(0.34*sqrt(ALT))+ (0.002*ALP)-(1.1*log(platelets)-(0.02*GGT).

RESULTS

Median age of the subjects was 12.5 yr (range: 7.0-17.0 yr). The frequency of MHO was 41 %. Subjects with MHO were significantly younger, had lower waist circumference (WC) and waist height ratio (WHtR) and lower HOMA-IR than those without MHO(p<0.05 for all). Frequencies of hyperuricemia, hypertransaminasemia, hepatosteatosis and PNFS p values≥8 were similar betwen the groups. When putatively influential factors associated with MHO status were assessed with logistic regression analysis, only WC(β=1.03) and HOMA-IR(β=1.166) emerged as significant factors(Nagelkerke R2=0.142). None of the investigated demographic factors were associated with MHO status.

CONCLUSIONS

We found a remarkably high frequency of MHO status. Nevertheless, the absence of decreased frequencies of hyperuricemia, hypertransaminasemia and PNFS in subjects with MHO may suggest the need to reconsider the validity of the criteria defining MHO.

Transcriptomic Differences Between Monozygotic Adolescent Twins Discordant For Metabolic Syndrome Following Weight Loss: A Case Study

This case reports peripheral blood mononuclear cell (PBMC) transcriptomic changes in a pair of male monozygotic pediatric twins with metabolic syndrome (MetS) undertaking assisted weight loss. These 14-year-old boys presented with similar baseline biochemistry and body composition. After a 16-week weight-loss intervention, percent body weight loss was similar (Twin A 12%, and Twin B 13%). MetS resolved in Twin A but Twin B maintained elevated triglycerides after weight loss. Analysis of the PBMC transcriptome before and after weight loss revealed very different changes in gene expression including differences in the direction of expression of genes related to immune cell activation. 48.7% of genes that were downregulated in Twin A were upregulated in Twin B. This case highlights a novel approach to report the influence of chronic low-grade inflammation and metabolic dysfunction on the PBMC transcriptome. It explores whether expression of genes related to immune functions may underlie the differences in response to weight loss or whether transcriptomic alterations in immune cells may precede more traditional biomarkers of chronic pro-inflammation. These monozygotic twins present an example of divergence of phenotypic outcomes despite identical genetic background and similar treatment response.

Visceral adiposity index performed better than traditional adiposity indicators in predicting unhealthy metabolic phenotype among Chinese children and adolescents

The relationship between visceral adiposity index (VAI) and unhealthy metabolic phenotype remained unclear in children and adolescents. This study aimed to investigate their association and compared the ability of VAI and traditional adiposity indicators (body mass index, waist circumference and waist-to-height ratio) to predict metabolically unhealthy phenotype among normal-weight, overweight and obese children and adolescents. In this cross-sectional study, 1722 children and adolescents aged 12-18 years were selected by cluster random sampling, underwent a questionnaire survey, physical examination and biochemical tests. Participants were divided into four phenotypes according to the combination of the weight status determined by body mass index (BMI) and metabolic syndrome components. Receiver operating characteristic (ROC) analysis and multivariate logistic regression were used to compare the predictive capacity between VAI and traditional adiposity indicators and their relationship with metabolically unhealthy phenotype. We found that VAI had better performance in predicting metabolically unhealthy phenotype than traditional adiposity indicators, the area under the receiver-operating characteristic curve (AUC) were 0.808 and 0.763 for boys and girls with normal-weight, 0.829 and 0.816 for boys and girls with overweight and obese (all P < 0.001). VAI was most strongly related to metabolically unhealthy phenotype whether or not to adjust the age, the adjusted OR and 95%CI was 6.15 (4.13-9.14) in boys with normal weight, and 5.90 (3.06-11.36), 4.95 (2.35-10.41) in boys and girls with overweight and obese, respectively (all P < 0.001). Our findings suggested VAI could be used as a comprehensive predictor to identify unhealthy metabolic phenotype in children and adolescents.

Cross-sectional comparisons of subgingival microbiome and gingival fluid inflammatory cytokines in periodontally healthy vegetarians versus non-vegetarians

BACKGROUND

Vegetarian diets are known to reduce inflammation. The objective of this study was to test the hypothesis that reduced inflammation associated with a vegetarian diet would promote a more commensal subgingival bacterial profile.

METHODS

A total of 39 periodontally healthy subjects (PD ≤3 mm, bleeding on probing <10%) were enrolled. Dietary intake was assessed by a food frequency questionnaire. A comprehensive periodontal examination was performed. Gingival crevicular fluid (GCF) and subgingival plaque samples were collected. GCF samples were assessed for interleukin-1β, interleukin-6, interleukin-8, tumor necrosis factor-alpha, and interleukin-10. Plaque samples were analyzed for bacteria using 16S rDNA sequencing on an Illumina platform. GenBank database was used for taxonomy classification.

RESULTS

Twenty-three subjects were categorized as vegetarian and 16 non-vegetarians. Clinical periodontal measures and GCF cytokine levels were statistically comparable between the two groups. Measures of microbial richness and alpha diversity were also comparable between the two dietary groups. Vegetarians harbored higher levels of phyla associated with gingival health (Actinobacteria, and Proteobacteria). Two species known to be associated with periodontitis (Mogibacterium timidum and Veillonella rogosae) were prominent in non-vegetarians. Pearson's correlations between GCF inflammatory cytokines and microbial taxa differed between vegetarians and non-vegetarians. In vegetarians, the anti-inflammatory cytokine IL-10 positively correlated with two species known to be associated with periodontal health (Peptidiphaga sp. HMT183 and Rothia aeria).

CONCLUSIONS

Diet is directly and indirectly associated with the microbial composition of subgingival plaque. A vegetarian diet may promote a subgingival microbiota associated with periodontal health.

Eight-year longitudinal study of whole blood gene expression profiles in individuals undergoing long-term medical follow-up

Blood circulates throughout the body via the peripheral tissues, contributes to host homeostasis and maintains normal physiological functions, in addition to responding to lesions. Previously, we revealed that gene expression analysis of peripheral blood cells is a useful approach for assessing diseases such as diabetes mellitus and cancer because the altered gene expression profiles of peripheral blood cells can reflect the presence and state of diseases. However, no chronological assessment of whole gene expression profiles has been conducted. In the present study, we collected whole blood RNA from 61 individuals (average age at registration, 50 years) every 4 years for 8 years and analyzed gene expression profiles using a complementary DNA microarray to examine whether these profiles were stable or changed over time. We found that the genes with very stable expression were related mostly to immune system pathways, including antigen cell presentation and interferon-related signaling. Genes whose expression was altered over the 8-year study period were principally involved in cellular machinery pathways, including development, signal transduction, cell cycle, apoptosis, and survival. Thus, this chronological examination study showed that the gene expression profiles of whole blood can reveal unmanifested physiological changes.

Distinct Associations of BMI and Fatty Acids With DNA Methylation in Fasting and Postprandial States in Men

We have previously shown that blood global DNA methylation (DNAm) differs between postprandial state (PS) and fasting state (FS) and is associated with BMI and polyunsaturated fatty acid (PUFA) (negatively and positively, respectively) in 12 metabolically healthy adult Mexican men (AMM cohort) equally distributed among conventional BMI classes. Here, we detailed those associations at CpG dinucleotide level by exploiting the Infinium methylation EPIC array (Illumina). We sought differentially methylated CpG (dmCpG) that were (1) associated with BMI (BMI-dmCpG) and/or fatty acids (FA) (FA-dmCpG) in FS or PS and (2) different across FS and PS within a BMI class. BMI-dmCpG and FA-dmCpG were more numerous in FS compared to PS and largely prandial state-specific. For saturated and monounsaturated FA, dmCpG overlap was higher across than within the respective saturation group. Several BMI- and FA-dmCpG mapped to genes involved in metabolic disease and in some cases matched published experimental data sets. Notably, SETDB1 and MTHFS promoter dmCpG could explain the previously observed associations between global DNAm, PUFA content, and BMI in FS. Surprisingly, overlap between BMI-dmCpG and FA-dmCpG was limited and the respective dmCpG were differentially distributed across functional genomic elements. BMI-dmCpG showed the highest overlap with dmCpG of the saturated FA palmitate, monounsaturated C20:1 and PUFA C20:2. Of these, selected promoter BMI-dmCpG showed opposite associations with palmitate compared to C20:1 and C20:2. As for the comparison between FS and PS within BMI classes, dmCpG were strikingly more abundant and variably methylated in overweight relative to normoweight or obese subjects (∼70–139-fold, respectively). Overweight-associated dmCpG-hosting genes were significantly enriched in targets for E47, SREBP1, and RREB1 transcription factors, which are known players in obesity and lipid homeostasis, but none overlapped with BMI-dmCpG. We show for the first time that the association of BMI and FA with methylation of disease-related genes is distinct in FS and PS and that limited overlap exists between BMI- and FA-dmCpG within and across prandial states. Our study also identifies a transcriptional regulation circuitry in overweight that might contribute to adaptation to that condition or to transition to obesity. Further work is necessary to define the pathophysiological implications of these findings.

Lifestyle Clusters and Cardiometabolic Risks in Adolescents: A Chinese School-Based Study Using a Latent Class Analysis Approach

Background: Unhealthy dietary and lifestyle behaviors are associated with a higher prevalence of non-communicable chronic diseases and higher mortality in adults. However, there remains some uncertainty about the magnitude of the associations between lifestyle behaviors and cardiovascular factors in adolescents. Methods: We conducted a school-based cross-sectional study of 895 Chinese adolescents aged 15-19 years. They participated in a questionnaire survey, physical examination, and blood sample collection. Latent class analysis (LCA) was used to identify heterogeneous subgroups of lifestyle behaviors. A set of 12 latent class indicators, which reflected lifestyle behaviors including dietary habits, physical activity, sleep duration, screen time, and pressure perception, were included in the analysis. Logistic regression analysis was performed to determine whether the derived classes were related to a cardiometabolic risk. Results: In total, 13.7 and 5.6% of the participants were overweight and obese, respectively, and 8.4 and 14.1% reported having pre-hypertension and hypertension, respectively. A two-class model provided the best fit with a healthy lifestyle pattern (65.8%) and a sub-healthy lifestyle pattern (34.2%). There were more female participants with a healthy lifestyle (56.2 vs. 43.8%), whereas there were more males with a sub-healthy lifestyle (45.4 vs. 54.6%), (all P = 0.002). Increased risk of cardiometabolic abnormality (BMI categories, blood pressure and lipids) was not significant across lifestyle patterns, except for waist circumference (70.5 vs 69.1 cm, P = 0.044). There was no significant difference in physical activity and intake of fruit and vegetable between the two patterns. Conclusion: Primary prevention based on lifestyle modification should target patterns of behaviors at high risk in adolescents. Due to the complex effect of lifestyle clusters on cardiometabolic risks, well-designed and prospective studies in adolescents are needed in the future.