Chromatin Regulation of Adipose Tissue
Obesity prevalences are increasing alarmingly, thus constituting a profound socioeconomic burden. At the individual level, obese are at risk for serious co-morbidities such as type 2 diabetes mellitus (T2D), cardiovascular disease, and certain types of cancer. Obesity is a consequence of chronic positive energy balances, leading to accumulation of dietary lipids in adipose tissue and chronic low-grade inflammatory processes. Interestingly, homeothermal mammals harbour two anatomically and functionally distinct types of fat: White adipose tissue (WAT) stores excess nutrient-derived energy and secretes circulating factors (adipokines) for coordinating systemic energy balances, whereas brown adipose tissue (BAT) dissipates nutrient excess as heat during cold stress, a process known as ‘non-shivering thermogenesis (NST)’.
Intriguingly, brown adipose activation results in exaggerated lipid catabolism and glucose oxidation, thereby mitigating the development of obesity and metabolic disease in rodents. The rediscovery of active BAT also in adult humans, and the tight positive correlations of BAT mass and function with lower body mass indices and amelioration of T2D in human subjects suggests that BAT activation can counteract Metabolic Diseases in people and first preclinical studies point into this encouraging direction.
One bottleneck towards safe and efficient ‘thermogenesis drugs’ is our limited understanding of the regulatory mechanisms - not only of the genetic, but also epigenetic regulatory mechanisms, for instance the role of post-translational reversible histone modifications, DNA methylation and mono-allelic gene expression (‘gene imprinting’) or Long Noncoding RNAs that orchestrate BAT gene regulation and function.
Long Noncoding RNA Regulation of Adipose Gene Imprinting
Increasing brown adipose tissue (BAT) thermogenesis in mice and humans improves metabolic health and understanding BAT function is of interest for novel approaches to counteract obesity. The role of nongenetic gene regulation in these processes remains elusive. In a recent study (Schmidt et al 2018 Nature Communications), we observed that monoallelically (maternally) expressed, imprinted lncRNA H19 increased upon cold activation and decreased in obesity in BAT. Inverse correlations of H19 with BMI were also observed in healthy human individuals. We could demonstrate that increasing H19 expression promoted, whereas short interfering RNA (siRNA) silencing of H19 impaired adipogenesis, oxidative metabolism and mitochondrial respiration in brown but not white adipocytes.
H19 overexpression in vivo protected against diet-induced obesity, improved insulin sensitivity and mitochondrial biogenesis, whereas tissue-selective loss of H19 sensitized towards high-fat diet induced weight gains. Strikingly, striktly paternally expressed monoallelic genes (PEG) were to a high degredd absent from BAT and were able to demonstrated that H19 recruits PEG-inactivating H19-MBD1 complexes, thereby acting as BAT-selective gatekeeper preventing PEG expression. Our findings have important implications for understanding how monoallelic gene expression affects metabolism in rodents and, potentially, humans.
Interplay of micronutrient metabolism and histone modification
We are interested in identifying epigenetic and transcriptional regulators correlating with thermogenic gene induction in Health and Metabolic Disease. To this end, we performed bulk RNA seq and H3K4me3 Chromatin immuno-precipitation coupled to sequencing (ChIP seq) in brown adipose tissue (BAT) of mice. Subsequently we exposed mice to chronic high fat diet (HFD) or low fat control diet (LFD) feeding, followed by cold exposure (5C) to induce non-shivering thermogenic, increase calorie turnover and thus improve metabolic homeostasis. Intriguingly, and by performing H3K4me3 ChIP seq analysis, we found genome-wide levels of H3K4me3 (a promoter-proximal chromatin mark believed to correlate positively with transcriptional activation) starkly increased after HFD feeding. Short term induction of BAT thermogenesis and concomitant increases in turn restored low H3K4me3 chromatin domains comparable to those found in lean mice housed (‘healthy chromatin’). Crucially, these reversible H3K4me3 alterations happened independently of concordant changes in gene expression, arguing for ‘non-canonical’, non-transcription relevant roles for (tri)-methylation of H3K4 and possibly other histone methylated residues during thermogenic activation of brown adipocytes.
An increasing body of evidence and overwhelming findings from genome-wide studies suggest that little transcription changes arise upon removal or addition of H3K4me3 under dynamically changing conditions, suggesting that these post-translational modifications could rather reflect memories of previous cellular states, or a reflection of intermediate metabolite levels capable of fuelling energetic reactions. We thus hypothesize that histone methylation levels in brown adipose tissue (a) reflect local, i.e. cell/tissue-level or systemic, i.e. organism-level energetic states and that (b) enzymatic removal of histone methyl groups support oxidative chemical conversions that ultimately serve a metabolic purpose.
(Illustration generated using BioRender)