High fat diet has sexually dimorphic effects on body composition, adiposity and glucose homeostasis in Poly(A)- binding protein 4 (Pabp4) knockout mice
Scanlon, Jessica Patricia
MetadataShow full item record
Obesity can lead to a range of health problems including type 2 diabetes (T2DM), cardiovascular disease and non-alcoholic fatty liver disease (NAFLD), and causes an estimated 2.8 million deaths annually (2016). It is a growing epidemic affecting over 600 million people worldwide (in 2014), with 26.8% of the adult population in England alone being obese, an increase of 10% in the last decade, and 62.9% overweight or obese. This trend is predicted to continue, and is attributed to an increasingly sedentary lifestyle, coupled with a high calorie “western diet”, which is estimated to cost >£25billion/year in the UK (2015), which is predicted to rise to £49.9 billion by 2050. It is clear that both sex and genetics affect the extent to which individuals exposed to a high fat diet develop adiposity and its associated morbidities, although the mechanisms underlying these differences are not well understood. Here we explore this aetiology, focusing on poly(A)-binding protein 4 (PABP4), an RNA-binding protein in which polymorphisms associated with altered cholesterol levels and cardiovascular disease risk were identified in human GWAS studies. To this end, I take advantage of an unpublished Pabp4 knock-out mouse, maintained on either normal (ND) or high fat diet (HFD), to explore the role of PABP4 in determining the response to high fat diet. PABP4 is a poorly characterised member of the PABP family, which are multifunctional central regulators of global and mRNA-specific translation, and stability. In cell lines, PABP4 is predominantly cytoplasmic, consistent with such functions. However, analogously to PABP1, the prototypical PABP family member, PABP4 is relocalised to stress granules or the nucleus in response to specific cellular stresses and/or viral infections, suggesting a role in altering gene expression programs in responses to changing cellular conditions. Whilst the expression pattern of PABP4 within tissues has not been previously characterised, western blotting of adult mouse tissues revealed that PABP4 is highly expressed in tissues relevant to obesity, T2DM and NAFLD, such as adipose, pancreas, liver and muscle, consistent with the idea that it may play a role in regulating gene expression programs in response to HFD. Immunohistochemistry of tissue sections provided additional insight, revealing a distinct cellular distribution of PABP4 in some tissues, when compared to the well characterised PABP1. Birth weight and post-birth growth can affect adult metabolism. In particular, low birth weight and catch-up growth, characterised by preferentially putting down adipose over lean mass, increases the risk of metabolic conditions in adulthood, such as obesity, T2DM and cardiovascular disease. Therefore, Pabp4-/- and wildtype mice were weighed at birth and daily until weaning. Interestingly this revealed that Pabp4-/- mice have a reduced weight at birth that is exacerbated to weaning (21days (P21)) (5.7% and 18.3% reductions respectively). This analysis also uncovered a reduced survival to weaning, with both male and female Pabp4-/- mice being present at sub-Mendelian ratios by P21 (p=0.0056). Whilst most death occurred neonatally, Pabp4-/- mice showed an increased rate of attrition until weaning, preceded in some cases by an arrest of weight gain. Weight gain was also tracked from 4 weeks to 12 weeks of age on normal diet showing that Pabp4-/- mice had reduced weight into adulthood (12% reduction at 12wks). Analysis of weight gain by sex uncovered a sexually dimorphic effect of Pabp4-deficiency, with female, but not male, Pabp4-/- mice remaining reduced in weight compared to wildtype after 8 weeks on ND (13.4% reduction in female weight). Body composition analysis showed that fat mass was equivalent to wildtype at 12 weeks of age in both sexes but that female Pabp4-/- mice had a 14.3% reduction in lean mass. Neither the catch-up growth in males nor the reduced lean mass in females was sufficient to result in a change in glucose homeostasis. As the risk of developing metabolic disorders in adult life is a consequence of both genetic and environmental factors, such as diet, Pabp4-/- were placed on a HFD at 4 weeks of age for 8 weeks. HFD models the ‘western’ diet, and has been shown to induce obesity, insulin resistance and glucose intolerance in wildtype mice. Whereas Pabp4-/- mice were only distinguishable from wild-type in terms of female lean mass on normal diet, pronounced sexually dimorphic differences were observed in HFD fed mice. Male Pabp4-/- mice appeared to be partially protected from the negative effects of an 8 week HFD regimen, with a 44% decrease in adipose mass gain compared to wildtype despite equal lean mass. Pabp4-/- male mice also had significantly reduced ectopic lipid stores, with an 81% decrease in hepatic triglyceride concentration compared to wildtype, meaning that NAFLD has not developed. Furthermore, Pabp4- /- male mice did not develop hyperinsulinemia on HFD and retained insulin sensitisation (assessed via glucose tolerance test (GTT) and insulin tolerance test (ITT)), although they displayed wildtype-like elevated plasma glucose concentrations (compared to ND). Western blotting had detected high PABP4 levels in the pancreas, indicating a possible pancreatic origin of these alterations. However, immunofluorescence revealed that PABP4 was confined to the exocrine portion of the pancreas, and was undetectable in the insulin producing pancreatic beta cells, suggesting this phenotype may not be beta cell in origin. This is consistent with the fact that the Pabp4-/- male mice retained an appropriate glucose-induced burst of insulin secretion, and therefore insulin production appears unimpaired. Thus, the primary defect may reside in the exocrine pancreas, which aids digestion, or in other key metabolism related tissues (e.g. muscle, liver, adipose and brain), or a combination thereof. In HFD fed wildtype mice, insulin resistance is caused by increased adiposity and ectopic lipid depots, which blunt insulin stimulated signalling cascades, meaning that the normal responses to insulin (e.g. cellular up take of glucose in muscle and arrested glucose production in liver, to decrease plasma glucose concentrations), are impaired. Therefore, the absence of insulin resistance in HFD fed Pabp4-/- male mice may be a consequence of the reduced increase in adipose mass and ectopic lipid deposits detected in these mice, and their consequent lack of inhibition on insulin signalling pathways. The reduced adiposity was not a result of reduced food intake or dietary fat absorption as male Pabp4-/- mice did not eat less nor exhibit apparent steatorrhea (fatty stools). These results highlight that the Pabp4-/- male mice appear to have an alteration in energy use/storage, and the investigation of this will form the basis of future work. When fed HFD, female Pabp4-/- mice revealed a divergent phenotype to that of wildtype female mice and Pabp4-/- male mice. HFD fed Pabp4-/- female mice showed no difference to HFD-fed wildtype mice in terms of weight, but still exhibited the reduction in lean mass seen on ND, but now with a 22.8% increase in volume of adipose tissue. Together, this means that HFD fed Pabp4-/- females have a higher body fat percentage (32.6% compared to 25.9 % for wildtype females). In contrast to the males, there was no difference in terms of hepatic triglycerides in HFD fed Pabp4-/- female mice and they showed greater hyperglycaemia than wildtype (GTT), although like males they retained insulin sensitisation (ITT). These potentially conflicting results in terms of insulin sensitivity and plasma glucose concentrations may result from the alterations in body composition, which can confound results when lean mass is altered and total body weight is used for calculating doses for GTT/ITT. Interestingly, adiponectin, an adipokine normally found in inverse proportion to adipose mass, was increased in plasma from HFD fed Pabp4-/- female mice (21% increase from HFD fed wildtype mice). Whilst surprising given the increase fat mass of Pabp4-/- females, the insulin sensitising properties of adiponectin may help to explain the retained insulin sensitivity detected in the female Pabp4-/- mice. The finding that HFD revealed metabolic differences in the Pabp4-/- mice lead to the question of whether Pabp4-/- mice have issues adapting to other situations which require modulation of energy storage and glucose homeostasis. One such event is pregnancy, when maternal regulation of insulin resistance is tightly modulated throughout gestation. We therefore characterised the maternal Pabp4-/- environment in late pregnancy (E18.5), when insulin sensitivity decreases to 40-60% lower than pre-pregnancy which results reduced maternal glucose uptake, freeing the glucose up for the rapidly developing foetus. Pregnant Pabp4-/- mice had elevated plasma insulin concentration post fasting (63.7% increase), however glucose homeostasis was wildtype-like, both in terms of plasma glucose and insulin concentrations, throughout a GTT. However, plasma glucose and insulin concentrations in E18.5 Pabp4-/- foetuses were significantly decreased (9% and 44.3% respectively). Pabp4-/- foetuses also had reduced foetal and placental weight/length parameters. This establishes that the differences in weight observed at birth were present by late gestation and secondly, that the reductions in both foetal glucose and insulin concentrations which may contribute to or underlie the reduced growth. It also suggests that the differences seen in adulthood on HFD may be a consequence of metabolic differences present during pregnancy. Taken together, these data support the hypothesis that PABP4 plays a key role in the regulation of mRNAs which are important in growth, post-natal survival and metabolic adaption to high fat diet. Pabp4-/- mice therefore not only provide a model for intrauterine growth restriction, as previously described, but add to an emerging picture (e.g. from genetic manipulation of 4EBP, PERK, eIF6 and eIF2 in mice) that translational control contributes to the regulation of growth, adiposity and glucose homeostasis. The Pabp4-/- mice model the gender and diet specific differences seen in human metabolic health, and thus provide a tool to identify pathways that may be open to modulation to tackle diet induced obesity.