Birth weight (BW) variation is influenced by fetal and maternal genetic and non-genetic factors, and has been reproducibly associated with future cardio-metabolic health outcomes. These associations have been proposed to reflect the lifelong consequences of an adverse intrauterine environment. In earlier work, we demonstrated that much of the negative correlation between BW and adult cardio-metabolic traits could instead be attributable to shared genetic effects. However, that work and other previous studies did not systematically distinguish the direct effects of an individual’s own genotype on BW and subsequent disease risk from indirect effects of their mother’s correlated genotype, mediated by the intrauterine environment. Here, we describe expanded genome-wide association analyses of own BW (n=321,223) and offspring BW (n=230,069 mothers), which identified 278 independent association signals influencing BW (214 novel). We used structural equation modelling to decompose the contributions of direct fetal and indirect maternal genetic influences on BW, implicating fetal- and maternal-specific mechanisms. We used Mendelian randomization to explore the causal relationships between factors influencing BW through fetal or maternal routes, for example, glycemic traits and blood pressure. Direct fetal genotype effects dominate the shared genetic contribution to the association between lower BW and higher type 2 diabetes risk, whereas the relationship between lower BW and higher later blood pressure (BP) is driven by a combination of indirect maternal and direct fetal genetic effects: indirect effects of maternal BP-raising genotypes act to reduce offspring BW, but only direct fetal genotype effects (once inherited) increase the offspring’s later BP. Instrumental variable analysis using maternal BW-lowering genotypes to proxy for an adverse intrauterine environment provided no evidence that it causally raises offspring BP. In successfully separating fetal from maternal genetic effects, this work represents an important advance in genetic studies of perinatal outcomes, and shows that the association between lower BW and higher adult BP is attributable to genetic effects, and not to intrauterine programming.

This study was designed to investigate the effects of physical conditioning on the expression of the insulin sensitive glucose transporter 4 protein (GLUT4) on mononuclear cells and HOMA-IR levels in dogs and compared to results reported in human skeletal muscle and the skeletal muscle of rodent models. Blood was sampled from conditioned dogs (n=8) and sedentary dogs (n=8). The conditioned dogs were exercised four months prior the experiment and were following a uniform training protocol, whereas the sedentary dogs were not. GLUT4 expression in mononuclear cells and plasma insulin levels were measured using commercially available enzyme-linked immunosorbent assay (ELISA). Blood glucose levels were determined using blood plasma. HOMA-IR was calculated using plasma insulin and blood glucose levels using the linear approximation formula. Our results indicate that the state of conditioning had a significant effect on the GLUT4 expression at the surface of mononuclear cells. HOMA-IR was also affected by conditioning in dogs. GLUT4 levels in mononuclear cells of sled dogs were inversely correlated with the homeostasis model assessment of insulin sensitivity. This study demonstrates that conditioning increases GLUT4 levels in mononuclear cells of sled dogs as it has been previously reported in skeletal muscle. Our results support the potential of white blood cells as a proxy tissue for studying insulin signaling and may lead to development of a minimally invasive and direct marker of insulin resistance. This may be the first report of GLUT4 in mononuclear cells in response to exercise and measured with ELISA.

Using sled dogs as exercise model, our objectives of this study were to (1) assess the effects of one acute bout of high-intensity exercise on surface GLUT4 concentrations on easily accessible peripheral blood mononuclear cells (PBMC) and (2) compare our findings with published research on exercise induced GLUT4 in skeletal muscle. During the exercise bout, dogs ran 5 miles at approximately 90% of VO2 max. PMBC were collected before exercise (baseline), immediately after exercise and after 24 h recovery.GLUT4 was measured via ELISA. Acute exercise resulted in a significant increase on surface GLUT4 content on PBMC. GLUT4 was increased significantly immediately after exercise (~50%; p<0.05) and reduced slightly by 24 h post-exercise as compared to baseline (~22%; p>0.05). An effect of acute exercise on GLUT4 levels translocated to the cell membrane was observed, with GLUT4 levels not yet returned to baseline after 24 h post-exercise. In conclusion, the present investigation demonstrated that acute high-intensity exercise increased GLUT4 content at the surface of PBMC of sled dogs as it has been reported in skeletal muscle in other species. Our findings underline the potential use of peripheral blood mononuclear cell GLUT4 protein content as minimally invasive proxy to investigate relationships between insulin sensitivity, insulin resistance, GLUT4 expression and glucose metabolism.

This study provides evidence to support exploration of PBMC as a proxy tissue for studying GLUT4 response to exercise or other noninsulin factors.

The insulin responsive glucose transporter, GLUT4 is found predominantly in muscle and adipose cells. Maratou and others (2007) reported that there is GLUT4 in white blood cells (WBC) collected from human subjects in response to insulin activation. This study was designed to validate the presence of GLUT4 in white blood cells of sled dogs and furthermore to investigate whether changes in levels of the GLUT4 protein might be associated with aging. Additionally, we examined the blood insulin concentration of two populations of dogs, young and old, before and after a meal to observe their insulin response. It is documented in skeletal muscle that GLUT4 expression is increased as a result of conditioning, making sled dogs an excellent model in the circumpolar north for studying the effects of exercise, nutrition and diabetes (Felsburg 2002; Kararli 2006). Blood was withdrawn from 11 healthy sled dogs: 6 young (1–5 years) and physically fit, conditioned for racing and 5 old (7–13 years), retired from racing. The insulin response was determined using blood plasma and ELISA. The buffy coat (containing WBC) was collected with a glass pipette after centrifugation and washed and suspended in 1x phosphate buffer. GLUT4 was measured using ELISA kits (USCN Life Sciences). The results validate that GLUT4 is present in white blood cells in sled dogs. Age had no significant effect in the concentration of GLUT4 between the populations of old and young dogs. A significant difference in insulin levels pre and post meal in young (0.13 ± 0.03 ng/mL (pre), 0.22 ± 0.04 ng/mL (post), p < 0.05) and old (0.13 ± 0.02 ng/mL (pre), 0.22 ± 0.03 ng/mL (post), p < 0.05) dogs was observed, displaying the typical postprandial insulin spike. No significant difference was found in insulin concentration comparing old versus young dogs. Our data shows that white blood cells in young (40.4 ± 2.4 ng/mL) and old (35.3 ± 8.8 ng/mL) sled dogs have quantifiable but non-significant different GLUT4 levels (p > 0.05). Detecting GLUT4 via an ELISA in white blood cells, opens up minimally invasive avenues for studying the underlying molecular mechanisms associated with insulin resistance in more complex, dynamic and physiological systems. This project was the first step in developing a protocol for this simple, technique with a potential clinical application for diagnosing insulin resistance.