Insulin Resistance and Environmental Pollutants: Experimental Evidence and Future Perspectives

Background: The metabolic disruptor hypothesis postulates that environmental pollutants may be risk factors for metabolic diseases. Because insulin resistance is involved in most metabolic diseases and current health care prevention programs predominantly target insulin resistance or risk factors thereof, a critical analysis of the role of pollutants in insulin resistance might be important for future management of metabolic diseases. Objectives: We aimed to critically review the available information linking pollutant exposure to insulin resistance and to open the discussion on future perspectives for metabolic disruptor identification and prioritization strategies. Methods: We searched PubMed and Web of Science for experimental studies reporting on linkages between environmental pollutants and insulin resistance and identified a total of 23 studies as the prime literature. Discussion: Recent studies specifically designed to investigate the effect of pollutants on insulin sensitivity show a potential causation of insulin resistance. Based on these studies, a summary of viable test systems and end points can be composed, allowing insight into what is missing and what is needed to create a standardized insulin resistance toxicity testing strategy. Conclusions: It is clear that current research predominantly relies on top-down identification of insulin resistance–inducing metabolic disruptors and that the development of dedicated in vitro or ex vivo screens to allow animal sparing and time- and cost-effective bottom-up screening is a major future research need. Citation: Hectors TL, Vanparys C, Van Gaal LF, Jorens PG, Covaci A, Blust R. 2013. Insulin resistance and environmental pollutants: experimental evidence and future perspectives. Environ Health Perspect 121:1273–1281; http://dx.doi.org/10.1289/ehp.1307082

Authors state that exposure to Pb and As contaminated drinking water induced IR in male rats. This conclusion was derived from calculation of HOMA-IR and confirmation by OGTT tests, in which especially male mice had higher glucose and insulin levels compared to the male control group. IVGTT and hyperinsulinemic-euglycemic clamp technology showed IGT and whole body IR for animals exposed to 30 and 300 µg KD atrazine fed on a regular diet. The insulin sensitivity index was also decreased for both atrazine treatments. In L6 muscle cells, atrazine abolished insulin mediated Akt phosphorylation.

No
Plasma insulin (↓); IPGTT (glucose ) (=) Khalil et al. (2010) report slightly decreased plasma insulin levels in HFD + BaP, while no further worsening of glucose intolerance was observed compared to HFD alone. Decreased expression of glucagon-like peptide 1, an incretin which stimulates insulin secretion, in BaP-treated mice was proposed to explain decreased insulin levels. Furthermore, BaP exacerbated HFD-induced inflammatory gene expression and expression of genes related type 2 diabetes in bowel and/or liver. As such, it was concluded that BaP might increase the risk of type 2 diabetes without worsening IR. injected with insulin, reduced phosphorylation of the insulin receptor β-subunit and of Thr 308 of the Akt protein, compared insulin Glucose stimulated to vehicle-treated insulin exposed mice was observed. secretion from insulin secretion Furthermore, BPA treatment also seemed to affect the MAPK isolated (insulin) (↑); Muscle signaling pathway as illustrated by absence of insulin pancreatic protein expression: IRS-mediated ERK phosphorylation. In liver, impairment of the islets) 1 (↑), PI3K (p85) (=), insulin signaling pathway by BPA could only be observed at Akt (=), pIRβ (↓), pAkt (Ser 473 ) (=), pAkt (Thr 308 ) (↓), pErk (↓); the level of insulin stimulated tyrosine phosporylation of the insulin receptor β-subunit. No further effects in downstream insulin signaling pathways were observed, nor changes in Liver protein MAPK signaling. A pyruvate tolerance test showed no effects expression: IRS-1 (↑), on pyruvate induced gluconeogenesis. In both liver and PI3K (p85) (=), Akt (=), pIRβ (↓), pAkt (Ser 473 ) (=), pAkt (Thr 308 ) (=), muscle, increased IRS-1 expression was observed, suggested to be a counter regulatory mechanism to overcome IR. Based on these observations, the authors conclude that BPA induced pErk (=); Pyruvate IR is mainly due to targeting of skeletal muscle. tolerance test (liver; gluconeogenesis) (=) Srinivasan et Bis ( Glucose uptake (adipose expression and translocation, which diminishes glucose uptake (DEHP) glucose uptake tissue) (↓); Glucose and oxidation. DEHP is suggested to cause glucose in adipose oxidation (adipose intolerance by reducing expression of insulin signaling tissue) tissue) (↓); Adipose intermediates (IRS-1/Akt pathway) or by inhibiting tissue gene expression:

WB
phosphorylations which activate intermediates in the insulin IRec (↓), IRS-1 (↓), signaling cascade. The mechanisms thought to be involved are GLUT4 (↑,=); Adipose increased production of ROS and lipid peroxidation. Based on tissue protein these findings, the authors concluded that DEHP is associated expression: IRec with IR in adipose tissue. Barnes and HgCl 2 Mouse; In vitro 0.5, 1, and 5 µM; 24 h Yes Glucose uptake (3T3-Mercury had no effect on the levels of insulin-mediated P(A) Kircher 2005 (3T3-L1 cell line) L1) (↓) glucose transport, but since mercury itself increased glucose transport in 3T3-L1 cells, the authors concluded that there was a decrease in the insulin-mediated component of glucose transport. The mechanism suggested to underlie the limited cellular insulin response, is induction of stress (increased p38 phosphorylation). Because HgCl 2 decreased insulin-mediated glucose transport, a characteristic of IR, the authors state that HgCl 2 might have an impact on glucose homeostasis. worsened by VHF/S. Furthermore, increased insulin production was observed in VHF/S-fed mice in response to a glucose challenge and insulin induced glucose clearance was reduced in VHF/S-fed relative to C-fed and VHF-fed mice, all features demonstrating a state of IR. Ex vivo evaluation of peripheral IR (muscle) showed that insulin stimulated glucose uptake was reduced in mice fed VHF and VHF/S compared to C-fed mice. In addition, insulin stimulated phosphorylation of Akt was decreased in skeletal muscle both for VHF and VHF/S fed mice. Reducing POP-levels in VHF/S diet, resulted, in general, to a better whole body insulin sensitivity (insulin tolerance test). Glucose levels were the same for both mice fed VHF/S as VHF/S -POPs comparing FPG and GTT, while insulin plasma levels were decreased for VHF/S -POPsmice and glucose stimulated insulin production was also reduced.
An alternative treatment Ex vivo muscle glucose uptake showed enhance insulin action consisted of exposure in VHF/S -POPs -fed mice. For the WD-experiment, it was through diet for 6 weeks.
shown that, in fed condition, mice fed WD/S had a mild Diets were composed of a increase of blood glucose and a dramatic increase in plasma control diet (chow low-fat insulin levels, suggesting whole body IR. GTT and ITT diet), a western diet (WD) confirmed IGT and systemic IR. Insulin stimulated glucose or WD containing farmed uptake was also significantly reduced in WD/S-fed mice. All salmon fillet (WD/S).
these results highlighted, according to the authors, that POPs may have a causal role in metabolic disorders. with that of mice of the control group. VHF-fed mice, however, showed increased plasma insulin concentrations, glucose and insulin intolerance (decreased insulin sensitivity), decreased insulin stimulated glucose uptake in skeletal muscle and more ectopic fat accumulation. Pancreatic β-cell function was considered similar in all groups. The authors conclude that mice challenged with VHF +POPs were protected from IR despite elevated POP accumulation in adipose tissue.

Chemical Study Dose and time Ins Endpoints
Observations suggested to be associated with IR IR Sun et al. 2009 PM 2.5 Mouse; In vivo; Ex vivo (insulin Mice on a high fat diet were exposed for 128 day Yes FPG (↑); FPI (↑); HOMA-IR (↑); IPGTT HOMA-IR index was significantly higher for PM 2.5 -exposed mice on a HF diet, compared to HF-diet mice, indicative for WB responses in to concentrated PM 2.5 for 6 (glucose) (↑); Aortic whole body IR. Insulin signaling in aortic segments from aorta segments) hours/day (5 days/week).
protein expression: PM 2.5 -exposed mice was impaired, as illustrated by decreased pAkt/total Akt (↓), PKC Akt phosphorylation and changes in PKC expression. isoforms (↑) Attenuation of the PI3K/Akt insulin signalling pathway was implied. Xu et al. 2011 PM 2.5 Mouse; In vivo Mice were exposed to ambient concentrated In vitro concentrated PM 2.5 at HOMA-IR (↑); IPGTT increased FPG and FPI and were insulin resistant as indicated (human, LX-2 nominal 10x (glucose) (↑); Liver by increased HOMA-IR. Furthermore IGT was observed in hepatic stellate concentrations for 6 glycogen content (↓); IPGTT. cell line) hours/day (5 days/week) Liver protein In liver, glycogen content was significantly decreased in for a total of 3 to 10 weeks. expression: pIRS1 (Ser 636 ) (↑), pIRS1 (Ser 1101 ) (↑), pAkt/total animals exposed for 10 weeks. Mechanistically, this is supported by increased (inhibitory) phosphorylation of IRS-1 at Ser 636 and Ser 1101 . In addition, phosphorylation (activation) Akt (↓) of Akt, a key regulator of insulin mediated glycogen metabolism downstream of IRS-1 was decreased. In vivo exposure to PM 2.5 thus reduces insulin signaling via the IRS-1/Akt pathway. This was confirmed in vitro in a human stellate cell model. Besides effects on insulin and glucose homeostasis, long term PM 2.5 -exposure also induced expression of inflammatory pathways in liver as well as hepatic steatosis. Brook et al. PM 2.5 Human; In vivo Transport of subjects from No HOMA-IR (↑) Sub-acute exposure to prevailing PM 2.5 concentrations are WB 2013 locations with background associated with reduced metabolic insulin sensitivity among levels of PM 2.5 (5-10 µg/m 3 ) to an urban site healthy adults. Conclusion based on increased HOMA-IR after translocation to more polluted area for 5 days. with high PM 2.5 levels (mean 11.5 ± 4.8 µg/m 3 ). Exposure occurred during 5 consecutive days, for 4 to 5 h. Although no change in glucose tolerance and FPI was observed, FPG and HOMA-IR were significantly increased in mice exposed to a mixture of concentrated PM 2.5 and nickel compared to the control condition, and synergistically increased compared to PM 2.5 alone. Increased HOMA-IR is indicative for induction of IR. Yes Glucose oxidation (↓) Isolated adipocytes from mice exposed for 4 weeks to 14 mg KD PBDE showed decreased insulin stimulated glucose oxidation. No effect was observed for mice exposed for 2 weeks, nor for isolated adipocytes from control mice, exposed in vitro to PBDEs. Reduction in insulin signaling is a proposed explanation for the observed effects. The decreased glucose oxidation rate and significant increase in lipolysis are suggested to be associated with IR. Pretreatment of cells with TCDD did not alter insulin responsiveness with regard to glucose uptake. Insulin stimulated glucose uptake in the presence of TCDD, however, was dose-dependently attenuated. Inhibition of AhR gave comparable results, indicating AhR-independent action of TCDD. The authors postulate that "the involvement of TCDD in interfering with glucose uptake by adipocytes may lead to IR and disruption of glucose homeostasis." a Columns give information on the compound studied, the species used and type of experiments (in vivo, ex vivo or in vitro) (Study), the dose and length of exposure (Dose and time), whether insulin was present in some or all of the assays used to assess the degree of IR (Ins), the endpoints which made authors decide on presence of IR after pollutant exposure and finally, the type of IR studied (WB = whole body; H = hepatic; P = peripheral (A = adipose tissue; M = muscle)).

Source
b Only the general trend is provided for pollutant effects. In some of the cases, co-treatment with different diet types may influence the results on insulin sensitivity related endpoints or not. For those studies, only the general results of the pollutant-treatments are given. Furthermore, if more doses were tested, compiled results which made the authors decide on the presence of IR are represented. Studies are sorted alphabetically based on the name of the chemical.
c Only changes of HFC or VHF/S diet with regard to insulin/glucose homeostasis are given compared to control diets. Comparison with HFR or VHF/S -POPs sometimes deviates from the observations reported here. For all results we refer to the original papers (Ibrahim et al. 2011;Ruzzin et al. 2010).