Introduction
Oxidative stress is involved in the pathophysiology of many chronic diseases and in particular contributes to the development of insulin resistance and its progression towards type 2 diabetes [
1‐
3]. Peroxidation of cell membrane phospholipids associated with oxidative stress produces deleterious reactive species. Peroxidation of
n-6 polyunsaturated fatty acids (PUFA) leads to the production of 4-hydroxy-2-nonenal (4-HNE), while 4-hydroxy-2-hexenal (4-HHE) is released during the oxidation of
n-3-PUFA [
4]. These lipid aldehydes are major by-products of lipid peroxidation of PUFA and exhibit potent electrophilic properties allowing them to form covalent adducts with phospholipids, proteins and nucleotides [
5,
6]. Because of their relative stability and high reactivity, these aldehydes are thought to interfere with crucial physiological processes such as cell cycle, apoptosis or metabolic pathways [
7‐
9]. Importantly, the production of 4-hydroxyalkenals is associated with hindered insulin responses: 4-HNE-adducts accumulate in liver and pancreatic beta cells of diabetic rats [
10‐
12], impairs glucose-induced insulin secretion in isolated beta cells [
13] and blunts insulin action in 3T3-L1 adipocytes and L6 muscle cells [
14,
15].
Increased consumption of
n-3 PUFA might be expected to produce beneficial effects [
16]; however, enhanced 4-HHE formation in conditions associated with oxidative stress might be harmful. For instance, consumption of oxidised
n-3 PUFA induces oxidative stress and inflammation of mice intestine [
17]. 4-HHE has, however, received little attention, despite its similarities in structure and its reactivity with related aldehydes. Indeed, 4-HHE is also produced under oxidative stress conditions, accumulates in tissues [
18,
19] and is able to form adducts on biological molecules [
20,
21]. Only a handful of studies have shown activation of stress signalling pathways by 4-HHE [
22,
23] but data regarding its pathophysiological effects remains scarce. Especially, the putative role of 4-HHE in the development of insulin resistance has not been investigated. In the present study we hypothesised that circulating levels of 4-HHE are elevated in diabetic individuals and that 4-HHE can impair insulin responses in skeletal muscle cells and contribute to insulin resistance in vivo.
Discussion
Diabetes is associated with increased oxidative stress in metabolic tissues and excessive production of reactive oxygen species negatively affects insulin responses [
2]. Oxidative stress can impair cell function through direct attack by reactive oxygen species and via the production of reactive oxidation by-products. Here, we demonstrate that one of the by-products of
n-3 PUFA peroxidation, 4-HHE, is increased in plasma during type 2 diabetes, induces insulin resistance in vivo and impairs glucose uptake and signalling in skeletal muscle cells in vitro.
Unlike radical oxygen species, lipid peroxidation by-products are long-lived and may spread from their site of production to exert their effects throughout the whole organism. Previous reports show that the plasma concentration of free 4-HNE ranges from 70–600 nmol/l at baseline to up to 2–10 μmol/l under pathological conditions [
11,
30,
31]. In our study, the 4-HNE plasma concentration in healthy volunteers was 50 nmol/l and was surprisingly unchanged in individuals with type 2 diabetes. On the contrary, 4-HHE concentration was doubled in individuals with type 2 diabetes compared with healthy individuals. The measurement of free 4-HHE or 4-HNE by GC does not take into account the amount of aldehydes bound to biomolecules. There are no commercially available ELISA kits for 4-HHE adducts, so we used dot blots to demonstrate that plasma proteins from individuals with type 2 diabetes exhibit higher levels of 4-HHE Michael adducts (sevenfold increase). The accumulation of protein adducts (carbonyls) has been reported in individuals with type 2 diabetes; urinary levels of acrolein adducts are increased and significantly correlated with control of blood glucose [
32]. A previous study has reported a fivefold increase in 4-HHE Michael adducts on phospholipids during diabetic retinopathy [
33]. Our study therefore reinforces the evidence for increased production of aldehyde by-products and their adducted targets during diabetes.
Daily supplementation with 800–1600 mg docosahexaenoic acid (DHA) in healthy volunteers significantly increased plasma free 4-HHE (from 9 to 93 nmol/l), likely resulting from increased lipid peroxidation [
34]. Interestingly, a DHA supplement that did not significantly increase plasma 4-HHE (400 mg DHA/day) had beneficial effects on platelet function and induced antioxidant effects [
35], more recently confirmed in individuals with type 2 diabetes [
36]. In human volunteers who take
n-3 PUFA supplements, the balance between the beneficial effects of DHA supplementation and the potential deleterious effects of its by-products is therefore difficult to assess and might depend on the concentration of DHA as well as the oxidative environment.
Data from animal models and cell culture suggest that oxidative stress plays a causative role in the development of type 2 diabetes. Reactive oxygen species and their by-products can have a negative impact on insulin sensitivity [
2] and we demonstrate here that 10 μmol/l 4-HHE is sufficient to impair insulin-induced glucose uptake. In rat L6 muscle cells, impaired glucose uptake likely results from an alteration of insulin-induced PKB/Akt phosphorylation on serine 473 as well as IRS1 tyrosine phosphorylation and p85 docking. Insulin-induced IRS1 phosphorylation can be counteracted by serine phosphorylation conducted for instance by mitogen-activated protein kinase [
37,
38]. However, extracellular signal-regulated kinase and c-Jun N-terminal kinases were only mildly activated by 4-HHE in L6 cells (data not shown), suggesting that these kinases do not play a major role in the impairment of insulin signalling in this context. On the contrary, we detected a significant accumulation of Michael adducts in L6 muscle cells exposed to 4-HHE, suggesting that the major effects of 4-HHE are due to protein carbonylation. Detoxification of aldehydes in cells is fulfilled by several enzymes and antioxidant systems. Compared with 4-HNE, 4-HHE is a poor substrate for aldehyde dehydrogenase 5A [
39] but is metabolised to GSH adducts more efficiently [
40], suggesting that GSH metabolism may be the major mechanism for detoxification of 4-HHE. In our study, prevention of carbonylation either by increasing glutathione pools with D3T or by using the by-product scavenger aminoguanidine reversed 4-HHE-induced insulin resistance. This favours the notion that 4-HHE impairs insulin signalling through the formation of covalent adducts on key proteins and that GSH is a major means by which to prevent its deleterious effects. In 3T3 adipocytes, 4-HNE directly binds IRS1 and promotes its degradation [
14]; however, we did not observe any change in the total amount of PKB/Akt and IRS1 proteins in response to 4-HHE. It is therefore likely that 4-HHE exerts its noxious effects through adduction of proteins but not through the specific degradation of proteins of the insulin signalling pathway.
In conclusion, we demonstrate that plasma levels of 4-HHE are significantly increased in type 2 diabetes and that 4-HHE can significantly impede insulin action in vitro and in vivo. We also report that increasing the GSH pool is an efficient way to prevent 4-HHE-induced carbonylation of cellular proteins and impairment of insulin signalling. These data support the idea that lipid peroxidation by-products, especially 4-HHE, can significantly contribute to the development of type 2 diabetes and could represent a therapeutic target to taper insulin resistance.
Acknowledgements
We gratefully acknowledge A. Geloën (CNRS, France) for fruitful discussions, R. Vella, A. Carravieri and Y.-H. Chionh (INSA-Lyon, France) for contributing to the experiments, A. Makino (INSA-Lyon, France) for advising on confocal microscopy, R. Colas (INSA-Lyon, France) for the kind gift of human plasma and P. Moulin and C. Pelletier (both from Hospices Civils de Lyon, France) for the selection of volunteers. Some of the data were presented as an abstract at the 53rd EASD Annual Meeting in 2017.
Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.