Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2017 Aug 2:8:182.
doi: 10.3389/fendo.2017.00182. eCollection 2017.

Is There a Role for Bioactive Lipids in the Pathobiology of Diabetes Mellitus?

Affiliations

Is There a Role for Bioactive Lipids in the Pathobiology of Diabetes Mellitus?

Undurti N Das. Front Endocrinol (Lausanne). .

Abstract

Inflammation, decreased levels of circulating endothelial nitric oxide (eNO) and brain-derived neurotrophic factor (BDNF), altered activity of hypothalamic neurotransmitters (including serotonin and vagal tone) and gut hormones, increased concentrations of free radicals, and imbalance in the levels of bioactive lipids and their pro- and anti-inflammatory metabolites have been suggested to play a role in diabetes mellitus (DM). Type 1 diabetes mellitus (type 1 DM) is due to autoimmune destruction of pancreatic β cells because of enhanced production of IL-6 and tumor necrosis factor-α (TNF-α) and other pro-inflammatory cytokines released by immunocytes infiltrating the pancreas in response to unknown exogenous and endogenous toxin(s). On the other hand, type 2 DM is due to increased peripheral insulin resistance secondary to enhanced production of IL-6 and TNF-α in response to high-fat and/or calorie-rich diet (rich in saturated and trans fats). Type 2 DM is also associated with significant alterations in the production and action of hypothalamic neurotransmitters, eNO, BDNF, free radicals, gut hormones, and vagus nerve activity. Thus, type 1 DM is because of excess production of pro-inflammatory cytokines close to β cells, whereas type 2 DM is due to excess of pro-inflammatory cytokines in the systemic circulation. Hence, methods designed to suppress excess production of pro-inflammatory cytokines may form a new approach to prevent both type 1 and type 2 DM. Roux-en-Y gastric bypass and similar surgeries ameliorate type 2 DM, partly by restoring to normal: gut hormones, hypothalamic neurotransmitters, eNO, vagal activity, gut microbiota, bioactive lipids, BDNF production in the gut and hypothalamus, concentrations of cytokines and free radicals that results in resetting glucose-stimulated insulin production by pancreatic β cells. Our recent studies suggested that bioactive lipids, such as arachidonic acid, eicosapentaneoic acid, and docosahexaenoic acid (which are unsaturated fatty acids) and their anti-inflammatory metabolites: lipoxin A4, resolvins, protectins, and maresins, may have antidiabetic actions. These bioactive lipids have anti-inflammatory actions, enhance eNO, BDNF production, restore hypothalamic dysfunction, enhance vagal tone, modulate production and action of ghrelin, leptin and adiponectin, and influence gut microbiota that may explain their antidiabetic action. These pieces of evidence suggest that methods designed to selectively deliver bioactive lipids to pancreatic β cells, gut, liver, and muscle may prevent type 1 and type 2 DM.

Keywords: arachidonic acid; bioactive lipids; lipoxin A4; maresins; polyunsaturated fatty acids; protectins; resolvins.

PubMed Disclaimer

Figures

Figure 1
Figure 1
Scheme showing probable mechanism by which diabetogenic viruses, streptozotocin (STZ), and alloxan induce the development of type 1 diabetes mellitus (type 1 DM). The same mechanism may occur in non-obese diabetic (NOD) and other animals that are known to develop type 1 DM. Bacterial endotoxin lipopolysaccharide (LPS), an agonist of toll-like receptor-4 (TLR-4), inhibits type 1 DM. LPS administered to NOD mice during the prediabetic state delays the onset and decreases the incidence of diabetes. A multiple-injection protocol of LPS is more effective than a single LPS intervention. LPS administration suppresses spleen T lymphocyte proliferation, increases the generation of T regulatory cells [indicated as (+) in the figure], and reduces the synthesis of T-helper 1 pro-inflammatory cytokines [indicated as (−) in the figure], and downregulates TLR-4 and its downstream MyD88-dependent signaling pathway and enhances IL-4 and IL-10. Multiple injections of LPS induce tolerogenic dendritic cell (DC) subset with low TLR-4 expression and, thus, prevent development of type 1 DM in NOD diabetic mice see text, Figure 2, and Wang et al. (47). Alloxan and STZ and other diabetogenic molecules, including viruses, may block activity of desaturases and, thus, decrease the formation of arachidonic acid, eicosapentaneoic acid, and docosahexaenoic acid that, in turn, leads to deficiency of lipoxins, resolvins, protectins, and maresins, potent anti-inflammatory substances. Polyunsaturated fatty acid and their products may alter gut microbiota and regulate Treg and Teff cells. Bioactive lipids inhibit production of pro-inflammatory cytokines and possess cytoprotective actions that may explain their ability to prevent type 1 DM (see Figures 6–8).
Figure 2
Figure 2
Multiple-injections of lipopolysaccharide (LPS) is effective in preventing type 1 diabetes mellitus (type 1 DM). LPS administration suppresses spleen T lymphocyte proliferation, increases the generation of T regulatory, reduces the synthesis of T-helper 1 pro-inflammatory cytokines [interleukin-2 (IL-2), interleukin-1 (IL-1), interferon-γ (IFN-γ), and tumor necrosis factor-α (TNF-α)], and downregulates toll-like receptor-4 (TLR-4) and its downstream MyD88-dependent signaling pathway and enhances IL-4 and IL-10 and antioxidant defenses. Multiple injections of LPS induce tolerogenic dendritic cell (DC) subset with low TLR-4 expression and, thus, prevent development of type 1 DM in non-obese diabetic mice [see text and Wang et al. (47)].
Figure 3
Figure 3
Scheme showing the metabolism of essential fatty acids, their role in inflammation and the effect of glucose, sucrose, and fructose on the activities of desaturases and formation of various eicosanoids, cytokines, and lipoxins (LXs), resolvins, and protectins. (+) Indicates increase in the activity or enhanced formation. (−) Indicates decrease in the activity or decreased formation. Glucose, sucrose, and fructose decrease activities of Δ6 and Δ5 desaturases and, thus, decrease the formation of arachidonic acid (AA), eicosapentaneoic acid (EPA), and docosahexaenoic acid (DHA) that are precursors of various eicosanoids and LXs, resolvins, and protectins. Glucose, sucrose, and fructose seem to enhance the formation of pro-inflammatory prostaglandins, leukotrienes, and thromboxanes and generation of free radicals and decrease the formation of LXs, resolvins, and protectins that have anti-inflammatory activities and prevent development of type 2 diabetes mellitus and metabolic syndrome and insulin resistance; they may also enhance the formation of pro-inflammatory cytokines and decrease those of anti-inflammatory cytokines. The pro-inflammatory activities of glucose, fructose, and sucrose may be in the order of fructose > sucrose ≥ glucose. Nitrolipids are formed due to interaction between polyunsaturated fatty acids and nitric oxide and these compounds have anti-inflammatory activity. Fibroblast growth factor 1 (FGF1) is a critical transducer of remodeling of adipose tissue in response to fluctuations in nutrient availability that is essential for maintaining metabolic homeostasis and is regulated by the nuclear receptor peroxisome proliferator-activated receptor-γ (PPAR-γ). PPAR-γ is an adipocyte master regulator and the target of the thiazolidinedione class of insulin-sensitizing drugs. FGF1 is the prototype of the 22-member FGF family of proteins and is involved in a range of physiological processes, including development, wound healing, and cardiovascular changes. FGF1 is highly induced in adipose tissue in response to a high-fat diet and that mice lacking FGF1 develop an aggressive diabetic phenotype coupled to aberrant adipose expansion when challenged with a high-fat diet. FGF1-deficient mice have abnormalities in the vasculature network, an accentuated inflammatory response, aberrant adipocyte size distribution, and ectopic expression of pancreatic lipases. It is interesting that withdrawal of the high-fat diet, inflamed adipose tissue fails to properly resolve, resulting in extensive fat necrosis that could be attributed to decreased production of LXs, resolvins, protectins, and maresins. Adipose induction of FGF1 in the fed state is regulated by PPAR-γ acting through a conserved promoter proximal PPAR response element within the FGF1 gene. These results suggest that the PPAR-γ–FGF1 axis is critical for maintaining metabolic homeostasis and insulin sensitization (48). In this context, FGF-19 has been shown to have hypoglycemic actions. Central nervous system responds to FGF-19 administered in the periphery. In mouse models of insulin resistance, leptin-deficiency and high-fat diet feeding and intracerebroventricular infusions of FGF-19 improved glycemic status, reduced insulin resistance and potentiated insulin signaling in the periphery. In addition, central action of FGF-19 included suppression of AGRP/neuropeptide Y neuronal activity (49). Furthermore, high-fat diet (HFD)-fed mice lacking lysosome-associated membrane protein-2 (lamp-2), which is essential for the fusion with lysosome and subsequent degradation of autophagosomes, showed a resistance against HFD-induced obesity, hyperinsulinemic hyperglycemia, and tissue lipid accumulation, accompanied with higher energy expenditure due to high expression levels of thermogenic genes in brown adipose tissue in HFD-fed lamp-2-deficient mice. Serum level of FGF-21 and its mRNA expression level in the liver were significantly higher in HFD-fed lamp-2-deficient mice in an ER stress-, but not PPAR-α-, dependent manner. These results suggest that a lamp-2-dependent fusion and degradation process of autophagosomes, and FGF-21 are involved in the pathogenesis of diabetes implicating a role for autophagy in this process (50). FGF activates phospholipases (–53) that leads to the release of polyunsaturated fatty acid (PUFAs) that, in turn, can be utilized for the formation of various eicosanoids, LXs, resolvins, protectins, and maresins. Thus, PUFAs and LXs resolvins, protectins, and maresins could mediate anti-obesity and antidiabetic actions of FGFs. Alloxan, streptozotocin, and HFD block the activity of Δ6 and Δ5 desaturases and, thus, lead to a decrease in the synthesis and plasma and tissue levels of GLA, DGLA, AA, EPA, and DHA and decreased formation of LXs, resolvins, protectins, and maresins (from AA, EPA, and DHA) that could lead to increase in inflammation [increase in IL-6 and tumor necrosis factor-α (TNF-α)] and failure of resolution of inflammation and tissue repair. This may result in increase in peripheral insulin resistance, inflammation of mesenteric tissue, gut, adipose tissue, and liver (including NAFLD = non-alcoholic fatty liver disease). It may also lead to inflammation of hypothalamic neurons.
Figure 4
Figure 4
Scheme showing possible role of Treg and Teff cells and cytokines associated with these cells and their role in the prevention or development of type 1 diabetes mellitus (type 1 DM). Low-dose IL-2 and tumor necrosis factor-α (TNF-α) seem to prevent insulitis and development of type 1 DM by augmenting immune tolerance and enhancing antioxidant defenses in β cells. High levels of IL-2, TNF-α, and interferon-γ (IFN-γ) are pro-inflammatory in nature, enhance reactive oxygen species (possibly, NO) generation, decrease immune tolerance and antioxidant defenses of β cells, cause insulitis, and eventually lead to the development of type 1 DM (see text for further details). High levels of IL-2 enhance the production of interleukin-1.
Figure 5
Figure 5
Scheme showing possible interaction among high and low doses of IL-2 in inducing and preventing type 1 diabetes mellitus (type 1 DM). It is proposed that high doses of IL-2 induce the activation of iPLA2 and COX-2 that leads to the synthesis and release of PGE2 and LTB4 and other pro-inflammatory molecules that enhance the formation of free radicals leading to apoptosis of pancreatic β cells and onset of type 1 DM. On the other hand, low doses of IL-2/tumor necrosis factor-α (TNF-α) activates sPLA2 and cPLA2 (cPLA2 > sPLA2) that leads to the formation of lipoxins, resolvins, and protectins; anti-inflammatory molecules, which decrease the formation of free radicals and enhance antioxidant capacity of pancreatic β cells and prevents type 1 DM.
Figure 6
Figure 6
Effect of pre-treatment with resolvin D2 and protectin and lipoxin A4 (LXA4) on streptozotocin (STZ)-induced cytotoxicity to RIN5F cells in vitro [these data are taken from Ref. (137)]. (A,B) RIN5F cells were pretreated with 1, 5, 10, and 50 ng/ml of resolvin D2 and protectin, respectively to study its modulatory action on STZ (21 mM)-induced cytotoxic action. (C) RIN5F cells were pretreated with 1, 5, 10, and 50 ng/ml of LXA4 to study its modulatory action on STZ (21 mM) induced cytotoxic action. All values are expressed as mean ± SEM. *P ≤ 0.05 compared to untreated control, #P ≤ 0.05 compared to STZ.
Figure 7
Figure 7
Effect of arachidonic acid (AA) on streptozotocin (STZ)-induced type 1 diabetes mellitus in Wistar rats [these data are taken from Ref. (137)]. These studies were approved by Institutional Animal Ethics committee. After 7 days of acclimatization, animals received 10 µg/ml of AA intraperitoneally (IP) or oral (OR) for 1 week and once in every week, whereas STZ 45 mg/kg of body weight was given only on day 1. (A) Plasma blood glucose levels in animals: blood glucose estimation was performed once in 10 days until the end of the study. All values are expressed as mean ± SEM. aP ≤ 0.05 compared to 10th day control values. bP ≤ 0.05 compared to 20th day control values. cP ≤ 0.05 compared to 30th day control values. dP ≤ 0.05 compared to plasma glucose levels seen on day 10 after STZ alone administration. eP ≤ 0.05 compared to plasma glucose levels seen day 20 after STZ administration. fP ≤ 0.05 compared to plasma glucose levels seen on day 30 after STZ administration. All the above set of experiments were done in triplicate on two separate occasions (n = 6) and values are expressed as mean ± SEM. *P ≤ 0.05 compared to untreated control. #P ≤ 0.05 compared to STZ. (B) Measurement of lipoxin A4 levels in plasma of AA ± STZ treated animals at the end of the study (day 30). (C) Plasma insulin levels in AA ± STZ treated Wistar rats. Insulin estimation was done in the plasma collected at the end of the study. All values are expressed as mean ± SEM. *P ≤ 0.05 compared to untreated control. #P ≤ 0.05 compared to STZ control (positive control group). (D) Plasma tumor necrosis factor-α (TNF-α) level in AA ± STZ treated rats: TNF-α measurement was done in plasma collected once in every 10 days till the end of the study. All values are expressed as mean ± SEM. aP ≤ 0.05 compared to the 10th day control; bP ≤ 0.05 compared to the 20th day control; cP ≤ 0.05 compared to the 30th day control; dP ≤ 0.05 compared to the 10th day STZ control; eP ≤ 0.05 compared to the 20th day STZ control; fP ≤ 0.05 compared to the 30th day STZ control. *P ≤ 0.05 compared to untreated control; #P ≤ 0.05 compared to STZ control. All values are expressed as mean ± SEM.
Figure 8
Figure 8
Effect of lipoxin A4 (LXA4) on streptozotocin (STZ)-induced type 1 diabetes mellitus (type 1 DM) (A–C) and resolvin D1 on STZ-induced type 1 DM (D) [these data are taken from Ref. (137) and unpublished data]. These studies were approved by Institutional Animal Ethics committee. T1D = type 1 DM. After 7 days of acclimatization, animals received 60 ng/ml LXA4 intraperitoneally for 5 days and 45 mg/kg body weight of STZ only on day 1. (A) Plasma LXA4 levels measured on day 30 of the study. (B) Plasma glucose levels: plasma glucose estimation was performed once in 10 days until the end of the study. All values are expressed as mean ± SEM. aP ≤ 0.05 compared to 10th day control values; bP ≤ 0.05 compared to 20th day control values; cP ≤ 0.05 compared to 30th day control values; dP ≤ 0.05 compared to 10th day STZ values; eP ≤ 0.05 compared to 20th day STZ values; fP ≤ 0.05 compared to 30th day STZ values; *P ≤ 0.05 compared to untreated control; #P ≤ 0.05 compared to STZ control. All values are expressed as mean ± SEM. (C) Plasma insulin levels: plasma insulin levels were estimated on day 30. All values are expressed as mean ± SEM. *P ≤ 0.05 compared to untreated control; #P ≤ 0.05 compared to STZ. (D) Plasma glucose levels in STZ-induced type 1 DM treated with resolvin D1 (derived from DHA). *P < 0.05 compared to control.
Figure 9
Figure 9
Effect of serotonin on the proliferation of RIN 5F cells in vitro and its modulatory effect on streptozotocin-induced inhibition (apoptosis) of RIN5F cells.
Figure 10
Figure 10
Scheme showing how tryptophan plays a role in the development of type 1 diabetes mellitus (type 1 DM) and type 2 DM. Tryptophan is an essential amino acid. Gut bacteria convert tryptophan into indole derivatives: indol-3-acetic acid, indoxyl-3-sulfate, indole-3-propionic acid, and indole-3-aldehyde that are ligands for the aryl hydrocarbon receptor (AHR). Tryptophan indole derivatives activate AHR in gut-resident T cells and innate lymphoid cells that produce IL-22, which protects against inflammation. Tryptophan metabolites by signaling through AHR influence a type 1 IFN signaling pathway that reduces NF-κB-driven inflammation (via SOCS2) and inhibits/ameliorates autoimmunity. Thus, tryptophan and its indole metabolites may have a role in autoimmune diseases (AID), such as type 1 DM and central nervous system AID. Tryptophan is also the precursor of serotonin that has immunomodulatory and cytoprotective actions (see Figure 9) and enhances lipoxin A4 production (unpublished data) and possibly that of resolvins, protectins, and maresins; anti-inflammatory bioactive lipids, which prevent type 1 DM (see Figure 8).
Figure 11
Figure 11
Scheme showing possible interaction(s) among gut microbiota, gut microbial metabolites, polyunsaturated fatty acid (PUFAs) and their metabolites [lipoxin A4 (LXA4), resolvins and protectins], and endocannabinoid system. It is possible that endocannabinoid receptors in the hypothalamus and other brain areas also play a role in DM. A stronger role for endocannabinoid system is seen in type 2 DM compared to its role in type 1 diabetes mellitus. It is likely (needs firm evidence) that gut microbiota metabolizes dietary linoleic acid and α-linolenic acid to arachidonic acid (AA) and eicosapentaneoic acid (EPA) and docosahexaenoic acid (DHA), respectively, that may enhance the formation of anti-inflammatory and antidiabetic molecules LXA4 (from AA), resolvins (from EPA and DHA), and protectins (from DHA, see Figures 7 and 8 also). AA, EPA, and DHA may enhance the proliferation of useful microbiota. Thus, there could be a two-way interaction between PUFAs and gut microbiota. PUFAs have antibiotic-like actions and so may suppress the proliferation of harmful bacteria that are associated with obesity. There could be an interaction between endocannabinoid system, gut microbiota metabolites, and hypothalamic neurotransmitters as shown in the figure.
Figure 12
Figure 12
Scheme showing relationship among diet, gut microbiota, vagus, exercise, polyunsaturated fatty acid (PUFAs), lipoxins (LXs), resolvins, protectins, and maresins, and blood glucose, insulin, and tissues/organs concerned with glucose homeostasis: pancreas, muscle, liver, adipose tissue, and brain. High calorie diet induces low-grade systemic inflammation, obesity, and insulin resistance. PUFAs decrease insulin resistance, suppress secretion of pro-inflammatory cytokines, and lead to the formation of (a) LCFAs-CoA; (b) enhance gut cholecystokinin (CCK) secretion; and (c) augment endocannabinoids formation that act via afferent vagal fibers on hypothalamus to induce satiety and decrease appetite. PUFAs lead to increase in the formation of LXs, resolvins, protectins, and maresins that reduce insulin resistance, protect β cells from toxins, inhibit IL-6 and tumor necrosis factor-α (TNF-α) production, augment brain-derived neurotrophic factor (BDNF) production and action, interact with incretins, CCK, and acetylcholine, and influence gut microbiota, and may act on hypothalamic neurons and modulate insulin response of hypothalamic neurons and enhance response of peripheral tissues to insulin (reduce insulin resistance) and, thus, ameliorate type 2 DM. PUFAs enhance the growth of Bacteroidetes and inhibit Firmicutes and, thus, reduce obesity. PUFAs may augment the production and secretion of gut incretins that, in turn, augment insulin secretion. PUFAs enhance BDNF production, which inhibits appetite and decrease obesity. Liver and pancreas talk with each other through vagal fibers. Exercise reduces insulin resistance and obesity by (i) suppressing production of pro-inflammatory cytokines; (ii) increasing BDNF production in the brain and enhanced levels in the plasma; (iii) enhancing the production of lipoxin A4 (and possibly that of resolvins, protectins, and maresins) from muscle and gut; (iv) upregulating glucose utilization; (v) increasing vagal tone and thus, is (vi) anti-inflammatory in nature. Adipose tissue of obese subjects is infiltrated by macrophages and lymphocytes that secrete high amounts of IL-6 and TNF-α that cause low-grade systemic inflammation resulting in insulin resistance. Leptin has pro-inflammatory actions. Bacteroidetes, the predominant bacteria in the gut of the lean subjects, whereas Firmicutes are dominant in the gut of obese. Firmicutes breakdown polysaccharides and, thus, provide higher amounts of energy source that enhances the probability of development of obesity. Firmicutes stimulate gut associated lymphocytes and macrophages and augment production of pro-inflammatory cytokines. Insulin has anti-inflammatory actions and, hence, is likely that hyperinsulinemia seen in obesity and type 2 DM could be a compensatory phenomenon in order to suppress low-grade systemic inflammation seen in them. Insulin enhances activity of desaturases that leads to increased formation of arachidonic acid, eicosapentaneoic acid and docosahexaenoic acid, the precursors of LXs, resolvins, protectins, and maresins. Though expression and genotype (including single nucleotide polymorphism) of UCPs, FOXC2, adiponectin, FTO, MC4R, and other related genes are closely associated with obesity, their expression and function is modulated by diet, exercise, and other life-style-related factors. Thus, a close interaction(s) exists among genes, gut, diet, microbiota, and exercise that is not only complex but also interesting in the pathogenesis of obesity and type 2 DM. It may be noted here that PUFAs and acetylcholine and exercise can influence production and action of various neurotransmitters, such as serotonin, dopamine, leptin, ghrelin, GABA, and α-MSH and, thus, muscle, gut, food, and brain interact with each other and determine development/amelioration/prevention of obesity and type 2 DM.
Figure 13
Figure 13
Effect of lipoxin A4 (LXA4) against streptozotocin (STZ)-induced type 2 DM in Wistar rats. Protocol of the study: after 7 days of acclimatization, type 2 DM was induced by intraperitoneal administration of STZ 175 mg/kg of body weight, which is considered as day 1 of the study. LXA4 60 ng/animal was given intraperitoneally on day 1 (the day STZ was administered) and daily for 5 days [these data are taken from Ref. (420)]. These studies were approved by Institutional Animal Ethics committee. (A) Plasma glucose levels: plasma glucose was estimated once in 10 days till day 30, the day study was concluded. All values are expressed as mean ± SEM. aP ≤ 0.05 compared to control values of day 10. bP ≤ 0.05 compared to control values of day 20. cP ≤ 0.05 compared to control values of day 30. dP ≤ 0.05 compared to STZ values of day 10.eP ≤ 0.05 compared to STZ values of day 20. fP ≤ 0.05 compared to STZ values of day 30. *P ≤ 0.05 compared to untreated control. #P ≤ 0.05 compared to STZ control. (B) Plasma insulin levels measured on day 30 of the study: all values are expressed as mean ± SEM. *P ≤ 0.05 compared to untreated control; #P ≤ 0.05 compared to STZ.
Figure 14
Figure 14
Scheme showing actions of polyunsaturated fatty acid and their anti-inflammatory products: lipoxins (LXs), resolvins, protectins, and maresins on various factors that have a role in the pathobiology of type 1 diabetes mellitus (type 1 DM) and type 2 diabetes mellitus (type 2 DM). (−) indicates inhibition of action or negative control. (+) indicates increase in action or synthesis or positive control. Exercise enhances the formation and action of brain-derived neurotrophic factor (BDNF), LXs, resolvins, protectins, and maresins and suppress inflammation and reduces insulin resistance. BDNF, LXs, resolvins, protectins and maresins suppress inflammation, reduce insulin resistance, and protect pancreatic β cells from the cytotoxic action various endogenous and exogenous cytotoxic molecules/agents. Alloxan, streptozotocin, and high-fat diet suppress the activities of desaturases and reduce the formation of LXs, resolvins, protectins, and maresins and their precursors and the formation and action of BDNF (for details see text).

References

    1. Nielsen JH. Affinity purified human interleukin-1 is cytotoxic to isolated islets of langerhans. Diabetologia (1986) 29:63–7.10.1007/BF02427283 - DOI - PubMed
    1. Dunger A, Cunningham JM, Delaney CA, Lowe JE, Green MH, Bone AJ, et al. Tumor necrosis factor-alpha and interferon-gamma inhibit insulin secretion and cause DNA damage in unweaned rat islets: extent of nitric oxide involvement. Diabetes (1996) 45:183–9.10.2337/diabetes.45.2.183 - DOI - PubMed
    1. Lukic ML, Stosic-Grujicic S, Shahin A. Effector mechanisms in low-dose streptozotocin-induced diabetes. Dev Immunol (1998) 6:119–28.10.1155/1998/92198 - DOI - PMC - PubMed
    1. Movahedi B, Van de Casteele M, Caluwé N, Stangé G, Breckpot K, Thielemans K, et al. Human pancreatic duct cells can produce tumour necrosis factor-alpha that damages neighbouring beta cells and activates dendritic cells. Diabetologia (2004) 47:998–1008.10.1007/s00125-004-1426-3 - DOI - PubMed
    1. Bojunga J, Kusterer K, Bacher M, Kurek R, Usadel KH, Renneberg H. Macrophage migration inhibitory factor and development of type-1 diabetes in non-obese diabetic mice. Cytokine (2003) 21:179–86.10.1016/S1043-4666(03)00076-0 - DOI - PubMed