12-Lipoxygenase Products Reduce Insulin Secretion and {beta}-Cell Viability in Human Islets

Abstract

Context: Inflammation is increasingly recognized as an important contributing factor in diabetes mellitus. Lipoxygenases (LOs) produce active lipids that promote inflammatory damage by catalyzing the oxidation of linoleic and arachidonic acid, and LO is expressed in rodent and human islets. Little is known about the differential effect of the various hydroxyeicosatetraenoic acids (HETEs) that result from LO activity in human islets.

Objective: We compared the effects of 12-LO products on human islet viability and function.

Design: Human islets were treated with stable compounds derived from LOs: 12(S)-HETE, 15HETE, 12HPETE, and 12RHETE and then examined for insulin secretion and islet viability. The p38-MAPK (p38) and JNK stress-activated pathways were investigated as mechanisms of 12-LO-mediated islet inhibition in rodent and human islets.

Results: Insulin secretion was consistently reduced by 12(S)-HETE and 12HPETE. 12(S)-HETE at 1 nm reduced viability activity by 32% measured by MTT assay and increased cell death by 50% at 100 nm in human islets. These effects were partially reversed with lisofylline, a small-molecule antiinflammatory compound that protects mitochondrial function. 12(S)-HETE increased phosphorylated p38-MAPK (pp38) protein activity in human islets. Injecting 12-LO siRNA into C57BL/6 mice reduced 12-LO and pp38-MAPK protein levels in mouse islets. The addition of proinflammatory cytokines increased pp38 levels in normal mouse islets but not in siRNA-treated islets.

Conclusions: These data suggest that 12(S)-HETE reduces insulin secretion and increases cell death in human islets. The 12-LO pathway is present in human islets, and expression is up-regulated by inflammatory cytokines. Reduction of 12-LO activity could thus provide a new therapeutic approach to protect human beta-cells from inflammatory injury.

Figure 1

HETE effects on human islet insulin secretion. Insulin secretion after treatment with several different LO products (HETEs) for 4 h in 11 mm glucose at either 1 nm (A) or 100 nm (B) concentrations. The 12-LO product 12(S)HPETE (12sHP) and its more stable derivative 12(S)-HETE (12sH) produced the most substantial reduction in insulin secretion. *, P < 0.05; **, P < 0.01.

Figure 2

Effects of 12(S)-HETE (12sH) on metabolic activity and insulin secretion are partially reversed by the small molecular antiinflammatory compound LSF. Human islets from four different donors were pretreated with 50 μm LSF overnight or untreated as indicated. Islets were then treated with 1 nm 12(S)-HETE for 4 h and incubated in 11 mm glucose for 1 h to measure islet metabolic activity by MTT assay (A) or insulin secretion (B). Insulin secretion and metabolic activity were significantly inhibited by 12sH, but these effects were partially reversed by LSF pretreatment. *, P < 0.05.

Figure 3

12-LO products induce cell death in human islets. A, Human islets (top) cultured overnight in untreated conditions show relatively little cell death measured by PI fluorescence intensity (bottom). B, Overnight treatment with 12(S)-HETE causes a significant increase in cell death at 100 nm. C, Dose-dependent cell death response to 12(S)-HETE. *, P < 0.01. 12sH, 12(S)-HETE.

Figure 4

Inflammatory cytokines induced 12-LO protein and gene expression in human islets. Human islets were treated with cytokines (10 ng/ml for TNF-α, 100 ng/ml for IFN-γ, and 5 ng/ml for IL-1β in PBS) and vehicle for 30 min and collected for Western blot (A) and for different time points and collected for real time RT-PCR. A, 12-LO protein levels after cytokine treatment compared with control, and its quantification was normalized to the actin control. The results are presented as the means ± sd. *, P < 0.05 compared with the corresponding control. B, 12-LO mRNA expression was determined by using Alox12 TaqMan probe by quantitative PCR at indicated time points. The data were normalized to total actin, and fold differences were calculated using the 2^−ΔΔCT method. 12sH, 12(S)-HETE.

Figure 5

Effects of 12(S)-HETE and 12-LO blockade on pp38-MAPK protein activity. Human islets were treated with cytokines (10 ng/ml for TNF-α, 100 ng/ml for IFN-γ, and 5 ng/ml for IL-1β in PBS), cytokines + CDC (1 μm), 12(S)-HETE (100 nm), or vehicle for 30 min and collected for Western blot. A, Representative Western blots showing pp38-MAPK, total p38-MAPK protein, ppJNK 1 and -2, and actin for each condition. B, Protein levels quantified for each protein in panel A and normalized to the actin control. The results are presented as the means ± sd. *, P < 0.05 compared with the corresponding control; #, P < 0.05 compared with the cytokine treatment group. 12sH, 12(S)-HETE.

Figure 6

siRNA knockdown of 12-LO reduces pp38-MAPK protein levels in mouse islets. Islets were isolated from mice that were ip injected on 3 consecutive days with a stabilized siRNA construct against leukocyte 12-LO. The siRNA against 12-LO (si-Alox15) considerably reduced 12-LO levels compared with untreated controls (vehicle) or mice injected with missense siRNA (si-Control). Similarly, pp38-MAPK was greatly reduced in the siRNA knockdown group, whereas the total p38-MAPK (p38) expression does not change. Phosphorylated JNK expression (ppJNK1 and -2) was similar in all groups.