NADPH oxidase-dependent production of reactive oxygen species induces endoplasmatic reticulum stress in neutrophil-like HL60 cells

Abstract

Reactive oxygen species (ROS) primarily produced via NADPH oxidase play an important role for killing microorganisms in neutrophils. In this study we examined if ROS production in Human promyelocytic leukemia cells (HL60) differentiated into neutrophil-like cells (dHL60) induces ER stress and activates the unfolded protein response (UPR). To cause ROS production cells were treated with PMA or by chronic hyperglycemia. Chronic hyperglycemia failed to induce ROS production and did not cause activation of the UPR in dHL60 cells. PMA, a pharmacologic NADPH oxidase activator, induced ER stress in dHL60 cells as monitored by IRE-1 and PERK pathway activation, and this was independent of calcium signaling. The NADPH oxidase inhibitor, DPI, abolished both ROS production and UPR activation. These results show that ROS produced by NADPH oxidase induces ER stress and suggests a close association between the redox state of the cell and the activation of the UPR in neutrophil-like HL60 cells.

Fig 1. Effect of hyperglycemia and PMA on ROS production in neutrophil-like HL60 cells (dHL60) and non-differentiated HL60 cells.

PMA (1μM) triggered the production of ROS only in dHL60 (A; C). Hyperglycemia did not affect the ROS production in neutrophil-like HL60 cells (B). DPI (10 μm) was used as NADPH oxidase inhibitor. [NG] normoglycemic media (5.5 mM of glucose); [MN] Mannitol enriched media (5.5 mM of glucose + 19.5 mM of mannitol); [HG] Hyperglycemic media (25 mM of glucose). DHR (10 μM) was used to monitor ROS production by flow cytometry. Graphs show median of fluorescence ± S.E.M. Results are from 6 independent experiments. (***) Indicate p <0.001.

Fig 2. Effect of PMA, fMLP and hyperglycemia on calcium influx and ER calcium content in neutrophil-like HL60 cells (dHL60).

PMA (1 μM) (A) did not alter calcium dynamics in dHL60 cells. fMLP (1 μM) (B-D) promoted calcium influx in neutrophil-like HL60 cells; and hyperglycemia (C) did not disturb the calcium intake by fMLP stimulation. ER calcium content in dHL60 was not altered by hyperglycemia (E; F). Calcium dynamics were monitored by Indo-1-AM (1 μM) using fluorometry. [NG] normoglycemic media (5.5 mM of glucose); [MN] Mannitol enriched media (5.5 mM of glucose + 19.5 mM of mannitol); [HG] Hyperglycemic media (25 mM of glucose). Graphs show fluorescence intensity during the time of analysis (A; B; C; E). Histograms show the mean intracellular calcium concentration ± S.E.M. (D) and the mean ± S.E.M. of ER calcium content (F). Results are from 4 independent experiments. (**) Indicate p <0.01 and (*) indicate p<0,05.

Fig 3. Effect of hyperglycemia and PMA on eIF2α phosphorylation and splicing of XBP1 mRNA in neutrophil-like HL60 cells (dHL60).

Hyperglycemia failed to modulate eIF2α phosphorylation (A; B) and splicing of XBP1 (C; D). Splicing of XBP1 mRNA was caused by PMA (1 μM) stimulus (1 and 4h) (C; D). [NG] normoglycemic media (5.5 mM of glucose); [MN] Mannitol enriched media (5.5 mM of glucose + 19.5 mM of mannitol); [HG] Hyperglycemic media (25 mM of glucose). Histogram (B) shows the mean ± S.E.M. optical density (OD) of the protein bands (A). γ Tubulin was used as loading control. (C) Complementary cDNA bands of unspliced XBP1 (uXBP1) (top band) and spliced (sXBP1) (bottom band). Results is representative of 3 independent experiments. [Tg] Positive control (1 μM Thapsigargin, 1h); [N1h] NG + 1h PMA; [M1h] MN + 1h PMA; [H1h] HG + 1h PMA; [N4h] NG + 4h PMA; [M4h] MN + 4h PMA; [H4h] HG + 4h PMA.

Fig 4. Effect of ROS production on UPR regulated proteins in neutrophil-like HL60 cells (dHL60).

PMA (1 μM) treatment for 1 and 4h increased phosphorylated eIF2α (A; C) and GRP78 protein levels (B; D). This was blocked by DPI (10 μM), a NADPH oxidase inhibitor (A; B; C; D). GADD34 and ATF4 protein content was not altered by PMA (E; F). Histograms (C; D; E; F) show the mean ± S.E.M. of the optical density (OD) of the protein bands. γ Tubulin was used as loading control. Results are from at least 3 independent experiments. (***) Indicates p <0.001 and (**) Indicates p <0.01.

Fig 5. Effect of ROS production on the IRE1α/XBP1 pathway in neutrophil-like HL60 cells (dHL60).

PMA (1 μM) treatment for 1h and 4h caused an increase in spliced XBP1 (sXBP1) mRNA content as monitored by RT-PCR (top panel). DPI (10 μM) inhibited the splicing of XBP1 mRNA. GAPDH was used as a loading control (bottom panel). Results are from 3 independent experiments. Thapsigargin (1μM, 1h) was used as a positive control for ER stress-induced XBP1 splicing.

Fig 6. Effect of ROS production on the expression of genes involved in the unfolded protein response (UPR).

GAPDH and 18S were used as reference control genes. Results are from 6 independent experiments. Thapsigargin (1μM) 1h and Tunicamycin (2 μg/ml) 16h were used as positive controls. (***) Indicates p <0.001; (**) Indicates p <0.01 and (*) Indicates p <0.05; (###) indicates p<0.001 vs control group; (##) indicates p<0.01 vs control group; (#) indicates p<0.05 vs control group.