Rin-5F cells, a clone derived from the RIN-m rat pancreatic islet cell line, which produces and secrete insulin, were cultured in RPMI-1640 with 10% fetal bovine serum in a 37°C humidified incubator in the presence of 5% CO₂—95% air. Once confluent (about 80%), cells were treated with bacterial endotoxin, LPS (1 μg/ml) for 24 h, and appropriate controls were used as described before (Nahdi et al., 2017; Al-Nahdi et al., 2018; Alnahdi et al., 2020). In addition, some of the cells were incubated with azadirachtin (AZD) (25 µM for 24 h), either alone or with LPS. Concentrations and time points for LPS and AZD used in the present study were based on literature search as well as our previous studies (Raza et al., 2014, 2016; John and Raza, 2021). After treatment, cells were harvested and homogenized in cold H-medium buffer (70 mM sucrose, 220 mM mannitol, 2.5 mM HEPES, 2 mM EDTA, and 0.1 mM phenylmethylsulfonyl fluoride, pH7.4). Sub-cellular fractionation was performed to isolate the nuclear, mitochondrial, and post-mitochondrial fractions as described previously (Nahdi et al., 2017; Al-Nahdi et al., 2018; Alnahdi et al., 2020), and protein concentration of the individual fractions was checked by the Bradford method (Bradford, 1976).

In our previous study, we observed LPS-induced caspase-mediated apoptosis in Rin-5F cells via the caspase-8 extrinsic pathway (John and Raza, 2021). To further elucidate the involvement of the intrinsic pathway, Rin-5F cells were treated with LPS and/or AZD and caspase-9 activity was measured as per the manufacturer’s protocol. Briefly, cell lysates from the control and treated cells were incubated with the caspase-specific substrate, LEHD conjugated to a chromophore, p-nitro aniline. The release of the chromophore was then measured spectrophotometrically at 405 nm as described in the manufacturer’s protocol (Abcam, Cambridge, England, United Kingdom).

Further confirmation of apoptosis was ascertained in the cells treated with LPS and/or AZD by evaluating the expression of apoptotic marker proteins, Bcl-2 and Bax, as well as the release of mitochondrial cytochrome c by SDS-PAGE and Western blot analysis (as described later).

Membrane lipid peroxidation was measured to assess the oxidative stress in LPS and/or AZD-treated Rin-5F cells using the LPO-586 kit and the concentration of malondialdehyde was calculated as described before (Alnahdi et al., 2019b).

Alterations in nitric oxide (NO) production in the control and treated cells were assessed based on NO synthase (NOS) enzyme activity, using the NO assay kit (R & D Systems, MN, United States) which measures the nitrite formation using Griess reagent.

Antioxidant enzymes in the control and LPS and/or AZD-treated cells were measured using appropriate kits according to the manufacturer’s protocols. The catalytic action of catalase was measured based on the oxidation of methanol by hydrogen peroxide, in the presence of a chromogen, Purpald. The formaldehyde formed was then measured colorimetrically at 540 nm.

The conversion of xanthine to uric acid and hydrogen peroxide by xanthine oxidase was used as the basis for the measurement of SOD. SOD inhibited the reduction of NBT (nitro blue tetrazolium) to NBT-diformazan by superoxide ions, which was linearly related to the xanthine oxidase activity. SOD activity was thus measured as the percentage inhibition of NBT-diformazan formation.

GSH-redox homeostasis and metabolism were measured in control and LPS and/or AZD-treated Rin-5F cells. The ratio of GSH/GSSG (reduced glutathione to oxidized glutathione) was measured using the GSH/GSSG-Glo kit as per the vendor’s instructions as described before (Alnahdi et al., 2019b). Glutathione peroxidase (GSH–Px), glutathione reductase (GSH-reductase), and glutathione S-transferase (GST) activities were measured using cumene hydroperoxide, GSSG/NADPH, and CDNB as respective substrates, using standard protocols (Alnahdi et al., 2019b).

Media from control and LPS and/or AZD-treated Rin-5F cells was used for the measurement of inflammatory markers, TNF-α, and IL-6 using standard ELISA kits from BD Pharmingen (BD Biosciences, San Jose, United States) as described in the manufacturer’s protocols.

Cell lysates from LPS and/or AZD-treated cells were added to phospho-NF-κB p65 antibody-coated wells and incubated with NF-κB p65 antibody to detect the phospho-NF-κB p65 protein using the PathScan^® Phospho-NF-κB p65 sandwich ELISA assay kit as per the manufacturer’s protocol. Absorbance was read at 450 nm.

The peroxidase activity of the Cox enzyme was used as a basis for the measurement of cyclooxygenase -2 (Cox-2) activity in the cell lysates from control and LPS and/or AZD-treated cells. The assay was carried out in the presence of isozyme-specific inhibitors to distinguish Cox-2 from Cox-1 activity. The appearance of TMPD (N, N, N′, N′-tetramethyl-p-phenylenediamine) was monitored at 590 nm and used as a measure of Cox-2 activity.

Mitochondrial membrane potential was measured using a fluorescent cationic dye, DePsipher ™ (R & D Systems, MN, United States)in control and LPS and/or AZD-treated Rin-5F cells, and quantitated by flow cytometry as described before (Alnahdi et al., 2019a). The dye which could not access the transmembrane space due to low membrane potential remained in its green fluorescent form and this was measured using a flow cytometer (Becton Dickinson FACSCanto II).

Freshly isolated mitochondria (5 µg protein) from control and LPS and/or AZD-treated Rin-5F cells were suspended in mitochondrial buffer (20 mM potassium phosphate buffer, pH 7.4 containing 0.2% lauryl maltoside). Mitochondrial respiratory enzymes, NADH-ubiquinone oxidoreductase (NADH-dehydrogenase, Complex I), succinate-ubiquinone oxidoreductase/ubiquinol-cytochrome c oxidoreductase (Complex II/III), and cytochrome c oxidase (Complex IV) were measured using their respective substrates, coenzyme Q_2, succinate/cytochrome c, and reduced cytochrome c, according to the method of Birch-Machin and Turnbull (Birch-Machin and Turnbull, 2001).

The ATP content in the LPS and/or AZD-treated cells was assayed using an ATP Bioluminescent cell assay kit according to the vendor’s protocol. The luminescence emitted was then measured using the TD-20/20 Luminometer (Turner Designs, Sunnyvale, CA, United States).

The activities of aconitase, and glutamate dehydrogenase (Krebs’ cycle enzymes), were measured using the aconitase assay kit (Oxis Int, Inc. Portland, OR, United States) and the glutamate dehydrogenase kit (Abcam, Cambridge, England, United Kingdom) respectively as per the vendor’s protocol as described before (Alnahdi et al., 2019a; Alnahdi et al., 2020).

Protein (5–10 µg) from the total, mitochondrial, nuclear, and post-mitochondrial extracts from control and treated Rin-5F cells were separated electrophoretically and transferred onto nitrocellulose membrane by Western blotting as described before (Alnahdi et al., 2019a; Alnahdi et al., 2020). Ponceau S staining was used to confirm equal loading of the transferred proteins and then probed with cytochrome c, Bax, Bcl-2, NF-κB p65, I-κB, Cox-2, cytochrome c oxidase (COX), and aconitase antibodies. Immunoreactive proteins were detected by enhanced chemiluminescence using the Sapphire Biomolecular Imager (Azure biosystems, Dublin United States) or using X-ray films. Loading controls used for total/post-mitochondrial, mitochondrial and nuclear fractions were beta-actin, VDAC, and histone H3, respectively. Densitometric analysis was done using Image Studio Lite Ver.5.2 (LI-COR Biosciences, Lincoln, Nebraska, United States) software, and histograms were plotted based on the relative ratios of the proteins normalized to their respective loading control.

The histograms plotted are expressed as mean ± SD of three independent experiments. The statistical difference between the groups was assessed by two-way ANOVA followed by LSD (least significant difference) posthoc analysis for multiple comparisons using SPSS software (version 23). p-values < 0.05 were considered statistically significant.

Our recent study on Rin-5F cells (John and Raza, 2021) showed LPS-induced caspase-3/8-dependent apoptosis followed by increased DNA fragmentation in LPS treated cells and AZD caused attenuation of the apoptotic damages by enhancing the autophagic repair process. Our present study has further demonstrated that intrinsic caspase-9, which is the initiator caspase of the apoptotic pathway in the mitochondria, was also activated by LPS treatment and AZD protected the cells against LPS–induced apoptosis (Figure 1). This was further confirmed by analyzing the expression of Bcl-2 and Bax, oxidative stress marker proteins. A decrease in Bcl-2 expression was observed after LPS treatment which increased after AZD treatment (Figure 2A). Contrary to this, Bax expression was elevated in LPS-treated cells and AZD treatment brought the Bax protein level close to normal values (Figure 2B). Our results also indicated an increase in the release of mitochondrial intramembranous cytochrome c in the post-mitochondrial supernatant (PMS) (Figure 2C), indicating increased mitochondrial membrane permeability. These results are indicative of mitochondrial intrinsic apoptosis after LPS treatment and AZD appeared to be protective against LPS-induced apoptosis.

In our previous study on Rin-5F cells (John and Raza, 2021), we also showed that LPS treatment caused a marked increase in ROS production, which leads to DNA damage and apoptosis. Our present study has further confirmed the consequences of high ROS production by measuring LPO and oxidative stress-sensitive enzymes, SOD, and catalase. Our results show that LPO was significantly increased (about 30%) after treatment with LPS and AZD brought the values close to the control level (Figure 3A). Cells treated with AZD only had minimal effect on LPO. Similarly, catalase activity was also elevated (almost 30%) in LPS-treated cells and the activity decreased significantly after AZD treatment, though still higher than control values (Figure 3B). Contrary to this, LPS treatment markedly decreased (almost 50%) the SOD activity, which increased significantly after AZD treatment, though slightly lower than control values (Figure 3C). Nitric oxide (NO), another marker for increased oxidative/nitrosative stress as well as inflammation was also found to be increased (about 30%) in LPS-treated cells and AZD protected the cells by decreasing the NO synthesis (Figure 3D).

Alterations in GSH-dependent redox homeostasis were also observed after LPS treatment. As shown in Figure 4A, the GSH/GSSG (reduced glutathione/oxidized glutathione) ratio was markedly reduced (almost 40%) after LPS treatment while AZD treatment showed slight recovery. AZD alone showed no change in the redox homeostasis. This was further evaluated by measuring the GSH-dependent metabolizing enzymes. Figure 4B shows almost a two-fold increase in GST activity after LPS treatment. The levels remained high even after AZD treatment. This could be due to the detoxifying activity of GST, since AZD alone also significantly increased GST activity, though not as much as LPS alone. We have also shown that LPS treatment inhibited the recycling of reduced GSH from oxidized GSSG by inhibiting the GSH-reductase enzyme activity (Figure 4C) and stimulating the GSH-Px activity (Figure 4D). Alterations in these GSH metabolizing enzyme activities, as well as redox homeostasis, appear to be protected by AZD treatment.

As seen earlier in Figure 3D, LPS treatment had increased NO production, which is a marker for increased oxidative/nitrosative stress as well as inflammation. To confirm this, we checked the levels of inflammatory markers, Cox-2, TNF-α, and IL-6. We observed a marked increase in Cox-2 activity after LPS treatment which was slightly inhibited by AZD treatment (Figure 5A). Increased levels of TNF-α (Figure 5B) and IL-6 (Figure 5C) were also observed in LPS-treated cells and AZD treatment significantly reduced the levels of these inflammatory cytokines. In support, we have also shown that LPS treatment increased NF-κB activation (as phosphorylated NF-κB) and this was protected by AZD treatment (Figure 5D). These results demonstrate that cells treated with LPS undergo immense inflammatory stress and AZD treatment partially protected the cells against LPS-induced inflammatory responses.

Significant disturbance in mitochondrial membrane potential was observed in LPS-treated Rin-5F cells, while AZD caused partial protection (Figure 6). LPS treatment also caused slight inhibition of mitochondrial complex I (around 20%) and a significant inhibition (around 50%) of complexes II/III and IV activities which were partially protected after AZD treatment (Figures 7A–C). In support, we also observed a significant reduction (almost 40%) in ATP production after LPS treatment (Figure 7D). However, only a moderate increase in ATP production was observed in cells treated with both LPS and AZD. A minimal decrease in ATP production was observed after AZD treatment alone.

A significant inhibition (almost 60%) in the activity of ROS-sensitive aconitase, a mitochondrial matrix enzyme, was observed after LPS treatment in Rin-5F cells, which almost normalized after treatment with AZD (Figure 8A). AZD alone had minimal effect on the enzyme activity. Similarly, the activity of another mitochondrial matrix enzyme, glutamate dehydrogenase, which plays an important role in glutamate metabolism and energy homeostasis, was also markedly inhibited (about 40%) in LPS-treated cells, and AZD treatment brought the values close to control values (Figure 8B). Again, minimal effect was seen on AZD-treated cells.

In support of our observation of the activities of mitochondrial enzymes, electrophoresis and Western blot analysis demonstrated a decrease in the expression of mitochondrial enzymes, cytochrome c oxidase (COX) (Figure 9A), and aconitase (Figure 9B) after treatment of Rin-5F cells with LPS, which were partially protected by AZD.

Treatment with LPS induced translocation of NF-κB p65 into the nucleus (Figure 10A), which decreased after AZD treatment. Treatment with AZD significantly reduced the expression of the inflammatory marker. We also observed an increase in the cytosolic expression of inhibitory I-κB protein (Figure 10B), in LPS-treated cells confirming the release of the inhibitory protein from the NF-κB/I-κB complex and translocation of NF-κB to the nucleus. Increased expression of the inflammatory marker, cyclooxygenase-2 (Cox-2) (Figure 10C) was also observed after LPS treatment, which partially reduced in the presence of AZD.