M. fortuitum (Strain 993) was purchased from Microbial Type Culture Collection and Gene Bank (MTCC), Chandigarh, India. The strain is sensitive to amikacin and resistant to ampicillin as suggested by the antibiogram. For infection studies, bacteria were grown to mid-log phase (120 h) in Middlebrook 7H9 broth (HiMedia), supplemented with 0.05% Tween-80, 0.50% glycerol, and 100 μg/ml ampicillin in a shaking incubator (120 rpm) at 30°C. Before infection, M. fortuitum clumping was removed by repeatedly passaging through a 26-G needle. Stocks were maintained at −80°C in 10% glycerol and in Lowenstein Jensen media (HiMedia) at 4°C for further use.

Catfish, Clarias gariepinus (Siluriformes: Clariidae, 100–150 g), were procured from local fish farms and maintained in 50-L glass tanks (2–3 fish per tank) under natural photoperiod. Prior to initiating the study, fish were acclimatized to laboratory conditions for 15 days, and fish health was monitored at regular intervals by morphological and pathological examinations (38).

Head kidneys were aseptically removed and placed in Roswell Park Memorial Institute (RPMI)-1640 (Gibco-Invitrogen) with phenol-red indicator supplemented with 25 mM HEPES (Gibco-Invitrogen) containing 1% penicillin–streptomycin. Single-cell suspensions of each pair of head kidney were prepared using 100-mm wire mesh. The cell suspension was centrifuged at 400×g for 10 min at 4°C, the supernatant was discarded and the pellet resuspended and then layered on a discontinuous Percoll density gradient (34/51) and centrifuged at 400×g for 20 min at 4°C. The phagocyte-rich fraction appearing above the 34/51 interface was collected, washed, and incubated overnight at 30°C under 5% CO₂ for adherence to sterile Petri dishes (Nunc). The non-adherent cells were removed carefully, and the adherent macrophages were obtained by incubation with 1% cell dissociation medium (59418C, Sigma) at 30°C for 20 min. The purity of the HKM was checked by staining with Wright Giemsa Stain (>90% pure), and viability was determined using 0.4% trypan blue dye exclusion method (>95% viable) (39, 40).

For infection, the HKM were washed in antibiotic-free RPMI supplemented with 10% fetal bovine serum (FBS) (Gibco-Invitrogen) and infected with M. fortuitum at a multiplicity of infection (MOI) of 1:10 (HKM/bacteria). The number of HKM used for different experiments is mentioned in the corresponding section. A short spin of 5 min was given to facilitate bacteria–HKM interactions, the cells were distributed in six-well tissue culture plates and incubated for 4 h at 30°C. Subsequently, amikacin (50 μg/ml, HiMedia) was added and the cells further incubated for 1 h to kill the extracellular bacteria. The concentration of amikacin effectively killed extracellular bacteria without affecting HKM viability (data not shown). Finally, the infected HKM were washed and resuspended in RPMI supplemented with 10% FBS containing amikacin (5 μg/ml) and incubated at 30°C for further studies (36, 37).

TLR-2 inhibitor (CUCPT-22, 1 μM), ER-stress inhibitor (4-phenyl butyric acid, 4-PBA, 10 µM), mtROS inhibitor (YCG063, 10 µM), mtROS inducer (antimycin A, Ant A, 50 µM), TNF-α biosynthesis inhibitor pentoxifylline (Pentox, 1 mM), and mt-Ca²⁺ uniporter blocker (Ru360, 10 µM) were purchased from Sigma. Caspase-8 inhibitor (Z-IETD-FMK, 10 µM) was purchased from Biovision. (Ca²⁺) _c monitoring dye (Fluo-3/AM, 2 µM) and mt-Ca²⁺ monitoring dye (Rhod-2/AM, 5 µM) were purchased from Invitrogen. Mitochondrial superoxide indicator (Mitosox, 5 μM) was purchased from Molecular Probes. HKM were pretreated with specific inhibitors for 1 h prior to infection with M. fortuitum. The doses of different inhibitors were selected on the basis of inhibitor specificity and cytotoxicity. The HKM treated with the indicated concentrations of the inhibitors remained as viable as control HKM at all-time points as determined by the trypan blue (0.4%) dye exclusion method and were maintained during the entire course of the experiment (16, 36, 37).

HKM (2 × 10⁷/ml) pretreated with or without inhibitors or transfected with scrambled (sc-) or specific small-interfering RNAs (siRNAs) were harvested at indicated time points p.i., total RNA isolated using TRIZOL (Sigma), and dissolved in diethyl pyrocarbonate (DEPC) water. One microgram RNA was used as a template using first-strand complementary DNA (cDNA) synthesis kit (MBI fermentas). Primers for CHOP, STIM1, Orai1, TNF-, and β-actin genes were already available in our laboratory ( Table 1 ). Fold change in CHOP, STIM1, Orai1, and TNF mRNA expression levels were studied using SYBR green PCR Master Mix (Applied Biosystems) by RT-qPCR (ABI ViiA, Applied Biosystems). cDNA (1:100 dilution), forward and reverse primers (0.20 µM each), and 5 µl SYBR green PCR master mix (Applied Biosystems) were used (total volume, 10 μl) for each assay. Expression levels of different genes were analyzed by the comparative ΔΔCT method wherein β-actin was taken as the endogenous control and uninfected HKM (0 h) was used as the calibrator (16).

siRNA ( Table 2 ) transfection was done using HiPerFect Transfection Reagent (Qiagen). Briefly, 5 μl each of siRNA and HiPerFect complex were mixed gently, added to 90 μl Opti-Mem (Invitrogen), and incubated for 20 min at 30°C to allow complex formation. The complex was then added to the HKM cultures maintained in Opti-MEM, mixed properly (total volume, 1 ml), and incubated at 30°C with 5% CO₂ for 16 h during which HKM viability was continuously monitored. Thereafter, HKM were infected with M. fortuitum and processed for subsequent studies. Targeted gene knockdown was confirmed by real-time quantitative PCR (RT-qPCR) using target-specific siRNAs. Scrambled or sc-siRNA (Sigma, 5 nM) was used as universal negative control in this study.

HKM (2 × 10⁶/ml) pretreated with or without inhibitors or transfected with sc- or targeted siRNAs were incubated with cell-permeable mt-Ca²⁺ dye Rhod-2/AM for 20 min at 30°C under dark conditions. Excess dye was removed by washing with phosphate-buffered saline (PBS) (1) and HKM infected with M. fortuitum. The changes in fluorescence intensity were measured at indicated time points p.i. in a fluorimeter (Spectramax, Molecular Devices) at excitation–emission of A₅₅₂ and A₅₈₁, respectively, and changes in mt-Ca²⁺ levels were plotted as the relative increase in fluorescence values.

HKM (2 × 10⁶/ml) pretreated with or without inhibitors or transfected with sc- or targeted siRNAs were infected with M. fortuitum. HKM were collected at indicated time point p.i. and incubated with MitoSOX for 20 min at room temperature under dark conditions. Excess dye was removed by washing with PBS (1), and the changes in fluorescence intensity were measured at excitation–emission of A₅₁₀ and A₅₈₀, respectively (Spectramax, Molecular Devices). The changes in mtROS levels were plotted as the relative increase in fluorescence values.

HKM (2 × 10⁶/ml) pretreated with or without inhibitors or transfected with sc- or targeted siRNAs were infected with M. fortuitum. Cell-free culture supernatant was collected at indicated time points p.i. and TNF-α levels measured with fish-specific TNF-α ELISA kit (MyBioSource, MBS704369). Briefly, 100 μl of culture supernatant was loaded to the antibody precoated wells and incubated at 37°C for 90 min. The supernatant was removed, and 100 μl biotinylated detection antibody was added followed by incubation at 37°C for 1 h. Wells were washed and 100 μl horseradish peroxidase (HRP) conjugate was added, and the plates were incubated at 37°C for 30 min. Following incubation, the wells were washed, and 90 μl substrate was added and incubated at 37°C for 15 min. Fifty microliters of stop solution was added to terminate the reaction and absorbance read at A₄₅₀ (Spectramax, Molecular Devices). The concentration of TNF-α in each sample was interpolated from the standard curve (11).

HKM (2 × 10⁶/ml) pretreated with specific inhibitors or transfected with sc-siRNA or targeted siRNAs, respectively, were infected with M. fortuitum. The cultures were terminated at indicated time point p.i., lysed with 0.1% Triton X-100, serially diluted, plated on 7 H11 Middlebrook agar plates supplemented with 0.05% Tween-80, 0.50% glycerol, and 100 μg/ml ampicillin. Intracellular bacteria (CFU) were enumerated following incubation at 30°C (16).

HKM (2 × 10⁶/ml) pretreated with or without specific inhibitors or transfected with sc-siRNA or targeted siRNAs were infected with M. fortuitum. Caspase-8 (Elabscience) and caspase-3 activity (Biovision) were studied using specific assay kits. Briefly, HKM collected at indicated time points p.i. were lysed, and to the cell lysate (50 μl), equal volume of 2 reaction buffer and caspase-8/caspase-3-specific substrate (5 μl) was added to separate wells and mixed gently, avoiding bubble formation. The plates were incubated at 37°C for 2 h, absorbance was measured at A₄₀₅ (Spectramax, Molecular Devices), and relative fold changes in caspase-8 and caspase-3 activity were plotted.

Mean ± SE were calculated using pairwise comparison by employing t-test: two samples using unequal variances to determine the statistical significance between the groups. A value of p < 0.05 was considered as statistically significant. Individual assays were done in triplicates, and the vertical bars represent mean ± SE of three independent observations (n = 9).

The role of mt-Ca²⁺-induced mtROS as an innate immune factor has been reported previously (7, 41). In absence of prior knowledge, our aim was to study this in M. fortuitum pathogenesis. HKM were infected with M. fortuitum, and mt-Ca²⁺ levels were monitored at indicated time points p.i. using mt-Ca²⁺-specific dye Rhod-2/AM. We observed a significant increase in mt-Ca²⁺ levels with maximum levels recorded at 2 h p.i. ( Supplementary Figure 1 ), and selected this time interval for subsequent studies. The mitochondrial uniporter MICU mediates the influx of (Ca²⁺) _c inside mitochondria, triggering mtROS production (42). To look into this, HKM were pretreated with MICU inhibitor Ru360 and then infected with M. fortuitum, and mt-Ca²⁺ levels were monitored at 2 h p.i. It was observed that mt-Ca²⁺ levels were significantly downregulated in the presence of Ru360 ( Figure 1 ), suggesting the involvement of MICU in the dynamics of mt-Ca²⁺ in M. fortuitum infection. Furthermore, we pre-treated the HKM with TLR-2 and ER-stress inhibitors (CUCPT-22 and 4-PBA) followed by infection with M. fortuitum and measured the mt-Ca²⁺ levels ( Figure 1 ). Our results affirmed the involvement of TLR-2-ER-stress-STIM1-Orai1 axis in triggering mt-Ca²⁺ elevation p.i. Additionally, we also measured mt-Ca²⁺ levels in the absence of STIM1 and Orai1 signaling. For that, the HKM were transfected with STIM1 and Orai1 siRNA, respectively, and then infected with M. fortuitum and the changes in mt-Ca²⁺ levels monitored. The significant reduction in mt-Ca²⁺ levels led us to conclude the involvement of SOCE in triggering mt-Ca²⁺ influx in M. fortuitum-infected HKM ( Figure 1 ).

Our next step was establishing the link between mt-Ca²⁺ dynamics and mtROS production. To study this, we used the specific dye MitoSOX (43). Mycobacterium fortuitum-infected HKM were stained with MitoSOX, and the changes in mtROS levels were monitored at indicated time points p.i. We observed maximum mtROS levels at 4 h p.i. ( Supplementary Figure 2 ) and selected this time interval for subsequent studies. In our next step, we pretreated the HKM with Ru360 and measured mtROS in infected HKM. We observed a significant reduction in mtROS levels in the presence of Ru360 ( Figure 2A ). Based on our kinetics and inhibitor studies, it is evident that the influx of mt-Ca²⁺ through MICU leads to downstream mtROS generation in M. fortuitum-infected HKM. Additionally, we also checked whether inhibition of mtROS is having any impact on the mt-Ca²⁺ levels. To this, we pretreated the HKM with mtROS inhibitor (YCG063) and measured the mt-Ca²⁺ levels. Our results suggested the existence of a positive feedback loop between mt-Ca²⁺ and mtROS elevation ( Figure 1 ).

mtROS is reported to trigger apoptosis of host immune cells (44). However, the same is not well reported in fish. Our previous studies demonstrated that M. fortuitum induces caspase-3-dependent HKM apoptosis (36). Here, we hypothesized that mtROS generated in response to M. fortuitum infection induces HKM apoptosis. To begin with, HKM were pretreated with the mtROS inhibitor YCG063 (45) and then infected with M. fortuitum and mtROS levels monitored at 4 h p.i. We noted that pretreatment with YCG063 significantly downregulated mtROS levels in infected HKM ( Figure 2A ). Next, HKM pretreated with YCG063 were infected with M. fortuitum and apoptosis studied by caspase-3 assay at 24 h p.i. Pretreatment with YCG063 attenuated caspase-3 activity ( Figure 2B ). Antimycin A (Ant A), an inhibitor of complex III of the ETC, was used as positive control, which very predictably led to the production of measurable quantities of mtROS in HKM ( Figure 2A ) and triggered caspase-3 activity ( Figure 2B ). Thus, our results confirmed that mtROS plays a proapoptotic role in M. fortuitum pathogenesis.

We concluded investigating the role of mtROS in regulating M. fortuitum growth. HKM pretreated with YCG063 were infected with M. fortuitum, and the growth of intracellular bacteria was studied at 24 h p.i. We observed that inhibiting mtROS production by YCG063 led to a significant increase in the number of intracellular M. fortuitum ( Figure 2C ). Furthermore, pretreatment with mtROS-inducer Ant A also resulted in a significant reduction in the number of intracellular M. fortuitum ( Figure 2C ). Collectively, our results implicate that mtROS induces HKM apoptosis and helps in the clearance of M. fortuitum.

We wanted to explore the upstream signaling events triggering mtROS production in infected HKM. Based on our own study (16) and previous reports (7), we hypothesized the primal role of the TLR-2-ER-stress axis in the process. To test this, HKM were treated separately with the TLR-2-specific inhibitor, CUCCPT-22 (46), or transfected with TLR-2-siRNA, then infected with M. fortuitum, and the changes in mtROS production (4 h p.i.) were monitored. We observed that inhibiting the TLR-2 signaling resulted in the downregulation of mtROS production in infected HKM. We followed this by pretreating the HKM with the ER-stress ameliorator, 4-PBA (47), or transfecting the HKM with CHOP-siRNA and monitoring the changes in mtROS levels. It was observed that inhibiting ER stress resulted in significant downregulation in the production of mtROS in M. fortuitum-infected HKM ( Figure 3A ). YCG063 was used as the control.

We had previously reported the proapoptotic role of SOCE in M. fortuitum pathogenesis (16). Here, we questioned the involvement of SOCE in mtROS generation in M. fortuitum-infected HKM. Towards that direction, we silenced STIM1 and Orai1 expression with specific siRNAs and measured mtROS levels in M. fortuitum-infected HKM. We observed that the silencing of STIM1 and Orai1 interfered with mtROS production in the infected HKM ( Figure 3A ). Based on these observations, we suggest that signaling via TLR-2–ER stress–SOCE axis induces mtROS generation in M. fortuitum-infected HKM.

We extended the study by pretreating HKM with the mtROS inhibitor, YCG063, and monitored the expression of CHOP, STIM1, and Orai1, respectively at 2 h p.i (16, 37). It was observed that inhibiting mtROS production downregulated CHOP, STIM1, and Orai1 mRNA expression in M. fortuitum-infected HKM ( Figure 3B ). However, pretreatment with mtROS inducer Ant A resulted in a significant increase in CHOP, STIM1, and Orai1 mRNA expression levels in M. fortuitum-infected HKM. To this, we concluded that the cross-talk between TLR-2–ER stress–SOCE axis and mtROS production potentiates M. fortuitum-induced HKM pathology.

mtROS is reported to induce the production of proinflammatory TNF-α with apoptotic implications (48). However, evidence of any link between ER stress–SOCE–mtDNA axis-dependent mtROS production and TNF-α synthesis remains unexplored to date. To begin with, HKM were infected with M. fortuitum and TNF-α mRNA expression studied at indicated time points p.i. We observed maximum TNF-α mRNA expression at 6 h p.i. ( Supplementary Figure 3A ). Complementing this, the changes in TNF-α protein levels were also studied using a specific assay kit. Maximum TNF-α production was observed at 24 h p.i. and selected for subsequent studies ( Supplementary Figure 3B ).

Next, HKM pretreated with YCG063 were infected with M. fortuitum, and changes in TNF-α mRNA expression and protein levels were monitored at 6 and 24 h p.i. respectively ( Figures 4A, B ). We observed that inhibiting mtROS production resulted in a significant decline in TNF-α mRNA expression and protein concentration. In addition to this, treatment with Ant A led to significant upregulation in TNF-α mRNA expression, which led us to conclude that M. fortuitum infection in fish triggers mtROS-dependent TNF-α production.

We extended our study wherein HKM pretreated with CUCPT-22, 4-PBA, or transfected separately with TLR-2-, CHOP-, STIM1-, and Orai1-siRNA were infected with M. fortuitum and the changes in TNF-α levels monitored. It was observed that silencing of TLR-2 signaling, ameliorating ER stress, inhibiting STIM1/Orai1 expression, and mt-Ca²⁺ influx led to significant downregulation in TNF-α mRNA and protein expression ( Figures 4A, B ) in the infected HKM. Pentox was used as a control for the study. Collectively, these results implicated the critical role of mtROS in inducing TNF-α production in M. fortuitum-infected HKM and that TLR-2/ER stress/SOCE-mt-Ca²⁺ axis plays a primal role in the process.

TNF-α is well known to induce the activation of caspase-8 under various conditions of stress (49, 50). We had previously observed that M. fortuitum triggers caspase-8-mediated apoptosis of HKM (36). We questioned the role of TNF-α in activating caspase-8 in M. fortuitum-infected HKM. For that, HKM pretreated with TNF-α inhibitor pentox or transfected with TNF-α siRNA were infected with M. fortuitum, and the changes in caspase-8 levels were studied 24 h p. i (36). It was observed that inhibiting TNF-α led to a significant decrease in caspase-8 activity ( Figure 5A ), which clearly suggested that TNF triggers caspase-8 activation in M. fortuitum-infected HKM. Caspase-8 inhibitor Z-IETD-FMK was used as a control in the study.

Caspase-8 activation leads to both caspase-3-dependent and caspase-3-independent apoptosis (51). To study this, HKM pretreated with caspase-8 inhibitor Z-IETD-FMK were infected with M. fortuitum, and caspase-3 activity was monitored at 24 p.i (36). We observed that inhibiting caspase-8 activation resulted in significant downregulation in caspase-3 activity. Furthermore, pretreatment with pentox or transfection with TNF-α siRNA also markedly inhibited caspase-3 activation in M. fortuitum-infected HKM ( Figure 5B ). Based on these results, it is evident that the activation of the TNF-α/caspase-8 axis leads to caspase-3-mediated apoptosis of M. fortuitum -infected HKM.

Next, we investigated the role of TNF in deciding the fate of intracellular M. fortuitum. To achieve this, HKM pretreated with TNF inhibitor pentox and TNF siRNA were infected with M. fortuitum and intracellular bacteria (CFU) enumerated at 24 p.i. Inhibition of TNF production led to significant improvement in intracellular bacterial number ( Figure 5C ), implicating the bactericidal role of TNF in M. fortuitum pathogenesis. Taken together, our findings established that TNF-induced HKM apoptosis aids in the clearance of M. fortuitum.