BRG1 rabbit monoclonal antibody (Catalog No ab110641) was purchased from Abcam, UK. STAT1α rabbit polyclonal antibody (Catalog No SAB3500364-100UG) and β-actin mouse monoclonal antibody (Catalog No A1978-100UL) were purchased from Sigma-Aldrich, USA. Anti-mouse IgG, HRP-linked antibody (Catalog No 7076), and anti-rabbit IgG, HRP-linked antibody (Catalog No 7074S) was purchased from Cell Signaling Technology, USA. Alexa green 488-conjugated goat anti-rabbit IgG (Catalog No A-11070) was purchased from Thermo Fisher Scientific, USA.

L. donovani Bob (LdBob/strain/MHOM/SD/62/1SCL2D) (Goyard et al., 2003; Debrabant et al., 2004) acquired from Dr Stephen Beverly (Washington University, St. Louis, MO) and THP-1 cells (202 TIB; American Type Culture Collection, Rockville, MD) were cultured as described previously (Roy et al., 2020).

THP-1 cells (10⁶ cells/ml) were differentiated and infected as previously described (Roy et al., 2020). After infection, cells were washed with phosphate-saline buffer (PBS) and rested for 2 h, followed by 1 ng/ml IFNγ stimulation (catalogue no SRP3058-100UG, Sigma-Aldrich, USA) for 30 min. The cells were harvested at time points - 0, 3, 6, and 24 h. Infection was confirmed by Giemsa (Sigma-Aldrich, USA) and Propidium Iodide (PI) (Sigma-Aldrich, USA) staining.

Total RNA was isolated and used for quantitative real-time RT-PCR (qPCR) as described in (Roy et al., 2020). Expression of various genes was analyzed using their specific primers ( Table S1 ). RNU6A was used as a housekeeping gene. The fold change values of different genes at 3, 6 and 24 h were normalized to the respective values at 0 h. The results were calculated by the 2^-ΔΔ CT method (Singh et al., 2015).

Recruitment of BRG1 and STAT1α proteins and H3 acetylation at the promoters of concerned genes was analyzed by ChIP assay using chromatin from infected and/or IFNγ stimulated THP-1 cells (10⁶ cells/ml). Further, qPCR was performed using promoter-specific primers ( Table S1 ). The cells were harvested and processed for ChIP analysis as reported previously (Roy et al., 2020). BRG1, STAT1α and acetylated histone (Ac-H3) bound DNA was immunoprecipitated overnight at 4°C using BRG1 (1 µg/25 µg chromatin extract), STAT1α (2 µg/25 µg chromatin extract) and Ac-H3 (1 µg/25 µg chromatin extract) antibodies. To quantify the DNA isolated by ChIP, qPCR was performed using primers spanning -288 to -99 of CIITA promoter I and -158 to +21 of CIITA promoter IV ( Table S1 ). The change in gene expression for relative quantification was calculated by 2^^−ΔΔCT method (Singh et al., 2015). For calculating relative enrichment of each DNA fragment, fold change difference of the C_T values concerning the no antibody control and 0 h chromatin extract control was used.

To study STAT1α and BRG1 protein expression, infected and/or stimulated THP-1 cells were lysed in urea buffer (90% 8.8 M urea, 2% 5 M NaH₂PO₄, 8% 1 M Tris-Cl pH 8) at 4°C. 80 μg of total protein was separated on 8% SDS-PAGE by electrophoresis (Bio-Rad Laboratories, USA). Immunoblotting was performed as described before (Roy et al., 2020) using STAT1α (1:500), BRG1 (1:1000) and β-actin (1:2000) specific antibodies. The membrane was then washed with Tris-buffered saline (TBS) and incubated with horseradish-peroxidase (HRP)-conjugated anti-mouse (1:3000) or anti-rabbit (1:1000) IgG antibody. The complexes were visualized by ECL chemiluminescence. Protein expression was normalized with the corresponding β-actin and quantitated by densitometry using ImageJ software.

THP-1 cells were differentiated on coverslips followed by infection and stimulation as mentioned above. After 6 h, the cells were fixed and permeabilized with 0.5% Triton X-100 and then blocked in 2% BSA. The cells were probed with STAT1α antibody (1:50) followed by incubation with Alexa green 488-conjugated goat anti-rabbit IgG (1:200). DAPI (1 μg/ml) was used to stain the host nuclei and parasite kinetoplastid DNA. All antibody incubations were followed by washes with 0.2% Triton X-100. The images were then visualized under a confocal laser scanning microscope (Olympus FluoViewTM FV1000) at 488 nm wavelength.

THP-1 cells (10⁶ cells/ml) were differentiated, followed by parasite infection and IFNγ stimulation as mentioned above. Subsequently, the cells were transiently transfected with 1.5 μg of plasmid overexpressing BRG1 (Patne et al., 2017) for 48 h, using lipofectamine 3000 (Catalog No L3000015, Thermo Fisher Scientific, USA). The transfection reagents were mixed according to the manufacturer’s protocol. The mRNA expression levels of concerned genes were examined by qPCR. THP-1 cells transfected with vector (pcDNA3.1 LAP-Zeo) alone was used as a negative control.

The PMA treated differentiated THP-1 cells (10⁵ cells/ml) were transiently transfected with 600 pmole (Garcia-Garcia et al., 2009) of siGENOME Human HDAC1 (3065) siRNA – SMARTpool (Dharmacon, USA) using lipofectamine 3000. The cells were incubated with siRNA for 24 h to allow gene silencing. Subsequently, the cells were washed for infection and stimulated as described earlier. THP-1 cells were harvested after 6 h and mRNA expression of HDAC1 and CIITA genes was analyzed by qPCR. ON-TARGET plus Control Pool (Dharmacon, USA) was used as a negative control. The basal level of HDAC1 and CIITA in uninfected Sc-siRNA transfected cells respectively were used for data normalization and were taken as 1.0. The transfection efficiency was calculated (ThermoFisher) to be >50%, as has also been reported by the manufacturer (Dharmacon, USA).

THP-1 cells (10⁵ cells/ml) were infected and stimulated with IFNγ as mentioned earlier. After 6 h, for visualization of intracellular parasites, Giemsa staining was performed, and the parasite load was calculated (Manhas et al., 2014; Roy et al., 2020).

GraphPad Prism (version 5.0) software (GraphPad Software, Inc.) was used for plotting data. Statistical analysis was measured using ANOVA. P ≤ 0.05 was considered significant [* (P ≤ 0.01 to 0.05), ** (P≤ 0.001), *** (P≤ 0.0001), **** (P≤ 0.0001), ns (P≥ 0.05)]. Error bars used in the figures specify standard error of the mean (Cheeseman and Weitzman).

In earlier reports and in our earlier studies, THP-1 cells were incubated with parasites for 3 h (Roy et al., 2020), followed by 2 h of resting period (Forget et al., 2005) prior to IFNγ stimulation (Forget et al., 2005, Lang et al., 2012). Significant IFNγ response has been demonstrated, within 30 min to 6 h of IFNγ stimulation (Blanchette et al., 1999; Forget et al., 2005; Forget et al., 2006; Matte and Descoteaux, 2010, Lang et al., 2012, Singh et al., 2015; Roy et al., 2020). Based on these studies, we designed all our experiments to study the effect of Leishmania infection in the host cells at an initial stage of infection. The impact of IFNγ on parasite load within the infected macrophages was analyzed by visually counting the intracellular amastigotes after Giemsa staining ( Figure S3 ). A comparable parasitemia count (6h: 8/macrophage and 6/macrophage; 24 h: 13/macrophage and 11/macrophage) between non-stimulated and stimulated cells at both 6 and 24 h was observed showing that IFNγ stimulation has no detrimental effect on the intracellular parasite load.

Hence, we first investigated the expression of BRG1 in response to IFNγ stimulation at 3, 6 and 24 h, post Leishmania donovani infection, by qPCR. The data showed that there was no significant alteration in the expression of BRG1 between infected and uninfected THP-1 cells under unstimulated condition ( Figure 1A ). However, the mRNA levels of BRG1 were significantly upregulated at 3 h (~ 18 fold, P = 0.01), 6 h (~ 8.3 fold, P = 0.019) and 24 h (~ 7.3 fold, P = 0.014) in uninfected and stimulated cells as compared to the resting macrophages. This is similar to the upregulation of STAT1α observed in the previous studies (Ray et al., 2000; Forget et al., 2005) and in Figure S1A . On infection and stimulation with IFNγ, the expression of BRG1 decreased significantly at 3 h (~ 86.6%, P = 0.016), 6 h (~ 94.4%, P = 0.012) and 24 h (~ 91%, P = 0.01) in comparison to the uninfected, stimulated cells ( Figure 1A ).

Next, we evaluated the expression of BRG1 protein in response to IFNγ stimulation in THP-1 cells. Cells were harvested at 6 h since at this time point maximal increase in BRG1 expression was observed in stimulated uninfected cells as compared to the resting macrophages ( Figure 1A ). BRG1 protein levels were checked by western blot and analysed by densitometry. As with the expression of mRNA levels, there was no significant change in the protein expression between uninfected and infected cells under unstimulated condition ( Figure 1B ). On stimulation with IFNγ, BRG1 expression was ~ 2 fold (P = 0.004) higher in uninfected cells as compared to the resting macrophages ( Figure 1B ). BRG1 protein expression decreased significantly (~ 48.3%, P = 0.0195) on infection as compared to the uninfected, stimulated cells ( Figure 1B ).

It is well established that normal macrophage functioning with Leishmania infection is disrupted at early time points (Blanchette et al., 1999; Forget et al., 2006; Singh et al., 2015; Roy et al., 2020). To confirm that the results obtained in our study are because of live parasite infection, we have used uninfected cells (Matte and Descoteaux, 2010) and 0 h time points as controls. These control macrophages were incubated with or without parasites for 3 h, followed by washes and 2 h resting period and then 30 min IFNγ stimulation and then harvested to count as 0 h infected or uninfected. We have performed Giemsa and propidium iodide staining to confirm internalization of parasites inside the macrophages by 3 h of incubation (data not shown). We have also performed a similar infection experiment with heat-killed L. donovani (HKLD) instead of live parasites. qPCR data clearly demonstrated that HKLD infection did not have any effect on the regulation of BRG1 expression ( Figure S4A ), thus, suggesting that live parasite infection plays a potential role for the downregulation of the BRG1 expression, even after IFNγ stimulation. Similar results, showing no regulatory effect of HKLD on STAT1α expression were observed ( Figure S4B ).

In mammals, CIITA expression is controlled by multiple promoters and is activated in a selective manner (Muhlethaler-Mottet et al., 1997) ( Figure 1A ). CIITA promoters I and III regulates the expression of MHC-II in dendritic cells and B cells, respectively (Muhlethaler-Mottet et al., 1997). Promoter IV has been demonstrated to mediate IFNγ inducible expression of MHC-II genes (Muhlethaler-Mottet et al., 1997). IFNγ induction is also reported to increase not only type IV but also type I CIITA mRNA levels in bone marrow-derived macrophage (BMM) cells (Pai et al., 2002). Since CIITA promoter IV (pIV) and promoter I (pI) both mediate IFNγ inducible gene expression, we hypothesized that the occupancy of BRG1 would be reduced on these promoters leading to the decreased expression of CIITA. Therefore, we checked the occupancy of BRG1 on CIITA pIV and also for the first time on pI ( Figure 2 ).

ChIP experiments showed that the occupancy of BRG1 on pIV and pI was not significantly altered between infected and uninfected cells under unstimulated conditions ( Figures 2D, E ). On stimulation, the occupancy of BRG1 increased on both CIITA pIV (~ 10 fold, P = 0.0001) and pI (~ 9 fold, P = 0.003) in uninfected cells as compared to the resting macrophages. Further, the occupancy of BRG1 decreased significantly on both pIV (~ 72.5%, P = 0.0016) and pI (~ 92.7%, P = 0.002) in infected, stimulated cells in comparison to uninfected, stimulated cells ( Figures 2D, E ). Thus, the binding pattern of BRG1 to the CIITA promoters is in concordance with the CIITA expression pattern ( Figure S2A ). The occupancy of BRG1 on the promoter of DNA topoisomerase 1 (TOP1) was also checked, which served as a negative control. No occupancy of BRG1 at TOP1 promoter region was observed ( Figure S5B ), suggesting a specific occupancy of BRG1 at CIITA promoters.

Thus, we conclude that in infected and stimulated THP-1 cells, disruption of the occupancy of BRG1 at both the CIITA promoters has a potential role in decreased CIITA transcription

Earlier studies have reported that JAK2/STAT1 signalling in host macrophages is negatively regulated on Leishmania infection (Nandan and Reiner, 1995; Blanchette et al., 1999; Bhardwaj et al., 2005). It is very well established that STAT1α, and its downstream IFNγ responsive genes CIITA and MHC-II (HLA-DR and HLA-DM) are downregulated in IFNγ stimulated host cells post Leishmania infection (Ray et al., 2000; Forget et al., 2005; Matte and Descoteaux, 2010; Singh et al., 2019). In Figure S1A, B , we also demonstrated that parasitic infection had an inhibitory effect on the STAT1α expression, both at the mRNA (at time points, 3, 6 and 24 h, post infection) and protein level at 6 h post infection, even in the presence of IFNγ. Further, the localization of STAT1α to the nucleus was also suppressed in infected and stimulated THP-1 cells as observed by immunofluorescence microscopy ( Figure S1C ). The anti-STAT1α fluorescence intensity in the host cell nuclei also showed a similar pattern ( Figure S1D ). We also observed downregulation of CIITA and MHC-II in parasite infected THP-1 cells ( Figure S2 ).

The transcription factor, STAT1α, has been shown to be important for the expression of IFNγ responsive genes like MHC-II (Darnell et al., 1994). CIITA has been reported to play an important role in the expression of MHC-II in a STAT1-dependent manner (Muhlethaler-Mottet et al., 1997). (Pai et al., 2002). Further, in our previous results, we observed reduced STAT1α ( Figure S1 ), CIITA ( Figure S2 ) and BRG1 expression ( Figure 1 ) in infected, IFNγ stimulated cells. Studies have shown that STAT1α binding to promoters of IFNγ responsive genes such as CIITA and GBP1, is dependent on BRG1 (Ni et al., 2005). Therefore, we hypothesized that the occupancy of STAT1α at the CIITA promoters in infected and stimulated THP-1 cells would also be reduced, leading to downregulation of CIITA expression.

Thus, the occupancy of STAT1α between infected and uninfected cells was checked at both CIITA pIV and pI ( Figure 3 ). No change in occupancy levels was observed in unstimulated cells. However, on stimulation, the occupancy of STAT1α was dramatically increased on CIITA pIV (~ 15 fold, P = 0.00001) and pI (~ 10 fold, P = 0.00001) in uninfected cells when compared to the resting macrophages. On infection, the occupancy of STAT1α decreased significantly at both the promoters (pIV: ~ 83%, P = 0.00004; pI: ~ 77%; P = 0.0001) as compared to the uninfected, stimulated cells. No binding of STAT1α at TOP1 promoter region was detected ( Figure S5A ), thus, suggesting a specific occupancy of STAT1α on CIITA promoters, Therefore, based on these results, we conclude that the decreased occupancy of STAT1α on CIITA promoters leads to decreased CIITA transcription.

Taken together, we conclude that in infected and stimulated THP-1 cells, disruption of the occupancy of both STAT1α ( Figure 3 ), and BRG1 ( Figure 2 ), at CIITA promoters leads to decreased CIITA transcription.

BRG1, a core subunit of the BAF complex, is required for the expression of certain IFNγ responsive genes (Chi, 2004). As observed that reduced BRG1 occupancy is correlated with reduced CIITA expression ( Figure 2 ), we hypothesized that the overexpression of BRG1 would lead to increased CIITA mRNA levels in parasite infected cells, and thus, negatively impact the survival of the parasite within the macrophages. To test this hypothesis, THP-1 cells were transiently transfected with the plasmid overexpressing BRG1 and the transcript levels of the genes of interest were analyzed by qPCR. As expected, BRG1 expression increased in all the conditions in the cells transfected with plasmid overexpressing BRG1 as compared to vector alone ( Figure 4A ). The expression of CIITA in unstimulated cells was comparable between uninfected and infected cells ( Figure 4B ). However, in stimulated cells, the CIITA mRNA was upregulated (~ 53.4 fold, P = 0.0235) significantly in infected cells as compared to the uninfected one ( Figure 4B ).

To determine the intracellular parasite load in these conditions, infected and stimulated cells were used for visually counting the amastigotes within the THP-1 cells after Giemsa staining ( Figure 4C ). A significant decrease in the parasitemia (~ 29.4%, P = 0.012) was observed in infected and stimulated cells overexpressing BRG1, compared to infected and stimulated cells overexpressing only vector ( Figure 4C ). The parasite load in control samples, unstimulated and infected cells, was unchanged between cells overexpressing BRG1 and vector alone ( Figure S6 ). Taken together, this data demonstrates that BRG1 has a potential negative effect on parasite survival.

Enhanced transcription of genes like MHC-II and GBP2 in response to IFNγ is dependent on the acetylation of histones like H3 and H4 (Zika et al., 2003; Ramsauer et al., 2007). BRG1 is known to bind to acetylated histones (Shen et al., 2007). Therefore, we hypothesized that the global levels of H3 acetylation would decrease on the CIITA promoters on infection under stimulated condition. THP-1 cells were infected and stimulated as mentioned before. Cells were harvested at 6 h and analyzed for global H3 acetylation levels by ChIP assay. In unstimulated cells, the acetylation levels were low in both uninfected and infected cells ( Figure 5A ). On stimulation, a significant increase in the total H3 acetylation at CIITA pIV (~ 9 fold, P = 0.00003) and pI (~ 49 fold, P = 0.0002) in uninfected cells was observed as compared to the resting macrophages ( Figures 5A, B ). Further, as expected, in stimulated cells the acetylation levels decreased significantly on both the promoters in infected cells as compared to uninfected cells (pIV: 90%, P = 0.00002; pI: 95.7%, P = 0.0002) ( Figures 5A, B ). Thus, these results confirm our hypothesis.

In our earlier study, we have reported significant upregulation of HDAC1 in THP-1 cells on Leishmania infection (Roy et al., 2020). Therefore, to investigate the role of HDAC1 in the reduced H3 acetylation levels, THP-1 cells were transfected either with HDAC1-siRNA or scrambled-siRNA (Sc-siRNA). In THP-1 cells transfected with Sc-siRNA, comparable levels of HDAC1 were observed in uninfected and infected conditions showing that Sc-siRNA has no inhibitory effect on host HDAC1 expression. When THP-1 cells were transfected with HDAC1-siRNA, a decrease in HDAC1 expression was observed in both infected (~ 62%, P = 0.00009) and uninfected (~ 30%, P = 0.0674) cells. This data confirms a specific silencing effect of HDAC1-siRNA on the expression of host HDAC1 ( Figure 5C ).

The expression of CIITA in cells transfected with Sc-siRNA showed basal levels ( Figure 5D ). In cells transfected with HDAC1-siRNA, the CIITA levels were upregulated significantly in both infected (~ 4.8 fold, P = 0.0177) and uninfected (~ 1.5 fold, P = 0.28) cells. Thus, suggesting that downregulation of HDAC1 is beneficial for the expression of CIITA.

To determine the role of HDAC1-siRNA on parasite load, visual counting of intracellular amastigotes was done after Giemsa staining ( Figure 5E ). In infected and stimulated cells, the parasite load (~ 45.25%, P = 0.04) was significantly downregulated in cells transfected with HDAC1-siRNA as compared to Sc-siRNA. This data demonstrates the positive effect of host HDAC1 on parasite survival.

The H3 acetylation pattern at both the CIITA promoters ( Figures 5A, B ) is in concordance with the mRNA expression pattern of CIITA in similar conditions ( Figure S2A ). Further in Figures 5C –E , a significant increase in CIITA expression whereas a decrease in the parasite viability upon HDAC1 silencing was observed in infected and stimulated cells. Taken together, these results suggest that parasite infection leads to reduced host H3 acetylation at CIITA pI and pIV with the help of HDAC1, leading to decreased CIITA transcription.