Methylomic Changes of Autophagy-Related Genes by Legionella Effector Lpg2936 in Infected Macrophages

Abstract

Legionella pneumophila (L. pneumophila) is a Gram-negative bacterium that infects the human respiratory tract causing Legionnaires' disease, a severe form of pneumonia. Recently, rising evidence indicated the ability of Legionella to regulate host defense via its type 4 secretion system including hundreds of effectors that promote intracellular bacterial replication. The host defense against such invaders includes autophagic machinery that is responsible for degradation events of invading pathogens and recycling of cell components. The interplay between host autophagy and Legionella infection has been reported, indicating the role of bacterial effectors in the regulation of autophagy during intracellular replication. Here, we investigated the potential impact of Legionella effector Lpg2936 in the regulation of host autophagy and its role in bacterial replication using mice-derived macrophages and human lung epithelial cells (A549 cells). First, monitoring of autophagic flux following infection revealed a marked reduction of Atg7 and LC3B expression profile and low accumulation levels of autophagy-related LC3-I, LC3-II, and the Atg12-Atg5 protein complex. A novel methyladenine alteration was observed due to irreversible changes of GATC motif to G(6 mA) TC in the promoter region of Atg7 and LC3B indicated by cleaved genomic-DNA using the N6 methyladenine-sensitive restriction enzyme DpnI. Interestingly, RNA interference (RNAi) of Lpg2936 in infected macrophages showed dramatic inhibition of bacterial replication by restoring the expression of autophagy-related proteins. This is accompanied by low production levels of bacterial-associated pro-inflammatory cytokines. Furthermore, a constructed Lpg2936 segment in the GFP expression vector was translocated in the host nucleus and successfully induced methyladenine changes in Atg7 and LC3B promoter region and subsequently regulated autophagy in A549 cells independent of infection. Finally, treatment with methylation inhibitors 5-AZA and (2)-Epigallocatechin-3-gallate (EGCG) was able to restore autophagy-related gene expression and to disrupt bacterial replication in infected macrophages. This cumulative evidence indicates the methylation effect of Legionella effector Lpg2936 on the host autophagy-related molecules Atg7 and LC3B and subsequent reduction in the expression levels of autophagy effectors during intracellular replication of L. pneumophila.

Keywords: Legionella pneumophila; Lpg2936 effector; N6-methyleadinine changes; autophagy; macrophages.

FIGURE 1

Depletion of autophagosome formation in response to Legionella infection. (A) The fold change in steady-state mRNA of autophagy Atg7 during the indicated time points upon Legionella infection of mice-derived macrophages compared with control-infected cells. (B) Fold change in steady-state mRNA of LC3B in Legionella-infected macrophages related to control-infected cells during the indicated time points. GAPDH-mRNA was used as an internal control for the qRT-PCR test. (C,D) Immunoblotting assays indicate Atg5–Atg12 protein complex, LC3-I, and LC3-II in infected macrophages for different time points compared with control-infected cells. Calreticulin served as an internal control. (E) Immunofluorescent assay represents accumulated LC3B protein (red) and Legionella protein biomarker (green) in mice-derived macrophages in response to Legionella infection. 4′,6-Diamidino-2-phenylindole, dihydrochloride (DAPI) was used for chromosome counterstaining (blue). (F) Scoring of the percentage of macrophages harboring more than five puncta in infected cells compared with other control cells using ImageJ software. Data are representative of 500 cells. Error panels shown represent three independent experiments displaying similar results. Comparison of groups for statistical differences was done using two-tailed Student t test. ^∗P ≤ 0.05 and ^∗∗P ≤ 0.01.

FIGURE 2

Autophagosome formation and bacterial replication in Lpg2936-knockdown cells. (A) Quantification of Lpg2936 steady-state mRNA in mice-derived macrophages that represented the knockdown efficiency during indicated time points following infection in comparison with control transfection (siRNA-Luci) using qRT-PCR test. (B) Colony-forming unit (CFU) assay of Legionella replication in siRNA-Lpg2936 transfected macrophages during the indicated time points post-infection compared to control-transfected and non-transfected cells. (C,D) Relative expression of autophagy-related genes Atg7 and LC3B in siRNA-transfected and -infected macrophages during the indicated time points using qRT-PCR. (E) Immunoblotting assays represent Legionella indicator protein, Atg5–Atg12 protein complex, LC3-I, and LC3-II in siRNA-transfected and -infected macrophages. Calreticulin served as an internal control. (F) Immunofluorescent assay represents accumulated LC3B puncta (red) and Legionella protein biomarker (green) in murine-derived macrophages in response to siRNA transfection-induced Lpg2936 knockdown. DAPI was used for chromosome counterstaining (blue). Error bars reveal the SD of three independent experiments. Student two-tailed t test was used for statistical differences of Ct values in different groups. ^∗P ≤ 0.05 and ^∗∗P ≤ 0.01.

FIGURE 3

Levels of produced pro-inflammatory cytokines in Lpg2936-knockdown and infected cells. The concentration of pro-inflammatory cytokines (pm/ml) that are produced in the fluid media of infected macrophages in response to siRNA transfection against Legionella effector Lpg2936 compared with control transfection (anti-Luciferase): (A) The concentration of IL-1α. (B) The concentration of IL-1β. (C) The concentration of IL-6. (D) The concentration of TNF-α.

FIGURE 4

Lpg2936-independent regulation of autophagy. (A) Schematic representation of GFP-LpDME constructs revealing cloned full-length cDNA of Lpg2936 downstream GFP cassette. (B) Immunoblotting analysis of resulting GFP and Lpg2936 protein levels in transfected A549 cells using specific antibodies against GFP. (C,D) Relative gene expression of Atg7 and LC3B, respectively, in transfected and rapamycin-treated cells compared with control-treated cells using qRT-PCR assay. (E) Immunoblot analysis detecting Atg5–Atg12 protein complex and conjugated LC3 protein in transfected and rapamycin-treated cells compared with control-treated cells. (F) Fluorescent confocal images represent the considerable levels and localization of GFP-Lpg2936 fusion protein (green) and accumulated LC3 puncta (red) in transfected cells upon rapamycin treatment. Error panels showed the SD of three independent experiments. Student two-tailed t test was used for statistical analysis of Ct values in different groups. ^∗P ≤ 0.05 and ^∗∗P ≤ 0.01.

FIGURE 5

Adenine N6 methylation changes in 3′-UTR of Atg7 and LC3 induced by Legionella effector Lpg2936. (A) Agarose gel electrophoresis represents genomic DNA that was isolated from Legionella-infected macrophages and digested with methylation-dependent DpnI for 2 h. (B) Fold change of amplified autophagy-related molecules, 3′-UTR-Atg7 and 3′-UTR-LC3B, in DpnI-cleaved DNA product for infected cells using qRT-PCR. (C,D) Agarose gel electrophoresis represents the amplified autophagy-related molecules 3′-UTR-Atg7 and 3′-UTR-LC3B, respectively. (E) Agarose gel electrophoresis reveals DpnI-cleaved genomic DNA isolated from LpDME-transfected A549 cells in comparison with cleaved genomic DNA isolated from control-transfected cells. (F) Fold change of amplified autophagy-related molecules, 3′-UTR-Atg7 and 3′-UTR-LC3B, in DpnI-cleaved DNA product for transfected cells using qRT-PCR. Error panels indicate the SD of three independent experiments. Student two-tailed t test was used for statistical differences of Ct values in different groups. ^∗P ≤ 0.05 and ^∗∗P ≤ 0.01.

FIGURE 6

Resting of Legionella replication via restoring autophagy-related genes in macrophages in response to methylation inhibitor treatment. (A,B) Cell viability rate of mice-derived macrophages that were overnight-treated with the indicated concentration of either 5-AZA or EGCG by using MTT kit. (C) Relative gene expression of autophagy-related Atg7 and LC3 in Legionella-infected macrophages that were pre-treated with 50 μM of either 5-AZA or EGCG compared with control-infected cells using qRT-PCR. (D) Immunoblotting assays represent autophagy-related Atg5–Atg12 protein complex, LC3-I, and LC3-II in infected macrophages that were pre-treated with 50 μM of either 5-AZA or EGCG compared with DMSO-treated and infected cells. Calreticulin served as an internal control. (E) Fold change in steady-state mRNA of the indicated pro-inflammatory cytokines in infected macrophages that were pre-treated with 50 μM of either 5-AZA or EGCG compared with DMSO-treated and infected cells using qRT-PCR. (F) CFU of Legionella particles in mice-derived macrophages that were pre-treated with 50 μM of either 5-AZA or EGCG during the indicated time points compared with DMSO-treated and infected cells. Error panels indicate the SD of three independent experiments. Student two-tailed t test was used for statistical differences of Ct values in different groups. ^∗P ≤ 0.05 and ^∗∗P ≤ 0.01.

FIGURE 7

Schematic model represents the function of Lpg2936 enzyme during Legionella replication. To modulate host defense before infection, L. pneumophila submits a specific agent known as Lpg2936 effector, which translocates in the host nucleus and begins its methylation activity. The first methylated molecules are autophagy-related; Atg7 and LC3B resulted in marked inhibition of autophagosome formation and high production levels of Legionella-associated pro-inflammatory cytokines. Such events modulate the cellular immune response and promote intracellular replication of L. pneumophila.