The Rab32/BLOC-3-dependent pathway mediates host defense against different pathogens in human macrophages

Abstract

Macrophages provide a first line of defense against microorganisms, and while some mechanisms to kill pathogens such as the oxidative burst are well described, others are still undefined or unknown. Here, we report that the Rab32 guanosine triphosphatase and its guanine nucleotide exchange factor BLOC-3 (biogenesis of lysosome-related organelles complex-3) are central components of a trafficking pathway that controls both bacterial and fungal intracellular pathogens. This host-defense mechanism is active in both human and murine macrophages and is independent of well-known antimicrobial mechanisms such as the NADPH (reduced form of nicotinamide adenine dinucleotide phosphate)-dependent oxidative burst, production of nitric oxide, and antimicrobial peptides. To survive in human macrophages, Salmonella Typhi actively counteracts the Rab32/BLOC-3 pathway through its Salmonella pathogenicity island-1-encoded type III secretion system. These findings demonstrate that the Rab32/BLOC-3 pathway is a novel and universal host-defense pathway and protects mammalian species from various pathogens.

Fig. 1. The Rab32/BLOC-3–dependent pathway mediates the killing of different pathogens.

(A and B) BMDMs were derived from control mice C57BL/6 (wt) or from HPS4^−/− mice, infected with S. aureus and (A) CFUs were enumerated at the times indicated or (B) cells were fixed at 3 hours post-infection (p.i.) and stained to show Rab32 localization. (C) wt or HPS4^−/− mice were infected with C. albicans, and fungal burden in kidneys was evaluated 72 hours p.i.

Fig. 2. The Rab32/BLOC-3–dependent pathway does not require oxidative burst to clear bacterial and fungal infections in murine cells.

BMDMs were infected with S. Typhi wild-type (WT) or expressing GtgE (::gtgE), and CFUs were enumerated at the times indicated. (A) BMDMs were derived from control mice (wt) or from NADPH oxidase^−/− mice (Phox^−/−). (B) BMDMs were infected in the presence or absence of the iNOS inhibitor 1400W. (C) BMDMs were derived from control mice (wt) or CRAMP^−/− mice.

Fig. 3. GtgE delivery from S. Typhi results in the cleavage of human Rab32.

(A) PMA-differentiated THP-1 cells were left uninfected or infected with either wild-type S. Typhi (WT) or an S. Typhi strain expressing GtgE (::gtgE). Cells were lysed 2.5 hours p.i. and analyzed by Western blot with a Rab32-specific antibody. (B and C) Peripheral blood monocyte–derived macrophages were infected with either wild-type S. Typhi (WT) or an S. Typhi derivative expressing GtgE (::gtgE), both carrying a chromosomal copy of the mCherry gene, fixed at 2.5 hours p.i. and analyzed by immunofluorescence with a Rab32-specific antibody. Scale bars, 10 μm. A.U., arbitrary units.

Fig. 4. Rab32 inactivation and BLOC-3 knockout results in S. Typhi over-replication in human macrophages.

(A) Peripheral blood monocyte–derived macrophages were infected with either wild-type S. Typhi (WT) or an S. Typhi strain expressing GtgE (::gtgE). Cells were lysed at the indicated time points to measure CFUs. (B) THP-1 cells were transduced with lentivirus to silence the indicated Rab, infected with wild-type S. Typhi, and lysed at the indicated time points to measure CFUs. (C) Human macrophages derived from WT or HPS4^−/− hiPSCs were infected with either wild-type S. Typhi (WT) or an S. Typhi strain expressing GtgE (::gtgE), lysed at the indicated time points for counting of intracellular CFUs. (D) Human macrophages derived from WT or HPS4^−/− hiPSCs were infected with wild-type S. Typhi (WT) and lysed at the indicated time points to measure CFUs. (E) Human macrophages derived from WT or HPS4^−/− hiPSCs were plated on glass coverslips, then infected with S. Typhi glmS∷Cm::mCherry or S. Typhi::gtgE glmS∷Cm::mCherry, and fixed at 1.5 and 24 hours p.i. Differentiated macrophages were identified by CD68 staining, and bacteria in CD68⁺ cells were counted. The whole populations are reported, and the infected versus total number of cells are indicated. Bars represent the mean and SD of the population. (F) Human macrophages derived from WT or HPS4^−/− hiPSCs were infected with S. Typhi glmS∷Cm::mCherry or S. Typhi::gtgE glmS∷Cm::mCherry, fixed at 24 hours p.i., and analyzed by flow cytometry. CD68⁺ cells were subgated in two populations containing respectively low or high mCherry signal (i.e., bacterial content). Errors bars represent SD between experiments.

Fig. 5. S. Typhi counteracts the Rab32/BLOC-3–dependent pathway through the SPI-1 type III secretion system.

(A) Human macrophages derived from WT or HPS4^−/− hiPSCs were infected with S. aureus and lysed at the indicated time points to measure intracellular CFUs. (B) Human macrophages derived from WT hiPSCs were infected with wild-type S. Typhi (WT) or E. coli O157 and lysed at the indicated time points to measure intracellular CFUs. (C) Human macrophages derived from WT or HPS4^−/− hiPSCs were infected with either wild-type S. Typhi (WT) or S. Typhi ∆invA (∆invA). Cells were lysed at the indicated time points to measure intracellular CFUs. (D) Human macrophages derived from WT or HPS4^−/− hiPSCs were infected with either wild-type S. Typhi (WT) or E. coli O157. Cells were lysed at the indicated time points to measure intracellular CFUs. Results in (B), (C), and (D) are reported as percentage of CFUs measured at the first time point (1.5 hours p.i.). Values are means ± SEM of at least three independent experiments performed in triplicate. P values were calculated using the Student’s t test and are indicated only when <0.05.