Mycobacterial P1-Type ATPases Mediate Resistance to Zinc Poisoning in Human Macrophages

Summary Mycobacterium tuberculosis thrives within macrophages by residing in phagosomes and preventing them from maturing and fusing with lysosomes. A parallel transcriptional survey of intracellular mycobacteria and their host macrophages revealed signatures of heavy metal poisoning. In particular, mycobacterial genes encoding heavy metal efflux P-type ATPases CtpC, CtpG, and CtpV, and host cell metallothioneins and zinc exporter ZnT1, were induced during infection. Consistent with this pattern of gene modulation, we observed a burst of free zinc inside macrophages, and intraphagosomal zinc accumulation within a few hours postinfection. Zinc exposure led to rapid CtpC induction, and ctpC deficiency caused zinc retention within the mycobacterial cytoplasm, leading to impaired intracellular growth of the bacilli. Thus, the use of P1-type ATPases represents a M. tuberculosis strategy to neutralize the toxic effects of zinc in macrophages. We propose that heavy metal toxicity and its counteraction might represent yet another chapter in the host-microbe arms race.


INVENTORY OF SUPPLEMENTAL INFORMATION
. Significant features of the zinc stress response following exposure of M. tuberculosis to high concentrations of zinc (0.5 mM), as revealed by microarray analysis Table S2. List of oligonucleotides used in the study Movie S1. Time-lapse fluorescence microscopy of FZ3 staining of E. coli vacuoles in a human macrophage.   were processed at 4hr p.i. either according to a conventional EM procedure (A) or to the AMG procedure (B). A) In control cells fixed according to the normal procedure for conventional electron microscopy, free zinc ions are too small to be visualized. As a consequence the organelles cited above are all free of electron dense deposits. In contrast, the fixation conditions are such that all the membranes can be well visualized. In such conditions, all the bacilli were found to reside within membrane-bound phagosomes (arrows). B) Free intracellular zinc ions were captured by fixing cells with 2.5% glutaraldehyde in presence of 0.1% sodium sulfide. This induced the formation of zinc sulfur nanocrystals that were further enhanced with silver by the AMG method. On this general view, the small dense precipitates corresponding to the free zinc ions formed during this procedure can be seen within mitochondria (M), lysosomes (Ly) and also in an M. tb-containing phagosome (Ph) in which they are located between the outer surface of the mycobacterium and the phagosome membrane. Due to the AMG procedure, the cell ultrastructure, and more especially the membrane of the different cellular organelles, including phagosomes, lacks definition, and is, therefore, usually difficult to visualize. Scale bar= 0.5µm.  Human macrophages were infected at a MOI of 10, stained without fixation with FZ3, and observed lived by confocal microscopy. Arrowheads point to FZ3 vesicular structures contacting FZ3-positive phagosomes. Accelerated 75 times.   Halle-Wittenberg, Germany), and were grown in the presence of chloramphenicol (20 µg/ml) or ampicillin (100 µg/ml) and chloramphenicol, respectively.

Plasmids and reagents
The

Construction of the M. tuberculosis ctpC mutant and complemented strains
Briefly, a DNA fragment overlapping the ctpC gene was amplified by PCR from genomic DNA with the oligonucleotides ctpCKOFd and ctpCKORv (Table S2). It was inserted, after the insertion of a pACYC177 (New England Biolabs)-derived kanamycin resistance cassette (Kan R ), into the unique XhoI restriction site of pPR23. The resulting plasmid was transferred by electroporation into M. tuberculosis, and an allelic exchange at the ctpC locus was screened by PCR analysis of the genomic DNA from several kanamycin-and sucroseresistant colonies, with the following primers: P1, P2, P3 and P4 (Table S2 and Figure   S3A,B). One M. tuberculosis clone with a pattern corresponding to the disruption of ctpC was selected and named ctpC::Kan R . A complemented strain was obtained by electroporation of to stain plasma membranes. After labeling, coverslips were set in Fluoromount G (SouthernBiotech) on microscope slides.

Time-Lapse Fluorescence Microscopy Analysis
Human macrophages were cultivated in 8-well Lab-Tek Chamber Slide System (Thermo Scientific). Cells were infected for 20 min at 37°C with Crimson-expressing E. coli at a MOI of 10 bacteria/cell, washed in PBS, and incubated for 20 min at 37°C with 5 µM FZ3 in Hepes buffer. Cells were then washed in PBS and observed in a 37°C temperature-controlled box in Hepes buffer. Confocal microscopy analysis was performed with a LSM710 microscope equipped with a x40 1.30 NA objective (Carl Zeiss, Inc.), recorded with Zen software (Carl Zeiss, Inc.) and movie was mounted with ImageJ software. Images were acquired every 15 s.

AMG silver enhancement and processing for electron microscopy
At selected time points p.i., cells were fixed overnight at 4°C with 2.5% glutaraldehyde in 0.1M Na-cacodylate buffer containing 0.1M sucrose and 0.1% sodium sulfide. Cells were thoroughly washed with sucrose-containing Na-cacodylate buffer, and incubated for 60 min at RT, in the dark, with 1.5 ml of AMG developer. The latter consisted of a 60 ml gum Arabic solution and 10 ml Na-citrate buffer (25.5 g citric acid, 1 H 2 O + 23.5 g Na-citrate, 2 H 2 O added to 100 ml distilled water), 15 ml reductor (0.85 g of hydroquinone dissolved in 15 ml distilled water at 40°C) and 15 ml of a solution containing silver ions (0.12 g silver lactate in 15 ml distilled water at 40°C) which were added immediately before use, while thoroughly stirring the AMG solution (Danscher, 1981). The AMG development was stopped by replacing the developer with a 5% Na-thiosulfate solution (in distilled water) for 10 min.
Cells were washed extensively with distilled water (5 successive baths for 1 min each) and further processed for electron microscopy. After fixation and treatment according to the AMG procedure described above, cells were post-fixed for 1h at RT with 1% osmium tetroxide in 0.1M Na-cacodylate buffer, pH 7.2, devoid of sucrose. They were washed with buffer, scraped off the dishes, concentrated in 2% agar in cacodylate buffer and treated for 1 h at room temperature with 1% uranyl acetate in Veronal buffer. Samples were dehydrated in a graded series of ethanol and embedded in Spurr resin. Thin sections (70nm-thick) were stained with 1% uranyl acetate in distilled water and then with lead citrate. As a control, infected cells were processed for conventional electron microscopy. In this case, cells were first fixed for 1h at RT with 2.5% glutaraldehyde in 0.1M Na-cacodylate buffer, pH 7.2, containing 0.1M sucrose, 5mM CaCl 2 and 5mM MgCl 2 before fixation with osmium tetroxide and subsequent processing as indicated above above. For quantification, 100-150 different bacilli per sample were examined for the distribution of zinc crystals on the outer surface of bacilli, the cell wall and the cytoplasm. Care was taken to avoid serial sections.

Mouse infection
Mice were housed in pathogen-free conditions and treated according to institutional animal care protocol no. 20080318/9, approved by the Regional Ethics Committee of Midi-Pyrénées (France) for Animal Experimentation (authorization no. MP/01/36/06/08). The survival of SCID mice was monitored over time, and mice were sacrificed when they showed signs of suffering.

qPCR and microarray analysis
Briefly, the bacterial pellet was resuspended in 700 µl of RLT buffer supplemented with 1%  (Table S2).
For microarray experiments to assess the effect of zinc on the mycobacterial transcriptome, mycobacteria were exposed to 0, 50 or 500 mM ZnSO 4 for four hours. RNAs were extracted using the procedure described above, were checked with the Nanodrop ND 1000 and Bioanalyzer 2100 Expert (Agilent), and were labeled with the Quick Amp Labeling kit (Agilent) where the oligodTprimer was replaced by a random primer T7 ( transcripts were excluded. The hybridised slides were scanned on an Axon 4000B (Molecular devices) and analyzed with FeatureExtraction V.9 (Agilent). Data were pre-processed, normalized, and analyzed using the LIMMA (linear models for microarray analysis) library of the Bioconductor R package (Smyth, 2005).

Phylogenetic analysis
We used Muscle software to align the sequences. Ambiguous regions (i.e. containing gaps and/or poorly aligned) were removed with Gblocks. The phylogenetic tree was reconstructed by the maximum likelihood method, as implemented in the PhyML program, with the WAG matrix and a gamma correction for variable rates of evolution. Internal branch reliability was assessed with the aLRT test. The values reported are aLRT values and only aLRT values >50 are shown.