High-affinity Zn2+ uptake system ZnuABC is required for bacterial zinc homeostasis in intracellular environments and contributes to the virulence of Salmonella enterica

Abstract

To investigate the relevance of zinc in host-pathogen interactions, we have constructed Salmonella enterica mutant strains in which the znuA gene, which encodes the periplasmic component of the ZnuABC high-affinity Zn2+ transporter, was deleted. This mutation does not alter the ability of Salmonella to grow in rich media but drastically reduces its ability to multiply in media deprived of zinc. In agreement with this phenotype, ZnuA accumulates only in bacteria cultivated in environments poor in zinc. In spite of the nearly millimolar intracellular concentration of zinc, we have found that znuA is highly expressed in intracellular salmonellae recovered either from cultivated cells or from the spleens of infected mice. We have also observed that znuA mutants are impaired in their ability to grow in Caco-2 epithelial cells and that bacteria starved for zinc display decreased ability to multiply in phagocytes. A dramatic reduction in the pathogenicity of the znuA mutants was observed in Salmonella-susceptible (BALB/c) or Salmonella-resistant (DBA-2) mice infected intraperitoneally or orally. This study shows that the amount of free metals available for bacterial growth within the infected animal is limited, despite the apparent elevated concentration of free metals within cells and in plasma and suggests that Salmonella exploits the ZnuABC zinc transporter to maximize zinc availability in such conditions. These results shed new light on the complex functions of zinc in vertebrate and bacterial physiology and pave the way for a better comprehension of pathogenic mechanisms in Salmonella infections.

FIG. 1.

Growth curves of S. enterica serovar Typhimurium in synthetic media. Wild-type serovar Typhimurium (triangles) and znuA::kan SA123 mutant (squares) were grown in LB medium (A) and minimal medium (B). Overnight cultures of both strains grown in LB were diluted 1:500 in fresh LB medium or MM, and the optical density at 600 nm (OD_600nm) was registered every hour during exponential growth.

FIG. 2.

Zinc-dependent ZnuA accumulation in S. enterica serovar Typhimurium (SA140). (A) Bacteria were grown in LB medium (lane 1) supplemented with 5 μM (lane 2) and 10 μM (lane 3) ZnSO₄ and in LB supplemented with EDTA at the following concentrations: 0.2 mM (lane 4), 0.4 mM (lane 5), 0.6 mM (lane 6), 0.8 μM (lane 7), 1 mM (lane 8), and 1.5 mM (lane 9). (B) Bacteria were grown in minimal medium (lane 1) or in minimal medium supplemented with various concentrations of ZnSO₄. Minimal medium was supplemented with the following concentrations of ZnSO₄: 1 μM (lane 2), 0.5 μM (lane 3), 0.1 μM (lane 4), 0.05 μM (lane 5), 0.01 μM (lane 6), 0.005 μM (lane 7), 0.001 μM (lane 8), and 0.0005 μM (lane 9). (C) Bacteria were grown in LB medium (lane 1) supplemented with 1 mM EDTA (lane 2), 0.2 mM 2,2′-bipyridyl (lane 3), and TPEN at the following concentrations: 0.05 mM (lane 4), 0.1 mM (lane 5), 0.15 mM (lane 6), and 0.2 mM (lane 7). The bacteria in lane 8 were grown in minimal medium. (D) Bacteria were grown in minimal medium. When the cultures reached an optical density at 600 nm of 0.5, the medium was supplemented with 3 μM ZnSO₄ (lane 2), 3 μM CuSO₄ (lane 3), 3 μM FeCl₂ (lane 4), and 3 μM MnCl₂ (lane 5), and bacteria were grown for 3 hours before harvesting. Lane 1 shows ZnuA accumulation in bacteria grown in standard minimal medium.

FIG. 3.

Intracellular growth of Salmonella strains. Intracellular growth of the znuA mutant strain SA123 (squares) compared to wild-type serovar Typhimurium (triangles) in differentiated THP-1 human monocytes (A and B) and Caco-2 colon epithelial cells (C). Prior to infection, bacteria were grown in LB medium (A and C) or in LB supplemented with 1mM EDTA (B). The reported CFU/ml values are the means ± standard deviations (error bars) of at least three independent experiments.

FIG. 4.

Zn(II)-Zinquin-dependent fluorescence of THP-1 and Caco-2 cells. Emission spectra of Zn(II)-Zinquin complexes from cellular lysates of differentiated THP-1 cells (solid black line), THP-1 cells precultivated with 100 μM TPEN (dashed line), Caco-2 cells (dotted line), and Caco-2 cells precultivated with 100 μM TPEN (dotted-dashed line). Zn(II)-Zinquin fluorescence intensity on the y axis is shown in arbitrary florescence units normalized to the protein concentration of the samples.

FIG. 5.

ZnuA accumulation in intracellular salmonellae. (A) ZnuA accumulation in bacteria grown in rich medium (LB) and under zinc-limiting conditions (MM) compared to ZnuA accumulation in bacteria extracted from infected macrophages (J774), differentiated monocytes (THP-1), and colon epithelial cells (Caco-2). Each gel shows the results from two independent experiments. (B) Accumulation of ZnuA in bacteria harvested from spleen homogenates of BALB/c mice infected with strain SA140 (znuA::3xFLAG cat::3xFLAG) (lanes 1 and 2) or strain MA7225 (sodCI::3xFLAG cat::3xFLAG) (lanes 3 and 4). Each lane shows the epitope-flagged proteins recovered from a different mouse.

FIG. 6.

Survival of BALB/c and DBA-2 mice infected intraperitoneally with different doses of S. enterica serovar Typhimurium wild-type and SA123 strains. Infection doses were 10 CFU/mouse (▪), 50 CFU/mouse (▴), 250 CFU/mouse (▾), 1,250 CFU/mouse (♦), and 6,250 CFU/mouse (•). Asterisks indicate mortality curves showing significant differences between the wild-type and mutant strains.

FIG. 7.

Survival of BALB/c and DBA-2 mice infected orally with different doses of S. enterica serovar Typhimurium wild-type and SA123 strains. Infection doses were 10⁵ CFU/mouse (▪), 10⁶ CFU/mouse (▴), 10⁷CFU/mouse (▾), 10⁸ CFU/mouse (♦), and 10⁹ CFU/mouse (•). Asterisks indicate mortality curves showing significant differences between the wild-type and mutant strains.