Antimicrobial Effectors Act Cooperatively to Stress Salmonella in the Macrophage Phagosome

Kim, Byoungkwan

Permalink

https://hdl.handle.net/2142/18605 Copy

Description

Title

Antimicrobial Effectors Act Cooperatively to Stress Salmonella in the Macrophage Phagosome

Author(s)

Kim, Byoungkwan

Issue Date

2011-01-21T22:51:24Z

Director of Research (if dissertation) or Advisor (if thesis)

Slauch, James M.

Doctoral Committee Chair(s)

Slauch, James M.

Committee Member(s)

• Cronan, John E.
• Wilson, Brenda A.
• Vanderpool, Carin K.

Department of Study

Microbiology

Discipline

Microbiology

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

Ph.D.

Degree Level

Dissertation

Keyword(s)

• Salmonella virulence
• Periplasmic superoxide dismutase
• Cationic antimicrobial peptide
• Peptidoglycan, Periplasmic stress

Abstract

Salmonella Typhimurium survives and replicates in host macrophages. The bacterium produces two Cu-Zn periplasmic superoxide dismutases, SodCI and SodCII. Although both enzymes are expressed during infection, only SodCI contributes to virulence in the mouse by combating phagocytic superoxide. We have shown that the differential contribution to virulence is due to inherent differences in the SodCI and SodCII proteins that are apparently independent of enzymatic activity. Our goal is to understand what features of these proteins are required for virulence in the animal. SodCII is a monomer, is protease sensitive, and is released by osmotic shock, like other known periplasmic proteins. In contrast, SodCI forms dimers, is protease resistant, and is retained within the periplasm after osmotic shock, a phenomenon that we term tethering. We are further characterizing SodCI tethering within the periplasm and its importance for Salmonella virulence. We hypothesize that during infection, cationic antimicrobial peptides (CAMPs) from the macrophage transiently disrupt the outer membrane. SodCII is released from the periplasm and degraded. SodCI is both tethered within the periplasm and is protease resistant. This hypothesis provides a rationale for why tethering is important for virulence. We have found that osmotic shock and polymyxin B treatment are analogous. SodCII is preferentially released by polymyxin B, while SodCI is retained. Likewise, mouse macrophage Cathelicidin-related antimicrobial peptide (CRAMP) treatment leads to SodCII release, while SodCI is tethered within the periplasm. We tested if SodCII, protected from release by antimicrobial peptides, could contribute to virulence in vivo. A Salmonella pmrAC mutant is resistant to CAMPs. We show that SodCII enhances virulence in a pmrAC background. These results suggest that SodC tethering within periplasm is important for Salmonella virulence. We are also addressing how SodCI is tethered within the periplasmic space. It was previously shown that SodCI was not found in the membrane fraction and is tethered even when it is over expressed. We also know that this is a non-covalent interaction. Therefore, we hypothesize that it binds to something abundant in periplasm. We can show that SodCI binds to purified peptidoglycan (PGN), whereas PhoA and monomeric SodCI, both of which were released by osmotic shock, do not bind. We also show SodCI does not bind to chitin, staphylococcus PGN but binds bacillus PGN which has same structure of Salmonella PGN. This result suggests that SodCI is tethered to peptidoglycan during osmotic shock or antimicrobial peptide treatment. Lastly, we investigated the relationship between the macrophage host factors and Salmonella periplasmic stress. Salmonella cope with various antimicrobial substances in macrophages, such as cationic antimicrobial peptides (CAMPs), proteases, and reactive oxygen species, by encoding various virulence factors such as the PmrAB regulon that modifies LPS, the protease inhibitor Ecotin, and the periplasmic superoxide dismutase SodCI. The phagocytic antimicrobial substances presumably initiate damage to the bacterium in the outer membrane or periplasmic space. The Sigma E and the Cpx regulons of Salmonella sense and respond to stress in the outer membrane and periplasm, respectively. It is known that these periplasmic stress response regulators are induced and important during the infection, but the nature of the host factors that induce these regulators has not been elucidated in vivo. To answer this question and gain insight into the mechanisms by which phagocytes kill bacteria, we constructed RpoE- and Cpx-regulated lacZ fusions to measure the level of stress during infection in the mouse model. To test the effect of cationic antimicrobial peptides (CAMPs) during infection, htrA-lacZ (Cpx-dependent) and rpoE-lacZ (RpoE-dependent) expression was compared in wild type and pmrA mutant backgrounds. The PmrA regulon is required to protect against CAMPs. The htrA fusion was specifically induced in the pmrA mutant during infection, whereas rpoE was not. Interestingly, in vitro, CRAMP (presumed to be the principal antimicrobial peptide and mouse macrophages) and the polymyxin derivative, polymyxin nonapeptide (PMNP), induce both htrA and rpoE, whereas cecropin induces htrA but not rpoE. These results suggest that different antimicrobial peptides induce you induce different types of damage to the cell. Extrapolating this result to our in vivo data suggests that the damage occurred in the phagosome is analogous to that used by the secretary in vitro rather than CRAMP. The effect of superoxide on the Salmonella periplasm in the phagosome was also tested by monitoring the expression of these genes in a sodCI deletion background. Our results show that rpoE expression was induced but htrA was not. Neither hydrogen peroxide nor chemically generated superoxide induced htrA or rpoE in vitro. The periplasmic stress seen in the sodCI deletion may be the indirect effect of reactive oxygen species. Lastly, the effect of macrophage proteases was tested in vivo by removing the periplasmic serine protease inhibitor Ecotin. Only rpoE was induced consistently during infection with the eco strain. In the future, the effects of proteases on periplasmic stress will be tested in vitro.

Graduation Semester

2010-12

Permalink

http://hdl.handle.net/2142/18605

Copyright and License Information

Copyright 2010 Byoungkwan Kim

Owning Collections