Vanderbilt Institute of Chemical Biology



Discovery at the VICB







Increasing Bacterial Virulence by Metabolic Collaboration


By: Carol A. Rouzer, VICB Communications
Published: October 20, 2014



Exchange of metabolites between antibiotic resistant strains of bacteria contributes to their growth and virulence.


The increasingly rapid emergence of antibiotic resistant bacteria is one of the most important public health threats of our time. Bacteria acquire resistance through multiple mechanisms, including expression of enzymes that destroy the drug, efflux pumps that eject the drug from the cell, and mutation of target proteins. In addition, some bacteria can acquire mutations in genes required for fundamental biosynthetic pathways, rendering them incapable of carrying out respiration. These bacteria must rely on fermentation to survive but are frequently resistant to many antibiotics. Staphylococcus aureus (Figure 1) uses this strategy, particularly in sites of chronic infections as found in the bones (osteomyelitis), in tissues of diabetics, and in the lungs of cystic fibrosis patients. These respiration-deficient S. aureus strains, referred to as small colony variants (SCVs), survive despite an inability to synthesize key biomolecules such as menaquinone or heme. However, Vanderbilt Institute of Chemical Biology (VICB) member Eric Skaar and his laboratory hypothesized that SCVs likely do not grown in isolation in vivo. Thus, a mechanism for survival could be the exchange of metabolites between different strains of SCVs or between SCVs and bacteria of other species. They now report key evidence that supports that hypothesis [N. D. Hammer, J. E. Cassat, et al. (2014) Cell Host & Microbe, 16, 531].



Figure 1. Figure 1. Electron micrograph of S. aureus bacteria (gold) escaping from a white blood cell (blue). Figure is reproduced from NIAID and is in the public domain.


For their initial studies, the investigators used S. aureus SCVs created in the laboratory by mutation of key genes in the pathways for menaquinone (Δmen) or heme (Δhem) biosynthesis. SCVs bearing mutations in either pathway grew more slowly than wild-type bacteria and exhibited resistance to aminoglycoside antibiotics. However, when Δmen and Δhem SCVs were cultured together, their growth rate increased markedly, approaching that of wild-type bacteria. Growth enhancement was evident either by a cross-streaking assay (Figure 2) or in a liquid culture assay. Bacteria containing mutations in both menaquinone and heme biosynthesis (ΔmenΔhem) were unable to enhance the growth of Δhem bacteria, confirming that the growth-enhancing effects of Δmen was due to their ability to provide heme to the Δhem strain. This was further confirmed by using ΔmenΔhem SCVs that were capable of  producing heme if the precursor δ-aminolevulinic acid (ALA) was provided. These bacteria could enhance growth ofΔhem SCVs but only in the presence of ALA. The investigators also found that addition of the heme binding protein Isdl to co-cultures of Δmen and Δhem SCVs suppressed growth. Together these data supported the hypothesis that exchange of menaquinone and heme between the two strains of SCVs was required for the growth enhancement observed in co-culture.


Figure 2.
Cross-streaking assay for growth enhancement in SCV co-culture. A vertical streak of Δhem SCVs is first placed on the plate, and then Δmen SCVs are streaked horizontally across the Δhem streak. The mixing of the bacteria as the loop crosses the Δhem streak results in abundant growth on the right side of the cross. Poor growth, where the SCVs remain isolated from each other, is seen at the other three arms of the cross. Image kindly provided by Eric Skaar, copyright 2014.



Bacteria possess a number of signaling pathways to sense the presence of other bacteria around them and to respond to that presence. This process, called quorum sensing is mediated by the accessory gene regulator (agr) system in S. aureus. The Skaar lab investigators found that agr-dependent signaling was markedly reduced in single cultures of Δmen or Δhem SCVs as compared to that in wild-type cultures. However, in co-cultures of Δmen and Δhem SCVs, quorum sensing approached that observed in wild-type bacteria. This increase in agr-mediated signaling in the co-cultures was accompanied by an increase in the production of virulence factors, which are controlled by the accessory gene regulator. Consistently, supernatants from Δmen and Δhem co-cultures were more toxic to cultured human osteoblasts than were supernatants from monocultures of Δmen or Δhem SCVs.


To determine if the presence of multiple SCV strains could enhance bacterial growth in vivo, the Skaar lab used a mouse model of osteomyelitis (Figure 3), an infection in which S. aureus SCVs are commonly found. They infected femurs of the mice with wild-type S. aureus, Δhem or Δmen SCVs alone, or the two SCVs together. Consistent with their hypothesis, bacterial growth was rapid in the wild-type S.aureus-infected bones, and very slow in the bones infected with just one strain of SCV, while bones infected with both strains of SCVs exhibited bacterial growth similar to that seen in wild-type infections. The amount of bone destruction correlated with the amount of bacterial growth, suggesting that cooperation between two genetically distinct SCVs can lead to greater virulence in this in vivo model.




Figure 3. In a mouse model of osteomyelitis, the femurs of mice are infected with S. aureus, leading to bone erosion as can be seen on the bone on the left. MicroCT imaging analysis can be used to estimate the amount of new bone growth (green) and cortical bone destruction (yellow) which occur in response to the infection. Right: a concatenated image showing microCT renderings overlying the infected murine femur. Image kindly provided by Eric Skaar, copyright 2013.



In the lungs of cystic fibrosis patients, S. aureus often grows in the presence of other bacteria, including Pseudomonas aeruginosa. P. aeruginosa produces pyocyanin, an exotoxin that inhibits respiration in S. aureus, leading to SCV formation. Frequent use of antibiotics in these patients also leads to SCV formation via other mechanisms. To mimic these conditions in the laboratory, the Skaar lab exposed S. aureus to pyocyanin or to the antibiotic gentamycin to create SCVs. They then demonstrated that co-culture of these two distinct SCV strains led to an enhancement of growth. Similarly, S. aureus frequently grows together with Enterococcus faecalis in diabetic tissue infections. The investigators demonstrated that co-cultures of an S. aureus Δmen SCV together with E. faecalis, which does not produce heme, resulted in growth enhancement of both species.


Bacteria isolated from the respiratory tract of a cystic fibrosis patient included mixtures of Staphylococcus epidermidis, streptococcal species, and diphtheroid-like species. All of these isolates enhanced the growth of laboratory-derived S. aureus SCVs. Similarly, gentamycin-resistant SCVs from the same patient enhanced the growth of the laboratory-derived SCVs. These results confirmed the clinical relevance of the co-culture growth enhancement phenomenon.


It is important to note that co-culture of antibiotic resistant SCVs did not eliminate their resistance. This finding is particularly important because clinical tests for antibiotic sensitivity are typically interpreted as reflecting the response of a single bacterial clone. Instead, the growth and antibiotic sensitivity of bacteria at a site of infection could well be the result of a complex interaction between multiple species. Failure to take this into account could well contribute to treatment failure of infections, especially in immunologically compromised hosts.





The Vanderbilt Institute of Chemical Biology, 896 Preston Building, Nashville, TN 37232-6304, phone 866.303 VICB (8422), fax 615 936 3884
Vanderbilt University is committed to principles of equal opportunity and affirmative action. Copyright © 2014 by Vanderbilt University Medical Center