Cell Host & Microbe
Previews occurs on a variety of cell types and is connected to multiple signaling networks. Restoring community diversity with obligate anaerobes, such as Barnsiella, Bacteroides, or Ruminococcus, could also be worthwhile. The findings of Pham and colleagues, however, suggest a direct and simpler approach: administration of fucosylated oligosaccharides, such as 20 -fucosyllactose, which may do much to restore community complexity and promote homeostasis. ACKNOWLEDGMENTS The authors wish to thank Jose´ T. Saavedra for assistance in developing the figure that accompanies this article. For work on enterococcal colonization, virulence, and antibiotic resistance, the authors gratefully acknowledge the support of
NIH grants AI072360 and AI108710, the Harvardwide Program on Antibiotic Resistance AI083214, and fellowship support from Grant AI109855 (to D.V.T.).
REFERENCES Deatherage Kaiser, B.L., Li, J., Sanford, J.A., Kim, Y.M., Kronewitter, S.R., Jones, M.B., Peterson, C.T., Peterson, S.N., Frank, B.C., Purvine, S.O., et al. (2013). PLoS ONE 8, e67155. Goto, Y., Obata, T., Kunisawa, J., Sato, S., Ivanov, I.I., Lamichhane, A., Takeyama, N., Kamioka, M., Sakamoto, M., Matsuki, T., et al. (2014). Science 345, 1254009. Jernberg, C., Lo¨fmark, S., Edlund, C., and Jansson, J.K. (2007). ISME J. 1, 56–66. Kinnebrew, M.A., Ubeda, C., Zenewicz, L.A., Smith, N., Flavell, R.A., and Pamer, E.G. (2010). J. Infect. Dis. 201, 534–543.
Lo Vecchio, A., and Cohen, M.B. (2014). Curr. Opin. Gastroenterol. 30, 47–53. Pham, T.A.N., Clare, S., Goulding, D., Arasteh, J.M., Stares, M.D., Browne, H.P., Keane, J.A., Page, A.J., Kumasaka, N., Kane, L., et al. (2014). Cell Host Microbe 16, this issue, 504–516. Shogan, B.D., Smith, D.P., Christley, S., Gilbert, J.A., Zaborina, O., and Alverdy, J.C. (2014). Microbiome 2, 35. Turnbaugh, P.J., Ridaura, V.K., Faith, J.J., Rey, F.E., Knight, R., and Gordon, J.I. (2009). Sci. Transl. Med. 1, 6ra14. Ubeda, C., Taur, Y., Jenq, R.R., Equinda, M.J., Son, T., Samstein, M., Viale, A., Socci, N.D., van den Brink, M.R., Kamboj, M., and Pamer, E.G. (2010). J. Clin. Invest. 120, 4332–4341. Van Tyne, D., and Gilmore, M.S. (2014). Annu. Rev. Microbiol. 68, 337–356.
Strength in Diversity Daniel J. Wolter1 and Lucas R. Hoffman1,2,* 1Department
of Pediatrics of Microbiology University of Washington, Seattle, Seattle, WA 98195, USA *Correspondence: [email protected] http://dx.doi.org/10.1016/j.chom.2014.09.018 2Department
In this issue of Cell Host & Microbe, Hammer et al. (2014) show that distinct, slow-growing bacteria have better in vitro and in vivo growth and virulence when cocultured than in isolation. They provide evidence that the observed inter- and intraspecies ‘‘complementation’’ involves the intercellular exchange of metabolites. Laboratory culture is a mainstay of medical bacteriology. The ability to grow a pathogen in pure culture is the basis for one of Koch’s postulates—the criteria presented by Robert Koch in 1890 to establish relationships between individual pathogens and disease (Fredericks and Relman, 1996). Since then, much of medical microbiology, including the identification of infecting bacteria, testing their in vitro susceptibilities to antibiotics, and probing myriad other behaviors, relies on laboratory growth in pure, single-species (i.e., ‘‘axenic’’) cultures. These methods have been refined for well over a century, resulting in a dizzying array of media and techniques, and allowing for everimproved capacity for bacterial cultivation, identification, and discrimination.
The usefulness of axenic cultures is undeniable; these techniques have been invaluable in diagnosing and choosing treatments for innumerable infections. These methods have also proven to be incredibly powerful for investigating the inner workings of specific pathogens, including their growth requirements, their patterns of gene and protein expression, and even their virulence. Axenic cultures are particularly effective when applied to acute infections by a single bacterial species. In such cases, it can be relatively straightforward to pinpoint the pathogen in question and to identify antibiotics that should be able to eradicate, or at least suppress, the infection. However, there are reasons to question the accuracy of axenic culture techniques
when they are applied to more complex infections. For example, a single-isolate culture may not faithfully reflect the behavior of an organism when multiple microbes are present in infected tissues, or during chronic infections, when even a single species may diversify as it adapts to the varying selective pressures it might encounter in those tissues. Pure cultures of individual isolates from these infections may underestimate the diversity of bacteria—both within and among species—that contribute to a particular disease. Without an inclusive accounting of these infecting microbes, it is difficult to understand the pathogenesis of polymicrobial or chronic infections and to treat them. Specifically, a single bacterial isolate could behave entirely differently
Cell Host & Microbe 16, October 8, 2014 ª2014 Elsevier Inc. 427
Cell Host & Microbe
Previews
In vivo
Inoculum
In vitro
Normal colony
SCV-1
SCV-2
Virulence Factor
Metabolites
Healthy Bone
Figure 1. Impact of Community Interactions on the Growth and Virulence of Slow-Growing Bacteria A metabolically deficient small colony variant (SCV, center) has significantly impaired growth and decreased virulence factor production in vitro and lower bacterial burdens and virulence in an osteomyelitis infection model (indicated by gray lesions in bone) compared to bacteria with normal growth (left). However, cocultures of two SCVs (SCV-1 and SCV-2, right) with distinct metabolic deficiencies results in enhanced growth and virulence in vitro and in vivo, probably through the intercellular transfer of metabolites.
when growing in pure culture than when in communities comprised either of other variants of its own species or with other species. Interactions between different microbes in such communities could alter each member’s metabolism, virulence, and susceptibilities to antibiotics. Many chronic infections persist despite treatment with antibiotics that effectively kill each identified pathogen isolated in pure cultures. Therefore, accurately describing and modeling community behaviors could help to identify improved treatments for chronic infections, which account for tremendous morbidity and mortality. The work described in this issue of Cell Host & Microbe by Hammer et al. (2014) provides a provocative example of bacterial community interactions at both intra- and interspecies levels that may be relevant for many chronic infections, particularly those that include the common animal pathogen Staphylococcus aureus. This gram-positive bacterium often chronically infects the bones, skin
and soft tissues, and cardiovascular and respiratory epithelia of diverse mammals (Lowy, 1998; Peton and Le Loir, 2014), frequently undergoing metabolic adaptation along the way to become small-colony variants (SCVs). SCVs grow slowly in axenic cultures, tend to be defective for producing specific respiratory cofactors, and have reduced virulence (Proctor et al., 2006). In many cases, such as in the respiratory tract, S. aureus infects in the presence of other bacteria with which it could interact (Boase et al., 2013; Harris et al., 2007). As shown by at least one experimental coculture system—with S. aureus and the gram-negative bacterium Pseudomonas aeruginosa—these interactions can select for S. aureus SCVs, which are in turn resistant to killing by certain antibiotics (Hoffman et al., 2006). Work with this and other simple, binary systems show that microbial community interactions can select for antibioticresistant variants with decreased fitness in vitro. Interestingly, SCVs have been
428 Cell Host & Microbe 16, October 8, 2014 ª2014 Elsevier Inc.
associated with persistent infections and worse disease outcomes despite their impairment in pure culture (Besier et al., 2007; Wolter et al., 2013), possibly suggesting that their behaviors in vitro may not accurately reflect those in vivo. Based on observations of this type, Hammer et al. (2014) wondered whether the reduction in fitness observed for SCVs in axenic cultures might be an artifact of removing these variants from naturally diverse microbial environments. The researchers hypothesized that community interactions between genetically distinct bacteria would offset the fitness costs associated with mutations that confer antibiotic resistance, such as mutations carried by SCVs. To test this hypothesis, Hammer et al. (2014) cocultured S. aureus SCVs with different types of causative mutations and measured the effects of coculture on growth, intercellular signaling activity, toxin production, cytotoxicity, antibiotic susceptibility, and ultimately in vivo pathogenesis in a murine bone infection model. These experiments showed that genetically distinct SCVs, each defective for in vitro and in vivo growth as single-isolate cultures, can grow, infect, damage bone tissue, and otherwise behave with nearly wild-type characteristics when cocultured with each other (Figure 1). By using SCVs with different defined mutations, and with the clever use of exogenously added reagents, Hammer et al. (2014) provide compelling evidence that the mechanism of this intraspecies ‘‘complementation’’ involves the intercellular exchange of metabolites, particularly heme. Therefore, the depiction of the behaviors of individual isolates provided by traditional, axenic culture techniques may not reflect their behaviors in a community context. The SCV cocultures used in this study may more accurately reflect the in vivo behaviors of infectious microbial communities containing these variants than do single-isolate cultures. However, while both types of laboratory-derived S. aureus SCVs used in this study have been isolated from chronic infections (such as from airway secretions of people with cystic fibrosis [CF]) (Proctor et al., 2006), there had been little evidence in the literature to date for their coisolation from the same specimens. Therefore, while these experiments provide convincing and provocative evidence for
Cell Host & Microbe
Previews Hammer et al. (2014)’s hypothesis in principle, their clinical relevance remained to be demonstrated. The researchers began to address this issue with their next experiments, in which Hammer et al. (2014) applied similar methods to study the effects of coculturing slow-growing isolates of different bacterial species (including S. aureus SCVs) from CF patients’ respiratory secretions. Intriguingly, these experiments revealed enhancement of growth by many of these isolates when cocultured. The bacterial pairs that grew better together than separately included not only different S. aureus SCVs isolated from the same patient, but also those same SCVs with isolates of nonstaphylococcal species, some of which are considered ‘‘upper respiratory flora’’ and are often not identified further by routine clinical diagnostic procedures. As with the paired S. aureus SCVs, Hammer et al. (2014) provide evidence, albeit less complete than with the engineered SCVs, that these growth-promoting interactions are due to the intercellular exchange of metabolites such as heme, as well as the respiratory cofactor menaquinone. Therefore, both intra- and interspecies community interactions can impact the fitness of bacteria, further underscoring how axenic cultures often incompletely reflect the behaviors of complex, chronic infections.
These results strongly suggest the benefits of a communal lifestyle for slow-growing organisms. However, as is often the case with provocative findings, they raise several as yet unanswered but compelling questions for future experiments. For example, do all members involved in these types of interactions benefit? What are the molecular mechanisms and dynamics of intercellular metabolite transfer; are export and uptake active or passive? What happens when community members who do not grow slowly, and who may therefore be less likely to benefit from metabolite exchange, are involved? How does the host influence these infectious community behaviors? And, perhaps the most daunting, how do we identify and model the interactions between members of a community larger than two constituents? As we struggle with these mechanistic details, an even more practical question emerges: how should we address the shortcomings of traditional, axenic cultures, and how can we better diagnose, study, and treat chronic infections and the diverse microbial communities that cause them? At present, such improved approaches remain to be developed or successfully tested. However, the results of studies such as this one, given the high global burden of chronic, persistent
infections, provide ample rationale for such efforts. REFERENCES Besier, S., Smaczny, C., von Mallinckrodt, C., Krahl, A., Ackermann, H., Brade, V., and Wichelhaus, T.A. (2007). J. Clin. Microbiol. 45, 168–172. Boase, S., Foreman, A., Cleland, E., Tan, L., Melton-Kreft, R., Pant, H., Hu, F.Z., Ehrlich, G.D., and Wormald, P.J. (2013). BMC Infect. Dis. 13, 210. Fredericks, D.N., and Relman, D.A. (1996). Clin. Microbiol. Rev. 9, 18–33. Hammer, N.D., Cassat, J.E., Noto, M.J., Lojek, L.J., Chadha, A.D., Schmitz, J.E., Creech, C.B., and Skaar, E.P. (2014). Cell Host Microbe 16, this issue, 531–537. Harris, J.K., De Groote, M.A., Sagel, S.D., Zemanick, E.T., Kapsner, R., Penvari, C., Kaess, H., Deterding, R.R., Accurso, F.J., and Pace, N.R. (2007). Proc. Natl. Acad. Sci. USA 104, 20529–20533. Hoffman, L.R., De´ziel, E., D’Argenio, D.A., Le´pine, F., Emerson, J., McNamara, S., Gibson, R.L., Ramsey, B.W., and Miller, S.I. (2006). Proc. Natl. Acad. Sci. USA 103, 19890–19895. Lowy, F.D. (1998). N. Engl. J. Med. 339, 520–532. Peton, V., and Le Loir, Y. (2014). Infect. Genet. Evol. 21, 602–615. Proctor, R.A., von Eiff, C., Kahl, B.C., Becker, K., McNamara, P., Herrmann, M., and Peters, G. (2006). Nat. Rev. Microbiol. 4, 295–305. Wolter, D.J., Emerson, J.C., McNamara, S., Buccat, A.M., Qin, X., Cochrane, E., Houston, L.S., Rogers, G.B., Marsh, P., Prehar, K., et al. (2013). Clin. Infect. Dis. 57, 384–391.
Cell Host & Microbe 16, October 8, 2014 ª2014 Elsevier Inc. 429
Previews occurs on a variety of cell types and is connected to multiple signaling networks. Restoring community diversity with obligate anaerobes, such as Barnsiella, Bacteroides, or Ruminococcus, could also be worthwhile. The findings of Pham and colleagues, however, suggest a direct and simpler approach: administration of fucosylated oligosaccharides, such as 20 -fucosyllactose, which may do much to restore community complexity and promote homeostasis. ACKNOWLEDGMENTS The authors wish to thank Jose´ T. Saavedra for assistance in developing the figure that accompanies this article. For work on enterococcal colonization, virulence, and antibiotic resistance, the authors gratefully acknowledge the support of
NIH grants AI072360 and AI108710, the Harvardwide Program on Antibiotic Resistance AI083214, and fellowship support from Grant AI109855 (to D.V.T.).
REFERENCES Deatherage Kaiser, B.L., Li, J., Sanford, J.A., Kim, Y.M., Kronewitter, S.R., Jones, M.B., Peterson, C.T., Peterson, S.N., Frank, B.C., Purvine, S.O., et al. (2013). PLoS ONE 8, e67155. Goto, Y., Obata, T., Kunisawa, J., Sato, S., Ivanov, I.I., Lamichhane, A., Takeyama, N., Kamioka, M., Sakamoto, M., Matsuki, T., et al. (2014). Science 345, 1254009. Jernberg, C., Lo¨fmark, S., Edlund, C., and Jansson, J.K. (2007). ISME J. 1, 56–66. Kinnebrew, M.A., Ubeda, C., Zenewicz, L.A., Smith, N., Flavell, R.A., and Pamer, E.G. (2010). J. Infect. Dis. 201, 534–543.
Lo Vecchio, A., and Cohen, M.B. (2014). Curr. Opin. Gastroenterol. 30, 47–53. Pham, T.A.N., Clare, S., Goulding, D., Arasteh, J.M., Stares, M.D., Browne, H.P., Keane, J.A., Page, A.J., Kumasaka, N., Kane, L., et al. (2014). Cell Host Microbe 16, this issue, 504–516. Shogan, B.D., Smith, D.P., Christley, S., Gilbert, J.A., Zaborina, O., and Alverdy, J.C. (2014). Microbiome 2, 35. Turnbaugh, P.J., Ridaura, V.K., Faith, J.J., Rey, F.E., Knight, R., and Gordon, J.I. (2009). Sci. Transl. Med. 1, 6ra14. Ubeda, C., Taur, Y., Jenq, R.R., Equinda, M.J., Son, T., Samstein, M., Viale, A., Socci, N.D., van den Brink, M.R., Kamboj, M., and Pamer, E.G. (2010). J. Clin. Invest. 120, 4332–4341. Van Tyne, D., and Gilmore, M.S. (2014). Annu. Rev. Microbiol. 68, 337–356.
Strength in Diversity Daniel J. Wolter1 and Lucas R. Hoffman1,2,* 1Department
of Pediatrics of Microbiology University of Washington, Seattle, Seattle, WA 98195, USA *Correspondence: [email protected] http://dx.doi.org/10.1016/j.chom.2014.09.018 2Department
In this issue of Cell Host & Microbe, Hammer et al. (2014) show that distinct, slow-growing bacteria have better in vitro and in vivo growth and virulence when cocultured than in isolation. They provide evidence that the observed inter- and intraspecies ‘‘complementation’’ involves the intercellular exchange of metabolites. Laboratory culture is a mainstay of medical bacteriology. The ability to grow a pathogen in pure culture is the basis for one of Koch’s postulates—the criteria presented by Robert Koch in 1890 to establish relationships between individual pathogens and disease (Fredericks and Relman, 1996). Since then, much of medical microbiology, including the identification of infecting bacteria, testing their in vitro susceptibilities to antibiotics, and probing myriad other behaviors, relies on laboratory growth in pure, single-species (i.e., ‘‘axenic’’) cultures. These methods have been refined for well over a century, resulting in a dizzying array of media and techniques, and allowing for everimproved capacity for bacterial cultivation, identification, and discrimination.
The usefulness of axenic cultures is undeniable; these techniques have been invaluable in diagnosing and choosing treatments for innumerable infections. These methods have also proven to be incredibly powerful for investigating the inner workings of specific pathogens, including their growth requirements, their patterns of gene and protein expression, and even their virulence. Axenic cultures are particularly effective when applied to acute infections by a single bacterial species. In such cases, it can be relatively straightforward to pinpoint the pathogen in question and to identify antibiotics that should be able to eradicate, or at least suppress, the infection. However, there are reasons to question the accuracy of axenic culture techniques
when they are applied to more complex infections. For example, a single-isolate culture may not faithfully reflect the behavior of an organism when multiple microbes are present in infected tissues, or during chronic infections, when even a single species may diversify as it adapts to the varying selective pressures it might encounter in those tissues. Pure cultures of individual isolates from these infections may underestimate the diversity of bacteria—both within and among species—that contribute to a particular disease. Without an inclusive accounting of these infecting microbes, it is difficult to understand the pathogenesis of polymicrobial or chronic infections and to treat them. Specifically, a single bacterial isolate could behave entirely differently
Cell Host & Microbe 16, October 8, 2014 ª2014 Elsevier Inc. 427
Cell Host & Microbe
Previews
In vivo
Inoculum
In vitro
Normal colony
SCV-1
SCV-2
Virulence Factor
Metabolites
Healthy Bone
Figure 1. Impact of Community Interactions on the Growth and Virulence of Slow-Growing Bacteria A metabolically deficient small colony variant (SCV, center) has significantly impaired growth and decreased virulence factor production in vitro and lower bacterial burdens and virulence in an osteomyelitis infection model (indicated by gray lesions in bone) compared to bacteria with normal growth (left). However, cocultures of two SCVs (SCV-1 and SCV-2, right) with distinct metabolic deficiencies results in enhanced growth and virulence in vitro and in vivo, probably through the intercellular transfer of metabolites.
when growing in pure culture than when in communities comprised either of other variants of its own species or with other species. Interactions between different microbes in such communities could alter each member’s metabolism, virulence, and susceptibilities to antibiotics. Many chronic infections persist despite treatment with antibiotics that effectively kill each identified pathogen isolated in pure cultures. Therefore, accurately describing and modeling community behaviors could help to identify improved treatments for chronic infections, which account for tremendous morbidity and mortality. The work described in this issue of Cell Host & Microbe by Hammer et al. (2014) provides a provocative example of bacterial community interactions at both intra- and interspecies levels that may be relevant for many chronic infections, particularly those that include the common animal pathogen Staphylococcus aureus. This gram-positive bacterium often chronically infects the bones, skin
and soft tissues, and cardiovascular and respiratory epithelia of diverse mammals (Lowy, 1998; Peton and Le Loir, 2014), frequently undergoing metabolic adaptation along the way to become small-colony variants (SCVs). SCVs grow slowly in axenic cultures, tend to be defective for producing specific respiratory cofactors, and have reduced virulence (Proctor et al., 2006). In many cases, such as in the respiratory tract, S. aureus infects in the presence of other bacteria with which it could interact (Boase et al., 2013; Harris et al., 2007). As shown by at least one experimental coculture system—with S. aureus and the gram-negative bacterium Pseudomonas aeruginosa—these interactions can select for S. aureus SCVs, which are in turn resistant to killing by certain antibiotics (Hoffman et al., 2006). Work with this and other simple, binary systems show that microbial community interactions can select for antibioticresistant variants with decreased fitness in vitro. Interestingly, SCVs have been
428 Cell Host & Microbe 16, October 8, 2014 ª2014 Elsevier Inc.
associated with persistent infections and worse disease outcomes despite their impairment in pure culture (Besier et al., 2007; Wolter et al., 2013), possibly suggesting that their behaviors in vitro may not accurately reflect those in vivo. Based on observations of this type, Hammer et al. (2014) wondered whether the reduction in fitness observed for SCVs in axenic cultures might be an artifact of removing these variants from naturally diverse microbial environments. The researchers hypothesized that community interactions between genetically distinct bacteria would offset the fitness costs associated with mutations that confer antibiotic resistance, such as mutations carried by SCVs. To test this hypothesis, Hammer et al. (2014) cocultured S. aureus SCVs with different types of causative mutations and measured the effects of coculture on growth, intercellular signaling activity, toxin production, cytotoxicity, antibiotic susceptibility, and ultimately in vivo pathogenesis in a murine bone infection model. These experiments showed that genetically distinct SCVs, each defective for in vitro and in vivo growth as single-isolate cultures, can grow, infect, damage bone tissue, and otherwise behave with nearly wild-type characteristics when cocultured with each other (Figure 1). By using SCVs with different defined mutations, and with the clever use of exogenously added reagents, Hammer et al. (2014) provide compelling evidence that the mechanism of this intraspecies ‘‘complementation’’ involves the intercellular exchange of metabolites, particularly heme. Therefore, the depiction of the behaviors of individual isolates provided by traditional, axenic culture techniques may not reflect their behaviors in a community context. The SCV cocultures used in this study may more accurately reflect the in vivo behaviors of infectious microbial communities containing these variants than do single-isolate cultures. However, while both types of laboratory-derived S. aureus SCVs used in this study have been isolated from chronic infections (such as from airway secretions of people with cystic fibrosis [CF]) (Proctor et al., 2006), there had been little evidence in the literature to date for their coisolation from the same specimens. Therefore, while these experiments provide convincing and provocative evidence for
Cell Host & Microbe
Previews Hammer et al. (2014)’s hypothesis in principle, their clinical relevance remained to be demonstrated. The researchers began to address this issue with their next experiments, in which Hammer et al. (2014) applied similar methods to study the effects of coculturing slow-growing isolates of different bacterial species (including S. aureus SCVs) from CF patients’ respiratory secretions. Intriguingly, these experiments revealed enhancement of growth by many of these isolates when cocultured. The bacterial pairs that grew better together than separately included not only different S. aureus SCVs isolated from the same patient, but also those same SCVs with isolates of nonstaphylococcal species, some of which are considered ‘‘upper respiratory flora’’ and are often not identified further by routine clinical diagnostic procedures. As with the paired S. aureus SCVs, Hammer et al. (2014) provide evidence, albeit less complete than with the engineered SCVs, that these growth-promoting interactions are due to the intercellular exchange of metabolites such as heme, as well as the respiratory cofactor menaquinone. Therefore, both intra- and interspecies community interactions can impact the fitness of bacteria, further underscoring how axenic cultures often incompletely reflect the behaviors of complex, chronic infections.
These results strongly suggest the benefits of a communal lifestyle for slow-growing organisms. However, as is often the case with provocative findings, they raise several as yet unanswered but compelling questions for future experiments. For example, do all members involved in these types of interactions benefit? What are the molecular mechanisms and dynamics of intercellular metabolite transfer; are export and uptake active or passive? What happens when community members who do not grow slowly, and who may therefore be less likely to benefit from metabolite exchange, are involved? How does the host influence these infectious community behaviors? And, perhaps the most daunting, how do we identify and model the interactions between members of a community larger than two constituents? As we struggle with these mechanistic details, an even more practical question emerges: how should we address the shortcomings of traditional, axenic cultures, and how can we better diagnose, study, and treat chronic infections and the diverse microbial communities that cause them? At present, such improved approaches remain to be developed or successfully tested. However, the results of studies such as this one, given the high global burden of chronic, persistent
infections, provide ample rationale for such efforts. REFERENCES Besier, S., Smaczny, C., von Mallinckrodt, C., Krahl, A., Ackermann, H., Brade, V., and Wichelhaus, T.A. (2007). J. Clin. Microbiol. 45, 168–172. Boase, S., Foreman, A., Cleland, E., Tan, L., Melton-Kreft, R., Pant, H., Hu, F.Z., Ehrlich, G.D., and Wormald, P.J. (2013). BMC Infect. Dis. 13, 210. Fredericks, D.N., and Relman, D.A. (1996). Clin. Microbiol. Rev. 9, 18–33. Hammer, N.D., Cassat, J.E., Noto, M.J., Lojek, L.J., Chadha, A.D., Schmitz, J.E., Creech, C.B., and Skaar, E.P. (2014). Cell Host Microbe 16, this issue, 531–537. Harris, J.K., De Groote, M.A., Sagel, S.D., Zemanick, E.T., Kapsner, R., Penvari, C., Kaess, H., Deterding, R.R., Accurso, F.J., and Pace, N.R. (2007). Proc. Natl. Acad. Sci. USA 104, 20529–20533. Hoffman, L.R., De´ziel, E., D’Argenio, D.A., Le´pine, F., Emerson, J., McNamara, S., Gibson, R.L., Ramsey, B.W., and Miller, S.I. (2006). Proc. Natl. Acad. Sci. USA 103, 19890–19895. Lowy, F.D. (1998). N. Engl. J. Med. 339, 520–532. Peton, V., and Le Loir, Y. (2014). Infect. Genet. Evol. 21, 602–615. Proctor, R.A., von Eiff, C., Kahl, B.C., Becker, K., McNamara, P., Herrmann, M., and Peters, G. (2006). Nat. Rev. Microbiol. 4, 295–305. Wolter, D.J., Emerson, J.C., McNamara, S., Buccat, A.M., Qin, X., Cochrane, E., Houston, L.S., Rogers, G.B., Marsh, P., Prehar, K., et al. (2013). Clin. Infect. Dis. 57, 384–391.
Cell Host & Microbe 16, October 8, 2014 ª2014 Elsevier Inc. 429
