A bacterium’s enemy isn’t your friend

The bacterium Staphylococcus aureus is a leading cause of hard-to-treat human infections. It now seems that, if the bacterium is infected by a virus, a viral enzyme helps the microbe to evade detection by the immune system.
Michael S. Gilmore is in the Departments of Ophthalmology and of Microbiology and Immunobiology, Harvard Medical School, Massachusetts Eye and Ear Infirmary, Boston, Massachusetts 02114, USA.

Search for this author in:

Ona K. Miller is in the Departments of Ophthalmology and of Microbiology and Immunobiology, Harvard Medical School, Massachusetts Eye and Ear Infirmary, Boston, Massachusetts 02114, USA.

Search for this author in:

Microorganisms thrive on our body’s surfaces. The species present are not just a random assembly; rather, they are a community of organisms that are particularly well adapted to the local conditions of temperature, moisture, nutrient availability and host defences1. Staphylococcus aureus is one of our most common bacterial residents. It usually lives in nasal, respiratory and reproductive tissues without causing disease, yet, unlike many other resident bacteria, S. aureus has the capacity to give rise to a potentially deadly infection2.

During the past 50 years2, the resistance of S. aureus to antibiotics has become an increasing problem, and strains of the bacterium termed methicillin-resistant S. aureus (MRSA), which are resistant to treatment with the antibiotic methicillin and other methylated penicillin-based antibiotics, cause both hospital- and community-acquired infections around the globe. Writing in Nature, Gerlach et al.3 describe a previously unknown mechanism whereby viruses influence whether MRSA is recognized by the immune system, shedding light on a process that might tip the balance in determining whether this bacterium will be harmless or disease-causing.

Staphylococcus aureus belongs to the Gram-positive group of bacteria, and has been described as existing on the borderline between being a normal human microbial resident and a disease-causing organism4. This bacterium seems to have the capacity to probe for signs of host weakness, such as reduced immune defences caused by disease. When this is detected, the bacterium can increase its population to a level that can cause the death of the host5. Factors that regulate host–microbial interactions are complex, and in addition to host defences, such inter-actions can be influenced by the presence or absence of other bacteria6. Gerlach and colleagues report that viruses can also be part of the mix that influences host–microbial interactions in the context of MRSA.

In Gram-positive bacteria, the cell wall contains polymers known as wall teichoic acids (WTA), which are made up of ribitol phosphate or glycerol phosphate molecules and can constitute up to half of the cell-wall mass6. Unlike the other main cell-wall component, peptidoglycan, which forms a porous and comparatively insoluble meshwork, WTA form a highly hydrated, gel-like material that fills much of the space between peptido-glycan strands. WTA provide a soluble matrix through which all substances pass before reaching the bacterial cell membrane, and therefore affect bacterial access to ions, nutrients, proteins and antibiotics7. In S. aureus, WTA are composed of units of d-ribitol phosphate, which are crosslinked to the peptidoglycan (Fig. 1). WTA function is tuned by attachments of the amino acid d-alanine and of N-acetylglucosamine (GlcNAc)7 molecules to the ribitol-phosphate polymer.

Figure 1 | Viral infection of a bacterium can alter the host’s immune response to the microbe.a, The bacterium Staphylococcus aureus is a resident of the human body. Its outer surface is coated with layers of the polymer peptidoglycan crosslinked to wall teichoic acids (WTA) — polymers of ribitol phosphate molecules. The bacterial enzyme TarS modifies WTA, generating a form that has the molecule GlcNAc attached at carbon atoms in the C4 position in the ribitol. Human antibodies against S. aureus often target WTA. b, Gerlach et al.3 report that some antibiotic-resistant strains of S. aureus, which are associated with difficult-to-treat infections, have been infected with a virus called a phage. The phage DNA encodes an enzyme called TarP that attaches GlcNAc to WTA at carbon atoms in the C3 position of ribitol rather than in the normal C4 position. In studies of mice and of human cells, the authors find that these TarP-modified WTA trigger an immune response that is weaker than the response against TarS-modified WTA.

Gerlach and colleagues decided to investigate whether bacterial evasion of immune-system defences might be one of the reasons that MRSA strains can reach high enough bacterial numbers to cause disease. The authors studied the genome sequences of MRSA strains to identify genes encoding enzymes that modify WTA. This revealed that some MRSA strains encode an enzyme called TarP that catalyses the addition of GlcNAc to d-ribitol phosphate at a particular carbon atom (known as C3) in the ribitol. Normally, GlcNAc is added at a different position, the C4 carbon, by the action of a related enzyme called TarS.

Surprisingly, the TarP-encoding sequence is of viral origin, and is found in S. aureus as a result of infection by a bacterial virus called a phage. TarP is dominant over its bacterial counterpart, TarS — that is, if both enzymes are present, the GlcNAc linkage is made on the C3 carbon of ribitol, rather than on the C4 carbon. S. aureus is normally held in check because the immune system has the ability to detect it. However, the authors found that, in mice, the form of WTA made by TarP action is less likely to trigger an immune response than is the form of WTA generated by TarS.

This virus-mediated change to the S. aureus cell wall reported by Gerlach and colleagues is important for two reasons. First, it highlights the fact that a fragile truce between host and resident microbe can be affected by the intervention of a third party with its own vested interests. Second, at a time that some8 have called the beginning of a ‘post-antibiotic era’ — given the rise in antibiotic-resistant bacteria and the limited development of new antibiotics reaching the clinic — there is a pressing need to develop new strategies to manage infection.

We are now at the dawn of a clinical era in which the goal will be to precisely manage human and microbial interactions to promote health and limit disease. Antibiotics will continue to have a key role, as undoubtedly will other approaches, including the replacement of a person’s gut microbes using techniques such as faecal transplants, or the use of phage-mediated elimination of undesirable microbes. Determining the best approach will be helped by the development of new diagnostic tools and a clearer understanding of the nature of human and microbial interactions. If deciding whether to take an approach based on a vaccine or possibly using phage treatments in the future, key considerations will include knowing how a bacterium’s susceptibility to phage infection varies, and determining whether the presence of phage DNA in a bacterial genome affects the dynamics between human cells and the microbes that colonize the body.

We do not yet know whether the phage-mediated alteration of WTA described by Gerlach and colleagues affects where the bacteria reside on the body or the number of bacterial cells present. We also lack a clear understanding of whether the anti-staphylo-coccal WTA-targeting antibodies that most people have, and which do not seem to be protective against infection in immune-deficient individuals, are a ‘distraction’ imposed by the presence of S. aureus. This distraction would keep the immune system busy generating antibodies that end up in ineffective locations such as the bloodstream and do not eliminate the microbe. Alternatively, this low-level immune warfare could represent a stalemate between the host and its resident bacteria.

It is clear that phage-encoded TarP changes the immune reactivity of S. aureus. In a model system of human immune cells grown in vitro, the authors found that S. aureus strains encoding TarP were cleared from the system less effectively than were S. aureus strains that lacked TarP. Similar phage-mediated changes in a bacterial cell surface that alter antibody recognition of the microbe have been reported9 for the disease-causing Gram-negative bacterium Shigella flexneri.

Gerlach and colleagues’ work, as well as that of others in this area, demonstrates that the balance between host and microbes is a dynamic one. The discovery that phages can have a role in tipping the delicate balance between S. aureus colonization and infection might one day affect the choice of approaches for treating MRSA infections.

Nature 563, 637-638 (2018)

doi: 10.1038/d41586-018-07414-w
Nature Briefing

Sign up for the daily Nature Briefing email newsletter

Stay up to date with what matters in science and why, handpicked from Nature and other publications worldwide.

Sign Up


  1. 1.

    Proctor, D. M. & Relman, D. A. Cell Host Microbe 21, 421–432 (2017).

  2. 2.

    Lee, A. S. et al. Nature Rev. Dis. Primers 4, 18033 (2018).

  3. 3.

    Gerlach, D. et al. Nature (2018).

  4. 4.

    Camargo, I. L. B. C. & Gilmore, M. S. J. Bacteriol. 190, 2253–2256 (2008).

  5. 5.

    Guerra, F. E., Borgogna, T. R., Patel, D. M., Sward, E. W. & Voyich, J. M. Front. Cell Infect. Microbiol. 7, 286 (2017).

  6. 6.

    Boldock, E. et al. Nature Microbiol. 3, 881–890 (2018).

  7. 7.

    Neuhaus, F. C. & Badilly, J. Microbiol. Mol. Biol. Rev. 67, 686–723 (2003).

  8. 8.

    Zucca, M. & Savoia, D. Int. J. Biomed. Sci. 6, 77–86 (2010).

  9. 9.

    Mavris, M., Manning, P. A. & Morona, R. Mol. Microbiol. 26, 939–950 (1997).

Download references