Fungus produces a toxic surprise

A protein fragment released by filaments of the fungus Candida albicans destroys host cells. This is the first demonstration that human fungal pathogens other than moulds can release toxic peptides. See Article p.64

The fungus Candida albicans, a common cause of infection in mucosal tissues, forms long filaments called hyphae, comprised of tubular cells, that are required for virulence in animals1. Hyphal-associated adhesion proteins bind to host tissue, which is then degraded by hydrolytic enzymes. However, until now, no hyphal toxin that damages host cells has been identified. Because of this, the fungus is considered to be an 'accidental' pathogen that benignly inhabits our mucosal tissues, causing tissue damage by happenstance rather than design. That view must now change drastically. On page 64 of this issue, Moyes et al.2 reveal a mechanism through which hyphae actively wage war on our cells — the production of a toxin that the authors call Candidalysin.

Interactions between C. albicans and epithelial cells, which line the body's cavities, occur during the early stages of mucosal infections such as thrush and vaginitis. Hyphae elicit3 several epithelial-cell responses, including the production of signalling molecules called cytokines that recruit cells of the immune system to defend tissues, and loss of cell integrity through cell-membrane deterioration. Moyes et al. discovered that a strain of C. albicans in which the gene ECE1 was mutated could not elicit epithelial-cell responses, despite growing apparently normal hyphae. Moreover, the authors validated these tissue-culture observations in vivo — the ECE1 mutant was unable to reliably infect mucosa in a zebrafish swimbladder model and a mouse model of thrush.

ECE1 was one of the first genes to be identified in hyphal-specific expression screens more than 20 years ago4, yet until now it has been one of the most poorly understood genes in C. albicans. In fact, ECE1 is among the most highly expressed genes in hyphae5, but its role has not previously been investigated thoroughly because mutants show no defects in hyphal morphology or cell proliferation4,6. Thus, the function of the Ece1 protein has remained a puzzle.

How does Ece1 promote epithelial-cell responses? The protein's amino-acid sequence suggests that it is secreted from hyphae as a group of eight short protein fragments, or peptides, and so would be well positioned to interact with host cells. Moyes and colleagues confirmed that all eight Ece1 peptides are secreted from hyphae. Analysis of synthetic versions of each peptide revealed that one, Ece1-III, elicits the same responses from epithelial cells as do hyphae. Moreover, precise deletion of the genetic region that codes for only Ece1-III created a mutant C. albicans that secreted the remaining seven peptides, but did not elicit epithelial-cell responses or cause mucosal disease in animal models. These results clearly demonstrate that Ece1-III mediates the pathogenic activity associated with ECE1.

By what mechanism does Ece1-III exert this activity? Certain chemical and structural features indicate that Ece1-III might function like peptide toxins, such as the bee-venom toxin melittin. Indeed, the authors show that the peptide causes rapid and transient permeabilization of artificial cell membranes in vitro. These activities are enhanced in the presence of cholesterol, a component of animal — but not fungal — membranes. The researchers conclude that Ece1-III acts as a peptide toxin, which they name Candidalysin.

Moyes and colleagues' study establishes that C. albicans hyphae evolved to damage host cells. When combined with our knowledge of hyphal adhesin proteins and enzymes, a simple program of tissue destruction emerges (Fig. 1). First, the hyphal-specific adhesin Hwp1 attaches to mucosal surfaces7. Second, the hyphal-specific invasion protein Als3, acting with the protein Ssa1, binds to receptors on the surface of the host cell, promoting engulfment of the hypha by the host cell8. Finally, Candidalysin accumulates in the invasion pocket around the hypha, attacking the host's cholesterol-containing membrane.

Figure 1: A toxic relationship.

The pathogenic fungus Candida albicans infects its host by forming filamentous structures called hyphae. Proteins on the hyphal surface — the adhesin Hwp1, the invasin Als3 and its partner Ssa1 — make contact with the host cell directly or through receptor proteins to promote adhesion and engulfment of the hypha by the host. Moyes et al.2 report that the protein Ece1 is secreted from the hypha as eight short peptides. One of these, Candidalysin, acts as a toxin that accumulates in the invasion pocket and attacks the host-cell membrane, leading to membrane permeabilization and the induction of host defences.

This attack leads to membrane permeabilization, leakage of cell contents and a defensive cytokine response, which serves to limit the size of the C. albicans population in healthy individuals. However, impaired defences in people with conditions such as AIDS, diabetes and some cancers permit C. albicans growth and consequent disease.

Every study raises fresh questions, and this one is no exception. For instance, why did a previous analysis6 find that ECE1 was not necessary for invasive infection? The study in question was rigorous and efficiently tested many C. albicans genes using a mixture of mutant strains. One possible explanation, given that Ece1 functions extracellularly, is that an ECE1 mutant was rescued by neighbouring cells that did express the gene.

Might other human pathogens of the genus Candida also produce peptide toxins? The ECE1 gene is found in only a few other genomes. However, other aspects of ECE1 might serve as good guides in the hunt for prospective toxin genes. For example, a small, highly expressed gene product with predicted chemical and structural features reminiscent of Ece1 would be an excellent candidate for a peptide toxin.

It remains unclear whether the seven Ece1 peptides that are secreted along with Candidalysin have a role in host interaction, or are just hitchhikers. Given that C. albicans spends most of its time as a commensal rather than a pathogenic fungus, it seems likely that Candidalysin or the other Ece1 peptides have roles in maintaining normal fungus–host interactions. Finally, Ece1 seems to be a promising drug target — inhibitors of its production or activity might be effective antifungal agents. Such an advance would be welcome, because there is currently a dearth of antifungal drugs, and the rates of disease and mortality associated with C. albicans infections are high9.Footnote 1


  1. 1.

    See all news & views


  1. 1

    Mayer, F. L., Wilson, D. & Hube, B. Virulence 4, 119–128 (2013).

    Article  Google Scholar 

  2. 2

    Moyes, D. L. et al. Nature 532, 64–68 (2016).

    ADS  CAS  Article  Google Scholar 

  3. 3

    Naglik, J. R., Richardson, J. P. & Moyes, D. L. PLoS Pathog. 10, e1004257 (2014).

    Article  Google Scholar 

  4. 4

    Birse, C. E., Irwin, M. Y., Fonzi, W. A. & Sypherd, P. S. Infect. Immun. 61, 3648–3655 (1993).

    CAS  PubMed  PubMed Central  Google Scholar 

  5. 5

    Bruno, V. M. et al. Genome Res. 20, 1451–1458 (2010).

    CAS  Article  Google Scholar 

  6. 6

    Noble, S. M., French, S., Kohn, L. A., Chen, V. & Johnson, A. D. Nature Genet. 42, 590–598 (2010).

    CAS  Article  Google Scholar 

  7. 7

    Staab, J. F., Bradway, S. D., Fidel, P. L. & Sundstrom, P. Science 283, 1535–1538 (1999).

    ADS  CAS  Article  Google Scholar 

  8. 8

    Filler, S. G. Trends Microbiol. 21, 389–396 (2013).

    CAS  Article  Google Scholar 

  9. 9

    Brown, G. D. et al. Sci. Transl. Med. 4, 165rv113 (2012).

    Article  Google Scholar 

Download references

Author information



Corresponding author

Correspondence to Aaron P. Mitchell.

Related links

Related links

Related links in Nature Research

Microbial genetics: Love the one you're with

Fungal biology: Multiple mating strategies

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Mitchell, A. Fungus produces a toxic surprise. Nature 532, 41–42 (2016). https://doi.org/10.1038/nature17319

Download citation

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.