Cell biology

A table for two

Autophagy, the process of cellular self-cannibalism, comes in various forms. It now emerges that two of these — mitophagy and xenophagy — share a common initiator protein, Parkin. See Article p.512

Infectious disease is not the inevitable consequence of exposure to a pathogen. Host factors have a crucial role in determining the outcome of such exposure, yet much remains unknown about how individuals vary in their capacity to resist pathogens. In this regard, genomic studies offer a tenable approach to identifying pathways involved in microbial handling. An intriguing example of how genomic findings can guide the mechanistic understanding of complex biological systems is Manzanillo and colleagues' paper1 published on page 512 of this issue. The study concerns the mechanism by which cells remove the human pathogen Mycobacterium tuberculosis through a process called autophagyFootnote 1.

Genomic studies provide lists of genes, both expected and unexpected. Whereas expected genes serve to reinforce existing models, it is the set of genes with no apparent link to mechanism that is the most daunting. For instance, how can one explain the observation that PARK2, a gene first identified in autosomal recessive early-onset Parkinson's disease, has been identified in genomic studies2,3 as being associated with susceptibility to the infectious diseases leprosy and typhoid fever? Manzanillo et al. provide a functional link. They show that the product of PARK2 — an E3 ubiquitin ligase enzyme called Parkin — has a role in xenophagy, an autophagic pathway that delivers intracellular bacteria for degradation in cellular organelles called lysosomes, and that is related to but distinct from mitophagy, an autophagic pathway that clears damaged cellular organelles known as mitochondria.

Mitochondria are a main source of cellular energy, which is generated in a process that depends on the presence of an electrochemical potential across the inner mitochondrial membrane. Mitochondria that fail to maintain this membrane potential are replaced as part of mitochondrial quality control4. Specifically, the kinase enzyme PINK1 senses dissipation of the inner-membrane potential and accumulates on the outer membrane of dysfunctional mitochondria (Fig. 1a). There, PINK1 is recognized by Parkin, which tags target mitochondrial proteins with the small protein ubiquitin for subsequent degradation by mitophagy5. At least in neurons, Parkin translocation to mitochondria also requires reactive oxygen species (ROS)6.

Figure 1: Parkin and autophagy.

a, The autophagic process of mitophagy is initiated by recruitment of the enzyme Parkin to dysfunctional mitochondria; this is mediated by both the enzyme PINK1 and reactive oxygen species (ROS; not shown). Subsequent activation of Parkin leads to addition of the small molecule ubiquitin to mitochondrial proteins and autophagy of the ubiquitinated mitochondria. This involves the engulfment of these organelles in autophagosomes and the fusion of autophagosomes with lysosomes, where the engulfed mitochondria are destroyed. b, Manzanillo et al.1 find that Parkin is the link between mitophagy and another autophagic process, xenophagy. Xenophagy also depends on ROS and the binding of Parkin to phagosomes containing ingested intracellular pathogens (such as the bacteria Mycobacterium tuberculosis and Salmonella Typhimurium). How Parkin binds to phagosomes is not known. But once bound, Parkin ubiquitinates certain proteins, targeting the phagosome for lysosomal destruction. For clarity, only one of the two double membranes of the autophagosome is shown.

Mutated Parkin from patients with autosomal recessive Parkinson's disease cannot facilitate mitophagy5, suggesting that defective mitophagy contributes to neuronal death and neurodegeneration in this disorder. Unexpectedly, Manzanillo et al. find that Parkin is also involved in the clearance of microbial pathogens, but in a manner that is independent of its mitochondrial function.

The authors investigated the role of Parkin in mouse and human macrophages — immune cells that specialize in engulfing extracellular material such as pathogens by a process known as phagocytosis. To trigger the assembly of the autophagy complex, Parkin ubiquitinates phagosomal vesicles containing M. tuberculosis; the identity of the phagosomal proteins to which ubiquitin binds through its lysine amino-acid residue at position 63 is unknown. Consistent with an essential role for Parkin in delivering M. tuberculosis to lysosomes, deletion of Park2 results in increased bacterial proliferation in infected macrophages in mice, and shorter survival times for these animals.

Autophagy has a key role in the destruction of invading bacteria such as mycobacteria and salmonella7. Manzanillo and colleagues' work adds Parkin recruitment as an essential event in initiating the clearance of these intracellular pathogens by autophagy (Fig. 1b). Intriguingly, the authors find that Parkin mediates the control of infection with different intracellular pathogens (mycobacteria, salmonella and listeria) and in different hosts (mice and fruitflies). This suggests an evolutionarily conserved role for Parkin in innate immunity. Yet, despite its broad spectrum of pathogen control, Parkin seems to act on a different cellular compartment from that involved in the direct ubiquitination of cytosolic bacteria that is mediated by LRSAM1 — another E3 ligase involved in the autophagic destruction of Salmonella bacteria8.

Phagocytosis of salmonella by macrophages results in recruitment of the enzyme NOX2 NADPH oxidase to the phagosome membrane and the production of ROS9. Generation of ROS is necessary for recruitment of the autophagy machinery to phagosomes and delivery of their bacterial cargo to the lysosome9. However, the factors that link ROS generation to autophagosome formation are not known. It is reasonable to propose that Parkin may be one such factor, given this enzyme's key role in both xenophagy and mitophagy1,5 and the requirement for ROS in both autophagy of salmonella-containing phagosomes and Parkin-dependent mitophagy in neurons.

The present paper raises further questions. For example, following infection of macrophages with M. tuberculosis, approximately one-third of the ingested bacteria are tagged by Parkin for autophagy. What is the fate of the two-thirds of bacteria that are not tagged, and do they benefit through a form of bacterial altruism? What is the membranous structure surrounding the bacteria that is tagged by Parkin for autophagy? Is it derived from the cell membrane as would be expected, or has it undergone structural and compositional changes to enable Parkin binding? Do bacterium-specific signals contribute to the autophagic response? Active Parkin ubiquitinates many proteins on damaged mitochondria10, and it will be interesting to contrast these with the unknown targets of phagosomal Parkin.

The capacity for functionally dissecting genomic findings has advanced tremendously since the completion of microbial and human genomes a decade or so ago. Manzanillo and colleagues' work reminds us that the unexpected results of genomic studies are often the most exciting.


  1. 1.

    *This article and the paper under discussion1 were published online on 4 September 2013.


  1. 1

    Manzanillo, P. S. et al. Nature 501, 512–516 (2013).

    ADS  CAS  Article  Google Scholar 

  2. 2

    Mira, M. T. et al. Nature 427, 636–640 (2004).

    ADS  CAS  Article  Google Scholar 

  3. 3

    Ali, S. et al. Clin. Exp. Immunol. 144, 425–431 (2006).

    CAS  Article  Google Scholar 

  4. 4

    Vincow, E. S. et al. Proc. Natl Acad. Sci. USA 110, 6400–6405 (2013).

    ADS  CAS  Article  Google Scholar 

  5. 5

    Narendra, D. P. et al. PLoS Biol. 8, e1000298 (2010).

    Article  Google Scholar 

  6. 6

    Joselin, A. P. et al. Hum. Mol. Genet. 21, 4888–4903 (2012).

    CAS  Article  Google Scholar 

  7. 7

    Gutierrez, M. G. et al. Cell 119, 753–766 (2004).

    CAS  Article  Google Scholar 

  8. 8

    Huett, A. et al. Cell Host Microbe 12, 778–790 (2012).

    CAS  Article  Google Scholar 

  9. 9

    Huang, J. et al. Proc. Natl Acad. Sci. USA 106, 6226–6231 (2009).

    ADS  CAS  Article  Google Scholar 

  10. 10

    Sarraf, S. A. et al. Nature 496, 372–376 (2013).

    ADS  CAS  Article  Google Scholar 

Download references

Author information



Corresponding authors

Correspondence to Marcel A. Behr or Erwin Schurr.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Behr, M., Schurr, E. A table for two. Nature 501, 498–499 (2013). https://doi.org/10.1038/nature12555

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.


Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing