Effector Specialization in a Lineage of the Irish Potato Famine Pathogen

See allHide authors and affiliations

Science  31 Jan 2014:
Vol. 343, Issue 6170, pp. 552-555
DOI: 10.1126/science.1246300

The Makings of a Choosy Pathogen

The oomycete Phytophthora infestans is responsible for potato blight. A closely related pathogen afflicts the 4 o'clock flower. To assess why such similar pathogens are restricted to one host or the other, Dong et al. (p. 552; see the Perspective by Coaker) analyzed similar effectors from both pathogens. The results suggest that the host specialization that led to evolutionary divergence depends on reciprocal single–amino acid changes that tailor the pathogen effector to a specific host protease that is being disabled. Thus, small changes can open the door for a pathogen to jump to another species of host and, itself, diversify into another species of pathogen.


Accelerated gene evolution is a hallmark of pathogen adaptation following a host jump. Here, we describe the biochemical basis of adaptation and specialization of a plant pathogen effector after its colonization of a new host. Orthologous protease inhibitor effectors from the Irish potato famine pathogen, Phytophthora infestans, and its sister species, Phytophthora mirabilis, which is responsible for infection of Mirabilis jalapa, are adapted to protease targets unique to their respective host plants. Amino acid polymorphisms in both the inhibitors and their target proteases underpin this biochemical specialization. Our results link effector specialization to diversification and speciation of this plant pathogen.

The potato blight pathogen, Phytophthora infestans, is a recurring threat to world agriculture and food security. This funguslike oomycete traces its origins to Toluca Valley, Mexico, where it naturally infects wild Solanum plants (1). In central Mexico, P. infestans co-occurs with closely related species in a tight phylogenetic clade known as clade 1c. These species evolved through host jumps followed by adaptive specialization on plants belonging to different botanical families (2, 3) (fig. S1). One species, Phytophthora mirabilis, is a pathogen of four-o’clock (Mirabilis jalapa). It split from P. infestans about 1300 years ago (1), and the two species have since specialized on their Solanum and Mirabilis hosts. Adaptive evolution after the host jump has left marks on the genomes of P. infestans and P. mirabilis (3). Comparative genomics analyses revealed signatures of accelerated evolution, structural polymorphisms, and positive selection in genes occurring in repeat-rich genome compartments (3). In total, 345 genes induced within plants show signatures of positive selection between the two sister species (3). These include 82 disease effector genes, rapidly evolving determinants of virulence that act on host target molecules. We lack a molecular framework to explain how plant pathogen effectors adapt and specialize on new hosts, even though this process affects pathogen evolution and diversification (46).

To gain insight into the molecular patterns of host adaptation after host jumps, we selected the cystatinlike protease inhibitor EPIC1, an effector protein of P. infestans that targets extracellular (apoplastic) defense proteases of the Solanum hosts (7, 8). The epiC1 gene and its paralogs epiC2A and epiC2B evolved relatively recently in the P. infestans lineage, most likely as a duplication of the conserved Phytophthora gene epiC3 (7) (Fig. 1). To reconstruct the evolution of these effectors in the clade 1c species, we aligned the epiC gene cluster sequences, performed phylogenetic analyses, and calculated variation in selective pressure across the phylogeny (Fig. 1, fig. S2, and table S1) (9). We detected a signature of positive selection in the branch of PmepiC1, the P. mirabilis ortholog of P. infestans epiC1 [nonsynonymous to synonymous ratio (ω) = 2.52] (Fig. 1B). This is consistent with our hypothesis that PmEPIC1 evolved to adapt to a M. jalapa protease after P. mirabilis diverged from P. infestans.

Fig. 1 Evolution of the epiC1 gene family in Phytophthora clade 1c sister species.

(A) Phylogeny, host range, and the presence or absence of polymorphisms of the epiC1 gene family of Phytophthora clade 1c sister species. The Phytophthora phylogenetic tree was adapted from a previous publication (15). The positions of P. mirabilis and P. ipomoeae as closest to P. infestans remain to be resolved. P. sojae and P. ramorum are species from remote clades that are shown for comparison and as outgroups. The presence or absence of epiC genes in clade 1c is a simplified diagram of fig. S2. Arrows with colors indicate the presence of the epiC1 homolog in genomes. The polymorphism in Phytophthora clade 1c and other Phytophthora species is consistent with the recent emergence of epiC1 and 2 from duplication of the widespread epiC3 gene (7). (B) Selection analysis of the epiC1 family genes in Phytophthora clade 1c. The phylogenetic tree was rooted with conserved P. infestans gene epiC3 using PhyML software (9). Rates of nonsynonymous to synonymous ratio (ω) across branches were investigated with the branch model using the codeml function provided by PAML software (9). Note that the PmEPIC1 branch is the only one with a strong signature of positive selection (ω = 2.52). ψ indicates a pseudogene that has lost its ability to produce the full-length protein.

To test our hypothesis, we first determined the inhibition spectra of the EPIC effectors using DCG-04 protease profiling, a method based on the use of a biotinylated, irreversible protease inhibitor that reacts with the active site cysteine of papainlike proteases in an activity-dependent manner (7, 10). We assayed effector activities against the tomato (Solanum lycopersicum) papainlike cysteine proteases RCR3, PIP1, and C14, which have been previously implicated in pathogen defense (7, 8, 11, 12). RCR3 was inhibited by effectors EPIC1, PipEPIC1, and EPIC2B but not by PmEPIC1 (fig. S3). C14 was inhibited by all four effector proteins. PIP1 was inhibited by EPIC2B but not by the three EPIC1 effectors (fig. S3). The reduced activity of PmEPIC1 on RCR3 is not due to instability, given that this protein remained as stable as EPIC1 after a 2-hour incubation in plant apoplastic extracts (fig. S4).

P. infestans originates from central Mexico, where it naturally infects the wild potato species Solanum demissum, S. verrucosum, and S. stoloniferum (2). To challenge our model with ecologically relevant host proteases, we assayed five RCR3 homologs from these three wild potato species for their interactions with pathogen effectors EPIC1 and PmEPIC1 (fig. S5). For all three species, EPIC1 was a better inhibitor of the potato RCR3 proteases than PmEPIC1 (fig. S6). This confirms that the two orthologous effectors have evolved different biochemical activities on Solanum RCR3 proteases.

To identify the target of PmEPIC1 in M. jalapa, the natural Mexican host of P. mirabilis, we coincubated M. jalapa intercellular fluids with the EPIC1 proteins and analyzed the interactions by coimmunoprecipitation and tandem mass spectrometry (fig. S7). We compared the peptides detected with an expressed sequence tag database generated from P. mirabilis–infected leaves of M. jalapa (tables S2 and S3 and fig. S7). We consistently detected peptides matching two RCR3-related cysteine proteases. We termed the two proteases Mirabilis RCR3-like protease 1 and 2 (MRP1 and MRP2). We recovered the full-length proteases and found that they share ~60% amino acid sequence similarity with Solanum RCR3 and carry common signatures of the papainlike cysteine protease SAG12 subfamily (13) (fig. S8).

We transiently expressed MRP1 and MRP2 proteins as a fusion to the histidine (His) epitope tag in M. jalapa leaves. However, only MRP2-His can be expressed and enriched (fig. S9A). Activity profiling of the MRP2-His protein revealed that it can be labeled with the cysteine protease probe DCG-04 (10), and this labeling can be blocked by preincubation with the irreversible cysteine protease inhibitor E-64, which indicates that MRP2 is an active cysteine protease (fig. S9B).

To independently confirm the interaction, we performed coimmunoprecipitation experiments on enriched MRP2-His incubated with FLAG epitope–tagged EPIC proteins (9). Both PmEPIC1 and EPIC1 coimmunoprecipitated with MRP2-His, and the interaction could be outcompeted by E-64, which suggests that the cystatinlike effectors most likely bind to the active site of the protease (fig. S10). We then determined the extent to which the EPIC proteins inhibit MRP2-His using the DCG-04 activity profiling assay. This revealed that PmEPIC1 is more effective than EPIC1 at inhibiting MRP2-His, results counter to the activity of these effectors on Solanum RCR3 proteases (Fig. 2A). To further investigate these findings, we conducted a time-course DCG-04 labeling experiment. Here too, the effectors showed differential inhibition activities on MRP2-His and S. demissum RCR3dms3-His (Fig. 2B). PmEPIC1 reduced DCG-04 labeling of MRP2-His for up to 60 min but had no noticeable effect on RCR3dms3. EPIC1, in contrast, had opposite effects on DCG-04 labeling of the two proteases, as it showed stronger inhibition of RCR3dms3-His than MRP2-His. These findings reveal an adaptive biochemical phenotype, with the effectors displaying increased inhibition of protease targets from their respective host plants.

Fig. 2 The effectors EPIC1 and PmEPIC1 are stronger inhibitors of proteases from their host plants.

(A) Inhibition activity of EPIC1 and PmEPIC1 toward Solanum RCR3 and M. jalapa MRP2 proteases. Activity profiling assays in the presence or absence of effectors revealed that FLAG-EPIC1, but not FLAG-PmEPIC1, inhibits RCR3lyc and RCR3dms3-His and that FLAG-PmEPIC1 exhibits stronger inhibition of MRP2-His than FLAG-EPIC1 does. Equal amounts of Nicotiana benthamiana apoplastic fluid containing RCR3lyc-His and RCR3dms3-His or MRP2-His enriched from M. jalapa apoplastic fluids were preincubated with excess E-64 (200 μM) or recombinant EPIC1 or PmEPIC1 (4 μM) for 1 hour, followed by DCG-04 (100 nM) labeling for 30 min (9). Biotinylated proteases were detected by streptavidin–horseradish peroxidase (Strep. HRP). α-His, antibody against histidine. Protease activities were quantified, normalized, and presented as means ± SEM. *P < 0.05 (Wilcoxon rank-sum test). (B) Time-course inhibition assay confirms inhibition specificity of EPIC1 and PmEPIC1. Apoplastic fluids containing RCR3dms3-His and enriched MRP2-His proteins were preincubated with an excess of FLAG-EPIC1 and FLAG-PmEPIC1 (4 μM) for 1 hour, followed by addition of DCG-04 (200 nM) to label the proteases. Reactions were stopped at different time points to record inhibition of protease activity over time.

We resurrected (14) the ancestral EPIC1 gene to test the hypothesis that the activity spectrum of EPIC1 has shifted over time. We inferred the sequence for ancestral EPIC1 using the maximum likelihood method, synthesized the corresponding gene, and determined the inhibition spectrum of the resurrected protein (fig. S11) (9). Ancestral EPIC1 displayed activities similar to those of modern EPIC1, i.e., stronger inhibition of RCR3dms3 than MRP2 (fig. S11). This indicates that the ability of PmEPIC1 to inhibit MRP2 and its inability to inhibit Solanum RCR3s were not characteristic of the ancestral protein.

To determine which of the variant amino acids underpin the alterations in specificity, we first took advantage of the crystal structure of tarocystatin in complex with a papain protease (Protein Data Bank ID: 3MIA) (fig. S12A). Structure-based sequence alignments of PmEPIC1 or EPIC1 with tarocystatin suggest that the inhibitors share a similar fold and are likely to bind proteases through three regions, two of which are polymorphic between the orthologous effectors (fig. S12). Next, we modified the proteins to determine which of the polymorphic residues contribute to the difference in biochemical activity. Protease inhibition assays with chimeric proteins and with single-site mutants revealed that the Gln-Arg polymorphism at position 111 is critical for specificity (Fig. 3 and figs. S13 and S14). In particular, EPIC1Q111R, carrying a Gln-to-Arg mutation, most closely recapitulated the function of PmEPIC1, with more effective inhibition of the M. jalapa protease MRP2 and less inhibition of the Solanum proteases RCR3dms3 and RCR3lyc (S. lycopersicum) (Fig. 3). Sequences of epiC1 or PmepiC1 alleles from 26 P. infestans isolates and 9 P. mirabilis isolates indicated that the key Gln or Arg residue is fixed in each population (table S1 and fig. S11).

Fig. 3 EPIC1Q111R mutant can inhibit MRP2 but is impaired in RCR3 inhibition.

(A) Amino acid sequence alignment of EPIC1 and PmEPIC1 highlighting the key residue 111. Residues predicted to bind to cysteine proteases are highlighted in boxes. Gln111/Arg112 is in red. Predicted signal peptides are in gray. (B) Inhibition of EPIC1Q111R toward RCR3lyc, RCR3dms3, and MRP2. Activity profiling assays in the presence or absence of effectors were performed as described in the methods (9). Protease activities were quantified and normalized. Data are presented as means ± SEM. *P < 0.05 (Wilcoxon rank-sum test).

We also investigated which variant amino acids determine specificity in the proteases. Inspection of the tarocystatin-papain complex identified a protease region that interacts with a tarocystatin residue equivalent to the EPIC1 or PmEPIC1 key Gln or Arg residue (fig. S15A). Structure-based sequence alignments of RCR3 or MRP2 with papain indicated that the inhibitor-binding region overlaps with a seven–amino acid region that is polymorphic between RCR3 and MRP2 (fig. S15, B and C). We constructed proteases altered in this region by swapping the entire seven–amino acid domain or by single–amino acid changes. The results revealed the polymorphic residue His148 or Asn147 in RCR3 and Asp152 in MRP2 as a key element of specificity (figs. S16 and S17). Unlike wild-type RCR3 proteases, RCR3H148D and RCR3N147D mutants with a single His to Asp or Asn to Asp mutation could be inhibited by PmEPIC1 (fig. S17). This suggests that effector PmEPIC1 adaptation to host protease MRP2 was in part driven by the occurrence of Asp152 in the M. jalapa protease.

Thus, in this case of oomycete infection of potato and four o’clock flower, a single–amino acid polymorphism in the host protease and a reciprocal single–amino acid change in the pathogen effectors underpin the ecological diversification (fig. S18). The arginine substitution found in the P. mirabilis effector may enhance effector inhibition of the M. jalapa protease. This same substitution would impair interaction with Asn147 of RCR3dms3 and His148 of RCR3lyc and so provide a molecular explanation for how this effector works on one protease but not the other (fig. S19).

Supplementary Materials

Materials and Methods

Supplementary Text

Figs. S1 to S19

Tables S1 to S5

References (1636)

  • * Present address: Division of Plant Sciences, University of Dundee, Invergowrie, Dundee DD2 5DA, UK.

  • Present address: Center for Proteomics and Bioinformatics, School of Medicine, Case Western Reserve University, Cleveland, OH 44106 USA.

  • Present address: Plant Breeding, Genetics, and Biotechnology, International Rice Research Institute (IRRI), Los Baños, Laguna, Philippines.

  • § Present address: Department of Plant and Environmental Protection Sciences, University of Hawaii, Honolulu, HI 96822, USA.

  • || Present address: School of Life Sciences, Gibbet Hill Campus, The University of Warwick, Coventry, CV4 7AL, UK.

References and Notes

  1. Materials and methods are available as supplementary material on Science Online.
  2. Acknowledgments: We thank D. G. O. Saunders for comments on drafts of the manuscript and C. Taylor, M. D. Coffey, and V. Vleeshouwers for providing biomaterial. This project was funded by the Gatsby Charitable Foundation, the U.K. Biotechnology and Biological Sciences Research Council, the Ohio Agricultural Research and Development Center at The Ohio State University, and a National Research Initiative of the U.S. Department of Agriculture grant OHO00963–SS. Sequences are deposited in GenBank under the submission accession numbers provided in (9).
View Abstract

Navigate This Article