Identification of a Papain-like Cysteine Protease Functioning as an Avirulence Factor in Striga-Cowpea Interactions

Abstract

While most cowpea cultivars are susceptible to parasitism by the root parasitic weed Striga gesnerioides (Willd.) Vatke, cultivar B301 is resistant to all Striga races except for SG4z. Resistance to Striga parasitism is manifested by the elicitation of a hypersensitive response (HR) at the site of parasite attachment on the host root followed by rapid death of the attached parasite. We isolated a papain-like cysteine protease (PLCP) designated SGCP1 that is highly expressed in the haustoria of S. gesnerioides race SG3 at the time of parasite attachment to the host root. SGCP1 contains an apoplast-targeting signal peptide, a Cathepsin pro-peptide inhibitory domain, a papain family cysteine protease domain, and a granulin domain. Full-length SGCP1 and a variant lacking the signal peptide (SGCP∆SP) were expressed in the roots of composite B301 plants. Expression of SGCP1 and SGCP∆SP resulted in activation of host innate immune responses exemplified by increased frequency of HR and decreased levels of parasite cotyledon expansion (CE), indicative of successful host parasitism, in transgenic compared to wild-type B301 roots parasitized by SG4z. These data indicate that SGCP1 functions as an avirulence factor capable of activating host innate immunity and furthers our understanding of how compatible and incompatible host-parasite interactions are controlled.

Keywords: Striga gesnerioides; avirulence; cowpea; cysteine protease; innate immunity.

Figure 1

Domain structure of SGCP1 and predicted protein structure. (A) The predicted protein structure of SGCP1 along with its identified functional domains. (B) The result of modeling the predicted protein structure of SGCP1 using the I-TASSER [39]. The figure depicts the highest ranked model (with a confidence score for estimating the quality of predicted model (C-score = −2.56) and estimated measures of the predicted model to obtain structures of TM-score = 0.42 ± 0.14; and RMSD = 13.4 ± 4.1Å. In the figure blue indicates an alpha-helix, red indicates a beta-sheet, and yellow represents a coil or irregular secondary structure.

Figure 1

Domain structure of SGCP1 and predicted protein structure. (A) The predicted protein structure of SGCP1 along with its identified functional domains. (B) The result of modeling the predicted protein structure of SGCP1 using the I-TASSER [39]. The figure depicts the highest ranked model (with a confidence score for estimating the quality of predicted model (C-score = −2.56) and estimated measures of the predicted model to obtain structures of TM-score = 0.42 ± 0.14; and RMSD = 13.4 ± 4.1Å. In the figure blue indicates an alpha-helix, red indicates a beta-sheet, and yellow represents a coil or irregular secondary structure.

Figure 2

Protein sequence alignment of SGCP1 with its S. asiatica homolog. Alignment of SGCP1 with a cysteine protease homolog from S. asiatica (STAS_16851, GenBank: GER40192.1). The alignment was generated by MultAlin version 5.4.1 [41]. Red lettering indicates identical amino acids between the two homologous proteins (100% identity); black lettering indicates residues that are not identical; and blue lettering indicates conserved residues. The Signal Peptide (SP), Cathepsin propeptide inhibitor domain (129), papain family cysteine protease domain, and Granulin domain are indicated by black lines.

Figure 3

Expression of SGCP1 and SGCP∆SP in B301 roots enhances HR and suppresses parasite CE when challenged by SG4z. (a) Ex vitro composite B301 plants were generated that express the SGCP and SGCP∆SP proteins in their roots. Transgenic and non-transgenic B301 roots were inoculated with 2 d pre-germinated SG4z seedlings, and the phenotypic responses (HR, hypersensitive response; CE, cotyledon expansion) were scored at 10- and 30-days post-inoculation (dpi). The interaction event frequencies of HR and CE were obtained by counting the number of each event category and dividing by the total number of attachment events on transgenic and non-transgenic roots of each host plant. Statistical analysis was performed using the unpaired two-tailed t test on >4 independent host plant replicates (non-transgenic roots 10 and 30 dpi, n = 5; SGCP overexpressing roots 10 and 30 dpi, n = 5; SGCP∆SP overexpressing roots 10 and 30 dpi, n = 4). Boxplots indicate the median (horizontal lines), 25th and 75th percentile range (boxes) and minimum and maximum values (whiskers). A minimum of 25 events per root and at least 400 events per plant were scored. The asterisks indicate significant difference (* p < 0.05; ** p < 0.01; *** p < 0.001). (b) Representative photographs illustrating the phenotypic response of non-transgenic and transgenic B301 roots overexpressing SGCP and SGCP∆SP when parasitized by SG4z at 30 dpi. In the left panel, the arrows indicate the development of the SG4z parasite on non-transgenic B301 roots. The arrows on the middle and right panel highlight where the SG4z parasite has resulted in an HR and parasite death due to the presence of the SGCP transgene.

Figure 4

Immunoblot analysis of SGCP1-HA and SGCP∆SP-HA expressed in transgenic cowpea roots. The full-length SGCP1 proteins with HA epitope tag fused to the C-terminus and a truncated version lacking the signal peptide (SGCP∆SP-HA) were expressed in ex vitro composite B301 plants. Immunoblotting was carried out on total root protein extracts transgenic roots using anti-HA antibody. Lanes 1–3, roots expressing SGCP-HA; lanes 4–5, roots expressing SGCP∆SP-HA. The molecular weights SGCP-HA and SGCP∆SP-HA were predicted by the Compute pI/MW tool to be 50.95 kDa and 48.66 kDa, respectively. The lower molecular weight forms seen on the gel correspond to the processed intermediate and mature forms generated in vivo. Molecular weight markers are given on the left.

Figure 5

Subcellular localization of SGCP-mCherry and SGCP∆SP-mCherry fusion proteins. The figure shows representative photographs of the subcellular localization of SGCP-mCherry and SGCP∆SP-mCherry fusion proteins in transgenic B301 cowpea roots as viewed by confocal microscopy. B301 roots were transformed using Agrobacterium rhizogenes R1000 containing the pK7WG2D-SGCP-mcherry and pK7WG2D-SGCP∆SP-mcherry plasmids. The left panel is a brightfield image, the second from left is the image channel showing the EGFP marker for transformation, the mCherry lane shows the signal from the target transgene expression and the right panel is the Merged image (an overlay of the two). Visualization of mCherry was performed by excitation at 587 nm and monitoring emission at 610 nm (using a 564–630 nm range filter), and EGFP visualization was performed by excitation at 488 nm and monitoring emission at 509 nm (using a 497–544 range filter). Arrows indicate signal in nucleus. The images were captured at 100× magnification; bars, 10 µm.

Figure 6

Model of compatible and incompatible interactions between Striga race SG4z and B301 with and without potentiation by transgenic SGCP1 and SGCP∆SP expression. The lower panel illustrates the current state of understanding of how SG4z overcomes host innate immunity by secretion of the suppressing effector SHR4z that leads to the proteolysis of VuPOB1 that is required for mounting an HR [18]. Expression of SGCP1 and SGCP∆SP leads to the formation of DAMPs that potentiate innate immunity, allowing the B301 root to shift the balance in favor of immunity versus susceptibility prior to SG4z attachment.