An effector from the Huanglongbing-associated pathogen targets citrus proteases

Abstract

The citrus industry is facing an unprecedented challenge from Huanglongbing (HLB). All cultivars can be affected by the HLB-associated bacterium 'Candidatus Liberibacter asiaticus' (CLas) and there is no known resistance. Insight into HLB pathogenesis is urgently needed in order to develop effective management strategies. Here, we use Sec-delivered effector 1 (SDE1), which is conserved in all CLas isolates, as a molecular probe to understand CLas virulence. We show that SDE1 directly interacts with citrus papain-like cysteine proteases (PLCPs) and inhibits protease activity. PLCPs are defense-inducible and exhibit increased protein accumulation in CLas-infected trees, suggesting a role in citrus defense responses. We analyzed PLCP activity in field samples, revealing specific members that increase in abundance but remain unchanged in activity during infection. SDE1-expressing transgenic citrus also exhibit reduced PLCP activity. These data demonstrate that SDE1 inhibits citrus PLCPs, which are immune-related proteases that enhance defense responses in plants.

Fig. 1

SDE1 interacts with citrus papain-like cysteine proteases. a Yeast-two-hybrid (Y2H) assays using the CLas effector SDE1 as the bait and full-length citrus papain-like cysteine proteases (CsPLCPs), representing different subfamilies as the prey. SDE1 was cloned into the vector pGBKT7 and individual CsPLCPs were cloned into the vector pGADT7. Growth of yeast cells on SD-3 selective media represents protein–protein interaction, growth of the same cells on SD-2 media confirms yeast transformation. Yeast transformed with the empty vectors served as negative controls. The initial PLCP found from Y2H screening (CsSAG12-1, XM_006495158) is indicated with an asterisk (*). The gene IDs of the other interacting PLCPs are CsSAG12-2 (XM_006470229), CsXCBP3 (orange1.1g012960), CsRD21a (XM_006473212), CsRD19 (orange1.1g017548), CsAALP (XM_006474664), and CsCTB (orange1.1g018568). b Phylogeny and subfamily classification of canonical PLCPs in the C. sinensis (sweet orange) genome. The phylogenetic tree was made with MEGA6.06 (100 bootstrap replicates, Maximum-Likelihood method, Jones–Taylor–Thornton model), using the Arabidopsis thaliana PLCP subfamily classification. The asterisk (*) indicates the initially found CsSAG12-1. c Y2H assay examining the interaction of SDE1 with the cysteine protease domain of CsPLCPs. d In vitro pull-down assay using the GST-tagged cysteine protease domain of CsPLCPs to immunoprecipitate SDE1 protein. Input and immunoprecipitated proteins (output) were visualized by western blotting using anti-GST and anti-SDE1 antibodies. Asterisks (*) indicate the protein bands that correspond to individual CsPLCPs. GST-tagged Arabidopsis double-stranded DNA-binding protein 4 (AtDRB4) was used as a negative control

Fig. 2

SDE1 inhibits PLCP activity in vitro and in plant cells. a Proteolytic activity of papain measured by digestion of a fluorescent casein substrate in the presence of E-64, purified SDE1 protein, or BSA (as a negative control). Fluorescence was measured at 485/530 nm excitation/emission. Mean ± standard deviation (n = 3) is shown. Asterisks (*) indicate statistically significant differences based on the two-tailed Student’s t-test. p < 0.01 = **, p < 0.001 = ***. b Inhibitory effect of SDE1 on the protease activity of papain examined by activity-based protein profiling (ABPP). Active papain was labeled by DCG-04 in the presence of 10 μM E-64 or 1.6 μM purified SDE1 protein and detected using streptavidin conjugated with horseradish peroxidase (HRP). c SDE1 inhibits the activity of CsRD21a. CsRD21a-Flag (with its N-terminal secretion signal) was expressed in N. benthamiana. Active protease in the apoplastic fluid was labeled via ABPP. ImageJ analysis of the signal intensity revealed approximately 9%, 62%, and 96% reduction of CsRD21a activity in the presence of 0.8, 1.6, or 3.2 μM purified SDE1 protein, respectively. d SDE1 inhibits PLCP activity in citrus. Total protein extracts from Navel orange (C. sinensis) leaves were labeled via ABPP in the presence of 120 nM purified SDE1 protein. Active proteases were enriched using streptavidin beads and detected using streptavidin-HRP conjugates. e Transgenic grapefruit (Duncan) seedlings expressing SDE1 exhibit reduced protease activity. Five individual lines were analyzed by ABPP. SDE1-10 does not have significant SDE1 protein accumulation and served as a negative control

Fig. 3

CsPLCPs accumulate during SA treatment and infection. a Abundance of PLCP genes was determined by quantitative RT-PCR after SA treatment. One-year-old Navel oranges (C. sinensis) were sprayed with 2 mM SA or water. Leaf samples were collected after 48 h for RNA extraction and PCR analyses. Cytochrome oxidase subunit 1 (COX, KF933043.1) was used as the internal standard. Graph shows mean ± standard error of three replicates. Asterisks (*) indicate statistically significant differences based on the two-tailed Student’s t-test. p < 0.05 = *, p < 0.01 = **. b Protein abundance of PLCPs was determined in healthy (−) or CLas-infected (+) citrus branches using an anti-AALP antibody. Freshly cut stems were stamped onto nitrocellulose membranes and PLCPs and SDE1 were detected using western blotting. The titer of CLas in each sample was evaluated by quantitative PCR with observed Ct values of 27.97 for symptomatic tissue (+S), and not detected for asymptomatic tissue from the same infected tree (+AS) or tissue from an uninfected tree (−). Ponceau-stained membrane was shown as a control

Fig. 4

CsSAG12s increase in abundance but not activity in infected citrus. a Diagram illustrating the experimental approach for detecting abundance and activity of CsPLCPs in healthy and infected Navel oranges (C. sinensis) from a Texas grove using mass spectrometry. b Abundance and protease activity of six PLCPs belonging to four subfamilies in citrus leaf samples. Leaf tissue was ground in Tris buffer and divided into two to assess abundance and activity. PLCP abundance was tested using an in-solution digest coupled with mass spectrometry. For activity, the addition of DCG-04 permits the labeling of active PLCPs. Active PLCPs were captured and identified by streptavidin IP coupled with mass spectrometry. Mean ± standard error of three replicates is shown. Asterisks (*) indicate statistically significant differences based on the two-tailed Student’s t-test. N.d = no difference, p < 0.05 = *, p < 0.01 = **, p < 0.001 = ***. The gene IDs are CsXBCP3 (orange1.1g012960m), CcXCP1 (Ciclev10001665m), CsAALP (orange1.1g036910m), CcSAG12-1 (Ciclev10005334m), CsSAG12-3 (orange1.1g018958m), and CsSAG12-4 (orange1.1g019063m). Jessica Franco created the citrus images in panel a by hand for use in this paper

Fig. 5

SDE1 promotes Pseudomonas syringae infection of Arabidopsis. Mature leaves of 5-week-old plants of Arabidopsis thaliana ecotype Col-0 were syringe-infiltrated with cell suspensions of P. syringae pv. tomato (Pto) strains including DC3000 (wild type), Δcip1, Δcip1(empty vector, EV), and Δcip1(SDE1). Bacterial titers were determined as colony forming units (cfu/cm²) at the time of infiltration (Day 0) and 3 days post infiltration (Day 3). Graph shows mean ± standard deviation of data from three independent experiments. Different letters (a, b, and c) indicate statistically significant differences (p < 0.05) based on a one-way ANOVA followed by Tukey’s HSD post hoc test

Fig. 6

SDE1 does not inhibit Solanaceous RCR3 activity. a SDE1 did not elicit cell death in tomato (Moneymaker, Solanum lycopersicum) leaves expressing the immune receptor Cf-2. Purified recombinant proteins FLAG-Avr2-6XHIS or FLAG-6XHIS-SDE1 were infiltrated in tomato leaves in three different concentrations: 1 μg, 100 ng, and 10 ng. Avr2 inhibits the protease activity of RCR3^pim (RCR3 from Solanum pimpinellifolium, which is required for Avr2-triggered cell death in tomato plants expressing Cf-2 produced by S. pimpinellifolium (left). Cf-0 (right) and Cf-2 rcr3 (middle, containing a rcr3 mutant with a premature stop codon) tomatoes were used as negative controls. Picture was taken after 7 dpi. b SDE1 does not inhibit the activity of RCR3^pim. Full length of RCR3^pim-HIS was transiently expressed in N. benthamiana and secreted into the apoplast. Apoplastic fluid was extracted and the active protease was labeled via ABPP in the presence of 0.8, 1.6 or 3.2 μM of purified SDE1 protein. Coomassie brilliant blue (CBB) served as a loading control