Fig 1.
The enzymatic activity of Avr3b is activated by plant factors.
(A) Avr3b only has Nudix hydrolase activity when produced in plants. Avr3b was ectopically expressed in E. coli (Ec) or N. benthamiana (Nb). Relative Nudix hydrolase activity was calculated by comparing enzymatic activity of purified GST-Avr3b with GST produced in E. coli or purified FLAG-Avr3b with FLAG-GFP in N. benthamiana. NADH was applied as substrate in Nudix hydrolase activity assays. Means and standard errors from three replicates are shown. ** representing t test P < 0.01. (B) E. coli produced Avr3b protein can be activated by incubation with plant protein extracts. Recombinant GST-Avr3b protein was purified from E. coli. The recombinant protein (2 μg) was incubated with 100 μg dialyzed soybean (Glycine max, Gm) extract, 100 μg dialyzed N. benthamiana extract, or 100 μg dialyzed P. sojae (Ps) mycelium extract at 25°C for 15 hours. The Nudix hydrolase activity was then determined using FLAG-GFP as a control. Means and standard errors from three measurements are shown. ** or *, representing t test P < 0.01 or 0.05, respectively.
Fig 2.
Avr3b interacts with plant cyclophilins.
(A) Avr3b interacts with GmCYP1 in yeast. Avr3b and GmCYP1 were cloned into pGBKT7 and pGADT7 vectors, respectively. Yeast transformants were grown on SD/-Trp/-Leu (SD-2) or the selective medium SD/-Trp/-Leu/-His/-Ade (SD-4). The plates were photographed 2 days after inoculation. (B) Avr3b physically interacts with GmCYP1 in vitro. GST-Avr3b or GST bound resins were incubated with E. coli cell lysate containing His-GmCYP1. Co-precipitation of His-GmCYP1 with the GST-binding resins was examined by western blots using anti-His antibody before (Input) and after resins incubation (Pull-down). (C) Avr3b interacts with GmCYP1 in plant cells. FLAG-Avr3b was transiently expressed with either GFP-GmCYP1 or GFP in N. benthamiana leaves. FLAG-tagged Avr3b was immunoprecipitated by anti-FLAG M2 affinity gel from total plant extracts. Input controls (Input) and M2 gel binding proteins (IP) were analyzed by western blots using anti-FLAG or anti-GFP antibody. (D) Avr3b interacts with NbCYP3 and NbCYP4 in N. benthamiana. FLAG-Avr3b was transiently expressed with GFP-tagged protein fusions of NbCYPs or GFP in N. benthamiana. Immunoprecipitates pulled down using anti-FLAG M2 affinity gel from total protein extracts were immunoblotted with anti-FLAG or anti-GFP antibodies. IP, immunoprecipitate. These experiments were repeated three times with similar results.
Fig 3.
Activation of Avr3b is dependent on the PPIase activity of GmCYP1.
(A) GmCYP1 possesses peptidyl-prolyl cis-trans isomerase (PPlase) activity. GmCYP1R62A (PPIase-deficient mutant) was generated by site-directed mutagenesis. The PPIase enzymatic activity of GST, GmCYP1, and GmCYP1R62A produced in E. coli was analyzed using a chymotrypsin-coupled assay. Enzyme activity can be assessed by analyzing the peak of the curve 15 s (dotted line indicated) after the adding of α-chymotrypsin. A higher absorbance in 390 nm indicates increased PPIase activity. 20 μM cyclosporine A (CsA) was added to the reaction when appropriate as a chemical inhibitor of the PPIas activity. Three independent replicates were performed with similar results. (B) Avr3b can be activated by GmCYP1 in vitro. GST-Avr3b proteins produced from E. coli were incubated with purified GmCYP1or GmCYP1R62A. CsA was supplemented to the reactions when appropriate. The Nudix hydrolase activity was then determined using GST as a control. Means and standard errors from three measurements are shown. ** or *, representing significantly different than the Avr3b t test P < 0.01 or 0.05, respectively. (C) GmCYP1 enhances the Nudix hydrolase activity of Avr3b in planta. Avr3b was expressed with GFP-GmCYP1, GFP-GmCYP1R62A or GFP in N. benthamiana leaves using Agrobacterium-mediated transient expression and Nudix hydrolase activity was analyzed 48 hours post Agro-infiltration (hpi). Bars represent standard errors from four biological replicates. The same letter indicates no significant difference between values, and different letters indicate significant differences between values (P < 0.01, nonparametric Kruskal-Wallis test). (D) GmCYP1 enhances the virulence activity of Avr3b in N. benthamiana. Leaves expressing Avr3b with GFP-GmCYP1, GFP-GmCYP1R62A or GFP were inoculated with zoospore suspension of P. capsici 48 hpi. DNA from P. capsici infected regions was isolated at 36 hours post P. capsici inoculation, and the biomass of P. capsici in infected tissues was determined using quantitative PCR comparing with GFP-expressing tissues. Bars represent standard errors from three biological replicates. The letters represent statistical significance (P < 0.01 nonparametric Kruskal-Wallis test).
Fig 4.
NbCYP3 and NbCYP4 are required for the Nudix hydrolase and the virulence activity of Avr3b in N. benthamiana.
(A) NbCYP3/4-silenced N. benthamiana can not activate the Nudix hydrolase activity of Avr3b. Avr3b and RFP (as a negative control) were transiently expressed in NbCYP3/4-silenced N. benthamiana leaves. Total proteins were extracted at 48 hpi, and the Nudix hydrolase activity was measured using NADH as substrate. Means and standard errors from four measurements are shown. **, t test P<0.01. (B) NbCYP3 and NbCYP4 are required for the virulence activity of Avr3b in N. benthamiana. FLAG-RFP or FLAG-Avr3b proteins were transiently expressed in N. benthamiana leaves by Agro-infiltration. The leaves were inoculated with P. capsici at 48 hours post Agro-infiltration. Infection was determined using quantitative PCR to measure the ratios of P. capsici and N. benthamiana DNA at 36 hours post inoculation. Means and standard errors from four measurements are shown. ** representing t test P < 0.01. (C) NbCYP3 and NbCYP4 can enhance the Nudix hydrolase activity of Avr3b in N. benthamiana. FLAG-Avr3b was co-expressed in N. benthamiana leaves with GFP-NbCYP3 or GFP-NbCYP4 by Agro-infiltration method, and the Nudix hydrolase activity was measured using NADH as substrate comparing with GFP-expressing leaves. Bars represent standard errors from three biological replicates. The same letter indicates no significant difference between values, and different letters indicate significant differences between values (P < 0.01, nonparametric Kruskal-Wallis test).
Fig 5.
Avr3b recognition in soybean is dependent on the PPlase activity of GmCYP1.
(A, B) CsA suppresses Avr3b-triggered cell death in soybean leaves. (A) Soybean cultivars Williams (rps3b) and PRX146-36 (Rps3b) were transformed by co-bombardment with a plasmid mixture consisting of a β-glucuronidase (GUS) expression vector and Avr3b or empty vector (EV). Soybean leaves were sprayed 20 μM CsA solution immediately after bombardment and incubated for 2 days in darkness at 28°C. The leaves were then stained using X-gluc. Scale bars, 3mm. (B) Percentage rate of GUS-positive blue spots following bombardment with Avr3b compared with EV. Bars represent standard errors from four independent replicates. **, t test P<0.01. (C, D) CsA could not suppress Avr1b-triggered cell death. (C) Avr1b was transiently expressed in soybean cultivars Williams (rps1b) and L77-1863 (Rps1b) by co-bombardment with GUS expression vector and Avr1b or EV. The leaves were incubated for 2 days in darkness at 28°C after bombardment, and then stained using X-gluc. Scale bars, 3mm. (D) Percentage rate of GUS-positive blue spots following bombardment with Avr1b compared with the empty vector. Bars represent standard errors from four independent replicates. **, t test P<0.01. (E, F) Avr3b failed to induce Rps3b-mediated cell death in GmCYP1-silenced soybean leaves. (E) Avr3b was delivered into leaves of soybean cultivars Williams (rps3b) and PRX146-36 (Rps3b) together with a GmCYP1-RNAi construct by co-bombardment. The leaves were incubated for 2 days in darkness at 28°C after bombardment, and then stained using X-gluc. Scale bars, 3mm. EV was used as a negative control. (F) Percentage rate of GUS-positive blue spots following bombardment with Avr3b compared with the EV in the presence of GmCYP1-RNAi. Bars represent standard errors from four independent replicates. **, t test P<0.01.
Fig 6.
Proline132 of Avr3b plays a critical role in activation by cyclophilins.
(A) Avr3bP132A mutant can not be activated in N. benthamiana. FLAG-GFP, FLAG-Avr3b or FLAG-Avr3bP132A were expressed in N. benthamiana leaves. Relative Nudix hydrolase activity was analyzed at 48 hpi. Bars represent standard errors from three independent replicates. * representing significantly different than FLAG-GFP (t test P<0.05). (B) Avr3bP132A is impaired in its interaction with GmCYP1. Yeast cells were co-transformed with pGADT7-GmCYP1 and pGBKT7 carrying Avr3b or Avr3bP132A. Transformants were grown on a selective medium.β-galactosidase activity in yeast cells was measured to determine the Avr3b-GmCYP1 interaction. Bars represent standard errors from three independent replicates. *, t test P<0.05. (C) The P132A mutation of Avr3b resulted in loss of virulence activity. FLAG-GFP, FLAG-Avr3b or FLAG-Avr3bP132A proteins were transiently expressed in N. benthamiana leaves by Agro-infiltration. The leaves were inoculated with P. capsici at 48 hours post Agro-infiltration. Biomass of P. capsici was determined at 36 hours post inoculation by quantitative PCR. Bars represent standard errors from three independent replicates. *, t test P<0.05. (D) Avr3bP132A failed to induce Rps3b-mediated cell death in soybean. Percentage rate of β-glucuronidase (GUS)-positive blue spots following bombardment with Avr3bP132A compared with the EV. Bars represent standard errors from four independent replicates. **, t test P<0.01. (E) Avr3bP132A could not suppress Avr1b-triggered HR in soybean producing Rps1b. Percentage rate of GUS-positive blue spots following co-bombardment of Avr1b with Avr3b or Avr3bP132A compared with EV. Data are the means of four independent experiments. Bars represent standard errors from four independent replicates. **, t test P<0.01.
Fig 7.
Avr3b specifically interacts with GmCYP1 in soybean.
(A, C) Avr3b does not interact with other GmCYP1 homologs in soybean or two CYPs of P. sojae. Selected cyclophilin genes were cloned into pGADT7. Yeast cells were co-transformed with the empty prey vector, pGADT7 or pGADT7 containing one of homologue genes and the bait vector pGBKT7-Avr3b. Yeast transformants were grown on the selective minimal SD medium lacking tryptophan, and leucine (SD-2) or the selective SD medium lacking adenine, tryptophan, histidine, and leucine (SD-4). The plates were photographed 2 days after inoculation. (B, D) Western blots showing protein expression of cyclophilin homologs in yeast cells. Anti-HA antibody was used to detect the expression of cyclophilin homologous proteins of soybean and P. sojae. The same protein gel was stained with Coomassie brilliant blue (CBB) to show loading. WB, western blot. (E) A schematic summary illustrating maturation of Avr3b by plant cyclophilin proteins during Phytophthora infection. Inactive Avr3b enters plant cell and interacts with plant cyclophilin to gain the Nudix hydrolase activity. Active Avr3b suppresses ETI-associated cell death triggered by Avr1b. However, in the presence of Rps3b, active Avr3b triggers Rps3b-induced HR.