Molecular genetic evidence for the role of SGT1 in the intramolecular complementation of Bs2 protein activity in Nicotiana benthamiana

Abstract

Pepper plants (Capsicum annuum) containing the Bs2 resistance gene are resistant to strains of Xanthomonas campestris pv vesicatoria (Xcv) expressing the bacterial effector protein AvrBs2. AvrBs2 is delivered directly to the plant cell via the type III protein secretion system (TTSS) of Xcv. Upon recognition of AvrBs2 by plants expressing the Bs2 gene, a signal transduction cascade is activated leading to a bacterial disease resistance response. Here, we describe a novel pathosystem that consists of epitope-tagged Bs2-expressing transgenic Nicotiana benthamiana plants and engineered strains of Pseudomonas syringae pv tabaci that deliver the effector domain of the Xcv AvrBs2 protein via the TTSS of P. syringae. This pathosystem has allowed us to exploit N. benthamiana as a model host plant to use Agrobacterium tumefaciens-mediated transient protein expression in conjunction with virus-induced gene silencing to validate genes and to identify protein interactions required for the expression of plant host resistance. In this study, we demonstrate that two genes, NbSGT1 and NbNPK1, are required for the Bs2/AvrBs2-mediated resistance responses but that NbRAR1 is not. Protein localization studies in these plants indicate that full-length Bs2 is primarily localized in the plant cytoplasm. Three protein domains of Bs2 have been identified: the N terminus, a central nucleotide binding site, and a C-terminal Leu-rich repeat (LRR). Co-immunoprecipitation studies demonstrate that separate epitope-tagged Bs2 domain constructs interact in trans specifically in the plant cell. Co-immunoprecipitation studies also demonstrate that an NbSGT1-dependent intramolecular interaction is required for Bs2 function. Additionally, Bs2 has been shown to associate with SGT1 via the LRR domain of Bs2. These data suggest a role for SGT1 in the proper folding of Bs2 or the formation of a Bs2-SGT1-containing protein complex that is required for the expression of bacterial disease resistance.

Figure 1.

AvrBs2-Elicited Bs2 Resistance in the N. benthamiana Pathosystem Prevents Pstab Wild Fire Symptoms. N. benthamiana plants with and without Bs2:HA were hand infiltrated with 10⁵ cfu/mL Pstab expressing either the empty vector control or the avrBs2 construct, pV(avrRpm1-avrBs2:HA). Wild fire symptoms are visible 3 d after infiltration. Leaf symptoms were photographed 3 d after inoculation.

Figure 2.

VIGS of Bs2 Signal Transduction Genes Disrupts Bs2-Induced Resistance in the N. benthamiana Pathosystem. Bs2, SGT1, or NPK1 silencing interferes with Bs2/AvrBs2–mediated disease resistance, but RAR1 silencing does not. Plants tested were N. benthamiana (Nb); Bs2:HA–expressing N. benthamiana [Nb(Bs2:HA)]; Nb(Bs2:HA) silenced (Δ) for SGT1, NPK1, RAR1, or Bs2; and a GUS control for virus replication. (A) Control in planta pathogen growth assay of Pstab containing empty vector (pEV), lacking AvrBs2 expression, after low-inoculum infiltration. (B) In planta pathogen growth assay of Pstab with the avrBs2 construct, pV(avrRpm1-avrBs2:HA), after low-inoculum infiltration.

Figure 3.

Assay for VIGS. Reduction of detectible GFP resulting from cosilencing of the 35S-GFP-UTR reporter containing the target sequence used for VIGS of Bs2 signal transduction genes. All samples for a given VIGS treatment were infiltrated in the same leaf and collected 24 h after inoculation.

Figure 4.

Subcellular Localization of Bs2 in the Plant Cytoplasm. The first three columns show immunoblots of total protein extracts (T) from N. benthamiana Bs2:HA plants before ultracentrifugation at 100,000g for 1 h. Soluble (S) and membrane (M) fractions are indicated. The last six columns show immunoblots of protein extracts from N. benthamiana Bs2:HA plants at 24 h after inoculation with 10⁸ cfu/mL of either Pstab containing empty vector (pVSP61) or Pstab with the avrBs2 construct, pV(avrRpm1-avrBs2:HA). Samples were processed and analyzed as described above. Blots were probed with anti-HA antibody for detection of Bs2:HA, anti-NptII antibody as a soluble control protein, and anti-H-ATPase antibody as a membrane-localized control protein. As shown above, Bs2:HA localization was comparable with that observed for the soluble control protein NptII, indicating soluble (i.e., cytoplasmic) localization. The infection of Pstab with or without AvrBs2 did not alter Bs2 localization.

Figure 5.

Intramolecular Interactions within the Bs2 Protein Domains. (A) The Bs2 NX-NB domain associates with the LRR but is not altered by the coexpression of AvrBs2. The Bs2 domain constructs NX-NB:HA and LRR:Flag were transiently expressed in N. benthamiana plants with and without AvrBs2:GFP. Total protein extracts were immunoprecipitated with anti-HA (left panels) and anti-Flag (right panels) antibodies. Immunoprecipitated proteins were detected by immunoblotting with anti-HA (top panels) and anti-Flag (bottom panels) antibodies. (B) The Bs2 NX domain does not associate with the NB-LRR. The Bs2 domain constructs NX-HA and NB-LRR:Flag were transiently expressed for 24 h in N. benthamiana plants. Total protein extracts were immunoprecipitated (IP) with anti-HA (left panels) and anti-Flag (right panels) antibodies. Immunoprecipitated proteins were detected by immunoblotting (IB) with anti-HA (top panels) and anti-Flag (bottom panels) antibodies.

Figure 6.

Silencing SGT1 Inhibits the Association of the Bs2 NX-NB and the LRR. The NX-NB of Bs2 does not interact with the LRR in SGT1-silenced plants. The Bs2 domain constructs NX-NB:HA and LRR:Flag were transiently expressed with and without AvrBs2:GFP in N. benthamiana plants silenced for SGT1. Total protein extracts were immunoprecipitated with anti-HA (left panels) and anti-Flag (right panels) antibodies. Immunoprecipitated proteins were detected by immunoblotting with anti-HA (top panels) and anti-Flag (bottom panels) antibodies. The Bs2 domain constructs NX-NB:HA and LRR:Flag coimmunoprecipitate when transiently expressed in viral replication control GUS-silenced N. benthamiana. Total protein extracts were immunoprecipitated with anti-HA (left panels) and anti-Flag (right panels) antibodies. Immunoprecipitated proteins were detected by immunoblotting with anti-HA (top panels) and anti-Flag (bottom panels) antibodies.

Figure 7.

The LRR of Bs2 Associates with N. benthamiana SGT1. The Bs2:HA and Bs2 domain constructs NX-NB:HA and LRR:HA were transiently expressed with SGT1:Flag in N. benthamiana plants. Total protein extracts were immunoprecipitated with anti-HA (A) and anti-Flag (B) antibodies. Immunoprecipitated proteins were detected by immunoblotting with anti-HA and anti-Flag antibodies. Both Bs2:HA and LRR:HA coimmunoprecipitate with SGT1:Flag.

Figure 8.

NbSGT1 Is Biologically Active. Agrobacterium-mediated transient expression of SGT1 and SGT1:Flag reactivates Bs2:HA-AvrBs2–specific HR in SGT1-silenced N. benthamiana Bs2:HA plants. N. benthamiana SGT1 (NbSGT1), C-terminal epitope-tagged NbSGT1:Flag, and NbSGT1:Flag_stop80 mutant, with the stop codon introduced at amino acid 80, were cloned in a binary vector with a chimeric UAS and MAS promoter for enough transient expression to overcome the effects of VIGS of SGT1. Agrobacterium coinfiltration mixtures were 5 × 10⁸ cfu/mL each, and HR reactions were photographed at 3 d after infiltration. Infiltrated plants were N. benthamiana viral replication control (Nb ΔGus), N. benthamiana Bs2:HA viral replication control [Nb(Bs2:HA) ΔGus], and N. benthamiana Bs2:HA with VIGS for SGT1 [Nb(Bs2:HA) ΔSGT1].