The interaction between geminivirus pathogenicity proteins and adenosine kinase leads to increased expression of primary cytokinin-responsive genes

Abstract

Pathogenicity proteins (AL2/C2) of begomo- and curtoviruses suppress silencing through inhibition of the methyl cycle, as a consequence of inhibiting adenosine kinase (ADK). ADK phosphorylates cytokinin nucleosides, helping maintain a pool of bioactive cytokinins through interconversion of free-bases, nucleosides and nucleotides. We provide evidence that inhibiting ADK affects expression of primary cytokinin-responsive genes. Specifically, we demonstrate increased activity of a primary cytokinin-responsive promoter in adk mutant Arabidopsis plants, and in response to silencing ADK expression or inhibiting ADK activity in transient assays. Similar changes in expression are observed in geminivirus infected tissue and when AL2/C2 are over-expressed. Increased cytokinin-responsive promoter activity may therefore be a consequence of an ADK/AL2/C2 interaction. Application of exogenous cytokinin increases susceptibility to geminivirus infection, characterized by a reduced mean latent period and enhanced viral replication. Thus, ADK appears to be a high value target of geminiviruses that includes increasing expression of primary cytokinin-responsive genes.

Fig. 1

Increased ARR5∷GUS promoter activity in response to geminiviruses. Panels A through K illustrate N.benthamiana leaf sections stained for the presence of GUS. Leaves were co-infused with Agrobacterium cultures containing the ARR5∷GUS promoter-reporter in the presence of: A) empty vector; B) 5 µM benzylamino purine; C) WT SCTV; D) WT TGMV DNA A; E) TGMV AL2; F) SCTV C2; G) TGMV AL1; H) TGMV AL3; I) dsADK; J) dsAKIN11; and K) ADK inhibitor A-134974. Tissue was analyzed six days post-infusion. Panels L, M and N illustrate GUS staining of inflorescences from transgenic Arabidopsis plants containing the ARR5∷GUS promoter-reporter, either mock inoculated (L), bolts exhibiting SCTV symptoms (M) or asymptomatic bolts from SCTV-infected plants (N).

Fig. 2

Geminivirus pathogenicity proteins induce increased expression of endogenous primary cytokinin responsive genes in Arabidopsis. The graphs indicate relative levels of ARR5 (A) and ARR16 (B) mRNA. Values were determined by quantitative real time RT-PCR analysis of RNA isolated from Arabidopsis plants infused with Agrobacterium cultures containing DNA capable of expressing CaLCuV AL2, SCTV C2 or vector alone (Mock). Columns represent the fold change calculated from the mean ΔΔCt value from two (SCTV) or three (CaLCuV) independent experiments using RNA isolated one and two days post-infusion. Asterisks indicate significant differences in expression as determined using the Student's t-test (P<0.05). P values for all samples are given.

Fig. 3

Geminivirus infection induces expression of endogenous primary cytokinin responsive genes in Arabidopsis. The graph represents the relative quantification of ARR16 mRNA levels observed in different treatments. Values were determined by quantitative real time RT-PCR analysis of RNA isolated from Arabidopsis rosette leaves systemically infected with Beet curly top virus (BCTV), Beet severe curly top virus (BSCTV) or Spinach curly top virus (SCTV). The columns represent relative differences in ARR16 mRNA levels compared to that present in mock-inoculated plants.

Fig. 4

The effect of exogenous cytokinin on accumulation of viral nucleic acid in N.benthamiana leaves. N.benthamiana leaves were infused with Agrobacterium cultures containing wild type or c2-4 mutant SCTV DNA (A), or mechanically inoculated with TBSV (B), in the absence or presence of 5µM Benzylaminopurine (−/+ BAP). Total DNA or RNA was isolated from leaves three to 12 days post-infusion and hybridized to probes specific for SCTV (A) or TBSV (B). To the left of each panel double-stranded supercoiled (SC), open circular (OC) and linear (LIN), and single-stranded (SS) viral DNA forms of SCTV, and genomic (g) and subgenomic (sg1) forms of TBSV are shown. Hybridization was normalized to either 18S rDNA (A) or 18S rRNA (B). The relative levels of total viral DNA or RNA are indicated below each respective panel. For SCTV a value of 1 was arbitrarily assigned to the total amount of viral DNA present in samples from leaves infused with wild type SCTV in the absence of BAP at six days post-infusion (A). For TBSV a value of 1 was arbitrarily assigned to the total amount of viral RNA present in samples from leaves inoculated in the absence of BAP at three days post-infusion (B).