HopX1 in Erwinia amylovora functions as an avirulence protein in apple and is regulated by HrpL

Abstract

Fire blight is a devastating disease of rosaceous plants caused by the Gram-negative bacterium Erwinia amylovora. This pathogen delivers virulence proteins into host cells utilizing the type III secretion system (T3SS). Expression of the T3SS and of translocated and secreted substrates is activated by the alternative sigma factor HrpL, which recognizes hrp box promoters upstream of regulated genes. A collection of hidden Markov model (HMM) profiles was used to identify putative hrp boxes in the genome sequence of Ea273, a highly virulent strain of E. amylovora. Among potential virulence factors preceded by putative hrp boxes, two genes previously known as Eop3 and Eop2 were characterized. The presence of functionally active hrp boxes upstream of these two genes was confirmed by β-glucuronidase (GUS) assays. Deletion mutants of the latter candidate genes, renamed hopX1(Ea) and hopAK1(Ea), respectively, did not differ in virulence from the wild-type strain when assayed in pear fruit and apple shoots. The hopX1(Ea) deletion mutant of Ea273, complemented with a plasmid overexpressing hopX1(E)(a), suppressed the development of the hypersensitivity response (HR) when inoculated into Nicotiana benthamiana; however, it contributed to HR in Nicotiana tabacum and significantly reduced the progress of disease in apple shoots, suggesting that HopX1(Ea) may act as an avirulence protein in apple shoots.

Fig 1

Expression assays of selected putative hrp boxes. The reporter plasmids pCPP1744 to pCPP1746, corresponding to EAM_2780, EAM_2189, and DspA/E hrp boxes, respectively, were transformed either into wild-type Ea273 or into Ea273-K49 (an hrpL mutant) as a negative control. pCPP1746 (DspA/E hrp box) was used as a positive control. The strains of E. amylovora were grown for 6 h at 20°C in Huynh's medium to induce HrpL production followed by 4 h of incubation at 37°C in reaction buffer. GUS activity was measured in pU/CFU. The bars represent the standard errors of four replicates. The graph shows the results for one of three replicate experiments.

Fig 2

Pathogenicity test in immature pears. Pears were inoculated with mutants of Ea273 as indicated, and symptoms were assessed 5 days after inoculation. Wild-type Ea273 was used as a positive control, and pears treated with potassium phosphate buffer served as negative controls. (A) Deletion of hopAK1_Ea did not have an effect on virulence of Ea273. (B) Deletion of hopX1_Ea did not have an effect on virulence of Ea273, but complementation of hopX1_Ea, including the chaperone and promoter region, in a low-copy-number plasmid (pCPP1738) or only hopX1_Ea in a high-copy-number plasmid (pCPP1737) had reduced virulence compared with the wild type or the hopX1_Ea mutant. The figure shows one of three replicates of one of three experiments with similar results.

Fig 3

HR test in tobacco plants. Nicotiana plants were inoculated with mutants of Ea273 at an OD₆₀₀ of 0.2. Wild-type Ea273 was used as positive control, and inoculations with potassium phosphate buffer served as negative control. HR elicitation was assessed 24 h postinoculation. (A) In N. tabacum, Ea273ΔhopX1_Ea does not elicit HR at 24 h. (B) None of the mutants introduced into N. benthamiana differed from the wild type; however, overexpression of HopX1_Ea by Ea273ΔhopX1_Ea(pCPP1737) suppressed HR.

Fig 4

HopX1_Ea functions as an HR suppressor in N. benthamiana; however, it contributes to HR in N. tabacum. Nicotiana plants were inoculated with EcoMC4100 strains harboring pCPP430, pCPP431, or pCPP432 alone or together with pCPP1737 (overexpressing HopX1_Ea) at an OD₆₀₀ of 0.8. Inoculations with potassium phosphate buffer served as negative control. HR elicitation was assessed 24 h postinoculation. (A) In N. tabacum, the expression of HopX1_Ea enhanced HR of EcoMC4100(pCPP431). (B) In N. benthamiana, the expression of HopX1_Ea reduced drastically the HR of EcoMC4100(pCPP430).

Fig 5

Virulence assay in apple shoots. Apple shoots of 20 to 40 cm were inoculated with Ea273 mutants and complemented strains as indicated. Wild-type Ea273 was used as a positive control, and inoculations with potassium phosphate buffer served as negative controls. Deletions of hopX1_Ea and hopAK1_Ea do not affect virulence of Ea273 on apple shoots; however, overexpression of hopX1_Ea in Ea273ΔhopX1_Ea(pCPP1737) reduced the severity of symptoms significantly. (A) Pictures of symptomatic apple shoots at 5 days after inoculation. (B) Disease progress at 8 days after inoculation measured as the percentage of the length of symptomatic shoots versus the total length of the shoot. The values for each strain are the means of nine samples per treatment from one experiment. The bars represent standard errors. Values with the same capital letters do not differ significantly at the P = 0.05 level. The experiment was repeated three times.