Ectopically expressed sweet pepper ferredoxin PFLP enhances disease resistance to Pectobacterium carotovorum subsp. carotovorum affected by harpin and protease-mediated hypersensitive response in Arabidopsis

Abstract

Plant ferredoxin-like protein (PFLP) is a photosynthesis-type ferredoxin (Fd) found in sweet pepper. It contains an iron-sulphur cluster that receives and delivers electrons between enzymes involved in many fundamental metabolic processes. It has been demonstrated that transgenic plants overexpressing PFLP show a high resistance to many bacterial pathogens, although the mechanism remains unclear. In this investigation, the PFLP gene was transferred into Arabidopsis and its defective derivatives, such as npr1 (nonexpresser of pathogenesis-related gene 1) and eds1 (enhanced disease susceptibility 1) mutants and NAHG-transgenic plants. These transgenic plants were then infected with the soft-rot bacterial pathogen Pectobacterium carotovorum subsp. carotovorum (Erwinia carotovora ssp. carotovora, ECC) to investigate the mechanism behind PFLP-mediated resistance. The results revealed that, instead of showing soft-rot symptoms, ECC activated hypersensitive response (HR)-associated events, such as the accumulation of hydrogen peroxide (H2 O2 ), electrical conductivity leakage and expression of the HR marker genes (ATHSR2 and ATHSR3) in PFLP-transgenic Arabidopsis. This PFLP-mediated resistance could be abolished by inhibitors, such as diphenylene iodonium (DPI), 1-l-trans-epoxysuccinyl-leucylamido-(4-guanidino)-butane (E64) and benzyloxycarbonyl-Val-Ala-Asp-fluoromethylketone (z-VAD-fmk), but not by myriocin and fumonisin. The PFLP-transgenic plants were resistant to ECC, but not to its harpin mutant strain ECCAC5082. In the npr1 mutant and NAHG-transgenic Arabidopsis, but not in the eds1 mutant, overexpression of the PFLP gene increased resistance to ECC. Based on these results, we suggest that transgenic Arabidopsis contains high levels of ectopic PFLP; this may lead to the recognition of the harpin and to the activation of the HR and other resistance mechanisms, and is dependent on the protease-mediated pathway.

Keywords: Erwinia; NPR1; PFLP; disease resistance; ferredoxin; harpin.

Figure 1

Characterization of plant ferredoxin‐like protein (PFLP)‐transgenic Arabidopsis. The genomic DNA was digested with EcoRI (E) or SacI (S) and detected by Southern blot analysis using a NPTII (NEOMYCIN PHOSPHOTRANSFERASE II) gene probe (A). The total RNAs isolated from non‐transgenic Arabidopsis (Wt 1–4) and transgenic lines (PFLP‐1 1–2 and PFLP‐2 1–2) were investigated by Northern blot analysis. The ribosomal RNA (rRNA), stained by ethidium bromide (EtBr), was used as the loading control (B). The total RNA (1 μg) was used in the semi‐quantifying reverse transcription‐polymerase chain reaction RT‐PCR) analysis together with specific primers for the PFLP, ATFD1, ATFD2 and ATELF1 α genes. (C). The crude extract proteins (10 μg) of Wt, PFLP‐1 and PFLP‐2 were identified by Western blot analysis with antiserum against PFLP. The monoclonal antiserum against actin was used as a loading control. The number indicated is the relative ratio of the PFLP signal after normalization by actin (D).

Figure 2

Immunolocalization of plant ferredoxin‐like protein (PFLP) in transgenic Arabidopsis. Images of the non‐transgenic line (A–C) and transgenic line 2 (D–F) were obtained from leaf tissue of 1‐month‐old plants. The green pseudo‐colour indicates the existence of protein recognized by the PFLP antiserum and fluorescein isothiocyanate (FITC) (A, D). The red pseudo‐colour indicates chloroplast autofluorescence (B, E). The merged images show the localization of protein recognized by the PFLP antiserum (C, F). The bar represents 5 μm in length.

Figure 3

Inoculation of Erwinia carotovora ssp. carotovora (ECC) in plant ferredoxin‐like protein (PFLP)‐transgenic Arabidopsis. The 1‐month‐old Arabidopsis plants were sprayed with ECC bacterial suspension [1 × 10⁵ colony‐forming units (cfu)/mL] and photographs were taken at 48 h (A) and 24 h (B) post‐inoculation. The rosette leaves of 1‐month‐old Arabidopsis were detached and immersed in a suspension of ECC (1.0 × 10³ cfu/mL) at 28 °C with shaking at 100 rpm. (C) Photograph taken at 24 h post‐inoculation. The absorption of the suspension was estimated by a spectrophotometer at OD ₆₀₀ (optical density at 600 nm) at 0, 12, 24, 36, 48 and 60 h post‐incubation. Data are presented as the mean ± standard error of the mean for six independent determinations (D). Wt, wild‐type.

Figure 4

The activation of the hypersensitive response (HR) in plant ferredoxin‐like protein (PFLP)‐transgenic Arabidopsis. RNA was extracted at 0, 6, 12 and 24 h post‐inoculation of Erwinia carotovora ssp. carotovora (ECC) and investigated by Northern blot analysis with probes for the HR marker genes ATHSR2 and ATHSR3. The ribosomal RNA (rRNA), stained by ethidium bromide (EtBr), was used as the loading control (A). The electrolyte leakage of plants was analysed at 0, 8, 16, 24 and 38 h post‐inoculation by ECC (B). The accumulation of H₂O₂ in leaf tissue was estimated at 8, 16, 24 and 36 h post‐inoculation by ECC. Data are presented as the mean ± standard error of the mean for six independent determinations (C). The accumulation of H₂O₂ in leaf tissue was assessed after staining by 3,3′‐diaminobenzidine (DAB) in vivo. The detached leaves were immersed in phosphate buffer (Mock), ECC bacterial suspension [1.0 × 10³ colony‐forming units (cfu)/mL] (ECC) and bacterial suspension of ECC containing 10 μm diphenylene iodonium (ECC + DPI) for 8 h (D). Wt, wild‐type.

Figure 5

The plant ferredoxin‐like protein (PFLP)‐mediated resistance was altered by inhibitors. The rosette leaves of non‐transgenic (Wt) and transgenic (PFLP‐2) Arabidopsis were immersed in a suspension of Erwinia carotovora ssp. carotovora (ECC) containing phosphate buffer (Mock), 10 μm diphenylene iodonium (DPI), 1‐l‐trans‐epoxysuccinyl‐leucylamido‐(4‐guanidino)‐butane (E64), benzyloxycarbonyl‐Val‐Ala‐Asp‐fluoromethylketone (z‐VAD‐fmk), myriocin and fumonisin, and incubated at 28 °C and shaken at 100 rpm. The photograph was taken at 24 h post‐incubation.

Figure 6

Inoculation of the harpin mutant strain in the plant ferredoxin‐like protein (PFLP)‐transgenic plants. Non‐transgenic Arabidopsis (Wt) and the transgenic lines (PFLP‐1 and PFLP‐2) were sprayed with bacterial suspension of the harpin‐defective strain, ECC AC5082 [1 × 10⁵ colony‐forming units (cfu)/mL]. The photograph was taken at 48 h post‐inoculation (A). The electrolyte leakage of plants was analysed at 0, 8, 16, 24 and 36 h post‐inoculation. Data are presented as the mean ± standard error of the mean for six independent determinations (B).

Figure 7

Inoculation by Erwinia carotovora ssp. carotovora (ECC) of the plant ferredoxin‐like protein (PFLP)‐transgenic resistant‐defective derivatives. The rosette leaves of non‐transgenic Arabidopsis (Wt), PFLP‐transgenic plants (PFLP‐1, PFLP‐2), the npr1 (nonexpresser of pathogenesis‐related gene 1) mutant and the npr1/PFLP transgenic lines (npr1/PFLP‐1, npr1/PFLP‐8 and npr1/PFLP‐9) were detached and immersed in a suspension of ECC. The photograph was taken at 24 h post‐incubation (A). The detached leaves prepared from NAHG‐transgenic Arabidopsis and its double transgenic lines (NAHG/PFLP‐4, NAHG/PFLP‐7 and NAHG/PFLP‐22) (C), or eds1 (enhanced disease susceptibility 1) mutant and its PFLP‐transgenic lines (eds1/PFLP‐3, eds1/PFLP‐5 and eds1/PFLP‐7) were treated as described previously (E). The degree of maceration was estimated using a spectrophotometer at OD ₆₀₀ (optical density at 600 nm). Data are presented as the mean ± standard error of the mean for six independent determinations (B, D, F). hpt, hours post‐treatment.

Figure 8

The protein accumulation of ferredoxin (Fd) varied with salicylic acid (SA), methyl jasmonate (MeJA), harpin and Erwinia carotovora ssp. carotovora (ECC) treatment. The crude extract proteins (10 μg) isolated from treated plants were identified by Western blot analysis with antiserum against plant ferredoxin‐like protein (PFLP) and monoclonal antiserum against actin (Ac) (A). The 10 μg of crude extracts of ECC‐treated plants were identified by Western blot analysis with antiserum against PFLP (PFLP). The sodium dodecylsulphate‐polyacrylamide gel electrophoresis (SDS‐PAGE) stain with Coomassie brilliant blue (CB) was served as loading control (B). The 3 μg of crude extracts isolated from harpin‐treated plants were identified as described previously (C). The number indicated is the relative ratio of the PFLP signal after normalization. hpi, hours post‐inoculation; hpt, hours post‐treatment.