kdgREcc negatively regulates genes for pectinases, cellulase, protease, HarpinEcc, and a global RNA regulator in Erwinia carotovora subsp. carotovora

Abstract

Erwinia carotovora subsp. carotovora produces extracellular pectate lyase (Pel), polygalacturonase (Peh), cellulase (Cel), and protease (Prt). The concerted actions of these enzymes largely determine the virulence of this plant-pathogenic bacterium. E. carotovora subsp. carotovora also produces HarpinEcc, the elicitor of the hypersensitive reaction. We document here that KdgREcc (Kdg, 2-keto-3-deoxygluconate; KdgR, general repressor of genes involved in pectin and galacturonate catabolism), a homolog of the E. chrysanthemi repressor, KdgREch and the Escherichia coli repressor, KdgREco, negatively controls not only the pectinases, Pel and Peh, but also Cel, Prt, and HarpinEcc production in E. carotovora subsp. carotovora. The levels of pel-1, peh-1, celV, and hrpNEcc transcripts are markedly affected by KdgREcc. The KdgREcc- mutant is more virulent than the KdgREcc+ parent. Thus, our data for the first time establish a global regulatory role for KdgREcc in E. carotovora subsp. carotovora. Another novel observation is the negative effect of KdgREcc on the transcription of rsmB (previously aepH), which specifies an RNA regulator controlling exoenzyme and HarpinEcc production. The levels of rsmB RNA are higher in the KdgREcc- mutant than in the KdgREcc+ parent. Moreover, by DNase I protection assays we determined that purified KdgREcc protected three 25-bp regions within the transcriptional unit of rsmB. Alignment of the protected sequences revealed the 21-mer consensus sequence of the KdgREcc-binding site as 5'-G/AA/TA/TGAAA[N6]TTTCAG/TG/TA-3'. Two such KdgREcc-binding sites occur in rsmB DNA in a close proximity to each other within nucleotides +79 and +139 and the third KdgREcc-binding site within nucleotides +207 and +231. Analysis of lacZ transcriptional fusions shows that the KdgR-binding sites negatively affect the expression of rsmB. KdgREcc also binds the operator DNAs of pel-1 and peh-1 genes and represses expression of a pel1-lacZ and a peh1-lacZ transcriptional fusions. We conclude that KdgREcc affects extracellular enzyme production by two ways: (i) directly, by inhibiting the transcription of exoenzyme genes; and (ii) indirectly, by preventing the production of a global RNA regulator. Our findings support the idea that KdgREcc affects transcription by promoter occlusion, i.e., preventing the initiation of transcription, and by a roadblock mechanism, i.e., by affecting the elongation of transcription.

FIG. 1

(A) Physical map of the 7.35 kb of ClaI DNA segment of strain Ecc71 containing the kdgR_Ecc gene. The location and direction of the gene are indicated by an arrow. The ogl gene is located upstream of kdgR_Ecc, as indicated by the broken arrow. The omega (Ω) fragment (Sp resistance cassette) was introduced at the BstEII site. B, BstEII; Bg, BglII; C, ClaI; E, EcoRI; P, PstI; S, SalI; V, EcoRV. (B) Nucleotide sequence of kdgR_Ecc and the 3′-terminal region of the ogl gene of strain Ecc71. The deduced amino acid sequence of KdgR_Ecc is also given. Palindromic sequences in between the ogl and kdgR_Ecc genes are indicated by inverted arrows. Sequences similar to the −10 and −35 consensus sequences are double underlined, and the transcriptional start site is indicated by “+1”. Transcriptional termination sequences represented by an inverted repeat beyond the 3′ end of kdgR_Ecc are indicated by double-lined inverted arrows. Several restriction endonuclease sites are also shown. Numbers on the right refer to the positions of the nucleotides.

FIG. 2

Alignment of the deduced amino acid sequence of KdgR_Ecc of strain Ecc71 with those of E. chrysanthemi EC3937 (KdgR_Ech) and E. coli (KdgR_Eco). The HTH motif is shown. Identical amino acids are not identified. Dots indicate conserved substitutions.

FIG. 3

Effects of KdgR_Ecc on the production of exoenzymes and Harpin_Ecc, on the transcription of exoenzyme genes and hrpN_Ecc, and on pathogenicity. (A and B) Agarose plate assays for Peh, Prt, and Cel activities of E. carotovora subsp. carotovora strains. Strains Ecc71 (KdgR_Ecc⁺, column A1) and AC5073 (KdgR_Ecc⁻, column A2) were grown at 28°C in minimal salts medium plus sucrose to an A₆₀₀ of 2.3, and the culture supernatants (20 μl) were used for the assays of enzymatic activities. AC5073 carrying pLAFR5 (cloning vector, column B1) or pAK1024 (KdgR_Ecc⁺, column B2) was grown in minimal salts medium plus sucrose and Tc to an A₆₀₀ of 2.3, and the culture supernatants (20 μl) were used for the assays of enzymatic activities. The plates were scored for activities after incubation for 24 h at 28°C. Halos around the wells are due to enzymatic activities. (C and D) Levels of transcripts of pel-1, peh-1, hrpN_Ecc, and celV. Bacteria were grown at 28°C in minimal salts medium plus sucrose or in this medium supplemented with Tc to an A₆₀₀ of 1.0 for RNA extraction. Total RNAs from strains Ecc71 (column C1), AC5073 (column C2), AC5073 carrying pLAFR5 (column D1), and AC5073 carrying pAKC1024 (column D2) were used for Northern blot analysis. Lanes 1 and 2 in parts C and D contained 10 and 20 μg of total RNA, respectively. (E) Plant tissue maceration induced by strain Ecc71 (site 2) and its KdgR_Ecc⁻ mutant, strain AC5073 (site 1). Each inoculation site of this celery petiole was injected with 2 × 10⁸ cells. Water was used as a control (site 3). The inoculated petiole was incubated in a moist chamber at 25°C for 24 h. (F) Western blot analysis of Harpin_Ecc produced by strain Ecc71 (lane 1) and its KdgR_Ecc⁻ derivative, strain AC5073 (lane 2). Each lane contained 20 μg of total bacterial protein.

FIG. 4

Effect of kdgR_Ecc on the transcription of rsmB in E. carotovora subsp. carotovora. Bacteria were grown in minimal salts medium plus sucrose or in this medium supplemented with Tc at 28°C to an A₆₀₀ of 1.0. Total RNAs were isolated and used for Northern blot analysis. Lanes: A1, Ecc71 (KdgR_Ecc⁺); A2, AC5073 (KdgR_Ecc⁻); B1, AC5073 carrying pLAFR5 (cloning vector); B2, AC5073 carrying pAKC1024 (KdgR_Ecc⁺). Each lane contained 5 μg of total RNA.

FIG. 5

Overexpression and purification of KdgR_Ecc-6His. Crude extracts and fractionated KdgR_Ecc-6His were analyzed by SDS-PAGE in a 12% (wt/vol) polyacrylamide gel. Lanes: 1, lysate of JM109(DE3) carrying the cloning vector, pET28b(+); 2, lysate of JM109(DE3) carrying pAKC1029; and 3, purified KdgR_Ecc-6His protein. Lanes 1 and 2 contained 10 μg of protein, whereas lane 3 contained 2 μg of protein.

FIG. 6

Gel mobility shift assays for binding of KdgR_Ecc-6His to the pel-1 (A), peh-1 (B), and rsmB (C) DNAs. ³²P-labeled rsmB (1 ng), pel-1 (2 ng), or peh-1 (2 ng) DNA was used. Lanes A1, B1, and C1, no protein was added; lanes A2 and B2, reaction mixtures contained 300 ng of KdgR_Ecc-6His; lanes C2, C3, C4, and C5, reactions were carried out with 300, 400, 500, or 600 ng of purified KdgR_Ecc-6His protein, respectively; lane C6, reaction was performed with 300 ng of KdgR_Ecc-6His in the presence of 200 ng of excess of cold rsmB DNA.

FIG. 7

(A) DNase I protection analysis of the rsmB DNA fragment by KdgR_Ecc. 5′ and 3′ refer to the ³²P-end-labeled portion of rsmB DNA. In the 50 μl of binding reaction mixture, an 11.7 nM concentration of the sense strand probe (5′) or a 5.0 nM concentration of the antisense strand probe (3′) was incubated with 0 (lane 1) or with 0.16, 0.32, 0.64, 0.96, and 1.28 μM purified KdgR_Ecc-6His protein (lanes 2 to 6, respectively). The G+A chemical sequence of the same labeled DNA fragments is shown in the leftmost lanes. Brackets indicate nucleotide positions relative to the transcriptional start site, which were protected from DNase I digestion by KdgR_Ecc-6His. (B) Nucleotide sequence alignment of the protected regions of rsmB and putative KdgR_Ecc-binding sites of pel-1 and peh-1.

FIG. 8

A speculative model depicting the regulatory effects of KdgR_Ecc on the production of exoenzymes and Harpin_Ecc. The proposed scheme postulates KdgR_Ecc to function via two different pathways: by directly repressing the transcription of the exoenzyme genes, i.e., pel and peh, and by affecting the transcription of rsmB, a global RNA regulator, which controls pel, peh, cel, prt, and hrpN_Ecc expression (28). While we have documented the inhibition of pel and peh transcription by kdgR_Ecc, we do not have similar evidence for a direct effect of kdgR_Ecc on the transcription of hrpN_Ecc, cel, or prt genes. However, the data presented here show that rsmB transcription is affected by a roadblock mechanism. We propose that as the level of active KdgR_Ecc drops, rsmB transcription is stimulated, producing RNA which binds RsmA. Since RsmA promotes transcript decay, the decrease in the free RsmA pool could contribute to mRNA stability. The formation of RsmA-rsmB RNA complex also facilitates rsmB RNA processing. The processed rsmB RNA (rsmB′) then activates Pel, Peh, Cel, Prt, and Harpin_Ecc production, although the mechanism by which this is brought about is not yet fully understood.