Growth phase regulation of Vibrio cholerae RTX toxin export

Abstract

Vibrio cholerae, the causative agent of the severe diarrheal disease cholera, secretes several "accessory" toxins, including RTX toxin, which causes the cross-linking of the actin cytoskeleton. RTX toxin is exported to the extracellular milieu by an atypical type I secretion system (T1SS), and we previously noted that RTX-associated activity is detectable only in supernatant fluids from log phase cultures. Here, we investigate the mechanisms for regulating RTX toxin activity in supernatant fluids. We find that exported proteases are capable of destroying RTX activity and may therefore play a role in the growth phase regulation of toxin activity. We determined that the absence of RTX toxin in stationary-phase culture supernatant fluids is also due to a lack of toxin secretion and not attributable to solely proteolytic degradation. We ascertained that the T1SS apparatus is regulated at the transcriptional level by growth phase control that is independent of quorum sensing, unlike other virulence factors of V. cholerae. Additionally, in stationary-phase cultures, all RTX toxin activity is associated with bacterial membranes or outer membrane vesicles.

FIG. 1.

The rtx locus of V. cholerae contains six genes in two divergently transcribed operons. (A) Schematic illustration of the rtx locus of V. cholerae. CHP, conserved hypothetical protein. (B) The rtx intergenic region contains 411 bp that lie between the rtx operon and the rtxBDE operon. The transcriptional start site of the rtxBDE transcript was determined by 5′ RACE analysis and is underlined. The putative −35 and −10 sites are underlined and were determined by visual inspection of the promoter region. The discriminator region is double underlined. Bases corresponding to promoter trim-backs are in bold and labeled above.

FIG. 2.

RTX toxin-associated activities are absent in supernatant fluids from stationary-phase cultures. Supernatant fluids collected from a culture of strain KFV43 (N16961ΔhapA) grown to various optical densities were assayed for RTX activity via actin cross-linking assay. LB was used as a mock treatment. Monomer (M), dimer (Di), trimer (Tri), and tetramer (Te) forms of actin are labeled at right. Numbers indicate molecular masses in kilodaltons.

FIG. 3.

Deletion of toxS alone or in combination with hapA reduces protease activity in strain Bah1. Secreted proteases in overnight culture supernatant fluids from strains Bah1 (E7946Δcore), BMV5 (Bah1ΔtoxS), BMV6 (BMV5ΔhapA), and Bah1P (Bah1ΔhapA) were detected by skim milk overlay zymogram. White areas on the gel denote protease activity. Numbers indicate molecular masses in kilodaltons.

FIG. 4.

Protease-containing supernatant fluids inactivate RTX toxin activity. Supernatant fluids were collected and filtered to remove bacteria. A total of 0.25 ml of filtered supernatant fluid containing active RTX toxin (from log phase [A₆₀₀ = 0.4] E7946mpc cultures) was mixed with 0.25 ml of LB or supernatant fluid from overnight cultures of the indicated strain and incubated at 37°C for 30 min. Treated supernatant fluids were added to HEp-2 cells, and cells were processed for the detection of actin cross-linking. LB was used as a mock treatment. Monomer (M), dimer (Di), trimer (Tri), and tetramer (Te) forms of actin are labeled at right. Numbers indicate molecular masses in kilodaltons.

FIG. 5.

RTX toxin-associated activities are absent in supernatant fluids but present in cell lysates from stationary-phase cultures. Equivalent numbers of bacteria of strain KFV43 (N16961ΔhapA) grown to various cell densities were collected and lysed. Filtered supernatant fluids or cell lysates were then evaluated for RTX activity by actin cross-linking assay. Luria broth and 2 mM Tris, 1 mM EDTA, pH 8.0 (TE), were used as mock treatments. Monomer (M), dimer (Di), trimer (Tri), and tetramer (Te) forms of actin are labeled at right. Numbers indicate molecular masses in kilodaltons.

FIG. 6.

The T1SS for export of V. cholerae RTX toxin is regulated by growth phase. (A) Equivalent numbers of bacteria from overnight cultures or log phase cultures (A₆₀₀ = 0.4) of V. cholerae El Tor strain N16961 (wild type), BBV16 (N16961 ΔhapA rtxB::Km), BBV21 (ΔhapA ΔrtxB), or CW149 (ΔhapA ΔrtxE) were collected and boiled in 1× SDS buffer. Proteins were separated on an 8% SDS-PAGE gel, and RtxE was subsequently detected by Western blotting. (B) Agarose gel analysis of RT-PCR of the rtxB, rtxD, and rtxE genes that encode the T1SS. RNA was isolated from wild-type N16961 grown to the indicated optical densities. recA was used as an internal control. (C) Relative expression of T1SS gene mRNA transcripts in log (A₆₀₀ = 0.4) and stationary (A₆₀₀ = 5) phases. Values are representative of at least two experiments.

FIG. 7.

β-galactoside activity produced by an N16961 strain containing an rtxBDE promoter::lacZ reporter plasmid peaks during log growth. (A) Growth curves of strains assayed for β-galactosidase activity. (B) N16961ΔlacZ strains containing a full-length rtxBDE promoter::lacZ reporter plasmid or a negative control plasmid containing only lacZ and no portions of the rtxBDE promoter (no pro) were assayed for β-galactosidase activity at various time points.

FIG. 8.

RTX toxin is regulated by growth phase, and activity in stationary-phase cultures is associated with the bacterial membrane. (A) Agarose gel analysis of RT-PCR of rtxA transcripts in RNA isolated from wild-type N16961 grown to the indicated optical densities. recA was used as an internal control. (B) HEp-2 cells were treated with filtered supernatant fluids (sup), total cell lysates (total lys), cytoplasm (cyto), or membrane (mem) fractions from log (A₆₀₀ = 0.4) or stationary-phase cultures (A₆₀₀ = 3.0) of strain KFV43 (N16961 ΔhapA) or BBV16 (N16961 ΔhapA, rtxB::Km). LB was used as a mock treatment. Monomer (M), dimer (Di), trimer (Tri), and tetramer (Te) forms of actin are labeled at right. (C) The separation of total cell lysate into cytoplasmic and membrane fractions was ensured by assaying for the cytoplasmic marker β-galactosidase, shown in Miller units (MU), and by Western blotting for the inner membrane protein EspL. Numbers indicate molecular masses in kilodaltons.

FIG. 9.

Model of growth phase regulation of RTX toxin export as described in the text.