Screening and verifying proteins interacting with CheA. (a) Protein samples obtained in pull-down assay and detected by SDS/PAGE. The “bait” protein CheA on the gel was indicated. Lanes 1, 2, and 3 are samples from the control column, and lanes 4, 5, and 6 are samples from the CheA-binding column. M represents a protein marker. (b) The volcano plot shows the P-value and fold-change of all proteins identified in MS analyses. Red spots represent proteins that showed two or higher folds in the experimental group compared with the control group (P < 0.05). Blue spots represent proteins with a higher amount in the control group (P < 0.05). Grey spot proteins showed no apparent difference between the two groups (P > 0.05). (c) Detect the interaction between CheA and indicated proteins via BTH. Blue indicates protein-protein interaction in the colony after 60 h of incubation, while white indicates no protein-protein interaction. A colony containing T25a-zip and T18C-zip plasmids was used as a positive control (CK+), and a colony containing empty T25a and T18C plasmids was used as a negative control (CK-). The LacZ activities of colonies were shown above the colonies. (d) The red fluorescence intensities in BiFC assay. The results in panels c and d are the average of three independent assays. Error bars represent standard deviations. The asterisks above the column denote significant differences (**P < 0.01) between indicated strains and CK- strain. “ns.” represents none statistically significant between indicated strain and CK- strain.

CsoR and PhaD inhibit CheA autophosphorylation. (a) Chemotaxis of indicated strains on semisolid plates. Photos of colonies on the top were taken after 16 h incubation at 28°C. Diameters of colonies (swimming zone) shown at down were calculated from three replicates. The asterisks above the column denote significant differences (*P < 0.05, **P < 0.01) between indicated strains and control strain (WT+pVec) analyzed by Student’s t-test. (b) Effect of the 14 proteins on the CheA autophosphorylation. The name/ID and the concentration of tested proteins added in each lane are indicated above the gel. CK represents CheA alone in the reaction mixture. BSA is used as a negative control. The ladder represents a protein marker. (c and d) CsoR (c) and PhaD (d) affect CheA autophosphorylation. The concentration of tested proteins added in each lane is indicated above the gel. The time represents the time of the CheA autophosphorylation reaction. The SDS-PAGE gels in panels b, c, and d were detected by Coomassie Blue Staining (Above) and autoradiograph (Below). The experiments for panels b, c, and d were repeated three times, and a representative photo was shown. The relative autoradiograph intensity of the CheA band was calculated using Image J software and shown below each lane.

CheA domains involved in interacting with CsoR and PhaD. (a) Schematic diagram of CheA and truncated CheA proteins. The predicted domains are based on the Pfam database and the amino acid positions where the predicted domains start and end are shown. (b) The interaction between CheA domains and CsoR/PhaD was tested using BTH. Blue indicates protein-protein interaction in the colony after 60 h of incubation, while white indicates no protein-protein interaction. A colony containing T25a-zip and T18C-zip plasmids was used as a positive control (CK+), and a colony containing empty T25a and T18C plasmids was used as a negative control (CK-). (c) Confirmation of BTH interactions in panel B by LacZ activity assay. The results are the average of three independent assays. Error bars represent standard deviations. The asterisks above the column denote significant differences (**P < 0.01) between indicated strains and CK- strain analyzed by Student’s t-test. “ns.” represents none statistically significant between indicated strain and CK- strain.

CsoR is a metal-binding repressor. (a) MST analysis of the interaction between CsoR-GFP and metal ions. CsoR-GFP (250 nM) was incubated with increasing concentrations of metal ions. (b) Analysis of relative transcription level of target genes in wild-type (WT+pVec), csoR mutant (ΔcsoR+pVec), complemented strain (ΔcsoR+pcsoR), and cheA mutant (ΔcheA+pVec) in the presence and absence of CuCl2 (10 μM) by qRT-PCR. The results are the average of three independent assays. Error bars represent standard deviations. The asterisks represent statistically significant differences between the two compared strains (*P < 0.05, **P < 0.01). “ns.” represents none statistically significant between two compared strains. (c) Analysis for interactions between CsoR and copA-I promoter DNA using EMSA. (d) The effect of indicated metal ions on the interaction between CsoR and copA-I promoter DNA. The concentrations of CsoR and metal ions in panels c and d added in each lane are shown above the gel. Free DNA and CsoR-DNA complex are indicated.

Copper inhibits the interaction between CheA and CsoR. (a) MST analysis of the interaction between CheA-GFP and CsoR in the presence of Cu2+. CheA-GFP (160 nM) was incubated with increasing concentrations of CsoR. (b) SDS-PAGE detected protein samples obtained in a pull-down assay. The “bait” protein Strep-CheA and the “prey” protein His-CsoR on the gel were indicated. The gel showed the influence of Cu2+ on the amount of “prey” protein His-CsoR. The concentration of Cu2+ added in each pull-down assay was displayed above the gel. The SDS-PAGE gel was detected by Coomassie Blue Staining. (c) CheA autophosphorylation in the presence of CsoR and Cu2+. The tested proteins and Cu2+ concentrations added in each lane are indicated above the gel. The SDS-PAGE gel was detected by Coomassie Blue Staining (Above) and autoradiograph (Below). The relative intensity of the CsoR band in panel b and the autoradiograph intensity of the CheA band in panel c were calculated using Image J software and shown below each lane. (d) Chemotaxis of indicated strains on semisolid plates supplied with or without CuCl2. Photos of colonies on the top were taken after 16 h (for the control plate) or 18 h (for the copper-adding plate) incubation at 28°C. The diameters of colonies were measured and normalized to the diameters of WT+pVec, shown below. The asterisks above the column denote significant differences (**P < 0.01) between two indicated strains analyzed by Student’s t-test. “ns.” represents none statistically significant between two compared strains.

Role of CsoR in bacterial chemotaxis to copper. (a) Chemotaxis of indicated strains in the absence and presence of copper gradient. The chemotaxis rings and chemotaxis distance (D1/D2) were indicated by arrows. The red arrows pointed at the agar plug with or without copper in the center of the plate. The green dots represented the sites where the bacteria were initially inoculated on the semisolid plate. The assay was performed with three repeats, and a representative photo was shown. (b) RI value (D1/(D1+D2)) of indicated strains shown in panel a. (c) Aggregated trajectories of individual tested strain cells in the absence and presence of copper gradient. The tracking data presented is a composite of two experiments performed in duplicate (n = 100 cells). The overall directionality of migration is depicted in the rose diagram in the upper right corner of each single-track summary. (d) Center of mass of tested strains in the presence of copper gradients. It represents the average of all single-cell endpoints. The results of panels b and d are the average of three independent assays. Error bars represent standard deviations. The asterisks represent statistically significant differences between the two indicated strains (*P < 0.05, **P < 0.01). “ns.” represents no statistically significant between the two indicated strains. (e) Velocity analysis of indicated strains in the presence or absence of copper gradient (n = 100 cells). The lowercase letters above each bar in panel e indicate significant differences (P < 0.05).

The interaction between CheA and CsoR from indicated bacterial species. The interaction between CsoR and CheA was tested by using BTH. The LacZ activities of colonies were shown above the colonies. The results are the average of three independent assays. Error bars represent standard deviations. The asterisks above the column denote significant differences (**P < 0.01) between indicated strains and CK- strain analyzed by Student’s t-test. “ns.” represents none statistically significant between indicated strain and CK- strain.

A proposed model for describing how CsoR coordinates chemotaxis and resistance to copper in P. putida.

Under low Cu2+ levels, more none Cu2+-binding CsoR molecules (free CsoR) exist in the cell, and the free CsoR forms tetramer and binds to promoters of copper-resistance genes (such as copA-I and copA-II), leading to repressed gene transcription and low copper resistance. Meanwhile, free CsoR interacts with CheA and inhibits its autophosphorylation activity, decreasing chemotaxis ability. In contrast, more Cu2+-binding CsoR molecules exist under high Cu2+ levels, and the Cu2+-binding changes the conformation of the CsoR tetramer and releases CsoR from target promoters, leading to increased gene transcription and copper resistance. Besides, Cu2+-binding of CsoR decreases the interaction between CsoR and CheA, which relieves the inhibition of CsoR on CheA autophosphorylation, resulting in increased chemotaxis ability.