Figure 1.
Pie chart representing the specific requirements for V. cholerae genes.
This data represents fitness values calculated using Tn-seq on a near-saturating transposon insertion library used in 4 independent experiments, 3 different environmental conditions, and 9–21 biological replicates for each condition.
Figure 2.
Flow diagram of experimental design.
A library of mTn5 transposon mutants of V. cholerae was used to infect infant rabbits. After 12 hrs of infection, the host-passaged V. cholerae were collected, pelleted, and resuspended in pond water for 48 hrs at 30°C. In addition, the library was also passaged in LB broth overnight and put directly into pond water for 48 hr at 30°C. After each selective condition the following steps were done: 1. Bacteria were minimally outgrown to a high titer in LB. 2. gDNA was prepared and sheared. 3. Poly-C tails were added to the 3′ ends using Terminal deoxynucleotidyl Transferase. 4. The transposon-genome junctions were amplified by PCR. 4. Barcodes were added to each sample by a second PCR. 5. Multiplexed samples were subjected to massively parallel sequencing.
Table 1.
Summary of data analysis.
Figure 3.
A correlation plot of fitness values from host-passaged V. cholerae.
Each dot represents the fitness values for one gene in the two host compartments shown on the x- and y-axes. Only genes that had calculated fitness values in both conditions were plotted, which represents 93% of all non-essential genes in V. cholerae. A linear regression analysis was used to determine the correlation.
Table 2.
Functional classification of genes important for infection and dissemination*.
Figure 4.
Growth-phase dependent survival of V. cholerae in pond water.
(A) Competition assays in pond water. Either a wild type or a ΔlacZ strain were grown overnight to stationary phase and competed 1∶1 with the test strain in pond water for 2 or 8 days as indicated at 30°C. The first column is a control competition between the wild type and ΔlacZ stationary phase cells. Exponential-phase cultures were collected at OD600 = 0.3. Host passaged wild type was collected from the cecal fluid of infant rabbits 12 hrs post-infection. Each dot represents a biological replicate, and open squares indicate the limit of detection. Both the exponential phase (P value = <0.0001) and host-passaged (P value = 0.01) at 8 days are statistically different than 1.0. (B) Survival comparison in pond water. Stationary phase, exponential phase and host-passaged V. cholerae were pelleted, washed and resuspended in pond water. The pond cultures were placed standing at 30°C for 6 days and the CFU/ml was plated each day. At least 6 biological replicates were used for each time point. The input CFU at time zero was used to determine 100% survival. The dotted line indicates the limit of detection.
Figure 5.
Validation of Tn-seq results by gene deletion.
(A) Schematic of FRT scar use for high throughput sequencing. In-frame gene deletions were made using FRT/FLP recombination. Nested PCR using reverse primers 1 and 2 and a forward primer (not shown) was used to amplify the junction between the deletion scar and the gDNA. The sequencing primer is shown. (B) Correlation plot of fitness values of deletions compared to original Tn-seq screen. Each dot represents the fitness value and standard error (SE) for one gene in both screens shown on the x- and y-axes. Shown are paired fitness values from 35 gene deletions tested in 4 different conditions: LB, Cecal fluid, Small Intestine and Pond. A linear regression analysis was used to determine a correlation. (C) Validation of Frt-marked deletions in a 1∶1 competition with a ΔlacZ wild-type for 48 hr in pond water. Median values are indicated. Using a one sample t-test all CI values were significantly different than 1.0 (P≤0.001).
Figure 6.
The VarS/VarA two component system regulates glycogen storage.
(A) Fitness values obtained from the Tn-seq screen. The data represents the mean and standard error for each gene in rich media (LB), cecal fluid (CF), and pond. (B) Competition assay in pond water. A Frt-marked deletion of varS (ΔvarS) and a suppressor strain (ΔvarS**) were competed 1∶1 with ΔlacZ wild-type for 48 hr in pond water. Median values are indicated. Using a one sample t-test the CI for ΔvarS was significantly different than 1.0 (P≤0.0001). (C) Glycogen was measured in WT, ΔvarS, and ΔvarS** after growth on Kornberg Media by enzymatic hydrolysis of glycogen into glucose monomers. Liberated glucose was subsequently measured using the tetrazolium blue reducing sugar assay. The mean and standard deviation of two independent experiments each performed in triplicate are shown. Using a two sample t test, the CI of ΔvarS was significantly different than wild-type (<0.0001). The assay control was treated identically to wild-type but does not contain the enzymatic hydrolysis step and therefore measures the amount of background glucose present in WT cells.