Transposon sequencing reveals the essential gene set and genes enabling gut symbiosis in the insect symbiont Caballeronia insecticola

Abstract

Caballeronia insecticola is a bacterium belonging to the Burkholderia genus sensu lato, which is able to colonize multiple environments like soils and the gut of the bean bug Riptortus pedestris. We constructed a saturated Himar1 mariner transposon library and revealed by transposon-sequencing that 498 protein-coding genes constitute the essential genome of Caballeronia insecticola for growth in free-living conditions. By comparing essential gene sets of Caballeronia insecticola and seven related Burkholderia s.l. strains, only 120 common genes were identified, indicating that a large part of the essential genome is strain-specific. In order to reproduce specific nutritional conditions that are present in the gut of Riptortus pedestris, we grew the mutant library in minimal media supplemented with candidate gut nutrients and identified several condition-dependent fitness-defect genes by transposon-sequencing. To validate the robustness of the approach, insertion mutants in six fitness genes were constructed and their growth deficiency in media supplemented with the corresponding nutrient was confirmed. The mutants were further tested for their efficiency in Riptortus pedestris gut colonization, confirming that gluconeogenic carbon sources, taurine and inositol, are nutrients consumed by the symbiont in the gut. Thus, our study provides insights about specific contributions provided by the insect host to the bacterial symbiont.

Keywords: Burkholderia sensu lato; Caballeronia insecticola; Riptortus pedestris; Tn-seq; essential gene; fitness gene; genome comparison; gut symbiosis.

Figure 1

Caballeronia insecticola essential genes; circular representation of the genome of C. insecticola, the markings outside the outer circle represent genome positions (in Mb) for each chromosome or plasmid; Chr 1 is Chromosome 1, Chr 2 is Chromosome 2, Chr 3 is Chromosome 3, Pl 1 is Plasmid 1, and Pl 2 is Plasmid 2; the second and third tracks represent CDS on the forward and reverse strand, respectively; subsequent tracks 4, 5, 6, and 7 represent, respectively, the essential genes that are common to the YG, MM glucose, and MM succinate conditions, the essential genes specific for YG, MM succinate, and MM glucose media; the innermost track 8 shows the number of TA sites per 1000 bp; the Venn diagram in the center represents the number of essential genes identified for each medium: top left, YG; top right, MM glucose; bottom, MM succinate.

Figure 2

Distribution of C. insecticola essential genes in COG categories; the categories are S, function unknown; ND, not determined; R, general function prediction only; Q, secondary metabolites biosynthesis—transport—catabolism; P, inorganic ion transport—metabolism; I, lipid transport—metabolism; H, coenzyme transport—metabolism; F, nucleotide transport—metabolism; E, amino acid transport—metabolism; G, carbohydrate transport—metabolism; C, energy production—conversion; N, cell motility; O, posttranslational modification—protein turnover—chaperones; U, intracellular trafficking—secretion—vesicular transport; M, cell wall—membrane—envelope biogenesis; T, signal transduction mechanisms; V, defense mechanisms; D, cell cycle control—cell division—chromosome partitioning; L, replication—recombination—repair; K, transcription; J, translation—ribosomal structure—biogenesis; for each category, the number of genes and the percentage that they represent are indicated; top histograms indicate the whole genome of C. insecticola, center histograms indicate the essential genome of C. insecticola, and lower histograms indicate the common essential genes to eight Burkholderia s.l. species.

Figure 3

IGV plots for genomic regions containing selected C. insecticola essential genes; (A) 30S and 50S ribosomal proteins encoding region; (B) Ara4N lipid A modification gene cluster; (C) ATP synthase subunits encoding region; tracks, from bottom to top: position of TA sites, region of interest and its flanking neighbors, histogram of insertion counts at TA sites for the indicated experimental conditions, genome positions (in kb) on Chromosome 1.

Figure 4

IGV plots of genomic regions carrying condition-specific fitness genes; (A) fructose-bisphosphatase encoding gene (fbp); (B) phosphoenolpyruvate synthase encoding gene (pps); (C) malic enzyme encoding gene (maeB); (D) phosphoenolpyruvate carboxylase encoding gene (ppc); (E) chloride channel protein and potential taurine transporter encoding gene (tauT); (F) myo-inositol utilization genes including the ABC transporter permease (inoT); tracks, from bottom to top: position of TA sites; gene organization in the region of interest with fitness genes and their flanking neighbors; histograms of insertion counts at TA sites for the indicated experimental conditions; genome positions (in kb) on Chromosome 1; taurine C, taurine as carbon source; taurine N, taurine as nitrogen source; taurine S, taurine as sulfur source; taurine CNS, taurine as carbon, nitrogen, and sulfur source.

Figure 5

Growth curves of C. insecticola WT and metabolic mutants in different media; (A) MM with glucose; (B) MM with HBA; (C) MM with taurine as carbon source; (D) MM with myo-inositol; (E) MM with succinate; (F) YG medium. X-axis, time of growth in hours; Y-axis, growth measured as OD₆₀₀; error bars are standard deviation.

Figure 6

Ability of C. insecticola mutants to colonize the R. pedestris M4 midgut region; midguts were analyzed at 5 dpi; (A) colonization capacity of WT and mutant strains in single-strain infection conditions were determined by microscopy observation with an epi-fluorescence microscope; one representative image is shown for each condition; infection rate (%) indicates the proportion of infected animals with indicated strains (n = 10); (B) colonization capacity of strains in coinfection conditions; Riptortus pedestris was infected with an equal mix of mScarlett-labeled C. insecticola WT and the indicated GFP-labeled WT or mutant strains; relative abundance of the two strains in the M4 midgut regions at 5 dpi was determined by flow cytometry on dissected intestines; the competition index expresses for all samples the ratio of the indicated mutant to WT, corrected by the ratio of the inoculum, which was in all cases close to 1 (mutant bacteria/WT bacteria)/(inoculum mutant bacteria/inoculum WT bacteria); each dot represents the competition index in an individual and the mean per mutant is indicated by a horizontal black line (n = 10); CI is competition index; different letters indicate statistically significant differences (P < .05); statistical significance was analyzed by Kruskal–Wallis test, Dunn post hoc test, and Benjamini–Hochberg correction; (C) microscopy observation with an epi-fluorescence microscope of competition assays between C. insecticola WT (RFP) and mutants (GFP) as in panel B; one representative image is shown for each condition. Scale bars in panels A and C are 40 μm.