Figure 6. Gut colonization phenotypes of different gene-deletion mutants of F. perrara.

(A) Colonization levels were assessed 10 d after inoculation by counting colony-forming units (CFUs) in dilutions of homogenized bee guts plated on Brain Heart Infusion (BHI) agar. Only the pylorus and ileum section of the gut were analyzed. Limit of detection (LOD) corresponds to the lowest colonization level detectable in our assay, that is, points below the LOD correspond to bees for which no CFUs were detected. Statistically significant differences of the colonization levels of each mutant relative to the wt of F. perrara were determined using the Wilcoxon rank-sum test with BH correction. Bees were inoculated with an OD₆₀₀ of 0.1. Data come from two independent experiments. Figure 6—figure supplement 5 shows the data points by experiments. *p<0.05, **p<0.01, ***p<0.001. Filled circle colors indicate whether a scab was detected during dissection (green = scab; yellow = no scab). (B) Location within the pylorus was assessed using FISH microscopy. Bees were inoculated with different F. perrara genotypes at OD₆₀₀ = 0.1, guts were dissected at day 10 after inoculation and sectioned using a microtome. Hybridizations were done with probes specific for F. perrara (magenta). DAPI counterstaining of host nuclei and bacteria is shown in blue. Images were generated by merging brightfield, F. perrara and DAPI images that were obtained for the same section of the gut. The composite images here shown were obtained by merging the images of each channel presented in Figure 6—figure supplements 6 and 7. These were obtained using the ×5 and ×40 objectives of the Zeiss LSM900. Scale bar for images obtained with ×5: 100 µm, for ×40: 20 µm. Figure 6—source data 1 contains the numeric values used to generate (A).

Figure 6—figure supplement 1. Scheme of the gene-deletion strategy based on a two-step homologous recombination procedure.

(A) A non-replicative plasmid pKS2 integrates via homologous recombination of one of the two cloned ‘homology regions’ (HRs) upstream or downstream of the gene that is targeted for deletion. (B) In a second step, a replicative plasmid harboring the restriction enzyme I-SceI is transformed. I-SceI targets corresponding recognition sites located on pKS2 resulting in the selection of either revertant (wt genotypes) or mutant genotypes that underwent a second homologous recombination event in the region targeted for deletion. PCR screening and replica plating on different selective allows to identify to correct clones.

Figure 6—figure supplement 2. Growth curves for the different Frischella strains.

(A) The six gene-deletion mutants, ihfA*, and wt strains were diluted to OD₆₀₀ = 0.05 and grown in Brain Heart Infusion (BHI) under anaerobic conditions at 35°C with continuous agitation. Absorbance was measured every 20 min for 48 hr. Per strain, four technical replicates were performed. (B) Max growth curve was calculated using the R package ‘growthcurver.’ For the hcp2 mutant, only three technical replicates were considered due to a contamination in one of the technical replicates. Figure 6—source data 1 contains the numeric values for the figures shown here.

Figure 6—figure supplement 3. Single-cell imaging of F. perrara strains.

(A) The six gene-deletion mutants, ihfA*, and wt F. perrara strains were grown in liquid Brain Heart Infusion (BHI), diluted to OD₆₀₀ = 0.1, plated in agar patches and imaged using a Nikon Ti inverted light microscope. Images were taken with a ×100 objective. Scale bar indicates 20 µm. (B) Cell length was quantified using the MicrobeJ plugin of ImageJ. (C) Electron microscopy images were obtained for the wt, ihfA*, and ΔpilE strains. Scale bar indicates 200 nm. Figure 6—source data 1 contains the numeric values for the figures shown here.

Figure 6—figure supplement 4. Correspondence between OD and colony-forming unit (CFU) for F. perrara genotypes.

For each F. perrara strain, a bacterial solution at OD₆₀₀ = 0.1 was prepared, serially diluted, and plated in BHIA medium. The number of CFUs present in 5 µl of solution at OD₆₀₀ = 0.1 was calculated based on the counts obtained from the serial dilutions. Statistics were calculated using a linear model with negative binomial distribution: ***p<0.0001, **p<0.001. Figure 6—source data 1 contains the numeric values for the figures shown here.

Figure 6—figure supplement 5. Same graph as shown in Figure 6 but data points are labeled by experiment.

Figure 6—source data 1 contains the numeric values for the figures shown here.

Figure 6—figure supplement 6. F. perrara colonization of the pylorus.

Composite images are the same as in Figure 6. These were obtained by merging the brightfield, DAPI, and F. perrara probe individual images. Images were obtained with the ×5 objective of the Zeiss LSM900. Hybridizations were done with probes specific for F. perrara (magenta). DAPI counterstaining of host nuclei and bacteria is shown in blue. The scale bar correspons to 100 µm.

Figure 6—figure supplement 7. F. perrara colonization of the pylorus.

Composite images are the same as in Figure 6. These were obtained by merging the brightfield, DAPI and F. perrara probe images. Images were obtained with the ×40 objective of the Zeiss LSM900. Hybridizations were done with probes specific for F. perrara (magenta). DAPI counterstaining of host nuclei and bacteria is shown in blue. The scale bar corresponds to 20 µm.