The bacterial strains and plasmids used in this study are listed in Table 1. The primer sequences used in this study are listed in Table 2.

The frozen bacterial stocks of R. anatipestifer ATCC 11845 were cultured on sheep blood agar plates for 16–18 h at 37°C. Then, the cells from a single colony were inoculated to GC broth (GCB). GCB medium was prepared by supplementing 1 L of H₂O with 1.5% peptone (Oxoid, China), 0.4% K₂HPO₄ (Sigma), 0.1% KH₂PO₄ (Sigma), and 0.5% NaCl (Sigma) plus 1% Kellogg’s supplements I and 0.1% Kellogg’s supplements II as described in a previous study (Liu et al., 2017b). Kellogg’s supplements I containing 40% glucose (Sigma), 1% L-glutamine (Sigma), and 0.002% vitamin B1; Kellogg’s supplements II containing 20 mM Fe (NO₃)₃. Kellogg’s supplements I was sterilized by a 0.45 μm filter, others were sterilized using an autoclave. GCB agar plates were prepared by GCB supplementation with 1.5% agar. Escherichia coli (E. coli) strains were routinely cultured in LB liquid medium or on LB agar plates at 37°C. Antibiotics were added at the following final concentrations: 1 μg/ml erythromycin (Erm), 1 μg/ml cefoxitin (Cfx), 20 μg/ml kanamycin (Kana), and 50 ng/m1 streptonigrin for R. anatipestifer and 100 μg/ml ampicillin (Amp) for E. coli.

Transforming DNA (tDNA) for natural transformation experiments was amplified from the strain R. anatipestifer ATCC 11845 ΔRA0C_1551 using the primers RA0C_1551 upP1 and RA0C_1551 downP2. The tDNA contains the upstream of RA0C_1551, the Erm antibiotic resistance cassette, and the downstream of RA0C_1551. The strain R. anatipestifer ATCC11845ΔRA0C_1551 was constructed by transformation-mediated recombination in our previous study (Liu et al., 2017b). The fragments were purified using the TianGEN Extract II Kit (TIANGEN, Beijing, China). The concentration of the fragments was measured by a NanoDrop 2000 spectrophotometer.

The standard natural transformation assay was performed as described previously (Liu et al., 2017b). Briefly, the bacteria were grown to the exponential phase (OD₆₀₀ = 1.5–2.0) under aerobic conditions with shaking at 37°C. Then, the bacteria were harvested and resuspended to an OD₆₀₀ of 1 in the GCB medium. Then, 0.3 ml of bacterial cells were transferred to sterilized tubes, and 1 μg of tDNA fragments was added. The transformation was allowed to proceed at 37°C for 1 h. Unabsorbed DNA was removed by washing, and then, the bacteria were plated on GCB plates containing 1 μg/ml Erm to obtain the transformants and on GCB plates to obtain the total viable bacteria, respectively. Transformation frequencies were calculated as the number of transformants divided by the total viable bacteria (Kristensen et al., 2012).

The exponential phase R. anatipestifer ATCC 11845 cultures (OD₆₀₀ = 1.5–2.0) were harvested, washed, and then resuspended to an OD₆₀₀ of 1 in fresh GCB medium without one of the following: glucose, L-glutamine, vitamin B1, Fe (NO₃)₃, NaCl, phosphate (K₂HPO₄ and KH₂PO₄), or peptone. Following preincubation at 37°C for 1 h with shaking, the bacterial suspensions (0.3 ml) were used to perform the natural transformation in these nutrient-limited medium. In parallel, the GCB medium without peptone or phosphate was supplemented with different concentrations of peptone (0.5, 1 and 1.5%) or phosphate (0.1–0.5%). Then, bacteria in exponential growth were harvested by centrifugation, washed, and resuspended to an OD₆₀₀ of 1 in these media. After preincubation for 1 h at 37°C with shaking, 0.3 ml of bacterial suspensions was used to perform the natural transformation in these media, respectively. In addition, sterile H₂O was supplemented with 1.5% peptone, 0.5% phosphate (0.4% K₂HPO₄ and 0.1% KH₂PO₄), 0.5% NaCl, 0.4% glucose, 0.01% L-glutamine, 0.00002% VB₁, 20 μM Fe (NO₃)₃, 1.5% peptone plus 0.5% phosphate, or 0.5% NaCl plus 0.5% phosphate, respectively. Then, the exponentially growing R. anatipestifer ATCC 11845 cells in GCB were collected by centrifugation, washed, and resuspended to an OD₆₀₀ of 1 in these media. Following 1 h of preincubation, the bacterial suspensions (0.3 ml) were used to perform natural transformation in these media, respectively.

The bacteria were grown to an exponential phase in the GCB medium (OD₆₀₀ = 1.5–2.0) and then resuspended to an OD₆₀₀ of 1 in the GCB medium supplemented with 100 μM 2,2′-dipyridyl (Dip), which is able to chelate iron. In parallel, the bacterial suspensions (containing 100 μM Dip) were divided into five fractions, and 0, 2, 4, 20, and 120 μM Fe (NO₃)₃ were added, respectively. Following 1 h of preincubation, 0.3 ml of the bacterial suspension was used to perform natural transformation in these media, respectively.

The in vitro growth rates of R. anatipestifer ATCC 11845 under the experimental conditions described above were determined at OD₆₀₀ in the spectrophotometer (Eppendorf Biophotometer, Germany). Briefly, the overnight cultured bacterial cells were inoculated into 20 ml of the fresh medium at OD₆₀₀ = 0.05 and incubated at 37°C with shaking at 180 rpm. The OD₆₀₀ value was measured every 2 h for 14 h. Cultures were diluted to bring the OD600 at ∼0.5 when measured in 1.0 cm path length cuvettes.

The knockout plasmids pOES:tonBA, pOES:tonBB, and pOES:tbfA were constructed using the primers listed in Table 2. The construction of tonB mutants was performed as described in a previous study (Liu et al., 2018). For complementation, the pLMF03 derivatives were transformed into relevant R. anatipestifer ATCC 11845 strains by conjugation as described previously (Liu et al., 2016).

The overnight cultured bacterial cells were inoculated into fresh GCB liquid medium at an OD₆₀₀ of 0.05. The cultures were incubated at 37°C with shaking (180 rpm) until they reached an OD₆₀₀ = 1.5–2.0, and then, the bacteria were harvested by centrifugation, washed, and resuspended in PBS to be adjusted to an OD₆₀₀ = 0.5. The total viable bacteria of 1 ml were counted by plating on the GCB plate. The final concentration of 50 ng/ml streptonigrin was added to 1 ml of the bacterial suspension, and the mixture was incubated at 37°C for 30 min. The incubated samples were washed and diluted serially to 10^–5, 10^–6, and 10^–7 with PBS; 100 μl of each dilution was spread onto GCB agar plates, and the viable bacteria were counted after incubation overnight at 37°C. The survival rate was calculated by the number of surviving bacteria divided by the total number of viable bacteria.

R. anatipestifer ATCC 11845 was grown in the GCB medium starting at OD₆₀₀ = 0.05 at 37°C with shaking at 180 rpm. When the cultures reached the exponential growth phase (OD₆₀₀ = 1.5–2.0), half of the bacteria were transferred to the GCB medium containing 100 μM Dip. After 1 h of incubation with shaking, the bacterial cells in GCB and GCB with 100 μM Dip were harvested for RNA extraction. Similarly, the exponentially bacterial cells were transferred to the GCB medium without phosphate, peptone, or phosphate plus peptone and incubated for 1 h at 37°C. Then, the cells were collected for RNA extraction. For the analysis of the transcription level of tonBs, the wild-type, tonBA mutant, tonBB mutant, and tbfA mutant strains were grown in the GCB medium until they reached the exponential growth phase (OD₆₀₀ = 1.5–2.0) and were harvested for total RNA extraction. Total RNA extraction was performed using the RNeasy Protect Bacteria Mini Kit (QIAGEN, Cat. Number 74524) as described in a previous study (Liu et al., 2016). cDNA was synthesized from each RNA sample according to our recent study (Liu et al., 2017b). Real-time PCR assays were conducted using the primers listed in Table 2. Quantitative PCR was performed on samples in triplicate using the standard curve protocol in which the calibration curve was generated using serial fivefold dilutions of 100 ng of total cDNA. The RNA quantity was normalized using a probe specific for the 16S rRNA gene. Quantitative measurements were performed on biological samples in triplicate.

Statistical analyses were performed using the GraphPad Prism 6 (GraphPad software, CA, United States) for Windows. Statistical significance was evaluated using the Student’s t-test, one-way ANOVA, or two-way ANOVA. p < 0.05 represents statistically significant differences. Error bars in all figures represent the standard deviations of three independent experiments.

Since R. anatipestifer is naturally competent in all growth phases, and the transformation frequency is highest in the exponential growth phase (Liu et al., 2017b) and cell density do not play a significant role in competence development (data not shown), we suspected that the nutritional environment affects bacterial natural transformation. Next, we studied the effect of nutrient components in the GCB medium on the natural transformation of R. anatipestifer ATCC 11845, since the GCB medium promotes optimal competence and DNA uptake (Liu et al., 2017b, 2018; Liu M. et al., 2019).

First, we subjected cells from exponential phase cultures to resuspension in the GCB medium without glucose, L-glutamine, vitamin B1, Fe (NO₃)₃, NaCl, phosphate, or peptone to perform natural transformation as described in section “Materials and Methods.” As shown in Figure 1A, no significant differences in the transformation frequency of R. anatipestifer ATCC 11845 were observed in the GCB medium without glucose, L-glutamine, vitamin B1, Fe(NO₃)₃, or NaCl. In contrast, the transformation frequency was significantly decreased when phosphate or peptone was removed (Figure 1A). To rule out the possibility of bacterial death, we also assessed the viability and growth in these nutrient-restricted media and found that R. anatipestifer ATCC 11845 survived well in these nutrient-restricted media during natural transformation assay (Figure 1B), but they did not grow in the GCB medium without vitamin B1, peptone, or phosphate (Figure 1B). Specifically, R. anatipestifer ATCC 11845 was observed to grow well in the GCB medium without Fe(NO₃)₃ (Figure 1C), suggesting that it was not an iron-limited medium. To create the iron-limited condition, the iron chelator 2,2’-dipyridyl (Dip) was added in the GCB medium without Fe(NO₃)₃. Thus, we compared the natural transformation frequency of R. anatipestifer ATCC 11845 in the GCB medium and GCB medium supplemented with 100 μM Dip as described in section “Materials and Methods.” As shown in Figure 1D, the transformation frequency decreased by approximately 100-fold when Dip was added. In addition to chelating Fe²⁺, Dip, as a divalent cation chelator, can also chelate other cations. To determine if the effect was caused by iron, we further performed natural transformation in GCB (100 μM Dip) supplemented with 0, 2, 4, 20, and 120 μM Fe (NO₃)₃, respectively. The result showed that the transformation frequency was gradually restored, suggesting that the natural transformation of R. anatipestifer ATCC 11845 requires iron. However, even though a high concentration of Fe (NO₃)₃ was added, the transformation frequency could not be completely restored to the original level. It was suggested that Dip may chelate other cations related to natural transformation (Figure 1D). We also measured the effect of Dip on the viability and growth of R. anatipestifer ATCC 11845, and the results showed that Dip did not affect bacterial survival during natural transformation assay (Figure 1E). However, the bacterial cells did not grow in the GCB medium supplemented with 100 μM Dip (data not shown). Thus, we chose to culture R. anatipestifer ATCC 11845 in the GCB medium containing 60 μM Dip and found that the growth of bacteria was significantly inhibited but gradually restored when Fe (NO₃)₃ was added (Figure 1F). These results indicated that phosphate, peptone, and iron affected the natural transformation of R. anatipestifer ATCC 11845 in the GCB medium. In contrast, depletion of vitamin B1, which is essential for the growth of R. anatipestifer ATCC 11845, did not have any effect on the natural transformation.

To further investigate the effect of phosphate and peptone on the natural transformation frequency of R. anatipestifer ATCC 11845, we transformed the tDNA into exponentially growing bacterial cells in the GCB medium depleted for or supplemented with different concentrations of phosphate or peptone, respectively. As shown in Figure 2A, the transformation ability of R. anatipestifer ATCC 11845 was gradually restored by the addition of 0.1 and 0.5% phosphate to the GCB medium without phosphate. To exclude the possibility that phosphate might affect the pH, we measured the pH value of the phosphate-free GCB medium and GCB medium supplemented with different concentrations of phosphate and found that the pH did not change significantly in these media (data not shown). These results implied that the effect of phosphate on the natural transformation was not due to changes in the pH. In addition, we also assessed the bacterial growth and the result showed that the growth of R. anatipestifer ATCC 11845 could be restored when phosphate was added, and when the concentration of phosphate was 0.5%, the growth could return to its original level (Figure 2B). Similarly, as shown in Figure 2C, the lowest concentration of peptone tested (0.5%) was able to restore the transformation ability of R. anatipestifer ATCC 11845. In addition, the growth was partially restored by adding 0.5 or 1% peptone and was completely restored by adding 1.5% peptone (Figure 2D).

We also measured whether iron, phosphate, and peptone have additive effects on the natural transformation of R. anatipestifer ATCC 11845. As shown in Supplementary Figure 1A, compared with the GCB medium without phosphate or peptone, the natural transformation frequency was decreased significantly when both phosphate and peptone were removed. Similarly, the natural transformation frequency was decreased significantly in the GCB medium without peptone when 100 μM Dip was added (Supplementary Figure 1B).

Next, to further verify the function of GCB components in natural transformation, we investigated whether a single component of GCB medium has an effect on the natural transformation of R. anatipestifer ATCC 11845. As described in section “Materials and Methods,” the transformation experiment was performed in sterile H₂O and sterile H₂O supplemented with the single component glucose, L-glutamine, vitamin B1, Fe (NO₃)₃, NaCl, phosphate, or peptone, respectively. The results showed that the natural competence of R. anatipestifer ATCC 11845 cannot occur in sterile H₂O and sterile H₂O supplemented with glucose, L-glutamine, vitamin B1, Fe (NO₃)₃, or phosphate, respectively. However, the transformants were detected after performing the natural transformation in sterile H₂O supplemented with peptone or NaCl (Figure 3A), although only 20 ± 5 transformants per ∼10⁹ CFU were detected when the natural transformation was performed in sterile H₂O supplemented with NaCl. In parallel, the bacterial cells survived well in these media during natural transformation (Supplementary Figure 2).

Although phosphate plays a crucial role in the natural transformation of R. anatipestifer ATCC 11845 in the GCB medium, we did not observe the occurrence of natural transformation in sterile H₂O supplemented with phosphate (Figure 3A). This prompted us to further investigate the natural transformation of R. anatipestifer ATCC 11845 in sterile H₂O containing peptone or NaCl by adding phosphate. The results showed that the transformation frequency increased significantly after the addition of phosphate in sterile H₂O with peptone or NaCl (Figure 3B). As a control, the bacterial cells survived well in these media during the natural transformation but did not have any growth (data not shown).

To further explore the mechanism affecting the natural transformation in R. anatipestifer ATCC 11845 in iron-limited GCB medium and in GCB medium without phosphate or peptone, we measured the transcription of conserved natural transformation genes in most competent bacteria, such as dprA, comEC, recA, and comM, which are involved in transporting ssDNA to the cytoplasm and facilitating RecA loading on internalized ssDNA and assisting homologous recombination (Johnston et al., 2014). The deletion of dprA, comEC, recA, and comM abolished the natural transformation of R. anatipestifer ATCC 11845 (Huang et al., 2019) (our unpublished data). As shown in Figure 4A, the transcription of dprA and comEC was significantly downregulated (twofold) under iron-limited conditions, whereas that of recA and comM was not changed significantly under the same conditions, indicating that the decreased transformation frequency of R. anatipestifer ATCC 11845 under iron-limited conditions could be related to lower dprA and comEC expression. Interestingly, the levels of dprA and comEC transcripts in the GCB medium without phosphate were observed to be increased by two and threefold, respectively, and those of recA and comM had almost no change (Figure 4B). This result suggested that the effect of phosphate on natural transformation was not caused by changes in dprA, comEC, recA, or comM transcription levels. Moreover, in the GCB medium without peptone, only the transcription of dprA was significantly downregulated (twofold), whereas that of comEC, recA, and comM showed no statistically significant changes (Figure 4C). As evidence, the qRT-PCR results showed that the transcription of comEC was upregulated (twofold), whereas that of dprA, recA, and comM were not significantly different in the GCB medium without phosphate and peptone (Figure 4D). These findings suggested that iron, phosphate, and peptone may act through different mechanisms or at different stages of the natural transformation process.

Since TonB of gram-negative bacteria is important for nutrient transportation, including iron (Noinaj et al., 2010), we further investigated whether TonB was involved in the natural transformation of R. anatipestifer ATCC 11845. In a previous study, it was shown that both TonBA and TonBB, but not TbfA, are required for hemin uptake in R. anatipestifer (Liao et al., 2015; Liu et al., 2016). As shown in Figure 5A, compared with the wild type, the natural transformation frequency was decreased significantly in the tonBA and tonBB mutants; however, the tbfA mutation did not have any effect on the natural transformation of R. anatipestifer ATCC 11845. Particularly, it was observed that the tonBB mutant inhibited the natural transformation more severely than the tonBA mutant. Similarly, the tonBA–tonBB–tbfA triple mutant of R. anatipestifer ATCC 11845 exhibited a decreased natural transformation frequency than the single mutants, and the complementation of TonBA or TonBB, but not TbfA, was able to partially restore the natural transformation (Figure 5B). As a control, it was shown that tonB single mutant did not have any effect on the survival of the bacteria in the GCB medium during the natural transformation (Supplementary Figure 3).

Previous studies have established that the expression of the tonB genes of R. anatipestifer is not regulated by iron availability (Liu et al., 2017a). We then determined whether TonBB is more important or highly expressed than the other two TonBs. The transcription of tonBA, tonBB, and tbfA in R. anatipestifer ATCC 11845 were measured and the results showed that tonBB had the highest transcriptional level, and tbfA had the lowest transcriptional level (Figure 6A). To identify whether the tonB mutants influence the transcription of the other two tonBs, we further measured the transcription level of the three tonBs in each tonB single mutant and found that the tonB single mutants did not have any effect on the transcription of the other two tonBs (Figures 6B–D). It is suspected that TonBA and TonBB, but not TbfA, are involved in the natural transformation of R. anatipestifer ATCC 11845.

TonBA and TonBB are involved in the natural transformation in the GCB medium, which can be reasoned by the deletion of tonBA or tonBB decreased the intracellular iron level in R. anatipestifer ATCC 11845. To verify this hypothesis, the sensitivity of the strains to streptonigrin, which exhibits an enhanced bacterial killing in the presence of iron (White and Yeowell, 1982), was compared. The results showed that R. anatipestifer ATCC11845 ΔtonBA did not exhibit a significant difference in survivability compared to the wild-type strain when treated with 50 g/ml of streptonigrin (Figure 7). However, the survival rate of R. anatipestifer ATCC11845 ΔtonBB was significantly higher than that of the wild-type strain when treated with 50 ng/ml streptonigrin (Figure 7). The survival rate of the tonBAtonBB double mutant did not change significantly compared to that of the tonBB single mutant when treated with 50 ng/ml streptonigrin (Figure 7). These results indicated that the R. anatipestifer tonBB mutant, but not the tonBA mutant, is defective in iron uptake.

If the concentration of intracellular iron is decreased, the putative iron uptake genes will be upregulated. Thus, we further measured the transcript levels of RA0C_RS09540 and RA0C_RS09840 in R. anatipestifer ATCC11845 ΔtonBA and R. anatipestifer ATCC11845 ΔtonBB, respectively, since their homologs were upregulated in R. anatipestifer CH-1 under iron-limited conditions (Liu et al., 2017b; Huang M. et al., 2021). As shown in Supplementary Figure 4A, compared to that of the wild type, RA0C_RS09540 and RA0C_RS09840 had higher transcript levels in R. anatipestifer ATCC11845 ΔtonBB, but not in R. anatipestifer ATCC11845 ΔtonBA, when the bacteria were grown in the GCB medium. As a control, the transcripts of RA0C_RS09540 and RA0C_RS09840 in the wild type were higher in an iron-limited medium than in iron-rich medium (Supplementary Figure 4B). Taken together, these results suggest that the tonBB mutant, but not the tonBA mutant, is defective in the iron uptake in R. anatipestifer ATCC11845 when grown in the GCB medium.