Basic Characterization of Natural Transformation in Avibacterium paragallinarum

Natural transformation is an important mechanism for bacteria to acquire exogenous DNA molecules during the process of evolution. Additionally, it can also be used as a method to introduce foreign genes into bacteria under laboratory conditions. ABSTRACT Avibacterium paragallinarum is the pathogen involved in infectious coryza (IC), an acute infectious upper respiratory disease in chickens. The prevalence of IC has increased in China in recent years. There is a lack of reliable and effective procedures for gene manipulation, which has limited the research on the bacterial genetics and pathogenesis of A. paragallinarum. Natural transformation has been developed as a method of gene manipulation in Pasteurellaceae by the introduction of foreign genes or DNA fragments into bacterial cells, but there has been no report on natural transformation in A. paragallinarum. In this study, we analyzed the existence of homologous genetic factors and competence proteins underlying natural transformation in A. paragallinarum and established a method for transformation in it. Through bioinformatics analysis, we identified 16 homologs of Haemophilus influenzae competence proteins in A. paragallinarum. We found that the uptake signal sequence (USS) was overrepresented in the genome of A. paragallinarum (1,537 to 1,641 copies of the core sequence ACCGCACTT). We then constructed a plasmid, pEA-KU, that carries the USS and a plasmid, pEA-K, without the USS. These plasmids can be transferred via natural transformation into naturally competent strains of A. paragallinarum. Significantly, the plasmid that carries USS showed a higher transformation efficiency. In summary, our results demonstrate that A. paragallinarum has the ability to undergo natural transformation. These findings should prove to be a valuable tool for gene manipulation in A. paragallinarum. IMPORTANCE Natural transformation is an important mechanism for bacteria to acquire exogenous DNA molecules during the process of evolution. Additionally, it can also be used as a method to introduce foreign genes into bacteria under laboratory conditions. Natural transformation does not require equipment such as an electroporation apparatus. It is easy to perform and is similar to gene transfer under natural conditions. However, there have been no reports on natural transformation in Avibacterium paragallinarum. In this study, we analyzed the presence of homologous genetic factors and competence proteins underlying natural transformation in A. paragallinarum. Our results indicate that natural competence could be induced in A. paragallinarum serovars A, B, and C. Furthermore, the method that we established to transform plasmids into naturally competent A. paragallinarum strains was stable and efficient.

IMPORTANCE Natural transformation is an important mechanism for bacteria to acquire exogenous DNA molecules during the process of evolution. Additionally, it can also be used as a method to introduce foreign genes into bacteria under laboratory conditions. Natural transformation does not require equipment such as an electroporation apparatus. It is easy to perform and is similar to gene transfer under natural conditions. However, there have been no reports on natural transformation in Avibacterium paragallinarum. In this study, we analyzed the presence of homologous genetic factors and competence proteins underlying natural transformation in A. paragallinarum. Our results indicate that natural competence could be induced in A. paragallinarum serovars A, B, and C. Furthermore, the method that we established to transform plasmids into naturally competent A. paragallinarum strains was stable and efficient.
KEYWORDS natural transformation, Avibacterium paragallinarum, uptake signal sequence, infectious coryza stable and effective genetic manipulation methods. The introduction of DNA into the cytoplasm is a prerequisite for genetic manipulation in bacteria. Although electroporation is a common and efficient method for transformation in bacteria (2), it is not always effective in A. paragallinarum (3,4). However, some genera of the Pasteurellaceae, such as Haemophilus, Pasteurella, Riemerella, and others, are naturally competent (5)(6)(7). These genera actively take up DNA from their surroundings as nutrients or incorporate DNA into their genomes via homologous recombination (5,(8)(9)(10).
Competent bacteria have a series of competence proteins that allow them to accomplish natural transformation. In Haemophilus influenzae, competence is induced when nutrients become limited. Cells of H. influenzae acquire competence once they are transferred from a nutrient-rich medium to a starvation medium (11)(12)(13). In addition, two regulatory proteins, cyclic AMP (cAMP) receptor protein (CRP) and transcriptional coactivator for CRP (Sxy [also called TfoX in H. influenzae]), are required to induce competence in H. influenzae (14,15).
The concentration of the signal molecule cAMP increases in the absence of nutrients, thus activating its receptor protein CRP and enabling the formation of the cAMP-CRP complex. When combined with Sxy, the activated CRP binds to a specific promoter sequence (CRP-S) that is common to the competence genes responsible for DNA uptake and processing. This binding leads to the induction of gene transcription by the recruitment of RNA polymerase (16,17). Additionally, certain specific sequences may enhance the effectiveness of the natural transformation process. The cell surface's DNA uptake mechanism has a preference for binding and ingesting fragments that contain the uptake signal sequence (USS) (18).
Natural transformation has been extensively employed for the generation of targeted mutants (18)(19)(20)(21)(22) and is a convenient tool for the study of bacterial genetics and pathogenesis. However, there is a lack of relevant research on A. paragallinarum. Therefore, the goals of this work were to analyze the homologous genetic factors and competence proteins underlying natural transformation in A. paragallinarum and to develop a method to induce its transformation.

RESULTS
Putative competence proteins and regulators identified in A. paragallinarum. The analysis identified 16 competence proteins and regulators in A. paragallinarum genomes (Table 1). CRP and Sxy are responsible for the regulation of natural transformation, while the other 14 competence proteins are closely associated with DNA uptake (ComE, ComF, PilA, PilB, PilC, PilD, PulG, ComEC, and PilW) and DNA processing (ComM, DprA, RadC, Ssb, and LigA) ( Table 2). The presence of homologous competence proteins and regulators indicated that A. paragallinarum may be naturally competent like H. influenzae. However, there are no homologs of some competence proteins in A. paragallinarum (such as ComA, ComB, ComC, and ComD). Further investigation is required to determine the reason for the lack of these proteins in A. paragallinarum and its influence.
The results of the sequence analysis demonstrated that the competence proteins among the A. paragallinarum strains exhibited .97% sequence identity, except for Ssb, whose sequence identity was .80.4% (Table 1). These findings clearly suggest that the proteins linked with natural transformation in A. paragallinarum are remarkably conserved.
In our study, an examination of the promoter regions of competence homologs in A. paragallinarum revealed elements resembling the CRP-S sites identified in other bacteria. We identified these consensus sequences in the upstream DNA sequences of 11 transcriptional units of competence homologs in A. paragallinarum (Table 3). These reverse-complement consensus sequences exhibit high levels of similarity to the CRP-S consensus sequences of H. influenzae and Haemophilus parasuis (Fig. 1).
In H. influenzae, 17 genes regulated by CRP-S are necessary for natural transformation (23). These CRP-S sites have been identified as binding sites for CRP under Sxy regulation. Notably, the CRP-S sites differ from canonical CRP-N sites in that the former require both CRP and Sxy proteins for transcription activation and the 6th base is cytosine instead of thymine (7,17).
Upon analyzing the genomes of the A. paragallinarum strains, it was discovered that there were 1,537 to 1,641 copies of the 9-bp short DNA sequence 59-ACCGCACTT-39 (Table 4). This high number suggests that this particular DNA sequence is enriched in the genome of A. paragallinarum. Conversely, the DNA sequence 59-ACAAGCGGT-39 was found to have only 38 to 45 copies in the genome, while the sequence 59-ACCGAACTC-39 had only 10 to 11 copies. Additionally, we compared the core USS of Neisseria (59-GCCGTCTGAA-39) and found that there were only 6 to 22 copies present in the genomes of the A. paragallinarum strains. It is worth noting that the genomes of bacteria with natural competence exhibit high levels of enrichment in their preferred sequences, meaning that one or more USSs would appear in any fragment larger than approximately 2 kb (5). Therefore, only the core DNA sequence of H. influenzae is consistent with the characteristics of the USS in A. paragallinarum.
Induction of natural competence in A. paragallinarum. In H. influenzae, competence is induced when exponentially growing cells are transferred from rich medium to the starvation medium M-IV (12). In the case of A. paragallinarum, we grew cells in tryptic soy broth (TSB) to an optical density at 600 nm (OD 600 ) of 0.2, incubated them in M-IV medium for 100 min, and then transformed 1 mg/mL of plasmid pEA-KU or pEA-K. These procedures resulted in abundant resistant colonies. Cells without added DNA did not produce any resistant colonies. Although we have explored a variety of electroporation conditions, no positive transformant was obtained.
Effects of the DNA concentration and incubation time on natural transformation. To investigate the effect of the DNA concentration on the frequency of natural transformation, we conducted a transformation process on A. paragallinarum by varying the concentrations of pEA-KU. Our findings revealed that the transformation frequency increased as the DNA concentration increased within limits. The highest transformation frequency was obtained at saturating DNA concentrations of $0.5 mg/mL ( Fig. 2A). Furthermore, the transformation frequency was affected by the time of coincubation of 1 mg/mL DNA with competent cells and was highest when the cells were incubated for 30 min (Fig. 2B). Natural transformation abilities of different A. paragallinarum strains. Strains 2019/HB64, 2019/HB68, and 2019/NX56 were prepared in competent cells for natural transformation. Briefly, fresh bacterial fluid was added to TSB medium at a ratio of 1:100, cultivation was performed until the OD 600 reached ;0.2, and the bacteria were collected by centrifugation. The collected cells were resuspended in fresh M-IV medium in equal volumes and incubated at 37°C with shaking (100 rpm) for 100 min. Next, 1 mg/mL plasmid pEA-KU or pEA-K was used for natural transformation. The corresponding transformation frequency was calculated using the plate colony counts after transformation.
The results showed that the transformation frequency of the plasmid carrying the USS was higher than that of the plasmid without the USS in all tested strains, which showed that the USS 59-ACCGCACTT-39 promoted the uptake of exogenous DNA by A. paragallinarum (Fig. 3). In addition, the transformation frequency of plasmid pEA-KU in strain 2019/HB64 was significantly higher than that for strains 2019/HB68 and 2019/NX57 (Fig. 3). The pEA-K plasmid led to a transformation frequency of strain 2019/HB64 that was significantly higher than those of the other strains (Fig. 3). These results indicated that there

DISCUSSION
In the Pasteurellaceae, electroporation is an essential method for the introduction of foreign genes into cells (39,40), but it is not always effective (41). In this study, we attempted to introduce shuttle plasmids into A. paragallinarum by electroporation. Nevertheless, despite exploring various experimental conditions, we could not obtain any transformants. Wang et al. previously found that A. paragallinarum strains 221 and H18 could not be successfully transformed by electroporation (3). Similarly, electroporation did not prove successful in introducing the transposons mini-Tn10 and Tn5 into A. paragallinarum strains TW-1 and 221 (4). This may be due to the obstruction or restriction of structures such as bacterial capsule and the influence of the modification system (42,43).
Natural transformation is an important mechanism for bacteria to obtain external DNA molecules in the process of evolution (18). It can also be used as a means to introduce foreign genes into bacteria under laboratory conditions. Natural transformation is simpler to perform, is closer to gene transfer under natural conditions, and does not require equipment such as a gene pulser transfection apparatus. In A. paragallinarum, 16 competence proteins related to natural competence, DNA uptake, and DNA processing were identified in this study. Some noncompetent bacterial strains have mutations in one or more of  the competence genes, which likely explains their failure to transform (17,21,44). But in Riemerella anatipestifer, only some homologs of ComE, ComM, DprA, RadC, and Ssb were identified (45). Similarly, we did not find some competence proteins in A. paragallinarum, such as ComA, ComB, ComC, and ComD, so it may have a new mechanism of DNA transport. In order to adapt to various environmental pressures, bacteria have different ways to start the establishment of natural competence to regulate the uptake of foreign genes. For example, Streptococcus pneumoniae forms a natural receptive state under the action of antibiotics such as fluoroquinolones to absorb foreign DNA in the environment (46). The natural competence of Vibrio cholerae can be established by chitosan in the environment, while peptone, phosphate, and iron promote the natural transformation of Riemerella anatipestifer (47). A change in solid-liquid culture conditions can promote the formation of natural competence in H. parasuis (48). When H. influenzae and Bacillus subtilis are transferred from nutrient-rich medium to barren medium, the regulation of their natural transformation competence is activated (12,49). We found that the competence of A. paragallinarum can be induced under the same conditions as those for H. influenzae by transfer to M-IV starvation medium, indicating that their regulations of natural competence share features.
In H. influenzae, homologous DNA fragments (including the USS) hinder the uptake of other fragments by bacteria (50). Neisseria is more likely to ingest DNA fragments  parasuis that was defective in the thy gene (48). Thus, USSs are enriched in the genome over a long period of evolution in bacteria with competence, so one or more USSs appear for every fragment larger than ;2 kb (5, 52). In H. influenzae Rd, the highly conserved USS is the 9-bp sequence 59-AAGTGCGGT-39, which has 1,471 copies in the 1.83-Mb genome (53). The USS of Neisseriaceae (59-GCCGTCTGAA-39) presents 1,891 copies in the 2.18-Mb genome of Neisseria meningitidis Z2491 (5). In this study, we found that there were 1,537 to 1,641 copies of the sequence 59-ACCGCACTT-39 in the genome of A. paragallinarum. We compared the transformation frequencies between plasmid pEA-KU carrying the USS (59-ACCGCACTT-39) and plasmid pEA-K without the USS in A. paragallinarum. The results showed that plasmids carrying the USS were more easily absorbed by A. paragallinarum.
There are differences in the efficiencies of natural transformation between different strains, even within the same bacteria. Bossé et al. identified only one highly transformable strain of Actinobacillus pleuropneumoniae among 16 strains (21). The other strains were either nontransformable or poorly transformable. Kristensen et al. tested the natural competence of 9 Gallibacterium anatis strains from different origins (53), among which the transformation frequency of the lowest-transformation-frequency strain, F149T, was almost 3 orders of magnitude lower than that of the highest-transformation-frequency strain, 12656-12.
There were differences in the frequencies of natural transformation among different A. paragallinarum strains. We found that the transformation frequency of 2019/HB64 was 10 to 100 times higher than those of 2019/HB68 and 2019/NX57. However, all tested strains could be successfully transformed by natural transformation.
In conclusion, natural competence in A. paragallinarum is induced when cells are transferred from a rich medium to a defined starvation medium. In this study, we established a method for transforming a plasmid into naturally competent A. paragallinarum strains stably and efficiently. This transformation method could be an important tool for introducing foreign DNA into A. paragallinarum and could provide a basis for constructing a genetic operation platform for this species.

MATERIALS AND METHODS
Strains and plasmids. Strains of A. paragallinarum were isolated from chickens with clinical signs of IC (39) ( Table 5). These strains were grown in TSB (Hopebio, China) or on tryptic soy agar (TSA) (Hopebio, China) supplemented with 10% inactivated bovine serum (Solarbio, China) and 0.0025% (wt/vol) NAD (Sangon Biotech, China). The plates were cultured at 37°C under 5% CO 2 for 24 to 36 h. Typical colonies were inoculated into TSB and cultured at 37°C in a shaker for 16 h.
To find and identify competence proteins for natural transformation, the genomes of A. paragallinarum were analyzed using the Prokaryotic Genome Annotation Pipeline (PGAP) (version 5.2) (57) and the BLAST function from the NCBI (58). They were subsequently used in a reciprocal search in H. influenzae Rd KW20.
CRP-S sequences were identified by scanning 300 nucleotides (nt) upstream of the start codons of competence transcription units predicted using FIMO software (version 5.4.1) (59). A logo map of the CRP-S locus of A. paragallinarum was drawn using WebLogo software (version 2.8.2) (60). The copies of H. influenzae USS cores or H. parasuis USS cores (5,48) in the A. paragallinarum genomes were determined using the search function of SnapGene (version 4.2.4) (http://www.snapgene.com).
M-IV medium induction of natural competence. Natural competence was induced by incubation in M-IV medium according to a method described previously by Poje and Redfield (61). Briefly, typical colonies of A. paragallinarum were inoculated into TSB and cultured at 37°C in a shaker for 16 h. They were added to TSB medium at a ratio of 1:100, and we continued to cultivate them until the OD 600 reached ;0.2. The bacteria were collected by centrifugation, and the collected cells were resuspended in fresh M-IV medium in equal volumes, followed by incubation at 37°C with shaking (100 rpm) for 100 min. The cells were then used for transformation or stored in 15% (vol/vol) glycerol at 280°C.
Procedure for natural transformation. Before transformation, competent cells were thawed on ice, centrifuged, collected, and resuspended in fresh M-IV medium. Cells (1 mL) were incubated with 1 mg of DNA for 30 min at 37°C. Next, we added 2 mL of TSB, incubated the cells at 37°C for 100 min, serially diluted them, and plated them onto selective plates (usually with kanamycin at 10 mg/mL). A nonselective plate was also coated with the cells as a control. Positive transformants were identified by PCR using primer pair Kan-jd F/Kan-jd R ( Table 6). The transformation frequency was calculated as follows: transformation frequency = (number of positive transformants)/(amount of total viable bacteria). The bacteria were cultured for at least 24 h after transformation.
Electroporation of A. paragallinarum. Exponentially growing cells were diluted 1:100 in TSB medium and incubated at 37°C in a shaker (220 rpm) until the OD 600 reached 0.6 to 0.8. The culture was incubated on ice for 15 min before electroporation. Cells were collected by centrifugation at 4°C and washed three times in precooled 10% (vol/vol) glycerol. Next, the cells were resuspended in fresh TSB medium (1% of the initial volume).  Suspensions (100 mL) were mixed with 1 mg of DNA and transferred to precooled electroporation cuvettes (gap size of 2 mm). Electroporation was performed using a gene introduction instrument (Scientz, China) with settings of 2.2 kV, 400 X, and 25 mF, resulting in time constants of 8 to 10 ms. A total of 1 mL of preheated (37°C) TSB was immediately added, and the cells were incubated for at least 2 h at 37°C. Selective plates containing kanamycin (10 mg/mL) were then coated with the culture. A nonselective plate was also coated with the cells as a control. The transformation frequency was calculated as described above.
Statistical analysis. Statistical analysis was performed using GraphPad Prism software (version 6) for Windows. The statistical significance of the data was analyzed using Student's t test. A P value of ,0.05 was considered significant.

ACKNOWLEDGMENTS
This work was supported by the Earmarked Fund for Modern Agroindustry Technology Research System (CARS-40-K16), supported by 111 Project D18007, and by a project funded by the Priority Academic Program Development of Jiangsu Higher Education Institutions (PAPD).