Peptidoglycan Acetylation of Campylobacter jejuni Is Essential for Maintaining Cell Wall Integrity and Colonization in Chicken Intestines

ABSTRACT

Peptidoglycan (PG) acetylation of Gram-positive bacteria confers lysozyme resistance and contributes to survival in the host. However, the importance of PG acetylation in Gram-negative bacteria has not been fully elucidated. The genes encoding putative PG acetyltransferase A (PatA) and B (PatB) are highly conserved in Campylobacter jejuni, the predominant cause of bacterial diarrhea worldwide. To evaluate the importance of PatA and PatB of C. jejuni, we constructed patA and patB isogenic mutants and compared their phenotypes with those of the parental strains. Although transmission electron microscopy did not reveal morphological changes, both mutants exhibited decreased motility and biofilm formation in vitro. The extent of acetylation of the PG purified from the patA and patB mutants was significantly lower than the PG acetylation in the parental strains. Both mutants exhibited decreased lysozyme resistance and intracellular survival in macrophage cells. In a chick colonization experiment, significant colonization deficiency was observed for both mutants. These results suggest that PatA and PatB of C. jejuni play important roles in maintaining cell wall integrity by catalyzing PG O-acetylation and that the loss of these enzymes causes decreased motility and biofilm formation, thus leading to colonization deficiency in chicken infection.

IMPORTANCE The importance of peptidoglycan (PG) acetylation in Gram-negative bacteria has not been fully elucidated. The genes encoding putative PG acetyltransferase A (PatA) and B (PatB) are highly conserved in Campylobacter jejuni, the predominant cause of bacterial diarrhea worldwide. We evaluated the importance of these enzymes using isogenic mutants. The results of this study suggest that PatA and PatB of C. jejuni play important roles in maintaining cell wall integrity. The loss of these factors caused multiple phenotypic changes, leading to colonization deficiency in chicken infection. These data should be useful in developing novel control measures to prevent chicken colonization by C. jejuni. Inhibitors of the PG acetylation enzymes PatA and PatB might serve as potent anti-C. jejuni agents.

INTRODUCTION

Campylobacter is one of the most common bacterial causes of diarrhea, of which there are approximately 400 million cases per year worldwide (1). Campylobacter jejuni colonizes the intestinal tracts of various wild and domestic animals, and avian species such as poultry are considered to be the main reservoir of this pathogen (2–5). Contaminated poultry meat is one of the principal sources of C. jejuni pathogenic to humans (6). The reduction of C. jejuni contamination in the food chain is an important step in the control of campylobacteriosis. One of the most important control points is the colonization of broiler chickens by C. jejuni. However, there are currently no measures in practical use to control Campylobacter infections in poultry.

The helical morphology of C. jejuni is responsible for its enhanced ability, compared with that of rod-shaped bacteria, to move through the mucus layer of the chicken intestinal tract. To maintain cell shape, most bacterial species are surrounded by a cell wall that consists mainly of the cross-linked polymer peptidoglycan (PG) (7). The normal glycan strands of bacterial PG consist of alternating N-acetylglucosamine (GlcNAc) and N-acetylmuramic acid (MurNAc) residues that are cross-linked by short peptides. In C. jejuni, the PG layer also plays an important role in maintaining the structural integrity of the cell wall. PG plays an important role in maintaining cell shape, and structural changes in PG affect various phenotypes, such as motility and biofilm formation (8, 9). Reduced motility and/or biofilm formation have been reported to reduce C. jejuni colonization in chicken intestines (10, 11).

O-Acetylation, N-deacetylation, and N-glycosylation are common secondary modifications of glycan strands of PG and confer resistance against degrading enzymes such as lysozyme (12). The reported base-labile acetate contents of PG glycan strands purified from two C. jejuni strains were 55.7% and 62.7% of the MurNAc concentration (13). Based on the results of homology searches of genome databases, Weadge et al. have identified a cluster of genes encoding PG O-acetyltransferase (PatA) and O-acetylpeptidoglycan esterase (Ape1a and Ape2) in Gram-negative bacteria, including C. jejuni (13). It was subsequently reported that Ape2 of Neisseria gonorrhoeae does not exhibit esterase activity but functions as a periplasmic protein for the O-acetylation of PG. Thus, the protein was renamed PG O-acetyltransferase B (PatB) (14). PatA and PatB have recently been proposed to function together in translocating acetate from cytoplasmic pools to the periplasm, where it is transferred to peptidoglycan in Gram-negative bacteria (15) (Fig. 1A).

FIG 1.

Model of O-acetylation of PG in Gram-negative bacteria and the C. jejuni 81-176 patA and patB gene locus. (A) In Gram-negative bacteria, acetyl is translocated from the cytoplasm to the periplasm by PatA for PatB-catalyzed transfer to MurNAc residues in PG. (B) The patA or patB mutants were constructed by deletion of most of the gene and insertion of the nonpolar cat Cm^r cassette. The region (from 55 bp downstream of cjj0639 to 134 bp upstream of cjj0640) cloned into the pRRK plasmid used for complementation is shown below the gene cluster.

Wang et al. have reported that 26 of 30 whole-genome-sequenced strains of Helicobacter pylori do not have both patA and patB, and a patA mutant exhibits no defect in mouse colonization ability (16). In contrast, sequence alignments using Basic Local Alignment Search Tool (BLAST; http://blast.ncbi.nlm.nih.gov/Blast.cgi) (17) have revealed that all 96 whole-genome-sequenced strains of C. jejuni, including subsp. doylei, harbor patA and patB (Fig. 1B). However, the importance of PatA and PatB in C. jejuni has not been evaluated. The purpose of this study was to demonstrate that both PatA and PatB are essential for efficient PG O-acetylation and maintaining several biological characteristics of C. jejuni and thus play important roles in survival in the chicken intestine.

MATERIALS AND METHODS

Bacterial strains, plasmids, and culture conditions.

The bacterial strains and plasmids used in this study and their sources are listed in Table 1. Mueller-Hinton (MH) broth and MH agar (both from Becton Dickinson and Company, Sparks, MD) were used to grow the C. jejuni strains at 42°C under microaerophilic conditions, which were generated using an AnaeroPack (Mitsubishi Gas Chemical Company, Inc., Tokyo, Japan) in an enclosed jar. Escherichia coli DH5α was grown in Luria-Bertani (LB) broth or on LB agar (both from Becton Dickinson and Company) with or without 100 μg/ml ampicillin and 30 μg/ml kanamycin (both from Wako Pure Chemical Industries, Ltd., Osaka, Japan) at 37°C. The nalidixic acid-resistant phenotype of the C. jejuni 11-164 strain facilitated the enumeration of viable cells in the cecal contents of experimentally infected chicks.

TABLE 1.

Bacterial strains and plasmids used in this study

Strain or plasmid	Description	Source or reference
Strains
C. jejuni
Wild type
81-176	Wild type, human isolate	32
11-164	Wild type, chicken isolate, Nalr	This study
Mutant
817patA	81-176 derivative, patA mutant	This study
817patB	81-176 derivative, patB mutant	This study
817patA-c	817patA derivative, rrs::patA-patB	This study
817patB-c	817patB derivative, rrs::patA-patB	This study
164patA	11-164 derivative, patA mutant	This study
164patB	11-164 derivative, patB mutant	This study
164patA-c	164patA derivative, rrs::patA-patB	This study
164patB-c	164patB derivative, rrs::patA-patB	This study
E. coli strain DH5α	Cloning strain	TaKaRa
Plasmids
pGem-T Easy	PCR cloning vector, Ampr	Promega
pRRK	Cloning vector used for complementation of mutants, Kmr	19
pRRKpatA-patB	pRRK::patA-patB	This study

Construction of patA or patB mutants and complementary strains.

The PCR primers used in this study are shown in Table 2. PCR was performed with Ex Taq DNA polymerase (TaKaRa Bio, Inc., Shiga, Japan) or Pyrobest DNA polymerase (TaKaRa Bio) according to the manufacturer's instructions. Isogenic mutants of patA or patB were constructed in C. jejuni strains 81-176 and 11-164 using a nonpolar chloramphenicol cassette as described previously (18). Transformants were selected on MH agar containing 6 μg/ml chloramphenicol (Wako Pure Chemical Industries).

TABLE 2.

Primers used in this study

Primer	Sequence (5′→3′)a	Target	Origin(s)b
CHEF	TGCTCGGCGGTGTTCCTTT	cat	pUOA18
CHER	GCGCCCTTTAGTTCCTAAAG
patA.1	TCGCTGTAAATTTTATTGAGC	Region downstream of patA	81-176 and 11-164
patA.2	AAAGGAACACCGCCGAGCATATGCCAGATGGAATTCC
patA.3	CTTTAGGAACTAAAGGGCGCACTAAGGCGAAATAAGGATTG	Region upstream of patA	81-176 and 11-164
patA.4	ACTTGGAAACATTTTGGAAAC
patB.1	ATGACACGCGGAAAGAAATC	Region downstream of patB	81-176 and 11-164
patB.2	AAAGGAACACCGCCGAGCAAAGTGGCGCAAGAGAAATGT
patB.3	CTTTAGGAACTAAAGGGCGCTACAACAACCAAGCCAAGCA	Region upstream of patB	81-176 and 11-164
patB.4	CCAGCTTTAGGAAGATTTTTAACC
rrs	CTGGAACTCAACTGACGCTAAG	rRNA	81-176
rrl	CTCTTGCACATTGCAGTCCTAC
aphA3_L	CGCCAATTGCCACAATGATAGAACCAACGA	aphA3	pMW10
aphA3_R	CGCCAATTGATAGGCAGCGCGCTTATCA
patA.comR	CCGCTCGAGATTGACAAGATTAAGTTAATCGG	patA-patB region	81-176
patB.comF	CCGCTCGAGATTTTGTGCATAAGAGCTAG

^aThe underlined regions indicate the complementary sequences of the CHEF or CHER primers. The restriction sites are shown in boldface.

^bOrigins of the genomic DNA or plasmids used as templates for PCR amplification.

The complemented strains of the patA or patB mutants were constructed as previously described (19) with some modifications. Briefly, the 2.5-kb patA-patB region was amplified from extracted C. jejuni 81-176 genomic DNA. The amplified fragment was digested with XhoI (TaKaRa Bio) and cloned into the pRRK vector. The complemented strains were selected using MH agar plates containing 30 μg/ml kanamycin and 6 μg/ml chloramphenicol.

PG isolation and quantitative determination of its acetylation.

PG was isolated from C. jejuni cells grown for 24 h in MH broth and purified using the boiling SDS method as described by Wang et al. (16). The lyophilized PG was resuspended in 100 mM Tris-HCl (pH 6.8) by using a W-375 nsonicator cell disruptor (Qsonica, LLC., Newtown, CT). The samples were treated with 500 mM NaOH (final concentration) for 30 min at room temperature to release any ester-linked acetate, neutralized using an appropriate volume of 500 mM H₂SO₄, and subjected to ultracentrifugation (100,000 × g). The released acetate was quantified using a Megazyme acetic acid assay kit (Megazyme International Ireland, Ltd., Wicklow, Ireland). The extent of acetylation is presented as a percentage of muramic acid content as described by Hadzija (20).

Electron microscopy.

Cellular morphology was assessed by transmission electron microscopy (TEM). C. jejuni strains were grown at 42°C for 18 h, and 100 μl of each culture was then pelleted by centrifugation at 8,000 × g for 10 min. The bacterial pellet was washed twice, resuspended in 1 ml of phosphate-buffered saline (PBS [pH 7.2]), and placed on collodion membrane attachment mesh (Nisshin EM, Tokyo, Japan). The mesh was washed with PBS and incubated with 1% ammonium acetate solution for 10 min. Excess fluid was removed using filter paper, and the mesh was dried at room temperature. The sample mesh was observed with an H-7500 TEM (Hitachi, Ltd., Tokyo, Japan) at 80 kV. Images were recorded at a 25,000× magnification.

Motility assay.

Motility was quantified as previously described (21) with some modifications. Briefly, an overnight C. jejuni culture was inoculated in 0.4% soft MH agar with a needle, and the agar was incubated for 48 h at 37°C under microaerobic conditions. After incubation, the diameter of the zone of motility was measured. Three independent experiments were performed by using the same strains and conditions. The standard deviations are indicated by error bars.

Biofilm formation assay.

Biofilm formation was assessed as previously described (22) with slight modifications. Briefly, overnight C. jejuni cultures were prepared and then diluted to an optical density at 600 nm (OD₆₀₀) of 0.05 using MH broth. A 1-ml aliquot of this dilution was added to new sterile borosilicate tubes and incubated for 5 days at 42°C under microaerobic conditions. After incubation, the medium was removed, and the tubes were stained with 1 ml of 1% (wt/vol) crystal violet (CV) for 30 min at room temperature. The tubes were washed twice with sterile distilled water to remove unbound CV and then dried for 1 h at 42°C. The bound CV was dissolved by addition of 1 ml of 20% acetone in ethanol for 15 min. Biofilm formation was quantified by measuring the absorbance at 570 nm.

Lysozyme susceptibility testing.

The lysozyme susceptibility of the C. jejuni strains was measured as previously described (16) with some modifications. Lactoferrin was used as a membrane permeabilizer. The C. jejuni strains were grown on MH agar plates for 18 h, and the cells were suspended in PBS at a concentration of 10⁶ cells/ml. The cell suspensions were incubated with 0.3 mg/ml lysozyme (Wako Pure Chemical Industries) and 3 mg/ml lactoferrin (Wako Pure Chemical Industries) at 37°C under microaerobic conditions. Samples were removed at 0 and 6 h after inoculation, serially diluted, and spread onto MH agar.

Gentamicin protection assay for internalization and intracellular survival.

RAW 264.7 cells (murine macrophage cell line) were cultured in minimal essential medium (MEM [Sigma-Aldrich Co., St. Louis, MO]) with 20% fetal bovine serum (FBS [Sigma-Aldrich]) at 37°C in a 5% CO₂ incubator. The cells were seeded at 3 × 10⁵ cells per well in a 24-well plate and allowed to grow for 20 h. The cell monolayer was rinsed once with MEM with 10% FBS, and bacteria were then added to the monolayer at a multiplicity of infection (MOI) of 100. To count the number of adherent and internalized bacteria at 3 h after inoculation (time point 3 h [t₃]), the infected cells were washed three times with PBS and lysed with 1 ml of 1% Triton X-100 (Wako Pure Chemical Industries) in PBS. The number of adherent bacteria was determined by counting the resultant colonies on the MH agar plates. To count the number of internalized bacteria, the infected monolayers were rinsed three times with MEM–10% FBS at t₃ and incubated for an additional 2 h in MEM–10% FBS containing 150 μg/ml gentamicin (Wako Pure Chemical Industries). The number of bacteria was measured as described above (time point 5 h [t₅]). To assess intracellular survival, the medium was replaced with MEM–10% FBS containing 10 μg/ml gentamicin at t₅, and the number of bacteria was measured after a 5-h incubation (time point 10 h [t₁₀]). We prepared different triplicate wells for each strain to count the number of bacteria at each time point. The rates of internalization and intracellular survival corresponded to (CFU at t₅/CFU at t₃) × 100 and (CFU at t₁₀/CFU at t₅) × 100, respectively.

Chicken colonization experiment.

Newly hatched 1-day-old chicks were obtained from Nisseiken Co., Ltd. (Tokyo, Japan). To compare the colonization of the strains, 50 4-day-old chicks were assigned to five groups (10 chicks/group). Each group was orally challenged with 10⁶ CFU of the 11-164 wild-type and derivative strains by a stomach tube. Five chicks from each group were sacrificed at 7 and 14 days after inoculation, and their cecal contents were collected, serially diluted, and spread on MH agar plates supplemented with 100 μg/ml nalidixic acid (Wako Pure Chemical Industries). These experiments were conducted in strict accordance with the guidelines of animal experimentation defined by the National Institute of Animal Health (NIAH) of Japan. The protocol was approved by the Committee on the Ethics of Animal Experiments of the NIAH (permit no. 15-082).

Statistical analysis.

Differences in the results were tested using two-tailed unpaired Student t tests. A P value of <0.05 was considered to be statistically significant (see the figure legends for specific values).

RESULTS

patA or patB mutation reduces PG acetylation in C. jejuni.

To confirm the involvement of PatA and PatB in PG acetylation in C. jejuni, we constructed isogenic deletion mutants of patA and patB and complementary strains derived from the wild-type strains 81-176 and 11-164. Table 3 shows the results of the quantitative determination of the PG acetylation of each strain. PG acetylation was significantly lower in the patA or patB mutants (49.0% to 69.0%) than in the wild-type strains (P < 0.05). No significant differences were observed between the patA and patB mutants of both wild-type strains (P > 0.05). Complementation fully restored the acetylation levels of PG in all of the mutants.

TABLE 3.

Extent of peptidoglycan acetylation in the C. jejuni strains used in this study

Strain	mol% acetylation (mean ± SD)a	% of wild type
81-176 (wild type)	61.7 ± 8.1	100.0
817patA	35.8 ± 7.3*	58.1
817patB	42.5 ± 3.8*	69.0
817patA-c	65.2 ± 6.6	105.8
817patB-c	64.7 ± 4.8	105.0
11-164 (wild type)	58.3 ± 8.1	100.0
164patA	28.6 ± 3.9*	49.0
164patB	35.4 ± 3.8*	60.7
164patA-c	56.1 ± 5.6	96.2
164patB-c	64.9 ± 2.7	111.3

^aResults are shown as the moles percent base-labile acetate relative to the MurNAc concentration in the isolated peptidoglycan (n = 3). Asterisks indicate statistically significant differences compared with the wild type by Student's t test (P < 0.05).

Reduced PG acetylation decreases motility and biofilm formation but does not affect cell morphology.

In various Gram-positive and Gram-negative bacteria, including C. jejuni, structural changes in PG affect cell morphology, motility, and/or biofilm formation (8, 9, 23). In this study, no differences in cell shape and size were observed between the wild-type 81-176 and 11-164 strains and the mutants (Fig. 2A). The patA and patB mutants both exhibited helical shapes and no flagellar structural defects. In motility tests, the halo diameters of 817patA and 817patB in soft agar plates were 64.7% and 70.6% of those of wild type, respectively. The diameters of 164patA and 164patB were 48.4% and 45.2% of those of wild type, respectively (Fig. 2B). The motility of all complementary strains was comparable to that of the wild-type strains. Rates of biofilm formation by the patA and patB mutants were 49.1% and 65.9% of that in 81-176 and 24.8% and 26.8% of that in 11-164, respectively (Fig. 2C). There were no significant differences in biofilm formation between the complementary strains 817patA-c, 817patB-c, and 164patB-c and their wild-type strains (P > 0.05). Biofilm formation by 164patA-c was 73.1% of that of wild type.

FIG 2.

Cell morphology, motility, and biofilm formation of the C. jejuni strains. (A) Negatively stained TEM images of the C. jejuni wild-type strains 81-176 and 11-164 (817wild and 164wild, respectively) and the mutants. (B) Motility was assayed by measuring halo diameters in 0.4% soft agar plates. The standard error of the mean was calculated from 3 measurements. (C) Biofilm formation was assessed by crystal violet staining of standing cultures in borosilicate tubes and quantification of the dissolved crystal violet at 570 nm. The standard error of the mean was calculated from triplicate cultures and is representative of three independent experiments. Asterisks indicate statistically significant differences compared with the wild type, using an unpaired Student t test (*, P < 0.05; **, P < 0.01).

Reduced PG acetylation increases susceptibility to lysozyme.

To investigate the effect of reduced PG acetylation on the membrane integrity of C. jejuni, we compared the susceptibility of each strain to lysozyme. As shown in Fig. 3, no significant differences were observed in the initial viable cell numbers of all strains. Six hours after lysozyme treatment, the viable cell numbers of 817patA and 817patB (10^2.5 and 10^2.7 cells/ml, respectively) were significantly lower than that of the wild-type strain (10^3.5 cells/ml) (P < 0.05). The viable cell numbers of 164patA and 164patB (10^2.8 and 10^3.5 cells/ml, respectively) were also significantly lower than that of the wild-type strain (10^4.5 cells/ml) (P < 0.05). There were no significant differences between the patA and patB mutants in both sets of derivative strains after treatment (P > 0.05). Complementation fully restored resistance to lysozyme treatment in all mutants.

FIG 3.

Lysozyme susceptibility in the presence of lactoferrin. C. jejuni cell suspensions (10⁶ cells/ml) were treated with 0.3 mg/ml lysozyme plus 3 mg/ml lactoferrin for the indicated times, and the number of surviving cells was determined. 817wild, wild-type strain 81-176; 164wild, wild-type strain 11-164. The data are means and standard deviations from three experiments. Asterisks indicate statistically significant differences compared with the wild type, using an unpaired Student t test (P < 0.05).

Reduced PG acetylation enhances internalization and decreases intracellular survival in macrophages.

To evaluate the importance of PG acetylation in resistance to phagocytosis, we compared the rate of internalization and short-term intracellular survival in RAW264.7 macrophages. As shown in Fig. 4, the rates of internalization of the patA or patB mutants were significantly higher (ranging from 149.5% to 226.9%) than those of the wild-type strains (P < 0.05 or P < 0.01). The intracellular survival rates of the mutants were significantly lower (ranging from 38.8% to 52.5%) than those of the wild-type strains (P < 0.05 or P < 0.01). Among the complementary strains, the internalization rate of 817patA-c was significantly lower (68.8%) than that of the wild type (P < 0.05), and the intracellular survival rates of 817patA-c and 817patB-c were significantly higher (134.6% and 130.9%, respectively) than that of the wild type (P < 0.05 or P < 0.01).

FIG 4.

Gentamicin protection assay. The rates of internalization and intracellular survival of the C. jejuni strains in RAW264.7 cells were measured in a 24-well plate assay. 817wild, wild-type strain 81-176; 164wild, wild-type strain 11-164. The values for percentage of recovery were normalized relative to that of wild type, which was designated 100% at each time point. The results represent an average of three independent experiments in triplicate. The standard errors are indicated by error bars. Asterisks indicate statistically significant differences compared with the wild type using an unpaired Student t test (*, P < 0.05; **, P < 0.01).

Reduced PG acetylation decreases intestinal colonization ability in chickens.

As shown in Fig. 5, wild-type strain 11-164 colonized all chickens tested at both 7 and 14 days after inoculation, and the viable bacterial counts in the cecal contents ranged from 10^6.5 to 10^8.2 CFU/g. The mutant strains 164patA and 164patB were not detected in any of the samples recovered during the experimental period, except in one chicken inoculated with 164patA. At 7 days after inoculation, 164patB-c colonized all chickens, although the cecal colonization level was not fully restored (ranging from 10^5.2 to 10^6.8 CFU/g). At 14 days after inoculation, complementation fully restored colonization levels in both mutants.

FIG 5.

Colonization of chickens by the C. jejuni 11-164 wild type and derivative strains. Each dot represents the log₁₀ CFU per gram in the cecal content of the individual infected chicken at (A) 7 and (B) 14 days after inoculation. The solid and dashed lines indicate the average log₁₀ CFU/g in each group and the lower limit of detection (2.0 log₁₀ CFU/g), respectively. 164wild, wild-type strain 11-164. Asterisks indicate statistically significant differences compared with the wild type using an unpaired Student t test (**, P < 0.01; ***, P < 0.001).

DISCUSSION

Gram-negative bacteria have a relatively thin cell wall composed of several layers of PG. The inactivation of PG O-acetyltransferase in Gram-negative bacteria has not been reported to have a large influence on the biological characteristics of these bacteria. The N. gonorrhoeae patA mutant does not exhibit any decrease in resistance to lysozyme or serum, but purified PG from the wild-type strain is significantly more resistant to human lysozyme than PG purified from the patA mutant (24). In H. pylori, the patA mutant is more susceptible to lysozyme than the wild-type strain but exhibits no defect in mouse colonization ability (16). The mutation of pgdA encoding N-deacetylase of GlcNAc residues in PG in addition to the patA mutation has been shown to reduce the mouse colonization by H. pylori. In this study, we demonstrated that the inactivation of patA or patB in C. jejuni reduced not only lysozyme resistance but also motility and biofilm formation, as well as the ability of C. jejuni to colonize the chicken intestinal tract.

In this study, O-acetylation of the PG purified from the patA and patB mutants of C. jejuni was lower than that of the wild-type strains but was not completely eliminated in the mutants. This result suggests that C. jejuni may have additional O-acetyltransferases. Lysozyme resistance is considered to be the representative biological function of PG O-acetylation in both Gram-positive and Gram-negative bacteria (12, 15). Lysozyme is an antimicrobial enzyme that is naturally present in the mucosal layer and in leukocytes, including macrophages (25). In this study, the patA or patB mutants were more susceptible to lysozyme and killing by macrophages than the wild-type strains, and these results support that PatA and PatB are PG O-acetyltransferases of C. jejuni. The mechanism of C. jejuni colonization in chicken is poorly understood, and the relationship between lysozyme resistance and chicken colonization is also unclear. However, the primary site of C. jejuni colonization is the mucosal layer close to the epithelial cells in the deep crypts of the cecum of the chicken gastrointestinal tract (26), and the mucous layer contains abundant lysozyme and lactoferrin (25). The increased susceptibility to lysozyme may be the cause of the colonization deficiency in chicken infection.

PG acetylation has been suggested to affect bacterial colonization by modulating macrophage function. For example, Shimada et al. have reported that the PG O-acetyltransferase of Staphylococcus aureus strongly suppresses inflammasome activation and inflammation and that phagocytosis and lysozyme-based cell wall degradation by macrophages are functionally coupled to inflammasome activation and interleukin-1β (IL-1β) secretion (27). In addition, the Nod protein of the host cell, including macrophages, senses bacterial PG fragments and initiates immune responses that aid in clearing the infection (28). The reduced intracellular survival of the patA and patB mutants compared with that of the wild-type strains in this study might be partly attributable to these mechanisms in macrophages.

Previous studies have demonstrated the importance of motility and/or biofilm formation in chicken intestinal colonization by C. jejuni (10, 11). In the present study, the patA or patB mutations of the C. jejuni stains exhibited decreased motility and biofilm formation, but TEM analysis revealed no differences in cell morphology compared with that of the wild-type strains. A number of components of the flagellum and secretion systems must interact with the bacterial PG layer (29). In Gram-negative bacteria, the flagellar stator proteins MotA and MotB form a complex and bind to the PG of the cell wall (30). In C. jejuni, the inactivation of PG peptidases 1 (Pgp1) and 2 (Pgp2), which are PG modification enzymes, causes morphological changes (from helical form to rod form) and results in defective motility and biofilm formation (8, 9). Frirdich et al. have suggested that changes in PG structure resulting from Pgp1 and Pgp2 activities might alter the interactions between MotB and the PG residues, thereby affecting the efficiency of flagellar rotation (9). In addition, Moe et al. have reported that the motility mediated by the flagella is required for biofilm formation because the C. jejuni motA mutant does not form biofilms (31). The small structural change in the PG resulting from reduced acetylation does not appear to have a significant impact on cell shape but might affect the MotB function of C. jejuni.

Both patA and patB were detected in 96 whole-genome sequences of C. jejuni by BLAST alignment (17), thus suggesting that these genes are widely distributed in C. jejuni. The results of this study suggest that PatA and PatB of C. jejuni play an important role in maintaining cell wall integrity, and the loss of these factors caused multiple phenotypic changes, leading to colonization deficiency during chicken infections. These data should be useful for the development of novel control measures to prevent chicken colonization by C. jejuni. Inhibitors of the PG acetylation enzymes PatA and PatB might serve as potent anti-C. jejuni agents.

Funding Statement

The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.