Arginine deiminase pathway is far more important than urease for acid resistance and intracellular survival in Laribacter hongkongensis: a possible result of arc gene cassette duplication

Background Laribacter hongkongensis is a Gram-negative, urease-positive bacillus associated with invasive bacteremic infections in liver cirrhosis patients and fish-borne community-acquired gastroenteritis and traveler’s diarrhea. Its mechanisms of adaptation to various environmental niches and host defense evasion are largely unknown. During the process of analyzing the L. hongkongensis genome, a complete urease cassette and two adjacent arc gene cassettes were found. We hypothesize that the urease cassette and/or the arc gene cassettes are important for L. hongkongensis to survive in acidic environment and macrophages. In this study, we tested this hypothesis by constructing single, double and triple non-polar deletion mutants of the urease and two arc gene cassettes of L. hongkongensis using the conjugation-mediated gene deletion system and examining their effects in acidic environment in vitro, in macrophages and in a mouse model. Results HLHK9∆ureA, HLHK9∆ureC, HLHK9∆ureD and HLHK9∆ureE all exhibited no urease activity. HLHK9∆arcA1 and HLHK9∆arcA2 both exhibited arginine deiminase (ADI) activities, but HLHK9∆arcA1/arcA2 double deletion mutant exhibited no ADI activity. At pH 2 and 3, survival of HLHK9∆arcA1/arcA2 and HLHK9∆ureA/arcA1/arcA2 were markedly decreased (p < 0.001) but that of HLHK9∆ureA was slightly decreased (p < 0.05), compared to wild type L. hongkongensis HLHK9. Survival of HLHK9∆ureA/arcA1/arcA2 and HLHK9∆arcA1/arcA2 in macrophages were also markedly decreased (p < 0.001 and p < 0.01 respectively) but that of HLHK9∆ureA was slightly decreased (p < 0.05), compared to HLHK9, although expression of arcA1, arcA2 and ureA genes were all upregulated. Using a mouse model, HLHK9∆ureA exhibited similar survival compared to HLHK9 after passing through the murine stomach, but survival of HLHK9∆arcA1/arcA2 and HLHK9∆ureA/arcA1/arcA2 were markedly reduced (p < 0.01). Conclusions In contrast to other important gastrointestinal tract pathogens, ADI pathway is far more important than urease for acid resistance and intracellular survival in L. hongkongensis. The gene duplication of the arc gene cassettes could be a result of their functional importance in L. hongkongensis.


Background
Laribacter hongkongensis is a Gram-negative, facultative anaerobic, motile, S-shaped, asaccharolytic, urease-positive bacillus that belongs to the Neisseriaceae family of βproteobacteria [1]. It was first isolated from the blood and thoracic empyema of an alcoholic liver cirrhosis patient in Hong Kong [1]. Recently, it was also recovered from the blood culture of a Korean patient with liver cirrhosis as a result of Wilson's disease [2]. These cases make chronic liver disease a distinct possible risk factor for invasive L. hongkongensis infections, where intestinal mucosal edema and local immunosuppression secondary to portal venous congestion vasculopathy due to liver cirrhosis predisposed the patients to L. hongkongensis invasion through the gastrointestinal mucosa. In addition to invasive bacteremic infections, L. hongkongensis is also associated with community-acquired gastroenteritis and traveler's diarrhea [3]. L. hongkongensis is likely to be globally distributed, as travel histories from patients suggested its presence in at least four continents: Asia, Europe, Africa and Central America [3][4][5][6]. L. hongkongensis has been found in up to 60% of the intestines of commonly consumed freshwater fish of the carp family [7,8]. It has also been isolated from drinking water reservoirs and Chinese tiger frogs in Hong Kong and little egrets in Hangzhou [9][10][11]. Pulsed-field gel electrophoresis and multilocus sequence typing showed that the fish and patient isolates fell into separate clusters, suggesting that some clones could be more virulent or adapted to human [8,12]. These data strongly suggest that this bacterium is a potential diarrheal pathogen that warrants further investigations.
For any gastrointestinal tract pathogen, after transmission through the oral route, the first challenge that the bacterium has to face is the hostile acidic environment of the stomach. When the bacterium invades the intestinal mucosa, it has to survive the attack of submucosal macrophages, which sometimes may be related to its resistance to the acidic environment in endocytic vacuoles. More importantly, for a successful pathogen, the ability of resisting acidic environments is definitely crucial for its survival in different environment and transition from environments to humans. Various gastrointestinal bacteria have developed different mechanisms to overcome this hostile environment and evade host defense. For example, Helicobacter pylori and verotoxigenic Escherichia coli O157 have developed unique mechanisms to overcome such an acidic environment [13][14][15]. For H. pylori, urease converts urea to carbon dioxide and ammonia and increases the local pH of the bacterium, which is essential for its pathogenesis [16]. During the process of analyzing the L. hongkongensis genome, a complete urease cassette, which includes eight open reading frames, encoding three urease structural proteins (UreA, UreB and UreC) and five accessory proteins (UreE, UreF, UreG, UreD and UreI) ( Figure 1A), was observed [17]. In addition, two adjacent arc gene cassettes, each of them consisting of four genes, arcA, arcB, arcC and arcD ( Figure 1A), were also found [17]. arcA, arcB and arcC encode the three enzymes, arginine deiminase (ADI), ornithine carbamoyltransferase and carbamate kinase, of the ADI pathway; and arcD encodes a membrane bound arginine-ornithine antiporter. These enzymes of the ADI pathway convert L-arginine to carbon dioxide, ATP, and ammonia, which increases the pH of the environment and have been shown to be important for survival of various bacteria in acidic environments [18][19][20][21]. We hypothesize that the urease cassette and/or the arc gene cassettes are important for L. hongkongensis to survive in acidic environments and macrophages. In this study, we tested this hypothesis by systematically knocking out genes in the urease cassette and the two arc gene cassettes in L. hongkongensis and examining their effects in the survival of the single, double and triple knockout mutants in acidic environment in vitro, in macrophages and in a mouse model.

Ethics statement
The experimental protocols were approved by the Committee on the Use of Live Animals in Teaching and Research, The University of Hong Kong, in accordance with the Guidelines laid down by the NIH in the USA regarding the care and use of animals for experimental procedures.

Bacterial strains and growth conditions
The bacterial strains and plasmids used in this study are listed in Table 1. The parental L. hongkongensis strain HLHK9, was a clinical isolate from a patient in Hong Kong [3], for which the complete genome has been sequenced [17]. Streptomycin (Sm)-resistant HLHK9 strain was obtained by serial passage of HLHK9 cells on Luria broth (LB) agar with increasing concentrations of Sm, starting at 10 μg/ml, and increased up to 100 μg/ml. Unless stated otherwise, all HLHK9 and its derivate strains used in this study were Sm resistant. HLHK9 and its derivatives were grown in brain heart infusion (BHI) broth or on BHI agar (BHA) plates (BBL, BD) whereas all other E. coli strains were grown in LB or on LB agar (LBA) plates (BBL, BD). Media were supplemented with antibiotics (Sigma-Aldrich) when appropriate: ampicillin (Amp) (100 μg/ml), kanamycin (Km) (50 μg/ml), chloramphenicol (Cm) (15 μg/ml), tetracycline (Tet) (12.5 μg/ml) and Sm (100 μg/ml). Growth phase and bacterial cell density were determined by measuring absorbance spectrophotometrically at optical density (OD) 600 .

Construction of non-polar deletion mutant strains
Primers used for deletion mutagenesis are listed in Table 2. To generate unmarked, non-polar deletion of ureA, suicide plasmid pDS132 was used for constructing in-frame deletion mutants by homologous recombination [22]. 5′-and 3′-flanking regions of ureA were amplified by PCR from chromosomal DNA of HLHK9, using primers ureA-UF/UR and ureA-DF/DR, respectively, and the individual PCR products were mixed to generate an in-frame deletion pattern of ureA by an overlapping PCR method. The overlapping amplicon containing the in-frame deletion pattern was cloned into pUC19 followed by pDS132, resulting in the final construct of pDS132-ureA, which was electro-transformed into pir-positive E. coli SM10 λ pir [23]. pDS132-ureA was transferred into HLHK9 from transformed SM10 λ pir by bacterial conjugation. Exconjugants having single recombination with suicide vector pDS132 were first selected on BHA with antibiotics (Cm, 15 μg/ml; Sm, 100 μg/ml). After that, the positive enconjugants with single recombination were further cultured and selected Primers for mutagenesis of arcA1 Primers for mutagenesis of arcA2 on LBA with 10% sucrose (without NaCl and antibiotics) at room temperature to obtain the final double recombinants [24]. Sucrose-resistant colonies were screened by PCR using primers ureA-UF/DR and Inner-ureA-F/R, which were specific for the deleted sequence. All mutant strains were confirmed by DNA sequencing. Similarly, non-polar deletion of the ureC, ureD and ureE were constructed respectively as described above ( Table 2). Instead of using suicide plasmid pDS132, arcA1, arcA2, arcA1/arcA2 double mutant and ureA/ arcA1/arcA2 triple mutant strains were constructed using suicide plasmid pCVD442 [25], and Amp was used as the selection marker.

Qualitative analysis of urease enzyme activity
Thirty human strains, including HLHK9, and mutant strains HLHK9ΔureA, HLHK9ΔureC, HLHK9ΔureD and HLHK9ΔureE, were grown at 37°C overnight. Bacterial cultures were diluted 1:50 in BHI containing Sm and further cultured at 37°C with shaking, until earlyexponential phase (about 0.6 at OD 600 ). One hundred microliter of bacterial cultures was used to inoculate 2 ml urease test broth [26]. The mixtures were incubated at 37°C without shaking. The color change in urease test broth was monitored at 4, 8, 24 and 48 h with the uninoculated urease test as negative control [27].

Qualitative analysis of ADI activity
A chemical colorimetric method, based on the production of L-citrulline from L-arginine, was used to measure ADI activity of whole-cell lysates of 30 human strains, including HLHK9, and mutant strains HLHK9ΔarcA1, HLHK9ΔarcA2 and HLHK9ΔarcA1/arcA2 [28,29]. Briefly, 10 ml overnight culture of test strains were re-suspended in 2 ml extraction solution (2% Triton X-100, 1% SDS, 100 mM NaCl, 10 mM Tris pH 8, 1 mM EDTA) and lysed by glass beads (Sigma-Aldrich). One milliliter of supernatants were mixed with 0.4 ml of 100 mM potassium phosphate buffer (containing 10 mM L-arginine) and incubated at 37°C for 1 h. Afterwards, 250 μl of 1:3 (vol/ vol) mixture of 95% H 2 SO 4 and 85% H 3 PO 4 , and 250 μl of 3% diacetylmonooxime solution were added into the samples, followed by boiling for 15 min. Citrulline standard and the uninoculated reagents were used as positive and blank controls, respectively. The development of an orange color was monitored among the tested strains.
In vitro susceptibility of L. hongkongensis to acid pH One hundred microliter of overnight cultures of HLHK9 and derivative mutant strains were inoculated into 5 ml of fresh BHI respectively and grown to exponential phase (OD 600 0.6 to 0.8), washed with sterile water, and harvested by centrifugation. The pH of the phosphate buffered saline (PBS, Sigma-Aldrich) was adjusted to 2, 3, 4, 5 and 6 by adding 1 N HCl in the presence or absence of 50 mM urea (for HLHK9, HLHK9ΔureA, HLHK9ΔureC, HLHK9ΔureD, HLHK9ΔureE and HLHK9ΔureA/arcA1/ arcA2) and 50 mM arginine (for HLHK9, HLHK9ΔarcA1, HLHK9ΔarcA2, HLHK9ΔarcA1/arcA2 and HLHK9ΔureA/ arcA1/arcA2). About 10 8 colony-forming units (CFUs) per ml of bacterial cells were resuspended in PBS of pH 2 to 6 respectively and incubated at 37°C for 1 h. Furthermore, survival of HLHK9, HLHK9ΔureA, HLHK9ΔarcA1/arcA2 and HLHK9ΔureA/arcA1/arcA2 were also monitored at pH 4 after 3 and 5 h incubation respectively. Following incubation, bacterial cells were washed three times in PBS (pH 7.4), and serial dilutions of each culture were spread in duplicate on BHA to determine the number of viable cells [20,30]. The experiments were performed in triplicate from three independent experiments.

Intracellular survival assays in J774 macrophages
J774 macrophages (Sigma-Aldrich) were grown in DMEM (Gibco) supplemented with 10% fetal bovine serum (FBS, Sigma-Aldrich) at 37°C in an atmosphere of 5% CO 2 . Infection assays were performed as described previously [31,32]. J774 macrophages were seeded to 24well tissue culture plates at 4 × 10 5 cells per well and  incubated at 37°C with 5% CO 2 for 24 h before infection. Log-phase bacterial cultures (OD 600 of 0.6 to 0.7) of the wild type L. hongkongensis HLHK9 and mutants were washed twice with sterile phosphate-buffered saline (PBS) and resuspended in antibiotic-free media. Infection was carried out by inoculating 1 × 10 7 bacterial cells to each well at a multiplicity of infection of about 10:1 and incubated at 37°C for 1 h to allow adhesion and invasion to occur. After that, the culture supernatants were aspirated and the cells were washed three times with sterile PBS. Gentamicin (Sigma-Aldrich) was then added to each well at a concentration of 100 μg/ml and incubated at 37°C for 1 h to kill the extracellular bacteria followed by washing with sterile PBS and replacing the medium with serum-free DMEM containing 25 μg/ml of gentamicin. After 2 and 8 h post-infection, macrophages were lysed with 1% Triton X-100 (Sigma-Aldrich) for CFUs counts. The CFUs recovered from cell lysates after 2 h of phagocytosis were considered as the initial inocula and were used as the baseline values for intracellular survival analysis. CFUs recovered at 8 h were used to calculate the recovery rate of bacterial cells in macrophages. Experiments were repeated in triplicate to calculate the mean of intracellular survival of bacteria.

RNA isolation and real-time quantitative RT-PCR
At 2 h and 8 h post infection, the macrophage monolayers were washed with PBS and lysed with 1% Triton X-100 (Sigma-Aldrich). Total RNA was then extracted respectively using RNeasy Mini kit (Qiagen), followed by treating with RNase-free DNase I (Roche) at 37°C for 20 min. Reverse transcription was performed using the SuperScript III kit (Invitrogen). Real-time RT-PCR assay was performed in ABI7900HT Fast Real Time PCR machine (Applied Biosystems) with FastStart DNA Master SYBR Green I Mix reagent kit (Roche), as described by the manufacturer. The sequences of the primers used in the quantitative reverse transcription-PCR (qRT-PCR) were listed in Table 2. The mRNA levels of arcA1 and arcA2 and ureA genes were measured by quantitation of cDNA and the calculated threshold cycle (CT) corresponding to the target gene was calculated as 2 (CtTarget − CtReference) and normalized to that of rpoB gene [33].

Survival of L. hongkongensis in mouse model
One hundred microliters of overnight cultures of HLHK9 and mutant strains HLHK9ΔureA, HLHK9ΔarcA1/arcA2 and HLHK9ΔureA/arcA1/arcA2 were inoculated into 5 ml of fresh BHI respectively and grown to exponential phase (OD 600 0.6 to 0.8). The bacteria were harvested by centrifugation at 5,000 g for 15 min and resuspended in PBS to about 10 9 CFUs/ml. Five hundred microliters of bacterial suspension were orally inoculated to groups (n = 5) of 6-to 8-week-old female BALB/c mice which were starved for 6 h previously. Mice were sacrificed 120 min after inoculation and the terminal ileum were removed aseptically and homogenized in 5 ml PBS. Serial dilutions of the homogenates were plated in duplicate on BHA with Sm (100 μg/ml) to determine the number of viable cells [30]. The data were collected from three independent experiments.
PCR amplification and DNA sequencing of arcA1 and arcA2 Extracted DNA from the 30 L. hongkongensis human strains previously isolated from stool specimens of patients with community-acquired gastroenteritis [3], was used as template for amplification of arcA1 and arcA2 genes, using specific primers LPW16076/16077 and LPW16078/16079, respectively. The PCR mixture (25 μl) contained L. hongkongensis DNA, 1× PCR buffer II, 2.0 mM MgCl 2 , 200 μM of each dNTPs and 1.0 unit AmpliTaq Gold DNA polymerase (Applied Biosystems). All samples underwent denaturation at 95°C for 10 min, followed by 40 cycles of 95°C for 1 min, 60°C for 1 min and 72°C for 1 min, with a final extension at 72°C for 10 min in an automated thermal cycler (Applied Biosystems). Five microliters of each amplified product was electrophoresed in 2% (wt/vol) agarose gel and Trisborate-EDTA buffer, with molecular size marker (Gen-eRuler 50-bp DNA ladder; Fermentas) in parallel, at 100 volts for 1 h. Five PCR products were randomly selected, gel-purified and sequenced with an ABI Prism 3700 DNA Analyzer (Applied Biosystems), using the PCR primers.

Examination of L. hongkongensis strains for urease activity
With the exception of native urease-negative L. hongkongensis HLHK30, the urease test broth incubated with all human strains, including HLHK9, began to turn pink after 4 h (Figure 2A), and the color became more intense after 24 h of incubation. Similar to the natural ureasenegative strain HLHK30, mutant strains HLHK9ΔureA, HLHK9ΔureC, HLHK9ΔureD and HLHK9ΔureE elicited no color change after prolonged incubation (Figure 2A). These results indicated that these four urease genes were all essential for the urease enzyme activity.

Examination of L. hongkongensis strains for ADI activity
In the qualitative assay, similar to the positive control (citrulline standard), cellular extracts prepared from all 30 human strains, including wild type L. hongkongensis HLHK9, also generated an orange color, confirming that citrulline was being produced ( Figure 2B). Cell extracts from both single knockout mutant strains, HLHK9ΔarcA1 and HLHK9ΔarcA2, also yielded an orange color, whereas deletion of both arcA1 and arcA2 abolished the ADI activity ( Figure 2B). These results showed that both the arcA1 and arcA2 genes encode functional ADI enzymes, which could complement the functions of each other.
In vitro susceptibility of urease-negative mutants to acid HLHK9 and mutant strains HLHK9ΔureA, HLHK9ΔureC, HLHK9ΔureD and HLHK9ΔureE were subjected to a range of acidic pHs (from pH 2 to 6) in the presence and absence of 50 mM urea, respectively. Since the four urease mutant strains exhibited similar survival abilities under different acidic conditions, only the viable counts of HLHK9ΔureA are shown. In the absence of urea, both the parental and mutant strains survived equally well at pH 4 and above, but all strains did not survive at pH 2 and 3 (data not shown). In the presence of urea, there were no significant difference in the survival levels of HLHK9 and urease mutant strains after incubation at pH 5 and 6 for 1 h, with viable counts of all strains declining slightly at pH 4 ( Figure 3A). When the pH was further decreased to pH 2 and 3, the survival counts of HLHK9 reduced about 6-log, and the mutant strain could barely be recovered (p < 0.05) ( Figure 3A). These demonstrated that the urease system has a contribution to the survival of L. hongkongensis at pH 3 and below.

In vitro susceptibility of ADI-negative mutants to acid
To study the role of the two arc loci of L. hongkongensis under acidic conditions, wild type L. hongkongensis HLHK9, HLHK9ΔarcA1, HLHK9ΔarcA2, HLHK9ΔarcA1/ arcA2 were exposed to different acidic pHs (pH 2 to 6) in the presence and absence of 50 mM of L-arginine, respectively. In the absence of L-arginine, survival of the three mutants were similar to that of HLHK9 at ≥pH 4, and they became susceptible at ≤pH 3 (data not shown). In the presence of L-arginine, wild type L. hongkongensis HLHK9, HLHK9ΔarcA1 and HLHK9ΔarcA2 survived well under all tested pHs, suggesting that the two copies of the arcA gene performed complementary functions in L. hongkongensis ( Figure 3B). On the other hand, the survival of HLHK9ΔarcA1/arcA2 decreased about 2-log at pH 4 (p < 0.05) and it was barely recovered at pH 2 and 3 (p < 0.01) ( Figure 3B). This indicated that the ADI pathway played a crucial role in the survival of L. hongkongensis under acidic conditions.
In vitro susceptibility of urease-and ADI-negative triple knockout mutant to acid Given the above results that both the urease and ADI pathway contribute towards the overall acid tolerance of L. hongkongensis, we constructed a triple knockout mutant strain HLHK9ΔureA/arcA1/arcA2 and compared its survival abilities with HLHK9, HLHK9ΔureA and HLHK9ΔarcA1/arcA2 under different acidic conditions in the presence of 50 mM each of L-arginine and urea. The parental and mutant strains displayed similar susceptibilities at pH 5 ( Figure 3C). At pH 4, the survival count of HLHK9ΔureA was similar to that of HLHK9 but there was about 2-log reduction in that of HLHK9ΔarcA1/ arcA2 (p < 0.01) and HLHK9ΔureA/arcA1/arcA2 (p < 0.01) ( Figure 3C), and the reduction trend became more pronounced after 3 and 5 h incubation ( Figure 3D). At pH 2 and 3, the survival counts of HLHK9ΔureA started to decrease (P <0.05), whereas there were dramatic decreases in the survival counts of HLHK9ΔarcA1/arcA2 (p < 0.001) and triple knockout mutant HLHK9ΔureA/ arcA1/arcA2 strains, which were almost completely killed (p < 0.001) ( Figure 3C). These showed that the ADI pathway of L. hongkongensis played a more important role than the urease in resisting acidic environments.

Intracellular survival in J774 macrophages and mRNA expression level analyses
Survival of wild type L. hongkongensis HLHK9, HLHK9ΔureA, HLHK9ΔarcA1/arcA2 and HLHK9ΔureA/ arcA1/arcA2 in J774 macrophages were shown in Figure 4A. Survival of HLHK9ΔureA/arcA1/arcA2 and HLHK9ΔarcA1/arcA2 in macrophages were markedly decreased (p < 0.001 and p < 0.01 respectively) but that of HLHK9ΔureA was slightly decreased (p < 0.05), compared to wild type L. hongkongensis HLHK9. The decrease of survival was more prominent in HLHK9ΔureA/arcA1/ arcA2, compared to HLHK9ΔarcA1/arcA2 (p < 0.05) and HLHK9ΔureA (p < 0.01); and in HLHK9ΔarcA1/arcA2, compared to HLHK9ΔureA (p < 0.05). Given the above results, we further investigated the expression level of ADI genes (arcA1 and arcA2) and ureA gene of wild type L. hongkongensis HLHK9 survived in macrophages using real-time quantitative RT-PCR assay. At 8 h post infection, the mRNA levels of arcA1, arcA2 and ureA genes were markedly increased compared to those at 2 h post infection (p < 0.05, p < 0.01 and p < 0.05 respectively) ( Figure 4B).

Survival of L. hongkongensis strains in BALB/c mice
To further investigate the role of urease and ADI pathway in acid tolerance of L. hongkongensis, we compared the survival ability of HLHK9, mutant strains HLHK9ΔureA, HLHK9ΔarcA1/arcA2 and HLHK9ΔureA/arcA1/arcA2 after transit through the stomach of mice. Using this mouse model, HLHK9ΔureA exhibited similar survival abilities as HLHK9 ( Figure 5). In contrast, the viable counts of HLHK9ΔarcA1/arcA2 and HLHK9ΔureA/ arcA1/arcA2 were reduced by 1.2-log and 1.3-log respectively, compared to that of HLHK9 (p < 0.01) ( Figure 5). This also indicated that the ADI pathway played a more significant role than urease in the survival of L. hongkongensis under the acidic conditions encountered during passage through the mouse gastric environment.
PCR amplification and DNA sequencing of arcA1 and arcA2 A specific 739-bp fragment of arcA1 and a specific 712bp fragment of arcA2 of L. hongkongensis were amplified from the DNA extracts of all 30 human strains, indicating that both arcA1 and arcA2 were present in all 30 human strains. DNA sequencing of the PCR products from five randomly selected L. hongkongensis strains confirmed that the amplified products were arcA1 and arcA2 respectively. Sequence analyses showed that there were 1 to 5 nucleotide differences and one amino acid difference between the 739-bp fragments and the deduced amino acid sequences of the arcA1 genes from these five selected strains and the corresponding region of HLHK9. Similarly, there were 1 to 4 nucleotide differences but no amino acid difference between the 712-bp fragments of the arcA2 genes from these five strains and the corresponding region of HLHK9. Sequence analysis also revealed that most of the conserved residues were present in the partial fragments of arcA1 and arcA2, compared to ADI sequences of other bacteria.

Discussion
We showed that the arc gene cassettes are more important than the urease gene cassette for acid resistance and survival in macrophages in L. hongkongensis. Although both urease and arc gene cassettes have previously been reported to play roles in acid resistance in bacteria, urease function appears to be more important in gastrointestinal tract bacteria such as H. pylori, Yersinia enterocolitica and Klebsiella pneumoniae [16,30,34]. In fact, the mechanisms of acid resistance are similar in both reactions, which result in production of ammonia, thereby increasing the pH of the immediate environment of the bacterium. As for survival in macrophages, ADI pathway has been shown to contribute to survival in macrophages in Salmonella Typhimurium [32], but not in Listeria monocytogenes [29]; and urease has been shown to contribute to survival in macrophages in H. pylori [35], but not in Brucella suis and Brucella abortus [30,36]. To the best of our knowledge, the present study is the first to compare the relative importance of these two acid resistance and intracellular survival mechanisms using in vitro and in vivo models, although these two gene cassettes are present in many gastrointestinal tract bacteria, such as Y. enterocolitica and Enterobacter cloacae. By constructing a series of urease knockout mutants, we found that both structural and accessory genes in the urease gene cassette are crucial for the urease activity; which is in line with previous studies performed in other bacterial species [15,30,37]. Contrary to our initial hypothesis, we observed only a small reduction in survival abilities of the urease knockout mutants in acidic media (pH 2 and 3) and macrophages as well as during gastric passage in the mouse model. This is consistent with our previous recovery of a strain of ureasenegative L. hongkongensis (HLHK30) from an 84-year old male with gastroenteritis. Sequencing of the urease cassette of HLHK30 showed that all eight of the component genes were present with no deletions or frame shift mutations; although there were a number of polymorphic sites that resulted in amino acid changes compared to gene homologues present in HLHK9 ( Figure 1B). On the other hand, the ADI-deficient mutant HLHK9ΔarcA1/arcA2 showed marked reduction in survival abilities in acidic media and macrophages as well as in the mouse model, indicating that arc gene cassettes play a more important role than urease gene cassettes for acid resistance in L. hongkongensis. In fact, the survival abilities of the triple knockout mutant strain HLHK9ΔureA/arcA1/arcA2 were only marginally lower than those of the ADI-deficient double mutant strain HLHK9ΔarcA1/arcA2 in acidic media and macrophages, and both mutant strains had equivalent survival abilities in the mouse model, which further supports the conclusion that ADI play a more important role.
The gene duplication of the arc gene cassettes could be a result of their functional importance in L. hongkongensis. One of the important mechanisms of virulence evolution in bacteria and fungi is gene duplication [38][39][40]. L. hongkongensis is the only bacterium known to possess two adjacent arc gene cassettes. The L. hongkongensis mutant strain containing deletions of the arcA genes in both arc cassettes exhibited a marked reduction in survival abilities compared to the mutant strains containing single deletion of either one of the two arcA genes, indicating that both arc gene cassettes are functional and contribute to acid resistance. Phylogenetic analysis showed that the two copies of arc in L. hongkongensis are clustered in all the four trees constructed using arcA, arcB arcC and arcD [41]. This strongly suggests that the two arc gene cassettes result from a gene cassette duplication event. Interestingly, in our previous study on differential gene expression in L. hongkongensis at different temperatures, it was observed that the two copies of argB, encoding two isoenzymes of N-acetyl-Lglutamate kinase from the arginine biosynthesis pathway, which have distinct biochemical properties, are also clustered phylogenetically [17]. This indicates that these two copies of argB probably also arose as a result of gene duplication. Subsequent evolution enabled the two copies of argB to adapt to different temperatures and habitats. These coincidental findings of gene duplication in two different pathways of arginine metabolism, enabling the bacterium to better adapt to different environmental conditions, argB for temperature adaptation and arc gene cassette for acid resistance, is intriguing.
The present study further strengthened the feasibility of using a conjugation mediated gene deletion system based on a suicide vector in the Neisseriaceae family of β-proteobacteria, a strategy which has been widely used in γ-proteobacteria, such as E. coli, Salmonella Typhimurium and Vibrio cholera [22,42,43]. In our previous studies on plasmid transformation and gene expression system in L. hongkongensis, we observed that plasmids commonly used for expression systems in E. coli did not replicate in L. hongkongensis [44]. Therefore, an E. coli-L. hongkongensis shuttle vector, based on a L. hongkongensis plasmid backbone and origin of replication, was constructed [44]. In our subsequent gene deletion experiments in L. hongkongensis, we used a pBK-CMV plasmid that harbored 1000 bp of genomic upstream and downstream of the target gene, but lacked the target gene, which was transformed into L. hongkongensis. This gene deletion system was successfully used to delete several L. hongkongensis genes, such as the flgG flagellar gene. However attempts to delete the ureA, ureB, ureC and ureI genes were all unsuccessful (unpublished data). Therefore, the present gene deletion system, which was first used in E. coli [42], and also recently used in Chromobacterium violaceum, another pathogenic bacterium of the Neisseriaceae family [45], was used for knocking-out genes from the urease and arc gene cassettes. Further experiments will elucidate whether this gene deletion system is also useful for knocking out genes in other important bacteria of the Neisseriaceae family, such as the Neisseria gonorrhoeae and Neisseria meningitidis.

Conclusions
ADI pathway is far more important than urease for acid resistance and intracellular survival in L. hongkongensis. The gene duplication of the arc gene cassettes could be a result of their functional importance in L. hongkongensis.