DNA Adenine Methyltransferase Influences the Virulence of Aeromonas hydrophila

Abstract

Among the various virulence factors produced by Aeromonas hydrophila, a type II secretion system (T2SS)-secreted cytotoxic enterotoxin (Act) and the T3SS are crucial in the pathogenesis of Aeromonas-associated infections. Our laboratory molecularly characterized both Act and the T3SS from a diarrheal isolate, SSU of A. hydrophila, and defined the role of some regulatory genes in modulating the biological effects of Act. In this study, we cloned, sequenced, and expressed the DNA adenine methyltransferase gene of A. hydrophila SSU (dam_AhSSU) in a T7 promoter-based vector system using Escherichia coli ER2566 as a host strain, which could alter the virulence potential of A. hydrophila. Recombinant Dam, designated as M.AhySSUDam, was produced as a histidine-tagged fusion protein and purified from an E. coli cell lysate using nickel affinity chromatography. The purified Dam had methyltransferase activity, based on its ability to transfer a methyl group from S-adenosyl-l-methionine to N⁶-methyladenine-free lambda DNA and to protect methylated lambda DNA from digestion with DpnII but not against the DpnI restriction enzyme. The dam gene was essential for the viability of the bacterium, and overproduction of Dam in A. hydrophila SSU, using an arabinose-inducible, P_BAD promoter-based system, reduced the virulence of this pathogen. Specifically, overproduction of M.AhySSUDam decreased the motility of the bacterium by 58%. Likewise, the T3SS-associated cytotoxicity, as measured by the release of lactate dehydrogenase enzyme in murine macrophages infected with the Dam-overproducing strain, was diminished by 55% compared to that of a control A. hydrophila SSU strain harboring the pBAD vector alone. On the contrary, cytotoxic and hemolytic activities associated with Act as well as the protease activity in the culture supernatant of a Dam-overproducing strain were increased by 10-, 3-, and 2.4-fold, respectively, compared to those of the control A. hydrophila SSU strain. The Dam-overproducing strain was not lethal to mice (100% survival) when given by the intraperitoneal route at a dose twice that of the 50% lethal dose, which within 2 to 3 days killed 100% of the animals inoculated with the A. hydrophila control strain. Taken together, our data indicated alteration of A. hydrophila virulence by overproduction of Dam.

Aeromonas hydrophila is both a human and an animal pathogen, and the diseases associated with this bacterium in humans include gastroenteritis, wound infections, and septicemia (36, 88). In immuno-compromised individuals, some of the more serious complications associated with A. hydrophila infections are cellulitis, myonecrosis, and ecthyma gangrenosum, which often are fatal (2, 36). A variety of virulence factors/mechanisms, such as the surface layer, capsule, different forms of pili (e.g., type IV [Tap] and bundle-forming pili), toxins, quorum sensing, biofilm formation and, more recently, the type III secretion system (T3SS), were identified in various Aeromonas species that are involved in the pathogenesis of Aeromonas infections (10, 11, 40, 52, 75, 82, 89). Currently, the set of virulence factors crucial in causing human diseases is unknown in this pathogen and, therefore, the search for new virulence factors and their regulation must continue.

Our laboratory defined the role of three enterotoxins of a diarrheal isolate, SSU of A. hydrophila, namely, cytotoxic enterotoxin (Act) and cytotonic enterotoxins (heat-labile [Alt] and heat-stable [Ast]), in causing gastroenteritis (1, 72), based on case-control human studies. The role of these enterotoxins in a mouse model of gastroenteritis was also shown by preparing specific enterotoxin gene-targeted mutants via marker exchange mutagenesis. Our studies indicated that Act that was secreted by the T2SS contributed maximally to fluid secretion, and this toxin also had hemolytic and cytotoxic activities, in addition to those associated with lethality in mice (27, 64, 72, 73). Further, we characterized the T3SS from an A. hydrophila SSU strain which contained 35 open reading frames (ORFs) (75). We recently noted minimal lethality (10%) in mice infected with the mutant of A. hydrophila SSU that had deletion of both the act and Aeromonas outer membrane protein B (aopB) genes at a dose two times the 50% lethal dose (LD₅₀). The aopB gene constitutes part of the T3SS and is involved in cytotoxicity of the host cell (75). This finding was compared to those in animals infected with act and aopB mutants alone (40 to 50% lethality) and thus signified that both the T2SS-associated Act and the T3SS played crucial roles in bacterial virulence (75). Further, we identified the role of the ferric uptake regulator (fur) and glucose-inhibited division gene (gidA) in modulating the biological effects of Act (73, 74). In the search for additional genes that could alter bacterial virulence, we now report characterization of the DNA adenine methyltransferase (Dam) from A. hydrophila SSU that modulates the function of both T3SS and T2SS-associated Act.

DNA methylation occurs in bacteria, plants, and mammals and, recently, methylated DNA was also reported in Drosophila melanogaster (47, 48). DNA methyltransferases (MTases) catalyze methylation of either the cytosine residues at the C-5 or N-4 position or at the N-6 position of an adenine residue within the DNA (85). Specifically, Dam exerts its function by DNA methylation at adenine residues in GATC sequences (59). Methylation is a postreplicative process, and the newly replicated DNA is methylated only on the parental strand. Therefore, this hemimethylated DNA is distinct from the rest of the chromosomal DNA. The hemimethylated status of newly synthesized DNA provides a timeframe during which cellular processes, such as DNA replication (43, 55, 68) and repair of mismatched bases as well as alteration of gene expression (50), occur. During DNA replication, the GATC hemimethylated sites have a high affinity for SeqA protein, which is a negative regulator of replication initiation and essential for sequestration, a process that blocks secondary replication initiation events (46, 83). Likewise, during methyl-directed mismatch repair, mutH binds to hemimethylated DNA and cleaves the nonmethylated strand (3, 51). Finally, activation of a promoter upon hemimethylation could result in a burst of transcription shortly after passage of the replication fork, thereby linking gene expression to the cell cycle. Indeed, in the case of the Tn10 transposase gene, an increase in the promoter activity was directly demonstrated when DNA was in the hemimethylated state (55).

In addition to the above-mentioned functions of the hemimethylated DNA, GATC methylation can also influence gene expression. For example, the methylated GATC sequences in gene promoter regions can alter the affinity of regulatory protein(s) to DNA target sites. Conversely, regulatory proteins may bind nonmethylated DNAs with high affinity, thus protecting specific DNA sequences from methylation, resulting in the formation of DNA methylation patterns (DMPs) (32, 45, 51). One of the best-studied examples for regulation of gene expression by DMPs is the pyelonephritis-associated pilus (pap) operon of uropathogenic Escherichia coli. The DMPs influence the binding of the regulatory proteins Lrp and PapI to the papBA pilin promoter, which correlates with the on and the off stages of pilus expression in this bacterium (33, 45).

Dam methylation has received significant interest recently because of its impact on the virulence of several pathogens, such as Vibrio cholerae (38, 49), Salmonella enterica serovar Typhimurium (19, 20, 28, 57), pathogenic E. coli strains (12, 43), Yersinia pseudotuberculosis (4, 39), Yersinia enterocolitica (25), Haemophilus influenzae (9, 90), and others. However, in Shigella flexneri, the dam mutants showed no attenuation of virulence (34). Overall, data seem to support the hypothesis that Dam could globally alter virulence gene expression in gram-negative bacteria (31, 45).

The role of DNA methylation has been reported in the viability of V. cholerae, Y. pseudotuberculosis, and Y. enterocolitica (25, 38). Further, it was shown that Dam overproduction led to attenuation in the virulence of these pathogens (25, 38). Dam was required for virulence in Pasteurella multocida, known to cause bovine respiratory disease (14). Recently, the gene encoding Dam was cloned and sequenced from Actinobacillus actinomycetemcomitans, which is implicated in causing human periodontal diseases (21). Therefore, Dam presents the exciting possibility that it may play a role in the virulence of a broad range of pathogens and, thus, further investigation is merited.

In this paper, we showed that a diarrheal isolate, SSU of A. hydrophila, harbored the dam gene and that its overexpression attenuated bacterial virulence, specifically that of T3SS-associated cytotoxicity, motility, and virulence in a mouse lethality model. This is the first report of characterization of the dam gene from Aeromonas species. According to the recent nomenclature for methyltransferases and their genes, we denoted the A. hydrophila SSU strain DNA adenine MTase as M.AhySSUDam and the dam gene as dam_AhSSU (66).

MATERIALS AND METHODS

Reagents.

Restriction enzymes, Taq DNA polymerase, T4 ligase, and N⁶-methyladenine-free lambda DNA were purchased from New England BioLabs, Beverly, MA. Proteinase K and unlabeled S-adenosyl-l-methionine (AdoMet) were obtained from Sigma, St. Louis, MO. We purchased [methyl-³H]AdoMet (15 Ci/mmol; 1 mCi/ml) from New England Nuclear, Boston, MA. ProBond resin charged with nickel was from Invitrogen, Carlsbad, CA.

DNA and RNA manipulations.

Plasmid DNA was isolated using a QIAprep Spin Miniprep kit (QIAGEN, Valencia, CA). A. hydrophila genomic DNA (gDNA) for sequencing was isolated using the published protocol with some modifications (63) or by utilizing a DNeasy tissue kit (QIAGEN) for PCR assays. DNA fragments were purified using a PCR purification kit or gel extraction kit (QIAGEN). Bacterial RNA was isolated using a RiboPure-bacteria RNA isolation kit (Ambion, Austin, TX) according to the manufacturer's protocol.

Bacterial strains, plasmids, and growth conditions.

Bacterial strains and plasmids used in the study are listed in Table 1. A. hydrophila SSU and E. coli cultures were grown at 37°C in Luria-Bertani (LB) medium and LB agar plates (69). The medium was supplemented with l-arabinose (0.2%) or d-glucose (0.2%) when the dam gene was expressed from the pBAD-dam_AhSSU plasmid (Table 1) under the control of an arabinose-inducible P_BAD promoter. d-Glucose was used to repress the expression of the dam gene from the pBAD-dam_AhSSU plasmid (29). LBNS solid medium (LB with no NaCl) was supplemented with 10% sucrose for levansucrase (sacB) counterselection and 0.001% arabinose when used for the isolation of dam mutants. Appropriate antibiotics were used at previously described concentrations (72).

TABLE 1.

Strains and plasmids used in this study

Strain or plasmid	Relevant characteristic(s)	Source or reference
A. hydrophila SSU strains		CDC, Atlanta, Ga.a
SSU-R	Rifampin-resistant (Rifr) strain of A. hydrophila SSU	Laboratory stock
Dam-overproducing strain	A. hydrophila SSU-R harboring plasmid pBAD-damAhSSU; Rifr Apr	This study
Control strain	A. hydrophila SSU-R harboring plasmid pBAD; Rifr Apr	This study
dam mutant	Chromosomal copy of dam gene was deleted from Dam-overproducing strain of A. hydrophila SSU-R (Δdam); Rifr Apr Kmr	This study
E. coli strains
TOP10	F−mcrA Δ(mrr-hsdRMS-mcrBC) φ80lacZΔM15 ΔlacX74 recA1 deoR araD139 Δ(ara-leu)7697 galU galK rpsL (Strr) endA1 nupG	Invitrogen
JM109	endA1 recA1 gyrA96 thi hsdR17 (rK− mK+) relA1 supE44 Δ(lac-proAB) [F′ traD36 proAB lacIqZΔM15]	Promega
ER2566	F− λ−fhuA2 ompT lacZ::T7gene1 gal1 sulA11 Δ(mcrC-mrr)114::IS10 (R9mcr-73::mini-Tn10-Tets) 2R(zgb-210::Tn10) (Tets) endA1	New England BioLabs
SM10(λpir)	Kmr; thi-1 thr leu tonA lacY supE recA::RP4-2-Tc::Mu pir	22
Plasmids
pCR2.1	TA cloning vector; Apr Kmr	Invitrogen
pCR2.1-dam	TA cloning vector pCR2.1 harboring a portion of the dam gene (400 bp) which was PCR amplified from A. hydrophila SSU-R gDNA; Apr Kmr	This study
pET-30a(+)	T7 promoter-based prokaryotic expression vector; Kmr	Novagen
pET-30a(+)-damAhSSU	dam gene of A. hydrophila cloned at NdeI/XhoI sites of pET-30a(+) vector for purification of DAM protein; Kmr	This study
pBAD/Thio-E	Arabinose-inducible araBAD promoter-based prokaryotic expression vector; Apr	Invitrogen
pBAD	Control vector, which was derived from pBAD/Thio-E by deleting the NcoI/PmeI fragment; Apr	This study
pBAD-damAhSSU	dam gene of A. hydrophila cloned at the NcoI/PmeI sites of pBAD/Thio-E vector for hyperexpression; Apr	This study
pKD4	Kmr Apr; template plasmid for kanamycin cassette (Km1)	16
pUC4K	A plasmid containing 1.2-kb kanamycin cassette (Km2); Kmr	Amersham
pDMS197	A suicide vector; oriT oriV sacB; Tcr	22
pDMS197upkm1down	Suicide vector pDMS197 containing Km1 cassette flanked by upstream and downstream sequences of the dam gene; Kmr Tcr	This study
pDMS197upkm1/Km2down	Suicide vector pDMS197 containing Km cassettes (Km1 and Km2) flanked by upstream and downstream sequences of the dam gene; Kmr Tcr	This study
pTP166	A plasmid containing E. coli dam gene; Apr	7

^aCDC, Centers for Disease Control and Prevention.

Primers and PCR assays.

The primers used for various experiments are indicated in Table 2 and were synthesized by Integrated DNA Technologies, Inc. (Coralville, IA). PCR assays were performed with 100 ng of gDNA and 50 ng of plasmid DNA. The PCR program used was as follows: 94°C for 5 min (denaturation) followed by 25 cycles at 94°C for 30 s, 55°C for 30 s, and 72°C for 1 to 3 min (depending on the size of the amplified DNA fragment). The PCR products were then held at 72°C for 7 min and stored at 4°C or at −20°C until they were subsequently run on a 0.8% agarose gel. We performed PCR assays from single bacterial colonies in TEP buffer (10 mM Tris-HCl, pH 8.0, 1 mM EDTA, and 50 μg/ml proteinase K). Briefly, a single colony was removed from the LB agar plate and resuspended in 12 μl of the TEP buffer. The mixture was incubated first at 55°C and then at 85°C for 15 min. After incubation on ice for 1 min, cellular debris was removed and 3 μl of the clarified extract was used in a PCR assay.

TABLE 2.

Primers used in this study

Primer	Sequence (5′-3′)	Purpose
dam1	GGTGCATAAGGCGGATCGCA	PCR amplification of a portion (400 bp) of the dam gene from A. hydrophila SSU
dam2	TACATCCTTGCCGATATCAA
dam3	GCGCCGAAGAGGATTGGGTCATCTATTGCGATCCG	DNA sequencing of A. hydrophila gDNA to obtain coding region of the dam gene
dam4	GGGTGGTTGTAAAGCCCGATCAGATCCGGATTGAT
damP1	CCATTTTAAAAAAGCGCGTGTTTTTTTCAT	Primer extension analysis
damP2	CCCGGCCCATTTTAAAAAAGCGCGTGTTTT
damP3	TTTCCCCCGGCCCATTTTAAAAAAGCGCGT
damN-NdeI	GGAATTCCATATGAAAAAAACACGCGCTTTTTTAA	Cloning of the dam gene into pET-30a(+) vector
damC-XhoI	CCGCTCGAGGCCGAGTGGCGCCAGTTCGGCGTCGCTCGGG
damN-NcoI	CATGCCATGGATGAAAAAAACACGCGCTTTTTTAAAATGG	Cloning of the dam gene into pBAD/Thio-E vector
damC-PmeI	AGCTTTGTTTAAACGCCGAGTGGCGCCAGTTCGGCGTCGC
KF	ATGATTGAACAAGATGGATTGCACG	PCR amplification of a Km cassette (Km1) from pKD4 plasmid
KR	TCAGAAGAACTCGTCAAGAAGGCGA
KpnI-F2	CGGGGTACCCTGTCGCCCGCCTTGCTGAAGGGTCAACCCTG	PCR amplification of an upstream flanking sequence to the dam gene
R2	AATCCATCTTGTTCAATCATGGAAATCCGGTGTTAGAAGG
R3-XbaI	GCTCTAGATCGGCGGAGAGGATCGAGGGGGCAATCAGAA	PCR amplification of a downstream flanking sequence to the dam gene
F3	TTCTTGACGAGTTCTTCTGAGCGGCACCTTGGGTGCCGTT
F1	GGCTGCCGCCGAGCCGGTGGCGACTCCGGCGCCCAAGGTG	PCR verification of the dam mutant
R1	TGACCATCAGGTGCACGTCGATGGGGGCTTCGATACCGTA

^aUnderlining indicates restriction endonuclease site.

^bBold sequence indicates primer's overlapping sequence with the Km1 cassette (Fig. 5). Bold italic nucleotides represent the start and stop codons of the kanamycin cassette (Km1) from pKD4 plasmid.

Southern blot analysis.

The digested gDNA of A. hydrophila or the PCR products were transferred to a nylon membrane (Gibco BRL, Gaithersburg, MD) and baked at 80°C in a vacuum oven for 2 h. The blots were prehybridized for 6 h and hybridized overnight under low-stringency conditions in a solution containing 30% formamide, 1 mM EDTA, 5× SSC (1× SSC is 0.15 M NaCl plus 0.015 M sodium citrate), 5× Denhardt's solution, 10 μg/ml salmon sperm DNA, and 0.1% sodium dodecyl sulfate (SDS) at 42°C (69). For hybridization, a [α-³²P]dCTP (ICN, Irvine, CA)-labeled E. coli dam gene probe (the 854-bp XbaI-PvuII fragment of pTP166 plasmid) was used (7). The membranes were washed at 54°C with 5× SSC plus 0.1% SDS for 40 min and then in 4× SSC plus 0.1% SDS for another 40 min. This analysis identified the initial 400-bp fragment of the dam gene, which was used to obtain the full-length dam_AhSSU gene.

Fosmid library construction.

A. hydrophila gDNA (5 μg) was sheared to generate fragments in the 25- to 40-kb size range. The sheared DNA was end repaired to generate blunt ends and size selected using a 1% low-melting-point agarose gel. The size-selected DNA was then ligated to the dephosphorylated blunt-ended pEpiFOS-5 Fosmid vector (Epicenter, Madison, WI) and packaged using MaxPlax Lambda packaging extracts. E. coli EPI100 plating cells (Epicenter) were used as the host. Fosmid colonies were lifted onto nylon filters from LB agar plates containing 12.5 μg/ml chloramphenicol (Cm). The gene encoding Cm^r was carried by the fosmid vector. The filters were screened with the [α-³²P]dCTP-labeled 400-bp dam gene probe by colony blot hybridization. Positive colonies were identified and grown overnight in LB medium with Cm and 10 mM MgSO₄ for DNA isolation.

Colony blot hybridization.

Nylon membrane filters used for the colony blots were removed from the agar plates and processed as described elsewhere (69). The filters were dried and baked at 80°C for 2 h. Filters were prewashed with 50 mM Tris-HCl (pH 8.0), 1 M NaCl, 1 mM EDTA, and 0.1% SDS at 42°C to remove cell debris, prehybridized, and then hybridized with the [α-³²P]dCTP-labeled 400-bp dam gene probe under high-stringency conditions. The filters were prehybridized using Quickhyb solution at 68°C as described by the manufacturer (Stratagene, La Jolla, CA) and hybridized with ³²P-labeled probe. After 2 to 3 h, the filters were washed with 2× SSC, 0.1% SDS for 40 min and 1× SSC, 0.1% SDS for 40 min at 68°C and exposed to autoradiograph film overnight at room temperature.

Primer extension analysis.

The primer extension system kit from Promega (Madison, WI) was used to determine the presumptive transcriptional start site of the dam_AhSSU gene. Primers (10 pM each) complementary and downstream to the initiation codon of the dam_AhSSU gene (Table 2 and Fig. 1) were end labeled with 30 μCi of [γ-³²P]ATP (ICN). For each extension reaction mixture, 10 pM of labeled primer was mixed with 50 μg of RNA, heated to 58°C for 20 min, and cooled to room temperature. Subsequently, the mixture was incubated at 42°C with 1 U of avian myeloblastosis virus reverse transcriptase (Promega). After 30 min of incubation, the products were mixed with loading buffer, heated at 90°C for 10 min, and analyzed by running on a denaturing polyacrylamide gel containing 8% acrylamide (19:1 acrylamide-bis) and 7 M urea as described in the protocol (Promega). The length of the cDNA on the gel reflected the number of bases between the labeled nucleotide of the primer and the 5′ end of the RNA and was compared with that of the labeled markers, which ranged in size from 24 to 311 bp.

FIG. 1.

Sequence of a 1,020-bp DNA fragment that encompasses the dam_AhSSU gene of A. hydrophila SSU. Presented is the nucleotide sequence of the coding region of the dam_AhSSU gene with its deduced amino acid sequence. The start and stop codons of the dam_AhSSU gene are in capital letters. Underlined DPPY amino acids are the conserved catalytic motif for N⁶-methyladenine MTase. The sequences of the primers used for primer extension analysis are indicated with arrows. The potential transcriptional start site (G) is indicated by the capital letter and is underlined. The rectangular boxes represent putative −10 and −35 boxes.

Transformation of E. coli and A. hydrophila.

Competent E. coli cells for transformation were prepared using the Z-Competent E. coli transformation kit from Zymo Research, Orange, CA. For electroporation, A. hydrophila SSU (50 ml) was grown until early log phase and resuspended in 300 mM of sucrose (sequentially in the same volume, in a 0.5 volume, and finally in a 0.01 volume) (77) and electroporated with the DNA (1 to 3 μg) by using GenePulser Xcell in 0.2-cm gap cuvettes (Bio-Rad, Hercules, CA).

Purification of DNA adenine methyltransferase.

PCR amplified from A. hydrophila SSU gDNA, the dam_AhSSU gene was cloned into the pET-30a(+) T7 promoter-based expression vector (Novagen, San Diego, CA) using primers with NdeI and XhoI restriction enzyme sites (Table 2). To overexpress the dam gene in E. coli, the recombinant plasmid was transformed into the E. coli ER2566 strain, which harbored the T7 RNA polymerase gene under the control of the lac promoter. The E. coli(pET-30a-dam_AhSSU) culture was grown in 500 ml of the LB medium with kanamycin (Km; 30 μg/ml) to an optical density at 600 nm (OD₆₀₀) of 0.4 to 0.5 before induction with a final concentration of 1 mM isopropyl-β-d-thiogalactopyranoside for 3 h. The bacterial cells were harvested, resuspended in 10 ml of appropriate buffer (Tris-HCl, pH 8.0, 100 mM NaCl, 1 mM MgCl₂, and 5% glycerol), and disrupted by sonication. The cell lysate was passed through ProBond resin (3-ml bed volume in a 5-ml column) charged with nickel according to the manufacturer's recommendations. The resin was washed with a five-column volume of the wash buffer, followed by elution of the proteins using different concentrations (0.25 to 1 M) of imidazole. The purity of Dam was examined by SDS-12% polyacrylamide gel electrophoresis followed by Coomassie blue staining of the gel. After dialysis and concentration, purified M.AhySSUDam was used to determine enzymatic activity by the DNA adenine methylation assays as described below.

Cloning of the dam gene under the araBAD promoter.

To regulate dam gene expression, pBAD-dam_AhSSU plasmid was generated using damN-NcoI and damC-PmeI primers (Table 2) by replacing the NcoI-PmeI fragment of the pBAD/Thio-E vector (Invitrogen) under its arabinose P_BAD promoter. To induce expression of the gene from the plasmid, arabinose (0.2%) was added to the medium (29). We referred to this culture as the Dam-overproducing A. hydrophila strain (Table 1). In some experiments, we also examined tight regulation of the dam gene in the pBAD system in A. hydrophila by either omitting arabinose or adding glucose and evaluating the effect of such a strain on bacterial virulence. To construct the pBAD plasmid for use as a control, the pBAD/Thio-E vector was digested with NcoI and PmeI restriction endonucleases, treated with DNA polymerase I (Klenow fragment), and ligated. Both plasmids were then subjected to transformation/electroporation in E. coli JM109 and A. hydrophila SSU strains. The latter with pBAD vector alone was designated the A. hydrophila control strain (Table 1).

DNA methylation assays.

The GATC-specific DNA methylation assay was conducted according to the published procedure (56). Briefly, the methylation mixture (20 μl) contained 20 mM Tris-HCl, pH 8.0, 50 mM NaCl, 7 mM 2-mercaptoethanol, 1 mM EDTA, 0.1 mg/ml bovine serum albumin (BSA), 1 μg N⁶-methyladenine-free lambda DNA, purified A. hydrophila Dam (3 μg), and 50 μM unlabeled AdoMet. After 1 h of incubation at 37°C, the reaction was stopped by heat inactivation at 65°C for 15 min. Then, methylated lambda DNA was examined by digestion with 5 U of restriction enzyme DpnI or DpnII in recommended buffers. The reaction mixtures were incubated for an additional 1 h at 37°C, and digested DNA was analyzed by 1% agarose gel electrophoresis after staining the gel with ethidium bromide.

We also measured DNA adenine MTase activity using [methyl-³H]AdoMet (38). Briefly, A. hydrophila SSU control and Dam-overproducing strains were grown for 5 h in the presence of 0.2% arabinose. Bacterial cultures were centrifuged, and pellets were resuspended in the buffer (50 mM NaPO₄, pH 7.0, 200 mM NaCl, and 0.1% Triton X-100 [TX-100]). After incubation for 30 min on ice with lysozyme (200 μg/ml), the cells were disrupted by sonication. The cell extracts were recovered, and the total amount of protein in the extracts was determined by a Bradford protein assay (6).

The activity of M.AhySSUDam was quantified by adding 5-μl aliquots of the cell extract dilutions to 45 μl of the methylation reaction mixture that contained 20 mM Tris-HCl, pH 8.0, 1 mM EDTA, 1 mM dithiothreitol, 5 μg lambda DNA N⁶-methyladenine free, and 1 μCi of [methyl-³H]AdoMet. The reaction mixtures were incubated for 30 min at 37°C, 40-μl aliquots were removed, and the reactions were stopped by transfer of these aliquots to Whatman DE81 filter paper. The filters were washed three times with 50 mM KH₂PO₄, twice with 80% ethanol, and once with 95% ethanol. The filters were allowed to dry, and the amount of methyl-³H incorporated into the lambda DNA was determined using 3 ml of ScintiVerse (Fisher Scientific, Pittsburg, PA) in a Beckman scintillation counter LS 6500. The MTase activity in each cell extract was calculated as counts per minute of transferred radioactive methyl groups to lambda DNA by A. hydrophila Dam per microgram of total protein in the cell extracts.

Northern blot analysis.

The total RNA was isolated from the control and Dam-overproducing strains by using an RNAqueous kit (Ambion). The RNA samples were subjected to Northern blot analysis as described previously (73). A 1.4-kb [α-³²P]dCTP-labeled act gene that was PCR amplified from the plasmid pXHC95 (88) was used as a probe to detect the act transcript, which was quantitated using a PhosphorImager Storm 860 (Molecular Dynamics, Sunnyvale, CA). The amount of RNA loaded onto each lane of the gel was determined by scanning 23S or 16S rRNA bands in ethidium bromide-stained gel using the Gel Doc 2000 system (Bio-Rad), and the intensity of the act signal was normalized to the RNA load. Likewise, PCR-amplified aopB, aopD, ascV, and acrV DNA fragments representing components of the T3SS (75) were used for Northern blot analysis.

ELISA.

Aliquots (100 μl) of fivefold-diluted culture supernatants from the Dam-overproducing and the control strains of A. hydrophila, which were grown in the absence or presence of 0.2% arabinose overnight at 37°C (with shaking at 180 rpm), were used for coating the wells on the first row of 96-well microtiter plates in sodium bicarbonate buffer (pH 9.6). Each of the tested samples was triplicated and followed by twofold serial dilution on the plates. After overnight incubation at 4°C, the plates were washed three times with the wash buffer (phosphate-buffered saline [PBS] containing 0.05% Tween 20 and 0.1% BSA) and blocked with 2% BSA in PBS for 2 h at room temperature. After blocking, the plates were washed two times with wash buffer, and affinity-purified antibodies to Act (diluted 1:1,000) were added. After 1 h of incubation at room temperature, the plates were washed three times with wash buffer, goat anti-rabbit secondary antibodies that were conjugated to horseradish peroxidase (Southern Biotechnology Associates, Inc., Birmingham, AL) were added at a dilution of 1:2,000 for 1 h, then the wells of the microtiter plates were washed four times with wash buffer, and TMB substrate (3,3′,5,5′-tetramethyl-benzidine; Sigma Chemical Co.) was added (71). The absorbance at 370 nm was measured using a VERSA_max microplate enzyme-linked immunosorbent assay (ELISA) reader (Molecular Devices Corporation, Sunnyvale, CA). The data were presented as ELISA titers/ml/10⁸ CFU.

Construction of the A. hydrophila SSU dam chromosomal deletion mutant.

To obtain the dam knockout mutant, we used a method for inactivating chromosomal genes as described recently (17) with some modifications. In order to generate the kanamycin resistance (Km^r) gene cassette flanked with upstream and downstream sequences to the dam gene, three PCR synthesis reactions were performed. A 794-bp PCR fragment (the Km^r gene cassette, designated Km1) was amplified using the pKD4 plasmid (16) as a template and KF/KR primers (Table 2; see also Fig. 5, below). The upstream (144-bp) and downstream (369-bp) sequences to the dam gene were also independently amplified using A. hydrophila gDNA, KpnI-F2/R2 pair primers, and R3-XbaI/F3 pair primers, respectively (Table 2; see also Fig. 5). Primers R2 and F3 contained at their 5′ ends an extension of 20 nucleotides that overlapped with KF and KR primers, respectively (Table 2; Fig. 5). Primers KpnI-F2 and R3-XbaI contained restriction sites to clone the final PCR product into a pDMS197 suicide vector (Table 2; Fig. 5).

FIG. 5.

Construction of the chromosomal dam deletion mutant of A. hydrophila SSU. By using primer pairs KpnI-F2/R2, F3/R3-XbaI, and KF/KR, the flanking sequences (gray boxes) to the dam gene (open box) as well as the coding region of a kanamycin resistance gene cassette Km1 (dotted box) were PCR amplified from the chromosome of A. hydrophila and the pKD4 plasmid, respectively. The primers R2 and F3 contained at their 5′ ends an extension of 20 nucleotides that were overlapping (small dotted boxes) with KF and KR primers, respectively. Through these overlapping regions, the three PCR products were joined via a second PCR amplification using KpnI-F2/R3-XbaI primers. The resulting PCR DNA was cloned into the pDMS197 suicide vector at the KpnI-XbaI sites. As the gene encoding Km1 did not express efficiently, we introduced another kanamycin cassette, Km2 (hatched box), from the pUC4K plasmid at the PstI site within the Km1 cassette. This construct was delivered into the A. hydrophila strain harboring a pBAD-dam_AhSSU plasmid (Table 1). By homologous recombination, the dam gene on the chromosome of A. hydrophila was replaced with the intact Km2, which was flanked by the partial sequences of the Km1 cassette. The arrows represent the direction and position of different primers. The figure is not drawn to scale.

These three PCR products were mixed in equimolar concentrations and used for the next PCR with KpnI-F2 and R3-XbaI primers to amplify a 1,307-bp up-km1-down DNA fragment (see Fig. 5, below). To test each region (up, km1, and down) of this construction and its respective integrity, PCR assays were also performed with different pairs of primers. To generate the pDMS197-upkm1down recombinant plasmid (Table 1), both the suicide vector and the 1,307-bp PCR fragment were digested with KpnI/XbaI restriction enzymes, ligated, and transformed into E. coli SM10 strain, which contained λpir, allowing replication of the suicide vector only in this strain (22).

A. hydrophila (with the pBAD-dam_AhSSU plasmid) and E. coli SM10λpir (with the pDMS197-upkm1down plasmid) strains were conjugated as described previously (88). A rifampin-resistant (Rif^r) spontaneous mutant of A. hydrophila (designated SSU-R) was used for conjugation (Table 1). Transconjugants were plated onto LBNS agar plates with Rif (200 μg/ml), Km (50 μg/ml), 10% sucrose, and 0.001% arabinose. The colonies that were resistant to Km, Rif, and sucrose but sensitive to tetracycline (Tc; the Tc-encoding gene was carried by the suicide vector) were picked up for further analysis. The cultures were identified as Aeromonas by a positive oxidase test (35). We noted that expression of the Km1 cassette was not high enough, resulting in false-positive clones during the screening process. Therefore, we introduced a second Km^r (Km2) cassette (1,240 bp, obtained by PstI restriction enzyme digestion of the pUC4K plasmid) at the PstI restriction site (that existed within the Km1 cassette) of the pDMS197-upkm1down plasmid, resulting in the generation of another recombinant, the pDMS197-upkm1/km2down plasmid. This plasmid was used for generation of the dam mutant of A. hydrophila (Table 1; see also Fig. 5, below).

Cell membrane integrity and binding assays.

The integrity of the cell envelope of both Aeromonas strains (with the pBAD vector alone or the pBAD-dam_AhSSU plasmid) was examined by analyzing sensitivity to TX-100 (1 to 2%) (13, 24, 70) and by evaluating the outer membrane permeability change with Torula yeast RNA (Sigma), which measured periplasmic RNase I leakage from bacteria. For membrane permeability changes, bacterial cells (equal number) were spotted onto LB agar plates containing 1.5% Torula yeast RNA and incubated overnight at 37°C. Subsequently, 10% trichloroacetic acid was added onto the plates, and the RNA leakage from bacterial cells was examined by measuring the diameter of the clear zones around the bacterial colonies.

We infected human HT-29 colonic epithelial cells (American Type Culture Collection, Manassas, VA) with the above-mentioned bacteria at a multiplicity of infection (MOI) of 10 and incubated them at 37°C for 1 h. Unbound bacteria were aspirated, cells were washed four times with 1× Dulbecco's PBS (DPBS) and lysed with 0.1% TX-100, and various dilutions of the cell lysates were plated onto 1.5% LB agar plates for determining CFU (24). The bacteria were grown in the presence of 0.2% arabinose.

To evaluate in vivo binding of A. hydrophila control and Dam-overproducing strains, 6- to 8-week-old female Swiss-Webster mice (Taconic Farms, California) were used. Briefly, mice were anesthetized under halothane and a ventral midline incision was made. A single 5-cm segment of small intestine was ligated with 00 suture and injected with 100 μl of the above-mentioned bacteria (1 × 10⁷ CFU). After 2 h of infection, the animals were euthanized by cervical dislocation, and the intestinal loops were removed. The injected loops were measured, extensively washed in PBS, homogenized, and examined for colonization by determining CFU on LB agar plates containing antibiotics rifampin and ampicillin (200 and 100 μg/ml, respectively) to select only for A. hydrophila.

Motility assay.

LB medium with 0.35% agar was used to characterize the motility phenotype of A. hydrophila control and Dam-overproducing strains. The overnight cultures grown in the presence of 0.2% arabinose were adjusted to the same optical density, and equal numbers of CFU (10⁶) were stabbed into 0.35% LB agar plates containing 0.2% arabinose. Plates were incubated at 37°C overnight, and the motility was assayed by examining migration of bacteria through the agar from the center towards the periphery of the plate (70).

Cytotoxicity assay.

RAW 264.7 murine macrophages (American Type Culture Collection) were seeded into 96-well plates (1 × 10⁵ cells/well) and infected with either the live A. hydrophila SSU control or the Dam-overproducing strain at an MOI of 10 for 3 to 4.5 h to examine the T3SS-associated cytotoxicity (70). Both strains were grown in the presence or absence of 0.2% arabinose. Host cells were also treated with 5 μl of filter-sterilized, overnight-grown bacterial culture supernatants (for determining T2SS Act-associated toxicity) (70). After incubation at 37°C for 2 h, the tissue culture medium was examined for the release of lactate dehydrogenase (LDH) enzyme using the CytoTox96 kit (Promega). The percentage of LDH released by RAW 264.7 macrophages infected with bacterial cells was determined following the manufacturer's instructions. In macrophages treated with culture supernatants, cytotoxicity was reported per milliliter of the culture filtrate per 10⁸ CFU and expressed as a fold change when compared between A. hydrophila SSU control and Dam-overproducing strains.

Hemolytic activity.

A. hydrophila SSU control and Dam-overproducing strains were grown overnight in LB medium with 0.2% arabinose and ampicillin (Ap; 100 μg/ml). The culture supernatants were collected and treated with trypsin to convert the precursor form of T2SS-associated Act to its mature form (13, 24, 70). For the hemolytic activity assay, 100 μl of 1× DPBS was added to each of the wells of a 96-well microtiter plate. The above-treated culture supernatants were added to the first well in each row of the microtiter plate followed by serial twofold dilution and the addition of 100 μl of 3% rabbit erythrocytes (Colorado Serum Co., Denver, CO). Our negative control included 1× DPBS and trypsin alone. The plate was incubated at 37°C for 1 h and observed for hemolytic activity associated with Act. The supernatants were taken from those wells that showed partial lysis of rabbit erythrocytes, and the hemoglobin release was recorded at 540 nm using a microplate ELISA reader. The hemolytic titers were calculated as the value of the hemoglobin release multiplied by the dilution of the culture supernatant. The hemolytic units were reported per milliliter of culture filtrate per 10⁸ CFU. For neutralization studies, culture filtrates were mixed with either preimmune or hyperimmune rabbit sera (laboratory stock; 1:10 dilution) containing antibodies to Act. After incubation at 37°C for 1 h, the hemolytic and cytotoxic activities were measured.

Proteinase activity.

An aliquot (200 μl) of overnight culture filtrates (in the presence of 0.2% arabinose) from A. hydrophila control and Dam-overproducing strains was added to disposable 6-ml, snap-cap tubes which contained 800 μl of the DPBS and 5 mg of Hide azure powder substrate (Calbiochem, La Jolla, CA). The tubes were incubated in a shaker incubator at 37°C for 1 to 3 h. As the proteinase in the culture filtrates catalyzed the substrate, blue color was released from the substrate and was quantified at OD₅₉₅. The proteinase activity was calculated per ml of culture filtrate per 10⁸ CFU. The substrate incubated with the LB medium alone served as a negative control.

Animal experiments.

Eight-week-old female Swiss-Webster mice were used to determine lethality induced by A. hydrophila SSU control and Dam-overproducing strains. Mice were inoculated intraperitoneally with a lethal dose (3 × 10⁷ CFU, representing 2 LD₅₀s) of both Aeromonas strains in a group of 10 mice each. One group of mice was inoculated with DPBS (n = 10) and served as a control. Mice were observed daily for signs of distress and mortality for up to 3 weeks. The animals were provided with 0.2% arabinose and ampicillin (40 mg/kg/day) in the drinking water for the first 2 to 3 days following challenge to retain the plasmid in bacteria. The treatment with arabinose and ampicillin started 1 day prior to the challenge.

Statistics.

Wherever applicable, at least three independent experiments were performed; the data were analyzed by using Student's t test, and P values of ≤0.05 were considered significant. The animal data were analyzed using Fisher's exact test.

Nucleotide sequence accession number.

The sequence of the A. hydrophila SSU dam gene was deposited in the GenBank database under accession number DQ067435.

RESULTS

Cloning and sequencing of the A. hydrophila SSU dam gene.

To determine whether A. hydrophila strain SSU contains the DNA adenine MTase gene, we initially used restriction endonucleases which are sensitive to methylation of adenine residues in 5′-GATC-3′ sequences to digest gDNA. It is known that the DpnI enzyme cuts the methylated GATC sequence, but not unmethylated GATC. Conversely, DpnII does not cut methylated GATC, while it cuts the unmethylated GATC sequence (67). We noted that gDNA of A. hydrophila was sensitive to DpnI restriction endonuclease digestion and resistant to digestion with DpnII restriction enzyme (data not shown). These findings indicated that A. hydrophila might possess MTase activity with GATC specificity.

As a result, we designed dam1 and dam2 primers (Table 2) that corresponded to the regions of highest conservation among the dam genes from different gram-negative bacteria, such as the Vibrio species (e.g., V. cholerae, V. fischeri, and V. parahaemolyticus), E. coli, S. enterica serovar Typhimurium, and Yersinia species (e.g., Y. pseudotuberculosis and Y. pestis), to PCR amplify a portion of the dam gene (a 400-bp fragment) from the gDNA of A. hydrophila SSU. Upon Southern blot analysis under low-stringency conditions, this fragment reacted with an E. coli dam gene probe. Consequently, this 400-bp fragment was cloned into a pCR 2.1 vector and transformed in TOP10 chemically competent E. coli cells (TA cloning kit; Invitrogen). The DNA sequence analysis of this 400-bp fragment revealed 61% and 63% identities with the corresponding regions of the dam gene of E. coli and V. cholerae, respectively. The coding regions of the dam gene in E. coli and V. cholerae contained 834 and 831 bp, respectively (7, 38).

We used two strategies to obtain a complete sequence of the A. hydrophila dam gene coding region as well as of the flanking DNA sequences to the dam gene. The nucleic acid sequence of the entire dam_AhSSU gene (Fig. 1) was determined by gDNA sequencing using the additional primers dam3 and dam4 (Table 2). The construction of a fosmid library of A. hydrophila SSU gDNA enabled us to confirm the DNA sequence of the dam gene and to obtain its flanking sequences for preparing isogenic mutants. The 400-bp dam gene fragment generated by PCR amplification that reacted with the E. coli dam gene probe was used to screen the fosmid library of A. hydrophila. This probe did not react with the E. coli dam gene under high-stringency conditions and thus prevented identification of false-positive fosmid clones. Five positive fosmid clones that contained inserts of approximately 25 kb were obtained out of 800 clones that were screened. Further, this 400-bp dam gene fragment hybridized specifically with the gDNA digests (cut with various restriction enzymes) of A. hydrophila SSU under high-stringency conditions (data not shown).

BLAST search of the DNA sequences obtained using the above-mentioned strategies at the National Center for Biotechnology Information (NCBI) revealed an 873-bp ORF that had a high degree of identity with previously published bacterial dam sequences (7, 38). The gene encoded a protein of 291 amino acid residues with a molecular mass of 32.7 kDa. The complete DNA sequence of the dam_AhSSU gene with its corresponding amino acid sequence is depicted in Fig. 1. The overall identity of the A. hydrophila SSU dam gene with those of V. cholerae, E. coli, S. enterica serovar Typhimurium, and Y. pseudotuberculosis was 65%, 61%, 59%, and 58%, respectively. At the amino acid level, a maximum homology (68%) of M.AhySSUDam was noted with that of V. fischeri.

Analysis of the dam_AhSSU gene presumptive transcriptional start site and promoter region.

Primer extension analysis was used to determine the presumptive transcriptional start site of the dam gene, and three primers were used. The damP1 primer was designed to nucleotide positions 1 to 30 (nucleotide A of the dam_AhSSU gene start codon [ATG] represented position 1). The damP2 primer represented nucleotide positions 7 to 36, while the damP3 primer spanned nucleotide positions 12 to 41 (Table 2 and Fig. 1). The primers were end labeled and hybridized to the RNA isolated from wild-type (WT) A. hydrophila SSU. The synthesized cDNA was analyzed on a denaturing polyacrylamide gel. The length of the cDNA reflected the number of bases between the labeled primer and the 5′ end of the dam_AhSSU gene transcript. cDNA products with lengths of 54, 60, and 65 bp, were obtained with damP1 (Fig. 2, lane 4), damP2 (Fig. 2, lane 3), and damP3 (Fig. 2, lane 2) primers, respectively. Based on the DNA sequence, primer extension analysis identified a presumptive transcriptional start site (G) at a position 24 nucleotides upstream of the ATG start codon of the dam gene (Fig. 1). Putative −10 (GGGTAGAAT) and −35 (TAGCCA) elements of the promoter were also identified in this region using a software program found at www.softberry.com.

FIG. 2.

Primer extension analysis to define the presumptive transcriptional start site of the A. hydrophila SSU dam gene. The products from primer extension reactions were separated by electrophoresis on an 8% denaturing polyacrylamide gel. Lanes: M, HinfI-digested φX174 DNA markers (Promega); 1, cDNA product of the control RNA (87 bp); 2, cDNA product of A. hydrophila RNA with damP3 primer (65 bp); 3, cDNA product of A. hydrophila RNA with damP2 primer (60 bp); 3, cDNA product of A. hydrophila RNA with damP1 primer (54 bp). The gel was exposed to the X-ray film overnight at −70°C.

Overproduction and purification of M.AhySSUDam.

The dam_AhSSU gene was overexpressed, and M.AhySSUDam as a His tag fusion protein was purified using ProBond resin charged with nickel. Purified Dam was eluted from the column using 1 M imidazole. Based on SDS-polyacrylamide gel electrophoresis analysis and Coomassie blue staining, a single protein band with a molecular mass of 34 kDa was detected. The molecular mass of purified Dam was in agreement with the predicted size of 291 amino acid residues (32,662 Da) based on the DNA sequence plus the additional amino acid residues derived from the His tag region of the pET-30a(+) vector. The purified enzyme was then employed to determine its MTase activity using N⁶-methyladenine-free lambda DNA.

GATC DNA methylation assay.

It is known that the enzyme DpnI digests methylated GATC sequences but does not digest unmethylated GATC (Fig. 3, lane 1). Conversely, DpnII does not cut methylated GATC, while it does cut unmethylated GATC sequences (Fig. 3, lane 2) (67). The N⁶-methyladenine-free lambda DNA was treated with purified M.AhySSUDam enzyme and then digested with DpnI and DpnII restriction endonucleases. Subsequently, the reaction products were separated by electrophoresis on a 1% agarose gel and visualized by ethidium bromide staining. As illustrated in Fig. 3 (lanes 3 and 4), phage lambda DNA (treated with purified M.AhySSUDam) was cut by DpnI but was not cut by DpnII, indicating that Dam was indeed functional in the A. hydrophila SSU strain, as it methylated adenine residues in GATC sequences of N⁶-methyladenine-free lambda DNA.

FIG. 3.

Methylation of N⁶-methyladenine-free lambda DNA by purified M.AhySSUDam alters the sensitivity of lambda DNA to restriction endonucleases DpnI and DpnII. Lanes: M, lambda DNA/HindIII markers (Promega); 1, lambda DNA N⁶-methyladenine free (-) digested with DpnI; 2, lambda DNA N⁶-methyladenine free (-) digested with DpnII; 3, digestion with DpnI of methylated (+) lambda DNA; 4, digestion with DpnII of methylated (+) lambda DNA. In lanes 3 and 4, the lambda DNA was methylated by M.AhySSUDam (3 μg). The samples were run on a 1% agarose gel and stained with ethidium bromide.

The ability of M.AhySSUDam enzyme to transfer methyl groups from [methyl-³H]AdoMet to N⁶-methyladenine-free lambda DNA was also confirmed. As noted from Fig. 4, cell lysate from the A. hydrophila control strain had a basal expression level of MTase activity. However, cell lysate prepared from the Dam-overproducing A. hydrophila strain showed a considerable increase in MTase activity which was dose dependent. Overproduction of M.AhySSUDam from the P_BAD promoter, after induction with arabinose, resulted in an up-to-30-fold increase in Dam activity (Fig. 4), indicating the functional regulation of the dam gene from the pBAD-dam_AhSSU plasmid by the araBAD promoter in the A. hydrophila SSU strain. The levels of MTase activity in uninduced cultures were similar to that of the A. hydrophila control strain.

FIG. 4.

MTase activity associated with the overproduced M.AhySSUDam of A. hydrophila SSU. The cell extracts from the A. hydrophila control strain (with the pBAD vector alone, designated as pBAD) and the Dam-overproducing strain (with pBAD-dam_AhSSU) were prepared, and the ability of M.AhySSUDam to transfer methyl-³H from AdoMet to N⁶-methyladenine-free lambda DNA was measured as described in Materials and Methods. Three independent experiments were performed, and the arithmetic mean ± standard deviation is plotted. An uninduced culture of A. hydrophila SSU with pBAD-dam_AhSSU exhibited a basal level of MTase activity.

The dam gene is essential for viability of A. hydrophila SSU.

Different genetic strategies to obtain the dam knockout mutant of A. hydrophila SSU were futile, indicating that the dam gene is essential for the viability of the bacterium. However, it was possible to inactivate a chromosomal copy of the dam gene when the corresponding gene was first provided on the plasmid to the A. hydrophila strain in trans before the chromosomal copy was deleted or inactivated. Therefore, to successfully generate a chromosomal deletion of the dam gene, a pBAD-dam_AhSSU plasmid was first electroporated into A. hydrophila. For homologous recombination, we opted to use flanking upstream and downstream sequences of the dam gene in the suicide vector pDMS197 (this construct was designated as pDMS197-upkm1/km2down [Table 1]) instead of using the coding region of the dam gene (Fig. 5). This strategy ensured specific targeting of the dam gene on the chromosome rather than on the plasmid pBAD-dam_AhSSU (Fig. 5). To confirm insertion of the Km^r cassettes and the chromosomal deletion of the dam gene in the gDNA of the mutants, we used PCR analyses with three combinations of primers (Table 2 and Fig. 6). As a control, we used gDNA of WT A. hydrophila.

FIG. 6.

Confirmation of the identity of the dam gene mutant of A. hydrophila SSU by PCR analysis, based on agarose gel electrophoresis of PCR products with gDNA of WT and dam mutant strains. Lanes: M, 1-kb DNA ladder (New England BioLabs); 1, WT gDNA with F1/R1 primers (1.8 kb); 2, WT gDNA with F1/dam2 primers (0.75 kb); 3, WT gDNA with R1/KF primers (no product); 4, gDNA from the dam gene knockout mutant with F1/R1 primers (3.0 kb); 5, gDNA from the dam knockout mutant with F1/dam2 primers (no product); 6, gDNA from the dam knockout mutant with R1/KF primers (2.6 kb).

PCR confirmed the chromosomal deletion of the dam gene with a set of primers, F1 and R1 (Fig. 5 and 6), that represented external sequences relative to the flanking up- and downstream sequences of this gene. Specifically, these primers were designed for regions located outside of 144-bp upstream and 369-bp downstream sequences from the dam structural gene that were used to construct the pDMS197-upkm1/km2down plasmid (Table 1). We observed PCR products with gDNA of both the mutant and WT bacteria. The mutant showed the expected size of the product (3 kb) when compared to that of the WT bacterium (1.8 kb) (Fig. 6, lanes 4 and 1). A pair of F1/dam2 primers (the dam2 primer was located within the dam gene coding region) verified the deletion of the dam gene in the mutant. A correctly sized band of 0.75 kb was amplified from the WT gDNA, and no PCR product was amplified from the gDNA of the mutant (Fig. 6, lanes 2 and 5). PCR amplification with R1/KF primers (the KF primer was located in the Km1 gene cassette) confirmed the insertion of Km^r gene cassettes instead of the dam gene in the mutant. No PCR product was amplified from the gDNA of the WT bacterium; however, a band of an expected size of 2.6 kb was amplified from the gDNA of the mutant (Fig. 6, lanes 3 and 6).

To further provide evidence that the dam gene was necessary for the viability of A. hydrophila, we demonstrated that the chromosomal dam gene-deleted mutant was not viable, based on determining colony counts, when plated on LB agar plates with glucose, which shuts off the expression of the dam gene from the pBAD-dam_AhSSU plasmid (data not shown).

Integrity of the cell membrane of the A. hydrophila SSU Dam-overproducing strain.

We determined growth rates of the A. hydrophila control strain and its Dam-overproducing strain in LB medium with 0.2% arabinose. No significant difference in growth curves was noted in these two strains. The presence or absence of arabinose did not affect the growth phenotype. Also, both of these strains equally tolerated bile salts present in MacConkey's agar plates as determined by colony counts. Further, the ability of control and Dam-overproducing strains to bind to HT-29 colonic epithelial cells was not affected, indicating the intactness of the bacterial cell envelope. Both the control and Dam-overproducing strains were resistant to the effect of TX-100 up to a concentration of 2%, and the release of periplasmic RNase I from the Dam-overproducing strain was not statistically different than that of the control A. hydrophila strain. These data indicated that increased MTase activity in the Dam-overproducing strain did not alter cell membrane integrity (data not shown).

Effect of M.AhySSUDam overproduction on A. hydrophila SSU virulence.

It was recently shown that alterations in the level of Dam attenuated the virulence of a number of pathogens, including S. enterica serovar Typhimurium, Y. pseudotuberculosis, H. influenzae (4, 76, 86), and others. Although Dam in E. coli and S. enterica serovar Typhimurium is not essential for bacterial growth, it is required for the viability of such pathogens as V. cholerae, Y. pseudotuberculosis (38), and Y. enterocolitica (25) as we also demonstrated in this study in A. hydrophila.

To determine whether altered Dam production affected the virulence potential of A. hydrophila SSU, we evaluated various biological activities associated with A. hydrophila control and Dam-overproducing strains. Motility is an important pathogenic factor for bacteria to reach the host target tissue, to colonize, and then to cause disease. We noted that overproduction of Dam significantly reduced (58%) the motility of the bacterium (Fig. 7A). To determine if overproduction of Dam would have an effect on the cytotoxicity associated with the T3SS of A. hydrophila, RAW 264.7 murine macrophage cells were infected with either the A. hydrophila control or the Dam-overproducing strain. As shown in Fig. 7B, T3SS-associated cytotoxicity (as measured by LDH release) was reduced significantly (55%; P < 0.0001) at 4.5 h after infection when the dam gene was overexpressed compared to that of the control A. hydrophila strain. The LDH release from macrophages by the control strain was similar to the positive control provided in the kit. No difference in the LDH release was noted between the control and Dam-overproducing strain at 3 h postinfection. However, at 3.5 and 4.0 h after infection, a decrease in the LDH release of 12% (P = 0.0004) and 30% (P = 0.0001) was noted between the control and dam-overproducing strains. Interestingly, when the cultures were grown in the absence of arabinose, no statistically significant difference was noted in LDH release from macrophages when the control strain was compared to that of the Dam-overproducing strain (Fig. 7B).

FIG. 7.

Overproduction of Dam reduces motility and T3SS-associated cytotoxicity of A. hydrophila SSU. A. Motility assay results, with the motility phenotype of the A. hydrophila control strain (with the pBAD vector alone and designated as pBAD) and that of A. hydrophila (with pBAD-dam_AhSSU). The bar graph shows the distances migrated by these two strains. B. T3SS-associated cytotoxicity induced by A. hydrophila control and Dam-overproducing strains in RAW 264.7 macrophages. After infection of macrophages with the bacterial cells (for 3 to 4.5 h) at an MOI of 10, LDH release was measured. Both arabinose-induced and uninduced cultures were used for infection. Data from three wells were averaged, three independent experiments were performed, and data are plotted ± standard deviation. The asterisks denote statistically significant differences (P ≤ 0.05, as determined by Student's t test) between control and Dam-overproducing strains.

In contrast, cytotoxicity associated with the T2SS-associated Act on macrophages showed a 10-fold increase when the culture supernatant of the Dam-overproducing A. hydrophila strain was used compared with those of the A. hydrophila control strain (Fig. 8A). Another major biological activity associated with Act is hemolysis of red blood cells. Rabbit erythrocytes treated with culture supernatant from the Dam-overproducing strain had a threefold higher hemoglobin release than the control strain (Fig. 8B), indicating that overproduction of Dam led to a concomitant increase in both the cytotoxicity and hemolysis associated with Act. To demonstrate that these hemolytic and cytotoxic activities were indeed associated with Act, we neutralized the culture supernatants of the above-mentioned strains with Act-specific antibodies, which resulted in abrogation of these activities.

FIG. 8.

Overproduction of Dam increases activity of T2SS-associated Act and proteinase. A. Act-associated cytotoxicity induced by culture supernatants of A. hydrophila control and Dam-overproducing strains in RAW 264.7 macrophages. After treatment of macrophages with the culture filtrates (for 2 h), the LDH release was measured in the tissue culture supernatants (see Materials and Methods). The results were reported as fold increases in activity in the Dam-overproducing strain compared to the control strain. B. Act-associated hemolytic activity in the culture supernatants of A. hydrophila control and Dam-overproducing strains, as measured by the release of hemoglobin (see Materials and Methods). C. Increased production of Act based on ELISA in the culture supernatants of A. hydrophila Dam-overproducing and control strains when grown in the absence (w/o) and presence (w/) of arabinose. D. Proteinase activity in the culture supernatants of A. hydrophila control and Dam-overproducing strains, as measured by hydrolysis of Hide azure powder. Data from three independent experiments were plotted with standard deviations. The asterisks denote statistically significant differences (P ≤ 0.05, as determined by Student's t test) between control and Dam-overproducing strains.

We confirmed these observations by performing Northern blot analysis and noted an increased transcript of the act gene in the Dam-overproducing strain compared to the control strain (data not shown). Further, based on the ELISA data, increased production of Act was noted when Dam was overproduced in A. hydrophila (containing the pBAD-dam_AhSSU plasmid) compared to the control strain harboring pBAD vector alone in a medium containing arabinose (Fig. 8C). The ELISA values using antibodies to Act increased from 2.4 to 9.7/ml/10⁸ CFU (P = 0.0003) when the Dam-overproducing strain was compared to that of the control strain. On the contrary, in the absence of arabinose, no difference in Act production was noted between the control and the Dam-overproducing strain. Expression of the genes (e.g., aopB, aopD, ascV, and acrV) that constitute the T3SS apparatus was not altered (data not shown).

The pathogenic and virulence characteristics of A. hydrophila are also associated with the production of T2SS-associated exoenzymes (e.g., proteases and lipases) (15, 36). We noted increased production of proteinase (2.4-fold) with the A. hydrophila Dam-overproducing strain compared to its appropriate control in the culture supernatant (Fig. 8D). These results indicated that overproduction of Dam enzyme did, indeed, alter the virulence potential of A. hydrophila based on in vitro assays.

To finally confirm the effect of Dam overproduction on bacterial virulence, we injected mice intraperitoneally with either the A. hydrophila control strain or the Dam-overproducing strain at a lethal dose of 3 × 10⁷ CFU. All of the animals infected with the A. hydrophila control strain died within 2 days. However, animals infected with the Dam-overproducing strain did not die over a tested period of 3 weeks, indicating bacterial attenuation as a result of Dam overproduction. We also noted that the ability of the Dam-overproducing strain to colonize the small intestine of mice remained unaltered compared to that of the control strain. After 2 h of infection, approximately 2 × 10³ to 7 × 10³ CFU of control and Dam-overproducing strains bound the intestinal epithelial cells compared to 1 × 10⁷ CFU that were injected into the ligated ileal loops (data not shown).

DISCUSSION

The role that Dam plays in altering bacterial virulence is only beginning to be understood (45), and studies have shown that adenine methylation can either directly or indirectly alter the interaction of regulatory proteins with DNA (18). Dam can act as a de novo methylase by methylating both nonmethylated and hemimethylated GATC sites (79) and plays a pivotal role in controlling gene expression by the formation of DMPs (80). In this study, we identified, cloned, and sequenced the dam gene from a diarrheal isolate, SSU of A. hydrophila. Further, the role of A. hydrophila Dam in altering bacterial virulence was evaluated using both in vitro and in vivo systems.

Previous studies indicated that Dam methylation played a role in S. enterica serovar Typhimurium cell envelope integrity (62). These investigators showed that Dam mutants enhanced the release of extracellular proteins to the medium, with no obvious alteration in the T3SS-associated secretion of proteins. We therefore examined the effect of Dam on cell membrane integrity of A. hydrophila SSU. We noted that the growth phenotypes of the control and Dam-overproducing A. hydrophila strains as well their abilities to adhere to HT-29 colonic or mouse small intestinal epithelial cells were similar. Likewise, membrane permeability remained unaltered in the Dam-overproducing strain of A. hydrophila compared to the control strain. These data indicated that the virulence defect in the Dam-overproducing strain of A. hydrophila was directly the result of alterations in gene expression and not the pleiotropic effects of Dam on cell physiology (30, 38, 45).

Our data indicated that Dam was functional (Fig. 3 and 4) and essential for the viability of A. hydrophila (Fig. 5 and 6). Similarly, in V. cholerae, Y. pseudotuberculosis, and Y. enterocolitica, deletion of the dam gene was lethal to the corresponding bacterium (25, 38). However, a recent study indicated that the dam gene could be deleted from another strain of Y. pseudotuberculosis without loss in viability (78). The reason for the growth requirement of Dam is not clear, but for V. cholerae, which has two chromosomes, Dam is required for initiation and replication of both (23). Studies with the E. coli dam mutants revealed that Dam MTase was not required for the viability of the bacterium; however, the viability was dependent on the increased expression of the SOS regulon, and more specifically of the genes recA⁺ and ruv⁺ within this regulon (61). It was noted that RecA facilitated cleavage of the LexA repressor, a negative regulator of approximately 20 unlinked operons involved in DNA repair and mutagenesis (84), indicating that methyl-directed mismatch repair played a crucial role in the cell viability of E. coli dam mutants (51, 60). In Y. enterocolitica, Falker et al. (25) speculated that the potential role of Dam lay in the expression of essential genes required for bacterial viability rather than in Dam's role in DNA mismatch repair when compared to the dam gene of E. coli. These results are in contrast to those in E. coli and Salmonella in which, for example, Dam is one of the three dispensable methyltransferases in addition to Dcm (methylates the internal cytosine residues in the sequences CCwGG) and EcoK (modifies adenine residues in the sequences AAC[N⁶]GTGC and GCAC[N⁶]GTT) (44).

Regardless of the role of Dam in bacterial viability, both deletion and overproduction of the dam gene have been shown to attenuate bacterial virulence (4, 31, 39). We explored the role of Dam in the pathogenesis of A. hydrophila and showed decreased motility and cytotoxicity associated with the T3SS of the Dam-overproducing strain (Fig. 7A and B). Motility is an important virulence factor of many gram-negative pathogens and a significant invasion-related factor for bacteria such as S. enterica serovar Typhi (42). Motility was shown to be decreased in a dam mutant of E. coli (58). Similarly, a decrease in motility might have contributed to the lack of invasiveness of S. enterica serovar Typhi dam mutants (42). Since invasiveness of S. enterica serovar Typhimurium to the host cells is also dependent upon the T3SS (26), it was noted that the dam mutants were indeed defective in the secretion of Salmonella pathogenicity island 1-encoded effector proteins, including those proteins essential for the invasion of the bacterium (28).

A possible mechanism by which overproduction of Dam leads to decreased T3SS-associated cytotoxicity in A. hydrophila could be explained by findings observed in Y. pseudotuberculosis. The overproduction of Dam in Y. pseudotuberculosis altered the expression and secretion of a T3SS-associated effector protein YopE (yersinia outer membrane protein E), which is secreted by the WT bacterium under low calcium and high temperature (37°C) conditions and is also known to be antigenic (39). Yops translocated into the host cell via the T3SS act to inhibit phagocytosis of the bacterium and to induce proinflammatory cytokine release (5). The overproduction of Dam in Y. pseudotuberculosis disrupted both the thermal and calcium regulation of YopE synthesis and relaxed the thermal but not the calcium dependence of YopE secretion (39). Currently, the effector proteins secreted by the A. hydrophila T3SS are not known. However, the phenomenon we observed of reduced T3SS cytotoxicity associated with Dam overproduction could be related to the altered secretion and/or synthesis of T3SS effectors. Our Northern blot analysis data indicated no alteration in the expression of T3SS apparatus genes aopB, aopD, ascV, and acrV in Dam-overproducing versus control A. hydrophila strains (data not shown). These data suggested that the T3SS machinery itself remained unaltered in the Dam-overproducing strain.

It is also possible that Dam overproduction in A. hydrophila may play a role in the increased expression of negative regulators of the T3SS or, conversely, the decreased expression of positive T3SS regulatory genes. An example of a T3SS regulator was recently elucidated in Pseudomonas aeruginosa. A specific locus was defined (sadARS) which was comprised of genes for a putative sensor histidine kinase and two response regulators (41). Among the genes regulated by this three-component SadARS system are those required for the T3SS. This report showed that SadS and SadA were important for controlling expression of T3SS genes. SadA contains a helix-turn-helix motif and may regulate T3SS gene expression at the transcription level. In Bordetella pertussis, the BvgA response regulator also contains a helix-turn-helix motif, and under activating conditions (in the Bvg⁺ phase), this protein binds to virulence gene promoters and activates transcription (54). Recent data suggest that the SadARS regulatory system may function to promote biofilm formation, possibly, in part, by repressing expression of the T3SS (41). Similarly, it was reported recently that ExsE is a negative regulator of the T3SS in P. aeruginosa (65) and is secreted via the T3SS under conditions of low calcium. Therefore, it is intriguing to determine whether a homolog of ExsE exists in A. hydrophila and whether overproduction of Dam might prevent release of ExsE via the T3SS and hence reduced expression of the T3SS-secreted effectors.

In addition to T3SS-associated cytotoxicity, the biological activities associated with the T2SS-associated Act, a potent virulence factor of A. hydrophila, were also affected by M.AhySSUDam. Interestingly, overproduction of Dam augmented the virulence potential of Act. Both the cytotoxic and hemolytic activities associated with Act were markedly increased in the culture filtrate of the A. hydrophila Dam-overproducing strain compared to the control strain (Fig. 8A and B), indicating a positive effect on bacterial virulence by Dam. Indeed, act gene expression was increased in the Dam-overproducing strain of A. hydrophila, based on Northern blot analysis (data not shown) and ELISA (Fig. 8C), for the act transcript and Act protein, respectively. A similar pattern was noticed for proteinase production (Fig. 8D), which was upregulated in the Dam-overproducing strain compared to the control strain of A. hydrophila.

Although the mechanism(s) by which Dam overproduction alters gene expression is far from clear, a recent study based on the microarray analysis of different mutants has implicated SeqA protein as playing an important role in the alteration of gene expression in the Dam-overproducing E. coli strain (43). These investigators noted that the absence of SeqA protein (seqA mutant) and high DNA methyltransferase levels (Dam-overproducing strain) affected global gene expression in an almost identical manner. However, a different pattern of gene expression was noted in the dam mutant of E. coli (43). In addition to DNA initiation and replication, SeqA has been shown to exert its function in nucleoid organization through interaction with hemimethylated DNA (43). Similar to other chromosome structure-maintaining proteins, such as H-NS, Fis, IHF, HU, etc. (53), a global regulatory role has been proposed for SeqA (43). It has been speculated that Dam and SeqA compete for binding to hemimethylated DNA behind the replication fork. Either deletion of the seqA gene or overproduction of Dam increases the negative superhelicity of the chromosome (43, 87), thus facilitating open complex formation by RNA polymerase on promoters in general, which leads to redistribution of the RNA polymerases in bacteria, resulting in up- and downregulation of certain genes (37, 43).

Dam overproduction leads to attenuation of V. cholerae in animals (38). We observed a similar pattern of attenuation of A. hydrophila when animals injected with the Dam-overproducing strain did not die at a dose (2 LD₅₀) that killed 100% of the animals infected with the A. hydrophila control strain.

Previous studies also indicated that Dam overproduction in Y. pseudotuberculosis caused the ectopic secretion of LcrV (low calcium response protein V) under conditions that are nonpermissive for synthesis and secretion in the WT strain (i.e., under high-calcium and low-temperature conditions). LcrV is a Yersinia T3SS virulence protein involved in the expression and translocation of Yop proteins, as well as in the suppression of host inflammatory activities via interleukin-10 by activation of toll-like receptor 2 (4, 8, 39, 81). It was demonstrated in Y. pseudotuberculosis that the protection conferred by the Dam-overproducing strain against the WT bacterium is highly dependent on the presence of LcrV (4). Such dependence on LcrV may be due to its role as a principal immunogen and/or its role in the synthesis and localization of Yops, which may also contribute to the immunity observed in Dam-overproducing, Yersinia-vaccinated hosts (4). A recent study indicated that oral immunization of mice with a dam mutant of Y. pseudotuberculosis protected them against infection with Y. pestis (78).

Taken together, the overproduction of Dam in A. hydrophila SSU altered the expression of two key virulence factors of this bacterium, namely T3SS- and T2SS-associated Act, in addition to motility and proteinase production. Overexpression of the dam gene might alter the expression of virulence genes in a positive or a negative way. Perhaps this dual nature of Dam in A. hydrophila may be responsible for causing diseases via aberrant virulence gene expression. Although expression of the act gene was increased in the Dam-overproducing strain compared to the WT A. hydrophila, overall virulence of the bacterium appears to depend upon the interplay between the T3SS- and/or T2SS-associated Act, and possibly other factors. Finally, in our studies, we provided evidence that the Dam-overproducing strain was avirulent in mice compared to the WT bacterium. Our in vivo binding studies also indicated that bacterial attenuation due to Dam overproduction was not related to a defect in colonization.

At present it is not clear whether the decreased virulence of Dam-overproducing A. hydrophila is attributable to the direct increase in MTase activity or whether it occurs indirectly through other A. hydrophila proteins that could have been affected by increased MTase activity. Our future studies will be focused on obtaining the mutated form of Dam with no MTase activity and comparing the effects of overproduction of inactive and active Dam on A. hydrophila virulence. Our studies also will be aimed at delineating whether animals immunized with the Dam-overproducing strain are protected against challenge with the WT A. hydrophila.