Detection of putative virulence genes in Aeromonas isolates from humans and animals

Introduction: Aeromonas are foodand water-borne bacteria that are considered to be zoonotic human pathogens. This study aimed to investigate the presence of genes associated with virulence in human and animal Aeromonas isolates and the potential role of animal isolates with regards to human Aeromonas infections. Methodology: The presence of aerA, hlyA, alt, ast, laf, ascF-G, stx1 and stx2 putative virulence genes in 40 human and animal Aeromonas isolates (16 human and 24 animal isolates) were examined by polymerase chain reaction (PCR). DNA fragments of expected sizes were purified and sequenced. BLAST in the NCBI was used to verify any amplified products. Results: PCR screening showed that hlyA, alt, and laf genes were determined at ratios of 6.25%, 50%, and 6.25%, respectively, in human isolates. The ratios of hlyA, alt, ascF-G, laf, stx2, and stx1 genes in animal isolates were 58.3%, 20.83%, 33.3%, 20.83%, 8.33%, and 4.17%, respectively. Neither aerA nor ast genes were detected in any isolates. Any one of eight putative virulence genes was not detected in seven human and eight animal isolates in the study. Conclusions: The current study is the first to investigate the presence of the virulence gene in gull Aeromonas isolates. The manifestation of the presence of the virulence gene and gene combinations was considerable, especially in fish and gull isolates when compared with clinical human isolates. The current study demonstrates the potential importance of fish and gulls in terms of human Aeromonas infections.


Introduction
Aeromonads are food-and water-borne bacteria.However, these bacteria are considered to be zoonotic human pathogens that can cause severe diarrhea, dysentery, and bacteremia.In human and veterinary medicine, Aeromonas strains are isolated mainly from fecal, wound, and abort samples [1].Detection of aeromonads in animals constitutes a potential risk [2,3].Cumulative data strongly allege that animals are subject to an ever-present reservoir [3].Most of the studies on putative virulence genes of animal Aeromonas isolates are related to fish isolates [4][5][6][7][8][9][10].Gulls are widespread coastal bird species that may contaminate coastal and lake water by their feces, which is a major source of contamination [11].In our previous studies, we determined a clonal relationship by pulsed-field gel electrophoresis between a cattle A. caviae isolate and a human A. caviae isolate [12].The determination of such a relationship reveals a possibility that animals other than fish have potential roles in human Aeromonas infections.The pathogenicity of Aeromonas spp. is multifactorial and complex and may include products of a number of different genes acting individually or collectively.The presence of aerA, hlyA, alt, and ast genes could contribute to diarrhea-related virulence [13].Shiga toxins (STX1 and STX2) coded by stx1 and stx2 genes are important virulence factors in the pathogenesis of gastroenteritis, hemorrhagic colitis, and hemolyticuremic syndrome (HUS) [14], and the ascF genecoded putative type III secretion system also plays important role in pathogenicity [15].The Aeromonas species that possess lateral flagella are generally related to persistent or dysenteric infections [16].It has been reported that the presence of aeromonads that carry the virulence gene in animals is a risk factor for public health [5,17].In the literature review, no study was found that investigated the presence of these genes in gull Aeromonas strains.Most studies focused on Aeromonas strains isolated from fish, but there are studies that investigated the presence of some of these genes in Aeromonas strains isolated from other animal species [4][5][6][7][8][9][10]17,18]; however, among the studies in which human and animal Aeromonas isolates were examined, no study was found that aimed to determine the presence of aerA, hlyA, alt, ast, ascF-G, laf, stx1 and stx2 genes that are targeted in the current study.The current study aimed to investigate the presence of the eight virulence genes in human and animal Aeromonas isolates and their potential importance in terms of human Aeromonas infections.

Aeromonas isolates
In this study, 40 Aeromonas strains isolated previously from humans (14 clinical A. caviae and 2 non-clinical A. sobria isolates) and animals (15 A. caviae, 5 A. sobria, 3 A. veronii and 1 A. media isolates) were used.Human clinical strains were isolated from diarrheal stool specimens.Non-clinical strains were isolated from healthy human stool specimens.Nine clinical human strains, cattle and sheep strains were isolated in 2002; other human strains (2 non-clinical, 5 clinical strains), chicken strain, gull and fish strains were isolated at 2005.Strains were isolated as previously described [12].

Detection of putative virulence genes by PCR
Using a GF-1 Bacterial DNA Extraction Kit (Vivantis Sdn Bhd, Selangor Darul Ehsan, Malaysia), bacterial genomic DNA isolation was conducted from colonies of Aeromonas isolates grown upon incubation on tryptic soy agar for 24 hours at 37°C and reproduced in a thermal cycler device.The following PCR protocol was followed: 5 μL bacterial DNA, 200 μM deoxy-nucleotide triphosphate (dNTP), 1.25 mM magnesium chloride (MgCl 2 ), 1.5 μM primer (forward-reverse for each primer), 5 μL 10X PCR buffer solution (500 mM KCl, 15 mM MgCl 2 , 100 mM Tris-HCl, pH 9.0) and 1.25 unit of Taq DNA Polymerase (New England BioLabs, Ipswich, USA,) were added 0.5mL PCR tubes autoclaved previously and were completed up to 50 μL with a final volume of distilled water.Samples were subject to the following standard cycles to detect the target virulence genes (aerA, hlyA, alt, ast, laf, ascF-G, stx1, and stx2).Different annealing temperatures of each primer are presented in Table 1.First, thermal cycles were applied for 3 minutes at 94°C, and then 30 cycles were applied individually as follows: 30 seconds at 92°C (denaturation), 30 seconds at an annealing temperature as presented in Table 1for each primer, 1 minute at 72°C (extension), and finally for 2 minutes at 72°C.PCR products to be used were stored at +4°C.Each product was loaded into a gel well for purposes of determining the presence of any amplified gene zone and verifying the presence or absence of a target gene in each PCR product.In TBE buffer, electrophoresis (80V, 70 minutes) was implemented in 1% agarose gel.After the gel was painted by ethidium bromide, it was monitored and analyzed under ultraviolet light (Figure 1).For this purpose, gel images obtained in this study were compared to lengths of base pairs given in Table 1by means of a marker.Representative PCR products selected arbitrarily from PCR products belonging to each gene were sequenced after PCR amplification.PCR products were purified for sequencing using the High Pure PCR Product Purification Kit (Roche, Mannheim, Germany), according to the protocol of the manufacturer.PCR products were sequenced on both strands using the virulence gene primers (2 pmol/μL) in an ABI 3130xl genetic analyzer (Applied Biosystems, Foster City, USA).BLAST in the NCBI was used for purpose of verifying the amplified products.

Statistical analysis
The statistical significance of the data was detected by a Chi-square test (χ 2 ) and probability value (p) < 0.05 was considered to be statistically significant.

Results
Of 40 isolates analyzed, 25 (62.5%) were positive for at least one of the putative virulence genes, 9/16 (56.25%) and 16/24 (66.66%) from human and animal samples, respectively.aerA and ast genes were not detected in any of isolates.When the sequence data were compared with the existing sequences present in GenBank (http://www.ncbi.nlm.nih.gov/) using the BLAST program, the sequence identity for hlyA, alt, ascF-G, laf, stx2, and stx1 genes were 90%, 88%, 90%, 94%, 97%, and 97%, respectively.In all isolates, hlyA, alt, ascF-G, laf, stx2 and stx1 genes were detected at ratios of 37.5% (15/40), 32.5% (13/40), 20% (8/40), 15% (6/40), 5% (2/40), and 2.5% (1/40), respectively.The stx2 gene was detected in only two gull isolates.Namely, both stx1 and stx2 genes (stx1 + /stx2 + ) were detected in one gull isolate, and the stx1 gene was detected in the other gull isolate.In other animal and human isolates, no stx1 and stx2 genes were detected.The alt gene was detectedin five animal isolates (5/24; 20.83%) and eight clinical isolates (8/14; 57.14%).The laf gene was detected in five (5/24; 20.83%) animal isolates and one (1/14; 7.14%) clinical isolate.The hlyA gene was detected in  2).The presence of hlyA and ascF-G genes in human isolates was found significantly less frequently than it was in animal isolates (p < 0.05).No statistically significant difference was determined for the other detected genes.Comparing the detection rates of genes in fish and gull isolates to the rates detected in human isolates, hlyA and ascF-G genes were detected considerably more frequently (p < 0.05) in fish isolates than in human ones.While the hlyA gene was detected considerably more frequently in gull isolates than in human isolates, the alt gene in gull isolates was detected considerably less frequently than in human isolates.No statistically significant difference was found between bacterial species and the presence of the virulence gene.Finally, when any findings were analyzed by considering combinations of putative virulence genes, isolates were determined to be divided into 12 genotypes (Table 3).Among significant genotypes in human and animal isolates, alt + /ast + and aerA + /hlyA + genotypes were not detected.Genotypes 1 and 12 were dominant in human isolates, and genotype 12 was dominant in animal isolates.While genotype 2 and 12 were detected in both human and animal isolates, genotype 1 and genotype 3 were found in only human isolates and genotypes 4 through 11 were found in only animal isolates.

Discussion
In our study, 62.5% of all isolates and 57.14% (8/14) of clinical isolates harbored at least one putative virulence gene, while 42.86% (6/14) of clinical isolates did not harbor any of these genes.Ottaviani et al. [19] found that 60.6% of the isolates from food, clinical and environmental isolates harbored at least one putative virulence gene in 142 Aeromonasisolates, and this percentage was 56.2% in clinical isolates in the studies to determine act, ast, alt and aerA genes.The same researchers determined that 14 (43.75%) of 32 clinical samples were negative in terms of all virulence aspects.The percentages obtained in our study matched the percentages specified in the study of Ottaviani et al. [19].
The presence of aerA, hlyA, alt, and ast genes encoding hemolytic, cytotoxic, cytotonic, and enterotoxic activities may contribute to diarrhealrelated virulence [13].Cytotoxins (aerolysin, hemolysin and multifunctional repeat-in-toxin) produced by A. hydrophila, a node-like receptor family, trigger caspase-1 activation by means of pyrin domain containing 3 (NLRP 3 ) inflammasomes in macrophages, and this triggers pyroptosis as a form of proinflammatory necrosis in macrophages [20].Gray et al. [21] found cytotoxins in 10 of 16 cattle isolates.In our study, no virulence genes were detected in two cattle isolates.
Aerolysin is the best-studied hemolysin, but its Aeromonas isolates may produce more hemolytic toxins having any virulence properties [22].HlyA is a hemolysin like β-hemolytic Vibrio cholera [13].In a previous study, Ottaviani et al. [19] found the detection rate of the aerA gene to be 92.7% and 88% in food and clinical Aeromonas isolates, respectively.Baloda et al. [8] reported that they detected an aerolysin-specific PCR product in 8 (66.66%) of 12 fish isolates.Nawaz et al. [10] noted that the detection rate of aerA gene as 96% in fish isolates.Wu et al. [23] detected aerA and hlyA genes, respectively, in 31% and 40% of 116 consecutive clinical Aeromonas isolates.Pablos et al. [13] reported that they detected the aerA and hlyA genes in 25.0% and 28.1% clinical isolates, respectively.Abdullah et al. [18] detected the hlyA gene in 7 of 8 (87.5%) isolates from children with diarrhea and in 21 of 32 (65.63%) chicken carcass isolates.Khajanchi et al. [24] detected the hlyA gene in 15.09% of 53 stool clinical Aeromonas isolates obtained from different regions of the world and from different geographical locations of the United States.Contrary to the findings of previous studies, in our study, we did not detect the aerA gene in human isolates and animal isolates, and we did not detect the hlyA gene in clinical strains.However, we did detect the hlyA gene in a chicken isolate.
Aeromonasstx1 and stx2 gene series are quite similar to most virulent stx gene variants of Shiga toxin-producing Escherichia coli.It has been suggested that a horizontal transfer can be realized among such microorganisms having these genes.Shiga toxins (STX1 and STX2) include significant virulence factors in pathogenesis of gastroenteritis, hemorrhagic colitis, and hemolytic-uremic syndrome (HUS) [14].In previous studies conducted to detect the stx1 and stx2 genes, Pablos et al. [13] did not detect the stx1 and stx2 genes in clinical isolates.Alperi and Figueras [14] reported that they detected the stx1 gene in 19 (23.8%) of 80 human isolates, and that 1 (1.3%) of these was positive for both stx1 and stx2 genes (stx1 + /stx2 + ).Contrary to any findings obtained in Alperi and Figueras's studies [14], in our study, we did not detect stx1 and stx2 genes in human isolates.However, the relevant findings obtained in our study were found to be compatible with the findings obtained in the studies of Pablos et al. [13].
The ascF-G gene is one of the genes that encode the components of the putative type III secretion system (TTSS).Since TTSS facilitates delivery of toxins directly into host cells, it plays a principal role in pathogenicity [15].In our study, no ascF-G gene was found in clinical isolates.This was compatible with the findings from the study of Pablos et al. [13].Contrary to the findings obtained in our study, Chacón et al. [15] detected the ascF-G genein 50%of 84 clinical isolates.Senderovich et al. [25] detected the same gene in 12% of 17 clinical isolates.Silver and Graf [26] detected the same gene in all 20 environmental and human isolates they studied.Wu et al. [23] detected the same gene in 18% of the isolates they studied.
Gavín et al. [4] noted that the presence of lateral flagella and swarmer motility was a pathogenic factor in mesophilic Aeromonas isolates.The same researchers reported that they detected the laf gene in 62% of clinical Aeromonas isolates and in 70% of fish isolates.Kirov et al. [16] detected the laf gene in approximately 50% of clinical, environmental, and reference strains.Aguilera-Arreola et al. [6] detected the same gene in 77.27% (17/22) of clinical isolates; they detected the lafA gene in two fish isolates as well.Pablos et al. [13], Senderovich et al. [25], and Aguilera-Arreola et al. [27] detected this gene in 9.4%, 41%, and 36.7% isolates, respectively.Contrary to previous studies, in our study, the laf gene was not detected in 14 clinical isolates.However, the same gene was detected in 28.57% (2/7) of fish isolates.In our study, the percentage of the same gene detected in fish isolates was found to be lower than that found by Gavín et al. [4] and Aguilera-Arreola et al. [6].
Aeromonads may produce any heat-labile cytotonic enterotoxin (Alt) and heat-stable cytotonic enterotoxins (Ast).These toxins cause extension of Chinese hamster ovary (CHO) cells, a liquid secretion in rat ligated ileal loop, and an increase in prostaglandins and cyclic adenosine monophosphate (AMP) levels of the intestinal mucosa [28].Abdullah et al. [18] detected the alt gene at ratios of 87.5% and 62.5% in isolates of children with diarrhea and chicken carcasses, respectively.Aguilera-Arreola et al. [6] detected the ast and alt genes at ratios of 95.45% (21/22) and 81.82% (18/22) in clinical isolates, respectively; furthermore, they detected the ast gene in two fish isolates, but did not detect the alt gene.Wu et al. [23] detected the presence of the ast and alt genes at ratios of 13% and 44%; Pablos et al. [13] detected them at ratios of 18.8% and 71.9%; Khajanchi et al. [24] detected them at ratios of 13.21% and 90.57%; Yi et al. [5] detected them at ratios of 27.1% and 55.7%; Senderovich et al. [25] detected them at ratios of 6% and 18%; and Aguilera-Arreola et al. [27] detected them at ratios of 28.4% and 34.9%, respectively.Morinaga et al. [29] detected the alt and ast genes in five and three of seven clinical Aeromonas isolates, respectively, from septicemia cases.Ottaviani et al. [19] reported that they detected the alt gene in 76.4% and 55.5% of food and clinical isolates, respectively, and the ast gene in 16.4% and 61.1% of food and clinical isolates, respectively.Albert et al. [30] detected the alt and ast genes at ratios of 16.5% and 15.7%, respectively, in isolates of children with diarrhea.Also, they detected the same two genes at ratios of 33.3% and 25.9%, respectively, in a control group of children.Nawaz et al. [10] did not detect the alt and ast genes.In our study, the alt gene was detected at a ratio of 57.14% in clinical isolates.While this percentage is compatible with the findings obtained in the studies of Yi et al. [5] and Ottaviani et al. [19], it was higher than the findings obtained in the studies of Senderovich et al. [25], Albert et al. [30], Aguilera-Arreola et al. [27] and Wu et al. [23], and lower than the findings obtained in the studies of Abdullah et al. [18], Pablos et al. [13], Khajanchi et al. [24] and Morinaga et al. [29].Contrary to any findings obtained by Nawaz et al. [10] and Aguilera-Arreola et al. [6], in our study, the alt gene was detected at a ratio of 57.14% (4/7) in fish isolates.Contrary to the findings of Aguilera-Arreola et al. [6], in our study, we did not detect the ast gene in fish isolates, in agreement with the findings of Nawaz et al. [10].However, we did not detect the ast gene in any human or animal isolates.
Putative virulence genes may occur in different combinations in Aeromonas.Ottaviani et al. [19] found that 33.7% of all isolates had three toxins, and when sources of the isolates were considered, the ratio was 38.8% in clinical isolates and 34.5% in food isolates.In these isolates, they found percentages of isolates with two toxins to be 22.2% and 41.8% in clinical isolates and food isolates, respectively.Contrary to the studies of Ottaviani et al. [19], in our study, the act gene was not examined, and the ast and aerA genes were not detected; the gene combinations detected by the other researchers were therefore not detected in our study.Senderovich et al. [25] reported that, as virulence genotypes, three isolates (18%) had five different genes; four isolates (24%) had three or four different genes; and two isolates (12%) had two different genes.Contrary to the findings of Senderovich et al. [25], in our study, any isolate having five different genes was not detected.While three or four different genes were detected in seven isolates (17.5%; 7/40), two different genes were detected in five isolates (12.5%; 5/40).In our study, the 12.5% isolates obtained from strains having two different genes was found to be compatible with the findings of Senderovich et al. [25]; however, the percentage obtained by them for the isolates having three or four genes was higher than the percentage obtained in the current study.Hu et al. [9] noted that all fish and water isolates had three or more genes in different combinations.Contrary to Hu et al. [9], in our study, these gene combinations were detected only in seven isolates.Aguilera-Arreola et al. [27] reported that they detected seven different gene combinationsalt alone, ast alone, aer/hem alone, alt+aer/hem, ast+aer/hem, alt/ast, alt+ast+aer/hem -at ratios of 0.9%, 0%, 61.5%, 6.4%, 0.9%, 7.3%, and 20.2%, respectively, by PCR and dot blot.Yi et al. [5] noted that the isolates investigated in their study were divided into seven genotypes based on the enterotoxin genes (act/alt/ast).In our study, 12 different gene combinations were detected.It has been reported that the aerA + /hlyA + genotype provided the best estimation of virulence for A. hydrophila in an animal model [31].However, a study found that, when the distribution of hemolytic genes was compared to extracellular products, three genotypes (aerA + /hlyA + , aerA + /hlyA -, aerA -/hlyA -) were enterotoxic in the suckling-mouse test and had a capacity to express other virulence properties at different temperatures [32].Pablos et al. [13] stated that they detected the aerA + /hlyA + gene combination in two clinical isolates (6.25%; 2/32).Albert et al. [30] suggested that the enterotoxigenic aeromonads having the alt + /ast + genotype could be true diarrheal pathogens.Pablos et al. [13] detected the alt + /ast + genecombination in four clinical isolates (12.5%; 4/32).Albert et al. [30] reported that they detected the alt + /ast + genotype at ratios of 55.7% and 22.2% in isolates of children with diarrhea and control children, respectively.Aguilera-Arreola et al. [6] found the alt + /ast + genotype in 4.55% (1/22) of isolates.Contrary to previous studies, in our study, no gene combinations were detected in aerA + /hlyA + and alt + /ast + genotypes.Abdullah et al. [18] stated they detected the alt + /hlyA + genotype in 75% and 37.5% of isolates from children with diarrhea and from chicken carcasses, respectively.In our study, the alt + /hlyA + genotype was detected in only one gull isolate.
In the current study, the virulence gene was not detected in 43.75% of human isolates and in 33.3% of animal isolates (in all cattle and sheep isolates).The multifactorial and complex pathogenicity of Aeromonas or its involvement in many different gene products, which work alone or in cooperation [13], necessitates detailed studies on this subject.In the present study, the alt gene was found as the main gene at a rate of 57.14% in human diarrhea-originated strains.This finding was consistent with the study of Aguilera-Arreola et al. [6].In the present study, at least one virulence gene was detected in all fish isolates.Nawaz et al. [10] reported that consumption of improperly cooked catfish contaminated with A. veronii, which carries the virulence gene, produces a potential health risk.Thus, the presence of virulence genes in fish is important for public health [5][6][7]10,30,32].Albert et al. [30] reported that the prevalence of the alt gene is similar among environmental isolates and isolates of children with diarrhea; however, the prevalence of the ast gene is significantly higher in environmental isolates when compared with the isolates from children with diarrhea.The present study revealed that the eight genes are not equally distributed in human and animal isolates and that the prevalence of the alt gene is higher in human isolates, whereas the prevalence of hlyA, stx1, stx2, laf, and ascF-G genes is higher in animal isolates.Lévesque et al. [33] reported that the detection of bacteria, including Aeromonas in gulls, would contribute to the microbiological contamination of recreational water and that there is a need for additional studies investigating the potential roles of the birds in the contamination of humans.In the literature review, no study was found related to the detection of virulence genes in gull Aeromonas strains.In the current study, for the first time, alt, hlyA, stx1, stx2, laf, and ascF-G virulence genes were detected at various ratios (9.09%, 72.73%, 9.09%, 18.18%, 27.27%, and 27.27%, respectively) in gull Aeromonas isolates.Furthermore, again in gull isolates, hlyA + /laf + /stx1 + /stx2 + , ascF-G + /hlyA + /stx2 + , ascF-G + /hlyA + /laf + , and alt + /hlyA + virulence gene combinations were found at rates of 9.09%, 9.09%, 18.18%, and 9.09%, respectively.

Conclusions
The current study investigated for the first time the presence of the virulence gene in gull Aeromonas isolates.The virulence gene and gene combinations were found in considerable numbers, especially in fish and gull isolates when compared with clinical human isolates.Fish and gulls are potentially important in human Aeromonas infections.

Table 1 .
Polymerase chain reaction primers used to detect the targeted genes, and applied annealing temperatures and PCR products anticipated for putative virulence genes

Table 2 .
Distribution of eight putative genes by species and isolation sources in Aeromonas