Development of A Multiplex Pcr for the Identification of Pathogenic Edwardsiella Tarda and Application to Edwardsiellosis Diagnostics
G. Y. Li,Z. L. Mo,J. Li,P. Xiao,B. Hao,Y. H. Guo
DOI: https://doi.org/10.1111/j.1365-2761.2012.01446.x
2012-01-01
Journal of Fish Diseases
Abstract:Edwardsiella tarda is a common fish pathogen that causes the disease edwardsiellosis, which, in fish, is characterized by septicaemia (Thune, Stanley & Cooper 1993). It is the only species in its genus that can cause disease in humans, with the signs of gastroenteritis and extra-intestinal infection (Janda & Abbott 1993; Mohanty & Sahoo 2007). Populations of E. tarda in water, sediment and aquatic invertebrates are considered reservoirs for fish edwardsiellosis (Mohanty & Sahoo 2007). Human edwardsiellosis is often associated with either eating raw seafood or the exposure of open wounds to aquatic environments and infected animals (Golub, Kim & Krol 2010). Strains isolated from fish are subdivided into pathotype and non-pathotype based on differences in biochemical, genetic and serological test results (Park, Wakabayashi & Watanabe 1983; Rao, Yamada & Leung 2003; Tan et al. 2005; Han et al. 2006). Although no link has been established between direct contact with pathogens and human infection, we hypothesize that contaminated aquatic environments and animals could constitute a risk to public health. The best way to diminish or to avoid bacterial infection is to constantly monitor the presence of pathogenic bacteria in the environment and maintain pathogen-free conditions for fish stocks. Current methods of diagnosing edwardsiellosis, which are based on pathological, immunological and biochemical techniques, are time-consuming and labour-intensive. PCR methods are faster and more efficient and have been developed for the identification of E. tarda at a species-specific level and the discrimination of pathogenic strains from the total E. tarda population (Park et al. 1983; Chen & Lai 1998; Sakai et al. 2007; Lan et al. 2008; Chang et al. 2009; Li et al. 2011). However, none of these existing PCR techniques could discriminate potentially pathogenic strains from clinical samples in a single PCR assay. Multiplex polymerase chain reaction (mPCR), however, can target multiple genes in a single PCR and has been proven to be a useful tool for the rapid identification and detection of various types of bacteria in different kinds of samples. The main object of the present work was to develop a specific mPCR system that would allow the user to simultaneously identify pathogenic and non-pathogenic E. tarda strains from bacterial samples. Other goals include adapting the mPCR protocol to edwardsiellosis diagnostics and monitoring the presence of this pathogen in a turbot culture environment. Two genes, rpoS and esaV, and one DNA fragment were selected as targets for designing PCR primers. rpoS encodes a ó S subunit of RNA polymerase that functions as a stress regulator. It is commonly found in proteobacteria. Previous studies have explored the applications of rpoS in the identification of bacteria at the species level (Kim et al. 2008a,b). In this study, rpoS sequences of Edwardsiella sp. and other bacterial species were obtained from GenBank and aligned using Clustal X. The regions conserved in Edwardsiella sp. but which varied in other species were selected to design the PCR primers. The primer set (RpoS-F: 5′-GGATGGCGTTGAGGACGG-3′; RpoS-R: 5′-CCGCTTACGCGATGCCTCA-3′) targeting rpoS was expected to be specific for identifying both E. tarda and Edwardsiella ictaluri. To differentiate both bacteria, a primer set of IVS/IRS (IVS: 5′-TTAAAGTCGAGTTGGCTTAGGG-3′; IRS: 5′-TACGCTTTCCTCAGTGAGTGTC-3′) targeting a specific DNA fragment of E. ictaluri was used (Williams & Lawrence 2010). Based on the above work and on our recent work regarding esaV as a good indicator for pathogenic strains (Li et al. 2011), a primer set (EsaV-F: 5′-GGTCAATAGCTGGCTACACAA-3′; EsaV-R: 5′-GCGCCTCAGCGAGTATGCGAT-3′) targeting esaV was used to differentiate the pathogenic and non-pathogenic E. tarda. For each amplification, primer set 16S-F/16S-R (16S-F: 5′-GAGTTTGMTCCTGGCTCAG-3′; 16S-R: 5′-CTGCTGGCACGGAGTTAG-3′), targeting the 16S rRNA gene, was included as an internal positive control to verify the PCR fidelity. All PCR was performed in 25 μL reaction volumes containing 1× PCR buffer (TransGen Biotech), 2 mM MgCl2, 200 μm each of dNTPs (TransGen), 1 U Taq DNA polymerase (TransGen) and 1 μL of DNA sample (cell lysates, extracted or purified DNA). Primers were added at the concentration of 0.4 μm each of Rpos-F, Rpos-R, EsaV-F and EsaV-R, 0.2 μm each of IVS and IVR and 0.05 μm each of 16S-F and 16S-R. A negative control (no template DNA) was included in each PCR batch. The PCR conditions were 94 °C for 3 min followed by 35 cycles of 94 °C for 45 s, 60.8 °C for 45 s and 68 °C for 1.5 min, and a final extension at 72 °C for 10 min. To test the specificity of the mPCR, a total of 36 E. tarda strains together with 30 strains of Vibrio species and related genera were used (Table 1). Genomic DNA (gDNA) was extracted and purified from each pure bacterial culture using a commercial genome extraction kit (OMEGA) and quantified by spectrophotometer (Nanodrop-2000; Thermo Scientific). Aliquots of 50 ng gDNA from each strain were used in each PCR. As shown in Fig. 1, the primer set targeting rpoS could produce 223-bp amplicons using DNA extracted only from the E. tarda and the E. ictaluri strains, the primer set of IVS/IRS could produce a 2050-bp amplicon only using DNA from E. ictaluri HSN-1 and the primer set targeting esaV could produce 955-bp amplicons only using DNA from the pathogenic E. tarda strains and E. ictaluri HSN-1. These results indicate that established mPCR techniques can specifically identify E. tarda and E. ictaluri, also allowing differentiation of the pathogenic E. tarda from various bacterial species in a single reaction. To adapt mPCR to edwardsiellosis diagnostics in turbot, three groups of 20 fish kept in tanks A, B and C were intramuscularly injected with pathogenic E. tarda LSE40 (105 CFU per fish), non-pathogenic E. tarda AL92448 (105 CFU per fish) and both LSE40 (105 CFU per fish) and AL92448 (105 CFU per fish), respectively. Three fish and two aliquots of sea water from three tanks were sampled simultaneously at 12-h intervals. Direct mPCR was performed using bacterial DNA extracted from the kidney and the sea water. The bacterial DNA in fish kidney was extracted using a commercial animal tissue DNA extraction kit (OMEGA). For bacterial DNA extraction from sea water, 1-L samples were filtered through a 0.22-μm nitrocellulose membrane (Bio-Rad). The filtered membrane was cut into small pieces with sterilized scissors, resuspended in 2 mL TE buffer (10 mm Tris–Cl, 1 mm EDTA, pH 8.0) containing 7.5% chelex 100 (Bio-Rad) and boiled at 100 °C for 10 min to release the DNA. Colony mPCR was performed using DNA extracted from bacterial colonies recovered from the sea water and kidneys by plating the suspensions on TSA (Difco) containing 12.5 mg mL-1 colistin, a selective medium designed for isolation of the E. tarda (Li et al. 2011). Colony DNA was extracted by boiling the cells suspension in 10 μL TE buffer for 5 min. In the direct mPCR assay, the presence of pathogenic E. tarda was only detected in fish kidney and in seawater samples from tanks A and C from 24 to 48 h and from 36 to 48 h post-injection, and non-pathogenic E. tarda was not detected in samples from all three tanks. In the colony mPCR assay, 30 colonies were collected randomly from plates with colony numbers from 30 to 300, and all colonies were collected from plates with colonies <30. Only the pathogenic but not the non-pathogenic E. tarda was detected from colonies isolated from samples of tanks A and C. Bacterial cells were not recovered from samples of tank B (Table 2). The fact that the non-pathogenic E. tarda cells cannot be detected by both direct mPCR and colony mPCR could be due to the disabled growth and multiplication of the bacteria in vivo the turbot. To test the sensitivity of the direct mPCR assay for pathogenic E. tarda, sterilized sea water and healthy kidney tissue were seeded with ten-fold dilutions of overnight culture of E. tarda LSE40 and processed in the same manner as described above. The detection sensitivity was 1.87 × 105 CFU mL−1 for sea water and 1.40 × 105 CFU g−1 for kidney tissue (Fig. 2). This is similar to the detection limits described by other authors (Savan et al. 2004; Lan et al. 2008; Chang et al. 2009). The mPCR protocol was then tested with turbot with natural edwardsiellosis from a fish farm in Tianjin, China. Kidneys, spleens, livers and blood were collected from the moribund fish. Sea water and fish excrement were simultaneously sampled from the culture ponds where moribund fish were found. The excrement was collected with a sterilized medical inhaler and processed for DNA extraction with the tissue DNA extraction kit. Samples were analysed via both direct mPCR and colony mPCR as described above. All samples were positive for the presence of pathogenic E. tarda by direct PCR and colony PCR analysis. Presence of non-pathogenic E. tarda was also detected in sea water and excrement samples by both methods (Table 2). These results demonstrate that the infected fish might shed cells into the culture pond, which may play an important role in the spread of edwardsiellosis among cultured fish (Matsuoka 2004). Thus, the direct mPCR protocol would allow users to diagnose natural turbot edwardsiellosis from fish tissue and to monitor the presence of pathogenic and non-pathogenic E. tarda in the turbot production environment in <5 h from the point of DNA extraction to observation on gel agarose. The multiplex PCR was then applied to detect the presence of E. tarda in a local turbot farm where edwardsiellosis had occurred 5 months before. Tank sea water and fish excrement were sampled and processed for mPCR analysis as described above. Fish skin was also selected for sampling because bacterial pathogens preferentially colonize this site (Muratori et al. 2000). Skin mucus was scraped off with sterilized cotton swabs and subjected to DNA extraction with DNA extraction kits and mPCR analysis. Presences of pathogenic and non-pathogenic E. tarda were not detected from the collected samples by direct mPCR. Then, aliquots of the collected samples were incubated for 24 h in colistin-Triptic Soy Broth (TSB) selective medium. DNA was extracted from the cultures and used in mPCR analysis. The pathogenic E. tarda was detected by mPCR from 4 of 20 skin mucus samples, one of six seawater and two of six fish excrement, and the non-pathogenic E. tarda was also detected in the skin mucus, sea water and fish excrement. Both the pathogenic and non-pathogenic colonies isolated from the enrichment cultures were detected (Table 2), confirming the feasibility of the protocol. Thus, the direct mPCR protocol would allow monitoring of the turbot farm environment and detection of carriers without killing the animals. In this case, PCR would be recommended after pre-enrichment in colistin-TSB for 24 h. However, it must be noted that the selective medium may enrich resistant bacterial species other than the E. tarda, which may affect the accuracy of the mPCR. In conclusion, the multiplex PCR techniques outlined here are a useful tool for the detection and identification of E. tarda from pure bacterial cultures. In addition, the assay could be used in monitoring the presence of potential pathogens to assess edwardsiellosis risk in turbot culture environments. However, much investigation is needed to support the mPCR as a tool for monitoring the presence of pathogenic and non-pathogenic E. tarda in any fish culture environment and for discrimination of those fish pathogens that might constitute a risk to public health. This work was supported by the Programme of Turbot and Flounder Industry Technology System (Grant No. NYCYTX-50), Project of National Natural Science Foundation of China (Grant No. 31072245) and the Knowledge Innovation Programme of the Chinese Academy of Sciences (Grant No. 20091101).