Correspondence: Marisa Manzano, Department of Food Science, University of Udine, via Sondrio 2/A, 33100 Udine, Italy. Tel.: +390 432 558 127; fax: +390 432 558 130; e-mail: email@example.com
Rainbow trout gastroenteritis has been related to the accumulation of segmented filamentous bacteria in the digestive tract of fish, which presents lethargy, reduced appetite and accumulation of mucoid faeces. Some authors associate the comparison of illness with the presence of viable filaments, which produce and release strings of endospores in the lumen of the gut. The segmented filamentous bacteria that could not be cultured in vitro have been related to Clostridium group I, and they have been named Candidatus arthromitus. Despite the various strategies that have been used to detect unculturable microorganisms, molecular methods have facilitated studies on culture-independent microorganisms. Direct DNA extraction from samples and subsequent study of 16S rRNA genes represent a tool for studying unculturable microbial flora. As direct detection of specific microorganisms is possible through the utilization of primers or probes annealing specific DNA sequences, the aim of this work was to design specific primers for the direct detection of C. arthromitus in fish using a nested PCR.
Gram-positive, endospore-forming, segmented filamentous bacteria (SFB) have been observed in the small intestine of many animals (e.g. rats, pigs, insects) and in the intestinal content of trouts (Oncorhynchus mykiss) affected by diarrhoea. Intensive fish-farming systems have been actively developed during recent decades. This intensification has resulted in an increase in the number of pathogens reported from these intensive aquaculture production systems. An enteritic syndrome affecting farmed rainbow trout [rainbow trout gastroenteritis (RTGE)] has been described and related to the accumulation of the SFB in the digestive tract of fish (Goodwin et al., 1991; Klaasen et al., 1993). Fish generally exhibit lethargy and reduced appetite, and accumulations of mucoid faeces.
Some papers report the nonpathogenic nature of these microorganisms, while other reports associate the occurrence of illness (with diarrhoea and malabsorption) with the presence of SFB (Del Pozo et al., 2009). The origin and the role of the SFB have not been elucidated completely (Michel et al., 2002) despite the presence of viable filaments producing and releasing strings of endospores in the lumen of the gut, as they could not be cultured in vitro. These unculturable bacteria, related to Clostridium group I, are named Candidatus arthromitus, as no formal taxonomic criteria are applicable due to the impossibility to obtain an in vitro culture (Murray & Stackebrandt, 1995; Snel et al., 1995; Urdaci et al., 2001). The microbial communities of the intestinal tract of fish include high densities of unculturable bacteria whose identity has not been reported, but lead to differences between viable counts and total microbial counts (Sugita et al., 2005; Shiina et al., 2006). Various strategies have been used to detect unculturable microorganisms. Klaasen et al. (1992) detected these microorganisms using light microscopy. They can be identified using electron or light microscopy on the basis of their morphology and habitat (Urdaci et al., 2001). Molecular methods have facilitated studies on culture-independent microorganisms. Most of them are based on direct DNA extraction from samples and a subsequent study of 16S rRNA genes. FISH (Langendijl et al., 1995), denaturing gradient gel electrophoresis (Muyzer et al., 1993) and DNA clone libraries for the study of microbial communities have been satisfactorily used (Kim et al., 2007). Also, direct detection of specific microorganisms is possible by the utilization of primers or probes annealing specific DNA sequences. The aim of this work was to design primers to directly detect C. arthromitus responsible for RTGE.
Materials and methods
Intestines from 35 asymptomatic and symptomatic brown trout (Salmo trutta fario) were obtained at 30, 60 and 90 days of growth. The fish intestines were examined at the laboratory within 2 h. The intestinal content was removed by squeezing it out. One gram was diluted into the buffer for the DNA extraction using the QIAamp DNA Stool Mini Kit (Qiagen, Hilden, Germany). The DNA obtained was stored at −20 °C before use. The intestines from samples at 90 days were divided into the initial ileum tract (I) and the final ileum tract (F). A drop of the fresh intestinal content aseptically collected from sample showing symptomatic behaviour (90 days) was examined in phase-contrast microscopy using a light microscopy Axiophot (Zeiss, Milan Italy) (× 1000 magnification). Each test was repeated three times.
The 16S rRNA gene sequences of various microbial flora from fish and C. arthromitus were retrieved from GenBank and aligned using the ‘multiple sequence alignment’ by Corpet (1988) to detect regions showing differences in base pair sequences. On the bases of the differences shown by the sequences, two primers (CAF) CA Forward 5′-ATGGAACTGAGATACGGTCCAT-3′ and (CAR) CA Reverse 5′-GGATTGTTCACAACCC-3′ were designed to anneal to the 16S rRNA gene. A second set of primers named NEST-F 5′-GGAAACCCTGATGCAGCA-3′ and NEST-R 5′-ACACAGTTTTATATGCTT-3′ annealing within the amplicon obtained by the primers CAF and CAR were designed to attest and improve first PCR protocol sensitivity. Primers were tested using the amplify 3 program (Bill Engels, University of Wisconsin, 2005).
PCR assays were carried out in a reaction mixture containing 10 mmol L−1 Tris-HCl, 50 mmol L−1 KCl, 200 μmol L−1 of each dNTP, 0.2 μmol L−1 of each primer, 1.5 mmol L−1 MgCl2, 1.25 UI TaqGold Polymerase (Applera Italia, Monza, Italy) and 2 μL of DNA (100 ng μL−1) in a final volume of 50 μL. Thermal cycler conditions consisted of 95 °C denaturation for 5 min, 30 cycles of 95 °C for 1 min, 57 °C for 1 min, 72 °C for 1 min and a final extension at 72 °C for 7 min in a Thermal Cycler (DNA Engine Dyad peltier Thermal Cycler, BioRad, Milan, Italy). To verify the specificity of the CAF and CAR primers for the C. arthromitus DNA sequence, the PCR protocol was tested on the microorganisms listed in Table 1. On the basis of the data reported by Spanggaard et al. (2000), mostly Gram-negative bacteria were chosen for the test. The bacteria were grown overnight at 30 °C on brain–heart infusion agar (Oxoid), and a single colony was picked from the plate and transferred into a 1.5-mL microfuge tube containing 0.3-g glass beads and DNA extraction was performed as described by Manzano et al. (2003). Primers CAF and CAR were used for PCR on the DNA extracted from fish gut. It was not possible to know the detection limit of the DNA concentration needed in order to obtain a positive result in the PCR assay because of the unculturable nature of the microorganism This led us to utilize heterogeneous DNA from fish as a template. For this reason, a nested PCR to attest the specificity of the positive result achieved and to decrease the detection limit of the protocol was prepared.
Table 1. Microorganisms used for the optimization of the PCR protocol
DSMZ, Deutsche Sammlung von Mikroorganismen und Zellkulturen GmbH (Braunschweig, Germany); CECT, Colección Española de Cultivos Tipo; DISTAM, Dipartimento di Scienze e tecnologie alimentari e microbiologiche (Milan, Italy); ATCC, American Type Culture Collection.
As PCR product visualization is difficult when the DNA concentration loaded in the agarose gel is below 20 ng, the new primers NEST-F 5′-GGAAACCCTGATGCAGCA-3′ and NEST-R 5′-ACACAGTTTTATATGCTT-3′ were used in a second PCR step (nested PCR). The PCR assay was performed in a reaction mixture containing 10 mmol L−1 Tris-HCl, 50 mmol L−1 KCl, 200 μmol L−1 of each dNTP, 0.2 μmol L−1 of each primer, 1.5 mmol L−1 MgCl2, 2.5 UI TaqGold Polymerase (Applera Italia) and 2 μL of PCR product in a final volume of 50 μL. The thermal cycler conditions consisted of 95 °C denaturation for 5 min, 30 cycles of 95 °C for 1 min, 45 °C for 1 min, 72 °C for 1 min and a final extension at 72 °C for 7 min in a Thermal Cycler.
To establish the limit of the sensitivity of the nested PCR, an amplicon obtained by the amplification protocol using NEST-F and NEST-R was measured using a spectrophotometer (SmartSpec 3000™, BioRad) and diluted to a final concentration of 0.08 fg μL−1 DNA. These DNA samples were then used as templates for the nested PCR.
A 5-μL aliquot of the PCR products was separated electrophoretically in a 2% agarose gel (Sigma, Milan, Italy) stained with ethidium bromide (0.5 μg mL−1) in 0.5 × TBE buffer (0.045 M Tris-borate; 0.001 M EDTA, pH 8) and compared with a Molecular Weight Marker (Sigma).
Amplicons obtained from 90-day samples (S90) and 60-day samples (S60) were purified using the QIAquick PCR Purification Kit (Qiagen), dried and sent to the MWG sequencing centre (Eurofins MWG GmbH, Martinsried, Germany) for sequencing.
The samples of freshly collected intestinal content of trout at 90 days were positive for the presence of segmented filamentous bacteria (SFB, C. arthromitus) under microscopic examination (Fig. 1). Filaments containing endospores were clearly visible in phase-contrast microscopy under × 1000 magnification. This result allowed us to consider these samples as positive reference samples.
The primer pair CAF–CAR showed specificity for C. arthromitus as no PCR products were obtained when the DNA from microorganisms reported in Table 1 were used, representing indigenous microbial communities of freshwater fish as a template in the PCR assay. The expected PCR products of 515 bp were obtained for the samples at 90 days, as reported in Fig. 2. They indicated the presence of C. arthromitus in the fish intestinal content either in the initial ileum tract or in the final ileum tract. This result confirmed the presence of the microorganism obtained by microscopic examination. No PCR products were obtained for control samples and for the samples at 30 and 60 days. The sensitivity tests results obtained by nested PCR were in agreement with the first PCR protocol, applied using CAF and CAR primers for all the S90 samples showing the presence of C. arthromitus and confirming the positive results obtained before. The expected amplicon of 270 bp shown in Fig. 3 for the 90-day samples was also obtained for 16 out of 18 60-day samples using the nested PCR. The samples were positive for the presence of C. arthromitus, showing the importance of a method able to decrease the detection limit in the presence of heterogeneous DNA as the template. The control samples (SC) and S30 samples were also negative after nested PCR, as summarized in Table 2. The sensitivity tests obtained by nested PCR are reported in Fig. 4. PCR products were obtained when 80 ng μL−1 to 0.08 pg μL−1 DNA were used as the template, whereas no amplicons were produced when 8–0.08 fg μL−1 DNA were used as the template. This can be considered a good result because the medium DNA content of a single prokaryote cell is 5.5–10 fg. The results suggest that the method can detect a single cell of the microorganism tested: C. arthromitus. In fact, nested PCR allowed for the detection of C. arthromitus in asymptomatic trout at 60 days of growth.
Table 2. Nested PCR results for control samples SC, samples at 30 days: S30, 60 days: S60 and at 90 days: S90; negative: absence of expected 270-bp amplicon, positive: presence of expected 270-bp amplicon
Number of samples
Positive 1° PCR
Positive nested PCR
All the DNA sequences obtained from positive samples (60 and 90 days) were sent to the MWG sequencing centre. All the sequences matched with the C. arthromitus sequence retrieved from GenBank (X87244, AY007720) when aligned, demonstrating that the sequence belonged to the C. arthromitus species (data not shown).
The detection and identification of microbial communities of the gastrointestinal tract of freshwater fish have been conducted for many years using culturing techniques, which limited the knowledge of the microbial intestinal content of fish. The unculturable nature of some microorganisms did not allow for their detection using culture methods through isolation procedures. The application of molecular methods allowed an improvement in microbial detection, leading to an increased understanding of the microbial composition of fish intestine. Molecular methods are important tools for the detection of a microorganism considered to be responsible for a form of summer mortality reported since 1995 in France and Spain in farmed rainbow trouts. The enteritic syndrome, RTGE, which affects farmed rainbow trout, occurs with inappetence of fish and is associated with the presence of SFB in the digestive tract. As the presumptive filamentous agents failed to grow on artificial media, the association of SFB with an enteritic syndrome in rainbow trout would represent an original finding (Michel et al., 2002). ‘C. arthromitus’ has been suggested as a possible aetiological agent for RTGE, because they are always observed in trout presenting RTGE clinical signs (Urdaci et al., 2001).
A specific PCR protocol, followed by a nested PCR, improved the sensitivity in the detection of the microorganism when it was present in low numbers, as well and not detectable by classical PCR protocols. In fact, the S90 samples (symptomatic) were positive for the presence of C. arthromitus by microscopic examination and by a classical PCR protocol, as shown in Figs 1 and 2. The S60 samples were negative upon microscopic examination and by the standard PCR protocol with CAF–CAR primers, but were positive by the nested PCR, as reported in Table 2. Candidatus arthromitus in these samples has only been detected by the utilization of a nested PCR, which was able to decrease the detection limit to 0.08 pg μL−1 of DNA, as shown in Fig. 4 (lane 8).
The authors would like to thank the Skretting Aquaculture Research Centre (ARC), which supported this research work.