A new real time PCR-based assay for diagnosing

Journal of Microbiological Methods 76 (2009) 75–80

Contents lists available at ScienceDirect

Journal of Microbiological Methods j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / j m i c m e t h

A new real time PCR-based assay for diagnosing Renibacterium salmoninarum in rainbow trout (Oncorhynchus mykiss) and comparison with other techniques Nabil Halaihel ⁎, Daniel Vendrell, Imanol Ruiz-Zarzuela, Ignacio de Blas, José Luis Alonso, Olivia Gironés, Tania Pérez, José Luis Muzquiz Laboratory of Fish Pathology, Faculty of Veterinary Sciences, University of Zaragoza, Zaragoza 50013, Spain

a r t i c l e

i n f o

Article history: Received 3 July 2008 Received in revised form 11 September 2008 Accepted 16 September 2008 Available online 26 September 2008 Keywords: Renibacterium salmoninarum Nested PCR Reverse transcriptase-Quantitative PCR Lux primers Immune-Chromatography strip P57 ELISA

a b s t r a c t Bacterial Kidney Disease of salmonid is caused by a slow-growing gram-positive bacterium, Renibacterium salmoninarum. This bacterium lives both extra-cellular and intra-cellular in the host. Serological and molecular diagnostic methods to detect the bacterium major surface protein antigen p57 have been developed. In the present work, a newly developed quantitative Reverse Transcriptase-PCR (RT-QPCR), using self-quenched fluorescent primer (Lux), a nested PCR (NPCR), a commercial ELISA and recently commercially available Immune-chromatographic strip test(IC-Strip) were compared for their ability to detect BKD in kidney tissue samples obtained from experimentally infected fish. ELISA test resulted to be rapid, simple and indicative for the bacterial load. The IC-Strip test had similar characteristics for bacterial detection. Both tests are a good option for rapid and relatively inexpensive screening studies, despite the one and two log decrease in bacterial detection limits compared to NPCR and RT-QPCR, respectively. The use of Lux primers in the newly developed RT-QPCR revealed to be a cost-effective alternative to other fluorescence-based PCR techniques. The option of generating a melting temperature curve with the real time PCR instrument confirmed the specificity of the PCR product. The RT-QPCR technique had the advantage of detecting low numbers of viable bacterial mRNA which implied a higher capacity of detecting chronically infected animals. For instance, some fish in the group infected by cohabitation had very low bacterial load and were only detected by this technique. © 2008 Elsevier B.V. All rights reserved.

1. Introduction Bacterial Kidney Disease (BKD), a disease of salmonids was first reported in Scotland in 1931 and has since been reported in many countries around the world, including England, France, Germany, Iceland, Italy, Japan, Spain, Turkey, Canada, the United States, Chile and in the Balkan peninsula (Inglis et al., 1993; Magnusson et al., 1994). The causative agent of BKD is a slow-growing Gram-positive bacterium, Renibacterium salmoninarum. This bacterium lives both extracellularly and intracellularly in the salmonid host and has been shown to survive and even to multiply within macrophages (Grayson et al., 2002). The ability of the disease to be transmitted vertically, and the lack of an effective vaccine, makes BKD difficult to control (Chase and Pascho, 1998; Hamel, 2005; Piganelli et al., 1999). The isolation and identification of R. salmoninarum in samples from clinically infected fish is relatively easy. However, the fact that R. salmoninarum is a fastidiously growing organism requiring prolonged incubation at 15 °C to produce colonies renders such tests of little utility in practice (Bandin et al., 1996). Thus, serological and

⁎ Corresponding author. Tel.: +34 9767612013; fax: +34 976761612. E-mail address: [email protected] (N. Halaihel). 0167-7012/$ – see front matter © 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.mimet.2008.09.014

molecular biology methods are usually employed to confirm the presence of the bacterium (Chase et al., 2006; Rhodes et al., 1998). Enzyme-linked immunosorbent assay (ELISA) is widely used for the detection of R. salmoninarum, and is commercially available from several biotech-companies. Most ELISAs in use, are based on a double antibody sandwich method (Bruno, 2004). An advantage of the ELISA test is that it yields a quantitative optical density value (OD) proportional to the amount of R. salmoninarum bacterial antigens, mostly the (major surface protein antigen) p57(Chase and Pascho, 1998; Meyers et al., 1999; Wiens et al., 1999). ELISA test limitations are related to the lethal sampling (kidney tissue) and the variability between test lots. Non-lethal sampling requirements and research into the vertical transmission of BKD led to the development of Polymerase Chain Reaction (PCR)-based methods. There are currently many PCR assays developed for detection of the p57 gene (Chase and Pascho, 1998; McIntosh et al., 1996; Miriam et al., 1997) or the detection by nested reverse transcription-PCR of 16S rRNA sequences (Konigsson et al., 2005; Magnusson et al., 1994; Rhodes et al., 1998). In the present study a real time RT-PCR assay using self-quenched primers labeled with a single fluorophore for the detection of R. salmoninarum was developed. Our main objective was to establish a protocol for routine BKD diagnosis in our pathology laboratory. Therefore we compared newly developed quantitative RT-QPCR

76

N. Halaihel et al. / Journal of Microbiological Methods 76 (2009) 75–80

2. Materials and methods

negative control) / (Mean of positive control − Mean of negative control). A standard curve from known bacterial concentrations was run within each ELISA plate. Bacterial presence and concentration in each sample were determined in comparison to the bacterial standard curve. For the test to be valid, the difference between mean OD value of the positive and negative control must be greater than 0.1.

2.1. Bacterial strain

2.4. Immunogold-chromatography strip test (IC-Strip)

The R. salmoninarum strain (Rs02-87) used throughout the study was obtained from the Spanish Agriculture Ministry reference laboratory in Madrid. The bacteria were cultured on selective kidney disease medium SKDM agar (Austin et al., 1983) for 3 weeks at 15 °C. Several colonies were diluted in PBS and adjusted to a concentration of 109 particles using McFarland turbidity test, confirmed by optical density (OD) measurement: an OD value of one corresponded to approximately 109 particles of BKD per ml (Miriam et al., 1997) as determined by a spectrophotometer set to 660 nm wave length. After serial log10 dilutions, two different concentrations,106 and 108 cells ml− 1 were used for fish inoculation. For further confirmation, hemocytometer counts of bacteria were done after suspension in 0.05% [vol/vol] Triton X-100/PBS.

This is a new commercially available quick test, based on a singlestep immunochromatographic vertical-flow assay from Certest, Spain. Monoclonal antibodies directed against the main bacterial membrane protein p57, are attached to colloidal gold particles. The test strip is dipped into the supernatant of a kidney homogenate sample. Solubilized p57 from tissue containing R. salmoninarum would form a complex with colored colloidal gold particle. This complex would accumulate in a specific collection zone where a red band appears. Five hundred micro-liters from the same homogenate used for ELISA were placed in an eppendorf and used for this BKD immunechromatographic test. Results were obtained after 10 min by the presence or absence of a visible red band. Another red band on the strip appears to indicate the validity of the assay. Consequently, a two red band strip means the sample is positive, where as only one band indicates a negative sample and that the test is valid. In order to compare band intensities from IC-strip test and correlate them with the sample's concentration (ie: cultured bacteria dilution curve), we acquired an image of the strips with a G-Box instrument from Syngene Co., U.K. Digitalized images were considered as if they were stained gel bands. The software gave us a raw intensity value for each band. Absolute intensity values were obtained by subtracting surrounding background value. Once the Values of the intensities were plotted against the dilutions logarithm, a correlation R2 value was calculated by linear data fitting.

versus already available nested PCR (NPCR), a commercial ELISA (Biotech CV, Gouda, Holland) and a recently available product for the detection of P57 antigen by an Immunogold Chromatographic strip test (IC-Strip) based on a monoclonal antibody (Certest S.A., Zaragoza, Spain).

2.2. Experimental design and animals A total of 104 rainbow trout fish (Oncorhynchus mykiss) (45–50 g) were used. They were obtained from a rearing farm for rainbow trout, using imported certified BKD free eggs from UK. One week before animal collection, we tested the facility for BKD presence in kidney samples from 50 fish randomly chosen. Samples were negative by nested PCR for R. salmoninarum, using P3/M21 and P4/M38 primers as described (Chase and Pascho, 1998). After 10 days of adaptation to the tanks in our animal facility, the fish were divided into four groups with four different treatments. A control group, designated as Group I, had the 24 fish injected intraperotineally (ip) with 0.2 ml of PBS as sham solution. Twenty four other fish inoculated with 0.2 ml PBS containing 106 particles of R. salmoninarum formed Group II, whereas fish in Group III received 108 particles in 0.2 ml inoculum. For the group IV, six fish inoculated with 108 particles were housed with another 24 fish injected with sham solution of PBS (0.2 ml). The fish used for cohabitation infection were marked by trimming their dorsal fins and were not included in the study. Animals were given commercial food twice daily. Each group was kept in a 150 l covered tank with a constant temperature of 15 °C. Water was recycled for each tank through canister filter that hung on the tank and had their own internal pump with a sealed container to trap waste materials from the water. Fifteen days post-infection, animals were sacrificed by immersion in 0.06% of 3-Aminobenzoic Acid Ethyl Ester (methanesulfonate salt, also known as MS-222, Sigma A-5040). None of them presented any symptoms of BKD disease, and no significant body weight differences were found between groups. Kidney, intestine, and liver samples were collected and frozen at −20 °C until processing. In addition to whole tissue sampling, two other kidney samples for RNA in RNAlater buffer (Sigma cat.#R0901-100) and for protein in RIPA buffer supplemented with complete protease inhibitor cocktail tablets and protease inhibitors from (Roche Diagnostics cat. #04693124001) were prepared.

2.5. RNA isolation and reverse transcriptase reaction (RT) Total RNA was extracted from tissue homogenate in Later RNAlater (sigma # 0901). Biotools speed RNA extraction kit was used and the sample's concentration was determined by the QubitR instrument from Invitrogen. Aliquots of 15 μl were treated with 10 U of RNase-free DNase (Promega Corp., Madison, Wis.) in a total reaction volume of 50 μl, containing 40 mM Tris–HCl (pH 8.0), 3 mM MgCl2, 10 U of RNasin (Promega Corp.), and 1 mM dithiothreitol incubated at 37 °C for 30 min. One microliter of 500 mM EDTA was added, and the reaction mixture was centrifuged at 12 000 g for 1 min, placed at 90 °C for 5 min to inactivate DNase, and placed on ice. The RT reaction volume of 20 μl contained 6 μl of the DNase-treated RNA extract, 1 μl of Reverse primer M21 5′-GCAACAGGTTTATTTGCCGGG-3′(Chase and Pascho, 1998), 1 μl of dNTPs mixture and 1 μl of RNasin (100 U). The reaction mixture was allowed to reach 56 °C for 5 min (MJ mini programmable thermal controller from Bio-Rad lab. Inc.), before the addition of 10 μl of a mixture containing 1 μl Reverse Transcriptase Enzyme (Invitrogen AMV kit 12328), 4 μl of AMV buffer and 5 μl of PCR quality water. The Reverse transcription mixture was kept at 42 °C for 45 min then heated to 94 °C for 3 min then hold at 4 °C and stored at −20 °C until further processing.

2.3. Sample preparation and ELISA Kit 2.6. Quantitative polymerase chain reaction (QPCR): primers and settings Kidney sample from each fish was diluted 1:2 (w/v) with PBS-T20 (0.01 M phosphate-buffered saline, pH 7.4, and containing 0.05% Tween 20). The sample was then homogenized, and heated to 94 °C for 5 min. Samples were stored at −20 °C until examined by ELISA following the manufacturer instructions. ELISA Kit (FK101) from bioCheck C.V, Holland, includes its own positive and negative standards. A sample is considered positive, when its SP value is greater than 0.15. The SP value being equal to: (Mean of Test sample − Mean of

We developed a Real Time PCR reaction with self-quenching fluorescent primers from Invitrogen. LUXR fluorogenic primers were designed from DNA sequence for msa1 gene (GenBank accession No. AF123890), to bind nucleotide sequence inside the same region of nested PCR (Chase and Pascho, 1998). This would enable us to compare the assay with the Nested PCR already in use in our laboratory. Light Upon eXtension (LUX) primers (Invitrogen) had a fluorophore


attached near the 3′ end and a short complementary sequence at the 5′ end that creates a hairpin structure. This structure effectively quenches the fluorophore. When the primer is incorporated into a double stranded PCR product and extended by DNA polymerase, the fluorophore is dequenched, resulting in a significant increase in fluorescent signal (Lowe et al., 2003). Custom designed Lux primers were obtained from Invitrogen web site: http://escience.invitrogen. com/lux/. Forward Primer containing JOE fluorophore sequence: 5′cggttGACCACCATTTACCAGTAAC(JOE) was ordered along with the reverse non-labeled 5′-GTTCCGCAACAGCAACTGACA-3′ primer. The primers pair would amplify a product 105 base-pairs (bp). The QPCR was performed on a 5 μl of the RT reaction product. For the quantitative PCR reaction, 25 μl PCR mix contained 12.5 μl Platinum Quantitative PCR SuperMix-UDG: (uracil DNA glycosylase) [30 U Platinum Taq DNA polymerase ml− 1, 40 mM Tris/HCl (pH 8.4), 100 mM KCl, 6 mM MgCl2, 400 mM each dGTP, dATP and dCTP, 800 mM dUTP, 40 U UDG ml− 1 and stabilizers], 2 μl JOE-labeled LUX primer and corresponding unlabelled primer (200 nM each final concentration), 0.5 μl ROX reference dye (Invitrogen) 5 μl of the cDNA template from the RT reaction and 5 μl of PCR quality water. An ABI PRISM 7000 was programmed as follows: 52 °C for 5 min followed by 45 cycles of 94 °C for 15 s, and 60 °C for 1 min. A temperature dissociation curve determination was set from 60 to 94 °C. Data were analyzed using ABI PRISM 7000 sequence detection software, version 1.2.3 (Applied Biosystems). The cycle threshold (Ct) was calculated as the cycle number at which the fluorescence intensity became exponential. The Ct of each sample was then compared with a standard curve from known weight of purified DNA-PCR product. The results were expressed as absolute amounts of Bacteria. Additional information on the QPCR product was obtained from the temperature dissociation curve, where the amplified product melting point can be determined directly through specifically setting this option of the AB Prism 7000 instrument. 2.7. A PCR product as a standard for QPCR In order to standardize the real time results, a known concentration of PCR product using P3/M21 primer set and DNA extracted from cultured bacterial sample was prepared. 200 μl of PCR product were gel purified by a kit from Invitrogen (Cat #K2100-12), and the final concentration was adjusted to 50 ng per ml by comparison to mass marker from Invitrogen (Cat # 10068-013). DNA determination was confirmed by the Qubit fluorometer from Invitrogen, following the manufacturer's instructions for DNA concentration determination.

77

infected. Correlation curves were plotted after logarithmic transformation, and the correlation coefficient was also calculated. Student's t test for paired samples was applied in order to detect significant differences between bacterial load in different groups as well as Pearson's correlation coefficient (R). Type I error lower than 0.05 was considered significant. Statistical analysis and graphics were performed with Microsoft Excel 2003 and SPSS 15.0 for Windows. 3. Results and discussion 3.1. BKD detection using the different techniques Kidney homogenate, DNA and RNA from each animal (n = 96) were processed by ELISA, IC-Strip, NPCR and RT-QPCR. Control group samples were negative in all four assays. Similarly, samples from Group II and Group III were positive in all four techniques used. The only discrepancies were observed in the Group IV, where seven, five and three samples were negative by ELISA, IC-Strip, and NPCR, respectively. All samples were positive when tested by RT-QPCR. The data from R. salmoninarum detection by the different tests for Group IV permitted the calculation of the following sensibilities: 83, 86, 90 and 100% for ELISA, IC-Strips, NPCR, and RT-QPCR, respectively (Table 1). 3.2. Detection and quantification of R. salmoninarum by ELISA and ICstrip tests Based on dilution assays starting at a 106 bacteria ml− 1, the ELISA assay had positive SP value with 500 and 1000 particles of cultured bacteria and positive tissue samples, respectively. The SP values obtained for the serially diluted cultured bacteria samples were highly correlated as well as the serially diluted kidney samples from infected fish (R2 = 0.969) (Fig. 1). IC-Strip test, after image processing and band intensity scoring, (see Material and methods), had similar correlation result from cultured bacteria samples R2 = 0.959, but not from Kidney samples R2 = 0.580 (data not shown). This latest difference in correlation is probably due to the heterogeneity of the background on the strips from kidneys homogenates. Therefore, we considered the IC-Strip test as only qualitative. ELISA test showed a correlated SP value for five fold sample dilutions from both, cultured bacteria and positive kidney samples (fish group III) (Fig. 1). However, IC-Strip test had one fold increase in detecting a serially diluted cultured bacterial sample homogenates compared to ELISA test. A clear positive band was observed at 1/250 dilution (Fig. 2). The ELISA SP value of the same concentration reached a value of 0.12 which renders a negative result.

2.8. Agarose gel electrophoresis for visualization of NPCR final product Prior to gel casting, dried agarose is dissolved in TBE buffer (Invitrogen #15581) by heating and the warm gel solution then is poured into a mold (Bio-Rad), which is fitted with a well-forming comb. The percentage of agarose in the gel was 1.7%. Ethidium bromide was included in the gel matrix to enable fluorescent visualization of the DNA fragments under UV light. Agarose gels are submerged in electrophoresis buffer (TBE) in a horizontal electrophoresis apparatus. The DNA samples are mixed with gel tracking dye and loaded into the sample wells. Electrophoresis was run at 100 mA for an hour at room temperature. Size markers of a 100 bp from Biotools, S.A., Spain, were run in separate lanes. After electrophoresis, the gel was placed on G-Box instrument from Syngene Co., U.K and a picture of the fluorescent gel was acquired and digitized bands were processed by the 2.7 image processing program from Syngene Co. UK., for product size determination. 2.9. Statistical analysis The sensitivity for each diagnostic test was calculated as percentage of positive results in infected groups, assuming that all animals were

3.3. Comparison between NPCR and RT-QPCR for R. salmoninarum detection capability From cultured R. salmoninarum, 106 particles ml− 1 samples with serial dilutions: 1/50, 1/125, 1/625, 1/3125 and 1/15625 were prepared in triplicate. After DNA and total RNA extraction, dilutions of 1/3 125 and 1/ 15625 were positive for NPCR and RT-QPCR, respectively. These Table 1 Sensitivity of Renibacterium salmoninarum detection by ELISA, IC-strips, NPCR and RT-QPCR

Group I Group II Group III Group IV Sensitivitya

n

ELISA

IC-Strip

Nested PCR

RT-QPCR

24 24 24 24

0/24 24/24 24/24 17/24 83%

0/24 24/24 24/24 19/24 86%

0/24 24/24 24/24 21/24 90%

0/24 24/24 24/24 24/24 100%

Group I had twenty four negative control fish. Group II and group III had twenty four fish each, intraperotineally inoculated with 106 and 108 bacteria of R. salmoninarum, respectively. Group IV had twenty four fish cohabitating with additional six fish intraperotineally inoculated 108 bacteria of R. salmoninarum. For details see Material and methods. a Only calculated for bacterial detection with Group IV.

78


Fig. 1. Linearity and detection limits of ELISA test. Bacterial numbers are plotted against the SP value. Correlation value was obtained from linear fitting of the log transformed data.

results would roughly point to a detection limit of about 160 for NPCR and 32 particles for RT-QPCR. In the literature, as low as 10 particles are being mentioned as detectable by real time PCR (Chase et al., 2006). This discrepancy might be related to the fact that at least 2 msa genes are present in the R. Salmoninarum genome (Coady et al., 2006), which probably made the number of bacteria detected by RT-QPCR to be twice the real number. Subsequently, one log difference is observed between the two techniques (16 versus 160 particles). In fact, NPCR technique failed to detect three samples from Group IV. RT-QPCR not only resulted in 100% positive detection but also gave an exact number of bacteria in each sample (Table 1 and Fig. 4).

Fig. 3. Product specificity of Lux primer PCR dissociation curve. Derivative of the fluorescence peaks at the same melting temperature, indicating a similar PCR product without any need for further confirmation (i.e.: sequencing).

3.4. Specificity of the nested and RT-QPCR assays We practically assumed that a NPCR product size of 320 bp was indicative of bacterial DNA presence in a given sample. Sequencing the final PCR product would definitely improve the specificity of the assay but would also increase its cost considerably. The RT-QPCR developed for this study, had the advantage of providing data over the fluorescence derivative curve of the final PCR product. The melting temperature is specific for each PCR product. There would be no need for further determination of the specificity of the result (i.e.: sequencing). As observed in (Fig. 3), only one product is being obtained from the dissociation curve obtained by the (Real Time Instrument AB Prizm 7000). 3.5. Bacterial load as determined by RT-QPCR versus ELISA tests

Fig. 4. Correlation curve between RT-QPCR dilutions from 500 µl RNA tissue of an ELISA positive kidney tissue. 1/50, 1/125, 1/625, 1/3125, 1/15625. Linear correlation was quite high R2 = 0.996. NB: see Material and methods for Ct product definition. RT product is a sample RNA reversely transcribed. Five micro-liters of the transcription reaction is used for second step Real Time PCR detection with Lux primers (see text for more details).

Five hundred μl of a 106 bacteria ml− 1 used previously for comparing ELISA and NPCR were used also for RT-PCR in order to compare their capabilities in detecting the bacterium. RT-PCR technique was able to detect (1/15 625) dilution with a CT value of 37 and Correlation coefficient value R2 = 0.996 was obtained after fitting the RT-QPCR data (Fig. 4). ELISA test, as mentioned earlier failed to detect lower dilution than 1/125. Bacterial load from both ELISA and RT-QPCR form kidney samples belonging to Group III showed an

average of 1 × 106 particles, compared to 3.5 × 104 from Group II and 3.4 × 103 particles from Group IV (Table 2). Statistically significant differences were not observed between the two assays. The bacterial quantification from Group II as determined by ELISA and RT-QPCR, showed the best correlation between paired sampled. This is reflected by Pearson's correlation coefficient (R = 0.756) with a p b 0.001, meaning

Fig. 2. Linearity and detection limits for IC-Strip test. 1, 1/5, 1/25, 1/125, 1/250, and 1/625 serial dilutions samples. Band intensity adjusted score after image digitalization from cultured bacteria correlated with bacterial concentration (R2 = 0.9595) (see text for detail).

N. Halaihel et al. / Journal of Microbiological Methods 76 (2009) 75–80 Table 2 Statistical comparison of the mean for dependent samples in each experimentally infected groups: Mean R. salmoninarum cell counts as determined by ELISA and RT-QPCR assays) Kidney sample

n

ELISA bacteria ml− 1

n

RT-QPCR bacteria ml− 1

p-valuea Pearson's R p-valueb

Group I Group II Group III Group IV

24 24 24 17

0 33685±14079 727390±184079 3288 ± 1714

24 24 24 24

0 37393±11335 1192764±1857281 3517 ± 1423

N/A 0.062 0.217 0.631

N/A 0.756 0.375 0.090

N/A b0.001 0.071 0.730

N/A: not applicable. See Material and methods or (Table 1) for Group description. Significance for Student's t test for dependent samples. b Significance for Pearson's correlation coefficient (R). a

79

or IC-Strip (Cook and Lynch, 1999). Meaning that, RT-QPCR could be applied non-lethally and can determine whether R. salmoninarum is transcriptionally active (Powell et al., 2005). However, it does require qualified personnel and higher inputs in laboratory equipment and expenditure. For the epidemiological surveillance that we are to conduct in several geographical locations in Spain, we are going to apply the NPCR technique as standard, ELISA or IC-strip test on positive samples to discard false positives. The new RT-PCR technique would be used to ensure a population is BKD pathogen free or to discard a BKD carrier status of any given female brooder. Acknowledgements

that at this range of bacterial load—around 3.5 × 104 ml− 1 homogenate of kidney tissue—both tests would give similar results (Table 2). We have to keep in mind that RT-QPCR would detect a two-digit bacteria per milliliter of sample, compared to a 1000 particles needed for ELISA assay. In fact, seven samples contained some 892, 116, 54, 97, 252, 454, and 19 particles as determined by RT-QPCR. These samples were not detected by ELISA. However, no statistically significant differences were found between the two tests(Table 2). The apparent differences observed in samples from Group III, (RT-QPCR = 1,119,764 versus ELISA = 727,390), are possibly due to the fact that the linearity of the ELISA test is on its upper limit in this group, with an OD average of 5.5 while the positive sample standard from the kit reached 4.5. 3.6. Overall comparison of the different diagnosis tests The newly tested Immune-chromatographic strip and ELISA tests showed similar sensitivity range. Both tests are a good option for rapid and relatively inexpensive screening. IC-Strip is very simple and fast (15 min at most). Even though, the red colored band intensity reflects the antigen concentration, quantitative determinations are not accurate. ELISA and IC-strip tests used in the study for p57 antigen detection had a 3-fold decrease in bacterial number detection compared to the NPCR method. RT-QPCR designed for the present study had one log increase in sensitivity over NPCR, detecting around 20 bacteria. Similar results are reported by others (Powell et al., 2005, Suzuki and Sakai, 2007). The use of Lux primers in the RT-QPCR revealed to be as a costeffective alternative to other fluorescence-based PCR techniques such as TaqMan probes, with the advantage of determining the PCR product melting temperature, that would confirm the specificity of the test (Nazarenko et al., 2002). The RT-QPCR used in this study, is significantly more expensive than the other assays, but it would detect specifically a low numbers of viable bacterial which implies a higher capacity of detecting chronically infected animals. 4. Conclusion Depending on the purpose and information needed from any diagnostic test, one or another assay used in this study might be used. For instance, in a disease prevalence study, an ELISA or the rapid ICstrip tests are equally useful. In this kind of study, NPCR would give more accurate estimate on the disease prevalence rate. The RT-QPCR would result expensive, but should be used in case higher degree of certainty is needed (i.e.: to declare fish population(s) as R. salmoninarum free. From our results, 15 days of cohabitation with infected fish, NPCR is not sensitive enough to detect low bacterial load in early infections. For information regarding the presence and load of the bacterium from kidney samples in a population, both ELISA and RTQPCR could be used. The newly developed RT-QPCR would be the “Gold Standard” technique for R. salmoninarum diagnosis in chronically infected or carrier female brood stock from ovarian liquid samples, since ovarian liquid cannot be screened effectively by ELISA

The authors would like to thank Dr. Jose luis Balcazar for his helpful suggestions, to Dr. Carlos Genzor and Dr. Barend van Dam for providing the IC-strip and ELISA Kits, respectively. Thanks to Ma del Carmen Uriel for her valuable technical assistance. Nabil H. K. has a Ramon y Cajal co-financed contract from the Spanish Minister of Education and Science and the University of Zaragoza. This work was supported by a grant from the Spanish ministry of agriculture (Project supra 07-08). References Austin, B., Embly, T.M., Goodfellow, M., 1983. Selective isolation of Renibacterium salmoninarum. FEMS Microbiol. Lett. 17, 111–114. Bandin, I., Santos, Y., Barja, J.I., Toranzo, A.E., 1996. Growth of the fish pathogen Renibacterium salmoninarum on different media. Microbiologia 12 (3), 439–442. Bruno, D.W., 2004. Prevalence and diagnosis of bacterial kidney disease (BKD) in Scotland between 1990 and 2002. Dis. Aquat. Organ. 59 (2), 125–130. Coady, A.M., Murray, A.L., Elliott, D.G., Rhodes, L.D., 2006. Both msa genes in Renibacterium salmoninarum are needed for full virulence in bacterial kidney disease. Appl. Environ. Microbiol. 72 (4), 2672–2678. Cook, M., Lynch, W.H., 1999. A sensitive nested reverse transcriptase PCR assay to detect viable cells of the fish pathogen Renibacterium salmoninarum in Atlantic salmon (Salmo salar L.). Appl. Environ. Microbiol. 65 (7), 3042–3047. Chase, D.M., Pascho, R.J., 1998. Development of a nested polymerase chain reaction for amplification of a sequence of the p57 gene of Renibacterium salmoninarum that provides a highly sensitive method for detection of the bacterium in salmonid kidney. Dis. Aquat. Organ. 34 (3), 223–229. Chase, D.M., Elliott, D.G., Pascho, R.J., 2006. Detection and quantification of Renibacterium salmoninarum DNA in salmonid tissues by real-time quantitative polymerase chain reaction analysis. J. Vet. Diagn. Invest. 18 (4), 375–380. Grayson, T.H., Cooper, L.F., Wrathmell, A.B., Roper, J., Evenden, A.J., Gilpin, M.L., 2002. Host responses to Renibacterium salmoninarum and specific components of the pathogen reveal the mechanisms of immune suppression and activation. Immunology 106 (2), 273–283. Hamel, O.S., 2005. Immunosuppression in progeny of chinook salmon infected with Renibacterium salmoninarum: re-analysis of a brood stock segregation experiment. Dis. Aquat. Organ 65 (1), 29–41. Inglis, V., Roberts, R.J., Bromage, N.R., 1993. Bacterial diseases of fish. Blackwell, Oxford (United Kingdom), pp. 177–195. Konigsson, M.H., Ballagi, A., Jansson, E., Johansson, K.E., 2005. Detection of Renibacterium salmoninarum in tissue samples by sequence capture and fluorescent PCR based on the 16S rRNA gene. Vet. Microbiol. 105 (3–4), 235–243. Lowe, B., Avila, H.A., Bloom, F.R., Gleeson, M., Kusser, W., 2003. Quantitation of gene expression in neural precursors by reverse-transcription polymerase chain reaction using self-quenched, fluorogenic primers. Anal. Biochem. 315 (1), 95–105. Magnusson, H.B., Fridjonsson, O.H., Andresson, O.S., Benediktsdottir, E., Gudmundsdottir, S., Andresdottir, V., 1994. Renibacterium salmoninarum, the causative agent of bacterial kidney disease in salmonid fish, detected by nested reverse transcription-PCR of 16S rRNA sequences. Appl. Environ. Microbiol. 60 (12), 4580–4583. McIntosh, D., Meaden, P.G., Austin, B., 1996. A simplified PCR-based method for the detection of Renibacterium salmoninarum utilizing preparations of rainbow trout (Oncorhynchus mykiss, Walbaum) lymphocytes. Appl. Environ. Microbiol. 62 (11), 3929–3932. Meyers, T.R., Thrower, F., Short, S., Lipson, K., Joyce, J., Farrington, C., et al., 1999. Different prevalences of Renibacterium salmoninarum detected by ELISA in Alaskan chinook salmon Oncorhynchus tshawytscha spawned from freshwater and seawater. Dis. Aquat. Organ. 35 (2), 101–105. Miriam, A., Griffiths, S.G., Lovely, J.E., Lynch, W.H., 1997. PCR and probe-PCR assays to monitor broodstock Atlantic salmon (Salmo salar L.) ovarian fluid and kidney tissue for presence of DNA of the fish pathogen Renibacterium salmoninarum. J. Clin. Microbiol. 35 (6), 1322–1326. Nazarenko, I., Lowe, B., Darfler, M., Ikonomi, P., Schuster, D., Rashtchian, A., 2002. Multiplex quantitative PCR using self-quenched primers labeled with a single fluorophore. Nucleic Acids Res. 30 (9), e37.

80


Piganelli, J.D., Wiens, G.D., Zhang, J.A., Christensen, J.M., Kaattari, S.L., 1999. Evaluation of a whole cell, p57-vaccine against Renibacterium salmoninarum. Dis. Aquat. Organ 36 (1), 37–44. Powell, M., Overturf, K., Hogge, C., Johnson, K., 2005. Detection of Renibacterium salmoninarum in chinook salmon, Oncorhynchus tshawytscha (Walbaum), using quantitative PCR. J. Fish Dis. 28 (10), 615–622. Rhodes, L.D., Nilsson, W.B., Strom, M.S., 1998. Sensitive detection of Renibacterium salmoninarum in whole fry, blood, and other tissues of pacific salmon by reverse transcription-polymerase chain reaction. Mol. Mar. Biol. Biotechnol. 7 (4), 270–279.

Suzuki, K., Sakai, D.K., 2007. Real-time PCR for quantification of viable Renibacterium salmoninarum in chum salmon Oncorhynchus keta. Dis. Aquat. Organ 74 (3), 209–223. Wiens, G.D., Chien, M.S., Winton, J.R., Kaattari, S.L., 1999. Antigenic and functional characterization of p57 produced by Renibacterium salmoninarum. Dis. Aquat. Organ 37 (1), 43–52.