RAPID DETECTION AND IDENTIFICATION OF FISH PATHOGENS - ALEXANDRA ADAMS

Institute of Aquaculture, University of Stirling FK9 4LA, Scotland, UK.

Introduction

Rapid identification of pathogens is crucial for effective disease control in aquaculture. Detection of pathogens is important not only in infected fish (clinically and sub-clinically), but also in the environment e.g. between harvesting and re-stocking, and as an 'early warning system'. The application of antibody probes and DNA primers/probes in pathogen detection and identification has made a significant impact on the development of such rapid diagnostic methods. Standardisation and validation of these methods, however, has in most cases not been addressed (Hiney and Smith, 1998).

Antibody-based tests

A variety of antibody-based tests and molecular tests have been developed to detect mainly bacterial and viral fish pathogens, although tests have also recently been reported for parasites and fungal agents. The antibody-based tests include slide agglutination, co-agglutination/latex agglutination, immunodiffusion, direct and indirect fluorescent antibody tests (FAT and IFAT), immunohistochemistry (IHC) and enzyme linked immunosorbent assay (ELISA), dot blot/dip stick and western blot (WB) (reviewed by Adams, 1999). The antibody-based test selected for the identification of pathogens depends on a variety of factors since each method has its merits and disadvantages. Although such methods are useful for the detection of pathogens in pure culture or/and in infected fish tissue, their sensitivity thresholds limit use in environmental samples, especially where pathogen levels are extremely low. DNA detection methods, however, such as polymerase chain reaction (PCR) and in situ hybridisation are ideally suited.

DNA detection methods

Many bacterial pathogens can now be detected in samples of various kinds without the need to first culture the organism. PCR methods are not only highly specific and quick but they also can lead to the detection of 'non-culturable' bacteria. (Brauns et al., 1991). PCR and in situ hybridisation methods are currently being developed for the detection of numerous fish pathogens. These include tests to identify numerous bacterial pathogens such as Renibacterium salmoninarum (Brown et al., 1994); Aeromonas hydrophila (Cascon et al., 1996); Aeromonas salmonicida (Hiney et al., 1992); Vibrio anguillarum (Hirono et al., 1996), Photobacterium damsela subspecies pisicida, formerly Pasteurella piscicida (Aoki et al., 1995; 1996), and many others are under development.

Examples of antibody and DNA based methods currently in use

A variety of rapid methods for pathogen detection and identification have been developed at the University of Stirling. These will be used to illustrate the application of such tests, and serve to pin-point areas requiring standardisation. All the tests require validation in the field. The tests developed include IFAT, IHC, ELISA (detection of pathogen/antibodies), PCR (one cycle/nested/reverse cross blot hybridisation/RT-PCR/quantitative PCR), and in situ hybridisation. These assays are performed on a variety of pathogens, including bacteria (Aeromonas salmonicida, Aeromonas hydrophila, Photobacterium damsela subspecies piscicida, Renibacterium salmoninarum, Mycobacterium spp., Vibrio anguillarum and Flavobacterium psychrophilum; Adams and Thompson, 1990; Adams et al., 1995; Bakopoulos et al., 1997; Adams et al., 1996), parasites (PKX; Morris et al., 1997; 1998), viruses (infectious salmon anaemia virus), and fish rickettsia (Piscirickettsia salmonis; Alday-sanz et al., 1994). In some instances the antibody-based methods are sufficient and achieve the necessary sensitivity and specificity. On other occasions, for example in water samples and sub-clinical disease, the increased sensitivity of DNA-based methods is required. Again, the method selected depends on a variety of factors.

PCR and in situ hybridisation

PCR is used for the identification of pathogens in blood, water, sediment and tissue samples due its high sensitivity and specificity. In our laboratory, we use this method to detect Mycobacterium spp., F. psychrophilum, R. salmoninarum and ISAV. This involves identification to species level in some cases e.g. M. marinum, M. fortuitum and M. chelonae can be identified using reverse cross blot PCR. This is normally performed on fresh or frozen samples. Fixed samples on the other hand are analysed using in situ hybridisation. Both PCR and in situ hybridisation are used in preference to antibody-based methods for the detection of the parasite PKX as detection with antibodies is life cycle dependent (Morris et al., 1998).

Detection of PCR products

PCR can be performed as a single or nested assay, and the products can be identified using a variety of methods. Our present tests include single/nested PCR (PKX, R. salmoninarum), reverse cross blot hybridisation to identify multiple species in a single test (Mycobacterium spp.), RT-PCR for RNA viruses (ISAV), and quantitative PCR (F. psychrophilum). Each assay has been optimised and the method selected according to the type of pathogen and the kinds of samples requiring analysis. For example, the detection of F. psychrophilum is required in environmental samples as well as fish tissue. Quantitation is also desired so that upward trends in pathogen titre can be detected. This is achieved with this particular assay by linking an ELISA to the PCR to detect the products.

Standardisation, validation and interpretation of results

Clearly, the standardisation of methods is crucial. The extraction method and primers used will influence the results of the PCR and these should be clearly defined. The use of PCR beads minimises the risk of contamination and extreme care should be taken with nested PCR. The inclusion of all the appropriate positive (sometimes a series of DNA positive standards is required) and negative controls is essential if the results are to be interpreted correctly. The analysis of results may be difficult in some cases, especially with tissue samples where tissue may inhibit the reaction. Table 1 illustrates the interpretation of results from a typical PCR test in our laboratory. As well as having the obvious positive and negative controls it is necessary to include spiked samples i.e. a known positive control is added to the sample. This controls for inhibition by the tissue and false negative results. Normally the non-spiked samples are run in duplicate, while the spiked samples are run singly. Thus, three results are generated from each sample. A decision is then taken on whether to accept the results as positive or negative, or to repeat the test. A positive result is expected from the spiked sample on each occasion, regardless of the result from the actual test sample (the non-spiked sample). A negative result can, however, sometimes be generated by the spiked sample if the DNA level is high, resulting in inhibition. If this is the case then comparison of the spiked results with those of the test sample (run in duplicate) results should lead to a definitive interpretation; whether this is to report positive, negative or to repeat the test. As shown in Table 1 the test would need to be repeated if all three samples were negative in the test, or if the non-spiked duplicate samples resulted in one positive and one negative.

The PCR tests developed in our laboratory are capable of detecting the presence of specific pathogen DNA. As long as the test is optimised, standardised and all the appropriate controls are included, the results can be interpreted and appear to be consistent. The actual meaning of the results with regard to disease, however, has not been established as the tests have not been fully validated in the field. It is important to note that both the antibody and DNA-based methods do not necessarily detect live pathogen. A positive result will also be obtained with non-viable pathogen (Josephson et al., 1993), or even only parts of a dead pathogen. Thus, validation of the methods in the field to see how a positive result correlates with infection and disease is vital. Blind trials need to be conducted to confirm that a negative is always negative in the test, and comparisons conducted with other methods. Some preliminary work has been conducted on the validation of the F. psychrophilum, Mycobacterium spp. and PKX tests in our laboratory. There are also many reports of the application of PCR-based techniques to the detection of fish pathogens and these also require validation (Hiney and Smith, 1998).

Table 1. Interpretation of results from a typical PCR test

* Dilute DNA and repeat PCR

References

Adams, A. (1999) Application of antibody probes in the diagnosis and control of fish diseases. Silver Jubilee Lecture, Mangalore, India (in press).

Adams, A., and Thompson, K. (1990). Development of an ELISA for the detection of Aeromonas salmonicida in fish tissue. Journal of Aquatic Animal Health 2, 281-288.

Adams, A., Thompson, K.D., McEwan, H., Chen, S.-C. and Richards, R.H. (1996). Development of monoclonal antibodies to Mycobacterium spp. isolated from Chevron snakehead and Siamese fighting fish. Journal of Aquatic Animal Health 8, 208-215.

Adams, A., Thompson, K.D., Morris, D, Farias, C. and Chen, S.C. (1995). Development and use of monoclonal antibody probes for immunohistochemistry, ELISA and IFAT to detect bacterial and parasitic fish pathogens. Journal of Fish and Shellfish Immunology 5, 537-547.

Alday-Sanz, V., Rodger, H., Turnbull, T., Adams, A. and Richards, R.H. (1994). Immunohistochemical identification of Pickirickettsia salmonis in Atlantic salmon, Salmo salar L. Journal of Fish Diseases 17, 189-192.

Aoki, T., Hirono, I. and Hayashi, A. (1995). The fish pathogenic bacterium Pasteurella piscicida detedted by polymerase chain reaction (PCR). In Diseases in Asian Aquaculture II. Shariff., Arthur, J.R., Subasinghe, R.P. (Eds.). Fish Health Section, Asian Fisheries Society, Manila, pp. 347-353.

Aoki, T., Ikeda, D., Katagari, T. and Hirono, I. (1996). Rapid detection of the fish-pathogenic bacterium Pasteurella piscicida by polymerase chain reaction targetting nucleotide sequences of the species-specific plasmid pZP1. Fish Pathology 32, 143-151.

Bakopoulos, V., Adams, A. and Richards, R.H. (1997). The production and characterisation of monoclonal antibodies against the fish pathogen Pasteurella piscicida. Journal of Fish Diseases 20, 307-315.

Brauns, L.A. Hudson, M.C. and Oliver, J.D. (1991). Use of the polymersae chain reaction in detection of culturable and nonculturable Vibrio vulnificus cells. Applied Envionmnetal Microbiology 57, 2651-2655.

Brown, L.L., Iwama, G.K., Evelyn, T.P.T., Nelson, W.S. and Levine, R.P. (1994). Use of polymersase chain reaction (PCR) to detect DNA from Renibacterium salmoninarum within individual salmonid eggs. Diseases of Aquatic Organisms 18, 165-171.

Campbell, R., Adams, A., Tatner, M.F., Chair, M. and Sorgeloos, P. (1993). Investigation of antigen uptake by Artemia during incubation with Vibrio anguillarum vaccine using enzyme linked immunosorbent assay (ELISA). Journal of Fish and Shellfish Immunology 3, 451-459.

Cascon, A., Anguita, J., Hernandez, C., Sanchez, M., Fernandez, M. and Naharro, G. (1996). Identification of Aeromonas hydrophila hybridisation group 1 by PCR assays. Applied Environmental Microbiology 62, 1167-1170.

Hiney, M.P. and Smith, P.R. (1998). Validation of polymerase chain reaction-based techniques for proxy detection of bacterial fish pathogens: framework, problems and possible solutions for environmental applications. Aquaculture 162, 41-68.

Hiney, M., Dawson, M.T., Heery, D.M., Smith, P.R., Gannon, F., and Powell, R. (1992). DNA probe for Aeromonas salmonicida. Applied Environmnental Microbiology 58, 1039-1042.

Hirono, I., Masuda, T. and Aoki, T. (1996). Cloning and detection of the hemolysin gene of Vibrio anguillarum. Microbial Pthogenesis, 20.

Josephson, K.L., Gerba, C.P. and Pepper, I.L. (1993). Polymerase chain reaction detection of non-viable bacterial pathogens. Applied Environmnetal Microbiology 59, 3513-3515.

Morris, D., Adams, A. and Richards, R.H. (1997). Studies on the PKX myxosporean in rainbow trout via immunohistochemistry and immunogold microscopy. Journal of Aquatic Animal Health 8, 219-235.

Morris, D., Adams, A. and Richards, R.H. (1998). In situ hybridisation of PKX, the causative organism of Proliferative Kidney Disease (PKD). Journal of Fish Diseases (in press).