Aquatic Animal Health Unit, Faculty of Veterinary Medicine,
University Putra Malaysia, 43400 Serdang, Selangor, Malaysia
Introduction
The use of any DNA-based diagnostic technology faces some fundamental problems, as implementation requires knowledge in the field of molecular diagnostics. Such knowledge is obtained by experience gained from research on molecular detection methods. It is important that true expertise is established in molecular biology and that technologies are not introduced with a "jump in the bandwagon" mentality. It should also be recognised that interpretation of results can be problematic when strict controls, guidelines and problem solving are not adhered to.
Furthermore, the implementation of DNA-based diagnostic programs must add real value. They should not be implemented for the sake of novelty, particularly when employed for diagnostic certification. The need for implementation needs to be determined in terms of practicality, ease of use, speed and technical capabilities of trained manpower. The programs must also be reliable and reproducible with very low frequency of false interpretation and variation between laboratories. Finally, the cost involved in running such programs should not be prohibitive as the outcome of results reflects the professionalism, responsibility and standards with which the tests were conducted.
Advantages of PCR diagnosis
PCR is at the forefront of molecular diagnostic technology today. It is highly sensitive, with a capacity to amplify from even a single molecule of DNA. It is also very specific due to the nature and orientation of the oligonucleotide primers that are required to allow amplification to proceed. PCR is also very rapid. In only a few hours, millions of copies of a single DNA segment can be produced by standard procedures. In my laboratory, we have achieved similar results in less than 5 minutes by using rapid cycle amplification in capillary tubes. The potential for semi-automation of the procedure would make it even more attractive, allowing genetic information to be acquired more quickly. There is also potential for DNA sequence analysis of the PCR product to confirm the identity of a disease or pathogen and allow examination for genetic variation.
Such advantages have been of great benefit to molecular diagnostics. In general, 2 areas that has benefited the most from this technology are disease gene mutation screening and of course the detection and characterisation of specific pathogens. Can there be any doubt that PCR should be used as a diagnostic tool when it can pick up a needle and amplify it in a haystack?
Potential obstacles to PCR standardization
Why are there potential problems with the implementation of PCR for the diagnosis of WSSV? The reason is very simple. Diagnosing WSSV with PCR can go awfully wrong if the technology is applied casually. There are some contributing factors that could lead us to arrive at such a conclusion.
Firstly, PCR requires some technical knowledge for effective implementation. Standardization between laboratories that may have no molecular biological research background will require the provision of support to develop technical capabilities. Secondly, due to the flexibility of PCR with its many amplification strategies, the selection of appropriate gold standards that will be critical for acceptance of diagnostic certification for WSSV requires careful consideration. The use of different strategies such as single or nested PCR leads to differences in the sensitivity of detection. There has been strong inclination for laboratories in Asia to adopt the nested approach, as the method is at least 100 to 1000 times more sensitive than single PCR. This extreme sensitivity comes with its own set of problems as will be discussed later.
Another important factor that needs to be addressed is that of reproducibility between laboratories. The assay procedure not only consists of performing the PCR but also reproducing the same sensitivity, eliminating of false interpretation and implementing contamination control procedures.
Questions have also been posed as to whether PCR results should be just a simple positive or negative. This issue stems from an interpretation that a PCR result represents an absolute presence or absence of WSSV in the test sample. Presently, routine PCR detection of WSSV can result in the same observation for both early and late stages of infection. The biological significance of such a PCR result is somewhat obscure. How do we assess the degree of infection? This problem has been demonstrated by the reports of farms having white spot infection but with varying degrees of mortality. The use of PCR in WSSV diagnosis is presently devoid of the ability to monitor the progress of the disease. There are ways to slightly overcome this problem, but the work and effort required for routine diagnosis may be prohibitive and the methods do not provide good accuracy in predicting the viral load of WSSV.
Another area of serious concern is the high probability of false interpretation of results. This is largely due to ignorance in establishing, employing and following strict PCR controls for verification of results. For example, failure to use negative controls can lead to the interpretation of negative WSSV detection in a test which was actually a failed PCR! This is frequently caused by PCR inhibition - a factor which deserves greater attention in WSSV diagnosis as shrimp tissues frequently contain PCR inhibitory substances. False positive interpretation is almost exclusively due to contamination from post-PCR amplification products. While sample to sample contamination do exist, it is contamination from PCR products that causes greatest concern and headache. This problem is even more serious when procedures such as nested PCR are used for routine WSSV detection. In view of these problems, it will be important to ensure that, in addition to the use of standard PCR procedures, appropriate positive and negative controls are also employed. The use of proper controls is seldom emphasised, and if controls are emphasised obtaining them is not easy.
Perhaps the greatest concern in the use of PCR for WSSV diagnosis is the lack of strict quality control on the results released. Without proper guidelines, consistent, standard and valid PCR diagnosis of WSSV will be virtually impossible to achieve.
Areas that require standardization
With so many factors that could go wrong in PCR, one could wonder where to begin with standardization. Perhaps a general view of the process involved to set up a PCR based diagnostic assay would be useful. This is certainly not an exhaustive list but may help the difficult process of finding focal points in establishing standard and valid PCR assays for WSSV diagnosis in Asia.
The areas we have identified that may lead to initiatives in standardization and validation programs are sample processing, PCR set-up, amplification strategies, controls, specificity and sensitivity.
Sample processing. The initial stages before performing PCR analysis involve sample processing, and correct sampling methods. It is important to ensure the correct sample size for reliable detection of pathogens as possible low prevalence of the agents can render a false negative interpretation. In determining sample size for PCR analysis, the statistical confidence levels provided by Amos (1985) are a useful guide. However, a practical issue to consider is the sampling method by the farmer. For example, we have had cases in which hatchery operators have requested PCR analysis on post-larval samples that have been pooled from various tanks. This sampling practice jeopardises the accuracy and statistical confidence of the PCR results.
The proper selection of DNA sources is also critical in obtaining accurate PCR results. From our own experience, use of gill, muscle and integument tissues yields good results, provided that the amount of DNA is appropriate. An excess of DNA may inhibit the PCR process. Haemolymph can also be used but it is usually not appropriate for routine diagnosis. PCR on post-larvae is sometimes more problematic as PCR inhibitory substances are present in the eyes which should be removed before DNA extraction. The inhibitory substances have not yet been properly identified but there is evidence suggesting the involvement of poryphrin which is a common component of pigment in shrimp. Little research has been conducted to resolve this problem.
The DNA extraction procedure may be a good area in which to commence standardisation. There are a number of methods to obtain DNA for PCR analysis and these will be discussed later. The key factors to consider in the development of a standardized procedure are time/manpower, safety, reliability, contamination risks, inhibition and DNA quality.
The significance time required to conduct the test is important, especially when a large sample number has to be screened. The availability of manpower and technical technical skills should also be considered. Worker safety when employing the procedures should not be compromised. The extraction procedure should be evaluated for its reliability and risk of contamination when processing a large number of samples. The extraction methods should minimise the use of PCR inhibitory substances as will be described later. There is also a need for some level of DNA quality for PCR analysis to generate accurate results.
Methods which can be employed for extraction of WSSV DNA include alkaline lysis, proteinase K digestion, treatment with guanidinium salts (passive) and boiling. Alkaline lysis is rapid and provides good quality DNA for PCR. However, a boiling step to properly liberate DNA from tissues and a neutralisation step are required. Depending on how the samples are handled, contamination risks can increase when boiling is employed. Proteinase K digestion is relatively slow as it involves several subsequent steps and the protocol also usually involves the use of hazardous materials such as phenol and chloroform. Detergents are also commonly used for the liberation of DNA. However, ionic detergents such as SDS can inhibit Taq polymerase at certain concentrations. Residual phenol can also cause inhibition of Taq polymerase.
Guanidinium salts have not been widely used but could provide some very useful advantages. Guanidinium salts are non-toxic and the method is rapid, requiring only a few steps involving ethanol washing and precipitation. The technique also allows passive lysis for release of DNA from tissues without the need to boil or homogenise. As such, contamination risks can be significantly reduced. Guanidinium salts also inhibit common DNA degrading enzymes so tissue can be stored in solution while being transferred or while awaiting extraction. The use of simple boiling methods in lysis buffers should be evaluated carefully in terms of contamination risks and DNA quality. It is our opinion that the disadvantages far outweigh its advantages of DNA extraction using this method.
In summary, a good DNA extraction protocol after going through all the above discussion should ideally be rapid, non-toxic, economical, and have a low risk of contamination.
PCR set-up. The second area for which standardisation is the set-up of the PCR laboratory. A properly equipped laboratory to perform PCR diagnostics is absolutely critical if routine WSSV diagnosis is to be conducted. The use of dedicated instruments such micro-pipettors and aerosol resistant tips for PCR must be strictly observed. A PCR facility must be clean and well managed, and the architecture of the laboratory must include separate rooms for pre- and post-amplification procedures. Samples must not be prepared in the same room as the PCR machine is located and PCR post-amplification analysis is performed. The laboratory in which PCR reagents are prepared should also clean benches and ideally should include a laminar flow cabinet with built-in UV lamps for decontamination procedures after every reaction set-up.
The need for trained manpower is critical, not just to perform the PCR assay but to ensure the reliable management of a PCR diagnostic laboratory. PCR reagents must be properly prepared and troubleshooting skills must be available so that the quality of result test results is consistently and strictly controlled. Decontamination protocols should also be established and strictly followed.
Amplification strategies. Despite all care, it is our experience that contamination can and will occur. For this reason, the use of low contamination risk amplification strategies is essential. The following PCR strategies are now in use.
PCR may be conducted using single step or nested primer methodologies. In our experience, as shared by many other laboratories in Asia, nested PCR can reach 1000 times the sensitivity as compared to single step PCR. The extreme sensitivity of nested PCR works well in many cases but the test is highly susceptible to contamination as it involves more time for sample manipulation. The most common source of contamination is the product from a previous PCR. In fact, one molecule of a contaminating PCR template in the first step reaction may be sufficient to obtain a false positive result by nested PCR. The nested PCR is also more costly as it involves the use of 2 separate reactions to arrive at one result. The cost will increase dramatically if assays are repeated when contamination occurs.
A new methodology which may give more biological significance to the test result is quantitative PCR. Qualitative PCR provides no useful quantitative assessment of the infection level or disease progression and does not indicate pathogen replication. Because of this deficiency, the application of primary cell lines have been suggested to be more appropriate pathogen detection than PCR, despite the difficulty in establishing shrimp cell lines. However, quantitative PCR technology has been available for some time and has been widely applied in diagnosis of human pathogens such as HIV and herpes simplex virus. The ability to quantify using PCR allows investigation of the molecular pathogenesis of an infection. Such information for WSSV has been very limited apart from studies conducted using DNA probes in Taiwan. In situ hybridisation (ISH) using DNA probes also available for disease investigation but the method is technically demanding and requires time for data interpretation. Although it does not provide a direct quantification of the virus, ISH may give information on the severity of infection. Quantification using PCR is more rapid and accurate, and can provide an absolute determination of the number or copies of the targeted DNA. The most reliable method for PCR quantification is by competitive PCR. This technique utilises a known amount of engineered internal standard which has the same primer binding sites as the target DNA or RNA. The internal standard is differentiated from the target by size on the basis of size by inclusion of a small deletion, or by including a single base mutation that allows separation after restriction enzyme digestion. Since the target DNA and internal standard are virtually identical, the efficiency of amplification should be equivalent, leading to a fair competition when co-amplified and an accurate measurement of the relative concentrations. This eliminate tube-to-tube and template variability. PCR quantification can be conducted in real time with specialised equipment which uses fluorescent dye quenching technology. The cost to purchase such equipment may be prohibitive for some laboratories. The procedure makes absolute quantification possible but is technically demanding in the initial stages of design and development of engineered standards for co-amplification. It is also more costly as it requires the amplification of multiple standards simultaneously and is less sensitive as compared to nested PCR.
In order to develop an easy, sensitive and robust method with lower risk of contamination for the routine diagnosis of WSSV infection, we have been developing a single sample load, single-tube nested PCR. By this method, we can reliably detect 1-10 copies of double-stranded viral DNA target. The assay is much easier, more user friendly and faster with very little hands-on manipulation of the PCR reaction preparation and is less susceptible to sporadic contamination. The assay is also half the cost of our normal 2 tube nested PCR as only require one reaction tube and set of PCR reaction components are required.
Semi-quantitative properties can also be incorporated into the system for an assessment of the WSSV infection level. The two tube nested PCR provides very little quantitative information. A positive PCR signal has very much the same overall intensity over a wide range of DNA target copies. The re-amplification of the first step PCR product quickly masks differences in the concentrations of amplified products due to the tendency of the PCR process to reach a plateau after 30 or 40 cycles, regardless of the amount of the input target template. This is also due in part to the presence of renewed sources for PCR amplification in the second tube. The characteristics of the single-tube nested PCR approach are different. There is a direct relationship between the amount of the first step PCR product and the amount of the nested PCR product. At low target concentration, the nested PCR step does not proceed to reach a plateau as in the 2 tube nested PCR. This phenomenon occurs only when high template target is present. As such, different levels of PCR signal intensity can be observed, reflecting the severity of the infection. This permits a semi-quantitative approach to be applied to the assessment of a positive result.
Although this requires extensive and intensive optimisation to achieve reliability, such approaches may be more beneficial in the long run. In our case, the use of more sophisticated PCR methods has been worth the investment in energy, money and manpower.
Diagnostic controls. Standardisation is also required in the use of appropriate diagnostic controls. Without proper controls for result interpretation, no PCR result should ever be accepted as valid. Appropriate controls, which are essential for elimination of false negative results, involve the simultaneous amplification of a fragment of host DNA using primers targeting conserved sequences in the host genome. This control indicates that the PCR has been successful and that template quality was adequate for PCR amplification. It also allows the recognition of PCR inhibition factors in the sample. The process will require the use of multiplex PCR as it will be more reliable when the control is amplified in the same diagnostic reaction.
Appropriate control reactions to indicate false positive results should also be performed. A negative control reaction without template must be used in every assay. The use of vapour barriers such a mineral oil overlay of the PCR reaction is advised even though most thermal cyclers now employ a heated bonnet to prevent vapour transfers. Strict adherence to contamination prevention and decontamination procedures must also be followed. The use of positive controls that can be differentiated from the target is also advantageous when it is necessary to distinguish a true positive result from contamination with a positive control. By knowing whether contamination is from a diagnostic target or from a positive control permits a more focused approach in eliminating the source of contamination. Weak positive controls (low amount of target) can also be employed if necessary to minimise the risk of contamination in the laboratory. A novel approach is the use of engineered controls that differ in properties from diagnostic targets. Differentiation can be accomplished by another round of PCR or by hybridisation with an internal probe that detects the difference in the PCR fragments. However, the fastest and easiest way is to use restriction enzymes that cut either the target or control fragments. By this method, the result can be determined by analysing the PCR products elctrophoretically.
Diagnostic specificity. In establishing standard PCR protocols for WSSV diagnosis, the specificity of PCR primers for conserved regions of the genome also should be considered. WSSV variants may occur that cannot be detected by certain primer sequences. For example, Korean researchers (Park et al., 1998) could amplify PCR products from WSSV-diseased shrimp using primers developed for RV-PJ but not when using primers developed by Lo and co-workers. It is also interesting that the Korean virus looks similar to WSSV found in Taiwan and not to RV-PJ from Japan. The sequence of the PCR product was identical to sequence of RV-PJ fragment. It seems likely that mutations in WSSV are occurring and this should be investigated. The possibility of having less virulent or virulent strains of WSSV should also be considered. This has been reported for yellow head viruses in research conducted in Australia (Dr. Peter Walker, CSIRO, Chiang Mai Meeting). However, as YHV is an RNA virus, it would be expected to mutate more frequently than WSSV which is a DNA virus.
Confirmation of PCR specificity is quite straightforward. The nested PCR approach is already an accepted method for confirmation of PCR specificity based on the distinct size of the nested PCR product. Other techniques such as RFLP analysis and the use of an internal ISH probe are more laborious.
Diagnostic sensitivity. Another area that sometimes evokes confusion that requires proper standardisation is the definition of PCR diagnostic sensitivity or the limit of PCR detection. Descriptions of detection limits currently vary from the number of DNA target copies to the number of virus particles. It has not been established for WSSV whether the identification and detection of single gene sequence reflects the detection of a single viable virus. It should be determined if sensitivity can be standardised with quantitative qualities such as infection status, disease progression, viral replication or antiviral therapies and strategies. The ideal WSSV diagnostic PCR assay would determine the infection status and allow more accurate monitoring of the progression of the disease. The monitoring of the disease has an important role in the management of WSSV infection on farms. There is clearly potential to reduce the economic impact of WSSV by developing disease management plans based on viral load information. In this approach, it will be important to determine whether the virus is replicating and causing the disease.
The most important potential application of PCR technology is the possibility of developing new antiviral therapies or strategies against WSSV. It is impossible to assess these accurately using the techniques presently available.
Conclusion
How do we begin to validate PCR protocols then for WSSV diagnosis? Is it first through standardisation or it is more practical to identify areas for validation that would allow a better standardisation exercise? After evaluating all the problems and technical issues associated with the PCR process, validation steps can be perhaps grouped into a few general areas.
Firstly, it is obvious that for a procedure to be valid, the protocol must be reliable, simple and sensitive enough to do the work required. It must be technically easy to perform to allow reliable replication between laboratories. This should encompass the entire process from DNA preparation to PCR amplification.
Secondly, the use of proper controls must be employed consistently to totally eliminate false negative and false positive results. In other words, very strict quality control must be imposed on laboratories performing diagnosis.
Thirdly, it may be important to include a quantitative element to add the biological significance dimension to diagnosis. This criterion has often been applied to diagnostic assays for infectious pathogens. In any case, it is easier to verify a result when degree of infection is incorporated into the assay. Again strict quality control must be implemented to allow the validation of a diagnostic result. This may also minimise legal concerns and certainly will reduce unnecessary economic losses on farms.
The use of a PCR test kit may help in the validation process. In this way, individual laboratories would need only to meet certain requirements and avoid the need to develop diagnostic methods that meet the imposed standards. However, a suitable PCR diagnostic kit must first be developed. This will help encourage private companies to provide commercial clinical diagnostic services without a huge investment in RandD. The kit in our opinion must satisfy basic criteria such as being simple to perform and interpret, have a low contamination risk, include an appropriate false negative and false positive control/indicators, have appropriate sensitivity controls and the test should be semi-quantitative.
In short, the test should be designed to meet the expectations of the novice clinician for whom the use of PCR as a diagnostic tool is still foreign. The test should also be sufficiently complete in its features to satisfy the experienced investigator. The interpretation of results should be similarly and readily understood by both groups. The complete package should be sufficiently attractive to justify its use despite a potentially higher cost. The test must be the gold standard if is to be implemented as a service to our local farmers who depend on us in seeing through their crop and their livelihood.
References
Botwell, D.D.L. (1987). Rapid isolation of eukaryotic DNA. Analytical Biochemistry 162, 463-465.
Chang, P.S., Lo, C.F., Wang, Y.C. and Kou, G.H. (1996). Identification of white spot syndrome associated baculovirus (WSBV) target organs in shrimp Penaeus momnodon by in situ hybridization. Diseases of Aquatic Organisms 27, 131-139.
Chomczynski, P., Mackey, K., Drews, R. and Wilfinger, W. (1997). DNAzolâ: A reagent for the rapid isolation of genomic DNA. BioTechniques 22, 550-553.
Clementi, M., Menzo, S., Bagnarelli, P., Manzin, A., Valenza, A. and Varaldo, P.E. (1993). Quantitative PCR and RT-PCR in virology. PCR Methods and Application 2, 191-196.
Clementi, M., Menzo, S., Manzin, A. and Bagnarelli, P. (1995). Quantitative molecular methods in virology. Archives of Virology 140, 1523-1539.
Flegel, T.W. (1997). Major viral diseases of the black tiger prawn (Penaeus monodon) in Thailand. World Journal of Microbiology and Biotechnology 13, 433-442.
Lo, C.F., Leu, J.H., Ho, C.H., Chen, C.H., Peng, S,E., Chen, Y.T., Chou, C.M., Yeh, P.Y., Huang, C.J., Chaou, H.Y., Wang, C.H. and Kou, G.H. (1996). Detection of baculovirus associated with white spot syndrome (WSBV) in penaeid shrimps using polymerase chain reaction. Diseases of Aquatic Organisms 27, 131-139.
Park, J-H., Lee, Y.S., Lee, S. and Lee, Y. (1998). An infectious viral disease of penaeid shrimp newly found in Korea. Diseases of Aquatic Organisms 34, 71-75.
Rolfs, A., Schuller, I., Finckh, U. and Weber-Rolfs, I. (1992). PCR: Clinical diagnostics and research. Springer-Verlag Berlin Heidelberg. 268 pp.
Saiki, R.K., Gelfand, D.H., Stoffel, S., Scharf, S.J., Higuchi, R., Horn, G.T., Mullis, K.B. and Erlich, H.A. (1988). Primer directed enzymatic amplification of DNA with a thermostable DNA-polymerase. Science 239, 487-491.
Wongteerasupaya, C., Wongwisansri, S., Boonsaeng, V., Panyim, S., Pratanpipat, P., Nash, G.L., Withyachumnarnkul, B. and Flegel, T.W. (1996). DNA fragment of Penaeus monodon baulovirus PmNOBII gives positive in situ hybridization with white spot viral infections in six penaeid shrimp species. Aquaculture 143, 23-32.
Xu, H., Jevnikar, A.M. and Rubin-Kelly, V.E. (1990). A simple method for the prepartion of chromosomal DNA from cell. Culture. Nucleic Acids Research 18, 4943.
