Here we present a procedure to quantify enterovirus and norovirus in environmental and drinking waters using reverse transcription-quantitative PCR. Mean virus recovery from groundwater with this standardized procedure from EPA Method 1615 was 20% for poliovirus and 30% for murine norovirus.
EPA Method 1615 measures enteroviruses and noroviruses present in environmental and drinking waters. This method was developed with the goal of having a standardized method for use in multiple analytical laboratories during monitoring period 3 of the Unregulated Contaminant Monitoring Rule. Herein we present the protocol for extraction of viral ribonucleic acid (RNA) from water sample concentrates and for quantitatively measuring enterovirus and norovirus concentrations using reverse transcription-quantitative PCR (RT-qPCR). Virus concentrations for the molecular assay are calculated in terms of genomic copies of viral RNA per liter based upon a standard curve. The method uses a number of quality controls to increase data quality and to reduce interlaboratory and intralaboratory variation. The method has been evaluated by examining virus recovery from ground and reagent grade waters seeded with poliovirus type 3 and murine norovirus as a surrogate for human noroviruses. Mean poliovirus recoveries were 20% in groundwaters and 44% in reagent grade water. Mean murine norovirus recoveries with the RT-qPCR assay were 30% in groundwaters and 4% in reagent grade water.
Quantitative PCR (qPCR; see supplemental materials for definitions of terms used in this manuscript) and reverse transcription-qPCR (RT-qPCR) are valuable tools for detecting and quantifying human enteric viruses in environmental and drinking waters, and especially for many viruses that do not replicate or replicate poorly in cell culture systems. Both tools have demonstrated that many virus types are present in environmental and drinking waters throughout the world1-6. Their use coupled with sequencing of amplified genomic fragments during disease outbreak investigations has provided evidence for waterborne virus transmission, as they have shown that the virus found in the drinking water is identical to that shed by outbreak patients7-10.
Both qPCR and RT-qPCR are useful public health tools. For example, data from studies conducted by the U.S. Environmental Protection Agency (EPA) showed a strong relationship between indicator measurements by qPCR and health effects in recreational waters. As a result, EPA's final 2012 Recreational Water Quality Criteria includes a qPCR method for monitoring recreational beaches11,12. Borchardt and colleagues also found a strong relationship between acute gastroenteritis in communities using untreated groundwater and virus in groundwater as measured by RT-qPCR1.
The purpose of this paper is to describe the molecular assay component of EPA Method 161513,14. This assay uses RT-qPCR to provide a quantitative estimate of enterovirus and norovirus genomic copies (GC) per liter based upon the original volume of the environmental or drinking water passed through an electropositive filter. An overview of the molecular procedure is shown in Figure 1. Protocol section 1 details the procedures for preparing the standard curve. These standards are prepared from a reagent that contains an RNA copy of the target sequence for all the primer/probe sets. Section 2 describes the tertiary concentration procedure. Section 3 gives the procedure for extracting RNA from the concentrated water and control samples. The RNA from each test sample is reverse transcribed using triplicate assays and random primers to prime the transcription (Section 4). The cDNA from each reverse transcription reaction is split into five separate virus-specific assays that are analyzed in triplicate by qPCR (Section 5; Figure 2). The assay uses primers and probes from the scientific literature (Table 1) that are designed to detect many enteroviruses and noroviruses and a reagent containing hepatitis G RNA to identify test samples that are inhibitory to RT-qPCR15.
Large-scale national studies of viral contamination of source and drinking waters require the use of multiple analytical laboratories. Under these conditions a standard method is needed to ensure that the data generated by the multiple laboratories is comparable. There are many published molecular methods for virus detection, but very few standardized molecular methods. EPA Method 1615 is a standardized method specifically designed for detection of enterovirus and norovirus in water matrices by RT-qPCR. Standardized molecular methods are available for virus detection in foods (CEN/ISO TS 15216-1 and CEN/ISO TS 15216-2; April 7, 2013)16,17 and have been applied to the detection of hepatitis A virus and norovirus in spring water16. All standard methods must include quality performance controls and criteria to minimize inter- and intra-laboratory variation and false positive data due to laboratory contamination. To further reduce false data, EPA Method 1615 follows EPA’s guidance on molecular methods,18 which stipulates separation of work during processing and one way work flow. It includes a hepatitis G1,19 internal control and procedures to minimize false negative results due to inhibitors of RT-qPCR15. It uses quantitative assays along with standardized volumes of both water sampled and water analyzed so that all field data is expressed in genomic copies per liter of the field or drinking water sampled. Although efficient single tube (one-step) RT-PCR assays are commercially available, the method intentionally uses separate assays. This has the disadvantage of minimizing the amount of sample that can be assayed in each reaction, but gives greater flexibility in use of multiple primer sets. RT-qPCR assays are limited by and only as good as the primers and probes used and likely no primer set will detect all virus variants within a group. The enterovirus primer set was chosen because it targets the conserved 5’-non coding region,20,21 detects a wide variety of enterovirus serotypes, and viruses detected by it are associated with health effects from consumption of untreated groundwaters1. Two primer sets are used for detection of genogroup I noroviruses22,23. The first was chosen due to the strong correlation between health effects in young children and detected virus1. The second genogroup I primer set and the primer set used for genogroup II noroviruses were chosen because they detect the widest variety of strains24,25.
In spite of the major advantages of qPCR and RT-qPCR procedures for detecting viral RNA in water, there are several limitations. First, both infectious and noninfectious virus particles, including those inactivated by disinfectants, can be amplified by these procedures. The results of Borchardt suggest that this is less of a problem for untreated groundwaters from aquifers similar to those in the communities studied than for disinfected surface waters1. For culturable viruses this problem can be overcome using PCR in combination with culture26,27. The problem has also been addressed for some viruses through use of nucleic acid cross-linking agents28-30. This latter approach is more effective for viruses inactivated by hypochlorite and not effective for those inactivated by UV.
A second limitation of these molecular procedures is that the volume of concentrated sample that can be assayed typically is much smaller than that used for culture procedures6,31. This problem is often handled by either substituting a polyethylene glycol-based procedure for the standard secondary concentration by organic flocculation, which allows the sample to be resuspended in a smaller volume, or by the addition of a tertiary sample concentration step6,32,33. Method 1615 uses centrifugal ultrafiltration to provide tertiary concentration. Centrifugal ultrafiltration removes water and components less than 30,000 Daltons resulting in both concentration of any virus in test samples and a reduction in small molecular weight inhibitors of molecular assays. This tertiary concentration step results in an overall concentration factor of >105 for any virus that was present in the water being tested.
A third limitation is the presence of inhibitors of molecular procedures in environmental samples. Although numerous approaches to remove inhibitors have been developed, no approach is effective for all water matrices and virus types6,34,35, making the use of internal controls designed to estimate the level of inhibition essential. The hepatitis G reagent used in this method satisfies this requirement by providing a constant level of viral RNA in all reactions and an RT-qPCR assay for estimating inhibition. When the best of the inhibitor removal approaches fail to remove inhibition, sample concentrates can be diluted as long as virus concentrations are higher than inhibitor concentrations14,15.
The standard curve procedure described herein has both advantages and a major limitation. An advantage is that the reagent used supplies all the necessary components in a single reagent, allowing a single control to be used for all assays. This reagent is especially an advantage for norovirus assays. Norovirus particles can only be obtained from infected individuals making it very difficult to obtain viral particles for use as standards. A more important advantage is that it provides an RNA standard for all targeted RNA viruses in one reagent, as having an RNA standard is essential for accurate quantification of RNA36. However, its ability to quantify virus accurately is limited by the fact that matrix effects are not taken into account. This means that genomic copy number values cannot be considered absolute and should only be considered in relative terms. It is recommended that a sufficient number of standard curve working stock aliquots (step 1.2) be prepared to cover complete studies. For example, each 250 µl aliquot provides sufficient reagent for 6 RT plates. If it is known that a study will require analysis of 500 samples, a minimum of 12 aliquots would be needed (500 samples/7 samples per RT plate/6 RT plates per aliquot).
EPA Method 1615 is a performance based method. Many manufacturers make equivalent reagents to those specified herein and these reagents can be substituted as long as the performance criteria are met. The standard curve, which also serves as a positive RT-qPCR control, is of value in troubleshooting performance issues. Performance can decline due to degradation of RNA, reagent shelf life, failure of freezers, instrument calibration, and technical error. Performance issues should be suspected if standard curves differ from that shown in Figure 4 or if they do not meet the performance specification for standard curves. The RT-qPCR assay is quite robust; complete failure is likely due to improper handling of RNA or technical error (e.g., a missing reagent). Great care should be taken in handling RNA samples between extraction and the RT step to reduce RNA degradation from ribonucleases.
Poliovirus recoveries from ground and reagent grade waters and murine norovirus recoveries from groundwater met the EPA Method 1615 performance acceptance criteria (Table 11) and are similar to those reported by others33,37,38. Murine norovirus recoveries from LFB samples were much lower than those of poliovirus and would not have met the poliovirus-specific acceptance criteria. The reasons for lower murine norovirus recovery from reagent grade water are unknown. Similar to the results herein, Karim and colleagues reported a recovery for norovirus GI.1 of 4% from tap water39. Lee et al.37 reported mean recoveries for murine norovirus and human norovirus GII.4 of 18% and 26% from distilled water using disc filters, respectively. Using similar conditions to Lee and colleagues, Kim and Ko observed recoveries of 46% and 43% for these viruses, respectively38. Gibbons et al.40 obtained around 100% recovery of human norovirus GII.4 from sea water, but Kim and Ko38 found that the addition of salt to distilled water at concentrations similar or higher than seawater significantly reduced recoveries of the murine virus and resulted in about a two-fold reduction in GII.4 recovery.
The authors have nothing to disclose.
The authors thank Dr. Eric Rhodes for preparing the clones used in the development of Standard curve reagent, Brian McMinn for assistance in sample processing, Larry Wymer for statistical analysis, Dr. Mark Borchardt, U.S. Department of Agriculture, Marshfield, WI, for supplying the Sabin poliovirus serotype 3 used in this study and Dr. H. W. Virgin, Washington University, St. Louis, MO, for murine norovirus. The authors also acknowledge Gretchen Sullivan for assistance in preparation of stock laboratory reagents, Dr. Mohammad Karim for propagation of murine norovirus stocks, and local private well owners and utilities for allowing us to collect water samples. Although this work was reviewed by EPA and approved for publication, it may not necessarily reflect official Agency policy. Mention of trade names or commercial products does not constitute endorsement or recommendation for use.
1.5 mL tube chamber | Diversified Biotech | CHAM-3000 | |
1°C cool brick | Diversified Biotech | BRIK-2501 | |
10X PCR Buffer II and 25-mM MgCl2 | Life Technologies | N8080130 | |
-20 °C freezer | VWR | 97043-346 | Must be a manual defrost freezer |
-70 °C or colder freezer | Thermo Scientific | MBF700LSAO-E | |
96-well chamber | Diversified Biotech | CHAM-1000 | |
Absolute ethanol | Fisher Scientific | BP2818-100 | |
Armored RNA EPA-1615 | Asuragen | Custom order | Used for quantifying the RT-qPCR assay |
Armored RNA Hepatitis G virus | Asuragen | 42024 | |
Autoclave | Steris | Amsco Lab Series | |
Biosafety cabinet | NuAir Laboratory Equipment Supply | Labgard 437 ES | |
Bovine serum albumin (BSA) | Affymetrix | 10856 | Crystalline grade or better |
Buffer AVE | Qiagen | 1026956 | Carrier RNA dilution buffer |
Buffer AVL | Qiagen | 19073 | Extraction buffer |
Carrier RNA | Qiagen | Not applicable | Use carrier RNA supplied with Buffer AVL |
Centrifuge bottles | Fisher Scientific | 05-562-23 or 05-562-26 | |
Centrifuge rotors | Beckman Coulter | 339080, 336380 | |
Cool safe box | Diversified Biotech | CSF-BOX | |
Dithiothreitol (DTT) | Promega | P1171 | |
LightCycler® 480 Probes Master kit | Roche Diagnostics | 4707494001 | |
Microcentrifuge tubes | Fisher Scientific | 02-682-550 | Use ribonuclease- and deoxyribonuclease-free tubes with snap caps |
Microcide III | Fitzgerald | 99R-103 | |
Microseal 'A' film | Bio-Rad Laboratories | HSA5001 | Heat resistant |
Microseal 'F' film | Bio-Rad Laboratories | MSA1001 | Freezer resistant |
Mini-plate spinner | Labnet International | MPS1000 | |
Multichannel pipette | Rainin | L8-20 | |
Multi-tube chamber | Diversified Biotech | CHAM-5000 | |
Optical reaction plate | Life Technologies | 4314320 | |
PCR nucleotide mix | Promega | U1515 | |
PCR plate | Bio-Rad Laboratories | HSS9601 | |
PCR-grade water | Roche | 3315932001 | |
Phosphate buffered saline (PBS) | U.S. Biological | D9820 | |
Plate mixer | Scientific Industries | MicroPlate Genie | |
Prionex gelatin | Sigma Aldrich | G0411 | |
QIAamp DNA Blood Mini Kit | Qiagen | 51104 | Includes Buffers AW1 (first wash buffer), AW2 (second wash buffer), AE (elution buffer); ethanol must be added to Buffers AW1 and AW2 before use; Do not use Buffer AL supplied with the kit |
Quantitative PCR thermal cycler | Life Technologies | 4351405 | |
Random primer | Promega | C1181 | |
Reagent Reservoir | Fisher Scientific | 21-381-27E | |
Refrigerated centrifuge | Beckman Coulter | 367501 | |
RNase Inhibitor | Promega | N2515 or N2615 | |
ROX reference dye | Life Technologies | 12223 | |
SuperScript II or III Reverse Transcriptase | Life Technologies | 18064-022 or 18080044 | |
Surgical gloves | Fisher Scientific | 19-058-800 | |
Thermal cycler | Life Technologies | 4314879 | |
Trisma base | Sigma Aldrich | T1503 | |
Vivaspin 20 centrifugal concentrator units | Sartorius-Stedim | VS2022 |