from Sewage and Ocean Water by Triplex Reverse - Europe PMC

Vol. 60, No. 7

APPLIED AND ENVIRONMENTAL MICROBIOLOGY, JUly 1994, p. 2400-2407

0099-2240/94/$04.00+0 Copyright © 1994, American Society for Microbiology

Detection of Poliovirus, Hepatitis A Virus, and Rotavirus from Sewage and Ocean Water by Triplex Reverse Transcriptase PCR YU-LI TSAI,* BICH TRAN, LOUIS R. SANGERMANO, AND CAROL J. PALMER Environmental Sciences Laboratory, County Sanitation Districts of Orange County, Fountain Valley, Califomia 92728 Received 31 January 1994/Accepted 6 May 1994

A triplex reverse transcriptase PCR (RT-PCR) was developed to simultaneously detect poliovirus, hepatitis A virus (HAV), and rotavirus in sewage and ocean water. Sewage and ocean water samples seeded with the three different viruses were concentrated by ultrafiltration. The unseeded ocean water and sewage samples were concentrated by vortex flow filtration and/or ultrafiltration. Random hexamers and a rotavirus downstream primer were used to initiate reverse transcription. Three different sets of primers specific for poliovirus, HAV, and rotavirus cDNAs were mixed in the PCR mixture to amplify the target DNA. Three distinct amplified DNA products representing poliovirus, HAV, and rotavirus were identified by gel electrophoresis as 394-, 192-, and 278-bp sequences, respectively. Dot blot and Southern analyses were used to confirm the amplified products for each virus present in the environmental samples. Except for poliovirus, the sensitivity of triplex RT-PCR for the detection of rotavirus and HAV was found to be similar to that of monoplex RT-PCR, which uses only one set of primers to amplify a single type of virus. The triplex RT-PCR has greater advantages over monoplex RT-PCR for virus detection, namely, the rapid turnaround time and cost effectiveness.

PCR has been used to detect pathogenic microorganisms in different environments and it could be used as an analytical tool to indicate the potential of disease outbreaks. Conventional reverse transcriptase PCR (RT-PCR) is monoplex and includes only one specific set of primers in the reaction mixture and can detect only one virus or one target RNA sequence in a sample (1, 3, 9, 12, 15, 22, 26). Multiplex PCR is the simultaneous PCR amplification of gene sequences associated with different organisms or different genes within the same organism. This technique has been used to detect dual genes from Escherichia coli or Legionella pneumophila (4) in water and to detect different Salmonella spp. in soil and water (29). In this study, we have developed a triplex RT-PCR method, which uses three different sets of primers in one reaction mixture, to detect poliovirus, HAV, and rotavirus simultaneously in sewage and ocean water. Because of the time savings in time and cost-effectiveness, the triplex RT-PCR method provides a more efficient way to detect various viral RNAs in environmental samples than the monoplex RT-PCR method.

Enteric viruses have been associated with many outbreaks of waterborne nonbacterial gastroenteritis (16, 18-20, 28) and are an important public health concern. It has been reported that at least 37 different human viruses have been isolated from drinking water all around the world (11). The traditional detection of viruses in water, including monitoring treatment removal efficiency, is done by labor-intensive tissue culture methodology (2). The problems with tissue culture methods are well-known and include the lack of susceptible cell lines for many important waterborne viruses (such as Norwalk virus), lack of sensitivity, lengthy analysis time (up to 6 weeks), and problems in detecting low virus numbers, which is the typical situation in environmental water samples. Faster and more sensitive technology is now available for the detection of viruses in both environmental and clinical samples. The use of molecular techniques, such as PCR and nucleic acid hybridization, for the detection of enteric viruses in environmental samples has been well documented in recent years (1, 3, 6, 8, 14, 15, 17, 26, 27, 30). Another rapid method, developed by Graff et al. (13), uses an antigen capture PCR technique to detect hepatitis A virus (HAV) from treated and untreated sewage sludge samples. These new methods offer several advantages over traditional tissue culture. These advantages include rapid turnaround time and a high degree of sensitivity. Molecular techniques do not require tissue culture or cell line maintenance. Through enzymatic amplification, the PCR is capable of detecting specific genes from one viable bacterial cell containing 10-15 g of DNA in 100 ml of water (5) and the viral genomic RNA in diluents containing less than 1 PFU (9, 26). Because the PCR does not require cell cultivation, it has been effectively used to detect viruses which are difficult to cultivate or are nonculturable (8, 14, 26, 27), such as HAV and Norwalk virus from sewage or human stools. The

MATERUILS AND METHODS

Samples. Ocean water samples were collected from both coastal and offshore waters in sterile 20-liter polypropylene containers. Sewage samples from primary influent and secondary effluent were collected in 250-ml sterile polypropylene containers. All samples were stored on ice. For virus detection, the ocean water samples (15 liters) and sewage samples (100 ml) were concentrated within 4 h after collection by vortex flow filtration and ultrafiltration as described previously (26). Virus strains. Poliovirus type 1 strain LSc was maintained in Buffalo green monkey kidney cells, and HAV strain HM175 was maintained on BS-C-1 cells (an African green monkey kidney-derived cell line) and/or FRhK-4 cells (fetal rhesus kidney-derived cell line). Two human rotaviruses (VR-2018 strain Wa and VR-970 strain D) were purchased from American Type Culture Collection, Rockville, Md. One human

* Corresponding author. Mailing address: Environmental Sciences Laboratory, County Sanitation Districts of Orange County, 10844 Ellis Ave., Fountain Valley, CA 92728. Phone: (714) 962-2411. Fax: (714) 962-2591.

2400

VOL. 60, 1994

rotavirus provided by National Institutes of Health and one animal strain rotavirus SA-1 1 were used for the determination of primer specificity. All rotaviruses were propagated on the MA-104 cell lines. Virus seeding. A virus mixture (30,ul) containing 1.0 x 104 PFU of HAV strain HM175, 6.0 x 104 PFU of poliovirus type 1 strain LSc, and 1.6 x 104 PFU of rotavirus strain Wa (ATCC VR-2018) was seeded into 100-ml portions of filtered and nonfiltered samples. The filtered samples, including both ocean water and secondary sewage effluent, were filter sterilized with 0.2-,um-pore-size Nalgene disposable filterware (Nalge Co., Rochester, N.Y.). The seeded and unseeded samples (100-ml portions) were concentrated by ultrafiltration with Centriprep-100 and Centricon-100 (Amicon, Inc., Beverly, Mass.) as described previously (26). The final concentrates (2,ul) were used as templates for the triplex RT-PCR. The remaining concentrates were stored at -20°C for future analysis. Triplex RT-PCR. The triplex RT-PCR was performed with a GeneAmp PCR System 9600 (Perkin-Elmer, Norwalk, Conn.). All the reagents required for RT-PCR were included in a GeneAmp RNA PCR kit (Perkin-Elmer). Each RT reaction mixture (16,ul) contained the following: 4 RI of 25 mM MgCl2; 2 RI of 1Ox PCR buffer 11 (500 mM KCl, 100 mM Tris-HCl, pH 8.3); 2 RI each of 10 mM dGTP, 10 mM dATP, 10 mM dTTP, and 10 mM dCTP; 1 ,ul of 50 ,uM random hexamers; and 1 RI of 10 ,uM End9 primer (12), 5'-GGTCACATCATA CAATTCTAATC TAAG-3'. Two microliters of concentrated sample was added to the RT reaction mixture and subsequently heated at 99°C for 5 min and cooled at 4°C for 5 min before the addition of 1 [I of RT (50 U/pA) and 1 pA of RNase inhibitor (20 U/,ul). The reverse transcription was completed after the following steps: 25°C for 10 min, 42°C for 30 min and 4°C for 5 min. The PCR mixture (80 pA) containing AmpliTaq DNA polymerase and three sets of oligonucleotide primers was added to the finished RT mixture (20 pI). Except for the primers, all other components used in the PCR process were kept at optimal concentrations as suggested by the manufacturer (Perkin-Elmer). Primers for poliovirus 5' noncoding region (3, 10, 24) (Polio-R, 5'-ACGGACACCCAAAGTA-3'; Polio-L, 5'-AGCACTTCTGTTTCCC-3'), for HAV capsid region (21) (HAVC-R, 5'-CTCCAGAATCATCTCCAAC-3'; HAVC-L, 5'-CAGCACATCAGAAAGGTGAG-3') and for a segment of rotavirus gene encoding the major outer capsid glycoprotein vp7 (End9 and Rota785 [5'-TTCGAAATTGTAAGAAATTAG-3']) were used to amplify 394-, 192-, and 278-bp sequences, respectively. The final concentrations for each set of primers in the reaction mixture were 50 nM. The PCR was carried out by the following protocol: initial denaturation step at 95°C for 2 min; 40 cycles, with 1 cycle consisting of 1 min at 95°C, 1 min at 55°C, and 1 min at 72°C; and final extension step at 72°C for 7 min. The PCR products were stored at 4°C before analysis. The internal oligonucleotide probes for poliovirus, HAV, and rotavirus PCR products were POLIO-IN (5'-ACATAAGAATCCTCCGGCCCCTGA3'), HAVC-IN (5'-7TfGCITCCFC-I-lTlATCATGCTAT-3') (9), and RTPB858 (5'-CATAACAGCAGATCCAACAAC-3'), respectively. All oligonucleotides were synthesized by a DNA/ RNA synthesizer (model 392; Perkin-Elmer, Applied Biosystems Division, Foster City, Calif.). The RT-PCR internal probes, POLIO-IN and RTPB858, and the PCR primer, Rota785, were designed by using OLIGO 4.1 computer software (National Biosciences, Plymouth, Minn.). DNA hybridization on PCR products. The amplified PCR products were identified under UV irradiation by electrophoresis on a 2% SeaKem GTG agarose gel (FMC BioProd-

TRIPLEX RT-PCR ON SEWAGE AND OCEAN WATER

2401

ucts, Rockland, Maine) stained with ethidium bromide (0.5 ,ug/ml). The amplified DNA was transferred onto Hybond-N+ positively charged nylon membranes (Amersham, Arlington Heights, Ill.) by Southern blotting (23) or dot blotting (7). A PosiBlot pressure blotter (Stratagene, La Jolla, Calif.) and a Minifold I dot blotter (Schleicher & Schuell, Keene, N.H.) were used for Southern blotting and dot blotting, respectively. All three PCR internal probes were nonradioactively labeled by using a Genius 5 DNA labeling kit (Boehringer Mannheim, Indianapolis, Ind.) by the protocols suggested by the manufacturer. The DNA hybridization on amplified target fragments and chemiluminescence detection were performed as described previously (25). The triplex PCR products on the membrane were first probed with HAVC-IN. The membrane was then boiled in deionized water for 30 min to strip off the old probe and reprobed with POLIO-IN. Last, the membrane was stripped off again and reprobed with RTPB858. The probes were tested for cross-reactivity with the triplex PCR products by dot blot analysis. Application of triplex RT-PCR on sewage and ocean water concentrates. The undiluted and serially diluted sewage and ocean water concentrates were used as templates in the triplex RT-PCR. Positive controls, including a virus mixture (poliovirus, HAV, and rotavirus), each type of virus, and a RNA fragment (a plasmid pAW109 transcript) from the RT-PCR kit (Perkin Elmer), were employed to confirm the amplification results. A negative control containing no template was used to ensure that carryover contamination did not occur in the reaction. In a separate experiment, serially diluted virus mixtures containing poliovirus, rotavirus, and HAV ranging from 2 x 102 to 2 x 105 PFU/ml were seeded into 100-ml portions of target virus-free primary influent and were concentrated to 200-pul for triplex RT-PCR and monoplex RT-PCR. The sensitivity of detection for these viruses present in primary influent was determined by dot blot analysis. RESULTS Triplex RT-PCR with various primer concentrations. Figure 1 shows the effects of multiple primer concentrations on the amplification efficiencies of different viral RNAs during the triplex RT-PCR. The amplified DNA products were 394, 278, and 192 bp long, which correspond to the sizes of the cDNA target fragments transcribed from viral RNAs of poliovirus, rotavirus, and HAV, respectively. The estimated initial virus numbers in each reaction mixture were 1,200, 200, and 320 PFU for poliovirus, HAV, and rotavirus, respectively. The composite primer concentrations for three different viruses used in the triplex RT-PCR are listed in Table 1. When each set of primers was initially maintained at a final concentration of 50 nM in the reaction mixture, the best amplification result for all three viruses was obtained (Fig. 1, lane 4; Table 1). Although the reduction of poliovirus and HAV primer concentrations from 50 nM to 30 nM also gave a similar result (Fig. 1, lane 10; Table 1), primer concentrations of 50 nM each were chosen for triplex RT-PCR throughout this study because of the simplicity of preparation. In addition, when the rotavirus primer concentration was decreased to 30 nM and the other two sets of primers were maintained at either 30 or 50 nM, significant reduction of rotavirus RT-PCR products was found (Fig. 1, lanes 5, 8, and 9; Table 1). Moreover, when the concentrations of one or two primer pairs were reduced to 10 nM, no amplified products from corresponding primers were observed (Fig. 1, lanes 11 to 16; Table 1). No cross-reactivity was found among the three primer sets for the target DNA.

2402

TSAI ET AL.

APPL. ENVIRON. MICROBIOL.

tion (1 to 10 PFU) during RT-PCR, the band resulting from amplification of that virus was much less intense in comparison to band intensity from the two viruses present in higher concentrations (10 to 1,000 PFU). The dot blot analysis was applied to verify the amplified products. Cross-reactivity and sensitivity. Figure 2 shows the specificity of each set of primers used in the triplex RT-PCR. Each set of primers amplified only one virus type (Fig. 2, lanes 1 to 9) and no cross-reactivity was found between the individual primers and nontarget virus during monoplex RT-PCR. This confirmed that each fragment found in the triplex RT-PCR was amplified from only one type of virus (Fig. 2, lane 1 ). Figure 3 is the dot blot analysis on the comparisL n of detection sensitivity between triplex RT-PCR and monoplex RT-PCR. Table 2 exhibits the detailed variable components of 1112131415161718 bp 20 21M RT-PCR and indicates which products were used as dot blot samples. The detection sensitivity of triplex RT-PCR was similar (3.2 PFU) to that of monoplex RT-PCR for the detection of rotavirus when only rotavirus was present at very I ~~~~~~~~~~~~~~-700 low numbers in the reaction mixture. However, when all three -400 2978 _ virus types were present, the triplex RT-PCR could detect 1 922300 rotavirus at a level of 0.32 PFU (Fig. 3, dots C4, F4, and 14). J ~~~~~~~~~~~~~~-100 Monoplex RT-PCR showed better sensitivity (0.12 PFU; Fig. 50 3, dot A4) than triplex RT-PCR did (1.2 PFU; Fig. 3, dots B3 and C3) in the detection of poliovirus. A similar sensitivity (0.2 FIG. 1. Effects of primer concentrations on the triplex RT-PCR for PFU) was obtained for both monoplex (dot G7) and triplex the detection of poliovirus, rotavirus, and HAV viral RNAs. The (dots H7 and 13) RT-PCR in the detection of HAV. Significant RT-PCR products were analyzed on a 2% agarose gel stained with cross-reactivity was observed between HAVC-IN and polioviethidium bromide by gel electrophoresis. Samples contained in lanes 1 rus RT-PCR products. POLIO-IN did not exhibit cross-reacto 21 are exhibited in Table 1. Lane M contains molecular size tivity against rotavirus and HAV RT-PCR products, and standards (BioMarker Low; BioVentures, Inc., Murfreesboro, Tenn.).

The yields of amplified products, based on the intensity of each fragment on the ethidium bromide-stained agarose gel, were similar by monoplex and triplex RT-PCR (Fig. 1, lanes 1 to 3 and 17 to 19; Table 1). RT-PCR product was not observed when template RNA was not present in the reaction mixture (Fig. 1, lane 21). The optimal primer concentration (50 nM) was tested against different ratios of each virus by triplex RT-PCR and monoplex RT-PCR. It was found in both cases that regardless of the ratio, if the concentration of each target virus were above 10 PFU, similar intensities of each amplified product were observed on an agarose gel (data not shown). However, if one virus strain was present at a lower concentra-

1

2

3 4

5

6 7

8

9

19

-

TABLE 1. Viral particles and final concentrations of RT primers and PCR primers used in the triplex RT-PCR reaction mixture Lane no.'

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21

Templateb P H R P+H+R P+H+R P+H+R P+H+R P+H+R P+H+R P+H+R P+H+R P+H+R P+H+R P+H+R P+H+R P+H+R P H R RNA None

Final concn (nM) of RT primer

Final concn (nM) of PCR primer

RH"

Ed

Poliovirus"

HAVf

Rotavirue

500 500 500 500 500 500 500 500 500 500 500 500 500 500 500 500 500 500 0 500 500

100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 0 0 100 0 100

50 50 50 50 50 50 30 50 30 30 50 50 10 50 10 10 50 0 0 0 50

50 50 50 50 50 30 50 30 50 30 50 10 50 10 50 10 0 50 0 0 50

50 50 50 50 30 50 50 30 30 50 10 50 50 10 10 50 0 0 50 0 50

IRh

0

0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 150 0

" The lane numbers correspond to the lanes in the ethidium bromide-stained agarose gel shown in Fig. 1. b Abbreviations: P, poliovirus type strain 1 LSc; H, HAV strain HM175, R, rotavirus strain Wa (ATCC VR-2018); P+H+R, a mixture of poliovirus (1,200 PFU), HAV (200 PFU), and rotavirus (320 PFU). C Random hexamers from a GeneAmp RNA PCR kit (Perkin-Elmer). d End9 primer for the 3'-end downstream rotavirus primer (12). ' Polio-R and Polio-L primers (3). f HAVC-R and HAVC-L primers (9). g Rota785 and End9 primers. h DM152 and DM151 interleukin-1 primers for reverse transcription and PCR of positive-control RNA (Perkin Elmer).

VOL. 60, 1994

FIG. 2. Analysis of primer specificity on target viruses in RT-PCR by gel electrophoresis. Primers included in each reaction mixture are rotavirus primers (Rota785 and End9) (lanes I to 3), HAV primers (HAVC-L and HAVC-R) (lanes 4 to 6), poliovirus primers (Polio-L and Polio-R) (lanes 7 to 9), and all three sets of primers (lane 10). The virus templates contained in each reaction are rotavirus (300 PFU [lanes 3, 5, and 7]), HAV (200 PFU [lanes 1, 6, and 8]), poliovirus (1,200 PFU [lanes 2, 4, and 9]), all three viruses (lane 10), RNA RT-PCR control from a RNA PCR kit (Perkin Elmer) (lane 11), and negative control containing no template (lane 12). Lane M contains molecular size standards (BioMarker Low; BioVentures, Inc.). The numbers to the left of the gel indicate the number of base pairs.

RTPB858 did not exhibit cross-reactivity against poliovirus and HAV RT-PCR products. Triplex RT-PCR on sewage and ocean samples. Figure 4 illustrates the application of triplex RT-PCR on ocean water and sewage. Poliovirus, HAV, and rotavirus were not detected in coastal water (Fig. 4, lanes 1 to 4). An amplified DNA fragment similar in size to the positive HAV product (lane 6) was found in ocean water 30 m deep around a sewage outfall area, but both rotavirus and poliovirus were not found (Fig. 4, lanes 5 to 8). Poliovirus, HAV, and rotavirus were not detected in filtered ocean water (lane 9), but suspicious HAV RNA was observed in unfiltered secondary effluent (lane 10). Both of the suspicious HAV PCR fragments found in outfall deep-ocean water and secondary effluent were probed with HAV internal probe and they were all negative (Fig. 4B). However, when unfiltered secondary effluent and filtered ocean water were seeded with poliovirus, HAV, and rotavirus, the expected amplified product was observed, as shown in lanes 11 and 12 (Fig. 4A). They were all positive when hybridized with the respective internal probes (Fig. 4B). The amplified rotavirus target fragments in the seeded samples were not clearly demonstrated in the agarose gel, but they hybridized with the RTPB858 internal probe after Southern blotting (Fig. 4B). Except for positive controls, none of the three viruses were detected in filtered ocean water and secondary effluent by monoplex RT-PCR (lanes 14 to 22). However, more background RNA or DNA from microflora in the ocean water and secondary sewage was amplified by triplex RT-PCR than by monoplex RT-PCR. This result suggested that the method (26) used in concentrating viruses from environmental water samples was successful in removing substances in the samples that could cause inhibition of the PCR. Figure 4B is a composite autoradiogram, showing three images of the Southern analysis when the same membrane was probed with three different


2403

probes separately. Each internal probe positively identified its triplex RT-PCR products amplified from seeded water samples and respective positive control. Unlike dot blot analysis (Fig. 3, dots GI to G4 and HI to H4) no cross-reactivity between HAVC-IN and poliovirus RT-PCR products was observed on the Southern analysis (Fig. 4B). The rotavirus amplified products were dual-sized fragments on Southern blots but only a single-sized fragment was observed on agarose gel (Fig. 4, lanes 11 to 13 and 16). Figure 5 shows additional application of the triplex RT-PCR procedure to two primary influent and two secondary effluent sewage samples. In this experiment, 15 liters (instead of 100 ml) of sewage sample was collected and further concentrated by vortex flow filtration and ultrafiltration. Several distinct PCR products amplified from background organisms were observed on an agarose gel. Poliovirus, rotavirus, and HAV RNAs were detected in one undiluted and 10-fold-diluted primary influent concentrate (Fig. 5, lanes 5 and 6), but they were not detected in the 100- or 1,000-fold dilutions. Only HAV RNA was detected in the second primary influent sample (Fig. 5, lane 1). The secondary effluent concentrates were negative (Fig. 5, lanes 9 to 16). These results were confirmed with subsequent Southern blot analysis (data not shown). Table 3 shows the dot blot hybridization results using both triplex and monoplex RT-PCR to amplify various concentrations of poliovirus, rotavirus, and HAV from seeded primary influent samples. Triplex RT-PCR detected viral RNAs equivalent to 10 PFU of poliovirus, rotavirus, and HAV seeded into 100 ml of the primary influent. However, except for rotavirus and HAV, monoplex RT-PCR was able to detect poliovirus levels as low as 1 PFU in the seeded primary influent. DISCUSSION This study has developed a rapid and efficient method to simultaneously detect three medically important viruses commonly transmitted via water and/or shellfish. The efficiency of triplex RT-PCR was carefully controlled by the concentration of each set of primers involved in the reaction. An optimal primer concentration of 50 nM for each primer was determined as yielding the best results for the triplex RT-PCR protocol for the detection of poliovirus, HAV, and rotavirus. Even when the viruses were present at the levels detectable by RT-PCR (200 to 1,200 PFU), no PCR products of target viruses could be found if primer concentrations were decreased to 10 nM. This result indicates that the lower limit of the primer concentration was 10 nM in the triplex RT-PCR. Conversely, when 10-fold-concentrated primers (500 nM) were used in the reaction mixture, primers-dimers appeared and only rotavirus RNA was amplified in the triplex reaction (data not shown). Therefore, these results suggest that the optimization of primer concentration is critical to determine the efficacy of triplex or multiplex PCR on the detection of the target organisms in a mixed sample. Under optimized primer concentration and fixed virus number (10' PFU), no difference in amplification efficiency was found by triplex or monoplex RT-PCR for detection of each virus type. Moreover, the amount of each virus did not affect the amplification of the other two viruses at the optimal primer concentration. In addition to random hexamers, the End9 primer was required in reverse transcription to obtain successfully triplex RT-PCR results. If the End9 oligonucleotide (downstream primer) was not present in the RT reaction mixture, insufficient amplification of rotavirus RNA was observed (data not shown). Therefore, the End9 primer was critical to the success of the RT reaction for the triplex PCR.


TSAI ET AL.

2404

FIG. 3. Dot blot analysis of sensitivity test on triplex RT-PCR and monoplex RT-PCR. The variable components used in both triplex and monoplex RT-PCR are listed in Table 2. The same Southern blottransferred membrane was probed with each of three different RTPCR internal probes in separate hybridizations. The hybridization signals shown in each row are POLIO-IN for rows A, B, and C, RTPB858 for rows D, E, and F, and HAVC-IN for rows G, H, and I.

The primer sets used in this study have proved to be very specific in the triplex PCR reaction. No cross-reactivity between the virus primers was observed. The published primers (HAVC-R, HAVC-L, End9, Polio-R, and Polio-L), which were used in the triplex RT-PCR, have been screened against other virus types and were determined to be type specific by other investigators (3, 8, 12). Therefore, it was feasible to use these primers in the triplex RT-PCR to specifically detect poliovirus, HAV, or rotavirus in environmental samples. The rotavirus primers used in this study were specific to rotaviruses TABLE 2. Major components used in the triplex and monoplex RT-PCR for the dot blot analysis. Dot number shown in this table corresponds to the dot blot analysis in Fig. 3 Dot

no."

Al, Dl, Gl A2, D2, G2 A3, D3, G3 A4, D4, G4 A5, D5, G5 A6, D6, G6 A7, D7, G7 A8, D8, G8 A9, D9, G9 AIO, D10, G10 All, Dll, Gll A12, D12, G12 Bi, El, HI B2, E2, H2 B3, E3, H3 B4, E4, H4 B5, E5, H5 B6, E6, H6 B7, E7, H7 B8, E8, H8 B9, E9, H9 B10, EIO, H10

Bl, Ell, H1l B12, E12, H12

Cl, Fl, II C2, F2, 12 C3, F3, 13 C4, F4, 14

C5, F5, I5 C6, F6, I6

TemplateO P P P P H H H H R R R R P P P P H H H H R R R R P H R P H R P H R P H R RNA NONE

RT

PCR

primer"

primee

120 12 1.2 0.12 20 2 0.2 0.02

RH+E RH+E RH+E RH+E RH+E RH+E RH+E RH+E RH+E RH+E RH+E RH+E RH+E RH+E RH+E RH+E RH+E RH+E RH+E RH+E RH+E RH+E RH+E RH+E RH+E

Ppr Ppr Ppr Ppr Hpr Hpr Hpr Hpr Rpr Rpr Rpr Rpr Ppr+Hpr+Rpr Ppr+Hpr+Rpr Ppr+Hpr+Rpr Ppr+Hpr+Rpr Ppr+Hpr+Rpr Ppr+Hpr+Rpr Ppr+Hpr+Rpr Ppr+Hpr+Rpr Ppr+Hpr+Rpr Ppr+Hpr+Rpr Ppr+Hpr+Rpr Ppr+Hpr+Rpr Ppr+Hpr+Rpr

M M M M M M M M M M M M T T T T T T T T T T T T T

RH+E

Ppr+Hpr+Rpr

T

RH+E

Ppr+Hpr+Rpr

T

RH+E

Ppr+Hpr+Rpr

T

RH RH+E

DM151+DM152 Ppr+Hpr+Rpr

T

3,200 320 32 3.2 120 12 1.2 0.12 20 2 0.2 0.02 3,200 320 32 3.2 120 20 320 12 2 32 1.2 0.2 3.2 0.12

0.02 0.32

a The dot numbers correspond to the dot blot analysis shown in Fig. 3. b p, poliovirus type 1 strain LSc; H, HAV strain HM175; R, rotavirus strain Wa (ATCC VR-2018). RH, random hexamers from a GeneAmp RNA PCR kit (Perkin-Elmer); E, End9 primer for the 3'-end downstream rotavirus primer (12). dPpr, Polio-R and Polio-L primers; Hpr, HAVC-R and HAVC-L primers; Rpr, Rota785 and End primers.

'

eM, monoplex; T, triplex.

RT-PCR TyPee

PFU

M

VOL. 60, 1994


2405

A.

394278' 192-

B. 2

3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 M

FIG. 4. Detection of poliovirus, rotavirus, and HAV on sewage and water samples by triplex RT-PCR and monoplex RT-PCR. Lanes: 1, coastal water concentrate (CWC), undiluted; 2, 10-folddiluted CWC; 3, 100-fold-diluted CWC; 4, 1,000-fold-diluted CWC; 5, outfall-depth ocean water concentrate (ODWC), undiluted; 6, 10-folddiluted ODWC; 7, 100-fold-diluted ODWC; 8, 1,000-fold-diluted ODWC; 9, filtered ocean water concentrate, undiluted; 10, secondary effluent concentrate, undiluted; 11, concentrate from seeded (poliovirus [1,200 PFU], rotavirus [320 PFU1, HAV [200 PFU]) filtered ocean water, undiluted; 12, concentrate from seeded (poliovirus [1,200 PFU], rotavirus [320 PFU], HAV [200 PFU]) secondary effluent, undiluted; 13, positive control containing poliovirus (1,200 PFU), rotavirus (320 PFU), and HAV (200 PFU); 14, 17, and 20, 10-fold-diluted CWC; 15, 18, and 21, 10-fold-diluted ODWC; 16, positive rotavirus control (320 PFU); 19, positive HAV control (200 PFU); 22, positive poliovirus control (1,200 PFU); 23, RNA control from a RNA PCR kit (Perkin Elmer); 24, negative control containing no template; M, molecular size standards (BioMarker Low, BioVentures, Inc.). Lanes 1 to 13 are results from triplex RT-PCR, and lanes 14 to 24 are the results from monoplex RT-PCR (lanes 14 to 16, rotavirus; lanes 17 to 19, HAV; lanes 20 to 22, poliovirus). (A) Triplex and monoplex RT-PCR products on an ethidium bromide-stained 2% Seakem agarose gel. (B) Autoradiogram of the panel A gel produced by Southern analysis against poliovirus, rotavirus, or HAV internal probe and subsequent chemiluminescence detection. ocean

of human origin. Another rotavirus primer (Pan-VG9) (21) in combination with End9 was able to amplify the viral RNA from both human and animal rotavirus strains, such as strain SA-11. However, because the amplified product (208 bp) was too close to the size of HAV product (192 bp), it became indistinguishable from the HAV products on the agarose gel. Therefore, the Rota785 primer was designed and used in the triplex RT-PCR to produce a distinct rotavirus-specific fragment (278 bp). The dot blot analysis indicated that monoplex RT-PCR was 10 times more sensitive than triplex RT-PCR in detection of poliovirus but monoplex and triplex RT-PCR have similar sensitivities when only rotavirus or HAV was present in the

FIG. 5. PCR analysis of triplex RT-PCR on primary influent and secondary effluent sewage samples for detection of poliovirus, rotavirus, and HAV on an ethidium bromide-stained agarose gel. Lanes: 1, primary influent concentrate 1 (PIC1), undiluted; 2, 10-fold-diluted PIC1; 3, 100-fold-diluted PIC1; 4, 1,000-fold-diluted PIC1; 5, primary influent concentrate 2 (PIC2), undiluted; 6, 10-fold-diluted PIC2; 7, 100-fold-diluted PIC2; 8, 1,000-fold-diluted PIC2; 9, secondary effluent concentrate 1 (SEC1), undiluted; 10, 10-fold-diluted SEC1; 11, 100fold-diluted SEC1; 12, 1,000-fold-diluted SEC1; 13, secondary effluent concentrate 2 (SEC2), undiluted; 14, 10-fold-diluted SEC2; 15, 100fold-diluted SEC2; 16, 1,000-fold-diluted SEC2; 17, positive controls containing 200 PFU each of poliovirus, rotavirus, and HAV; 18, RNA control from a RNA PCR kit (Perkin Elmer); 19, negative control containing no template; M, molecular size standards (BioMarker Low; BioVentures, Inc.). The numbers to the right of the gel are in base pairs.

reaction mixture. The triplex RT-PCR was able to detect three types of viruses when each virus was present at levels less than 1 PFU. The multiple primers present in the PCR mixture could cause competition between primers for the target cDNA and thus affect the detection sensitivity, especially when the cDNA is present at very low concentrations. This could also impact the detection sensitivity on the environmental water samples, because only low numbers of viruses are present in these samples. Therefore, steps involving the concentration of a large volume of water are required to concentrate viruses before the application of triplex RT-PCR. The HAVC-IN probe slightly cross-reacted with the poliovirus product in the dot blot analysis, but this cross-reactivity was not found in the Southern analysis. The HAVC-IN probe did confirm the correct HAV fragment in the triplex RT-PCR. Experimental results shown in Fig. 2 indicated that each set of primers used in this study was very specific to one virus type, and hence the cross-reactivity found in the dot blot analysis was not due to nonspecific primer annealing during the RT-PCR. Because the amplified poliovirus PCR product was blotted onto a very limited area (13 mm2) in the dot blot analysis, this could increase the chances of a cross-reaction between HAVC-IN and poliovirus PCR product. In order to solve this problem, a more stringent wash condition, raising the low-salt wash temperature to 55°C, was performed for the HAVC-IN, but it reduced the detection sensitivity for HAV. Therefore, the cross-reaction might be due to nonspecific hybridization between probe and nontarget DNA, which was commonly found in the nucleic acid hybridization. The viruses used for seeding were stored at -20°C. Viruses were propagated on each cell line and counted before they were stored. The die-off rate after storage was not determined. Therefore, the PFU on the

2406


TSAI ET AL.

TABLE 3. Triplex and Monoplex RT-PCR and dot blot analyses on a primary influent sewage seeded with different densities of poliovirus, rotavirus, and HAV' Result' of dot blot hybridization and RT-PCR

Seeded PFU of each virus'

2x105(1,000) 2x104(100) 2 x 103 (10) 2 x 102 (1)

Triplex

Monoplex

_

Poliovirus

* Rotavirus

HAV

Poliovirus

Rotavirus

HAV

+ + +

+ + +

+ + +

+ + +

+ + +

-

-

+ + + +

-

0 a The primary influent sewage was tested for poliovirus, rotavirus, and HAV and found to be free of these virus before the seeding. Only 1/100th of the aliquot (2 ,ul) of the final concentrate (200 ,ul) was used for RT-PCR. The value shown in parentheses are equivalent to the number of actual

b

viral particles of each virus being amplified. c +, dot plot probed positive with RT-PCR internal probe; -, dot plot probed negative with RT-PCR internal probe.

sensitivity experiment was estimated from the initial PFU. The actual PFU used in the sensitivity test could be less. The application of triplex RT-PCR on some concentrated environmental samples showed positive results for the target viruses. Although amplified background DNA was observed in both deep-sea water and secondary effluent sewage samples, it was negative for poliovirus, HAV, or rotavirus after Southern analysis. This result underscores the importance of additional testing by Southern blotting to confirm the correct amplification products. The amplified rotavirus target fragments showed two bands on the Southern analysis but only one band (278 bp) on an ethidium-stained agarose gel. This observation indicated that a smaller fragment homologous to the rotavirus internal probe was produced during the triplex reaction. However, the smaller fragment, which could have resulted from incomplete primer extension, did not affect the interpretation of the results. Because there are many unidentified substances present in the environmental samples, these impurities could interfere with the RT-PCR during the amplification process. The detection sensitivity of both triplex and monoplex RTPCR on seeded primary influent was lower than that on seeded virus samples (Table 3, Fig. 3). In a previous report (27), we have detected enteroviruses and HAV in primary and secondary effluent samples using monoplex RT-PCR. In addition, there are other reports in the literature that document the detection of viral RNA in groundwater (1), sewage sludge (13), surface waters (15), and river water (9a) by monoplex RTPCR. In this current study, the application of a triplex RT-PCR amplification method was clearly an improvement over previously reported monoplex RT-PCR methods, since analysis time was greatly reduced and three types of virus could be detected with one amplification step. In conclusion, this study focused primarily on the development and application of triplex RT-PCR for the detection of poliovirus, HAV, and rotavirus from environmental water samples. The success of this method is an improvement over current monoplex RT-PCR because it provides a more rapid and efficient way to detect these three medically important viruses. This sensitive technology could help researchers to investigate the presence of virus in environmental samples and could also be of benefit in clinical diagnoses. ACKNOWLEDGMENTS This study is based upon work supported by the National Water Research Institute under awards WQ-91-01 and HR-92-06. Matching fund supports were received from County Sanitation Districts of Orange County. We thank Mark Sobsey for providing HAV strain HM175 and Moy Yahya and Marylin Koval for cultivating poliovirus type 1 and rotavirus

SA-11. We are grateful to G. Fred Bonilla for sample collection and concentration and to Patrick McNelly for editing the manuscript. REFERENCES 1. Abbaszadegan, M., M. S. Huber, C. P. Gerba, and I. L. Pepper. 1993. Detection of enteroviruses in groundwater with the polymerase chain reaction. Appl. Environ. Microbiol. 59:1318-1324. 2. American Public Health Association. 1992. Standard methods for the examination of water and wastewater, 18th ed. American Public Health Association, Washington, D.C. 3. Atmar, R. L., T. G. Metcalf, F. H. Neill, and M. K. Estes. 1993. Detection of enteric viruses in oysters by using the polymerase chain reaction. Appl. Environ. Microbiol. 59:631-635. 4. Bej, A. K., M. H. Mahbubani, R. Miller, J. L. DiCesare, L. Hal, and R. M. Atlas. 1990. Multiplex PCR amplification and immobilized capture probes for detection of bacterial pathogens and indicators in water. Mol. Cell. Probes 4:353-365. 5. Bej, A. K., R. J. Steffan, J. DiCesare, L. Haff, and R. M. Atlas. 1990. Detection of coliform bacteria in water by polymerase chain reaction and gene probes. Appl. Environ. Microbiol. 56:307-314. 6. Chapman, N. M., S. Tracy, C. J. Gauntt, and U. Fortmueller. 1990. Molecular detection and identification of enteroviruses using enzymatic amplification and nucleic acid hybridization. J. Clin. Microbiol. 28:843-850. 7. Davis, L. G., M. D. Dibner, and J. F. Battey. 1986. Basic methods in molecular biology, p. 147-149. Elsevier Science Publishing Co., New York. 8. De Leon, R., S. M. Matsui, R. S. Baric, J. E. Herrmann, N. R. Blacklow, H. B. Greenberg, and M. D. Sobsey. 1992. Detection of Norwalk virus in stool specimens by reverse transcriptase-polymerase chain reaction and nonradioactive oligoprobes. J. Clin. Microbiol. 30:3151-3157. 9. De Leon, R., C. Shieh, R. S. Baric, and M. D. Sobsey. 1990. Detection of enteroviruses and hepatitis A virus in environmental samples by gene probes and polymerase chain reaction, p. 833853. In Advances in water analysis and treatment. Proceedings of the Water Quality Technology Conference, San Diego, Calif. American Water Works Association, Denver. 9a.Divizia, M., G. Morace, R. Gabrieli, G. Pisani, and A. Pana. 1993. Application of the PCR technique to the detection of hepatitis-A virus in the environment. Water Sci. Technol. 27:223-225. 10. Gama, R. E., P. R. Horsnell, P. J. Hughes, C. North, C. B. Bruce, W. Al-Nakib, and G. Stanway. 1989. Amplification of rhinovirus specific nucleic acids from clinical samples using the polymerase chain reaction. J. Med. Virol. 28:73-77. 11. Gerba, C. P., and J. B. Rose. 1990. Viruses in source and drinking water, p. 380-395. In Drinking water microbiology. SpringerVerlag, New York. 12. Gouvea, V., R. I. Glass, P. Woods, K. Taniguchi, H. F. Clark, B. Forrester, and Z.-Y. Fang. 1990. Polymerase chain reaction amplification and typing of rotavirus nucleic acid from stool specimens. J. Clin. Microbiol. 28:276-282. 13. Graff, J., J. Ticehurst, and B. Flehmig. 1993. Detection of hepatitis A virus in sewage sludge by antigen capture polymerase chain

VOL. 60, 1994

reaction. Appl. Environ. Microbiol. 59:3165-3170. 14. Jiang, X., J. Wang, D. Y. Graham, and M. K. Estes. 1992. Detection of Norwalk virus in stool by polymerase chain reaction. J. Clin. Microbiol. 30:2529-2534. 15. Kopecka, H., S. Dubrou, J. Prevot, J. Marechal, and J. M. L6pez-Pila. 1993. Detection of naturally occurring enteroviruses in waters by reverse transcription, polymerase chain reaction, and hybridization. Appl. Environ. Microbiol. 59:1213-1219. 16. Morse, D. L., J. J. Guzewich, J. P. Hanrahan, R. Stricof, M. Shayegani, R. Deibel, J. C. Grabau, N. A. Nowak, J. E. Herrmann, G. Cukor, and N. R. Blacklow. 1986. Widespread outbreaks of clam- and oyster-associated gastroenteritis. Role of Norwalk virus. N. Engi. J. Med. 314:678-681. 17. Muir, P., F. Nicholson, M. Jhetam, S. Neogi, and J. E. Banatvala. 1993. Rapid diagnosis of enterovirus infection by magnetic bead extraction and polymerase chain reaction detection of enterovirus RNA in clinical specimens. J. Clin. Microbiol. 31:31-38. 18. Murphy, A. M., G. S. Grohmann, P. J. Christopher, W. A. Lopez, G. R. Davey, and R. H. Millsom. 1979. An Australia-wide outbreak of gastroenteritis from oysters caused by Norwalk virus. Med. J. Aust. 2:329-333. 19. Paul, M. O., and E. A. Erinle. 1982. Influence of humidity on rotavirus prevalence among Nigerian infants and young children with gastroenteritis. J. Clin. Microbiol. 15:212-215. 20. Richards, G. P. 1985. Outbreaks of shellfish-associated enteric virus illness in the United States: requisite for development of viral guidelines. J. Food Prot. 48:815-823. 21. Schwab, K. J., R. De Leon, R. S. Baric, and M. D. Sobsey. 1991. Detection of rotaviruses, enteroviruses and hepatitis A virus by reverse transcriptase-polymerase chain reaction, p. 475-491. In Advances in water analysis and treatment. Proceedings of the Water Quality Technology Conference, Orlando, Fla. American Water Works Association, Denver. 22. Schwab, K. J., R. De Leon, and M. D. Sobsey. 1993. Development


23. 24.

25.

26.

27.

28.

29.

30.

2407

of PCR methods for enteric virus detection in water. Water Sci. Technol. 27:211-218. Southern, E. M. 1975. Detection of specific sequences among DNA fragments separated by gel electrophoresis. J. Mol. Biol. 98:503-517. Toyoda, H., M. Kohara, Y. Kataoka, T. Omata, N. Imura, and A. Nomoto. 1984. Complete nucleotide sequences of all three poliovirus serotype genomes. Implication for genetic relationship, gene function and antigenic determinants. J. Mol. Biol. 174:561-585. Tsai, Y.-L., C. J. Palmer, and L. R. Sangermano. 1993. Detection of Escherichia coli in sewage and sludge by polymerase chain reaction. Appl. Environ. Microbiol. 59:353-357. Tsai, Y.-L., M. D. Sobsey, L. R. Sangermano, and C. J. Palmer. 1993. Simple method of concentrating enteroviruses and hepatitis A virus from sewage and ocean water for rapid detection by reverse transcriptase-polymerase chain reaction. Appl. Environ. Microbiol. 59:3488-3491. Tsai, Y.-L., M. T. Yahya, C. D. McGee, L. R. Sangermano, and C. J. Palmer. 1993. Detection of enteroviruses and hepatitis A virus in sewage and ocean water by reverse transcriptase-polymerase chain reaction, p. 115-122. In Proceedings of the Water Environment Federation 66th Annual Conference & Exposition, vol. 7, Anaheim, Calif. Water Environment Federation, Alexandria, Va. Wanke, C. A., and R. L. Guerrant. 1987. Viral hepatitis and gastroenteritis transmitted by shellfish and water. Infect. Dis. Clin. N. Am. 1:649-664. Way, J. S., K. L. Josephson, S. D. Pillai, M. Abbaszadegan, C. P. Gerba, and I. L. Pepper. 1993. Specific detection of Salmonella spp. by multiplex polymerase chain reaction. Appl. Environ. Microbiol. 59:1473-1479. Wilde, J., R. Van, L Pickering, J. Eiden, and R. Yolken. 1992. Detection of rotaviruses in the day care environment by reverse transcriptase polymerase chain reaction. J. Infect. Dis. 166:507-511.