Histo-Blood Group Antigen Assay for Detecting Noroviruses in Water

We evaluated a novel, magnetic-bead-based histo-blood groupantigen assay for the recovery of low numbers of norovirus particles.Using this assay, with Norwalk virus seeded in environmentalwaters as a model, we were able to recover 30 to 300 genomiccopies of the virus.

INTRODUCTION

Top

INTRODUCTION

In recent years, human noroviruses (NoVs) have been recognizedas the leading cause of outbreaks of acute gastroenteritis worldwide.Large NoV outbreaks have been reported in association with waterused for drinking, recreation, shellfish harvesting, irrigation,and washing produce (1, 2, 3, 5, 7, 10, 13, 16, 20). NoVs recognizehuman histo-blood group antigens (HBGAs) as receptors (4, 12,14). Since NoVs cannot be cultured in vitro, the virus is detectedby reverse transcription-PCR (RT-PCR) assays, which are susceptibleto inhibitors often found in environmental waters. Here, wereport a novel method for the rapid recovery of low numbersof NoVs by the use of a magnetic-bead-based HBGA assay.

HBGA binding assay optimization.

Top

HBGA binding assay optimization.

Using a previously described HBGA binding assay (9) we evaluatedthree blocking buffers (5% [wt/vol] Blotto, 5% [vol/vol] fetalbovine serum, and SuperBlock [Pierce Biotechnology, Rockford,IL]) for their ability to block nonspecific binding to uncoatedmagnetic beads without interfering with specific binding ofvirus to HBGA or interfering with the reproducibility of results.Based on these criteria, SuperBlock blocking buffer was chosenfor subsequent experiments. In brief, 25 µl of washedMyOne streptavidin-coated magnetic beads (Dynal Biotech, Olso,Norway) was incubated for 90 min at room temperature with 50µl of synthetic biotinylated H type 1 HBGA (1 mg/ml) (GlycoTech,Rockville, MD), followed by overnight incubation with blockingbuffer. RNA copy numbers of the Norwalk virus (NV) stool suspensionswere determined by comparison to a standard curve, using NVstrain 8FIIb RNA transcripts. Dilutions of 10% stool suspensionscontaining 3,000, 300, 30, or 3 NV (8FIIb) copy numbers wereadded to 1 ml of blocking buffer, environmental water concentrate,or phosphate-buffered saline prior to incubation (4 h at roomtemperature) on an end-over-end rotator (Dynal Biotech). Aftereight washes, the beads were suspended in 50 µl of phosphate-bufferedsaline and subjected to heat (5 min at 99°C) to releasethe viral RNA. For the HBGA assay, 2.5 µl or 1 µlof heat-released NV RNA was analyzed by use of a conventionalRT-PCR (Qiagen OneStep RT-PCR kit; Qiagen, Valencia, CA) (6)or a TaqMan real-time RT-PCR (QuantiTect probe RT-PCR kit; Qiagen)on an Applied Biosystems 7500 real-time PCR system platform(Foster City, CA) (22). The detection limit for the real-timeassay was 10 RNA copy numbers.

The method is sensitive and specific for detecting NoV.

Top

The method is sensitive...

A median of 300 copy numbers (n = 10) per milliliter of blockingbuffer was detected by the HBGA assay (Table 1). In the presenceof other enteric viruses (rotavirus group A serotype 1 [strainWA] or human astrovirus type 1 [Oxford strain]), 300 NV copynumbers were recovered by the assay (n = 2).

View this table:
[in this window]
[in a new window]

TABLE 1. Detection limits of the HBGA binding assay for Norwalk virus in the presence of an environmental water matrix or SuperBlock blocking buffer

The method is sensitive in the context of environmental waters.

Top

Surface water (n = 4) and influent (n = 2) and effluent (n =4) wastewater samples were collected from drinking-water andwastewater facilities in Columbus and Atlanta, GA, respectively.Surface water samples were concentrated by precipitation with8% polyethylene glycol (PEG) 8000 and 0.3 M NaCl (23). Aftercentrifugation (7,280 x g, 30 min), the PEG pellet was extractedwith 50% (vol/vol) chloroform. Supernatants (5 to 20 ml) werestored at –70°C until use. The wastewater samples(850 ml) were processed similarly; only an additional centrifugationstep (7,280 x g, 15 min at 4°C) and chloroform extractionof the pellet (50% [vol/vol]) were performed prior to PEG precipitationof the combined aqueous phases. Water concentrates were seededwith 3, 30, 300, or 3,000 copy numbers of NV and analyzed, usingthe HBGA binding assay. The detection limit for each water typetested was between 30 and 3,000 (median, 300) (Table 1).

The method detects viruses with an intact capsid.

Top

The method detects viruses...

To determine if the NV particles were intact and potentiallyinfectious, we bound 30,000 or 6,000 NoV copy numbers to H type1 HGBA-coated beads before treatment with 10 ng of RNase A (Invitrogen,Carlsbad, CA) in Tris-EDTA buffer and incubation for 30 minat 37°C. To neutralize any remaining RNase activity, RNaseinhibitor (160 U/50 µl sample) (Invitrogen) was added,and the sample was incubated for 15 min at 37°C. Beforeand after RNase treatment, GII.4 RNA (1 x 10⁵ copy numbers)was added to each reaction mixture to check the efficacy ofthe RNase A or the RNase inhibitor. The mean cycle thresholdvalues for the treated (35.8; range, 34.5 to 37.0) and untreated(35.2; range, 34.5 to 35.9) samples did not differ significantly(P $≥$ 0.2; Student's t test) at either input level. As expected,RNase treatment of the GII.4 control RNA resulted in completeloss of signal, when analyzed by TaqMan real-time RT-PCR (22).

In summary, we report a novel HBGA magnetic bead separationmethod for human NoVs. The method is sensitive and specificfor detecting NoVs with an intact HBGA receptor binding site.Furthermore, since we demonstrated only a 1-log decrease inmethod sensitivity upon application to sewage samples, thisassay can be used to remove RT-PCR inhibitory compounds presentin environmental waters.

Several research groups have developed sensitive methods forconcentrating NoVs by use of immunomagnetic separation or gastricmucins from pigs (8, 17, 18, 19, 21). Our method can detectlow numbers of NoVs, but it also detects NoV bound to its specificHBGA receptor, which may be a surrogate for detecting infectiousvirus. Previous studies indicate that many, but not all, NoVsbind to HBGAs (9, 11, 15), and the requirement of a secondaryreceptor during infection is not clear. Elucidating the roleof HBGA and NoV infectivity will be needed to further validatethe value of this assay as a surrogate for detecting infectiousvirus. In conclusion, we report an assay that may serve as arapid detection method for potentially infectious NoVs in complexmatrices, such as environmental waters.

ACKNOWLEDGMENTS

We thank Christine Moe (Emory University) and Robert Atmar (BaylorCollege of Medicine) for kindly providing Norwalk virus stocks,Linda Vermeulen for her assistance with the RNase A experiments,and Claudia Chesley for critical reading of the manuscript.

This project was supported by American Water Works and ResearchFoundation project no. 2860.

The findings and conclusions in this article are those of theauthors and do not necessarily represent the views of the Centersfor Disease Control and Prevention (CDC). This article did receiveclearance through the appropriate channels at the CDC priorto submission.

FOOTNOTES

* Corresponding author. Present address: University of Georgia, Center for Food Safety, 1109 Experiment St., Melton Bldg., Griffin, GA 30223. Phone: (770) 467-6094. Fax: (770) 229-3216. E-mail: jcannon{at}uga.edu

Published ahead of print on 5 September 2008.

REFERENCES

Top