Isolation of West Nile Virus from Urine Samples of Patients with Acute Infection

Abstract

This study demonstrated that West Nile virus (WNV) excreted in the urine of patients with acute infection can be isolated in cell cultures. In addition, the protocols for WNV isolation from urine samples were standardized, and factors that may affect the efficiency of WNV isolation were identified.

TEXT

West Nile virus (WNV), a mosquito-borne flavivirus, has become a public health emergence in recent years, since it has been responsible for large outbreaks of neuroinvasive disease and fever in humans in many European countries and in America (1). In humans, WNV infection is asymptomatic in approximately 80% of cases, while it presents as West Nile fever (WN-F) in 20% of cases and as neuroinvasive disease (WN-ND) in less than 1% of cases (2).

Laboratory diagnosis of acute WNV infection is based on the detection of WNV RNA in blood and cerebrospinal fluid (CSF), viral isolation in cell culture, and the demonstration of WNV-specific antibodies. This laboratory diagnosis is challenging because the WNV load in blood is generally very low or undetectable at the time of symptom onset and because WNV antibodies cross-react with other flaviviruses, thus requiring confirmation by neutralization assays. A relevant improvement to the diagnosis of WNV infection was recently provided by WNV RNA testing in urine, since viral RNA is detectable in urine at a higher load and for a longer time than in blood or CSF (3–11). In addition, the isolation of infectious WNV in cell cultures from urine samples has been reported in some patients with acute infection (8–10). However, the protocols for WNV isolation from urine have not been standardized, and the infectivity of WNV excreted in urine has not been investigated in detail. Thus, the aim of this study was to investigate the infectivity of WNV excreted in urine from patients with acute infection. To this aim, methods for the isolation of WNV in cell cultures from urine samples were set up.

The urine samples investigated in this study were collected from 17 patients from northern Italy with acute WNV infection in 2013. The cases included 7 patients with WN-ND, 6 patients with WN-F, and 4 asymptomatic blood donors in whom WNV RNA was detected in urine by real-time reverse transcription (RT)-PCR, according to previously reported methods (8). Thirteen patients were infected with WNV lineage 2 (Italy/2013 strain), and four were infected with WNV lineage 1 (Livenza strain). Genome sequence information on these strains was previously reported (9). Urine samples were stored at 2 to 8°C for up to 72 h and at −80°C thereafter. This study was approved by the institutional review board and ethics committee of Padua University Hospital.

All of the procedures for WNV isolation were performed within certified biosafety cabinets under biosafety level 3 (BSL3) containment. BHK21 (C-13; American Type Culture Collection [ATCC], Manassas, VA) and Vero E6 (ATCC) cell lines, which are permissive to WNV infection (12), were used to isolate WNV from the urine samples. Cells were grown in three different cell culture supports, i.e., shell vials, 6-well plates, and T25 tissue culture flasks. All isolation conditions were tested in triplicate for all of the urine samples.

For viral isolation, Vero E6 and BHK21 cells were seeded in cell culture supports in Dulbecco modified Eagle medium (DMEM) with 10% fetal bovine serum (FBS). After 48 h, when the cells were at 70% to 90% confluence, a final volume of 0.5 ml/shell vial, 1 ml/well, or 2 ml/flask of urine diluted 1:3 and 1:5 with DMEM (without FBS and added with 100 μg/ml ampicillin, 50 μg/ml gentamicin sulfate, 4 μg/ml ciprofloxacin, and 2.5 μg/ml amphotericin B) and with 0.01 M TRIS buffer (pH 8) to adjust the urine pH to 7.4 to 8.0 was inoculated onto cell monolayers. The shell vials were centrifuged at 290 × g for 2 h, followed by incubation for 60 min at 37°C in 5% CO₂, while the 6-well plates and the flasks were incubated for 90 min at 37°C in 5% CO₂. After incubation, DMEM with 10% FBS was added to the cells, but the urine inoculum was not removed. Cells were then cultured at 37°C in 5% CO₂ for up to 7 days. Since day 2 post urine inoculation, the presence of a cytopathic effect (CPE) was monitored daily, and quantitative real-time RT-PCR was done in the supernatant of cell cultures to detect increased WNV RNA load. Positive cell cultures were detached by scraping and centrifuged at 700 × g for 10 min to separate cell debris, and the supernatant was used for virus propagation in the cell culture.

Out of the 17 WNV RNA-positive urine specimens that were analyzed in this study (Table 1), WNV was isolated in 6 cases, including 2 with WNV lineage 1 infection and 4 with WNV lineage 2, and of these 6 patients, 3 had WN-ND, 1 had WN-F, and 2 were asymptomatic blood donors, thus demonstrating that infectivity of WNV in urine was not restricted to viral lineages or strains and was independent of the severity of disease.

TABLE 1.

Clinical and laboratory findings in patients with WNV RNA detected in urine

Case no.	Diagnosisa	WNV lineage	Days since symptom onset	Serum anti-WNV antibodies	WNV load in plasma (copies/ml)	WNV load in urine (copies/ml)	Urine storage conditions	WNV isolated from urine	Cell line for virus isolationb	Day of CPE appearanceb	Mean WNV RNA copies/ml in cell supernatantb
1	WN-ND	1	8	IgM+/IgG−	8,000	2,500,000	Frozen	Yes	Vero E6	3	8.0 × 108
									BHK21	5	2.4 × 109
2	WN-ND	1	8	IgM+/IgG+	Undetectable	2,300,000	Frozen	No
3	WN-ND	2	3	IgM+/IgG−	1,300	8,300,000	Unfrozen	Yes	Vero E6	2	1.6 × 109
									BHK21	4	8.9 × 109
4	WN-ND	2	2	IgM+/IgG+	Undetectable	15,000,000	Unfrozen	Yes	Vero E6	2	2.0 × 109
									BHK21	3	1.4 × 1010
5	WN-ND	2	4	IgM+/IgG+	100	1,300,000	Frozen	No	NAc	NA	NA
6	WN-ND	2	4	IgM+/IgG−	400	350,000	Frozen	No	NA	NA	NA
7	WN-ND	2	5	IgM+/IgG+	Undetectable	100	Unfrozen	No	NA	NA	NA
8	WN-F	2	8	IgM+/IgG−	Undetectable	180,000	Unfrozen	Yes	Vero E6	3	1.2 × 109
									BHK21	4	9.0 × 109
9	WN-F	2	10	IgM+/IgG+	Undetectable	1,200,000	Frozen	No	NA	NA	NA
10	WN-F	2	21	IgM+/IgG+	Undetectable	<100	Frozen	No	NA	NA	NA
11	WN-F	2	16	IgM+/IgG+	Undetectable	28,000	unfrozen	No	NA	NA	NA
12	WN-F	2	14	IgM+/IgG+	Undetectable	160	Unfrozen	No	NA	NA	NA
13	WN-F	2	3	IgM−/IgG−	Undetectable	28,000	Unfrozen	No	NA	NA	NA
14	Blood donor	1	5	IgM−/IgG−	3,900	1,000	Unfrozen	Yes	Vero E6	3	1.7 × 1010
									BHK21	5	9.2 × 109
15	Blood donor	1	6	IgM+/IgG−	14,000	100,000	Frozen	No	NA	NA	NA
16	Blood donor	2	4	IgM+/IgG−	50,000	3,200	Unfrozen	Yes	Vero E6	3	1.6 × 108
									BHK21	5	2.0 × 109
17	Blood donor	2	4	IgM−/IgG−	2,500	37,000	Frozen	No	NA	NA	NA

^aWN-ND, West Nile neuroinvasive disease; WN-F, West Nile fever.

^bData from WNV isolation in 6-well tissue culture plates.

^cNA, not applicable.

In all 6 cases, WNV was isolated from urine in Vero E6 cells and in BHK21 cells with equal efficiency. The use of spin inoculation in shell vial cultures did not appear to enhance viral isolation efficiency compared with that of standard culture in well plates or tissue culture flasks. In positive cell cultures, CPEs appeared at 2 to 3 days postinfection in Vero E6 cells and at 3 to 5 days postinfection in BHK21 cells. The CPEs in Vero E6 cells appeared as typical foci of infected cells rounding and detaching from the substrate, while the CPEs in BHK21 cells were characterized by diffuse lysis of the cell monolayer (Fig. 1). No differences in CPE features were observed between WNV lineage 1 and lineage 2 isolates. Although the number of WNV lineage 2 and lineage 1 isolates was very low in our experience, WNV lineage 2 seemed to grow more easily in cell culture than did lineage 1. The WNV RNA loads determined by quantitative real-time RT-PCR in positive cell cultures at 3 to 5 days postinoculation ranged from 10⁸ to 10¹⁰ copies/ml and were generally 10-fold higher in BHK21 than in Vero E6 cells (Table 1). WNV isolates were expanded in cell subcultures at high titers, similar to those obtained in the first passage culture.

FIG 1.

Isolation of WNV lineage 2 in cell culture from a urine specimen: microphotographs of Vero E6 and BHK21 cells seeded in 6-well tissue culture plates and inoculated with a WNV RNA-positive urine sample at 1:3 dilutions or mock infected. Images were taken at 3 and 4 days postinoculation (pi) with the urine sample and at 4 days after the mock infection. Magnification, ×10. The illustrated WNV strain was fully sequenced and named Italy/2013/Padova/34.1 (GenBank accession no. KF647251).

Since WNV was not isolated from all the urine specimens in which WNV RNA was detected, we investigated which factors (host, virus, and culture conditions) might have influenced the efficiency of WNV isolation.

Urine samples from which the virus was isolated were collected 2 to 8 days after symptom onset, while WNV RNA could be detected in urine up to 1 month after symptom onset, suggesting that the isolation of WNV from urine was more successful during the first days postinfection. Most (5 out of 6) of the patients from whom WNV was isolated from urine were already WNV IgM positive, indicating that the virus can be isolated from the urine of seropositive patients (Table 1).

The characteristics of the urine samples and cell culture conditions were more relevant to WNV isolation. In fact, the use of low-passage cells (i.e., <20 passages) and urine samples with a high WNV RNA load (>1,000,000 copies/ml) and stored at 2 to 8°C without freezing gave the best results (Table 1). An analysis of a larger number of urine samples from patients with infection, however, is required to further standardize and optimize the protocols for WNV isolation.

The possibility of isolating WNV from urine certainly has relevant implications in the diagnosis of WNV infection and in WNV research. In fact, urine can be collected noninvasively in large amounts from patients, thus representing a very convenient sample for molecular testing and viral isolation. By using urine for viral isolation, it is expected that the number of human WNV isolates available for virological studies will greatly increase in the future. In fact, WNV isolates allow full viral genome sequencing and phylogenetic analysis for the molecular epidemiology of outbreaks. In addition, they can be used in experimental studies on the pathogenicity of WNV and to test novel antiviral drugs and vaccines.

The relevance of WNV infectivity in urine for virus transmission remains unknown (13). It is probably not relevant in regard to human transmission, even though the virus is concentrated in urine and excreted at high titers. However, the presence of infectious WNV in urine could represent a risk for interhuman transmission through kidney transplantation. In fact, as demonstrated in the present study, infectious virus may be present in the urine of patients in whom WNV RNA is undetectable in blood by nucleic acid amplification tests (NAAT). Thus, a negative result by WNV RNA NAAT screening of blood in kidney donors might not exclude the risk of WNV transmission to the recipient, as occurred in our experience (14).

In conclusion, this study demonstrated that WNV excreted in the urine of patients with acute infection can be isolated in cell culture. This study also standardized protocols for the isolation of WNV from urine samples and identified factors that may affect the efficiency of WNV isolation.