Skip Navigation

In Vitro Evaluation of the Protective Role of Human Antibodies to West Nile Virus (WNV) Produced during Natural WNV Infection

  1. M. Rios1,
  2. A. I. Dayton1,
  3. O. Wood1,
  4. I. K. Hewlett1,
  5. J. S. Epstein1,
  6. S. Caglioti3 and
  7. S. L. Stramer2
  1. 1 Laboratory of Molecular Virology-Division of Emerging Transfusion Transmitted Diseases, Office of Blood Research and Review, Center for Biologics Evaluation and Research, Food and Drug Administration, Bethesda
  2. 2 American Red Cross, Gaithersburg, Maryland
  3. 3 Blood Systems Laboratories, Tempe, Arizona
  1. Correspondence or reprints: Maria Rios, 8800 Rockville Pike, Bldg. 29 Rm. 131, HFM-310, Bethesda, MD (Maria.Rios{at}fda.hhs.gov).
 
Next Section

Abstract

Background. West Nile virus (WNV) is endemic in the United States and transmissible by transfusion. Since 2003, the US blood supply has been screened by nucleic-acid tests (NAT) for WNV in minipools (MP-NAT) of 6 or 16 specimens. WNV infection begins with low-level viremia detectable only by individual testing (ID-NAT) and no detectable WNV antibodies. Viremia then increases to levels detectable by MP-NAT, and antibodies become detectable; later, viremia decays to levels detectable only by ID-NAT before becoming undetectable. All but 1 documented WNV transmission by transfusion involved blood components negative for WNV antibodies, raising the question whether WNV antibody-positive blood components with low levels of WNV RNA are infectious.

Methods. Specimens from 102 viremic donors with and withoutWNVantibodies were used to investigate infectivity in cultures of Vero cells and human monocyte-derived macrophages (MDMs).

Results. In Vero cell culture, 54 (74%) of 73 WNV antibody-negative specimens and 10 (36%) of 28 WNV antibody-positive specimens were infectious. In a random subset of 20 specimens tested inMDMculture, 7 (88%) of 8 WNV antibody-positive specimens and 12 (100%) of 12 WNV antibody-negative specimens were infectious.

Conclusion. WNV antibodies do not always protect susceptible cells from WNV infection in vitro. RNA positivity in the presence of antibody cannot be ignored as a theoretical risk for blood recipients and needs further investigation.

West Nile virus (WNV) is transmitted to humans primarily through mosquito bites, and the outcome of infection depends on the age and immune status of the exposed individual. WNV is endemic in the United States, and there have been recurring outbreaks of infection for 9 consecutive years. Infecting between 1.7 and 3.9 million people since 1999, WNV has caused >27,500 documented cases of disease and 1086 deaths reported to the Centers for Disease Control and Prevention (CDC), and it has become the most common cause of viral encephalitis in the country [1–4].

Human-to-human transmission by blood transfusion was identified in 2002 [5–8], which led to the rapid development and implementation of nucleic-acid testing (NAT) for blood screening under Food and Drug Administration-approved investigational new drug protocols in 2003. Because of logistical limitations, screening of blood donations for WNV is now performed in minipools (MP-NAT) of 6 or 16 plasma specimens (depending on the test kit manufacturer) until a seasonal resurgence of WNV is detected, whereupon NAT screening of individual donations (ID-NAT) is instituted [9].

Retrospective studies identified 23 cases of WNV transmission by transfusion in 2002, associated with blood components from 16 donations that had retention samples or retrieved plasma cocomponents that tested reactive forWNVRNA by use of a research-based polymerase chain reaction (PCR) assay [8]. WNV RNA was also detected in most samples by using investigational assays at 1:16 dilutions to simulate MP-NAT. With the exception of 1 borderline-reactive, IgM-containing sample, all were negative for WNV antibody [10]. To date, the CDC has documented 32 cases of WNV transmission by transfusion [8, 10–13]. In all cases investigated, except for 1 that involved an IgM-positive donation with a low WNV RNA load [10], the donations were viremic and negative for WNV antibody, raising the question of whether components containing low levels of WNV RNA in the presence ofWNVantibodies are infectious to recipients. Studies have shown that MP-NAT fails to detect 25% of infected units of blood components with low viral loads (i.e., viral loads detectable only by ID-NAT), of which 90% areWNVantibody positive [10, 14, 15].

There are no precise measurements of the time of appearance of viremia after human infection with WNV through mosquito bites. WNV is estimated to become detectable by ID-NAT 1–2 days after infection and detectable by MP-NAT 3–5 days after infection. The appearance of WNV antibodies in 7–10 days coincides with a decline in viremia level, which becomes detectable only by ID-NAT. Although most donors appear to clear the virus in weeks, low levels of viremia may persist for several months after seroconversion, eventually dipping below the level of detectability by ID-NAT [16]. Although donor screening forWNV by MP-NAT is carried out year-round, blood collection facilities use algorithms based on the incidence of MP NAT-reactive donations to determine when to implement ID-NAT during outbreaks, and most reactive units of blood components identified only by ID-NAT contain WNV antibodies [10].

In 32 documented cases ofWNVtransmission by transfusion, only a single case was associated with a unit of blood components that had a low WNV RNA load and was WNV antibody positive, suggesting that antibody-positive blood components are less efficient transmitters of the virus than antibody-negative blood components. We further investigated whether NATreactive specimens that contained WNV antibodies were infectious in vitro and determined that the presence of WNV antibodies in WNV RNA-positive specimens reduced the incidence of infection in susceptible cell cultures, but did not prevent infection.

Previous SectionNext Section

Methods

Specimens. This study included 110 blood donation specimens, collected between 2002 and 2006. There were 102 specimens that were NAT reactive for WNV (in 2003, there were 16 specimens collected; in 2004, there were 22; in 2005, there were 21; and in 2006, there were 43); these specimens were initially reactive either by ID-NAT or MP-NAT and confirmed reactive in the index donation or during donor follow-up studies by use of 2 NAT assays and/or by the presence of WNV antibodies in ELISAs from Abbott Laboratories and/or Focus Technologies [17–20]. Eight specimens from 2002 were negative for WNV by NAT and ELISA [8]; these constituted a negative control group.

Evaluation of plasma volume requirement for infectivity assay. The optimal volume of 0.5 mL of plasma used to infect cell cultures was determined in parallel experiments that used 11 WNV-positive specimens in increasing volumes of 0.1, 0.5, and 1 mL. The Vero cell cultures that received 0.1 mL of plasma showed lower levels of cytopathic effect (CPE) than the cultures that received 0.5 mL. However, when volume was increased to 1 mL of plasma, the level of CPE was reduced, and 2 specimens that had been positive forWNVin cultures infected with 0.5 mL were negative for WNV at the greater plasma volume. The reason for decreased infectivity with the higher volume remains unclear but could result from nonspecific toxicity or, most likely, from observable formation of a clot in the presence of Ca++ in the tissue culture medium.

Infectivity assays using Vero cells. Cells were grown to 85% confluence in T75 flasks in Dulbecco's modified Eagle's medium (Gibco BRL) containing 5% fetal bovine serum (Hyclone) and 10µg/mL of penicillin-streptomycin (Gibco). The medium was removed, 0.5 mL of the specimen was added, and the volume was adjusted to 5 mL with fresh medium. Cultures were incubated for 2 h with plasma, either at room temperature with gentle rocking or in an incubator at 37°C with occasional mixing. After incubation, 10 mL of fresh medium were added; cultures were incubated at 37°C in 5% CO2 and examined daily for CPE. The absence of infection was defined by the absence of observable CPE during 10 days of culture (figure 1A and 1B). Independent infectivity assays for each specimen were carried out at least twice, and each experiment included positive and negative controls, as described below. Of the 102 specimens (74 WNV antibody-negative specimens and 28 WNV antibody-positive specimens), 101 were tested in Vero cells.

Figure 1
View larger version:
Figure 1

West Nile virus infection of Vero cell and monocyte-derived macrophage (MDM) cultures. A, uninfected Vero cells; B, infected Vero cells with observable cytopathic effect on day 4 of culture; C, uninfected MDMs; D, infected MDMs on day 4 of culture.

In several instances, culture supernatants from a first passage (P1) were added to uninfected Vero cell cultures for a second passage (P2) to confirm results. Supernatants from all day 10, CPE-negative cultures were subjected to multiple passages to confirm negative results.

Infectivity assays using monocyte-derived macrophages (MDM). Human peripheral blood monocytes from blood components suitable for transfusion were kindly provided by the National Institutes of Health Blood Bank. Monocytes were obtained either by elutriation [21] or prepared by adherence of human peripheral blood mononuclear cells separated by standard Ficoll-gradient centrifugation using Histopaque 1077 (Sigma). Monocytes were added to T25 flasks in 5-mL volumes at a concentration of 1.7 × 106 cells/mL and cultivated in the presence of macrophage colony-stimulating factor for 1 week for differentiation into macrophages, as described elsewhere [21]. After 1 week, the medium was removed and individual cultures were treated with 0.5 mL of specimens that were NAT reactive for WNV plus 1.5 mL of fresh medium for 2 h at 37°C with occasional mixing, then 3 mL of fresh medium were added, and cultures were incubated at 37°C in 5% CO2. Infectivity experiments included negative and positive controls. Twenty of the 102 specimens (12 WNV antibody-negative specimens and 8 WNV antibody-positive specimens) were tested in MDM culture.

Because MDM infection with WNV does not result in CPE [21]—that is, because infected MDMs are not distinguishable from uninfected MDMs (figures 1C and 1D) under light microscopy—MDM infection was determined by testing supernatants with TaqMan reverse-transcriptase PCR (RT-PCR) for the WNV 3′ noncoding (3′NC) region. Of 20 specimens, 16 were tested multiple times (i.e., 2–6 times) in cultures of MDMs from different donors, and 4 specimens were tested in culture of MDMs from only 1 donor. For several culture supernatants that were positive for WNV by TaqMan RT-PCR, infection was confirmed by the addition of 0.1 mL MDM culture supernatant to uninfected Vero cells plated in 6-well plates, followed by daily observation. In a few experiments for which larger volumes of supernatant were available, 1 mL or 2 mL of supernatant were used.

Controls for infectivity assays. For both Vero cell and MDM cultures, negative controls were mock-infected by using an equal volume of virus-free medium, and the positive controls were infected with a well-characterized WNV isolate (AY646354) at an MOI of 1 pfu/100 cells, under the same conditions used for either plasma or culture supernatants.

WNVTaqMan RT-PCR. RNA extracts were obtained from 140µL of supernatant by using the QIAamp Viral RNA Mini Kit (Qiagen) in accordance with the manufacturer's instructions and tested in duplicate by TaqMan RT-PCR for the WNV 3′NC region; the primers and conditions of testing are described elsewhere [21]. WNV RNA of known copy number concentration [22] was used as a standard for viral load determination.

Estimation of viral titer. Viral titers were not formally determined for all specimens in this study; some specimens had their viral loads determined at the National Genetics Institute [14]. Specimens were classified on the basis of their WNV load. Specimens for which viral load had been determined were classified as having a high viral load if they had a quantifiable viral load (⩾100 copies/mL) and classified as having a low viral load if they had an unquantifiable viral load (<100 copies/mL). If viral load had not been determined, specimens were classified as having a high viral load if the specimen was reactive for WNV in both MP-NAT and ID-NAT and a low viral load if the specimen was not reactive for WNV in MP-NAT and reactive in ID-NAT or produced erratic results in ID-NAT.

To facilitate presentation of results, the specimens were sorted into 4 groups on the basis of estimated viral load and the presence of WNV antibodies (figure 2). Group 1 contained specimens with a low viral load and no detectable WNV antibodies, suggestive of the early phase of infection; group 2 contained specimens with a high viral load and no detectable WNV antibodies; group 3 contained specimens with a high viral load and detectable WNV antibodies; and group 4 contained specimens with a low viral load and detectable WNV antibodies, suggestive of the late phase of infection. Statistical significance was determined by χ2 and Fisher's exact tests.

Figure 2
View larger version:
Figure 2

Flow chart showing distribution of results among the panel of specimens that were nucleic-acid test (NAT) reactive for West Nile virus (WNV). Specimens are classified according to serological status, WNV RNA load, and infectivity results. *One specimen not tested in Vero cell culture. **Low viral load in the absence of antibody, suggestive of the early stage of infection (prior to seroconversion); ***Low viral load in the presence of antibody, suggestive of the late stage of infection (after seroconversion)

Previous SectionNext Section

Results

Characterization of the specimens according to viral load and presence of WNV antibodies. Of 102 specimens reactive for WNV in NAT, 74 were negative for WNV antibodies (figure 2). Of these 74 specimens, 61 (82%) had high viral loads (group 1), and 13 (18%) had low viral loads (group 2). Of the 28 specimens that were NAT-reactive for WNV and positive for WNV antibody, 14 (50%) had high viral loads (group 3), and 14 (50%) had low viral loads (group 4). The 102 specimens from all 4 groups were tested for their ability to infect Vero cell cultures and/or MDM cultures.

Infection of Vero cell cultures. Of the 102 specimens that were NAT reactive for WNV, 101 were tested in Vero cell culture. A total of 64 specimens (63%) infected Vero cells (58 of the specimens produced CPE on P1 by day 7 after inoculation), including 54 specimens that were WNV antibody negative and 10 that were WNV antibody positive, which showed that the presence of WNV antibodies in WNV RNA-positive specimens did not prevent the infection of Vero cells (table 1 and figure 2). Of the 37 CPE-negative specimens, 18 were WNV antibody positive and 19 were WNV antibody negative.

Table 1
View larger version:
Table 1

Infectivity of 102 specimens that were nucleic-acid test-reactive for West Nile virus (WNV), according to viral load and presence of WNV antibodies.

There was no significant correlation between antibody class and infectivity (table 2): 4 (20%) of 20 of IgM-positive specimens and 8 (37%) of 21 IgG-positive specimens were infectious in Vero cells. The 8 specimens in the negative control group did not infect Vero cell cultures after 3 sequential passages.

Table 2
View larger version:
Table 2

Infectivity of 28 specimens that contained antibodies to West Nile virus, according to viral load.

Infection of MDM cultures. Twenty ID NAT-reactive specimens were randomly selected for MDM culture, and 19 (95%) of these specimens infected MDMs (tables 1 and 3). Of the 19 that infected MDMs, 12 specimens were WNV antibody negative and 7 were WNV antibody positive. The specimen that did not infect MDMs in cultures of cells from 3 different donors or in Vero cell culture was positive for WNV antibodies. This specimen (number 16) was not MP-NAT reactive for WNV, which was suggestive of low viral load in late infection.

Table 3
View larger version:
Table 3

Infectivity of a random subset of 20 specimens that were reactive for West Nile virus in nucleic-acid testing (NAT), tested in monocyte-derived macrophage (MDM) culture.

MDM susceptibility to WNV infection is donor-dependent. The ability of specimens to infect MDM cultures obtained from different donors was variable. Some specimens infected MDMs from some donors, but not from others (table 3). This variability was more evident for WNV antibody-positive specimens with low viral loads, as shown in tables 2 and 3: specimens 15, 17, and 18 infected MDMs from 3 of 6, 2 of 3, and 5 of 8 different donors, respectively. However, some WNV antibody-negative specimens with high viral loads also failed to consistently infect MDMs from all donors (table 3, specimens 1–4). Similar variability has been previously observed [21].

MDM supernatants positive for WNV RNA infect Vero cells. Infection of MDM cultures was determined by PCR because WNV infection in MDM does not produce CPE. We investigated whether MDM supernatants positive for WNV RNA by PCR contained infectious WNV. Depending on the amount available, either 0.1 or 0.5 mL of supernatant from each of 18 of the 20 specimens tested on MDMs were added to individual cultures of Vero cells (table 3). Ten of the 18 specimens induced CPE; 4 of these specimens were positive for WNV antibody and 6 were negative for WNV antibody. Two WNV antibody-positive specimens (8 and 26) did not infect Vero cells but infected MDM. Interestingly, the MDM supernatants were capable of infecting Vero cells, suggesting that the antibodies present in the original specimen were neutralizing for Vero cells but not for MDMs. These results indicate that MDM cultures are more sensitive than Vero cell cultures to WNV infection by human specimens and that MDM infection generates viral particles capable of infecting Vero cells.

Infectivity of WNV antibody-negative specimens with low and high WNV RNA loads. Eleven of 13 (85%) WNV antibody-negative specimens with low viral loads were infectious. Nine specimens infected Vero cells, but 2 of them required P2 to induce CPE (table 1), and 2 infected only MDMs (table 3, specimens 1 and 2). Forty-nine (80%) of 61 specimens with high viral loads were infectious for either or both cell culture systems (45 specimens infected Vero cells and 4 infected MDMs only) (table 3, specimens 3, 5, 6, and 20). In Vero cells, 41 (68%) of 60 were positive in P1, and 4 (7%) required P2 (table 3, specimens 7–10). One specimen was not tested in Vero cells but was positive in MDM culture.

Infectivity of WNV antibody-positive specimens with low and high WNV RNA loads. Of 28 WNV antibody-positive specimens tested in Vero cell culture (table 2), 14 had high viral loads and 14 had low viral loads. Of the 14 specimens with high viral loads, 6 produced CPE. Of the 14 with low viral loads, 4 produced CPE. Of 8 WNV antibody-positive specimens tested in MDMs, 4 had high viral loads and 4 had low viral loads. All 4 with high viral loads infected MDMs, and 3 of the 4 with low viral loads infected MDMs. Overall, 15 (54%) of 28 WNV antibody-positive specimens—including 8 with high viral loads and 7 with low viral loads—were infectious for Vero cell culture and/or MDM culture (tables 2 and 3).

In Vero cells, 10 (36%) of 28 specimens were infectious: 4 of 14 with low viral loads (2 of these specimens required P2) and 6 of 14 with high viral loads. Eight specimens (4 with high viral loads and 4 with low viral loads) were tested in MDM cultures (table 3, specimens 11–18). Of these 8 specimens, 7 (88%) infected MDMs, including 2 that infected Vero cells (specimens 11 and 12); the specimen that did not infect MDMs had a low viral load (specimen 16). Culture supernatants from 6 of the 7 specimens that were infectious in MDM culture were tested in Vero cells; 4 produced CPE within 5 days (specimens 11, 12, 14, and 18), and 2 remained CPE negative at day 10 (specimens 13 and 5). Both specimens that remained CPE negative had been shown to have lower viral loads by TaqMan RT-PCR and had small amounts of MDM supernatant available for testing in Vero cells.

Multiple passages to confirm negative results. Forty-four specimens that were NAT reactive forWNVand CPE negative in Vero cell culture on P1 were subjected to multiple passages (P2–P5) to confirm the negative results. Of 24 antibody-negative cultures, 5 were positive for WNV antibodies on P2 and not tested in MDMs, and 19 remained CPE negative during multiple passages. Six of these 19 specimens tested positive for WNV in ⩾2 independent MDM experiments that used cells from different donors. For 4 of the 6 specimens positive for WNV in MDM culture, supernatant from 1 positive culture of each specimen was tested in Vero cells, and 3 of 4 specimens caused CPE within 5 days (table 3).

Two of the 20 antibody-positive specimens that tested CPE negative in Vero cells on P1 were CPE positive on P2 and were not tested in MDMs (table 2, specimens 39 and 40). The other 18 were consistently CPE negative during multiple passages. Five specimens infected MDMs in ⩾2 independent experiments that used cells from different donors. MDM culture supernatants infected with 4 of these 5 specimens were used to infect Vero cell cultures; 2 of these were CPE positive within 5 days, and 2 remained CPE negative at day 10 (table 3). Each of the specimens that remained CPE negative had lower viral loads and limited amounts of supernatant available for testing in Vero cells.

Previous SectionNext Section

Discussion

The data show that 15 (54%) of 28 WNV RNA-positive, WNV antibody-positive specimens could infect susceptible cells (table 3). Of 20 specimens positive for WNV-specific IgM, 8 (40%) could infect susceptible cells; of 21 specimens positive for WNVspecific IgG, 13 (62%) could infect susceptible cells. Both of these results suggest that antibodies toWNVare not always protective in vitro (figure 2 and tables 2 and 3). Overall, there was no statistical significance associated with antibody class and protection from infection (P = .217). MDM culture detected more specimens containing WNV RNA than did Vero cell culture. Overall, 64 (63%) of 101 WNV RNA-positive specimens induced CPE in Vero cells, whereas 19 (95%) of 20 specimens were identified as WNV positive by TaqMan RT-PCR of MDM culture supernatants. Nineteen specimens were tested in both systems; 18 of these specimens (95%) were positive for WNV in MDM culture, and only 9 specimens (47%) induced CPE in Vero cells (figure 2). This difference was highly significant (P ⩽ .001).

When infectivity results on Vero cell andMDMcultures were combined, 15 (54%) of 28 antibody-positive specimens were infectious, whereas 73 (99%) of 74 antibody-negative specimens showed infectivity. This difference was highly significant (P ⩽ .001). When analyzed independently, the presence of antibody significantly inhibited WNV infection of Vero cells in culture (P ⩽ .001) but it did not inhibit infection of MDMs (P > .05). We speculate that MDMs' higher susceptibility to infection in the presence of antibodies is the result of expression of immunoglobulin receptors on the macrophage membrane.

WNV antibody-positive blood components with low WNV RNA loads are found in later stages of infection and are only detected by ID-NAT. It is generally believed that these components pose a lower threat of infection to transfusion recipients because of the low number of documented transfusiontransmitted infections that have been the result of such blood components. We tested 14 such specimens and observed infection rates of 50% when the results of Vero cell and MDM cultures were combined.

The results presented clearly show that frequently, WNV RNA-positive specimens remain infectious in vitro despite the presence of WNV antibodies. Four of 14 WNV antibody-positive specimens, all with low levels of WNV RNA (i.e., not quantifiable), were capable of infecting Vero cell cultures (these specimens were not tested in MDM cultures, and 2 of the 4 required P2 to produce CPE). Three other WNV antibody-positive specimens with low levels of WNV RNA infected MDM cultures. One of these did not infect Vero cells directly, but did so after passage through MDMs, demonstrating that the TaqManpositive results for MDM supernatants reflect the presence of infectious WNV particles.

In conjunction with the one reported case of WNV transmission by blood components that contained low levels of WNV RNA and was WNV antibody-positive, our data suggest that blood components that have low levels of WNV RNA and are WNV antibody-positive can transmit WNV to recipients. The lack of identified cases of transmission of WNV infection by blood transfusion may be the result of several factors, including the fact that most WNV infections are benign and asymptomatic. Additionally, recipients of blood components are not routinely tested for WNV, and mild symptoms of WNV infection may be mistaken for other common conditions including colds or posttransfusion reactions. Nevertheless, the relative scarcity of reports of transfusion transmission by WNV antibody-positive blood components, when compared with our frequent finding of transmissibility to cells in vitro, suggests that if there is protection in the in vivo situation, it may involve not only humoral immunity but also additional immune pathways.

Antibody cross-reactivity among flaviviruses of the Japanese encephalitis complex is well known and requires dilutional testing to differentiate specificity [23]. In the present study, the WNV specificity of the IgG antibodies was not verified by the plaque reduction neutralizing test, which is used as the gold standard for differential diagnosis of infection in diagnostic settings [24, 25]. The observed infection of Vero cells and MDMs by WNV antibody-positive specimens should not be considered the result of lack of antibody specificity, because the specimens were collected from donors who were NAT positive for WNV RNA in communities that had had documented WNV outbreaks.

Additionally, not all specific antibodies are neutralizing. Thus, it is unlikely that these antibodies appeared as the result of infection by other members of the Japanese encephalitis serocomplex in the absence of recognized outbreaks. In several cases, the same plasma specimen was capable of infecting cultures of MDMs from some donors but not others. Although we cannot determine the reason(s) for this finding, similar results were previously observed in attempts to infect cultures of MDMs from different donors with WNV [21]. Variable susceptibility to infection does not correlate with the presence of antibodies. Alternatively, the variability in donor susceptibility may be associated with viral load and concentration of infectious particles in the specimen, or the in vivo situation may involve other biological pathways.

Infectivity in vitro does not always equate to infectivity in vivo. Also, the minimum infectious dose of WNV in humans is not known, which raises the possibility that the specimens we studied have an insufficient number of viral particles to be capable of infecting humans. However, our data show that specimens positive for both WNV RNA and WNV antibody contain live, infectious WNV, raising theoretical concerns about the risk of transmission by transfusion.

Ideally, recipient lookback studies could resolve this issue. However, such studies are logistically very difficult and limited in scope. Alternatively, this issue could be studied in appropriate animal models. However, small animal models are not suitable because of limits in specimen volume that can be administered, compared to the volume of blood components administered to humans in transfusion practice. Blood components that contain low levels of WNV RNA may only be infectious when large volumes are transfused. Nonhuman primate models could help clarify the relevance of our findings in vivo.

This study demonstrates that some specimens positive for both WNV RNA and WNV antibody can infect cells in vitro. Thus, WNV antibodies do not always protect susceptible cells from WNV infection. These results indicate that donations that have a low WNVRNA load and are antibody-positive should not be ignored as a potential risk for transfusion recipients and need further investigation in vivo.

Previous SectionNext Section

Acknowledgments

We would like to thank, Gregory Foster, David Krysztof, Letitia Fielding, and Sara Allmond, from the American Red Cross; and Valerie Winkelman from the Blood System Laboratories, for their technical assistance.

Previous SectionNext Section

Footnotes

  • Potential conflicts of interest: none reported.

  • Financial support: Center for Biologics Evaluation and Research, Food and Drug Administration.

  • The findings and conclusions in this article have not been formally disseminated by the Food and Drug Administration and should not be construed to represent any Agency determination or policy.

  • Received March 26, 2008.
  • Accepted May 22, 2008.
Previous Section
 

References

  1. 1.↵
    1. van der Meulen KM,
    2. Pensaert MB,
    3. Nauwynck HJ
    . West Nile virus in the vertebrate world. Arch Virol 2005;150:637-57.
    CrossRefMedlineWeb of Science
  2. 2.
    Centers for Disease Control and Prevention. Outbreak of West Nile-like viral encephalitis—New York, 1999. MMWR Morb Mortal Wkly Rep 1999;48:845-9.
    Medline
  3. 3.
    Centers for Disease Control and Prevention. Statistics, surveillance and control for West Nile virus. [Accessed 4 March 2007]. Available at: http://www.cdc.gov/ncidod/dvbid/westnile/index.htm.
  4. 4.↵
    1. O'Leary DR,
    2. Marfin AA,
    3. Montgomery SP,
    4. et al
    . The epidemic of West Nile virus in the United States, 2002. Vector Borne Zoonotic Dis 2004;4:61-70.
    CrossRefMedlineWeb of Science
  5. 5.↵
    Centers for Disease Control and Prevention. Update: investigations of West Nile virus infections in recipients of organ transplantation and blood transfusion—Michigan, 2002. MMWR Morb Mortal Wkly Rep 2002;51:833-6.
    Medline
  6. 6.
    Centers for Disease Control and Prevention. Possible West Nile virus transmission to an infant through breast-feeding—Michigan, 2002. MMWR Morb Mortal Wkly Rep 2002;51:877-8.
    Medline
  7. 7.
    Centers for Disease Control and Prevention. Intrauterine West Nile virus infection—New York, 2002. MMWR Morb Mortal Wkly Rep 2002;51:1135-6.
    Medline
  8. 8.↵
    1. Pealer LN,
    2. Marfin AA,
    3. Petersen,
    4. et al
    . ransmission of West Nile virus through blood transfusion in the United States in 2002. N Engl J Med 2003;349:1236-45.
    CrossRefMedlineWeb of Science
  9. 9.↵
    1. Busch MP,
    2. Tobler LH,
    3. Saldanha J,
    4. et al
    . Analytical and clinical sensitivity of West Nile virus RNA screening and supplemental assays available in 2003. Transfusion 2005;45:492-9.
    CrossRefMedlineWeb of Science
  10. 10.↵
    1. Stramer SL,
    2. Wagner AG,
    3. Paolillo JE,
    4. et al
    . West Nile Virus (WNV) tests identify units implicated in transfusion transmission [abstract S7-030B]. Transfusion, special abstract supplement, 56th Annual AABB Meeting; San Diego. 2003 [Accessed 25 August 2008]. Available at: http://www3.interscience.wiley.com/cgi-bin/fulltext/118875080/PDFSTART.
  11. 11.
    Centers for Disease Control and Prevention. Update: West Nile Virus Screening of blood donations and transfusion-associated transmission— United States, 2003. MMWR Morb Mortal Wkly Rep 2004;53:281-4.
    Medline
  12. 12.
    Centers for Disease Control and Prevention. Transfusion-associated transmission of West Nile virus—Arizona, 2004. MMWR Morb Mortal Wkly Rep 2004;53:842-4.
    Medline
  13. 13.↵
    Centers for Disease Control and Prevention. West Nile Virus Transmission Through Blood Transfusion—South Dakota, 2006. MMWR Morb Mortal Wkly Rep 2007;56:76-9.
    Medline
  14. 14.↵
    1. Stramer SL,
    2. Fang CT,
    3. Foster AG,
    4. Wagner AG,
    5. Brodsky JP,
    6. Dodd RY
    . West Nile virus among blood donors in the United States, 2003 and 2004. N Engl J Med 2005;353:451-9.
    CrossRefMedlineWeb of Science
  15. 15.↵
    1. Busch MP,
    2. Caglioti S,
    3. Robertson EF,
    4. et al
    . Screening the blood supply for West Nile virus RNA by nucleic acid amplification testing. Engl J Med 2005;353:460-7.
    CrossRefMedlineWeb of Science
  16. 16.↵
    1. Stramer SL,
    2. Foster GA,
    3. Paolillo JE,
    4. Wagner AG,
    5. Dodd RY,
    6. Brodsky JP
    . Dynamics of WNV infection and resolution among blood donors [abstract S14-030D]. Transfusion, special abstract supplement, 57th Annual AABB Meeting; Baltimore. 2004 [Accessed 25 August 2008]. Available at: http://www3.interscience.wiley.com/cgi-bin/fulltext/118764403/PDFSTART.
  17. 17.↵
    1. Muerhoff AS,
    2. Dawson GJ,
    3. Dille B,
    4. et al
    . Enzyme-linked immunosorbent assays using recombinant envelope protein expressed in COS-1 and drosophila S2 cells for detection of West Nile virus immunoglobulin M in serum or cerebrospinal fluid. Clin Diagn Lab Immunol 2004;11:651-7.
  18. 18.
    1. Ratterree MS,
    2. Gutierrez RA,
    3. Travassos da Rosa APA,
    4. et al
    . Experimental infection of rhesus macaques with West Nile virus: level and duration of viremia and kinetics of the antibody response after infection. J Infect Dis 2004;189:669-76.
    Abstract/FREE Full Text
  19. 19.
    1. Prince HE,
    2. Lape'-Nixon M,
    3. Moore RJ,
    4. Hogrefe WR
    . Utility of the focus technologies West Nile virus immunoglobulin M capture enzymelinked immunosorbent assay for testing cerebrospinal fluid. J Clin Microbiol 2004;42:12-5.
    Abstract/FREE Full Text
  20. 20.↵
    1. Prince HE,
    2. Hogrefe WR
    . Detection of West Nile virus (WNV)-specific immunoglobulin M in a reference laboratory setting during the 2002 WNV season in the United States. Clin Diagn Lab Immunol 2003;10:764-8.
    CrossRefMedline
  21. 21.↵
    1. Rios M,
    2. Zhang MJ,
    3. Srinivasan K,
    4. Daniel S,
    5. Hewlett I, Dayton A
    . Monocytes/macrophages are a potential target in human infection with West Nile Virus (WNV) through blood transfusion. Transfusion 2006;46:659-67.
    CrossRefMedline
  22. 22.↵
    1. Rios M,
    2. Sirnivasan K,
    3. Wood O,
    4. Daniel O,
    5. Tomioka K,
    6. Williams A,
    7. Hewlett I
    . West Nile Virus: correlation between infectivity and RNA copy number [abstract S9-030B]. Transfusion, special abstract supplement, 56th Annual AABB Meeting; San Diego. 2003 [Accessed 25 August 2008]. Available at: http://www3.interscience.wiley.com/cgi-bin/fulltext/118875080/PDFSTART.
  23. 23.↵
    1. Kuno G
    . Serodiagnosis of flaviviral infections and vaccinations in humans. Adv Virus Res 2003;61:3-65.
    CrossRefMedline
  24. 24.↵
    1. Martin DA,
    2. Biggerstaff BJ,
    3. Allen B,
    4. Johnson AJ,
    5. Lanciotti RS,
    6. Roehrig JT
    . Use of immunoglobulin M cross-reactions in differential diagnosis of human flaviviral encephalitis infections in the United States. Clin Diagn Lab Immunol 2002;9:544-9.
    CrossRefMedline
  25. 25.↵
    1. Wong SJ,
    2. Boyle RH,
    3. Demarest VL,
    4. et al
    . Immunoassay targeting nonstructural protein 5 to differentiate West Nile virus infection from Dengue and St. Louis encephalitis virus infections and from flavivirus vaccination. J Clin Microbiol 2003;41:4217-23.
| Table of Contents

This Article

  1. J Infect Dis. (2008) 198 (9): 1300-1308. doi: 10.1086/592277
  1. AbstractFree
  2. » Full Text (HTML)Free
  3. Full Text (PDF)Free

Classifications

Share

  1. Email this article

Navigate This Article

Current Issue

  1. March 1, 2012 205 (5)
  1. Journal of Infectious Diseases
  1. Alert me to new issues

Most