Abstract

Zika (ZIKV) and dengue (DENV) virus infections elicit a robust but cross-reactive antibody response against the viral envelope protein, while antibody responses against non-structural proteins (NS) are more virus specific. Building on this premise, we have previously developed a flavivirus multiplex microsphere immunoassay (MIA) for the serologic diagnosis of ZIKV and DENV infections. This assay significantly improved diagnostic accuracy; however, MIA could not differentiate more recent from past infections, which still represents a major diagnostic challenge. Therefore, an immunoglobulin G (IgG) based avidity assay was developed and its diagnostic performance evaluated. Specimens from New York State residents were submitted to the Wadsworth Center New York State Department of Health (NYSDOH) for routine clinical testing by Zika IgM ELISA and plaque reduction neutralization test (PRNT). Using our previously developed flavivirus MIA as a platform, we developed an IgG avidity assay to discriminate recent ZIKV from past DENV infections. Zika IgM positive specimens had an average Zika IgG avidity index of 14.8% (95% CI: 11.0–18.4%), while Zika IgM negative but flavivirus MIA and PRNT positive samples had an average Zika IgG avidity index of 34.9% (95% CI: 31.1–38.7%). Specimens positive for dengue antibodies by flavivirus MIA and PRNT had an average dengue IgG avidity index of 68.7% (95% CI: 62.7–75.0%). The IgG avidity assay accurately distinguished recent ZIKV from past DENV infections in patients who traveled to dengue endemic regions. This assay could be very useful in patients with high risk of Zika complications such as pregnant women and monitoring immune responses in vaccine trials.

Introduction

ZIKV and DENV are members of the genus Flavivirus, family Flaviviridae, which also includes Yellow fever virus (YFV) and West Nile virus (WNV). ZIKV, DENV, and YFV are transmitted by the Aedes mosquito species that are widely dispersed in the tropical and subtropical regions of the Americas. In March 2016, however, the Center for Disease Control (CDC) expanded its vector surveillance map to show that the Aedes mosquitoes now occupy a larger portion of the U.S., including the heavily populated northeast corridor [1].

With respect to both clinical presentation and diagnostic methods, ZIKV is very difficult to distinguish from DENV infections, particularly if specimens submitted for routine diagnostic testing are taken beyond the viremic window. Unlike DENV infections, ZIKV has been associated with congenital microcephaly, stillbirth, and neurological complications in adults such as Guillain-Barre syndrome (GBS) [2–5]. Current clinical ZIKV testing relies on RT-PCR or, IgM capture ELISA followed by confirmatory diagnostic PRNT. However, approximately 80% of Zika infections are mild or asymptomatic, and therefore patients often do not seek medical help while their viremia is still detectable by RT-PCR [6]. While, Zika IgM ELISA is recommended for specimens collected more than 4 days after disease onset and continuing for up to 12 weeks post symptom onset or exposure, this assay is confounded by significant levels of cross-reactivity with other flaviviruses primarily dengue [7–10]. In addition, ZIKV testing is complicated by the transient nature of IgM antibodies, which may not be detectable 65 days after disease onset [11]. Although, a more recent report suggests that IgM antibodies may persist for more than 12 weeks after infection [12]. To overcome any potential false positives, Zika IgM positive samples need to be referred for PRNT testing, a very time-consuming test with limited availability and a long turnaround time. PRNT measures neutralizing anti-envelope glycoprotein antibodies in human sera and is the most virus-specific serologic assay amongst flaviviruses; therefore, it is considered the laboratory standard for flavivirus testing [13]. Unfortunately, however, when Zika infection follows a previous dengue or other flavivirus infection, PRNT results are often difficult to interpret due to the presence of highly cross-reactive anti-envelope neutralizing antibodies.

The majority of New York State residents, whose samples were submitted to the Wadsworth Center New York State laboratory for routine diagnostic Zika testing, were from returning travelers from dengue endemic regions. Unfortunately, many people seek ZIKV testing beyond the RT-PCR window and often long after the IgM window had passed. Nevertheless, there is a heightened need to determine if the patients were ever exposed to ZIKV, especially during pregnancy. Therefore, we have recently developed a highly specific and sensitive flavivirus multiplex microsphere immunofluorescence assay (flavivirus MIA) to detect and discriminate ZIKV and DENV total antibodies (IgM, IgG, and IgA) in patient sera. In this assay, we combined the diagnostic power of the viral envelope protein with the discriminatory power of the viral non-structural proteins of Zika and the four serotypes of dengue viruses (DENV 1–4) [14]. This assay filled the need for ZIKV IgG testing and provided enhanced specificity to differentiate ZIKV from DENV.

Despite the improvement in ZIKV diagnostics, there is still a need to better differentiate recent ZIKV infection from past DENV. Therefore, we have developed an IgG avidity test for Zika and dengue by incorporating a urea wash step in our previously developed flavivirus MIA assay [14]. Avidity assays have been used in the past to differentiate current or acute from past infections [15–19]. The term avidity refers to the net-binding force of antibodies to antigens. IgG antibodies develop shortly after infection and are detectable within the first 7 days of illness [11]. Early in infection, IgG antibodies have low avidity for antigens; this gradually increases during subsequent weeks and months due to the immunological process of affinity maturation [18]. The aim of the present study was to use IgG avidity testing to differentiate recent ZIKV from past DENV infections that would particularly benefit pregnant women who might have been exposed to dengue in the past. In addition, it would also be beneficial for monitoring immune responses in vaccine trials.

Materials and methods

Serum samples

Serum samples were obtained from leftover clinical specimens submitted for routine testing to the Wadsworth Center of NYSDOH for Zika RT-PCR, IgM ELISA, Arbovirus MIA, and PRNT testing. A total of 46 serum specimens were grouped into two categories as per the CDC laboratory guidance and diagnostic testing: recent phase and late phase Zika specimens based on PCR and Zika IgM ELISA results [20]. Three of the recent phase specimens were either PCR positive, but all the samples were Zika IgM positive for Zika virus infection. Specimens in the late phase group on the other hand were PCR and Zika IgM ELISA negative. All samples were PRNT positive for anti-Zika virus antibodies and almost half the specimens were also PRNT positive for antidengue antibodies. In addition, most specimens were also positive by the Arbovirus MIA assay which was used as a surrogate for flavivirus exposure testing during the peak season of Zika virus testing [10] and before the development of the flavivirus microsphere immunofluorescence assay in 2017 [14]. The Arbovirus MIA test is also a microsphere-based assay developed in our lab that utilizes West Nile envelope protein to detect human antiflavivirus antibodies in patient sera [21]. All results are summarized in Supplementary Table S1. Recent phase dataset and Supplementary Table S2. Late phase dataset.

Additional samples (n=24) were kindly provided by BARDA-Centers for Disease Control and Prevention Zika Repository. These single timepoint samples are convalescent only specimens that were obtained from Puerto Rico residents and New York State travelers approximately 6–24 weeks (1.4–5.5 months) after Zika PCR positivity. All samples were PRNT confirmed for ZIKV and/or DENV antibodies. These results are summarized in Supplementary Table S3.

Reagents

Wash buffer (PBS), 0.05% Tween 20, pH 7.4) and storage buffer (PBS, 1% BSA, 0.05% azide, pH 7.4) were purchased from Sigma (Sigma–Aldrich, St. Louis, MO). A total of 8 M urea (prepared in PBS, pH 7.2) was kindly supplied by the New York State Wadsworth Center Media department. Microspheres, calibration microspheres, and sheath fluid were obtained from Luminex Corporation (Luminex Corp., Austin, TX). All antigens, Zika envelope, Zika non-structural protein 1 (NS1), dengue non-structural protein 1 serotype used in the assay were purchased from Meridian Life Science, Inc. (Memphis, TN).

Coupling of antigen proteins to Luminex microsphere beads

Recombinant ZIKV NS1 and DENV serotypes 1-4 NS1 proteins (Meridian Life Science, Memphis, TN) were covalently coupled as previously described in [22]. Briefly, 50 µg of purified protein was used to couple to the surface of 6.25 × 106 carboxylated polystyrene microparticles in a two-step carbodiimide process. First, microspheres aka beads were activated using 10 µl of N-hydroxysuccinimide (sulfo-NHS, 50 mg/ml), (Pierce Chemicals, Dallas, TX) followed by 10 µl of 1-ethyl-3-(3-dimethylamino-propyl) carbodiimide-HCl (EDC, 50 mg/ml), (Pierce Chemicals, Dallas, TX). After 20 min incubation at room temperature, each recombinant protein was added to microspheres color coded into spectrally distinct sets or fluorescently distinct regions. Finally, after 3 h of incubation in the dark on a LabTech tube rotator (Barnstead/Thermolyne, Dubuque, IA), the microspheres were then washed twice and resuspended in 1 ml of PBS-TN (PBS, pH 7.4, 0.05% sodium azide, 1% bovine serum albumin), and stored at 4°C.

Plaque reduction neutralization test

PRNT was performed to detect and differentiate neutralizing antibodies to ZIKV and DENV as described previously [23]. Briefly, serial dilutions of patient sera were mixed with equal amounts of viral suspension containing 200 PFU/100 µl and incubated for 1 h at 37°C. Each of the virus-serum sample mixture was then used to inoculate one well of a six-well plate containing a monolayer of Vero cells. After the plate was incubated for 1 h at 37°C, an agar overlay was added and further incubated at 37°C until virus plaques became visible. Cells were then stained with neutral red and plaques were counted. The antibody titer reported was the reciprocal of the dilution of the serum that inhibited 90% of the virus inoculum. For some samples, a modified version of the PRNT was performed in which only a single dilution (1/10) was tested. A positive or negative test result was reported based on 90% virus plaque inhibition. The viruses used for the PRNT test were: ZIKV Puerto Rico strain PRVABC59 and DENV-2 New Guinea strain.

Flavivirus IgG avidity assay

For the flavivirus IgG avidity assay, we modified our flavivirus microsphere immunofluorescence assay as described in detail in [14] by introducing a urea wash step. Furthermore, we used antihuman-IgG conjugate to restrict detection to only IgG antibodies. Briefly, 10 µl serum samples diluted 1:101 were dispensed in duplicate wells of a 96-well filter plate (Millipore Corporation, Burlington, MA). A total of 50 µl of ZIKV and DENV antigen coated beads were added to all the wells and samples were incubated at 37°C for 30 min. After incubation samples were washed three times with 190 µl wash buffer, half the samples were treated with storage buffer (PBS) while the other half were treated with 8 M urea solution. After a 10-min incubation at room temperature samples were washed three times with 190 µl wash buffer. A total of 50 µl (1 µg/ml) of Phycoerythrin (PE) labeled F(ab′) 2-Goat Anti-Human IgG Fc secondary antibody (Thermo Fisher Scientific, Waltham, MA) was added and samples were incubated at 37°C for 30 min. After incubation samples were washed twice with 190 µl wash buffer, and 125 µl of PBN was added to each of the wells. A total of 120 µl of the samples were then transferred to a flat bottom 96-well plate (Corning Incorporated, Kennebunk, ME). Analysis was completed using a Luminex 200 Analyzer configured to count 100 beads per bead class in a 100 µl sample size.

Calculation of avidity indexes

The avidity index expressed as a percentage was calculated as the ratio of the median fluorescence intensity (MFI) of 100 beads counted on the Zika and dengue NS1 antigens of the wells treated with urea to the MFI of the wells treated with buffer control multiplied by 100. Avidity index was calculated only for samples that tested positive for the presence of antibodies against ZIKV and/or DENV in the flavivirus microsphere immunofluorescence antibody detection assay.

Statistical analysis

Mann–Whitney test was used to determine statistical significance of differences between two groups (recent phase ZIKV vs. late phase ZIKV, recent phase ZIKV vs. DENV, and Late phase ZIKV vs. DENV). One-way ANOVA was performed to compare the variance in the means of the following groups: 1.4–2, 4–4.5, and 5–5.5 months after PCR positivity. Statistical analyses were performed using GraphPad Prism 7 software (San Diego, CA).

Results

In order to discriminate recent Zika from past dengue infections, Zika and dengue IgG antibody avidity was measured by incorporating a urea wash step in our flavivirus MIA assay to remove weakly bound antibodies. This left behind only antibodies with strong affinity for the antigens coated to the microspheres. The percent ZIKV and DENV antibody avidity was calculated by dividing the IgG levels as measured with the urea wash step by IgG levels measured in PBS buffer control with no urea wash multiplied by 100. A low avidity index is indicative of weak antibody binding and thereby a more recent infection, whereas a high avidity index indicates a remote or past infection.

Serum samples from 46 patients were submitted to the Wadsworth Center of the NYSDOH for routine ZIKV testing, therefore subjected to Zika RT-PCR, IgM-capture ELISA, Arbovirus MIA (WN envelope-based MIA), and PRNT testing [22]. These serum samples were obtained from asymptomatic or symptomatic returning travelers with possible exposure to Zika virus. Generally, 80% of individuals exposed to ZIKV are asymptomatic; therefore, symptom onset date is usually unknown in most cases. However, onset dates are crucial to better differentiate recent from past infections and to establish an estimated Zika infection timeline. For the present study, therefore, we attempted to obtain patient specimens with a known disease onset date. Some specimens unfortunately lack the date of onset. Three of such specimens, however, tested positive by ZIKV RT-PCR suggesting an acute infection. Based on symptom onset date and previous PCR and IgM ELISA results the samples were categorized as recent phase (n=28) and late phase (n=18) respectively (Supplementary Table S1 and S2). All the specimens in the recent phase category tested positive by the ELISA IgM test suggesting more recent infection. Specimens in the late phase group were by definition negative using RT-PCR as well as Zika IgM ELISA. Six out of 18 specimens have no onset date. These specimens were included in the late phase category due to either loss of IgM antibodies in specimens collected at a later timepoint or a decline in antibody titers as tested by PRNT (data not shown). Serum samples were first tested by the flavivirus MIA. Those positive for Zika and/or dengue antibodies were then subject to the IgG avidity component of the flavivirus MIA assay. For recent phase specimens, the mean time in days from disease onset to specimen collection was 34 days. ZIKV infected patients in the recent phase category showed low (<25%) ZIKV IgG avidity levels with a mean avidity index of 14.8% (95% CI of the mean: 11.0–18.4%), which agrees with the fact that these patients were either ZIKV-RNA-PCR positive and/or IgM positive, suggesting a recent ZIKV infection (Figure 1A). The trend of increasing ZIKV IgG avidity (25%<AI<55%) with increasing mean time of 102 days after illness onset can be seen in patients with ZIKV infection in the late phase category (Figure 1A). The mean avidity index for late phase infections was 34.9% (95% CI of the mean: 31.1–38.7%), significantly higher than that of recent phase infections (P-value < 0.0001). This finding agrees with the fact that these patients are past the IgM positivity window. PRNT results, however, confirmed their exposure to ZIKV (Supplementary Table S2). ROC of the threshold in those who have had ZIKV infection suggests that an avidity index threshold of >24.5% would give a 94.44% sensitivity with an 85.71% specificity (Figure 1B) in distinguishing more recent phase from late phase ZIKV infection. This threshold, however, is based on a small sample size. A larger reference sample would be needed to achieve adequate power to be able to extend the results of the statistical analysis [24,25]. Considering that DENV is endemic in Central and South America and all our patients traveled to such area, it was likely that they had been previously exposed to DENV. Indeed seven and 11 individuals in the ZIKV recent and late phase group respectively had antibodies against DENV (dengue avidity index % in Supplementary Table S1 and S2). We therefore assessed the ZIKV and DENV avidity indices to determine if past DENV could be distinguished from recent ZIKV infection. We detected high DENV IgG avidity level antibodies in the ZIKV recent group with an average avidity index of 69.9% (95% CI of the mean: 65.1–74.7%). The avidity indices for anti-DENV antibodies were significantly higher in comparison with anti-ZIKV antibodies, which had a mean avidity of 20.9% (P-value 0.0006) (Figure 2A). Similarly, antibodies against DENV were of high avidity in the ZIKV late phase group with a mean index of 68.2% (95% CI of the mean: 58.4–77.9%), significantly higher than the mean avidity index of anti-ZIKV antibodies of 32.5% (P-value < 0.0001) (Figure 2B). These results demonstrate that low avidity anti-ZIKV antibodies are indicative of a recent ZIKV infection, whereas high avidity anti-DENV antibodies confirm a past dengue infection (Figure 2 and Supplementary Figure S1). Furthermore, this is in line with the fact that these patient travelers contracted DENV before ZIKV.

Recent and late phase ZIKV infection is characterized by low versus intermediate avidity index IgG antibodies respectively

Figure 1
Recent and late phase ZIKV infection is characterized by low versus intermediate avidity index IgG antibodies respectively

(A) ZIKV NS1 based IgG antibody avidity of recent phase group of patients (n=28, PCR+ and/or IgM+) with a mean time of 34 days from disease onset was compared with late phase group of patients (n=18, PCR- and IgM-) with a mean time of 102 days from disease onset using microsphere immunofluorescence assay. Mann–Whitney test was performed to determine P-value. ****P<0.0001. (B) ROC analysis of ZIKV IgG avidity index (AUC 0.9306 ± 0.039).

NS1, viral non-structural protein 1; AUC, Area under the curve.

Figure 1
Recent and late phase ZIKV infection is characterized by low versus intermediate avidity index IgG antibodies respectively

(A) ZIKV NS1 based IgG antibody avidity of recent phase group of patients (n=28, PCR+ and/or IgM+) with a mean time of 34 days from disease onset was compared with late phase group of patients (n=18, PCR- and IgM-) with a mean time of 102 days from disease onset using microsphere immunofluorescence assay. Mann–Whitney test was performed to determine P-value. ****P<0.0001. (B) ROC analysis of ZIKV IgG avidity index (AUC 0.9306 ± 0.039).

NS1, viral non-structural protein 1; AUC, Area under the curve.

ZIKV IgG avidity is lower than DENV IgG avidity of the same individual

Figure 2
ZIKV IgG avidity is lower than DENV IgG avidity of the same individual

(A) ZIKV NS1 based IgG avidity of the recent stage of infection was compared with the same patient’s DENV IgG avidity levels (n=7), and (B) ZIKV NS1 based IgG avidity of the late stage of infection was compared with the same patient’s DENV IgG avidity levels (n=11) using microsphere immunofluorescence assay. Mann–Whitney test was performed to determine P-value. ***P<0.0006, ****P<0.0001.

Figure 2
ZIKV IgG avidity is lower than DENV IgG avidity of the same individual

(A) ZIKV NS1 based IgG avidity of the recent stage of infection was compared with the same patient’s DENV IgG avidity levels (n=7), and (B) ZIKV NS1 based IgG avidity of the late stage of infection was compared with the same patient’s DENV IgG avidity levels (n=11) using microsphere immunofluorescence assay. Mann–Whitney test was performed to determine P-value. ***P<0.0006, ****P<0.0001.

It is widely accepted that recent infections are characterized by low avidity antibodies, whereas past infections are characterized by high avidity antibodies. While our findings comparing IgG avidity indices of recent and late phase ZIKV infection further support the immunological process of affinity maturation, it also has its limitations: (1) only three of the recent phase specimens were PCR positive for ZIKV, (2) onset dates were not available for all the specimens, (3) the sample size was small for both recent and late phase groups, and (4) sequential samples from longitudinal studies were not available for testing. To confirm our findings, we obtained additional specimens from BARDA. These samples are single timepoint convalescent specimens collected 1.4–5.5 months after the initial ZIKV PCR positive result. All samples were also confirmed to have ZIKV and/or DENV antibodies by the gold standard PRNT (Supplementary Table S3). To evaluate these samples, we have performed an IgG avidity test by microsphere immunofluorescence assay. Indeed, we detected lower avidity ZIKV IgG antibodies in the 1.4–2 months group with an average avidity index of 44.2% (95% CI of the mean: 39.2–49.4%), while the 4–4.5 months group had a slightly higher avidity index of 53.5% (95% CI of the mean: 46.7–60.4%). Finally, ZIKV IgG avidity index of serum samples collected 5–5.5 months from the initial PCR positive results show an even higher level of avidity with an average of 68.6% (95% CI of the mean: 60.5–76.8%) compared with the 1.4–2 and 4–4.5 groups (Figure 3). These results also suggest that ZIKV IgG avidity increases significantly with time after illness onset (P=0.01). Considering that Puerto Rico is a dengue endemic region and New York residents travel to such areas, it was likely that these individuals were exposed to DENV. Indeed, 18 of the 24 samples were also positive for anti-DENV antibodies (Supplementary Table S3). Consistent with our results, this set of patients also exhibited high avidity antibodies against DENV with a mean avidity index of 79.2% that was significantly higher (P<0.0001) than the overall ZIKV avidity index of 47.9% (Supplementary Figure S2). Overall, these results are consistent with low affinity IgG responses post recent ZIKV infection that increases steadily over time, while the observed higher avidity anti-DENV IgG antibodies are consistent with more remote exposures.

ZIKV IgG avidity increases with time after illness onset

Figure 3
ZIKV IgG avidity increases with time after illness onset

ZIKV NS1 IgG avidity levels was measured for serum samples 1.4–2 (n=18), 4–4.5 (n=4), and 5–5.5 (n=2) months after PCR positivity using microsphere immunofluorescence assay. One-way ANOVA was performed to compare the variance in the group means. P=0.0103.

Figure 3
ZIKV IgG avidity increases with time after illness onset

ZIKV NS1 IgG avidity levels was measured for serum samples 1.4–2 (n=18), 4–4.5 (n=4), and 5–5.5 (n=2) months after PCR positivity using microsphere immunofluorescence assay. One-way ANOVA was performed to compare the variance in the group means. P=0.0103.

Discussion

Zika is a relatively new virus in the Americas. Although the virus was discovered in Uganda in 1947 [26], it has not been a global concern until recently. The first major Zika virus outbreak occurred on Yap Island (Federated States of Micronesia) in 2007 [6,27]. This was followed by the outbreak in French Polynesia in 2012–2013 where 10,000 cases were recorded before moving into South America in particularly Brazil in May 2015 [28,29]. Because people exposed to ZIKV encountered the virus for the first time in their lives, it is expected that these people will have low avidity antibodies. Our and others’ results concur with this assumption. In the study by Zhang et al., low-avidity Zika antibodies with <50% AI corresponded to less than 100 days after illness onset [11]. Consistent with their findings, we also detected low avidity ZIKV IgG antibodies in the sera of both recent phase (ZIKV IgM ELISA and/or PCR positive) and late phase (ZIKA IgM ELISA negative) group of patients. For example, recent phase group of patients had low IgG avidity indices with a mean index of 14.8%, whereas the late phase group had slightly higher IgG avidity indices with a mean index of 34.8%, albeit still significantly lower compared with the combined (patients of both recent and late phase groups) mean dengue avidity index of 68.9%. Given that avidity of antibodies increases with time after initial infection, it is not surprising to detect a small but statistically significant increase in ZIKV IgG avidity indices in the late phase group. To definitively evaluate the performance of the ZIKV IgG avidity test for the ‘past infection’ category, it would be necessary to obtain serum samples from the same patients presented in the present study at a later timepoint from the initial infection. Various studies suggest that antibodies produced by 6 months post-infection exhibit high avidity [11,15,30–32]. Unfortunately, however, collecting a follow-up serum specimen for ZIKV IgG avidity testing will not be possible. This is due to the fact that the specimens submitted for ZIKV and DENV diagnostic testing to the Wadsworth Center (New York State’s public health reference laboratory) during the height of the Zika epidemic were from patients with recent travels to Zika and dengue endemic areas. Once diagnostic testing was completed, clinicians did not summon patients for further testing. Therefore, no follow-up serum samples are available on the patients presented in the present study. Avidity assay performed on samples collected 1.4–5.5 months after the initial PCR positive result; however, show a gradual increase in antibody avidity after illness onset. Nevertheless, as we look to the future serum samples coming in for ZIKV and DENV serologic testing would enable us to better characterize past ZIKV infections.

In addition to ZIKV antibodies, we also detected antibodies against DENV in 18 out of the 46 patient specimens submitted to the Wadsworth Center and 18 out of the 24 specimens obtained from BARDA. Considering that DENV is endemic in Central and South America and all the patients have traveled or reside in such areas, it was likely that these individuals had been previously exposed to DENV. Indeed, we detected high (>60% AI) DENV IgG avidity level antibodies in these patients confirming a past dengue infection (Figure 2 and Supplementary Figure S2). Work by Zhang et al. also found high IgG antibody levels (>60%) to all four serotypes of DENV in their cohort of Colombian patients [11]. Similarly, Matheus et al. and de Souza et al. have assessed IgG antibody avidity to discriminate between primary and secondary DENV infections. Their studies have shown that the mean percent avidity was significantly lower in primary infections (25–27%), than in secondary infections (50–66%) [19,33].

Cross-reactive antibodies induced by closely related viruses like Zika, dengue, West Nile, Japanese encephalitis, Yellow fever, or vaccination complicate flavivirus serology. Secondary infections that trigger the rapid reappearance of cross-reactive antibodies further complicate diagnostic testing. In addition, anamnestic responses resulting from previous flavivirus infections boost high avidity antibody titers that make serology even more complicated. The avidity testing by microsphere immunofluorescence assay of anti-ZIKV IgG antibodies has several advantages: (1) anti-ZIKV IgG antibodies appear early in infection, (2) cross-reactivity of ZIKV NS1 IgG to DENV NS1 IgG is relatively low and (3) IgG antibodies tend to persist in sera of infected patients, (4) less than 4 h turnaround time. This is in stark contrast with IgM testing, which is troubled by cross-reactivity in addition to the transient nature of anti-ZIKV IgM antibodies particularly in persons with a history of past flavivirus infection. The addition of the IgG avidity assay to the current ZIKV and DENV antibody detection methods is a potentially valuable diagnostic tool for evaluating recent and past flavivirus infections in dengue-endemic areas and could be applied in clinical laboratories as potential routine test for ZIKV and DENV virus infections. Because our avidity assay exploits the idea that antibodies against flavivirus non-structural protein 1 (NS1) are highly virus specific, it allows for the better differentiation of acute flavivirus infection from past infection or vaccination with high specificity and sensitivity despite antibody cross-reactivity [34]. Although our results point to a more recent ZIKV and past DENV infections, it is important to note that avidity assay can also detect low avidity anti-DENV antibodies indicative of a more recent infection. Therefore, avidity assay would be a valuable diagnostic tool for pregnant women living in Zika endemic areas and also for those who travel to such areas and face the dilemma concerning conception, an issue discussed in greater detail by Graciaa et al. [35]. Finally, the assay has the potential to be used as an important research tool for serosurveillance.

Clinical perspective

  • We developed a novel serological assay to distinguish recent Zika from past dengue infection.

  • We detected low avidity Zika IgG antibodies and high avidity dengue IgG antibodies confirming recent Zika and past dengue infection.

  • This assay could be very useful in patients with high risk of Zika complications such as pregnant women and monitoring immune responses in vaccine trials.

Compliance with ethical standards

The present study was performed retrospectively using only leftover diagnostic specimens that were submitted to the Wadsworth Center laboratory with general consent for routine clinical testing and deidentified prior to the study. This research was conducted according to protocol 03-037 which was approved as an exempt study by NYSDOH Institutional Review Board (IRB). The NYSDOH IRB operates under the United States Department of Health and Human Services regulations for the protection of human subjects in research [45 CFR 46.101(b)(4)] which incorporate all ethical principles of the 1964 Declaration of Helsinki. In accordance with these regulations, the present study protocol met all criteria for a waiver of informed consent.

Acknowledgements

Authors would like to acknowledge and thank the members of the diagnostic immunology laboratory for their technical assistance namely: W. Lee, K. Kulas, V. Demarest, M. Marchewka, K. Carson, S. Bush, S. Casterlin; the virology laboratory namely: A. Dean, M. Fuschino, R. Hull, D. Lamson, P. Bryant, M. Popowich, S. Brunt, L.Zeng; arbovirology laboratory namely: A. Ciota, S. Jones, J.Maffei. We thank the NYSDOH Division of Epidemiology ZIKV Virus group namely: N. Ahmad, B. Backenson, D. Blog, E. Dufort, P. Huthfor, D. Kuhles, L. Lance, S. Ostrowski, L. Smith, J. Sommer, J. White for their contributions. Authors would like to thank Drs. W. Lee and R. Limberger for the critical reading of this manuscript and their helpful discussions. We would like to acknowledge the Wadsworth Center core facilities that contributed to this work: WC Tissue Culture and Media. Additional samples tested using our laboratory developed flavivirus microsphere immunofluorescence avidity assay were kindly provided by the BARDA-Centers for Disease Control and Prevention Zika Repository.

Funding

This work was supported by the New York State Department of Health (NYSDOH). A portion of the work described in this publication was supported by Co-operative Agreement Number NU50CK000423 from the Centers for Disease Control and Prevention. Its contents are solely the responsibility of the authors and do not necessarily represent the official views of the Centers for Disease Control and Prevention.

Author contribution

Conceived and designed the experiments: A.K.M.F. and S.J.W. Performed the experiments: A.K.M.F. and D.H. Analyzed the data: A.K.M.F., D.H., K.S.G., A.P.D., L.D.K., P.Y.S. and S.J.W. Wrote the paper: A.K.M.F., K.S.G. and S.J.W.

Competing interests

The authors declare that there are no competing interests associated with the manuscript.

Abbreviations

     
  • BARDA

    Biomedical Advanced Research and Development Authority

  •  
  • CDC

    Centers for Disease Control

  •  
  • DENV

    dengue virus

  •  
  • IgG

    immunoglobulin G

  •  
  • IRB

    Institutional Review Board

  •  
  • MFI

    median fluorescence intensity

  •  
  • MIA

    microsphere immunoassay

  •  
  • NYSDOH

    New York State Department of Health

  •  
  • PBN

    phosphate buffered saline containing 0.05% sodium azide

  •  
  • PRNT

    plaque reduction neutralization test

  •  
  • ROC

    receiver operating characteristic

  •  
  • RT-PCR

    Reverse Transcription Polymerase Chain Reaction

  •  
  • WNV

    West Nile virus

  •  
  • YFV

    yellow fever virus

  •  
  • ZIKV

    Zika virus

References

References
1.
Patterson
J.
,
Sammon
M.
and
Garg
M.
(
2016
)
Dengue, Zika and chikungunya: emerging arboviruses in the new world
.
West J. Emerg. Med.
17
,
671
679
[PubMed]
2.
Brasil
P.
et al.
(
2016
)
Zika virus infection in pregnant women in Rio de Janeiro
.
N. Engl. J. Med.
375
,
2321
2334
[PubMed]
3.
Calvet
G.
et al.
(
2016
)
Detection and sequencing of Zika virus from amniotic fluid of fetuses with microcephaly in Brazil: a case study
.
Lancet Infect. Dis.
16
,
653
660
[PubMed]
4.
Mlakar
J.
et al.
(
2016
)
Zika virus associated with microcephaly
.
N. Engl. J. Med.
374
,
951
958
[PubMed]
5.
Broutet
N.
et al.
(
2016
)
Zika virus as a cause of neurologic disorders
.
N. Engl. J. Med.
374
,
1506
1509
[PubMed]
6.
Duffy
M.R.
et al.
(
2009
)
Zika virus outbreak on Yap Island, Federated States of Micronesia
.
N. Engl. J. Med.
360
,
2536
2543
[PubMed]
7.
Prevention, C.f.D.C.a. Diagnostic Tests for Zika Virus. 2017 December 12, 2017 [cited 2018 05/15/2018]; https://www.cdc.gov/zika/hc-providers/types-of-tests.html
8.
Felix
A.C.
et al.
(
2017
)
Cross reactivity of commercial anti-dengue immunoassays in patients with acute Zika virus infection
.
J. Med. Virol.
89
,
1477
1479
[PubMed]
9.
Prevention, C.f.D.C.a., Zika MAC-ELISA, For use under an emergency use authorization only. 2016 www.cdc.gov: https://www.cdc.gov/zika/pdfs/zika-mac-elisa-protocol-eua-english.pdf
10.
Lee
W.T.
et al.
(
2018
)
Development of Zika virus serological testing strategies in New York state
.
J. Clin. Microbiol.
56
,
e01591
17
[PubMed]
11.
Zhang
B.
et al.
(
2017
)
Diagnosis of Zika virus infection on a nanotechnology platform
.
Nat. Med.
23
,
548
550
[PubMed]
12.
Oduyebo
T.
et al.
(
2017
)
Update: interim guidance for health care providers caring for pregnant women with possible zika virus exposure - United States (including U.S. territories), July 2017
.
Morb. Mortal. Wkly. Rep.
66
,
781
793
[PubMed]
13.
Roehrig
J.T.
,
Hombach
J.
and
Barrett
A.D.
(
2008
)
Guidelines for plaque-reduction neutralization testing of human antibodies to dengue viruses
.
Viral Immunol.
21
,
123
132
[PubMed]
14.
Wong
S.J.
et al.
(
2017
)
A multiplex microsphere immunoassay for Zika virus diagnosis
.
EBioMedicine
16
,
136
140
[PubMed]
15.
Prince
H.E.
and
Lape-Nixon
M.
(
2014
)
Role of cytomegalovirus (CMV) IgG avidity testing in diagnosing primary CMV infection during pregnancy
.
Clin. Vaccine Immunol.
21
,
1377
1384
[PubMed]
16.
Parekh
B.S.
et al.
(
2002
)
Quantitative detection of increasing HIV type 1 antibodies after seroconversion: a simple assay for detecting recent HIV infection and estimating incidence
.
AIDS Res. Hum. Retroviruses
18
,
295
307
[PubMed]
17.
Gassmann
C.
and
Bauer
G.
(
1997
)
Avidity determination of IgG directed against tick-borne encephalitis virus improves detection of current infections
.
J. Med. Virol.
51
,
242
251
[PubMed]
18.
Fox
J.L.
et al.
(
2006
)
Immunoglobulin G avidity in differentiation between early and late antibody responses to West Nile virus
.
Clin. Vaccine Immunol.
13
,
33
36
[PubMed]
19.
de Souza
V.A.
et al.
(
2004
)
Use of an immunoglobulin G avidity test to discriminate between primary and secondary dengue virus infections
.
J. Clin. Microbiol.
42
,
1782
1784
[PubMed]
20.
Prevention, C.f.D.C.a. Dengue Laboratory Guidance and Diagnostic Testing. 2017 August 3, 2017 [cited 2018 Januray 15, 2018]; https://www.cdc.gov/dengue/clinicallab/laboratory.html
21.
Wong
S.J.
et al.
(
2004
)
Detection of human anti-flavivirus antibodies with a west nile virus recombinant antigen microsphere immunoassay
.
J. Clin. Microbiol.
42
,
65
72
[PubMed]
22.
Wong
S.J.
et al.
(
2003
)
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.
41
,
4217
4223
[PubMed]
23.
Lindsey
H.S.
,
Calisher
C.H.
and
Mathews
J.H.
(
1976
)
Serum dilution neutralization test for California group virus identification and serology
.
J. Clin. Microbiol.
4
,
503
510
[PubMed]
24.
Hajian-Tilaki
K.
(
2014
)
Sample size estimation in diagnostic test studies of biomedical informatics
.
J. Biomed. Inform.
48
,
193
204
[PubMed]
25.
Faber
J.
and
Fonseca
L.M.
(
2014
)
How sample size influences research outcomes
.
Dental Press J. Orthod.
19
,
27
29
[PubMed]
26.
Dick
G.W.
,
Kitchen
S.F.
and
Haddow
A.J.
(
1952
)
Zika virus. I. Isolations and serological specificity
.
Trans. R. Soc. Trop. Med. Hyg.
46
,
509
520
[PubMed]
27.
Lanciotti
R.S.
et al.
(
2008
)
Genetic and serologic properties of Zika virus associated with an epidemic, Yap State, Micronesia, 2007
.
Emerg. Infect. Dis.
14
,
1232
1239
[PubMed]
28.
Oehler
E.
et al.
(
2014
)
Zika virus infection complicated by Guillain-Barre syndrome–case report, French Polynesia, December 2013
.
Euro Surveill.
19
,
pii: 20720
[PubMed]
29.
Pan American Health Organization, W.H.O., Regional Office for the Americas. Increase of microcephaly in the northeast of Brazil. Epidemiological Alert 2015 Tuesday, 17 November 2015 [cited 2017 December 4]; http://www.paho.org/hq/index.php?option=com_docman&task=doc_view&Itemid=270&gid=32285&lang=en
30.
Revello
M.G.
and
Gerna
G.
(
2002
)
Diagnosis and management of human cytomegalovirus infection in the mother, fetus, and newborn infant
.
Clin. Microbiol. Rev.
15
,
680
715
[PubMed]
31.
Hazell
S.L.
(
2007
)
Clinical utility of avidity assays
.
Expert Opin. Med. Diagn.
1
,
511
519
[PubMed]
32.
Tsai
W.Y.
et al.
(
2018
)
Use of urea wash ELISA to distinguish zika and dengue virus infections
.
Emerg. Infect. Dis.
24
,
1355
1359
[PubMed]
33.
Matheus
S.
et al.
(
2005
)
Discrimination between primary and secondary dengue virus infection by an immunoglobulin G avidity test using a single acute-phase serum sample
.
J. Clin. Microbiol.
43
,
2793
2797
[PubMed]
34.
Stettler
K.
et al.
(
2016
)
Specificity, cross-reactivity, and function of antibodies elicited by Zika virus infection
.
Science
353
,
823
826
[PubMed]
35.
Graciaa
D.S.
,
Collins
M.H.
and
Wu
H.M.
(
2018
)
Zika in 2018: advising travelers amid changing incidence
.
Ann. Intern. Med.
169
,
337
338
[PubMed]