Diagnostics for Zika virus on the horizon

See allHide authors and affiliations

Science  19 Aug 2016:
Vol. 353, Issue 6301, pp. 750-751
DOI: 10.1126/science.aah6187

Embedded Image

Well-characterized antibodies to ZIKV (pink; transmission electron micrograph shown) are needed.


Zika virus (ZIKV) is a mosquito-transmitted flavivirus that is related to other pathogens of clinical importance, including yellow fever and dengue (DENV) viruses. Although once infrequently associated with human disease, ZIKV has emerged as a global health threat with its introduction into South America during 2014 and 2015. Of concern, recent ZIKV outbreaks are linked to severe neuro-developmental complications in the children of women infected while pregnant, as well as Guillain-Barré syndrome in adults (1). Management of this epidemic has been complicated by extensive serological cross-reactivity among flaviviruses and the cocirculation of ZIKV and DENV in regions experiencing the greatest disease burden. Current serological diagnostics have a limited capacity to distinguish between DENV and ZIKV. On page 823 of this issue, Stettler et al. (2) characterize monoclonal antibodies (mAbs) isolated from ZIKV-infected humans that hold promise as diagnostics or therapeutics, and advance our understanding of the repertoire of antibodies elicited by ZIKV infection.

Flaviviruses are assembled from three viral structural proteins [capsid, premembrane (prM), and envelope (E)], a host-derived lipid envelope, and the genomic viral RNA (3). Flavivirus-infected cells also secrete a nonstructural protein 1 (NS1), which has multiple roles in viral replication and pathogenesis in vivo (4). Both NS1 and the structural proteins are immunogenic. Virus-neutralizing antibodies most commonly target the E protein, may be highly protective in vivo, and are a correlate of protection for many flavivirus vaccines (5). NS1 antibodies are non-neutralizing, yet they contribute to protection via antibody heavy chain–mediated effector functions (6). Whereas the functional characteristics of antibodies in ZIKV-immune individuals have been studied (7), human ZIKV mAbs have not been reported.

Accordingly, Stettler et al. have now isolated a panel of 119 human mAbs from the memory B cells of four ZIKV-infected donors; two of these subjects had been infected previously by DENV (2). Roughly two-thirds of the mAbs produced bound epitopes within the E protein. Antibodies specific for the immunoglobulin (Ig)–like domain III (DIII) had considerable neutralizing activity and were largely specific for either ZIKV or DENV. Numerous cross-reactive mAbs with modest neutralization capacity mapped to E protein domains I or II. Domain II is the location of the highly conserved fusion loop frequently targeted by antibodies elicited by other flaviviruses. Although more study is required, cross-reactive fusion loop–specific antibodies may also be common in ZIKV-immune individuals. Of considerable interest, Stettler et al. found that the most potent neutralizing mAbs bound efficiently to intact virions but not to soluble forms of the E protein, which suggests that antibodies that bind quaternary epitopes composed of more than a single ZIKV E protein may be desirable. In agreement, three recent studies detail the recognition and functional properties of neutralizing mAbs that bind a quaternary epitope shared by ZIKV and DENV (810). E protein antigens and immunogens that include the antiparallel dimers found on virions may be required to capture the full complexity of the humoral immune response when used as diagnostics and vaccines, respectively.

Neutralizing mAbs have potential as therapeutics capable of preventing or limiting disease, when administered after infection. Stettler et al. demonstrated that at least one neutralizing ZIKV type–specific DIII-reactive mAb, genetically modified to prevent antibody-FcγR interactions, was protective when administered 1 day before or after lethal challenge of immune-compromised mice. Similar murine ZIKV DIII–specific mAbs with therapeutic potential were also recently reported (11). In this second study, mAbs were produced from mice immunized with ZIKV and recombinant DIII. Structural studies revealed that the two most potent mAbs were specific for a DIII lateral ridge (DIII-LR) epitope on the surface of intact mature virions. Strikingly, this same DIII-LR epitope is also the target of a potent neutralizing West Nile virus–specific antibody that was developed as a potential therapeutic (12). Whether robustly neutralizing mAbs will have utility for limiting infection and disease in the unborn requires further study in recently developed animal models.

The complex serology of ZIKV virus infection

The envelope proteins of ZIKV and DENV share a considerable degree of amino acid homology. Both ZIKV and DENV infection result in the production of antibodies specific for only one of these two viruses [type-specific (TS) antibody] as well as cross-reactive (CR) antibodies capable of binding both viruses to varying degrees. The presence of CR antibody complicates the development of specific and sensitive serological tests to diagnose ZIKV infection in populations frequently exposed to other flaviviruses such as DENV.


NS1-reactive mAbs were also present in the panel of mAbs described by Stettler et al. Binding studies with recombinant NS1 from ZIKV and the four DENV serotypes revealed that most of these mAbs were virus type–specific. Antibody competition studies identified two sites on NS1 recognized by ZIKV-specific mAbs. While cross-reactive antibodies that target NS1 exist (2), the use of NS1 as a more specific antigen in diagnostics has been proposed (13).

The extensive serologic cross-reactivity among flaviviruses, and especially with DENV, poses an obstacle for the specific diagnosis of infection and management of disease, particularly in pregnant women. Existing diagnostic assays have limitations. Multiple commercial real-time reverse transcription–polymerase chain reaction assays for the detection of ZIKV RNA are now in use but are only effective during the brief window when viral RNA is detectable in blood, urine, or other fluids. Serological assays, such as the IgM-capture enzyme-linked immunosorbent assay (MAC-ELISA, with virus particles as antigen) and the plaque reduction neutralization test, have the potential to detect prior infection over longer intervals, but the specificity of these techniques is uncertain (14).

A detailed understanding of epitopes recognized by ZIKV-specific and cross-reactive antibodies has the potential to improve the specificity of diagnostics in multiple ways. For example, recent structural insights into ZIKV type–specific and cross-reactive epitopes on the E protein may inform the design of novel antigens in which cross-reactive determinants are destroyed (9, 11, 15). More detailed mapping of the E and NS1 epitopes bound by members of the large panel of mAbs described here will advance such efforts considerably (2). The availability of ZIKV-specific mAbs may be used directly to enhance the specificity of diagnostic tests. Here, Stettler et al. provide proof-of-principle for the utility of ZIKV NS1–specific mAbs in a competition ELISA format to more stringently detect virus-specific patterns of recognition in the polyclonal antibodies of ZIKV- and DENV-exposed individuals. Thus, beyond their therapeutic potential, well-characterized mAbs will accelerate the development of approaches to decipher the complex serology of flaviviruses that limits our ability to diagnose and study infection in the field.

References and Notes

Acknowledgments: Supported by the intramural program of NIAID.

Navigate This Article