Dengue virus antibodies enhance Zika virus infection

Abstract

For decades, human infections with Zika virus (ZIKV), a mosquito-transmitted flavivirus, were sporadic, associated with mild disease, and went underreported since symptoms were similar to other acute febrile diseases. Recent reports of severe disease associated with ZIKV have greatly heightened awareness. It is anticipated that ZIKV will continue to spread in the Americas and globally where competent Aedes mosquito vectors are found. Dengue virus (DENV), the most common mosquito-transmitted human flavivirus, is both well-established and the source of outbreaks in areas of recent ZIKV introduction. DENV and ZIKV are closely related, resulting in substantial antigenic overlap. Through antibody-dependent enhancement (ADE), anti-DENV antibodies can enhance the infectivity of DENV for certain classes of immune cells, causing increased viral production that correlates with severe disease outcomes. Similarly, ZIKV has been shown to undergo ADE in response to antibodies generated by other flaviviruses. We tested the neutralizing and enhancing potential of well-characterized broadly neutralizing human anti-DENV monoclonal antibodies (HMAbs) and human DENV immune sera against ZIKV using neutralization and ADE assays. We show that anti-DENV HMAbs, cross-react, do not neutralize, and greatly enhance ZIKV infection in vitro. DENV immune sera had varying degrees of neutralization against ZIKV and similarly enhanced ZIKV infection. Our results suggest that pre-existing DENV immunity may enhance ZIKV infection in vivo and may lead to increased disease severity. Understanding the interplay between ZIKV and DENV will be critical in informing public health responses and will be particularly valuable for ZIKV and DENV vaccine design and implementation strategies.

Figure 1

ELISA and immunostaining cross-reactivity of anti-DENV HMAbs against ZIKV. Anti-DENV HMAbs 1.6D and D11C that recognize the DENV E protein fusion loop cross-react with ZIKV MR766 strain E surface glycoprotein as shown by ELISA (a, 1.6D; b, D11C) and recognize ZIKV MR766 infected cells in an immunostained focus-forming assay (c, 1.6D; d, D11C). DENV E is serotype 2, strain New Guinea-C. Absorbance from no antigen control reactions was used to normalize the results. ELISA data shown are representative of two independent assays each done in triplicate. 10 and 100 ng ml⁻¹ concentrations were statistically significantly higher than no antigen controls by analysis of variance with Dunnett's post hoc, P<0.05.

Figure 2

Infectivity neutralizing activity of anti-DENV HMAbs against ZIKV MR766. Broadly neutralizing anti-DENV HMAbs 1.6D and D11C do not inhibit ZIKV MR766 infection in LLC-MK2 cells at the concentrations tested. However, a positive control neutralizing anti-ZIKV antibody (ZKA64) showed clear concentration-dependent inhibition. The results shown are the average±the s.d. of six replicates. None of the concentrations of HMAbs 1.6D or D11C tested showed statistically significant neutralizing activity compared to no antibody controls by analysis of variance with Dunnett's post hoc, P>0.05.

Figure 3

Enhancing activity of anti-DENV HMAbs against ZIKV by qRT–PCR. Broadly neutralizing anti-DENV HMAbs 1.6D and D11C show strong ZIKV MR766 infection enhancing activity. Independent assays were repeated twice in triplicate. All concentrations above 0.2 μgml⁻¹ were statistically higher than no antibody controls by analysis of variance with Dunnett's post hoc, P<0.05.

Figure 4

Infectivity neutralizing activity of anti-DENV human sera against DENV. All anti-DENV human sera showed broad neutralizing activity against multiple DENV serotypes 1–4. (a) Singapore 1, (b) Singapore 2, (c) Jamaica 1 and (d) Jamaica 2. All concentrations above 1:1000 dilution showed statistically significant inhibition of infection compared to no serum controls by analysis of variance with Dunnett's post hoc, P<0.05.

Figure 5

Infectivity neutralizing activity of anti-DENV human sera against ZIKV. Human anti-DENV sera from Singapore and Jamaica showed both non-neutralizing and neutralizing activity against ZIKV MR766. Singapore 1 serum strongly neutralized ZIKV MR766, suggesting prior ZIKV infection, while Singapore 2 serum had no neutralizing activity. Jamaica 1 serum neutralized ZIKV MR766 at high serum concentrations, while Jamaica 2 serum showed no neutralizing activity at the dilutions tested. Control serum from Canada showed no ZIKV neutralizing activity. The results shown are the average±the s.d. of six replicates. All dilutions of Singapore 1 serum as well as 1:100 and 1:50 dilutions of Jamaica 1 serum showed statistically significant inhibition of infectivity by analysis of variance with Dunnett's post hoc, P<0.05. No other sera showed statistically significant infectivity neutralizing activity.

Figure 6

Enhancing activity of anti-DENV human sera against ZIKV by qRT–PCR and infectious particle assay. (a) The effect of anti-DENV human sera on enhancement of ZIKV MR766 infection was determined by qRT–PCR in the human FcRII bearing cell line K562. All human anti-DENV sera tested showed strong infection enhancing activity of ZIKV MR766. At high serum concentrations, Singapore 1 serum blocked enhancement due to its strong neutralizing activity. Independent assays were repeated twice in triplicate. Compared with no serum controls, Singapore 1 serum showed statistically significant levels of enhancement at 1:100 000–1:10 000 dilutions, Singapore 2 serum at 1:50 000 and lower dilutions, Jamaica 1 serum at 1:10 000 and lower dilutions, and Jamaica 2 serum at 1:1000 and lower dilutions by analysis of variance with Dunnett's post hoc, P<0.05. Canadian serum showed no statistically significant effect at any dilution. (b) The effect of anti-DENV human sera on the production of infectious progeny viral particles was determined in the human FcRII bearing cell line K562 infected with ZIKV MR766 in the presence (1:50 000 dilution) or absence (0=no serum control) of Singapore 1 serum. The presence of enhancing serum resulted in a large and statistically significant increase in the amount of infectious progeny virus released (T-test, P=0.00016).

Figure 7

Anti-FcRII antibody blocks ZIKV enhancement activity of anti-DENV serum. K562 cells were pre-incubated with increasing concentrations of mouse anti-FcRII (CD32) MAb prior to infection with ZIKV MR766 that had been pre-incubated with a highly enhancing dilution (1:50 000) of Singapore 1 serum. The results indicate that the ZIKV enhancement effect can be effectively blocked in a dose-responsive manner with an anti-FcRII MAb. This effect was statistically significant at all concentrations by analysis of variance with Dunnett's post hoc, P<0.05.

Figure 8

Enhancing activity of anti-DENV human sera against ZIKV PRVABC59 by infectious particle and qRT–PCR assays. The effect of anti-DENV human sera on the production of (a) replication of viral RNA and on (b) infectious progeny viral particles was determined in the human FcRII bearing cell line K562 infected with ZIKV PRVABC59 in the presence (1:50 000 dilution) or absence (0=no serum control) of Singapore 1 serum. The presence of enhancing serum resulted in a large and statistically significant increase in the amount of viral RNA (T-test, P=0.000015) as well as infectious progeny virus released (T-test, P=0.00077).