Pre-Existing Dengue Immunity Drives a DENV-Biased Plasmablast Response in ZIKV-Infected Patient

Abstract

The re-emergence of Zika virus (ZIKV) in the western hemisphere has most significantly affected dengue virus (DENV) endemic regions. Due to the geographical overlap between these two closely related flaviviruses, numerous individuals who suffered ZIKV infection during recent outbreaks may have also previously been exposed to DENV. As such, the impact of pre-existing dengue immunity on immune responses to ZIKV has been an area of focused research and interest. To understand how B cell responses to a ZIKV infection may be modulated by prior dengue exposures, we compared and contrasted plasmablast repertoire and specificity between two ZIKV-infected individuals, one dengue-naïve (ZK018) and the other dengue-experienced (ZK016). In addition to examining serological responses, we generated 59 patient plasmablast-derived monoclonal antibodies (mAbs) to define the heterogeneity of the early B cell response to ZIKV. Both donors experienced robust ZIKV-induced plasmablast expansions early after infection, with comparable mutational frequencies in their antibody variable genes. However, notable differences were observed in plasmablast clonality and functional reactivity. Plasmablasts from the dengue-experienced donor ZK016 included cells with shared clonal origin, while ZK018 mAbs were entirely clonally unrelated. Both at the mAb and plasma level, ZK016 antibodies displayed extensive cross-reactivity to DENV1-4, and preferentially neutralized DENV compared to ZIKV. In contrast, the neutralization activity of ZK018 mAbs was primarily directed towards ZIKV, and fewer mAbs from this donor were cross-reactive, with the cross-reactive phenotype largely limited to fusion loop-specific mAbs. ZK016 antibodies caused greater enhancement of DENV2 infection of FcRγ-expressing cells overall compared to ZK018, with a striking difference at the plasma level. Taken together, these data strongly suggest that the breadth and protective capacity of the initial antibody responses after ZIKV infection may depend on the dengue immune status of the individual. These findings have implications for vaccine design, given the likelihood that future epidemics will involve both dengue-experienced and naïve populations.

Keywords: B-cell; DENV; ZIKV; antibodies; cross-reactivity; plasmablast.

Figure 1

ZIKV neutralizing antibody titers detectable within days after ZIKV infection. (A) ZIKV-specific plasma IgM and IgG titers for patient ZK016 at 3 and 7 DPO and for ZK018 at 8 DPO. Titers were determined by ZIKV IgM antibody capture enzyme linked immunosorbent assay (MAC-ELISA) and modified IgG ELISA. The values are represented as ratios of the optical densities of patient plasma to healthy control plasma. For IgM, a ratio >3 = positive (dark gray), 2–3 = equivocal (light gray), >2 = negative (white) and for IgG, a ratio >1.5 = positive (dark gray), 1.2–1.5 = equivocal (light gray), >1.2 = negative (white). (B) Plasma endpoint IgG binding titers to ZIKV lysate as determined by ELISA. The dotted line indicates the starting plasma dilution of 1:100. (C) The Focus Reduction Neutralization Test (FRNT) titers of plasma from patients against ZIKV and DENV1-4. The values plotted are the plasma dilutions required for 50% neutralization of ZIKV foci, termed FRNT50. The FRNT assay for each sample was repeated in two or more independent experiments, and the mean values are plotted. The dotted line represents the starting plasma dilution of 1:30.

Figure 2

Virus-specific plasmablasts appear in circulation early after ZIKV infection. (A) Flow cytometry plots showing patient plasmablast (CD3− CD19+ CD20−/low CD38high CD27high) frequencies as a fraction of total CD19+ B cells. PBMCs were isolated from whole blood collected from patient ZK016 (7 DPO) and patient ZK018 (8 DPO) for flow cytometry analysis and plasmablast sorting. The workflow chart follows the generation and characterization of mAbs from single-cell sorted plasmablasts. (B) Representative ELISpot showing ZIKV E protein and NS1 protein-specific antibody secreting cells (ASCs). The wells shown contain 7 DPO PBMCs from patient ZK016. The percentages beside the wells represent the frequency of antigen-specific ASCs relative to the total number of IgG + ASCs.

Figure 3

Clonality and mutational characteristics of the ZIKV plasmablast response. (A) Clonal relatedness of heavy chain variable region (VH) sequences amplified from patient plasmablasts. The number at the center of each pie chart is the total number of heavy chain sequences analyzed, including unpaired VH sequences that were not pursued for mAb synthesis. The numbers within the pie slices denote the number of clones with a shared VH rearrangement and junction. The percentage of total clonal sequences per donor is shown to the bottom right of the pie charts. (B) Average per donor VH mutation frequencies of the two ZIKV patients compared to historical data. Each circle represents the average number of VH nucleotide mutations per donor. Somatic hypermutation frequencies in naive, memory and germinal center B cells, and peripheral B cells from influenza and dengue infected donors were derived from previously published data [33,42]. (C) The range of VH mutation frequencies in the two ZIKV patients. Each circle represents the number of VH mutations per mAb sequence. The binding specificities of the mAbs to ZIKV and DENV are color coded: Cross-reactive mAbs are shown in blue, ZIKV-reactive in green and mAbs of undetermined specificity in orange. (D) VH CDR3 amino acid length distribution of all mAb sequences per patient.

Figure 4

The recall ZIKV plasmablast response is highly cross-reactive against DENV. Heat map showing the binding potency of the ZK016 and ZK018 mAbs against ZIKV and DENV. The mAbs generated from single-cell sorted plasmablasts were tested for reactivity against ZIKV and DENV1-4 recombinant E protein (E) and whole virion by ELISA. The mAbs that are fusion loop (FL)-specific are indicated with asterisks. In terms of scale, brown represents the highest and white the lowest potency of binding based on minimum effective concentration. The minimum effective concentration is the minimum mAb concentration required to obtain three times the background signal. The results plotted are mean values from two or more independent ELISA experiments.

Figure 5

Plasmablast-derived mAbs from dengue-experienced individual preferentially neutralize DENV compared to ZIKV. The mAbs generated from patient ZK016 (left panels) and ZK018 (right panels) were tested for neutralization activity against ZIKV (white bar) and DENV2 (black bar). FL-specific mAbs are indicated by asterisks. The values plotted represent mAb concentration required for 50% reduction of viral foci, termed FRNT50. The vertical dotted lines indicate the maximum concentrations of mAb tested (10 μg/mL). The horizontal dotted line separates the positive control, dengue plasmablast-derived mAb 33.3A06 reported previously [33]. The results plotted are mean values from two or more independent experiments.

Figure 6

ZIKV plasmablast responses enhance DENV2 infection in vitro. (A) Representative flow cytometry plots comparing ADE activity of plasma from ZK016 vs. healthy control. Boxed population represents DENV2-infected cells based on positive 4G2 staining. Plots shown display ADE at plasma dilution of 1:3000. (B) DENV2 ADE activities of plasma from ZK016, ZK018 and a flavivirus-naïve healthy donor (dotted line). (C) DENV2 ADE activities of FL-specific and non FL-specific mAbs from ZK016 and ZK018.

Figure 7

ZIKV infection drives activation of NS1-specific plasmablast responses. (A) Binding curves of ZK016 and ZK018 NS1-specific mAbs. Antibodies were tested for binding to ZIKV and DENV2 recombinant NS1 proteins by ELISA. The dotted lines represent three times the background signal obtained with plain blocking buffer. (B) Binding activity of mAbs to ZIKV (white bar) or DENV2 (black bar) recombinant NS1 protein by ELISA. The values plotted represent the minimum concentrations required to obtain three times the background signal. The dotted line indicates the maximum concentration of mAbs tested in ELISA (10 μg/mL). Results plotted are representative of two or more independent ELISA experiments.