Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2025 Oct 23.
Published in final edited form as: Sci Transl Med. 2025 Sep 24;17(817):eadx7231. doi: 10.1126/scitranslmed.adx7231

Afucosylation of anti-dengue IgG is associated with enhanced susceptibility to dengue virus infection postvaccination

Usama Ashraf 1,†, Saborni Chakraborty 1,†, Courtney Scallan 1, Nathan C Lo 2, Maria Theresa Alera 3, Aaron Farmer 3, Mary Noreen Cabalfin-Chua 4, Nelson L Michael 5, Alan L Rothman 6, Taia T Wang 1,2,7,*
PMCID: PMC12543083  NIHMSID: NIHMS2114910  PMID: 40991727

Abstract

Dengue viruses (DENVs) cause 390 million infections annually, although only ~25% of these infections are symptomatic. Whereas antibody features linked to severe DENV disease are well studied, factors influencing infection susceptibility remain less clear. Here, we examined immunoglobulin G (IgG) characteristics before and after DENV vaccination (Dengvaxia) in individuals with a history of prior DENV exposure, comparing those who developed postvaccination infections to those who remained infection free. Elevated anti-DENV afucosylation, present before or after vaccination, was associated with increased likelihood of infection after vaccination. These data were further supported by mechanistic studies, which revealed that nonneutralizing, afucosylated, post-Dengvaxia IgG enhanced DENV replication in mice. This enhancement was dependent on CD16, the receptor for the afucosylated IgG Fc domain. Together, these findings support a model in which the presence of afucosylated IgG promotes virus replication, increasing the likelihood of productive infection upon DENV exposure. Moreover, these results highlight that IgG1 fucosylation is a predictor of risk for breakthrough DENV infection despite vaccination and support the importance of investigating strategies to regulate Fc fucosylation during vaccination.

INTRODUCTION

Dengue viruses (DENVs), which cause dengue disease, comprise four distinct serotypes (DENV1 to DENV4) and represent one of the most substantial arthropod-borne viral threats to global public health. These flaviviruses are transmitted primarily by Aedes aegypti and Aedes albopictus mosquitoes and are endemic in more than 100 countries, particularly in tropical and subtropical regions (1). Although most DENV infections remain asymptomatic, even these infections contribute substantially to transmission dynamics and represent an often overlooked component of disease epidemiology (2).

The pathogenesis of dengue disease is complex and influenced by both viral and host factors. Previous research has extensively characterized antibody features associated with severe dengue disease, particularly the phenomenon of antibody-mediated enhancement of disease, when preexisting cross-reactive antibodies bind virus without neutralizing it, instead facilitating viral entry into Fc receptor–bearing cells and potentially increasing viral burden and inflammatory responses (3). Specific antibody characteristics, including subneutralizing concentrations and afucosylated immunoglobulin G (IgG) Fc glycan structures, have been implicated in this process (4, 5).

IgG glycosylation, particularly fucosylation of the Fc domain N-glycan, represents a key posttranslational modification that modulates dengue immunity. Afucosylated IgG antibodies exhibit enhanced binding to Fcγ receptor IIIa (FcγRIIIa; or CD16). CD16 is an immunoreceptor tyrosine-based activation motif–containing receptor expressed on subsets of leukocytes that mediates antibody-dependent cellular cytotoxicity and potentiates the release of inflammatory mediators. Fc fucosylation affects antibody function across diverse biological contexts. In DENV infections, specifically, afucosylation can be detrimental by exacerbating inflammatory responses and potentially promoting enhanced viral replication (5–9). Despite substantial progress in understanding how antibody signaling affects dengue pathophysiology, the immunological factors that influence initial susceptibility to infection, rather than progression to severe disease, remain inadequately characterized. This distinction is crucial from both clinical and epidemiological perspectives, because preventing infection altogether represents the most effective strategy for reducing transmission and disease burden.

Dengvaxia, the first licensed dengue vaccine, showed efficacy against symptomatic, virologically confirmed dengue infections, with correlates including age at first dose, seropositivity, and postvaccination neutralizing antibody titers (10–13). Efficacy against subclinical infections was lower than against symptomatic infections, especially in people who were seronegative upon vaccination (14). Although Dengvaxia is no longer in broad use, studying correlates of protection after Dengvaxia provides an opportunity to learn about immunological factors that contribute to DENV immunity. Further, identifying antibody features associated with infection postvaccination may clarify protective mechanisms relevant to both natural and vaccine-induced protection against DENV. Traditional correlates of protection against dengue, such as neutralizing antibody titers measured by plaque reduction neutralization tests (PRNTs), provide incomplete explanations for observed patterns of protection. Although higher titers generally correlate with reduced symptomatic disease, exceptions are common, suggesting that additional qualitative features of the antibody response may critically influence outcomes (13).

We therefore examined antibody characteristics before and after Dengvaxia administration that correlate with protection against DENV infection. By focusing on infection rather than symptomatic disease, we address a critical gap in understanding host immune factors that contribute to DENV transmission. By identifying antibody features that predict susceptibility to infection, this study provides insights into dengue immunity and highlights immune factors that may be targeted in future vaccine strategies.

RESULTS

Participant cohort

This study focused on defining anti–DENV1–4 envelope reactive IgG antibodies at baseline (BL) and after the third dose of Dengvaxia in two groups: those who showed no clinical or serological evidence of DENV infection for 6 years and those who exhibited signs of DENV infection within the first 3 years postvaccination. The 3-year time frame for the infected group was chosen on the basis of a prior study, which indicated that this period offers the greatest protection after Dengvaxia vaccination (13). Thus, infection during the initial 3 years represented a relatively poor vaccine response. In contrast, the 6-year time frame for the uninfected group was chosen because it represents an immune phenotype of durable protection after vaccination. By comparing these groups, the study aimed to examine the extremes of postvaccination immunity to DENV. No distinction was made between symptomatic and subclinical infections in the analysis.

Samples were randomly selected except for the inclusion of all participants with a known infection who had an available BL/prevaccination (n = 41) sample with a paired serum sample from 1 month after three doses of Dengvaxia [postvaccination (T1) (n = 88)] (Fig. 1A). In all, 67 participants were studied who developed evidence of DENV infection in the 3 years after vaccination, and the 21 available participants who remained uninfected for 6 years postvaccination were also studied (Fig. 1A).

Fig. 1. Dengvaxia modulated structural determinants of IgG effector functions.

Fig. 1.

Open in a new tab

(A) Schematic of the Dengvaxia (CYD-TDV) vaccine study cohort. (B and C) Distributions of anti-E DENV IgG subclasses (B) and relative abundances (RAs) of IgG1 Fc glycoforms (C) were measured in study participants at BL and 1 month after three doses of vaccine (T1) (n = 41). Proportion of IgG subclasses and geometric mean values of glycans are shown for the two time points. (D) RAs of anti-E DENV IgG1 and anti-E DENV-depleted total IgG1 with the indicated Fc glycoforms were measured in participants at BL and T1 (n = 10). Box and whisker plots in (C) and (D) represent the distribution of the data with the box depicting the interquartile range and the whiskers depicting the minimum and the maximum values. Lines in (B) to (D) connect samples from the same participant. (E) Heatmap representing the normalized mean values for each antibody feature. Statistics were obtained by two-tailed students’ Wilcoxon matched-pairs signed-rank test in (B), (C), and (E) or by two-way ANOVA with Geisser-Greenhouse correction in (D); *P < 0.05; **P < 0.01; ***P < 0.001.

Analyses were conducted to investigate changes in IgG Fc structural features after Dengvaxia vaccination and to identify specific IgG characteristics associated with dengue infection risk. The analysis focused on Fc glycosylation and IgG subclass distribution, which influence interactions with Fcγ receptors and are known correlates of dengue immunity (5–7). Neutralizing antibody titers were included to contextualize the relevance of Fc features associated with infection risk postvaccination. It was assumed that IgG features followed a normal distribution. Age and sex were not significantly different between the infected and uninfected groups (P = 0.81 and P = 0.26 respectively) (table S1). Other epidemiologic data to assess for confounders apart from age and sex were not available.

Three doses of Dengvaxia increased the proportion of afucosylated IgG1 and IgG3/4

The IgG subclass distribution and Fc glycan composition of anti-DENV IgGs present at BL and T1 were profiled. There was considerable heterogeneity in features profiled among participants, yet notable differences in some IgG features between the pre- and postvaccination time points were observed. Vaccination led to changes in the anti-DENV IgG subclass distribution: a decrease in IgG1 and an increase in IgG2, with no differences in IgG3 or IgG4 (Fig. 1B). Among IgG Fc glycans, a relative decrease in fucosylation of IgG1 [corresponding to an increase in afucosylated species, specifically those also lacking a bisecting N-acetylglucosamine (N), F0N0] was observed from BL to T1, despite the overall reduction in IgG1 from BL to T1; detectable changes in other Fc glycoforms were not observed (Fig. 1C). The abundance of anti-DENV IgG1 afucosylation did not correlate with age or sex (fig. S1, A and B). There were no changes in IgG2 glycoforms, whereas the IgG3 and IgG4 populations were enriched for afucosylation (P < 0.0001) and reduced in sialylation (P = 0.02) after vaccination (fig. S1, C and D).

To assess whether the change in IgG1 afucosylation after vaccination was specific to DENV-reactive IgG, we analyzed Fc glycan profiles of both anti-DENV IgG and paired total IgG depleted of DENV-specific antibodies. At both BL and T1 time points, DENV-specific IgG1 was enriched for afucosylation compared with total IgG1 lacking DENV-reactive antibodies (P = 0.0062 at BL and P = 0.0025 at T1) (Fig. 1D). Furthermore, afucosylation of DENV-specific IgG1 increased from BL to T1 (P = 0.017), whereas no such change was observed in total IgG1. In addition to afucosylation, DENV-specific IgG1 exhibited higher sialylation at both time points (P = 0.029 at BL and P = 0.02 at T1) and reduced galactosylation at T1 relative to total IgG1 (P = 0.019). No Fc glycan features changed between BL and T1 aside from IgG1 afucosylation (Fig. 1D and fig. S1, E and F).

Next, neutralizing antibody titers against four DENV serotypes were assessed at BL and T1 (13). All participants included in this study were seropositive at BL against at least one DENV serotype, and three doses of Dengvaxia elicited an increase in PRNT against all Dengvaxia components except DENV2 (fig. S2A). Further, the fold change between BL and T1 neutralizing titers was higher among participants with low PRNT at BL (fig. S2B). Because of the selective increase in anti-DENV IgG1 afucosylation after vaccination (Fig. 1, C and D), we assessed whether there was an association between the neutralizing and afucosylated antibody responses. Neither the degree of increase in afucosylation from BL to T1 nor the absolute abundance of afucosylation at BL or T1 correlated with the neutralizing antibody response (fig. S2, C to E). Overall, three doses of Dengvaxia elicited an increase in the proportion of the IgG2 subclass, an increase in the proportion of afucosylated IgG1 and IgG3/4, and an increase in PRNT against DENV serotypes 1, 3, and 4 (Fig. 1E).

Elevated anti-DENV IgG1 afucosylation was associated with increased likelihood of infection after vaccination

We next compared antibodies from individuals who became infected within the first 3 years after receiving the Dengvaxia vaccine with those from individuals who showed no evidence of infection for a full 6 years after vaccination. Analysis of Fc structural features revealed that DENV infection after vaccination was associated with a higher abundance of afucosylated Fc glycan species, either total afucosylation (F0) or afucosylation without bisection (F0N0), on anti-DENV IgG1 antibodies present at BL or T1 (P = 0.014 at BL and P = 0.007 at T1) (Fig. 2A). However, the absolute change in Fc afucosylation from BL to T1 did not differ between those who became infected and those who remained uninfected (fig. S3A), indicating that the basal abundance of afucosylation, rather than a vaccine-induced change, was the primary correlate of infection after vaccination. No other changes in Fc structural features, including glycan or IgG subclass distribution, correlated with infection after vaccination; additionally, age was not correlated with infection risk (Fig. 2, A and B; fig. S3B; and Table 1). Regarding neutralizing antibody titers, there were no differences in BL PRNT values between groups except for DENV3, where individuals who remained uninfected had higher titers (Fig. 2C). At T1, PRNT values were higher across all four DENV serotypes in the uninfected group compared with those who later became infected (Fig. 2C).

Fig. 2. IgG1 afucosylation is higher in participants that go on to develop dengue infection after vaccination.

Fig. 2.

Open in a new tab

(A and B) Abundances of IgG1 afucosylation (A) and IgG subclasses (B) were measured at BL and at T1 and in individuals who either remained uninfected during the study period (n = 21, Uninf.) or who became infected (n = 67, Infect.) after three doses of vaccine. (C) Distribution of anti-DENV neutralization titers at BL and T1 against the four DENV serotypes. P values were obtained by two-tailed t test with Welch’s correction in (A) and by two-way ANOVA with Šidák’s correction in (B) and (C). Black lines represent geometric means.

Table 1.

Anti-DENV IgG1 Fc glycans.

Uninfected (n = 21) Infected (n = 67)
Anti-E IgG1 glycan (RA, %) Mean SD Mean SD P
Baseline
G0F0 0.21 0.12 0.31 0.20 n.s.
G1F0 2.86 1.17 4.04 2.03 0.03
G2F0 3.35 1.27 4.96 2.32 0.01
lgG1 total F0N0 6.41 2.46 9.32 4.33 0.01
lgG1 total F0 8.84 2.59 12.11 4.44 0.008
Total GS0 80.87 2.08 80.93 1.81 n.s.
Total N 10.43 1.37 10.55 1.60 n.s.
Total S 13.27 2.01 13.33 2.29 n.s.
T1
G0F0 0.31 0.17 0.49 0.26 0.001
G1F0 3.62 1.49 4.85 2.17 0.005
G2F0 3.74 1.35 4.58 1.97 0.03
lgG1 total F0N0 7.68 2.84 9.91 4.14 0.007
lgG1 total F0 10.76 2.89 12.62 4.46 0.03
Total GS0 80.68 2.42 79.72 2.01 n.s.
Total N 10.92 1.72 10.18 1.24 n.s.
Total S 12.93 2.53 13.28 2.21 n.s.
Open in a new tab

The relative abundance of afucosylated (F0)/fucosylated (F), bisected (N), galactosylated (GS0), and sialylated (S) anti-E antibodies was determined at baseline and T1. P values were obtained by two-tailed t test with Welch’s correction. n.s., not significant.

Next, receiver-operating characteristic (ROC) analyses were conducted to evaluate how well afucosylated anti-DENV IgG1 or PRNT (geometric mean) at BL and at T1 predicted vaccine efficacy, defined by protection against infection. At BL, both afucosylation and PRNT showed similar predictive performance [area under the curve (AUC) = 0.726 and 0.721, respectively], and combining the two provided a modest improvement (AUC = 0.767). At T1, afucosylation was a weaker predictor of infection (AUC = 0.663), whereas PRNT was more predictive (AUC = 0.791). Combining both features at T1 further improved predictive accuracy (AUC = 0.829) (fig. S4).

We also evaluated specific afucosylation thresholds for their ability to identify individuals at higher risk of infection after vaccination. Compared with a geometric mean PRNT of <400 at T1, which has previously been linked to increased infection risk after Dengvaxia (13), elevated IgG1 afucosylation (>9% at BL or >12% at T1) was a less sensitive but more specific predictor of infection after vaccination (Table 2). The afucosylation thresholds in Table 2 were empirically derived from the current dataset to best correlate with postvaccination infection risk.

Table 2.

Sensitivity and specificity test of vaccine failure using IgG1 afucosylation (>9% at BL and >12% at T1) and geometric mean PRNT (<400 at T1).

>9% F0N0 BL >12% F0N0 T1 <400 mean PRNT T1
Value 95% CI Value 95% CI Value 95% CI
Relative risk 2.35 1.29 to 4.39 1.33 1.04–1.62 2.35 1.35–5.15
Sensitivity 0.55 0.34 to 0.74 0.27 0.18–0.39 0.93 0.84–0.97
Specificity 0.85 0.65 to 0.95 0.95 0.77–0.99 0.43 0.25–0.63
Open in a new tab

CI, confidence interval.

Nonneutralizing, afucosylated, post-Dengvaxia IgG increased infection in vitro and in vivo

To study biologic activity associated with afucosylation of post-Dengvaxia IgG, we assessed whether this Fc modification affects DENV infection in vitro and in vivo. DENV2 (New Guinea C) was used in experiments with postvaccination IgG pools from participants with low (~5%) or high (~15%) anti-DENV IgG1 afucosylation (F0); the pools had equivalent binding activity to DENV2 envelope protein and were selected for having no detectable DENV2 neutralizing activity at concentrations used for the study (fig. S5, A to C). First, the activities of the low- or high-F0 IgG pools were assessed for their relative ability to mediate antibody-dependent DENV infection in U937 cells. Because wild-type (WT) U937 cells express CD64, CD32a, and CD32b but lack CD16, the receptor for the afucosylated IgG, we also studied the activity of the IgG pools in a U937 cell line engineered to express CD16a (U937-CD16+) (7). DENV2 was incubated with either IgG pool followed by infection of U937 or U937-CD16+ cells. Seventy-two hours postinfection, viral titers in the culture supernatants were quantified by plaque assay (5, 7). The high-F0 IgG pool was associated with increased virus production in U937-CD16+ cells compared with WT U937 cells (fig. S5, D and E) (7, 15).

We next studied the activity conferred by nonneutralizing, afucosylated, post-Dengvaxia IgGs in mice that express human, but not murine, FcγRs (hFcγRs) (16). hFcγR mice are immunocompetent and were made permissive to DENV infection by coadministration of blocking monoclonal antibodies (mAbs) against a subunit of the type I interferon (IFN) receptor (IFNAR1) and anti–IFN-γ (17). Mice were then treated with either high-or low-F0 post-Dengvaxia IgG 24 hours before DENV2 (New Guinea C) infection (Fig. 3A). Viral RNA was subsequently quantified in sera, livers, spleens, and kidneys by quantitative reverse transcription polymerase chain reaction (qRT-PCR) targeting the envelope (E) region of DENV2. The pretreatment with high-F0, post-Dengvaxia IgG was associated with higher DENV genome copies in all organs studied (Fig. 3B). Because in vitro studies indicated that CD16 plays a role in enhancement of virus replication, we also performed DENV2 infections in CD16 knockout hFcγR mice (CD16−/− mice). These experiments revealed that the elevation in viral RNA in the presence of high-F0 Dengvaxia IgG was CD16 dependent, given that the increase in infection with F0 IgG was not observed in hFcγR mice lacking CD16 expression (Fig. 3B, CD16−/− mice).

Fig. 3. Highly afucosylated IgGs, produced after Dengvaxia immunization, enhanced DENV2 infection in hFcγR mice.

Fig. 3.

Open in a new tab

(A) Schematic of experimental design to test effect of post-Dengvaxia IgG on DENV2 infection in vivo. (B) Viral titers from blood, livers, spleens, and kidneys were measured by qRT-PCR and reported as PFU (PFU Eq); n = 8 mice per treatment group and n = 4 for the control (PBS) group. Representative data from two independent experiments are shown. P values were obtained by one-way ANOVA with Tukey’s correction. Means ± SEM indicated by the black horizontal line and error bars.

Although viral titers were highest in mice given highly afucosylated IgG before infection, they were also higher in the spleens of mice given the poorly afucosylated IgG when compared with the mice that received phosphate-buffered saline (PBS) as a control (P = 0.017). Similarly, CD16−/− mice administered highly afucosylated IgG showed increased viral titers in the spleens as compared with the no antibody group (P = 0.004), suggesting that the presence of reactive, nonneutralizing IgG could enhance infection in the spleen independent of the CD16-afucosylated IgG axis.

Afucosylated, post-Dengvaxia IgG increased anti-DENV IgG titers in vivo

Last, we probed whether higher viral titers in mice that received the nonneutralizing, high-F0 IgG pool were associated with a differential host immune response. hFcγR mice were treated with either high or low-F0 IgG 24 hours before infection with a high [106 focus-forming units (FFU)]– or low (102 FFU)–dose virus challenge (Fig. 4A). Consistent with higher viral titers, mice that received high-F0 IgG generated higher titers of murine anti-DENV IgGs (Fig. 4B). Overall, the presence of nonneutralizing and highly afucosylated post-Dengvaxia IgGs enhanced virus production in vitro and in vivo. This enhanced virus production was also correlated with a more robust host antibody response upon infection.

Fig. 4. Highly afucosylated IgGs, produced after Dengvaxia immunization, enhanced the anti-DENV antibody response in mice.

Fig. 4.

Open in a new tab

(A) Schematic of experimental design to study the host response. (B) AUC of binding by murine anti-DENV2 antibodies present on day 10 or 21 after infection with either 102 or 106 PFU DENV2; n = 6 mice per group. Representative data from two independent experiments are shown. P values were obtained by one-way ANOVA with Tukey’s correction. Means ± SEM indicated by the black horizontal line and error bars.

DISCUSSION

This study offers insights into the correlates of vaccine-mediated protection against DENV infection, highlighting a link between elevated Fc afucosylation and increased risk of infection after Dengvaxia administration. We also performed analyses to evaluate the predictive performance of IgG1 afucosylation for vaccine failure (here defined as infection after vaccination) and compared it with PRNT. A postvaccination PRNT cutoff of <400 demonstrated high sensitivity, accurately identifying most individuals who later experienced breakthrough infection; however, its limited specificity led to many false positives among those who remained uninfected. In contrast, a prevaccination anti-E IgG1 afucosylation threshold of >9% showed high specificity, correctly classifying individuals who eventually became infected, although its lower sensitivity resulted in missed identification of some at-risk individuals. These findings suggest that measuring prevaccination Fc afucosylation may be a highly specific tool for identifying a subset of individuals at elevated risk of infection after vaccination, potentially informing targeted interventions such as booster doses or antibody-based therapies. Fc afucosylation can be readily measured using a nanobody, which can be incorporated into an enzyme-linked immunosorbent assay (ELISA) format (18); this represents a practical alternative to neutralizing antibody titers, which requires titration of infectious virus. Fc afucosylation measurement could also be integrated into prospective dengue studies to define the relationship between neutralizing and afucosylated DENV antibodies across cohorts.

In this study, we observed that anti-DENV E IgG1 exhibited a modest increase in afucosylation after vaccination, a change not observed in matched total IgG. The biological relevance of this small shift, if any, remains uncertain, and prospective studies in larger cohorts would be required to determine any potential impact on infection. Postvaccination changes in IgG subclass distribution and Fc glycosylation have also been reported after administration of other vaccines, such as those for seasonal influenza and severe acute respiratory syndrome coronavirus 2 (19, 20). However, direct comparisons are limited by differences in vaccine platforms, study populations, analytical methods, and sampling time points. Beyond vaccination, DENV infection itself is associated with elevated Fc afucosylation, particularly among individuals who progress to severe disease (5, 8). Notably, secondary, but not primary, DENV infections are linked to pronounced increases in afucosylated IgG in those who develop severe outcomes (6). Additional host factors, such as age and sex, also influence IgG fucosylation, with higher IgG afucosylation often observed in older individuals and in males (21, 22).

The effect of polyclonal, afucosylated anti-DENV IgGs on in vivo virus replication may differ from that of DENV mAbs with enhanced CD16 affinity. Prior studies have shown a nonsignificant trend toward increased virus production in the presence of mAbs engineered for enhanced CD16 affinity, although any effect was less pronounced than what we observed with polyclonal IgG (15). Our findings indicate that polyclonal, afucosylated IgGs can promote DENV infection, which may be relevant to their association with increased disease severity in humans (5–7). This well-established association likely involves a combination of facilitated virus replication and CD16-mediated inflammatory activity, both of which may contribute to more severe clinical outcomes.

To that end, we mechanistically demonstrated that highly afucosylated post-Dengvaxia IgGs can enhance virus production both in vitro and in vivo and were associated with a stronger host antibody response upon infection. In hFcγR mice, the presence of afucosylated post-Dengvaxia IgG during infection led to increased viral RNA in multiple organs in a CD16-dependent manner. These findings raise the intriguing hypothesis that individuals with elevated anti-DENV Fc afucosylation may be more likely to develop detectable infections, given similar virus exposure, because of the role of afucosylated IgG-CD16 interactions in promoting infection (fig. S6).

A limitation of this study is that we did not have a means to confirm DENV exposure for all participants during the study period. However, we previously reported the high DENV infection rate among participants in the placebo arm of this trial (13). Furthermore, misclassification of DENV exposure would increase the likelihood of type II error rather than false-positive associations, supporting the robustness of our findings. Further, given the relatively modest sample size in this study, assessment of the association between eleated IgG afucosylation and infection risk is warranted in larger cohorts and in other geographic locations.

Overall, these findings emphasize the importance of considering both neutralizing and nonneutralizing antibody characteristics in vaccine development and evaluation. Furthermore, they highlight the potential for targeted approaches to improve dengue immunity by targeting CD16-mediated mechanisms. Such approaches could mitigate both symptomatic and asymptomatic DENV infections, contributing to reduced transmission and global disease burden.

MATERIALS AND METHODS

Study design

Samples analyzed were deidentified and were not collected for this specific research study. Samples were drawn from a study conducted as an ancillary component of a phase 3 Dengvaxia trial in Cebu City, Philippines (13). Participants from the phase 3 trial were invited to join the ancillary study, and recruitment occurred at two time points: before administration of the first vaccine dose and 1 month after the third dose of vaccine. Participants enrolled before vaccination provided blood samples at enrollment and 1 month after each vaccination. Those enrolled postvaccination provided a blood sample at the time of their enrollment in the ancillary study.

Infections were identified on the basis of clinical and serological criteria, as previously described (13). Briefly, symptomatic individuals who tested positive for DENV by RT-PCR or showed an increase in IgM titers were classified as having confirmed infections. Subclinical infections were detected using computational models that probabilistically identified infections during periods without reported symptomatic cases, incorporating neutralizing antibody activity from annual blood samples (13). Participants with a greater than 80% probability of DENV infection in at least one interval, without any reported symptomatic episodes, were classified as having asymptomatic DENV infection. The uninfected group included individuals who did not meet the criteria for either symptomatic or asymptomatic DENV infection throughout the 6-year postvaccination observation period.

Murine studies were conducted under an Institutional Animal Care and Use Committee (IACUC)–approved animal use protocol in an American Association for Accreditation of Laboratory Animal Care (AAALAC) International–accredited facility with a Public Health Services Animal Welfare Assurance and in compliance with the Animal Welfare Act and other federal statutes and regulations relating to laboratory animals. Power analysis to calculate sample size was not performed. Mice were randomized to various experimental groups on the basis of age, gender, and weight, ensuring that mean age and weight were comparable among the various groups. Researchers were not blinded to the experimental groups.

Neutralizing titer measurements from human participants

PRNTs were performed on all serum samples obtained from annual blood draws, as well as on acute and convalescent samples from confirmed infections, as previously described (13, 23). Briefly, a monolayer of Macaca mulatta kidney cells (LLC-MK2) was infected with 30 to 50 plaque-forming units (PFU) of DENV1–4 that were preincubated with fourfold serial dilutions of heat-inactivated serum sample in a 12-well plate at a starting dilution of 1:10. For each dilution, the number of virus plaques was counted manually and compared to the number of plaques in a control where no serum sample was added. Reference strains were as follows: DENV1 (Thailand/16007/1964), DENV2 (Thailand/16681/1964), DENV3 (Philippines/16562/1964), and DENV-4 (Thailand/C0036/2006). Neutralization titers were expressed as half-maximal PRNT titers.

Purification of IgGs

IgGs from sera were purified by affinity chromatography using protein G Sepharose 4 fast flow beads (GE Healthcare, 17–0618-05) following the manufacturer’s instruction. IgGs that bound to mixed envelope proteins from DENV1–4 were studied for subclass and Fc glycan distributions. For generating high and low afucosylated IgG pools, polyclonal IgG was purified from sera of eight vaccinated participants at T1 time point and then were pooled on the basis of the frequency of afucosylated anti-DENV IgG1 (>15% or <5%) in equimolar amounts.

Mass spectrometry analysis

Methods for relative quantification of Fc glycans and IgG subclasses have been previously described (22). Antigen-specific IgG were isolated on N-hydroxysuccinimide agarose resin (Thermo Fisher Scientific, catalog no. 26196) coupled to DENV E protein from four serotypes in equimolar concentrations (ProSpec). After tryptic digestion of purified IgG bound to antigen-coated beads, nanoscale liquid chromatography coupled to tandem mass spectrometry (nanoLC-MS/MS) analysis for characterization of glycosylation sites was performed on an UltiMate 3000 Nano LC (Dionex) coupled with a hybrid triple quadrupole linear ion trap mass spectrometer, the 4000 QTRAP (SCIEX). MS data acquisition was performed using Analyst 1.6.1 Software (SCIEX) for precursor ion scan–triggered information-dependent acquisition analysis for initial discovery-based identification. For quantitative analysis of the glycoforms at the N297 site of IgG1, multiple-reaction monitoring (MRM) analysis for selected target glycopeptides and their glycoforms was applied using the nanoLC-4000 QTRAP platform to the samples, which had been digested with trypsin. The mass/charge ratio (m/z) of four-charged ions for all different glycoforms as Q1 and the fragment ion at m/z 366.1 as Q3 for each of transition pairs were used for the MRM assays. A native IgG tryptic peptide (131-GTLVTVSSASTK-142) with Q1/Q3 transition pair of 575.9 + 2/780.4 was used as a reference peptide for normalization.

IgG subclass distribution was quantitatively determined by nanoLC-MRM analysis of tryptic peptides after removal of glycans from purified IgG with peptide N-glycosidase F. Here, the m/z value of fragment ions for monitoring transition pairs was always larger than that of their precursor ions with multiple charges to enhance the selectivity for unmodified targeted peptides and the reference peptide. All raw MRM data were processed using MultiQuant 2.1.1 (SCIEX). All MRM peak areas were automatically integrated and inspected manually. In the case where the automatic peak integration by MultiQuant failed, manual integration was performed using the MultiQuant Software.

DENV IgG depletion

IgG was purified from paired serum samples of 10 vaccinated participants at BL and T1 time points based on the frequency of afucosylated anti-DENV IgG1 (n = 5 high F0 and n = 5 low F0). To deplete DENV-specific antibodies, IgGs were incubated with mixed envelope proteins from DENV1–4–conjugated beads overnight, and the flow-through was collected. Flow-through from first round of depletion was subjected to two more rounds of depletion by serial incubation with DENV1–4 envelope proteins. ELISAs with equal amounts of DENV-depleted and undepleted IgG were performed to confirm DENV-specific IgG depletion (fig. S7). Fc glycans were quantified by MS analyses on DENV-specific IgG on beads and paired depleted total IgG samples as described above.

Cells and viruses

A WT human promonocytic cell line [U937-WT; American Type Culture Collection (ATCC), CRL-1593.2] was cultured and maintained in RPMI 1640 medium with l-glutamine (Gibco, 11875–093) supplemented with 10% heat-inactivated fetal bovine serum (FBS; Avantor Seradigm, 97068–085), and penicillin and streptomycin (100 U/ml; Gibco, 15140–122). U937 cells overexpressing CD16 (7) were cultured in the same medium as used for U937-WT cells with an addition of puromycin (1 μg/ml; InvivoGen, ant-pr-1) for clonal selection. African green monkey kidney cells (Vero; ATCC, CCL-81) were cultured and maintained in Dulbecco’s modified Eagle medium (DMEM) with d-glucose (4.5 g/liter), l-glutamine, and sodium pyruvate (110 mg/liter; Gibco, 11995–065) supplemented with 10% heat-inactivated FBS and penicillin and streptomycin (100 U/ml). Both cell lines were incubated at 37°C in a 5% CO2 atmospheric condition. DENV2 [New Guinea C strain; Biodefense and Emerging Infections (BEI) Research Resources Repository NR-84] was used for the experiments. Viral stock was propagated and titered on Vero cells.

Plaque assay

Vero cells were seeded in 96-well tissue-culture plates at a density of 3 × 104 cells per well and incubated overnight at 37°C to make a confluent monolayer. Cell culture supernatants containing viral particles were serially diluted and then used to inoculate monolayers of Vero cells. After 2 hours postincubation at 37°C, unbound viral particles were removed by washing cells three times with Dulbecco’s PBS (DPBS; Gibco, 14190–144). Next, 125 μl of overlay medium was added to wells and cells were incubated at 37°C for 72 hours. Overlay medium was prepared with one part liquid medium and one part 2% methylcellulose (Sigma-Aldrich, M0512–250G). Liquid medium was made with 2× minimum essential medium containing l-glutamine and sodium bicarbonate (Gibco, 11935–046), 20 mM Hepes (Gibco, 15630–080), and penicillin-streptomycin (200 U/ml). After 72 hours of incubation, cells were fixed with 4% formaldehyde solution (Thermo Fisher Scientific, 28908) and washed with 1× PBS with 0.1% Tween 20 (PBST) solution. Plaques were visualized by immunostaining with mouse anti-flavivirus envelope primary antibody (4G2; ATCC, HB-112). Primary antibody was diluted in staining buffer (0.5% bovine serum albumin and 0.5% saponin in PBS) to a concentration of 2 μg/ml and added to the Vero cells for 2 hours, followed by rinsing with PBST. Last, a secondary goat anti-mouse peroxidase-conjugated secondary antibody (SouthernBiotech, 1033–05) was added for 1 hour at a dilution of 1:2500 in staining buffer, followed by rinsing with PBST. Plaques were visualized by incubating cells with 100 μl of KPL TrueBlue peroxidase substrate (SeraCare, 5510–0030) at room temperature for 15 min and counted manually. Viral titers are expressed as PFU per milliliter.

Antibody-mediated infection in vitro

Polyclonal IgGs were threefold serially diluted in RPMI 1640 medium supplemented with 2% FBS and penicillin and streptomycin (R2; 100 U/ml). Diluted IgGs were immune complexed with an equal volume of DENV2 diluted in R2 at a multiplicity of infection (MOI) of 1 per well at 37°C for 1 hour. U937 and U937-CD16+ cells (7) were added to the immune complexes at a concentration of 3.5 × 104 per well and incubated at 37°C. Infection of cells in the absence of IgGs was used as a control. After 2 hours, cells were pelleted by centrifuging plates at 300g for 5 min. Cell supernatants were discarded and cells were washed twice by resuspending them in R2. Cells were repelleted by centrifuging at 300g for 5 min. Supernatants were discarded, and cells were resuspended in 150 μl per well of R2 and incubated at 37°C for 36 hours. Supernatants were collected by centrifuging plates at 300g for 5 min and stored at −80°C until further use. Viral titers in the supernatants were determined by plaque assay on Vero cells as described above.

Virus neutralization assay

Vero cells were seeded in 96-well tissue-culture plates at a density of 3 × 104 cells per well and were grown overnight at 37°C. Polyclonal IgGs purified from serum samples were threefold serially diluted in nonsupplemented DMEM (D0). Diluted IgGs were complexed with an equal volume of DENV2 diluted in D0 medium at 100 PFU per well at 37°C for 1 hour. Then, immune complexes were added onto the confluent monolayer of DPBS-washed Vero cells, followed by incubation at 37°C. After 2 hours of incubation, supernatants were removed, cells were washed three times with DPBS, and 125 μl of overlay medium was added to each well. Cells were further incubated at 37°C for 72 hours. Viral plaques were quantified as described above. Percent inhibition values were calculated by subtracting the percent infection from 100.

Antibody treatment and DENV infection of mice

Research was conducted under an IACUC-approved animal use protocol in an AAALAC International–accredited facility with a Public Health Services Animal Welfare Assurance and in compliance with the Animal Welfare Act and other federal statutes and regulations relating to laboratory animals. The effect of IgGs purified from human serum samples on the enhancement of DENV infection was assessed in a humanized FcγR mouse model. Mice were provided by J. V. Ravetch (Rockefeller University) and were bred in a specific pathogen–free facility at Stanford University. All experiments were performed in animal biosafety level 2 containment at Stanford University according to institutional guidelines. Three- to 4-week-old, sex-matched, fully FcγR-humanized or CD16a-deficient C57BL/6 mice were intraperitoneally administered human IgGs (5 μg per mouse) together with anti-mouse IFNAR1 (3 mg per mouse; clone MAR1–5A3; Leinco Technologies, I-1188) and anti–IFN-γ (250 μg per mouse; clone H22; Leinco Technologies, I-438) mAbs. At 24 hours posttreatment, all mice were anesthetized with 4% isoflurane and were subjected to DENV2 infection at a dose of 106 PFU per mouse by the intraperitoneal route. On day 3 postinfection, mice were bled and euthanized, and body organs (liver, spleen, and kidneys) were harvested. Subsequently, organs were homogenized in D0 medium with 3.0-mm zirconium beads (Sigma-Aldrich; Z763802) using a BeadBug homogenizer (Benchmark Scientific). Tissue homogenates were clarified by centrifuging at 10,000 RCF (relative centrifugal force) for 10 min at 4°C. Supernatants were collected and stored at −80°C until further use. The enhancement of viral infection was calculated by quantifying the viral RNAs using the qRT-PCR method described below.

Extraction and quantification of viral RNAs by qRT-PCR

Total RNA from infected cells was isolated using the RNeasy Mini kit (QIAGEN, 74104). Viral RNAs in the blood and supernatants of mice tissue homogenates were extracted using the QIAamp Viral RNA Mini Kit (QIAGEN, 52906). A total of 300 to 800 ng of RNA was used to synthesize cDNA using the RevertAid First Strand cDNA Synthesis Kit (Thermo Fisher Scientific, K1622) with random hexamers. qPCR was performed on diluted reverse-transcribed reactions using the GoTaq Probe qPCR Master Mix (Promega, A6102) and the StepOnePlus Real-Time PCR system. The virus strain-specific probes and primers were as follows: DENV2-Primer 1, 5′-GGA AGT ACT GTA TAG AGG CAA-3′; DENV2-Primer 2, 5′-ATA ATC CAC ATC CAT TTC CC-3′; and DENV2-Probe, 5′–/56-FAM/CAA CAC AAG/ZEN/GAG AAC CCA G/3IABkFQ/-3′. Amplification was performed at 95°C for 2 min, followed by 40 cycles of 95°C for 15 s and 60°C for 1 min. Cycle threshold (Ct) values obtained from qRT-PCR were converted to genome copy numbers using a standard curve generated from 10-fold serial dilutions of viral RNA extracted from a plaque-titered DENV2 stock. The coefficient of determination (R2) was calculated from the linear regression between Ct values and log10 genome copies to assess the quality of the standard curve. To relate genome copies to infectious units, genome copies/ml from each dilution were paired with the corresponding PFU/ml values of the same virus stock and plotted to generate a regression equation. This equation was then applied to convert genome copy numbers from experimental samples into PFU equivalents (PFU Eq), assuming a constant genome copy-to-PFU ratio across samples. For blood samples, viral titers were expressed as PFU Eq/ml, whereas, for organ samples, titers were normalized to PFU Eq per gram of tissue.

Binding assay of anti-DENV antibodies

The binding activities of purified IgGs or serum samples were determined by a cell-based ELISA method. In brief, a confluent monolayer of Vero cells was infected with DENV2 at an MOI of 1. After 2 hours of virus absorption, virus inoculum was removed, cells were washed with DPBS, 125 μl of overlay medium was added to each well, and cells were incubated at 37°C for 96 hours. After the 96-hour incubation, cells were fixed with a 4% formaldehyde solution, the overlay was removed, and cells were washed with PBST. Serially twofold diluted IgGs at a starting concentration of 100 μg/ml or sera at starting dilution of 1:50 were incubated with fixed cells for 2 hours at room temperature. Later, cells were washed with PBST and further incubated with alkaline phosphatase–conjugated goat anti-human IgG (SouthernBiotech, 2048–04) at a 1:2000 dilution or goat anti-mouse IgG (SouthernBiotech, 1030–04) at 1:2500 dilution for 1 hour at room temperature. All IgGs and sera were diluted in permeability wash buffer. The chromogenic reaction was performed using the p-nitrophenyl phosphate substrate (Thermo Fisher Scientific, 34047) in a diethanolamine substrate buffer (Thermo Fisher Scientific, 34064). Signals at optical density at 405 nm were quantified by using a SpectraMax iD3 multimode microplate reader (Molecular Devices).

Statistical analysis

Individual-level PRNT and glycan data are shown in data file S1. Individual-level data for experiments where n < 20 are presented in data file S2. All analyses were conducted using GraphPad Prism, version 9.0. Statistical analyses were calculated by one-way or two-way analysis of variance (ANOVA) with various corrections including Šidák’s, Geisser-Greenhouse, and Tukey’s corrections for multiple comparisons. Shapiro-Wilk’s test was used to test normality of data. For comparison between two groups, two-tailed student’s paired and unpaired t tests with Wilcoxon matched-pairs signed-rank test or Welch’s correction were used. For all tests, a P value of <0.05 was considered significant. ROC curves were generated using logistic regression with pROC package in R Studio.

Supplementary Material

Supp file 2
Supp file 1
Supp file 3

Funding:

We thank the Department of Medicine, Division of Infectious Diseases, Stanford University and the Institute for Immunity, Transplantation and Infection, Stanford University. The Wang laboratory was supported by NIH grants U19AI181960 (ReVAMPP/FLARE to T.T.W.), 5U19AI057229 (CCHI to T.T.W.), and R01AI186371 (to T.T.W.). Additional support was received from NIH P01 AI034533 (to A.L.R.) and the Military Infectious Diseases Research Program. This material has been reviewed by the Walter Reed Army Institute of Research. There is no objection to its presentation and publication. The opinions or assertions contained herein are the private views of the authors and are not to be construed as official or as reflecting true views of the Department of the Army, the Department of Defense, or NIH.

Footnotes

Competing interests: N.C.L. reports compensation from the World Health Organization related to guidelines on neglected tropical diseases, which are outside the scope of the present work. A.L.R. has received consulting fees from Takeda Inc. and Moderna Inc.; these entities are engaged in activity related to the subject matter of this contribution (flavivirus vaccine development). All other authors declare that they have no competing interests.

Data and materials availability:

All data associated with this study are present in the paper or the Supplementary Materials. Sharing of human participants’ material and identifiable data is restricted by the Institutional Review Boards and would require submission and approval of a separate human participants’ research protocol. Some of the study participants did not consent to future use of their specimens.

REFERENCES AND NOTES

  • 1.Bhatt S, Gething PW, Brady OJ, Messina JP, Farlow AW, Moyes CL, Drake JM, Brownstein JS, Hoen AG, Sankoh O, Myers MF, George DB, Jaenisch T, Wint GR, Simmons CP, Scott TW, Farrar JJ, Hay SI, The global distribution and burden of dengue. Nature 496, 504–507 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Ten Bosch QA, Clapham HE, Lambrechts L, Duong V, Buchy P, Althouse BM, Lloyd AL, Waller LA, Morrison AC, Kitron U, Vazquez-Prokopec GM, Scott TW, Perkins TA, Contributions from the silent majority dominate dengue virus transmission. PLoS Pathog. 14, e1006965 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Guzman MG, Alvarez M, Halstead SB, Secondary infection as a risk factor for dengue hemorrhagic fever/dengue shock syndrome: An historical perspective and role of antibody-dependent enhancement of infection. Arch. Virol. 158, 1445–1459 (2013). [DOI] [PubMed] [Google Scholar]
  • 4.Katzelnick LC, Gresh L, Halloran ME, Mercado JC, Kuan G, Gordon A, Balmaseda A, Harris E, Antibody-dependent enhancement of severe dengue disease in humans. Science 358, 929–932 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Wang TT, Sewatanon J, Memoli MJ, Wrammert J, Bournazos S, Bhaumik SK, Pinsky BA, Chokephaibulkit K, Onlamoon N, Pattanapanyasat K, Taubenberger JK, Ahmed R, Ravetch JV, IgG antibodies to dengue enhanced for FcγRIIIA binding determine disease severity. Science 355, 395–398 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Bournazos S, Vo HTM, Duong V, Auerswald H, Ly S, Sakuntabhai A, Dussart P, Cantaert T, Ravetch JV, Antibody fucosylation predicts disease severity in secondary dengue infection. Science 372, 1102–1105 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Thulin NK, Brewer RC, Sherwood R, Bournazos S, Edwards KG, Ramadoss NS, Taubenberger JK, Memoli M, Gentles AJ, Jagannathan P, Zhang S, Libraty DH, Wang TT, Maternal anti-dengue IgG fucosylation predicts susceptibility to dengue disease in infants. Cell Rep. 31, 107642 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Gonzalez JC, Chakraborty S, Thulin NK, Wang TT, Heterogeneity in IgG-CD16 signaling in infectious disease outcomes. Immunol. Rev. 309, 64–74 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Yamin R, Kao KS, MacDonald MR, Cantaert T, Rice CM, Ravetch JV, Bournazos S, Human FcγRIIIa activation on splenic macrophages drives the in vivo pathogenesis of dengue disease. bioRxiv 514909 [Preprint] (2022). 10.1101/2022.11.02.514909. [DOI] [Google Scholar]
  • 10.Dayan GH, Langevin E, Forrat R, Zambrano B, Noriega F, Frago C, Bouckenooghe A, Machabert T, Savarino S, DiazGranados CA, Efficacy after 1 and 2 doses of CYD-TDV in dengue endemic areas by dengue serostatus. Vaccine 38, 6472–6477 (2020). [DOI] [PubMed] [Google Scholar]
  • 11.Moodie Z, Juraska M, Huang Y, Zhuang Y, Fong Y, Carpp LN, Self SG, Chambonneau L, Small R, Jackson N, Noriega F, Gilbert PB, Neutralizing antibody correlates analysis of tetravalent dengue vaccine efficacy trials in Asia and Latin America. J. Infect. Dis. 217, 742–753 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Capeding MR, Tran NH, Hadinegoro SR, Ismail HI, Chotpitayasunondh T, Chua MN, Luong CQ, Rusmil K, Wirawan DN, Nallusamy R, Pitisuttithum P, Thisyakorn U, Yoon IK, van der Vliet D, Langevin E, Laot T, Hutagalung Y, Frago C, Boaz M, Wartel TA, Tornieporth NG, Saville M, Bouckenooghe A, CYD14 Study Group, Clinical efficacy and safety of a novel tetravalent dengue vaccine in healthy children in Asia: A phase 3, randomised, observer-masked, placebo-controlled trial. Lancet 384, 1358–1365 (2014). [DOI] [PubMed] [Google Scholar]
  • 13.Salje H, Alera MT, Chua MN, Hunsawong T, Ellison D, Srikiatkhachorn A, Jarman RG, Gromowski GD, Rodriguez-Barraquer I, Cauchemez S, Cummings DAT, Macareo L, Yoon IK, Fernandez S, Rothman AL, Evaluation of the extended efficacy of the Dengvaxia vaccine against symptomatic and subclinical dengue infection. Nat. Med. 27, 1395–1400 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Thomas SJ, Yoon IK, A review of Dengvaxia: Development to deployment. Hum. Vaccin. Immunother. 15, 2295–2314 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Yamin R, Kao KS, MacDonald MR, Cantaert T, Rice CM, Ravetch JV, Bournazos S, Human FcγRIIIa activation on splenic macrophages drives dengue pathogenesis in mice. Nat. Microbiol. 8, 1468–1479 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Smith P, DiLillo DJ, Bournazos S, Li F, Ravetch JV, Mouse model recapitulating human Fcγ receptor structural and functional diversity. Proc. Natl. Acad. Sci. U.S.A. 109, 6181–6186 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Chen RE, Diamond MS, Dengue mouse models for evaluating pathogenesis and countermeasures. Curr. Opin. Virol. 43, 50–58 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Kao KS, Gupta A, Zong G, Li C, Kerschbaumer I, Borghi S, Achkar JM, Bournazos S, Wang LX, Ravetch JV, Synthetic nanobodies as tools to distinguish IgG Fc glycoforms. Proc. Natl. Acad. Sci. U.S.A. 119, e2212658119 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Chakraborty S, Gonzalez JC, Sievers BL, Mallajosyula V, Chakraborty S, Dubey M, Ashraf U, Cheng BY, Kathale N, Tran KQT, Scallan C, Sinnott A, Cassidy A, Chen ST, Gelbart T, Gao F, Golan Y, Ji X, Kim-Schulze S, Prahl M, Gaw SL, Gnjatic S, Marron TU, Merad M, Arunachalam PS, Boyd SD, Davis MM, Holubar M, Khosla C, Maecker HT, Maldonado Y, Mellins ED, Nadeau KC, Pulendran B, Singh U, Subramanian A, Utz PJ, Sherwood R, Zhang S, Jagannathan P, Tan GS, Wang TT, Early non-neutralizing, afucosylated antibody responses are associated with COVID-19 severity. Sci. Transl. Med. 14, eabm7853 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Wang TT, Maamary J, Tan GS, Bournazos S, Davis CW, Krammer F, Schlesinger SJ, Palese P, Ahmed R, Ravetch JV, Anti-HA glycoforms drive B cell affinity selection and determine influenza vaccine efficacy. Cell 162, 160–169 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Bakovic MP, Selman MH, Hoffmann M, Rudan I, Campbell H, Deelder AM, Lauc G, Wuhrer M, High-throughput IgG Fc N-glycosylation profiling by mass spectrometry of glycopeptides. J. Proteome Res. 12, 821–831 (2013). [DOI] [PubMed] [Google Scholar]
  • 22.Chakraborty S, Gonzalez J, Edwards K, Mallajosyula V, Buzzanco AS, Sherwood R, Buffone C, Kathale N, Providenza S, Xie MM, Andrews JR, Blish CA, Singh U, Dugan H, Wilson PC, Pham TD, Boyd SD, Nadeau KC, Pinsky BA, Zhang S, Memoli MJ, Taubenberger JK, Morales T, Schapiro JM, Tan GS, Jagannathan P, Wang TT, Proinflammatory IgG Fc structures in patients with severe COVID-19. Nat. Immunol. 22, 67–73 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Buddhari D, Aldstadt J, Endy TP, Srikiatkhachorn A, Thaisomboonsuk B, Klungthong C, Nisalak A, Khuntirat B, Jarman RG, Fernandez S, Thomas SJ, Scott TW, Rothman AL, Yoon IK, Dengue virus neutralizing antibody levels associated with protection from infection in thai cluster studies. PLoS Negl. Trop. Dis. 8, e3230 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supp file 2
NIHMS2114910-supplement-Supp_file_2.zip (44.4KB, zip)
Supp file 1
NIHMS2114910-supplement-Supp_file_1.pdf (459.9KB, pdf)
Supp file 3
NIHMS2114910-supplement-Supp_file_3.pdf (4.1MB, pdf)

Data Availability Statement

All data associated with this study are present in the paper or the Supplementary Materials. Sharing of human participants’ material and identifiable data is restricted by the Institutional Review Boards and would require submission and approval of a separate human participants’ research protocol. Some of the study participants did not consent to future use of their specimens.

RESOURCES