Dengue Literature Review

Cross-Reactive Antibodies

11 min read

Cross-Reactive Antibodies

Overview

Cross-reactive antibodies are immunoglobulins generated against one dengue serotype (or another flavivirus) that bind — but do not necessarily neutralise — a different serotype. They are the molecular basis for both antibody-dependent enhancement (ADE) and serological cross-reactivity in diagnostics. After a primary dengue infection, only a small fraction of circulating antibodies are truly neutralising; the majority are cross-reactive with variable affinity across serotypes (see Guzman2016 - Dengue Infection). Whether a cross-reactive antibody protects or harms depends on its concentration, affinity, target epitope, isotype, and the time elapsed since the priming infection.

Cross-reactive antibodies are distinct from autoantibodies generated through molecular mimicry (see NS1 Molecular Mimicry in Dengue): cross-reactive antibodies bind different serotypes of the same virus, while mimicry-derived autoantibodies bind host proteins. Both contribute to dengue pathogenesis, but through different mechanisms.

Key Points from Literature

Neutralising vs. enhancing cross-reactivity

The functional outcome of cross-reactive antibodies depends on concentration relative to the neutralisation threshold:

  • Above the threshold: cross-reactive Abs neutralise the heterotypic virus — cross-protection
  • Below the threshold: the same Abs enhance viral uptake into FcR-bearing cells — ADE
  • After primary infection, heterotypic cross-protection lasts approximately 2 months; protection against severe disease persists for approximately 2 years; after this window, waning neutralising Ab concentrations shift the balance toward enhancement (see Guzman2016 - Dengue Infection)
  • A longer interval between primary and secondary infection increases DHF risk — in Cuba, a 20-year interval produced ~8x higher DHF rates than a 4-year interval — consistent with progressive waning below the neutralisation threshold (see Guzman2016 - Dengue Infection, Secondary Dengue Infection)

Maternal antibody cross-reactivity

Maternally transferred IgG antibodies in infants born to dengue-immune mothers follow the same threshold model: initially protective, then enhancing as they decay (~40-day half-life). This explains why primary dengue can cause DHF in infants aged 6-12 months — a natural experiment demonstrating that partial cross-reactive immunity, not prior personal infection, is sufficient for ADE (see Guzman2016 - Dengue Infection, Antibody-Dependent Enhancement).

Isotype determines function: IgG1 vs. IgA

Single-cell transcriptomic and V(D)J sequencing data from the DENFREE Thailand cohort reveal a striking isotype divergence (see Sungnak2025 - Distinct Immune Responses Asymptomatic Symptomatic Dengue):

  • IGHG1+ (IgG1) plasmablasts were significantly enriched in symptomatic dengue (P < 0.05); IgG1 has high FcgammaR affinity — the isotype most likely to mediate ADE
  • IGHA1+ (IgA1) plasmablasts were significantly enriched in asymptomatic dengue (P < 0.05); DENV-specific IgA has been separately shown to neutralise virus and antagonise IgG-mediated ADE
  • A public BCR clonotype network (79 plasmablasts from 8 symptomatic donors; 83.5% using IGHV4-3901 and IGKV1-901) was identified exclusively in symptomatic plasmablasts — biased V(D)J gene usage suggesting antigen-driven clonal expansion of potentially ADE-mediating IgG1 antibodies
  • IL-10 from CD4 proliferating T cells promotes plasma cell differentiation and class switching, driving IGHG1+ plasmablast expansion — a proposed axis for in vivo amplification of ADE-capable antibody

Anti-prM as the dominant non-neutralising cross-reactive antibody class (Dejnirattisai2010)

A crucial revision to the cross-reactive antibody framework: human monoclonal antibody isolation from 7 DENV-infected Thai donors reveals that anti-prM antibodies constitute ~60% of the structural humoral response — more than anti-E — and are fully cross-reactive across all four DENV serotypes (see Dejnirattisai2010 - Anti-prM Antibodies Enhance Dengue ADE, n=3,020 B cell lines screened, experimental).

  • Mechanism of poor neutralisation: Anti-prM antibodies plateau at 10–60% PRNT inhibition regardless of concentration. The reason: incomplete furin cleavage during virion maturation produces a mixed population of prM-positive (immature) and prM-negative (mature, classically infective) particles. Anti-prM binds the immature fraction but cannot neutralise the mature fraction — setting an intrinsic neutralisation floor.
  • ADE magnitude: Anti-prM hmAbs mediated up to 10^5-fold enhancement in primary human monocytes and DCs — substantially exceeding the typical anti-E ADE range. Enhancement was FcγR-dependent.
  • Dengue specificity: Anti-JEV serum showed minimal DENV prM cross-reactivity, distinguishing the anti-prM response as dengue-specific rather than a broad flavivirus phenomenon.

This finding reframes the cross-reactive antibody landscape: the cross-reactive population that most threatens to mediate ADE upon secondary exposure is not a sub-threshold anti-E antibody pool, but a numerically dominant anti-prM pool that cannot neutralise regardless of titre. The two mechanisms are not mutually exclusive and may act additively in vivo.

Broadly neutralising antibodies as vaccine targets

Not all cross-reactive antibodies are harmful. Some antibodies that bind the DI-DII domain hinge of the E protein broadly neutralise all four serotypes without mediating ADE — representing a potential target for vaccines capable of eliciting protective rather than enhancing cross-serotype immunity (see Guzman2016 - Dengue Infection, E Protein).

CYD-TDV as clinical proof-of-concept

The Dengvaxia (CYD-TDV) experience in seronegative children provides the clearest human evidence that vaccine-induced cross-reactive antibodies can replicate ADE: seronegative children aged ⇐5 years had 5x the hospitalisation rate of controls upon subsequent natural infection — the vaccine created an ADE-like state analogous to a first natural infection in a sensitised host (see Guzman2016 - Dengue Infection, CYD-TDV).

Anti-NS1 cross-reactivity with host proteins

Anti-NS1 antibodies represent a special case of cross-reactivity — not between serotypes, but between viral and host antigens. Anti-NS1 Abs cross-react with PDI, vimentin, ATP synthase beta-chain, and HSP60 on platelets and endothelial cells, causing complement-mediated platelet lysis and endothelial damage. The responsible domain maps to C-terminal NS1 aa 311-352; within this region, aa 311–330 (P311–330) is the PDI-specific cross-reactive epitope — confirmed by correlation of anti-P311–330 IgG with anti-PDI IgG (r = 0.732, P < 0.0001). HSP60 cross-reactivity maps to a distinct, unidentified NS1 epitope (see Lin2006 - Autoimmune Pathogenesis in Dengue Virus Infection, Lin2011 - Molecular Mimicry Virus Host Dengue Pathogenesis, Cheng2015 - NS1 P311-330 Anti-PDI Autoantibodies in DHF, NS1 Molecular Mimicry in Dengue).

Longitudinal kinetics — cross-reactive E-IgG rises, NS1-IgG wanes (Bos2025)

Bos2025 - Longitudinal Antibody Dynamics After Dengue provides the most granular longitudinal antibody kinetics data in this wiki, tracking 84 antibody features at <1M, 3M, 6M, and 18M post-dengue in a Nicaraguan pediatric cohort (n=79, DENV-1/DENV-3).

PREPRINT — not peer reviewed.

Key findings that directly revise the classical antibody-waning model:

  • XR E-IgG rises 6–18M post-primary infection: Calculated t½ = −2.13 years (negative sign indicating growth, not decay). This directly contradicts the assumption that cross-reactive antibodies uniformly wane in the inter-infection window. The classical model holds that ADE risk arises passively as neutralising antibodies decay — Bos2025 suggests the ADE risk window may instead reflect an active rise in cross-reactive non-neutralising antibodies.
  • Domain specificity: EDI/II-targeting cross-reactive IgG drives the rise; EDIII-targeting antibodies (more serotype-specific, more likely to be neutralising) remain flat. The antibody pool expanding in the 6–18M window is thus disproportionately composed of cross-reactive, non-EDIII antibodies — precisely the class most likely to bind heterotypic virus without neutralising it.
  • NS1-IgG wanes (t½ ≈ 2.1 years): NS1 responses are predominantly type-specific and follow the classical decay model. This kinetic divergence between NS1-IgG (waning) and XR E-IgG (rising) implies that antibody trajectories are antigen-specific, not a uniform property of the post-dengue immune response.
  • Isotype persistence at 18M: IgA seropositivity ≈ 100%, IgM ≈ 50% post-primary, IgG3 substantially positive, IgG4 rising long-term. The durability of IgM through 18 months substantially exceeds standard expectations and may reflect long-lived IgM plasma cells or ongoing antigen stimulation.

Diagnostic cross-reactivity

Flavivirus cross-reactive antibodies create significant diagnostic challenges:

  • Zika virus (flavivirus) causes IgM cross-reactivity with dengue serology; dual DENV/Zika testing is now recommended where Zika circulates (see Guzman2016 - Dengue Infection)
  • IgG:IgM ratio > 1.2 is used to distinguish secondary from primary infection, but cross-reactive IgG from prior flavivirus exposure can confound this classification
  • PRNT remains the gold standard for serotype-specific confirmation precisely because it measures functional neutralisation rather than binding alone (see PRNT)

Arbovirus cross-reactivity in differential diagnosis (Farias2024)

Farias2024 - Dengue Mimickers (narrative review, Brazil) documents the clinical scope of cross-reactivity in the context of arbovirus co-circulation:

  • CHIKV: ~6% serological cross-reactivity with dengue on commercial IgM platforms; distinguished from dengue by more intense polyarthralgia, chronic arthritis phase, and PCR/specific serology
  • ZIKV: High cross-reactivity with dengue (both flaviviruses sharing epitopes); serological differentiation unreliable; PCR required during acute phase
  • YFV: Cross-reactive with dengue on flavivirus serology; epidemiological competition with DENV hypothesised (mosquito co-infection exclusion may limit co-endemic transmission)
  • SLE/RA autoantibodies: Commercial dengue IgM kits generate false-positives in RF-positive patients (~15%; Santosa2012) and potentially in other autoantibody-positive patients; the cross-reactive signal is non-viral (low-affinity polyclonal IgM binding dengue antigen non-specifically)
  • Brazil has >200 described arboviruses; many flaviviruses circulating in Brazil (Bussuquara, Cacipacoré, Iguape, Ilhéus, Rocio, Saint Louis encephalitis) can produce serological cross-reactions with dengue, requiring RT-PCR for definitive differential (narrative review — secondary source, citing primary literature)

Contradictions & Debates

  • Anti-prM neutralisation ceiling vs. anti-E below-threshold model: The classical cross-reactive ADE model (Guzman2016) holds that cross-reactive anti-E antibodies below the neutralisation threshold cause ADE — implying that high anti-E titres are protective while low titres are enhancing. Dejnirattisai2010 introduces a distinct mechanism: anti-prM antibodies cannot neutralise above a 60% ceiling regardless of titre — the ceiling is structural, not a function of concentration. These are parallel ADE pathways, not competing explanations. Whether anti-prM or sub-threshold anti-E dominates in natural secondary infection is unresolved.
  • Rising XR E-IgG vs. classical waning ADE model: The standard model holds that ADE risk arises because cross-reactive antibodies wane below the neutralisation threshold in the inter-infection period (Guzman2016). Bos2025 shows XR E-IgG actually rises 6–18M post-primary infection — which either (a) means the rising antibodies are non-neutralising regardless of concentration (ADE is driven by the binding/non-neutralising antibody pool, not the waning neutralising pool), or (b) the classical model requires revision. The two interpretations are not mutually exclusive and have not been resolved (Bos2025 is a preprint).
  • Bruhns paradox: Bruhns et al. (2009) found IgG1 binds less efficiently to FcgammaRIIa-R131 than to FcgammaRIIa-H131 — the opposite of what the standard ADE model predicts. If RR individuals have lower IgG1/FcgammaRIIa affinity, the protective effect of the RR genotype (see FcγRIIa Receptor) cannot be explained by reduced ADE through this receptor alone. The discrepancy may reflect differences between monoclonal and polyclonal antibody systems and has not been resolved with DENV-specific immune complexes (see Garcia2010 - Asymptomatic Dengue FcγRIIa Polymorphism).
  • Antigen target of public clonotype unknown: The IGHV4-3901/IGKV1-901 public BCR clonotype in symptomatic dengue has not been mapped to a specific DENV antigen — whether these antibodies are cross-reactive, serotype-specific, or autoimmune is unknown.
  • IgA protection mechanism unclear: The enrichment of IGHA1+ plasmablasts in asymptomatic dengue is correlational. Whether IgA actively antagonises ADE in vivo (as shown in vitro), or merely marks a different immune trajectory, has not been established.

Sources

Graph View

Backlinks