EFB Dengue Literature Review

IgG

17 min read

IgG

Overview

IgG is the predominant antibody isotype in human serum and the primary product of class switch recombination from IgM in germinal centre and extrafollicular B cell responses. Surface IgG expression on B cells marks isotype-switched memory cells. IgG subclasses (IgG1–4) differ in effector functions (complement activation, Fc receptor binding).

Key Points from Literature

  • ~44% of DN (IgD⁻CD27⁻) B cells in healthy subjects express surface IgG, comparable to ~43% of CD27⁺ memory cells — demonstrating that isotype switching occurred in these cells despite their proposed extrafollicular origin (see Wei2007 - DN Memory B Cells in SLE, n=29 healthy cross-sectional).

  • IgG⁺ DN cells were the sorted population used for VH3 BCR sequencing in Wei et al., establishing the somatic hypermutation profile of the DN compartment (see Wei2007 - DN Memory B Cells in SLE).

  • IgG + IgA = majority of SLE ASC sequences: In circulating ASCs from both SLE patients and vaccinated healthy controls, IgG and IgA sequences are the predominant isotypes. IgM sequences contribute 5.42–19.53% in SLE ASCs. The predominance of switched isotypes in ASCs confirms that class switch recombination has occurred even in the EF-derived, low-SHM fraction of ASCs (see Tipton2015 - ASC Diversity and Origin in SLE).

  • DN2 cells have lower surface IgG and IgG3 enrichment: Surface IgG expression on DN2 cells is ~50% lower than on SWM or DN1 cells, consistent with the lower surface Ig levels expected in cells poised for PC differentiation (where Ig shifts from surface-bound to secreted). DN2 cells also show a higher frequency of IgG3⁺ cells compared with SWM or DN1 in both healthy donors and SLE patients. IgG3 enrichment may reflect the default isotype produced in TLR7-driven, T-bet-dependent class switching — IFN-γ promotes switching to IgG3 (and IgG1) in humans (see Jenks2018 - DN2 B Cells and EF Pathway in SLE, flow cytometry + in vitro differentiation).

  • IgG1 dominant switched isotype in COVID-19 ASC repertoire: Single-cell V(D)J sequencing of ASCs from a CoV-A patient showed IgG1 as the major switched isotype, with ongoing class switching from IgM to IgG1 (and IgA1). The majority of IgG1 clonotypes had germline (unmutated) VH genes, consistent with EF-derived, newly recruited ASCs rather than memory recall (see Woodruff2020 - EF B Cell Responses in COVID-19, 10x Chromium scV(D)J).

  • Anti-SARS-CoV-2 RBD IgG elevated in severe COVID-19: Anti-RBD IgG was significantly higher in ICU-C than OUT-C (P ≤ 0.001), with elevated titers by day 5 post-symptom onset. Despite being produced by a germline-dominant, EF-driven ASC response, these antibodies had confirmed neutralizing activity in vitro (see Woodruff2020 - EF B Cell Responses in COVID-19).

  • VH4-34/9G4 autoreactive IgG enriched in severe COVID-19: 85% of VH4-34-expressing ASC clonotypes retained the germline FR1 hydrophobic patch (defective clonal redemption). Serum 9G4 IgG concentrations were significantly elevated in ICU-C vs. HD (P ≤ 0.001) — paralleling the 9G4 enrichment seen in SLE (see Woodruff2020 - EF B Cell Responses in COVID-19).

  • First quantification of IgG dominance in dengue plasmablast response: Wrammert2012 demonstrated that the acute dengue plasmablast response is almost entirely IgG-secreting. ELISpot showed ≥70% (often ≥80%) of all IgG-secreting plasmablasts were DENV-specific, with several donors producing >10⁵ DENV-specific IgG spots per 10⁶ PBMC. IgA was ~100-fold lower; IgM was near-absent except in 4 primary responders. Serum DENV-specific IgG correlated with plasmablast frequency (r² = 0.3–0.4), but total serum IgG was not elevated — implying most plasmablasts are short-lived and do not contribute durably to the serum IgG pool (see Wrammert2012 - Plasmablast Responses in Acute Dengue, n=46 cohort, ELISpot + ELISA).

  • Anti-NS1 and anti-prM/M/E IgG elevated in severe dengue without increased neutralization: Binding IgG antibodies to DENV non-structural (NS1) and structural (prM/M/E) proteins are significantly higher in severe dengue, while FRNT₅₀ neutralizing titers do not differ between severity groups. This replicates the neutralizing Ab paradox from COVID-19 and suggests that EF-derived IgG may contribute to pathogenesis through non-neutralizing mechanisms (see Ansari2025 - Peripheral T Helper Subset Drives B Cell Response in Dengue, ELISA + FRNT, n=170 acute dengue).

  • DENV-specific IgG output is massive but not neutralizing in severe dengue: ELISpot showed ~40,000 DENV-reactive IgG-secreting cells per 10⁶ PBMC in secondary DFC (72% of all IgG ASCs). These IgG-producing plasmablasts cross-reacted with heterotypic serotypes (DENV-1, -2, -3) with 3-fold preference for the infecting serotype. Despite this massive DENV-specific IgG output, PRNT₅₀ neutralizing titers did not correlate with plasmablast frequency — the earliest demonstration of the plasmablast–neutralization disconnect in dengue (see GarciaBates2013 - Plasmablast Response and Dengue Severity, ELISpot + PRNT, n=84 cohort).

  • Plasmablast IgG mAbs are 85% E-specific and more potently neutralising than MBC-derived mAbs: In a dengue clonal analysis, PB-derived IgG mAbs were predominantly E protein-specific (85.3%) and neutralised DENV-1, -2, -3 at 0.1–10 µg/ml. MBC-derived IgG mAbs had broader specificity (55.6% complex epitope, 24.4% prM, 17.8% E) and required >10 µg/ml for neutralisation. No shared IgG CDR3 sequences were found between PBs and MBCs, reinforcing the clonal disconnect between these IgG-expressing populations (see Appanna2016 - Plasmablasts as Subset of Memory B Cell Pool, mAb cloning + neutralisation, n=12 dengue).

  • IgG-switched but poorly mutated — “natural” IgG antibodies in acute dengue: GodoyLozano2016 showed that the acute-phase IgG B cell repertoire is dominated by class-switched (IgG) clones with paradoxically low SHM. These IgG antibodies use biased IGHV segments (IGHV1-2, IGHV1-69) associated with innate-like “natural antibody” recognition. The authors propose these represent rapidly produced, germline-coded, cross-reactive IgG that are intrinsically poly-reactive — consistent with natural IgG antibodies described in other viral infections (VSV, Polyomavirus, Rotavirus). 78.9% of patients had cross-neutralizing serum antibodies to all 4 DENV serotypes, consistent with broad cross-reactivity of the hypomutated IgG response. These poorly mutated cross-reactive IgG antibodies are implicated as candidates for ADE-mediating antibodies in secondary dengue (see GodoyLozano2016 - Lower IgG SHM Rates in Acute Dengue, n=19 acute dengue, DENV RVP neutralisation + 454 pyrosequencing).

  • Cross-reactive E-specific IgG mAbs with near-universal ADE capacity: 53 mAbs cloned from single-cell sorted secondary DHF plasmablasts were predominantly IgG; 37/53 (70%) targeted E protein, all cross-reactive to ≥2 serotypes. 45/53 mAbs enhanced DENV infection of U937 cells (5–161-fold) regardless of neutralisation potency — even the most potently neutralising IgG mAbs caused ADE at sub-neutralising concentrations or against heterotypic serotypes. This establishes that the IgG produced during the secondary dengue plasmablast wave is intrinsically ADE-competent (see Priyamvada2016 - Cross-Reactive Memory Plasmablasts in Secondary Dengue, n=4 secondary DHF, mAb cloning + ADE assay).

  • OAS demonstrated at the IgG mAb level: In 2/4 secondary DENV2 patients, plasmablast-derived IgG mAbs preferentially neutralised DENV1 (median FRNT₅₀ 0.16 µg/ml) over the infecting DENV2 (median FRNT₅₀ 1.2 µg/ml), providing the strongest functional evidence for original antigenic sin in the dengue IgG response. These DENV1-biased IgG mAbs bound DENV2 weakly — the precise profile predicted to mediate ADE during secondary heterotypic infection (see Priyamvada2016 - Cross-Reactive Memory Plasmablasts in Secondary Dengue, n=4 secondary DHF, FRNT + capture virus ELISA).

  • IgG3 overrepresented in alternative lineage B cells in malaria: In malaria-exposed donors, alternative lineage B cells (transcriptomically defined by scRNA-seq) were enriched for IgG3 isotype compared to classical lineage cells, consistent with the IgG3 enrichment on DN2 cells reported in SLE by Jenks2018. IgG3 is the dominant IFN-γ-driven subclass and the default product of T-bet-dependent class switching (see Sutton2021 - Alternative Lineage B Cells in Vaccination and Infection, n=4, 10x Chromium scRNA-seq).

  • IgG⁺ memory B cells are predisposed to plasmablast fate: The isotype-fate segregation model (Seifert et al. 2015, cited in Bhattacharya2016 - Memory B Cell Subset Selection in Secondary Dengue, commentary) proposes that IgG⁺ memory B cells preferentially differentiate into plasmablasts and plasma cells upon recall, while IgM⁺ memory B cells re-initiate germinal centers. This explains the overwhelming IgG dominance of the dengue PB wave (Wrammert2012: ≥70% DENV-specific IgG; IgM near-absent in secondary infection) and the E-specific IgG⁺ bias of PB-derived mAbs vs. the prM/complex epitope specificity of MBC-derived mAbs (Appanna2016) — the IgG⁺ E-specific memory subset is selectively recruited to PB fate.

  • IgG is the predominant isotype of the ABC/atypical population. The majority of ABCs are class-switched, predominantly expressing IgG (or IgA), although scRNA-seq shows the population also retains an unswitched IgD⁺ fraction. This is consistent with the switched-isotype dominance of the EF/atypical compartment across SLE (DN2: IgG/IgG3), COVID-19 (IgG1), and malaria (IgG3) (see Lamprinou2026 - ABCs and DN B Cells, opinion, citing Ambegaonkar 2022 / Tangye 2023). The same claim is made independently in Glaros2025 - Multilayered Identity of B Cell Memory (review), which adds that some ABCs carry somatically hypermutated BCRs — so switched isotype here is not by itself evidence of a GC-independent route (see Class Switch Recombination).

  • IgG1⁺ memory is biased toward plasma-cell differentiation on recall — and is enriched for GC origin. IgG1⁺ MBCs (enriched for GC-derived cells) are biased toward PC differentiation, whereas IgM⁺ MBCs (enriched for GC-independent cells) more readily reenter secondary GCs (see Glaros2025 - Multilayered Identity of B Cell Memory, review, no original data). This independently supports the isotype-fate segregation model the wiki carries from Bhattacharya2016 - Memory B Cell Subset Selection in Secondary Dengue — though the review stresses isotype cannot be used to infer origin (see Class Switch Recombination).

  • IgG is a minority isotype among lung resident memory but drives the faster local recall. After respiratory infection the lung memory pool is predominantly IgA⁺ with a smaller IgG⁺ fraction; IgG⁺ memory is distributed across both lung and mediastinal lymph nodes. Circulating IgG⁺ memory showed delayed differentiation into PCs and was less able to generate PCs near infected alveoli than lung-resident IgG⁺ counterparts, which mounted a rapid local response (see Glaros2025 - Multilayered Identity of B Cell Memory, review, mouse). See Tissue-Resident Memory B Cell.

  • ABC/atypical isotype skewing toward IgG2a/c (mouse) and IgG1 (human), favouring ADCC as the preferred effector mechanism. Adoptive transfer of virus-specific IgG2a only partially restored viral control in a chronic-infection model, indicating additional functional distinctions exist beyond isotype alone (see Cancro2020 - Age-Associated B Cells, review — no original data; mouse and human).

  • Anti-dengue IgG and IgG1 are skewed high in DHF — but IgG3 runs the opposite way. In an adult Mexican acute-phase cohort, anti-DENV IgG was detectable in 93% of DF and 96.9% of DHF sera. On tertile-split indices, high total IgG was present in 55.4% of DHF versus 16% of DF, and high IgG1 in 39.0% versus 28.2%; high IgG4 was likewise commoner in DHF (35.6% vs 30.8%). High IgG3, however, was commoner in DF (38.5%) than DHF (25.4%), and IgG2 was negative in the great majority of both groups (see Posadas-Mondragon2020 - TLR Polymorphisms and Clinical Forms of Dengue, n=254 case-control, DF n=100 / DHF n=65, WHO 1997, capture ELISA plus in-house subclass EIA — no B cell data of any kind). A severity-linked rise in IgG quantity is consistent with the wiki’s claim that the EF/atypical route yields abundant low-fidelity antibody (GodoyLozano2016 - Lower IgG SHM Rates in Acute Dengue) but does not test it: nothing there measures affinity, mutation load, neutralisation or cell of origin. The IgG3 inversion is unexplained by the authors. Subclass serology is canonical to dengue-wiki/; recorded here only for the severity contrast.

  • ★ [2026-08-27] DN3 B cells are the only B cell subset transcriptomically enriched for IgG4. Bulk RNA-seq of sorted DN1–DN4 from IgG4-related disease blood (n=4) found high IGHG4 message specifically in DN3, alongside proliferation and unfolded-protein-response signatures and other antibody-secreting-cell genes (MZB1, TNFRSF17/BCMA, TXNDC5). In a disease defined by an IgG4 isotype skew, the DN subset carrying that message is the one with the pre-plasmablast programme — an isotype-to-subset link the wiki holds for no other isotype (see Allard-Chamard2023 - DN3 B Cells Infiltrate Inflamed Tissues, n=4 bulk RNA-seq).

  • [2026-08-27] The IgG4 class switch in this disease is attributed to pre-GC Tfh help outside germinal centres. The paper’s framing (citing the same group’s earlier work) is that IgG4 switching in IgG4-RD is driven by IL-4- and IL-10-expressing pre-GC T follicular helper cells acting outside germinal centres — a GC-independent route to a specific isotype. This is background in that paper rather than data reported in it (see Allard-Chamard2023 - DN3 B Cells Infiltrate Inflamed Tissues, introduction; no primary data on switching). See Class Switch Recombination, Extrafollicular T Cell Help.

  • Table 1 assigns IgG to both of the DN rows that carry an isotype annotation at all — so isotype is not what separates the DN subsets. DN1 and DN2 are both given as IgM/IgG/IgA⁺ (the third, FcRL4⁺ DN row carries no isotype annotation); what distinguishes them is CD38/CD24/CD21 plus CXCR5, CD11c, FCRL5 and FcRL4. IgG and IgA are listed among the additional rather than core markers (see Sanz2019 - Consistent Classification of Human B Cell Populations, review — no original data). See Double-Negative B Cell.

  • Monocyte-driven IgG secretion in vitro — with an isotype asymmetry in which blockade arms reached significance. DENV-infected monocytes cocultured 6 days with resting allogeneic CD19⁺ B cells stimulated IgG and IgM (but not IgA) secretion to levels significantly higher than monocyte-derived DCs or B-cell-only controls. In the blocking experiments, however, every arm that reached significance named IgM — TACI-Fc “significantly reduced production of IgM”, and anti-IL-10 “significantly blocked plasmablast differentiation and IgM secretion” — while no significant reduction in IgG is reported for any blocking condition (see Kwissa2014 - Monocytes Drive Plasmablast Differentiation in Dengue, in vitro, 4 independent experiments with 4 healthy blood donors). Whether the BAFF/APRIL and IL-10 requirement is genuinely isotype-selective, or the IgG arm was simply underpowered at n=4, is not addressed; the underlying data are in supplementary Figure S5, which is not in the wiki’s PDF.

  • ★ Autoreactive and protective IgG dissociate after B cell depletion — the memory compartments are not equivalent. Autoreactive VH4.34 IgG memory B cells (detected by the 9G4 anti-idiotype) were expanded in SLE at 16.2 ± 11.9% versus 1.3 ± 0.3% in healthy controls (P=0.03) and returned to near-normal (1.92 ± 0.7%) after effective rituximab depletion. In the same patients, anti-tetanus and anti-pneumococcal IgG levels were maintained at 6 months, and serum anti-dsDNA (IgG and IgM) frequently did not normalise — only 4 of 8 evaluable patients improved by 1 year, some requiring 24–36 months (see Anolik2004 - Rituximab and B Cell Abnormalities in SLE, n=15-17 SLE + 7 healthy controls, phase I/II open-label trial, 4-colour flow). Two readings the authors draw: protective IgG memory is sustained by a compartment that CD20-targeted depletion spares, and persistent autoantibody despite normalised autoreactive memory implicates long-lived plasma cells as a separate source. ⚠ VH4.34/9G4 serology is background here under the wiki’s standing scope note; the propagated claim is the cellular one about IgG memory compartments.

  • Class switching to IgG proceeds normally outside the germinal centre, and the switched output is already functional before any TLR stimulus. Ig repertoire sequencing found isotype usage of fate-mapped GC-independent T-bet⁺CD11c⁺ B cells not different from GC B cells (~80% of GC sequences IgG; the T-bet⁺CD11c⁺ population predominantly IgG2b/IgG2c). By ELISpot the population already contained anti-LCMV IgG-secreting cells before R848 stimulation, rising after — unlike memory B cells, which were reliant on the TLR ligand. On sorted transfer into MD4 recipients it generated a rapid, robust anti-LCMV IgG recall response at days 5 and 10, higher than naive and comparable to phenotype-negative memory (see Song2022 - Tfh Outside Germinal Centers Drive T-bet CD11c B Cells, mouse, LCMV-Armstrong, Ig-seq n=3 mice).

Contradictions & Debates

  • ★ A directional prediction that does not hold: DN2 cells are IgG3-enriched, but severe dengue is IgG3-depleted. Jenks2018 - DN2 B Cells and EF Pathway in SLE reports a higher frequency of IgG3⁺ cells among DN2 than among switched memory or DN1, in both healthy donors and SLE patients, and attributes it to IFN-γ-driven switching in the TLR7-driven EF programme. The wiki’s spine claim is that this EF/DN2 route drives the antibody surge in severe dengue. If both held straightforwardly, serum IgG3 should be higher in DHF. It is not: high IgG3 was commoner in DF (38.5%) than DHF (25.4%), moving opposite to IgG, IgG1 and IgG4, which are all skewed high in DHF (see Posadas-Mondragon2020 - TLR Polymorphisms and Clinical Forms of Dengue, n=165 patients, tertile-split indices, acute phase 1–8 d, WHO 1997). The authors do not discuss it.

    Held open, not resolved, and deliberately not treated as a refutation. Four reasons the inference does not close: (i) these are different cohorts and different measurements — Jenks2018 counts surface-IgG3⁺ cells by flow in SLE, Posadas measures serum anti-DENV IgG3 by EIA in dengue; (ii) serum isotype does not report cell of origin — total serum IgG3 integrates output from long-lived plasma cells, GC-derived plasmablasts and any EF contribution, and no ingested source has sorted DN cells and measured their antibody output in any infection; (iii) the indices are tertile splits of OD ratios, not concentrations, so “high IgG3” is a within-cohort rank; (iv) IgG3 has the shortest serum half-life of the IgG subclasses, so the acute-phase window may not be where an EF-derived IgG3 signal would appear. Recorded because it is the first ingested observation that runs against the direction the spine claim predicts, and it should not be quietly dropped. See also the Cancro/Sutton2021 isotype tension immediately below — the wiki now holds three partly incompatible statements about which IgG subclass the atypical lineage favours.

  • ★ Apparent isotype tension: Cancro’s human ABC IgG1 skew vs. Sutton2021’s IgG3 enrichment. Cancro states human ABC/atypical isotype skewing favours IgG1 (paralleling murine IgG2a/c), from a general effector-function argument about the dominant anti-hemagglutinin isotypes in original-antigenic-sin contexts. Sutton2021 - Alternative Lineage B Cells in Vaccination and Infection instead reports IgG3 enrichment in the alternative lineage relative to classical lineage cells in malaria-exposed donors, at the level of cluster-specific transcriptomic enrichment. The two claims likely rest on different measurement bases — a general isotype-skew argument versus a cluster-level transcriptomic enrichment — and are not reconciled in either source; both positions are recorded here rather than smoothed into consensus (see Cancro2020 - Age-Associated B Cells, review — no original data; human; and Sutton2021 - Alternative Lineage B Cells in Vaccination and Infection, n=4, 10x Chromium scRNA-seq).

Double-Negative B Cell, DN2 B Cell, Age-Associated B Cell, Plasmablast, Memory B Cell, Class Switch Recombination, IgM, IgA, Somatic Hypermutation, Original Antigenic Sin, Antibody-Dependent Enhancement, Early Memory B Cell, Tissue-Resident Memory B Cell, DN3 B Cell

Sources

Graph View

Backlinks