EFB Dengue Literature Review

Atypical B Cell Effector Output

14 min read

Atypical B Cell Effector Output

Overview

This page collects what atypical / DN2 / age-associated B cells do to other cells and tissues — as distinct from the signals that create them, which are covered under Toll-like Receptor Signaling in B Cells, B Cell Receptor Signaling, IFN-gamma and Extrafollicular T Cell Help.

Four candidate effector routes appear in the literature: antibody-secreting-cell differentiation, cytokine secretion, antigen presentation to T cells, and tissue localisation effects. They are not equally evidenced, and the difference matters more than any individual claim.

★ Evidence-quality warning — read before citing anything below. Only the ASC/antibody arm rests on primary human data. The cytokine-output arm rests almost entirely on a single murine review (Cancro2020 - Age-Associated B Cells), and the antigen-presentation arm is asserted by three reviews and mechanised by none. No source in this wiki has measured cytokine secretion by sorted human DN2 cells. Statements about atypical B cells “driving inflammation through cytokine production” are, at the current state of this evidence base, extrapolations from mouse.

Key Points from Literature

Route 1 — ASC differentiation and antibody output (best evidenced)

Route 2 — Cytokine secretion (review-carried and murine — treat with caution)

  • ABCs are reported to produce higher IFN-γ and exceptionally high IL-10 compared with follicular B cells (see Cancro2020 - Age-Associated B Cells, review, murine, zero original data — sole source)
  • ABC accumulation is linked to IL-6 in the context of inflammaging, and TNF-α is described as suppressing B lymphopoiesis via pre-B cell apoptosis (see Cancro2020 - Age-Associated B Cells, review, murine)
  • “Pro-inflammatory cytokine and chemokine production” is attributed to ABC/DN cells (see Lamprinou2026 - ABCs and DN B Cells, review — itself citing un-ingested work)
  • TNF-α has a second, opposite-direction role: excess TNF-α is implicated in the blockade of TFH differentiation and loss of germinal centres in severe COVID-19 (see Kaneko2020 - GC Loss and TFH Block in COVID-19, human tissue n=11, inferential — the mechanism is proposed, not demonstrated)
  • Gap: there is no human primary in this wiki measuring cytokine secretion by sorted DN2/atypical B cells. The wiki has no IL-6, IL-10 or TGF-beta entity pages because no ingested primary supports them.

Route 3 — Antigen presentation and T cell activation (asserted, never mechanised)

  • ABCs are described as “powerful antigen-presenting cells… potentially driving TFH induction” (see Sanz2025 - Human Atypical B Cells Overview, review, zero original data)
  • ABCs are said to express high MHC class II and to skew naive CD4 T cells toward Th17 (see Cancro2020 - Age-Associated B Cells, review, murine)
  • “Antigen presentation with strong phagocytic capacity” is attributed to ABCs (see Lamprinou2026 - ABCs and DN B Cells, review)
  • TLR7 stimulation upregulates HLA-DR and CD86 on DN2 cells while downregulating the inhibitory receptors CD72 and CD32b (see Jenks2018 - DN2 B Cells and EF Pathway in SLE, human phospho-flow n=5–10) — this is the only primary evidence in the wiki that the APC machinery is actually induced, and it measures surface expression, not function
  • Gap: Glaros2025 - Multilayered Identity of B Cell Memory explicitly names the plasma-cell-versus-APC effector route as unresolved. The wiki records the question and has no data to answer it. There is no MHC class II entity page because no ingested source supplies more than a phenotype mention.

Route 4 — Positional / tissue effects

  • As infection resolves, murine T-bet⁺CD11c⁺ B cells relocate to the splenic marginal zone and are retained there by LFA-1 and VLA-4, a site chosen for exposure to blood-borne antigen; the authors propose this “potentially enables a protective mechanism through which an antigen-experienced population is placed at the marginal zone to sample antigen and mount rapid response to systemic reinfection” (see Song2022 - Tfh Outside Germinal Centers Drive T-bet CD11c B Cells, murine). See Follicular Exclusion.
  • The same relocation means blood frequency and tissue pool are not in equilibrium — a caution Cancro2020 - Age-Associated B Cells raises independently from splenic/blood disequilibrium and the fall in human blood ABCs on HIV ART.

What the output is driven by, when it happens

Human primary data on the non-ASC outputs (partially closing this page’s largest gap)

  • ★ Granzyme B — a cytotoxic output measured in human DN B cells. DN B cells of young and elderly healthy donors and of MS patients produced granzyme B after in vitro stimulation with IL-21 + anti-BCR. This is human, not murine, and is the wiki’s first direct measurement of a cytotoxic effector molecule from this compartment (see Beckers2023 - Origins and Functions of DN B Cells, review, own data — Claes 2016 — + citing Bulati 2014; see IL-21, In Vitro B Cell Stimulation).

  • ★ Lymphotoxin-α and TNF-α — human, and DN cells out-produce switched memory on TNF-α. CD40L stimulation of total B cells from HD and MS patients produced LTα and TNF-α from DN cells, with LTα⁺ frequencies similar to and TNF-α⁺ frequencies higher than the SM compartment (see Beckers2023 - Origins and Functions of DN B Cells, review, own data; see TNF-alpha).

  • ⚠ But an earlier study found no cytokine output at all from the same compartment. DN B cells could not be induced to express IL-10 or TNF-α after total B cell stimulation with anti-CD40 + IL-4, or with CpG/PMA/ionomycin. The two results differ in stimulation, not in cell type — which means cytokine output from DN cells is stimulation-conditional and no single negative result closes the question (see Beckers2023 - Origins and Functions of DN B Cells, review, citing Bulati 2011; see In Vitro B Cell Stimulation).

  • Senescence-associated secretory output in healthy donors. Circulating DN cells of young and aged HD express the SASP markers TNF-α, IL-6 and IL-8 together with p16^INK4 and the inflammatory microRNAs miR-155, miR-16, miR-96, alongside low anti-apoptotic Bcl2. This is a constitutive pro-inflammatory output attributed to senescence rather than an induced effector response — a distinct category from the stimulation-driven outputs above (see Beckers2023 - Origins and Functions of DN B Cells, review, citing Frasca 2017).

  • Antigen presentation — the phenotypic case, stated at its strongest and its weakest. In HD and MS, DN expression of HLA-DR, CD80 and CD86 sits intermediate between naive and SM cells; in SLE, DN2 cells express higher HLA-DR, CD69 and CD86 than SM cells. The review states plainly that “no direct evidence is available for the induction of T cell responses by DN B cells” — the APC route remains an inference from surface phenotype (see Beckers2023 - Origins and Functions of DN B Cells, review, own data + citing Jenks 2018).

  • Migration as an effector-enabling property. DN cells express CXCR3 (young HD) and CCR6 (elderly HD), with both in HIV infection; CXCR3⁺ DN frequencies are increased in SLE and axSpA; and DN cells have been recovered from inflamed RA synovial tissue and MS cerebrospinal fluid. Whatever these cells secrete, they can deliver it at the inflamed site rather than only into blood (see Beckers2023 - Origins and Functions of DN B Cells, review, own MS/CSF data + citing Bulati 2014 / Moir 2008; see CXCR3).

  • ★ [2026-08-27] The DN3 transcriptome is the wiki’s clearest evidence that an atypical DN subset carries an antibody-secretion programme. Sorted DN1–DN4 from IgG4-related disease blood (n=4, SMART-Seq2): DN3 alone showed strong proliferation and unfolded-protein-response enrichment together with plasmablast/antibody-producing-cell features and high IGHG4 message — “the only B cells transcriptomically enriched for IgG4.” ASC-associated genes MZB1, TNFRSF17 (BCMA) and TXNDC5 appear in the same block. ⚠ n=4, bulk, and transcriptomic resemblance is not differentiation — no ASC assay, pseudotime, or clonal-connectivity evidence is presented (see Allard-Chamard2023 - DN3 B Cells Infiltrate Inflamed Tissues, n=4 bulk RNA-seq).

  • [2026-08-27] A non-antibody effector arm is proposed and left untested: antigen presentation to CD4⁺ T cells in tissue. The paper’s central proposal is that tissue-infiltrating DN cells drive inflammation and fibrosis by re-activating CD4⁺ T cells in situ via antigen presentation, upstream of macrophage and myofibroblast activation — a mechanism in which the effector output is T cell help received-and-returned rather than antibody. Support is the conjugate data plus CD86 and HLA class II staining. ⚠ The HLA class II result is internally contradictory in that paper (Results say DN3 ≈ switched memory and higher than plasmablasts; Fig. S3 legend says DN3 downregulates HLA class II), and it is the measurement the mechanism depends on. Quarantined to the source page under the [2026-08-23] inconsistent-reporting rule — recorded here as a proposed mechanism, not as a supported claim (see Allard-Chamard2023 - DN3 B Cells Infiltrate Inflamed Tissues, n=13).

  • [2026-08-27] DN2 and DN3 share a cytotoxic gene module whose function is untested. Both express GZMA, GZMH, GZMB, GNLY, NKG7, KLRB1, KLRD1, KLRF1 and FCGR3A — genes characteristic of cytotoxic T and NK cells — with perforin generally low in DN3. No cytotoxicity assay was performed. If real at the protein level this would be a third effector arm for the atypical compartment, but on present evidence it is a transcriptional signature only (see Allard-Chamard2023 - DN3 B Cells Infiltrate Inflamed Tissues, n=4).

  • [2026-08-29] The effector subsets move as one coordinated family, and that coordination is what stratifies patients. Unsupervised hierarchical clustering of 253 subjects on B cell subset frequency alone resolved five patient groups, built from three co-varying families — early, memory, and effector (activated naive, DN2, DN3, plasmablasts). Healthy donors fell exclusively into the two memory/early-dominant clusters; SLE patients concentrated in the effector-dominant clusters, with only 15–16% appearing healthy-like. The effector programme is therefore legible at the level of composition, without any functional or secretory assay (see Jenks2021 - B Cell Subset Composition in Cutaneous Lupus, n=207 patients + 46 healthy controls, cross-sectional, 8-marker conventional flow).

  • [2026-08-29] Effector-subset magnitude tracked more autoantibody, but the paper does not connect a subset to a specificity. Cutaneous-lupus patients with the effector-high profile carried more historical ANA, anti-dsDNA, anti-Ro and anti-La, more contemporaneous anti-dsDNA and anti-chromatin, and higher anti-RNP and anti-Ro52 titres than those with healthy-like profiles. No cell-level link was tested — this is a between-patient association between composition and serum, with no sorting, no BCR sequencing and no ASC assay (see Jenks2021 - B Cell Subset Composition in Cutaneous Lupus, n=69 primary CCLE). Relevant as a design template for the bridge-wiki/ cells→autoantibody thesis, and as a warning about what this design cannot establish.

  • ★ The brake-release model of effector output. The question this page tracks — whether atypical B cells can produce antibody-secreting output, and under what conditions — gains a candidate mechanism. In mouse Be1 cultures, terminal differentiation is gated not by whether the PC transcription-factor programme can be induced (IFN-γR signalling does that, T-bet-independently) but by whether the IFN-γ-induced inflammatory programme is switched off in time. T-bet’s function is to switch it off; sustained NF-κB or TLR7/9 signalling in wild-type cells is sufficient to block ASC formation (see Stone2019 - T-bet Promotes ASC Differentiation by Limiting IFN-gamma Inflammation, mouse, in vitro + ATAC-seq/RNA-seq). Effector output on this model is a timing and restraint problem, not a capacity problem — which is consistent with the wiki’s “conditional capacity” position and specifies a different condition than antigen-presentation form.

  • Output requirements and persistence requirements dissociate. Inducible deletion of T-bet from established memory B cells left the memory pool numerically intact over 10 days but cut antigen-specific ASC recall ~10-fold after heterologous challenge (see Stone2019 - T-bet Promotes ASC Differentiation by Limiting IFN-gamma Inflammation, mouse). Whether the atypical/DN2 pool shows the same dissociation — persisting on one transcriptional programme and secreting on another — is untested and is a live question for interpreting dengue convalescent timepoints.

Contradictions & Debates

Is the effector route plasma cell or antigen presentation? Glaros2025 - Multilayered Identity of B Cell Memory names this as unresolved. The wiki’s evidence is lopsided: the ASC route has human primary data from two independent groups; the APC route has three review assertions and one surface-marker measurement. This is an evidence asymmetry, not a resolved question — do not write as though the ASC route is established to the exclusion of the APC route.

Do these cells secrete autoantibody, or only contain autoreactive clones? Jenks2018 - DN2 B Cells and EF Pathway in SLE reports SLE DN2 cells producing autoantibodies. Song2022 - Tfh Outside Germinal Centers Drive T-bet CD11c B Cells finds murine T-bet⁺CD11c⁺ cells enriched for autoreactive clones but not secreting autoantibody without further TLR-driven differentiation. These are compatible if SLE DN2 cells are already further along the differentiation path than resting murine memory-phase cells — but the distinction matters for whether DN2 frequency predicts autoantibody titre.

Is cytokine output even a real effector function of these cells in humans? Every claim traces to murine work via one review. Flagged as the largest single evidence gap on this page.

Atypical B Cell, Double-Negative B Cell, DN2 B Cell, Plasmablast, Extrafollicular Response, Follicular Exclusion, Extrafollicular T Cell Help, Toll-like Receptor Signaling in B Cells, B Cell Receptor Signaling, IFN-gamma, IL-21, IL-21R, BAFF, TNF-alpha, Why DN B Cells Matter - Disease Relevance and Infectious Disease Case, DN3 B Cell

Sources

Graph View

Backlinks