Shuyi, A Name After Dendritic Cell-mediated Immunological Memory

Yiqiang Wang

REVIEW 4 major objections 5 minor 14 references

Shuyi, A Name After Dendritic Cell-mediated Immunological Memory

T0 review · 4 major / 5 minor · reviewed 2026-08-14 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Dendritic cells may form a third class of immune memory

desk verdict A memoir-style essay with a plausible but unproven APC-memory hypothesis; residual antigen is never really excluded, so treat it as perspective, not evidence. read the letter →

arxiv 1908.03686 v2 pith:JCDIZLIY submitted 2019-08-10 q-bio.CB

Yiqiang Wang This is my paper · ORCID

classification q-bio.CB

keywords dendriticcellsimmunologicalmemoryantigen-presentingSCIDmicetrainedimmunityinnateovalbuminadoptivetransfer

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This essay reports the author's late-1990s experiments suggesting that antigen-presenting cells, especially dendritic cells, can build up a memory of an antigen that lasts at least three months even when no T or B cells are present. In SCID mice primed with ovalbumin, transferring naive lymphocytes months later produced a secondary-like antibody response without a fresh challenge, and the memory could be carried by splenic or lymph-node cells but not bone-marrow cells. The author argues these findings, together with recent evidence for trained immunity, make antigen-presenting cells a plausible third class of immune memory. The personal story about naming his daughter Shuyi frames the hypothesis as an unfinished project that modern tools could now test.

What carries the argument

The key experimental system is the SCID mouse, which has no functional T or B lymphocytes, so any antigen-specific memory observed after reconstitution must originate from the host's antigen-presenting cells. The mechanism proposed is that APCs activated during priming retain antigenic epitopes or an enhanced presentation state for at least three months, and when they encounter naive T cells they present these epitopes more efficiently than newly-challenged APCs would. The effective transfer of the memory by splenic and lymph-node cells, but not bone-marrow cells, and its loss when adherent cells are removed, points to dendritic cells as the operative cell type. The paper also uses 125I-labeled OVA to estimate residual antigen at the injection site, an attempt to rule out the simplest alternative explanation.

What would settle it

Measure the 125I radioactivity in regional lymph nodes, spleen, liver, and other organs at three months after priming in SCID mice; if intact OVA or immunogenic fragments above the priming threshold exist anywhere, the memory-like response may be a primary response boosted by persistent antigen rather than APC memory. Alternatively, adoptively transfer primed APCs that have been fixed with paraformaldehyde or treated with acid to strip surface peptides; if memory survives such treatment, it would argue against epitope presentation as the mechanism.

Watch

Extended reading notes

Core claim

The author claims that antigen-presenting cells (APCs) constitute an independent, antigen-specific memory system. The central evidence comes from SCID mice, which lack functional T and B cells: mice primed with OVA and rested for three months produced a secondary-pattern anti-OVA IgG response after receiving naive splenocytes, even without further OVA challenge. Measured 125I-labeled OVA at the footpad injection site was minimal and below the threshold needed to trigger a de novo response, arguing against residual antigen. Transfer of splenocytes or draining-lymph-node cells from primed SCID mice into naive mice reproduced the memory-like response, whereas bone-marrow cells did not, and removing adherent cells (presumed APCs) abolished it. In a further step, bone-marrow-derived dendritic cells primed in vitro and transferred into SCID mice also conferred higher anti-OVA titers than non-primed dendritic cells. The author reads these results as evidence that dendritic cells and other APCs can present their epitope loads to naive T cells in an enhanced, memory-like manner for months, independent of lymphocytes.

Load-bearing premise

The load-bearing premise is that the amount of antigen remaining three months after priming is truly negligible everywhere in the body, but only the footpad injection site was measured, so antigen trapped in other tissues could still be present and cause the secondary-like response.

Editorial extensions

If this is right

Vaccines could be designed to induce long-lived memory in antigen-presenting cells, potentially extending protection beyond T and B cell memory.
Immunological memory would include a third cellular class, changing how immune history is modelled and measured.
APC memory might explain some heterologous immunity and cross-protection against unrelated pathogens.
If APCs carry antigenic epitopes for months, strategies that deplete or modulate dendritic cells could alter long-term vaccine responsiveness.
The SCID-mouse design offers a clean assay for memory that is independent of lymphocytes.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

The 1990s data, if reproduced with modern lineage tracing and single-cell sequencing, could distinguish genuine antigen-specific APC memory from trained immunity driven by epigenetic or metabolic changes.
A testable extension would determine whether the memory-like state requires retained peptide–MHC complexes on the APC surface or instead reflects a permanently altered activation program.
The author's proposal implies APC memory could be transferred between individuals by cell transfer, which might eventually be exploited therapeutically.
If APC memory exists in humans, it would be relevant to vaccine dosing schedules and to immune monitoring of patients with depleted lymphocytes.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

Read this so you know what it is: a personal essay, not a research preprint. Wang revisits unpublished experiments from the late 1990s that suggested dendritic cells/APCs can mediate antigen-specific memory in the absence of T and B cells, connects the idea to the trained immunity literature, and tells the story of naming his daughter Shuyi after the hypothesis. There is no new data, no protocols, and no statistics; the experiments cannot be checked from the text. If you expect a research claim, this is thin. But as a documented early hypothesis it is unusually honest: Wang states plainly that the original manuscript was rejected, never resubmitted, and is being described from memory.

Credit where due: the framing against current work (especially Hole et al. on memory-like DCs) is accurate, and the proposed modern experiment with GFP-labeled DCs and single-cell sequencing is concrete and sensible. The personal narrative is not fluff; it records priority and motivation without overclaiming. The reference list is appropriate, and the author does not lean excessively on his own old papers.

The soft spot is load-bearing. The old data cannot distinguish APC memory from persistent antigen. The author's observation that reconstituted SCID mice produced anti-OVA IgG even without OVA challenge is direct evidence that immunogenic OVA-derived material was still present somewhere. The residual-antigen control measured 125I only in the footpad; it did not rule out antigen depot in spleen, lymph nodes, bone marrow, or on transferred cells. The pure-DC experiment is the most confounded: DCs pulsed with OVA in vitro and transferred into SCID mice are a vehicle for antigen and peptide-MHC, so their persistence and ability to activate naive T cells three months later is exactly what long-lived antigen carriage would predict. Adherent-cell depletion also removes antigen-bearing cells rather than separating memory from transport. The manuscript itself contains the tension: APCs are said to act 'independent of priming antigens' while also presenting 'their antigenic epitope loads.' Those two statements cannot both carry the argument.

So the central hypothesis remains plausible but unproven, and the tests described do not exclude the obvious alternative. That makes this a perspective/history piece rather than a research result. It could be useful for someone thinking about trained immunity or the history of the idea; it is not citable as evidence. The citation pattern is fine.

My recommendation: as a research manuscript, desk reject. As a perspective or historical correspondence, one serious referee could help make the limitations explicit. I would not send it out as a primary research claim.

Referee Report

4 major / 5 minor

Summary. The manuscript is a personal essay by Yiqiang Wang describing his hypothesis, developed in the late 1990s, that antigen-presenting cells (APCs), including dendritic cells (DCs), constitute a third class of immune memory cells operating independently of T and B lymphocytes. The author recounts experiments in SCID mice in which OVA-primed mice, after a three-month rest and subsequent reconstitution with naive splenocytes, produced a secondary-pattern anti-OVA IgG response without further antigenic challenge. He also describes attempts to exclude residual OVA by 125I-labeling, adherent-cell depletion, and transfer of OVA-pulsed DCs into SCID mice. The essay argues that these findings support a memory-like state in APCs, connects the hypothesis to the recent literature on trained immunity (including Hole et al.), and proposes future experiments using GFP-labeled DCs and single-cell sequencing. It concludes with a personal story about naming his daughter Shuyi after the hypothesis.

Significance. If the central claim were established, it would represent a significant conceptual advance, adding APCs as a third arm of immunological memory independent of T/B cells. The manuscript clearly articulates the hypothesis and, importantly, proposes a concrete and feasible experimental design (GFP-labeled DCs, single-cell sequencing) that could rigorously test it. The connection to recent work on trained immunity and memory-like DC responses gives the hypothesis contemporary relevance. However, the paper provides no verifiable data: all experimental recollections are presented without raw data, statistics, methods details, or independent confirmation, and the original manuscript was never published. The evidentiary basis is anecdotal. The main constructive contribution is the proposal of future experiments; if those experiments are performed, they could provide the needed test.

major comments (4)

[Paragraph beginning 'To test if this apparent memory was actually due to potential residual OVA antigens'] The exclusion of residual OVA is incomplete, and this incompleteness directly undermines the central inference. The author reports measuring 125I radioactivity only at the footpad injection site, yet the key observation that primed SCID mice produce anti-OVA IgG after reconstitution with naive splenocytes even without OVA challenge indicates that immunogenic OVA-derived material persists somewhere in the animal. No measurements are described for regional lymph nodes, spleen, bone marrow, or cell-associated antigen depots. The alternative explanation that the 'secondary-like' response is a normal primary response boosted by persisting antigen is therefore not ruled out, which contradicts the claim that APCs 'gained and maintained the ability' to activate T cells independently of residual antigen.
[Paragraph beginning 'For explaining the memory-like status observed in OVA-primed SCID mice'] The pure-DC transfer experiment is confounded by antigen carriage. DCs pulsed with OVA and transferred into SCID mice carry both intact antigen and peptide-MHC complexes on their surface, so the enhanced anti-OVA response after three months could be due to long-lived antigen presentation rather than an intrinsic memory phenotype. The author's own statement that APCs 'present their antigenic epitope loads' is in tension with the claim that the memory is 'independent of priming antigens.' Without an experiment that separates antigen carriage from altered APC function, the described data cannot distinguish between these possibilities.
[Full text, experimental descriptions throughout] The manuscript contains no raw data, error bars, statistical tests, or methodological details (e.g., numbers of mice, antigen doses, time courses, ELISA protocols). Because the original manuscript was rejected and never published, the reader has no independent means to verify the described findings. This is load-bearing, since the paper's central claim rests entirely on these unreported experiments; the recent literature on trained immunity and Hole et al. provides plausibility but does not validate the specific observations recounted here.
[Final paragraph before Competing interests] The author acknowledges that the earlier manuscript was rejected, the project discontinued, and the hypothesis now needs to be demonstrated 'with more convincingness.' This admission should be reflected in the structure of the essay: the preliminary, confounded observations should be clearly labeled as historical background and hypothesis-generating, and the proposed future experiments should be presented as the definitive test. As written, the manuscript's language overstates the level of support for the memory-DC hypothesis.

minor comments (5)

[Abstract] The abstract contains a typo: 'dendtitic' should be 'dendritic.'
[Full text] The text uses 'naï ve' with a combining diaeresis and a space; it should be 'naive' in standard English formatting.
[Figure captions] Figures 1 and 2 are photographs of documents, not data figures. They would be better labeled as plates or supplementary material, and the narrative reference to them should be clarified accordingly.
[Final paragraph] The description of the daughter's current university and major is personal information that, while touching, is not relevant to the scientific content and could be moved to an acknowledgments section or omitted.
[References] The reference list could usefully include earlier work on antigen persistence and follicular dendritic cells, which would help readers evaluate the residual-antigen confound discussed in the major comments.

Circularity Check

0 steps flagged · score 0.0 of 10

No circular derivation: the essay's interpretive claims rest on external trained-immunity findings and historical experiments, not on self-referential reductions.

full rationale

I find no circular derivation in this manuscript. The essay does not derive a quantitative prediction from fitted parameters; the 1990s experiments are presented as historical supporting observations (anti-OVA IgG titers, 125I residual-antigen control, adherent-cell depletion, and DC transfer), and the interpretive claim that APCs can act as a third class of memory cells is explicitly a hypothesis rather than a consequence of equations. The cited prior work on trained immunity (refs 1-7) is external, and the author's own publications (refs 12-13) are cited only as unrelated completed projects, not as load-bearing support for the DC-memory claim. The closest thing to a constructed quantity, the 125I footpad measurement, is used as a control to exclude residual antigen at the injection site, not as an input that is later relabeled a prediction. Any concern about residual antigen in other tissues is an experimental confound or validity criticism, not a circularity of the manuscript's own derivation chain. Accordingly the circularity score is 0.

Assumptions & free parameters 0 free parameters · 3 assumptions · 0 invented entities

The paper introduces no free parameters or invented entities. It rests on standard domain assumptions about the SCID mouse model, antigen clearance, and cell depletion.

assumptions (3)

domain assumption SCID mice lack functional T and B lymphocytes, so immune responses after reconstitution reflect APC activity.
Invoked in the experimental description to attribute antibody responses in OVA-primed SCID mice to APC-mediated memory rather than lymphocyte memory.
domain assumption The 125I-measured OVA remaining at the injection site after three months is below the threshold for inducing an immune response, excluding residual antigen as the cause.
This control is load-bearing for the claim that APCs, not leftover antigen, drive the secondary-like response; the measurement is described but not shown.
domain assumption Depletion of adherent cells from splenocytes removes most APCs and therefore the memory-mediating population.
Used to attribute the transferred memory to APCs; adherent cell depletion is a coarse surrogate and not definitive for dendritic cells.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Shuyi, A Name After Dendritic Cell-mediated Immunological Memory." pith.science (2026). https://pith.science/paper/JCDIZLIY

@misc{pith2026190803686,
  author       = {Pith},
  title        = {Pith review of: Shuyi, A Name After Dendritic Cell-mediated Immunological Memory},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/JCDIZLIY}},
  note         = {Machine review of arXiv:1908.03686}
}

read the original abstract

Immunological memory is a fundamental theory of modern immunology, which is traditionally believed to be mediated only by B and T lymphocytes that recognize antigen epitopes in a receptor-restricted manner. During the last decade data accumulated to show that monocytes and macrophages, the two main initiators of innate immune response, also built up a "memory" to antigens they encountered, though in most concerned publications a different wording (i.e. "train" or"educate") was utilized to describe this feature. More recently, Hole et al demonstrated a "memory-like" response of dendritic cells (DCs). In brief, if fungal-challenged mice could develop a protective immune response, DCs immediately (in 3 weeks) isolated from those mice would manifest a pro-inflammatory phenotype. Even after the mice were allowed to rest for 10 weeks, DCs from them still exhibited an enhanced immune activation profile in their transcriptome and cytokine productions upon re-challenge with same pathogens. Lastly, Hole showed that the "training" or memory-building in DCs was achieved by histone modification. All above findings obtained in monocytes, macrophages or DCs emphasized the necessity for rechecking the questions whether antigen presenting cells (APCs) as a whole could be classified the third class of cells that would mediate immunological memory. In this essay, the author described the effort he made in late 1990s to identify dendtitic cell-mediated memory, and how he named his daughter SHUYI to memorize that hypothesis.

Figures

Figures reproduced from arXiv: 1908.03686 by the authors.

**Figure 2.** Statement about naming the baby Shuyi after dendritic [PITH_FULL_IMAGE:figures/full_fig_p006_2.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

14 extracted references · 14 canonical work pages

[1]

Metabolic Induction of Trained Immunity through the Mevalonate Pathway

Bekkering S, et al. Metabolic Induction of Trained Immunity through the Mevalonate Pathway. Cell 172, 135- 146 e139 (2018)

work page 2018
[2]

Alternative activation of macrophages: mechanism and functions

Gordon S, Martinez FO. Alternative activation of macrophages: mechanism and functions. Immunity 32, 593- 604 (2010)

work page 2010
[3]

BCG Educates Hematopoietic Stem Cells to Generate Protective Innate Immunity against Tuberculosis

Kaufmann E, et al. BCG Educates Hematopoietic Stem Cells to Generate Protective Innate Immunity against Tuberculosis. Cell 172, 176-190 e119 (2018)

work page 2018
[4]

Modulation of Myelopoiesis Progenitors Is an Integral Component of Trained Immunity

Mitroulis I, et al. Modulation of Myelopoiesis Progenitors Is an Integral Component of Trained Immunity. Cell 172, 147-161 e112 (2018)

work page 2018
[5]

Trained immunity: a memory for innate host defense

Netea MG, Quintin J, van der Meer JW. Trained immunity: a memory for innate host defense. Cell Host Microbe 9, 355-361 (2011)

work page 2011
[6]

Immune Training Unlocks Innate Potential

Song WM, Colonna M. Immune Training Unlocks Innate Potential. Cell 172, 3-5 (2018)

work page 2018
[7]

Induction of memory-like dendritic cell responses in vivo

Hole CR, et al. Induction of memory-like dendritic cell responses in vivo. Nat Commun 10, 2955 (2019)

work page 2019
[8]

Enhanced antitumor effects of tumor antigen-pulsed dendritic cells by their transfection with GM-CSF gene

Cao X, Zhang W, Ma S, Zhang M, Wang J, Ye T. Enhanced antitumor effects of tumor antigen-pulsed dendritic cells by their transfection with GM-CSF gene. Sci China C Life Sci 40, 539-545 (1997)

work page 1997

Show all 14 references

[9]

Granulocyte-macrophage colony -stimulating factor induces the differentiation of murine erythroleukaemia cells into dendritic cells

Cao X , et al. Granulocyte-macrophage colony -stimulating factor induces the differentiation of murine erythroleukaemia cells into dendritic cells. Immunology 95, 141-147 (1998)

1998
[10]

Fas ligation induces IL -1beta-dependent maturation and IL -1beta-independent survival of dendritic cells: different roles of ERK and NF-kappaB signaling pathways

Guo Z , et al. Fas ligation induces IL -1beta-dependent maturation and IL -1beta-independent survival of dendritic cells: different roles of ERK and NF-kappaB signaling pathways. Blood 102, 4441-4447 (2003)

2003
[11]

Splenic stroma drives mature dendritic cells to differentiate into regulatory dendritic cells

Zhang M, et al. Splenic stroma drives mature dendritic cells to differentiate into regulatory dendritic cells. Nat Immunol 5, 1124-1133 (2004)

2004
[12]

Anti-DNA antibodies exhibit different binding motif preferences for single stranded or double stranded DNA

Wang Y , Mi J, Cao X. Anti-DNA antibodies exhibit different binding motif preferences for single stranded or double stranded DNA. Immunol Lett 73, 29-34 (2000)

2000
[13]

Activation of antigen-presenting cells by immunostimulatory plant DNA: a natural resource for potential adjuvant

Wang Y , Wang W, Li N, Yu Y , Cao X. Activation of antigen-presenting cells by immunostimulatory plant DNA: a natural resource for potential adjuvant. Vaccine 20, 2764-2771 (2002). 5 Figure 1. Picture of the cover of Wang’s report for ending his postdoctoral training in September

2002
[1998]

2. Immunostimulatory activities of plant DNA

After about 21 years since the production of the report book , the black prints on the cover are somewhat blurring but still readable. The titles of the three worked-on projects were originally provided bilingually. For other Chinese characters on the cover , English translati...

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.

[1] [1]

Metabolic Induction of Trained Immunity through the Mevalonate Pathway

Bekkering S, et al. Metabolic Induction of Trained Immunity through the Mevalonate Pathway. Cell 172, 135- 146 e139 (2018)

work page 2018

[2] [2]

Alternative activation of macrophages: mechanism and functions

Gordon S, Martinez FO. Alternative activation of macrophages: mechanism and functions. Immunity 32, 593- 604 (2010)

work page 2010

[3] [3]

BCG Educates Hematopoietic Stem Cells to Generate Protective Innate Immunity against Tuberculosis

Kaufmann E, et al. BCG Educates Hematopoietic Stem Cells to Generate Protective Innate Immunity against Tuberculosis. Cell 172, 176-190 e119 (2018)

work page 2018

[4] [4]

Modulation of Myelopoiesis Progenitors Is an Integral Component of Trained Immunity

Mitroulis I, et al. Modulation of Myelopoiesis Progenitors Is an Integral Component of Trained Immunity. Cell 172, 147-161 e112 (2018)

work page 2018

[5] [5]

Trained immunity: a memory for innate host defense

Netea MG, Quintin J, van der Meer JW. Trained immunity: a memory for innate host defense. Cell Host Microbe 9, 355-361 (2011)

work page 2011

[6] [6]

Immune Training Unlocks Innate Potential

Song WM, Colonna M. Immune Training Unlocks Innate Potential. Cell 172, 3-5 (2018)

work page 2018

[7] [7]

Induction of memory-like dendritic cell responses in vivo

Hole CR, et al. Induction of memory-like dendritic cell responses in vivo. Nat Commun 10, 2955 (2019)

work page 2019

[8] [8]

Enhanced antitumor effects of tumor antigen-pulsed dendritic cells by their transfection with GM-CSF gene

Cao X, Zhang W, Ma S, Zhang M, Wang J, Ye T. Enhanced antitumor effects of tumor antigen-pulsed dendritic cells by their transfection with GM-CSF gene. Sci China C Life Sci 40, 539-545 (1997)

work page 1997