Conventional (myeloid-related) dendritic cell - the DC lineage RELB actually builds, and the thymic part of that phenotype is architecture rather than a cell-intrinsic requirement
also called conventional DC · cDC · cDC2 · CD8alpha-negative DC · myeloid-related dendritic cell · interdigitating dendritic cell · thymic medullary dendritic cell
Dendritic cells are the immune system's messengers, and they come in kinds that are built by different genes. This entry records the kind that needs the gene RELB - and it corrects a near-miss in this graph, where RELB had been attached to a different dendritic cell type because no node for this one existed.
The same thing, for a scientist
This node exists to replace a stated approximation: `gene-relb` points at `cell-type-plasmacytoid-dendritic-cell` at weak confidence with the mismatch declared in its own caveats, and RELB's published requirement is in the **conventional, myeloid-related** DC lineage, not in pDC. The decisive experiment is genetic and clean: RelB mRNA is expressed strongly in CD8alpha-negative DEC-205-negative DC and only weakly in CD8alpha-positive DEC-205-positive DC; CD8alpha-positive DC are **present and functional** in RelB-null mice; and bone-marrow chimeras show that the deficiency in splenic CD8alpha-negative DC is a direct, stem-cell-intrinsic effect of the RelB mutation, while **the partial deficiency in thymic DC is a SECONDARY effect of disrupted thymic architecture** [E1]. That last clause matters for how this graph words the RELB claim, because the dispatch that produced this node described RELB's role as 'thymic/conventional DC' - the literature supports the conventional half as cell-intrinsic and assigns the thymic half to the stroma. The original germline knockout is consistent: relB-null mice have multiorgan inflammatory infiltration, myeloid hyperplasia, splenomegaly from extramedullary haematopoiesis, a reduced population of thymic dendritic cells and impaired contact sensitivity [E2]. Expression mapping agrees and came first - relB transcripts are restricted to the thymic medulla, the periarterial lymphatic sheaths of the spleen and the deep cortex of lymph nodes, and double labelling identifies interdigitating dendritic cells as the RelB-expressing cell [E3]. Modern nomenclature classifies these cells primarily by ontogeny (cDC1/cDC2 within the mononuclear phagocyte system) rather than by the CD8alpha marker the 1998 work used, and the mapping between the two schemes is approximate [E4].
The correction this node makes, stated plainly
gene-relb records a cell-type answer against cell-type-plasmacytoid-dendritic-cell at weak confidence and says in its own caveats that 'the dendritic-cell node used is an imperfect match and is labelled as such'. That was the right thing to do with the vocabulary available. It is now wrong to keep, because the evidence separates the two lineages explicitly:
- CD8alpha-negative DEC-205-negative DC: strong RelB expression; deficient in RelB-null mice; the deficiency is stem-cell-intrinsic by bone-marrow chimera 1.
- CD8alpha-positive DEC-205-positive DC: weak RelB expression; present and functional in RelB-null mice 1.
- Thymic DC: partially deficient, but the chimera experiment attributes it to disrupted thymic architecture, i.e. a stromal effect, not a DC-intrinsic one 1.
pDCs are a separate lineage again, whose development runs through TCF4/E2-2 and IRF8; nothing in the RelB literature cited here assigns them a RelB requirement. Pointing a RELB residency claim at pDC therefore asserts something the sources do not.
What the knockout phenotype is, and what it is not
The germline relB knockout has a striking systemic phenotype - multifocal mixed inflammatory infiltration in several organs, myeloid hyperplasia, splenomegaly due to extramedullary haematopoiesis, reduced thymic dendritic cells, and impaired cellular immunity in contact sensitivity - and the authors conclude that RelB's absence cannot be functionally compensated by any other NF-kB/Rel family member 2.
Two cautions this graph should carry. First, a germline knockout with multiorgan inflammation cannot separate cell-intrinsic requirements from consequences of the inflammation; the 1998 chimera work exists precisely because of that 1. Second, the myeloid expansion and extramedullary haematopoiesis in that animal are what gene-relb correctly declined to call a cell state - they are a change in compartment size, and this graph has no 'myeloid expansion' node. Writing this cell-type node does not fill that second gap and should not be read as doing so.
Nomenclature, and why the marker scheme has moved under this evidence
The 1998 result is framed on CD8alpha and DEC-205, and the phrase used is 'myeloid-related DC lineage'. The field has since reorganised the mononuclear phagocyte system by ontogeny rather than by surface phenotype or presumed myeloid/lymphoid origin, giving cDC1 and cDC2 as the conventional DC subsets, with location, function and phenotype as secondary criteria 4.
The mapping is approximate and this entry does not pretend otherwise: murine CD8alpha-positive DC correspond broadly to cDC1 and CD8alpha-negative DC broadly to cDC2, but the correspondence is not exact across tissues or species, and the 'myeloid-related' framing of the original paper is precisely the ontogenetic assumption the newer scheme was written to replace. A reader transferring the RelB result into cDC1/cDC2 language should say that they did.
Why a leukaemia graph needs this node, honestly stated
The direct justification is vocabulary hygiene: this graph holds gene-relb, gene-rela and gene-rel with opposite directions of effect in this disease, RELB is reported to suppress MLL-AF9 leukaemia, and its only measured normal cell-type requirement had nowhere correct to point.
The indirect justification is that the compartment RelB builds is the antigen-presenting compartment, and this graph already carries an antigen-presentation arm (the W6/32 and Bw6 clone traps of rule 74 sit there). An interpretation that a leukaemia is restrained by a transcription factor whose cell-autonomous job is building conventional DC has an obvious immune reading and no evidence behind it in this disease - so the reading is recorded here as an open question, not as a claim.
What this node does not support: that KMT2A-rearranged leukaemia has any measured relationship to conventional DC. Nobody has looked. cell-type-plasmacytoid-dendritic-cell is in this graph for a different and better-evidenced reason and is not superseded.
| ontology | CL:0000990 |
|---|---|
| ontology note | CL:0000990 is 'conventional dendritic cell'. CL:0002465 (CD8alpha-negative dendritic cell) is closer to the population in E1 but is a murine marker-defined term; CL:0000784 (plasmacytoid dendritic cell) is the DIFFERENT cell that cell-type-plasmacytoid-dendritic-cell holds. |
| species | mouse for every functional result cited here |
| location | thymic medulla, periarterial lymphatic sheaths of the spleen, deep cortex of lymph nodes |
| nomenclature caveat | The load-bearing genetics is framed on CD8alpha and DEC-205 in a myeloid-versus-lymphoid-origin scheme that the field replaced with an ontogeny-based cDC1/cDC2 classification (PMID 25033907). CD8alpha+ maps broadly to cDC1 and CD8alpha- broadly to cDC2, approximately and not across all tissues or species. Say which scheme you are using. |
| shown for fusion | NO - nobody has looked. No study has measured conventional dendritic cells in KMT2A-rearranged leukaemia, in any model or patient cohort. RELB's relevance to this graph is as an NF-kB subunit reported to SUPPRESS MLL-AF9 leukaemia; the connection between that and its DC-building function is unexamined. |
| why this node exists | gene-relb attached its cell-type answer to cell-type-plasmacytoid-dendritic-cell at weak confidence with the mismatch stated in its own caveats, because no conventional-DC node existed. The sources separate the two lineages explicitly, so the approximation asserted something they do not. |
markers
[ "CD11c+", "CD8alpha- DEC-205- for the RelB-dependent murine subset (broadly cDC2 in current nomenclature)", "MHC class II high on activation", "RelB protein high in interdigitating DC of the thymic medulla, splenic PALS and lymph-node deep cortex" ]
relb requirement
{
"cell_intrinsic": "YES for splenic CD8alpha-negative DEC-205-negative DC, established by bone-marrow chimera (PMID 9881974)",
"not_required_for": "CD8alpha-positive DEC-205-positive DC, which are present and functional in RelB-null mice (PMID 9881974)",
"thymic_dc": "PARTIALLY DEFICIENT BUT NOT CELL-INTRINSIC - attributed to disrupted thymic architecture in the same chimera experiment (PMID 9881974). This corrects the 'thymic DC' framing in the dispatch that commissioned this node.",
"plasmacytoid_dc": "NO RelB requirement is asserted by any source cited here; pDC development runs through TCF4/E2-2 and IRF8 and is a separate lineage"
}- RelB builds the compartment that presents antigen. Any therapy that inhibits non-canonical NF-kB signalling - and several are in development - is predicted by this biology to deplete conventional dendritic cells, which is a direct hit on the arm of immunity an infant needs during and after cytotoxic therapy.
- Presence is not function: residual DC in RelB-deficient animals fail to upregulate MHC class II and costimulatory molecules and fail to prime T cells, so a flow-cytometry count showing preserved DC numbers under such a therapy would not establish preserved immunity.
- The germline RelB knockout also shows myeloid hyperplasia and extramedullary haematopoiesis. If RELB restrains myeloid output, pharmacological RELB inhibition in a myeloid leukaemia would be pushing in the wrong direction, and the gene's reported suppression of MLL-AF9 leukaemia says the same.
- The thymic component of this phenotype is stromal. A therapeutic or genetic manipulation of RELB in a child's thymus could disorganise thymic architecture rather than only affecting the DC, and thymic output matters more in an infant than at any later age.
- Does RELB restrain KMT2A-rearranged leukaemia through the cells it builds, or inside the leukaemic cell? RELB is reported to suppress MLL-AF9 leukaemia and its only established normal requirement is building conventional DC. Those could be the same mechanism (an antigen-presenting compartment that restrains the disease) or two unrelated facts about one gene, and no experiment separates them - a RelB-null recipient versus RelB-null donor transplant of the same MLL-AF9 cells would.
- Is the conventional DC compartment altered in KMT2A-rearranged disease at all? No measurement exists in any model or cohort. The reagents are routine and the question has never been asked.
- Does the RelB requirement hold in HUMAN conventional DC? Every functional result cited here is murine, and human cDC2 development has its own dependencies. A human system - RELB knockout in CD34-positive cells differentiated to DC - has not been reported in this context.
- Which of cDC1 and cDC2 the 1998 CD8alpha-negative population maps to, in which tissues, is an approximation this entry declines to resolve. Re-running the RelB dependency with current ontogeny-based markers would close it and nobody has.
- E1strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
IN MICE, RelB mRNA is expressed strongly in CD8alpha-negative DEC-205-negative dendritic cells and only weakly in CD8alpha-positive DEC-205-positive DC; CD8alpha-positive DEC-205-positive DC are present and functional in RelB-null mice with the DC deficiency mainly in the CD8alpha-negative DEC-205-negative population; and bone-marrow chimeras show that the partial deficiency in RelB-null thymic DC is a secondary effect of disrupted thymic architecture whereas the splenic CD8alpha-negative DEC-205-negative deficiency is a direct, stem-cell-intrinsic effect of the RelB mutation.
RelB is essential for the development of myeloid-related CD8alpha- dendritic cells but not of lymphoid-related CD8alpha+ dendritic cells. · Wu L, D'Amico A, Winkel KD, Suter M, Lo D, Shortman K · Immunity 1998What this evidence cannot show
Mouse, 1998, and the classification is built on CD8alpha and DEC-205 in a 'myeloid-related versus lymphoid-related' framework that the field has since replaced with an ontogeny-based cDC1/cDC2 scheme (E4). The bone-marrow chimera is the strong part and is what licenses the intrinsic-versus-stromal split. Nothing here is human, leukaemic or KMT2A-related. Read: abstract only. Not previously cited anywhere in this graph.
Evidence that argues against this (1)
- The residual DC in RelB-deficient animals are functionally impaired even where they are present, which complicates a clean 'CD8alpha+ DC are unaffected' reading. In bone-marrow chimeras from relB-null mice, the small number of residual spleen and lymph-node DC - mainly CD8alpha-positive - failed to upregulate MHC class II and costimulatory molecules after stimulation, had a four-fold decrease in uptake and processing of soluble ovalbumin, and failed to prime CD4 and CD8 T cells in vitro and in vivo. Presence is not function. Read: abstract only. Castiglioni P et al. 2003
- E2established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper
IN MICE carrying a germline disruption of relB, homozygotes lack RelB protein and show a dramatic reduction of constitutive kappa-B-binding activity in thymus and spleen, with multifocal mixed inflammatory-cell infiltration in several organs, myeloid hyperplasia, splenomegaly due to extramedullary haematopoiesis, a reduced population of thymic dendritic cells, and impaired cellular immunity in contact-sensitivity experiments; the authors conclude that RelB's absence cannot be functionally compensated by any other NF-kB/Rel family member.
Multiorgan inflammation and hematopoietic abnormalities in mice with a targeted disruption of RelB, a member of the NF-kappa B/Rel family. · Weih F, Carrasco D, Durham SK, Barton DS, Rizzo CA, Ryseck RP, et al., Bravo R · Cell 1995What this evidence cannot show
Germline knockout with multiorgan inflammation, so cell-intrinsic requirements and inflammation-driven consequences are not separable in this animal; the chimera work in E1 exists to do that separation and reassigns the thymic component to the stroma. Mouse only, fusion-free, no leukaemia. Already cited in gene-relb. Read: abstract only.
Evidence that argues against this (1)
- The thymic dendritic-cell deficiency in this animal is not a DC-intrinsic requirement. Bone-marrow chimera experiments assign the partial thymic DC deficiency in RelB-null mice to disrupted thymic architecture, a secondary stromal effect, while only the splenic CD8alpha-negative DC deficiency is stem-cell-intrinsic - so a germline-knockout phenotype list that reads 'reduced thymic dendritic cells' should not be transferred as a cell-autonomous claim about DC. Read: abstract only. Wu L et al. 1998
- E3strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
IN MOUSE DEVELOPMENT, relB transcripts are very low or undetectable in early and late embryos, low in the late-embryonic thymus and rapidly increasing after birth; in adult lymphoid tissues relB is detected in the medullary region of the thymus, the periarterial lymphatic sheaths of the spleen and the deep cortex of lymph nodes, and double immunofluorescent labelling of thymic cell suspensions identifies interdigitating dendritic cells as the cells expressing RelB.
Expression of relB transcripts during lymphoid organ development: specific expression in dendritic antigen-presenting cells. · Carrasco D, Ryseck RP, Bravo R · Development 1993What this evidence cannot show
In situ hybridisation and immunocytochemistry in mouse, 1993, so it is an expression map rather than a functional requirement, and 'interdigitating dendritic cell' is a histological term that predates the cDC1/cDC2 scheme. It is cited here because it establishes the residency claim - where in the lymphoid organs RelB-expressing DC sit - independently of the knockout phenotype. Read: abstract only.
Evidence that argues against this (1)
- An expression map does not resolve which DC subset is being seen, and the functional split that followed is subset-specific: RelB mRNA is strong in CD8alpha-negative DEC-205-negative DC and only weak in CD8alpha-positive DEC-205-positive DC. A tissue-level in situ signal in the thymic medulla and splenic PALS is therefore compatible with expression in a subset that is not the one the histology names, and cannot by itself support a requirement claim. Read: abstract only. Wu L et al. 1998
- E4strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.review
The mononuclear phagocyte system, historically split into monocytes, dendritic cells and macrophages on functional and phenotypic grounds that frequently overlap, has been reorganised under a proposed unified nomenclature in which these cells are classified primarily by ONTOGENY and only secondarily by location, function and phenotype - the framework within which conventional dendritic cells are designated cDC1 and cDC2.
Dendritic cells, monocytes and macrophages: a unified nomenclature based on ontogeny. · Guilliams M, Ginhoux F, Jakubzick C, Naik SH, Onai N, Schraml BU, et al., Yona S · Nat Rev Immunol 2014What this evidence cannot show
This is an Opinion/Review article proposing a nomenclature, not a primary result, and it is cited here only as the map for translating the 1998 CD8alpha framing into current usage. The correspondence CD8alpha+ ~ cDC1 and CD8alpha- ~ cDC2 is approximate and not uniform across tissues or species; this entry does not assert the RelB requirement in cDC2 terms, it records that the translation is approximate. Read: abstract only.
Evidence that argues against this (1)
- The nomenclature this replaces is the one the load-bearing genetics was done in, and the replacement is ontogeny-based precisely because 'myeloid-related' versus 'lymphoid-related' - the framing of the RelB result's own title - turned out to be the wrong axis. So the newer scheme does not simply rename the 1998 populations; it rejects the inference the 1998 title draws from them. The genetic result stands; the lineage interpretation attached to it in that title does not transfer. Read: abstract only. Wu L et al. 1998
RESIDENCY: relB transcripts are detected in the thymic medulla, splenic PALS and lymph-node deep cortex, and double immunofluorescence identifies interdigitating dendritic cells as the RelB-expressing cell; RelB mRNA is strong in CD8alpha-negative DEC-205-negative DC and weak in CD8alpha-positive DC
REQUIREMENT, separate from the residency edge: the splenic CD8alpha-negative DEC-205-negative DC deficiency in RelB-null mice is stem-cell-intrinsic by bone-marrow chimera, while CD8alpha-positive DC are present and functional
the two DC lineages gene-relb had to be forced to choose between: RelB is required for the conventional myeloid-related subset and no cited source assigns it a pDC requirement. cell-type-plasmacytoid-dendritic-cell is not superseded, it is simply not RELB's cell
required for cDC differentiation; DC-specific Topbp1 knockout causes cDC deficiency with pre-DC accumulation (PMID 42104014)
Human heterozygous SPI1 mutation: patients lack circulating B cells and possess FEW conventional dendritic cells (PMID 33951726) - an allele-resolved human cell-type requirement, not an inference from mouse (Q91 2026-09-22)