Oncofusion

start / graph / gene-nlrc5

Genemoderate confidence 3 of 5: One strong study, or several smaller consistent ones.Q54-immune

NLRC5 - the MHC class I transactivator, and the route by which interferon activates class I when KMT2A and KMT2B are both gone

also called CITA · class I transactivator · NOD27 · NOD4 · CLR16.1 · caterpiller protein 16.1

NLRC5 is the master switch that turns class I identity cards on when interferon arrives. In cells with both KMT2A-family genes deleted, it is the only way the switch still works - which makes it the back-up route for the whole mechanism at the centre of this beat.

The same thing, for a scientist

NLRC5 (UniProt **Q86WI3**, NLRC5_HUMAN, **1,866 aa**, primary gene name NLRC5, synonyms NOD27 and NOD4) is the class I transactivator: the class I counterpart of CIITA, recruited to MHC class I promoters through the MHC enhanceosome in which RFX5 is a core component [E1][E2]. It matters to this beat because of one experiment. In Sparbier's system, **IFN-gamma-induced MHC class I expression in KMT2A/KMT2B double-knockout cells was entirely dependent on the NLRC5-RFX5 enhanceosome**, and once H3K27me3 was removed, IFN-gamma induced MHC-I just as robustly in KMT2A/KMT2B-null cells as in controls - with SETD1A/SETD1B supplying the H3K4me3, since their additional depletion blocked RFX5-dependent activation [E3]. **So the KMT2A requirement is a requirement for opposing polycomb, not a requirement for transcription.** Remove polycomb and the transcription-factor route works without KMT2A at all. That has a direct consequence for the index disease, where KMT2A dosage is exactly what is in question. **Added after a genome-wide join:** Zhou's 9,292-locus menin/H2AK119ub target set places **NLRC5 in the MENIN-SOLO group** while CIITA sits in the polycomb-marked, menin-unbound group [E5] - which would make the class I transactivator a direct menin target and the class II transactivator an indirect one, the reverse of the arrangement this beat had assumed.

Go deeper
Why this is the load-bearing detail for a KMT2A-rearranged cell

The worry raised in mechanism-mhc-class-i-bivalency-under-menin-inhibition is that the menin-inhibitor class I effect requires KMT2A/KMT2B catalysis, and that a KMT2A-rearranged cell has lost one allele's SET domain to the fusion. This entry is the reason that worry is not decisive.

The requirement is conditional on polycomb. In Sparbier's hands:

  • KMT2A/KMT2B double knockout impaired IFN-gamma-induced MHC-I 3.
  • Pre-treating with a menin inhibitor and/or an EZH2 inhibitor to strip H3K27me3, then adding IFN-gamma, gave equally robust MHC-I in control and KMT2A/KMT2B-null cells 3.
  • Expressing the H3.3 K27M oncohistone, which blocks H3K27me3 spreading, produced the same bypass 3.
  • The bypass was entirely NLRC5-RFX5-dependent, and blocked by additional SETD1A/SETD1B depletion 3.

Two routes, then. Route one needs KMT2A/B to out-compete polycomb at a bivalent promoter. Route two needs polycomb gone and a transcription factor present, and runs on SETD1A/B. A KMT2A-rearranged blast under a menin inhibitor is in an odd position with respect to both, and which route dominates is an experiment nobody has done.

The query that returns zero and the conclusion that would be false

(KMT2A OR MLL) AND leukemia AND NLRC5[tiab] returns 0 records as of 2026-09-20; NLRC5[tiab] AND leukemia returns 2, and title-checking both excludes them - one is about chicken NLRC5 and avian leukosis virus subgroup J, the other a Mendelian-randomisation study of beta-2 microglobulin and B-cell malignancy 4.

A correct measurement. And the conclusion 'NLRC5 has no established relationship to KMT2A' is false, because PMID 36635503 establishes a direct epistatic relationship between the KMT2A/KMT2B genes and the NLRC5-RFX5 enhanceosome - in a figure, with the word NLRC5 nowhere in its abstract. This is the same failure mode as rule 71's PHGDH example and as this beat's HLA-A example: the finding is in the paper and unreachable from the symbol.

The practical instruction: when a gene's relationship to the index biology is likely to be mechanistic rather than topical, search the phrase the mechanism travels under - here bivalent, MHC class I, enhanceosome - before recording a symbol-level zero.

NEW (from the Zhou join): NLRC5 is a menin-bound locus, and CIITA is not

A genome-wide intersection adds a fact this entry did not have. In Zhou's 9,292-locus menin/H2AK119ub target set from KMT2A-rearranged AML, NLRC5 is in the MENIN-SOLO group (chr16:56989555) - menin-bound, no H2AK119ub - while CIITA is in the H2AK119ub-SOLO group, polycomb-marked and menin-unbound 5.

If that holds, the two transactivators are regulated in opposite ways in this disease, and the consequence is the reverse of what this beat assumed:

class Iclass II
transactivatorNLRC5 - menin-boundCIITA - menin-unbound, polycomb-marked
promoters of the presented genesmenin absent at bivalent MHC-I promoters (PMID 36635503)menin reported present at HLA-DR by ChIP-qPCR (PMID 39704147, contested)
implied route under a menin inhibitorpossibly direct, via the transactivatorpossibly indirect, via HERV/interferon

This is a genuinely new hypothesis and it is not what the class I literature proposes. Sparbier's model is entirely indirect - menin is not at the MHC-I promoters, and the effect is KMT2A redistribution. A menin-bound NLRC5 would supply a second, direct route to class I that nobody has looked for: evict menin from the NLRC5 locus, change NLRC5, change class I at every heavy-chain promoter at once.

The obvious cautions. A menin-solo occupancy call is not a functional result; Sparbier found VTP50469 induced remarkably few expression changes despite globally reducing menin occupancy; and the system is adult myeloid KMT2A-rearranged AML, not infant B-lineage disease. But it is a one-qPCR test - measure NLRC5 transcript under a menin inhibitor - and no published experiment has done it.

And it sharpens the query trap above. (KMT2A OR MLL) AND leukemia AND NLRC5[tiab] returns 0. NLRC5 now has two independent, figure-level or supplementary-level relationships to the KMT2A axis - the KMT2A/KMT2B epistasis in PMID 36635503 and the menin occupancy in PMID 39158067's Table 4 - and neither is reachable from the symbol.

Facts and figures
gene symbolNLRC5
gene primaryNLRC5
uniprotQ86WI3
length AA1866
uniprot entry nameNLRC5_HUMAN
roleMHC class I transactivator; binds class I promoters through the MHC enhanceosome with RFX5, RFXANK, RFXAP, CREB and NFY at the SXY module
subnuclear localizationUNPUBLISHED. NLRC5's nuclear pool has never been resolved to a subnuclear compartment in any cell type; SXY-module promoter occupancy is a ChIP result, not an address.
shown for fusionNO for localisation. Nobody has asked where NLRC5 is - cytoplasm or nucleus - in any KMT2A-rearranged cell; the bounded query returned 0. The functional link this graph holds runs the other way: menin inhibition disrupts the KMT2A/B-polycomb balance at bivalent class I promoters and hands the job to IRF1 and NLRC5-RFX5 (PMID 36635503, already E3 in this entry). That is a transcriptional result in KMT2A/KMT2B-depleted cells, not a localisation experiment, and it says nothing about whether NLRC5 protein reaches the nucleus in those cells.
gene synonyms
[
  "NOD27",
  "NOD4"
]
two routes to class I
{
  "route_1_chromatin": "KMT2A/KMT2B oppose PRC2 at bivalent class I promoters; required when H3K27me3 is present",
  "route_2_transcription_factor": "once H3K27me3 is removed, IRF1 and NLRC5-RFX5 drive class I with SETD1A/B supplying H3K4me3, and KMT2A/B are dispensable",
  "pmid": "36635503",
  "title": "Targeting Menin disrupts the KMT2A/B and polycomb balance to paradoxically activate bivalent genes.",
  "authors": "Sparbier CE, Gillespie A, Gomez J, et al.",
  "venue": "Nat Cell Biol",
  "year": 2023,
  "title_provenance": "Title/authors/venue backfilled from PubMed esummary on 2026-09-20 because this citation stored a bare PMID with no title, which made it unscoreable by verify-citations.py. BACKFILLING IS NOT VERIFICATION: it confirms the PMID resolves to a real paper, not that this is the right paper for the claim. That still requires reading."
}
counted absence
{
  "query": "(KMT2A OR MLL) AND leukemia AND NLRC5[tiab]",
  "count": 0,
  "date": "2026-09-20",
  "control": {
    "NLRC5[tiab] AND leukemia": 2
  },
  "on_topic_after_title_check": 0,
  "verdict": "accurate measurement, false conclusion - the relationship is figure-level in PMID 36635503"
}
zhou locus set
{
  "NLRC5": "Menin solo (chr16:56989555)",
  "CIITA": "H2AK119ub solo (chr16:10877200)",
  "source_pmid": "39158067",
  "join": "analyses/zhou-locus-set-join/results.json",
  "implication": "a possible DIRECT route from menin to class I via the transactivator, absent from every published model",
  "system_caveat": "adult myeloid KMT2A-r AML, gene-level, absence != unbound"
}
depmap kmt2a selectivity
{
  "chronos_mean_kmt2a_rearranged": -0.0724,
  "chronos_mean_other_leukaemia": -0.0467,
  "difference": -0.0257,
  "pan_essentiality_all_lines": -0.0713,
  "selectivity_rank": 5861,
  "rank_denominator": 17728,
  "provenance": "Joined 2026-09-21 from analyses/depmap-kmt2a-selective-scan-full (generated 2026-09-19T10:12:06Z), the project's own genome-wide scan over 11 KMT2A-rearranged and 43 other leukaemia DepMap lines. NOT re-fetched and NOT independent of that scan: every caveat it carries applies here, including that the KMT2A-rearranged arm is a small number of lines drawn from nine individuals, that the rank denominator is 17,728 genes, and that a difference is not a dependency. Added because only 26 of 308 gene entries carried this number, which made any KMT2A-selectivity question answerable for 8% of the corpus.",
  "how_to_read": "A negative difference means MORE required in KMT2A-rearranged lines than in other leukaemia lines. A rank near 17,728 is ANTI-selective. Neither is evidence of a fusion-driven dependency on its own: check pan_essentiality_all_lines before calling anything selective, and note that a paralogue-redundant gene can score near zero for reasons that have nothing to do with this disease - see ana-the-paralogue-redundancy-excuse-is-not-tested-and-my-proxy-decided-the-answer."
}
localization
{
  "subcellular": "CYTOPLASMIC AT REST, with a REGULATED NUCLEAR POOL that is the functional one. UniProt Q86WI3 records only Cytoplasm (ECO:0000269 PubMed:20061403, 20434986 - both antiviral/NF-kB papers, neither a class-I study). HPA immunofluorescence agrees on the resting state: main location Cytosol, with Centrosome additional, reliability Supported. But a dedicated 2024 study shows that nuclear IMPORT and EXPORT mechanisms for NLRC5 coexist and counterbalance, that GCN5 retains NLRC5 in the nucleus, and that the nuclear distribution of NLRC5 is what its MHC class I transactivation activity depends on (PMID 38519032).",
  "WHY_THE_DATABASE_ANSWER_IS_THE_WRONG_ANSWER_HERE": "'Cytoplasm' is true and useless for NLRC5. The protein is named CITA, MHC class I transactivator; it works by binding the SXY module at class I promoters with RFX5/RFXANK/RFXAP/CREB/NFY - which requires it to be in the nucleus. The steady-state cytoplasmic pool is a RESERVOIR, and the regulated step this project should care about is the import/export balance, not the resting address. A pathogen has independently confirmed the point: Orientia tsutsugamushi Ank5 inhibits MHC class I expression by forcing NLRC5 CYTOPLASMIC RETENTION and degradation (PMID 39277599) - evolution has found that the way to switch off class I through NLRC5 is to keep it out of the nucleus.",
  "subnuclear": "UNPUBLISHED. Once NLRC5 is in the nucleus nobody has asked where in it. No nucleolar, speckle, body or chromatin-compartment observation exists; the bounded subnuclear query returned 1 record and it is an NPM1/antigen-presentation paper, not an NLRC5 subnuclear study. Promoter occupancy at the SXY module is a ChIP result and is not a subnuclear address.",
  "how_shown": "domain-by-domain import/export analysis with GCN5 dependence (PMID 38519032); effector-driven cytoplasmic retention (PMID 39277599); antibody immunofluorescence (HPA)",
  "tissue_of_evidence": "transfected and tumour cell lines; the Orientia work is in infected human cells. Nothing leukaemic and nothing in a KMT2A-rearranged cell.",
  "node": "localization-partner-baseline-compartment-is-often-miscalled"
}
cell type
{
  "haematopoietic_answer": "LYMPHOID TISSUE AND BONE MARROW are its two highest-expressing tissues. HPA RNA tissue-enhanced: bone marrow 33.8 nTPM and lymphoid tissue 31.8 nTPM - the only gene in this tranche whose top tissues are both haematopoietic. Within blood, HPA records low immune cell specificity and detection in ALL immune cell types; UniProt's tissue comment lists spleen, thymus, lung, brain, tonsil, heart and prostate.",
  "why_this_matters_here": "NLRC5 is not a borrowed solid-tumour protein. Its expression is centred on the haematopoietic compartment, which strengthens the case for reading the class I axis in this disease through NLRC5 rather than through the heavy chain alone.",
  "cross_reference": "Read with gene-hla-a (the heavy chain) and gene-b2m (the light chain). This entry holds the transactivator arm."
}
cell state
{
  "state": "INTERFERON-INDUCED. NLRC5 is an interferon-inducible transactivator and the route by which interferon activates class I when KMT2A and KMT2B are both gone (this entry's two_routes_to_class_I). The localisation corollary, new here, is that induction is necessary but not sufficient: the protein must also be imported, and import is separately regulated by GCN5 (PMID 38519032).",
  "a_second_regulated_layer_nobody_has_measured_in_leukaemia": "If class I is low in a KMT2A-rearranged blast, there are now three candidate lesions and the graph can distinguish them: transcription of NLRC5, IMPORT of NLRC5, or assembly/trafficking of the heavy chain. Only the first has ever been measured in this disease."
}
localization bounded absence
{
  "date": "2026-09-21",
  "tool": "NCBI E-utilities esearch, db=pubmed",
  "shape": "gene clause AND localisation clause AND KMT2A/MLL-fusion clause, one [tiab] tag per term (brief rule 94), every cytogenetic term quoted (brief rule 95)",
  "querytranslation_fusion_clause": "\"KMT2A\"[Title/Abstract] OR \"MLL\"[Title/Abstract] OR \"MLL1\"[Title/Abstract] OR \"MLL-AF4\"[Title/Abstract] OR \"MLL-AF9\"[Title/Abstract] OR \"MLL-ENL\"[Title/Abstract] OR \"KMT2A-rearranged\"[Title/Abstract] OR \"MLL-rearranged\"[Title/Abstract] OR \"t 4 11\"[Title/Abstract] OR \"t 9 11\"[Title/Abstract] OR \"MLL fusion\"[Title/Abstract]",
  "count": 0,
  "controls": {
    "fusion_clause_alone": 8743,
    "gene_clause_alone": 574
  },
  "reading": "Counted in TITLES AND ABSTRACTS only; supporting figures were not searched (brief rule 80)."
}
How this could bite us back
  • If the NLRC5-RFX5 route is the operative one, then the intervention that actually raises class I is polycomb removal plus interferon, and the menin inhibitor is a substitutable and weaker way of getting there - which would make the immune rationale for menin inhibition an argument for an EZH2 inhibitor instead.
  • A transactivator route depends on cytokine supply. In a lymphopenic infant, or inside the CNS niche where T-cell activity is suppressed in this disease, the interferon needed to drive NLRC5 may simply not be present.
  • NLRC5 loss is a documented tumour route to class I silencing in other cancers, so any pressure applied through this axis selects for cells that have lost the transactivator - and unlike B2M loss, that is invisible to a heavy-chain sequencing panel.
  • The delay and loss of uniformity seen in KMT2A/B-null cells is a heterogeneity phenotype, and heterogeneity is what antigen-directed therapies fail against; a mean fluorescence intensity endpoint cannot see it.
Still open
  1. Does NLRC5 transcript change under a menin inhibitor? Zhou's table places NLRC5 in the menin-bound group, which would give class I a direct route that no published model includes. It is one qPCR and nobody has run it.
  2. Which route dominates in a KMT2A-rearranged blast under a menin inhibitor - the KMT2A/B-dependent chromatin route or the polycomb-stripped NLRC5-RFX5 route? The two predict different combination partners: the first argues against anything that further reduces KMT2A function, the second argues for pairing a menin inhibitor with an EZH2 inhibitor and interferon.
  3. Is NLRC5 itself expressed in KMT2A-rearranged blasts, and is its own promoter bivalent? A transactivator that is itself silenced would make both routes moot, and nobody has looked.
  4. Does the 'variable delay in MHC-I activation' in KMT2A/B-null cells translate into a selectable antigen-low fraction under real immune pressure? That is a competition experiment, not a mean-expression experiment.
  5. MHC class I is low in KMT2A-rearranged leukaemia and this graph attributes that to the chromatin route. There are now three separable lesions - NLRC5 transcription, NLRC5 NUCLEAR IMPORT (GCN5-dependent, PMID 38519032), and heavy-chain assembly/trafficking - and only the first has been measured in this disease. A nuclear/cytoplasmic fractionation of NLRC5 in a KMT2A-rearranged blast, with and without a menin inhibitor, would separate them.
Every source behind this entry (9)
  1. E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database

    The reviewed human UniProt record for NLRC5 is Q86WI3 (NLRC5_HUMAN), primary gene name NLRC5, synonyms NOD27 and NOD4, 1,866 aa, protein name 'Protein NLRC5' with alternative names Caterpiller protein 16.1, CLR16.1 and nucleotide-binding oligomerization domain proteins 27 and 4.

    UniProtKB reviewed human entry Q86WI3 (NLRC5_HUMAN) · UniProt Consortium · UniProtKB 2026
    What this evidence cannot show

    Database fact, fetched 2026-09-20, gene_primary read back. A single reviewed record; at 1,866 aa the length check discriminates cleanly. The alias NOD4 is a caution in itself - the NOD/NLR family is large and NOD1/NOD2 are innate-immunity genes with unrelated functions, so a source saying only 'NOD4' should be resolved before use.

    Evidence that argues against this (1)
    • NLRC5's alternative names sit inside an inflammasome-and-innate-immunity naming space (NOD, NLR, CLR, caterpiller) that gives no hint of its antigen-presentation function; a reader checking 'is this the class I transactivator?' from the UniProt protein name alone gets no confirmation, which is the rule 65 failure mode applied to a functional rather than a family question. Kobayashi KS et al. 2012
  2. E2established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.review

    NLRC5 is the transactivator of MHC class I-dependent immune responses, the class I counterpart of the class II transactivator CIITA.

    NLRC5: a key regulator of MHC class I-dependent immune responses. · Kobayashi KS, van den Elsen PJ · Nat Rev Immunol 2012
    PMID 23175229DOI
    What this evidence cannot show

    A 2012 Nature Reviews Immunology review, included as a map of the field rather than as primary evidence. Nothing in it is about leukaemia or about KMT2A. Read: title and PubMed record fetched 2026-09-20; the functional statement it supports here is also stated independently in the primary source E3.

    Evidence that argues against this (1)
    • Reviews of a transactivator tend to present it as necessary and sufficient; Sparbier's data show that in cells with intact polycomb, NLRC5-RFX5 is not sufficient - the chromatin state gates it, and KMT2A/B are needed to open that gate. The transactivator model on its own predicts the wrong answer in a bivalent promoter. Sparbier CE et al. 2023
  3. E3strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    In K-562 cells, IFN-gamma-induced MHC class I expression in dual KMT2A/KMT2B knockout cells was entirely dependent on the NLRC5-RFX5 enhanceosome; removing H3K27me3 with a menin inhibitor and/or an EZH2 inhibitor, or expressing H3.3 K27M, allowed IFN-gamma to induce MHC-I as robustly in KMT2A/KMT2B knockout cells as in controls, and additional depletion of SETD1A and SETD1B in those cells blocked RFX5-dependent MHC-I activation; KMT2A/KMT2B knockout cells nevertheless showed a variable delay in MHC-I activation.

    Targeting Menin disrupts the KMT2A/B and polycomb balance to paradoxically activate bivalent genes. · Sparbier CE, Gillespie A, Gomez J, Kumari N, Motazedian A, Chan KL, et al., Dawson MA · Nat Cell Biol 2023
    PMID 36635503DOI
    What this evidence cannot show

    K-562 only; nothing about a KMT2A-rearranged cell, and the knockouts remove both KMT2A alleles rather than modelling a fusion. The bypass requires H3K27me3 to be removed first, a condition supplied pharmacologically here and of unknown status in a patient's blast. RFX5 rather than NLRC5 itself was the primary genetic handle in the bypass experiment, with NLRC5 characterised through the enhanceosome; the two are treated together by the authors. Read: full text via Europe PMC fullTextXML (PMC7614190), 2026-09-20.

    Evidence that argues against this (1)
    • The same experiment cuts against reading KMT2A/B as dispensable: maximal MHC-I was reached in the knockouts but activation was variably delayed, and the authors conclude KMT2A/B 'may facilitate uniform transcriptional activation'. A delay and a loss of uniformity in a population of blasts under immune pressure is not a null result - it is exactly the kind of heterogeneity from which an antigen-low subclone is selected. Sparbier CE et al. 2023
  4. E4moderate confidence 3 of 5: One strong study, or several smaller consistent ones.dataset

    As of 2026-09-20 `(KMT2A OR MLL) AND leukemia AND NLRC5[tiab]` returns 0 PubMed records, and the control `NLRC5[tiab] AND leukemia` returns 2, both excluded on title check - one on chicken NLRC5 signalling in avian leukosis virus subgroup J infection and one a Mendelian-randomisation study of beta-2 microglobulin and B-cell malignancies.

    What this evidence cannot show

    Bounded absence that is accurate as a measurement and misleading as a conclusion, and is recorded here for that reason. Name list: NLRC5 only; not searched were CITA, NOD27, NOD4 or CLR16.1, nor the phrase 'class I transactivator'. The relationship that does exist - KMT2A/KMT2B knockout making IFN-gamma-induced MHC-I entirely NLRC5-RFX5-dependent - lives in a figure of PMID 36635503 and is unreachable from any symbol query.

    Evidence that argues against this (1)
    • PMID 36635503 is the refutation of the conclusion this zero invites. It is a direct epistatic experiment between the KMT2A family and the NLRC5-RFX5 enhanceosome, it is already cited five times in this beat, and no symbol-keyed sweep in this disease would ever find it. Sparbier CE et al. 2023
  5. E5moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper

    In Zhou's published 9,292-locus menin and H2AK119ub target set from KMT2A-rearranged AML, NLRC5 falls in the MENIN-SOLO group at chr16:56989555 - menin-bound without H2AK119ub - while CIITA, the class II counterpart transactivator, falls in the H2AK119ub-solo, menin-unbound group in the same table.

    Epigenetic regulation of noncanonical menin targets modulates menin inhibitor response in acute myeloid leukemia. · Zhou X, Zhang L, Aryal S, Veasey V, Tajik A, Restelli C, Moreira S, Zhang P, et al. · Blood 2024
    PMID 39158067DOI
    What this evidence cannot show

    Bounded by the source system: Zhou's locus set is KMT2A-rearranged ADULT MYELOID AML, largely OCI-AML2 (CVCL_1619, KMT2A::AFDN) and MV4-11 plus mouse models, not infant B-lineage KMT2A::AFF1. The intervals are 2 bp TSS anchor points carrying a gene id, so this supports a GENE-LEVEL intersection and not a base-resolution one. Most important: ABSENCE FROM THE TABLE IS NOT ABSENCE OF BINDING - it is absence from Zhou's called set under Zhou's thresholds in Zhou's cells. Symbol-to-ENSG resolution refuses any symbol mapping to more than one ENSG, which is why HLA-A, HLA-B, HLA-DRA, HLA-DRB1 and B2M are all unresolved in this join and the class I layer is invisible to it. Join run 2026-09-20; the Zhou PubMed record and DOI were fetched independently by this beat rather than copied. Zhou's paper never discusses NLRC5; this is the join's readout of its supplementary Table 4. A menin-solo call is an occupancy call, not a functional one - it does not show that menin regulates NLRC5 transcription, and Zhou's own finding is that menin-solo canonical targets behave differently from the bivalent noncanonical ones the paper is about.

    Evidence that argues against this (1)
    • Menin occupancy at a promoter is not evidence that menin inhibition changes that gene: Zhou's whole point is that menin-bound loci divide into classes with different responses, and Sparbier separately found that VTP50469 induced remarkably few significant gene expression changes despite globally reducing menin chromatin occupancy. A menin-solo NLRC5 could be unchanged by the drug. Sparbier CE et al. 2023
  6. E6strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    Nuclear import and export mechanisms for NLRC5 coexist and counterbalance each other, GCN5 retains NLRC5 in the nucleus, and it is the nuclear distribution of NLRC5 - not merely its abundance - that determines MHC class I transactivation.

    PMID 38519032DOI
    What this evidence cannot show

    Domain-deletion and transfection analysis in cell lines; abstract-level reading. No leukaemic and no KMT2A-rearranged system. Overexpressed constructs can distort nucleocytoplasmic ratios, so the quantitative balance in an endogenous setting is not established by this work.

    Evidence that argues against this (1)
    • The size of the physiological nuclear pool remains unknown, because this analysis is domain-deletion and overexpression based and overexpressed constructs saturate import receptors and mislocalise. Both of UniProt's coded citations and HPA's independent antibody immunofluorescence report an endogenous steady state that is cytoplasmic, with no nuclear call at all. The defensible reading is that a regulated import route EXISTS, not that NLRC5 is substantially nuclear at rest. Human Protein Atlas 2026
  7. E7strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.database

    UniProt Q86WI3 records NLRC5 only as cytoplasmic, on two coded citations that are both antiviral/NF-kB signalling papers rather than class-I or localisation studies, and HPA immunofluorescence likewise gives main location Cytosol (Centrosome additional, reliability Supported).

    What this evidence cannot show

    Brief rule 91: the ECO:0000269 codes point at PMID 20061403 (IFN-dependent antiviral responses) and PMID 20434986 (negative regulation of NF-kB and type I interferon signalling). Neither is about where NLRC5 is, and the record carries no nuclear line at all despite a dedicated 2024 study of NLRC5 nuclear import. Reading this record alone would leave a reader believing the MHC class I transactivator never enters the nucleus.

    Evidence that argues against this (1)
    • A dedicated study of NLRC5 nucleocytoplasmic trafficking establishes a regulated nuclear pool that UniProt does not record. J Biol Chem 2024
  8. E8strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    A bacterial effector, Orientia tsutsugamushi Ank5, inhibits MHC class I expression by promoting cytoplasmic retention and degradation of NLRC5 - independent confirmation that keeping NLRC5 out of the nucleus is sufficient to switch off class I.

    PMID 39277599
    What this evidence cannot show

    An infection model in non-haematopoietic host cells; abstract-level reading. It establishes that the import step is a controllable choke point, not that any human tumour uses it.

    Evidence that argues against this (1)
    • A pathogen effector shows the import step CAN be blocked, not that it is rate-limiting in uninfected cells - and the class I readout is not cleanly attributable to nuclear exclusion, because NLRC5 also has cytoplasmic signalling functions, negatively regulating NF-kB and type I interferon pathways. Forcing NLRC5 into the cytoplasm and degrading it perturbs both arms at once, so the experiment cannot separate 'class I fell because NLRC5 left the nucleus' from 'class I fell because interferon signalling was disturbed'. Cell 2010
  9. E9strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.database

    Human Protein Atlas v25.1 records NLRC5 RNA as tissue-enhanced in bone marrow (33.8 nTPM) and lymphoid tissue (31.8 nTPM), its two highest tissues, and detected in all immune cell types.

    What this evidence cannot show

    RNA rather than protein, normal tissue rather than leukaemia. The subcellular arm of the same record (Cytosol main, Centrosome additional) reports the resting state and cannot see a regulated nuclear pool.

    Evidence that argues against this (1)
    • A resting-tissue RNA atlas measures basal tone, not capacity, and for this gene that distinction is large: NLRC5 is interferon-inducible, so its abundance in an inflamed or interferon-exposed marrow is not the number reported here. The confound also runs the other way - bone marrow and lymphoid tissue are the most immune-cell-dense tissues sampled, so a high tissue-level nTPM partly reports cellular composition rather than per-cell expression. The haematopoietic centring of this gene is likely real and is not established by these two numbers. J Immunol 2010
Connected to