Oncofusion

start / graph / gene-ccnt1

Proteinmoderate confidence 3 of 5: One strong study, or several smaller consistent ones.Q5c-interactome2

CCNT1 / cyclin T1 — the regulatory half of P-TEFb, and the one part of the elongation core with any dependency variance

also called CCNT1 · cyclin T1 · cyclin-T1 · CycT1

Cyclin T1 is the partner that switches on CDK9, the kinase that tells RNA polymerase to start moving. The KMT2A fusion pulls this pair in through its own scaffold protein, and unlike CDK9 itself, cyclin T1 is not needed by every cell — which makes it the more interesting half.

The same thing, for a scientist

CCNT1 (O60563, 726 aa) is the regulatory cyclin of positive transcription elongation factor b (P-TEFb, CDK9/cyclin T1), an obligate module of the Super Elongation Complex and of Yokoyama's AEP. It is recruited to the KMT2A fusion via the AFF1/AFF4 scaffold rather than by direct contact with KMT2A. Unlike its catalytic partner CDK9, which is DepMap common-essential at -1.54 and depleted in 99.3% of lines, CCNT1 is not common-essential (mean Chronos -0.206, 8.4% depleted) and shows a nominal leukemia-restricted correlation with MEN1 of r = 0.349 that fails Holm correction.

What it looks like
CCNT1726 aa
Nuclear localization signal; also the Tat-TAR RNA interaction motif · residues 253–270CoilCoiled coil · residues 384–4250200400600
loading AF-O60563-F1…
AlphaFold AF-O60563-F1 ↗
colourshape

This is a prediction, not a measurement. No one has seen this protein in this shape. The colours say how much the software trusts itself, and that is worth reading before anything else — the orange stretches are places the model is effectively saying this part has no fixed shape.

very high (90+)confident (70–90)low (50–70)very low (<50)
I have never read one of these pictures before

A protein is one long chain of amino acids that folds up. The ribbon traces that chain from one end to the other; it is not a wire, it is a simplification of where the backbone runs. Drag to rotate, scroll to zoom.

Domains are the stretches of chain that do a particular job, and they are what the coloured blocks above the picture mark out. Click one and the view flies to it. In this project the important question about any domain is usually whether the fusion still has it — the labels say lost where it does not.

Anything in yellow sticks is a bound small molecule, which is usually the drug.

Go deeper
The asymmetry inside P-TEFb

This graph already carries protein-ptefb-cdk9-ccnt1 as a module and target-cdk9-inhibition-kmt2a-r as a target dossier, and has concluded that the hijacked elongation machinery is pan-essential and therefore offers no therapeutic window. That conclusion is correct for CDK9. It is not obviously correct for cyclin T1 3:

Genemean Chronos% lines depletedcommon essentialr vs MEN1 (leukemia)
CDK9-1.53699.3%yes-0.087
CCNT1-0.2068.4%no0.349

Cyclin T1 has an order of magnitude less essentiality and, unlike CDK9, enough variance across lines to produce a signal. Its leukemia-restricted correlation with MEN1 is 0.349 (95% CI 0.038–0.585, permutation p = 0.012), which fails Holm correction across 16 genes but does not cross zero 4.

Two cautions before anyone gets excited. First, a cyclin can be non-essential in a CRISPR screen because a paralogue substitutes — cyclin T2 exists and can pair with CDK9 — so a low Chronos score here may mean redundancy rather than dispensability. Second, a nominal correlation that fails correction is not a finding. What this justifies is testing CCNT1 in a larger leukemia panel, nothing more.

How it reaches the fusion

Cyclin T1 does not touch KMT2A. P-TEFb is brought in by the AFF1/AFF4 scaffold: Lin and colleagues defined AFF4 as a component of the ELL/P-TEFb elongation complex and a shared subunit of MLL chimeras 1, and Yokoyama and colleagues independently defined a higher-order complex containing AF4 and ENL family proteins with P-TEFb that facilitates oncogenic and physiologic MLL-dependent transcription 2. Since AFF1 is the 3' partner in KMT2A::AFF1, the fusion protein itself carries the P-TEFb docking module.

The cyclin T1–AFF4 contact is structurally resolved, though in an HIV Tat context: 2PK2, 3BLH (CDK9/cyclin T1, 2.48 Å) and the Tat–P-TEFb–AFF4 structures define the geometry 5. Twenty-nine PDB entries cover this protein, many of them CDK9/cyclin T1 with inhibitors bound — chemical matter aimed at the kinase, not at the cyclin.

Constraint and expression

gnomAD: pLI ~1.0, LOEUF 0.397, 18 observed against 67.3 expected loss-of-function variants 3. Constrained.

Expression is low-specificity and modest: bone marrow 27.8, thymus 12.1, testis 10.7, spleen 9.9 nTPM, with classical monocytes at 2.6 and naive B cells at 5.4 3. No selectivity handle. No verified mouse knockout; we searched and did not find one, and none is asserted here.

Caveat added: the KMT2A-rearranged dependency is myeloid-driven

Added 2026-09-20 after a lineage split that this entry did not carry. On DepMap via Open Targets, re-queried the same day, CCNT1's KMT2A-rearranged requirement is -0.485 in the myeloid half against -0.085 in the lymphoid half.

That is close to the whole effect sitting in the wrong lineage for this project. The index disease is infant B-ALL, and a dependency that is five times stronger in myeloid lines is a dependency of the comparator arm. This graph has made exactly this mistake before in the other direction - inheriting the differentiation narrative, which is a myeloid story, into a lymphoid disease.

The entry's MEN1-correlation claim is untouched by this and is not being revised.

Facts and figures
uniprotO60563
gene symbolCCNT1
ensemblENSG00000129315
length AA726
alphafold IDAF-O60563-F1
PDB count29
complexP-TEFb (CDK9/cyclin T1); recruited to the KMT2A fusion via the AFF1/AFF4 scaffold, not by direct KMT2A contact
druggabilityChemical matter exists for the CDK9 ATP pocket, not for cyclin T1. The cyclin T1-AFF1 interface is unaddressed. No ChEMBL count quoted because a target-level count for this heterodimer would conflate the kinase pocket with the cyclin.
subnuclear localizationnuclear speckles, with cyclin T1 as the targeting subunit: a central region of cyclin T1 is required for speckle accumulation and speckle-localised cyclin T1 recruits CDK9 and Tat (PMID 11282025). The corollary is the informative part - cyclin T1 is exclusively nuclear while CDK9 shuttles and is exported, and raising cyclin T1 drives CDK9 nuclear (PMID 12115727), so the ratio of the two subunits sets where the kinase is.
shown for fusionNO. Measured 2026-09-21 with NCBI esearch: (CCNT1[tiab] OR cyclin T1[tiab] OR cyclin-T1[tiab]) AND (MLL[tiab] OR KMT2A[tiab] OR MLL-AF4[tiab] OR MLL-AF9[tiab] OR mixed lineage leukemia[tiab]) AND (localiz*[tiab] OR localis*[tiab] OR subcellular[tiab] OR nuclear body[tiab] OR nuclear bodies[tiab] OR speckle*[tiab] OR condensat*[tiab] OR nucleolar[tiab] OR immunofluorescence[tiab] OR compartment*[tiab]) returns 0. Operands: gene clause 477, gene AND fusion 4 - so the CCNT1-and-KMT2A literature is four papers wide and none makes a spatial claim. Positive control through the identical shape with (MLLT1[tiab] OR ENL[tiab]) returns 23. Zero in titles and abstracts; supporting figures not searched (brief rule 80). This is the single largest spatial gap in this graph's elongation model: every mechanistic account of KMT2A::AFF1 terminates at P-TEFb recruitment through AFF4/SEC, the SEC-condensate literature makes explicit spatial claims, and NEITHER subunit of P-TEFb has ever been imaged in a KMT2A-rearranged cell.
PDB verified
[
  "3BLH",
  "6Z45",
  "7NWK"
]
domains
[
  {
    "name": "Nuclear localization signal; also the Tat-TAR RNA interaction motif",
    "start": 253,
    "end": 270,
    "pfam": null
  },
  {
    "name": "Coiled coil",
    "start": 384,
    "end": 425,
    "pfam": null
  }
]
partner comparison
[
  {
    "gene": "CDK9",
    "chronos_mean": -1.5359,
    "frac_depleted": 0.9934,
    "common_essential": true,
    "r_leukemia_men1": -0.0868
  },
  {
    "gene": "CCNT1",
    "chronos_mean": -0.2059,
    "frac_depleted": 0.0844,
    "common_essential": false,
    "r_leukemia_men1": 0.3486
  }
]
essentiality
{
  "depmap_common_essential": false,
  "chronos_mean": -0.2059,
  "chronos_mean_leukemia_lines": -0.2,
  "n_lines": 1258,
  "fraction_lines_depleted": 0.0844,
  "gnomad_pLI": 0.99998,
  "gnomad_loeuf": 0.3969,
  "gnomad_obs_lof": 18,
  "gnomad_exp_lof": 67.3,
  "mouse_ko": "Not established; no Ccnt1 knockout citation verified and none asserted",
  "hpa_ntpm": {
    "bone_marrow": 27.8,
    "thymus": 12.1,
    "lymph_node": 10.9,
    "testis": 10.7,
    "spleen": 9.9,
    "liver": 7.9,
    "heart_muscle": 6.4,
    "naive_B_cell": 5.4,
    "classical_monocyte": 2.6
  },
  "hpa_specificity": "Low tissue specificity"
}
codependency men1
{
  "r_leukemia_lines": 0.3486,
  "ci_leukemia": [
    0.0375,
    0.5847
  ],
  "crosses_zero": false,
  "p_permutation": 0.0122,
  "p_holm": 0.1464,
  "survives_holm": false
}
localization
{
  "subcellular": "EXCLUSIVELY NUCLEAR - and it is the subunit that makes P-TEFb nuclear. UniProt O60563 annotates Nucleus only, at ECO:0000269. Cyclin T1 is exclusively nuclear and speckled while CDK9 shuttles and is actively exported; enforced cyclin T1 expression pulls CDK9 into the nucleus. That asymmetry is the localisation fact about P-TEFb that matters, and it belongs to CCNT1.",
  "subnuclear": "NUCLEAR SPECKLES, and cyclin T1 is the address-bearing subunit. Cdk9 and cyclin T1 are distributed through the non-nucleolar nucleoplasm with foci, colocalising with splicing-factor markers - more strongly for cyclin T1 - and a region in the CENTRAL portion of cyclin T1 is required for speckle accumulation. Cyclin T1 that accumulates at speckles recruits CDK9 and HIV Tat there. Both proteins show only limited colocalisation with phosphorylated Pol II, which argues against reading speckles as the elongation site itself.",
  "how_shown": "high-resolution immunofluorescence with splicing-factor and phospho-Pol II co-staining, transcription-inhibitor treatment, and transient expression of cyclin T1 deletion mutants (PMID 11282025); leptomycin B export block and cyclin T1 overexpression with CDK9 compartment readout (PMID 12115727); UniProt O60563 SUBCELLULAR LOCATION comment.",
  "systems": "cultured human cells in an HIV-transcription setting; no leukaemia and no KMT2A-rearranged cell has ever been used to localise cyclin T1",
  "node": "localization-nuclear-speckles-kmt2a-r"
}
cell type
{
  "expressed_in": "RESTING CD4-positive T lymphocytes at low level, rising with cytokine exposure: IL-2 plus IL-6 plus TNF-alpha raise cyclin T1 and CDK9 protein and TAK/P-TEFb activity in purified resting CD4-positive cells, apparently without requiring proliferation [E_q66_cell1]. This is the best-characterised cell-type-and-state regulation of any elongation-core subunit, and it comes entirely from the HIV-latency field.",
  "required_in": "Not established as a cell-type-restricted requirement. CCNT1 is not DepMap common-essential - mean Chronos -0.206 across 1,258 lines with 8.4 percent depleted - while its catalytic partner CDK9 is pan-essential [E3], and CCNT1's KMT2A-rearranged dependency splits by lineage with a mean gene effect of -0.485 in the myeloid half of the index arm [E-ccnt1-lineage]. So the requirement, such as it is, is lineage-conditioned rather than cell-type-specific.",
  "not_required_in": "The lymphoid half of the KMT2A-rearranged DepMap arm is the weaker side of the split [E-ccnt1-lineage] - which is the half the index disease belongs to.",
  "how_shown": "cytokine stimulation of purified resting primary human CD4-positive T lymphocytes with protein-level and kinase-activity readouts (PMID 11689614); genome-scale CRISPR fitness with a Cellosaurus-based lineage split (this entry's E3 and E-ccnt1-lineage)",
  "nodes": [
    "cell-type-early-t-precursor",
    "cell-type-monocyte"
  ]
}
cell state
{
  "expressed_in_state": "QUIESCENCE is the state cyclin T1 is LOW in. Cyclin T1 protein and P-TEFb activity are low in resting CD4-positive T lymphocytes and rise on cytokine exposure without proliferation [E_q66_cell1]. Because this graph invokes quiescence to explain measurable residual disease, a P-TEFb regulatory subunit that is down in quiescent cells is directly relevant: a CDK9-directed drug would meet less of its target in exactly the cells that survive therapy.",
  "maintains_state": "not established. No experiment shows cyclin T1 maintaining a cell identity; the readouts are protein level, kinase activity and fitness.",
  "state_dependent_localisation": "P-TEFb's compartment is itself state-dependent through this subunit: cyclin T1 is exclusively nuclear, CDK9 shuttles, and raising cyclin T1 drives CDK9 into the nucleus [E_q66_loc1 counter-evidence]. A cytokine-poor, cyclin-T1-low quiescent cell should therefore hold a larger cytoplasmic CDK9 fraction, and nobody has looked.",
  "how_shown": "cytokine induction in resting primary lymphocytes (PMID 11689614); leptomycin B export block with cyclin T1 overexpression (PMID 12115727)",
  "nodes": [
    "cell-state-quiescence"
  ]
}
How this could bite us back
  • Redundancy: cyclin T2 can pair with CDK9, so removing cyclin T1 may simply be compensated, and a CRISPR screen cannot distinguish that from dispensability.
  • The complex serves physiologic MLL-dependent transcription as well as oncogenic transcription, which is the structural reason this module has no therapeutic window however it is approached.
  • Chemical reality: every molecule aimed at this heterodimer hits the CDK9 ATP pocket, and CDK9 is required by 99.3% of cell lines.
  • Target abundance is a state variable, not a constant: cyclin T1 is low in resting CD4-positive lymphocytes and induced by IL-2, IL-6 and TNF-alpha without proliferation (PMID 11689614). A CDK9/P-TEFb inhibitor therefore meets a different amount of assembled complex in a quiescent marrow-resident blast than in a cycling one, and no CDK9-inhibitor study in this disease reports cyclin T1 level as a covariate.
  • Compartment as a resistance route: CDK9 is actively exported and cyclin T1 abundance is what retains it in the nucleus (PMID 12115727). A cell that lowers cyclin T1 would shift its kinase partly to the cytoplasm - reducing the nuclear target pool without changing total CDK9, which is what a western blot would measure.
  • The speckle rationale is weaker than it sounds: in the same experiment that put cyclin T1 in speckles, both subunits showed only LIMITED colocalisation with phosphorylated Pol II (PMID 11282025). Using speckle residence as evidence that P-TEFb acts there overreads the result, and a condensate-disrupting strategy justified that way would be justified by co-staining with splicing factors rather than with the polymerase.
Still open
  1. Is CCNT1's low essentiality real, or is cyclin T2 substituting? A CCNT1/CCNT2 double-knockout comparison in a KMT2A-r line would settle whether the cyclin slot has a window that the kinase slot does not.
  2. Does the nominal CCNT1-MEN1 correlation survive in a larger leukemia panel? It has now been reproduced at r = 0.349 in two independent runs of this project and has never reached significance in either.
  3. Is the cyclin T1-AFF1 interface distinct enough from the cyclin T1-AFF4 interface to be targeted selectively? The structural work is all on AFF4 and on HIV Tat, and AFF1 is the fusion partner that matters here.
  4. Image P-TEFb in a KMT2A-rearranged cell. Neither CDK9 nor cyclin T1 has ever been localised in one, while every mechanistic account of this fusion ends at P-TEFb recruitment and the SEC-condensate literature makes spatial claims. One co-stain of cyclin T1 with SC35 and with the fusion in an index line is the cheapest missing measurement in this graph's elongation model.
  5. Is cyclin T1 lower in the quiescent, therapy-surviving fraction of KMT2A-rearranged blasts than in the bulk? If it is, CDK9 inhibitors are least effective where they are most needed, and the fix is scheduling rather than potency.
  6. Does the cyclin T1:CDK9 ratio differ between the myeloid and lymphoid halves of the KMT2A-rearranged arm? The DepMap split in this entry is already lineage-dependent, and a subunit-ratio difference would explain it without invoking a fusion-partner difference.
Every source behind this entry (8)
  1. E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper

    AFF4 is a component of the ELL/P-TEFb elongation complex and a shared subunit of MLL chimeras, linking transcription elongation to leukemia.

    AFF4, a component of the ELL/P-TEFb elongation complex and a shared subunit of MLL chimeras, can link transcription elongation to leukemia. · Lin C, Smith ER, Takahashi H, Lai KC, Martin-Brown S, Florens L, Washburn MP, Conaway JW, Conaway RC, Shilatifard A · Molecular Cell 2010
    PMID 20159561DOI
    What this evidence cannot show

    Establishes P-TEFb as part of the complex; does not test cyclin T1 individually as a dependency.

    Evidence that argues against this (1)
    • Complex composition is not requirement. The AEP/SEC papers establish what co-purifies and what a reconstituted complex does; they do not establish that each subunit is rate-limiting in a patient's leukaemia, and this project's own dependency data separates the members by an order of magnitude. Nature 2012
  2. E2established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper

    A higher-order complex containing AF4 and ENL family proteins with P-TEFb facilitates both oncogenic and physiologic MLL-dependent transcription.

    PMID 20153263DOI
    What this evidence cannot show

    The phrase 'oncogenic and physiologic' matters: the same complex serves normal MLL-dependent transcription, which is precisely why targeting it has no window.

    Evidence that argues against this (1)
    • Complex composition is not requirement. The AEP/SEC papers establish what co-purifies and what a reconstituted complex does; they do not establish that each subunit is rate-limiting in a patient's leukaemia, and this project's own dependency data separates the members by an order of magnitude. Facilitating BOTH oncogenic and physiological KMT2A-dependent transcription is the clause that bounds the therapeutic reading. Nature 2012
  3. E3strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.database

    CCNT1 is not DepMap common-essential (mean Chronos -0.206 across 1,258 lines, 8.4% depleted) while its catalytic partner CDK9 is common-essential at -1.536 with 99.3% depleted; gnomAD gives CCNT1 pLI ~1.0 and LOEUF 0.397 with 18 observed against 67.3 expected loss-of-function variants; HPA gives bone marrow 27.8 nTPM with low tissue specificity.

    What this evidence cannot show

    DepMap release version not stated by the API. Cyclin T2 can substitute for cyclin T1 in P-TEFb, so a low CCNT1 Chronos score may reflect paralogue redundancy rather than dispensability of the cyclin function. This is an important alternative explanation that the screen cannot distinguish.

    Evidence that argues against this (1)
    • P-TEFb has already been tested in the clinic in acute leukemia and the result was toxicity, not selectivity. In a randomised phase II in relapsed/refractory AML, the arm containing alvocidib (flavopiridol), a CDK9/P-TEFb inhibitor, gave a 28% overall response, better than the two comparator regimens, but 7 of the 9 treatment-related deaths across the whole trial occurred in that arm, 4 of them in elderly patients, and the authors advise using the regimen with caution in the elderly. Cyclin T1 is the regulatory half of the same enzyme. Whatever dependency variance CCNT1 shows in DepMap, the pharmacology of shutting down P-TEFb in a patient is dominated by the general transcription requirement, not by the fusion. Litzow MR et al. 2018
  4. E4weak confidence 2 of 5: Preliminary, small, indirect, or not yet peer reviewed.dataset

    Across 54 DepMap leukemia lines CCNT1 correlates with MEN1 at r = 0.349 (95% CI 0.038-0.585, permutation p = 0.012), which does not survive Holm correction across the 16 genes tested (Holm p = 0.146).

    Leukemia-restricted co-dependency with MEN1, bootstrap and permutation testing; analyses/wrad-discrimination/ in this project · Broad Institute DepMap; Open Targets Platform; analysis run by this project · Database and derived analysis 2026
    What this evidence cannot show

    Our own computation in 54 lines, failing multiple-testing correction. The cycle-9 bootstrap independently put CCNT1 at r = 0.349 with the same conclusion, so the value is reproducible but still not significant. Do not cite CCNT1 as a co-dependency.

    Evidence that argues against this (1)
    • This nominal correlation should be read against the deeper null, not the 16-gene panel. Re-run against a 2,743-gene variance-matched pool with Holm correction across the same 54 leukemia lines, the surviving set is KMT2A, PSIP1, DOT1L, FLT3, MEF2C and KAT7; MEIS1 (r 0.468) fails at Holm 0.067 and DPY30, MLLT1 and BCL2 fail as well. CCNT1's r = 0.349 is below all of those, and the 16-gene Holm p of 0.146 already failed. There is no CCNT1 co-dependency signal to act on, and the separate clinical record for P-TEFb inhibition in AML is one of toxicity. Broad Institute DepMap; Open Targets Platform; analysis run by this project 2026
  5. E5established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database

    Verified crystal structures of the CDK9/cyclin T1 heterodimer exist at 2.48 A (3BLH) and with inhibitors bound at 3.37 A (6Z45) and 2.81 A (7NWK).

    What this evidence cannot show

    The inhibitor structures target the CDK9 ATP pocket, not cyclin T1. No small molecule is known to us that targets cyclin T1 itself or the cyclin T1-AFF1 interface.

    Evidence that argues against this (1)
    • Counting PDB entries measures how tractable a protein was to CRYSTALLISE, which correlates with being small, ordered and well-behaved rather than with being druggable. What decides design value is whether the deposited density covers the site that matters, at a resolution that supports modelling, in the relevant state — and this graph has repeatedly found structures described in the direction that overstates what is resolved. Two inhibitor complexes at 3.37 A and 2.81 A are at the resolution where ligand placement is interpretable and water structure is not, which limits what they support for optimisation. RCSB PDB 2026
  6. E-ccnt1-lineagemoderate confidence 3 of 5: One strong study, or several smaller consistent ones.database

    CCNT1's KMT2A-rearranged dependency splits by lineage: mean gene effect -0.485 in the myeloid half of the index arm against -0.085 in the lymphoid half.

    What this evidence cannot show

    Re-queried 2026-09-20. **Moderate**: the lymphoid half of the DepMap KMT2A-rearranged arm is four cell lines, and this project has documented that the arm is myeloid-enriched and contains no infant line. A four-line mean is a reason to stop quoting an unsplit number, not a measurement of lymphoid biology.

    Evidence that argues against this (1)
    • A lineage split is not automatically a confound - this graph's Fanconi work found the signal surviving within lineage and the lymphoid half STRONGER, which is the opposite shape. The right conclusion here is that CCNT1's number should not be quoted unsplit, not that CCNT1 is irrelevant to B-lineage disease. PNAS 2014
  7. E_q66_loc1strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    Cyclin T1 is the subunit that carries P-TEFb to nuclear speckles, not the passenger: high-resolution immunofluorescence places Cdk9 and cyclin T1 throughout the non-nucleolar nucleoplasm with increased signal at numerous foci, with significant colocalisation with splicing-factor antibodies marking nuclear speckles - ESPECIALLY for cyclin T1 - a region in the CENTRAL portion of cyclin T1 is required for speckle accumulation, and cyclin T1 proteins that accumulate at speckles recruit Cdk9 AND HIV Tat to that compartment; the distribution is altered by transcription inhibitors.

    PMID 11282025DOI
    What this evidence cannot show

    Cultured human cells in an HIV-transcription setting, not haematopoietic and not leukaemic, and the recruitment half of the claim comes from OVEREXPRESSION of cyclin T1 deletion mutants, so the domain mapping is a gain-of-function result. Cdk9 and cyclin T1 showed only limited colocalisation with phosphorylated Pol II forms in the same experiment, which sits awkwardly with reading speckles as the site of elongation. Read: abstract only. Identifiers copied field by field from the efetch record fetched 2026-09-21, ArticleId anchored per brief rule 81.

    Evidence that argues against this (1)
    • The partner's compartment is not fixed either, and cyclin T1 is what fixes it: CDK9 shuttles between nucleus and cytoplasm, is actively exported in a leptomycin-B-sensitive manner, and ENFORCED EXPRESSION OF CYCLIN T1 enhances its nuclear localisation, while cyclin T1 itself is exclusively nuclear and speckled. So the P-TEFb compartment is set by cyclin T1 abundance, which is precisely the quantity that varies across haematopoietic cell types and activation states - meaning a speckle assignment made in one cell state does not transfer. Napolitano G; Licciardo P; Carbone R; Majello B; Lania L 2002
  8. E_q66_cell1strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    Cyclin T1 is a CELL-STATE-REGULATED protein in primary human lymphocytes, and its induction does not require proliferation: treating peripheral blood lymphocytes and purified RESTING CD4-positive T lymphocytes with the combination of IL-2, IL-6 and TNF-alpha increased Cdk9 and cyclin T1 protein levels and increased TAK/P-TEFb enzymatic activity, and the cytokine induction in resting CD4-positive cells did not appear to require lymphocyte proliferation.

    PMID 11689614DOI
    What this evidence cannot show

    Primary human resting CD4-positive T lymphocytes in an HIV-latency setting - a normal lymphoid cell type, not a leukaemic one, and not B-lineage. The readout is protein level plus kinase activity in bulk, not per-cell. 'Did not appear to require proliferation' is a negative inference from the experiment's design rather than a direct measurement of cycling status. Read: abstract only.

    Evidence that argues against this (1)
    • Making cyclin T1 the state-regulated subunit cuts against the graph's use of P-TEFb as a fixed target: if the regulatory subunit's abundance rises and falls with the cytokine environment, then CDK9-directed inhibitor potency in vivo depends on marrow niche signalling, and no CDK9-inhibitor study in this disease reports cyclin T1 level. The dependency data in this entry already show CCNT1 is not DepMap common-essential (mean Chronos -0.206 across 1,258 lines, 8.4 percent depleted) while its catalytic partner is pan-essential - a variance that a state-regulated subunit would produce. Oncofusion knowledge graph; DepMap Chronos via Open Targets 2026
Connected to