BGPT: Paper Review: CDK11 in TREX/THOC Regulates HIV mRNA 3′ End Processing

Explore by Goal

Quick Explanation Copied

Central claim (from the paper): CDK11—embedded in the TREX/THOC mRNA-processing/export machinery—promotes HIV 3′ end processing by increasing RNAPII Ser2 phosphorylation near the viral 3′ end, recruiting cleavage/polyadenylation (CPA) factors, shortening readthrough transcripts, and yielding longer, more stable HIV poly(A) tails.

Long Explanation

Paper Review (Visual + Skeptical): CDK11 in TREX/THOC Regulates HIV mRNA 3′ End Processing

Cell Host & Microbe (2015) • DOI: 10.1016/j.chom.2015.10.012

One-sentence mechanistic model (as stated by the authors): TREX/THOC recruits CycL2:CDK11 to HIV transcription sites → CDK11 raises RNAPII Ser2P (especially near the HIV 3′ end) → CPA machinery (CstF77, PAP, etc.) is assembled at the poly(A) region → efficient cleavage/polyadenylation reduces 3′ readthrough and increases poly(A) tail length and mRNA stability → increased HIV gene expression and replication.

Figure A — Reported magnitude of HIV output changes

Figure B — CDK11 depletion shifts HIV transcript abundance (US vs spliced)

Skeptical note: the paper reports fold-changes (e.g., “US reduced ~2-fold” and spliced “~5-fold”) rather than normalized values; the bars above are a visual approximation for readability.

Figure C — CDK11/THOC1 depletion increases HIV 3′ readthrough

Approximation policy: the text states “20-to 30-fold” for the HeLa P4 readthrough assay; the figure uses the midpoint (25) purely for visualization.

Figure D — CDK11 extends HIV poly(A) tails and increases transcript stability

Skeptical caution: tail length is reported in nt ranges and stability as hours; the combined grouped bar is for qualitative co-visualization only.

Evidence chain (what the paper did, and what it supports)

1) Physical association: CDK11 (p110) co-purifies with TREX/THOC-associated proteins by affinity purification + mass spectrometry; most recovered interactors are TREX/THOC subunits; co-immunoprecipitation is shown for tagged and endogenous CDK11 with TREX/THOC components.

2) Functional impact on HIV output: CDK11 overexpression boosts HIV transcription reporter activity and p24 levels in 293T; it also increases HIV-related protein expression and replication readouts; CDK11 affects other promoters too (suggesting a broad transcriptional role, not HIV-specificity).

3) HIV transcript class effects: CDK11 depletion lowers all major HIV transcript classes (US, SS, MS), with a stronger relative impact on spliced transcripts; this is consistent with CDK11 influencing mRNA maturation steps, though the authors argue the major mechanistic effect is 3′ end processing rather than splicing.

4) Mechanistic kinase logic (CTD phosphorylation): CDK11 can phosphorylate RNAPII CTD Ser2P and Ser5P in vitro using a GST-CTD substrate; the authors further test CK2 dependence via siCK2 depletion.

5) In-cell 3′ end processing and CPA factor recruitment: ChIP assays show CDK11 depletion decreases Ser2P specifically toward the HIV 3′ end, while RNAPII levels are not strongly altered; CPA factors (CstF77, PAP) are reduced at/near the 3′ end; THOC1 depletion mirrors CDK11 effects for Ser2P and CPA factors.

6) Readthrough and poly(A) tail outcomes: CDK11 or THOC1 depletion increases uncleaved/readthrough species; CDK11 overexpression increases poly(A) tail length and increases transcript stability (actinomycin D). Together these support the CPA/termination model.

Skeptical critique (what could weaken the causal interpretation)

Association ≠ direct causation: The TREX/THOC–CDK11 binding is supported by AP-MS and co-IP, but the paper does not identify which specific THOC subunit bridges recruitment to HIV (they explicitly state they do not know the subunit). That leaves room for alternate recruitment routes or indirect effects.

CTD phosphorylation specificity: The authors show CDK11 can phosphorylate Ser2 and Ser5 in vitro, but ChIP suggests the dominant in-cell effect is Ser2P at the 3′ end. This mismatch could reflect assay sensitivity, epitope availability, indirect regulation, or timing/kinetics not captured by static ChIP.

System limitation (models & expression contexts): The study relies on transformed cell lines and transient reporter systems; they also note that endogenous CDK11 is low in resting CD4+ T cells, which prevents primary-cell repetition. Therefore, the findings may not fully translate to natural latency/reactivation dynamics.

Off-target effects & redundancy: siRNA knockdowns and overexpression can introduce off-targets or non-physiological stoichiometries; additionally, CTD kinases show redundancy/hierarchy (which the authors acknowledge broadly), so CDK11 effects could be context-dependent. The paper partially addresses CK2 independence for in vitro CTD phosphorylation but does not fully eliminate compensation by other CTD kinases in all contexts.

CPA machinery incompleteness: The paper measures key factors (CstF77, PAP) and uses poly(A) outcomes, but does not exhaustively examine all CPSF/CstF subunits. Thus, the mechanistic “recruitment program” is partially mapped rather than fully reconstructed.

Confidence-weighted bottom line

High confidence: CDK11 level causally affects HIV reporter activity, viral protein readouts, HIV 3′ end readthrough, poly(A) tail length, and transcript stability in the studied cell models.
Moderate-to-high confidence: Mechanistically, the strongest “middle layer” is the Ser2P–CPA-factor axis: CDK11 depletion reduces Ser2P at HIV 3′ ends and reduces CPA factors (CstF77, PAP) at/near the 3′ end.
Moderate uncertainty: Identification of the exact THOC subunit and the precise timing/kinetics of RNAPII mark changes at viral 3′ sites remain unresolved.
Moderate uncertainty: Translation to resting CD4+ T cell biology/latency reactivation is limited by low physiological CDK11 levels and by reliance on transformed cell systems.

Go deeper on BGPT (bespoke actions)

Author reviews (BGPT links)

Feedback:

Updated: April 28, 2026