SEC-L2 — the AFF2-scaffolded super-elongation-like complex: P-TEFb plus ENL and AF9, and NO ELL
also called SEC-like 2 · SEC-L2 · AFF2 complex · FMR2 complex
SEC-L2 is one of the cell's several near-identical transcription-speeding machines. It is built on AFF2 instead of the AFF1 that the infant-leukaemia fusion uses, and it carries the same two mark-reading proteins, ENL and AF9 — but it is missing the ELL speed-up subunits the main machine has.
The same thing, for a scientist
SEC-L2 is the AFF2-scaffolded member of the super elongation complex family, isolated by Flag affinity purification of AFF2 from tetracycline-inducible 293 Flp-In-TRex cells and characterised by MudPIT, endogenous co-immunoprecipitation and Superose 6 size-exclusion chromatography [E1]. Reproducibly recovered components are CDK9, cyclin T1/T2 (P-TEFb) and the two YEATS-domain readers ENL/MLLT1 and AF9/MLLT3; a low level of AFF4 was also detected in the AFF2 purification. **ELL1, ELL2 and ELL3 were NOT reproducibly detected** in the Flag-AFF2 preparation by MudPIT, and endogenous co-IP placed ELL2 and ELL3 only in AFF1-, AFF4- and CDK9-precipitated material, never in AFF2 [E2]. The complex elutes in Superose 6 fractions 11-14, overlapping the AFF4/SEC profile, which the authors attribute to AFF2 homodimerisation rather than to an ELL-bearing composition. Functionally SEC-L2 has Pol II CTD kinase activity comparable to SEC and SEC-L3 in vitro, is NOT required for HSP70 induction on heat shock, and its RNAi signature (226 differentially expressed genes) is largely non-overlapping with those of AFF3 and AFF4 [E3]. This entry exists because AFF2 is the AFF-family paralogue that is NOT in the index fusion and is therefore the natural comparator for any argument about what the AFF1 moiety contributes.
Composition as purified, with the negative stated
| subunit | in SEC-L2? | how established |
|---|---|---|
| AFF2 (P51816) | scaffold | Flag bait |
| CDK9 / cyclin T1-T2 (P-TEFb) | yes | MudPIT, endogenous co-IP, size-exclusion WB 1 |
| ENL / MLLT1 | yes | MudPIT, endogenous co-IP, size-exclusion WB 1 |
| AF9 / MLLT3 | yes | MudPIT, size-exclusion WB 1 |
| ELL1, ELL2, ELL3 | NOT DETECTED | absent from MudPIT; ELL2/ELL3 absent from endogenous anti-AFF2 IP while present in anti-AFF1, anti-AFF4 and anti-CDK9 2 |
| EAF1 / EAF2 | not tested | EAF subunits enter SEC through ELL; the figure legend states their presence is expected from sequence conservation but not experimentally confirmed 2 |
| BRD4, HEXIM1 | absent | recovered in the CDK9 purification but not in the AFF purifications, consistent with distinct P-TEFb pools 1 |
The ELL negative is the compositional fact that most changes the picture and it is the one the rest of this corpus does not carry. SEC-L2 is not 'SEC with a different AFF'. It is a P-TEFb plus YEATS-reader particle without the ELL elongation-stimulatory module.
What SEC-L2 does, and what it does not do
- CTD kinase: Flag-AFF2 material phosphorylates GST-Pol II CTD with activity similar to Flag-AFF3 and Flag-AFF4 preparations at matched CDK9 input 1.
- HSP70: no significant AFF2 ChIP enrichment over the HSP70 gene before or after heat shock, and AFF2 siRNA does not compromise heat-shock induction of HSP70 by RT-qPCR or Northern blot — only AFF4 depletion does 3.
- Its own gene set: 226 genes differentially expressed on AFF2 shRNA in HEK293T (151 down), of which 55.3% are unique to AFF2 depletion 3. Full cross-tabulation in
ana-q187-sec-family-target-set-overlap-measured.
The authors themselves flag the limiting caveat: AFF2 and AFF3 mRNA are about 10-fold lower than AFF4 in HEK293T, so a failure to detect ELL, or a failure to ChIP AFF2 at HSP70, is confounded with abundance 2.
Why this matters to a YEATS inhibitor
ENL and AF9 are in SEC-L2 and they are in SEC. A small molecule that occupies the ENL/AF9 YEATS acyl-lysine pocket does not distinguish which scaffold the reader is sitting on, because the YEATS domain is not the scaffold-binding surface — the AHD is (mechanism-kmt2a-mllt1-engages-ptefb-through-the-enl-ahd-494-559). The consequence is developed in mechanism-q187-target-specificity-is-a-scaffold-property-while-the-reader-is-shared and hyp-q187-yeats-inhibition-is-scaffold-blind.
| complex name | SEC-L2 (super elongation complex-like 2) |
|---|---|
| scaffold | AFF2 |
| method class | Flag affinity purification with benzonase plus MudPIT; endogenous co-IP; Superose 6 size exclusion |
| system | HEK293 Flp-In-TRex / HEK293T; not haematopoietic, not leukaemic, not a fusion context |
| epigenetic readout | carries two YEATS-domain acyl-lysine readers (ENL, AF9) as shared subunits; AFF2 itself writes, reads and erases nothing |
| shown for fusion | NO — no KMT2A fusion experiment anywhere in the defining paper; AFF2 is a paralogue of the fusion's 3' moiety and is not a component of it |
subunits
{
"AFF2": {
"accession": "P51816",
"length_aa": 1311,
"role": "scaffold; Flag bait in the defining purification"
},
"CDK9": {
"accession": "P50750",
"role": "P-TEFb kinase"
},
"CCNT1": {
"accession": "O60563",
"role": "P-TEFb cyclin"
},
"MLLT1": {
"accession": "Q03111",
"role": "YEATS-domain acyl-lysine reader; shared with SEC and SEC-L3"
},
"MLLT3": {
"accession": "P42568",
"role": "YEATS-domain acyl-lysine reader; shared with SEC and SEC-L3"
}
}subunits looked for and not found
[ "ELL", "ELL2", "ELL3" ]
subunits not tested
[ "EAF1", "EAF2" ]
marks named
[ "H3K27ac", "H3K9ac" ]
measurements
[
{
"quantity": "differentially_expressed_genes",
"subject": "AFF2 shRNA knockdown, HEK293T (SEC-L2 proxy)",
"value": 226,
"units": "genes",
"n": 2,
"universe": "HEK293T, RefSeq via TopHat/Cufflinks, FDR-adjusted p<0.05, 2 shRNA replicates vs 4 non-targeting replicates",
"null": null,
"note": "151 of the 226 are downregulated; 55.3% are unique to AFF2 knockdown"
},
{
"quantity": "unique_target_fraction",
"subject": "AFF2 shRNA DEGs not shared with AFF3 or AFF4 DEGs",
"value": 0.553,
"units": "fraction",
"n": 226,
"universe": "same HEK293T mRNA-Seq experiment; 'unique' means not differentially expressed in either of the other two knockdowns",
"null": null,
"note": "stated as 55.3% in Results; 33.6% of AFF2 DEGs are shared with AFF3, leaving ~11% shared with AFF4 only"
}
]uniprot members
[ "P51816", "P50750", "O60563", "Q03111", "P42568" ]
- If SEC-L2 really lacks ELL then a therapeutic aimed at the SEC ELL slot would leave SEC-L2 intact while a YEATS-pocket inhibitor would not. The two intervention points have opposite selectivity profiles across the family and nobody has stated that trade-off.
- AFF2 loss in humans causes FRAXE intellectual disability. A drug that functionally disables the SEC-L2 particle by occupying its reader subunit is, on paper, partially phenocopying a neurodevelopmental syndrome in an infant. There is no measurement anywhere of a YEATS inhibitor's effect on any AFF2-dependent gene.
- Is AFF2 protein present at all in a fetal-liver or pro-B progenitor, the cell of origin for KMT2A::AFF1? If not, SEC-L2 is irrelevant to this disease and the AFF family has three members here, not four. gene-aff2 already asks this and it is still open.
- Is the ELL negative real or an abundance artefact? The decisive experiment is an endogenous anti-AFF2 IP from a cell type where AFF2 is abundant — brain, or a FRAXE-relevant neuronal model — blotted for ELL2 and ELL3.
- What fraction of the cellular ENL and AF9 pool sits in SEC-L2? Nothing in the defining purification is quantitative in that direction, and the answer decides whether SEC-L2 is a meaningful sink for the readers or a trace species.
- E1strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
In HEK293 Flp-In-TRex cells, Flag affinity purification of AFF2 recovered CDK9, cyclin T1/2 and the YEATS readers ENL/MLLT1 and AF9/MLLT3 by MudPIT, confirmed by endogenous co-immunoprecipitation and by Superose 6 size-exclusion chromatography in which the Flag-AFF2 complex elutes in fractions 11-14 together with ENL, AF9, CDK9 and cyclin T1; the complex was named SEC-like 2 (SEC-L2).
The super elongation complex family of RNA polymerase II elongation factors: gene target specificity and transcriptional output. · Luo Z, Lin C, Guest E, Garrett AS, Mohaghegh N, Swanson S, Marshall S, Florens L, Washburn MP, Shilatifard A · Molecular and Cellular Biology 2012What this evidence cannot show
HEK293/293T only, with the bait over-expressed from a tetracycline-inducible transgene; no haematopoietic cell, no leukaemia line, no fusion protein. dNSAF spectral values are given only in a figure panel and are not recoverable from the text, so the STOICHIOMETRY of ENL/AF9 in SEC-L2 is not established by this record. Benzonase was present during purification, so the recovery is not DNA-bridged; but co-elution across fractions 11-14 is consistent with several particle compositions.
Evidence that argues against this (1)
- No second laboratory has reproduced the ENL/AF9 membership of SEC-L2. Lu and colleagues characterise the SEC family by tandem affinity purification and argue explicitly that it is combinatorial, but their antibody panel is AFF1, AFF4, ELL1, ELL2, ENL, AF9, CycT1 and CDK9 — AFF2 and AFF3 are not in it, so their work neither supports nor contradicts SEC-L2's composition. A PubMed search for the complex name run 2026-09-24 returns two records in total and both are from the originating group. Lu H et al. 2015
- E2moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper
ELL1, ELL2 and ELL3 could not be reproducibly detected by MudPIT in the purified AFF2 complex, and in endogenous co-immunoprecipitation ELL2 and ELL3 were detected only in AFF1-, AFF4- and CDK9-immunoprecipitated samples and not in the AFF2 sample; the authors state that the presence of ELL- and EAF-related proteins in SEC-L2 has not been experimentally confirmed, and offer the roughly 10-fold lower AFF2/AFF3 expression in HEK293T as one possible reason for the failure to detect.
The super elongation complex family of RNA polymerase II elongation factors: gene target specificity and transcriptional output. · Luo Z, Lin C, Guest E, Garrett AS, Mohaghegh N, Swanson S, Marshall S, Florens L, Washburn MP, Shilatifard A · Molecular and Cellular Biology 2012What this evidence cannot show
This is a NEGATIVE from one cell type at one abundance. The authors explicitly decline to rule out that AFF2 interacts with ELLs under different conditions or in different cell types, and they name the abundance confound themselves. A negative from MudPIT on a roughly 10-fold under-expressed bait is weak evidence of absence. It is recorded because every existing entry in this corpus describing SEC-L2 lists what was found and not what was looked for and missed.
Evidence that argues against this (2)
- The same paper's Figure 1F schematic hedges in the opposite direction from its own blots: it draws SEC-L2 and SEC-L3 and states that the presence of ELL- and EAF-related proteins 'has not been experimentally confirmed but is expected based on sequence conservation among AFF family members'. So the paper simultaneously reports the negative and predicts it will not hold, and a reader taking the schematic rather than the blot would record ELL as present. Luo Z et al. 2012
- An ELL that is absent from SEC-L2 is not an ELL that is idle: ELL1 is also a subunit of the little elongation complex with an AFF-independent snRNA-gene function, which is this corpus's preferred explanation for ELL's pan-essentiality (protein-little-elongation-complex). So 'no ELL in SEC-L2' constrains SEC-L2 and says nothing about how much ELL is doing elsewhere. Smith ER et al. 2011
- E3strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
AFF2 is not recruited to the HSP70 gene on heat shock by ChIP-qPCR and AFF2 siRNA does not compromise heat-shock induction of HSP70 while AFF4 depletion does; genome-wide, AFF2 shRNA in HEK293T changes 226 genes (151 downregulated), of which 55.3% are unique to AFF2 depletion.
The super elongation complex family of RNA polymerase II elongation factors: gene target specificity and transcriptional output. · Luo Z, Lin C, Guest E, Garrett AS, Mohaghegh N, Swanson S, Marshall S, Florens L, Washburn MP, Shilatifard A · Molecular and Cellular Biology 2012What this evidence cannot show
HEK293T, FDR<0.05, two shRNA replicates against four non-targeting replicates, 80-90% knockdown. A knockdown-response set is not an occupancy set: 226 genes is what changes when AFF2 goes, not what SEC-L2 binds. The negative HSP70 ChIP carries the same abundance confound as E2 and the authors say so.
Evidence that argues against this (1)
- The AFF2 DEG set is the smallest of the three (226 against AFF3's 423 and AFF4's 1,116) and AFF2 is the lowest-expressed scaffold in the cell type used, so the apparent restriction of SEC-L2's target set is confounded with detection power. A set that is small because the scaffold is scarce is not the same as a set that is small because the complex is selective, and nothing in the paper separates them. Luo Z et al. 2012
AFF2 is the scaffold subunit of SEC-L2
ENL is a shared subunit of SEC-L2 (MudPIT, endogenous co-IP, size exclusion)
AF9 is a shared subunit of SEC-L2
P-TEFb (CDK9/cyclin T1-T2) is a SEC-L2 subunit
- related to →ELL2 — a SEC subunit that almost no cell needs, next to a paralogue almost every cell needs
ELL2 was looked for in SEC-L2 by MudPIT and endogenous co-IP and NOT found
the ELL module absent from SEC-L2 has its own AFF-independent complex
- related to →Step 3 — The hijack: the fusion drags the super elongation complex to paused RNA polymerase II
SEC-L2 is a paralogous particle of the complex the fusion hijacks
paralogous SEC-family particles; 76 shared knockdown-responsive genes, Jaccard 0.133
full-text composition detail, plus the ELL negative the abstract does not carry
- related to →SEC — the AFF1/AFF4-scaffolded super elongation complex, the particle KMT2A::AFF1 is a permanent subunit of
SEC-L2 is the AFF2-scaffolded paralogue of SEC; shares the readers and P-TEFb, lacks detectable ELL
AFF2 arm: 226 DEGs, 125 unique, 39 shared with both siblings
SEC-L2 carries the same YEATS readers and cannot be distinguished by a pocket ligand
no YEATS antagonist has ever been tested against SEC-L2 in any system
AFF2 is FMR2; the FRAXE arm of the same liability
the entry that names this hypothesis in its prose