Plasma cell - the end of the B lineage, and the only cell type in which ELL2 has a documented job
also called plasma cell · antibody-secreting cell · ASC · plasmablast (immediate precursor)
A plasma cell is a B cell that has finished growing up and turned into a factory that pumps out antibody. It is here because ELL2, one of the elongation factors this graph tracks inside the KMT2A fusion complex, has its entire known day job in this one cell type and nowhere else.
The same thing, for a scientist
The plasma cell (CL:0000786) is the terminal, non-dividing, antibody-secreting stage of the B lineage, reached via the plasmablast through a Blimp-1/XBP1-driven secretory reprogramming. It enters this graph for one reason. The super-elongation-complex subunit **ELL2** is induced on plasma-cell differentiation and is the factor that switches immunoglobulin heavy-chain pre-mRNA from the membrane form to the **secreted** form, by loading with CstF-64 on RNA polymerase II across the Igh locus, driving use of the weak proximal poly(A) site and skipping of the non-consensus splice site in the final secretory-specific exon [E1]. Its paralogue **ELL3** occupies the earlier B-cell compartment, expressed in B cells before ELL2 and driving proliferation and survival there [E2]. The relevance to KMT2A::AFF1 is a negative one and worth stating plainly: the index disease is arrested at the **pro-B** stage, at the opposite end of B-cell development, and no experiment places ELL2's plasma-cell function anywhere near a KMT2A-rearranged blast.
Why a plasma cell is in a leukaemia graph at all
Two reasons, and the second is the honest one.
First, it is the cell type that defines ELL2. ELL2 appears in this graph as a component of the super elongation complex that KMT2A::AFF1 hijacks. But if you ask the literature what ELL2 does in a cell, the answer is almost entirely about plasma cells: ELL2 is induced when a B cell becomes a plasma cell, and it redirects immunoglobulin heavy-chain mRNA processing toward the secreted form 1. Lowering ELL2 lowers secretory-form IgH mRNA. That is a cell-type-specific function for a supposedly general elongation factor, and it is exactly what the PI's cell-type dimension exists to capture.
Second, the graph had no node for it. Of the 49 cell-type and cell-state entries that existed before this one, not one was a terminally differentiated B cell, so a well-evidenced cell-type-specific function of a core interactome member had nowhere to attach.
What is actually shown, and in what system
In B-cell and plasma-cell systems, ELL2 and the polyadenylation factor CstF-64 track together with RNA polymerase II across the Igh mu and gamma segments; ELL2-specific siRNA blocks the association of both, and lowering ELL2 lowers secretory-form heavy-chain mRNA 1. This is an expressed-in plus required-in result for the plasma cell: the protein is induced there, and removing it changes the cell's defining output.
It is not a maintains-a-state result. Nothing in that work shows that a plasma cell stops being a plasma cell when ELL2 goes; it shows the secretory output collapses. The three evidence classes are kept apart throughout this tranche for exactly this reason - expression is not dependency is not identity.
The paralogue split matters too. ELL3 is selectively expressed in B cells prior to ELL2 and drives proliferation and survival at that earlier stage 2. The B lineage therefore runs two different ELL-family factors at two different developmental positions, and a reagent or an inference that does not separate them is reading two cells at once. A third family member, ELL (ELL1), is itself a KMT2A fusion partner.
The counted absence against the index disease
Measured 2026-09-21 on PubMed via E-utilities.
| query | records |
|---|---|
ELL2[tiab] AND (KMT2A[tiab] OR MLL[tiab] OR "mixed-lineage leukemia"[tiab]) | 6 |
the same plus ("plasma cell"[tiab] OR "cell type"[tiab] OR differentiation[tiab]) | 0 |
operand KMT2A[tiab] OR MLL[tiab] OR "mixed-lineage leukemia"[tiab] | 8,160 |
The operands are large and the identical query shape returns non-zero counts for other genes in this tranche, so the zero is a real empty intersection and not a non-composing query.
Zero in titles and abstracts; supporting figures not searched. A cell-type assignment for ELL2 in a KMT2A-rearranged blast could sit in a supplementary panel of any SEC paper and this query would not see it.
| ontology | CL:0000786 |
|---|---|
| position in b lineage | terminal; downstream of the germinal-centre B cell and the plasmablast, at the opposite end of B-cell development from the pro-B arrest of KMT2A::AFF1 |
| why in this graph | the only cell type in which ELL2, a super-elongation-complex subunit this graph carries, has a documented cell-type-specific function |
| ell2 function here | induced on plasma-cell differentiation; drives the membrane-to-secreted switch of immunoglobulin heavy-chain mRNA by proximal poly(A) site use and terminal-exon skipping, loading with CstF-64 on Pol II across Igh |
| evidence class for ell2 | expressed_in AND required_in (for the secretory output). NOT maintains_a_state: no experiment shows a plasma cell losing plasma-cell identity on ELL2 loss. |
| shown for fusion | NO. Measured 2026-09-21: ELL2[tiab] AND (KMT2A[tiab] OR MLL[tiab] OR "mixed-lineage leukemia"[tiab]) returns 6 records; the same query with a cell-type or differentiation term returns 0. Operand KMT2A/MLL = 8,160. Zero in titles and abstracts; supporting figures not searched. |
| paralogue note | ELL3 occupies the earlier B-cell compartment, ELL2 the plasma cell, and ELL (ELL1) is itself a KMT2A fusion partner. Three family members sitting in three different places in this graph. |
markers
[ "CD138 (SDC1)", "CD38 high", "CD19 low or negative", "surface immunoglobulin low", "PRDM1/Blimp-1", "XBP1s" ]
- ESSENTIAL-FUNCTION TOXICITY — Anything that lowers ELL2 function systemically is predicted to impair antibody secretion, because the secreted-form switch is ELL2's documented job in the plasma cell. In an infant already immunosuppressed by chemotherapy and often by blinatumomab-driven B-cell aplasia, a further hit to antibody output is not a hypothetical toxicity. No ELL2-directed agent exists, so this is a design constraint rather than an observed harm.
- COMPENSATION — ELL1 and ELL3 are present in the same lineage and the family is partly interchangeable inside the SEC, so ELL2 loss may be buffered. A single-paralogue perturbation producing no phenotype must not be read as ELL2 being unimportant.
- KMT2A::AFF1 arrests B-cell development at the pro-B stage, upstream of everything the plasma cell does. Is ELL2's presence in the SEC of a pro-B blast purely the general elongation function, or is any part of the plasma-cell mRNA-processing programme aberrantly running there? A 3' end-sequencing comparison of KMT2A-rearranged blasts against pro-B controls would settle it cheaply.
- Does the ELL3-to-ELL2 handover happen at a defined developmental checkpoint, and is a KMT2A-rearranged blast fixed on one side of it? If the blast is ELL3-high and ELL2-low, the ELL2 in its SEC is a minority species and this graph's ELL2 claims are weighted wrongly.
- Is the multiple-myeloma risk locus at 5q15 acting through the plasma cell's own ELL2 dosage, and if so what does it say about how much ELL2 reduction other B-lineage cells tolerate?
- E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper
In B-lineage cells, the RNA polymerase II elongation factor ELL2 is induced in plasma cells and directs immunoglobulin secretion by enhancing use of the weak promoter-proximal poly(A) site and skipping of the non-consensus splice site in the final secretory-specific exon of the heavy-chain pre-mRNA; ELL2 and CstF-64 track together with Pol II across the Igh mu and gamma segments, and lowering ELL2 lowers secretory-form heavy-chain mRNA.
Transcription elongation factor ELL2 directs immunoglobulin secretion in plasma cells by stimulating altered RNA processing. · Martincic K, Alkan SA, Cheatle A, Borghesi L, Milcarek C · Nat Immunol 2009What this evidence cannot show
B-cell and plasma-cell systems including reporter constructs, hnRNP F transfection and siRNA knockdown; not a KMT2A-rearranged system and not a leukaemia experiment of any kind. Read: abstract only. The result is expression plus requirement for an output, not a demonstration that ELL2 maintains plasma-cell identity.
Evidence that argues against this (1)
- The B lineage runs two ELL-family factors at two different stages, so 'ELL2 is the B-lineage ELL' would be wrong: ELL3 is selectively expressed in B cells BEFORE ELL2 and drives proliferation and survival there. Any inference drawn from an ELL-family reagent that does not separate the paralogues is reading two cell states at once. Alexander LMM et al. 2017
- E2moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper
ELL3 is selectively expressed in B cells prior to ELL2 and drives proliferation and survival at that earlier B-cell stage.
Selective expression of the transcription elongation factor ELL3 in B cells prior to ELL2 drives proliferation and survival. · Alexander LMM, Watters J, Reusch JA, Maurin M, Nepon-Sixt BS, Vrzalikova K, Alexandrow MG, Murray PG, Wright KL · Mol Immunol 2017What this evidence cannot show
B-cell lines and primary B cells; no KMT2A-rearranged system. Read: abstract only. Establishes a stage-wise paralogue split, not a mechanism, and does not measure whether the two factors are functionally interchangeable.
Evidence that argues against this (1)
- A stage-wise expression split does not establish non-redundancy. The definitive ELL2 result in the plasma cell came from knockdown in cells that also express other ELL-family members, so part of the ELL2 phenotype could be buffered or shared; no double-depletion experiment resolving ELL2 against ELL3 in one cell was found. Martincic K et al. 2009
- ← expressed inELL2 — a SEC subunit that almost no cell needs, next to a paralogue almost every cell needs
ELL2 is induced on plasma-cell differentiation and directs the membrane-to-secreted switch of immunoglobulin heavy-chain mRNA; expressed_in plus required_in for that output, NOT identity maintenance, and no KMT2A-rearranged arm