Overlapping expression of early B-cell factor and basic helix-loop-helix proteins as a mechanism to dictate B-lineage-specific activity of the lambda5 promoter

Abstract

The basic helix-loop-helix (bHLH) transcription factors are a large group of proteins suggested to control key events in the development of B lymphocytes as well as of other cellular lineages. To examine how bHLH proteins activate a B-lineage-specific promoter, I investigated the ability of E47, E12, Heb, E2-2, and MyoD to activate the lambda5 surrogate light chain promoter. Comparison of the functional capacity of the E2A-encoded E47 and E12 proteins indicated that even though both were able to activate the lambda5 promoter and act in synergy with early B-cell factor (EBF), E47 displayed a higher functional activity than E12. An ability to act in synergy with EBF was also observed for Heb, E2-2, and MyoD, suggesting that these factors were functionally redundant in this regard. Mapping of functional domains in EBF and E47 revealed that the dimerization and DNA binding domains mediated the synergistic activity. Electrophoretic mobility shift assay analysis using the 5' part of the lambda5 promoter revealed formation of template-dependent heteromeric complexes between EBF and E47, suggesting that the synergistic mechanism involves cooperative binding to DNA. These findings propose a unique molecular function for E47 and provide overlapping expression with EBF as a molecular mechanism to direct B-cell-specific target gene activation by bHLH proteins.

FIG. 1

E2A proteins are partially redundant in their ability to activate the λ5 promoter in synergy with EBF. The left part of panel A shows schematic drawings indicating the structure of the E2A proteins and the E2A forced dimers. The right part shows an autoradiogram from an SDS-PAGE gel with the products obtained after in vitro translation of the E2A proteins and the forced dimers in the presence of [³⁵S]methionine. The left part of panel B shows a Western blot obtained with a 9E10 anti-myc tag antibody and 10 μg of nuclear extracts from HeLa cells transfected with tagged E47 or E12 proteins as indicated. The second gel shows an EMSA analysis with the μE5 E-box from the IgH intron enhancer and 5 μg of the same nuclear extracts as in the Western blot. The right part shows a diagram indicating the luciferase activity relative to that of 200 ng of λ5 promoter-controlled reporter plasmid in the presence of 200 ng of empty expression plasmid, after Lipofectin-mediated transfection of 50 ng of tagged E47 or E12 with 150 ng of empty (−) or EBF-encoding (+) cDNA3 into HeLa cells. The induction obtained with 300 ng of E12 was related to the activity of the reporter transfected with 450 ng of empty cDNA3. The data shown are based on three representative transfection experiments. Error bars indicate standard deviations. The left part of panel C shows the resulting Western blot when nuclear extracts from transiently transfected HeLa cells were probed for the presence of E2A forced dimers by an anti-E2A antibody. The second radiogram shows an EMSA using a labeled μE5 E-box and the same nuclear extracts as in the Western blot. The right part of the panel shows diagrams indicating the luciferase activity obtained after transient transfections of 200 ng of the λ5 reporter plasmid with 150 ng of either empty or E47-E47- or E47-E12-encoding plasmids and 150 ng of empty (−) or EBF-encoding (+) expression plasmid in HeLa cells. The reporter activity obtained with 300 ng of empty expression plasmid was set as 1, and inductions were calculated from three representative transfection experiments. Error bars indicate standard deviations.

FIG. 2

MyoD and class I bHLH proteins have the ability to activate the λ5 promoter together with EBF. Panel A shows a diagram of the relative luciferase activity after transfection of 200 ng of λ5 promoter-controlled luciferase reporter plasmid with 300 ng of expression plasmid encoding either Heb, E2-2, or MyoD and 150 ng of either empty (−) or EBF-encoding (+) expression plasmids into HeLa cells. The fold induction was based on the activity obtained when the reporter plasmid was transfected together with 450 ng of empty expression plasmid or 900 ng of empty plasmid when 900 ng of MyoD plasmid was used. The data shown are based on three transfection experiments. Error bars indicate standard deviations. Panel B shows the relative induction of 200 ng of luciferase reporter gene under the control of a basal promoter and two copies of a μE5-μE2 combination from the IgH intron enhancer after transfection of 600 ng of empty, E47-expressing (58), or MyoD-encoding expression plasmid into HeLa cells. The induction was based on data from three representative transfections, and the error bars indicate standard deviations. (C) Relative inductions of 200 ng of reporter constructs controlled either by a basal fos promoter or two copies of the E3-E2 combination (58) from the 5′ part of the λ5 promoter cloned 5′ of the basal fos promoter after transfection with E47 or MyoD expression plasmids in HeLa cells as indicated. The data are collected from three representative transfections, and the error bars indicate the standard deviations.

FIG. 3

The transactivation domains of E47 are essential for full function but not for cooperation with EBF. Panel A shows a schematic drawing of the full-length and truncated E47 proteins that were fused to an amino-terminal 9E10 myc tag. Functional domains in E47 are indicated by black boxes and represent the transactivation domains AD1 and AD2 and the bHLH domain as indicated. The numbers indicate the amino acid positions for the truncations. The figure is not drawn to scale. (B) The left panel shows Western blot analysis with 9E10 anti-myc antibody and 10-μg nuclear extracts from HeLa cells transiently transfected with 800 ng of the indicated E47 protein. The right panel shows an autoradiogram from EMSA using 5-μg nuclear extracts from transiently transfected HeLa cells and an end-labeled oligonucleotide encompassing the μE5 E-box as indicated. (C) Diagram indicating the relative luciferase activity obtained after transient transfections of 200 ng of the λ5 reporter plasmid with 300 ng of E47-encoding plasmids as indicated. The reporter activity obtained with 300 ng of empty expression plasmid was set as one, and the data were calculated from three representative transfection experiments. Error bars indicate standard deviations. (D) Diagram representing the relative luciferase activity obtained when 200 ng of the λ5 reporter plasmid was transfected with 50 ng of E47-encoding plasmids in combination with 150 ng of empty (−) or 9E10-tagged EBF-encoding (+) expression plasmid in HeLa cells. The black bars indicate that these data were collected from an independent transfection experiment. The reporter activity obtained with 200 ng of empty expression plasmid was set as 1, and data are collected from three representative transfections. Error bars indicate standard deviations.

FIG. 4

The carboxy-terminal transactivation domain of EBF is important for full functional activity, but DNA binding and dimerization domains are sufficient to mediate synergy with E47. Panel A shows a schematic drawing of the full-length and truncated EBF proteins that were fused to an amino-terminal 9E10 myc tag. Functional domains in EBF are indicated by black boxes and represent the DNA binding, dimerization, and transactivation domains TSI and TSII as indicated. The numbers indicate the amino acid positions for the introduced truncations. The figure is not drawn to scale. (B) The left panel shows Western blot analysis with 9E10 anti-myc antibody and 10-μg nuclear extracts from HeLa cells transiently transfected with 800 ng of the indicated EBF protein. The right panel shows an autoradiogram from EMSA using 5-μg nuclear extracts from transiently transfected HeLa cells and an end-labeled oligonucleotide encompassing the mb-1 promoter EBF binding site as indicated. (C) Diagram indicating the relative luciferase activity obtained after transient transfections of 200 ng of the λ5 reporter plasmid with 800 ng of EBF encoding expression plasmids as indicated. The reporter activity obtained with 800 ng of empty expression plasmid was set as 1, and data were calculated from three representative transfections. Error bars indicate standard deviations. The radiogram presents a Western blot using 9E10 antibody and the same protein extract as used for the luciferase assays. (D) Diagram representing the relative luciferase activity obtained when 200 ng of the λ5 reporter plasmid was transfected with 150 ng of EBF-encoding plasmids in combination with 50 ng of empty (−) or 9E10-tagged E47-encoding (+) expression plasmid in HeLa cells. The black bars indicate that these data were collected from an independent transfection experiment. The reporter activity obtained with 200 ng of empty expression plasmid was set as 1, and data are collected from three representative transfections. Error bars indicate standard deviations.

FIG. 5

Maximal synergy between EBF and E47 is dependent on multiple binding sites within the λ5 promoter. The top part of panel A shows a schematic drawing of the λ5 promoter. Relevant restriction sites as well as binding sites for EBF (gray boxes) and E47 (black boxes) are indicated. The lower part shows the resulting luciferase activity when 200 ng of the indicated λ5 reporter constructs were transiently cotransfected with the indicated expression plasmids (150 ng of EBF and/or 300 ng of E47) into HeLa cells. ND, not done. The reporter activity obtained with 450 ng of empty expression plasmid was set as 1, and data are collected from three representative transfections. Error bars indicate standard deviations.

FIG. 6

EBF and E47 form a stabilized ternary complex on the λ5 promoter. (A) DNA sequence of the 5′ region of the λ5 promoter with indicated EBF binding sites and E-boxes. Asterisks indicate base pair matches to the consensus binding sites. (B) Autoradiograms from EMSA with an end-labeled λ5 5′ NcoI-EcoRV fragment and increasing amounts of in vitro-translated EBF or/and E47 proteins as indicated. The middle panel shows an EMSA with the same probe and increasing amounts of in-vitro-translated EBFΔ1 and/or E47. The amount of protein in each reaction was normalized by the addition of nonprogrammed reticulocyte lysate. The right panel displays an EMSA with the λ5 promoter probe and 7.5 μg of nuclear extracts from transfected HeLa cells in combination with 1.5 μl of in vitro-translated EBF. The gels displayed are representative of two experiments. F, free DNA. (C) A diagram compiling the data obtained by densitometric analysis of the obtained EMSA complexes. The bars represent the total relative amount of radioactivity in complex with protein and are divided into sections according to the presence of the protein in homo- (white) or heteromeric (black and grey) complexes as indicated. (D) Autoradiogram of an EMSA where the preformed complexes between E47 and/or EBF on the λ5 promoter have been distorted by the addition of increasing amounts of unlabeled oligonucleotides spanning the E2 and E3 boxes in the λ5 promoter as indicated. The gel displayed is representative of three experiments. The probe is not shown, since the gel has been cut to save space.

FIG. 7

The formation of ternary complexes and functional synergy between EBF and E47 demands functional binding sites for both factors. (A) Autoradiograms from EMSAs with end-labeled wild-type or mutated 5′ NcoI-EcoRV fragments and 4 μl of in vitro-translated recombinant EBF and/or E47 as indicated. The gels displayed each represent one out of three experiments. F, free DNA. (B) The resulting luciferase activity when 200 ng of the indicated λ5 promoter reporter constructs was transiently cotransfected with 150 ng of EBF- and 50 ng of E47-encoding expression plasmids into HeLa cells. The reporter activity obtained with the wild-type promoter was set as 1, and data were collected from three representative transfections. Error bars indicate standard deviations.