BTG2 bridges PABPC1 RNA-binding domains and CAF1 deadenylase to control cell proliferation

While BTG2 plays an important role in cellular differentiation and cancer, its precise molecular function remains unclear. BTG2 interacts with CAF1 deadenylase through its APRO domain, a defining feature of BTG/Tob factors. Our previous experiments revealed that expression of BTG2 promoted mRNA poly(A) tail shortening through an undefined mechanism. Here we report that the APRO domain of BTG2 interacts directly with the first RRM domain of the poly(A)-binding protein PABPC1. Moreover, PABPC1 RRM and BTG2 APRO domains are sufficient to stimulate CAF1 deadenylase activity in vitro in the absence of other CCR4–NOT complex subunits. Our results unravel thus the mechanism by which BTG2 stimulates mRNA deadenylation, demonstrating its direct role in poly(A) tail length control. Importantly, we also show that the interaction of BTG2 with the first RRM domain of PABPC1 is required for BTG2 to control cell proliferation.


CAF1 (CNOT7 paralog) is less than 4 residues long.
A 5'-fluorescein-labeled synthetic RNA of 17 residues (100nM) was incubated 5 min at 50°C with decreasing concentrations of Nuclease P1 as indicated. In parallel, a 5'-fluoresceinlabeled synthetic RNA of 20 A residues (100nM) was incubated with 6His-CNOT7 (1.8µM) during the indicated times. Both reactions were electrophoresed on a 15% denaturing polyacrylamide gel. An asterisk "*" indicates a distortion of migration due to the presence of HEPES in CAF1 deadenylase buffer. Note that on a 15% polyacrylamide gel, the end products of the reactions migrate as a single visible band whereas two bands appear with time on 8% polyacrylamide gels (see Fig. 3). This is due to the hydrophobic nature of the fluorescein tag that affects migration of oligonucleotides shorter than 4 residues. An asterisk "*" indicates a distortion of migration due to the presence of HEPES in CAF1 deadenylase buffer. Deadenylation rate is 18.9 nucleotides per minute in these conditions. The region shown starts at the initiator methionine and is truncated after the Apro domain.

Supplementary
Coloring emphasizes amino acid properties conservation. The location of BoxC 1 is indicated.
Alpha helices and beta-strands determined in the structure of Tob1 (PDB:2DR5) are depicted below the sequences. The following protein sequences were used: BTG1: human Darkness and/or contrast were modified to reveal blot dimensions.

Transcriptional pulse-chase experiments
HEK293 Tet-Off cells were transfected with 0.8 µg of the pTet-β-globin plasmid (pBS2800) and 1.6 µg of the BTG2 expressing plasmids in 6-cm diameter culture dishes. Immediately after transfection, doxycyclin (1ng/ml) was added in the medium of the cells to block transcription of the reporter. Two days after transfection, cells were washed and a 3-hour transcriptional pulse was performed before re-addition of doxycyclin (2 µg/ml). Chase times correspond to hours after doxycyclin addition and to times of RNA extraction.

Northern blot analysis
10 µg of total RNA was electrophoresed onto 1.4% agarose/6% formaldehyde gels and transferred to Hybond-N+ membranes (GE Healthcare). For poly(A)-controls, 10 µg of total RNA sample was digested with RNase H (Invitrogen) and oligo(dT) as recommended by the supplier. After phenol extraction and ethanol precipitation, 1:5 of the digestion was loaded on agarose gels. After transfer, blots were stained with methylene blue to check for equal loading and hybridized to probes synthesized by in vitro transcription with the T7 RNA polymerase (Promega). Hybridization signals were visualized with Typhoon 8600 (GE Healthcare) and quantified as followed: 15 identical adjacent rectangles encompassing the mRNA migration region from fully adenylated to deadenylated mRNAs were drawn to quantify the signal intensities that were plotted as a function of the distance of migration (indicative of poly(A) tail length).

Analytical ultracentrifugation
Analytical ultracentrifugation sedimentation velocity experiments were done at 4°C in a Beckman-Coulter ProteomeLab XL-I analytical ultracentrifuge at 50,000 rpm in a AN-50 Ti rotor with absorbance and interference detections. Sedimentation data were collected in 7 min intervals for formation of complexes between BTG2(APRO) and 6His-PABPC1(1-190) and in 5 min intervals for formation of complexes between [BTG2(APRO);6His-CNOT7] and 6His-PABPC1(1-190). The fitting of data was performed using SEDFIT software, version 14.1 and continuous sedimentation coefficient distribution model. The distributions obtained for each protein sample were integrated to determine the weight-average sedimentation coefficients as a function of protein concentrations and to generate Sw isotherms. The Sw isotherms were loaded into SEDPHAT for fitting with the hetero-association model A+B <-> AB to obtain an estimate of the Kd. Buffer density, buffer viscosity and protein partial specific volumes were calculated using SEDNTERP software. The software GUSSI was used to plot and integrate the sedimentation coefficient distributions.

Electrophoretic Mobility Shift Assays
1 picomole of the fluorescently labeled poly(A) substrate was incubated 10min at 30°C with increasing concentrations of proteins in the same buffer as for in vitro deadenylation assays in a final volume of 10µl. 2.5µl of loading buffer (50% glycerol, 0.05% bromophenol blue, TBE 1X) was added and the samples were electrophoresed at 4°C on native 6% polyacrylamide gels (ProtoGel, national diagnostics) that were visualized with Typhoon 8600 (GE Healthcare).