Figure 1.
Establishing the LMO4-binding region of DEAF1.
A. Schematic domain structure of the mouse DEAF1 protein containing a DNA-binding SAND (Sp100, AIRE-1, NucP41/75, DEAF11), a Helix-Loop-Helix (HLH) domain, a predicted coiled coil region (depicted as a helix), a protein-binding MYND (myeloid translocation protein 8, Nervy, DEAF1) domain, a nuclear localization signal (NLS) and a nuclear export signal (NES). The previously identified LMO4-binding region (335–545) [4] is indicated (thin grey line). B. Yeast two-hybrid experiments where Saccharomyces cerevisiae (AH109) were co-transformed with full-length DEAF1 fused to a transcriptional activator domain (pGAD10) and LMO4 fused to a DNA binding domain (pGBT9). Co-transformants were serially diluted and spotted on growth (−L/−W; growth) and high stringency interaction plates (−L/−W/−H/−A; selection). Left-most panels show controls. Schematic on right shows corresponding domain truncations of DEAF1 constructs used in the assays. Growth of yeast or its absence on selection plates indicates an interaction (ticks) or an abrogation of the interaction (crosses) with LMO4 respectively. C. DEAF1 internal deletion mutants were tested for interaction with LMO4 by yeast two-hybrid assays; interactions are represented as above. D. Yeast two-hybrid data for DEAF1/LMO4 interactions to assay replacement of the DEAF1 coiled coil domain by the dimeric GCN4 leucine zipper. Selection was medium/high stringency (−L/−W/−H+3AT)/(−L/−W/−H/−A). Yeast two-hybrid spot test results are shown. Three dilutions (A600nm = 0.2, diluted serially 2×1-in-10) are spotted left to right to show differences in growth under each selection condition.
Figure 2.
Characterising DEAF1 and the coiled coil domain.
A. Schematic showing DEAF1 constructs used in yeast two-hybrid self-association experiments. Selection was medium/high stringency as in Figure 1D; +++ indicates strong growth, - indicates no growth, ND indicates not determined. B. SEC-MALLS analysis of full length and DEAF1335–485 constructs (left panel) and DEAF1404–479 and the coiled coil domain (right panel). DEAF1 proteins (∼200 µg) were applied to a Superose 12 column with an in line MALLS detector to determine weight-averaged molecular weight in solution. The elution (continuous line) and light-scattering (▪) are shown. C. Summary of the theoretical monomeric and experimentally determined molecular weight of DEAF1 proteins in A and B were used to calculate the oligomeric state. D. Far-UV circular dichroism spectropolarimetry (CD) spectrum of the DEAF1 coiled coil domain.
Figure 3.
Nuclear localisation of EGFP-NLS4-DEAF1404–479 in the presence of LMO4.
A. DEAF1 construct in pEGFP-C2 that was used for transfection. It has an N-terminal EGFP tag followed by the altered DEAF1 NLS4 and DEAF1404–479. The NLS4 and NES protein sequences and DEAF1 numbering are shown. B. HEK293 cells grown on cover slips in 6 well plates were transfected with a total of 4 µg of DNA: control pEGFP (panel 1), 2 µg EGFP-DEAF1404–479+2 µg empty pCMV vector (panel 2), 2 µg EGFP-NLS4-DEAF1404–479+2 µg empty pCMV (panel 3) and EGFP-NLS4-DEAF1404–479+2 µg pCMV LMO4 (panel 4). After 24 h transfection, cells were fixed with paraformaldehyde and nuclei stained with Hoechst dye. Cells were imaged for EGFP fluorescence (green) and nuclear staining (blue) by fluorescence microscopy. C. Quantification of A. The two-dimensional areas of n = 8 fields of view were measured for % nuclear localisation of EGFP-NLS4-DEAF1404–479 in the presence and absence of LMO4. Difference is statistically significant as p<0.05.