An intrinsically disordered transcription activation domain increases the DNA binding affinity and reduces the specificity of NFκB p50/RelA

Abstract

Many transcription factors contain intrinsically disordered transcription activation domains (TADs), which mediate interactions with coactivators to activate transcription. Historically, DNA-binding domains and TADs have been considered as modular units, but recent studies have shown that TADs can influence DNA binding. Whether these results can be generalized to more TADs is not clear. Here, we biophysically characterized the NFκB p50/RelA heterodimer including the RelA TAD and investigated the TAD's influence on NFκB-DNA interactions. In solution, we show the RelA TAD is disordered but compact, with helical tendency in two regions that interact with coactivators. We determined that the presence of the TAD increased the stoichiometry of NFκB-DNA complexes containing promoter DNA sequences with tandem κB recognition motifs by promoting the binding of NFκB dimers in excess of the number of κB sites. In addition, we measured the binding affinity of p50/RelA for DNA containing tandem κB sites and single κB sites. While the presence of the TAD enhanced the binding affinity of p50/RelA for all κB sequences tested, it also increased the affinity for nonspecific DNA sequences by over 10-fold, leading to an overall decrease in specificity for κB DNA sequences. In contrast, previous studies have generally reported that TADs decrease DNA-binding affinity and increase sequence specificity. Our results reveal a novel function of the RelA TAD in promoting binding to nonconsensus DNA, which sheds light on previous observations of extensive nonconsensus DNA binding by NFκB in vivo in response to strong inflammatory signals.

Keywords: DNA-protein interaction; NF-kB transcription factor; cooperativity; hydrogen-deuterium exchange; intrinsically disordered protein; small-angle X-ray scattering; structural model; transcription.

Figure 1

Characterization of NFκB constructs used in this work.A, NFκB subunits p50 and RelA contain well-folded N-terminal domains (NTDs) and dimerization domains (DDs) that make up the Rel-homology domain (RHD). Additionally, RelA contains an intrinsically disordered transcription activation domain (TAD), which contains two regions important for protein–protein interactions, TA1 and TA2. B, the disorder propensity of full-length RelA (blue, solid) and the predicted AlphaFold2 pLDDT score (black, dashed) were predicted using MetaPredict. The TAD (residues 320–549) is largely predicted to be disordered except for two short stretches corresponding to the TA1 and TA2 motifs. Note that the graph residue numbers are aligned with the RelA_FL (residues 19–549) cartoon in panel (A). C, the three protein constructs used in this work, p50/RelA_FL (blue), p50/RelA_RHD (red), and RelA_TAD (green) were analyzed by analytical SEC. Solid lines represent 10 μM p50/RelA_FL and p50/RelA_RHD and 20 μM RelA_TAD. Dashed lines represent 5 μM p50/RelA_FL and p50/RelA_RHD and 10 μM RelA_TAD. D, SAXS Kratky plots p50/RelA_FL (blue), p50/RelA_RHD (red), and RelA_TAD (green). E, pairwise distance distribution for p50/RelA_FL (blue), p50/RelA_RHD (red), and RelA_TAD (green). The R_g and D_max values of each construct based on SAXS data analysis are in Table 1. F, a model of RelA_TAD consistent with the SAXS data was generated using the AWSEM and BilboMD. G, a model of p50/RelA_RHD consistent with the SAXS data was generated using BilboMD. H and I, two of the models of p50/RelA_FL that were generated using FoXSDock and BilboMD represent two possible conformations of this protein that agree well with the SAXS data. SAXS, small-angle X-ray scattering; SEC, size-exclusion chromatography.

Figure 2

Hydrogen-deuterium exchange analysis of NFκB constructs alone and bound to DNA.A, the constructs p50/RelA_FL and p50/RelA_RHD alone and bound to a DNA hairpin containing the HIV-LTR κB sequence were analyzed using hydrogen-deuterium exchange mass spectrometry (HDX-MS). Representative deuterium uptake plots are shown mapped to the structural model of p50/RelA_FL. Peptides in the DNA-binding cavity showed reduced deuterium incorporation in the presence of DNA (green and yellow peptides highlighted above, representing the DBD and NTD respectively), whereas peptides outside the cavity did not show significant changes in deuterium incorporation in the presence of DNA (gray and orange peptides highlighted above, representing the DBD and NTD respectively). In general, there were no significant differences in deuterium incorporation when comparing p50/RelA_FLversus p50/RelA_RHD alone or bound to DNA. B, the deuterium uptake of peptides in the RelA TAD in the presence (light blue) and absence (blue) of DNA. Horizontal bars represent the span of residues in each peptide, and the fraction uptake represents the number of deuterons incorporated after 10 s relative to the number of exchangeable hydrogen atoms in the peptide. Vertical error bars represent the SD of percent uptake based on three technical replicates. NTD, N-terminal domain; TAD, transcription activation domain.

Figure 3

EMSA analysis of binding stoichiometries of p50/RelA_FLand p50/RelA_RHDto tandem DNA sequences.A, a 33 bp segment of the HIV LTR promoter sequence was used in these experiments. B, binding of p50/RelA_RHD (0, 62.5, 125, 250, 500, 1000, and 2000 nM) to the HIV LTR promoter DNA (250 nM) was detected using EMSA. Bands are visible corresponding to free DNA and DNA bound by 1, 2, or 3 p50/RelA_RHD dimers. C, binding of varying concentrations of p50/RelA_FL to the HIV LTR promoter DNA (250 nM) was detected using EMSA. D and E, the intensities of the bands in the EMSA gels detecting binding of p50/RelA_RHD (D) and p50/RelA_FL (E) to the HIV LTR DNA were quantified using ImageJ and plotted as a function of NFκB concentration. Data points represent the mean and SD of two biological replicates. F, a 59 bp segment of the NFKBIA promoter sequence was used in these experiments. G, binding of p50/RelA_RHD (0, 62.5, 125, 250, 500, 1000, and 2000 nM) to the NFKBIA promoter sequence (250 nM) was analyzed using EMSA. The smear present near the top of the gel at higher p50/RelA_RHD concentrations likely represents DNA bound by four or more p50/RelA_RHD dimers. H, binding of p50/RelA_FL to the NFKBIA promoter sequence was analyzed using EMSA. I and J, the intensities of the bands in the EMSA gels detecting binding of p50/RelA_RHD (I) and p50/RelA_FL (J) to the NFKBIA DNA were quantified using ImageJ and plotted as a function of NFκB concentration. Data points represent the mean and SD of two independent biological replicates.

Figure 4

EMSA analysis of binding stoichiometries of p50/RelA_FLand p50/RelA_RHDto mutated DNA sequences.A and B, binding of p50/RelA_RHD and p50/RelA_FL to the HIV LTR sequence in which the first κB site is scrambled was detected using EMSA. Varying concentrations of p50/RelA_RHD and p50/RelA_FL were mixed with 250 nM HIV-LTR DNA in which the first κB site is scrambled and analyzed using EMSAs. The fraction of DNA bound by 0, 1, 2, or 3 p50/RelA dimers was quantified using ImageJ and plotted as a function of p50/RelA concentration. C and D, when the second κB site in the HIV-LTR promoter DNA segment is scrambled, 21 bp of nonspecific DNA are present following the first κB site. Varying concentrations of p50/RelA_RHD and p50/RelA_FL were mixed with 250 nM HIV-LTR DNA in which the second κB site is scrambled and analyzed using EMSAs. Compared to WT HIV-LTR DNA and HIV-LTR DNA in which the first site is scrambled, both p50/RelA_RHD and p50/RelA_FL formed more complexes in which 3 p50/RelA dimers were bound to the DNA. E and F, similar experiments were conducted using a DNA segment consisting of the NFKBIA promoter sequence with the first κB site scrambled. G and H, binding of p50/RelA_RHD and p50/RelA_FL to the NFKBIA sequence with the second κB site scrambled was also monitored using EMSAs. Overall, scrambling either site in the NFKBIA promoter led to an increase in higher order complex formation at high concentrations of p50/RelA_RHD and p50/RelA_FL. All data points represent the mean and SD of two independent experiments.

Figure 5

Competition between specific and nonspecific DNA sequences for p50/RelA_FLbinding. 250 nM double-stranded HIV-LTR DNA was incubated with 500 nM p50/RelA_FL, and 250 nM hairpin DNA containing either the HIV-LTR κB sequence or a scrambled sequence was added to the sample. The κB hairpin was able to efficiently compete with the HIV LTR dsDNA for p50/RelA_FL binding (comparing lanes 5 & 8), whereas the scrambled DNA hairpin was unable to compete with the HIV LTR dsDNA for p50/RelA_FL binding (comparing lanes 5 & 9).

Figure 6

Binding affinities of p50/RelA_FLand p50/RelA_RHDfor hairpin DNA.A–G, DNA hairpins were labeled on the 5′ ends with fluorescein, and 5 nM DNA was incubated with varying concentrations of p50/RelA_FL (blue) and p50/RelA_RHD (red) to determine binding affinity via fluorescence anisotropy. Note that the same data are graphed twice in panel (G) to enable clear visualization of both binding curves. Data points represent the mean and SEM of three independent experiments, and the reported K_d values are the mean and SEM of the K_d values derived from each of the experiments individually. Lines represent the expected binding curve based on the mean K_d values.

Figure 7

Binding affinities of p50/RelA_FLand p50/RelA_RHDfor κB sites within tandem DNA sequences.A, dsDNA containing the sequence of the HIV LTR promoter was labeled with fluorescein on the 5′ end of the forward strand to detect binding to the first κB site or the 5′ end of the reverse strand to detect binding to the second κB site. B and C, binding of p50/RelA_RHD to the HIV LTR DNA labeled on the 5′ end of forward (B) or reverse (C) strand was detected using fluorescence anisotropy and fit well to a model in which it binds each site independently with a K_d of 1.7 ± 0.4 nM. D and E, binding of p50/RelA_FL to the HIV LTR DNA labeled on the 5′ end of forward (D) or reverse (E) strand was detected using fluorescence anisotropy and fit well to a model in which it binds each site independently with a K_d of 1.3 ± 0.4 nM. F, dsDNA containing the sequence of the NFKBIA promoter was labeled with fluorescein on the 5′ end of the forward or reverse strand to detect binding to the first or second κB site, respectively. G and H, binding of p50/RelA_RHD to the NFKBIA DNA labeled on the 5′ end of forward (G) or reverse (H) strand was detected using fluorescence anisotropy and fit to a model in which it binds each κB site independently, with a K_d of 1.7 ± 0.5 nM for site 1 and 10 ± 2 nM for site 2. I and J, binding of p50/RelA_FL to the NFKBIA DNA labeled on the 5′ end of forward (I) or reverse (J) strand was detected using fluorescence anisotropy and fit to a model in which it binds each κB site independently, with a K_d of 1.5 ± 0.7 nM for site 1 and 5.1 ± 0.4 nM for site 2. Data points shown in this figure are the mean and SD of three independent experiments. K_d values are the mean and SEM of the best-fit values determined for each of the three experiments.