Multivalency regulates activity in an intrinsically disordered transcription factor

Abstract

The transcription factor ASCIZ (ATMIN, ZNF822) has an unusually high number of recognition motifs for the product of its main target gene, the hub protein LC8 (DYNLL1). Using a combination of biophysical methods, structural analysis by NMR and electron microscopy, and cellular transcription assays, we developed a model that proposes a concerted role of intrinsic disorder and multiple LC8 binding events in regulating LC8 transcription. We demonstrate that the long intrinsically disordered C-terminal domain of ASCIZ binds LC8 to form a dynamic ensemble of complexes with a gradient of transcriptional activity that is inversely proportional to LC8 occupancy. The preference for low occupancy complexes at saturating LC8 concentrations with both human and Drosophila ASCIZ indicates that negative cooperativity is an important feature of ASCIZ-LC8 interactions. The prevalence of intrinsic disorder and multivalency among transcription factors suggests that formation of heterogeneous, dynamic complexes is a widespread mechanism for tuning transcriptional regulation.

Interactions by Disorder – A Matter of Context

“…The flanking regions may furthermore contribute mechanistically to disorder-based interactions. Cooperativity driven by flanking regions is elegantly demonstrated by the IDRs constituting the activation domain of the zinc finger TF ataxia telangiectasia mutated substrate Chk2-interacting Zn 2+ -finger protein (ASCIZ) (Clark et al, 2018). ASCIZ uses 11 out of 17 highly conserved TQT SLiMs to bind the dimeric hub protein LC8 (Ranaldi et al, 1994;Rapali et al, 2011), whereas Chia, another LC8 binding IDP uses three out of four TQT SLiMs to bind, whose affinities depend on the flanking regions in a non-predictable manner (Clark et al, 2016).…”

“…ASCIZ binding to LC8 generates a scaffold (Clark et al, 2015) onto which additional LC8s bind with increased affinity ( Figure 1H and Table 1). Then, negative cooperativity regulates the formation of higher-order LC8 assemblies to ensure that low-occupancy complexes dominate at saturating concentrations of LC8 to prevent switching off transcription completely (Clark et al, 2018). Thus, for the disordered ASCIZ and Chia, both flanks and context affect binding affinity and in vivo regulation of activity.…”

Discriminating between negative cooperativity and ligand binding to independent sites using pre-equilibrium properties of binding curves

“…On a different context, negative cooperativity has shown to play an important role in tuning transcriptional regulation. The prevalence of intrinsic disorder and multivalency among transcription factors suggests that formation of heterogeneous, dynamic complexes is a widespread mechanism, and negative cooperativity is an important feature in this scenario [13]. Finally, a direct correlation between negative cooperativity and receptor oligomerization was reported for a particular subfamily of GPCR [14].…”

Continuum dynamics and statistical correction of compositional heterogeneity in multivalent IDP oligomers resolved by single-particle EM

“…For use in method development and validation studies, the training dataset was curated by the microscopist, who is familiar with the LC8-IDP structure (see Figure 1A) and the NSEM dataset, 15 to manually classify a representative set of LC8 oligomers as 2-mers, 3-mers, 4-mers, etc. To minimize ambiguity, the microscopist selected complexes that were well separated from neighboring particles on the micrograph (see Figure 1B-D).…”

“…The combination of conformational and compositional heterogeneity of multivalent LC8-IDP duplexes -i.e., a continuum of shape fluctuations and differences in the number of LC8s per complex -rules out common EM analysis software, which predominantly rely on class averages that suppress conformational fluctuations by construction, and clustering/classification methods that presume the existence of discrete states. [12][13][14] Traditional 2D classification methods were shown explicitly to fail in our analysis of the 11-site LC8-ASCIZ duplex system, due to extreme conformational heterogeneity, 15 thus requiring painstaking manual curation of the EM image dataset. The commonly used single particle EM image processing software, RELION, has a 'multibody' scheme, 16 but it requires establishing orientations for the individual 'bodies' which is not possible for the 20 kD LC8 dimers, which are far below the detection limit of current Cryo-EM methods and just at the limit of resolvability by negative stain EM.…”