Structure of the HapX:CCAAT-binding protein complex with DNA: a piece of the puzzle revealed

Summary

DNA recognition by the HapX transcription factor from Aspergillus species requires the presence of a heterotrimeric DNA-binding protein called the CCAAT-binding complex (CBC). In this issue of Structure, Huber et al (2022) illuminate the structural basis for the multivalent binding of the CBC, HapX and the DNA target site.

The CCAAT-binding complex (CBC) is found throughout the eukaryotic kingdom but requires a specialized partner protein in filamentous fungi to control iron homeostasis. In the Aspergillus spp, the CBC requires an additional basic region-leucine zipper (bZIP) transcription factor called HapX to properly control the response to both iron excess and limitation. In Aspergillus fumigatus, the CBC controls more than 2500 genes while HapX impacts only a fraction of these. Disruption strains lacking CBC components have a generally more severe phenotype than analogous strains without a functional hapX gene Hortschansky, Eisendle et al. (2007), consistent with the expanded range of target genes. Binding of the CBC to DNA has been analyzed in detail and requires an element closely related to CCAAT as its recognition site (Huber, Scharf et al. 2012). The requirements for binding of HapX to DNA has been more difficult to determine as no obvious consensus sequence emerged from early studies on HapX-regulated genes (Gsaller, Hortschansky et al. 2014).

Association of HapX with the CBC was shown to be mediated by interaction between the N-terminus of the HapE subunit of the CBC with an amino-terminal segment of the HapX protein (see (Hortschansky, Haas et al. 2017) for a review). Important insights into the molecular details of DNA-binding by the CBC:HapX complex came from the characterization of the binding of this complex to the cccA gene in A. fumigatus (Gsaller, Hortschansky et al. 2014). These experiments established a 1:2:1 stoichiometry for CBC:HapX:DNA complex formation. While this earlier work provided strong evidence for an association of both the CBC and HapX with different recognition elements embedded in promoters of regulated genes, DNA binding by HapX remained somewhat mysterious. Surface plasmon resonance measurements of cccA promoter binding by the CBC and HapX, either together or individually, clearly indicated that these factors bound DNA quite differently. The CBC was able to bind a probe from the cccA promoter with high affinity while HapX bound with relatively low affinity when each protein was present alone in the binding reaction. Addition of HapX to preformed CBC:DNA complexes led to a dramatic 87-fold increase in HapX affinity. A potential HapX recognition site was suggested but clearly this element was not sufficient to support high affinity binding of HapX unless associated with bound CBC.

This unsolved problem of determining a consensus sequence for HapX binding to DNA was addressed by using a genomic approach to evaluate all HapX recognition sites. Chromatin immunoprecipitation coupled with high throughput sequencing (ChIP-seq) was used to compare the binding of HapC and HapX to genes under control of these transcription factors across the entire A. fumigatus genome (Furukawa, Scheven et al. 2020). Combined with biochemical analysis of in vitro binding of recombinant forms of the CBC and HapX, along with cross-species comparisons of the relevant promoters, a new consensus sequence emerged that suggested a potential binding motif for the CBC and HapX. This motif contained the familiar CBC binding site (CSAAT) and two new elements involved in HapX DNA-binding. An A/T-rich trinucleotide with the consensus RWT (R=any purine and W=either A or T) located at a fixed 12 bp distance 3’ to the core CBC site, along with either single or overlapping T^T/_GAn bZIP half-sites positioned at variable distances relative to the CBC binding site. This latter site exhibited a high degree of orientation and spacing differences between various HapX-regulated promoters. This was suggested to underlie the wide ranges of responses seen for the different genes under control of HapX to iron-dependent signals.

These findings provided important clarification to the likely mechanism of DNA recognition by the CBC:HapX complex but several important questions remained. What parts of the CBC and/or HapX bind to these newly recognized features? How is HapX recruited to these novel CBC-binding sites?

The work of Huber et al (Huber, Hortschansky et al. 2022) provides important insights into precisely how the HapX:CBC:DNA complex is formed. In combination with the previously determined three-dimensional structure of the CBC bound to DNA, their new structures elaborate and clarify novel molecular details of the association of HapX with this complex. HapX binds to a region close to the CBC recognition element as a homodimer, as would be expected from a bZIP transcription factor. These data confirm the conclusions of a previous study that HapX requires two different sequence elements to bind to DNA (Furukawa, Scheven et al. 2020). These elements include the bZIP half-site and the RWT sequence located 12 bp downstream of the CCAAT CBC-binding motif.

The contribution of this RWT element is likely to be organization of the N-terminal CBC-interacting domain on one of the HapX monomers. Since both monomers are identical in sequence, the difference in positioning over the bZIP half site allows the amino-terminus of one monomer to interact with the RWT element and subsequently with the CBC. This organization of the N-terminus of HapX allows the CBC-interacting domain to make contacts with both HapC and HapE in the CBC. In contrast, the N-terminal region of the second HapX monomer is disordered in the crystal structure, which is consistent with the presence of only a single HapX-binding site in the CBC.

Taken together, these structural details help to explain the observed stoichiometry and the high affinity binding of HapX in the presence of the CBC as well as the remarkable cooperativity between these factors. DNA-binding of the CBC is unusual as it forms a histone-like fold and interacts via the minor groove but this complex is able to recognize its binding site without any requirement for another protein (Huber, Scharf et al. 2012). HapX is recruited into the CBC:DNA complex through a combination of different interactions: binding to the bZIP half-site, association with the RWT element via the HapX N-terminus and finally interaction between two CBC subunits and the CBC interaction domain of HapX. This model was supported by mutagenesis of the CBC interaction domain and a nearby residue involved in DNA binding. All of these mutant HapX proteins exhibited in vitro defects in association with CBC:DNA complexes and also failed to restore in vivo function when introduced into a hapXΔ strain. Interestingly, while responses to iron limitation and excess were defective, these mutant proteins still restored function above that of a strain lacking the hapX gene. These data are consistent with the central importance of this region of HapX in all its physiological roles. While the protein:protein interaction supported by the CBC interaction domain is important in HapX function, the remaining modes of interaction with the DNA region or the CBC can still provide some function. These multiple modes of interaction provide a robust framework to ensure that HapX function can be retained even when mutationally compromised.

While the data provided by Huber (Huber, Hortschansky et al. 2022) provide a detailed understanding of the formation of the HapX:CBC complex on DNA, more remains to be uncovered in terms of how this protein complex acts to regulate gene expression. While some repression mediated by HapX:CBC may involve steric effects (Gsaller, Hortschansky et al. 2016), the disparate architecture of iron-regulated promoters indicates this type of a repressive mechanism is unlikely to explain all HapX:CBC-modulated transcription. Since HapX:CBC can also activate the expression of genes involved in iron acquisition, it seems reasonable to imagine that this complex must engage components of the transcriptional machinery that can lead to gene activation. The high resolution picture of HapX:CBC:DNA formation will allow these questions to be addressed experimentally with a firm understanding of how this complex is formed on target genes.

Figure 1. Model for the association of HapX with the CBC and DNA.

A cartoon depicting the interactions of the HapX homodimer with its DNA target site and the CBC is shown. The CBC binds to DNA via the association of the HapB subunit with the minor groove and interaction of the other subunits with the DNA backbone. HapX binds to the bZIP half-site present in the major groove of DNA while also interacting via one of the two N-terminal regions with the minor groove-localized RWT sequence, in addition to protein:protein interactions with the CBC. See the text for a summary of this interaction and the Huber et al article (Huber, Hortschansky et al. 2022) in this issue of Structure for details.