The fungal CCAAT-binding complex and HapX display highly variable but evolutionary conserved synergetic promoter-specific DNA recognition

Abstract

To sustain iron homeostasis, microorganisms have evolved fine-tuned mechanisms for uptake, storage and detoxification of the essential metal iron. In the human pathogen Aspergillus fumigatus, the fungal-specific bZIP-type transcription factor HapX coordinates adaption to both iron starvation and iron excess and is thereby crucial for virulence. Previous studies indicated that a HapX homodimer interacts with the CCAAT-binding complex (CBC) to cooperatively bind bipartite DNA motifs; however, the mode of HapX-DNA recognition had not been resolved. Here, combination of in vivo (genetics and ChIP-seq), in vitro (surface plasmon resonance) and phylogenetic analyses identified an astonishing plasticity of CBC:HapX:DNA interaction. DNA motifs recognized by the CBC:HapX protein complex comprise a bipartite DNA binding site 5'-CSAATN12RWT-3' and an additional 5'-TKAN-3' motif positioned 11-23 bp downstream of the CCAAT motif, i.e. occasionally overlapping the 3'-end of the bipartite binding site. Phylogenetic comparison taking advantage of 20 resolved Aspergillus species genomes revealed that DNA recognition by the CBC:HapX complex shows promoter-specific cross-species conservation rather than regulon-specific conservation. Moreover, we show that CBC:HapX interaction is absolutely required for all known functions of HapX. The plasticity of the CBC:HapX:DNA interaction permits fine tuning of CBC:HapX binding specificities that could support adaptation of pathogens to their host niches.

Figure 1.

(A) Schematic illustration of the A. fumigatus HapX domain organization. The N-terminal Hap4-like CBC binding domain (CBC-BD) is shown in blue. Basic region and coiled coil parts of the bZIP domain are depicted in red and green, respectively. Cysteine-rich regions are marked in yellow. Cysteine-rich regions A and B are essential for HapX function during iron excess but not during iron starvation, while the C-terminus is crucial during iron starvation (shown in turquois). Residues N65, Q69, F72 and R73, which mediate specific DNA contacts in the Schizosaccharomyces pombe bZIP Pap1:DNA complex (17), are labeled with a black triangle. The location of a putative monopartite nuclear localization sequence (NLS) motif identified within the Aspergillus oryzae HapX basic region (18) is shown in grey.

Figure 2.

Identification of common direct targets of HapX and the CBC by genome-wide ChIP-seq analysis and de novo discovery of a CBC:HapX motif. Venn diagrams showing comparison of ChIP-seq target sites of HapX and the CBC were depicted using a modified version of Cistrome software (47). For comparison of common ChIP-seq target sites, 150 bp sequences centered at the peak summit regions were used. The total number of ChIP-seq target sites is reported in parentheses. Numbers in white indicate how many ChIP sequence targets overlap in the respective comparison. (A) Comparison of the ^VenusHapX ChIP-seq target sites (>3-fold enrichment, found within the 1.5-kb of the 5′-upstream region) during iron sufficiency (+Fe) versus iron starvation (−Fe) conditions. (B) Comparison of the HapC^GFP and the ^VenusHapX ChIP-seq target sites (>2-fold enrichment for the CBC, >3-fold enrichment for HapX, found within 1.5 kb of the 5′-upstream region) during +Fe conditions. (C) Comparison of the HapC^GFP and the ^VenusHapX ChIP-seq target sites during −Fe conditions. CBC:HapX consensus DNA recognition motifs identified by MEME from the ^VenusHapX ChIP-seq peak summit regions for +Fe (D) and −Fe (E) conditions.

Figure 3.

Binding of HapX and the CBC on the promoter region of the selected 33 HapX-regulated genes. In vivo binding of ^VenusHapX and HapC^GFP on the 5′-upstream region of the 33 HapX-regulated genes are shown. In cases where no A. fumigatus gene designation was assigned, we used the gene name of the respective S. cerevisiae ortholog. Tracks for the ChIP-seq (ChIP), their input DNA control (Input), and the binding specificity controls for HapX (NLS-Venus ChIP) are visualized in the UCSC genome browser together with annotated gene models and their transcript, which are expressed in RPMI-1640 culture conditions. RNA-seq data were taken from (49). Direction of the target gene and the proximal genes are shown in red arrows and blue arrows, respectively. The positions of the nucleotide motif matched with the consensus binding sequence for the CBC identified from our ChIP-seq analysis (5′-CCAATVR-3′) are also shown.

Figure 4.

(A) MEME analysis of 36 A. fumigatus CBC:HapX sites identified an A/T-rich trinucleotide (5′-RWT-3′) located 12 bp downstream of the CBC binding site. (B) The sequence logo of Yap1-like (TTAN) and Gcn4-like half-sites (TGAN) found in the 3′-submotifs of 32 CBC:HapX sites was generated by WebLogo (54). The distance between the CCAAT box and HapX 5′-T(T/G)AN-3′ half-site consensus motifs shows a remarkable variation (C), whereas their DNA strand orientation tends to be distance specific with preferred positions at 14(+) and 20(−) nt downstream of the CBC consensus motif (D).

Figure 5.

Comparative SPR analysis of CBC-binding to DNA and HapX-binding to preformed CBC:DNA complexes on 36 individual A. fumigatus HapX regulon promoter sites. (A) Sites are ranked from the top to the bottom according to their apparent HapX binding affinity to CBC-bound DNA (red bars); the CBC affinity to the sites is shown as black bars. Black and white dots indicate the position of the 5′-thymine base of Gcn4-like (5′-TGAN-3′) and Yap1-like (5′-TTAN-3′) half-sites relative to the CCAAT consensus motif as well as their DNA strand orientation (black lines). Blue and light blue colored boxes indicate two 7 bp spanning DNA regions which contain 5′-T(T/G)AN-3′ half-sites at their 5′- or 3′-border. (B) The in vitro HapX DNA-binding affinity correlates with the in vivo^VenusHapX ChIP-seq peak fold enrichment on 33 A. fumigatus CBC:HapX promoter sites.

Figure 6.

HapX-binding depends in vitro on both the Gcn4-like half sites and A/T-rich submotifs within the bipartite CBC:HapX consensus sites as shown by SPR analysis. SPR analyses included binding of the CBC to DNA (panel 1), HapX to DNA (panel 2) and HapX to preformed CBC:DNA complexes (panel 3). The SPR sensorgrams are shown from sensor-immobilized duplexes covering evolutionary conserved bipartite CBC:HapX binding sites from A. fumigatus leuA and sreA promoters (A and D) as well as duplexes carrying mutations within their 3′ Gcn4-like half-site (B and E) or A/T-rich submotif (C and F). Nt underlined in black are covered by the CBC and nt marked in blue represent the A. fumigatus conserved submotifs. Gcn4-like half-sites are underlined in blue. Substituted nt relative to the wild type sequence are written in red and shown in lower case. Binding responses of the indicated CBC or HapX concentrations injected in duplicate (black lines) are shown overlaid with the best fit derived from a 1:1 interaction model including a mass transport term (red lines). Binding responses of CBC:DNA:HapX ternary complex formation (panel 3, blue lines) were obtained by concentration dependent co-injection of HapX on preformed binary CBC:DNA complexes after 200 seconds within the steady-state phase. Sensorgrams in panel 4 depict the association/dissociation responses of HapX on preformed CBC:DNA and were generated by CBC response subtraction (co-injection of buffer) from HapX co-injection responses.

Figure 7.

In vitro and in vivo analysis of the regulatory function of HapX cccA promoter binding motifs. (A–D) SPR co-injection analysis of HapX binding to CBC bound A. fumigatus cccA promoter DNA duplexes carrying mutations within their conserved 5′-RWT-3′ and 5′-TKAN-3′ submotifs. Data are presented as described in figure legend 6. (E) Graphical representation of native or mutated A. fumigatus cccA promoter E. coli lacZ fusions integrated in single copy at the A. fumigatus pksP gene locus. Numbers indicate the positions of the three evolutionary conserved CBC:HapX binding sites (F) Effect of site 1 (-369) mutations on PcccA-lacZ expression after 18 h growth under iron depleted (−Fe) conditions followed by a 3 h shift of identical cultures to iron sufficiency (0.03 mM Fe). Iron dependent PcccA-lacZ expression was determined as β-galactosidase specific activity from soluble cell extracts. The host strains, AfS77 (wt) and ΔhapX, were used as negative controls. Data represent the mean ± SD of three independent biological replicates.

Figure 8.

Loss of cooperative CBC:HapX binding is elicited by HapX DNA-BD mutations, deletion or modifications of the CBC-BD. SPR sensorgrams of HapX (A, B), HapX/CgYap5 hybrids (C, D) or Yap1 (E, F) binding to DNA (panel 1), and bZIP peptides to preformed CBC:DNA complexes (panel 2) are shown. Aa residues that differ from the native HapX sequence are marked in gray. Nt underlined in black are covered by the CBC and nt marked in blue represent the A. fumigatus HapX DNA-binding site in the cyp51A promoter. The overlapping Yap1-like half-sites in the cyp51A promoter are underlined in blue. Binding responses of the indicated bZIP concentrations injected in duplicate (black lines) are shown overlaid with the best fit derived from a 1:1 interaction model including a mass transport term (red lines). Binding responses of CBC:DNA:bZIP ternary complex formation (panel 2, blue lines) were obtained by concentration dependent co-injection of bZIPs on preformed binary CBC:DNA complexes. Sensorgrams in panel 3 depict the association/dissociation responses of bZIPs on preformed CBC:DNA and were generated by CBC response subtraction (co-injection of buffer) from bZIP co-injection responses. (G) Amino acid sequence comparison of the HapX CBC-BD and DNA-BD spanning region with A. fumigatus bZIPs carrying the basic region signature sequence of the fungal Yap1/Pap1 subfamily. Sequences are aligned according to similarity of the basic regions. The HapX CBC-BD and a rudimentary CBC-BD in the N-terminal region of Yap1 are underlined.

Figure 9.

HapX DNA-binding and CBC:HapX interaction is essential for its function during iron starvation and overload. (A) Growth pattern of A. fumigatus wild type (wt, AfS77), ^venushapX, ΔhapX and ^venushapX strains expressing hapX alleles with either a deletion of the HapX CBC-BD (^venushapX ΔCBC-BD) or aa substitutions within the HapX DNA-BD (^venushapX DNA-BDm) on solid minimal medium at 37°C for 48 h. (B) Production of biomass during iron starvation (−Fe), iron sufficiency (0.03 mM FeSO₄, +Fe) and iron excess (5 mM FeSO₄, hFe) after liquid growth at 37°C for 24 h. Biomass production of the mutant strains was significantly different compared to wt and ^venushapX during −Fe and hFe, but not +Fe (P <0.001). (C) Western blot analysis of wild type and mutant HapX protein levels from strains grown for 24 h under iron deficiency. Actin was used as loading control. (D) Northern blot analyses were performed after liquid growth under iron starvation (−Fe) and high-iron availability (5 mM FeSO₄, hFe) at 37°C for 20 h or from mycelia shifted for 30 min from −Fe to iron sufficiency (0.03 mM FeSO₄, sFe). (E) Nuclear localization of Venus-tagged wild type HapX and HapX mutant proteins in iron limitation conditions.

Figure 10.

Production of TAFC is impaired by HapX-deficiency as well as deletion of the HapX CBC-BD or mutations within the HapX DNA-BD in vivo. (A) TAFC production of the mutant strains was significantly reduced compared to wt and ^venushapX under iron starvation (P <0.001). (B) HapX-CBC as well as HapX-DNA interaction is crucial for activation of genes that promote TAFC biosynthesis in A. fumigatus.

Figure 11.

Domain architecture of fungal transcription factors that require the CBC as DNA-binding scaffold for their function. (A) The Hap4 activators of respiratory gene expression in S. cerevisiae (Sc) and C. glabrata (Cg) contain the complete 16 aa Hap4-like CBC-BD, but lack a basic region DNA-BD (25,63). The same applies to Php4 in S. pombe (Sp), which represses iron-dependent pathways in response to iron deficiency (64). A. fumigatus (Afu) HapX and its ortholog Hap43 in C. albicans (Ca) are bZIP transcription factors that include the full 16 aa Hap4-like CBC-BD. Both HapX and Hap43 have dual functions in regulation of iron homeostasis, thereby acting as both repressor and activator depending on iron availability (3,10,20). Yap5 bZIPs contain a degenerated Hap4-like domain and activate the high iron stress response in C. glabrata and S. cerevisiae (21,25). Full and rudimentary Hap4-like domains required for CBC-binding are shown in blue. Basic regions, coiled coil domains and Cysteine-rich regions are depicted in red, green and yellow, respectively. (B) Amino acid sequence alignment of the Hap4-like CBC-BD and basic region DNA-BD spanning regions in the N-Termini of regulatory CBC subunits shown in (A). (C) Amino acid sequence alignment of the Hap4 recruitment domain of fungal Hap5 orthologs. Secondary structures (α: helix) are indicated for the HapE subunit from A. nidulans (30). Identical residues are marked in yellow, residues conserved in 50% of the sequences are shaded in light blue and blocks of similar residues are marked in green. Alignments were performed with AlignX (Vector NTI Advance 11).