Biochemical Journal

Research article

Solution structure of the Taf14 YEATS domain and its roles in cell growth of Saccharomyces cerevisiae

Wen Zhang , Jiahai Zhang , Xuecheng Zhang , Chao Xu , Xiaoming Tu

Abstract

Chromatin modifications play important roles in cellular biological process. A novel conserved domain family, YEATS, has been discovered in a variety of eukaryotic species ranging from yeasts to humans. Taf14, which is involved in a few protein complexes of chromatin remodelling and gene transcription, and is essential for keeping chromosome stability, regular cell growth and transcriptional regulation, contains a YEATS domain at its N-terminus. In the present study, we determined the solution structure of the Taf14 YEATS domain using NMR spectroscopy. The Taf14 YEATS domain adopts a global fold of an elongated β-sandwich, similar to the Yaf9 YEATS domain. However, the Taf14 YEATS domain differs significantly from the Yaf9 YEATS domain in some aspects, which might indicate different structural classes of the YEATS domain family. Functional studies indicate that the YEATS domain is critical for the function of Taf14 in inhibiting cell growth under stress conditions. In addition, our results show that the C-terminus of Taf14 is responsible for its interaction with Sth1, which is an essential component of the RSC complex. Taken together, this implies that Taf14 is involved in transcriptional activation of Saccharomyces cerevisiae and the YEATS domain of Taf14 might play a negative role in cell growth.

  • chromatin remodelling
  • NMR
  • Saccharomyces cerevisiae
  • Taf14
  • transcriptional regulation
  • YEATS domain

INTRODUCTION

Chromatin modification is a general biological process in many eukaryotes. The covalent modification of core histones, as a primary style of chromatin modification, regulates different states of gene expression. The covalent modifications of histone tails include acetylation, methylation, phosphorylation, SUMOylation, ubiquitination and so on. These histone modifications are carried out by a variety of protein complexes, for example NuA4, RSC and INO80. In these multi-subunit protein complexes, an increasing number of conserved domains have been identified to be crucial for establishing the different chromatin modifications.

The YEATS domain is distributed widely in 59 different eukaryotes [1] and was discovered recently as one of the domains related to chromatin modification and transcription [2]. However, the structures and functions of YEATS domains are far from clear. Interestingly, unlike other domains involved in chromatin modification complexes, the YEATS domain is present in a single copy in YEATS-domain-containing proteins, always located at the N-terminus. In addition, most YEATS-domain-containing proteins do not have any other annotated domain [2]. This implies that the YEATS domain might play critical roles in the functions of YEATS-domain-containing proteins.

YEATS-domain-containing proteins are usually involved in chromatin remodelling complexes. Among these proteins, Taf14 is the only one which is involved in both chromatin remodelling and transcription regulation. Taf14 was first identified to be involved in actin cytoskeletal function in Saccharomyces cerevisiae and therefore was named ANC1 [3]. Although Taf14 is a non-essential gene, cells devoid of Taf14 were thermo- and osmo-sensitive and had defects in actin organization [4]. Thereafter, Taf14 was reported to be important for chromosome stability [5] and was associated with several chromatin remodelling and transcription complexes as follows: Swi/Snf (switch/sucrose non-fermentable), which has DNA-stimulated ATPase activity and activates transcription by helping transcription factors access their binding sites [68]; RSC, which is essential for cell cycle progression [9,10]; INO80, which has helicase activity and is involved in transcription, replication and DNA repair [1113]; and NuA3, which stimulates transcription or replication elongation through nucleosomes by providing a coupled acetyltransferase activity [14].

Besides roles in chromatin remodelling, Taf14 is involved in complexes related to transcription. Taf14 is a subunit of the transcription factors TFIID [15] and TFIIF [6,16] in S. cerevisiae. A two-hybrid screen revealed that Taf14 interacts with Tsm1 in TFIID complexes and Tfg1 in TFIIF complexes [17]. Moreover, Taf14 is associated with transcription initiation. There is evidence that Taf14 is recruited to the promoter region in the ADH1 gene [18] and the GAL1 gene, suggesting its role in the assembly of the RNA polymerase II preinitiation complex [17]. Taf14 is also involved in cell cycle regulation by interaction with Sth1, which plays an important role in the RSC complex.

In the present study, we determined the solution structure of the Taf14 YEATS domain using NMR. The Taf14 YEATS domain exhibits a β-sandwich structure that contains two parallel β-sheets. Structural comparison between the Taf14 YEATS domain and Yaf9 YEATS domain revealed significant differences. In vivo, deletion of the Taf14 YEATS domain enhanced cell growth under stress conditions. In addition, our results indicate that the C-terminus of Taf14 has an important role in the interaction of Taf14 with Sth1.

EXPERIMENTAL

Cloning, expression and purification of recombinant Taf14 YEATS domain

The Taf14 YEATS domain was obtained from the S. cerevisiae DNA gene library by PCR and was cloned into the NdeI/XhoI-cleaved plasmid pET22b(+) (Novagen). The recombinant vector was transformed into the expression host BL21(DE3). The recombinant Taf14 YEATS domain was expressed and purified as described previously [19]. Uniformly 15N- and 13C-labelled protein was prepared with medium containing 0.5 g/l 99% ammonium chloride and 2.5 g/l 99% 13C-glucose as the sole nitrogen and carbon source respectively. The NMR sample contained 0.8 mM Taf14 YEATS domain, 20 mM sodium phosphate (pH 6.5), 100 mM sodium chloride and 1.5 mM dithiothreitol in either 90% H2O/10% 2H2O or 100% 2H2O.

NMR experiments and structure calculations

All NMR data were collected at 298 K on a Bruker DMX500 spectrometer. A set of standard triple-resonance spectra was recorded for backbone and side-chain assignments. NOE (nuclear Overhauser effect) distance restraints were obtained from 3D (three-dimensional) 15N- and 13C-edited NOESY (nuclear Overhauser enhancement spectroscopy) spectra acquired with a mixing time of 130 ms. After these experiments, the sample was lyophilized and redissolved in 99.96% 2H2O. A series of 15N-HSQC (15N-heteronuclear single-quantum coherence) experiments were performed to monitor the disappearance of NH signals to obtain the hydrogen-bond information. NMR data were processed with NMRPipe and analysed with Sparky 3 software. Chemical shift index was carried out for Cα, Cβ, C′ and Hα. The information on the ϕ and ψ backbone dihedral angles was obtained using the TALOS program [20]. Hydrogen-bond restraints were obtained by assignment of slow-exchange amide protons located in regular SSEs (secondary structure elements). The CNS program [21] was used to calculate the 3D structure of the Taf14 YEATS domain by distance restraints using the ARIA setup and protocols. Short-range NOEs and long-range NOEs were first used to determine the SSEs of the Taf14 YEATS domain. ϕ and ψ backbone dihedral angles and hydrogen-bond restraints were added in consecutive steps to constrain the 3D structure better. The 20 structures with the lowest energy were analysed with MOLMOL [22]. The Ramachandran plot was analysed with PROCHECK [23].

15N Longitudinal (T1) and transverse (T2) relaxation times and the heteronuclear 1H-15N NOE were detected at 298 K on a Bruker DMX500 spectrometer. For the T1 measurements, eight time points were collected with delays of 11.15, 61.30, 141.54, 241.84, 362.20, 522.68, 753.37 and 1144.54 ms. For T2, seven time points with delays of 0, 17.60, 35.20, 52.80, 70.40, 105.60 and 140.80 ms were obtained. The heteronuclear 1H-15N NOE was measured from duplicate pairs of 1H-15N spectra recorded with and without amide proton saturation.

Yeast strains, plasmids, media and genetic methods

Yeast strains used in the present study were BY4742 (wild-type) and BY4742 (ΔTaf14). Plasmids used in the present study are listed in Supplementary Table S1 (at http://www.BiochemJ.org/bj/436/bj4360083add.htm). Media such as YPD [1% (w/v) yeast extract/2% (w/v) peptone/2% (w/v) glucose] and synthetic complete (YC) were made according to standard procedures [24]. Formamide was added at a final concentration of 2%. Hydroxyurea was added at a concentration of 30 μg/ml. Yeast cells were transformed using lithium acetate [25]. For the growth studies on plates, cells in exponential phase were diluted. Five concentrations with D600 of 10−1, 10−2, 10−3, 10−4 and 10−5 were made. The same volume of cells was applied on to YC-URA (synthetic medium lacking uracil) plates. Cells were grown at 30°C.

SPR (surface plasmon resonance)

Real-time interactions of Sth1 with Taf14, Taf14 YEATS domain and the C-terminus of Taf14 were measured on a Biacore 3000 system. Taf14 and the C-terminus of Taf14 were cloned, expressed and purified in the same way as the Taf14 YEATS domain. Sth1 was cloned into the NdeI/XhoI-cleaved plasmid pGEX-4T-1 and was expressed and purified by standard affinity chromatography procedures. The CM5 (carboxymethylated dextran) chip was activated using an amine coupling-reagent mixture containing 0.4 M EDC [N-ethyl-N-(3-dimethylaminopropyl)carbodi-imide] and 0.1 M NHS (N-hydroxysuccinimide) at a flow rate of 5 μl/min for 10 min. Sth1 was diluted to a concentration of 2 mg/ml with running buffer (20 mM NaH2PO4, pH 7.0, containing 100 mM NaCl) and immobilized on the chip at a flow rate of 5 μl/min for 10 min. Ethanolamine (1.0 M, pH 8.5) was used to neutralize unbound activated sites on the chip at a flow rate of 5 μl/min for 10 min. The chip was washed twice with regeneration buffer (40 mM NaOH) and then with running buffer. Purified Taf14, Taf14 YEATS domain and the C-terminus of Taf14 were diluted with running buffer. The kinetic analysis of interaction between Sth1 and Taf14 was performed at six concentrations of Taf14 (7.00 μM, 3.50 μM, 1.75 μM, 0.88 μM, 0.44 μM and 0 μM) at a flow rate of 10 μl/min for 2 min. Kinetic analysis of interaction between Sth1 and C-terminus of Taf14 was also performed at six concentrations of the C-terminus of Taf14 (5.30 μM, 2.65 μM, 1.33 μM, 0.67 μM, 0.34 μM and 0 μM) at a flow rate of 10 μl/min for 2 min. Regeneration was performed at a flow rate of 30 μl/min for 2 min. The analysis was performed three times for each concentration. Kinetic analysis of SPR data was performed using BIAevaluation 4.1 (Biacore). Curves were fitted to the 1:1 (Langmuir) binding model. The dissociation constant (Kd) was derived from the kinetic analysis.

RESULTS

Sequence analysis of Taf14 and Taf14 YEATS domain

Taf14, 244 residues in length, can be divided into two parts: a YEATS domain at the N-terminus which is evolutionarily conserved from yeasts to humans, a C-box and another region at the C-terminus (Figure 1A). The Taf14 YEATS domain shares 18–44% sequence identity and 40–65% sequence similarity with other YEATS domain family members. Amino acid sequence analysis for the YEATS-domain-containing proteins shows YEATS domains contain a few conserved residues and have similar secondary structure patterns composed of eight β-strands. Interestingly, the sequence in the eighth β-strand and the loop connecting the seventh and the eighth β-strands are not conserved (Figure 1B). The highly divergent sequence in this region might imply the existence of different classes within the YEATS domain family.

Figure 1 The Taf14 YEATS domain

(A) Schematic representation of the Taf14 YEATS domain and other elements. (B) Sequence alignment of the Taf14 YEATS domain with other YEATS domains by ClustalW [30] and annotated using ESPript [31].

Solution structure of the Taf14 YEATS domain

The Taf14 YEATS domain, containing residues 1–123 of the protein, was recombinantly expressed and purified. The solution structure of the Taf14 YEATS domain was calculated based on a series of NMR spectra. The NMR data used for structure calculations are summarized in Table 1. In total, approximately 95% of backbone atoms and 90% of side-chain atoms were assigned. The chemical shift assignments of the Taf14 YEATS domain have been deposited in the Biological Magnetic Resonance Data Bank (http://www.bmrb.wisc.edu; accession number 17352). Depending on chemical shifts of various atoms, secondary structure prediction was carried out using CSI. The result shows the secondary structures of the Taf14 YEATS domain encompass eight β-strands (see Supplementary Figure S1 at http://www.BiochemJ.org/bj/436/bj4360083add.htm). The calculated structures of the Taf14 YEATS domain have been deposited in the PDB under accession code 2L7E. The best representative structure and the assembly of the 20 lowest-energy structures are shown in Figures 2(A) and 2(B) respectively. The statistical parameters (Table 1) indicate a high-quality structure of the Taf14 YEATS domain. The structure looks like a β-sandwich, which contains two β-sheets. One β-sheet consists of strand 1 (residues 6–18), strand 2 (residues 29–38), strand 5 (residues 76–80) and strand 8 (residues 110–120). The other β-sheet consists of strand 3 (residues 52–57), strand 4 (residues 66–69), strand 6 (residues 84–91) and strand 7 (residues 99–106). The two β-sheets are fairly parallel and close to each other (Figure 2A).

View this table:
Table 1 NMR and structural statistics

i, and j are random atoms.±values are ±S.D.

Figure 2 NMR structure of the Taf14 YEATS domain

(A) Ribbon representation of the minimized averaged structure of the Taf14 YEATS domain with the SSEs highlighted. The structure of the Taf14 YEATS domain has been deposited in the PDB under accession code 2L7E. (B) Backbone superposition of 20 selected conformers with the lowest energy from the final CNS calculation. This Figure was produced with MOLMOL.

The structure of the Taf14 YEATS domain exhibits a few unique features, except for the β-sandwich pattern. There are a number of hydrogen bonds between adjacent β-strands within the β-sheets, which are indicated by two-dimensional 1H-2H exchange experiments. The hydrogen bonds enable the β-sheets to pack compactly. The second feature is the flexible loops of the Taf14 YEATS domain. The Taf14 YEATS domain has several loops, e.g. the loop (Leu-His-Xaa-Ser/Thr-Tyr-Phe, where Xaa is any amino acid) connecting strands β3 and β4, and the loop (Gly-Trp-Gly) connecting strands β5 and β6. These loops are conserved in various YEATS-domain-containing proteins. The loops of the Taf14 YEATS domain are highly flexible in solution, as demonstrated by the lack of distance constrains in 15N-NOESY and 13C-NOESY spectra. The third feature is that the surface of the Taf14 YEATS domain is distributed with a large amount of negative charge (see Supplementary Figure S2 at http://www.BiochemJ.org/bj/436/bj4360083add.htm). As Taf14 is involved in several protein complexes, this electrostatic distribution pattern may be relevant to its interactions with other protein regulators.

Dynamic analysis of the Taf14 YEATS domain

The Taf14 YEATS domain contains eight β-strands and several loops. The dynamic characteristics of the β-strands and the loops were investigated by relaxation experiments (Figure 3). The heteronuclear NOE data show relatively smaller values for residues 4–6, 18–26, 41–50, 62–64, 70–71 and 95–98 (Figure 3D), signifying more backbone flexibility in these regions, consistent with their location near the N-terminus or within the loops. Correspondingly, the T1 values at these sites were bigger than others in the Taf14 YEATS domain (Figure 3A), indicating higher motion on the pico- to nano-second time scale in these regions. Notably, residues 21, 27, 44, 48, 71, 84, 95 and 96 exhibit relative smaller T2 (Figure 3B), implying the milli- to micro-second time-scale motion at these sites. The slow motion might result from the cistrans isomerism of the proline residues immediately following these residues or the conformational exchange of the long flexible loop (residues 90–100).

Figure 3 The heteronuclear 1H-15N NOEs, 15N T1 and 15N T2 relaxation rates of the Taf14 YEATS domain

The black bars indicate the values of T1 (A), T2 (B), T1/T2 (C) and the heteronuclear 1H-15N NOEs (D).

Structural comparison between the Taf14 and Yaf9 YEATS domains

The structure of the Taf14 YEATS domain was submitted to the structure recognition program DALI [26] to search for its similar structures. Not surprisingly, the result shows that the structure most similar to the Taf14 YEATS domain is that of the Yaf9 YEATS domain (PDB code 3FK3), which is the only solved structure of a YEATS domain [27]. The Cα RMSD (root mean square deviation) between the Taf14 and Yaf9 YEATS domain is 5.7 Å (1 Å=0.1 nm), with a Z-score of 6.0. Sequence identity and similarity between the Taf14 and Yaf9 YEATS domains are 18% and 40% respectively. The two YEATS domains share some common features. Besides adopting a similar elongated β-sandwich fold and possessing similar flexible loops, they both contain eight β-strands, and adopt the same topology (Figure 4A). Relative to the residues in structured regions, those in flexible loops are more conserved. The sequence conservation and structure flexibility of the loops in the two YEATS domains might imply their similar roles in function, as flexible loops are usually the executors of biological functions.

Figure 4 Structural comparison between the Taf14 and Yaf9 YEATS domains

(A) Topology diagram of the β-sandwich structure from the Taf14 and Yaf9 YEATS domains. (B) Structural comparison of the Taf14 and Yaf9 YEATS domains. Left-hand image: Taf14 YEATS domain (PDB code 2L7E); right-hand image: Yaf9 YEATS domain (PDB code 3FK3). The ribbon diagram is produced by MOLMOL.

However, the Taf14 YEATS domain differs significantly from the Yaf9 YEATS domain in some aspects. The Yaf9 YEATS domain adopts a trimeric form in crystal structure (although it was detected to be monomeric in solution), whereas the Taf14 YEATS domain exhibits a monomeric form in solution. In addition, the eight β-strands of Taf14 YEATS domain are composed of 123 residues. Compared with this, in the Yaf9 YEATS domain, the eight β-strands encompass 164 residues. This discrepancy was largely due to the much longer loop between β7 and β8 in the Yaf9 YEATS domain (Figures 1B and 4). In addition, in the Yaf9 YEATS domain, the two β-sheets are not completely parallel and relatively far from each other, and the β-strands within the same β-sheets are not parallel with each other. In contrast, the two β-sheets of the Taf14 YEATS domain are basically parallel and close to each other (Figure 4B). Moreover, the Taf14 YEATS domain lacks the two short α-helices in Yaf9 YEATS domain and a number of hydrogen bonds between adjacent β-strands stabilize the compact structure of the Taf14 YEATS domain.

The Taf14 YEATS domain plays a negative role in cell growth under different stress conditions

S. cerevisiae devoid of Taf14 had defects in cell growth. Furthermore, cells devoid of Taf14 were viable but thermo- and osmo-sensitive. To study the biological function of the Taf14 YEATS domain, two mutant cell strains with the YEATS domain or the C-terminus of Taf14 deleted were constructed in vivo. Growth rates were compared between the wild-type cells and the two mutant cells. In YC-URA plates at 30°C, the C-terminus-deleted cells and the YEATS-deleted cells showed similar growth rates to that of the wild-type cells (Figure 5A).

As Taf14-deleted yeast cells are sensitive to different genotoxic stressors, growth rates of the strains expressing different Taf14 constructs in the presence of several stressors were also compared to study the function of Taf14 in cell cycle and cell growth. Hydroxyurea blocks the synthesis of deoxynucleotides, which inhibits DNA synthesis and induces synchronization or cell death in S-phase, and formamide destabilizes nucleic acid duplexes to affect the cell cycle. Thus these two genotoxic stressors both affect the cell cycle. In the presence of hydroxyurea, the cells expressing YEATS-deleted Taf14 exhibited a similar growth rate to that of the wild-type cells, whereas the C-terminus-deleted cells could not grow at all (Figure 5B). In the presence of formamide, the C-terminus-deleted cells could not grow at all, whereas the YEATS-deleted cells exhibited an enhanced growth rate compared with wild-type cells (Figure 5C). These observations indicate that the YEATS domain and C-terminus of Taf14 might execute different functions in the cells and the YEATS domain might play a negative role in cell growth under the stress conditions.

Figure 5 Growth rates of wild-type cells and Taf14-deficient cells expressing different Taf14 mutants

(A) Growth rates in YC-URA at 30°C. (B) Growth rates in YC-URA with hydroxyurea at 30°C. (C) Growth rates in YC-URA with formamide at 30°C. WT, wild-type cells; Taf14, Taf14-deficient cells expressing intact Taf14; C-terminus of Taf14Δ, Taf14-deficient cells expressing the YEATS domain of Taf14; Taf14 YEATS domainΔ, Taf14-deficient cells expressing the C-terminus of Taf14.

The C-terminus of Taf14 is responsible for the interaction between Taf14 and Sth1

The RSC complex is essential for cell cycle progression and chromatin remodelling. Taf14 has been found to be involved in the RSC complex. Two-hybrid screening and GST (glutathione transferase) pull-down indicate that Taf14 interacts with Sth1, an essential component of the RSC complex [17]. In the present study, we show that the YEATS domain of Taf14 does not interact with Sth1 directly, whereas the C-terminus is responsible for this interaction (Figure 6A). The affinity constant was measured to be 0.79±0.03 μM for the interaction between intact Taf14 and Sth1, and 0.15±0.02 μM for the interaction between the C-terminus of Taf14 and Sth1 (Figures 6B and 6C), illustrating that binding of Taf14 to Sth1 is strong in the RSC protein complex and mediated by the C-terminus of Taf14.

Figure 6 Kinetic analysis of interactions between Sth1 and Taf14, the Taf14 YEATS domain or the C-terminus of Taf14

(A) Interactions between Sth1 and Taf14, the Taf14 YEATS domain or the C-terminus of Taf14. (B) Interactions between Sth1 and Taf14 with different concentrations of Taf14. (C) Interactions between Sth1 and the C-terminus of Taf14 with different concentrations of the C-terminus of Taf14.

DISCUSSION

The YEATS domain is an evolutionarily conserved domain involved in various protein complexes and related to a number of human cancers. However, the biological functions and structures of YEATS domains are poorly understood. Interestingly, proteins containing a YEATS domain are generally involved in multi-subunit protein complexes that can recognize and change different chromatin modifications. A universal distribution of YEATS domains in chromatin modification complexes might imply a critical role of YEATS domains in chromatin remodelling and regulation of gene expression. In the present study, we determined the solution structure of the Taf14 YEATS domain and conducted dynamic analysis using NMR. The calculated Taf14 YEATS domain structure shows that it adopts an elongated β-sandwich fold formed by eight β-stands. The fold is compact and stabilized by a number of hydrogen bonds between adjacent β-strands. Dynamic data show the loops between β-strands are relatively flexible and exhibit slow time-scale motion at some sites, which might result from cistrans isomerism of the proline residues or conformation exchange. The conformational flexibility and heterogeneity of the loops probably contribute to the interaction of the Taf14 YEATS domain with its partners.

So far, the solution NMR structure of the monomeric Taf14 YEATS domain in the present study and the crystal structure of trimeric Yaf9 YEATS domain are the only two solved structures for YEATS domains. The Taf14 YEATS domain has a similar global fold to the Yaf9 YEATS domain. However, the Taf14 YEATS domain differs significantly from the Yaf9 YEATS domain in many aspects, which includes the composition of SSEs, the distance and spatial orientation of the two β-sheets, and the length of the loop between β7 and β8. Sequence analysis of YEATS domains indicates that YEATS domains are obviously conserved in the first seven β-strands, whereas they are highly divergent in the eighth β-strand and the loop connecting β7 and β8. This divergence of the loop is reflected in both the amino acid composition and the length, as exemplified by the comparison of the Taf14 and Yaf9 YEATS domains. The structural distinctions of the two YEATS domains might imply their different biological functions and represent the different classes of YEATS domain family.

The YEATS domain exists in many species as a conserved domain superfamily. However, its biological function is so far unclear. Taf14 is involved in several chromatin remodelling and transcription regulation complexes. In the present study, we showed enhanced yeast cell growth rate when the Taf14 YEATS domain was deleted in the presence of formamide, and the extremely slow growth of cells expressing the Taf14 YEATS domain alone in several different stress conditions. Formamide and hydroxyurea both affect the cell cycle by causing DNA damage. The results indicate that the YEATS domain and the C-terminus of Taf14 might be involved in distinct functions and the YEATS domain has a negative role in cell growth under the stress conditions. As formamide and hydroxyurea both affect the cell cycle by causing DNA damage, the Taf14 YEATS domain might be involved in a few biological events, such as DNA repair and cell cycle regulation.

It has been shown that Taf14 interacts with Sth1 in the RSC complex [17]. The results from the present study indicate that the YEATS domain of Taf14 does not interact with Sth1, whereas the C-terminus is responsible for this interaction with a strong binding affinity. As the C-terminus of Taf14 has a highly flexible conformation, which is indicated by disorder prediction analysis using PONDR [28] (results not shown), and disordered regions are often involved in recognition and interaction between proteins [29], it is not surprising that the interaction between Taf14 and Sth1 is mediated by the C-terminus of Taf14. Possibly, by adopting a compact and rigid fold, the Taf14 YEATS domain might function as a scaffold in chromatin remodelling and transcription regulating complexes and interact with other subunits via the conformationally flexible C-terminus.

AUTHOR CONTRIBUTION

Wen Zhang and Jiahai Zhang performed the experimental work. Xuecheng Zhang, Chao Xu and Xiaoming Tu wrote the manuscript. Xiaoming Tu supervised the project and all authors discussed the results and commented on the manuscript.

FUNDING

This work was supported by the Knowledge Innovation Program of the Chinese Academy of Science [grant number KSCX2-EW-Q-4], the National High-Tech R&D program [grant number 2006AA02A315] and the National Basic Research Program of China (973 Program) [grant numbers 2007CB914503 and 2009CB918804].

Acknowledgments

We thank F. Delaglio and A. Bax (Laboratory of Chemical Physics, National Institute of Diabetes and Digestive and Kidney Disease, Bethesda, MD, U.S.A.) for providing NMRPipe and NMRDraw; T.D. Goddard and D. Kneller (University of California, San Francisco, San Francisco, CA, U.S.A.) for providing Sparky; A.T. Brünger (Howard Hughes Medical Institute and Department of Molecular Biophysics and Biochemistry, Yale University, New Haven, CT, U.S.A.) for providing CNS; and R. Koradi and K. Wuthrich (Institut für Molekularbiologic und Biophysik, Eidgenössische Technische Hochschule-Honggerberg, Zurich, Switzerland) for providing MOLMOL.

Footnotes

  • The structural co-ordinates reported for the Taf14 YEATS domain will appear in the PDB under accession code 2L7E.

Abbreviations: 3D, three-dimensional; NOE, nuclear Overhauser effect; NOESY, nuclear Overhauser enhancement spectroscopy; RMSD, root mean square deviation; SPR, surface plasmon resonance; SSE, secondary structure element; YC-URA, synthetic medium lacking uracil

References

View Abstract