- Email: [email protected]
Structural insight into the role of VAL1 B3 domain for targeting to FLC locus in Arabidopsis thaliana
Biochemical and Biophysical Research Communications xxx (2018) 1e8
Contents lists available at ScienceDirect
Biochemical and Biophysical Research Communications journal homepage: www.elsevier.com/locate/ybbrc
Structural insight into the role of VAL1 B3 domain for targeting to FLC locus in Arabidopsis thaliana Baixing Wu a, *, 1, Mengmeng Zhang a, 1, Shichen Su a, Hehua Liu b, Jianhua Gan b, Jinbiao Ma a, ** a
State Key Laboratory of Genetic Engineering, Collaborative Innovation Centre of Genetics and Development, Department of Biochemistry, Institute of Plant Biology, School of Life Sciences, Fudan University, Shanghai 200438, China State Key Laboratory of Genetic Engineering, Collaborative Innovation Center of Genetics and Development, Department of Physiology and Biophysics, School of Life Sciences, Fudan University, Shanghai, 200438, China b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 28 April 2018 Accepted 1 May 2018 Available online xxx
Vernalization is a pivotal stage for some plants involving many epigenetic changes during cold exposure. In Arabidopsis, an essential step in vernalization for further flowering is successful silence the potent floral repressor Flowering Locus C (FLC) by repressing histone mark. AtVal1 is a multi-function protein containing five domains that participate into many recognition processes and is validated to recruit the repress histone modifier PHD-PRC2 complex and interact with components of the ASAP complex target to the FLC nucleation region through recognizing a cis element known as CME (cold memory element) by its plant-specific B3 domain. Here, we determine the crystal structure of the B3 domain in complex with Sph/RY motif in CME. Our structural analysis reveals the specific DNA recognition by B3 domain, combined with our in vitro experiments, we provide the structural insight into the important implication of AtVAL1-B3 domain in flowering process. © 2018 Elsevier Inc. All rights reserved.
Keywords: Vernalization Flowering FLC locus PRC2 B3 domain DNA binding
1. Introduction Vernalization is the main method used by some plants to cope with the environmental stimuli during cold exposure [1,2]. In Arabidopsis, a key regulating mechanism of vernalization is through silencing the potent floral repressor Flowering Locus C (FLC) [2e5]. Sets of histone modifiers and a long noncoding RNA known as COOLAIR were indicated as regulators that dynamically modulated the expression of FLC [1,2,5]. The N-terminal tails of histone proteins carry various reversible post-translational modifications. One of these modifications, H3K27me3, is validated to be a repress mark to FLC that dynamically installed and uninstalled by a series of modifiers including Polycomb group (PcG) and Trithorax group [6]. Polycomb-mediated silencing typically involves PcG complex
* Corresponding author. State Key Laboratory of Genetic Engineering, Collaborative Innovation Centre of Genetics and Development, Department of Biochemistry, Institute of Plant Biology, School of Life Sciences, Fudan University, Shanghai 200438, China. ** Corresponding author. E-mail addresses: [email protected] (B. Wu), [email protected] (J. Ma). 1 Co-first authors.
recruitment and spreading of repression mark at target loci [4,7]. An interacting partner of PcG known as Polycomb repressive complex 2 (PRC2), is pointed to catalyze the repressive H3K27me3 in animals and in plants [8]. Cold exposure will induce the formation of a PHDPRC2 complex to localize to FLC chromatin involving the expression of two plant homeodomain (PHD) protein called VERNALIZATION INSENSITIVE 3 (VIN3) and VIN3-like protein 1(VRN5) [9,10]. This PHD-PRC2 complex accumulates specifically in a region, termed the “nucleation region” that covers the first exon and part of the first intron of FLC, to catalyze the H3K27me3 during vernalization [11]. After winter, when the temperature rises, the H3K27me3 modification spreads to cover the entire FLC locus, resulting in stable FLC silencing to render the vernalized plants competent to flower [11]. Thus, the PHD-PRC2 complex targeting to the “nucleation region” of FLC is a key step for downstream functions. A cis DNA element in FLC locus containing two Sph/RY motifs (50 -TGCATG-3’; referred to hereafter as RY motifs) were recently discovered crucial for FLC repression during vernalization [12,13]. The two RY motifs containing region that essential for the maintenance of FLC silencing upon return to warm temperatures were then named cold memory element (CME) as its capability to remember the past cold [14]. Furthermore, H3K27me3 levels were
https://doi.org/10.1016/j.bbrc.2018.05.002 0006-291X/© 2018 Elsevier Inc. All rights reserved.
Please cite this article in press as: B. Wu, et al., Structural insight into the role of VAL1 B3 domain for targeting to FLC locus in Arabidopsis thaliana, Biochemical and Biophysical Research Communications (2018), https://doi.org/10.1016/j.bbrc.2018.05.002
2
B. Wu et al. / Biochemical and Biophysical Research Communications xxx (2018) 1e8
not found to increase when mutated the RY motif during vernalization, manifesting that the RY motifs enable PHD-PRC2 nucleation. Meanwhile, two B3 domain family protein AtVAL1 and AtVAL2 but not AtVAL3 were found as the transacting protein binding to the RY motifs and confirmed that the binding was mediated by its B3 domain [12,13], which was previously shown to recognize the RY motif [15]. These two homologs belong to the ABSCISIC ACID INSENSITIVE 3 (ABI3), FUSCA3 (FUS3) and LEAFY COTYLEDON 2 (LEC2) family of transcriptional repressor proteins [16]. AtVAL1/2 were shown to function with redundancy, e.g. AtVAL1 and AtVAL2 were previously shown to function together during seedling development with sugar signaling to repress ectopic expression of seed maturation genes [17]; simultaneous disruption of val1 and val2 resulted in seedlings that developed embryonic traits and accumulated seed storage proteins and oils, whereas individual val1 and val2 mutants had no apparent morphological phenotypes [17,18]. Moreover, AtVAL1/2 were also shown to function with redundancy and physically interact with the Polycomb component LIKE HETEROCHROMATIN PROTEIN 1 (LHP1) [13], and interact with components of the ASAP complex and promote histone deacetylation during vernalization [12]. AtVAL1/2 proteins seem to be sequence-specific epigenome readers that, through multivalent recognitions and interactions, can be tethered to a particular allele or locus [13]. Whereas, the structural basis of these domains of AtVAL1/2 in the regulation of gene expression is yet to be elucidated. Here, we report the crystal structure of B3 domain of AtVAL1. Our AtVAL1 B3 domain in complex with RY motif and biochemical studies unveil the specific DNA recognition by AtVAL1/2 and give a structural evidence that why AtVAL1/2 but not AtVAL3 can recognize the RY motif. We also illustrate the structural basis for different sequence selectivity among B3 domain-containing proteins in plants. 2. Materials and methods 2.1. Protein expression and purification The PCR-amplified cDNA fragments encoding the AtVAL1 B3
domain (273e400) were cloned into a modified pSUMO vector encoding 6xHis tag at N-terminus following a SUMO fusion tag and Ulp1 protease site. The plasmid containing the DNA insert of interest was transformed into Escherichia coli strain BL21 (DE3) grown in LB medium supplemented with 50 mg/ml kanamycin. The recombinant protein expression was induced by 0.2 mM IPTG at 37 C, followed by 16e18 h incubation at 18 C. The cell pellets were resuspended in buffer containing 20 mM Tris-HCl pH 8.0, 500 mM NaCl, 25 mM imidazole pH 8.0 and lysed using the high press and further clarified by centrifugation at 18,000 rpm. Supernatants were purified with nickel-chelating affinity column HisTrap (GE Healthcare), the target protein was washed with lysis buffer and then eluted with a buffer containing 20 mM Tris-HCl, pH 8.0, 500 mM NaCl and 500 mM imidazole pH 8.0. Ulp1 protease was added to remove the N-terminal tag and the fusion protein of the recombinant protein and dialyzed with lysis buffer containing 20 mM Tris pH 8.0, 500 mM NaCl 3 h. The mixture was applied to another Ni-NTA resin to remove the protease and uncleaved proteins with the buffer containing 20 mM HEPES pH7.5, 500 mM NaCl, 25 mM Imidazole pH 8.0. Eluted proteins were concentrated by centrifugal ultrafiltration, loaded onto a pre-equilibrated HiLoad € 16/60 Superdex 75-pg column in an Akta-purifier (GE Healthcare), eluted at a flow rate of 1 ml/min with the buffer containing 10 mM HEPES pH 7.5, 200 mM NaCl. Peak fractions were analyzed by SDSPAGE (15%, w/v) and stained with Coomassie Brilliant Blue R-250. Purified fractions were pooled together and concentrated by centrifugal ultrafiltration. The concentration was determined by A280. 2.2. ITC measurements For B3 domain and DNA interactions, ITC assays were carried out on a MicroCal iTC200 calorimeter (GE Healthcare) at 25 C. The buffer used for testing the binding affinity between proteins and DNA was 10 mM HEPES pH 7.5, 100 mM NaCl. The concentrations of proteins were determined spectrophotometrically. The DNA substrates were diluted in the reaction buffer. The ITC experiments involved 25 injections of protein into DNA. The sample cell was loaded with 250 ml of DNA at 50 mM and the syringe with 80 mL of
Table 1 Data collection and refinement statistics.
Data collection Space group Cell dimensions a, b, c (Å) a, b g, ( ) Wavelength(Å) Resolution (Å) Rmerge I/sI Completeness (%) Redundancy Refinement Resolution (Å) No. reflections Rwork/Rfree No. atoms Protein DNA Water/Ligands Average B-factor (Å) R.m.s. deviations Bond lengths (Å) Bond angles ( ) Ramachandran plot Favored/allowed (%)
B3 (273e400) and 13bp-DNA (Hg-II) (5YZY)
B3 (273e400) and 13bp-DNA (5YZZ)
I4122
I4122
87.780, 87.780, 148.735 90, 90, 90 0.97853 75.60e2.608 (2.69e2.60) 0.096 37.3 (2.25) 100.0(100.0) 16.3
86.896, 86.896, 148.237 90, 90, 90 0.97853 74.97e2.585 (2.67e2.58) 0.077 29.4 (2.33) 99.5(95.7) 13.9
30.00e2.61 7578 0.2103/0.2567 1394 866 527 1 37.787
30.00e2.58 7414 0.2260/0.2552 1393 866 527 0 44.338
0.013 1.70
0.013 1.89
93.58/6.42
92.66/6.42
Please cite this article in press as: B. Wu, et al., Structural insight into the role of VAL1 B3 domain for targeting to FLC locus in Arabidopsis thaliana, Biochemical and Biophysical Research Communications (2018), https://doi.org/10.1016/j.bbrc.2018.05.002
B. Wu et al. / Biochemical and Biophysical Research Communications xxx (2018) 1e8
protein at 1 mM. Curve fitting to a single binding site model was performed by the ITC data analysis module of Origin 7.0 (MicroCal) provided by the manufacturer. DGo of protein-DNA binding was computed as RTln(1/KD), where R, T, and KD are the gas constant, temperature and dissociation constant, respectively. 2.3. Oligonucleotides All oligonucleotides used in this study were synthesized from Sangon Biotech (Shanghai).
3
2.5. Data collection and structure determination The B3 domain in complex with 13-bp DNA was collected from a single crystal at 100 K at SSRF-BL18U1. HKL3000 was used for data processing [19]. The phase was determined by SAD with the anomalous signal of mercury using hkl2map [20]. The model was manually built using COOT [21] and refined using REFMAC [22]. Data collection and refinement statistics are shown in Table 1. All molecular representations were prepared using PyMOL (DeLano Scientific).
2.4. Crystallization
3. Results
The B3 domain in a buffer 10 mM HEPES pH 7.5, 200 mM NaCl was concentrated to 10 mg/ml. And then the protein and DNA were mixed with the molar ratio of 1:1.2. B3 with 13bp-DNA was crystallized in the solution containing 30% (w/v) PEG 5,000 MME, 0.1 M Tris base/Hydrochloric acid pH 8.0, 0.2 M Lithium sulfate. An Hgsoak was obtained by soaking crystals of the B3 domain-DNA complex for 3 min in the crystallization solution supplemented with 1 mM methylmercury chloride. The B3-DNA complex crystals were cryoprotected by adding the crystallization solution supplemented with 25% glycerol and flash frozen in liquid nitrogen.
3.1. AtVAL1 B3 domain specifically recognize the Sph/RY motifs in CME Previous studies identified that the AtVAL1 and AtVAL2 can specifically bind to a 47-bp FLC silencing element containing two RY motifs [12,13]. Thus, we first set out to test the in vitro binding affinity with wildtype or mutant DNA substrates using ITC experiment (Fig. 1B). As expected, the truncated B3 domain (273e400) that we designed could bind to the CME-wildtype with a high binding affinity about 8.47 mM with a molar ratio about 2:1 (Fig. 1C).
Fig. 1. Interactions between B3 domain and RY motif are sequence-specific. (A) Domain organization of AtVAL1/2. The PHD-like domain containing region is colored in green; the B3 domain is colored in sky blue; the CW domain is colored in pink; the zinc finger of unknown function is colored in red and the EAR domain is colored in yellow. (B) Schematic of the wildtype CME region, CME mutants and the sequence used for crystallization. The core RY motif is circled by a black box. The mutations are colored in red. (C-F) ITC results of B3 domain titrate into wildtype or mutated CME DNA sequence. (C) ITC results of wildtype CME and B3 domain. (D) ITC results of CME-M1 and B3 domain. (E) ITC results of CME-M2 and B3 domain. (F) ITC results of CME-M1/M2 and B3 domain. (G) ITC results of B3 domain titrate to 13bp DNA substrate. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)
Please cite this article in press as: B. Wu, et al., Structural insight into the role of VAL1 B3 domain for targeting to FLC locus in Arabidopsis thaliana, Biochemical and Biophysical Research Communications (2018), https://doi.org/10.1016/j.bbrc.2018.05.002
4
B. Wu et al. / Biochemical and Biophysical Research Communications xxx (2018) 1e8
And then, according to the studies that the transition of the cytidine to thymine in the RY motif will destroy the binding between RY motif and AtVAL1 [12], we design another three substrates containing one or two mutations to confirm these results (Fig. 1B). Our ITC results showed that mutations in either RY motifs will destroy the binding between B3 domain and the mutated RY motif, but not impact the binding affinity with both CME-M1 and CME-M2, molar ratio of which are about 1:1 (Fig. 1D and E). Simultaneously mutate the two RY motifs in CME will cause the binding between B3 domain and CME lost (Fig. 1F). These results confirmed the specific binding between B3 domain and RY motif and the molar ratio depends on the number of RY motifs. 3.2. Crystal structure of AtVAL1 B3 domain in complex with RY motif Based on these ITC results, we design another set of DNA substrates with overhang for crystallization with a high binding affinity with the B3 domain (Fig. 1G). And then, we successfully obtained the crystal structure of B3 domain (273e400) in complex with 13bp-DNA (50 -ATTCTGCATGGAT-30 , 30 -AAGACGTACCTAT-50 ) that diffracted to 2.6 Å (Table 1). This complex structure of AtVAL1 B3 domain with DNA belongs to the space group I4122. The structure was solved by single-wavelength anomalous diffraction methods and refined to high resolution with excellent geometry (Table 1). AtVAL1 B3 domain adopts a pseudobarrel fold as reported before in
other plant B3 domain-containing proteins (Fig. 2A) [23e26], which reveals a seven-stranded b-barrel capped at either edge of the barrel by a short helix (Fig. 2A). There are one B3 domain and a double strand DNA in the asymmetric unit. A positive-charge riched region faced with the DNA duplex (Fig. 2B). 3.3. Identification of the B3 domain DNA-binding region The DNA duplex is mainly recognized by two regions in the B3 domain, where we termed as loop1 (from 300 to 309 that contains a short helix a2), and a loop2 between b4 and b5 (from 349 to 354) (Fig. 2C). The two loops of AtVAL1 B3 domain insert into the major groove of the double strand DNA (Fig. 2C and D). Residues in the two loops are account for most of the specific recognition (Fig. 2D). There are appropriate contacts with the six base-pairs. In the T5: A22 pair, the methyl-group of T5 interacts with the side chain of I307 via hydrophobic contacts (Fig. 3A). In the G6: C22 pair, the O6, and N7 of G6 are recognized by Arg309 through hydrogen bonds (Fig. 3B). In the C7: G21 pair, the C7 is recognized through hydrophobic interactions with Trp349 with about 3.5 Å (Fig. 3C). The N7 of A8 in A8: T20 pair is hydrogen bonded to the side chain of Asn351 and the phosphate group hydrogen bonds to the side chain of Ser302, the methyl-group of T20 interacts with the side chain of V311 via hydrophobic contacts (Fig. 3D). Asn351 also denotes to the binding of T9 through its side chain by forming a hydrogen bond with the O4 in T9: A19 pair, the phosphate group of A19 is
Fig. 2. The overall structure of B3 domain in complex with 13bp-DNA containing RY motif. (A) The overall structure of B3 domain (colored in wheat) in complex with DNA (colored in white and black). Omit FO-FC electron density for the RNA at 2.6 A resolution, contoured level is 1.0 s. (B) Surface representation of B3-DNA complex. (C) The interacting region between B3 domain and RY motif. The contacting loops are colored in green indicated by the black arrow and the RY motif region is indicated by the black dash line. (D) The interacting loops are inserted into the major groove of target DNA. The double strand DNA is shown as spheres colored like Fig. 2A, the residues involved in the interaction with DNA was shown as stick colored in green. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)
Please cite this article in press as: B. Wu, et al., Structural insight into the role of VAL1 B3 domain for targeting to FLC locus in Arabidopsis thaliana, Biochemical and Biophysical Research Communications (2018), https://doi.org/10.1016/j.bbrc.2018.05.002
B. Wu et al. / Biochemical and Biophysical Research Communications xxx (2018) 1e8
5
Fig. 3. Specific recognition of RY motif by B3 domain. (A-F) Detailed interactions between residues of the B3 domain and DNA base-pair. (A) T5:A23; (B) G6:C22; (C) C7:G21; (D) A8:T20; (E) T9:A19; (F) G10:C18. (G) Sequence alignment between AtVALs B3 domain. The amino acids participated in recognizing the phosphate group is indicated by a green circle, recognizing the base through hydrophobic contacts and hydrogen bonds are shown as a blue and red circle, respectively. (H) Comparison of ITC results of mutations in each site of RY motif. The wildtype 13bp-DNA are colored in blue, and the mutants are colored in green, red, and purple. ND: not detectable. WB: weak binding. (I) ITC results of B3 mutants including R309D, N351S, and N352S. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)
Please cite this article in press as: B. Wu, et al., Structural insight into the role of VAL1 B3 domain for targeting to FLC locus in Arabidopsis thaliana, Biochemical and Biophysical Research Communications (2018), https://doi.org/10.1016/j.bbrc.2018.05.002
6
B. Wu et al. / Biochemical and Biophysical Research Communications xxx (2018) 1e8
recognized by Lys297 (Fig. 3E). In the G10: C18 pair, the O6 of G10 forms hydrogen bond with the side chain of Asn352, the N4 and phosphate group of C18 are hydrogen bonded to the side chain of Asn351 and the main chain of Lys314, respectively (Fig. 3F). These residues of AtVAL1 B3 domain involved in the specific recognition of the DNA play important role in binding the double strand DNA, especially three keynote residues: I307, Asn351 and Asn352, which are conserved in AtVAL1 and AtVAL2 but not in AtVAL3 (Fig. 3G). These contacts may explain why AtVAL3 cannot recognize the RY motif [13].
We also produced protein mutants R309D, N351S, and N352S to assess theirs in vitro DNA-binding ability using ITC experiment (Fig. 3I). The two mutants N351S and R309D show dramatically decrease of binding ability, and the N352S just slightly decrease the binding affinity, because the distance between the N352S and the O6 of G10 is 3.8 Å, there may exist water molecue that cannot be seen in our structure mediates this interaction (Fig. 3I). These mutations in both the DNA and the B3 domain confirm the importance of sequence specificity in the RY motif and the residues that involving the binding for protein-DNA interaction.
3.4. Mutagenesis studies of interactions between B3 domain and target DNA
3.5. Comparison of the structure of AtVAL1 B3 domain with homologous
To confirm the binding properties of these residues and nucleotides, mutations were first taken out in the RY motif 50 -TGCATG-30 except T5: A22. In each site, we mutate the sequence to another three base-pairs to confirm the sequence-specific binding (Fig. 3H). These mutants almost destroyed the binding ability to the B3 domain. Mutants of A6: T22, T6: A22 and C6: G22 all loss the binding ability with the B3 domain (Fig. 3H), the main interaction between G6 group and Arg309 is through O6, which is a lack in the other three types (Fig. 3H). Mutations in the C7: G21 has the same tendency like G6: C22 pair, although T7: A21 keeps some binding ability because the Trp349 denotes the steric hindrance to the C7 site, any mutation to purine in this site will clash with the side chain of Trp349 (Fig. 3H). The mutations of the other sites all present dramatically decrease binding ability (Fig. 3H).
There are four structures of plant B3 domain-containing proteins including two crystal structures, AtVRN1 B3 domain (PDB code: 4L1K) and AtARF1 in complex with DNA (PDB code: 4LDX), and two NMR structures, AtRAV1 (PDB code: 1WID) and At1g16640 (PDB code: 1YEL) have been reported previously [23e26]. These proteins share low sequence identity between each other, whereas they have very similar architectures except AtVRN1, which has a ahelix at its N-terminal (Fig. 4A). A DALI search when AtVAL1 B3 domain was used as a search model found that AtARF1 and AtRAV1 structures are the top hits: AtARF1 gives a Z-score of 16.3, with 32% sequence identity and an r.m.s.d. of 1.8 A, and AtRAV1 gives a Zscore of 13.3, with 38% sequence identity and an r.m.s.d. of 2.0 A [27]. AtVAL1 share more similarities with AtARF1 and AtRAV1 than AtVRN1 and At1g16640, the latter two are similar to each other.
̊ ̊
Fig. 4. Comparisons of B3 domain-containing proteins in plants (A) Superposition of all B3 domains that structures have been solved in plants. AtVAL1 is colored in wheat; AtARF1 is colored in purple-blue; AtRAV1 is colored in cherry, AtVRN1 is colored in cyan and At1g16640 is colored in deep blue. (B) Sequence alignment of B3 domain-containing proteins in Arabidopsis mentioned above. Sequence alignment of the B3 domain among AtVAL1-3, AtARF1, AtRAV1, AtVRN1 and At1g16640. Secondary structure alignment of AtVAL1 B3 domain and AtARF1 B3 domains is shown above and below the sequences, respectively. Blue rectangles indicate the residues involved in recognizing the RY motif by AtVAL1. (CF) Superposition of AtVAL1 B3 domain with structural homologs. The AtVAL1 is colored in wheat shown as cartoon and the loop responsible for interactions are colored in green. The others are colored like Fig. 4A. The RMSD between these structures are calculated by PyMOL. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)
Please cite this article in press as: B. Wu, et al., Structural insight into the role of VAL1 B3 domain for targeting to FLC locus in Arabidopsis thaliana, Biochemical and Biophysical Research Communications (2018), https://doi.org/10.1016/j.bbrc.2018.05.002
B. Wu et al. / Biochemical and Biophysical Research Communications xxx (2018) 1e8
7
These comparisons of structural similarities are reflected to some degree in the DNA binding properties of these proteins. AtVAL1, AtARF1 and AtRAV1 bind specific DNA sequence [12,13,24,25], whereas, AtVRN1 binds DNA sequence non-specifically, and At1g16640 has been reported to not bind DNA [23,26]. Comparison of the structures and sequences of AtVAL1(273e400) with other B3 domains in plants revealed several features in the former that could contribute to DNA binding (Fig. 4B). The stretch of residues in loop1 and loop2 forms a continuous surface of all proteins (Fig. 2C, D, E, F). In AtVAL1, AtARF1, and AtRAV1, this region contributes the DNAbinding (Fig. 4B, C, D). By comparison, this region of AtVRN1 (208e341) lacks the DNA binding residues of AtVAL1 for nucleotides selectivity (Fig. 4E). Lacking the residues corresponding to the loop2 of AtVAL1 B3 domain in At1g16640 may explain why it cannot bind DNA like the previous study showed (Fig. 4F) [26].
Acknowledgements
4. Discussion
References
Previous studies reported that AtVAL1/2 binds to the CME in the nucleation region at FLC locus through its B3 domain and is required for H3K27 tri-methylation at FLC during vernalization [1,2]. A point mutation in the Sph/RY motif in the CME prevents long-term cold-mediated PHD-PRC2 nucleation at FLC [1,2]. In addition, AtVAL1 was also validated to interact with PHD-PRC2 complexes in coordination with ASAP or a histone deacetylase complex to establish a repressive chromatin environment at FLC for its silencing during vernalization [12,13]. Recruiting these complexes are pivotal to take off certain functional processes. Our studies illustrate the molecular basis for the AtVAL1/2 protein B3 domain specifically recognize the Sph/RY motif, and our structure complex gives the insight that why AtVAL1/2 but not AtVAL3 have the binding ability to the RY motif. Our plant B3 domain-related protein analysis also illustrates the structural basis of the target DNA recognition difference through two important loops (loop1 and loop2), which supply structure evidence to explore more specific or non-specific cis element recognition properties by plant B3 domain-containing proteins.
[1] S. Berry, C. Dean, Environmental perception and epigenetic memory: mechanistic insight through FLC, Plant J.: For Cell and Molecular Biology 83 (2015) 133e148. [2] R. Amasino, Seasonal and developmental timing of flowering, Plant J.: For Cell and Molecular Biology 61 (2010) 1001e1013. [3] C.A. Helliwell, R.S. Anderssen, M. Robertson, E.J. Finnegan, How is FLC repression initiated by cold? Trends Plant Sci. 20 (2015) 76e82. [4] V.H. Meller, S.S. Joshi, N. Deshpande, Modulation of chromatin by noncoding RNA, Annu. Rev. Genet. 49 (2015) 673e695. [5] J.B. Heo, S. Sung, Vernalization-mediated epigenetic silencing by a long intronic noncoding RNA, Science 331 (2011) 76e79. [6] S. Pien, U. Grossniklaus, Polycomb group and trithorax group proteins in Arabidopsis, Biochim. Biophys. Acta 1769 (2007) 375e382. [7] C. Lanzuolo, V. Orlando, Memories from the polycomb group proteins, Annu. Rev. Genet. 46 (2012) 561e589. [8] I. Mozgova, C. Kohler, DNA-sequence-specific erasers of epigenetic memory, Nat. Genet. 48 (2016) 591e592. [9] S. Sung, R.M. Amasino, Vernalization in Arabidopsis thaliana is mediated by the PHD finger protein VIN3, Nature 427 (2004) 159e164. [10] F. De Lucia, P. Crevillen, A.M. Jones, T. Greb, C. Dean, A PHD-polycomb repressive complex 2 triggers the epigenetic silencing of FLC during vernalization, Proc. Natl. Acad. Sci. U.S.A. 105 (2008) 16831e16836. [11] A. Angel, J. Song, C. Dean, M. Howard, A Polycomb-based switch underlying quantitative epigenetic memory, Nature 476 (2011) 105e108. [12] J.I. Questa, J. Song, N. Geraldo, H. An, C. Dean, Arabidopsis transcriptional repressor VAL1 triggers Polycomb silencing at FLC during vernalization, Science 353 (2016) 485e488. [13] W. Yuan, X. Luo, Z. Li, W. Yang, Y. Wang, R. Liu, J. Du, Y. He, A cis cold memory element and a trans epigenome reader mediate Polycomb silencing of FLC by vernalization in Arabidopsis, Nat. Genet. 48 (2016) 1527e1534. [14] C. Li, Y. Cui, A DNA element that remembers winter, Nat. Genet. 48 (2016) 1451e1452. [15] M. Santos-Mendoza, B. Dubreucq, S. Baud, F. Parcy, M. Caboche, L. Lepiniec, Deciphering gene regulatory networks that control seed development and maturation in Arabidopsis, Plant J. : For Cell and Molecular Biology 54 (2008) 608e620. [16] H. Tsukagoshi, T. Saijo, D. Shibata, A. Morikami, K. Nakamura, Analysis of a sugar response mutant of Arabidopsis identified a novel B3 domain protein that functions as an active transcriptional repressor, Plant Physiol. 138 (2005) 675e685. [17] H. Tsukagoshi, A. Morikami, K. Nakamura, Two B3 domain transcriptional repressors prevent sugar-inducible expression of seed maturation genes in Arabidopsis seedlings, Proc. Natl. Acad. Sci. U.S.A. 104 (2007) 2543e2547. [18] M. Suzuki, H.H. Wang, D.R. McCarty, Repression of the LEAFY COTYLEDON 1/ B3 regulatory network in plant embryo development by VP1/ABSCISIC ACID INSENSITIVE 3-LIKE B3 genes, Plant Physiol. 143 (2007) 902e911. [19] W. Minor, M. Cymborowski, Z. Otwinowski, M. Chruszcz, HKL-3000: the integration of data reduction and structure solutionefrom diffraction images to an initial model in minutes, Acta crystallographica. Section D, Biol. Crystallogr. 62 (2006) 859e866. [20] T. Pape, T.R. Schneider, HKL2MAP : a graphical user interface for macromolecular phasing with SHELX programs, J. Appl. Crystallogr. 37 (2004) 843e844. [21] P. Emsley, K. Cowtan, Coot: model-building tools for molecular graphics, Acta crystallographica. Section D, Biol. Crystallogr. 60 (2004) 2126e2132. [22] G.N. Murshudov, P. Skubak, A.A. Lebedev, N.S. Pannu, R.A. Steiner, R.A. Nicholls, M.D. Winn, F. Long, A.A. Vagin, REFMAC5 for the refinement of macromolecular crystal structures, Acta crystallographica. Section D, Biol. Crystallogr. 67 (2011) 355e367.
Data availability Coordinates have been deposited in the Protein Data Bank with accession codes for 5YZY for B3 domain (273e400) in complex with 13bp-DNA(Hg-II). 5YZZ for the native B3 domain complex with 13bp-DNA and 5Z00 for B3 domain (285e400) with 15bp-DNA complex. Author contributions B.W. conducted all the crystallographic and biochemical experiments under the supervision of J.M. B.W., M.Z., S. S., and H. L. expressed, purified, and grew crystals and performed the biochemical assays. B.W., M.Z., and J.G. collected X-ray diffraction data. B.W. solved the structures. B.W. and J.M. wrote and revised the manuscript. Funding This work was supported by the National Basic Research Program of China (2011CB966304 and 2012CB910502), and by the National Natural Science Foundation of China (31230041).
We thank the staffs from BL18U1 and BL19U1 beamlines of National Facility for Protein Science in Shanghai (NFPS) at Shanghai Synchrotron Radiation Facility, for assistance during data collection. Transparency document Transparency document related to this article can be found online at https://doi.org/10.1016/j.bbrc.2018.05.002. Appendix A. Supplementary data Supplementary data related to this article can be found at https://doi.org/10.1016/j.bbrc.2018.05.002.
Please cite this article in press as: B. Wu, et al., Structural insight into the role of VAL1 B3 domain for targeting to FLC locus in Arabidopsis thaliana, Biochemical and Biophysical Research Communications (2018), https://doi.org/10.1016/j.bbrc.2018.05.002
8
B. Wu et al. / Biochemical and Biophysical Research Communications xxx (2018) 1e8
[23] G.J. King, A.H. Chanson, E.J. McCallum, M. Ohme-Takagi, K. Byriel, J.M. Hill, J.L. Martin, J.S. Mylne, The Arabidopsis B3 domain protein VERNALIZATION1 (VRN1) is involved in processes essential for development, with structural and mutational studies revealing its DNA-binding surface, J. Biol. Chem. 288 (2013) 3198e3207. [24] K. Kowalski, R. Czolij, G.F. King, M. Crossley, J.P. Mackay, The solution structure of the N-terminal zinc finger of GATA-1 reveals a specific binding face for the transcriptional co-factor FOG, J. Biomol. NMR 13 (1999) 249e262.
[25] T. Ulmasov, G. Hagen, T.J. Guilfoyle, ARF1, a transcription factor that binds to auxin response elements, Science 276 (1997) 1865e1868. [26] J.K. Waltner, F.C. Peterson, B.L. Lytle, B.F. Volkman, Structure of the B3 domain from Arabidopsis thaliana protein At1g16640, Protein Sci. 14 (2005) 2478e2483. [27] L. Holm, P. Rosenstrom, Dali server: conservation mapping in 3D, Nucleic Acids Res. 38 (2010) W545eW549.
Please cite this article in press as: B. Wu, et al., Structural insight into the role of VAL1 B3 domain for targeting to FLC locus in Arabidopsis thaliana, Biochemical and Biophysical Research Communications (2018), https://doi.org/10.1016/j.bbrc.2018.05.002
Contents lists available at ScienceDirect
Biochemical and Biophysical Research Communications journal homepage: www.elsevier.com/locate/ybbrc
Structural insight into the role of VAL1 B3 domain for targeting to FLC locus in Arabidopsis thaliana Baixing Wu a, *, 1, Mengmeng Zhang a, 1, Shichen Su a, Hehua Liu b, Jianhua Gan b, Jinbiao Ma a, ** a
State Key Laboratory of Genetic Engineering, Collaborative Innovation Centre of Genetics and Development, Department of Biochemistry, Institute of Plant Biology, School of Life Sciences, Fudan University, Shanghai 200438, China State Key Laboratory of Genetic Engineering, Collaborative Innovation Center of Genetics and Development, Department of Physiology and Biophysics, School of Life Sciences, Fudan University, Shanghai, 200438, China b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 28 April 2018 Accepted 1 May 2018 Available online xxx
Vernalization is a pivotal stage for some plants involving many epigenetic changes during cold exposure. In Arabidopsis, an essential step in vernalization for further flowering is successful silence the potent floral repressor Flowering Locus C (FLC) by repressing histone mark. AtVal1 is a multi-function protein containing five domains that participate into many recognition processes and is validated to recruit the repress histone modifier PHD-PRC2 complex and interact with components of the ASAP complex target to the FLC nucleation region through recognizing a cis element known as CME (cold memory element) by its plant-specific B3 domain. Here, we determine the crystal structure of the B3 domain in complex with Sph/RY motif in CME. Our structural analysis reveals the specific DNA recognition by B3 domain, combined with our in vitro experiments, we provide the structural insight into the important implication of AtVAL1-B3 domain in flowering process. © 2018 Elsevier Inc. All rights reserved.
Keywords: Vernalization Flowering FLC locus PRC2 B3 domain DNA binding
1. Introduction Vernalization is the main method used by some plants to cope with the environmental stimuli during cold exposure [1,2]. In Arabidopsis, a key regulating mechanism of vernalization is through silencing the potent floral repressor Flowering Locus C (FLC) [2e5]. Sets of histone modifiers and a long noncoding RNA known as COOLAIR were indicated as regulators that dynamically modulated the expression of FLC [1,2,5]. The N-terminal tails of histone proteins carry various reversible post-translational modifications. One of these modifications, H3K27me3, is validated to be a repress mark to FLC that dynamically installed and uninstalled by a series of modifiers including Polycomb group (PcG) and Trithorax group [6]. Polycomb-mediated silencing typically involves PcG complex
* Corresponding author. State Key Laboratory of Genetic Engineering, Collaborative Innovation Centre of Genetics and Development, Department of Biochemistry, Institute of Plant Biology, School of Life Sciences, Fudan University, Shanghai 200438, China. ** Corresponding author. E-mail addresses: [email protected] (B. Wu), [email protected] (J. Ma). 1 Co-first authors.
recruitment and spreading of repression mark at target loci [4,7]. An interacting partner of PcG known as Polycomb repressive complex 2 (PRC2), is pointed to catalyze the repressive H3K27me3 in animals and in plants [8]. Cold exposure will induce the formation of a PHDPRC2 complex to localize to FLC chromatin involving the expression of two plant homeodomain (PHD) protein called VERNALIZATION INSENSITIVE 3 (VIN3) and VIN3-like protein 1(VRN5) [9,10]. This PHD-PRC2 complex accumulates specifically in a region, termed the “nucleation region” that covers the first exon and part of the first intron of FLC, to catalyze the H3K27me3 during vernalization [11]. After winter, when the temperature rises, the H3K27me3 modification spreads to cover the entire FLC locus, resulting in stable FLC silencing to render the vernalized plants competent to flower [11]. Thus, the PHD-PRC2 complex targeting to the “nucleation region” of FLC is a key step for downstream functions. A cis DNA element in FLC locus containing two Sph/RY motifs (50 -TGCATG-3’; referred to hereafter as RY motifs) were recently discovered crucial for FLC repression during vernalization [12,13]. The two RY motifs containing region that essential for the maintenance of FLC silencing upon return to warm temperatures were then named cold memory element (CME) as its capability to remember the past cold [14]. Furthermore, H3K27me3 levels were
https://doi.org/10.1016/j.bbrc.2018.05.002 0006-291X/© 2018 Elsevier Inc. All rights reserved.
Please cite this article in press as: B. Wu, et al., Structural insight into the role of VAL1 B3 domain for targeting to FLC locus in Arabidopsis thaliana, Biochemical and Biophysical Research Communications (2018), https://doi.org/10.1016/j.bbrc.2018.05.002
2
B. Wu et al. / Biochemical and Biophysical Research Communications xxx (2018) 1e8
not found to increase when mutated the RY motif during vernalization, manifesting that the RY motifs enable PHD-PRC2 nucleation. Meanwhile, two B3 domain family protein AtVAL1 and AtVAL2 but not AtVAL3 were found as the transacting protein binding to the RY motifs and confirmed that the binding was mediated by its B3 domain [12,13], which was previously shown to recognize the RY motif [15]. These two homologs belong to the ABSCISIC ACID INSENSITIVE 3 (ABI3), FUSCA3 (FUS3) and LEAFY COTYLEDON 2 (LEC2) family of transcriptional repressor proteins [16]. AtVAL1/2 were shown to function with redundancy, e.g. AtVAL1 and AtVAL2 were previously shown to function together during seedling development with sugar signaling to repress ectopic expression of seed maturation genes [17]; simultaneous disruption of val1 and val2 resulted in seedlings that developed embryonic traits and accumulated seed storage proteins and oils, whereas individual val1 and val2 mutants had no apparent morphological phenotypes [17,18]. Moreover, AtVAL1/2 were also shown to function with redundancy and physically interact with the Polycomb component LIKE HETEROCHROMATIN PROTEIN 1 (LHP1) [13], and interact with components of the ASAP complex and promote histone deacetylation during vernalization [12]. AtVAL1/2 proteins seem to be sequence-specific epigenome readers that, through multivalent recognitions and interactions, can be tethered to a particular allele or locus [13]. Whereas, the structural basis of these domains of AtVAL1/2 in the regulation of gene expression is yet to be elucidated. Here, we report the crystal structure of B3 domain of AtVAL1. Our AtVAL1 B3 domain in complex with RY motif and biochemical studies unveil the specific DNA recognition by AtVAL1/2 and give a structural evidence that why AtVAL1/2 but not AtVAL3 can recognize the RY motif. We also illustrate the structural basis for different sequence selectivity among B3 domain-containing proteins in plants. 2. Materials and methods 2.1. Protein expression and purification The PCR-amplified cDNA fragments encoding the AtVAL1 B3
domain (273e400) were cloned into a modified pSUMO vector encoding 6xHis tag at N-terminus following a SUMO fusion tag and Ulp1 protease site. The plasmid containing the DNA insert of interest was transformed into Escherichia coli strain BL21 (DE3) grown in LB medium supplemented with 50 mg/ml kanamycin. The recombinant protein expression was induced by 0.2 mM IPTG at 37 C, followed by 16e18 h incubation at 18 C. The cell pellets were resuspended in buffer containing 20 mM Tris-HCl pH 8.0, 500 mM NaCl, 25 mM imidazole pH 8.0 and lysed using the high press and further clarified by centrifugation at 18,000 rpm. Supernatants were purified with nickel-chelating affinity column HisTrap (GE Healthcare), the target protein was washed with lysis buffer and then eluted with a buffer containing 20 mM Tris-HCl, pH 8.0, 500 mM NaCl and 500 mM imidazole pH 8.0. Ulp1 protease was added to remove the N-terminal tag and the fusion protein of the recombinant protein and dialyzed with lysis buffer containing 20 mM Tris pH 8.0, 500 mM NaCl 3 h. The mixture was applied to another Ni-NTA resin to remove the protease and uncleaved proteins with the buffer containing 20 mM HEPES pH7.5, 500 mM NaCl, 25 mM Imidazole pH 8.0. Eluted proteins were concentrated by centrifugal ultrafiltration, loaded onto a pre-equilibrated HiLoad € 16/60 Superdex 75-pg column in an Akta-purifier (GE Healthcare), eluted at a flow rate of 1 ml/min with the buffer containing 10 mM HEPES pH 7.5, 200 mM NaCl. Peak fractions were analyzed by SDSPAGE (15%, w/v) and stained with Coomassie Brilliant Blue R-250. Purified fractions were pooled together and concentrated by centrifugal ultrafiltration. The concentration was determined by A280. 2.2. ITC measurements For B3 domain and DNA interactions, ITC assays were carried out on a MicroCal iTC200 calorimeter (GE Healthcare) at 25 C. The buffer used for testing the binding affinity between proteins and DNA was 10 mM HEPES pH 7.5, 100 mM NaCl. The concentrations of proteins were determined spectrophotometrically. The DNA substrates were diluted in the reaction buffer. The ITC experiments involved 25 injections of protein into DNA. The sample cell was loaded with 250 ml of DNA at 50 mM and the syringe with 80 mL of
Table 1 Data collection and refinement statistics.
Data collection Space group Cell dimensions a, b, c (Å) a, b g, ( ) Wavelength(Å) Resolution (Å) Rmerge I/sI Completeness (%) Redundancy Refinement Resolution (Å) No. reflections Rwork/Rfree No. atoms Protein DNA Water/Ligands Average B-factor (Å) R.m.s. deviations Bond lengths (Å) Bond angles ( ) Ramachandran plot Favored/allowed (%)
B3 (273e400) and 13bp-DNA (Hg-II) (5YZY)
B3 (273e400) and 13bp-DNA (5YZZ)
I4122
I4122
87.780, 87.780, 148.735 90, 90, 90 0.97853 75.60e2.608 (2.69e2.60) 0.096 37.3 (2.25) 100.0(100.0) 16.3
86.896, 86.896, 148.237 90, 90, 90 0.97853 74.97e2.585 (2.67e2.58) 0.077 29.4 (2.33) 99.5(95.7) 13.9
30.00e2.61 7578 0.2103/0.2567 1394 866 527 1 37.787
30.00e2.58 7414 0.2260/0.2552 1393 866 527 0 44.338
0.013 1.70
0.013 1.89
93.58/6.42
92.66/6.42
Please cite this article in press as: B. Wu, et al., Structural insight into the role of VAL1 B3 domain for targeting to FLC locus in Arabidopsis thaliana, Biochemical and Biophysical Research Communications (2018), https://doi.org/10.1016/j.bbrc.2018.05.002
B. Wu et al. / Biochemical and Biophysical Research Communications xxx (2018) 1e8
protein at 1 mM. Curve fitting to a single binding site model was performed by the ITC data analysis module of Origin 7.0 (MicroCal) provided by the manufacturer. DGo of protein-DNA binding was computed as RTln(1/KD), where R, T, and KD are the gas constant, temperature and dissociation constant, respectively. 2.3. Oligonucleotides All oligonucleotides used in this study were synthesized from Sangon Biotech (Shanghai).
3
2.5. Data collection and structure determination The B3 domain in complex with 13-bp DNA was collected from a single crystal at 100 K at SSRF-BL18U1. HKL3000 was used for data processing [19]. The phase was determined by SAD with the anomalous signal of mercury using hkl2map [20]. The model was manually built using COOT [21] and refined using REFMAC [22]. Data collection and refinement statistics are shown in Table 1. All molecular representations were prepared using PyMOL (DeLano Scientific).
2.4. Crystallization
3. Results
The B3 domain in a buffer 10 mM HEPES pH 7.5, 200 mM NaCl was concentrated to 10 mg/ml. And then the protein and DNA were mixed with the molar ratio of 1:1.2. B3 with 13bp-DNA was crystallized in the solution containing 30% (w/v) PEG 5,000 MME, 0.1 M Tris base/Hydrochloric acid pH 8.0, 0.2 M Lithium sulfate. An Hgsoak was obtained by soaking crystals of the B3 domain-DNA complex for 3 min in the crystallization solution supplemented with 1 mM methylmercury chloride. The B3-DNA complex crystals were cryoprotected by adding the crystallization solution supplemented with 25% glycerol and flash frozen in liquid nitrogen.
3.1. AtVAL1 B3 domain specifically recognize the Sph/RY motifs in CME Previous studies identified that the AtVAL1 and AtVAL2 can specifically bind to a 47-bp FLC silencing element containing two RY motifs [12,13]. Thus, we first set out to test the in vitro binding affinity with wildtype or mutant DNA substrates using ITC experiment (Fig. 1B). As expected, the truncated B3 domain (273e400) that we designed could bind to the CME-wildtype with a high binding affinity about 8.47 mM with a molar ratio about 2:1 (Fig. 1C).
Fig. 1. Interactions between B3 domain and RY motif are sequence-specific. (A) Domain organization of AtVAL1/2. The PHD-like domain containing region is colored in green; the B3 domain is colored in sky blue; the CW domain is colored in pink; the zinc finger of unknown function is colored in red and the EAR domain is colored in yellow. (B) Schematic of the wildtype CME region, CME mutants and the sequence used for crystallization. The core RY motif is circled by a black box. The mutations are colored in red. (C-F) ITC results of B3 domain titrate into wildtype or mutated CME DNA sequence. (C) ITC results of wildtype CME and B3 domain. (D) ITC results of CME-M1 and B3 domain. (E) ITC results of CME-M2 and B3 domain. (F) ITC results of CME-M1/M2 and B3 domain. (G) ITC results of B3 domain titrate to 13bp DNA substrate. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)
Please cite this article in press as: B. Wu, et al., Structural insight into the role of VAL1 B3 domain for targeting to FLC locus in Arabidopsis thaliana, Biochemical and Biophysical Research Communications (2018), https://doi.org/10.1016/j.bbrc.2018.05.002
4
B. Wu et al. / Biochemical and Biophysical Research Communications xxx (2018) 1e8
And then, according to the studies that the transition of the cytidine to thymine in the RY motif will destroy the binding between RY motif and AtVAL1 [12], we design another three substrates containing one or two mutations to confirm these results (Fig. 1B). Our ITC results showed that mutations in either RY motifs will destroy the binding between B3 domain and the mutated RY motif, but not impact the binding affinity with both CME-M1 and CME-M2, molar ratio of which are about 1:1 (Fig. 1D and E). Simultaneously mutate the two RY motifs in CME will cause the binding between B3 domain and CME lost (Fig. 1F). These results confirmed the specific binding between B3 domain and RY motif and the molar ratio depends on the number of RY motifs. 3.2. Crystal structure of AtVAL1 B3 domain in complex with RY motif Based on these ITC results, we design another set of DNA substrates with overhang for crystallization with a high binding affinity with the B3 domain (Fig. 1G). And then, we successfully obtained the crystal structure of B3 domain (273e400) in complex with 13bp-DNA (50 -ATTCTGCATGGAT-30 , 30 -AAGACGTACCTAT-50 ) that diffracted to 2.6 Å (Table 1). This complex structure of AtVAL1 B3 domain with DNA belongs to the space group I4122. The structure was solved by single-wavelength anomalous diffraction methods and refined to high resolution with excellent geometry (Table 1). AtVAL1 B3 domain adopts a pseudobarrel fold as reported before in
other plant B3 domain-containing proteins (Fig. 2A) [23e26], which reveals a seven-stranded b-barrel capped at either edge of the barrel by a short helix (Fig. 2A). There are one B3 domain and a double strand DNA in the asymmetric unit. A positive-charge riched region faced with the DNA duplex (Fig. 2B). 3.3. Identification of the B3 domain DNA-binding region The DNA duplex is mainly recognized by two regions in the B3 domain, where we termed as loop1 (from 300 to 309 that contains a short helix a2), and a loop2 between b4 and b5 (from 349 to 354) (Fig. 2C). The two loops of AtVAL1 B3 domain insert into the major groove of the double strand DNA (Fig. 2C and D). Residues in the two loops are account for most of the specific recognition (Fig. 2D). There are appropriate contacts with the six base-pairs. In the T5: A22 pair, the methyl-group of T5 interacts with the side chain of I307 via hydrophobic contacts (Fig. 3A). In the G6: C22 pair, the O6, and N7 of G6 are recognized by Arg309 through hydrogen bonds (Fig. 3B). In the C7: G21 pair, the C7 is recognized through hydrophobic interactions with Trp349 with about 3.5 Å (Fig. 3C). The N7 of A8 in A8: T20 pair is hydrogen bonded to the side chain of Asn351 and the phosphate group hydrogen bonds to the side chain of Ser302, the methyl-group of T20 interacts with the side chain of V311 via hydrophobic contacts (Fig. 3D). Asn351 also denotes to the binding of T9 through its side chain by forming a hydrogen bond with the O4 in T9: A19 pair, the phosphate group of A19 is
Fig. 2. The overall structure of B3 domain in complex with 13bp-DNA containing RY motif. (A) The overall structure of B3 domain (colored in wheat) in complex with DNA (colored in white and black). Omit FO-FC electron density for the RNA at 2.6 A resolution, contoured level is 1.0 s. (B) Surface representation of B3-DNA complex. (C) The interacting region between B3 domain and RY motif. The contacting loops are colored in green indicated by the black arrow and the RY motif region is indicated by the black dash line. (D) The interacting loops are inserted into the major groove of target DNA. The double strand DNA is shown as spheres colored like Fig. 2A, the residues involved in the interaction with DNA was shown as stick colored in green. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)
Please cite this article in press as: B. Wu, et al., Structural insight into the role of VAL1 B3 domain for targeting to FLC locus in Arabidopsis thaliana, Biochemical and Biophysical Research Communications (2018), https://doi.org/10.1016/j.bbrc.2018.05.002
B. Wu et al. / Biochemical and Biophysical Research Communications xxx (2018) 1e8
5
Fig. 3. Specific recognition of RY motif by B3 domain. (A-F) Detailed interactions between residues of the B3 domain and DNA base-pair. (A) T5:A23; (B) G6:C22; (C) C7:G21; (D) A8:T20; (E) T9:A19; (F) G10:C18. (G) Sequence alignment between AtVALs B3 domain. The amino acids participated in recognizing the phosphate group is indicated by a green circle, recognizing the base through hydrophobic contacts and hydrogen bonds are shown as a blue and red circle, respectively. (H) Comparison of ITC results of mutations in each site of RY motif. The wildtype 13bp-DNA are colored in blue, and the mutants are colored in green, red, and purple. ND: not detectable. WB: weak binding. (I) ITC results of B3 mutants including R309D, N351S, and N352S. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)
Please cite this article in press as: B. Wu, et al., Structural insight into the role of VAL1 B3 domain for targeting to FLC locus in Arabidopsis thaliana, Biochemical and Biophysical Research Communications (2018), https://doi.org/10.1016/j.bbrc.2018.05.002
6
B. Wu et al. / Biochemical and Biophysical Research Communications xxx (2018) 1e8
recognized by Lys297 (Fig. 3E). In the G10: C18 pair, the O6 of G10 forms hydrogen bond with the side chain of Asn352, the N4 and phosphate group of C18 are hydrogen bonded to the side chain of Asn351 and the main chain of Lys314, respectively (Fig. 3F). These residues of AtVAL1 B3 domain involved in the specific recognition of the DNA play important role in binding the double strand DNA, especially three keynote residues: I307, Asn351 and Asn352, which are conserved in AtVAL1 and AtVAL2 but not in AtVAL3 (Fig. 3G). These contacts may explain why AtVAL3 cannot recognize the RY motif [13].
We also produced protein mutants R309D, N351S, and N352S to assess theirs in vitro DNA-binding ability using ITC experiment (Fig. 3I). The two mutants N351S and R309D show dramatically decrease of binding ability, and the N352S just slightly decrease the binding affinity, because the distance between the N352S and the O6 of G10 is 3.8 Å, there may exist water molecue that cannot be seen in our structure mediates this interaction (Fig. 3I). These mutations in both the DNA and the B3 domain confirm the importance of sequence specificity in the RY motif and the residues that involving the binding for protein-DNA interaction.
3.4. Mutagenesis studies of interactions between B3 domain and target DNA
3.5. Comparison of the structure of AtVAL1 B3 domain with homologous
To confirm the binding properties of these residues and nucleotides, mutations were first taken out in the RY motif 50 -TGCATG-30 except T5: A22. In each site, we mutate the sequence to another three base-pairs to confirm the sequence-specific binding (Fig. 3H). These mutants almost destroyed the binding ability to the B3 domain. Mutants of A6: T22, T6: A22 and C6: G22 all loss the binding ability with the B3 domain (Fig. 3H), the main interaction between G6 group and Arg309 is through O6, which is a lack in the other three types (Fig. 3H). Mutations in the C7: G21 has the same tendency like G6: C22 pair, although T7: A21 keeps some binding ability because the Trp349 denotes the steric hindrance to the C7 site, any mutation to purine in this site will clash with the side chain of Trp349 (Fig. 3H). The mutations of the other sites all present dramatically decrease binding ability (Fig. 3H).
There are four structures of plant B3 domain-containing proteins including two crystal structures, AtVRN1 B3 domain (PDB code: 4L1K) and AtARF1 in complex with DNA (PDB code: 4LDX), and two NMR structures, AtRAV1 (PDB code: 1WID) and At1g16640 (PDB code: 1YEL) have been reported previously [23e26]. These proteins share low sequence identity between each other, whereas they have very similar architectures except AtVRN1, which has a ahelix at its N-terminal (Fig. 4A). A DALI search when AtVAL1 B3 domain was used as a search model found that AtARF1 and AtRAV1 structures are the top hits: AtARF1 gives a Z-score of 16.3, with 32% sequence identity and an r.m.s.d. of 1.8 A, and AtRAV1 gives a Zscore of 13.3, with 38% sequence identity and an r.m.s.d. of 2.0 A [27]. AtVAL1 share more similarities with AtARF1 and AtRAV1 than AtVRN1 and At1g16640, the latter two are similar to each other.
̊ ̊
Fig. 4. Comparisons of B3 domain-containing proteins in plants (A) Superposition of all B3 domains that structures have been solved in plants. AtVAL1 is colored in wheat; AtARF1 is colored in purple-blue; AtRAV1 is colored in cherry, AtVRN1 is colored in cyan and At1g16640 is colored in deep blue. (B) Sequence alignment of B3 domain-containing proteins in Arabidopsis mentioned above. Sequence alignment of the B3 domain among AtVAL1-3, AtARF1, AtRAV1, AtVRN1 and At1g16640. Secondary structure alignment of AtVAL1 B3 domain and AtARF1 B3 domains is shown above and below the sequences, respectively. Blue rectangles indicate the residues involved in recognizing the RY motif by AtVAL1. (CF) Superposition of AtVAL1 B3 domain with structural homologs. The AtVAL1 is colored in wheat shown as cartoon and the loop responsible for interactions are colored in green. The others are colored like Fig. 4A. The RMSD between these structures are calculated by PyMOL. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)
Please cite this article in press as: B. Wu, et al., Structural insight into the role of VAL1 B3 domain for targeting to FLC locus in Arabidopsis thaliana, Biochemical and Biophysical Research Communications (2018), https://doi.org/10.1016/j.bbrc.2018.05.002
B. Wu et al. / Biochemical and Biophysical Research Communications xxx (2018) 1e8
7
These comparisons of structural similarities are reflected to some degree in the DNA binding properties of these proteins. AtVAL1, AtARF1 and AtRAV1 bind specific DNA sequence [12,13,24,25], whereas, AtVRN1 binds DNA sequence non-specifically, and At1g16640 has been reported to not bind DNA [23,26]. Comparison of the structures and sequences of AtVAL1(273e400) with other B3 domains in plants revealed several features in the former that could contribute to DNA binding (Fig. 4B). The stretch of residues in loop1 and loop2 forms a continuous surface of all proteins (Fig. 2C, D, E, F). In AtVAL1, AtARF1, and AtRAV1, this region contributes the DNAbinding (Fig. 4B, C, D). By comparison, this region of AtVRN1 (208e341) lacks the DNA binding residues of AtVAL1 for nucleotides selectivity (Fig. 4E). Lacking the residues corresponding to the loop2 of AtVAL1 B3 domain in At1g16640 may explain why it cannot bind DNA like the previous study showed (Fig. 4F) [26].
Acknowledgements
4. Discussion
References
Previous studies reported that AtVAL1/2 binds to the CME in the nucleation region at FLC locus through its B3 domain and is required for H3K27 tri-methylation at FLC during vernalization [1,2]. A point mutation in the Sph/RY motif in the CME prevents long-term cold-mediated PHD-PRC2 nucleation at FLC [1,2]. In addition, AtVAL1 was also validated to interact with PHD-PRC2 complexes in coordination with ASAP or a histone deacetylase complex to establish a repressive chromatin environment at FLC for its silencing during vernalization [12,13]. Recruiting these complexes are pivotal to take off certain functional processes. Our studies illustrate the molecular basis for the AtVAL1/2 protein B3 domain specifically recognize the Sph/RY motif, and our structure complex gives the insight that why AtVAL1/2 but not AtVAL3 have the binding ability to the RY motif. Our plant B3 domain-related protein analysis also illustrates the structural basis of the target DNA recognition difference through two important loops (loop1 and loop2), which supply structure evidence to explore more specific or non-specific cis element recognition properties by plant B3 domain-containing proteins.
[1] S. Berry, C. Dean, Environmental perception and epigenetic memory: mechanistic insight through FLC, Plant J.: For Cell and Molecular Biology 83 (2015) 133e148. [2] R. Amasino, Seasonal and developmental timing of flowering, Plant J.: For Cell and Molecular Biology 61 (2010) 1001e1013. [3] C.A. Helliwell, R.S. Anderssen, M. Robertson, E.J. Finnegan, How is FLC repression initiated by cold? Trends Plant Sci. 20 (2015) 76e82. [4] V.H. Meller, S.S. Joshi, N. Deshpande, Modulation of chromatin by noncoding RNA, Annu. Rev. Genet. 49 (2015) 673e695. [5] J.B. Heo, S. Sung, Vernalization-mediated epigenetic silencing by a long intronic noncoding RNA, Science 331 (2011) 76e79. [6] S. Pien, U. Grossniklaus, Polycomb group and trithorax group proteins in Arabidopsis, Biochim. Biophys. Acta 1769 (2007) 375e382. [7] C. Lanzuolo, V. Orlando, Memories from the polycomb group proteins, Annu. Rev. Genet. 46 (2012) 561e589. [8] I. Mozgova, C. Kohler, DNA-sequence-specific erasers of epigenetic memory, Nat. Genet. 48 (2016) 591e592. [9] S. Sung, R.M. Amasino, Vernalization in Arabidopsis thaliana is mediated by the PHD finger protein VIN3, Nature 427 (2004) 159e164. [10] F. De Lucia, P. Crevillen, A.M. Jones, T. Greb, C. Dean, A PHD-polycomb repressive complex 2 triggers the epigenetic silencing of FLC during vernalization, Proc. Natl. Acad. Sci. U.S.A. 105 (2008) 16831e16836. [11] A. Angel, J. Song, C. Dean, M. Howard, A Polycomb-based switch underlying quantitative epigenetic memory, Nature 476 (2011) 105e108. [12] J.I. Questa, J. Song, N. Geraldo, H. An, C. Dean, Arabidopsis transcriptional repressor VAL1 triggers Polycomb silencing at FLC during vernalization, Science 353 (2016) 485e488. [13] W. Yuan, X. Luo, Z. Li, W. Yang, Y. Wang, R. Liu, J. Du, Y. He, A cis cold memory element and a trans epigenome reader mediate Polycomb silencing of FLC by vernalization in Arabidopsis, Nat. Genet. 48 (2016) 1527e1534. [14] C. Li, Y. Cui, A DNA element that remembers winter, Nat. Genet. 48 (2016) 1451e1452. [15] M. Santos-Mendoza, B. Dubreucq, S. Baud, F. Parcy, M. Caboche, L. Lepiniec, Deciphering gene regulatory networks that control seed development and maturation in Arabidopsis, Plant J. : For Cell and Molecular Biology 54 (2008) 608e620. [16] H. Tsukagoshi, T. Saijo, D. Shibata, A. Morikami, K. Nakamura, Analysis of a sugar response mutant of Arabidopsis identified a novel B3 domain protein that functions as an active transcriptional repressor, Plant Physiol. 138 (2005) 675e685. [17] H. Tsukagoshi, A. Morikami, K. Nakamura, Two B3 domain transcriptional repressors prevent sugar-inducible expression of seed maturation genes in Arabidopsis seedlings, Proc. Natl. Acad. Sci. U.S.A. 104 (2007) 2543e2547. [18] M. Suzuki, H.H. Wang, D.R. McCarty, Repression of the LEAFY COTYLEDON 1/ B3 regulatory network in plant embryo development by VP1/ABSCISIC ACID INSENSITIVE 3-LIKE B3 genes, Plant Physiol. 143 (2007) 902e911. [19] W. Minor, M. Cymborowski, Z. Otwinowski, M. Chruszcz, HKL-3000: the integration of data reduction and structure solutionefrom diffraction images to an initial model in minutes, Acta crystallographica. Section D, Biol. Crystallogr. 62 (2006) 859e866. [20] T. Pape, T.R. Schneider, HKL2MAP : a graphical user interface for macromolecular phasing with SHELX programs, J. Appl. Crystallogr. 37 (2004) 843e844. [21] P. Emsley, K. Cowtan, Coot: model-building tools for molecular graphics, Acta crystallographica. Section D, Biol. Crystallogr. 60 (2004) 2126e2132. [22] G.N. Murshudov, P. Skubak, A.A. Lebedev, N.S. Pannu, R.A. Steiner, R.A. Nicholls, M.D. Winn, F. Long, A.A. Vagin, REFMAC5 for the refinement of macromolecular crystal structures, Acta crystallographica. Section D, Biol. Crystallogr. 67 (2011) 355e367.
Data availability Coordinates have been deposited in the Protein Data Bank with accession codes for 5YZY for B3 domain (273e400) in complex with 13bp-DNA(Hg-II). 5YZZ for the native B3 domain complex with 13bp-DNA and 5Z00 for B3 domain (285e400) with 15bp-DNA complex. Author contributions B.W. conducted all the crystallographic and biochemical experiments under the supervision of J.M. B.W., M.Z., S. S., and H. L. expressed, purified, and grew crystals and performed the biochemical assays. B.W., M.Z., and J.G. collected X-ray diffraction data. B.W. solved the structures. B.W. and J.M. wrote and revised the manuscript. Funding This work was supported by the National Basic Research Program of China (2011CB966304 and 2012CB910502), and by the National Natural Science Foundation of China (31230041).
We thank the staffs from BL18U1 and BL19U1 beamlines of National Facility for Protein Science in Shanghai (NFPS) at Shanghai Synchrotron Radiation Facility, for assistance during data collection. Transparency document Transparency document related to this article can be found online at https://doi.org/10.1016/j.bbrc.2018.05.002. Appendix A. Supplementary data Supplementary data related to this article can be found at https://doi.org/10.1016/j.bbrc.2018.05.002.
Please cite this article in press as: B. Wu, et al., Structural insight into the role of VAL1 B3 domain for targeting to FLC locus in Arabidopsis thaliana, Biochemical and Biophysical Research Communications (2018), https://doi.org/10.1016/j.bbrc.2018.05.002
8
B. Wu et al. / Biochemical and Biophysical Research Communications xxx (2018) 1e8
[23] G.J. King, A.H. Chanson, E.J. McCallum, M. Ohme-Takagi, K. Byriel, J.M. Hill, J.L. Martin, J.S. Mylne, The Arabidopsis B3 domain protein VERNALIZATION1 (VRN1) is involved in processes essential for development, with structural and mutational studies revealing its DNA-binding surface, J. Biol. Chem. 288 (2013) 3198e3207. [24] K. Kowalski, R. Czolij, G.F. King, M. Crossley, J.P. Mackay, The solution structure of the N-terminal zinc finger of GATA-1 reveals a specific binding face for the transcriptional co-factor FOG, J. Biomol. NMR 13 (1999) 249e262.
[25] T. Ulmasov, G. Hagen, T.J. Guilfoyle, ARF1, a transcription factor that binds to auxin response elements, Science 276 (1997) 1865e1868. [26] J.K. Waltner, F.C. Peterson, B.L. Lytle, B.F. Volkman, Structure of the B3 domain from Arabidopsis thaliana protein At1g16640, Protein Sci. 14 (2005) 2478e2483. [27] L. Holm, P. Rosenstrom, Dali server: conservation mapping in 3D, Nucleic Acids Res. 38 (2010) W545eW549.
Please cite this article in press as: B. Wu, et al., Structural insight into the role of VAL1 B3 domain for targeting to FLC locus in Arabidopsis thaliana, Biochemical and Biophysical Research Communications (2018), https://doi.org/10.1016/j.bbrc.2018.05.002