Solution structure of the Set2–Rpb1 interacting domain of human Set2 and its interaction with the hyperphosphorylated C-terminal domain of Rpb1

Li et al. 10.1073/pnas.0506350102.

Supporting Information

Files in this Data Supplement:

Supporting Figure 5
Supporting Text
Supporting Table 3
Supporting Figure 6
Supporting Figure 7

Supporting Figure 5

Fig. 5. The hSRI domain binds PCTD directly and interacts specifically with doubly phosphorylated CTD repeats. (a) Domain composition and organization of human and yeast Set2 proteins; human HYPB, hSet2. (b) Thrombin cleavage and far-Western analysis of a GST-HYPB (1884–2061) fusion protein. The fusion protein is schematically represented above Upper, with sites cleaved by thrombin indicated by downward-pointing arrows (strong and weak thrombin sites are indicated by solid and dotted arrows, respectively). The fusion protein was digested with thrombin for varying amounts of time (0 min to 6 days), and the cleavage products were electrophoresed, transferred to nitrocellulose, and analyzed by Ponceau S staining (Upper) followed by far-Western blotting with the PCTD (Lower). Digestion times are indicated above the respective lanes; molecular-mass markers are indicated on the left, and the arrows to the right of the blots indicate where the intact fusion protein and its thrombin cleavage products migrate (the various fragments are both labeled and schematically represented). (c) Biacore sensorgrams illustrate the binding in real time of the hSRI domain to three-repeat CTD peptides phosphorylated on either Ser-2 of each repeat (2,2,2), Ser-5 of each repeat (5,5,5), or Ser(2 + 5) of each repeat (2,5,2,5,2,5) (control surface = 6PC; Table 1 and Interaction (Biacore) assays, above). (d) Equilibrium binding (Response Units at equilibrium, RU_eq) as a function of concentration of hSRI domain was used to determine the apparent K_D for the 2,5,2,5,2,5 peptide (see Interaction (Biacore) Assays in Supporting Text).

Supporting Figure 6

Fig. 6. The hSRI domain binds preferentially to CTD repeats with four contiguous phosphates. Equilibrium binding (RU_eq) as a function of concentration of hSRI domain was used to determine the apparent K_D values for the 2,5,2,5 peptide (a), the 2,5,2 peptide (b), and the 5,2,5 peptide (c) (see Interaction (Biacore) Assays in Supporting Text).

Supporting Figure 7

Fig. 7. Point mutations that decrease the binding affinity of the hSRI domain toward CTD phosphopeptides. Equilibrium binding curves of wild-type hSRI domain and five single-point mutations that diminish the affinity of the hSRI domain toward the 2,5,2 peptide (a) and the 5,2,5 peptide (b). Equilibrium binding (RU_eq) as a function of concentration of wild-type and mutant proteins was used to determine the apparent K_D values for the 2,5,2 and the 5,2,5 peptides (see Interaction (Biacore) Assays in Supporting Text).

Table 3. Structural statistics for the hSRI domain (20 structures)

NOE distance restraints	2,600
Intraresidue	610
Sequential (\|i - j\| =1)	558
Medium-range (\|i – j\| ≤ 4)	689
i, i + 2	187
i, i + 3	324
i, i + 4	178
Long-range (\|i - j\| ≥ 5)	663
Hydrogen bonds*	80
Dihedral angle constraints^†	178
Residual dipolar couplings (¹D_NH)	61
Residual dipolar couplings (¹D_CH)	59
Dipolar coupling R factor of ¹D_NH,^‡ %	6.7 ± 0.3
Dipolar coupling R factor of ¹D_CH,^‡ %	9.5 ± 0.4
Ramachandran plot^§
Favored region (98%)	98.6
Allowed region (>99.8%)	100.0
Deviations from idealized geometry
Bonds, Å	0.0136 ± 0.0003
Angles, ˚	1.6408 ± 0.0497
Impropers, ˚	1.5420 ± 0.0741
Mean pairwise rmsd
Backbone (residues 10-109), Å	0.43
Heavy atoms (residues 10-109), Å	1.29

None of these structures exhibit distance violations of >0.4 Å or dihedral angle violations of >4°.

*Two constraints per hydrogen bond (d_HN-O ≤ 2.0 Å and d_N-O ≤ 3.0 Å) are implemented for amide protons protected from solvent exchange.

^†Dihedral angle constraints were generated by TALOS based on backbone atom chemical shifts and by the ³J_HNHa couplings (1, 2).

^‡R factor for residual dipolar coupling is the ratio of the rms deviation (rmsd) between observed and calculated values to the expected rmsd if the vectors were randomly distributed (3).

^§MOLPROBITY was used to assess the quality of the structures (4, 5).

1. Wang, A. C. & Bax, A. (1996) J. Am. Chem. Soc. 118, 2483–2494.

2. Cornilescu, G., Delaglio, F. & Bax, A. (1999) J. Biomol. NMR 13, 289–302.

3. Clore, G. M. & Garrett, D. S. (1999) J. Am. Chem. Soc> 121, 9008–9012.

4. Davis, I. W., Murray, L. W., Richardson, J. S. & Richardson, D. C. (2004) Nucleic Acids Res. 32, W615–W619.

5. Lovell, S. C., Davis, I. W., Arendall, W. B., 3rd, de Bakker, P. I., Word, J. M., Prisant, M. G., Richardson, J. S. & Richardson, D. C. (2003) Proteins 50, 437–450.

Supporting Text

DNA Constructs, Purification of Recombinant Proteins, and Initial Studies

A GST-fusion construct containing hSet2/HYPB (1884–2061) with a WW domain and a putative SRI domain was transformed into BL21(DE3)STAR cells (Invitrogen). The fusion protein was overexpressed in LB medium and purified by using glutathione resin following standard procedures (Sigma–Aldrich). Purified GST-fusion protein was subjected to thrombin cleavage in a buffer containing 25 mM sodium phosphate, 100 mM KCl, and 0.1% 2-mercaptoethanol at pH 7.0 to produce the target protein. Overdigestion, however, produced two smaller fragments of hSet2, 1948–2061 and 1954–2061 as determined by mass spectrometry (QSTAR XL quadrupole time-of-flight tandem mass spectrometer equipped with an electrospray source, ABI/MDSSciex, Toronto), corresponding to the proposed SRI domain. The shorter fragment, 1954–2061 of hSet2/HYPB, was subsequently cloned into the pET15b vector (EMD Biosciences, Madison, WI) and overexpressed as an N-terminal His-6-tagged protein using BL21(DE3)STAR cells. The fusion protein was purified by a Ni-NTA column and exchanged into the thrombin cleavage buffer. The N-terminal His-6-tag was then removed by thrombin cleavage; the resulting fragment contains GSHM at the N terminus and residues 1954–2061 of hSet2. This fragment was renumbered 1-112 and was used for NMR studies. After the thrombin cleavage, the hSRI domain was further purified by size-exclusion chromatography. Far-Western blotting was performed as described in ref. 1.

NMR

Sequential assignments were determined from HNCA, HN(CO)CA, HN(CA)CB, HN(COCA)CB, and HNCO experiments (2), along with ¹H-¹⁵N HSQC of selectively ¹⁵N-lysine-labeled samples, using the PACES program (3), and confirmed by manual analysis. Side-chain resonances were assigned by using HA(CA)NH, HA(CACO)NH, HCCH-TOCSY, and HC(CCO)NH-TOCSY experiments (4, 5). Aromatic resonances were assigned based on 2D-homonuclear NOESY, 2D-homonuclear TOCSY, and ¹³C-aromatic NOESYHSQC experiments (4). Stereo-specific assignment of valine and leucine methyl groups was obtained by means of a high-resolution ¹H-¹³C HSQC spectrum of a 10% ¹³C-labeled hSRI domain (6). Distance constraints were derived from 3D ¹⁵N-NOESY-HSQC, 3D ¹³C-NOESY-HSQC, and homonuclear 2D NOESY with mixing times of 60, 80, and 50 ms, respectively. Dihedral angle restraints were derived from ³J_HNHa couplings and from TALOS analysis of the chemical shift information (7, 8). Two constraints per hydrogen bond (d_HN-O £ 2.0 Å and d_N-O £ 3.0 Å) were added for amide protons protected from solvent exchange. Initial structures were calculated by using DYANA starting from random conformations (9). After the global fold was obtained, automated data analysis by CYANA2.0 was used to obtained additional constraints (10). These structures were further refined against residual dipolar couplings using a water refinement protocol adapted for X-PLOR-NIH (11, 12).

GI Accession Numbers of SRI Domains Used in Sequence Alignment (Fig. 1c)

The SRI domains of Set2-like proteins from Pan troglodytes (gi55620141) and Rattus norvegicus (gi62654363) are identical to the hSRI domain in primary sequence and are not included in the alignment. The listed species include Homo sapiens (Hs, gi30410779), Gallus gallus (Gg, gi50732165), Apis mellifera (Am, gi66551118), Anopheles gambiae str. PEST (Ang, gi58394584), Yarrowia lipolytica CLIB99 (Yl, gi49649966), Ashbya gossypii ATCC 10895 (Asg, gi44983421), Debaryomyces hansenii CBS767 (Dh, gi49656436), Kluyveromyces lactis NRRL Y-1140 (Kl, gi49640425), Cryptococcus neoformans var. neoformans B-3501A (Cn, gi50257086), Candida glabrata CBS138 (Cg, gi49524498), Ustilago maydis 521 (Um, gi46098305), Drosophila melanogaster (Dm, gi24641786), Saccharomyces cerevisiae (Sc, gi6322293), Schizosaccharomyces pombe (Sp, gi2408044), Candida albicans (Ca, gi46435962), Aspergillus fumigatus Af293 (Af, gi66851668), and Magnaporthe grisea (Mg, gi38110264). Sequence alignment was generated by MAFFT (Version 5.667) using default parameters (13).

Interaction (Biacore) Assays

General. Biotinylated synthetic CTD phosphopeptides (Table 1) were immobilized on the surface of a streptavidin-coated SA sensor chip, and purified hSRI domain was interacted with the peptides. Response data from a "blank" injection of buffer were subtracted before obtaining the sensorgrams for each individual flow cell. From these sensorgrams, the response data from the control surface (6PC or NP peptide; Table 1) were in turn subtracted to obtain the specific response for each CTD phosphopeptide surface. After interaction, dissociation was allowed to proceed for 300 s, and the flow cell surface was regenerated between injections with 20 ml of 1 M MgCl₂ at a flow rate of 100 ml/min. By using wild-type or mutant hSRI domains, apparent K_D values were determined for a number of phosphopeptides: Equilibrium binding responses (RU_eq) for serial dilutions of each hSRI protein were plotted as a function of protein concentration (C) and the data were fitted (using KALEIDAGRAPH) to a single-site binding isotherm, RU_eq = R_max/((K_D/C) + 1), where R_max = maximum surface-binding capacity, to extract the apparent K_D values.

Experiment-Specific. For Fig. 5 c and d, approximately 500 Response Units (RU) of peptide were immobilized as follows: the 6PC peptide (control) on Flow Cell (FC) 1, the 2,2,2 peptide on FC 2, the 5,5,5 peptide on FC 3, and the 2,5,2,5,2,5 peptide on FC 4 (Fig. 5). The hSRI domain (in 25 mM sodium phosphate buffer at pH 7.0, with 100 mM KCl and 0.1% 2-mercaptoethanol) was interacted with the peptides at concentrations ranging from 0.1 to 200 mM in 8 steps (25 ml of each concentration was injected at a flow-rate of 10 ml/min.; regeneration were as described above). Response curves at one protein concentration (10 mM hSRI domain) are illustrated in Fig. 5c. Response curves at a series of protein concentrations (Fig. 5d) were used to obtain apparent K_D values as described above.

Sensorgrams in Fig. 3a were obtained when a 2 mM solution of the hSRI domain was interacted with different phosphoCTD peptides; »150 RU of each peptide had been immobilized on the sensor chip (control = NP) (Fig. 3a). The protein (100 ml) was injected at a flow rate of 50 ml/min; then, the surface was regenerated.

Binding of wild-type and mutant hSRI constructs to CTD phosphopeptides was compared by using new sensor chips with »30 RU of the NP peptide in FC 1 and the 2,5,2,5 peptide in FC 4, and »100 RU of the 2,5,2 peptide in FC 2 and the 5,2,5 peptide in FC 3 (Figs. 3b, 6, and 7 and Table 2). One hundred microliters of twofold serial dilutions of the various proteins, from 6 to 0.2 mM for wild-type protein (Fig. 6) and from 48 to 0.2 mM for mutant proteins (Fig. 7), in a buffer containing 25 mM sodium phosphate (pH 7.0), 100 mM KCl, and 1 mM DTT, were titrated over this surface at a flow rate of 50 ml/min. Dissociation was allowed to proceed for 120 s, the surface was regenerated as above, and there was a 2-min wait before the next cycle. Response curves were corrected (using NP for control), and apparent binding affinities (K_D) were derived as described above.

1. Phatnani, H. P., Jones, J. C. & Greenleaf, A. L. (2004) Biochemistry 43, 15702–15719.

2. Yamazaki, T., Lee, W., Arrowsmith, C. H., Muhandiram, D. R. & Kay, L. E. (1994) J. Am. Chem. Soc. 116, 11655–1666.

3. Coggins, B. E. & Zhou, P. (2003) J. Biomol. NMR 26, 93–111.

4. Clore, G. M. & Gronenborn, A. M. (1998) Trends Biotechnol. 16, 22–34.

5. Lin, Y. & Wagner, G. (1999) J. Biomol. NMR 15, 227–239.

6. Szyperski, T., Neri, D., Leiting, B., Otting, G. & Wüthrich, K. (1992) J. Biomol. NMR 2, 323–334.

9. Güntert, P., Mumenthaler, C. & Wüthrich, K. (1997) J. Mol. Biol. 273, 283–298.

10. Herrmann, T., Güntert, P. & Wüthrich, K. (2002) J. Mol. Biol. 319, 209–227.

11. Schwieters, C. D., Kuszewski, J. J., Tjandra, N. & Clore, G. M. (2003) J. Magn. Reson. 160, 65–73.

12. Nabuurs, S. B., Nederveen, A. J., Vranken, W., Doreleijers, J. F., Bonvin, A. M., Vuister, G. W., Vriend, G. & Spronk, C. A. (2004) Proteins 55, 483–486.

13. Katoh, K., Kuma, K., Toh, H. & Miyata, T. (2005) Nucleic Acids Res. 33, 511–518.