Intrinsic Differences in Backbone Dynamics between Wild Type and

Subscriber access provided by JAMES COOK UNIVERSITY LIBRARY

Article

Intrinsic differences in backbone dynamics between wild type and DNAcontact mutants of p53 DNA binding domain revealed by NMR spectroscopy Juhi Augusta Rasquinha, Aritra Bej, Shraboni Dutta, and Sujoy Mukherjee Biochemistry, Just Accepted Manuscript • DOI: 10.1021/acs.biochem.7b00514 • Publication Date (Web): 24 Aug 2017 Downloaded from http://pubs.acs.org on August 26, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Biochemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

Intrinsic differences in backbone dynamics between wild type and DNA-contact mutants of p53 DNA binding domain revealed by NMR spectroscopy Juhi A. Rasquinha‡, Aritra Bej‡, Shraboni Dutta and Sujoy Mukherjee* Structural Biology and Bioinformatics Division, CSIR-Indian Institute of Chemical Biology, West Bengal, Kolkata 700032, India

*To whom correspondence should be addressed. E-mail: [email protected], [email protected]. ‡These authors have contributed equally. Keywords. p53 DNA binding domain, NMR spectroscopy, protein dynamics, cancer hotspot mutation.

ACS Paragon Plus Environment

1

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 32

ABSTRACT Mutations in p53’s DNA binding domain (p53DBD) are associated with 50% of all cancers, making it an essential system to investigate in order to understand the genesis and progression of cancer. In this work, we studied the changes in structure and dynamics of wild type p53DBD in comparison with two of its “hotspot” DNA contact mutants, R248Q and R273H, by analysis of backbone amide chemical shift perturbations and 15N spin relaxation measurements. The results of amide chemical shift changes indicated significantly more perturbations in the R273H mutant in comparison to the wild type and R248Q p53DBD. Analysis of

15

N spin relaxation rates and

the resulting NMR order parameters suggest that for most parts, the R248Q mutant exhibits limited conformational flexibility and is similar to the wild type protein. In contrast, R273H showed significant backbone dynamics extending up to its β-sandwich scaffold in addition to motions along the DNA binding interface. Furthermore, comparison of rotational correlation times between the mutants suggest that R273H mutant, with a higher correlation time, forms an enlarged structural fold in comparison to the R248Q mutant and wild type p53DBD. Finally, we identify three regions in these proteins that show conformational flexibility to varying degrees, which suggests that R273H mutant, in addition to being a DNA-contact mutation, exhibits properties of a conformational mutant as well.

INTRODUCTION The tumor suppressor p53 is an important transcription factor that is induced under physiological conditions leading to the further activation of various downstream genes that result in cell cycle arrest and apoptosis. Consequently, p53 plays a pivotal role in preventing the potential transformation of healthy cells into cancer cells (1). Mutations in the TP53 gene that encodes wild type p53, is the most common genetic alteration and is involved in approximately half of all


2

Page 3 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

incidences of human cancers (2). Among the three distinct domains of p53, most of these mutations are located in the DNA-binding domain comprising residues 92-312 of the wild type p53 (wt-p53DBD), suggesting the importance of this domain in cancers. While most of these mutations result in the loss of p53’s tumor suppressive functions, it is increasingly being found that many of these mutant p53 gain additional functions (gain-of-function) that promote tumorigenesis. Of the many possible gain-of-function mutants, six residues have been identified as “hotspots” for mutations that are most frequently mutated in cancers (3). Among the various mutations in these residues, the R248Q and R273H mutations hold one of the highest rate of frequency making them important mutants to investigate (4, 5). Moreover, most of the TP53 mutations can also be classified into two main categories and referred to as “structural” or “conformational” mutations and “DNA-contact” mutations. The former group includes mutations that cause changes in the local or global conformation of the protein, which thermodynamically destabilize p53, culminating in its inactivation (6-8). However, the latter group consists of mutations within residues that form direct contacts with the DNA and these mutations abrogate p53 function by altering its DNA binding ability without significantly changing the overall structure of the protein (9). DNA-contact mutations account for nearly 20% of the total reported cases of cancer mutations, of which mutations in residues R248 and R273 have the highest incidence (9). For example, R248 mutations are highly prevalent in breast, colon and skin cancers, mutations in R273 have shown the highest incidence in lung, cervical and ovarian cancers (10-15). While the DNA-contact mutations explain the loss-of-function of p53 in a straightforward manner, there is no clear explanation of the molecular mechanism for the gainof-functions in p53 mutants. Structural elucidation of mechanism have been stymied by lack of structures of important, single point mutants of the full length or p53DBD, although structures of


3

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 32

mutants are available for a thermostabilized quadruple mutant of p53DBD (16). Of the two aforementioned DNA-contact mutants, the structure of R248Q is not known, whereas the R273H structure has been solved (17). However, its comparison with wt-p53DBD did not yield a clear understanding on how mutant p53 gains new function. Indeed, superimposition of the R273H (and R273C) mutant with the wt-p53DBD revealed that their three dimensional structures were similar with minor local changes, demonstrating that the DNA-contact mutations induced negligible changes in the structure of the p53DBD (17). On the other hand, loss of DNA contacts could impart new conformational flexibility to the protein and may help explain p53’s gain of new functions. We performed solution NMR spectroscopy to investigate the role of protein dynamics, which, in addition to providing residue specific details of a protein’s conformational flexibility, helps overcome constraints induced on dynamic structures by crystal packing (18, 19). We measured the spin relaxation rates of the wild type and two DNA-contact mutants (R248Q and R273H) and obtained quantitative order parameters that are considered to be accurate reporters of protein conformational flexibility in the fast timescale regime spanning picoseconds (ps) to nanoseconds (ns). We showed the existence of local flexibility in three distinct regions for all the three proteins but to varied degrees. These results not only reveal important distinction between the wt-p53DBD and the two mutations, we also report relevant differences between the mutants themselves.

EXPERIMENTAL PROCEDURES Cloning and site-directed mutagenesis of R248Q and R273H mutants of p53 DNA binding domain. The plasmid encoding human wt-p53DBD, comprising of residues 92-312 and cloned in a pET11b vector was procured from BioBharati LifeScience Pvt. Ltd., India. Prior to inserting


4

Page 5 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

the DNA-contact mutations, the wt-p53DBD fragment was amplified by PCR along with the insertion of NdeI and BamHI restriction sites into the amplified gene products. Subsequently, the mutation was inserted by overlapping extension PCR into the wild type gene sequence using primers containing the appropriate point mutation, i.e. R248Q and R273H. Finally, the purified fragment was ligated into the pET11b vector. Fidelity of the DNA sequence of wt-p53DBD and both the mutants was verified by DNA sequencing (GCC Biotech, India). Expression and purification of wt-p53DBD and its mutants. In order to over-express the wild type and mutant p53DBD proteins, E. coli Rosetta cells were transformed with the appropriate plasmid by electroporation. The cells were grown in LB medium, 180 rpm, 37 ºC, supplemented with 100 µg/ml carbenicillin and induced with 1 mM IPTG (final concentration). In order to increase the over-expression of the wt-p53DBD in the soluble fraction of cell lysate, the cell density at the time of induction, concentration of IPTG, post-induction growth time and temperatures were thoroughly optimized. Cells were harvested by centrifuging at 6,000 × g, 4 ºC. Cells from 1L culture was lysed by resuspending the pellet in 15 ml of equilibration buffer containing 20 mM potassium phosphate (pH 6.8), 50 mM KCl and 5 mM DTT. After sonicating on ice for 14 min (8 s ON, 22 s OFF), the lysate was centrifuged at 91,000 × g, 4 °C for 1 h. The lysate obtained thereafter was loaded onto a HiTrap SP HP cation exchange column (GE Healthcare) that was pre-equilibrated with equilibration buffer and eluted by a KCl gradient up to 500 mM. Purity and integrity of the eluted protein fractions were ascertained by SDS-PAGE and MALDI-ToF, respectively. Bradford assay was performed to determine the concentration of the purified protein. In order to incorporate 13C and 15N NMR isotopes, the proteins were over-expressed in modified M9 minimal medium (20) containing 1 g/L 15NH4Cl and 4 g/L 13C-glucose (Cambridge Isotopes


5

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 32

Laboratories) as the sole carbon and nitrogen sources, respectively. Perdeuterated (i.e. 2H, 15

13

C,

N labeled) cultures were prepared by growing the cells in the aforementioned media using 99

% D2O (Cambridge Isotopes Laboratories) instead of H2O. NMR sample preparation. All NMR samples were prepared in 50 mM potassium phosphate (pH 6.8), 150 mM KCl and 5 mM DTT containing 5 % (v/v) D2O and 0.02 % (w/v) NaN3 and packed into Shigemi NMR tubes (Shigemi Inc). The following isotopically labeled samples were prepared: 15N (320 µM), 13C,15N (320 µM) and 2H, 13C, 15N (290 µM) wt-p53DBD, 15N-R248Q (320 µM) and

13

C,15N-R273H (320 µM) p53DBD. Unlabeled samples for DOSY experiments

were packed into regular 5 mm NMR tubes. NMR spectroscopy. NMR experiments to assign the p53DBD proteins were performed on a Bruker Avance 600 MHz (14.1 T field strength) NMR spectrometer using either a room temperature (TXI) or cryogenic (TCI) 5 mm triple resonance probe, both of which were equipped with z-axis gradients. The 15N-1H HSQC spectrum for wt-p53DBD was assigned using a suite of triple resonance experiments, including HNCA, HN(CO)CA and HN(CA)CB at 20 ºC. Wherever significant sensitivity enhancements were observed, TROSY (21) optimized pulse sequences were used instead of decoupled sequences. For R273H mutant, a set of HNCA and HN(CO)CA experiments were recorded at 18 ºC to complete unambiguous assignments. Residue specific backbone 15N longitudinal (R1) and transverse (R2) relaxation rates and {1H}-15N steadystate heteronuclear NOE were acquired at 18 ˚C on Bruker Avance III 700 MHz (16.47 T field) spectrometer using a 5 mm QCI cryogenic probe equipped with z-axis gradient using the pulse scheme of Kay and co-workers (22). Suitable modifications were incorporated into the pulse sequences, including the usage of interleaved sequences and heat compensation pulses, as described previously (23). For R1 and R2 rate measurements,


15

N-1H HSQC spectra were

6

Page 7 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

collected with relaxation delays of 5, 65, 145, 245, 365, 525, 750, 1200 ms and 8.5, 17, 25.5, 34, 51, 68, 85, 102 ms, respectively. For {1H}-15N NOE experiments, the reference spectrum was acquired with a 5 s relaxation delay whereas the NOE spectrum was recorded with a 1 s relaxation delay followed by a 4 s proton presaturation. DOSY spectra were recorded using Bruker’s stebpgp1s19 pulse sequence with gradient pulse length (δ/2) and diffusion time (∆) of 2.2 ms and 200 ms, respectively. A set of 32 1H spectra were recorded by linearly incrementing gradient strength from 0.67 G/cm to 32.03 G/cm.

NMR data processing and analysis. All NMR data were processed using NMRPipe (24) and Sparky was used to visualize and assign the resonances (25). Relaxation spectra were processed in a manner described previously (26) to obtain the R1, R2 spin relaxation rates and {1H}-15N NOE values. Errors in R1, R2 and NOE were calculated from the root-mean-square spectral noise obtained from NMRPipe. NMR order parameters (S2) were subsequently obtained as described previously (23). Briefly, an estimate of rotational diffusion tensor (27) was obtained from the R2/R1 ratio and an appropriately modified wt-p53DBD structure (PDB ID 2FEJ) (28) was used as inputs. This estimated diffusion tensor along with the R1, R2 and NOE values were used to obtain the NMR order parameters (29, 30) using Modelfree program (31). The model selection procedures of Modelfree were iterated using the FASTModelfree (32) code using a 1H-15N bond distance and

15

N CSA of 1.015 Å and -179 ppm, respectively (33, 34). Residue specific

conformational entropy (Sconf) was derived from S2 for each NH bond vector using (35, 36): = (3 − 1 + 8 )


7

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 32

where kB is Boltzmann’s constant. The uncertainty of Sconf was obtained from the procedure described previously by Lee and co-workers (37). DOSY spectra were processed using programs inbuilt in Bruker’s Topspin v1.3 by integrating peaks that appear at the following 4 regions and corresponds to signals exclusively from protein: 0.84 – 1.03 ppm; 1.26 – 1.79 ppm; 1.79 – 2.17 ppm; and, 2.17 – 2.47 ppm. However, similar results were obtained by fitting individual peaks in this region. The integrated peak area (I) for each region was plotted as function of increasing gradient strength (g) and fit to the Stejskal-Tanner equation (38): =

(∆ )

!

where, I0 is peak area in absence of gradient, D is the diffusion coefficient, γ is the gyromagnetic ratio, δ is half the gradient pulse length and ∆ is the diffusion time. The diffusion coefficients for each region were extracted by fitting the equation using GraphPad Prism v5.

RESULTS Optimization of over-expression and stabilization of p53DBD proteins for NMR studies. While wt-p53DBD has been expressed and purified for structural biology applications, in many cases the protein was expressed into inclusion bodies due to the inherent instability of the protein, which required appropriate refolding of the protein after purification (39). Given the inherent difficulty in refolding proteins as well as verification of the accuracy of proper folding, it is highly desirable to over-express large quantities (~mg) of wt-p53DBD in soluble form for structural studies. Moreover, previous over-expression of p53DBD has made use of N-terminal tags to assist purification (28). Although tags can be cleaved by proteases after purification, the process is often inefficient, leading to loss of protein yield as well as contamination with tagged


8

Page 9 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

protein. Thus, high yield over-expression of tag-less p53DBD in soluble fraction is often desirable. Hence, we optimized a construct to express the p53DBD proteins without tags and to over-express the wild type and mutant proteins in soluble form followed by the optimization of purification conditions to obtain high quantities of p53DBD suitable for structural studies. We found an optimal expression of soluble wt-p53DBD by inducing the culture at O.D.600nm ~0.8, followed by a low temperature, post induction growth at 18 ºC, 120 rpm for 18 h, leading to minimal amount of p53DBD accumulating as inclusion bodies. Using this construct and the optimized over-expression and purification conditions, we obtained up to 30 mg of

13

C,

15

N

labeled protein per liter of minimal medium for the wild type and R273H p53DBD, although lower yield was observed for R248Q (14 mg/L culture) as well as for perdeuterated wild type p53DBD (~5 mg/L culture). In previous studies with p53DBD, NMR experiments were performed at pH 7.1 and 20 °C (5), which in our case did not yield a sufficiently stable condition for extended NMR studies and lead to precipitation of the p53DBD proteins. Hence, we optimized the buffer composition and temperature which showed that the wild type and mutant p53DBD remained sufficiently stable at pH 6.8 (50 mM potassium phosphate, 150 mM KCl and 5 mM DTT) and 18 ºC for prolonged time required for NMR studies. Hence, all relaxation measurements were performed under these conditions. Backbone resonance assignment of wild type and mutant p53DBD proteins. 1H-15N HSQC spectrum of wild type p53DBD was recorded (Figure 1a) which was similar to the spectrum previously published by Fersht and co-workers (5). By comparing the spectra, we were able to assign a small fraction (~30%) of the resonances. To verify these assignments and overcome ambiguities of the remaining assignments, we recorded triple resonance HNCA and HN(CA)CB experiments using a uniformly 13C,15N-labeled wt-p53DBD protein, of which the former could


9

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 32

Figure 1. Comparison of NMR chemical shift differences for R248Q and R273H with wt-p53DBD, at pH 6.8 (a)

15

N-1H HSQC spectrum of wt-p53DBD at pH 6.8 with backbone resonance assignments. Bar

graph plots indicating residue specific amide chemical shift differences (∆δamide) for (b) R248Q and (c) R273H mutants, where, ∆δHN and ∆δN are the changes in 1HN and 15N chemical shifts, respectively. Red dashed lines indicate the point of respective mutations. (d,e) Amide chemical shift differences obtained from (b) and (c) are mapped onto the x-ray crystal structure of wt-p53DBD (PDB ID 3KMD) (40) for (d)


10

Page 11 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

R248Q and (e) R273H mutants with the maximum and minimum changes shown in red and cyan, respectively. Residues for which ∆δamide could not be determined are colored in gray while key amino acids located in various parts of the p53DBD are shown in spheres for illustration.

be used to assign many additional residues by sequential backbone walk approach. However, the HN(CA)CB spectrum did not yield sufficient signal probably due to fast transverse relaxation from a relatively large (~25 kDa) protein. Hence, we prepared 2H,13C,15N labeled wt-p53DBD and recorded the HN(CA)CB spectrum, along with a HN(CO)CA spectrum and finally, out of the 202 non-proline amino acids, backbone amide resonances for 186 residues (~92%) could be unambiguously assigned. For the R248Q mutant, the spectrum was essentially identical to the wild type protein with minor chemical shift perturbations near the site of mutation, as can be observed from its spectrum overlaid atop the wt-p53DBD spectrum (Figure S1). Here, ~82% of the resonances were found to be overlapping with the wild type protein, which could be assigned by direct comparison between the two proteins. However for R273H mutant, there were relatively larger shifts in comparison to R248Q mutant, even though a majority of the backbone amide resonances were still overlapping with the wild type p53DBD (Figure S2). Nonetheless, we recorded a set of HNCA and HN(CO)CA spectra on a uniformly

13

C,15N-labeled R273H

p53DBD, using which ~85% of the non-proline backbone amide resonances could be unambiguously assigned. Estimation of structural perturbation in DNA-contact mutants by chemical shift analysis. The HSQC spectrum of wt-p53DBD is shown in Figure 1a, along with the spectra of R248Q (Figure S1) and R273H (Figure S2) mutants which were overlaid atop wt-p53DBD. As noted before, the majority of residues for both mutants displayed overlapping amide resonances with significant deviations observed at the site of the mutations. Moreover, the R273H mutant had


11

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 32

more residues with extensive chemical shift perturbations with respect to the wild type p53DBD than the R248Q mutant. To analyze this more quantitatively, we plotted the amide chemical shift perturbation (∆δamide) of residues that could be assigned for each of the mutants with respect to the wt-p53DBD using the equation: ∆δ#$%&' = (∆δ() ) + (∆δ) ⁄5) where, ∆δN and ∆δHN are the observed

15

N and 1HN chemical shift differences between each of

the mutants and the wild type, respectively. These chemical shift differences were plotted for R248Q and R273H p53DBD mutants in Figures 1b and 1c, respectively. Between the two mutants, R248Q shows an average (±standard deviation) ∆δamide of 0.01 (±0.01) ppm with most of the chemical shift changes taking place in the vicinity of the mutation, although most of these residues could not be assigned due to weak signal. It is noteworthy that in addition to residues in proximity to the site of mutation (R248) which include β-strands S9, S10 and loop L3, there are also significant chemical shift changes noted in strands S2, S2’and helix H2 (Figure 1d). In comparison to R248Q mutant, we detected significantly more perturbations in R273H with an average (±SD) ∆δamide of 0.03 (±0.04) ppm (Figure 1c). Although the overall locations of the perturbations follow the same pattern as with R248Q mutant, the spread and extent of the perturbations are higher for R273H mutant with more number of residues in the β-sandwich motif reporting chemical shift changes (Figure 1e). The chemical shift perturbations in R273H mutant are localized near its DNA binding surface and could suggest modest structural changes. However, since

13

Cα chemical shifts are strong reporters of secondary structural changes in

proteins (41), we compared the residue-specific

13 α

C chemical shift difference (∆δCα) between

wt-p53DBD and R273H mutant. We found an excellent correlation between 13Cα chemical shifts of wt-p53DBD and R273H mutant (r2 = 0.99, Figure S3a) with |∆δCα | yielding an average (±SD)


12

Page 13 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

of 0.025 ± 0.096 ppm (Figure S3b). However, residues between E271 to D281 near the DNA binding surface, comprising strand S10-loop-helix H2 exhibit appreciable differences between the δCα values, suggesting the possibility of some rearrangement of secondary structural units in this region. Overall, the cumulative results of chemical shift analysis suggest that while R248Q has minimal backbone amide chemical shift changes compared to wt-p53DBD, the perturbations are more in magnitude and spread for R273H, although most parts of both the mutant proteins stay unperturbed. This is consistent with previous studies stating that DNA-contact mutations do not significantly alter the overall structural conformation of p53DBD (9). However, chemical shift perturbations alone do not accurately reflect subtle changes in the overall structure of the protein and the resulting mutation can impart significant change in the backbone dynamics in the DNA-contact mutations which may be related to the reduced stabilities noted in these systems (6). Analysis of spin relaxation rates of wt-p53DBD. To understand the effects of DNA-contact mutations on the backbone dynamics of mutant p53DBD, we performed spin relaxation NMR studies which helps us probe the effect of mutation on the dynamic properties of the protein backbone. Table 1. Summary of 15N spin relaxation parameters of R1, R2 and NOE for wt-p53DBD and its DNAcontact mutants, R248Q and R273H, where, R1 and R2 are the backbone amide longitudinal and transverse relaxation rates, respectively.

p53DBD

R1 (s-1)

R2 (s-1)

NOE

wild type (wt)

1.18 ± 0.32

24.13 ± 6.57

0.68 ± 0.27

R248Q

1.18 ± 0.34

24.34 ± 7.11

0.65 ± 0.29

R273H

0.79 ± 0.22

22.01 ± 6.26

0.66 ± 0.36


13

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 32

Figure 2. Qualitative analysis of backbone dynamics of wild type and mutant p53DBD. (a-c) Residue specific plot of R2/R1 ratios of (a) wild type, (b) R248Q and (c) R273H mutants of p53DBD. For each protein, the gray area identifies residues lying within one standard deviation from the mean of R2/R1 ratio and is indicative of residues that exhibit rigid backbone of the p53DBD. The key regions that are predicted to have fast and slower conformational dynamics in comparison to the rest of the protein are indicated with red and blue dashed lines, respectively. R2/R1 ratios of respective proteins are mapped onto the X-ray structure of wt-p53DBD (PDB entry 3KMD) (40); with the lowest and highest values indicated in colors green and magenta, respectively, while residues with no information of R2/R1 ratio are shown in gray.

We extracted the backbone 15N longitudinal (R1) and transverse (R2) spin relaxation rates, along with the values of NOE for the wt-p53DBD (Table S1 and Figure S4, black circles). Out of a


14

Page 15 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

total of 186 assigned residues, relaxation rates could be obtained for 170 backbone spins whereas for many residues these values could not be extracted due to significant overlap or poor signal. The mean (±SD) values for these rates are summarized in Table 1. The value of NOE and the ratio of R2/R1 relaxation rates could be used as a qualitative indicator for backbone dynamics (42), wherein a lowering of these values with respect to neighbouring amino acids suggest higher conformational flexibility in the picoseconds (ps) – nanosecond (ns) timescale of motion. Moreover, residues that undergo chemical exchange induced line-broadening due to conformational dynamics in the slower microsecond (µs) – millisecond (ms) regime are expected to have higher R2 values, and hence, increased R2/R1 values (43). From the NOE values (Figure S4c) and residue specific plot of R2/R1 values of wt-p53DBD (Figure 2a), it appears that three regions may undergo fast conformational dynamics (red dashed lines, Figure 2a), in addition to residues located at both termini that are reasonably expected to have high flexibility. The residues in these three regions include loops L1 and L2 as well as parts of the β-strands S8 whereas for the region between strands S5 to S7 is predicted to undergo slower motions in the µs-ms timescale. Analysis of NMR spin relaxation rates of p53 DNA-contact mutants. For the DNA-contact mutants, the relaxation rates could be obtained for 143 and 151 residues for R248Q and R273H, respectively (Tables S2 and S3). The transverse relaxation rate (R2) and NOE, averaged (±SD) over all residues for which rates could be extracted, are shown in Table 1. From the summary of these rates, it is evident that the values are quite similar between the wt-p53DBD (24.1 ± 6.6 Hz; 0.68 ± 0.27) and both mutants, i.e. R248Q (24.3 ± 7.1 Hz; 0.65 ± 0.29) and R273H (22.0 ± 6.3 Hz; 0.66 ± 0.36). While there is a slight drop in the average R2 value for R273H, this may not be significant. However, while the longitudinal relaxation rate (R1) for wt-p53DBD and R248Q


15

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 32

averaged (±SD) over all residues were (1.18 ± 0.32 Hz) and (1.18 ± 0.34 Hz), respectively, the rate for R273H mutant (0.79 ± 0.22 Hz) was significantly and systematically lower for most of the residues (Figure S4a). This suggests a significant change in rotational diffusion property of R273H mutant in comparison to the wild type, possibly due to an increase in its hydrodynamic radius of the mutant with respect to wt-p53DBD. To verify this, we measured the diffusion coefficients (D) for all three proteins (Ttable S4), which revealed that while the coefficient for wild type and R248Q were similar at 9.12±0.14 x10-11 m2/s and 9.13±0.52 x10-11 m2/s, respectively, the coefficient reduced to 8.31±0.11 x10-11 m2/s for R273H mutant. Since the diffusion coefficient is inversely proportional to the hydrodynamic radius of the molecule, the reduction of diffusion coefficient indicates an enlargement of the R273H mutant. From the R2/R1 ratios, in R248Q mutant the β-sandwich scaffold appears to adopt a rigid conformation, except for the DNA-binding interface comprising loops L1, L2 and β-strand S8, which is likely to undergo fast time scale dynamics (Figures 2b). In case of R273H, the spread and extent of backbone dynamics appears to extend throughout the β-sandwich scaffold, especially around its site of mutation (Figures 2c). Additionally, the loop region between strands S7 and S8 appear to be uniformly mobile in all three proteins with a reduced R2/R1 and NOE values for all the proteins (Figures 2 and S4c). Cumulatively, a comparison of the relaxation parameters between the mutants suggest that the overall pattern of residue specific backbone dynamics predicted for the mutants is similar to that of wt-p53DBD, although most residues of R273H exhibit higher R2/R1 values due to lowering of R1.

Estimation of quantitative backbone dynamics parameters in wild type p53DBD and DNAcontact mutants. While the relaxation rates can provide a qualitative evaluation of the nature


16

Page 17 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

and extent of dynamics, there are many global (diffusion tensor) and local (conformational flexibility) parameters that contribute to these rates. In order to obtain more quantitative evaluation of conformational dynamics in the ps – ns regime, the relaxation rates were used to extract the generalized order parameters (S2) along with the diffusion tensor and motional parameters using Lipari-Szabo Modelfree formalism (44-46). Initially, the relaxation data for the wt-p53DBD was fit to both isotropic and axially symmetric diffusion tensor models, which predicted a significantly better fit for an axially symmetric tensor based on F-value statistical analysis (47). Fitting the data to an asymmetric tensor did not yield statistically significant

Figure 3. Quantitative characterization of fast time scale dynamics of wt-p53DBD and DNA-contact mutants. Plots of simulated order parameters (S2) of (a) wt-p53DBD, (b) R248Q, and (c) R273H as a function of residue obtained from Modelfree analysis of spin relaxation data at 18 ºC. Vertical red dashed lines denote key regions within p53DBD where significant conformational perturbations are noticed.


17

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 32

Values of S2 are also mapped on to the x-ray structure (PDB ID 3KMD) (40) into the respective proteins to depict the location of the dynamic regions. Red and blue colors indicate residues with S2 values lower than 0.49 and higher than 0.99, respectively, whereas residues for which S2 could not be determined were colored in gray. Some of the important residues are shown in spheres for illustration.

improvement in the fit, and hence, the axially symmetric tensor parameters along with the appropriately modified structure of wt-p53DBD obtained after the diffusion tensor estimation (47) were used to perform Modelfree simulations (31). Calculations were initiated with 156 residues for which relaxation data were available and models could be assigned to 134 residues for wt-p53DBD (Figure 3a and Table S5). Table 2. Summary of order parameter and diffusion tensor obtained from Modelfree analysis of wtp53DBD and its DNA-contact mutants, R248Q and R273H. D-ratio is the ratio of major (Dǁ) and minor (D┴) axis axially symmetric model. The mean of order parameter obtained over all the residues has been denoted as S2avg.

p53DBD

Rotational correlation time, τm (ns)

D-ratio (Dǁ/┴)

S2avg (mean ± SD)

wild type (wt)

13.5 ± 0.2

0.67 ± 0.06

0.88 ± 0.10

R248Q

14.0 ± 0.2

1.23 ± 0.09

0.88 ± 0.10

R273H

15.7 ± 0.1

0.79 ± 0.03

0.86 ± 0.08

The results show that the wild type forms an axially symmetric tensor ((Dǁ/┴ = 0.67 ± 0.06) with a rotational correlation time (τm) of 13.5 ± 0.2 ns and has an average S2 of 0.88 ± 0.10 (Table 2) for the 134 residues which could be assigned to one of the five models (31). A similar procedure was followed to perform the Modelfree calculations for both the mutant proteins and the results


18

Page 19 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

are shown in Tables S6 and S7 for R248Q and R273H mutants, respectively, along with a summary of key motional parameters shown in Table 2. Importantly, in comparison to the wild type, R273H exhibits a significant increase in its rotational correlation time (τm = 15.7 ± 0.1 ns). This could be due to a relatively larger molecular size (hydrodynamic radius) resulting from an enlargement in its overall global fold upon mutation and is consistent with a reduction of the diffusion coefficient with respect to wt-p53DBD. However, since wt-p53DBD is known to form oligomers, albeit at concentrations over 400 µM (28), the possibility for some inter-molecular association leading to low populated dimers in R273H exists. The average (±SD) S2 for 112 residues in R248Q (S2 = 0.88 ± 0.10) and 119 residues in R273H (S2 = 0.86 ± 0.08) that could be assigned to one of the five possible models, are remarkably similar for both mutants as well as the wild type p53DBD (Figures 3b and 3c). In spite of similarity of their S2 values, a closer look at the internal time scale parameter (τe) along with chemical exchange correction factor (Rex) indicate interesting differences. Both wild type and R248Q p53DBD exhibits appreciable τe and Rex values (Figures S5 and S6), wherein a significant number of residues, mostly located in strands S1, S2 and S7, exhibit slower internal motions (i.e. large τe values) (31) . In contrast, R273H exhibits comparatively faster internal motions (i.e. lower τe values) for most part of the protein. In addition, there is indication of chemical exchange in residues around strand S7 (for R248Q) along with loop L3 (for wild type), which is less dispersed among residues in R273H. Taken together, for the wt-p53DBD, most of the backbone dynamics was located in the loops L1 and L2 as well as for residues in and round β-strand S8, whereas most of the β-sandwich appears to be fairly rigid. While both the R248Q and R273H mutants exhibit an overall similar pattern of dynamic character, there are important differences between the mutants. For example, the flexibility of residues in strand S8 is practically quenched in R273H but not in R248Q mutant or


19

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 32

the wild type (Figure 3). In a similar contrast, the R273H mutant exhibits limited flexibility in the β-sandwich core, encompassing strands S5 and S6, which is not observed for the other proteins. Overall, while all three proteins exhibit similar backbone dynamics, the motions of R248Q appears to be closer to the wild type than the R273H mutant.

DISCUSSION Mutations in the DNA-binding domain of p53 (p53DBD) have been associated with a majority of human cancers. Due to its low intrinsic thermodynamic stability, even weakly destabilizing mutations can have severe effects on the overall structure and functionality of p53DBD (48). Moreover, a study on breast cancer patients has led to the observation that different p53 mutations are associated with different prognostic values (49). Therefore, to understand the role of p53 mutations in cancers, it is imperative to study the changes in structure and dynamics associated with such mutations. Since, among the three p53 domains, the DNA-binding domain has most of these mutations, we studied the changes associated in the backbone dynamics of wild type and two of the most commonly occurring hot-spot DNA-contact mutations, R248Q and R273H. For this, we designed a scheme to over-express p53DBD without tags and in soluble form but in sufficient quantities to allow NMR studies. This approach overcomes many difficulties associated with tagged proteins or refolding them from inclusion bodies and provides a reliable protocol for in vitro studies with p53DBD. Previous studies have shown that R248Q and R273H mutations abrogate the DNA-binding ability of p53DBD without significantly changing the overall structure of the DNA-binding domain (9), which is in agreement with our HSQC data that shows minor changes in the chemical shifts of 1H,15N resonances when compared to wt-p53DBD (Figure S1, S2). Simultaneously, we examined the chemical shift


20

Page 21 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

differences (∆δamide) between wt-p53DBD and the DNA-contact mutants which indicated subtle changes in ∆δamide for R248Q primarily in strands S2, S2′ and between strand S10 and helix H2 but also in loops L2 and L3, with most of these changes taking place away from its site of mutation. The location of chemical shift changes observed here is in agreement with prior work by Fersht and co-workers (5) which suggest that R248Q may also be considered as a “structure mutant” in addition to being a DNA-contact mutant, even though the magnitude of ∆δamide changes found here do not suggest major changes in its structural fold. In contrast, R273H showed greater ∆δamide perturbations especially in region S2′, S10 and helix H2, signifying more backbone amide chemical shift differences and suggests that more extensive structural changes may take place in R273H mutation compared to R248Q. Prior studies (5) have indicated that R273H introduces chemical shift changes in the loop-sheet-helix motif and L3 loop, while R248Q exhibits chemical shift changes in loops L2 and L3 of the core domain. This is in partial agreement with our observation wherein for R273H we notice chemical shift change in the loopsheet-helix motif. However, we were unable to detect these changes in loop L3 due to the unavailability of amide resonances for a large number of residues in this region, although chemical shift variation in loop L3 is certainly possible.

In order to understand whether structural perturbations are accompanied with changes in the backbone dynamics of p53DBD, we performed spin relaxation measurements which typically influence molecular dynamics in the fast, ps – ns timescale regime. Earlier studies with wtp53DBD and a structure-mutant have shown that the ratio of spin relaxation rates, R2/R1, can be used as a predictor of conformational dynamics (43). Here, we applied a similar technique to qualitatively evaluate the presence of backbone flexibility. Our results show that R248Q behaves


21

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 32

as a classic DNA-contact mutant whose β-sandwich scaffold adopts a rigid conformation, similar to wt-p53DBD, with bulk of increased backbone dynamics located near the DNA binding interface. On the other hand, certain residues mostly present in and around strand S7 and between S10 and H2 seem to exhibit higher R2/R1 values, suggesting the possibility of slower motions in the µs – ms regime, although this has not been explicitly studied here (Figure 2b). On the contrary, R273H showed a marked difference in R2/R1 ratios in comparison to the wtp53DBD where large parts of its β-sandwich scaffold exhibited substantial increase in R2/R1 value and increased backbone motions across the entire DNA binding surface. This was more quantitatively studied by the Modelfree calculations and extraction of various motional parameters, importantly the generalized order parameters (S2). While appreciable dynamics were observed for loops L1 and L2 for all the proteins, there appeared to be differences between the mutants. Interestingly, strand S8 is dynamic in R248Q mutant and the wild type p53DBD, however, for R273H, this is absent. Instead it exhibits additional dynamics for residues in strand S6 (Y205) which is relatively rigid in both wt-p53DBD and R248Q mutant (Figure 3). The observed mobility of the loop regions is consistent with solvent protection factors observed in previous 1H/2H exchange studies on wt-p53DBD (28). In order to understand if the dynamic propensities of the p53DBD proteins correlate with their known thermodynamic stabilities, we compared our results with urea-mediated equilibrium denaturation experiments on these proteins (6) that report on their stability. Comparative analysis of the wild type and the two DNA contact mutants show that at 10 ºC wt-p53DBD and R273H mutant higher similar stability and sensitivity to urea denaturation compared to R248Q, which is less stable, even though both . /

mutants have lower free energies of unfolding (∆,) in comparison to the wild type (6). While

the lower stability of mutants appears to correlate with increased dynamics observed here, wt-


22

Page 23 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

p53DBD was found to have similar stabilities with R273H but not R248Q, although this could be due to differences in experimental conditions, including temperature and pH. We extracted the residue specific conformational entropy (Sconf) from the backbone

15

N order parameters (S2) for

all the three proteins to understand if the reduced stability for mutants were concomitant with an increase of conformational entropy. The average Sconf of residues for which

15

N S2 values were

available revealed a mean (±SD) value of -3.2 ± 1.3 cal/mol/K (wt-p53DBD), -3.2 ± 1.2 cal/mol/K (R248Q) and -2.7 ± 1.1 (R273H) cal/mol/K. This shows that while the backbone conformational entropy is similar for wild type p53DBD and R248Q, there is an appreciable increase of entropy for the R273H mutant. The crystal structure of wt-p53DBD (Figure 4a) shows the quaternary structural arrangement of its four sub-units bound to DNA. To summarize the effect of DNA-contact mutations in p53DBD, key residues that perturb the structure or dynamics of the mutants from this work are shown in red (Figure 4b and 4c) While a number of residues in the DNA binding interface exhibit perturbations either in chemical shift or backbone mobility, few residues far away from this interface appear to be important and could play important roles in the overall stability of the p53DBD molecule. For example, β-strands S6 and S7 form a pair on one edge of the β-sandwich scaffold, which provides stability to the core of the DNA-binding domain, whereas strands S1 and S3 form the other edge of this scaffold. As shown in Figure 3, while strand S1 appears rigid in the wild type, it becomes flexible in both the mutants, extending up to loop L1 (residues L114 and S116). As a consequence of this elevated dynamics, residue S116 that has a stabilizing effect on strand S2, which in turn stacks with strand S2’ in the wild type protein, appear to become more flexible in both the mutants. Given the high R2/R1 value of S116 in R273H but not in R248Q or wt-p53DBD (Figure 2), it is also possible that these motions lead to more extensive


23

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 24 of 32

dynamics in the slow, µs – ms timescale in L1 loop in R273H, although this has not been verified in this study.


24

Page 25 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

Figure 4. Representative residues playing key role in modulating the structure and dynamics of p53DBD mutants. (a) Cartoon representation of the crystal structure of wt-p53DBD (PDB ID 3KMD) (40) depicting the assembly of the 4 sub-units around the dsDNA, where helices, strands and loops are shown in red, yellow and blue, respectively. (b) The DNA binding interface of a monomeric sub-unit of p53DBD and zoomed in (c) with key residues highlighted in red. (d) The 1H,15N chemical shift perturbations of important the residues highlighted in (b,c) are shown, where the resonances from wild type, R248Q and R273H p53DBD are shown in black, blue and red, respectively.

Similarly, residue Y205 in strand S6, along with adjacent residue S215 in strand S7 gain significant fast motions in R273H mutant but not in R248Q protein based on S2 values (Figure 3c). Interestingly, while being dynamic, these residues appear not to cause significant structural perturbations as evidenced by their overlapping resonances in 1H,15N HSQC spectra (e.g. Y205 and L114 in Figure 4d). Overall, the change in dynamics observed in L1, in addition to those in strand S10 and helix H2 are important since they form an integral part of the loop-sheet-helix motif which in turn plays an essential role in binding to the DNA (40). In addition to the observed dynamics in loop L1, it is quite possible that loop L3, which forms part of the proteinDNA interface, may also undergo slow timescale dynamics. However, this could not be verified directly in our work due to low intensity of peaks in this region, possibly due to intermediate timescale chemical exchange induced line-broadening. This appears to be a possibility due to the overall lower intensity of resonances and much lesser number of residues that could be assigned in this region for the mutants than the wild type.

In summary, this study has led to the identification of three key regions and residues of importance in the two DNA-contact mutants, R248Q and R273H, which show significant


25

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 26 of 32

conformational dynamics. We show that R248Q is structurally closer to the wild type p53DBD than the R273H and is a classic DNA-contact mutant. R273H on the other hand, exhibits comparatively extensive structural perturbations in addition to changes in its dynamics, which suggests that in addition to being a DNA-contact mutation, it also has properties of a structural mutant. Our results show distinct regions that are differentially dynamic in the wild type as well as in both the mutants, which in turn suggest how changes in these fast timescale motions could be responsible for destabilization of the core β-sheet scaffold of p53DBD and hence its thermodynamic destabilization.


26

Page 27 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

ASSOCIATED CONTENT Supporting Information. Figures S1-S2 showing the overlaid

15

N-1H HSQC NMR spectra of

R248Q and R273H mutants overlaid atop wt-p53DBD. Figure S3 showing

13

Cα chemical shift

changes in R273H mutant with respect to wt-p53DBD and Figures S4 – S6 showing plots of 15N R1, R2, NOE, τe and Rex of all the proteins. Tables S1-S8 containing

15

N spin relaxation data,

DOSY NMR, results of model-free analysis and conformational entropy for the p53DBD proteins. AUTHOR INFORMATION Corresponding Author *Email: [email protected] Phone: +91-33-24995715 Author Contributions J.A.R. performed mutagenesis, protein over-expression, purification and preparation of NMR samples with S.D., A.B. performed the NMR experiments, analyzed NMR data, S.M. conceived the project, supervised research and performed NMR experiments. J.A.R., A.B. and S.M. wrote the manuscript. Funding Sources This work was supported by funding from DST (Ramanujan fellowship to S.M. and INSPIRE fellowship to A.B.) and research fellowships from UGC (J.A.R.) and CSIR (S.D.). Funding for the 700 MHz NMR was obtained from DBT (Grant Number BT/PR3106/INF/22/138/2011) ACKNOWLEDGMENT


27

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 28 of 32

We thank Prof. Siddhartha Roy for stimulating discussions on the research theme, Prof. Arthur G. Palmer and Prof. J. Patrick Loria for access to NMR data analysis programs, the NMR facility at Bose Institute for access to 700 MHz NMR spectrometer funded by the DBT (Grant Number BT/PR3106/INF/22/138/2011). This work was supported by funding from DST (Ramanujan fellowship to S.M. and INSPIRE fellowship to A.B.) and research fellowships from UGC (J.A.R.) and CSIR (S.D.). ABBREVIATIONS NMR, nuclear magnetic resonance; HSQC, heteronuclear single quantum coherence; 2D, two-dimensional; NOE, nuclear Overhauser enhancement; ps, picosecond; ns, nanosecond; ms, millisecond; µs, microsecond; SD, standard deviation; wt-p53DBD, wild type p53 DNA binding domain;


28

Page 29 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19.

Vousden, K. H., and Lu, X. (2002) Live or let die: the cell's response to p53, Nature reviews 2, 594-604. Vogelstein, B., Lane, D., and Levine, A. J. (2000) Surfing the p53 network, Nature 408, 307-310. Hollstein, M., Sidransky, D., Vogelstein, B., and Harris, C. C. (1991) p53 mutations in human cancers, Science 253, 49-53. Brosh, R., and Rotter, V. (2009) When mutants gain new powers: news from the mutant p53 field, Nature reviews 9, 701-713. Wong, K. B., DeDecker, B. S., Freund, S. M., Proctor, M. R., Bycroft, M., and Fersht, A. R. (1999) Hot-spot mutants of p53 core domain evince characteristic local structural changes, Proc Natl Acad Sci U S A 96, 8438-8442. Bullock, A. N., Henckel, J., and Fersht, A. R. (2000) Quantitative analysis of residual folding and DNA binding in mutant p53 core domain: definition of mutant states for rescue in cancer therapy, Oncogene 19, 1245-1256. Bullock, A. N., Henckel, J., DeDecker, B. S., Johnson, C. M., Nikolova, P. V., Proctor, M. R., Lane, D. P., and Fersht, A. R. (1997) Thermodynamic stability of wild-type and mutant p53 core domain, Proc Natl Acad Sci U S A 94, 14338-14342. Bullock, A. N., and Fersht, A. R. (2001) Rescuing the function of mutant p53, Nature reviews 1, 68-76. Joerger, A. C., and Fersht, A. R. (2008) Structural biology of the tumor suppressor p53, Annu Rev Biochem 77, 557-582. Toyooka, S., Tsuda, T., and Gazdar, A. F. (2003) The TP53 gene, tobacco exposure, and lung cancer, Human mutation 21, 229-239. Tommasino, M., Accardi, R., Caldeira, S., Dong, W., Malanchi, I., Smet, A., and Zehbe, I. (2003) The role of TP53 in Cervical carcinogenesis, Human mutation 21, 307-312. Giglia-Mari, G., and Sarasin, A. (2003) TP53 mutations in human skin cancers, Human mutation 21, 217-228. Schuijer, M., and Berns, E. M. (2003) TP53 and ovarian cancer, Human mutation 21, 285-291. Iacopetta, B. (2003) TP53 mutation in colorectal cancer, Human mutation 21, 271-276. Gasco, M., Yulug, I. G., and Crook, T. (2003) TP53 mutations in familial breast cancer: functional aspects, Human mutation 21, 301-306. Joerger, A. C., Ang, H. C., and Fersht, A. R. (2006) Structural basis for understanding oncogenic p53 mutations and designing rescue drugs, Proc Natl Acad Sci U S A 103, 15056-15061. Eldar, A., Rozenberg, H., Diskin-Posner, Y., Rohs, R., and Shakked, Z. (2013) Structural studies of p53 inactivation by DNA-contact mutations and its rescue by suppressor mutations via alternative protein-DNA interactions, Nucleic Acids Res 41, 8748-8759. Palmer, A. G., 3rd. (2004) NMR characterization of the dynamics of biomacromolecules, Chemical reviews 104, 3623-3640. Jarymowycz, V. A., and Stone, M. J. (2006) Fast time scale dynamics of protein backbones: NMR relaxation methods, applications, and functional consequences, Chemical reviews 106, 1624-1671.


29

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

20. 21. 22.

23. 24. 25. 26.

27. 28. 29. 30. 31. 32. 33.

34.

Page 30 of 32

Cai, M. L., Huang, Y., Yang, R. B., Craigie, R., and Clore, G. M. (2016) A simple and robust protocol for high-yield expression of perdeuterated proteins in Escherichia coli grown in shaker flasks, Journal of Biomolecular Nmr 66, 85-91. Salzmann, M., Pervushin, K., Wider, G., Senn, H., and Wüthrich, K. (1998) TROSY in triple-resonance experiments: New perspectives for sequential NMR assignment of large proteins, Proc Natl Acad Sci (USA) 95, 13585-13590. Farrow, N. A., Muhandiram, R., Singer, A. U., Pascal, S. M., Kay, C. M., Gish, G., Shoelson, S. E., Pawson, T., Forman-Kay, J. D., and Kay, L. E. (1994) Backbone dynamics of a free and phosphopeptide-complexed Src homology 2 domain studied by 15N NMR relaxation, Biochemistry 33, 5984-6003. Das, J. K., Mall, S. S., Bej, A., and Mukherjee, S. (2014) Conformational flexibility tunes the propensity of transthyretin to form fibrils through non-native intermediate States, Angew Chem Int Ed Engl 53, 12781-12784. Delaglio, F., Grzesiek, S., Vuister, G. W., Zhu, G., Pfeifer, J., and Bax, A. (1995) NMRPipe: a multidimensional spectral processing system based on UNIX pipes, J Biomol NMR 6, 277-293. Goddard, T. D., and Kneller, D. G. SPARKY 3, University of California, San Francisco. Mukherjee, S., Pondaven, S. P., and Jaroniec, C. P. (2011) Conformational flexibility of a human immunoglobulin light chain variable domain by relaxation dispersion nuclear magnetic resonance spectroscopy: implications for protein misfolding and amyloid assembly, Biochemistry 50, 5845-5857. Tjandra, N., Feller, S. E., Pastor, R. W., and Bax, A. (1995) Rotational diffusion anisotropy of human ubiquitin from N-15 NMR relaxation, Journal of the American Chemical Society 117, 12562-12566. Canadillas, J. M., Tidow, H., Freund, S. M., Rutherford, T. J., Ang, H. C., and Fersht, A. R. (2006) Solution structure of p53 core domain: structural basis for its instability, Proc Natl Acad Sci U S A 103, 2109-2114. Lipari, G., and Szabo, A. (1982) Model-free approach to the interpretation of nuclear magnetic resonance relaxation in macromolecules. 1. Theory and range of validity, Journal of the American Chemical Society 104, 4546-4559. Lipari, G., and Szabo, A. (1982) Model-free approach to the interpretation of nuclear magnetic resonance relaxation in macromolecules. 2. Analysis of experimental results, Journal of the American Chemical Society 104, 4559-4570. Mandel, A. M., Akke, M., and Palmer, A. G. (1995) Backbone dynamics of Escherichia coli ribonuclease HI: Correlations with structure and function in an active enzyme, J Mol Biol 246, 144-163. Cole, R., and Loria, J. P. (2003) FAST-Modelfree: a program for rapid automated analysis of solution NMR spin-relaxation data, J Biomol NMR 26, 203-213. Yao, L., Grishaev, A., Cornilescu, G., and Bax, A. (2010) The impact of hydrogen bonding on amide 1H chemical shift anisotropy studied by cross-correlated relaxation and liquid crystal NMR spectroscopy, Journal of the American Chemical Society 132, 10866-10875. Yao, L., Grishaev, A., Cornilescu, G., and Bax, A. (2010) Site-specific backbone amide (15)N chemical shift anisotropy tensors in a small protein from liquid crystal and crosscorrelated relaxation measurements, Journal of the American Chemical Society 132, 4295-4309.


30

Page 31 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46.

47. 48. 49.

Yang, D. W., and Kay, L. E. (1996) Contributions to conformational entropy arising from bond vector fluctuations measured from NMR-derived order parameters: Application to protein folding, J Mol Biol 263, 369-382. Zidek, L., Novotny, M. V., and Stone, M. J. (1999) Increased protein backbone conformational entropy upon hydrophobic ligand binding, Nat Struct Biol 6, 1118-1121. Sabo, T. M., Bakhtiari, D., Walter, K. F., McFeeters, R. L., Giller, K., Becker, S., Griesinger, C., and Lee, D. (2012) Thermal coefficients of the methyl groups within ubiquitin, Protein Sci 21, 562-570. Stejskal, E. O., and Tanner, J. E. (1965) Use of spin echoes in the presence of a timedependent field gradient, J. Chem. Phys. 42, 288-292. Klein, C., Hesse, F., Dehner, A., Engh, R. A., Schwaiger, M., and Hansen, S. (2004) In vitro folding and characterization of the p53 DNA binding domain, Biol Chem 385, 95102. Cho, Y., Gorina, S., Jeffrey, P. D., and Pavletich, N. P. (1994) Crystal structure of a p53 tumor suppressor-DNA complex: understanding tumorigenic mutations, Science 265, 346-355. Spera, S., and Bax, A. (1991) Empirical correlation between protein backbone conformation and Cα and Cβ 13C nuclear magnetic resonance chemical shifts, J. Am. Chem. Soc. 113, 5490-5492. Kay, L. E., Torchia, D. A., and Bax, A. (1989) Backbone dynamics of proteins as studied by 15N inverse detected heteronuclear NMR spectroscopy: application to staphylococcal nuclease, Biochemistry 28, 8972-8979. Friedler, A., DeDecker, B. S., Freund, S. M., Blair, C., Rudiger, S., and Fersht, A. R. (2004) Structural distortion of p53 by the mutation R249S and its rescue by a designed peptide: implications for "mutant conformation", J Mol Biol 336, 187-196. Lipari, G., and Szabo, A. (1982) Model-free approach to the interpretation of nuclear magnetic resonance relaxation in macromolecules. 1. theory and range of validity, J Am Chem Soc 104, 4546-4559. Lipari, G., and Szabo, A. (1982) Model-free approach to the interpretation of nuclear magnetic resonance relaxation in macromolecules. 2. Analysis of experimental results, J Am Chem Soc 104, 4559-4570. Clore, G. M., Szabo, A., Bax, A., Kay, L. E., Driscoll, P. C., and Gronenborn, A. M. (1990) Deviations from the simple two-parameter model-free approach to the interpretation of nitrogen-15 nuclear magnetic relaxation of proteins, J Am Chem Soc 112, 4989-4991. Tjandra, N., Feller, S. E., Pastor, R. W., and Bax, A. (1995) Rotational diffusion anisotropy of human ubiquitin from N-15 NMR relaxation, J. Am. Chem. Soc. 117, 12562-12566. Joerger, A. C., Ang, H. C., Veprintsev, D. B., Blair, C. M., and Fersht, A. R. (2005) Structures of p53 cancer mutants and mechanism of rescue by second-site suppressor mutations, The Journal of biological chemistry 280, 16030-16037. Olivier, M., Langerod, A., Carrieri, P., Bergh, J., Klaar, S., Eyfjord, J., Theillet, C., Rodriguez, C., Lidereau, R., Bieche, I., Varley, J., Bignon, Y., Uhrhammer, N., Winqvist, R., Jukkola-Vuorinen, A., Niederacher, D., Kato, S., Ishioka, C., Hainaut, P., and Borresen-Dale, A. L. (2006) The clinical value of somatic TP53 gene mutations in 1,794 patients with breast cancer, Clin Cancer Res 12, 1157-1167.


31

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 32 of 32

For Table of Contents Use Only


32

Intrinsic Differences in Backbone Dynamics between Wild Type and

Recommend Documents