Phe71 in Type III Trypanosomal Protein Arginine Methyltransferase 7

Jan 29, 2018 - Tamar B. Cáceres†, Abhishek Thakur‡, Owen M. Price†, Nicole Ippolito‡, Jun Li§∥, Jun Qu§∥ , Orlando Acevedo‡ , and Joa...
0 downloads 3 Views 2MB Size
Subscriber access provided by UNIVERSITY OF TOLEDO LIBRARIES

Article

Phe 71 in type III Trypanosomal protein arginine methyltransferase 7 (TbPRMT7) restricts the enzyme to monomethylation Tamar Caceres, Abishek Thakur, Owen M. Price, Nicole Ippolito, Jun Li, Jun Qu, Orlando Acevedo, and Joan Michelle Hevel Biochemistry, Just Accepted Manuscript • DOI: 10.1021/acs.biochem.7b01265 • Publication Date (Web): 29 Jan 2018 Downloaded from http://pubs.acs.org on February 3, 2018

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 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 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 24 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

Determinants of TbPRMT7 product specificity Phe 71 in type III Trypanosomal protein arginine methyltransferase 7 (TbPRMT7) restricts the enzyme to monomethylation Tamar B. Cáceres 1, Abhishek Thakur2, Owen M. Price 1, Nicole Ippolito2, Jun Li3,4, Jun Qu3,4, Orlando Acevedo2, and Joan M. Hevel1*. 1

Department of Chemistry and Biochemistry, Utah State University, 0300 Old Main Hill, Logan, UT 84322 2 Department of Chemistry, University of Miami, Coral Gables, FL 33146 3 Department of Pharmaceutical Sciences, University at Buffalo, State University of New York, Kapoor 318, North Campus, Buffalo, New York 14260 4 New York State Center of Excellence in Bioinformatics and Life Sciences, 701 Ellicott Street, Buffalo, New York 14203 *To whom correspondence should be addressed: Joan M. Hevel, Utah State University Chemistry and Biochemistry Department, 0300 Old Main Hill, Logan, UT 84322. Tel: (435)-797-1622; Fax: (435)-7973390; E-mail: [email protected] Keywords: Histone modification; Protein Methylation; Post-translational modification (PTM); Enzyme mechanism; S-Adenosylmethionine (AdoMet); Protein arginine methyltransferase (PRMT); product specificity; MMA; ADMA; SDMA Abstract Protein arginine methyltransferase 7 (PRMT7) is unique within the PRMT family as it is the only isoform known to exclusively make monomethylarginine (MMA). Given its role in epigenetics, the mechanistic basis for the strict monomethylation activity is under investigation. It is thought that PRMT7 enzymes are unable to add a second methyl group because of steric hindrances in the active site which restrict them to monomethylation. To test this, we probed the active site of Trypanosomal PRMT7 (TbPRMT7) using accelerated molecular dynamics, site-directed mutagenesis, kinetic, binding, and product analyses. Both the dynamics simulations and experimental results show that the mutation of Phe71 to Ile converts the enzyme from a type III methyltransferase, into a mixed type I, II; that is, an enzyme that can now perform dimethylation. In contrast, the serine and alanine mutants of Phe71 retain the type III behavior of the native enzyme. These results are inconsistent with a sterics-only model to explain product specificity. Instead, molecular dynamic simulations of these variants bound to peptides show that hydrogen bonding between would-be substrates and Glu172 of TbPRMT7. Only in the case of the Phe71 to Ile mutation is this interaction between MMA and the enzyme maintained, and the geometry for optimal SN2 methyl transfer obtained. The results of these studies highlight the benefit of combined computational/experimental methods in providing a better understanding for how product specificity is dictated by PRMTs. Introduction Arginine methylation of proteins is an important posttranslational modification involved in remodeling of chromatin, signal transduction, gene transcription, DNA repair, RNA processing and other essential biological pathways (1-5), reviewed in (6,7). Consequently the dysregulation of arginine methylation is connected with a wide variety of conditions such as: lung and kidney disease (8,9), hepatitis B (10), menopause (11), cocaine addiction (12), as well as cardiovascular disease and carcinogenesis (9,13). This post-translational modification is catalyzed by the family members of the protein arginine methyltransferases (PRMTs). The PRMTs transfer a methyl group from the donor molecule S-adenosylL-methionine (AdoMet) to the basic amino acid arginine in the substrate protein. The substrate arginine residue can be methylated in three distinct ways on the guanidino group, forming three different products, according to which the PRMTs can be classified into three types. The arginine guanidino group can be 1 ACS Paragon Plus Environment

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 24

Determinants of TbPRMT7 product specificity monomethylated and further asymmetrically dimethylated by type I PRMTs (forming ADMA) or symmetrically dimethylated by type II PRMTs (forming SDMA). All but one of the known isoforms has the capability of forming a dimethylated product. PRMT7 is unique in that it is only capable of monomethylating targeted proteins (forming only MMA), classifying it as the only type III PRMT (Figure 1).

Figure 1. PRMT catalysis and classification. Three different types of PRMTs catalyze the methylation of arginyl residues in targeted proteins using S-adenosylmethionine (AdoMet) as the methyl donor. PRMT7 is the only exclusive monomethyltransferase.

Each of the methylation marks (MMA, ADMA, and SDMA) can have different functional consequences in the cell. The arginine methylation of histones illustrates this concept quite nicely. For example, in yeast, methylation of histone H3Arg2 with the ADMA mark contributes to transcriptional repression by inhibiting the trimethylation of the adjacent Lys4. However the MMA mark on this same arginine does not have this inhibitory effect on H3 Lys4 trimethylation, and promotes transcription activation (14,15). Strikingly, asymmetric dimethylation could have the opposite effect on transcription, as it has been also associated with the activation of genes. For example, deposition of an ADMA mark on Arg3 of Histone H4 leads to transcriptional activation in yeast, while deposition of an SDMA mark at the same site leads to transcriptional repression (16). Similar to histone H4R3 methylation, symmetric and asymmetric dimethylation marks antagonize each other in a non-histone substrate by regulating the E2F-1 transcription factor. Asymmetric dimethylation (ADMA) by PRMT1 hinders methylation by PRMT5, which augments E2F-1-dependent apoptosis, whereas PRMT5-dependent symmetric dimethylation (SDMA) promotes cell-cycle progression by antagonizing methylation by PRMT1 (17). It is clear that functional specificity is governed by the distinct methylation status of an arginyl group in a protein, highlighting the importance of understanding the molecular and structural basis for how PRMTs achieve product specificity. One of the less well characterized members of the PRMT family is PRMT7. After some initial controversy, studies have now concluded that this is a type III PRMT, which is only capable of producing MMA (18-21). PRMT7 is expressed in mouse embryonic stem cells, where it has been suggested to contribute to cellular differentiation (22). Recently, a study using Prmt7-null mice showed a significantly reduced oxidative metabolism and endurance exercise capacity, linking the deficiency of this enzyme with age-related obesity (23). Two novel homozygous mutations in the Prmt7 gene have been associated to SBIDDS (Short Stature, Brachydactyly, Intellectual Developmental Disability, and Seizures) 2 ACS Paragon Plus Environment

Page 3 of 24 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

Determinants of TbPRMT7 product specificity syndrome, pointing to a possible correlation between the mutations and the severity of this phenotype (24). As with other members of the PRMT family, PRMT7 has been found to undergo automethylation (25-27). Automethylation of PRMT7 seems to play a role in inducing and promoting the migratory and invasive behavior of breast cancer cells, and metastasis (28). Some studies also have implicated PRMT7 with the DNA damage response (29). These are just some examples of the many important biological pathways this enzyme is involved in by placing the MMA mark; however, the structural and mechanistic basis for the inability of PRMT7 to further dimethylate a substrate are still not clear. In the case of human PRMT7, the activity has been noted not to be particularly robust, making it a difficult model to study. However, a homolog from the parasite Trypanosoma brucei TbPRMT7 possesses robust PRMT activity toward multiple substrates (21). In an effort to understand the basis of the strict monomethylation activity of this enzyme, the crystal structures of different orthologs of PRMT7 have been solved (30-33). Comparison of the active sites of TbPRMT7 with other types of PRMTs identified differences in the substrate guanidine binding pockets (30,33,34). In TbPRMT7, the substrate arginine is buried in a very small, narrow pocket, whereas in type I and II PRMTs, which can dimethylate the substrate, the substrate arginine is bound in a larger cavity (Fig. 2). Based on this static structural information, sterics were hypothesized to prevent PRMT7 from forming a dimethylated product (30,32,33). Consistent with this idea, the Glu181Asp and Gln329Ala variants of TbPRMT7 produce SDMA (33,34). However, Glu181 is conserved between type III monomethylating and type I/II dimethylating PRMTs and the homologous residue to Gln329 in PRMT1 is a histidine, which is also very bulky. Thus, the origin of product specificity between the natural isoforms is still unclear prompting us to investigate if the overall volume of the PRMT7 active site was the basis for the product specificity of this enzyme.

Figure 2. Substrate arginine in the binding pockets of ratPRMT1 (PDB: 1OR8) and TbPRMT7 (PDB: 4M38). Static crystal structures show a more congested active site in TbPRMT7 (right) compared to PRMT1 (left). Substrate arginine (blue), S-adenosylhomocysteine (red).

In order to investigate the molecular basis of TbPRMT7 product specificity we conducted molecular dynamics analyses and experimental studies using the natural variation observed between type I and III PRMT isoforms as a guide. Using this integrated approach, we have identified Phe71 as a key residue for dictating MMA formation in TbPRMT7. The generated Phe71Ile variant is able to produce the three different arginine products (MMA, ADMA and SDMA) as demonstrated both in the dynamic simulations, and experimentally, making it the first time a type III PRMT has been remodeled into a mixed type I and 3 ACS Paragon Plus Environment

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 24

Determinants of TbPRMT7 product specificity II and PRMT. Interestingly by mutating this residue to a serine (Phe71Ser), the corresponding amino acid in human PRMT7, the enzyme behaves as a type III PRMT, producing only MMA despite having a more open cavity in the active site for the addition of a second methyl group. Molecular dynamic simulations of these variants bound to peptides show hydrogen bonding between would-be substrates and Glu172 of TbPRMT7. However, only in the case of the Phe71 to Ile mutation is this interaction between MMA and the enzyme maintained, and the geometry for optimal SN2 methyl transfer obtained. This integrated computational and experimental study provides new insights into the catalytic mechanism and product specificity of type III PRMTs.

Materials and methods Expression and purification of wild type TbPPRMT7 and variants TbPRMT7 (UniProt accession ID: Q582G4) variant proteins were generated using the QuikChange® sitedirected mutagenesis kit (Stratagene) with sets of complementary oligonucleotide primers spanning the desired site of mutation. For each PCR, the pET28b vector (Novagen) containing the gene that codes for N-terminal histidine-tagged rat WT-TbPRMT7 plasmid (pET28a TbPRMT7) was used as a template (provided by Dr.Laurie read University at Buffalo). Desired mutations were confirmed through DNA sequencing. WT TbPRMT7 and variants were expressed in Nico21 DE3 cells (NEB) and protein expression was induced at 0.6-0.8 OD with 0.5 mM IPTG. Cells were harvested by centrifugation and resuspended in lysis/wash buffer (30 mM Tris [pH 8], 0.5M NaCl and 20 mM imidazole). Cells were lysed by sonication and centrifuged at 18,000 RPM at 4 °C for 20 minutes. The resulting crude supernatant was incubated with Ni–Sepharose High Performance resin (BioRad) for 2 h at 4 °C. The resin was washed with wash buffer at increasing concentrations of imidazole (20, 50, 70 mM), and the protein was eluted with elution buffer (30mM Tris [pH8], 0.1M NaCl and 250 mM Imidazole). The elutions were then purified using a Mono Q column (GE Healthcare). Purified proteins were more than 95% pure by SDS–PAGE. Methylation Assays Enzymatic activity toward different peptide substrates was measured with 100 nM TbPRMT7 or 500nM variant enzymes, 1 µM AdoMet (Sigma), and 1 µM [methyl-3H]AdoMet (Perkin Elmer), 10 nM AdoHcy nucleosidase, 50 mM sodium phosphate [pH 7.5], 2 mM DTT, and initiated with 200 µM peptide substrates at 37°C. Radiolabel incorporation over time was measured using a highly sensitive discontinuous kinetic assay (35,36). Briefly, 5 µl of the reaction samples were removed at different time points and quenched with 8 M guanidine HCl, 2.5% trifluoroacetic acid solutions and processed with ZipTipsC18 to purify the methylated peptide from the unreacted radiolabeled AdoMet. Radiolabeled, methylated peptide products were eluted into scintillation cocktail (Fisher Scientific) and counted in a liquid scintillation counter. Reverse phase-HPLC analysis of methylated arginines The three methylated arginine products (MMA, ADMA, and SDMA) were identified as described previously (37). Briefly, assays containing 0.5 µM WT or mutant PRMT1 proteins, 2.67 µM 3H labeled AdoMet, 6.9 µM unlabeled AdoMet, 10 nM AdoHcy nucleosidase, 2mM dithiothreitol, and 50 mM sodium phosphate buffer (pH 7.5) were incubated at 37 °C for 2 min. Reactions were initiated with 200 µM RKK (Ac-GGRGGFGGKGGFGGKW) peptide and were terminated after 4 h with 10% (v/v, final concentration) trifluoroacetic acid (TFA) and incubated at 4 °C for 10 min. TFA precipitated protein was removed through centrifugation and the supernatant (containing the peptide) was added to a glass vial containing 11 M HCl. Vials were sealed and heated to 110 °C overnight for acid hydrolysis. Hydrolyzed amino acids from the enzyme reactions were filtered using a Durapore®-PVDF 0.65 µm centrifugal filter. The amino acids were derivatized using o-phthaldialdehyde and separated with a Gemini 5 µm C18 110Å LC column 250 × 4.6 mm (Phenomenex). Mobile phase A consisted of 40 mM sodium phosphate buffer, pH 7.8, and mobile phase B was acetonitrile/methanol/H2O (45:45:10, v/v). Fractions were 4 ACS Paragon Plus Environment

Page 5 of 24 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

Determinants of TbPRMT7 product specificity collected, and radioactivity was counted in 5 ml of scintillation mixture (Fisher). MMA, ADMA, or SDMA standard amino acids were used to verify the identity of the methylated products generated.

Dissociation Constant Measurement by Intrinsic Fluorescence Quenching A PC1 spectrofluorophotometer model 95110 was used for fluorescence measurements. For AdoMet and AdoHcy affinity determinations, an excitation wavelength of 295 nm was used, and emission at 350 nm was collected. The excitation and emission slit was 0.5 nm. The total reaction volume was 500 µl containing 1.4 uM TbPRMT7 or variants in 150 mM sodium phosphate buffer, pH 7.5 and 2mM dithiothreitol. Increasing concentrations from 1 to 60 µM AdoMet or AdoHcy were added and mixed. Data from at least two titrations were averaged and analyzed using the modified Stern Volmer plots as previously described (37). Data were evaluated by nonlinear regression analysis using Kaleidagraph to obtain the dissociation constant (KD) using the following equation: Fc = F (10ϵcd/2), where Fc is the corrected fluorescence; ϵ is the extinction coefficient of AdoMet; c is the concentration of AdoMet, and d is the path length. Finitial/(Finitial − Fc) was then plotted against 1/[AdoMet], and the data were fit to a linear function where the y-intercept = 1/fa, the slope = 1/fa·KQ, and the KQ = 1/KD. Peptide Dissociation Constant Measurement by Fluorescence The KD values for the TbPRMT7 constructs and RKKK peptides (AcGGRGGFGGKGGCGGKGGFGGKGGFG) were determined using a modified version of the anisotropy assay described by Fang and coworkers (38), where a Cyanine 3 (Cy3) label replaced the fluorescein. The Cy3 group was linked to the cysteine residue of the RKKK peptide, and the arginine residue was either naked, monomethylated, or asymmetrically dimethylated. Each peptide was incubated at a constant concentration (50 nM) with increasing concentrations of eitherWT, Phe71Ser, or Phe71Ile PRMT7 in binding buffer (50 mM NaPO4 PH 7.6, 2 mM DTT). Fluorescence intensity was monitored as a function of PRMT7 concentration using a Biotek Synergy H4 Hybrid Reader with a 550/15 excitation filter, and a 575LP emission filter. Fluorescence intensity values were normalized to the highest and lowest values in each dataset and the KD values were calculated using KaleidaGraph 4.1.1 by iterative fitting to the following equation: I= Bmax x E/ (KD+E), where I is the normalized intensity, Bmax is the predicted fluorescence intensity of peptide when fully saturated with PRMT7, KD is the dissociation constant between the peptide and PRMT7, and E is the concentration of PRMT.

Mass Spectrometry Analysis of Methylated Arginine in RmKK peptide A Nano-RPLC system consisted of a Spark Endurance autosampler (Emmen, Holland) and an ultra-high pressure Eksigent (Dublin, CA) Nano-2D Ultra capillary/nano-LC system, which features low void volume and high chromatographic reproducibility was employed for peptide separation (39) Mobile phase A and B were 0.1% formic acid in 2% acetonitrile and 0.1% formic acid in 88% acetonitrile, respectively. Original samples were diluted to 20 fold with Tris-buffer (50 mM, pH 8.5) and acidified with formic acid. 1 µL sample was loaded onto a trap (300 µm ID×5 mm, packed with Zorbax 5 µm C18 material) with 1% B at a flow rate of 10 µL/min, and washed for 3 min. Then a series of nanoflow gradients (flow rate was 250 nL/min) was used to back-flush the trapped peptides onto Nano-LC column (75 µm ID, 75 cm length, packed with Pepmap 3-µm C18 material) for separation. The column was heated at 52°C to improve both chromatographic resolution and reproducibility. Gradient was as: (1) a linear increase from 3 to 8% B over 5 min; (2) an increase from 8 to 27% B over 65 min; (3) an increase from 27 to 45% B over 30 min; (4) an increase from 45 to 98% B over 20 min; and (5) isocratic at 98% B for 20 min. An LTQ/Orbitrap-ETD hybrid mass spectrometer (Thermo Fisher Scientific, San Jose, CA) was used for methyl cite(s) and types identification of the peptide. The instrument was operated under data-dependent product ion mode. One scan cycle included an MS1 scan (m/z 300-2000) at a resolution of 60,000 followed by MS2 scans by alternating CID and ETD activation to fragment the three most abundant 5 ACS Paragon Plus Environment

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 24

Determinants of TbPRMT7 product specificity precursors detected in the MS1 scan. The AGC (Automatic Gain Control) target for MS1 by Orbitrap was set at 6x106. For CID activation, the activation time was 30 ms, isolation width was 1.5 amu, the normalized activation energy was 35%, and the activation q was 0.25. For ETD activation, the reaction time was set at 110 ms and the isolation width was 2 amu for the precursor and 10 amu for the fluoranthene anions; supplemental activation was employed to enhance the fragmentation efficiency for doubly charged precursors. The AGC value of fluoranthene anions was set at 5x105, and singly charged precursors were rejected for ETD. Spectra were processed using BioWorks software (3.3.1, Thermo Scientific), which incorporates the SEQUEST algorithm. Briefly, the charge states of the precursors of ETD spectra were assigned by the Charger program (Thermo Scientific) and then searched against the database containing the sequence of the RmKK peptide: ac-GGR(CH3)GGFGGKGGFGGKW Mw:1536.7422 (mono) peptide. Differential modifications of MMA (+14.0156 Da) and DMA (+28.0313 Da) on Arg residues, and a static modification of N-terminus acetylation (+42.01057 Da) were considered. Mass tolerances were 10 ppm and 1.0 Da, for the precursor and fragments respectively. A stringent set of criteria were employed to filter the data, including 10 ppm precursor mass tolerance and high Xcorr and delta-CN cut-off values (Xcorr > 1.8 for 1+ charge (CID), > 2.1 for 2+ charge, > 3 for 3+ charge and >4 for 4+ charge, and deltaCN >0.1), probability = 0.05 for CID; and then using the Sf scores (final score, > 0.85) for ETD. In cases where more than one methylation pattern resulted for one ETD spectrum, final confirmation of the most probable assignment was obtained by manual inspection of the spectrum for c and z ions. For putatively identified methylated peptides, the charge states and accurate m/z of precursors were obtained by the Orbitrap, and any identification with incorrect charge state assignment or precursor mass error larger than 10 ppm was eliminated. The sequence-informative ions (b and y ions for CID and c and z ions for ETD) of each identified methylation types were carefully inspected to evaluate the reliability of sequencing and localization of the methyl site(s). Suspected false positives (including these with ambiguous localizations) were eliminated. Based on the database search, the methylation types were identified via characteristic neutral losses under CID activation, combined with high-resolution product ion scan (40). For each methylated peptide identified by ETD, the corresponding CID spectrum was manually inspected to determine the symmetry of the methylation. The Relative Quantification of different MA-Peptides was calculated by extracting ion currents (XIC) of the precursors obtained by Orbitrap (41). For each identified MA-peptide, the XICs were extracted in a narrow m/z window (± 0.02 u) around the monoisotopic m/z for each available charge state. The areaunder-curve (AUC) for each precursor at each charge state was calculated using Qualbrowser (Thermo Scientific), and then calculated the percentage of each product. The calculation of different methylation type of the same molecular weight was based on the spectra count information generated by BioWorks software. Computational enzyme preparation Initial Cartesian coordinates for the Trypanosoma brucei PRMT7 system were generated from a 2.04 Å resolution crystal structure (PDB ID: 4M38)(30). The structure was modeled as a monomer containing amino acids 41-374, with no clear density available for N-terminal region. The co-crystallized AdoHcy was methylated to form (S,S)-AdoMet and an alanine residue was added to the SGRG substrate (ASGRG) bound at the active site. MD simulation protocol Molecular dynamics simulations were performed on WT and mutant TbPRMT7 in 15 different enzyme/substrate combinations. Missing hydrogen atoms were added using the tleap module of the AMBER 16 package. The system was solvated explicitly using an orthorhombic water box of TIP3 model that extended 10 Å beyond the protein. Sodium ions were added to maintain charge neutrality. The ff14SB force field (42) was used to construct the topology files for the protein and peptide, while the 6 ACS Paragon Plus Environment

Page 7 of 24 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

Determinants of TbPRMT7 product specificity parameters for AdoMet were obtained from the generalized AMBER force field (GAFF)(43). For each system, the initial structure was conjugate gradient (CG) minimized for 200 steps for the water molecules and Na+ ions only, followed by 10,000 steps of CG optimization of the entire system to remove any bad contacts. After minimization, the full system was gradually heated from 0 to 300 K over 50 ps of MD with a constant NVT ensemble using the weak-coupling algorithm and a temperature coupling value of 2.8 ps. The system was then switched to a constant NPT ensemble at 300 K and 1 atm using a coupling value of 2.0 ps for both temperature and pressure and ran for 500 ps. The system was returned to NVT and equilibrated for an additional 500 ps. Following equilibration, a 10 ns MD simulation was carried out in order to calculate the boost parameters (44) necessary to perform an accelerated molecular dynamics (aMD) simulation (degrees of freedom boosted per system and their respective values are provided in the Supporting Information Table S3). The aMD method allows for enhanced sampling as compared to unbiased MD(45). Finally, 100 ns of aMD production data was collected at constant NVT for each protein complex using the GPU-accelerated version of AMBER 16 (46). All MD simulations utilized the particle mesh Ewald method to compute the long-range Coulomb force, the SHAKE algorithm to restrict all covalent bonds involving hydrogen atoms, periodic boundary conditions with a non-bonded cutoff distance of 12 Å, and a time step of 1.0 fs. The root-mean-square deviations (rmsd) and root-mean-square fluctuations (rmsf) of the trajectories were calculated for the WT and mutant PRMT7 systems in order to monitor the structural stability of each simulation. The rmsd and rmsf values of the backbone protein atoms are provided in the Supporting Information Figure S1 and yielded steady average values over all 100 ns for all complexes, confirming stable structures.

Results The sequence alignment of the N-terminus of type I and type III PRMTs shows little sequence conservation in this region (residues 1-71) (Fig. S2). One noteworthy amino acid is residue Phe71 in TbPRMT7 (Fig. 3), which is shown in the crystal structure to contribute to the active site, and rotates upon substrate binding to interact with the substrate arginine through van der Waals contact (30). In type I PRMTs the corresponding amino acid at this position is an isoleucine. It has been suggested that the narrow arginine binding pocket entrance of PRMT7 may explain its function as a type III enzyme, due to the steric hindrance produced by some of the bigger residues (32-34). It is therefore possible that Phe71 produces some steric hindrance not present in type I PRMTs, which prevents dimethylation. In order to investigate the impact of individual residues of the active site on product formation, this residue has been previously mutated to an alanine (Phe71Ala). Strikingly, this variant displayed similar levels of MMA as WT TbPRMT7 despite having a smaller side chain at this position, and no ADMA or SDMA product was formed using TbRBP16 as a substrate (33). Interestingly, the corresponding amino acid in the human, mouse and rat PRMT7 sequence is a serine. A serine residue should also allow more space in the active site to accommodate a second methyl group. However, human and mouse PRMT7 have been reported to be exclusive type III PRMTs (20,47). The current data highlight shortcomings of a sterics-only model, which would predict dimethylarginine formation by the Phe71Ala TbPRMT7, and does not explain why human, mouse and rat PRMT7 variants are exclusive type III enzymes. This suggests that other factors influence the exclusive type III activity of this enzyme. These inconsistencies prompted us to use computational approaches to investigate what those other missing forces might be. Therefore, to test if amino acid substitutions with the residues found in the human isoform and type I PRMTs at position 71 would change the exclusive type III activity of PRMT7, accelerated molecular dynamics (aMD) simulations were performed on the WT, Phe71Ser, Phe7Ile, and Phe71Ala enzymes bound with various peptide substrates in their respective methylated arginine states. Computational simulations In the subsequent discussion, the peptide substrate will be simply referred to as Arg. For example, WTArg denotes the WT TbPRMT7 enzyme bound with unmethylated peptide and AdoMet. The protonated 7 ACS Paragon Plus Environment

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 24

Determinants of TbPRMT7 product specificity monomethylated peptide substrate will be represented as MMA-Nη1 when the methyl group is covalently bound at the Nη1 nitrogen and as MMA-Nη2 when bound to the Nη2 nitrogen. Upon formation of MMA, there is the potential for the methyl group to be oriented in four unique positions within the active site, i.e., Nη1-‘up’, Nη1-‘down’, Nη2-‘up’, and Nη2-‘down’ (see Fig.4). Molecular mechanics/Poisson– Boltzmann surface area (MM/PBSA) calculations were performed to compare the binding free energy (∆Gbind) differences between the four MMA conformations in the WT PRMT7 active site (Supporting Information Table S1). The relative ∆Gbind differences computed between the four possible MMA orientations in the active site (Fig. 4) found the MMA-Nη2-‘up’ conformation to be the most energetically favorable in the WT PRMT7; the same MMA-Nη2-‘up’orientation was also energetically preferred in the F71I, F71S, and F71A mutant active sites (Supporting Information Table S2).

Figure 3. Sequence alignment of the F71 region of Trypanosoma brucei PRMT7 to several type III and type I PRMT variants. Identical residues are highlighted in black while similar residues are highlighted in grey. F71 and homologous residues are highlighted in blue. GenbBank accession numbers are given in Fig S2.

8 ACS Paragon Plus Environment

Page 9 of 24 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

Determinants of TbPRMT7 product specificity

Figure 4. The peptide substrate bound in the PRMT7 active site where the methyl group can be chemically transferred to either the Nη1 or Nη2 guanidino nitrogen on arginine.

The thermodynamic stabilization of near-attack conformers (NACs) resembling transition states affects enzymatic catalysis and likely product specificity (48). Bruice has characterized NACs as possessing a reacting atoms distance of 3.2 Å and an approach angle of ±15° of the bonding angle in the transition state for reactions involving O, N, C, and S atoms. In this work, the geometric orientations of the substrate in the active site were monitored to determine the probability of sampling near the SN2 transition states leading to the formation of a second turnover product, i.e., ADMA or SDMA. Evaluation of the entire trajectory was carried out by comparing the distributions of d + 0.5(cosθ) for all systems bound in the MMA-Nη2-‘up’ conformation, where d is the attack distance between the nitrogen atom in arginine (Nη1 or Nη2) and the methyl carbon in AdoMet, and θ is the respective attack angle [N…CH3…S]. An angle of θ = π radians (or 180°) is most ideal for an SN2 reaction and will favorably scale down the final value of d by -0.5. In this way, scaling rewards productive SN2 angles between 90-180° and penalizes θ