SEARCH

SEARCH BY CITATION

Keywords:

  • ribonuclease A;
  • total chemical synthesis;
  • native chemical ligation;
  • solid phase peptide synthesis;
  • X-ray structure

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS AND CONCLUSIONS
  6. CONCLUSIONS
  7. REFERENCES

The total chemical synthesis of RNase A using modern chemical ligation methods is described, illustrating the significant advances that have been made in chemical protein synthesis since Gutte and Merrifield's pioneering preparation of RNase A in 1969. The identity of the synthetic product was confirmed through rigorous characterization, including the determination of the X-ray crystal structure to 1.1 Angstrom resolution. © 2007 Wiley Periodicals, Inc. Biopolymers (Pept Sci) 90:278–286, 2008.

This article was originally published online as an accepted preprint. The “Published Online” date corresponds to the preprint version. You can request a copy of the preprint by emailing the Biopolymers editorial office at biopolymers@wiley.com


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS AND CONCLUSIONS
  6. CONCLUSIONS
  7. REFERENCES

The introduction of solid phase peptide synthesis by Merrifield in 19631 led to the first total chemical synthesis of an enzyme molecule, ribonuclease A (RNase A) in 1969.2, 3 Since that time, solid phase peptide synthesis has been integral in the total chemical synthesis of a number of other enzyme molecules including HIV-1 protease (HIV-1 PR)4; a tethered dimer of HIV-1 PR5; 4-oxalocrotonate tautomerase (4-OT)6; human type II secretory phospholipase A2 (PLA2)7; barnase8; and the electron transport protein cytochrome b562.9 Unlike Gutte and Merrifield's original groundbreaking stepwise synthesis of RNase A,2, 3 all of these enzyme syntheses (except for the synthesis of the 62 residue 4-OT6) have relied on some sort of peptide ligation chemistry in addition to solid phase peptide synthesis.

The use of ligation chemistries is advantageous in the synthesis of enzyme molecules, the smallest of which are typically ∼125 amino acid residues, because stepwise solid phase methods are typically limited to peptides of ∼50 residues. Ligation methods were originally introduced as a modular synthetic approach to overcoming this length limitation10: whatever size peptide can be made by stepwise SPPS, chemical ligation allows the immediate doubling or more of the synthetically accessible size. Chemical ligation involves the chemoselective joining, by means of unique mutually reactive moieties–one on each peptide–of two unprotected peptide segments. Native chemical ligation11 has proven to be the most practical of the ligation chemistries and has found the most use in the total chemical synthesis of proteins. In native chemical ligation a native peptide bond is formed between an unprotected peptide and a peptide-αthioester, both prepared by solid phase peptide synthesis. In this report, we use RNase A, the enzyme originally synthesized by Merrifield, as a model system to demonstrate the utility of chemical protein synthesis using native chemical ligation of synthetic peptides made by stepwise solid phase peptide synthesis.

The synthetic approach presented here uses six peptide segments ranging in size from 11 to 30 amino acid residues to assemble the full 124 amino acid sequence of RNase A. These shorter peptides were more readily prepared by stepwise solid phase peptide synthesis and were less susceptible to solubility issues than longer peptide segments examined in exploratory syntheses. Two consecutive sets of “one-pot” ligations and associated chemical transformations12 were used to assemble these six peptides; this greatly simplified the synthesis and increased overall yields by minimizing the need for isolation and handling of intermediate products. Additionally, the improved aryl thiol catalyst, (4-carboxymethyl)thiophenol13 was used to facilitate faster native chemical ligation reactions. The synthetic protein product, [D83A]Ribonuclease A, was thoroughly characterized by stringent analytical methods and had the correct covalent and three dimensional molecule structure and displayed full enzymatic activity.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS AND CONCLUSIONS
  6. CONCLUSIONS
  7. REFERENCES

Materials

2-(1H-Benzotriazol-1-yl)-1,1,3,3-tetramethyluronium hexafluorophosphate (HBTU), S-trityl-mercaptopropionic acid, L-cystine·2 HCl, and Boc-amino acids were obtained from Peptides International (Louisville, KY), except Boc-(4R)-1,3-thiazolidine-4-carboxylic acid (Boc-Thz) which was obtained from NovaBiochem, San Diego. Boc-aminoacyl-OCH2-Pam-resins were obtained preloaded from Applied Biosystems (Foster City), or synthesized from the Boc-aminoacyl-(4-carboxymethyl)benzyl esters (NeoMPS, San Diego) and aminomethyl-resin (Rapp Polymer, Germany) as described previously.14 Side chain protecting groups used were: Arg(Tos), Asp(OcHex), Asn(Xan), Cys(4MeBzl), Glu(OcHex), His (Bom), Lys(2ClZ), Ser(Bzl), Thr(Bzl), and Tyr(BrZ). N,N-diisopropylethylamine (DIEA) was obtained from Applied Biosystems, Foster City. N,N-Dimethylformamide (DMF), dichloromethane, and HPLC grade acetonitrile was purchased from VWR. Diethyl ether was purchased from Fisher. RNase-free buffer reagents (MES, NaCl, NaOH, and water) from Fisher were used in the enzyme assays and kinetic studies. Trifluoroacetic acid (TFA) was obtained from Halocarbon Products, NJ. HF was purchased from Matheson Tri-Gas. Tris(2-carboxyethyl)phosphine hydrochloride (TCEP), p-cresol, triisopropylsilane, (4-carboxymethyl)thiophenol (4-mercaptophenylacetic acid, MPAA), polycytidylic acid [poly(C)], tris buffer reagents, L-cysteine, cytidine 2′,3′-cyclic phosphate (C > p), and native RNase A type XII-A were purchased from Sigma-Aldrich.

Peptide Synthesis

Peptide-αCOSCH2CH2CO-Leu (“peptide-αCOSR”) thioesters were synthesized on a HSCH2CH2CO-Leu-OCH2-Pam-resin, as described previously.15 Arginine6-tagged segment 65–83 was synthesized on Boc-Arg-OCH2-Pam-resin to which five arginines and a HSCH2CH2CO- moiety were coupled. Peptides were synthesized by stepwise solid phase synthesis on a 0.5 mmol scale, using the manual Boc chemistry “in situ neutralization”/HBTU protocol described previously.16 After chain assembly was complete, the Nα-Boc group was removed with TFA, and the peptide-resin was washed with dichloromethane and dried under a stream of nitrogen. The peptide was cleaved from the resin and the side-chain protecting groups were simultaneously removed by treatment with anhydrous HF at 0°C for 1 h, with 5–10% (v/v) p-cresol added as a scavenger. After thorough removal of HF by evaporation under reduced pressure, the peptide was precipitated by addition of cold diethyl ether, triturated with cold diethyl ether, then dissolved in 50% aqueous acetonitrile + 0.1% TFA, and lyophilized.

Peptide Analysis and Purification

Analytical reverse-phase HPLC was performed on an Agilent 1100 system with either a Microsorb C-18 (5 μm, 300 Å, 2.1 mm × 50 mm) silica column packed in-house or a Vydac C-4 (5 μm, 300 Å, 2.1 mm × 50 mm) silica column. A 4% per minute gradient of buffer B (CH3CN + 0.08% TFA) versus buffer A (H2O + 0.1% TFA) was used at a flow rate of 250 μL per minute. The column effluent was monitored at 214 nm. Where applicable, peptide masses were obtained using on-line electrospray mass spectrometric detection (Agilent 1100 Series LC/MSD Trap). Preparative HPLC was performed on C-4 or C-18 columns (sizes 10 mm × 250 mm, 22 mm × 250 mm, 50 mm × 250 mm) from Vydac, depending on peptide quantities and retention characteristics. A preliminary analytical run was performed to verify the integrity of the preparative column and to establish the elution behavior of the desired peptide. The preparative separation was performed by pumping the acidified peptide solution onto the column at a fixed concentration of buffer B in buffer A until all nonadsorbed components had eluted and a smooth baseline had been reestablished. The separation was then carried out with a shallow gradient of buffer B versus buffer A (0.3% B per minute over a 30% range of buffer B, starting 15% below the expected elutropic concentration of B). Fractions containing the desired purified product were identified by analytical LC/MS, or by analytical HPLC and MALDI-TOF MS, and were combined and lyophilized.

Synthesis of the Polypeptide [D83A]RNase(SH)6

The synthesis of the full length 124 amino acid polypeptide chain of the target [D83A]RNase A was achieved using two sets of sequential one-pot native chemical ligations, starting from the C-terminal segment.12 Briefly, for the first series of one pot ligations approximately equimolar amounts of RNase A(Thz84-94)-αCOSR (42.6 mg, 29.3 μmol) and (Cys95-124)RNase A (91.0 mg, 27.8 μmol) were dissolved to a concentration of 4 mM in 6M guanidine·HCl, 0.1M Na phosphate pH 7.0 containing 200 mM (4-carboxymethyl)thiophenol (“MPAA”) and 20 mM TCEP–termed “ligation buffer”. Upon complete reaction (∼2 h), the N-terminal Thz-moiety of the product was converted to a Cys-moiety by addition of 0.2M methoxylamine· HCl, adjustment of the pH to 4.0, and overnight reaction. The next ligation was performed in the same reaction solution simply by readjusting to pH 7.0, adding 1.3 equivalents (113.8 mg, 36.3 μmol) of RNase A(Thz65-A83)-αCOSR' (containing a C-terminal Arg6 tag in the thioester leaving group) and an additional 20 mM TCEP (no additional MPAA was added; MPAA is not used up, and TCEP keeps it all reduced). Upon complete reaction (∼5 h), the Thz65-124 polypeptide was converted to Cys65-124 overnight by lowering the pH to 4.0 without additional methoxylamine·HCl. The product, (Cys65-124)RNase A, was purified by preparative HPLC and lyophilized.

The second series of one-pot ligations was initiated by dissolving RNase A(Thz40-64)-α (59.3 mg, 20.8 μmol) and (Cys65-124)RNase A (113.5 mg, 17.2 μmol) to a concentration of ∼2.5 mM in ligation buffer. After 4 h reaction, the Thz40-124 product was converted to Cys40-124 by adding 0.2M methoxylamine·HCl at pH 4.0 and overnight reaction as described earlier. The next ligation was initiated by adjusting the solution to pH 7.0, adding 56.3 mg (29.9 μmol) of RNase A(Thz26-39)-αCOSR, and an additional 20 mM TCEP (no additional MPAA). After ∼8 h reaction, the product Thz26-124 was converted to Cys26-124 by lowering the pH of the reaction to 4.0 without additional methoxylamine·HCl and overnight reaction. The final ligation was initiated by adjusting the solution to pH 7.0 and adding 54.6 mg (18.79 μmol) of RNase A(1-25)-αCOSR and an additional 20 mM TCEP; no additional MPAA was added. The full length product RNase A(1-124)(SH)6 was purified from the completed ligation reaction by preparative HPLC.

Folding/Disulfide Formation to Give [D83A]Ribonuclease A

Purified polypeptide [D83A]RNase A(1-124)(SH)6 was dissolved in 6M guanidine·HCl, 0.1M Tris pH 8.0 to a concentration of 0.5 mM (6.8 mg/mL). This solution was diluted with a buffered solution containing L-cysteine and L-cystine·2 HCl to final concentrations of 0.5M guanidine·HCl, 0.1M Tris pH 8.0, 8 mM L-cysteine, 1 mM L-cystine·2 HCl, 0.04 mM (0.57 mg/mL) [D83A]RNase A(1-124)(SH)6 and was stirred overnight. The reaction was monitored by analytical HPLC. Folded [D83A]RNase A (12.6 mg, 0.92 μmol) was purified from the reaction mixture by preparative HPLC, and was characterized by LC/MS.

Chemical Characterization of [D83A]RNase A(1–124)(SH)6 and [D83A]Ribonuclease A

Both the full length polypeptide chain and the folded protein molecule were characterized by analytical LC/MS using an Agilent 1100 Series LC/MSD Ion Trap instrument. The known mass of natural RNase A was used to calibrate the mass spectrometer in the mass range used for these measurements.

Enzyme Assays/Kinetics of [D83A]RNase A

All kinetics measurements were carried out on an Agilent 8453 UV-Vis diode array spectrophotometer. Initial velocities were determined using the manufacturer-supplied ChemStation software. Enzyme was dissolved in 10 mM MES/NaOH buffer, pH 6.0 and was dialyzed to a concentration of 100 mM NaCl. The final solutions were filtered through a 0.2 μm nylon filter using a syringe. Enzyme concentrations were determined using an extinction coefficient of 0.98 × 104M−1 cm−1 at λ 277.5 nm.17 The values of kcat, KM, and kcat/KM were determined from initial velocity data using the Enzyme Kinetics extension of SigmaPlot 9.0 (Systat Software).

First Step Assay

Steady state kinetic parameters for the cleavage of poly(C) were determined using the procedures and extinction coefficients described previously.18 Poly(C) was purified by precipitation from aqueous ethanol (70% v/v) before use. Assays were performed at 25°C in 0.1M MES/NaOH pH 6.00 containing NaCl (0.1M), substrate (3.8–130 μM), and enzyme (4.8–5 nM); the reaction was monitored at λ 250 nm.

Second Step Assay

Steady state kinetic parameters for the hydrolysis of C > p were determined using a modification of the procedures previously reported.19 Assays were performed at 25°C in 0.1M MES/NaOH pH 6.00 containing NaCl (0.1M), substrate (75 μM–1.5 mM), and enzyme (410–730 nM) and monitored at λ 290 nm. The measured Δε coefficient for this reaction was 792 M−1 cm−1 at λ 290 nm.

Crystallization, X-ray Data Collection, and Structure Refinement for Native RNase A and Synthetic [D83A]RNase A

RNase A enzyme solutions for crystallization were made by dissolving enzyme to final concentrations of 10–20 mg/mL in deionized water. All crystals were grown using hanging drop vapor diffusion methods. RNase A crystals were obtained from drops consisting of 2 μL enzyme solution and 2 μL crystallization solution. The crystallization solution used for native RNase A consisted of 30% saturated ammonium sulfate, 3M NaCl, 0.1M sodium acetate, pH 5.5. The crystallization solution used for synthetic [D83A]RNase A was made by mixing Hampton Index Reagents 3 and 9 (Hampton Research, Aliso Viejo, CA) in a 1:1 ratio. The resulting solution was 0.1M bis-tris, 1.0M ammonium sulfate, 1.5M NaCl, with a pH of 5.89. Crystals of native RNase A were observed within 3 weeks. Crystals of synthetic [D83A]RNase A took considerably longer to grow (months). Both enzymes crystallized in the same P 32 2 1 space group and were isomorphous to the previously published trigonal structure of [F46L]RNase A (PDB code 1IZP).20 For low temperature data collection, crystals were transferred to the cryoprotectant (paraffin oil) for a few seconds and flash-frozen in liquid nitrogen. Atomic resolution X-ray data were collected using 0.5° beam oscillation width at BioCARS station 14 BM-C (native RNase A, λ = 0.90020 Å, 10 s exposure time) and at GM/CA CAT station 23ID-D (synthetic [D83A]RNase A, λ = 0.97934 Å, 2 s exposure time) at the Advance Photon Source (APS), Argonne, IL. Images were processed and scaled with HKL2000.21 Both structures were solved by molecular replacement using MOLREP22 with the 1IZP structure as a starting model. The rigid body refinement, restrained positional and anisotropic temperature factor refinement, as well as search for water molecules were carried out in REFMAC523 implemented in CCP4.24 A Ramachandran plot calculated with PROCHECK25 indicates that 100% of the non-Gly and non-Pro residues in the final models lie in the most favored and additional allowed regions. The SigmaA-weighted 2Fo-Fc and Fo-Fc Fourier maps were calculated in CCP4.24 The Fourier maps were displayed and examined in TURBO-FRODO.26 Alignment of the two structures was performed with CCP4.24 Figures showing the electron maps and three-dimensional structures were prepared using TURBO-FRODO26 and the UCSF Chimera package,27 respectively. Chloride ions were located using anomalous X-ray data collected at 1.54980 Å on beam line 23-ID at the APS. Anomalous difference maps were generated with CNS1.128 and displayed in TURBO-FRODO26 to show the location of the chloride ions. The coordinates and structure factors have been deposited in the Protein Data Bank with entry codes 2E3W (native RNase A) and 2NUI (synthetic [D83A]RNase A).

RESULTS AND CONCLUSIONS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS AND CONCLUSIONS
  6. CONCLUSIONS
  7. REFERENCES

Bovine pancreatic RNase A is a small enzyme molecule that catalyzes the hydrolytic cleavage of RNA phosphodiester bonds. The protein has 124 amino acid polypeptide chain (Figure 1) that contains eight cysteine residues which form four disulfide bonds, providing ample options for thioester-mediated native chemical ligation at cysteine residues.11 In exploratory syntheses we evaluated several possible synthetic schemes using a variety of these potential ligation sites.

thumbnail image

Figure 1. Amino acid sequence of [D83A]RNase A. The enzyme molecule contains eight cysteines which form four disulfide bridges in the folded enzyme molecule. Native chemical ligation sites are underlined in bold italics.

Download figure to PowerPoint

Solid Phase Peptide Synthesis

Our first synthetic scheme required the synthesis of just two peptides (residues 1–64 and 65–124). These two target sequences were prepared by highly optimized, stepwise solid phase synthesis using the “in situ neutralization” protocol for tert-butoxycarbonyl (Boc) chemistry SPPS.16 The resulting crude products showed the statistical accumulation of byproducts characteristic of attempts to make long peptides by stepwise methods without purification of the resin-bound products, and were not of high enough quality to be purified to molecular homogeneity. These two peptides were each divided in half, resulting in a four-peptide synthetic scheme (peptide segments comprising residues 1–25, 26–64, 65–94, and 95–124). Unexpected synthetic difficulties were encountered during the synthesis of two of these peptides (residues 26–64 and 65–94). These difficulties were ascribed to sequence-dependent intermolecular aggregation of resin-bound peptide intermediates during peptide synthesis.29, 30 This interpretation was confirmed by pilot syntheses incorporating Pro residues at selected positions to disrupt aggregation arising from intermolecular beta sheet-type hydrogen bonding (Figure 2). Based on our understanding of the molecular mechanism of these difficult stepwise solid phase syntheses, a number of tactics were attempted to overcome the problems, including the incorporation of pseudo-proline residues.31 We were unable to find a useful way of making these two peptide-thioester segments in sufficient purity.

thumbnail image

Figure 2. (a) LC/MS analysis of crude RNase A(65–94)-αCOSR. (b) LC/MS analysis of crude [C72P, C84P]RNase A(65–94)-αCOSR. A significant improvement in the quality of the crude product resulted from the incorporation of the structure-disrupting amino acid, proline, at positions 72 and 84. This confirmed that the difficulty in synthesis was due to sequence-dependent intermolecular aggregation of resin-bound peptide intermediates during solid phase synthesis.

Download figure to PowerPoint

In the final successful synthetic scheme (Scheme 1), each of the difficult peptides was further divided in half, resulting in a six peptide segment synthetic scheme (residues 1–25, 26–39, 40–64, 65–83, 84–94, 95–124). Unprotected peptide segments were prepared by highly optimized, stepwise solid phase synthesis using the manual “in situ neutralization” protocol for tert-butoxycarbonyl (Boc) chemistry,16 which is optimal for the production of peptide αthioesters. Internal peptides (i.e. 26–39, 40–64, 65–83, 84–94) that contained both an N-terminal Cys and a C-terminal αthioester had their N-terminal cysteine protected as a (4R)-1,3-thiazolidine-4-carboxylic acid (Thz)32 to prevent undesired side-product formation from cyclization33 and oligomerization under ligation conditions. To overcome limited solubility, the peptide Thz65-83-COαthioester was synthesized with a hydrophilic tag containing six arginine residues in the thioester leaving group.34 The crude and purified synthetic peptides prepared by stepwise solid phase peptide synthesis are shown in Figure 3.

thumbnail image

Figure 3. Analytical HPLC analyses (214 nm) of the six peptide and peptide-αthioesters used in the synthesis of [D83A]RNase A. The peptides are shown before (top chromatogram), and after (bottom chromatogram) purification.

Download figure to PowerPoint

thumbnail image

Scheme 1. Strategy for the total chemical synthesis of [D83A]RNase A. Two sets of sequential one-pot native chemical ligation reactions gave the full length 124 residue polypeptide chain, which was folded with concomitant formation of disulfides. Peptide building blocks were synthesized by manual “in situ neutralization” protocols for Boc chemistry stepwise solid phase peptide synthesis. N-terminal cysteines on peptidethioesters were protected as Thz. Peptide 65–83 has a 6-arginine tag in the thioesterleaving group, to aid solubility. Peptides were ligated sequentially from C- to N-terminus. NCL refers to a native chemical ligation reaction, -Thz refers to the conversion of Thz to cysteine. The purified product from one-pot ligation series #1 was used as the starting material for one-pot ligation series #2. After purification, the full length 124 residue polypeptide was folded to generate the correct tertiary structure containing four disulfides.

Download figure to PowerPoint

Chemical Synthesis of the Protein Molecule [D83A]Ribonuclease A

Our target was the protein molecule [D83A]RNase A. The recombinant mutant enzyme [D83A]RNase A has been previously reported, and was found to catalyze reactions of cytidine-containing substrates with kinetic parameters similar to those of wild-type RNase A.35 This analogue protein was made in order to prevent a side reaction that is known to occur during native chemical ligation at Asp-Cys sites.36 The C-terminal half of the target polypeptide (i.e. 65–124) was assembled by the sequential one-pot ligation12 of peptides 95–124, 84–94, and 65–83, followed by HPLC purification of the resulting product polypeptide Cys65-124. In all native chemical ligation reactions, the water soluble, odor free, improved aryl thiol catalyst (4-carboxylmethyl)thiophenol (mercaptophenylacetic acid, MPAA)13 was used at high concentrations (200 mM), resulting in rapid, high-yield reactions. The assembly of the full length sequence was completed in a second series of one-pot ligation reactions in which peptides 40–64, 26–39, and 1–25 were sequentially added to 65–124. HPLC purification gave full length polypeptide RNase A(1–124)(SH)6 (observed mass: 13,647 ± 2 Da; calculated mass 13,646.2 Da (average isotope composition)).

After purification, the full-length polypeptide was folded in 0.5M guanidine·HCl in the presence of a standard cysteine–cystine redox system37, 38 to give the correctly folded protein molecule in good yield (Figure 4). The synthetic protein had an observed mass of 13,639 ± 2 Da (calculated mass 13,638.2 Da (average isotopes), Figure 5); the observed loss of 8 Da on folding is consistent with the formation of four disulfides. A final HPLC purification step gave the synthetic protein. The synthesis was repeated a number of times in order to optimize recoveries and yields. A single synthesis gave ∼10 mg amounts of purified protein, more than sufficient to carry out the subsequent kinetic and structural characterizations.

thumbnail image

Figure 4. Analytical reverse phase HPLC chromatograms (monitored at 214 nm) of three timepoints in the folding of [D83A]RNase A. Analysis was performed on a Microsorb C-18 column packed in-house.

Download figure to PowerPoint

thumbnail image

Figure 5. Analytical reverse phase HPLC of purified synthetic [D83A]RNase A performed on a Microsorb C-18 column. (Inset) electrospray mass spectrum of synthetic [D83A]RNase A.

Download figure to PowerPoint

Characterization of Synthetic [D83A]RNase A

Complete characterization of the synthetic product requires demonstration of structural and functional characteristics that match those of the native enzyme. The synthetic protein had acceptable purity as shown by analytical HPLC, and within experimental uncertainty had the correct mass as shown by ESI-MS.

The crystal structure of synthetic [D83A]RNase A was determined by X-ray diffraction to atomic resolution (1.10 Å) (Figure 6). For comparison purposes, the structure of natural RNase A (Sigma) was also determined to a similar resolution. The crystal data, X-ray data collection, and refinement statistics for both ribonuclease A molecules are shown in Table I. The molecular structures of the synthetic and native enzymes were essentially identical and are shown superimposed in Figure 6a. In both enzymes the active sites are empty of substrate ligand and contain water molecules and a previously observed conserved chloride ion involved in a hydrogen bond with the main chain nitrogen of Thr45 and in a possible electrostatic interaction with the side chain of Lys41 (Figure 6b). The only significant difference between the two enzyme structures is the deletion of the aspartic acid 83 carboxylic acid side chain group in the synthetic enzyme. The two deleted oxygen atoms are replaced in the synthetic enzyme structure with two well-defined water molecules located in approximately the same position in van der Waals contact with the Ala83 methyl group (Figure 6c). In the synthetic [D83A]RNase A structure, these two water molecules replace the Asp83 side chain in making hydrogen bonds to the side chains of Thr45 and Arg85. Additionally, the correct disulfide bonds can be directly observed in the atomic resolution X-ray structure of synthetic [D83A]RNase A (Figure 6d).

thumbnail image

Figure 6. (a) Structural alignment of the polypeptide backbone of natural RNase A (blue) and synthetic [D83A]RNase A (red) crystallized in the P 32 2 1 space group. (b) Both molecules contained a conserved chloride ion (magenta) in the active site, as well as a number of conserved water molecules. (c) Electron density map of residue 83 in the natural and synthetic structures. In the synthetic [D83A]RNase A structure the side chain oxygens of aspartic acid 83 are replaced by two well-defined water molecules which replace the natural amino acid in hydrogen bonding interactions with the side chains of Thr45 and Arg85. (d) Electron density map of the disulfide bond between Cys26 (lower) and Cys84 (upper) in the structure of synthetic [D83A]RNase A. The distance between the two sulfur atoms was measured at 2.03Å. The outer electron density contour is drawn at 1 σ, and the inner contour is drawn at 2 σ.

Download figure to PowerPoint

Table I. Data Collection and Refinement Statistics for RNase A Molecules in the Hexagonal Crystal Form
 Natural RNase ASynthetic [D83A] RNase A
  • a

    Highest resolution shell is shown in parenthesis.

Data Collection  
 Space groupP 32 2 1P 32 2 1
 Cell dimensions  
  a = b (Å)64.16763.59
  c (Å)63.62663.365
  α = β (°)9090
  γ (°)120120
 Resolution (Å)a50–1.05 (1.09–1.05)50–1.05 (1.09–1.05)
 Rmerge0.067 (0.371)0.046 (0.429)
 II20.7 (1.9)29.9 (1.7)
 Completeness, %98.6 (all)94.4 (all)
  1.18–1.13 Å98.397.9
  1.13–1.09 Å97.786.7
  1.09–1.05 Å96.663.6
 Redundancy4.1 (3.1)2.6 (1.3)
Refinement  
 Resolution (Å)20–1.0520–1.10
 No. reflections  
  Work/free set66,348/352356,514/2998
 Rwork/Rfree19.3/20.817.7/19.3
 No. Residues  
  Protein124124
  Chloride33
  Bis-Tris01
  Water178215
 B-factor (Å2)15.4214.452
 R.m.s. deviations  
  Bond lengths (Å)0.0140.014
  Bond angles (°)1.571.63

The expected ribonuclease catalytic activity was observed for the synthetic [D83A]RNase A. Full kinetic characterization of the first and second steps of RNase A catalysis was carried out, using the transphosphorylation of polycytidylic acid [poly(C)] and hydrolysis of cytidine 2′,3′–cyclic phosphate (C > p),19 respectively. Kinetic parameters obtained for the cleavage of poly(C) and the hydrolysis of C > p by synthetic [D83A]RNase A and by wild type (WT) RNase A (Sigma) are given in Table IIA. Literature values35 reported for the cleavage of the same two substrates by recombinant [D83A]RNase A and wild type (WT) RNase A, assayed under similar conditions, are given in Table IIB). The kcat values obtained for the synthetic and recombinant [D83A]RNase A are in reasonable agreement, given that the assays were performed a decade apart in different laboratories using different enzyme preparations. The KM values for the synthetic [D83A]RNase A were significantly lower than the KM values reported for the recombinant enzyme, resulting in kcat/KM values (normally taken as a measure of catalytic efficiency) ∼3-fold higher for the synthetic enzyme. These differences are probably the result of minor differences in assay conditions; this is supported by the ∼2-fold higher kcat/KM values observed for WT RNase A in our measurements when compared with the reported values.35 Complete details of our assay conditions are given in Materials and Methods. As was observed by delCardayre and Raines35 for recombinant [D83A]RNase A, the synthetic [D83A]RNase A had kcat/KM values for hydrolysis of the poly(C) and C>p substrates that were similar to WT RNase A.

Table 2. (A) Measured Steady-State Kinetic Parameters for the Cleavage of Poly(C) and the Hydrolysis of C > p by WT RNase A (Sigma) and by Synthetic [D83A]RNase A
RNase ASubstratekcat (s−1)KMM)kcat/KM (×106M−1 s−1)
  • a

    Data taken from delCardayré and Raines, 1995.35

WTpoly(C)652 ± 1021.1 ± 1.730.9 ± 3
 C > p2.48 ± 0.05312 ± 160.0080 ± 0.0004
Synthetic [D83A]poly(C)198 ± 35.1 ± 0.539 ± 4
 C > p2.0 ± 0.07285 ± 200.0070 ± 0.0006
(B) Literature Values for the Steady-State Kinetic Parameters for the Cleavage of Poly(C) and the Hydrolysis of C > p by WT RNase A and by Recombinant [D83A]RNase A
RNase ASubstratekcat (s−1)KMM)kcat/KM (×106M−1 s−1)
WTapoly(C)520 ± 1034 ± 0.215 ± 1
 C > p3.7 ± 0.71000 ± 6000.004 ± 0.002
Recombinant [D83A]apoly(C)240 ± 1022 ± 411 ± 2
 C > p2.5 ± 0.21200 ± 3000.0023 ± 0.0005

CONCLUSIONS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS AND CONCLUSIONS
  6. CONCLUSIONS
  7. REFERENCES

In this article, we have reported a practical and reproducible total chemical synthesis of the enzyme [D83A]Ribonuclease A by a combination of stepwise solid phase peptide synthesis and native chemical ligation methods. The resulting synthetic protein has been rigorously characterized in terms of catalytic activity, and both covalent and three-dimensional structures. The synthetic methods presented here illustrate the significant advances that have been made in chemical protein synthesis since Gutte and Merrifield's pioneering preparation of RNase A in 1969.

We have recently extended the total chemical synthesis reported here to design and make novel analogues of Ribonuclease A that incorporate non-natural, non-coded amino acids (unpublished data). These enzyme analogues will be used to illuminate facets of the mechanism and structure–activity relationships of this classical enzyme molecule that are not accessible by traditional biochemical means.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS AND CONCLUSIONS
  6. CONCLUSIONS
  7. REFERENCES