HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 275, Issue 22, 16408-16413, June 2, 2000

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Poland, B. W. Articles by Quiocho, F. A. Search for Related Content PubMed PubMed Citation Articles by Poland, B. W. Articles by Quiocho, F. A. Social Bookmarking                      What's this?

Structural Insights into the Protein Splicing Mechanism of PI-SceI^*

Bradley W. Poland, Ming-Qun Xu^§, and Florante A. Quiocho^¶

From the  Howard Hughes Medical Institute and Department of Biochemistry, Baylor College of Medicine, Houston, Texas 77030 and ^§ New England Biolabs, Inc., Beverly, Massachusetts 01915

Received for publication, January 11, 2000, and in revised form, February 4, 2000

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION REFERENCES

PI-SceI is a member of a class of proteins (inteins) that excise themselves from a precursor protein and in the process ligate the flanking protein sequences (exteins). We report here the 2.1-Å resolution crystal structure of a PI-SceI miniprecursor (VMA29) containing 10 N-terminal extein residues and 4 C-terminal extein residues. Mutations at the N- and C-terminal splicing junctions, blocking in vivo protein splicing, allowed the miniprecursor to be purified and crystallized. The structure reveals both the N- and C-terminal scissile peptide bonds to be in distorted trans conformations (  100°). Modeling of the wild-type PI-SceI based on the VMA29 structure indicates a large conformational change (movement of >9 Å) must occur to allow transesterification to be completed. A zinc atom was discovered at the C-terminal splicing junction. Residues Cys⁴⁵⁵, His⁴⁵³, and Glu⁸⁰ along with a water molecule (Wat⁵³) chelate the zinc atom. The crystal structure of VMA29 has captured the intein in its pre-spliced state.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION REFERENCES

Posttranslational autoprocessing involving peptide bond rearrangement has been observed in a variety of different systems including protein splicing, autoprocessing of hedgehog proteins, autocleavage of amidohydrolase, and pyruvoyl enzyme formation (1). Protein splicing element was first discovered in the VMA¹ gene of Saccharomyces cerevisiae (2). The chemical mechanism of protein splicing has been extensively studied (3-7). Protein splicing begins with an acyl rearrangement at the N-terminal junction whereby the conserved Ser or Cys attacks the carbonyl carbon of the preceding residue forming an ester or thioester intermediate (Fig. 1a). The conserved Cys/Ser/Thr residue at the C-terminal splicing site attacks the ester/thioester intermediate resulting in the formation of the branched intermediate. The next step couples the cyclization of a conserved Asn adjacent to the C-terminal splicing junction with the cleavage of the peptide bond to the branched ester (thioester) intermediate. The excised intein containing an aminosuccinimide residue at the C terminus and the ligated N- and C-terminal exteins are formed. Finally, a spontaneous O  N/S  N acyl rearrangement of the branched ester linkage completes the splicing event.

View larger version (20K): [in this window] [in a new window]   Fig. 1.   a, schematic diagram of the chemical mechanism for protein splicing. The four-step reaction couples the excision of the intein (red) from the precursor protein with the ligation of the two exteins (blue and blue-green) via a native peptide bond. b, diagram of the new motif structure of PI-SceI according to Pietrokovski (9). Blocks N1-N4 and C1 and C2 (blue) contain residues involved in protein splicing; blocks EN1-EN4 (black) contain residues associated with the endonuclease/linker domain. Nucleophilic residues are highlighted below the block diagram (yellow letters in purple box), and highly conserved residues are shown in red. c, sequence alignment of the wild-type PI-SceI (VMA) and mutant miniprecursor (VMA29). The red dash and arrow illustrate the N- and C-terminal splicing sites. Red residues in the VMA29 sequence indicate mutations made to the wild-type sequence (blue). Cys1 and Asn454 were mutated to Ala in order to block in vivo protein splicing.

Currently, over 100 protein splicing elements, inteins, have been deposited into the New England Biolabs Intein Data base (8). The primary amino acid sequence of the inteins can be divided into three conserved domains, N-terminal, C-terminal, and an optional endonuclease (9). The N-terminal domain is subdivided into four motifs, N1-N4. Likewise the C-terminal and endonuclease domains are divided into two and four motifs, C1 and C2 and EN1-EN4, respectively (Fig. 1b). The splicing motifs are located in blocks N1-N4, and C1 and C2 with strictly conserved residues (His⁷⁹ and Asn⁴⁵⁴) residing in blocks N3 and C1, respectively. The penultimate histidine of PI-SceI is highly conserved (>90% of the inteins contain this amino acid) but is not essential for splicing of the PI-SceI intein (3, 10, 11). Asn⁴⁵⁴ and Cys⁴⁵⁵, however, are absolutely necessary for protein splicing (3, 10, 11).

Crystal structures exist for two of these inteins, PI-SceI homing endonuclease (12) and GyrA splicing domain (13) as well as two mechanistically related proteins, hedgehog autoprocessing domain (14) and glycosylasparaginase (15). The four structures all exhibit the -scaffold termed the HINT (Hedgehog, INTein) module (14). The GyrA structure contains a single Ala⁰ preceding the mutated Cys¹  Ser¹ of the intein. The peptide linkage between Ala⁰ and Ser¹ is in the energetically less favorable cis conformation with the hydroxyl group of Ser¹ positioned for a nucleophilic attack on the carbonyl carbon of Ala⁰ (13). The structure suggests Thr⁷², His⁷⁵, and His¹⁹⁷ play catalytic roles in the splicing reaction.

Glycosylasparaginase activates itself by an N  O acyl shift followed by hydrolysis of the ester linkage (15). As a result, the N-terminal side of a loop covering the active site is cleaved to allow the substrate to enter. In contrast to the GyrA structure, the structure of glycosylasparaginase has revealed a high energy distorted trans peptide at the N-terminal junction (15). Residues involved in the N  O acyl rearrangement (Asp¹⁵¹ and Thr¹⁵²) form an unusual tight turn that forces the scissile peptide bond from an  = 180° (planar conformation) to an  <160°. Along with the main chain dihedral angle distortion, the  angle of Asp¹⁵¹ deviates ~9° from the ideal 110°. The authors conclude that these main chain distortions around the cleavage site could raise the energy 5 kcal/mol for each distorted residue.

The crystal structures of GyrA and glycosylasparaginase give structural support to the proposed splicing mechanism. However, the former lacks a full complement of residues from the N- and C-exteins including the absolutely required Thr¹⁹⁹ at the C-terminal splicing junction. The latter structure seems applicable only to the initial acyl rearrangement step in the protein splicing pathway. In the current study, we have cloned and expressed a mutant of the VMA1 intein (identified as VMA29) in which Cys¹ and Asn⁴⁵⁴ have been mutated to Ala in order to block protein splicing. In addition, 10 residues (MKAEEGKLEG) have been added to the N terminus of the intein as well as four residues (CGER) to the C terminus (Fig. 1c). This structure is the first example of a true precursor of an intein.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION REFERENCES

Protein Expression and Purification-- Plasmid pVMA29 encoding the VMA mutant C1A/N454A (VMA29) with the sequence MKAEEGKLEG positioned upstream to the N-terminal cleavage site and the sequence CGER positioned downstream to the C-terminal splicing site was used to transform the Escherichia coli strain B834(DE3), a methionine auxotroph. Cells containing the pVMA29 were grown in LeMaster media to an A₆₀₀ = 0.5, induced with 0.5 mM isopropyl-1-thio--D-galactopyranoside, and grown for an additional 15 h at 20 °C. Cells were lysed by the addition of 20 mg of lysozyme to cells suspended in 50 mM NaCl, 20 mM HEPES, pH 7.0. After 30 min incubation at 20 °C, the sample was pulse-sonicated for 3 min. The lysate was clarified by centrifugation at 18,000 rpm for 30 min in a Beckman J25 centrifuge. Clarified lysate was filtered through a 0.22-µm filter and loaded directly onto a Porus Heparin column. Samples containing the 50-kDa VMA29 protein were eluted from the column with a gradient run from 50 mM NaCl to 1 M NaCl. These fractions were pooled and concentrated, and the buffer was exchanged to 50 mM NaCl, 20 mM HEPES, pH 7.0. Pooled fractions were loaded onto Source 15S column from Amersham Pharmacia Biotech and eluted with 50 mM to 1 M NaCl gradient. Fractions containing VMA29, as determined by SDS-polyacrylamide gel electrophoresis, were pooled and loaded on to an S200 gel filtration column equilibrated with 500 mM NaCl, 20 mM HEPES, pH 7.0. SDS-polyacrylamide gel electrophoresis was used to locate the fractions containing the mutant protein; these fractions were pooled, concentrated, and the buffer exchanged to 500 mM NaCl, 20 mM HEPES, pH 7.0. Light-scattering experiments performed on these samples (data not shown) indicated the protein sample was monodispersed with an apparent molecular mass of 115 kDa.

Crystallization and Data Collection-- VMA29 was crystallized using the hanging drop vapor diffusion method. Crystals were grown from droplets containing 2 µl (31 mg/ml) of protein and 2 µl of polyethylene glycol 3350 (10% w/v, 3 mM CdCl₂, 1 mM MgCl₂, 10 mM -mercaptoethanol, 100 mM Tris-HCl, pH 8.5). The resulting 4-µl droplet was equilibrated against a well containing 500 µl of the above polyethylene glycol solution. The crystals belong to the space group P2₁ (unit cell a = 59.00 Å, b = 101.68 Å, c = 86.77 Å,  = 93.51^o) and contain two molecules in the asymmetric unit. Crystals (0.3 mm in each direction) were stabilized in a mother liquor containing 25% glycerol, flash-cooled to 170 °C in liquid nitrogen, and used for data collection. A 2.0-Å data set was collected at the HHMI X4A beam line of the National Synchrotron Light Source (NSLS), and data were reduced using DENZO and SCALEPACK (16). The synchrotron data encompassed 180^o of reciprocal space and were >95% complete (see Table I).

Structure Determination and Refinement-- Data collected on the VMA29 mutant were completely isomorphous with the previously determined structure of PI-SceI. Model phases for the VMA29 mutant were calculated based on the refined crystal structure of PI-SceI (12) (Protein Data Bank code 1VDE) with the water molecules removed. XTALVIEW (17) was used to examine the initial electron density map which showed a rough fit of the model to the experimental data with strong density present at the N and C terminus corresponding to the additional sequence present in the miniprecursor mutant. Refinement was carried out employing the simulated annealing protocol in the crystallographic and NMR system (18) software package, followed by a 30-step individual B_factor refinement.

The N- and C-terminal addition sequences as well as water molecules were built into the electron density map based on Fourier coefficients (|F_obs|  |F_calc|)e^icalc. Water molecules were placed in difference electron density peaks only if the peaks were above 2.5  and if acceptable hydrogen bonds to atoms in the model could be made. Deletion of water molecules occurred if the water molecules refined position was farther than 3.3 Å from its nearest neighbor or if their B_factors exceeded 70 Å². The statistics for the current refined structure are summarized in Table I.

Zinc Determination-- Zn²⁺ concentration was determined using Perkin-Elmer 2380 atomic absorption spectrometer in the laboratory of Dr. D. Giedroc at Texas A&M University. A standard curve from 0.5 µM Zn⁺² to 6.0 µM Zn²⁺ was used to quantitate the concentration of zinc in the protein sample. The protein solution was diluted. The atomic absorption readings from 1:50, 1:100, and 1:200 dilutions of the protein solution were averaged to give a mean zinc concentration of 161 µM. Using the molar extinction coefficient of 4.68 × 10⁴/mol for VMA29,² the protein concentration was determined to be 123 µM. The zinc present in the protein sample co-purified with the protein since neither the protein storage buffer (50 mM NaCl, 20 mM HEPES, pH 7.0) nor the crystallization solution described above contained any measurable zinc.

	RESULTS AND DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION REFERENCES

Structure of the N-terminal Splicing Junction-- The refined crystal structure of VMA29 consists of two independent molecules (A and B) in the asymmetric unit. The VMA29 molecule from residues 1 to 454 (the intein) has a fold identical to the PI-SceI intein (12). Of the 10 additional residues at the N terminus of VMA29, the last 4 (Lys⁴, Leu³, Glu², and Gly¹) have well defined electron density in molecule B (Fig. 2). Leu³, Glu², and Gly¹ have been shown to support proficient splicing (3). The side chain of Glu² forms a hydrogen bond to the backbone amide of Leu³, which fixes the main chain atoms of Lys⁴ and Leu³. The amide nitrogen of Gly¹ hydrogen bonds to the carbonyl oxygen of Ile⁴³⁴, and its carbonyl oxygen is hydrogen-bonded to the amide of the same residue. The backbone amide of Glu² and the side chain of Leu³ are in van der Waal contact with S- from Cys⁴⁵⁵. The C-2 atom of Leu³ packs against the nonpolar side chain atoms of Glu⁴⁵⁷. The arrangement of hydrogen bonds around Gly¹ and the packing of the 2 and 3 residues against residues 455 and 457 distort the main chain geometry of Gly¹. Its  angle (N-C-C) measures 100°, 10° less than the ideal angle. The scissile peptide bond between Gly¹ and Ala¹ is in the trans conformation, in agreement with the structure of glycosylasparaginase, but contrary to the GyrA structure (13). The Sce VMA intein seems more tolerant at the 1 position (19) than some mini-inteins (20-22). The mutational evidence provides support for the possibility of a distorted trans conformation to exist in the full-length Sce VMA precursor.

View larger version (49K): [in this window] [in a new window]   Fig. 2.   Stereoview of the electron density from a 2Fobs  Fcalc map associated with the N- and C-terminal additional extein residues present in the B molecule of VMA29. The contour level is 2 with a cover radius of 2.0 Å. Gly1 through Lys4 are associated with the N-terminal extein, whereas Cys455 through Arg458 are associated with the C-terminal extein. This figure was generated with XTALVIEW (17) and RASTER3D (25).

Molecule A in the asymmetric unit differs from B in that only residues Glu² and Gly¹ are defined by strong electron density. The main chain distortion observed in the molecule B is reduced here ( angle of 105°), and the scissile peptide bond is also in the trans conformation. A 50° rotation about the 1 torsion angle (C₁-C'₁) will superimpose Ala¹ from molecule A onto B. In the A molecule, Gly¹ has no hydrogen bond interaction with other surrounding residues. The backbone NH and carbonyl oxygen of Ala 1 make hydrogen bonds with the O-1 and NH, respectively, of Asn⁷⁶. Glu² contributes to the structural stability of this area with a hydrogen bond between its carbonyl oxygen and the side chain hydroxyl of Thr⁷⁸.

Structure of the C-terminal Splicing Junction-- The four native extein residues (Cys⁴⁵⁵, Gly⁴⁵⁶, Glu⁴⁵⁷, and Arg⁴⁵⁸) at the C-terminal splicing junction of molecule B are clearly defined by electron density (Fig. 2). However, in molecule A, electron density defines only Cys⁴⁵⁵. The most notable feature at the C-terminal splicing site is the unexpected presence of a zinc atom in both molecules, chelated by Glu⁸⁰, the highly conserved His⁴⁵³, Cys⁴⁵⁵, and a water molecule. The geometry of zinc coordination in molecule B is T₄ (tetrahedral) (23) (Fig. 3) with all three center angles involving the zinc atom measuring ~110°. It is unclear whether the zinc binding is structural or catalytic; however, the coordinating angles and bond distances are closer to those characteristic of a structural zinc (23). The zinc-binding site in the molecule A exhibits elongated electron density consistent with at least two different zinc binding modes. One involves His⁴⁴², Cys⁴⁵⁵, and His⁴⁵³ (₂ = 61), and a water molecule, whereas the other is similar to the site in the B molecule. Atomic absorption spectrometry identified the presence of one zinc atom per protein molecule (see "Materials and Methods"), but we found no measurable zinc in the storage or crystallization buffers. From these data, we conclude that the zinc atom co-purified with the VMA29 protein.

View larger version (28K): [in this window] [in a new window]   Fig. 3.   Close-up view of the zinc coordination at the C-terminal splicing junction of molecule B of VMA29. Zinc atom is the purple sphere.

Residues that chelate the zinc atom are important for protein splicing in the wild-type VMA protein. Cys⁴⁵⁵ receives the N-terminal extein during the transesterification reaction resulting in the formation of the branched intermediate (3). His⁴⁵³ is highly conserved among inteins deposited into the New England Biolabs intein data base (~90% of the inteins contain a histidine at this position). It is also adjacent to Asn⁴⁵⁴ (an Ala in the current structure) which is absolutely conserved and essential for excising the spliced exteins from the intein (3, 6). Glu⁸⁰ is not a conserved residue, but it is adjacent to the strictly conserved, catalytically required, His⁷⁹ (24) which is hydrogen-bonded to the amide of the scissile peptide bond at the N-terminal splice junction (or at Gly¹). His⁴² is a conserved residue in motif C2. The main chain distortion found in Gly¹ ( = 100°) is also present in Ala⁴⁵⁴. The chelation of the zinc atom by His⁴⁵³ and Cys⁴⁵⁵ bends the main chain at this residue (Fig. 3) thereby causing the observed change in  angle resulting in increase in energy of the peptide bonds surrounding this residue. In the wild-type protein, Asn⁴⁵⁴ undergoes succinimide formation whereby its side chain cyclizes with its own backbone carbonyl carbon. The distortion at this residue could help lower the energy barrier for succinimide formation. We conclude Cys⁴⁵⁵, His⁴⁵³, and Asn⁴⁵⁴ are poised to receive the thioester formed at Cys¹ and undergo transesterification and succinimide formation.

Model of Wild-type VMA Intein-- The wild-type model of the VMA precursor was generated by replacing Ala¹ and Ala⁴⁵⁴ in the VMA29 structure with Cys and Asn, respectively, and energy-minimizing the structure (see "Materials and Methods"). Side chain atoms were modeled such that their corresponding ₁ angles were equal to the most favored, Cys ₁ = 60 and Asn ₁ = 180. Following energy minimization, these angles undergo only very slight changes (61° and177°, respectively). The S- atom of the modeled Cys¹ is in position to hydrogen bond to the side chain O-1 of Asn⁷⁶ and is also within 4.4 Å of its own carbonyl carbon. The O- atom of Asn⁴⁵⁴ is 3.5 Å from Thr⁴³⁵ backbone amide, and its N- atom is 2.7 Å from both Thr⁴³⁵ backbone carbonyl oxygen and its own backbone carbonyl carbon. The model reveals a hydrophobic surface to the "back" of Asn⁴⁵⁴ (away from the N-terminal splicing junction) comprised of side chain atoms from Ile⁴³⁴, Phe⁴⁴⁴, Leu⁴³⁶, Val³⁷, and Val²⁷. This surface constrains the ₁ values between 117° and 180° (i.e. in an orientation consistent with cyclization). The distance between the S- atom of Cys¹ and the same atom of Cys⁴⁵⁵ is ~9 Å in both molecules in the asymmetric unit. It is unclear how this distance is traversed during transesterification.

Comparison of VMA29 and GyrA Structures-- The model for GyrA (Protein Data Bank code 1AM2 (13)) precursor was superimposed onto the structure of the VMA29 precursor (Fig. 4a) using the sequence alignment (HINT module) found in Klabunde et al. (13). Fig. 4b illustrates residues involved in protein splicing occupy the same relative position in each protein even though VMA29 contains a bound zinc atom and residues comprising the N- and C-terminal exteins. The largest deviation in -carbon position, 1.46 Å, occurs between GyrA Ser¹ and VMA29 Ala¹. The catalytically required GyrA His⁷⁵ is 1.22 Å from the position of its equivalent residue in VMA29 (His⁷⁹). Likewise the conserved His¹⁹⁷ is 1.05 Å from the position of VMA29 His⁴⁵³; however, the side chain ₂ differs by 136° (78 and 58, respectively). Similarly, Thr⁷² and Asn⁷⁶ are within 1.03 Å of each other, and the strictly conserved Asn¹⁹⁸ and Ala⁴⁵⁴ are 1.30 Å from one another. The structural superposition illustrates zinc binding at the C-terminal splicing junction does not influence the relative positions of the amino acids proven to be involved in protein splicing. It does, however, influence the side chain orientation of the penultimate histidine. The additional N- and C-extein residues also do not appear to perturb the important splicing residues. The Sce VMA inteins with Leu³, Glu², Gly¹, Cys⁴⁵⁵, and Gly⁴⁵⁶ have previously been shown to undergo efficient splicing (6). Thus, this is a splicing relevant structure.

View larger version (45K): [in this window] [in a new window]   Fig. 4.   a, superpositioned ribbon diagram of the VMA29 (blue) and GyrA (red) HINT modules based on the sequence alignment found in Klabunde et al. (13). For clarity, the endonuclease domain (Endo/Linker) and DNA recognition region (DRR) of VMA29 were not included in the diagram. Purple sphere is the zinc atom from VMA29. b, stereoview of the superposition of residues involved in protein splicing from VMA29 (blue) and GyrA (red). The superposition was performed as indicated in a.

VMA29 Structure and Protein Splicing-- Based on the wild-type model for the VMA precursor, we propose the following molecular mechanism (Fig. 5). The side chain carbonyl oxygen of Asn⁷⁶ and the S- of Cys¹ are within hydrogen bonding distance (2.91 Å). This interaction would help polarize the sulfhydryl such that its lone pair electrons would be oriented toward the carbonyl carbon of scissile peptide bond. Much like the GyrA structure (13), there does not appear to be a good general base in close proximity to Cys¹. Glu² is within 3.5 Å of S- atom, but the amino acid at this position has not been shown to be critical for protein splicing (3). It is not clear from this structure how the thiolate anion is generated. Regardless of the exact mechanism responsible for sulfhydryl activation, a tetrahedral intermediate would be formed during the acyl shift reaction. The intermediate could be stabilized by interactions of N- from Asn⁷⁶ and O- from Thr⁷⁸ (Fig. 5a). The imidazole ring of His⁷⁹ is in position to protonate the amide nitrogen of Cys¹ promoting the break down of the tetrahedral intermediate to form the thioester. Relieving the distorted/strained main chain atoms of Gly¹ may help drive the N  S acyl rearrangement.

View larger version (26K): [in this window] [in a new window]   Fig. 5.   a-d, proposed mechanism of protein splicing involving a structural zinc. Dashed lines represent hydrogen bonds, and numbers above dashes are distances in Angstroms. The red arrows indicate points of nucleophilic attack. Blue residues are part of N-extein; aqua residues are part of the C-extein, and black residues are part of the intein. The energy- minimized wild-type model was used to generate this illustration.

The transesterification reaction requires the thioester between Gly¹ and Cys¹ to be in close proximity to the acceptor, Cys⁴⁵⁵. As stated above, the distance between the modeled side chain of Cys¹ and that of Cys⁴⁵⁵ is ~9 Å (Fig. 5b). It is not possible at this time to determine exactly how this gap is traversed, but the tight type II turn between residues 3 and 6 could relax and allow Cys¹ to move toward Cys⁴⁵⁵. His⁴⁵³  Gln substitution provides experimental evidence for a conformational change occurring during splicing of the PI-SceI intein. C-terminal cleavage in the resulting mutant was shown to be absolutely dependent on N-terminal cleavage to liberate the N terminus (19).

The coordination between Cys⁴⁵⁵ and the zinc atom would be disrupted when the thioester is presented. The zinc atom could move away from the newly formed thioester and take up a position between His⁴⁵⁴ and His⁴⁴² (much like the position observed in molecule A of this structure) or it could diffuse away from the protein (Fig. 5c). If the former occurred, a main chain carbonyl oxygen from either Gly⁴⁵⁶ or Asn⁴⁵⁴ could replace the sulfur atom from Cys⁴⁵⁵ in the coordination of the zinc. His⁴⁵³ would then rotate ~120° about 2 to orient its N- toward the new position of zinc and in doing so would position its N- such that it could donate a proton to the amide nitrogen of Cys⁴⁵⁵ after asparagine cyclization. The resulting orientation of His⁴⁵³ would be consistent with that observed in the GyrA structure as well as the side chain orientation found in one of the zinc binding modes of molecule A.

Experimental data suggest that succinimide formation could occur independently of the initial steps in the splicing reaction. Self-splicing inteins carrying mutations of Cys¹ inhibited cleavage at the N-terminal splice junction but still retained cleavage activity at the C-terminal splice junction (11, 20, 21). However, in the normal splicing pathway prior cleavage at the N-terminal junction would be favored to occur before asparagine cyclization coupled peptide bond cleavage could proceed. Since transesterification most likely involves the relaxation of the type II tight turn between residues 3 and 6, subsequently the residues involved in cyclization of Asn⁴⁵⁴ would take favorable positions (Fig. 5d). His⁴⁵³ would be in position to protonate the amino group of Cys⁴⁵⁵. The side chain nitrogen of Asn⁴⁵⁴ refined to a position where it hydrogen bonds to the carbonyl oxygen of Ile⁴³⁴. These interactions coupled with the hydrophobic surface orient the -amide such that it could attack its own backbone carbonyl carbon. As a result the amide bond between Asn⁴⁵⁴ and Cys⁴⁵⁵ would be broken down forming the C-terminal aminosuccinimide which would be hydrolyzed to form a C-terminal asparagine or isoasparagine. A spontaneous S  N acyl shift would result in a stable amide bond between Gly¹ and Cys⁴⁵⁵, thus completing protein splicing.

The VMA29 miniprecursor structure presented here provides the first image of a true intein precursor containing multiple residues from both N- and C-terminal exteins. The N-terminal scissile peptide bond is in the trans conformation with a distorted  angle in the 1 residue. These observations are in agreement with the recently solved crystal structure of glycosylasparaginase (15) but contradict the findings of Klabunde et al. (13) concerning GyrA. An unexpected finding is a bound zinc atom chelated by residues located at the C-terminal splicing junction. We propose a splicing mechanism consistent with experimental evidence in which the PI-SceI intein may utilize a structural zinc atom.

	ACKNOWLEDGEMENTS

We thank Drs. Z. Wang, G. Hu, and C. Ogata for assistance with data collection at Brookhaven National Laboratory facility; F. Feng and M. Scott for technical assistance; Dr. D. Giedroc for assistance with zinc determination; and D. Comb for support and encouragement.

	FOOTNOTES

^* The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

The atomic coordinates and structure factors (code 1EF0) have been deposited in the Protein Data Bank, Research Collaboratory for Structural Bioinformatics, Rutgers University, New Brunswick, NJ (http://www.rcsb.org).

^¶ To whom correspondence should be addressed: Howard Hughes Medical Inst., Baylor College of Medicine, One Baylor Plaza, Houston, TX 77030. Tel.: 713-798-6565; E-mail: faq@bcm.tmc.edu.

² F. S. Gimble, personal communication.

	ABBREVIATIONS

The abbreviations used are: VMA, vacuolar membrane ATPase; HINT, hedgehog intein; VMA29, PI-SceI miniprecursor; VMA1, PI-SceI intein from the vacuolar ATPase subunit of S. cerevisiae; C-extein, intein C-terminal flanking region; N-extein, intein N-terminal region.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				K. Michalska, A. Hernandez-Santoyo, and M. Jaskolski The Mechanism of Autocatalytic Activation of Plant-type L-Asparaginases J. Biol. Chem., May 9, 2008; 283(19): 13388 - 13397. [Abstract] [Full Text] [PDF]

				Y. Sun and H.-C. Guo Structural constraints on autoprocessing of the human nucleoporin Nup98 Protein Sci., March 1, 2008; 17(3): 494 - 505. [Abstract] [Full Text] [PDF]

				J. Prieto, J.-C. Epinat, P. Redondo, E. Ramos, D. Padro, F. Cedrone, G. Montoya, F. Paques, and F. J. Blanco Generation and Analysis of Mesophilic Variants of the Thermostable Archaeal I-DmoI Homing Endonuclease J. Biol. Chem., February 15, 2008; 283(7): 4364 - 4374. [Abstract] [Full Text] [PDF]

				M. A. Johnson, M. W. Southworth, T. Herrmann, L. Brace, F. B. Perler, and K. Wuthrich NMR structure of a KlbA intein precursor from Methanococcus jannaschii Protein Sci., July 1, 2007; 16(7): 1316 - 1328. [Abstract] [Full Text] [PDF]

				J. Prieto, P. Redondo, D. Padro, S. Arnould, J.-C. Epinat, F. Paques, F. J. Blanco, and G. Montoya The C-terminal loop of the homing endonuclease I-CreI is essential for site recognition, DNA binding and cleavage Nucleic Acids Res., May 11, 2007; 35(10): 3262 - 3271. [Abstract] [Full Text] [PDF]

				P. Shemella, B. Pereira, Y. Zhang, P. Van Roey, G. Belfort, S. Garde, and S. K. Nayak Mechanism for Intein C-Terminal Cleavage: A Proposal from Quantum Mechanical Calculations Biophys. J., February 1, 2007; 92(3): 847 - 853. [Abstract] [Full Text] [PDF]

				W. Sun, J. Yang, and X.-Q. Liu Synthetic Two-piece and Three-piece Split Inteins for Protein trans-Splicing J. Biol. Chem., August 20, 2004; 279(34): 35281 - 35286. [Abstract] [Full Text] [PDF]

				A. Romanelli, A. Shekhtman, D. Cowburn, and T. W. Muir Semisynthesis of a segmental isotopically labeled protein splicing precursor: NMR evidence for an unusual peptide bond at the N-extein-intein junction PNAS, April 27, 2004; 101(17): 6397 - 6402. [Abstract] [Full Text] [PDF]

				Y. Ding, M.-Q. Xu, I. Ghosh, X. Chen, S. Ferrandon, G. Lesage, and Z. Rao Crystal Structure of a Mini-intein Reveals a Conserved Catalytic Module Involved in Side Chain Cyclization of Asparagine during Protein Splicing J. Biol. Chem., October 3, 2003; 278(40): 39133 - 39142. [Abstract] [Full Text] [PDF]

				J.-C. Epinat, S. Arnould, P. Chames, P. Rochaix, D. Desfontaines, C. Puzin, A. Patin, A. Zanghellini, F. Paques, and E. Lacroix A novel engineered meganuclease induces homologous recombination in yeast and mammalian cells Nucleic Acids Res., June 1, 2003; 31(11): 2952 - 2962. [Abstract] [Full Text] [PDF]

				Y. Wang and H.-C. Guo Two-step Dimerization for Autoproteolysis to Activate Glycosylasparaginase J. Biol. Chem., January 24, 2003; 278(5): 3210 - 3219. [Abstract] [Full Text] [PDF]

				E. Werner, W. Wende, A. Pingoud, and U. Heinemann High resolution crystal structure of domain I of the Saccharomyces cerevisiae homing endonuclease PI-SceI Nucleic Acids Res., September 15, 2002; 30(18): 3962 - 3971. [Abstract] [Full Text] [PDF]

				F. B. Perler InBase: the Intein Database Nucleic Acids Res., January 1, 2002; 30(1): 383 - 384. [Abstract] [Full Text] [PDF]

				I. Ghosh, L. Sun, and M.-Q. Xu Zinc Inhibition of Protein trans-Splicing and Identification of Regions Essential for Splicing and Association of a Split Intein* J. Biol. Chem., June 22, 2001; 276(26): 24051 - 24058. [Abstract] [Full Text] [PDF]

				K. V. Mills and H. Paulus Reversible Inhibition of Protein Splicing by Zinc Ion J. Biol. Chem., March 30, 2001; 276(14): 10832 - 10838. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Poland, B. W. Articles by Quiocho, F. A.