JBC INTERFERin siRNA transfection reagent

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Originally published In Press as doi:10.1074/jbc.M112365200 on February 15, 2002

J. Biol. Chem., Vol. 277, Issue 18, 16257-16264, May 3, 2002
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
277/18/16257    most recent
M112365200v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Guhan, N.
Right arrow Articles by Muniyappa, K.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Guhan, N.
Right arrow Articles by Muniyappa, K.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

Mycobacterium tuberculosis RecA Intein Possesses a Novel ATP-dependent Site-specific Double-stranded DNA Endonuclease Activity*

N. GuhanDagger and K. Muniyappa§

From the Department of Biochemistry, Indian Institute of Science, Bangalore 560 012, India

Received for publication, December 26, 2001, and in revised form, February 13, 2002

    ABSTRACT
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES

Mycobacterium tuberculosis recA harbors an intervening sequence in its open reading frame, presumed to encode an endonuclease (PI-MtuI) required for intein homing in inteinless recA allele. Although the protein-splicing ability of PI-MtuI has been characterized, the identification of its putative endonuclease activity has remained elusive. To investigate whether PI-MtuI possesses endonuclease activity, recA intervening sequence was cloned, overexpressed, and purified to homogeneity. Here we show that PI-MtuI bound both single- and double-stranded DNA with similar affinity but failed to cleave DNA in the absence of cofactors. Significantly, PI-MtuI nicked supercoiled DNA in the presence of alternative cofactors but required both Mn2+ and ATP to generate linear double-stranded DNA. We observed that PI-MtuI was able to inflict a staggered double-strand break 24 bp upstream of the insertion site in the inteinless recA allele. Similar to a few homing endonucleases, DNA cleavage by PI-MtuI was specific with an exceptionally long cleavage site spanning 22 bp. The kinetic mechanism of PI-MtuI promoted cleavage supports a sequential rather than concerted pathway of strand cleavage with the formation of nicked double-stranded DNA as an intermediate. Together, these results reveal that RecA intein is a novel Mn2+-ATP-dependent double-strand specific endonuclease, which is likely to be important for homing process in vivo.

    INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES

The prototype Escherichia coli RecA protein plays a central role in homologous recombination, DNA repair, restoration of stalled replication forks and SOS response (reviewed in Refs. 1-4). The process of homologous recombination, which is the main mechanism of genetic exchange and RecA, a ubiquitous multifunctional protein, is substantially conserved among a range of organisms (reviewed in Ref. 5). In contrast, the recA of Mycobacterium tuberculosis and Mycobacterium leprae contain an in-frame insertion of an intein-coding sequence (6, 7). Following the synthesis of RecA precursor, an internal domain, termed the intein, is excised from the precursor and the two flanking domains called exteins are ligated together to generate functionally active RecA protein (8). Previously, we have reported the recombination-like activities and x-ray structure of the spliced form of M. tuberculosis RecA protein (9-11). However, the biochemical function of M. tuberculosis RecA intein and its possible role in homologous recombination promoted by its cognate RecA remains obscure.

Inteins are protein-splicing elements that contain conserved domains with sequence homology to a diverse family of homing endonucleases (reviewed in Refs. 12-16). The hallmark of homing endonucleases, like restriction enzymes, is their ability to cleave double-stranded DNA at specific target sites (12, 17). First identified in a mobile group I intron of yeast mitochondria (18, 19); genes for homing endonucleases have since been found in group II introns and intein-coding sequences of unicellular eukaryotes, archaea, and eubacteria (12). On the basis of conserved motifs, these enzymes fall into four classes: (a) LAGLIDADG, (b) GIY-YIG, (c) H-N-H, and (d) His-Cys box or zinc finger enzymes (12, 16, 20). The LAGLIDADG class is the largest, more widespread with >200 known members, each containing one or two copies of the dodecapeptide motif. These enzymes are highly specific, with very long, rare recognition sites that span intein integration sites in homologous alleles that lack the intron or intein-coding sequence. Biochemical and structural studies have demonstrated that homing endonucleases composed of a single copy of LAGLIDADG motif function as homodimers, whereas double-motif enzymes act as monomers. The catalytic centers carry two essential aspartate residues in the LAGLIDADG motif. These residues function by coordinating a divalent cation necessary for catalysis, Mg2+ being the preferred metal ion (12-16).

It is clear from the scrutiny of complete genome sequence of M. tuberculosis that, in addition to recA (6, 7), as defined by both in vivo and in vitro studies, intein-coding sequences are embedded in the open reading frames of dnaA and Rv1461 (21). The deduced amino acid sequence of M. tuberculosis recA intervening sequence disclosed the existence of two copies of LAGLIDADG motifs. The protein-splicing ability of PI-MtuI1 has been studied extensively (22-25); however, the endonuclease activity has not been demonstrated. To this end, PI-MtuI was cloned, overexpressed, and purified, to explore its biochemical functions. Significantly, the results presented here reveal that RecA intein is a novel Mn2+-ATP-dependent, double-strand-specific endonuclease, which is likely to be important for intein homing in vivo.

    MATERIALS AND METHODS
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES

All the chemicals used in this study are of analytical grade. Buffers were prepared using deionized water. Restriction enzymes, Nylon N+ membrane, Sephacryl S-200 were purchased from Amersham Biosciences, Inc., Asia Pacific Pvt. Ltd, Hong Kong. Phage T4 polynucleotide kinase was purchased from Invitrogen, New York, NY. Hydroxyapatite was obtained from Bio-Rad Laboratories, Hercules, CA. Circular single-stranded and negatively supercoiled plasmid pEJ244 DNA were prepared by sucrose density gradient centrifugation as described previously (26). The DNA was dissolved in 10 mM Tris-HCl buffer (pH 7.5) containing 1 mM EDTA, and the concentrations are expressed in moles of nucleotide residues.

Construction of PI-MtuI-GST Fusion Plasmid-- Plasmid pEJ135 (kindly provided by E. O. Davis and M. J. Colston, National Institutes for Medical Research, London) bearing M. tuberculosis recA gene was digested with KpnI and HindIII. The 1.6-kb DNA fragment purified from agarose gel was incubated with T4 DNA polymerase (27). The KpnI-HindIII fragment contained the RecA intein coding sequence and a portion of recA corresponding to 15 amino acid residues of N-extein and 75 amino acid residues of C-extein flanking the intein autoproteolytic cleavage sites. The fragment was subcloned into expression vector pGEX-2T at the SmaI site. The recombinant plasmid was referred to as pGRI. The plasmid contains in-frame fusion of the RecA intein-coding sequence with GST. DNA sequence of the insert was determined using the automated ABI Prism DNA sequencer (PerkinElmer Life Sciences) to confirm the sequence. The recombinant plasmid was transformed and amplified in E. coli strain DH5alpha .

Expression and Purification of PI-MtuI Endonuclease-- Optimal expression of PI-MtuI was obtained in E. coli strain DH5alpha by growth at 37 °C in LB broth until A600 = 0.4, followed by addition of isopropyl-1-thio-beta -D-galactopyranoside to a final concentration of 0.5 mM. Cells were further incubated for 4 h and harvested later by centrifugation at 5000 × g for 10 min and stored at -70 °C. All subsequent steps were performed at 4 °C unless otherwise stated. Cells (30 g) were resuspended in buffer A (50 mM Tris-HCl, pH 8.0, 10% sucrose) and incubated with 20 mg of lysozyme on ice for 30 min. Cells were lysed by sonication (Vibra cell sonicator, Sonics and Material Inc., Danbury, CT) at setting 9 and 60% duty cycles in a pulse mode for 15 min. The supernatant was obtained by centrifugation at 30,000 rpm for 1 h in a Ti-45 Beckman rotor. Polyethyleneimine acetate (pH 6.5) was added to the supernatant to a final concentration of 0.2% over a period of 15 min with constant stirring. Precipitates were recovered by centrifugation at 12,000 rpm for 15 min. To the supernatant, (NH4)2SO4 was added to a final concentration of 0.3 g/ml with continuous stirring over a period of 1 h. The suspension was centrifuged at 15,000 rpm for 15 min, and the pellet was resuspended in buffer B (20 mM sodium phosphate, pH 6.5, 10% glycerol, 0.1 mM EDTA, 5 mM 2-mercaptoethanol). Protein solution was dialyzed against 1 liter of buffer B with three changes over 18 h. The dialysate was centrifuged at 15,000 rpm for 15 min, and the supernatant was loaded onto a hydroxyapatite column (2 × 11 cm), which had been equilibrated with buffer B. The column was washed with buffer B until the eluate contained no material that absorbed light at 280 nm. The bound proteins were eluted with a linear gradient of 20 to 250 mM sodium phosphate in buffer B. Fractions containing PI-MtuI (ascertained by immunoblot assay using antibodies raised against M. tuberculosis RecA precursor) were pooled and precipitated by the addition of (NH4)2SO4 to a final concentration of 0.3 g/ml. The precipitate was collected by centrifugation as described above and resuspended in buffer C (20 mM Tris acetate, pH 7.5, 1 M NaCl, 5 mM 2-mercaptoethanol, 0.1 mM EDTA, 10% glycerol), loaded onto a Sephacryl S 200 column (1.7 × 105 cm), which had been equilibrated with buffer C. The fractions containing PI-MtuI were combined, precipitated with (NH4)2SO4, and dialyzed against buffer C and loaded onto a Superdex 200 column (26 × 60 cm) (Amersham Biosciences, Inc.). Peak fractions containing PI-MtuI were pooled, dialyzed against storage buffer (20 mM Tris-HCl, pH 7.5, 100 mM NaCl, 1 mM DTT, 50% glycerol). Aliquots of the dialysate were stored at -20 °C. The final yield of PI-MtuI was in the range of 2.5 mg/liter of culture. Protein purity was ascertained by 10% SDS-PAGE/Coomassie Blue staining and was generally >98% pure. The concentration of protein was determined by dye binding method using bovine serum albumin as the standard (28) and expressed in moles of monomers per liter. PI-MtuI stored in 50% glycerol at -20 °C was stable for 6 months.

Electrophoretic Mobility Shift Assay-- Reaction mixtures (10 µl) contained 25 mM Tris-HCl (pH 7.5), 0.4 mM DTT, 3 µM 32P-labeled 75-mer DNA containing the PI-MtuI cleavage site (underlined sequence) 5'-dACGCTCAAGGACGGTACCAACGCGGTCGGCAACCGCACCCGGGTCAAGGTCGTCAAGAACAAGTGTTCGCCCCCC-3' or 75-bp DNA (75-mer DNA annealed to its complementary strand) with increasing concentrations of PI-MtuI. After incubation at 37 °C for 10 min, the reactions were terminated by the addition of 1 µl of loading buffer (20% glycerol containing 0.12% (w/v) each of bromphenol blue and xylene cyanol) to each reaction mixture. Samples were separated on an 8% polyacrylamide gel by electrophoresis in 40 mM Tris acetate buffer (pH 7.5) containing 0.1 mM EDTA at 12 V/cm for 5 h at 4 °C. The gel was dried on a Whatman 3MM filter paper and visualized by autoradiography.

DNA Binding and Kinetic Studies-- Surface plasmon resonance (SPR) measurements were performed using BIAcore 2000 system with streptavidin-coated sensor chips. All procedures were automated by using repetitive cycles of sample injection, washing, and regeneration. The chip contained streptavidin covalently immobilized on a carboxymethylated dextran layer at the surface. The DNA surface was prepared by immobilizing a biotinylated 60-mer DNA (5'-dAATTCTGGGTGTGTGGGTGTGTGGTGTGTGGGTGTGGTCAAGTTGACTACGTATACATC-Biotin-3') or 27-bp DNA (5'-Biotin-dTCGAGATCAAGGTCAAAGGTCACTCGC-3' annealed to its complementary strand) according to manufacturer's instructions. The chips produced response signals of ~400 response units. An empty flow cell served as a control. PI-MtuI was diluted in binding buffer containing 25 mM Tris-HCl (pH 7.5), 100 mM NaCl, 0.4 mM DTT, and 0.1 mM EDTA. Solutions containing increasing concentrations of PI-MtuI, designated as the "analyte," was injected onto the biosensor chip for 10 min at a flow rate of 5 µl/min for association, followed by a 5-min buffer injection for dissociation. Under these conditions, nonspecific binding of PI-MtuI to the control flow cell was insignificant. Specific binding in real-time and physical parameters of interaction was determined by analyzing the sensograms using the BIAcore evaluation software version 3.0. Specific SPR was plotted as function of time obtained at different analyte concentrations. The kinetic rate constants and the equilibrium binding constants were determined. After each injection, the surface was regenerated using 0.15% SDS, followed by buffer wash.

Endonuclease Assay-- Reaction mixtures (25 µl) contained 25 mM Tris-HCl buffer (pH 7.5), 3 mM MnCl2, 1.5 mM ATP, 0.4 mM DTT, and 16 µM of negatively supercoiled pEJ244 plasmid DNA. The concentration of PI-MtuI in each reaction is given in the figure legends. Plasmid pEJ244 bears one intein-less M. tuberculosis recA allele [In-:recA] cloned into pTZ18R. After appropriate time of incubation at 37 °C, the reaction was stopped by the addition of 0.1% SDS followed by deproteinization with proteinase K (0.2 mg/ml) for 15 min at 37 °C. To the samples, 3 µl of gel loading buffer was added. The samples were then separated by electrophoresis through 0.8% agarose gel at 3 V/cm in 89 mM Tris borate buffer (pH 8.3). The bands were visualized by staining with 0.5 µg/ml ethidium bromide. The gel was transferred to Nylon N+ membrane and developed by Southern hybridization (27). Quantification of bands was performed by UVI-BandMap software and plotted using GraphPad Prism 2.0 (UVI-Tech gel documentation system).

Immunological Techniques-- Antibodies to M. tuberculosis RecA precursor (9) and PI-MtuI were prepared in rabbits and characterized using standard procedures. Protein samples were electrophoresed on SDS-PAGE, transferred to nitrocellulose membrane, and visualized by chemiluminescence method as described (29, 30). Immunodepletion assay was performed by incubation of preimmune IgG-Sepharose or anti-PI-MtuI IgG-protein A-Sepharose with 20 µg of PI-MtuI in the endonuclease assay buffer at 4 °C for 2 h with continuous stirring. The slurry was centrifuged at 6000 rpm at 4 °C for 2 min, and the supernatant was assayed for the presence of Mn2+-ATP-dependent PI-MtuI endonuclease as described above.

Mapping of the PI-MtuI Cleavage Sites in Target DNA-- The DNA fragment containing the recA intein insertion site (inteinless recA allele) was generated from plasmid pEJ244 by digestion with SmaI and HinfI. The fragment was 5'-labeled using [gamma -32P]ATP and polynucleotide kinase. The reaction products were separated by electrophoresis on 8%polyacrylamide gels in 89 mM Tris borate buffer (pH 8.3) containing 2 mM EDTA. The bands corresponding to 44- and 77-bp DNA were excised from the gel, ground into small pieces, eluted with 10 mM Tris-HCl buffer (pH 7.5) containing 1 mM EDTA, extracted with phenol:chloroform:isoamyl alcohol solution, and DNA was precipitated by ethanol. About 0.1 µM PI-MtuI was incubated with ~100 ng of 32P-labeled DNA in the presence of 1 mM ATP and 3 mM MnCl2 in reaction conditions similar to those described above. The protein was digested with proteinase K (0.2 mg/ml) for 15 min at 37 °C. The reaction products were extracted with phenol:chloroform:isoamyl alcohol solution, and DNA was precipitated by ethanol. The pellets were dissolved in gel loading buffer (95% formamide, 10 mM NaOH, 0.2% bromophenol blue, and 0.2% xylene cyanol) and boiled for 10 min followed by immediate cooling to 4 °C. Aliquots of the sample was subjected to the Maxam-Gilbert chemical sequencing reaction, and electrophoresed through a 14% polyacrylamide gel in the presence of 8 M urea in 89 mM Tris borate buffer (pH 8.3) at 1800 V for 3.5 h (31). The gel was dried onto a 3MM Whatman filter paper and exposed to a phosphorimaging screen for 3 h for visualization of the bands. Signals were quantified using an Image Gauge version 3.0 (Fuji Science Laboratory, Japan).

    RESULTS
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES

Expression and Purification of PI-MtuI Endonuclease from E. coli-- A DNA fragment containing M. tuberculosis recA intein-coding sequence and a portion of recA gene corresponding to 15 and 75 amino acid residues from N- and C-exteins, respectively, flanking the intein autoproteolytic cleavage sites, was subcloned in-frame with GST in an expression vector, pGEX-2T. The resultant plasmid, pGRI, was transformed into E. coli strain DH5alpha , and expression of fusion protein was induced by the addition of isopropyl-1-thio-beta -D-galactopyranoside. Soluble extracts from induced cells were used to ascertain expression of fusion protein by SDS-polyacrylamide gel electrophoresis, visualized by Coomassie Blue staining and immunoblot analysis with polyclonal antibodies directed against M. tuberculosis RecA precursor. A band corresponding to the predicted size of fusion protein (84 kDa) was not detectable in the induced cell lysates. In contrast, we observed two bands: one identical to the native size of M. tuberculosis RecA intein (48 kDa) as calculated from deduced amino acid sequence, and the second corresponding to the combined size of GST plus portions of N- and C-exteins of RecA (36 kDa) (Fig. 1A). Consistent with these results, an immunoreactive species corresponding to the size of the fusion protein was not detectable in cell lysates. However, we observed a slower migrating immunoreactive band of the size of PI-MtuI-GST fusion protein in an enriched fraction, thereby confirming the synthesis of fusion protein (lane 3, Fig. 1B).


View larger version (29K):
[in this window]
[in a new window]
  		Fig. 1.   SDS-PAGE analysis showing induced expression of PI-MtuI in E. coli and at various stages during its purification. A, approximately 10 µg of protein was separated by SDS-PAGE and visualized by staining with Coomassie Blue. Lanes: 1, molecular mass markers (Invitrogen); 2, uninduced cell lysate; 3, induced cell lysate; 4, polymin P pellet fraction; 5, polymin P supernatant fraction; 6, (NH4)2SO4 pellet fraction; 7, hydroxyapatite eluate fraction; 8, Sephacryl S 200 fraction; 9, Superdex 200 gel filtration fraction. B, immunoblot analysis of PI-MtuI. After SDS-PAGE as in A, the proteins were transferred to a nitrocellulose membrane, stained with anti-PI-MtuI polyclonal antibodies, and visualized as described under "Materials and Methods." Lanes: 1, uninduced cell lysate; 2, induced cell lysate; 3, polymin P pellet fraction; 4, (NH4)2SO4 pellet fraction; 5, hydroxyapatite eluate fraction; 6, Sephacryl S 200 fraction; 7, Superdex 200 gel filtration fraction.

The soluble PI-MtuI was purified to homogeneity by precipitation with polymin P, (NH4)2SO4 fractionation, chromatography over hydroxyapatite and gel filtration on Superdex 200 column (Fig. 1A). Estimates of molecular mass for the intact protein based on SDS-PAGE and gel filtration on Superdex 75 column, suggested that PI-MtuI is a 48-kDa species, consistent with deduced amino acid sequence. The identity of the expressed and purified PI-MtuI was verified by sequencing 10 amino acid residues at the N-terminal end. The determined amino acid sequence was as follows: NH2-CLAEGTRIFD-COO-, identical to that predicted from the nucleotide sequence. The presence of cysteine at the N-terminal end indicated correct splicing of PI-MtuI in E. coli. Purified PI-MtuI was examined in an in vitro assay for the presence of exonuclease and was observed to be devoid of detectable 3'right-arrow5' or 5'right-arrow3' exonuclease activities.

Interaction of PI-MtuI with Single- and Double-stranded DNA-- To investigate the ability of PI-MtuI to bind DNA, we used electrophoretic mobility shift assay with substrates containing the PI-MtuI cleavage site (see below). The formation of PI-MtuI·DNA complexes was assayed by mobility shifts in polyacrylamide gels relative to free DNA and analyzed by autoradiography. As shown in Fig. 2, PI-MtuI bound to single- and double-stranded DNA as indicated by the reduced mobility of the band corresponding to free DNA as increasing amounts of PI-MtuI were added. However, the amount of PI-MtuI required for incorporating >90% of DNA into protein·DNA complexes differed between single- and double-stranded DNA. A 2-fold higher amount of PI-MtuI was required to achieve a comparable effect with single-stranded DNA. Similar results were also obtained with PI-MtuI with DNA substrates lacking the cleavage site (data not shown). Varying the period of incubation prior to electrophoresis revealed that binding occurred maximally with briefest incubation times tested at 37 °C (data not shown).


View larger version (28K):
[in this window]
[in a new window]
  		Fig. 2.   PI-MtuI forms a stable binary complex with single- or double-stranded DNA. Reaction mixtures containing 3 µM 32P-labeled 75-mer or 75-bp DNA was mixed with increasing concentrations of PI-MtuI in the absence of a metal ion and incubated as described under "Materials and Methods." Lanes 1 and 7, substrate DNA lacking PI-MtuI. Lanes 2-6, 75-mer DNA with PI-MtuI at concentration indicated on top of each lane; lanes 8-11, 75-bp DNA with PI-MtuI at the concentrations indicated on top of each lane. The positions of unbound DNA and PI-MtuI·DNA complexes are indicated on the left.

The kinetics of interaction of PI-MtuI with DNA in real-time was monitored by SPR spectroscopy using DNA lacking the cleavage site. Under these conditions, binding of PI-MtuI to DNA was independent of the presence of the cleavage site. Streptavidin sensor chip was derivatized with 60-mer DNA tagged with a 3'-terminal biotin, or 27-bp DNA bearing biotin at the 5'-end of the upper strand. The DNA on the chip was designated arbitrarily as the ligand, whereas a solution containing PI-MtuI, designated as the "analyte" was passed over the sensor chip in a microfluidics chamber. To collect kinetic data, a range of concentrations of PI-MtuI was injected over the immobilized DNA and reference surface for a period of 10 min. The chip was then washed with the binding buffer for additional 5-min. Specific binding of PI-MtuI to single- or double-stranded DNA resulted in a substantial enhancement of mass in a concentration-dependent manner (data not shown). Specific binding of PI-MtuI to the immobilized DNA was qualitatively detected by the increase in resonance units relative to control surface. Responses from the latter were used to correct for refractive index changes and instrument noise. A detailed kinetic analysis was performed by globally fitting the association and dissociation phase data from each surface. This approach has provided quantitative estimates of binding affinities and on/off rates of reaction (Table I). The apparent equilibrium dissociation constant for both single- and double-stranded DNA was in the range of 28 nM. These Kd values are in the same range as reported for PI-PfuI (13 nM) and PI-SceI (3.7 nM) (32, 33). Also, these results suggest that PI-MtuI could bind tightly to DNA in the absence of divalent cations, similar to PI-PfuI or PI-SceI (32, 33). Although our results are insufficient to allow determination of the stoichiometric ratio of binding of PI-MtuI to DNA, following conclusions could be drawn. First, binding of PI-MtuI to DNA does not require a target substrate containing the inteinless recA allele. Second, PI-MtuI binds to single- and double-stranded DNA to a similar extent and, more importantly, is independent of topology, divalent metal ions and NTPs.

                              
View this table:
[in this window]
[in a new window]
 
Table I
Kinetic and steady-state parameters for the interaction of PI-MtuI with single- and double-stranded DNA

PI-MtuI Displays Endonuclease Activity-- To test whether binding of PI-MtuI to DNA ensues in endonuclease activity, we used negatively supercoiled (form I) pEJ244 DNA containing inteinless recA allele as the substrate. The site for insertion of recA intervening sequence occurs once in pEJ244 plasmid DNA. The rationale for using form I DNA (likely to be the in vivo target) rather than linear double-stranded DNA would be to permit detection of even nicking endonuclease activity of PI-MtuI on supercoiled plasmids. Single-strand nicking would result in loss of negative supercoiling and, consequently, nicked circular duplex DNA (form II) could readily be distinguishable from that of form I DNA in a gel assay. Cleavage assays were performed by mixing PI-MtuI with DNA in the presence of alternative divalent cations. After incubation, the reaction mixtures were deproteinized and the reaction products were separated by gel electrophoresis. The DNA substrate and cleaved products were visualized by staining with ethidium bromide, followed by Southern hybridization and autoradiography (Fig. 3). Although PI-MtuI could interact with DNA in the absence of divalent cations (Fig. 2) but failed to cleave DNA (lane 1, Fig. 3). Significantly, PI-MtuI displayed weak nicking endonuclease activity in the presence of a variety of cofactors resulting in the conversion of form I DNA to form II DNA (compare lane 1 with lanes 2-14, Fig. 3). However, in the presence of ATP and MnCl2, PI-MtuI inflicted a double-strand break on form I DNA leading to the generation of form III DNA with no additional cleavages (lane 8, Fig. 3). To examine the role of ATP, we asked whether dATP in the presence of Mg2+ or Mn2+ could assist in the PI-MtuI promoted double-stranded DNA cleavage. In parallel, we also tested whether ATP hydrolysis was required for the generation of form III DNA. In both cases, addition of dATP or ATPgamma S in the presence of either Mg2+ or Mn2+ abrogated the generation of form III DNA (lanes 5, 6, 9, and 10, Fig. 3). Essentially, similar results were obtained with Mg2+, Ca2+, or Zn2+ alone or together with ATP. Thus, these results indicate that PI-MtuI requires Mn2+, and ATP hydrolysis for cleavage of double-stranded DNA.


View larger version (26K):
[in this window]
[in a new window]
  		Fig. 3.   Identification of endonuclease activity of PI-MtuI on cognate substrate in the presence of alternative cofactors. Lane 1, substrate DNA lacking PI-MtuI. The remaining lanes contained pEJ244 plasmid DNA (16 µM), PI-MtuI (1 µM), and the following cofactors: 5 mM Mg2+ (lane 2); 1.5 mM ATP (lane 3); 5 mM Mg2+-1.5 mM ATP (lane 4); 5 mM Mg2+-1.5 mM ATPgamma S (lane 5); 5 mM Mg2+-1.5 mM dATP (lane 6); 3 mM Mn2+ (lane 7); 3 mM Mn2+-1.5 mM ATP (lane 8); 3 mM Mn2+-1.5 mM dATP (lane 9); 3 mM Mn2+-1.5 mM ATPgamma S (lane 10); 3 mM Zn2+ (lane 11); 3 mM Zn2+-1.5 mM ATP (lane 12); Ca2+ (lane 13), and 3 mM Ca2+-1.5 mM ATP (lane 14). Following incubation, reaction mixtures were deproteinized, separated on an agarose gel, and visualized as described under "Materials and Methods." The positions of substrates and products are indicated on the left.

Intriguingly, PI-MtuI showed an unusual dependence for ATP. One factor that might account for the presence of Mn2+-ATP-dependent endonuclease could be attributable to a contaminating endonuclease. The identity of PI-MtuI was established by SDS-PAGE, and sequencing 10 amino acid residues at the N-terminal end. To seek further confirmation, immunodepletion assay was performed with anti-PI-MtuI IgG-protein A-Sepharose or preimmune IgG-protein A-Sepharose. The resultant supernatant was assayed for Mn2+-ATP-dependent endonuclease. Although the supernatant obtained from incubation with preimmune IgG displayed Mn2+-ATP-dependent endonuclease activity, the same from anti-PI-MtuI-treated sample lacked such an activity (data not shown). In addition, Mn2+-ATP-dependent endonuclease cofractionated exactly with PI-MtuI after Superdex 200 gel filtration. Together, these results argue that Mn2+-ATP-dependent endonuclease activity is intrinsic to PI-MtuI.

PI-MtuI Is Double-strand DNA-specific Endonuclease-- Our experiments demonstrate that PI-MtuI interacted with both single- and double-stranded DNA with similar affinity (Fig. 2 and Table I). The stable binding of PI-MtuI to ssDNA indicated that it might ensue in nicking of single-stranded DNA. To seek functional relationship between binding and cleavage, a fixed amount of pEJ244 form I or ssDNA was mixed with increasing amounts of PI-MtuI in the presence of Mn2+ and ATP as described above. Following incubation, reaction mixtures were deproteinized, and the resulting DNA was analyzed by agarose gel electrophoresis. The data in Fig. 4 indicate that PI-MtuI cleaved form I but not ssDNA. These results establish the notion that PI-MtuI is a double-stranded DNA-specific endonuclease.


View larger version (26K):
[in this window]
[in a new window]
  		Fig. 4.   PI-MtuI is a double-strand specific endonuclease. Reaction mixtures (25 µl) containing pEJ244 DNA (16 µM) was incubated with increasing concentrations of PI-MtuI in the presence of 3 mM Mn2+and 1.5 mM ATP. Reactions were terminated and analyzed as described under "Materials and Methods." Lanes 1 and 5, substrate DNA lacking PI-MtuI. The remaining lanes contained PI-MtuI at: lanes 2-4 at 1, 2.5, and 5 µM; lanes 6 and 7 at 1 and 2.5 µM of PI-MtuI, respectively. The positions of substrates and products are indicated on the left. ssDNA, single-stranded DNA; dsDNA, double-stranded DNA.

Specific Requirements for PI-MtuI Endonuclease Activity-- The parameters that influence the efficiency of endonuclease activity of PI-MtuI were tested by monitoring the cleavage of pEJ244 form I DNA relative to Mn2+, ATP, pH, and temperature conditions. We observed that both Mn2+ and ATP were essential to evoke double-stranded DNA cleavage activity of PI-MtuI. Both cleavage and nicking of double-stranded DNA was maximal in a buffer containing 3 mM Mn2+ (Fig. 5A), 1.5 mM ATP (Fig. 5B), pH 7.5 (Fig. 5C), at 37 °C (Fig. 5D). Interestingly, we observed significant nicking endonuclease activity across a wide range of ATP concentrations. It is possible that the lack of corresponding increase in the formation of form III DNA may be due to changes in the concentration of ADP in the reaction mixture. However, the likely effect of ADP was excluded by performing the reactions under ATP regeneration conditions, which failed to increase the conversion of form II to form III DNA (data not shown).


View larger version (40K):
[in this window]
[in a new window]
  		Fig. 5.   Characterization of double-stranded DNA endonuclease activity of PI-MtuI. Reactions were performed in an assay buffer (25 µl) containing 16 µM pEJ244 DNA, 1 µM PI-MtuI plus the indicated treatments. A, reaction mixtures contained 1.5 mM ATP plus Mn2+ at concentrations as indicated above each lane. B, reaction mixtures contained 3 mM Mn2+ plus ATP at concentrations as indicated above each lane. C, reaction mixtures contained 1.5 mM ATP and 3 mM Mn2+ in a buffer of pH as indicated above each lane. D, reaction mixtures containing 1.5 mM ATP and 3 mM Mn2+ and were incubated at the temperature as indicated above each lane. In C and D, lane M represents markers. The positions of substrates and products are indicated on the left.

We next examined the effect of added NaCl and potassium glutamate on the endonuclease activity of PI-MtuI. A representative experiment performed in the presence of increasing concentrations of NaCl or potassium glutamate is shown in Fig. 6. Addition of 50 mM either NaCl (lane 3, Fig. 6) or potassium glutamate (lane 10, Fig. 6) reduced cleavage but not nicking activity of PI-MtuI. However, further increase in ionic strength abolished double-strand cleavage activity (lanes 4-9 and 11-16, Fig. 6).


View larger version (33K):
[in this window]
[in a new window]
  		Fig. 6.   Effect of NaCl or potassium glutamate on the endonuclease activity of PI-MtuI. Reactions were performed as described in the legend to Fig. 5. Lane 1, substrate DNA lacking PI-MtuI; lane 2, complete reaction. The remaining lanes contained complete reaction mixture plus NaCl (lanes 3-9) or potassium glutamate (lanes 10-16) at 50, 100, 150, 200, 300, 400, and 500 mM, respectively.

Kinetics and the Pathway of PI-MtuI Cleavage Reaction-- We wished to investigate the kinetics and reaction pathway of PI-MtuI. Given a covalently closed circular DNA containing the cleavage site, PI-MtuI can cleave the substrate by two alternative mechanisms: generate linear double-stranded DNA sequentially by nicking one strand of form I DNA, followed by cutting in the second strand. Alternatively, convert supercoiled DNA directly to its linear form by cutting both the strands in a concerted manner. The data in Fig. 7A show the products of PI-MtuI cleavage reaction as a function of time. Under these conditions, a relatively slow phase was observed in the first 30 min of the reaction, during which the amount of form I DNA declined with corresponding increase in the levels of form II DNA. A small amount of form III DNA was also produced before the reaction reached steady state by 3 h. Overall, the reaction yielded more of form II relative to form III DNA. The increase in the amount of form II and III DNA coincided with a corresponding decrease in the amount of form I DNA. To analyze the rate of the reaction, the percentage of product in the reaction was quantified and plotted as a function of time. Interestingly, the plot of product accumulation shows a considerable lag followed by linear increase, and then plateaus after 250 min (Fig. 7B). The physiological relevance of lag in product accumulation is unclear and remains to be investigated. These results, taken together with the data from Figs. 5 and 6, argue that the reaction mechanism of cleavage of target DNA by PI-MtuI follows a sequential pathway. Although this mechanism is not unique, it is one of the simplest mechanisms that can fit the data presented in this report.


View larger version (22K):
[in this window]
[in a new window]
  		Fig. 7.   Kinetics of cleavage of target DNA by PI-MtuI. A mixture of negatively supercoiled and nicked circular duplex was incubated with PI-MtuI in a standard assay buffer containing 16 µM pEJ244 DNA, 1 µM PI-MtuI, 3 mM Mn2+, and 1.5 mM ATP. A, aliquots were removed at the indicated time intervals, and the reaction was terminated. Following deproteinization, products were separated on an agarose gel as described in the legend to Fig. 5. The positions of substrates and products are indicated on the left. B, kinetics of accumulation of form III DNA as a function of time. The extent of formation of form III DNA is shown as the mean values from six independent experiments determined by scanning the autoradiograms with UVItech gel documentation system.

In regard to the slow rate of catalysis, we considered two possibilities: first, the formation of enzyme·substrate complex or the release of product from the enzyme, likely to be the rate-limiting step. In an effort to gain insights into these possibilities, we measured the rate of formation of the PI-MtuI·DNA complex by SPR spectroscopy. The association rate constant (kon) was in the range of 104 to 105 M-1 s-1 indicating that the formation of enzyme·DNA complex with both single- as well as double-stranded DNA was rapid (Table I). Alternatively, the overall catalytic rates may likely to be affected by the release of the product from the enzyme. A key prediction is that decrease in rate should be observed upon addition of product to an ongoing reaction. To test this possibility, kinetics of the reaction was measured in the presence of excess of product (form III DNA). Significantly, at concentrations of product from 0.05 to 1 µM, we observed no detectable inhibition in the rate of cleavage (data not shown). Second, it is possible, if not likely, that the rate of catalysis is intrinsically slow, which remains to be investigated. However, it is formally possible that the low specific activity of PI-MtuI is because of the nature of the recognition site rather than the intrinsic property of the enzyme (see below).

Mapping of PI-MtuI Cleavage Sites-- Cleavage sites were mapped at the nucleotide level by using 32P-labeled DNA fragments produced from restriction digestion of pEJ244 plasmid DNA. The end-labeled fragments were incubated in the presence or absence of PI-MtuI in standard assay buffer containing Mn2+-ATP, purified on polyacrylamide gels. The cleavage products together with Maxam-Gilbert chemical sequencing ladder were separated on denaturing polyacrylamide gels (Fig. 8, A-C). Interestingly, all the incisions made by PI-MtuI are focused in a region encompassing 22 bp around the cleavage site. On the upper strand, nicks were made at four contiguous positions (lane marked "+", Fig. 8A) in the left boundary 24 bp away from the site of insertion of recA intervening sequence (open vertical arrow in Fig. 8D). Whereas on the lower strand, nicks were made at three prominent positions, a spacer of 10 bp separates the first from succeeding sites, at 34, 45, and 46 bp again in the left boundary from the place of insertion of recA intervening sequence (lane marked "+", Fig. 8B). Intriguingly, we found no detectable cleavage at the exact site of insertion of recA intervening sequence (arrowhead in Fig. 8C, lane marked "+"). Quantitation of signals corresponding to the cleavage sites on both the upper and lower strands is summarized in Fig. 8D. The cleavage pattern resulted in the generation of fragments with 3' overhangs. The DNA ends produced by PI-MtuI possessed 5'-phosphate and 3'-hydroxyl groups, because they could be labeled by polynucleotide kinase and extendable by terminal deoxynucleotidyl transferase, respectively (data not shown). While this report was under preparation, Saves et al. (34) reported that the recognition and cleavage sequence site of Pps1 intein endonuclease (designated PI-MgaI) of Mycobacterium gastri spans 22 bp, similar to PI-MtuI.


View larger version (39K):
[in this window]
[in a new window]
  		Fig. 8.   Sequence-specific cleavage of DNA by PI-MtuI. A-C, to determine the exact position of cleavage, DNA fragments centered surrounding the insertion site of intein-less recA allele were generated and individually labeled either at 3'- or 5'-end (shown as asterisks). The fragments were incubated either with PI-MtuI or subjected to Maxam-Gilbert sequencing reaction (C, G, T+C, or G+A) as described under "Materials and Methods." The reaction mixtures were electrophoresed in a denaturing polyacrylamide sequencing gel. The + and - symbols on the top of each panel correspond to reactions performed in the presence or absence of PI-MtuI, respectively. The nucleotide sequence surrounding the site of cleavage is shown by horizontal arrows on the right side of each panel. D, summary of cleavage of the target site sequence. The radioactivity in the bands at the target sequence was quantified by PhosphorImager analysis. The open vertical arrow denotes the site of insertion of intervening sequence in intein-less recA allele. Solid and dotted vertical arrows on the upper and lower strands denote the position of cleavage sites, and thicker arrows indicate the end result of more intense cleavage.


    DISCUSSION
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES

In contrast to E. coli, the pathway of homologous genetic recombination is poorly understood in mycobacteria: however, the protein components involved are just beginning to emerge. A growing body of evidence has emphasized the complexity of this process in M. tuberculosis (35, 36). Thus, previous studies have highlighted the unusual structural organization of M. tuberculosis recA and the mechanistic aspects of homologous recombination promoted by its gene product (6-10). Here, we report that recA intervening sequence encodes a novel Mn2+-ATP-dependent, double-strand-specific endonuclease, which is likely to be important for homing process in vivo.

DNA Binding Properties of PI-MtuI-- Whereas the cleavage mechanism utilized by homing endonucleases is beginning to be understood (12, 16), their interaction with DNA has remained obscure. Much of the work on DNA binding characteristics of homing endonucleases has focused on probing the interaction between the DNA sequence at the homing site and enzyme. However, there are still numerous details that we do not fully understand. Remarkably, PI-MtuI bound to both single- and double-stranded DNA to somewhat similar extents in the band shift assay. More surprisingly, SPR measurements showed that PI-MtuI bound both single- and double-stranded DNA with equal affinity and that the rate of association (kon) was similar in magnitude. However, it was surprising to discover that binding to single-stranded DNA failed to evoke single-strand nicking activity of PI-MtuI (see below). Therefore, it is possible that binding of PI-MtuI to double-stranded DNA is of primary importance, whereas its binding to single-stranded DNA is ancillary.

Target Recognition by PI-MtuI-- The biochemical activities associated with one or more eubacterial intein-encoded homing endonucleases are less well understood. Nevertheless, it is important to compare them with eukaryotic and archaeal homing endonucleases in regard to target site selectivity and cleavage specificity. Several studies have revealed that homing endonucleases recognize and cleave widely divergent target sites ranging from 15 to 40 bp. The long length of cleavage site is believed to confer high specificity for their target sites and avoid inadvertent cleavage of host genome (12, 15, 16). The sequence alignment of target sites of representative endonucleases from eukaryotic, eubacterial, and archaeal origin is shown in Table II. It is evident from the sequence alignment that the target sites are asymmetric in sequence and lacks a consensus motif. On the other hand, I-PpoI, I-CreI, and I-CeuI exhibit limited symmetry near the intron insertion site (12). Interestingly, comparison of target sites revealed distinct differences in sequence as well as length. PI-MtuI recognized a long target site in DNA.

                              
View this table:
[in this window]
[in a new window]
 
Table II
Sequence alignment of cleavage sites of homing endonucleases
The vertical arrows at the top of the table, and in the PI-MtuI column, denote intein insertion sites. The "diamond" symbols contiguous with the nucleotide sequence indicate the position of cleavage on the upper strand, and "underscore" at the lower portion of the nucleotide sequence denote the position of cleavage on the lower strand of the canonical target site. "I-" corresponds to intron-encoded endonuclease, and "PI-" corresponds to protein intein encoded endonuclease.

Critical insights into the molecular mechanism of DNA recognition has been inferred from the cocrystal structures of I-CreI (37, 38) and I-PpoI (39, 40) with their target sites, and from the apo-structures of I-DmoI (41), PI-PfuI (42), and PI-SceI (43). Notably, sequence-specific recognition is achieved by direct intrusion of an anti-parallel beta -sheet comprised of four beta -strands with dimensions and curvature similar to that of the major groove. Further specificity for DNA binding is mediated by the unique pattern of hydrogen bond donors and acceptors presented by the base pairs in the major groove. However, the structures of several members of homing endonucleases must be solved with both cognate and non-cognate sites to fully understand whether all of them recognize DNA through one or more similar or totally different mechanisms.

Target Cleavage and Divalent Cation Specificity-- Initially, we failed to detect endonuclease activity of PI-MtuI with linear duplex DNA or oligonucleotide substrates containing the inteinless recA allele as substrate. This could be construed by the fact that PI-MtuI inflicts a staggered double-strand break encompassing 22 bp. The lack of reports demonstrating endonuclease activity of PI-MtuI, including our own previous difficulties contrasts the facile success with negatively supercoiled DNA. Perhaps this could be attributable to non-ideal behavior of PI-MtuI with substrates used in previous studies.

In general, homing endonucleases show relaxed specificity in target site cleavage and often cleave variant sites as efficiently as wild-type sequences. PI-MtuI cleaved target DNA less efficiently and required both Mn2+ and ATP. The sharing of an essential divalent cation among active sites has been a subject of fascination to the nuclease field (44). Recent studies, particularly the determination of the crystal structure of I-CreI, has unveiled a novel DNA cleavage mechanism in which two catalytic active centers share three Mg2+ ions (37). In this regard, we show that PI-MtuI inflicts double-strand breaks using Mn2+-ATP cofactors. We speculate that PI-MtuI harbors a catalytic center with a subset of amino acid residues active in the presence of thiophilic Mn2+. This is, to our knowledge, the first example of a homing endonuclease exhibiting a preference for thiophilic metal and ATP for the display of cleavage activity.

The members of the LAGLIDADG class of homing endonucleases cleave target DNA in the presence of either Mg2+ or Mn2+, with Mg2+ being the preferred metal ion, and generate fragments of the same size (12, 15, 16). For example, several independent methods have demonstrated that PI-SceI and PI-PfuI contain two distinct active sites. This implies that the enzyme bears functionally and mechanistically different active centers for cofactors, and possibly for substrates, in the same subunit or at the interface of enzyme subunits. The strong structural conservation implies strong functional conservation. In common with a few members of the family of homing endonucleases, however, a major functional difference is in regard to NTPase properties that might be of physiological significance: PI-MtuI showed a marked dependence upon ATP for cleavage of target DNA. Correspondingly, PI-MtuI contains a putative ATP-binding motif (125GWVGGKT133) (Walker A motif) found in a wide variety of NTP, generally ATP, cleavage enzymes. The importance of ATP binding and its hydrolysis for PI-MtuI cleavage remains unknown. However, as in many cases, it is possible that ATP hydrolysis might accelerate the product release after cleavage reaction by disrupting interaction between the elements of the PI-MtuI·DNA complex. Alternatively, or in addition, cleavage of ATP might drive conformational transitions in PI-MtuI that move appropriate domains closer or farther apart relative to each other to facilitate cleavage on both the strands. Indeed, the kinetic mechanism of the PI-MtuI reaction supports a sequential rather than concerted mechanism of strand cleavage. Interestingly, the high resolution mapping of the cleavage sites in target DNA by PI-MtuI revealed that cleavage sites are asymmetric in sequence.

Comparison of PI-MtuI with Other Homing Endonucleases-- Homing endonucleases have been something of an enigma: The recognition and cleavage sites are exceptionally long and often asymmetric. Most of the homing endonucleases cleave their target sites adjacent to the intron or intein insertion sites to generate three to five nucleotide 3' overhangs. Interestingly, PI-MtuI cuts the canonical site in the left flanking sequence 24 bp away from the intein insertion site. Similarly, I-TevI and I-TevIII cut at 23 and 16 bp, respectively, in the left flanking sequence away from the intron insertion site, whereas I-TevII cuts at 15 bp in the right flanking sequence away from the intron insertion site. More recently, it has been demonstrated that recognition and cleavage sequence for M. gastri Pps1 endonuclease spans 22 bp (Table II). What would be the biological significance of inflicting breaks in the flanking sequence of the intein or intron insertion points? One obvious benefit would be to protect the integration site from exonucleolytic degradation. The additional differences between PI-MtuI and other homing endonucleases presumably reflect evolutionary modifications to optimize its propagation and survival (45). Regardless of the precise mechanism, the data reported here uncover important differences between PI-MtuI and the class of LAGLIDADG homing endonucleases. This study initiates further structural and functional analyses aimed at understanding the physiological role(s) of PI-MtuI in M. tuberculosis.

    ACKNOWLEDGEMENTS

We thank Drs. M. J. Colston and E. O. Davis of National Institutes for Medical Research, Mill Hill, London for plasmids.

    FOOTNOTES

* This work was supported in part by grants from the Wellcome Trust, UK, and Indian Council of Medical Research, New Delhi, India.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.

Dagger Supported by a fellowship from the Council of Scientific and Industrial Research, New Delhi, India.

§ To whom correspondence should be addressed. Tel.: 91-80-360-0278; Fax: 91-80-360-0814; E-mail: kmbc@biochem.iisc.ernet.in.

Published, JBC Papers in Press, February 15, 2002, DOI 10.1074/jbc.M112365200

    ABBREVIATIONS

The abbreviations used are: PI-MtuI, RecA intein endonuclease of M. tuberculosis; ATPgamma S, adenosine 5'-O-(thiotriphosphate); DTT, dithiothreitol; form I DNA, negatively supercoiled DNA; form II DNA, nicked circular double-stranded DNA, form III DNA, linear double-stranded DNA; SPR, surface plasmon resonance; ssDNA, single-stranded DNA; GST, glutathione S-transferase.

    REFERENCES
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES

1. West, S. C. (1992) Annu. Rev. Biochem. 61, 603-640[CrossRef][Medline] [Order article via Infotrieve]
2. Kowalczykowski, S. C., Dixon, D. A., Eggleston, A. K., Lauder, S. D., and Rehrauer, W. M. (1994) Microbiol. Rev. 58, 401-465[Abstract/Free Full Text]
3. Kuzminov, A. (1999) Microbiol. Mol. Biol. Rev. 63, 751-813[Abstract/Free Full Text]
4. Cox, M. M. (1999) Prog. Nucleic Acids Res. Mol. Biol. 63, 311-366[Medline] [Order article via Infotrieve]
5. Roca, A. I., and Cox, M. M. (1997) Prog. Nucleic Acids Res. Mol. Biol. 56, 129-223[Medline] [Order article via Infotrieve]
6. Davis, E. O., Sedgwick, S. G., and Colston, M. J. (1991) J. Bacteriol. 173, 5653-5662[Abstract/Free Full Text]
7. Davis, E. O., Thangaraj, H. S., Brooks, P. C., and Colston, M. J. (1994) EMBO J. 13, 699-703[Medline] [Order article via Infotrieve]
8. Davis, E. O., Jenner, P. J., Brooks, P. C., Colston, M. J., and Sedgwick, S. G. (1992) Cell 71, 201-210[CrossRef][Medline] [Order article via Infotrieve]
9. Kumar, R. A., Vaze, M. B., Chandra, N. R., Vijayan, M., and Muniyappa, K. (1996) Biochemistry 35, 1793-1803[CrossRef][Medline] [Order article via Infotrieve]
10. Vaze, M. B., and Muniyappa, K. (1999) Biochemistry 38, 3175-3186[CrossRef][Medline] [Order article via Infotrieve]
11. Datta, S., Prabu, M. M., Vaze, M. B., Ganesh, N., Chandra, N. R., Muniyappa, K., and Vijayan, M. (2000) Nucleic Acids Res. 28, 4964-4973[Abstract/Free Full Text]
12. Belfort, M., and Roberts, R. J. (1997) Nucleic Acids Res. 25, 3379-3388[Abstract/Free Full Text]
13. Liu, X. Q. (2000) Annu. Rev. Genet. 34, 61-76[CrossRef][Medline] [Order article via Infotrieve]
14. Noren, C. J., Wang, J., and Perler, F. B. (2000) Angew. Chem. Int. Ed. Engl. 39, 450-466[CrossRef][Medline] [Order article via Infotrieve]
15. Gimble, F. S. (2000) FEMS Microbiol. Lett. 185, 99-107[CrossRef][Medline] [Order article via Infotrieve]
16. Chevalier, B. S., and Stoddard, B. L. (2001) Nucleic Acids Res. 29, 3757-3774[Abstract/Free Full Text]
17. Mueller, J. E., Bryk, M., Loizos, N., and Belfort, M. (1993) in Nucleases (Linn, S. M. , Lloyd, R. S. , and Roberts, R. J., eds) , pp. 111-144, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY
18. Bos, J. L., Heyting, C., Borst, P., Arnberg, A. C., and Van Bruggen, E. F. (1978) Nature 275, 336-338[CrossRef][Medline] [Order article via Infotrieve]
19. Dujon, B. (1980) Cell 20, 185-197[CrossRef][Medline] [Order article via Infotrieve]
20. Belfort, M., and Perlman, P. S. (1995) J. Biol. Chem. 270, 30237-30240[Free Full Text]
21. Cole, S. T., Brosch, R., Parkhill, J., Garnier, T., Churcher, C., Harris, D., Gordon, S. V., Eiglmeier, K., Gas, S., Barry, C. E., III, Tekaia, F., Badcock, K., Basham, D., Brown, D., Chillingworth, T., et al.. (1998) Nature 393, 537-544[CrossRef][Medline] [Order article via Infotrieve]
22. Derbyshire, V., Wood, D. W., Wu., W., Dansereau, J. T., Dalgaard, J. Z., and Belfort, M. (1997) Proc. Natl. Acad. Sci. U. S. A. 94, 11466-11471[Abstract/Free Full Text]
23. Mills, K. V., Lew, B. M., Jiang, S., and Paulus, H. (1998) Proc. Natl. Acad. Sci. U. S. A. 95, 3543-3548[Abstract/Free Full Text]
24. Shingledecker, K., Jiang, S. Q., and Paulus, H. (1998) Gene 207, 187-195[CrossRef][Medline] [Order article via Infotrieve]
25. Paulus, H. (2000) Annu. Rev. Biochem. 69, 447-496[CrossRef][Medline] [Order article via Infotrieve]
26. Cunningham, R. P., DasGupta, C., Shibata, T., and Radding, C. M. (1980) Cell 20, 223-235[CrossRef][Medline] [Order article via Infotrieve]
27. Sambrook, J., Fritsch, E. F., and Maniatis, T. (1989) Molecular Cloning: A Laboratory Manual , 2nd. Ed. , Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY
28. Bradford, M. M. (1976) Anal. Biochem. 72, 248-254[CrossRef][Medline] [Order article via Infotrieve]
29. Laemmli, U. K. (1970) Nature 227, 680-685[CrossRef][Medline] [Order article via Infotrieve]
30. Bakshi, R. P., Galande, S., Bali, P., Dighe, R., and Muniyappa, K. (2001) J. Mol Endocrinol. 26, 193-206[Abstract]
31. Maxam, A. M., and Gilbert, W. (1980) Methods Enzymol. 65, 499-560[Medline] [Order article via Infotrieve]
32. Komori, K., Ichinayagi, K., Morikawa, K., and Ishino, Y. (1999) Nucleic Acids Res. 27, 4175-4182[Abstract/Free Full Text]
33. Gimble, F. S., and Stephens, B. W. (1995) J. Biol. Chem. 270, 5849-5856[Abstract/Free Full Text]
34. Saves, I., Westrelin, F., Daffe, M., and Masson, J.-M. (2001) Nucleic Acids Res. 29, 4310-4318[Abstract/Free Full Text]
35. Pelicic, V., Reyrat, J.-M., and Gicquel, B. (1998) Mol. Microbiol. 28, 413-420[CrossRef][Medline] [Order article via Infotrieve]
36. Norman, E., Dellagostin, O. A., McFadden, J., and Dale, J. W. (1995) Mol. Microbiol 16, 755-760[CrossRef][Medline] [Order article via Infotrieve]
37. Chevalier, B. S., Monnat, R. J., Jr., and Stoddard, B. L. (2001) Nat. Struct. Biol. 8, 312-316[CrossRef][Medline] [Order article via Infotrieve]
38. Heath, P. J., Stephens, K. M., Monnat, R. J., Jr., and Stoddard, B. L. (1997) Nat. Struct. Biol 4, 468-476[CrossRef][Medline] [Order article via Infotrieve]
39. Galburt, E. A., Chevalier, B., Tang, W., Jurica, M. S., Flick, K. E., Monnat, R. J., Jr., and Stoddard, B. L. (1999) Nat. Struct. Biol. 6, 1096-1099[CrossRef][Medline] [Order article via Infotrieve]
40. Galburt, E. A., Chadsey, M. S., Jurica, M. S., Chevalier, B., Erho, D., Tang, W., Monnat, R. J., Jr., and Stoddard, B. L. (2000) J. Mol. Biol. 300, 877-887[CrossRef][Medline] [Order article via Infotrieve]
41. Silva, G. H., Dalgaard, J. Z., Belfort, M., and Van Roey, P. (1999) J. Mol. Biol. 286, 1123-1136