Document Detail

Paired mutations abolish and restore the balanced annealing and melting activities of ORF1p that are required for LINE-1 retrotransposition.
Jump to Full Text
MedLine Citation:
PMID:  21441536     Owner:  NLM     Status:  MEDLINE    
Abstract/OtherAbstract:
Retrotransposition amplifies LINE-1 (L1) to high copy number in mammalian genomes. The L1 protein encoded by ORF1 (ORF1p) is required for retrotransposition. This dependence on ORF1p was investigated by mutating three highly conserved residues, R238, R284 and Y318 to alanine, thereby inactivating retrotransposition. R284A and Y318A were rescued by further substituting the alanine with the appropriate conservative amino acid, e.g. lysine or phenylalanine, respectively, whereas R238K remained inactive. Quantification of the steady-state levels of L1 RNA and ORF1p failed to discriminate active from inactive variants, indicating loss of L1 retrotransposition resulted from loss of function rather than reduced expression. The two biochemical properties known for ORF1p are high-affinity RNA binding and nucleic acid chaperone activity. Only R238A/K exhibited significantly reduced RNA affinities. The nucleic acid chaperone activities of the remaining paired mutants were assessed by single-molecule DNA stretching and found to mirror retrotransposition activity. To further examine ORF1p chaperone function, their energetic barriers to DNA annealing and melting were derived from kinetic work. When plotted against each other, the ratio of these two activities distinguished functional from non-functional ORF1p variants. These findings enhance our understanding of the requirements for ORF1p in LINE-1 retrotransposition and, more generally, nucleic acid chaperone function.
Authors:
James D Evans; Suresh Peddigari; Kathy R Chaurasiya; Mark C Williams; Sandra L Martin
Related Documents :
6580756 - Effects of xylitol on the acid production activity from sorbitol by streptococcus mutan...
17178806 - The effect of fermentation quality on the voluntary intake of grass silage by growing c...
3717806 - Organic acids in ethylene glycol intoxication.
11302166 - Growth of nisin-producing lactococci in cooked rice supplemented with soybean extract a...
531496 - Fatty acid metabolism during hypothermic perfusion of the isolated dog kidney.
2382576 - Excretion of 3-mercaptolactate-cysteine mixed disulfide, sulfate and taurine in human u...
Publication Detail:
Type:  Journal Article; Research Support, N.I.H., Extramural; Research Support, U.S. Gov't, Non-P.H.S.     Date:  2011-03-26
Journal Detail:
Title:  Nucleic acids research     Volume:  39     ISSN:  1362-4962     ISO Abbreviation:  Nucleic Acids Res.     Publication Date:  2011 Jul 
Date Detail:
Created Date:  2011-07-25     Completed Date:  2011-10-03     Revised Date:  2013-06-30    
Medline Journal Info:
Nlm Unique ID:  0411011     Medline TA:  Nucleic Acids Res     Country:  England    
Other Details:
Languages:  eng     Pagination:  5611-21     Citation Subset:  IM    
Affiliation:
Department of Cell and Developmental Biology, University of Colorado School of Medicine, Aurora, CO 80045, USA.
Export Citation:
APA/MLA Format     Download EndNote     Download BibTex
MeSH Terms
Descriptor/Qualifier:
Amino Acid Substitution
Animals
DNA / chemistry
Long Interspersed Nucleotide Elements*
Mice
Mutation
Nucleic Acid Denaturation
RNA / metabolism
Retroelements / genetics*
Grant Support
ID/Acronym/Agency:
P30 CA046934/CA/NCI NIH HHS; R01 GM072462-08/GM/NIGMS NIH HHS; R01GM40367/GM/NIGMS NIH HHS; R01GM75965/GM/NIGMS NIH HHS
Chemical
Reg. No./Substance:
0/Retroelements; 63231-63-0/RNA; 9007-49-2/DNA
Comments/Corrections

From MEDLINE®/PubMed®, a database of the U.S. National Library of Medicine

Full Text
Journal Information
Journal ID (nlm-ta): Nucleic Acids Res
Journal ID (publisher-id): nar
Journal ID (hwp): nar
ISSN: 0305-1048
ISSN: 1362-4962
Publisher: Oxford University Press
Article Information
Download PDF
© The Author(s) 2011. Published by Oxford University Press.
creative-commons:
Received Day: 12 Month: 1 Year: 2011
Revision Received Day: 9 Month: 3 Year: 2011
Accepted Day: 10 Month: 3 Year: 2011
collection publication date: Month: 7 Year: 2011
Print publication date: Month: 7 Year: 2011
Electronic publication date: Day: 26 Month: 3 Year: 2011
pmc-release publication date: Day: 26 Month: 3 Year: 2011
Volume: 39 Issue: 13
First Page: 5611 Last Page: 5621
ID: 3141268
PubMed Id: 21441536
DOI: 10.1093/nar/gkr171
Publisher Id: gkr171

Paired mutations abolish and restore the balanced annealing and melting activities of ORF1p that are required for LINE-1 retrotransposition
James D. Evans1
Suresh Peddigari1
Kathy R. Chaurasiya2
Mark C. Williams23*
Sandra L. Martin14*
1Department of Cell and Developmental Biology, University of Colorado School of Medicine, Aurora, CO 80045, 2Department of Physics, 3Center for Interdisciplinary Research on Complex Systems, Northeastern University, 111 Dana Research Center, Boston, MA 02115 and 4Program in Molecular Biology, University of Colorado School of Medicine, Aurora, CO 80045, USA
Correspondence: *To whom correspondence should be addressed. Tel: +303 724 3467; Fax: +303 724 3420; Email: Sandy.Martin@UCDenver.edu
Correspondence: *Correspondence may also be addressed to Mark C. Williams. Tel: +617 373 7323; Fax: +617 373 2943; Email: mark@neu.edu

INTRODUCTION

Long interspersed element-1 (LINE-1, or L1) is a successful and active retrotransposon in mammals. L1 sequence directly accounts for 17% of the human and 19% of the mouse genomes. L1 also mobilizes non-autonomous elements including short interspersed elements (SINEs) and processed pseudogenes, making it responsible for at least 30% of mammalian DNA (1). The structure and function of the genome is impacted by L1 at multiple levels, which range from insertional mutagenesis to altering gene expression and promoting recombination [reviewed in ref. (2)].

A majority of the 660 000 L1 elements in the mouse genome are truncated and therefore cannot retrotranspose, but approximately 3000 are estimated to be competent for retrotransposition (3). Active mouse L1 is characterized by a series of monomer promoter motifs followed by a 5′-UTR, two open reading frames (ORF1 and ORF2), a 3′-UTR, poly-A signal and an A-rich region, all embedded within a short target site duplication. Retrotransposition begins with transcription of one of the active elements, followed by translation of the two L1-encoded proteins, ORF1p and ORF2p. Both L1 proteins are essential for retrotransposition (4), and both are required in cis (5). The L1 RNA ultimately serves as a template for target-primed reverse transcription [TPRT, (6)] resulting in a new L1 insertion in the genome.

The role of ORF1p in L1 retrotransposition is incompletely understood. The protein forms a stable trimer via an N-terminal coiled-coil domain (7), followed by an RRM domain (8) and a CTD [(9), Figure 1A]. Two biochemical properties of ORF1p are known: RNA binding and nucleic acid chaperone activity. ORF1p from mouse L1 is a high-affinity, non-sequence-specific RNA binding protein (10), which is likely critical for packaging the L1 RNA during retrotransposition (11–14). The RNA binding region of the protein is C-terminal of the coiled-coil, encompassing the RRM and CTD domains, which are both required for high-affinity RNA binding (8,15). ORF1p is also a nucleic acid chaperone protein, a property which is required for retrotransposition (16), and may facilitate TPRT (17,18). Mutations exist in ORF1p that do not disrupt the high-affinity interaction of the protein with RNA but nevertheless abolish retrotransposition (16,17,19), consistent with a critical role for the nucleic acid chaperone activity of ORF1p in L1 replication.

Nucleic acid chaperones paradoxically promote both annealing of complementary single stranded (ss) and melting of double-stranded (ds) nucleic acids. A consequence of these two fundamental activities is rearrangement of nucleic acid secondary structure, a property that is essential in a variety of biological processes including retroviral replication (20,21), functional splicing of the T4 phage thymidylate synthase gene (22) and proper folding of a yeast tRNA anticodon (23).

Single-molecule stretching is a sensitive assay of nucleic acid chaperone activity. When single dsDNA molecules are stretched, the applied force facilitates melting of the DNA. When the DNA molecule is relaxed, DNA reannealing occurs. By measuring DNA stretching and relaxation in the presence of nucleic acid chaperone proteins, the extent to which the proteins facilitate DNA melting and annealing, important components of nucleic acid chaperone activity, can be determined [reviewed in refs (24,25)].

Here, we investigate the effect of six amino acid substitutions in ORF1p on L1 retrotransposition, expression, protein stability, RNA binding, the cooperativity of DNA melting and the kinetics of ORF1p-mediated annealing and strand exchange. We present a novel analysis of the latter data, and use it to quantify the balance between these two opposing components of nucleic acid chaperone activity by comparing wild-type (wt) to mutant proteins. Taken together, these data demonstrate that L1 retrotransposition is exquisitely sensitive to changes in the biophysical properties of ORF1p that determine its nucleic acid chaperone activity.


MATERIALS AND METHODS
Autonomous retrotransposition assay

Mutations were introduced into L1 ORF1 by site-directed mutagenesis and assayed for effects on retrotransposition in 143 B cells using eGFP expression as a marker of retrotransposition as described (5,16).

DNA, RNA and protein analysis

Whole-cell lysates (WCL) were prepared from L1 transfected 143 B cells every 24 h post-transfection as described (17). Protein was quantified (Bradford assay, Bio-Rad) and ORF1p was detected by western blotting in 20 μg of WCL as described (17). hnRNP-Q was detected on the same blots using a mouse monoclonal hnRNP-Q antibody (Abcam). Blot images were captured on a GE Typhoon 9400. L1 RNA was quantified by quantitative real time PCR using primers with a probe specific to the ORF2 region. Total RNA was isolated from WCL using TRIzol-LS (Invitrogen) and quantified (Nanodrop, Thermo Scientific). RNA (2 μg) was treated with RQ1 DNase (Promega) and cDNA was synthesized from 1 µg using a High-Capacity cDNA Reverse Transcription kit (Applied Biosystems). Two microliter cDNA were used for quantitative PCR using TaqMan probe assay with 5′-CTCAGAATGAAAGGCTGGAAAAC as the forward primer, 5′-AGGATGGCTACTCCTGCTTGTT as the reverse primer and 5′-FAM-CCAAGCAAATGGTATGAAG-NFQMGB as the probe. A standard curve was generated using 10-fold dilutions of L1-containing plasmid. 18S rRNA was quantified using the TaqMan Gene Expression assay for 18S rRNA (Applied Biosystems). The quantities of L1 and 18S rRNA in the cDNAs were calculated from their respective standard curves using the absolute standard curve method (Applied Biosystems) and then L1 RNA values were normalized to 18S rRNA values.

Recombinant ORF1p purification

ORF1p coding sequences with single residue mutations were cloned into a modified pFastBac bacmid and used for recombinant expression in SF9 cells as before (18), according to the Bac-to-Bac baculovirus expression system protocol (Invitrogen). We replaced the 6His to ORF1 AUG sequence of pFastBac with the corresponding sequence from pBluBac to facilitate ORF1p purification as described previously (7), except that the ammonium sulfate precipitation was 27% NH4(SO4)2 followed by 73% NH4(SO4)2.

Circular dichroism and thermal melt

Circular dichroism (CD) and melt scans were performed on a Jasco 815 spectropolarimeter with purified ORF1p between 2.1 and 4.2 μM trimer concentration in 50 mM phosphate, pH 7.6; 250 mM NaCl and 0.1 mM EDTA, and data were collected as the observed ellipticity in millidegree. Thermal stability was determined by observing the differential absorption at 222 nm while the temperature was raised at a rate of 2°C per minute, from 4 to 70°C. Data were normalized to the 4°C data point.

RNA filter binding

Assays were performed in 250 mM NaCl and analyzed as described previously (10). Differences in KDapp of 2- to 3-fold were not significant because of experimental variation (16).

Single-molecule DNA stretching

A dual beam optical tweezers instrument was used to stretch 5′ biotin-end-labeled bacteriophage λ DNA as described elsewhere in detail (26–28). Briefly, the DNA was captured on streptavidin-coated polystyrene beads (Bangs Labs), then stretched and relaxed after rinsing out other DNA molecules. All stretching experiments were performed at a pulling rate of 100 nm/s in 10 mM HEPES; 50 mM Na+; pH 7.5 buffer. Specific protein solutions were exchanged with the buffer around the DNA molecule to probe protein–DNA interactions. To quantify the effects of protein on the DNA stretching curves, the transition width was determined from the transition slope at its midpoint as described previously (29). Significant differences among protein groups were assessed by one way ANOVA followed by Tukey's pairwise comparisons using KaleidaGraph 4.03 (Synergy Software).

Annealing and displacement assays

For annealing, ORF1p and two complementary, 48-nt DNA oligonucleotides, one end-labeled with 32P, were incubated at equimolar concentrations in 30 μl reaction buffer (20 mM HEPES pH 7.6; 25 mM NaCl; 1 mM EDTA pH 8.0; 1 mM MgCl2; 1 mM dithiothreitol and 0.1% Triton-X 100 w/v). Reactions with ORF1p were incubated at 1, 15, 22 and 37°C; those without protein were incubated at 15, 22, 37 and 42°C. Aliquots (3 μl) were taken at the noted times, mixed with 3 μl stop buffer (40 mM HEPES pH 7.6; 0.4 mg/ml tRNA; 0.2% sodium dodecyl sulfate w/v; 10 mM EDTA pH 8.0; 3% Ficoll-400 w/v; 0.25% bromophenol blue w/v and 0.25% xylene cyanol w/v) and placed on ice. Samples were fractionated through a 15%, 19:1 bis:acrylamide gel at 4°C in 0.5× TBE, the gel dried and exposed to a storage phosphor screen. Images were collected on a Typhoon 9400 and bands containing ss and ds oligonucleotides were quantified with ImageQuaNT 5.2 (GE). KaleidaGraph was used for curve fitting and graphical representation of data.

Full duplex exchange experiments to quantify melting catalysis facilitating strand exchange were performed by mixing 1 nM 38-bp dsDNA, 5′ end-labeled with 32P on one of the strands, with 10 nM of the identical unlabeled ss oligonucleotide competitor and 10 nM ORF1p trimer in reaction buffer. This strand exchange assay differs from a strand-exchange assay used previously to study ORF1p because the large thermodynamic advantage that drove the previous reactions [replacement of a short (18) or imperfectly matched (16) strand in a preformed duplex with the perfect complement, provided at 50× molar excess] has been removed. Reactions were incubated with ORF1p at 22, 27, 32 and 37°C, or 37, 40, 42, 45, 47 and 50°C for reactions without protein, and then treated as described for the annealing reactions.


RESULTS

Previous studies demonstrated that retrotransposition of mouse L1 was reduced or abolished by the substitutions D159H and R297K in ORF1p (16,17). These residues lie in the coiled-coil and CTD, respectively, far apart in the primary sequence of ORF1p (Figure 1A). To further probe the functional requirements for ORF1p in L1 retrotransposition we substituted alanine for three conserved residues in the RRM and CTD domains of the protein in the context of an otherwise active L1: R238A, R284A or Y318A (Figure 1B). All three of these mutations rendered L1 inactive for retrotransposition in a cultured cell assay. Replacing the alanine to make the variants R284K and Y318F nearly or completely restored retrotransposition, respectively, but R238K remained inactive (Figure 1B). Retrotransposition detected by eGFP fluorescence using flow cytometry was confirmed by PCR amplification of the spliced eGFP from genomic DNA (data not shown).

Single residue mutations may decrease protein stability (30,31). Mutations in ORF1p would likely interfere with L1 retrotransposition if the protein became unstable or unable to interact with L1 RNA, which might in turn cause instability of L1 RNA. Therefore the steady-state levels of L1 RNA and ORF1p were measured daily from cells transfected with wt and mutant L1 constructs. Based on qRT-PCR, L1 RNA was not reduced below wt levels in any of the inactive mutants. Moreover, differences in abundance of L1 RNA did not distinguish among active and inactive elements (Figure 2A). The highest steady-state levels of ORF1p detected in the lysates by western blotting were consistently observed 2 days post-transfection. Small fold changes of ORF1p abundance among mutants again failed to distinguish between active and inactive ORF1p variants (Figure 2B). These data indicate the defects in retrotransposition must be due to something other than instability or loss of L1 RNA or ORF1p.

Mutant proteins were purified (Supplementary Figure S1A) in order to study their biochemical properties. The final purification step is size exclusion chromatography. All of the mutant proteins eluted in the same fraction as wt ORF1p, indicating that they share its homotrimeric asymmetric dumbbell-shaped structure (7).

Mutant proteins were characterized by their differential absorption of circularly-polarized light. All mutant and wt ORF1p produced absorption patterns consistent with a generally helical molecule (data not shown). If local secondary structure were perturbed by some of these mutations, it was insufficient to affect the overall absorption characteristics of the molecule. Additionally, the effect of increasing temperature on elliptical absorption at 222 nm revealed that the proteins denatured between 38 and 46°C (Supplementary Figure S1B). Although R238K and Y318F were more temperature sensitive than wt ORF1p, this property was not correlated with retrotransposition competence.

A nitrocellulose filter binding assay was used to assess whether any of the ORF1p mutants were compromised for RNA binding. As shown in Figure 3, the relative affinities were: Y318F ≥ Y318A ≥ wt ≥ R284K ≥ R284A >> R238K ≥ R238A. Because of experimental variation, however, only the two substitutions at R238 were significantly different than wt, each exhibiting an ∼10-fold reduction in their apparent affinity for RNA. It is likely that the retrotransposition defects in R238A and R238K are due to their reduced affinity for RNA (19), however, altered RNA binding does not explain the loss of retrotransposition in R284A or Y318A. The inability of these two mutants to retrotranspose cannot be explained by the loss of L1 RNA or ORF1 protein, a failure to trimerize, gross perturbation of secondary structure, decreased thermal stability or altered KDapp for RNA. Thus, some other property of ORF1p must be affected by these mutations.

Single-molecule stretching experiments were used to assess the nucleic acid chaperone activities of the paired mutants at R284 and Y318. Figure 4A shows a stretching and relaxation cycle for bacteriophage λDNA in the absence of protein and in the presence of wt ORF1p using optical tweezers. In the absence of protein, very little force is required to stretch the DNA to its full contour length. Near the contour length the force increases dramatically, reflecting the elasticity of the double helix. The approximately constant force plateau of ∼60 pN represents a cooperative transition from dsDNA to ssDNA, or a force-induced melting transition (26,27,32,33). In addition, the force-extension curve in the absence of protein was almost completely reversible, showing little hysteresis, i.e. disagreement between stretching and relaxation. The amount of hysteresis is characterized by the area between the stretching and relaxation curves. In the presence of wt ORF1p, the transition was altered significantly, as also shown in Figure 4A. The DNA force-induced melting plateau was sloped, indicating that the transition cooperativity was significantly reduced. A reduction in melting cooperativity indicates that DNA melting can be more easily initiated and smaller regions of DNA can be melted at once to facilitate DNA rearrangements. This change in shape of the transition was quantified by measuring the transition width, as described in Figure 4A. A greater transition width indicates a less cooperative transition. Therefore, when the transition width is greater, DNA will more easily undergo conformational rearrangements, as expected for a nucleic acid chaperone protein (28,34). The amount of hysteresis observed in the presence of wt ORF1p is greater than that observed for the nucleic acid chaperone HIV-1 NC (25), but significantly less than that observed for the slower nucleic acid chaperone protein HTLV-1 NC (35) and ssDNA binding proteins like T4 gene 32 protein (36). Thus, wt ORF1p exhibits moderately rapid kinetics, intermediate between other nucleic acid chaperones.

While Y318F was similar to wt ORF1p, Y318A had markedly less effect on DNA stretching (average ΔF = 14.7, 17.8 and 6.9, respectively, Figure 4B and D), consistent with the relative retrotransposition results for this mutant pair. Likewise, R284A, which was also defective for retrotransposition, had the least effect on DNA stretching while R284K strongly altered the DNA stretching curve (average ΔF = 2.1 and 9.7, respectively, Figure 4C and D). Thus, as observed with other mutations in ORF1p (16,17), inactivation of L1 retrotransposition in Y318A and R284A was correlated with loss of nucleic acid chaperone activity as measured by single-molecule stretching. Similarly, mutations that restored retrotransposition also restored chaperone activity measured by DNA stretching.

Although the results from single-molecule DNA stretching for the paired mutations at each site correlated well with retrotransposition, the single-molecule data were not able to distinguish within error between two mutants that clearly differed in retrotransposition, R284K and Y318F (Figure 4D). This result may be partially explained by the fact that R284K is less active than wt and Y318F (Figure 1B); however, our understanding of the relationship between the nucleic acid chaperone activity of ORF1p and L1 retrotransposition activity was clearly incomplete.

Two aspects of nucleic acid chaperone function (37), the kinetics of DNA annealing and melting, were then examined in detail. To accomplish this, we developed assays to measure the magnitude of the energetic barriers to annealing and melting of complementary DNA oligonucleotides in the presence of ORF1p. First, complementary DNA oligonucleotides were incubated for 2.5 min at 21°C in the presence of increasing concentrations of ORF1p trimer to determine the mid-point and saturating nt:ORF1p ratio for annealing under these conditions. The annealing rate saturated when the concentration of protein to nucleic acid was ∼9:1, and the half-maximal point was 1.15 nM (Supplementary Figure S2). Thus, all further experiments were performed with equimolar protein and oligonucleotides.

The effect of wt and mutant ORF1p on DNA annealing was determined at varying temperatures. The reaction can be expressed as:

[Formula ID: M1]
(1) 
Here S1 and S2 are the ssDNA strands that anneal to form duplex S1S2 with total rate kanneal. The fraction of labeled oligonucleotide incorporated into duplex was plotted against time and these points were fitted to a capped, inverse exponential decay (Figure 5A):
[Formula ID: M2]
(2) 
where fr is the fraction of labeled oligonucleotide incorporated into duplex as a function of time, is the equilibrium fraction of annealed oligonucleotide, t is the reaction time in seconds and kanneal is the reaction rate constant in s−1.

To find the temperature dependence of the reaction rate, the natural logarithm of the resulting rate coefficient was then plotted against the inverse temperature according to transition state theory (Figure 5B):

[Formula ID: M3]
(3) 
where R is the ideal gas constant, 8.31 J mol−1 K−1, T is the reaction temperature in Kelvin, is Boltzmann's constant, h is Planck's constant, ΔSa is the entropic contribution to the activation energy for annealing in J mol−1 K−1, and ΔHa is the enthalpic contribution to the activation energy for annealing in J mol−1. A line was fitted to these points, allowing us to determine the enthalpic and entropic activation energies for this process. Strand exchange assays were used to determine the energetic barrier to melting and subsequent strand exchange in order to probe the other fundamental property of nucleic acid chaperones. A complementary DNA duplex, one strand labeled with 32P, was incubated with a 10-fold excess of competing, unlabeled oligonucleotide and equimolar ORF1p, again varying only temperature. The excess of unlabeled complementary oligonucleotide assured that the displaced 32P -labeled strand remained displaced for quantification. The time dependence of the reaction could only be fit to two exponentials, and was therefore biphasic:
[Formula ID: M4]
(4) 
here S1 is the unlabeled ssDNA strand and S2S3 is the duplex that contains the labeled ssDNA strand as well as one unlabeled strand. The intermediate complex will be formed when an unlabeled strand encounters a partially melted duplex. This represents the first step in the reaction, which occurs with the faster rate k1. When the final duplex S1S3 is formed, S2 must be dissociated with rate k2. Based on this model, the data points were fit to a two-factor exponential decay (Figure 6A):
[Formula ID: M5]
(5) 
where P is the fractional contribution to the overall rate from the first rate, k1 and k2 are the first and second rates, respectively, and the remaining factors are the same as Equation (2).

Based on previous observations of a biphasic reaction for DNA and RNA hairpin annealing (21,37), the fast rate was assumed to be the primary DNA melting step required for initial annealing of the complex. This hypothesis is also supported by the observation that the fast rate is similar to that observed for other reported chaperone-mediated DNA-duplex melting reactions (21), whereas the rate constant of the slow phase was two to three orders of magnitude slower than that observed for other nucleic acid chaperones. We will show later that the analysis of the fast rates correlate with mutations in ORF1p, further supporting our assumption that the fast rate involves DNA melting needed to initiate strand transfer, a process facilitated by nucleic acid chaperones. In contrast, the slow rates did not depend significantly on the presence of ORF1p. The temperature dependence of the fast rate was plotted and fit to the transition state model as before (Figure 6B).

The enthalpic and entropic contributions to the activation energy for product formation were quantified from Equation (3) (Table 1) for both annealing and exchange and these terms were combined into a total free energy of activation according to:

[Formula ID: M6]
(6) 
where δΔHa is the difference in enthalpic contributions between a particular mutant and the no protein condition, δΔSa is the difference between entropic contributions between a mutant and no protein, T is the temperature in K (here set to 310, or 37°C), and δΔGa is the total difference in the free energy of activation to product formation in the presence of ORF1p.

All mutant and wt ORF1 proteins catalyzed both annealing and strand displacement, although to differing magnitudes. The activation energy for annealing was reduced by 400–900 J mol−1 bp−1 compared to reactions without protein, and the energetic barrier to strand displacement was reduced by 9–15 kJ mol−1 bp−1.

The change in free energy of melting was plotted as a function of that of annealing (Figure 7). This plot revealed an optimum magnitude for ORF1p-mediated annealing and strand-displacement catalytic activities in the retrotransposition-competent elements. All of the tested ORF1p variants exhibited an enhanced reduction in the energetic barrier to annealing, i.e. improved catalysis of annealing, compared to wt independent of retrotransposition activity. However, the retrotransposition-active mutants have both very similar reductions in the barrier to melting and annealing. In contrast, R284A shows strongly enhanced annealing and reduced melting. This strongly enhanced annealing is unable to overcome the reduction in the protein's ability to melt nucleic acids, which is clearly required for chaperone activity. Perhaps most interesting is the result for Y318A, which shows strongly enhanced melting and annealing, yet does not facilitate retrotransposition. This demonstrates that optimum melting and annealing is necessary for retrotransposition. This optimum melting and annealing is restored by substituting to form Y318F, which shows melting and annealing kinetics in these studies that are similar to wt.


DISCUSSION

ORF1p from mouse L1 is a robust nucleic acid chaperone protein that is required for retrotransposition. Previously, substitutions that compromise L1 retrotransposition were mapped to single amino acids in either the coiled-coil (17) or the CTD (16) domains (D159H and R297K in Figure 1A, respectively). Both of these substitutions affected nucleic acid chaperone activity and not RNA binding affinity.

Here, an additional three highly conserved amino acids, two (R238, R284) in the recently described RRM domain (8) and the third, Y318, in the CTD (9), were replaced with alanine, and also found to inactivate L1 retrotransposition. Two of these three new alanine mutations were rescued by substitution of the appropriate amino acid to restore the basic (R284K) or hydrophobic (Y318F) character of the original arginine or tyrosine, whereas the third (R238K) was not (Figure 1). Retrotransposition deficits were independent of changes observed in steady-state levels of L1 RNA or ORF1p, or aberrant ORF1p structure (Figure 2 and Supplementary Figure S1B). Both RNA binding and nucleic acid chaperone activities of ORF1p appear to be critical for L1 retrotransposition because R238A and R238K compromised high-affinity RNA binding (Figure 3), and R284A and Y318A compromised nucleic acid chaperone activity based on reduced cooperativity of DNA melting (Figure 4), and poorly optimized annealing and strand displacement catalysis (Figure 7).

Residues R238 and R284 lie in the RRM domain; the structure of the corresponding fragment from human L1 ORF1p has been solved by X-ray crystallography, revealing that R202 (homologous to R238 in mouse L1) forms a salt bridge together with E169. Disruption of the salt bridge is expected to destabilize ORF1p (8). In the case of mouse L1, this salt bridge does not appear to be important for ORF1p stability because eliminating it with R238A did not alter the thermal melting behavior compared to wt (Supplementary Figure S1B) or reduce its abundance in cells (Figure 2B). Moreover, the salt bridge partner of R202, E169 (D204 in mouse), is rotated to the outside of ORF1p in the modeled structure (Supplementary Figure S3); thus it is unlikely that a salt bridge can form between these two residues in the mouse protein. Interestingly, R238K fails to restore the RNA affinity of ORF1p. Taken together, the available data indicate that R238 is important for interactions between ORF1p and nucleic acids, although the interactions involve more than simple electrostatics.

The second residue of the RRM domain studied here, R284, has not been examined previously. Based upon the crystal structure of the human homolog, R284 is on the surface of the RRM domain in the vicinity of several other arginine residues that form electrostatic interactions with RNA (8). The interpretation that this residue is important for electrostatic interactions is supported by the observed inactivation of retrotransposition by R284A and restoration of activity by R284K. We were unable to measure a significant reduction in RNA affinity in R284A, although a reproducible trend in this direction was observed. The nucleic acid chaperone activity of R284A is, however, significantly altered by an imbalance of annealing and displacement activities because of a markedly decreased ability to catalyze the displacement reaction. In contrast, R284K restores the wt balance between annealing and strand displacement, suggesting that the electrostatic interaction between R284 and nucleic acids is important for the melting component of the nucleic acid chaperone activity.

The final residue examined, Y318 in the CTD, is conserved in placental mammals, marsupials and frogs. The homologous site, however, is substituted with phenylalanine in swimmer, a LINE element in Japanese medaka (8), consistent with the observed retrotransposition with Y318F here. Y318F also restored the balance between the melting and annealing activities of ORF1p that was lost in Y318A. Because annealing catalysis was not significantly different between the alanine and phenylalanine variants at residue 318, the biochemical problem with Y318A appears to be its excessive basepair melting catalysis. This residue is likely involved in hydrophobic or steric interactions because the terminal hydroxyl group on tyrosine is dispensable.

Based upon the results of earlier studies of wt and mutant ORF1 proteins, we proposed that at least some of the interactions involved in high-affinity binding of ORF1p to RNA are distinct from those involved in the nucleic acid chaperone activity (16,17), i.e. that these reactions involve distinct sites on the protein. The present results provide further support for this hypothesis. If the melting activity in the strand displacement assay is functionally equivalent to the RNA binding affinity, then the paired variants at residues 284 and 318 should have been indistinguishable in the strand displacement assay as they were in the RNA binding assay, but they were not.

A 3D structure of the intact trimer with nucleic acids will be necessary to fully understand the relationship between the structure of ORF1p and its RNA binding and nucleic acid chaperone activities. Nevertheless, it seems unlikely that all of the residues that affect the chaperone activity of ORF1p are closely juxtaposed in the tertiary structure of the protein, because they are so widely distributed along its primary structure. Instead we predict that residues responsible for chaperone activity are distributed in the folded structure as well as the primary sequence of this protein.

When taken together and compared with retrotransposition experiments, the in vitro single molecule and solution experiments demonstrate the importance of the finely balanced nucleic acid chaperone capabilities of ORF1p for biological activity. They also show that the nucleic acid chaperone activity of ORF1p requires optimal interactions with ssDNA and dsDNA that result in a reduction in melting cooperativity, as well as an optimal enhancement of strand annealing and strand displacement capabilities. This enhancement of annealing and displacement is likely a result of the combination of the ability of ORF1p to facilitate strand attraction as well as nucleic acid unwinding. Specific single amino acid substitutions in ORF1p inhibit this chaperone activity by interfering with at least one of these capabilities, as demonstrated by the properties of the mutations at R284 and Y318. A mutation that strongly enhanced both melting and annealing (Y318A) inhibited retrotransposition, but a compensatory mutation that reduced the melting and annealing capabilities of ORF1p and restored these properties to wt levels also restored retrotransposition. Similarly, a mutation that enhanced annealing but significantly reduced the ability of ORF1p to facilitate melting also inhibited retrotransposition. A compensatory mutation that restored the kinetics to wt values also restored retrotransposition activity (Figure 7).

Other groups have published kinetic data on DNA or RNA annealing (21,35,37–39), DNA melting as an intermediate step to annealing (21), steady-state duplex destabilization, and chaperone effects on force-induced DNA melting (29,40,41). Although it has been suggested that the melting and annealing activities of nucleic acid chaperones are in a tightly constrained balance (42) this is, to our knowledge, the first work to quantitatively determine an optimum kinetics of strand displacement and annealing that can be destroyed and restored with specific point mutations.

Finally, it is noteworthy that all of the ORF1p mutations that retained wt RNA affinity were nevertheless nucleic acid chaperone proteins—all ORF1p variants tested accelerated both base pair annealing and melting, and presumably allow for the rearrangement of secondary structure as a consequence, regardless of their retrotransposition competence. The property that discriminates between biologically active and inactive variants of ORF1p, however, is the balanced ratio of the energetic barriers to annealing and melting. This finding implies that L1 retrotransposition depends upon an ORF1p that provides the two optimized energetic reductions in specific kinetic properties, consistent with observations that dispersed mutations in ORF1p could disrupt this finely tuned balance and inactivate retrotransposition. These quantities must have a basis in the specific nucleic acid rearrangements that are required for L1 retrotransposition. It may be a general feature of nucleic acid chaperone proteins to maintain specifically tuned energetic reductions to annealing and melting that satisfy their biological function, as determined here for L1 ORF1p.


SUPPLEMENTARY DATA

Supplementary Data are available at NAR Online.


FUNDING

National Institutes of Health Grants (R01GM40367 to S.L.M.); (P30 CA046934 for the Protein Production Core of the Univ Colorado Cancer Center); (R01GM75965 to M.C.W.); National Science Foundation (MCB-0744456 to M.C.W.). Funding for open access charge: National Institutes of Health Grant (R01GM40367 to S.L.M.).

Conflict of interest statement. None declared.



ACKNOWLEDGEMENTS

We thank D. Branciforte, D. Bushman and J. Cummiskey for constructs, and Dr I. Rouzina, Dr D. Bain and Dr K. Maluff for valuable discussion.


REFERENCES
1. Cordaux R,Batzer MA. The impact of retrotransposons on human genome evolutionNat. Rev. Genet.Year: 20091069170319763152
2. Han JS,Boeke JD. LINE-1 retrotransposons: modulators of quantity and quality of mammalian gene expression?BioessaysYear: 20052777578416015595
3. Goodier JL,Ostertag EM,Du K,Kazazian HHJ. A novel active L1 retrotransposon subfamily in the mouseGenome Res.Year: 2001111677168511591644
4. Moran JV,Holmes SE,Naas TP,DeBerardinis RJ,Boeke JD,Kazazian HH Jr. High frequency retrotransposition in cultured mammalian cellsCellYear: 1996879179278945518
5. Wei W,Gilbert N,Ooi SL,Lawler JF,Ostertag EM,Kazazian HH,Boeke JD,Moran JV. Human L1 retrotransposition: cis preference versus trans complementationMol. Cell. Biol.Year: 2001211429143911158327
6. Luan DD,Korman MH,Jakubczak JL,Eickbush TH. Reverse transcription of R2Bm RNA is primed by a nick at the chomosomal target site: a mechanism for non-LTR retrotranspositionCellYear: 1993725956057679954
7. Martin SL,Branciforte D,Keller D,Bain DL. Trimeric structure for an essential protein in L1 retrotranspositionProc. Natl Acad. Sci. USAYear: 2003100138151382014615577
8. Khazina E,Weichenrieder O. Non-LTR retrotransposons encode noncanonical RRM domains in their first open reading frameProc. Natl Acad. Sci. USAYear: 200910673173619139409
9. Januszyk K,Li PW,Villareal V,Branciforte D,Wu H,Xie Y,Feigon J,Loo JA,Martin SL,Clubb RT. Identification and solution structure of a highly conserved C-terminal domain within ORF1p required for retrotransposition of long interspersed nuclear element-1J. Biol. Chem.Year: 2007282248932490417569664
10. Kolosha VO,Martin SL. High affinity, non-sequence-specific RNA binding by the open reading frame 1 (ORF1) protein from long interspersed nuclear element 1 (LINE-1)J. Biol. Chem.Year: 20032788112811712506113
11. Basame S,Li PW-L,Howard G,Branciforte D,Keller D,Martin SL. Spatial assembly and RNA binding stoichiometry of a protein essential for LINE-1 retrotranspositionJ. Mol. Biol.Year: 200635735135716434051
12. Hohjoh H,Singer MF. Cytoplasmic ribonucleoprotein complexes containing human LINE-1 protein and RNAEMBO J.Year: 1996156306398599946
13. Kolosha VO,Martin SL. In vitro properties of the first ORF protein from mouse LINE-1 support its role in ribonucleoprotein particle formation during retrotranspositionProc. Natl Acad. Sci. USAYear: 19979410155101609294179
14. Martin SL. Ribonucleoprotein particles with LINE-1 RNA in mouse embryonal carcinoma cellsMol. Cell. Biol.Year: 199111480448071715025
15. Martin SL,Li J,Weisz JA. Deletion analysis defines distinct functional domains for protein-protein and nucleic acid interactions in the ORF1 protein of mouse LINE-1J. Mol. Biol.Year: 2000304112011071806
16. Martin SL,Cruceanu M,Branciforte D,Wai-Lun Li P,Kwok SC,Hodges RS,Williams MC. LINE-1 retrotransposition requires the nucleic acid chaperone activity of the ORF1 proteinJ. Mol. Biol.Year: 200534854956115826653
17. Martin SL,Bushman D,Wang F,Li PW-L,Walker A,Cummiskey J,Branciforte D,Williams MC. A single amino acid substitution in ORF1 dramatically decreases L1 retrotransposition and provides insight into nucleic acid chaperone activityNucleic Acids Res.Year: 2008185845585418790804
18. Martin SL,Bushman FD. Nucleic acid chaperone activity of the ORF1 protein from the mouse LINE-1 retrotransposonMol. Cell. Biol.Year: 20012146747511134335
19. Kulpa DA,Moran JV. Ribonucleoprotein particle formation is necessary but not sufficient for LINE-1 retrotranspositionHum. Mol. Gen.Year: 2005143237324816183655
20. Liu HW,Cosa G,Landes CF,Zeng Y,Kovaleski BJ,Mullen DG,Barany G,Musier-Forsyth K,Barbara PF. Single-molecule FRET studies of important intermediates in the nucleocapsid-protein-chaperoned minus-strand transfer step in HIV-1 reverse transcriptionBiophys. J.Year: 2005893470347916100256
21. Stewart-Maynard KM,Cruceanu M,Wang F,Vo MN,Gorelick RJ,Williams MC,Rouzina I,Musier-Forsyth K. Retroviral nucleocapsid proteins display nonequivalent levels of nucleic acid chaperone activityJ. Virol.Year: 200882101291014218684831
22. Zhang A,Derbyshire V,Salvo JL,Belfort M. Escherichia coli protein StpA stimulates self-splicing by promoting RNA assembly in vitroRNAYear: 199517837937493324
23. Chakshusmathi G,Kim SD,Rubinson DA,Wolin SL. A La protein requirement for efficient pre-tRNA foldingEMBO J.Year: 2003226562657214657028
24. Chaurasiya KR,Paramanathan T,McCauley MJ,Williams MC. Biophysical characterization of DNA binding from single molecule force measurementsPhys. Life Rev.Year: 2010729934120576476
25. Wu H,Rouzina I,Williams MC. Single-molecule stretching studies of RNA chaperonesRNA Biol.Year: 201077384
26. McCauley MJ,Williams MC. Mechanisms of DNA binding determined in optical tweezers experimentsBiopolymersYear: 20078515416817080421
27. Williams MC,Rouzina I,Bloomfield VA. Thermodynamics of DNA interactions from single molecule stretching experimentsAcc. Chem. Res.Year: 20023515916611900519
28. Williams MC,Rouzina I,Wenner JR,Gorelick RJ,Musier-Forsyth K,Bloomfield VA. Mechanism for nucleic acid chaperone activity of HIV-1 nucleocapsid protein revealed by single molecule stretchingProc. Natl Acad. Sci. USAYear: 2001986121612611344257
29. Cruceanu M,Urbaneja MA,Hixson CV,Johnson DG,Datta SA,Fivash MJ,Stephen AG,Fisher RJ,Gorelick RJ,Casas-Finet JR,et al. Nucleic acid binding and chaperone properties of HIV-1 Gag and nucleocapsid proteinsNucleic Acids Res.Year: 20063459360516449201
30. Spector S,Wang M,Carp SA,Robblee J,Hendsch ZS,Fairman R,Tidor B,Raleigh DP. Rational modification of protein stability by the mutation of charged surface residuesBiochemistryYear: 20003987287910653630
31. Goldberg AL. Protein degradation and protection against misfolded or damaged proteinsNatureYear: 200342689589914685250
32. Williams MC,Rouzina I. Force spectroscopy of single DNA and RNA moleculesCurr. Opin. Struct. Biol.Year: 20021233033612127451
33. Shokri L,McCauley MJ,Rouzina I,Williams MC. DNA overstretching in the presence of glyoxal: structural evidence of force-induced DNA meltingBiophys. J.Year: 2008951248125518424499
34. Williams MC,Gorelick RJ,Musier-Forsyth K. Specific zinc-finger architecture required for HIV-1 nucleocapsid protein's nucleic acid chaperone functionProc. Natl Acad. Sci. USAYear: 2002998614861912084921
35. Qualley DF,Stewart-Maynard KM,Wang F,Mitra M,Gorelick RJ,Rouzina I,Williams MC,Musier-Forsyth K. C-terminal domain modulates the nucleic acid chaperone activity of human T-cell leukemia virus type 1 nucleocapsid protein via an electrostatic mechanismJ. Biol. Chem.Year: 201028529530719887455
36. Shokri L,Rouzina I,Williams MC. Interaction of bacteriophage T4 and T7 single-stranded DNA-binding proteins with DNAPhys. Biol.Year: 2009602500219571366
37. Vo MN,Barany G,Rouzina I,Musier-Forsyth K. Mechanistic studies of mini-TAR RNA/DNA annealing in the absence and presence of HIV-1 nucleocapsid proteinJ. Mol. Biol.Year: 200636324426116962137
38. Sharma K,Didier P,Darlix JL,de Rocquigny H,Bensikaddour H,Lavergne JP,Penin F,Lessinger JM,Mely Y. Kinetic analysis of the nucleic acid chaperone activity of the hepatitis C virus core proteinNucleic Acids Res.Year: 2010383632364220167640
39. Urbaneja MA,Wu M,Casas-Finet JR,Karpel RL. HIV-1 nucleocapsid protein as a nucleic acid chaperone: spectroscopic study of its helix-destabilizing properties, structural binding specificity, and annealing activityJ. Mol. Biol.Year: 200231874976412054820
40. Cruceanu M,Gorelick RJ,Musier-Forsyth K,Rouzina I,Williams MC. Rapid kinetics of protein-nucleic acid interaction is a major component of HIV-1 nucleocapsid protein's nucleic acid chaperone functionJ. Mol. Biol.Year: 200636386787716997322
41. McCauley MJ,Williams MC. Optical tweezers experiments resolve distinct modes of DNA-protein bindingBiopolymersYear: 20099126528219173290
42. Levin JG,Guo J,Rouzina I,Musier-Forsyth K. Nucleic acid chaperone activity of HIV-1 nucleocapsid protein: critical role in reverse transcription and molecular mechanismProg. Nucleic Acid Res. Mol. Biol.Year: 20058021728616164976

Figures

[Figure ID: F1]
Figure 1. 

Mutations to conserved ORF1p residues in the murine ORF1p sequence perturb retrotransposition. (A) Schematic of ORF1p with single amino acid substitutions studied here (bold) or shown previously to affect nucleic acid chaperone activity (16,17). (B) Average percent (±SD, n = 3) eGFP-expressing cells 6 days after transfection with antisense intron-containing eGFP reporter in wt or ORF1p mutant L1.



[Figure ID: F2]
Figure 2. 

Effect of ORF1p mutations on steady-state levels of L1 RNA and ORF1p in cells transfected with retrotransposition assay vectors. (A) L1 RNA normalized to 18 S rRNA measured by qRT-PCR on Days 1–5 following transfection of 143 B cells, average (±SD, n = 3). (B) Western blots of ORF1p. Top panel, triplicate wt extracts Days 1–4 post-transfection as indicated; ORF1p is not detectable on Days 5 and 6 (data not shown). Right arrow points to ORF1p, left arrow to signal obtained from 2.5 ng his-tagged recombinant ORF1p (rORF1p); (−) untransfected control cell extract. Middle panel shows hnRNP-Q detected on the same blot. Below are triplicate lanes containing protein extracts on Day 2 post-transfection with the indicated ORF1 mutants.



[Figure ID: F3]
Figure 3. 

Binding of ORF1p variants to RNA. Nitrocellulose filter binding assay with 25–50 pM 111 nt RNA and increasing amounts of wt and mutant ORF1p. Points are averages of (±SD, n = 3). Lines are fits as described previously (10). Most KDapp values fell in the low nanomolar range (5.7–28 nM), with the exception of R238K and R238A (130 and 230 nM, respectively).



[Figure ID: F4]
Figure 4. 

Single-molecule stretching. Typical force-extension (solid) and relaxation (dashed) curves of DNA only (black) and in the presence of the indicated ORF1p variants; blue lines are used for retrotransposition active variants and red lines for retrotransposition inactive variants. (A) wt, ; red lines illustrate the calculation of the transition width, δF. (B) R284A, ; and R284K, . (C) Y318A, and Y318F, . (D) Change in the transition width of DNA force-induced melting in the presence of wt and variant ORF1 proteins, where we have also subtracted the value of the transition width in the absence of protein ( ), with pN (29). Error bars are standard error for at least three measurements. Significant differences among the five ORF1p variants were found by ANOVA (P < 0.001); the active and inactive pairs at each site are different, Y318F ≠ Y318A (P = 0.036), R284K ≠ R284A (P = 0.017), but the active R284K is not distinguishable from the inactive Y318A (P = 0.658).



[Figure ID: F5]
Figure 5. 

Temperature dependence of ORF1p-mediated annealing catalysis. (A) Phosphorimage of complementary oligonucleotides, one labeled with 32P, annealing in the presence (inset, top), or absence (inset, bottom), of wt ORF1p, respectively. The fraction of duplex DNA formed in the presence of wt ORF1p was plotted against time for each temperature and fit to an inverse exponential decay, Equation (2). (B) Natural logarithms for the reaction rates were plotted against the inverse temperature, and points were fitted to a line for presence (circles) and absence (squares) of wt ORF1p.



[Figure ID: F6]
Figure 6. 

Temperature dependence of ORF1p-mediated strand displacement catalysis. (A) Phosphorimages of a 32P labeled oligonucleotide melting from duplex in the presence of wt ORF1p (top inset) or without protein (bottom inset) at 37°C for the indicated times. Graph plots duplex melted at the four temperatures indicated in the presence of wt ORF1p, lines show fit to a two-factor exponential decay, Equation (5). (B) Plot of temperature dependence of overall reaction rate. The natural logarithm of the found overall rate plotted against the inverse temperature, with fitted line for presence (circles) and absence (squares) of wt ORF1p.



[Figure ID: F7]
Figure 7. 

Retrotransposition requires a precise ratio between the reductions in the energetic barriers to strand displacement and annealing. Average energetic reductions for displacement and annealing (values from Table 1, multiplied by −R (gas constant) and normalized to oligonucleotide length); ovals represent standard deviations (thin arcs) and standard errors (thick arcs) in the x and y dimensions. The three retrotransposition-competent ORF1p variants (bold, circles) group closely together, with a generally observed upward trend given by the regression line, slope = 2.57, R = 0.29, distinguishing them from the inactivating mutants (squares) by at least one standard error.



Tables
[TableWrap ID: T1] Table 1. 

Slopes and y-intercepts of lines fitted to points plotted according to transition state theory


ORF1p Annealing (K·ln s−1) Annealing 1/T = 0 Exchange (K·ln s−1) Exchange 1/T = 0
Wt −4060 ± 340 (200) 8 ± 1 (1) −16 800 ± 900 (500) 48 ± 3 (2)
R284A −3070 ± 740 (430) 5 ± 2 (1) −20 300 ± 2200 (1300) 60 ± 7 (4)
R284K −3670 ± 210 (120) 7 ± 1 (1) −17 800 ± 3000 (1700) 52 ± 10 (6)
Y318A −3180 ± 500 (290) 6 ± 2 (1) −8200 ± 3900 (2300) 20 ± 13 (7)
Y318F −3540 ± 300 (170) 7 ± 1 (1) −14 800 ± 3700 (1800) 42 ± 12 (6)
No protein −5970 ± 350 (200) 11 ± 1 (1) −42 700 ± 4100 (2900) 125 ± 13 (9)

TF1Slopes of lines fitted to points as graphed from Equation (3) express the temperature dependence of the reaction rate; more negative values denote a more temperature-dependent reaction and thus a higher enthalpic barrier to product formation. Intercepts are used according to Equation (3) to determine the relative magnitude of entropic contributions to energetic barriers. Values are expressed as ±standard deviation and (standard error). The points comprising these lines were derived for each mutant from experimental data collected as in Figures 5 and 6.



Article Categories:
  • Nucleic Acid Enzymes


Previous Document:  Simplified quantification of myocardial flow reserve with flurpiridaz F 18: validation with microsph...
Next Document:  Cleavage of a model DNA replication fork by a methyl-specific endonuclease.