endonucleolytic cleavage of rna at 5′ endogenous stem structures by human flap endonuclease 1

Endonucleolytic Cleavage of RNA at 59 EndogenousStem Structures by Human Flap Endonuclease 1

Audrey Stevens1

Life Sciences Division, Oak Ridge National Laboratory, Oak Ridge, Tennessee 37831-8080

Received September 14, 1998

Structure-specific nucleases called 5* flap endo-nucleases cleave unannealed 5* arms of template-primer DNA model substrates at the start of theduplex and are involved in Okazaki fragment pro-cessing during DNA synthesis. To determine the pos-sible use of the enzymes in RNA structure analysis,the cleavage of synthetic and native RNAs was ex-amined using flap endonuclease 1 (Fen1) of HeLacells. RNAs are cleaved at about 20% of the rate ofDNA model substrates, and most of the cleavage sitesare within 200 nucleotides of the 5* end. Hydrolysisof MFA2 mRNA of yeast shows that the cleavages areat the start of five possible stem structures of afolded secondary structure predicted on the basis ofboth chemical and enzymatic structure probing. 16Sribosomal RNA of Escherichia coli is cleaved at sev-eral 5* stem structures of its phylogenetically pre-dicted folded structure. This type of RNA cleavagespecificity may be very useful in secondary structureanalysis in the future and also may be used by cellsfor specific 5* end-geared RNA cleavages. © 1998

Academic Press

Mammalian 59 flap endonucleases were first identi-fied and purified as enzymes required for the synthesisof mature DNA replication products in eukaryotic cell-free extracts. The studies showed that a 59 exonucleaseis functional in lagging-strand DNA synthesis andlikely involved in the removal of Okazaki fragments(1–6). In prokaryotic extracts, the same reaction iscatalyzed by the 59 nuclease domain of DNA poly-merases and there is low level homology between theeukaryotic enzymes and the 59 nuclease domain ofprokaryotic DNA polymerases (7). It has been reportedthat the genes encoding the 59 flap endonucleases fromdifferent eukaryotic sources are quite homologous (8–10). As investigated by genetic analysis in yeast, theenzyme is involved in Okazaki fragment processingand possibly in DNA repair (8, 11–15). DNA model

substrates have been used for enzyme characterization(7, 16, 17). The model substrates are composed of oli-gonucleotides that form a DNA duplex with an unan-nealed displaced 59 arm. It has been shown that theflap endonucleases migrate (migration is blocked by anoligonucleotide annealed at different sites on the arm)from the 59 end of the arm for a structure-dependentcleavage at the start of the duplex (18, 19).

Lyamichev et al. (20), using prokaryotic DNA poly-merases, showed that RNA model substrates arecleaved in a manner similar to the DNA model sub-strates. Murante et al. (21) also showed that calf Fen1hydrolyzed a flap RNA strand of model Okazaki frag-ments. Both have suggested that site-specific RNAcleavages could be catalyzed by the enzyme followingannealing of an appropriate oligonucleotide to theRNA. Recently, cleavase I, an engineered protein con-sisting of the 59 nuclease domain of Taq DNA polymer-ase, has been used for hydrolyzing ssDNAs to analyzefeatures of sequence divergence and mutations (22).This use is possible since the structural features of thefolded ssDNAs may be different, resulting in alteredcleavage sites. Interest in the secondary structure ofRNA led to our testing the ability of human Fen1 tocleave synthetic and native RNAs of different lengthsat endogenous stem structures, as reported here.

EXPERIMENTAL PROCEDURES

Purification of HeLa Fen1. Washed, frozen HeLa cells (from 6 l of0.6 3 106 cells/ml) were obtained from the Cell Culture Center,Cellex Biosciences Inc. Fen1 was purified from the cells using stepssimilar to those described by Harrington and Lieber (16). The initialsteps to obtain a nuclear lysate were as described by them. Thenuclear extract (about 80 mg of protein) was dialyzed against one l of10% glycerol, 20 mM Tris–HCl buffer, pH 7.6, 0.2 mM dithiothreitol,and 0.2 mg/ml of antipain and leupeptin (Sigma) (Buffer A) contain-ing 100 mM ammonium sulfate. The dialysate was applied (flow rate,1 ml/min) to a TosoHaas TSKgel Heparin-5 PW column (8,7.5) andthe column was eluted with 40 ml of a linear gradient of 80-400 mMammonium sulfate in buffer A. One ml fractions were collected andassayed. The peak heparin agarose fractions were dialyzed against10% glycerol, 25 mM sodium phosphate buffer, pH 7.2, 0.2 mMdithiothreitol, and 0.2 mg/ml of antipain and leupeptin (buffer B)containing 20 mM KCl. The dialysis was for 5 h with one change of1 Fax: (423) 574-1274. E-mail: [email protected].

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS 251, 501–508 (1998)ARTICLE NO. RC989499

501 0006-291X/98 $25.00Copyright © 1998 by Academic PressAll rights of reproduction in any form reserved.

buffer at 3 h. The dialysate was applied to a Pharmacia Mono S HRcolumn (5/5) (flow rate, l ml/min) and the column was eluted with 30ml of a linear gradient of no to 1 M KCl in buffer B. One ml fractionswere collected and assayed. The peak Mono S column fractions werecombined and an equal volume of buffer containing 20 mM Tris–HCl,pH 7.6, 10 mM potassium phosphate, pH 7.6, 10% glycerol (buffer C),with supplements as in buffer A was added. The diluted fraction was

applied (flow rate, 1 ml/min) to a TosoHaas TSKgel hydroxylapatitecolumn (HA-1000, 8/7.5). The column was eluted with a 30 ml lineargradient of 10 to 400 mM potassium phosphate in buffer C. Peakfractions were used in all the analysis. The enzyme was assayedusing [32P]oligo(A)12-18 z poly(dT) as an exonuclease substrate asdescribed by Ishimi et al. (1). The specific activity of the purified Fen1 was the same as reported by Harrington and Lieber (16) for mouse

FIG. 1. Hydrolysis of RNAs by Fen 1. (A) m7G[32P]pppTAT RNA, 0.6 pmol, was incubated with no (lane 1) and 0.15 unit (lane 2) of Fen1for 30 min at 37°C as described under Experimental Procedures. The reaction mixtures were extracted with phenol and aliquots wereexamined on an 8% polyacrylamide sequencing gel. (B) A 59-labeled luciferase RNA was analyzed as in A. In A and B, the numbers on theright are the size and position of the 100- to 400-nt RNA markers of Ambion. (C) Human b-globin mRNA, 1 pmol, was incubated with no (lane1) and 0.2 unit of Fen1 (lane 2) for 30 min at 43°C. The reaction mixtures were extracted with phenol and 0.08 pmol of RNA from each mixturewas used for primer extension analysis as described under Experimental Procedures using an oligonucleotide complementary to nt 111–130.Lanes 3 and 4 are marker lanes using ddATP and ddCTP, respectively. Fen1 cleavage sites are labeled on the left and marker bands on theright.

Vol. 251, No. 2, 1998 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

502

Fen 1 using this substrate (180,000 units per milligram). The puri-fied protein migated as a single band of 46 kDa when examined bypolyacrylamide gel electrophoresis. Analysis of the band by massspectrometry (Beckman Research Institute, Duarte, California)showed that it is Fen1 since four similar peptides were identified.

RNAs and model substrates. TAT RNA was transcribed using theplasmid described previously (23). MFA2 RNA of yeast was tran-scribed with an SP6 plasmid constructed for analysis of the second-ary structure of the RNA (Doktycz, M. J., Larimer, F. W., Pasternak,M., and Stevens, A., manuscript submitted for publication). A humanb-globin gene in a transcription cassette (24) was the kind gift ofJames Malter (Univ. of Wisconsin, Madison). The TAT plasmid wascut with PstI, the MFA2 plasmid with ApaI, and the b-globin plasmidwith HindIII. The DNAs were then trimmed with mung bean nucle-ase (BRL) to trim off the overhangs. RNAs were transcribed using

the Ribo-Max high-level RNA production system for SP6 polymerase(Promega). SP6 control DNA, encoding a luciferase RNA, from theRibo-Max kit was also transcribed. Ribosomal RNA from Escherichiacoli was obtained from Sigma.

The 59 ends of TAT RNA and the luciferase RNA were labeled with[a-32P]GTP (DuPont) using guanylyl transferase (BRL) according tothe procedure of Gehrke (25). To obtain an m7G cap structure, 1 mMadenosylmethionine was added to the reaction mixtures. A TAT RNAwith a 59 triphosphate end was prepared using [g-32P]GTP (ICN) inthe transcription reactions. TAT RNA with a monophosphate endgroup was prepared by dephosphorylation of the transcribed RNAfollowed by phosphorylation with [g-32P]ATP (ICN) and T4 polynu-cleotide kinase (Pharmacia Biotech).

The DNA model substrate, Flap Substrate 1, was prepared asdescribed by Harrington and Lieber (16). Oligonucleotides were ob-tained from DNA Integrated Technologies. [32P]oligo(A)12-18 z poly(dT)was prepared as described by Ishimi et al. (1) using oligo(A)12-18(prelabeled with [g-32P]ATP and polynucleotide kinase) and poly(dT)obtained from Pharmacia Biotech.

Reaction mixtures for assay of Fen1. The reactions mixtures (50ml) contained 2.5 mM Tris–HCl buffer, pH 7.6, 1 mM MgCl2, 0.5 mMdithiothreitol, 10 mg of acetylated albumin (BRL), and the amount ofsubstrate and enzyme described in the Figure legends. Incubationtimes were as described.

DMS and CMCT modifications. The chemical modifications of Aand C nucleotides with dimethyl sulfate (DMS, Aldrich ChemicalCo.) and of G and U nucleotides with 1-cyclohexyl-3-(2-morpholinoethyl) carbodiimide metho-p-toluene sulfonate (CMCT, AldrichChemical Co.) were carried out as described (26, 27) and analyzed byprimer extension.

Primer extension analysis and sequencing gels. Approximately 62ng of each of the oligonucleotide primers were pre-labeled at the 59end with [g-32P]ATP and polynucleotide kinase. Each primer (1-2 ng)was combined with 0.08–0.4 pmol of RNA and hybridization andreverse transcription were carried out as described (28) using murineleukemia virus reverse transcriptase (BRL). Marker lanes were gen-erated by adding ddATP and ddCTP to reaction mixtures of controlRNA. Following ethanol precipitation, the samples were dissolved inloading buffer and analyzed on 8% polyacrylamide sequencing gelscontaining 7.5 M urea. Products derived from 59 end labeled RNAsand the DNA model substrate were also examined on polyacrylamidesequencing gels.

RESULTS AND DISCUSSION

Hydrolysis of RNAs by Fen1. Figure 1A shows theproducts of Fen1 hydrolysis of a synthetic 59 cap-labeled TAT RNA (540 nt) as examined on an 8%polyacrylamide gel. Arrows on the right point to themajor Fenl cleavage sites (shown in lane 2). A predom-inant oligonucleotide product having a size of 8 nt isformed. An oligonucleotide of 9 nt is a lesser product.(Sizing was done with a 20% sequencing gel). Otherweaker cleavage sites are found and products of about70 and 140 nt are labeled. With 30-50% substrate hy-drolysis, no hydrolysis products of greater than 140 ntare detected as shown in Fig. 1A and as examined bothby use of 6% gels and by hydrolysis of 39 poly(A)-labeledTAT RNA. Figure 1B shows the hydrolysis of a lucif-erase RNA of about 1800 nt. Predominant products inthis case are oligonucleotides of length 13 and 15 nt.Another strong cleavage yields an 80 nt product.Again, the results show the major cleavages are within

FIG. 2. Comparison of the cleavage efficiency of a DNA modelsubstrate and TAT RNA. (A) The [32P]DNA model substrate, 0.2pmol, was incubated with no (lane 1), 0.025 unit (lane 2) and 0.05unit (lane 3) of Fen1 for 20 min at 37°C as described under Experi-mental Procedures. (B) TAT RNA, 0.6 pmol with a G[32P]ppp label atthe 59 end, was incubated with no (lane 1), 0.075 unit (lane 2) and0.15 unit (lane 3) of Fen1 for 20 min at 37°C as just described. BothA and B reaction mixtures were extracted with phenol and aliquotswere taken for analysis on a 12% polyacrylamide sequencing gel. Thenumbers on the right depict the position and size of oligonucleotidemarkers.


503

200 nt of the 59 end. Figure 1C shows a primer exten-sion analysis of the Fen1 cleavage products of synthetichuman b-globin mRNA in the sequence from nt 1-110.The results show that cleavages occur at approxi-mately nt 15, 42, and 50 in the 59 untranslated region.(The AUG initiation codon starts at nt 51.) Codingsequence cleavages are predominantly at nt 62, 79, 82,and 83.

With the use of 59 end-labeled TAT RNAs with dif-ferent 59 terminal structures, i.e., a triphosphate ter-minus, a monophosphate terminus, a capped terminuswithout a methyl group, and the m7G-capped terminusshown in Fig. 1A, it was found that Fen1 cleaves allthese RNAs at approximately the same rate. Muranteet al. (21) have described similar results with 59 termi-nal structures of RNA in model Okazaki fragments.The reaction conditions that are optimal for RNA hy-drolysis have been examined using TAT RNA, and themonovalent cation requirement is similar to that de-

scribed by Harrington and Lieber (16) for model DNAsubstrates using mouse Fen1. The divalent metal re-quirement differs. The optimal Mg12 concentration is 1mM, and the reaction is inhibited 67% by 5 mM and87% by 10 mM Mg12.

Comparison of endonuclease activity with a modelDNA substrate and RNA. A model DNA substratewas compared with TAT RNA for cleavage efficiency.Figure 2A shows the hydrolysis of the model substrate,the same as used by Harrington and Lieber (16) intheir studies of mouse Fen1. The results show that it ishydrolyzed about 50% by 0.025 unit and 70% by 0.05unit of Fen1. Figure 2B shows the hydrolysis of theTAT RNA by 0.075 and 0.15 units of Fen1. TAT RNA ishydrolyzed about 30 and 60% by these amounts ofFen1. This comparison and analysis of the hydrolysis ofother RNAs as well as a second model DNA substratesuggest that the overall hydrolysis rate of the RNAs

FIG. 3. Analysis of the sites of cleavage of MFA2 mRNA by Fen1 compared to its folded secondary structure. In A is shown the foldedstructure of MFA2 mRNA as determined with the Wisconsin Package, Version 9.0, Genetic Computer Group, Madison, WI, using bothenzymatic and chemical probing data as constraints. B. MFA2 mRNA, 4.5 pmol, was incubated with no (lane 1) and 0.2 unit (lane 2) of Fen1for 30 min at 37°C. The reaction mixtures were extracted with phenol and about 0.4 pmol of RNA was used for primer extension as describedunder Experimental Procedures. Lanes 3–6 are DMS and CMCT modifications as described in the text. Overall band labeling (middle) isbased on the DMS and DMCT modification data. Lanes 7 and 8 are a repeat of the lanes 4 and 3 results with DMS, respectively, and lanes9 and 10 are marker lanes using ddATP and ddCTP, respectively (bands identified at the right). An oligonucleotide complementary to nt132–151 was used for the primer extension analysis.


504

used in these studies is about 20% of the rate of themodel DNA substrates previously used in other labo-ratories (16, 18).

Evidence for cleavage of RNA at 59 endogenous stemstructures. The folded secondary structure of MFA2mRNA has been investigated in this laboratory withthe use of synthetic RNAs made with T7 and SP6 RNApolymerases. The analyses have included both enzy-matic and chemical probing of the secondary structure(submitted for publication). A structural model thatbest fits the probing data for a synthetic SP6 transcriptis shown in Fig. 3A, as determined using the Mfoldprogram of Zuker (29). Figure 3B (lanes 1 and 2) showsthe analysis of Fen1 cleavage sites in the same MFA2mRNA. Primer extension analysis of 59 termini, aftercleavage, was done with the use of an oligonucleotidecomplementary to nt 132-151. The results in lane 2show that Fen1 cleavages occur at nt 45, 61, 78, 102,and 124 (arrows on the left). When the Fen1 cleavagesites are located in the structure shown in Fig. 3A

(arrows), one can see that they occur at the start ofstem structures in the sequence from 1 to 131 nt.Chemical modification data with DMS and CMCT areshown in Fig. 3B, lanes 3 (with DMS), 4 (no DMS), 5(with CMCT) and 6 (no CMCT). The DMS data showthe unannealed A and C residues and the CMCT data,the unannealed G and U residues in the sequence from1-130 nt. These results support the structure shown inFig. 3A, and the stem regions can be detected by ex-amining these modification data. Oligonucleotidescomplementary to nt 192–211 and nt 61–80 were alsoused in the primer extensions, and the same cleavageswere found. No cleavage past nt 124 was identified.The strongest stem structure is the one at nt 124-131and the nt 124 site is cleaved best. A weak stem at nt102 is cleaved very poorly. The results show that withMFA2 RNA, Fen1 cleavages occur at the start of en-dogenous stem structures in the RNA, suggesting amigration of the enzyme from the 59 end to these sites.

The hydrolysis of 16S rRNA of E. coli was also mea-

FIG. 3—Continued


505

sured and cleavage sites identified by primer extensionanalysis. Figure 4A shows the folded secondary struc-ture for the terminal 59 sequence of the 16S rRNAbased on phylogenetic comparisons (30). Figure 4Bshows an analysis of the Fen1 cleavage sites using aprimer complementary to nt 100–119. The resultsshow no significant cleavage of the RNA from nt 60 tont 95. Figure 4C shows the results using an oligonu-cleotide complementary to nt 61–80 to identify thesites of cleavage from nt 1 to nt 60. The results (lanes1 and 2) show cleavages occur at nt 9, 33, 40, and 52, as

designated by arrows on the left and also with arrowsin Fig. 4A. Again, the cleavages occur close to the startof stem structures and are found to be localized to the59 end of the RNA. Lanes 3–6 (see figure legend) showmarker bands.

The studies reported here show that human Fen1can hydrolyze RNA and evidence is presented thatcleavages occur at the start of stem structures in thefolded RNA secondary structures. The structures werepredicted by chemical and enzymatic probing of oneRNA and phylogenetic comparisons of a second RNA.

FIG. 4. Analysis of the sites of cleavage of 16S rRNA compared to its folded secondary structure. In A is shown the folded structure of16S rRNA as deduced by phylogenetic comparisons (24). In B and C, the RNA was hydrolyzed with Fen1 using the conditions described inthe legend to Fig. 3. In B, primer extension analysis was done using an oligonucleotide primer complementary to nt 100–119. The numberson the right identify bands from DMS modifications (not shown). In C, primer extension analysis was done using an oligonucleotide primercomplementary to nt 61-80. Lanes 1 and 2 show no and Fen 1 cleavages, lanes 3 and 4 show no and DMS modification data, and lanes 5 and6 are marker band identification reactions using ddATP and ddCTP, respectively. The Fen1 cleavages are identified by arrows on the left andband identity from DMS modification data as well as marker band results are shown on the right.


506

That all cleavages occur close to the 59 end suggests thesame mechanism of cleavage as with DNA model sub-strates, i.e., that the enzyme migrates from the 59 enduntil it encounters a hydrolyzable annealed structure.As shown, Fen1 cleaves TAT RNA and luciferasemRNA close to the 59 termini, and these cleavagesappear to be the dominant cleavages. Further cleav-ages are much more limited. This is what might beexpected if a strong stem is found close to the 59 end,since a melted stem or open RNA structure may berare, thus not allowing much migration of the enzymepast the site. Studies of the flap endonucleases withDNA model substrates suggest that the enzyme doesnot pass over strong stem structures and may nothydrolyze at the start of stem structures dependingupon specific structural and sequence requirementsnot clearly defined at this time (10, 11). It seems likelythat Fen1 may be useful in RNA structure analysesespecially in the identification of the location of stemstructures in the 59 terminal sequence. The studies

reported here show cleavages occur at several stemstructures in each RNA. More specific requirements forcleavage have not yet been analyzed.

Protein–RNA interactions in vivo may protect RNAfrom hydrolysis by Fen1. Murante et al. (10) showedthat if flap arms of DNA model substrates are biotin-ylated on the 59 terminal nucleotide, they are not hy-drolyzed upon addition of streptavidin. Lim et al. (31,32) have found that human b-globin mRNA containingnonsense codons may be cleaved to yield three short-ened RNAs in transgenic mice. The RNAs are short-ened at the 59 end. The authors suggested that struc-tural features may be important and that a structure-dependent endonuclease or the stalling of a 59exoribonuclease may be involved in the formation ofthe shortened RNAs. Fen1 is such a structure-dependent endonuclease. Of considerable interest atthis time is whether the alphaherpes virus host shutoffproteins may degrade RNAs in this manner. It hasbeen reported that these proteins have some homology

FIG. 4—Continued


507

to DNA repair proteins (33), and also that specificamino acid sequences are present that predict a 59exonuclease function (34).

ACKNOWLEDGMENTS

This work was supported by the Office of Health and Environmen-tal Research, U.S. Department of Energy under Contract DE-AC05-96OR22464 with the Lockheed Martin Energy Research Corpora-tion.

REFERENCES

1. Ishimi, Y., Claude, A., Bullock, P., and Hurwitz, J. (1988) J. Biol.Chem. 263, 19723–19733.

2. Kenny, M. K., Balogh, L. A., and Hurwitz, J. (1988) J. Biol.Chem. 263, 9801–9808.

3. Goulian, M., Richards, S. H., Heard, C. J., and Bigsby, B. M.(1990) J. Biol. Chem. 265, 18461–18471.

4. Turchi, J. J., and Bambara, R. A. (1993) J. Biol. Chem. 268,15136–15141.

5. Turchi, J. J., Huang, L., Murante, R. S., Kim, Y., and Bambara,R. A. (1994) Proc. Natl. Acad. Sci. USA 91, 9803–9807.

6. Waga, S., Bauer, G., and Stillman, B. (1994) J. Biol. Chem. 269,10923–10934.

7. Robins, P., Pappin, D. J. C., Wood, R. D., and Lindahl, T. (1994)J. Biol. Chem. 269, 28535–28538.

8. Murray, J. M., Tavassoli, M., Al-Harithy, R., Sheldrick, K. S.,Lehmann, A. R., Carr, A. M., and Watts, F. Z. (1994) Mol. Cell.Biol. 14, 4878–4888.

9. Harrington, J. J., and Lieber, M. R. (1994) Genes Dev. 8, 1344–1355.

10. Hiraoka, L. R., Harrington, J. J., Gerhard, D. S., Lieber, M. R.,and Hsieh, C.-L. (1995) Genomics 25, 220–225.

11. Sommers, C. H., Miller, E. J., Dujon, B., Prakash, S., andPrakash, L. (1995) J. Biol. Chem. 270, 4193–4196.

12. Johnson, R. E., Kovvali, G. K., Prakash, L., and Prakash, S.(1995) Science 269, 238–240.

13. Reagan, M. S., Pittenger, C., Siede, W., and Friedberg, E. C.(1995) J. Bacteriol. 177, 364–371.

14. Tishkoff, D. X., Filosi, N., Gaida, G. M., and Kolodner, R. D.(1997) Cell 88, 253–263.

15. Kim, K., Biade, S., and Matsumoto, Y. (1998) J. Biol. Chem. 273,8842–8848.

16. Harrringon, J. J., and Lieber, M. R. (1994) EMBO J. 13, 1235–1246.

17. Murante, R. S., Huang, L., Turchi, J. J., and Bambara, R. A.(1996) J. Biol. Chem. 269, 1191–1196.

18. Murante, R. S., Rust, L., and Bambara, R. A. (1995) J. Biol.Chem. 270, 30377–30383.

19. Wu, X., Li, J., Li, X., Hsieh, C., Burgers, P. M. J., and Lieber,M. R. (1996) Nucleic Acids Res. 24, 2036–2043.

20. Lyamichev, V., Brow, M. A. D., and Dahlberg, J. E. (1993) Sci-ence 260, 778–783.

21. Murante, R. S., Rumbaugh, J. A., Barnes., C. J., Norton, J. R.,and Bambara, R. A. (1996) J. Biol. Chem. 271, 25888–25897.

22. Brow, M. A., Oldenburg, M. C., Lyamichev, V., Heisler, L. M.,Lyamicheva, N., Hall, J. G., Eagen, N. J., Olive, D. M., Smith,L. M., Fors, L., and Dahlberg, J. E. (1996) J. Clin. Microbiol. 34,3129–3137.

23. Stevens, A. (1988) Mol. Cell. Biol. 8, 2005–2010.24. Rajagopalan, L. E., and Malter, J. S. (1996) J. Biol. Chem. 271,

19871–19876.25. Gehrke, L. (1986) Gene Anal. Technol. 3, 45–52.26. Thisted, T., Nielsen, A. K., and Gerdes, K. (1994) EMBO J. 13,

1950–1959.27. Ehresmann, C., Baudin, F., Mougel, M., Romby, P., Ebel, J. P.,

and Ehresmann, B. (1987) Nucleic Acids Res. 15, 9109–9128.28. Sambrook, J., Fritsch, E. F., and Maniatis, T. (1989) Molecular

Cloning: A Laboratory Manual, 2nd ed., Cold Spring HarborLaboratory Press, Cold Spring Harbor, NY.

29. Zuker, M. (1989) Science 244, 48–52.30. Woese, C. R., and Pace, N. R. (1993) in RNA Structure

(Gesteland, R. F., and Atkins, J. F., Eds.), pp. 91–117, ColdSpring Harbor Laboratory Press, Cold Spring Harbor, NY.

31. Lim, S.-K., Sigmund, C. D., Gross, K. W., and Maquat, L. E.(1992) Mol. Cell. Biol. 12, 1149–1161.

32. Lim, S.-K., and Maquat, L. E. (1992) EMBO J. 11, 3271–3278.33. Read, G. S. (1997) in mRNA Metabolism and Post-

Transcriptional Gene Regulation (Harford, J. B., and Morris,D. R., Eds.), pp. 311–320, Wiley–Liss, New York.

34. Doherty, A. J., Serpell, L. C., and Ponting, C. P. (1996) NucleicAcids Res. 24, 2488–2497.


508

endonucleolytic cleavage of rna at 5′ endogenous stem structures by human flap endonuclease 1

Documents