Amplifiable hybridization probes containing a molecular switch

Molecular and Cellular Probes (1997) 11, 187–194

Amplifiable hybridization probes containing a molecular switch Herman J. Blok∗ and Fred Russell Kramer Department of Molecular Genetics, Public Health Research Institute, 455 First Avenue, New York, NY 10016, USA (Received 1 November 1996, Accepted 11 February 1997) In order to reduce background signals in Qb replicase-mediated bioassays, a target-dependent probe amplification strategy has been proposed that utilizes recombinant RNA hybridization probes that contain an inserted molecular switch.1 A molecular switch is an internal region of the probe that undergoes a conformational change when the probe hybridizes to its target. We investigated whether non-hybridized probes (which cause background signals) could be selectively destroyed by incubating the probe-target hybrids with ribonuclease III, which should cleave the non-hybridized probes and leave the hybridized probes intact. Two problems with this assay design were observed. First, ribonuclease III cleaved probe-target hybrids non-specifically when the target was an RNA, thereby destroying all of the bound probes. And second, the expected conformational change in the molecular switch did not occur when the probes were bound to their targets, apparently because the hairpin stem formed by the molecular switch was too long. Although these results demonstrated that the original assay design could not work, they provided insights that have led to better designs for target-dependent amplification assays. In these assays, the probes will be DNA molecules containing short-stemmed molecular switches. Non-hybridized probes will be selectively destroyed by incubation with a restriction endonuclease.  1997 Academic Press Limited

KEYWORDS: Qb replicase, recombinant MDV-1 RNA, ribonuclease III recognition site.

INTRODUCTION Qb amplification assays2,3 utilize amplifiable hybridization probes,4,5 which consist of a probe sequence embedded at an appropriate site within the sequence of MDV-1 (+) RNA,6,7 which is a template for the RNA-directed RNA polymerase of bacteriophage Qb. These recombinant RNAs8 hybridize to their targets as do ordinary probes, and the resulting probe-target hybrids are separated from non-hybridized probes as in ordinary hybridization assays. However, unlike most amplification assays, including those that utilize the polymerase chain reaction9,10 (PCR), the ligase chain reaction,11,12 the nucleic acid sequence-based amplification reaction,13,14 the similar

self-sustained sequence replication reaction,15,16 and the strand displacement amplification reaction,17 in our assay, the probes, rather than the targets, are amplified exponentially in a brief isothermal incubation with Qb replicase. In principle, these assays should be extremely sensitive, since a single probe molecule can initiate exponential amplification, and more than a billion copies of each probe can be synthesized in a 30 min reaction.4 In practice, however, the limiting factor that determines the sensitivity of these assays is not how well the replicatable probes can be amplified, but rather, how well the probe-target hybrids can be

∗ Author to whom all correspondence should be addressed.

0890–8508/97/030187+08 $25.00/0/ll970103

 1997 Academic Press Limited

188

H. J. Blok and F. R. Kramer

Target Probe

Probe

Fig. 1. Schematic representation of an amplifiable RNA probe containing a molecular switch. In the absence of target (lower figure), complementary sequences (indicated by bold lines) hybridize to each other to form a double-stranded binding site for RNase III. The sites at which RNase III cleaves the probe are indicated by dark arrows. When the probe is bound to its target (upper figure), a conformational change occurs that results in the separation of the complementary sequences, eliminating the RNase III binding site. Therefore, when the probe-target hybrids are incubated with RNase III, non-hybridized probes are selectively destroyed.

separated from the excess of non-hybridized probes that are used to ensure that hybridization occurs rapidly. Separation of hybridized probes from nonhybridized probes is achieved by capturing the probetarget hybrids on a solid support (such as a microtitre plate or paramagnetic particles) and then washing the support to remove all other components of the hybridization mixture.18 However, even in assays that utilize advanced hybrid isolation techniques,19,20 the number of non-hybridized replicatable probes that persist (despite extensive washing) is not reduced below 10 000 molecules.2 Because non-hybridized probes are amplified as well as hybridized probes, a low-level background signal is generated that obscures the presence of rare targets.5 In order to reduce this background, a target-dependent amplification strategy has been proposed in which recombinant RNA hybridization probes are utilized that contain a molecular switch inserted in a region that does not interfere with the replication of the probes. Molecular switches are oligonucleotides that contain a probe sequence embedded within a hairpin stem, and which undergo an allosteric conformational change when the probe hybridizes to its target.1 Recombinant RNAs containing a molecular

switch that consists of a 44-nucleotide probe sequence surrounded by complementary 21-nucleotide sequences were synthesized (Fig. 1). These sequences pair with each other to form a hairpin stem that is a recognition site for Escherichia coli ribonuclease III (RNase III).21–23 When these recombinant molecules are cleaved by RNase III, the resulting RNA fragments cannot be amplified by Qb replicase. When the probe sequence binds to its target, the probe-target hybrid is longer and more stable than the hairpin stem formed by the sequences surrounding the probe. The rigidity of the probe-target hybrid24 should prevent the simultaneous existence of the probe-target helix and the hairpin stem surrounding the probe. Consequently, upon hybridization of the probe to its target, a conformational change should occur that causes the hairpin stem to unwind, leaving the sequences surrounding the probe in a single-stranded conformation. Since RNase III does not bind to single-stranded RNA, hybridized probes should be protected from cleavage. Incubation of the isolated probe-target hybrids with RNase III prior to Qb replicase-mediated amplification should result in the selective destruction of persistent nonhybridized probes, thus reducing or eliminating back-

Amplifiable hybridization probes

189

UU C C C U U G G A U A C C C A G G U CG G AA U U UA C G C U CG Probe G U A U U UU C G U RNase III U UA GU C C GC UA recognition G G UA CG G C C C UA C A C G G U U G G G A G G CU A C UU C U A G G U U U A U A G C AGUGA AUUUUGUGAUGUGGUG U GU C A CG G C U G A UUA CUUA A A A C A CUA CGUC A C GA CC G C C C CA GU GC C G G G A A UU U G U U G C UU A GCU G UAU G MDV-1 (+) G C A G A G C A C A CUC GG G G UG G A C GC A C C G G G G A A G G G C C CA C CC G G U GG A C G UG C C C C G U CG G G GU C C CU G A U C A A GC GA U C C C G G G U C C CG G C GC G C CG C G G GC A U UA G U CG GC C G GC CC GC

Fig. 2. Structure and sequence of probe MDV-int-B. The nucleotide sequence was folded into the secondary structures predicted to be most stable by a computer program.27 The HIV-1 probe sequence is embedded within a hairpin stem (indicated by bold letters) that forms a recognition site for RNase III.28 The sites at which RNase III cleaves the probe are indicated by arrows.

ground signals. This strategy was investigated, and the insights provided by the results were used to design better target-dependent amplification assays. MATERIALS AND METHODS Replicatable HIV-1 probes Three different recombinant MDV-1 (+) RNAs containing inserted probe sequences that are complementary to a conserved region of the integrase gene in human immunodeficiency virus type 1 (HIV1) mRNA were prepared in vitro by transcription from PCR products.25 The three replicatable probes were: MDV-int-A which contained an HIV-1 specific probe sequence embedded within the sequence of MDV-1 (+) RNA; MDV-int-B, which was identical to MDVint-A, except that it contained additional sequences surrounding the probe sequence that formed a molecular switch; and MDV-int-C, which was identical to MDV-int-B, except that it contained a different HIV-1 specific probe sequence and the sequences surrounding the molecular switch were modified. The 44-nucleotide-long HIV-1 specific probe sequences in MDV-int-A and MDV-int-B were complementary to nucleotides 4532–4575 and 4530–4573, respectively, in HIV-NL43.26 The probe sequence in MDV-int-B was embedded within complementary 21-nucleotide sequences that pair with each other to form a hairpin stem that is a recognition site for RNase III (Fig. 2).

MDV-int-C contained a 44-nucleotide HIV-1 specific probe sequence that was complementary to nucleotides 4934–4977 in HIV-NL43 and that was embedded within the same 21-basepair hairpin stem. This probe sequence was chosen so that the sequence of the probe-target duplex would be as different as possible from the canonical RNase III recognition site. In addition, seven nucleotides were deleted from each of the sequences flanking the molecular switch so that the stem surrounding the probe sequence would be shorter.

Simulated HIV-1 targets Four different target nucleic acids were prepared: a synthetic oligodeoxyribonucleic target (5′-CTTTTAAAATTAGCAGGAAGATGGCCAGTAAAAACAATACATACAGACAATGG-3′) that was prepared on a model 380A DNA synthesizer (Perkin Elmer); an oligoribonucleotide target that was prepared by transcription from a synthetic DNA fragment, obtained by annealing primer 5′-TAATACGACTCACTATAG-3′ to the HIV-1 oligodeoxyribonucleotide 5′-CCATTGTCTGTATGTATTGTTTTTACTGGCCATCTTCCTGCTAATTTTAAAAGCCTATAGTGAGTCGTATTA-3′, and then extending the hybrid by incubation with the Klenow fragment of E. coli DNA polymerase I; a single-stranded bacteriophage M13 DNA containing

190

H. J. Blok and F. R. Kramer

the complete HIV-1 integrase sequence; and a fulllength HIV-1 integrase RNA, corresponding to positions 4230–5091 in HIV-NL43, that was prepared by in vitro transcription.25

Hybridization and incubation conditions Hybrids were formed by incubating each probe with different targets. After hybridization in the presence and in the absence of RNase III, the size and quantity of the cleavage products was determined electrophoretically. Ten nanograms of the probes [32P]MDV-int-A and [32P]MDV-int-B, and 1·2 ng of [32P]MDV-int-C, were incubated for 1 h at 37°C in 10 ll of 20 m Tris–HCl (pH 8), 10 m MgCl2, and 150 m NH4Cl, with a 10-fold excess of each target for probes [32P]MDV-int-A and [32P]MDV-int-B, and a 1000-fold excess of either full-length integrase DNA or full-length integrase RNA for probe [32P]MDV-int-C. After hybridization, the reaction mixtures containing [32P]MDV-int-A or [32P]MDV-int-B were incubated for another 15 min at 37°C with 564 ng RNase III, or with a mock buffer (20 m Tris–HCl, pH 8), and the reaction mixtures containing [32P]MDV-int-C were incubated with 65 ng RNase III, or with the mock buffer. The reactions were stopped by adding 50 ll of an ice-cold solution containing 0·1% (w/v) bromophenol blue, 0·1% (w/v) xylene cyanol, and 10 m EDTA (pH 8) dissolved in deionized formamide. Twenty microlitre aliquots were analysed by electrophoresis through an 8% polyacrylamide gel containing 89 m Tris-borate (pH 8·3), 2 m EDTA, and 7  urea. The gels were dehydrated in a vacuum dryer, and the cleavage products were visualized by autoradiography.

targets, nor does it prevent the recombinant molecules from serving as templates for exponential amplification by Qb replicase.4 Both recombinant RNA probes were hybridized to three different complementary targets: an oligodeoxyribonucleotide (53 nucleotides), an oligoribonucleotide (55 nucleotides), and a full-length transcript of the HIV-1 integrase gene (879 nucleotides). Each hybrid was incubated with E. coli RNase III and the digestion products were analysed by polyacrylamide gel electrophoresis (Fig. 3). As expected, in the absence of target, RNase III did not cleave probe MDV-int-A, while probe MDV-intB was cleaved into three fragments at its RNase III recognition site. Probe MDV-int-B should not have been cleaved when it was hybridized to each of the three target molecules, because it should have undergone a conformational change upon hybridization that would have eliminated its RNase III recognition site. Surprisingly, however, probe MDVint-B was cleaved by RNase III in each of the hybrids. Even more surprising, probe MDV-int-A, which did not have an RNase III recognition site, was cleaved into two fragments in those hybrids in which the target was an RNA molecule. It was apparent that two types of unexpected cleavage had occurred. First, the molecular switch that had been incorporated into probe MDV-int-B had not undergone the conformational change upon hybridization that would have eliminated its RNase III recognition site. And second, the hybridization of either probe to an RNA target created an RNA:RNA probe-target duplex that served as an RNase III recognition site. A close examination of these results provided useful insights which will be discussed in turn.

Sequence specificity of ribonuclease III RESULTS AND DISCUSSION Incubation of probes and probe-target hybrids with ribonuclease III Two different recombinant RNA probes were synthesized by in vitro transcription.25 The probes were identical, except that one probe contained a molecular switch, MDV-int-B (Fig. 2), and the other, MDV-int-A, did not. Both recombinant RNAs contained a probe sequence that was complementary to a conserved region of the integrase gene in HIV-1 RNA. This probe sequence was inserted into MDV-1 (+) RNA,7 which is a naturally occurring template for Qb replicase.6 The presence of a probe sequence within MDV-1 RNA does not interfere with the ability of the recombinant molecules to hybridize to their

An RNase III recognition site was created when either RNA probe hybridized to its complementary RNA target sequence. It is known that RNase III cannot cleave RNA:DNA hybrids.22,29 Indeed, when probe MDV-int-A was hybridized to its DNA target, no cleavage occurred (Fig. 3). However, when probe MDV-int-A was hybridized to either RNA target, it was cleaved into two fragments. From the lengths of these fragments (92 and 174 nucleotides), it can be inferred that cleavage occurred within the RNA:RNA probe-target helix. Furthermore, when probe MDVint-B was hybridized to its DNA target, it was cleaved into the same three fragments that were recovered from the digest of probe MDV-int-B alone. These fragments were identical to those expected for cleavage at the RNase III recognition site in MDV-int-B,

Amplifiable hybridization probes 1

2

3

4

5

6

7

191 8

9

10

314 nt 266 nt

171 nt

77 nt 66 nt

Fig. 3. Analysis of the RNase III digestion products of probes MDV-int-A and MDV-int-B in the presence and absence of targets. Lane 1 contained undigested probe MDV-int-A (266 nucleotides), and lane 2 contained undigested probe MDV-int-B (314 nucleotides). Lane 3 contained digested probe MDV-int-A, and lane 4 contained digested probe MDVint-B. Lanes 5, 7, and 9 contained the digestion products of hybrids formed with probe MDV-int-A, and lanes 6, 8, and 10 contained the digestion products of hybrids formed with probe MDV-int-B. The digested hybrids in lanes 5 and 6 were formed with an oligodeoxyribonucleotide target (53 nucleotides), the digested hybrids in lanes 7 and 8 were formed with an oligoribonucleotide target (55 nucleotides), and the digested hybrids in lanes 9 and 10 were formed with a full-length transcript of the HIV-1 integrase gene (879 nucleotides).

namely a 171-nucleotide 3′ fragment, a 77-nucleotide 5′ fragment, and a 66-nucleotide internal fragment containing the inserted HIV-1 probe sequence. However, when probe MDV-int-B was hybridized to either RNA target, only the 171-nucleotide fragment and the 77-nucleotide fragment were recovered. From the absence of the 66-nucleotide fragment, we infer that an additional cleavage occurred within the RNA:RNA probe-target duplex that resulted in the creation of two smaller fragments that were so small that they ran out of the gel. To investigate whether the inserted HIV-1 probe sequence had coincidentally created an RNase III recognition site when it hybridized to its complementary RNA target sequence, another recombinant RNA probe, MDV-int-C, was prepared that was similar to MDV-int-B, except that it contained a different 44-nucleotide-long probe sequence. This new HIV-1 probe sequence was chosen to be as different as possible from the published consensus RNase III recognition sequence.30 Probe MDV-int-C was hybridized to full-length integrase RNA. The hybrids were incubated with RNase III and the digestion products were analysed electrophoretically

(Fig. 4). Cleavage still occurred within the probetarget helix. These results suggest that RNase III is non-specific in its ability to bind and cleave relatively long double-stranded RNA duplexes. Although many researchers believe that the processing of RNA by RNase III is sequence-specific,31–33 our results support the conclusion of other researchers that RNase III can hydrolyse any long double-stranded RNA in a sequence-independent manner.21,22,34

Functioning of the molecular switch It is surmised that the molecular switch that was constructed did not undergo the expected conformational change because its flanking sequences formed structures that forced the probe into an intermediate conformation in which the molecular switch was only partially open and the probe sequence was incompletely hybridized. An additional modification in probe MDV-int-C was therefore included in which two 7-nucleotide-long regions of the MDV-1 RNA sequences that surrounded the molecular switch in MDV-int-B were deleted. These sequences were not

192

H. J. Blok and F. R. Kramer 1

2

3

4

300

164

70 66

Fig. 4. Analysis of the RNase III digestion products of probe MDV-int-C in the presence and absence of targets. Lane 1 contained undigested probe MDV-int-C (300 nucleotides). Lane 2 contained digested probe MDV-intC. Lanes 3 and 4 contained the digestion products of the hybrids of probe MDV-int-C and its targets. The target in lane 3 was a single-stranded circular bacteriophage M13 DNA containing the complete HIV-1 integrase sequence, while the target in lane 4 was a full-length HIV-1 integrase RNA transcript (879 nucleotides).

required for replication. When this probe was hybridized to its DNA target and incubated with RNase III, the molecular switch still remained closed, since cleavage occurred at the existing RNase III recognition site (Fig. 4). When this probe was hybridized to its RNA target, however, another surprising result occurred. Instead of obtaining the fragments expected from cleavage at both the new RNase III recognition site created by formation of the RNA:RNA probetarget helix and from cleavage at the existing RNase III recognition site formed by the closed molecular switch, the majority of the fragments resulted from exclusive cleavage at the newly created site in the probe-target helix. These results suggest that the molecular switch did indeed undergo a conformational change when the probe hybridized to its RNA target. The few other fragments that were seen in the autoradiograph were probably due to secondary cleavage at recreated RNase III sites that occur when the primary cleavage products anneal to each other. This interpretation of the results is supported by the observation that in all the other experiments that were

performed no significant cleavage products were recovered as a result of exclusive cleavage at the RNase III recognition site formed by the probe-target helix. Thus, the deletions in the MDV-1 RNA sequences flanking the molecular switch significantly improved its functioning. These results also indicate that a significant factor affecting the functioning of the molecular switch was whether the target was RNA or DNA. RNA:RNA hybrids are more stable than RNA:DNA hybrids. Consequently, an RNA target is more likely to force a molecular switch to open than a DNA target. In the hybrid formed by probe MDV-int-C and its DNA target the probe was fully digested at its existing RNase III recognition site (Fig. 4), while in the hybrid formed by probe MDV-int-C and its RNA target, the probe was mainly digested at the RNase III recognition site created by the formation of the probe-target helix, indicating that the switch had undergone a conformational change when the target was an RNA. Taken together, these results imply that a molecular switch functions more effectively if its hairpin stem is shorter. In the design that we tested, the hairpin stem is rather long in order to accommodate the RNase III recognition site. Recent experiments performed in our laboratory indicate that in the presence of magnesium, which strongly stabilizes double-stranded helices, molecular switches that possess hairpin stems as short as five base pairs are able to function effectively.35 It is concluded that for a molecular switch to function when it is embedded within a replicatable probe it is desirable that the target be an RNA, that the sequences flanking the molecular switch do not contribute to the formation of stable secondary structures, and that the hairpin stem be shorter than the one used in the experiments reported in this paper.

Improved design for probes containing a molecular switch Our results demonstrate that RNase III does not have the specificity required for exclusive cleavage at the recognition site within the molecular switch. In all the experiments that we performed it cleaved the RNA:RNA duplex that was created by hybridization of the probe to its target, including the one chosen to be as different as possible from the consensus RNase III recognition site. In addition, our results indicate that the functioning of a molecular switch embedded within a replicatable hybridization probe would be markedly improved if its hairpin stem were shorter than those that form an RNase III recognition site.

Amplifiable hybridization probes

Based on these results, we propose a new probe design in which MDV-1 probes are constructed from deoxyribonucleotides and contain a molecular switch that possesses short hairpin stems that serve as a recognition site for a restriction endonuclease. Restriction endonucleases, such as TaqI or EcoRI, are extremely specific and are able to cleave short doublestranded sequences.36,37 These probes also contain an RNA polymerase promoter sequence at their 5′ end. After hybridization of these probes to their target sequences, the hybrids are washed to remove nonhybridized probes and are then incubated with the appropriate restriction endonuclease to selectively destroy the remaining non-hybridized probes. The hybridized probes are then transcribed into replicatable reporter RNAs, which are then amplified exponentially by incubation with Qb replicase. The amount of reporter RNA that is synthesized indicates the number of target molecules that were present in the original sample. When the targets are RNA molecules, unexpected cleavage sites cannot form within the probe-target helices, because RNA:DNA heteroduplexes are not substrates for restriction endonucleases. Moreover, the shorter hairpin stems that can accommodate a restriction site should easily undergo the desired conformational change when the probe hybridizes to its target. Assays that utilize this smart probe strategy should have very low backgrounds and should be extremely sensitive.

6.

7.

8.

9.

10.

11.

12. 13.

14.

15.

ACKNOWLEDGEMENTS We thank Dr Sanjay Tyagi (Public Health Research Institute) for insightful discussions and Dr Donald Court (National Cancer Institute) for the gift of E. coli RNase III. This work was supported by the National Institutes of Health (Grants HL-43521 and AI-37015) and by the Eindhoven University of Technology.

REFERENCES 1. Kramer, F. R. & Lizardi, P. M. (1989). Replicatable RNA reporters. Nature 339, 401–2. 2. Lomeli, H., Tyagi, S., Pritchard, C. G., Lizardi, P. M. & Kramer, F. R. (1989). Quantitative assays based on the use of replicatable hybridization probes. Clinical Chemistry 35, 1826–31. 3. Pritchard, C. G. & Stefano, J. E. (1991). Detection of viral nucleic acids by Qb replicase amplification. Medical Virology 10, 67–80. 4. Lizardi, P. M., Guerra, C. E., Lomeli, H., Tussie-Luna, I. & Kramer, F. R. (1988). Exponential amplification of recombinant-RNA hybridization probes. Bio/Technology 6, 1197–202. 5. Kramer, F. R. & Lizardi, P. M. (1990). Amplifiable

16.

17.

18.

19.

20.

21.

193

hybridization probes. Annales de Biologie Clinique (Paris) 48, 409–11. Kacian, D. L., Mills, D. R., Kramer, F. R. & Spiegelman, S. (1972). A replicating RNA molecule suitable for a detailed analysis of extracellular evolution and replication. Proceedings of the National Academy of Science, USA 69, 3038–42. Mills, D. R., Kramer, F. R. & Spiegelman, S. (1973). Complete nucleotide sequence of a replicating RNA molecule. Science 180, 916–27. Miele, E. A., Mills, D. R. & Kramer, F. R. (1983). Autocatalytic replication of a recombinant RNA. Journal of Molecular Biology 171, 281–95. Saiki, R. K., Scharf, S., Faloona, F. et al. (1985). Enzymatic amplification of b-globin genomic sequences and restriction site analysis for diagnosis of sickle cell anemia. Science 230, 1350–4. Mullis, K. B. & Faloona, F. A. (1987). Specific synthesis of DNA in vitro via a polymerase-catalyzed chain reaction. Methods in Enzymology 155, 335–50. Barany, F. (1991). Genetic disease detection and DNA amplification using cloned thermostable ligase. Proceedings of the National Academy of Science, USA 88, 189–93. Barany, F. (1991). The ligase chain reaction in a PCR world. PCR Methods and Applications 1, 5–16. Kievits, T., van Gemen, B., van Strijp, D. et al. (1991). NASBA isothermal enzymatic in vitro nucleic acid amplification optimized for the diagnosis of HIV-1 infection. Journal of Virological Methods 35, 273–86. van Gemen, B., Kievits, T., Schukkink, R. et al. (1993). Quantification of HIV-1 RNA in plasma using NASBA during HIV-1 primary infection. Journal of Virological Methods 43, 177–88. Guatelli, J. C., Whitfield, K. M., Kwoh, D. Y., Barringer, K. J., Richman, D. D. & Gingeras, T. R. (1990). Isothermal, in vitro amplification of nucleic acids by a multienzyme reaction modeled after retroviral replication. Proceedings of the National Academy of Science, USA 87, 1874–8. Fahy, E., Kwoh, D. Y. & Gingeras, T. R. (1991). Selfsustained sequence replication (3SR): an isothermal transcription-based amplification alternative to PCR. PCR Methods and Applications 1, 25–32. Walker, G. T., Little, M. C., Nadeau, J. G. & Shank, D. D. (1992). Isothermal in vitro amplification of DNA by a restriction enzyme/DNA polymerase system. Proceedings of the National Academy of Science, USA 89, 392–6. Syva¨nen, A.-C., Laaksonen, M. & So¨derlund, H. (1986). Fast quantification of nucleic acid hybrids by affinity-based hybrid collection. Nucleic Acids Research 14, 5037–48. Morrissey, D. V., Lombardo, M., Eldredge, J. K., Kearney, K. R., Groody, E. P. & Collins, M. L. (1989). Nucleic acid hybridization assays employing dA-tailed capture probes. I. Multiple capture methods. Analytical Biochemistry 181, 345–59. Hunsaker, W. R., Badri, H., Lombardo, M. & Collins, M. L. (1989). Nucleic acid hybridization assays employing dA-tailed capture probes. II. Advanced multiple capture methods. Analytical Biochemistry 181, 360–70. Robertson, H. D., Webster, R. E. & Zinder, N. D. (1968). Purification and properties of ribonuclease III

194

22. 23. 24.

25. 26.

27.

28.

H. J. Blok and F. R. Kramer from Escherichia coli. Journal of Biological Chemistry 243, 82–91. Crouch, R. J. (1974). Ribonuclease III does not degrade deoxyribonucleic acid hybrids. Journal of Biological Chemistry 249, 1314–6. Dunn, J. J. (1982). Ribonuclease III. In The Enzymes (Boyer, P. D., ed.) pp. 485–99. New York: Academic Press. Shore, D., Langowski, J. & Baldwin, R. L. (1981). DNA flexibility studied by covalent closure of short fragments into circles. Proceedings of the National Academy of Science, USA 78, 4833–7. Blok, H. J. (1992). Target-dependent amplifiable nucleic acid hybridization probes. Doctoral Dissertation. Eindhoven University of Technology. Adachi, A., Gendelman, H. E., Koenig, S. et al. (1986). Production of acquired immunodeficiency syndromeassociated retrovirus in human and nonhuman cells transfected with an infectious molecular clone. Journal of Virology 59, 284–91. Zuker, M. & Stiegler, P. (1981). Optimal computer folding of large RNA sequences using thermodynamics and auxiliary information. Nucleic Acids Research 9, 133–48. Faubladier, M., Cam, K. & Bouche´, J.-P. (1990). Escherichia coli cell division inhibitor DicF-RNA of the dicB operon. Evidence for its generation in vivo by transcription termination and by RNase III and RNase E-dependent processing. Journal of Molecular Biology 212, 461–71.

29. Robertson, H. D. (1982). Escherichia coli ribonuclease III cleavage sites. Cell 30, 669–72. 30. Krinke, L. & Wulff, D. L. (1990). The cleavage specificity of RNase III. Nucleic Acids Research 18, 4809– 15. 31. Dunn, J. J. & Studier, F. W. (1973). T7 early RNAs are generated by site-specific cleavages. Proceedings of the National Academy of Science, USA 70, 1559–63. 32. Gegenheimer, P. & Apirion, D. (1981). Processing of procaryotic ribonucleic acid. Microbiological Reviews 45, 502–41. 33. Bardwell, J. C. A., Re´gnier, P., Chen, S.-M., Nakamura, Y., Grunberg-Manago, M. & Court, D. L. (1989). Autoregulation of RNase III operon by mRNA processing. EMBO Journal 8, 3401–7. 34. Robertson, H. D. & Dunn, J. J. (1975). Ribonucleic acid processing activity of Escherichia coli ribonuclease III. Journal of Biological Chemistry 250, 3050–6. 35. Tyagi, S. & Kramer, F. R. (1996). Molecular beacons: probes that fluoresce upon hybridization. Nature Biotechnology 14, 303–8. 36. Rosenberg, J. M. (1991). Structure and function of restriction endonucleases. Current Opinion on Structural Biology 1, 104–13. 37. Cao, W., Mayer, A. N. & Barany, F. (1995). Stringent and relaxed specificities of TaqI endonuclease: interactions with metal cofactors and DNA sequences. Biochemistry 34, 2276–83.