RNA-CLEAVING DNA ENZYMES WITH ALTERED REGIO- OR ENANTIOSELECTΓVITY
BACKGROUND OF THE INVENTION FIELD OF THE INVENTION [0001] The present invention relates to nucleic acid enzymes or catalytic
(enzymatic) DNA molecules that are capable of cleaving non-naturally occurring nucleotides, particularly ribonucleotides.
BACKGROUND INFORMATION
[0002] Two important properties of an enzyme are its catalytic rate enhancement and its specificity for a particular chemical transformation. There are several kinds of specificity, including that for a particular class of substrate, that for a particular regioisomer, and that for a particular stereoisomer. Naturally occurring enzymes, composed of either proteins or nucleic acids, exhibit these various types of selectivity as a consequence of their complex structure. Even simple peptides or oligonucleotides are capable of operating with a high degree of specificity1""3. [0003] The ability to obtain novel protein and nucleic acid enzymes through directed evolution has enabled the development of artificial enzymes that are capable of performing regio- or stereoselective chemical reactions4-7. Evolved protein enzymes have found industrial applications as chiroselective catalysts8. RNA enzymes have been obtained that catalyze a Diels-Alder cycloaddition reaction and operate with an enantiomeric excess (ee) of greater than 95% (ref. 9). When that same RNA enzyme was prepared from L- rather than D-nucleotides, it of course produced the opposite enantiomeric product with the same ee value. [0004] Nucleic acid enzymes also have been shown to operate in a regiospecific manner. For example, the class I RNA ligase ribozyme selectively catalyzes the formation of a 3',5'- rather than 2',5'-phosphodiester linkage10,11. The hammerhead ribozyme cleaves 3 ',5'- but not 2',5'-phosphodiester linkages of RNA12. The hepatitis delta vims ribozyme preferentially cleaves the natural 3 ',5 '-linkage, but also operates with about 100-fold reduced rate in cleaving a 2',5'-phosphodiester12.
[0005] There remains a need for enzymes which specifically cleave non- natural polyncucleotides, for example polynucleotides that include 2',5'- phosphodiester linkages or polynucleotides which are L-enantiomers. One of the advantages of an enzyme of this type, is that it can be used in diagnostic assays with reduced levels of non-specific cleavage of naturally-occurring polynucleotides from a sample.
SUMMARY OF THE INVENTION
[0006] In order to identify new nucleic acid enzymes with regio- and enantiospecificity, in vitro evolution methods were employed to isolate novel DNA enzymes. One group of isolated DNA enzymes cleaves a 2 ',5 '-linked β-D- ribonucleotide, and another group cleaves a 3 ',5 '-linked β-L-ribonucleotide (Figure 1 A). The DNA enzymes were discovered using in vitro evolution methods starting with separate populations of ~1015 random-sequence DNA molecules. Each population gave rise to Mg2+-dependent DNA enzymes that performed the target reaction. Both groups of enzymes were made to operate on a separate nucleic acid substrate with multiple-turnover. The 2',5'-phosphodiester-cleaving DNA group of enzymes exhibits a rate enhancement of about 20,000-fold compared to the uncatalyzed reaction, whereas the L-ribonucleotide-cleaving group of DNA enzymes exhibits a catalytic rate enhancement of about 600-fold. The former operates with a regioselectivity of about 6,000-fold, while the latter operates with an enantioselectivity of about 50-fold.
[0007] In one aspect, the present invention provides a catalytic single-stranded
DNA molecule that has site-specific endonuclease activity that is specific for a cleavage site in a substrate nucleic acid sequence, wherein the cleavage site includes a non-naturally occurring single-stranded ribonucleic acid, or a composite thereof. The catalytic DNA molecule includes one or more loop regions and one or more binding regions, wherein the binding regions bind to complementary sequences of the substrate nucleic acid sequence.
[0008] The catalytic single-stranded DNA molecules have site-specific endonuclease activity for a substrate nucleic acid sequence that can include a wide variety of non-naturally occurring single-stranded nucleic acids at the cleavage site.
These non-naturally occurring single-stranded nucleic acids in preferred embodiments include a 2',5' linked residue, for example a 2,,5'-linked adenylate or guanylate ribonucleotide residue, or an L enantiomer residue, for example an L-enantiomer 3',5'- linked adenylate ribonucleotide residue.
[0009] In one aspect, the present invention provides a catalytic single-stranded
DNA molecule that cleaves a substrate nucleic acid sequence that includes a 2', 5'- linked ribonucleotide, preferably a 2',5'-linked adenylate or guanylate ribonucleotide residue. The catalytic DNA molecule of this aspect of the invention has a catalytic domain and a recognition domain. The catalytic domain includes a downstream region that includes the sequence 5'-XιX2ACTCGGAGX3 -3' (SEQ TD NO:28) which is capable of forming a loop; a central stem region immediately 5' to the downstream loop and having the sequence δ'-Zi Z2 Z3Z4-3', and an upstream region immediately 5' to the central stem region, capable of forming a loop having the sequence 5'-GGGA- 3'. Xι is an optional cytidine residue, X2 is a cytidine or a thymidine residue, and X3 binds a complementary nucleotide on the substrate nucleic acid sequence that is two nucleotides upstream from the cleavage site. Z4 binds a complementary nucleotide on the substrate nucleic acid sequence that is immediately downstream from a cleavage site. In preferred embodiments Z2 is a cytidine residue and Z3 is a guanidine residue. [0010] The recognition domain includes an upstream flanking region and a downstream flanking region. The upstream flanking region is immediately 5' to the upstream loop and the downstream flanking region is immediately 3' to the downstream loop. In various embodiments, the individual nucleotides in the flanking regions are able to biNd to flanking regions of the substrate nucleic acid sequence. [0011] In certain embodiments of this aspect of the invention, the downstream loop has the sequence 5'-CCACTCGGAG-3' (SEQ ID NO:22).- In certain embodiments of this aspect of the invention, the upstream loop has the sequence 5'-YGGGA-3' wherein Y is 0 to 5 nucleotides. hi these embodiments, Y can have virtually any nucleotide sequence of 0 to 5 nucleotides. For example, Y can include the sequence 5 '-TTA-3 ', for example Y can be 5 '-GTTTA-3 ' (SEQ ID NO: 19), 5 '- GCTTA-3' (SEQ TD NO:20), 5'-GTTA-3' (SEQ ID NO:21). [0012] In another aspect as illustrated in Figure 14B, the present invention provides a catalytic single-stranded DNA molecule that cleaves a substrate nucleic
acid sequence at an L ribonucleotide residue, most preferably a 3 ',5 ' linked L- adenylate ribonucleotide residue. In this aspect, the catalytic DNA molecule includes a recognition domain and a catalytic domain capable of forming a loop that includes the nucleic acid sequence 5,-XιX2X3GX4CX5X6X7GACX8X9-3' (SEQ ID NO:29). Xt binds a complementary nucleotide on a substrate nucleic acid sequence that is immediately downstream from a cleavage site on the substrate nucleic acid sequence, X2 is a thymidine or a guanidine residue, X3 is a cytidine or a guanidine residue, X4 is a cytidine or a thymidine residue, X5 is a cytidine or a thymidine residue, X6 is a cytidine or a thymidine residue, X7 is an adenosine or a guanidine residue, X8 is an adenosine or a thymidine residue, and X9 binds a complementary nucleotide on the substrate nucleic acid sequence that is two nucleotides upstream from the cleavage site.
[0013] The recognition domain comprising an upstream flanking region and a downstream flanking region, the upstream flanking region being immediately 51 to the catalytic domain and the downstream flanking region being immediately 3' to the catalytic domain, as described above for catalytic DNA molecules that cleave 2', 3' ribonucleotides.
[0014] In some embodiments of the catalytic DNA of this aspect of the invention, X2 is a thymidine residue, X3 is a cytidine residue, X4 is a thymidine residue, X5 is a thymidine residue, X6 is a thymidine residue, X7 is an adenosine residue, and X8 is an adenosine residue. In a specific example of a preferred embodiment of the catalytic DNA of this aspect of the invention, the loop comprises 5' TCGTCTTAGACA 3' (SEQ ID NO:30).
[0015] In another aspect, the present invention provides a substrate nucleic acid sequence that includes a non-naturally occurring ribonucleotide immediately upstream from a cleavage site that is flanked by complementary sequences that bind to binding regions, also referred to herein as flanking regions, of the catalytic DNA molecule. In certain embodiments, the non-naturally occurring ribonucleotide of the substrate nucleic acid sequence is a 2', 5' ribonucleotide, preferably an adenylate or a guanylate residue, most preferably a guanylate residue. In other embodiments, the non non-naturally occurring ribonucleotide of the substrate nucleic acid sequence is a L-enantiomer of a 3', 5', ribonucleotide, most preferably an adenylate residue.
[0016] In another aspect, the present invention provides a non-naturally occurring single-stranded nucleic acid substrate, as discussed above, that includes a pair of interactive labels consisting of a first label and a second label, separated from each other by a cleavage site. The labels are attached to the single-stranded nucleic acid substrate either directly or indirectly. In some embodiments, the first label is a fluorescent moiety and the second label is a quencher that quenches the fluorescent moiety when both the fluorescent moiety and the quencher are attached to the single- stranded nucleic acid substrate.
[0017] In another aspect the present invention provides a method for detecting a target nucleic acid sequence that uses the catalytic nucleic acid molecules and substrate nucleic acid sequences of the present invention. The method is related to a method for quantitative PCR, termed "DzyNA-PCR."
[0018] Accordingly, the method includes admixing in an amplification buffer, the following components: i) a nucleic acid sample; ii) a polymerase; iii) a substrate non-naturally occurring single-stranded nucleic acid sequence comprising a pair of interactive labels consisting of a first label and a second label being attached to the oligonucleotide directly or indirectly, wherein the first label is separated from the second label by a non-naturally occurring ribonucleotide cleavage site, as discussed above; iv) a forward primer capable of binding to a 3' portion of a first strand of the target nucleic acid sequence and directing polynucleotide synthesis of the target nucleic acid sequence in a forward direction; and v) a reverse primer comprising a region capable of binding to a 3' portion of a second strand of the target nucleic acid sequence and directing polynucleotide synthesis of the target nucleic acid sequence in a reverse direction, and including the complement of a catalytic single-stranded DNA molecule, wherein the catalytic single-stranded DNA molecule is capable of cleaving the substrate nucleic acid sequence at the cleavage site.
[0019] The method then includes incubating the admixed components under amplification conditions to amplify the target nucleic acid sequence. This results in
synthesis of the catalytic single-stranded DNA molecule. The synthesized catalytic single-stranded DNA molecule then cleaves the substrate nucleic acid sequence, thereby releasing the interaction of the first label and the second label. The first label is then detected, thereby detecting the target nucleic acid sequence. [0020] In exemplary embodiments of the method, the non-naturally occurring ribonucleotide cleavage site includes a 2',5' linked residue or an L-enantiomer residue, as described above for the substrate nucleic acid sequences of the present invention. Furthermore, the catalytic single-stranded DNA molecule is preferably one of the catalytic single-stranded DNA molecules described above.
[0021] The present invention also contemplates methods that utilize directed evolution to produce nucleic acid molecules having a predetermined catalytic activity for cleaving a single stranded nucleic acid that includes a non-naturally occurring ribonucleotide. Accordingly, in another aspect, the present invention provides a method for identifying a catalytic DNA molecule having site-specific endonuclease activity that is specific for a non-naturally occurring ribonucleotide cleavage site. The method includes constructing a library of double stranded nucleic acid molecules that includes the non-naturally occurring ribonucleotide cleavage site and that includes a region of random-sequence nucleotides that are potentially capable of interacting with the region of the cleavage site. Then one strand of the library of double stranded nucleic acid molecules is captured, thereby providing a library of captured single- stranded nucleic acid molecules. Next, the library of captured single-stranded nucleic acid molecules is incubated under cleavage conditions to permit cleavage at the cleavage site and release of cleaved nucleic acid molecules. Then, the cleaved nucleic acid molecules are isolated, thereby identifying a catalytic DNA molecule having site- specific endonuclease activity that is specific for a non-naturally occurring ribonucleotide cleavage site.
[0022] In illustrative embodiments, the method further includes amplifying the cleaved nucleic acid molecules and repeating the capture, cleavage, and isolation steps above between 1 and 50 times, typically between 1 and 20 times. The selectively amplified cleaved nucleic acid molecules are randomly mutagenized to form mutagenized cleaved nucleic acid molecules. The mutagenized cleaved nucleic
acid molecules are then amplified the capture, cleavage, and isolation steps above are performed between 1 and 20 times.
[0023] In illustrative embodiments, as demonstrated in the Example below, the cleavage conditions can be changed during the method such that in subsequent repeat occurrences of the cleaving step, the cleavage reaction must be more efficient for nucleic acid cleavage to occur.
[0024] In several embodiments the non-naturally occurring ribonucleotide cleavage site is a 2 ',5 '-linked guanylate ribonucleotide cleavage site or an L-adenylate ribonucleotide cleavage site.
[0025] The present invention also provides kits that include the catalytic DNA molecules and or substrates with non-naturally occurring ribonucleotides. In other embodiments, the kits also include primers, polymerase, and other reagents useful for the methods of the invention.
BRIEF DESCRIPTION OF THE FIGURES
[0026] Figures 1A and IB show compounds employed in the development of
DNA enzymes that cleave unnatural ribonucleotide analogs. Figure 1 A shows chemical structure of a 2 ',5 '-linked β-D-guanylate (left) and 3 ',5 '-linked β-L- adenylate (right). Figure IB diagrammatically shows the structure of the starting library of DNA molecules that were used to obtain DNA enzymes with the desired activity. Each molecule contained a 5 '-terminal biotin (encircled B), either a 2 ',5'- linked D-nucleotide or 3 ',5 '-linked L-nucleotide at the target cleavage site (X), a fixed hairpin loop region downstream from the cleavage site (sequence shown) (SEQ ID NO: 31), and 50 random-sequence deoxynucleotides (N50). [0027] Figures 2A-2C show the putative secondary structure of the
(Figure 2A) 2':10-16 (SEQ LD NOS: 32, 9), (Figure 2 B) 2':15-2 (SEQ ID NOS: 32, 10), and (Figure 2 C) L: 15-30 (SEQ ID NOS: 33, 18) catalytic DNA molecules, each shown bound to its substrate in the intermolecular reaction format. Bold letter G or A indicates a 2 ',5 '-linked β-D-guanylate or 3 ',5 '-linked β -L-adenylate, respectively. Arrow indicates the cleavage site.
[0028] Figure 3 shows a time course of the cleavage reaction catalyzed by the
2': 10-16 (solid circle), 2': 15-2 (solid square), and L: 15-30 DNA enzymes (solid
triangle), measured under single-turnover conditions. Inset shows a detail of the first 1,000 min of the reaction, demonstrating its biphasic nature for the two 2 ',5'- phosphodiester-cleaving enzymes. Data were fit to either a single- or double- exponential equation (see Experimental Section of the Example). Reaction conditions: 25 mM MgCl2, 150 mM NaCl, pH 7.5, and 37 °C.
[0029] Figures 4A-4D show the catalytic activity of DNA enzymes that cleave unnatural ribonucleotide analogs, measured under multiple-turnover conditions. (Figure 4 A) 2': 10-16 DNA enzyme with full-length stem regions surrounding the cleavage site; (Figure 4B) 2': 10-16 DNA enzyme with the stem regions shortened by one base pair each; (Figure 4C) 2':15-2 DNA enzyme; (Figure 4D) L:15-30 DNA enzyme. Data were fit to a curve based on the Michaelis-Menten equation: v = kcat [substrate] / (Km + [substrate]). Reaction conditions: 25 mM MgCl2, 150 mM NaCl, pH 7.5, and 37 °C.
[0030] Figures 5A-5C show the catalytic activity of DNA enzymes that cleave unnatural ribonucleotide analogs, measured under single-turnover conditions. &0bs was determined for various concentrations of enzyme and the data were fit to a curve based on the equation: &0bs= &max [enzyme]/^ + [enzyme]), where kmsκ= &cat t saturating concentrations and K_ is the apparent dissociation constant for the enzyme- substrate complex. (Figure 5A) 2':10-16 DNA enzyme, with a kcat of 0.011 ± 0.0004 min"1 andXd of 0.11 ± 0.01 nM; (Figure 5B) 2':15-2 DNA enzyme, with a kcat of 0.034 ± 0.001 min"1 andXd of 0.12 ± 0.01 nM; (Figure 5C) L:15-30 DNA enzyme, with a kc_t of 0.0016 ± 0.0001 min"1 anάK_ of 3.2 ± 0.5 nM. Reaction conditions: 25 mM MgCl2, 150 mM NaCl, pH 7.5, and 37 °C.
[0031] Figures 6A-6B show pH dependence of the DNA-catalyzed reactions.
(Figure 6A) 2':10-16 DNA enzyme; (Figure 6B) L:15-30 DNA enzyme. The buffer was either MES (circles), EPPS (squares), or CHES (triangles). [0032] Figure 7 shows an autoradiogram depicting the cleavage reaction catalyzed by either the L: 15-30 DNA enzyme or 2': 10- 16 DNA enzyme, each with either its corresponding unnatural ribonucleotide substrate or a substrate in which the unnatural ribonucleotide was replaced by a standard ribonucleotide. Reaction conditions: 25 mM MgCl2, 150 mM NaCl, pH 7.5, and 37 °C, incubated in either the presence (+) or absence (-) of the DNA enzyme for 6 hr. The unnatural ribonucleotide
substrates also were subjected to alkaline hydrolysis (OFT) by incubating them in the presence of 0.1 N NaOH for 6 hr at 37 °C.
[0033] Figure 8 shows the kinetics of DNA-catalyzed cleavage of substrates that contained a natural ribonucleotide in place of the unnatural ribonucleotide.
Reactions were carried out in the presence of saturating concentrations of DNA enzyme, employing either the 2': 10-16 DNA enzyme (circle), 2': 15-2 DNA enzyme
(square), or L: 15-30 DNA enzyme (triangle). The catalytic rate was obtained from a best-fit line of the data plotted as a function of time. Reaction conditions: 25 mM
MgCl2, 150 mM NaCl, pH 7.5, and 37 °C.
[0034] Figures 9A-9B show the sequences of the variable region of individual clones isolated following the 15 round of in vitro selection for either 2', 5'- phosphodiester- or L-ribonucleotide-cleaving activity (SEQ ID NOS 1-8 and 11-17). Boxes indicate the regions with high sequence similarity. [0035] Figures 10 A- 10B show the dependence of catalytic rate on the concentration of Mg2+ for the (Figure 10A) 2':10-16 and (Figure 10B) L-15-30 DNA enzymes. The curve represents a best-fit line to the data based on the equation: λr0bs = &max [Mg2+]/([Mg2+] + Kd), where kmax is &0bS in the presence of saturating Mg2+, and k is the apparent dissociation constant for Mg2+.
[0036] Figures 11 A-l IB show the temperature dependence of the DNA- catalyzed reaction for the (Figure 11A) 2':10-16 and (Figure 11B) L-15-30 DNA enzymes. Reaction conditions: 25 mM MgCl2, 150 mM NaCl, and pH 7.5. [0037] Figures 12A-12B show divalent metal dependence of the DNA- catalyzed reaction for the (Figure 12A) 2':10-16 and (Figure 12B) L-15-30 DNA enzymes. Reaction conditions: 10 mM M2+, 150 mM NaCl, PH 7.5, and 37 °C. The metals are listed from left to right in order of decreasing pka of the corresponding metal hydrate.
[0038] Figures 13A-13D show MALDI mass spectra of oligonucleotide products resulting from reactions catalyzed by the 2':10-16 and L-15-30 DNA enzymes. (Figure 13 A) 5'-product from the 2', 5'-phosphodiester-containing substrate, with an expected m/z for the principle product ion of 3,346; (Figure 13B) 3'-product from the 2',5'-phosphodiester-containing substrate, with an expected m/z of 4,200; (Figure 13C) 5'- product from the L-ribonucleotide-containing substrate, with an
expected m/z of 2,971; (Figure 13D) 3 '-product from the L-ribonucleotide-containing substrate, with an expected m/z of 2,804.
[0039] Figure 14 shows uncatalyzed cleavage of the substrates employed in this study, (solid circle) 2',5'-phosphodiester-containing substrate, with a &unCat of 1 x 10"6 min"1; (open circle) 3',5'-phosphodiester-containing substrate, with a &unCat of 2xl0"6 min"1; (solid square) L-ribonucleotide-containing substrate, with a &unCat of 3 x 10"6 min-1; (open square) D-ribonucleotide-containing substrate, with a &unCat of 4 x 10"6 min-1. The cleavage rate was obtained from a best-fit line of the data plotted as a function of time. Reaction conditions: 25 mM MgCl2, 150 mM NaCl, pH 7.5, and 37 °C.
[0040] Figures 15A-15B show a diagrammatic representation of catalytic
DNA molecules (bottom nucleic acid sequence) (SEQ LD NOS 34, 29) according to the present invention, aligned with a corresponding substrate (top nucleic acid sequence). In both panels the arrow represents the cleavage site and regions 10 and 20 represent substrate nucleic acid sequence regions that are complementary to an upstream (25) and a downstream (15) flanking region of the catalytic DNA molecules of the present invention. Figure 15A illustrates a catalytic DNA molecules which cleaves a substrate ribonucleotide having a 2', 5' linkage. Figure 15B illustrates a catalytic DNA molecule which cleaves a substrate L-enantiomer ribonucleotide having a 3', 5' linkage.
[0041] Figure 16 shows DzyNA-PCR strategy for homogeneous amplification and detection of specific nucleic acid sequences.
DETAILED DESCRIPTION OF THE INVENTION
[0042] The present invention provides a catalytic single-stranded DNA molecule that has site-specific endonuclease activity that is specific for a cleavage site in a substrate nucleic acid sequence, wherein the cleavage site includes a non- naturally occurring single-stranded ribonucleic acid, or a composite thereof. The catalytic DNA molecule includes one or more loop regions and one or more binding regions, wherein the binding regions bind to complementary sequences of the substrate nucleic acid sequence. The non-naturally occurring single-stranded nucleic acids in preferred embodiments include a 2',5' linked residue, more preferably a 2',5'-
linked adenylate or most preferably a 2',5'-linked guanylate ribonucleotide residue, or an L enantiomer residue, most preferably an L-enantiomer 3',5'-linked adenylate ribonucleotide residue.
[0043] As used herein, the term "catalytic DNA molecule" is used to describe a DNA-containing nucleic acid that is capable of functioning as an enzyme. In the present disclosure, the term "deoxyribozyme" includes endoribonucleases and endodeoxyribonucleases, although deoxyribozymes with endoribonuclease activity. Other terms used interchangeably with deoxyribozyme herein are "enzymatic DNA molecule", "DNAzyme", or " "deoxyribozyme", which terms should all be understood to include enzymatically active portions thereof, whether they are produced synthetically or derived from organisms or other sources.
[0044] Catalytic DNA molecules of the present invention typically include
DNA molecules that have complementarity in a substrate-binding region to a specified oligonucleotide target or substrate; such molecules also have an enzymatic activity which is active to specifically cleave the oligonucleotide substrate. The enzymatic DNA molecule is capable of cleaving the oligonucleotide substrate intermolecularly. This complementarity functions to allow sufficient hybridization of the enzymatic DNA molecule to the substrate oligonucleotide to allow the intermolecular cleavage of the substrate to occur. While one-hundred percent (100%) complementarity is preferred, complementarity in the range of 75-100%, for example 90% complementarity, 95% complementarity, and 99% complementarity, is also useful and contemplated by the present invention.
[0045] Enzymatic DNA molecules of the present invention may alternatively be described as having nuclease or ribonuclease activity. These terms are used interchangeably herein.
[0046] The term "enzymatic nucleic acid" as used herein encompasses enzymatic RNA or DNA molecules, enzymatic RNA-DNA polymers, and enzymatically active portions or derivatives thereof, although enzymatic DNA molecules are a class of enzymatically active molecules according to the present invention.
[0047] hi one aspect, the present invention provides a catalytic single-stranded
DNA molecule that cleaves a substrate nucleic acid sequence that includes a 2', 5'-
linked ribonucleotide, such as a 2',5'-linked adenylate or guanylate ribonucleotide residue, as shown in Figure 14 A. The catalytic DNA molecule of this aspect of the invention has a catalytic domain and a recognition domain (15, 25). The catalytic domain includes a downstream region that includes the sequence 5'- X,X2ACTCGGAGX3 -3' (SEQ ID NO:28) which is capable of forming a loop; a central stem region immediately 5' to the downstream loop and having the sequence 5'-Z]. Z2 Z3Z4-3', and an upstream region immediately 5' to the central stem region, capable of forming a loop having the sequence 5'-GGGA-3'. The loops formed in the catalytic DNA molecules of this aspect of the invention are internal bulge loops. Xj. is an optional cytidine residue, X2 is a cytidine or a thymidine residue, and X3 binds a complementary nucleotide on the substrate nucleic acid sequence that is two nucleotides upstream from the cleavage site. Z4 binds and/or forms a wobble pair with a nucleotide on the substrate nucleic acid sequence that is immediately downstream from a cleavage site. In some embodiments Z2 is a cytidine residue and Z3 is a guanidine residue.
[0048] The recognition domain includes an upstream flanking region (25) and a downstream flanking region (15). The upstream flanking region (25) is immediately 5' to the upstream loop and the downstream flanking region (15) is immediately 3' to the downstream loop. In various embodiments of the present invention, the flanking regions vary in length. Thus, for example, a flanking region may comprise a single nucleotide to seventy-five nucleotides. However, it is understood that flanking regions of about 3-25 nucleotides in length, preferably about 3-15 nucleotides in length, and more preferably about 3-10 nucleotides in length are particularly preferred. In various embodiments, the individual nucleotides in the flanking regions are able to form complementary base pairs with the nucleotides of the substrate molecules; in other embodiments, non-standard pairing interactions are formed. A mixture of complementary and nonstandard pairing is also contemplated as falling within the scope of the disclosed embodiments of the invention. [0049] As mentioned above in certain embodiments, the present invention provides a catalytic single-stranded DNA molecule that cleaves a substrate nucleic acid sequence that includes a 2', 5'-linked ribonucleotide, preferably a 2',5'-linked adenylate or guanylate ribonucleotide residue, as shown in Figure 14A. However,
methods disclosed herein related to using directed evolution can be used to identify catalytic single-stranded DNA molecules that cleave other non-naturally occurring nucleic acids. For example, catalytic single-stranded DNA molecules that cleave for example, 2'-deoxy-2'aminonucleotides, 2'-deoxy-2'-fluoronucleotides, nucleoside 5'- phosphorothioates, nucleoside (3' -> 5') phosphoramidates, arabinonucleosides, and various RNA analogs that contain modified bases.
[0050] In another embodiment, a catalytic DNA molecule of the present invention may further comprise a third flanking region. In certain embodiments of the present invention the catalytic DNA molecule can further include one or more variable or "spacer" regions between the flanking regions.
[0051] In certain embodiments of this aspect of the invention, the downstream loop has the sequence 5'-CCACTCGGAG-3' (SEQ ID NO:22). In certain embodiments of this aspect of the invention, the upstream loop has the sequence 5'-YGGGA-3' wherein Y is 0 to 5 nucleotides. In these embodiments, Y can have virtually any nucleotide sequence of 0 to 5 nucleotides. For example, Y can include the sequence 5 '-TTA-3 ', for example Y can be 5 '-GTTTA-3 ' (SEQ ID NO: 19), 5 '- GCTTA-3' (SEQ ID NO:20), 5'-GTTA-3' (SEQ ID NO:21). [0052] In certain embodiments, Z is a guanidine residue, Xi is a cytidine residue, X2 is a cytidine residue, and X3 is a guanidine residue. In other preferred embodiments as illustrated in Figures 2a, 2b and 9a, the catalytic DNA molecule includes SEQ ID NO:l, SEQ ID NO:2, SEQ ID NO:3, SEQ ID NO:4, SEQ ID NO:5, SEQ ID NO:6, SEQ ID NO:7, SEQ TD NO:8, SEQ ID NO:9, or SEQ ID NO:10. SEQ ID NOs:l-8 do not include the upstream flanking region because this region was in the fixed-sequence region that was not allowed to mutate during the course of evolution, as described in the Example. In a particular embodiment, the catalytic DNA includes SEQ ID NO: 10.
[0053] In another aspect as illustrated in Figure 14B, the present invention provides a catalytic single-stranded DNA molecule that cleaves a substrate nucleic acid sequence at an L ribonucleotide residue, such as a 3 ',5 ' linked L-adenylate ribonucleotide residue. In this aspect, the catalytic DNA molecule includes a recognition domain and a catalytic domain capable of forming a loop that includes the nucleic acid sequence 5'-X1X2X3GX4CX5X6X7GACX8X9-3' (SEQ ID NO:29). The
loop is an internal bulge loop. Xi binds a complementary nucleotide on a substrate nucleic acid sequence that is immediately downstream from a cleavage site on the substrate nucleic acid sequence, X2i is a thymidine or a guanidine residue, X3 is a cytidine or a guanidine residue, X is a cytidine or a thymidine residue, X5 is a cytidine or a thymidine residue, X6 is a cytidine or a thymidine residue, X7 is an adenosine or a guanidine residue, X8 is an adenosine or a thymidine residue, and X9 binds a complementary nucleotide on the substrate nucleic acid sequence that is two nucleotides upstream from the cleavage site.
[0054] The recognition domain comprising an upstream flanking region and a downstream flanking region, the upstream flanking region being immediately 5' to the catalytic domain and the downstream flanking region being immediately 3' to the catalytic domain, as described above for catalytic DNA molecules that cleave 2', 3' ribonucleotides.
[0055] In preferred embodiments of the catalytic DNA of this aspect of the invention, X2 is a thymidine residue, X3 is a cytidine residue, 4 is a thymidine residue, X5 is a thymidine residue, X6 is a thymidine residue, X7 is an adenosine residue, and X8 is an adenosine residue. In a specific example of a preferred embodiment of the catalytic DNA of this aspect of the invention, the loop comprises 5' TCGTCTTAGACA 3' (SEQ LD NO:30).
[0056] As illustrated in figures 2C and 9B, in certain preferred embodiments the catalytic DNA molecule of this aspect of the invention includes SEQ ID NO: 11, SEQ ID NO:12, SEQ LD NO:13, SEQ ID NO:14, SEQ ID NO:15, SEQ ID NO:16, SEQ LD NO:17, or SEQ ID NO:18. SEQ LD NOs:ll-17 do not include the upstream flanking region because this region was in the fixed-sequence region that was not allowed to mutate during the course of evolution, as described in the Example. In a particularly preferred embodiments, the catalytic DNA includes SEQ ID NO: 18. [0057] The endonuclease activity of the catalytic DNA molecules of the present invention typically involves hydrolytic cleavage of a phosphoester bond at the cleavage site. In certain embodiments of the present invention, the substrate nucleic acid sequence and the catalytic DNA molecule are part of the same DNA molecule. [0058] The endonuclease activity of the catalytic DNA molecules is typically enhanced by the presence of certain metals, especially divalent cations, as illustrated
in Figure 12. In certain embodiments, the catalytic DNA molecule catalyzes the cleavage at a 2', 5' ribonucleotide and the activity of the catalytic DNA molecule is enhanced in the presence of Mg2+, Sr24", Ca2+, or Ba2+, especially Mg2+. In other embodiments, the catalytic DNA molecule catalyzes the cleavage at an L-ribonucleotide and the activity of the catalytic DNA molecule is enhanced in the presence of Mn2+, Mg2+, Sr2+, Ca2+, Pb2+, or Zn2+. especially Mn2+ andMg2+. [0059] The catalytic DNA molecules and the substrate nucleic acid sequences of the present invention can include nucleotide analogs. "Nucleotide analog" generally refers to a purine or pyrimidine nucleotide that differs structurally from A, T, G, C, or U, but is sufficiently similar to substitute for the normal nucleotide in a nucleic acid molecule. As used herein, the term "nucleotide analog" encompasses altered bases, different or unusual sugars (i.e. sugars other than the "usual" ribofuranose) such as L-ribose instead of D-ribose, or a combination of altered bases and different or unusual sugars.
[0060] In various embodiments, a catalytic DNA molecule of the present invention may combine one or more modifications or mutations including additions, deletions, and substitutions, of the specific catalytic DNA molecule sequences disclosed herein that are introduced using methods well known in the art. In alternative embodiments, such mutations or modifications may be generated using methods which produce random or specific mutations or modifications. These mutations may, for example, change the length of, or alter the nucleotide sequence of, a loop, a spacer region or the recognition sequence (or domain). One or more mutations within one catalytically active enzymatic DNA molecule may be combined with the mutation(s) within a second catalytically active enzymatic DNA molecule to produce a new enzymatic DNA molecule containing the mutations of both molecules. [0061] Catalytic DNA molecules of the present invention may be of varying lengths and folding patterns, as appropriate, depending on the type and function of the molecule. For example, catalytic DNA molecules can be about 15 to about 400 or more nucleotides in length, although a length not exceeding about 250 nucleotides is typical, to avoid limiting the therapeutic usefulness of molecules by making them too large or unwieldy. In certain embodiments, a catalytic DNA molecule of the present invention includes not more than about 100 nucleotides. In still other embodiments,
catalytic DNA molecules of the present invention are about 20-75 nucleotides in length, more preferably about 20-65 nucleotides in length. Other preferred catalytic DNA molecules are about 10-50 nucleotides in length. In other applications, catalytic DNA molecules may assume configurations similar to those of "hammerhead" ribozymes. Such catalytic DNA molecules are typically no more than about 75-100 nucleotides in length, or a length of about 20-50 nucleotides.
[0062] It is also to be understood that a catalytic DNA molecule of the present invention may include enzymatically active portions of a catalytic DNA molecule or may include a catalytic DNA molecule with one or more mutations, e.g., with one or more base-pair-forming sequences or spacers absent or modified, as long as such deletions, additions or modifications do not adversely impact the molecule's ability to perform as an enzyme.
[0063] "Oligonucleotide or polynucleotide" generally refers to a polymer of single- or double-stranded nucleotides. As used herein, "oligonucleotide" and its grammatical equivalents includes the full range of nucleic acids. An oligonucleotide will typically refer to a nucleic acid molecule comprised of a linear strand of ribonucleotides or deoxyribonucleotides. The exact size will depend on many factors, which in turn depends on the ultimate conditions of use, as is well known in the art. [0064] Catalytic nucleic acid molecules of the present invention also include those with altered recognition sites or domains. In various embodiments, these altered recognition domains confer unique sequence specificities on the enzymatic nucleic acid molecule including such recognition domains. The exact bases present in the recognition domain determine the base sequence at which cleavage will take place. Cleavage of the substrate nucleic acid occurs within the recognition domain. This cleavage leaves a 2', 3', or 2',3'-cyclic phosphate group on the substrate cleavage sequence and a 5' hydroxyl on the nucleotide that was originally immediately 3' of the substrate cleavage sequence in the original substrate. Cleavage can be redirected to a site of choice by changing the bases present in the substrate binding region (also called the recognition sequence (internal guide sequence)) of the catalytic DNA molecule. See Murphy et al, Proc. Natl. Acad. Sci. USA 86: 9218-9222 (1989). [0065] Moreover, it may be useful to add a polyamine to facilitate recognition and binding between the catalytic DNA molecule and its substrate. Examples of
useful polyamines include spermidine, putrescine or spermine. A spermidine concentration of about 1 mM may be effective in particular embodiments, while concentrations ranging from about 0.1 mM to about 10 mM may be useful. [0066] In various alternative embodiments, a catalytic DNA molecule of the present invention has an enhanced or optimized ability to cleave nucleic acid substrates, preferably RNA substrates. As those of skill in the art will appreciate, the rate of an enzyme-catalyzed reaction varies depending upon the substrate and enzyme concentrations and, in general, levels off at high substrate or enzyme concentrations. Taking such effects into account, the kinetics of an enzyme-catalyzed reaction may be described in the following terms, which define the reaction.
[0067] The enhanced or optimized ability of a catalytic DNA molecule of the present invention to cleave an RNA substrate may be determined in a cleavage reaction with varying amounts of labeled RNA substrate in the presence of catalytic DNA molecule. The ability to cleave the substrate is generally defined by the catalytic rate (kcat) divided by the Michaelis constant (KM). The symbol kcat represents the maximal velocity of an enzyme reaction when the substrate approaches a saturation value. KM represents the substrate concentration at which the reaction rate is one-half maximal.
[0068] For example, values for KM and kcat may be determined in this invention by experiments in which the substrate concentration [S] is in excess over catalytic DNA molecule concentration [E]. Initial rates of reaction (v0) over a range of substrate concentrations are estimated from the initial linear phase, generally the first 5% or less of the reaction. Data points are fit by a least squares method to a theoretical line given by the equation: V=-KM (V0 /[S])+Vmax. Thus, kcat and KM are determined by the initial rate of reaction, v0, and the substrate concentration [S]. [0069] In various alternative embodiments, a catalytic DNA molecule of the present invention has an enhanced or optimized ability to cleave nucleic acid substrates with non-naturally occurring ribonucleotides. In several embodiments, the enhanced or optimized ability of a catalytic DNA molecule to cleave RNA substrates shows a regioselectivity or enantiomer of at least 50-fold. Typically, the catalytic DNA molecule exhibits a Km of less than about 1 μM, or in certain embodiments less than about 100 nM.
[0070] One skilled in the art will appreciate that the enhanced or optimized ability of a catalytic DNA molecule to cleave nucleic acid substrates may vary depending upon the selection constraints applied during the in vitro evolution procedure of the invention.
[0071] Various methods of modifying deoxyribozymes and other enzymatic
DNA molecules and nucleases of the present invention are further described in the Example hereinbelow.
[0072] The invention also provides compositions containing one or more types or populations of catalytic DNA molecules of the present invention; e.g., different types or populations may recognize and cleave different nucleotide sequences. Compositions may further include a ribonucleic acid-containing substrate. Compositions according to the present invention may further comprise lead ion, magnesium ion, or other divalent or monovalent cations, as discussed herein. [0073] Preferably, the catalytic DNA molecule is present at a concentration of about 0.05 μM to about 2 μM. Typically, the catalytic DNA molecule is present at a concentration ratio of catalytic DNA molecule to substrate of from about 1:5 to about 1:50. More typically, the catalytic DNA molecule is present in the composition at a concentration of about 0.1 μM to about 1 μM. Even more typically, compositions contain the catalytic DNA molecule at a concentration of about 0.1 μM to about 0.5 μM. The substrate maybe present in the composition at a concentration of about 0.5 μM to about 1000 μM.
[0074] Magnesium ion, lead ion, or another suitable monovalent or divalent cation, as described previously, may also be present in the composition, at a concentration ranging from about 1-100 mM. More preferably, the preselected ion is present in the composition at a concentration of about 2 mM to about 50 mM, with a concentration of about 5 mM being particularly preferred. One skilled in the art will understand that the 100 mM ion concentration is only constrained by the limits of solubility of its source (e.g. magnesium) in aqueous solution and a desire to have the catalytic DNA molecule present in the same composition in an active conformation. The optimal cationic concentration to include in the nucleic acid cleaving conditions can be easily determined by determining the amount of single-stranded nucleic acid cleaved at a given cation concentration. One skilled in the art will understand that the
optimal concentration may vary depending on the particular catalytic DNA molecule employed.
[0075] The disclosed methods allow cleavage at any nucleotide sequence by altering the nucleotide sequence of the recognition domains of the catalytic DNA molecule. This allows cleavage of single-stranded nucleic acid in the absence of a restriction endonuclease site at the selected position.
[0076] An effective amount of a catalytic DNA molecule is the amount required to cleave a predetermined base sequence present within the single-stranded nucleic acid. Typically, the catalytic DNA molecule is present at a molar ratio of
DNA molecule to substrate cleavage sites of 1 to 20. This ratio may vary depending on the length of treating and efficiency of the particular catalytic DNA molecule under the particular nucleic acid cleavage conditions employed.
[0077] Thus, in one embodiment, treating typically involves admixing, in aqueous solution, the RNA-containing substrate and the enzyme to form a cleavage admixture, and then maintaining the admixture thus formed under RNA cleaving conditions for a time period sufficient for the catalytic DNA molecule to cleave the
RNA substrate at any of the predetermined nucleotide sequences present in the RNA.
In various embodiments, a source of ions is also provided — e.g., monovalent or divalent cations, or both.
[0078] In one embodiment of the present invention, the amount of time necessary for the catalytic DNA molecule to cleave the single-stranded nucleic acid has been predetermined. The amount of time is from about 1 minute to about 24 hours and will vary depending upon the concentration of the reactants and the temperature of the reaction. Usually, this time period is from about 10 minutes to about 2 hours such that the catalytic DNA molecule cleaves the single-stranded nucleic acid at any of the predetermined nucleotide sequences present.
[0079] The present invention further contemplates that the nucleic acid cleaving conditions include a pH of about pH 6.0 to about pH 9.0. In one embodiment, the pH ranges from about pH 6.5 to pH 8.0. In another embodiment, the pH emulates physiological conditions, i.e., the pH is about 7.0-7.8, with a pH of about
7.5 being typical.
[0080] One skilled in the art will appreciate that the methods of the present invention will work over a wide pH range so long as the pH used for nucleic acid cleaving is such that the catalytic DNA molecule is able to remain in an active conformation. A catalytic DNA molecule in an active conformation is easily detected by its ability to cleave single-stranded nucleic acid at a predetermined nucleotide sequence.
[0081] hi various embodiments, the nucleic acid cleaving conditions also include a variety of temperature ranges. As noted previously, temperature ranges consistent with physiological conditions are especially preferred, although temperature ranges consistent with industrial applications are also contemplated herein. In one embodiment, the temperature ranges from about 15° C. to about 60° C. hi another variation, the nucleic acid cleaving conditions include a temperature ranging from about 30° C. to about 56° C. In yet another variation, nucleic acid cleavage conditions include a temperature from about 35° C. to about 50° C. In a preferred embodiment, nucleic acid cleavage conditions comprise a temperature range of about 37° C. to about 42° C. The temperature ranges consistent with nucleic acid cleaving conditions are constrained only by the desired cleavage rate and the stability of that particular catalytic DNA molecule at that particular temperature. [0082] The present invention also features expression vectors including a nucleic acid segment encoding a catalytic DNA molecule of the present invention situated within the vector, preferably in a manner which allows expression of that catalytic DNA molecule within a target cell (e.g., a plant or animal cell). [0083] As mentioned above, the catalytic DNA molecules of the present invention cleave a substrate nucleic acid sequence (Figures 14A and 14B (top strand)) that includes a non-naturally occurring ribonucleotide immediately upstream from a cleavage site (arrow), that is flanked by complementary sequences that bind to binding regions, also referred to herein as flanking regions, of the catalytic DNA molecule. The substrate nucleic acid sequences themselves are another aspect of the present invention. In certain preferred embodiments, the non-naturally occurring ribonucleotide of the substrate nucleic acid sequence is a 2', 5' ribonucleotide, e.g., an adenylate or a guanylate residue. In other embodiments, the non non-naturally
occurring ribonucleotide of the substrate nucleic acid sequence is a L-enantiomer of a 3', 5', ribonucleotide, most preferably an adenylate residue. [0084] In another aspect, the present invention provides a non-naturally occurring single-stranded nucleic acid substrate, as discussed above, that includes a pair of interactive labels consisting of a first label and a second label, separated from each other by a cleavage site. The labels are attached to the single-stranded nucleic acid substrate either directly or indirectly. As discussed above, the cleavage site may include a 2,,5'-linked residue or an L-enantiomer residue. The first label may be fluorescent moiety and the second label a quencher that quenches the fluorescent moiety when both the fluorescent moiety and the quencher are attached to the single- stranded nucleic acid substrate.
[0085] The substrate nucleic acid sequences and the catalytic DNA molecules of the present invention do not have counterparts in nature. They are particularly useful as biochemical tools in cleaving a reporter molecule, such as a double-labeled substrate nucleic acid sequence that contains an unnatural ribonucleotide described above. Such a reporter is not cleaved by biological nucleases. [0086] Therefore, one application of the catalytic nucleic acid molecules and substrate nucleic acid sequences of the present invention, pertains to a method for quantitative PCR, termed "DzyNA-PCR" (shown diagrammatically in figure 16; see Todd, A. V., et al, Clin. Chem. 46, 625-630 (2000), incorporated herein in its entirety by reference; and Applegate et al. Clin. Chem., 48; 13399-1488 (2002), incorporated herein in its entirety by reference). This method employs an RNA-cleaving DNA enzyme to cleave a reporter oligonucleotide that contains a fluorescent label and quencher on either side of the cleavage site. The sequence of the DNA enzyme is encoded by a complementary sequence that is attached to the 5 ' end of one of the two PCR primers. As PCR amplification proceeds, functional copies of the DNA enzyme are produced that can cleave the reporter molecule, thereby separating the fluorescent label and quencher, and generating a fluorescent signal. In existing methods because the reporter contains natural ribonucleotides, it is susceptible to cleavage by biological ribonucleases. This is not the case, however, when the method utilizes a catalytic DNA molecule of the present invention that cleaves an unnatural ribonucleotide
within the reporter molecule. There is no known biological nuclease that cleaves either a 2 ',5 '-linked guanylate or an L-ribonucleotide.
[0087] Accordingly, another aspect of the present invention provides a method that includes admixing in an amplification buffer, the following components: i) a nucleic acid sample; ii) a polymerase; iii) a substrate non-naturally occurring single-stranded nucleic acid sequence comprising a pair of interactive labels consisting of a first label and a second label being attached to the oligonucleotide directly or indirectly, wherein the first label is separated from the second label by a non-naturally occurring ribonucleotide cleavage site, as discussed above; ι iv) a forward primer capable of binding to a 3' portion of a first strand of the target nucleic acid sequence and directing polynucleotide synthesis of the target nucleic acid sequence in a forward direction; and v) a reverse primer comprising a region capable of binding to a 3' portion of a second strand of the target nucleic acid sequence and directing polynucleotide synthesis of the target nucleic acid sequence in a reverse direction, and including the complement of a catalytic single-stranded DNA molecule according to the invention wherein the catalytic single-stranded DNA molecule is capable of cleaving the substrate nucleic acid sequence at the cleavage site. [0088] The method then includes incubating the admixed components under amplification conditions to amplify the target nucleic acid sequence. This results in synthesis of the catalytic single-stranded DNA molecule. The synthesized catalytic single-stranded DNA molecule then cleaves the substrate nucleic acid sequence, thereby releasing the interaction of the first label and the second label. The first label is then detected, thereby detecting the target nucleic acid sequence. [0089] In certain embodiments of the method, the non-naturally occurring ribonucleotide cleavage site includes a 2', 5' linked residue or an L-enantiomer residue, as described above for the substrate nucleic acid sequences of the present invention. Furthermore, the catalytic single-stranded DNA molecule is typically one of the preferred catalytic single-stranded DNA molecules described above. For example, where the ribonucleotide cleavage site is a 2',5'-linked adenylate or guanylate
ribonucleotide residue, the catalytic DNA molecule may include a recognition domain and a catalytic domain capable of forming a loop that includes the nucleic acid sequence
![Figure imgf000024_0001](https://patentimages.storage.***apis.com/fd/a2/b0/de7c567f33c461/imgf000024_0001.png)
(SEQ LD NO:29). [0090] The amplification procedure used in methods of the present invention for detecting a target nucleic acid sequence is typically the polymerase chain reaction (PCR), as is well known in the art and described for example in U.S. Pat. Nos. 4,683,195, 4,683,195, and 4,800,159, incorporated herein in their entirety by reference. Where PCR is used as the amplification methodology, the polymerase is a Thermus aquaticus (Taq) DNA polymerase.
[0091] The nucleic acid sample can be virtually any nucleic acid. In some embodiments, the nucleic acid sample is isolated from a natural source. For example, the nucleic acid can be genomic DNA or RNA, such as mRNA (See e.g., Applegate et al. (2002)).
[0092] In addition to PCR, other amplification methodologies can be used in the methods of the present invention for detecting a target nucleic acid sequence, provided that the amplification method utilizes a nucleic acid molecule that includes both an amplification primer and a complement of a catalytic DNA. [0093] In one method, the PCR process is carried out as an automated process that utilizes a thermostable enzyme. In this process the reaction mixture is cycled through a denaturing step, a primer annealing step, and a synthesis step, whereby cleavage of the nucleic acid substrate occurs upon synthesis of the catalytic DNA molecule by the polymerase. Commercially available machines for performing this process from Perkin-Elmer Cetus Instruments, which is specifically designed for use with a thermostable enzyme, may be employed.
[0094] Labels and label pairs that can be included on the substrate nucleic acid sequences of the present invention for use in the methods of detecting a target nucleic acid are reported in U.S. Pat. No. 6,214,979, incorporated herein in its entirety by reference. A variety of labels which are appropriate for use in the invention, as well as methods for their inclusion in the substrate nucleic acid sequence, are known in the art and include, but are not limited to, enzymes (e.g., alkaline phosphatase and horseradish peroxidase) and enzyme substrates, radioactive atoms, fluorescent dyes, chromophores, chemiluminescent labels, electrochemiluminescent labels, such as
Origin™ (Igen), ligands having specific binding partners, or any other labels that may interact with each other to enhance, alter, or diminish a signal. [0095] The term "label" as used herein refers to any atom or molecule which can be used to provide a detectable (preferably quantifiable) signal, and which can be attached to a nucleic acid or protein. Labels may provide signals detectable by fluorescence, radioactivity, colorimetry, gravimetry, X-ray diffraction or absorption, magnetism, enzymatic activity, and the like.
[0096] Fluorophores for use as labels in constructing labeled probes of the invention include rhodamine and derivatives, such as Texas Red, fluorescein and derivatives, such as 5-bromomethyl fluorescein, Lucifer Yellow, IAEDANS, 7-Me2, N-coumarin-4-acetate, 7-OH-4-CH3 -coumarin-3-acetate, 7-NH -4CH3 -coumarin-3- acetate (AMCA), monobromobimane, pyrene frisulfonates, such as Cascade Blue, and monobromotrimethyl-ammoniobimane. In general, fluorophores with wide Stokes shifts are preferred, to allow using fluorimeters with filters rather than a monochromometer and to increase the efficiency of detection. [0097] As mentioned above, typically two interactive labels are used on a single substrate nucleic acid sequence, with due consideration given for maintaining an appropriate spacing of the labels on the substrate nucleic acid molecule to permit the separation of the labels during substrate nucleic acid sequence hydrolysis. In preferred embodiments, both a fluorophore and quenching agent are used to label the substrate nucleic acid sequence. When the substrate nucleic acid sequence is intact, the fluorescence of the fluorophore is quenched by the quencher. During the present method, the substrate nucleic acid sequence is cleaved between the fluorophore and the quencher, allowing full expression of the fluorophore fluorescence. Interaction of labels, such as quenching of fluorescence, involves transfer of energy between the first and second label, such as the fluorophore and the quencher. Therefore, the emission spectrum of the fluorophore and the absorption spectrum of the quencher must overlap. A preferred combination for this aspect of the invention is the fluorophore rhodamine 590 and the quencher crystal violet.
[0098] Detection of the hydrolyzed labeled probe can be accomplished using, for example, fluorescence polarization, a technique to differentiate between large and small molecules based on molecular tumbling. Large molecules (e.g., intact labeled
probed) tumble in solution much more slowly than small molecules. Upon linkage of a fluorescent moiety to the molecule of interest (e.g., the 5' end of a labeled probe), this fluorescent moiety can be measured (and differentiated) based on molecular rumbling, thus differentiating between intact and digested probe. In methods of the present invention that utilize the labeled nucleic acid sequence substrates in a PCR reaction, the label may be measured directly during PCR or may be performed post PCR.
[0099] Where a fluorescent label is used as one of the labels of the label pair, a number of commercial instruments can be used which are designed for analysis of fluorescently labeled materials. For instance, the ABI Gene Analyzer can be used to analyze attomole quantities of DNA tagged with fluorophores such as ROX (6carboxy-X-rhodamine), rhodamine-NHS, TAMRA (5/6-carboxytetramethyl rhodamine NHS), and FAM (5'-carboxyfluorescein NHS). These compounds can attached to the probe by an amide bond through a 5'-alkylamine on the probe. Other useful fluorophores include CNHS (7-amino-4-methyl-coumarin-3-acetic acid, succinimidyl ester), which can also be attached through an amide bond. [0100] Using commercially available phosphoramidite reagents, one can produce substrate nucleic acid sequences containing functional groups (e.g., thiols or primary amines) at either the 5' or the 3' terminus via an appropriately protected phosphoramidite, and can label them using protocols described in, for example, PCR Protocols: A Guide to Methods and Applications (Innis et al., eds. Academic Press, Inc., 1990).
[0101] The present invention also contemplates methods that utilize directed evolution to produce nucleic acid molecules having a predetermined catalytic activity for cleaving a single stranded nucleic acid that includes a non-naturally occurring ribonucleotide. For an overview of directed in vitro evolution of enzymatic DNA molecules see Example 1 of U.S. Pat. No. 6,326,174, incorporated herein in its entirety by reference.
[0102] In vitro selection and in vitro evolution techniques allow new catalysts to be isolated without a prior knowledge of their composition or structure. Such methods have been used to obtain RNA enzymes with novel catalytic properties. For example, ribozymes that undergo autolytic cleavage with lead cation have been
derived from a randomized pool of tRNAphe molecules (Pan and Uhlenbeck, Biochemistry 31: 3887-3895 (1992)). Group I ribozyme variants have been isolated that can cleave DNA (Beaudry and Joyce, Science 257: 635-641 (1992)) or that have altered metal dependence (Lehman and Joyce, Nature 361: 182-185 (1993)). Starting with a pool of random RNA sequences, molecules have been obtained that catalyze a polymerase-like reaction (Bartel and Szostak, Science 261: 1411-1418 (1993)). [0103] Accordingly, in another aspect, the present invention provides a method for identifying a catalytic DNA molecule having site-specific endonuclease activity that is specific for a non-naturally occurring ribonucleotide cleavage site. The method includes constructing a library of double stranded nucleic acid molecules that includes the non-naturally occurring ribonucleotide cleavage site and that includes a region of random-sequence nucleotides that are potentially capable of interacting with the region of the cleavage site. Next, one strand of the library of double stranded nucleic acid molecules is captured, thereby providing a library of captured single- stranded nucleic acid molecules. Next, the library of captured single-stranded nucleic acid molecules is incubated under cleavage conditions to permit cleavage at the cleavage site and release of cleaved nucleic acid molecules. Next, the cleaved nucleic acid molecules are isolated, thereby identifying a catalytic DNA molecule having site- specific endonuclease activity that is specific for a non-naturally occurring ribonucleotide cleavage site.
[0104] In certain embodiments, the method further includes amplifying the cleaved nucleic acid molecules and repeating the capture, cleavage, and isolation steps above between 1 and 50 times, typically between 1 and 20 times. The selectively amplified cleaved nucleic acid molecules are randomly mutagenized to form mutagenized cleaved nucleic acid molecules. The mutagenized cleaved nucleic acid molecules are then amplified the capture, cleavage, and isolation steps above are performed between 1 and 20 times.
[0105] In some embodiments the non-naturally occurring ribonucleotide cleavage site is a 2 ',5 '-linked guanylate ribonucleotide cleavage site or an L-adenylate ribonucleotide cleavage site.
[0106] The Example section herein illustrates a method for constructing a library of double stranded nucleic acid molecules that includes the non-naturally
occurring ribonucleotide cleavage site and that includes a region of random-sequence nucleotides that are potentially capable of interacting with the region of the cleavage site. In this non-limiting example, the double stranded nucleic acid molecules of the library can be generated by chemically synthesizing pairs of overlapping oligonucleotides. A first oligonucleotide of the pair is synthesized with the non- naturally occurring ribonucleotide cleavage site, and a second oligonucleotide of the pair can have a region of sequence complementarity with the first oligonucleotide of the pair as well as the region of random-sequence nucleotides. The first and second oligonucleotides are combined under conditions that allow binding of the complementary nucleic acid sequences. The first oligonucleotide can be extended in a primer extension reaction using a DNA-dependent DNA polymerase such as, but not limited to, reverse transcriptase to synthesize a double-stranded molecule of the library.
[0107] The polymerase chain reaction can also be used to construct the library. However, care should be taken to assure that the ribo linkage does not break during the high-temperature conditions of PCR. As an alternative method, the library can be constructed by synthesizing both strands chemically.
[0108] As shown in Figure IB, oligonucleotide pairs can be constructed such that the double stranded nucleic acid molecule formed using an oligonucleotide pair forms a hairpin loop and such that the region of random-sequence nucleotides is potentially capable of interacting with the cleavage site. In certain embodiments, the hairpin loop forms 6 nucleotides from the cleavage site. However, it is not necessary that the double stranded nucleic acid molecule forms a hairpin loop (See e.g., Breaker & Joyce, 1994, incorporated in its entirety herein by reference; See Breaker & Joyce, 1995, incorporated in its entirety herein by reference, for another example of a double stranded nucleic acid molecule that included a hairpin). Not intended to be limited by theory, it is believed however that the hairpin provides an advantage because it is believed to bias evolution toward finding answers that involve Watson-Crick recognition domains, thus making it easier to generalize the resulting enzyme to different substrate sequences.
[0109] As mentioned above, methods for identifying a catalytic DNA according to the present invention typically include mutagenizing a selectively
amplified cleaved nucleic acid molecule after at least one at least one time during repeated cycles of selection. Many methods are know in the art for mutagenizing a nucleic acid molecule. For example, a nucleic acid molecule can be mutagenized by chemical modification, incorporation of randomized mutagenic oligodeoxynucleotides, or inaccurate copying by a polymerase (See, e.g., Cadwell and Joyce, in PCR Methods and Applications 2: 28-33 (1992); Cadwell and Joyce, PCR Methods and Applications 3 (Suppl.): S136-S140 (1994); Chu, et al., Virology 98: 168 (1979); Shortle, et al., Meth. Enzymol. 100: 457 (1983); Myers, et al., Science 229: 242 (1985); Matteucci, et al., Nucleic Acids Res. 11 : 3113 (1983); Wells, et al., Gene 34: 315 (1985); McNeil, et al., Mol. Cell. Biol. 5: 3545 (1985); Hutchison, et al., PNAS USA 83: 710 (1986); Derbyshire, et al., Gene 46: 145 (1986); Zakour, et al., Nature 295: 708 (1982); Lehtovaara, et al., Protein Eng. 2: 63 (1988); Leung, et al., Technique 1: 11 (1989); Zhou, et al., Nucl. Acids Res. 19: 6052 (1991)), such as hypermutagenic PCR .
[0110] As mentioned above, the method of this aspect of the invention includes capturing one strand of the library of double stranded nucleic acid molecules. Many methods are known in the art for capturing double stranded nucleic acid molecules. The gene product can be captured or otherwise selected, for example, by its ability to bind a ligand or to carry out a chemical reaction (See, e.g., Joyce, Id. (1989); Robertson and Joyce, Nature 344: 467 (1990); Tuerk, et al., Science 249: 505 (1990)). For example, the double stranded nucleic acid molecules can include a biotin moiety such that they are captured by a Streptavadin-containing solid support. [0111] As mentioned above, the method for identifying a catalytic DNA molecule according to methods of the present invention includes amplification of nucleic acid molecules. Many methods are known in the art for amplifying nucleic acid molecules. For example, nucleic acid molecules can be amplified by a reciprocal primer method, such as the polymerase chain reaction (PCR). (See, e.g., Saiki, et al., Science 230: 1350-54 (1985); Saiki, et al., Science 239: 487-491 (1988).) Alternatively, nucleic acid amplification may be carried out using self-sustained sequence replication (3SR) (See, e.g., Guatelli, et al., PNAS USA 87: 1874 (1990), the disclosures of which are incorporated by reference herein).
[0112] As mentioned above, methods for identifying a catalytic DNA according to the present invention typically include incubating a library of single- stranded nucleic acid molecules under conditions that permit cleavage at the cleavage site. As will be apparent to a skilled artisan, conditions that permit cleavage at the cleavage site depend on the specific catalytic activity sought. For example, where the catalytic activity is the ability to cleave non-naturally-occurring ribonucleotides, conditions can include 1 hr at 37 °C with three 300-μL volumes of reaction buffer (10 mM MgCl2, 0.5 M NaCl, 50 mM EPPS (pH 7.5)), as discussed in the attached Example.
[0113] In certain embodiments, as demonstrated in the Example below, the cleavage conditions can be changed during the method such that in subsequent repeat occurrences of the cleaving step, the cleavage reaction must be more efficient for nucleic acid cleavage to occur. That is, the method of this aspect of the present invention can further include steps for obtaining enzymes with improved catalytic properties via alteration of the selection constraints during an in vitro evolution. Thus, for example as illustrated in the Example, during subsequence rounds, for example rounds 7-10, the reaction buffer can be changed to 5 mM MgCl2, 0.2 M NaCl, and 50 mM EPPS (pH 7.5), and the reaction time can be reduced, for example to 30 min for round seven, 5 min for round eight, and 1 min for rounds nine and ten. Finally, after mutations are introduced, the reaction buffer for the remaining cleavage reactions can be even more stringent, for example 5 mM MgCl2, 0.15 M NaCl, and 50 mM EPPS (pH 7.5), and the reaction time can be reduced, for example to 0.5 min for round 11 and to no more than the time required for elution for rounds 12-15. [0114] The present invention also provides kits for detecting a target nucleic acid sequence, that include the catalytic DNA molecules and/or substrates with non- naturally occurring ribonucleotides described in detail above. In kits can include primers, polymerase, and other reagents useful for the methods of the invention. [0115] . In certain embodiments the kits of the present invention for detecting a target nucleic acid sequence, includes a substrate non-naturally occurring single- stranded nucleic acid sequence as described above; a forward primer and reverse primer that includes the complement of a catalytic single-stranded DNA molecule as
discussed above, wherein the catalytic single-stranded DNA molecule is capable of cleaving the substrate nucleic acid sequence at the cleavage site.
[0116] The kits can include a substrate wherein the non-naturally occurring ribonucleotide cleavage site includes a 2',5' linked residue. In other embodiments the kit can include a substrate wherein the non-naturally occurring ribonucleotide cleavage site includes an L-enantiomer residue.
[0117] The following example is intended to illustrate but not limit the invention.
EXAMPLE 1 ISOLATION AND CHARACTERIZATION OF DNA ENZYMES THAT CLEAVE
NON-NATURALLY OCCURRING RIBONUCLEOTLDES [0118] The following example illustrates the use of directed evolution to generate and isolate DNA enzymes that cleave non-naturally occurring ribonucleotides, and the characterization of the isolated DNA enzymes. These enzymes illustrated in this example cleave a substrate nucleic acid sequence at a 2', 5' phosphodiester following a D-ribonucleotide, or a 3 ', 5 ' phosphodiester following an L-ribonucleotide.
Experimental Section
[0119] Chemical Synthesis of Oligonucleotides. β-D-3 -t-butyl- dimethylsilyl- and β-L-2'-t-butyl-dimethylsilyl-ribonucleoside phosphoramidites were obtained from ChemGenes (Ashland, MA), and all other nucleoside phosphoramidites were obtained from Glen Research (Sterling, VA). All oligonucleotides were prepared by automated synthesis using an Applied Biosystems Expedite Nucleic Acid Synthesizer. A 15-min coupling step was employed for the 3 '-t-butyl-dimethylsilyl- and L-ribonucleoside phosphoramidites. The resulting oligonucleotides were deprotected by incubation in anhydrous saturated NH3:ethanol for 36 hr at 37 °C, followed by an overnight incubation at room temperature in a solution of 1 M tetrabutylammonium fluoride in THF. All other oligonucleotides were synthesized and deprotected using standard procedures. All oligonucleotides were purified by denaturing polyacrylamide gel electrophoresis (PAGE) and desalted on a NAP-25 column (Pharmacia Biotech, Piscataway, NJ).
[0120] In Vitro Selection. Starting pools of ~1015 DNA molecules were constructed by extension of 4 nmol of 5'-biotin- dfTTTTAGAGACGATGACGATGCAXTCGGACAGTCGCGAGACTG>3 ' (SEQ LD NO:23)(primer 1; X = 2',5 '-rG or L-rA) on 6 nmol of 5 '- dfGTGCCAAGCTTACCG-Nsn-CAGTCTCGCGACTGTCCGAV3' (SEQ LD NO:24) (N = A, C, G, or T; complementary sequences underlined). The 1-mL reaction mixture contained 5 units μL"1 Superscript II reverse transcriptase (Life Technologies, Gaithersburg, MD), 3 mM MgCl2, 75 mM KC1, 50 mM tris(hydroxymethyl)- aminomethane (Tris, pH 8.3), and 0.25 mM each of dATP, dGTP, dCTP, and TTP. The extension reaction was performed by annealing the two oligonucleotides at 85 °C for 4 min, cooling to room temperature, then adding MgCl2 and reverse transcriptase and incubating at 37 °C for 1 hr. The extension products were purified by non- denaturing PAGE, eluted from the gel, precipitated with ethanol, then dissolved in a 1-mL solution containing 1-2 μM extension product, 0.5 M NaCl, 0.2 mM Na EDTA, and 50 mM N-(2-hydroxyethyl)-piperazine-N'-3-propane sulfonic acid (EPPS, pH 7.0). This material was applied to an affinity column containing 300 μL of UltraLink Immobilized Streptavidin PLUS gel (Pierce, Rockford, IL) that had previously been equilibrated with four 400-μL volumes of wash buffer (0.5 M ΝaCl, 0.1 mM Νa2EDTA, 50 mM EPPS (pH 7.0)). The column was rinsed with five 400-μL volumes of wash buffer, five 400-μL volumes of ice-cold 0.1 N NaOH/150 mM NaCl, and five 400-μL volumes of wash buffer at 37 °C, then eluted over 1 hr at 37 °C with three 300-μL volumes of reaction buffer (10 mM MgCl2, 0.5 M NaCl, 50 mM EPPS (pH 7.5)). Molecules that eluted from the column were precipitated with ethanol in the presence of 150 pmol of the primer 5'- d(TCGGACAGTCGCGAGACTG)-3 ' (SEQ ID NO:25) (primer 2) and 250 pmol of the primer 5 '-d(AACAACAACYYYGTGCCAAGCTTACCG)-3 ' (SEQ ID NO:26) (primer 3; Y = abasic nucleotide analog), then PCR amplified in a 500-μL volume. The three abasic analogs created a stop site for Taq polymerase, which caused one of the PCR product strands to be 12 nucleotides shorter than the other. [0121] The amplified products were precipitated with ethanol, and the longer of the two strands was isolated by denatoing PAGE, eluted from the gel, and again
precipitated with ethanol. One-half of the eluted DNA (~80 pmol) then was used in a template-directed extension reaction employing 200 pmol of primer 1, under the same conditions as described above. In this and all subsequent rounds of selective amplification, the extension products were immobilized on 50 μL of Streptavidin Plus gel, rinsed with five 200-μL volumes of wash buffer, five 200-μL volumes of ice-cold 0.1 N NaOH/150 mM NaCl, and five 200-μL volumes of wash buffer at 37 °C, then eluted with three 40-μL volumes of reaction buffer over 1 hr. During round two the reacted molecules additionally were selected based on their electrophoretic mobility in a denaturing polyacrylamide gel. During rounds 7-10 the reaction buffer was changed to 5 mM MgCl2, 0.2 M NaCl, and 50 mM EPPS (pH 7.5), and the reaction time was reduced to 30 min for round seven, 5 min for round eight, and 1 min for rounds nine and ten. Following round ten, random mutations were introduced by hypermutagenic PCR13. Another five rounds of selective amplification were carried out, changing the reaction buffer to 5 mM MgCl2, 0.15 M NaCl, and 50 mM EPPS (pH 7.5), and reducing the reaction time to 0.5 min for round 11 and to no more than the time required for elution for rounds 12-15.
[0122] Analysis of Individual Clones. Following the
rounds, the
DNA molecules were amplified by PCR using primer 2 and a truncated version of primer 3 having the sequence 5'-d(GTGCCAAGCTTACCG)-3' (SEQ ID NO:27). The PCR products were cloned using the TA cloning kit and LNVαF' competent cells (Invitrogen, Carlsbad, CA). individual colonies were isolated on agar plates and amplified by either colony PCR or inoculation of 2-mL cultures. The DNA was isolated and sequenced by the dideoxy chain termination method14. Cleavage assays were performed under similar conditions to those employed during in vitro selection. The reactions were quenched by the addition of an equal volume of a mixture containing 10 M urea and 50 mM Na2EDTA, and the reaction products were separated by denaturing PAGE and analyzed using a Molecular Dynamics Phosphorimager.
[0123] Kinetic Analysis. Intermolecular cleavage reactions were carried out in the presence of 25 mM MgCl2, 150 mM NaCl, and 50 mM EPPS (pH 7.5) at 37 °C. The reactions were initiated by the addition of substrate to enzyme, each contained
within a mixture corresponding to the final reaction buffer. The reaction products were separated by denaturing PAGE and analyzed using a Molecular Dynamics Phosphorimager. Nalues for &0bS were obtained under single-turnover (enzyme excess) conditions, employing various concentrations of enzyme and a trace of [5 '- P]- labeled substrate. Experimental data for the L-ribonucleotide-cleaving DΝA enzyme were fit to a single exponential equation:
Ft = F∞ (l - kob&-t) +F0 , [0124] where Ft is the fraction cleaved at time t, F∞ is the fraction cleaved at the maximum extent of the reaction, and E0 is the fraction cleaved at time zero. Experimental data for the 2 ',5 '-ribonucleotide cleaving DΝA enzymes were fit to a double-exponential equation:
Ft = Fx (\ - Q *obs1' ') + F2 (1 - e fobs2' ) + Fo , [0125] where Ft is the fraction cleaved at time t, F\ and F2 are the amplitudes of the two phases of the reaction, £0bsι and &0bs2 are the corresponding rates of each phase, and FQ is the fraction cleaved at time zero. These parameters were estimated by nonlinear regression using the Levenberg-Marquardt algorithm (Deltagraph 4.5, SPSS Science).
[0126] Values for &0bs also were obtained under multiple-turnover (substrate excess) conditions for a range of concentrations of substrate that spanned Km. Data were obtained over the first ~15% of the reaction and were fit to a line typically based on six data points. The parameters kcat and Km were obtained from a standard Michaelis-Menten saturation plot consisting of 11-21 data points, always with [S] in at least 10-fold excess over [E], and [E] at least 5-fold below the Km. The data were adjusted to take into account the maximum extent of the reaction for the L- ribonucleotide-cleaving DΝA enzymes and the amplitude of the first phase of the reaction for the 2',5'-ribonucleotide-cleaving DΝA enzyme. Standard error values were calculated using SigmaPlot (SPSS Science).
[0127] The uncatalyzed rate of cleavage was determined by incubating 1 nM
[5'-32P]-labeled substrate under standard reaction conditions. Aliquots were taken over a 5-day period and analyzed by denaturing PAGE. The value for &unCat was obtained from the slope of a best-fit line of the fraction cleaved versus time.
[0128] Metal, Temperature, and pH Dependence. All &0bs values were obtained under single-turnover conditions employing 90 nM enzyme and 1 nM [5 '- 32P]-labeled substrate, which were incubated under standard reaction conditions as described above. MgCl2 dependence was assessed over a range of MgCl concentration of 1-100 mM for the 2',5'-ribonucleotide-cleaving DNA enzyme and 0.1-30 mM for the L-ribonucleotide-cleaving DNA enzyme. Metal ion requirements were tested using 10 mM M2+, except Pb2+ which was tested at 1 mM concentration. Temperature dependence was measured over a range of 10-65 °C, employing a temperature block and heated lid to control evaporation at the elevated temperatures. The pH dependence was assessed over a range of 6.0-9.5, employing three different buffers: 2-[N-morpholino]-ethanesulfonic acid (MES) for pH 6.0-7.0, EPPS for pH 7.0-8.5, and 2-[Ν-cyclohexylamino]-ethanesulfulfonic acid (CHES) for pH 8.5-9.5. [0129] Characterization of Cleavage Products. Large-scale reactions were carried out, employing 1 nmol of either 2 ',5'- or L-ribonucleotide-containing substrate, with 1 nmol of the corresponding DNA enzyme under the standard reaction conditions described above. The reactions were quenched after 24 hr and the cleavage products were purified by denaturing PAGE. Prior to the reaction, the L- ribonucleotide-containing substrate was 5 '-phosphorylated using T4 polynucleotide kinase and ATP. This permitted separation of the two 9mer cleavage products based on their differing elecfrophoretic mobility. The gel-purified products were desalted on a Nensorb-20 column (NEN Life Sciences) and analyzed by MALDI-TOF mass spectrometry, using a PerSeptive Biosystems Noyager-STR mass spectrometer. Results
[0130] In Vitro Selection. Two separate libraries of ~1015 DΝA molecules each were constructed, one containing a single 2 ',5 '-linked guanylate ribonucleotide and the other a single L-adenylate ribonucleotide embedded within an otherwise all- DΝA sequence. The libraries were constructed by primer extension, using a 5 '- biotinylated primer that contained the unnatural ribonucleotide linkage. The primer was hybridized to a DΝA template that contained 50 random-sequence deoxynucleotides flanked by residues of defined sequence that served as primer binding sites. Similar to a strategy that has been employed previously15, a DΝA
hairpin was engineered into the pool of molecules to favor base-pairing interactions surrounding the target ribonucleotide analog (Figure IB).
[0131] Primer extension was carried out employing reverse transcriptase as a
DNA-dependent DNA polymerase to generate a double-stranded product. DNA- catalyzed cleavage could not occur during the primer extension reaction because the two strands were maintained in duplex form. The full-length, double-stranded product was purified by non-denaturing PAGE and quantified based on its UN absorbance. The purified material was immobilized on a streptavidin-containing solid support and the non-biotinylated strand was removed by brief washing with an ice-cold solution of 0.1 Ν ΝaOH. The biotinylated single-stranded molecules that remained bound to the support then were challenged to cleave the embedded ribonucleotide linkage, thereby becoming released from the support, initially, the reaction conditions were chosen to favor duplex formation, with high salt concentrations of 10 mM MgCl2 and 500 mM ΝaCl at pH 7.5 and 37 °C. The released molecules were collected and amplified by PCR, thus enriching the population with reactive molecules. [0132] A total of 15 rounds of selective amplification were performed to obtain the most active catalysts. During the first six rounds, the reaction conditions were as described above, with a reaction time of 1 hr. During rounds 7-10, the reaction conditions were changed to 5 mM MgCl2 and 200 mM ΝaCl at pH 7.5 and 37 °C. During the 7th and 8th rounds, the time allowed for the reaction was reduced to 5 min to increase the stringency of selection; during the 9th and 10th rounds the time was further reduced to 1 min. Individual molecules were cloned from the population following the 10* round, and were sequenced and tested for catalytic activity. The population then was randomly mutagenized at a frequency of -10% per nucleotide position and five additional rounds of selective amplification were carried out, employing reaction conditions of 5 mM MgCl2 and 150 mM ΝaCl at pH 7.5 and 37 °C. The amount of time allowed for the reaction was reduced from 0.5 min for round 11 to no more than the amount of time required for elution during rounds 12-15. Individuals again were cloned from the population and sequenced, revealing a high degree of sequence similarity within the formerly random-sequence region (see Supplemental Material).
[0133] Identification of Catalytic Motifs. One of the cloned individuals isolated following the 10l round of selection for 2',5'-phosphodiester-cleaving activity had especially high activity and was chosen for further study. It was designated as "2 ': 10- 16", referring to the fact that it was the 16 clone isolated following the 10 round. Analysis of its sequence suggested a plausible secondary structure, as shown in Figure 2A. The enzyme and subsfrate strands were prepared separately by extending the regions of presumed base pairing surrounding the cleavage site and repairing any base mismatches. The cloned individuals isolated following the 15 round of selection for 2 ',5 '-phosphodiester-cleaving activity had approximately the same level of activity as the 2': 10-16 clone. A high degree of sequence similarity was noted among the clones isolated following round 15 (see FIG. 9A). A representative clone, designated "2': 15-2", was chosen for further analysis. It was prepared by chemical synthesis, separating the enzyme and substrate strands so that cleavage would occur in an intermolecular reaction format (Figure 2B). As will be discussed in more detail below, the 2': 10-16 and 2': 15-2 DNA enzymes were able to cleave a separate substrate with multiple-turnover at a rate of ~0.01 min"1. [0134] In an effort to confirm the proposed secondary structure of the 2 ',5 '- phosphodiester-cleaving DNA enzymes, a variety of nucleotide substitutions and deletions were made within the central unpaired region, also referred to herein as internal bulge loops (see figure 15), most of which resulted in a complete loss of catalytic activity. The putative flanking regions on either side of the two internal bulge loops were shown to be interchangeable with any base-paired nucleotides with little or no effect on the catalytic rate. When the internal bulge loop that lies furthest from the cleavage site, also referred to herein as the upstream loop region, was replaced by a single T residue, forming a continuous stretch of base pairs downstream from the cleavage site, catalytic activity was abolished. However, the sequence of this internal bulge loop could be altered somewhat, with 3 '-AGGGATTTG-5 ' (Figure 2A), 3'-AGGG-5' (Figure 2B), 3'-AGGGATTCG-5', and 3'-AGGGATTG-5', all resulting in full catalytic activity. Finally, on the substrate molecule the unpaired G residue located immediately upstream of the cleavage site could be changed to an A with only a slight reduction in activity, but when changed to either a C or U resulted in a complete loss of activity.
[0135] The cloned individuals isolated following the 15th round of selection for L-ribonucleotide-cleaving activity were much more active compared to those isolated following the 10th round. There again was a high degree of sequence similarity among the clones isolated after the final round (see FIG. 9b). A representative clone, designated "L: 15-30" was chosen for further analysis. It was prepared by chemical synthesis, separating the enzyme and substrate strands, and extending the regions of presumed base pairing surrounding the cleavage site while repairing any base mismatches (Figure 2C). As with the 2 ',5 '-phosphodiester- cleaving DNA enzymes, the base-paired nucleotides surrounding the cleavage site could be replaced by any paired nucleotides with little or no effect on the catalytic rate. No further mutational analysis was carried out on the L-ribonucleotide-cleaving motif.
[0136] Biochemical Properties of the DNA Enzymes. The catalytic properties of all three of the above-described DNA enzymes were studied in the intermolecular reaction format (Figure 2A-C). Time-course experiments revealed that the 2': 10-16 and 2': 15-2 DNA enzymes exhibited biphasic kinetics, with a fast initial rate, followed by a slower second phase of the reaction (Figure 3). Consequently, the data were fit to a double-exponential equation and the catalytic rate constant was determined for each phase of the reaction (see Experimental Section). The L: 15-30 DNA enzyme, in contrast, exhibited monophasic kinetics that fit well to a single exponential (Figure 3).
[0137] Under multiple-turnover conditions, product release was shown to be rate limiting for the 2':10-16 and 2':15-2 DNA enzymes. This was evident by comparing the multiple-turnover reaction with a single-turnover reaction carried out under conditions of enzyme excess, the latter being about 10-fold faster. The regions of base pairing between the enzyme and substrate were shortened on either or both sides of the cleavage site to favor product release. However, product release remained rate limiting until the paired regions were made so short that catalytic activity became impaired (data not shown). The highest value for £catwas obtained when the 2': 10-16 enzyme was shortened by one base pair at each end of the enzyme-substrate complex. Under multiple-turnover conditions, this construct exhibited a kcat of 0.0036 ± 0.0001 min"1 andKm of 0.21 ± 0.03 nM (Figure 4B). By comparison, the full-length construct
exhibited a feat of 0.0022 ± 0.0001 min"1 andKm of 0.042 ± 0.008 nM (Figure 4A). Under conditions of enzyme excess, the rate of cleavage for the full-length 2 ' : 10- 16 enzyme was 0.011 ± 0.0004 min"1 (Figure 5 A), five-fold higher than that obtained under multiple-turnover conditions. The 2': 15-2 enzyme had a slightly faster rate than the 2': 10-16 enzyme under both single- and multiple-turnover conditions. The kcat of the 2': 15-2 enzyme was 0.012 ± 0.0004 min"1 and Km was 0.064 ± 0.009 nM (Figure 4C). This corresponds to a catalytic efficiency, kc Km, of ~108 M"1 min'1. Under conditions of enzyme excess, the rate of the 2': 15-2 enzyme was 0.034 ± 0.001 min"1 (Figure 5B), which is about three-fold higher than that obtained under multiple- turnover conditions.
[0138] The regions of base pairing between the L: 15-30 DNA enzyme and its substrate could be shortened so that product release was not rate limiting and without causing a reduction of the catalytic rate. Under multiple-turnover conditions, the shortened enzyme exhibited Michaelis-Menten saturation kinetics, with a &cat of 0.0012 ± 0.0001 min"1 and .Km of 2.9 ± 0.8 nM (Figure 4D). Under conditions of enzyme excess, the catalytic rate was 0.0016 ± 0.0001 min"1 (Figure 5C), which is very similar to the value for kcat obtained under multiple-turnover conditions. [0139] The effects of temperature, pH, and Mg2+ concentration on the DNA- catalyzed reaction were explored for both the 2': 10-16 and L: 15-30 DNA enzymes under single-turnover conditions. The optimal temperature for both enzymes was -42 °C, which is slightly higher than that employed during the in vitro selection process (see Supplemental Material). The optimal pH for the 2': 10-16 DNA enzyme was about 7.5, with reduced activity below pH 6.5 and above 8.5 (Figure 6A). The catalytic rate for the L: 15-30 DNA enzyme was not dependent on pH over the range of 6.0-9.0 (Figure 6B), suggesting that the rate-determining step of the reaction is not the chemical step. The catalytic rates of both the 2': 10-16 and L: 15-30 DNA enzymes were dependent on the concentration of Mg2+, exhibiting saturation behavior in both cases. The apparent i d(Mg2+) for the 2': 10-16 enzyme was -4 mM, while that for the L: 15-30 enzyme was ~0.6 mM (see Figure 10).
[0140] Several different divalent metal cations were tested for their ability to support catalysis by the 2':10-16 and L:15-30 DNA enzymes. The 2':10-16 DNA enzyme showed the highest level of activity in the presence of Mg2+, with
progressively lower activity in the presence Ca2+, Sr24*, or Ba2+, and little or no activity in the presence of Mn2+, Pb2+, Cd2+, Co2+, or Zn2+. The L: 15-30 DNA enzyme was most active in the presence of Mn2+, with progressively lower activity in the presence
9-1- 9-1- 9-1- — ._ 9-1- 9-t- I of Mg , Ca , or Pb , and little or no activity in the presence of Ba , Sr , Cd , Co2+, or Zn2+ (see Figure 12). Neither enzyme exhibited activity in the presence of Co(NH3)6 (data not shown).
[0141] The cleavage products resulting from the reaction with both the 2 ' : 10-
16 and L: 15-30 DNA enzymes were analyzed by high-resolution PAGE and MALDI mass spectrometry. In both cases the 5 '-cleavage product was an oligonucleotide of the expected length, terminating in either a 2 ',3 '-cyclic phosphate or a 2'- or 3 '- monophosphate (see Figure 7 and Figure 13). The 3 '-cleavage product also was of the expected length and terminated in a free 5 '-hydroxyl, as confirmed by MALDI mass spectrometry (see Figure 13).
[0142] The regio- and enantiospecificity of the 2 ',5 '-phosphodiester-cleaving and L-ribonucleotide-cleaving DNA enzymes, respectively, were determined under single-turnover conditions, comparing substrates that contained either the unnatural or a natural ribonucleotide at the cleavage site (compare Figures 3 and 8). The uncatalyzed rate of cleavage also was measured for the various substrates (see Figure 14). The 2':15-2 DNA enzyme exhibited a kcat/kunc_t of -20,000 for the substrate containing a 2 ',5 '-linked ribonucleotide and a kc_t/kuncat of 3.3 for the corresponding substrate containing a 3 ',5 '-linked ribonucleotide, reflecting a regioselectivity of about 6,000-fold in favor of the unnatural substrate. The 2': 10-16 DNA enzyme exhibited a regioselectivity of about 2,000-fold. The L: 15-30 DNA enzyme was less selective than the 2 ',5 '-cleaving DNA enzymes. For the substrate containing an L- ribonucleotide, kcat/kuncat was -500, while for the corresponding substrate containing a D-ribonucleotide, feat/femcat was -13. This corresponds to an enantioselectivity of about 40-fold in favor of the unnatural substrate. The regio- or enantioselectivity of the 2', 5 '-phosphodiester- or L-ribonucleotide-cleaving DNA enzymes, respectively, is demonstrated in the autoradiogram shown in Figure 7.
Discussion
[0143] The substrate specificity of an enzyme is determined by its ability to discriminate both at the step of substrate binding and at the chemical step of the reaction. For small-molecule substrates it generally is more difficult to achieve a high degree of discrimination compared to macromolecular substrates because of the smaller number of potential interactions between the enzyme and small molecule. The interaction between two nucleic acid molecules can be highly specific, based on sequence recognition involving Watson-Crick base pairing as well as non-standard pairing interactions. The ability of nucleic acid molecules to distinguish one another based on their regio- or enantioisomeric composition also has been explored. For example, 2 ',5 '-linked RNA is able to form stable duplexes with either 2 ',5 '- or 3 ',5 '- linked RNA, but not with 3 ',5 '-linked DNA16. Similarly, 2 ',5 '-linked DNA can form stable duplexes with 3 ',5 '-linked RNA, but not with 3 ',5 '-linked DNA17. An all-L- oligodeoxynucleotide composed of six adenylate residues was shown to pair with a complementary all-D-RNA strand, but not with the corresponding all-D-DNA18. Another study, however, reported that all-L-DNA is unable to form duplexes with either all-D-RNA or all-D-DNA19. In the present study the substrates contained a single unnatural ribonucleotide, embedded within an otherwise all-natural DNA molecule, posing a more difficult challenge for either regio- or enantiospecific recognition.
[0144] The DNA enzymes that were disclosed in this Example are highly specific for substrates that contain a single unnatural ribonucleotide. One enzyme was able to distinguish between a 2 ',5'- and 3 ',5 '-linked residue with a regiospecificity of 6,000-fold. Another could distinguish between an L- and D- residue with an enantioselectivity of 40-fold. The catalytic rate of the 2 ',5 '-phosphodiester-cleaving DNA enzyme was -0.01 min"1, and that of the L-ribonucleotide-cleaving DNA enzyme was about ten-fold slower. These rates are significantly slower than the rate of other reported RNA-cleaving DNA enzymes that cleave natural ribonucleotides20- 3
. The "10-23" DNA enzyme, for example, can achieve a catalytic rate of up to 10 min"1 under optimal reaction conditions24. Perhaps the DNA enzymes that cleave unnatural ribonucleotides have difficulty folding into an active conformation or
positioning a divalent metal cation to assist in the cleavage of the target phosphodiester. By analogy with known ribonucleases, the mechanism of cleavage likely involves deprotonation of the free 2'- or 3 '-hydroxyl followed by attack of the resulting oxyanion on the adjacent phosphate. An in-line orientation is required for this attack, which can be achieved by forcing the nucleotide that precedes the cleavage site into an extraheh •cal position 9^ . A single unpaired purine nucleotide within an otherwise complete duplex structure is especially amenable to achieving this orientation through local conformational changes, primarily involving the ε and ζ
9fi backbone torsion angles . It may be more difficult for a 2 ',5 '-linked D- ribonucleoti.de, and especially a 3 ',5 '-linked L-ribonucleotide, to achieve the required orientation within the context of a Watson-Crick duplex. Thus, additional catalytic assistance may be required to bring about the cleavage of these unnatural ribonucleotides.
[0145] The uncatalyzed rate of cleavage of a 3 ',5 '-phosphodiester of RNA has been measured for both a single ribonucleotide embedded within an otherwise all- DNA molecule27 and for an all-RNA oligomer28. The uncatalyzed rate of hydrolysis for the 2 ',5 '-phosphodiester of RNA, either in the presence or absence of a divalent metal cation, is similar to that of a 3 ',5 '-phosphodiester12'29-32, except when the RNA is bound to a complementary strand. In that case, the 2 ',5 '-linkage is about seven-fold more labile, whereas the 3 ',5 '-linkage is about five-fold more stable . In order to determine whether this "duplex effect" was partially responsible for the catalytic rate enhancement observed with the 2',5 '-cleaving DNA enzyme, the substrate containing the 2',5 '-linked ribonucleotide was hybridized to a complementary DNA strand and its hydrolysis rate in that context was compared to the hydrolysis rate of the substrate alone. No difference was seen in the uncatalyzed rate of cleavage under these two conditions (data not shown). The uncatalyzed rate of cleavage measured for a single embedded ribonucleotide was similar to that reported previously14,28. [0146] The DNA enzymes developed in this study do not have any counterpart in nature and would not be able to function with any known biological substrate. They could, however, be useful as biochemical tools in cleaving a reporter molecule that contains an unnatural ribonucleotide. Such a reporter would not be cleaved by biological nucleases. One potential application of this activity, as
described in more detail hereinabove, pertains to a method for quantitative PCR, termed "DzyNA-PCR"34. This method employs an RNA-cleaving DNA enzyme to cleave a reporter oligonucleotide that contains a fluorescent label and quencher on either side of the cleavage site. The sequence of the DNA enzyme is encoded by a complementary sequence that is attached to the 5' end of one of the two PCR primers. As PCR amplification proceeds, functional copies of the DNA enzyme are produced. These can cleave the reporter molecule, separating the fluorescent label and quencher, generating a fluorescent signal. In existing methods because the reporter contains natural ribonucleotides, it is susceptible to cleavage by biological ribonucleases. This would not be the case, however, if one employed a DNA enzyme that cleaves an unnatural ribonucleotide within the reporter molecule. There are naturally-occurring ribonucleases that cleave 2 ',5 '-linked oligoadenylates35 — these molecules are generated as part of the interferon response pathway. However, there is no known biological nuclease that cleaves either a 2 ',5 '-linked guanylate or an L-ribonucleotide. [0147] hi addition to their potential application as biochemical tools, the DNA enzymes described here illustrate that nucleic acid enzymes can exhibit substrate regio- and enantioselectivity comparable to that of their natural protein counterparts. Snake venom phosphodiesterase I, for example, can cleave either a D- or L- ribonucleotide, but is 1, 800-fold more active in cleaving the natural D-RNA substrate36. Further in vitro evolution experiments, especially those employing functionally enhanced nucleic acid analogs37'38, may lead to the development of novel catalysts with even greater regio- or enantioselectivity.
References
(1) Joyce, G. F.; Nisser, G. M.; van Boeckel, C. A. A.; van Boom, J. H.; Orgel, L. E.; van Westrenen, J. Nature 1984, 310, 602-604.
(2) Kozlov, I. A.; Pitsch, S.; Orgel, L. E. Proc. Natl. Acad. Sci. USA 1998, 95, 13448-13452.
(3) Saghatelian, A.; Yokobayashi, Y.; Soltani, K.; Ghadiri, M. R. Nature 2001, 409, 797-801.
(4) Breaker, R. R. Chem. Rev. 1997, 97, 371-390.
(5) Jenne, A.; Famulok, M. Topics Curr. Chem. 1999, 202, 101-131.
(6) Petrounia, I. P.; Arnold, F. H. Curr. Opin. Biotechnol. 2000, 11, 325- 330.
(7) Kolkman, J. A.; Stemmer, W. P. C. Nature Biotechnol. 2001, 19, 423- 428.
(8) Reetz, M. T. Angew. Chem. Int. Ed. Engl. 2001, 40, 284-310.
(9) Seelig, B.; Keiper, S.; Stuhlmann, F.; Jaschke, A. Angew. Chem. Int. Ed. Engl. 2000, 39, 4516-4579.
(10) Bartel, D. P.; Szostak, J. W. Science 1993, 261, 1411-1417.
(11) Ekland, E. H.; Szostak, J. W.; Bartel, D. P. Science 1995, 269, 364- 370.
(12) Shih, I.-H.; Been, M. D. RNA 1999, 5, 1140-1148.
(13) Nartanian, J.-P.; Henry, M.; Wain-Hobson, S. Nucleic Acids Res. 1996, 24, 2627-2631.
(14) Sanger, F.; Νickelson, S.; Coulson, A. R. Proc. Natl. Acad. Sci. USA 1977, 74, 5463-5467.
(15) Breaker, R. R; Joyce, G. F. Chem. Biol. 1995, 2, 655-660.
(16) Wasner, M.; Arion, D.; Borkow, G; Νoronha, A.; Uddin, A. H.; Pamiak, M. A.; Damha, M. J. Biochemistry 1998, 37, 7478-7486.
(17) Alul, R; Hoke, G. D. Antisense Res. Dev. 1995, 5, 3-11.
(18) Fujimori, S.; Shudo, K. J. Am. Chem. Soc. 1990, 112, 7436-7438.
(19) Garbesi, A.; Capobianco, M. L.; Colonna, F. P.; Tondelli, L.; Arcamone, F.; Manzini, G.; Hilbers, C. W.; Aelen, J. M. E.; Blommers, M. J. J. Nucleic Acids Res. 1993, 21, 4159^1165.
(20) Santoro, S.; Joyce, G. F. Proc. Natl. Acad. Sci. USA 1997, 94, 4262- 4266.
(21) Sugimoto, N.; Okumoto, Y. Nucleic Acids Symp. Series 1999, 42, 281- 282.
(22) Li, J.; Zheng, W.; Kwon, A. H.; Lu, Y. Nucleic Acids Res. 2000, 28, 481^188.
(23) Feldman, A. R; Sen, D. J. Mol. Biol. 2001, 313, 283-294.
(24) Santoro, S.; Joyce, G. F. Biochemistry 1998, 37, 13330-13342.
(25) Husken, D.; Goodall, G.; Blommers, M. J. J.; Jahnke, W.; Hall, J.; Haner, R.; Moser, H. E. Biochemistry 1996, 35, 16591-16600.
(26) Tereshko, V.; Wallace, S. T.; Usman, N.; Wincott, F. E.; Egli, M. SNA 2001, 7, 405-420.
(27) Jenkins, L. A.; Autry, M. E.; Bashkin, J. K. J. Am. Chem. Soc. 1996, 118, 6822-6825.
(28) Li, Y.; Breaker, R. R. J. Am. Chem. Soc. 1999, 121, 5364-5372.
(29) Oivanen, M.; Kuusela, S.; Lδnnberg, H. Chem. Rev. 1998, 98, 961- 990.
(30) Oivanen, M.; Schnell, R.; Pfleiderer, W.; Lδnnberg, H. J. Org. Chem. 1991, 56, 3623-3628.
(31) Jarvinen, P.; Oivanen, M.; Lδnnberg, H. J. Org. Chem. 1991, 56, 5396-5401.
(32) Anslyn, E.; Breslow, R. J. Am. Chem. Soc. 1989, 111, 4473^1482.
(33) Usher, D. A.; McHale, A. H. Proc Natl. Acad. Sci. USA 1976, 73, 1149-1153.
(34) Todd, A. N.; Fuery, C. J.; Impey, H. L.; Applegate, T. L.; Haughton, M. A. Clin. Chem. 2000, 46, 625-630.
(35) Nilsen, T. W.; Wood, D. L.; Baglioni, C. J. Biol. Chem. 1982, 257, 1602-1605.
(36) Moyroud, E.; Biala, E.; Strazewski, P. Tetrahedron 2000, 56, 1475- 1484.
(37) Santoro, S.; Joyce, G. F.; Sakthivel, K.; Gramatikova, S.; Barbas, C. F. J. Am. Chem. Soc. 2000, 122, 2433-2439.
(38) Perrin, D. M.; Garestier, T.; Helene, C. J. Am. Chem. Soc. 2001, 123, 1556-1563.
[0148] Although the invention has been described with reference to the above examples, it will be understood that modifications and variations are encompassed within the spirit and scope of the invention. Accordingly, the invention is limited only by the following claims.