AU2008200974B2 - Peptide acceptor ligation methods - Google Patents

Abstract

Described herein are methods and reagents for the ligation of a peptide acceptor to an RNA, as well as the RNA-peptide acceptor products.

Description

AUSTRALIA Patents Act 1990 COMPLETE SPECIFICATION Standard Patent Applicant: Adnexus Therapeutics, Inc. Invention Title: PEPTIDE ACCEPTOR LIGATION METHODS The following statement is a full description of this invention, including the best method for performing it known to us: PEPTIDE ACCEPTOR LIGATION METHODS Background of the Invention In general, the present invention relates to ligation methods, in particular, for joining 5 peptide acceptors to nucleic acids. Methods currently exist for the preparation of RNA-protein fusions. An RNA-protein fusion is created by attaching a peptide acceptor to the 3' end of an RNA molecule, followed by in vitro or in situ translation of the RNA. The product is a peptide attached to the 3' end of the RNA encoding it. The generation of these RNA-protein fusions facilitates the isolation of o proteins with desired properties from large pools of partially or completely random amino acid sequences, and solves the problem of recovering and amplifying protein sequence information by covalently attaching the RNA coding sequence to its corresponding protein molecule. All references, including any patents or patent applications, cited in this specification s are hereby incorporated by reference. No admission is made that any reference constitutes prior art. The discussion of the references states what their authors assert, and the applicants reserve the right to challenge the accuracy and pertinency of the cited documents. It will be clearly understood that, although a number of prior art publications are referred to herein, this reference does not constitute an admission that any of these documents forms part of the o common general knowledge in the art, in Australia or in any other country. Summary of the Invention A first aspect provides an RNA molecule crosslinked to a peptide acceptor through a psoralen moiety. 2s A second aspect provides an RNA molecule chemically ligated to a peptide acceptor, wherein said RNA molecule is crosslinked to said peptide acceptor and said RNA molecule contains a psoralen photocrosslink target site. The present invention features methods for the attachment of a peptide acceptor to an RNA molecule as well as the RNA-peptide acceptor products. These methods facilitate the 30 production of RNA-protein fusions which can be used, for example, for the isolation of proteins or nucleic acids with desired properties from large pools of partially or completely random amino acid or nucleic acid sequences. This inventive method may be carried out by a variety of strategies for affixing a peptide acceptor to a nucleic acid molecule. These various approaches differ from one another in the types of bonds formed by the attachment of 3s the peptide to the nucleic acid, and in the reagents used to achieve the attachment. Disclosed herein is a method for affixing a peptide acceptor to an RNA molecule involving providing an RNA molecule having a 3' sequence which forms a hairpin structure, -2 3159724_1 (GHMatters) P44319.AU.2 21-Feb-12 providing a peptide acceptor covalently bonded to a nucleic acid linker molecule, and hybridizing the RNA molecule to the nucleic acid linker molecule under conditions which allow covalent bond formation to occur between the peptide acceptor and the RNA molecule. Also disclosed is a method for affixing a peptide acceptor to an RNA molecule s involving providing a peptide acceptor having a linker with a 5' sequence that forms a hairpin, hybridizing the peptide acceptor to the RNA molecule, and covalently bonding the peptide acceptor to the RNA. In one embodiment of the above aspects of the invention, the peptide acceptor is bonded to the RNA molecule using T4 DNA ligase. Also disclosed is a method for attaching a peptide acceptor to an RNA molecule, by .o providing an RNA molecule and a peptide acceptor covalently bonded to a linker molecule, where the linker molecule initiates with a deoxynucleotide triphosphate or dideoxynucleotide triphosphate, and contacting the RNA molecule and peptide acceptor with terminal deoxynucleotidyl transferase to covalently bond the peptide acceptor to the RNA molecule. Also disclosed is a method of affixing a peptide acceptor to an RNA molecule by .5 chemically ligating the RNA molecule to the peptide acceptor. In one embodiment, the peptide acceptor is joined to a psoralen moiety and crosslinked to the RNA molecule via the psoralen moiety. The psoralen moiety may be attached to either the 5' or 3' end of a linker molecule which is itself attached to the peptide acceptor, or the psoralen moiety may be located at an internal position of the linker molecule. 0 According to this technique, the peptide acceptor is crosslinked to the RNA molecule using UV irradiation. In further embodiments, the psoralen is attached to the peptide acceptor through a C6 alkyl chain and/or the RNA molecule contains a stop codon positioned proximal to its 3' end. Preferably, the linker is between 25 and 40 nucleotide units in length. In addition, prior to crosslinking the peptide acceptor to the RNA molecule, the RNA may be 25 hybridized to a linker that further includes a photocleavable moiety. The hybridized RNA may then be immobilized to a solid support through the photocleavable moiety. Preferably, the photocleavable moiety is biotin. In another embodiment, the RNA molecule is functionalized and is attached to a peptide that has been suitably modified to permit chemical bond formation between the 30 peptide acceptor and the RNA molecule. Preferably, the RNA molecule is functionalized through 104 oxidation. The peptide acceptor may be functionalized by attaching a molecule to the peptide acceptor chosen from the group consisting of amines, hydrazines, (thio)hydrazides, and (thio)semicarbazones. In yet another embodiment, the chemical ligation is carried out in the absence of an 35 external template. Alternatively, the chemical ligation reaction can be carried out in the presence of an external template. This second method involves aligning the RNA molecule and the linker portion of a peptide acceptor using a template, so that the 5' end of the template -3 3159724_1 (GHMalters) P44319.AU.2 21.Feb-12 hybridizes to the linker portion of the peptide acceptor and the 3' end of the template hybridizes to the RNA molecule. The chemical ligation of an RNA molecule to a peptide acceptor can also be carried out in the absence of an external template by hybridizing the linker molecule itself, which is covalently bonded to the peptide acceptor, to the RNA 5 molecule. This hybridization brings the peptide acceptor and RNA molecule into close proximity for ligation. Preferably, the functional group is at the 5' end of the linker region of the peptide acceptor, or is flanked by a hybridization domain on one side and the peptide acceptor on the other side. In a further embodiment, the chemical ligation of the peptide acceptor to the RNA .0 molecule involves attaching a functional group to the RNA molecule through reductive arnination of the RNA, followed by modification of the peptide acceptor to react with the RNA molecule. The two molecules are then joined through formation of a covalent bond. Preferably, the functional group attached to the RNA molecule is a thiol, maleimide, or amine. 5 Also disclosed is a method for attaching a peptide acceptor to an RNA molecule through a non-covalent bond. In one embodiment, the attachment is achieved by covalently bonding a peptide nucleic acid (PNA) to the peptide acceptor and non-covalently bonding the peptide acceptor to the RNA molecule through the PNA. In this embodiment, the RNA molecule may contain a stop codon. o Also disclosed are RNA molecules chemically or non-covalently ligated to peptide acceptors as well as the nucleic acid-protein fusions generated by transcription and translation (and, if desired, reverse transcription and/or amplification) of these RNA molecules. In one embodiment, the peptide acceptor is ligated at the 3' end of the RNA molecule. 2s Also disclosed are methods for the selection of a desired protein or nucleic acid using the RNA-peptide acceptor molecules of the invention. The selection techniques utilize the present molecules for RNA-protein fusion formation, and subsequent selection of proteins or nucleic acids of interest. The selection methods may be carried out by any of the approaches described, for example, in Szostak et al., WO 98/31700, and Szostak et al, U.S.S.N. o 09/247,190, hereby incorporated by reference. Also disclosed is a method of generating an RNA-protein fusion. This method involves providing an RNA molecule hybridized to a linker, where the linker contains a photocleavable moiety, a psoralen moiety, and a peptide acceptor; immobilizing the RNA to a solid support under conditions in which non-immobilized RNA are substantially removed 5 from the support; crosslinking the peptide acceptor to the RNA, through the psoralen moiety, whereby this crosslinking simultaneously releases the crosslinked RNA from the solid support; and translating the crosslinked RNA to form an RNA-fusion protein. In one -4 3159724_1 (GHMatters) P44319.AU.2 21-Feb-12 embodiment, the photocleavable moiety is biotin. The RNA molecule may include a translation initiation sequence and a start codon operably linked to a candidate protein coding sequence. In addition, one preferred peptide acceptor is puromycin, a nucleoside analog which adds to the C-terminus of a growing 5 peptide chain and terminates translation. In one embodiment, the peptide acceptor includes puromycin attached to a linker, for example, a nucleotide linker. This linker facilitates the alignment of the peptide acceptor to the RNA molecule for attachment. In a further embodiment, the linker region of the peptide acceptor includes non-nucleotide moieties, for example, PEG. Other possible choices for acceptors include tRNA-like structures at the 3' .0 end of the RNA, as well as other compounds that act in a manner similar to puromycin. Such compounds include, without limitation, any compound which possesses an amino acid linked to an adenine or an adenine-like compound, such as the amino acid nucleotides, phenylalanyl-adenosine (A-Phe), tyrosyl adenosine (A-Tyr), and alanyl adenosine (A-Ala), as well as amide-linked structures, such as phenylalanyl 3' deoxy 3' amino adenosine, alanyl 3' .5 deoxy 3' amino adenosine, and tyrosyl 3' deoxy 3' amino adenosine; in any of these compounds, any of the naturally-occurring L-amino acids or their analogs may be utilized. In addition, a combined tRNA-like 3' structure-puromycin conjugate may also be used in the invention. Also disclosed is a DNA sequence that is included between the end of the message 0 and the peptide acceptor. This sequence is designed to cause the ribosome to pause at the end of the open reading frame, providing additional time for the peptide acceptor (for example, puromycin) to accept the nascent peptide chain before hydrolysis of the peptidyl-tRNA linkage. During in vitro translation the ribosome may also pause at the site of chemical ligation, especially at a psoralen crosslinking site or at a PNA clamp. 25 Predominantly non-nucleotide linker moieties may be used in place of the nucleotide linkers attached to the peptide acceptor. This design facilitates the ligation of a peptide acceptor to an RNA molecule. For example, the linker may contain triethylene glycol spacers. The linker may also contain 2'-OMe-RNA phosphoramidites. In some cases where hybridization is a prerequisite for chemical or enzymatic ligation, a sufficient portion of the 30 linker next to the ligation site must be comprised of nucleic acids. Furthermore, the RNA or linker of the invention may contain a sequence (e.g., a poly(A) sequence) for use in purification, for example, affinity purification of the RNA or an RNA-protein fusion molecule formed from such an RNA or linker. The RNA molecule affixed to a peptide acceptor may be in vitro or in situ translated 35 to produce an RNA-protein fusion molecule. The RNA-protein fusion molecule is then incubated in the presence of high salt and/or incubated at low temperature (e.g., overnight at -20*C) as described by Szostak et al. (09/247,190). The RNA-protein fusion molecule may -5 3159724_1 (GHMatters) P44319.AU 2 21-Feb-12 be also purified, for example, using standard poly(A) purification techniques. In the claims which follow and in the description of the invention, except where the context requires otherwise due to express language or necessary implication, the word "comprise" or variations such as "comprises" or "comprising" is used in an inclusive sense, 5 i.e. to specify the presence of the stated features but not to preclude the presence or addition of further features in various embodiments of the invention. As used herein, by a "protein" is meant any two or more naturally occurring or modified amino acids joined by one or more peptide bonds. "Protein" and "peptide" are used interchangeably. LO By an "RNA" is meant a sequence of two or more covalently bonded, naturally occurring or modified ribonucleotides. One example of a modified RNA included within this term is phosphorothioate RNA. By a "translation initiation sequence" is meant any sequence that is capable of providing a functional ribosome entry site. In bacterial systems, this region is sometimes .5 referred to as a Shine-Dalgarno sequence. By a "start codon" is meant three bases which signal the beginning of a protein coding sequence. By a "stop codon" is meant three bases which signal the termination of a protein coding sequence. Generally, start codons are AUG (or ATG) and stop codons are UAA (or TAA), UAG (or TAG), or UGA (or TGA); however, any other base triplets capable of being o utilized as start or stop codons may be substituted. By "covalently bonded" is meant joined together either directly through a covalent bond or indirectly through another covalently bonded sequence (for example, DNA corresponding to a pause site). -6 3159724_1 (GHMauers) P44319AU. 21-Feb-12 THIS PAGE HAS BEEN LEFT INTENTIONALLY BLANK -7 3159724_1 (GHMatters) P44319.AU.2 21-Feb-12 By "non-covalently bonded" is meant joined together by means other than a covalent bond. By a "hairpin structure" is meant a double-stranded region formed by a single nucleic acid strand. Preferably, such hairpin structures are at least 8 5 base pairs in length, and more preferably, between 8 and 15 base pairs in length. By "chemically ligating" is meant the joining together of two molecules without the use of an enzyme. Chemical ligation can result in non covalent as well as covalent bonds. 10 By a "peptide acceptor" is meant any molecule capable of being added to the C-terminus of a growing protein chain by the catalytic activity of the ribosomal peptidyl transferase function. Typically, such molecules contain (i) a nucleotide or nucleotide-like moiety, for example adenosine or an adenosine analog (di-methylation at the N-6 amino position is acceptable), (ii) 15 an amino acid or amino acid-like moiety, such as any of the 20 D- or L-amino acids or any amino acid analog thereof including O-methyl tyrosine or any of the analogs described by Ellman et al. (Meth. Enzymol. 202:301, 1991), and (iii) a linkage between the two (for example, an ester, aide, or ketone linkage at the 3' position or, less preferably, the, 2' position). Preferably, this linkage 20 does not significantly perturb the pucker of the ring from the natural ribonucleotide conformation. Peptide acceptors may also possess a nucleophile, which may be, without limitation, an amino group, a hydroxyl group, or a sulfhydryl group. In addition, peptide acceptors may be composed of nucleotide mimetics, amino acid mimetics, or mimetics of the combined 25 nucleotide-ari no acid structure. By a "linker" or "linker molecule" is meant a sequence that includes deoxyribonucleotides, ribonucleotides, or analogs thereof. -8- By "functionalize" is meant to modify in a manner that results in the attachment of a functional group or moiety. For example, an RNA molecule may be functionalized through 104 oxidation or amination, or a peptide acceptor may be functionalized by attaching an amine, hydrazine, 5 (thio)hydrazide, or (thio)semicarbazone group. By an "external template," is meant a nucleic acid sequence which is added to a ligation reaction mixture, but which is not a part of the final product of the ligation reaction. By "high salt" is meant having a concentration of a monovalent 10 cation of at least 200 mM, and, preferably, at least 500 mM or even 1 M, and/or a concentration of a divalent or higher valence cation of at least 25 mM, preferably, at least 50 mM, and, most preferably, at least 100 mM. By "affinity purification sequence" is meant a nucleotide sequence that is utilized in the purification of a nucleic acid or a nucleic acid-protein 15 fusion molecule. For example, an affinity purification sequence may be a poly(A) sequence, such as A 8 2 0 , which can be used for purification of nucleic acid or fusion molecules on oligo-dT cellulose. An affinity purification sequence may also be a polypeptide sequence that is used to purify a nucleic acid-protein fusion molecule. Other exemplary purification techniques are 20 described by Szostak et al. U.S.S.N. 09/247,190. The present invention provides a number of advantages. For example, the methods described herein facilitate the efficient ligation of peptide acceptors to RNA molecules, in some aspects, without the need for an external template to bring the RNA and peptide acceptor together. The 25 invention also reduces the cost associated with the generation of an RNA protein fusion. -9- Other features and advantages of the invention will be apparent from the following detailed description, and from the claims. Detailed Description The drawings will first briefly be described. 5 Brief Description of the Drawings FIGURE 1 is a schematic representation of exemplary steps involved in the ligation of a peptide acceptor linker to an RNA using T4 DNA ligase on a hairpin template. Either the RNA 3' sequence or the peptide acceptor linker is designed to form a hairpin structure, and the peptide acceptor linker 10 hybridizes to the RNA, thus bringing the RNA and peptide acceptor into close proximity to each other. FIGURE 2 is a schematic representation of exemplary steps involved in the ligation of a peptide acceptor linker to an RNA using terminal deoxynucleotidyl transferase. 15 FIGURE 3A is a schematic representation of exemplary steps involved in the 3' modification of an RNA through hydrazide and semicarbazone formation. FIGURE 3B is a schematic representation of exemplary steps involved in the 3' modification of an RNA through reductive amination. 20 FIGURES 4A-4D are a series of diagrams showing exemplary strategies for the chemical ligation of a functionalized puromycin linker to a functionalized RNA. One strategy is template-independent (Figure 4A). Other strategies involve using external oligo templates to align the RNA and puromycin linker. In the strategies involving external oligo templates, the 25 oligo template may hybridize to both the RNA and puromycin linker (Figure -10- 4B), or be attached to the puromycin linker and hybridize to the RNA (Figure 4D). In addition, the functional group may be at the 5' end of (Figures 4B and 4C) or internal to (Figure 4D) the puromycin linker. FIGURE 5 is a schematic representation of exemplary steps involved 5 in the synthesis of a fully protected carbohydrazide phosphoramidite for use in a modified puromycin linker. FIGURE 6 is a schematic representation of exemplary steps involved in the ligation of a peptide acceptor linker to an RNA by attaching a functional group to the 3' end of the RNA followed by chemical ligation. The functional 10 group can then react with a suitably modified linker molecule to covalently bond the RNA and puromycin linker, through a thiol (A), maleimide (B), or amine group (C). FIGURES 7A-7C are a series of schematic representations of exemplary steps involved in the ligation of a peptide acceptor to an RNA using 15 photocrosslinking. In this general method, a psoralen moiety attached to the puromycin linker crosslinks the puromycin linker to the RNA upon exposure to UV irradiation. The psoralen moiety can be at the 5' end of (Figure 7A), internal to (Figure 7B), or at the 3' end of (Figure 7C) the puromycin linker. FIGURE 7D is a schematic representation of mRNAs used for 20 photo-crosslink formation. FIGURE 7E is a schematic representation of a duplex formed between the constant 3-sequence of the mRNA (const. region of the mRNAs of Figure 7D)) and the photolinker showing the putative psoralen intercalation site. Also shown is the structure of psoralen, which was attached through a C6 25 alkyl chain to the 5'-phosphate of the linker. The variables x and y determine the number of dA-nucleotides and triethyleneglycol-units (TEG) in the linker, respectively. The autoradiograph on the left depicts the gel-electrophoretic -11analysis of the photo-crosslinking reaction between mRNA 1 and linker B. FIGURE 7F is an image of a gel illustrating in vitro translation and fusion formation. Protein synthesis from mRNA template 2 and template 3 is shown in lanes I and 3, respectively. Translation of mRNA templates 5 photo-crosslinked to linker B produced mRNA-protein fusions (lanes 2 and 4). Template 3 carries a stop codon following the linker crosslink-site. Lane 5 shows the fusion product after purification on oligo-dT cellulose. A schematic representation of fusion formation using an mRNA molecule containing a stop codon is depicted above the image of the gel. 10 FIGURE 7G is an image of a gel illustrating the dependence of mRNA-protein fusion molecule yield on linker composition and on salt concentration. FIGURE 7H is a schematic representation of the steps involved in the ligation of mRNA to photocleavable biotin-based hybridized linkers and 15 their subsequent fusion formation. FIGURE 8A is a schematic representation of the attachment of a puromycin linker to an RNA through (PNA)2-RNA triplex formation. The puromycin linker is attached to two PNA molecules which bind to RNA forming a triple helix through strong non-covalent bonds. 20 FIGURE 8B is a schematic representation of steps involved in the synthesis of PNA-linker conjugates. The linker is attached to a solid support and modified to attach to the PNA. The PNA is then coupled to the desired linker, and the PNA linker molecule is deprotected and separated from the solid support. 25 Described herein are various methods of attaching a peptide acceptor to an RNA molecule. The RNA may be generated by any standard approach, including normal cellular synthesis, recombinant techniques, and chemical - 12 synthesis, and includes, without limitation, cellular RNA, mRNA libraries, and random synthetic RNA libraries. The peptide acceptor (for example, puromycin) is typically bonded to a DNA or RNA linker. Such peptide acceptor molecules may be generated by any standard technique, for example, 5 the techniques described in Roberts and Szostak (Proc. Natl. Acad. Sci. USA 94:12297, 1997), Szostak et al. (WO 98/31700), and Szostak et al., U.S.S.N. 09/247,190. Techniques for carrying out each method of the invention are now described in detail, using particular examples. These examples are provided for the purpose of illustrating the invention, and should not be 10 construed as limiting. Example 1: Enzymatic Ligation Methods Involving T4 DNA Ligase on a Hairpin Template In one particular approach according to the invention, T4 DNA ligase is used to attach a peptide acceptor to an RNA molecule using a hairpin 15 containing template, for example, as shown in Figure 1. The ligation reaction is carried out in a self-templating manner and does not utilize a splint oligo. Either the RNA 3' sequence or the 5' end of the linker region of the peptide acceptor is designed to form a hairpin structure. Close positioning of the 5' end of the linker region of the peptide acceptor and the 3' end of the RNA in 20 the context of a hairpin structure facilitates enzymatic ligation with T4 DNA ligase, as described, for example, in Sambrook, Fritsch, & Maniatis, Molecular Cloning, Cold Spring Harbor, New York, Cold Spring Harbor Laboratory Press, 1989. The ratio of the reactants is approximately 1:1, although a slight excess (1:1.2) of either reactant is acceptable. Optimal ligation conditions may 25 vary slightly from reaction to reaction and may be determined experimentally using techniques known to those skilled in the art. - 13 - Example 2: Enzymatic Ligation Methods Involving Terminal Transferase Another enzymatic method for the attachment of a peptide acceptor to an RNA molecule involves the modification of the 3' end of the RNA followed by ligation using terminal deoxynucleotidyl transferase, for example, 5 as shown in Figure 2. The TdT reaction is carried out as generally described in Sambrook, Fritsch, & Maniatis (supra). This enzyme extends the 3' end of nucleic acids with deoxy- (or dideoxy-) nucleotide triphosphates. These (d/dd)NTPs may be chemically modified on their base moieties to carry the desired linker structures (as described, for example, in Meyer, Methods in 10 Molecular Biology, Agrawal, ed., vol 26, Totowa: Humana Press, 1994, pages 73-91; Kumar et al., Anal. Biochem. 169:376, 1988; Riley et al., DNA 5:333, 1986; and Schmitz et al., Anal. Biochem. 192:222, 1991). As shown in Figure 2, the linker molecule must initiate with a (d/dd) NTP. The remainder of the linker composition, however, may vary. A special feature of this terminal 15 transferase ligation method is that the DNA linker region of the peptide acceptor can be as short as one (d/dd)NTP unit. Example 3: Chemical Ligation Methods Involving Functionalization of the 3' End of an RNA Through IO4. Oxidation Followed by Chemical Ligation According to the methods of the invention, a peptide acceptor can 20 also be attached to an RNA molecule by functionalization of the 3' end of the RNA via 104 oxidation, followed by chemical ligation of the peptide acceptor to the RNA, as shown in Figure 3A. The 104 oxidation is carried out as described below and according to the methods of Agrawal (Methods in Molecular Biology, Agrawal, ed., vol 26, Totowa: Humana Press, 1994, pages 25 93-120), Proudnikov and Mirzabekov, Nucleic Acids Res. 24:4535, 1996; Gosh et al., Anal. Biochem. 178:43,1989; Bauman et al., J. Histochem. - 14 - Cytochem. 29:227, 1981; and Wu et al., Nucleic Acids Res. 24:3472, 1996). Since the IO 4 oxidation step strictly requires a 1,2-diol for reaction, only the terminal nucleotide is modified, leaving internal residues of the RNA molecule unaffected. The resulting dialdehyde can be subjected to a further reaction 5 with various nucleophiles such as amines, hydrazines, carbo(thio)hydrazides, or (thio)semicarbazides, yielding Schiff base-like structures. These reactions are carried out as described, for example, in Agrawal (supra), Proudnikov and Mirzabekov (supra), Gosh et al. (supra), Bauman et al. (supra), and Wu et al. (supra). While the (thio)hydrazides and (thio)semicarbazones obtained after 10 reaction with carbo(thio)hydrazides and (thio)semicarbazides, respectively, are fairly stable, the initial adducts containing amines or hydrazines usually require a subsequent reduction step, such as reductive amination, as shown in Figure 3B and as described in Agrawal (supra), Proudnikov and Mirzabekov (supra), Gosh et al. (supra), Bauman et al. (supra), and Wu et al. (supra), to render the 15 newly formed bonds stable toward hydrolysis. The above-described coupling reactions can be executed using a number of different strategies, as shown in Figures 4A-4D. For example, the ligation can be carried out in an external template-independent manner as shown in Figure 4A. This strategy requires a large excess of modified peptide 20 acceptor (for example, 100 to 1000-fold excess) to achieve successful ligation to the RNA molecule. To increase the efficiency of the RNA ligation, external template oligos may be used for substrate alignment as shown in Figure 4B. Preferably, such oligos, which are complementary in sequence to the RNA and peptide acceptor linker sequences, are at least approximately ten nucleotides in 25 length. Typically, this oligo will include at least approximately ten nucleotides which are complementary to the RNA molecule and approximately ten nucleotides which are complementary to the peptide acceptor linker. - 15- Alternatively, the reactive sites may be brought into close proximity by di-ect hybridization of linker and RNA domains as shown in Figures 4C-4D. Here, the extent of hybridizing sequence is at least ten to fifteen nucleotides. A number of different constructs for the attachment of peptide 5 acceptors to RNA via crosslink formation may be utilized. For example, one type of exemplary construct includes a peptide acceptor attached to a linker carrying a modification at its 5' end as shown in Figures 4A-4C. An alternative construct may comprise an internal functional group flanked by a hybridization domain on one side and a puromycin-linker portion on the other 10 (Figure 4D). The synthesis of these modified linkers involves standard automated DNA synthesis using commercially available phosphoramidites (Glen Research, Sterling, VA) for assembling the main body of nucleotides or spacer moieties. The 3' puromycin may be introduced through the use of 15 puromycin-CPG (Glen Research, Sterling, VA) as a solid support for synthesis. The attachment of the reactive functional groups may be achieved using commercially available reagents such as amino terminus-modifiers (Glen Research, Sterling, VA) or uni-link amino modifiers (Clontech, Palo Alto, CA). Other functional groups may be incorporated utilizing appropriate 20 phosphoramidites. One exemplary method to generate a carbohydrazide phosphoramidite is described in Figure 5. A carbohydrazide phosphoramidite was generated by combining a lactone with hydrazine and methanol and placing the reactants under reflux for 2 hours to produce a carbohydrazide moiety. The yield for this step of the synthesis was 97%. The resulting 25 product was then reacted with the salt of a dimethoxytrityl group and pyridine/toluene at room temperature for 2.5 hours to yield a protected carbohydrazide moiety. The product yield for this step was 88%. This product - 16was further reacted with a phosphoranidite moiety in the presence of tetrahydrofuran and diisopropylamine at room temperature for 30 minutes, which yielded (72%) a reaction product of a fully protected carbohydrazide phosphoramidite. 5 In addition, the reactivity of the peptide acceptor toward the 10 4 -oxidized RNA may be further enhanced through introduction of multiple copies of reactive groups. One exemplary ligation reaction was carried out as follows. One nmole of RNA, consisting of a transcript encoding a flag epitope and a strep 10 tag, of the sequence: 5' G GGA CAA UUA CUA UUU ACA AUU ACA AUG GAC UAC AAG GAC GAU GAC GAU AAG GGC GGC UGG UCC CAC CCC CAG UUC GAG AAG GCA UCC GCU (SEQ ID NO:1) was combined with 20 pI of 500 mM NaOAc (pH 5.2), 10tl of 5 mM NaIO 4 , and brought up to a final volume of 100 sl with water. The reaction mixture was incubated for 15 15 minutes at room temperature. Next, 10 pl of 10 mM Na 2

SO

3 was added, and the reaction mixture was incubated again for 15 minutes at room temperature. Forty pA of I M phosphate buffer (pH 8.0), 1.5 nmole of the peptide acceptor linker Uni-A1/8, having the sequence: 5' X CGC GGA TGC AAA AAA AAA AAA AAA AAA AAA AAA AAA CC Pu (SEQ ID NO:2) 20 (where X is a Uni-link amino modifier [Clontech), and Pu is Puromycin-CPG [Glen Research]), and 20 pAl of NaCNBH 3 were added to the reaction mixture. The mixture was then incubated for 18 hours at room temperature, precipitated, purified on a 6% TBE-Urea gel, and crush-soaked overnight to obtain the RNA-protein fusion molecule. This reaction yielded 230 pmole of product. - 17 - Example 4: Chemical Ligation Methods Involving Attachment of Functional Groups to the 3' End of an RNA Molecule Followed by Chemical Ligation A peptide acceptor can also be affixed to an RNA molecule by attaching a functional group to the 3' end of the RNA followed by chemical 5 ligation, as shown in Figure 6. In a variation of the process described above, reductive amination and related reactions are used to attach functional groups to the 3' terminus of an RNA. These newly introduced groups then react further with suitably modified linkers on the peptide acceptor, leading to covalent bond formation between the RNA and peptide acceptor. Exemplary 10 reactive groups include thiols (for disulfide formation or reactions with thiolphilic reagents such as pyridyl disulfides; Figure 6, Reaction A) or maleimides (Figure 6, Reaction B). Other possible reactive groups for the functionalization of the 3' end of an RNA followed by attachment of a peptide acceptor are amines. For 15 example, N-hydroxysuccinimide-esters (NHS-esters) can be generated on 5' amino-modified linkers of the peptide acceptor by reaction with disuccinimidyl glutarate (DSG) or related reagents as described in Cox et al. (J. Immunol. 145:1719, 1990) and http://www. piercenet.com/Products/linksearch.cfm (Pierce, Rockford, IL; Figure 6, Reaction C). This modified linker can then be 20 reacted with the amino functional group of the modified RNA. This type of ligation reaction may be carried out in either an external template-independent or -dependent manner as described above, using the same general approaches. Example 5: Chemical Ligation Methods Involving Photochemical Methods 25 A peptide acceptor can also be attached to an RNA molecule using photochemical methods as shown in Figures 7A-7H. Peptide acceptor linker - 18molecules that carry psoralen groups allow the introduction of crosslinks to complementary RNA strands upon irradiation with long wave UV-light. This technique may be carried out generally as described in Pieles and Englisch, (Nucleic Acids Res. 17:285, 1989); and Godard et al. (Nucleic Acids Res. 5 22:4789, 1994). Attachment of the psoralen moiety to the 5' terminus of the linker region of the peptide acceptor can be accomplished using commercially available psoralen amidites, 2'-OMe-RNA phosphoramidites, psoralen C6 phosphoramidites, and triethylene glycol (TEG) phosphoramidites (Glen Research, Sterling, VA) on a standard DNA synthesizer. 10 In one exemplary approach, this method was carried out as follows. One nmole of RNA consisting of an RNA transcript encoding a flag epitope, a strep tag, and a photochemical target site of the sequence: 5' G GGA CAA UUA CUA UUU ACA AUU ACA AUG GAC UAC AAG GAC GAU GAC GAU AAG GGC GGC UGG UCC CAC CCC CAG UUC GAG AAG AAC 15 GGC UAU A (SEQ ID NO:3), 1.2 nmole of Photolinker 30/10 consisting of the sequence: 5' Pso TAG CCG TTC T AAA AAA AAA AAA AAA AAA AAA AAA AAA CC Pu (SEQ ID NO:4) (where Pso is a psoralen C2 amidite [Glen Research], and Pu is Puromycin-CPG [Glen Research]), synthesized according to standard manufacturer protocols, or 30/15 consisting of the 20 sequence: 5' Pso TAG CCG TTC TTC TCG AAA AAA AAA AAA AAA AAA AAA AAA AAA CC Pu (SEQ ID NO:5) (where Pso is a psoralen C2 amidite, and Pu is Puromycin-CPG), 10x buffer (250 mM Tris pH 7.0; 1 M NaCl), and water (bringing the final volume to 360 pl) were combined and heated to 80"C for 2 minutes, The reaction mixture was then slowly cooled to 25 room temperature. The reaction mixture was next irradiated for 15 minutes at 0 0 C with X greater than 310 nm using a 450 W immersion lamp (medium pressure; ACE Glass, cat. no. 7825-34), equipped with a Pyrex absorption - 19sleeve (ACE Glass, cat. no. 7835-44) in a Quartz immersion well (ACE Glass, cat. no. 7854-25), with the sample in a microcentrifuge tube strapped to the immersion well, and cooled in ice-water. The sample was then precipitated with 40 yil of 3 M NaOAc and 1000 pl of ethanol, and resuspended in 75 t of 5 water. Next, 75 gl of 2x loading buffer (Novex) was added to the sample, and the sample was purified on a precast 6% TBE-Urea gel (Novex). The product was recovered using a crush and soak method (0.3 M NaOAc, overnight at room temperature) followed by ethanol precipitation. This photocrosslinking method yielded 272 pmole of RNA-protein fusion product using Photolinker 10 10/30, and 227 pmole using Photolinker 15/30. For this photocrosslinking method of chemical ligation, various parameters of the reaction were evaluated. First, the salt dependence of the photocrosslink formation was tested. A set of crosslinking experiments with buffers containing 100-1000 mM NaCl were performed. No difference in 15 ligation efficiencies were observed between the various reactions. In addition, a change of the RNA target sequence to: 5'...GAC UAC AAG GAC GAG GCA UCC GCU CUU UCA CUA UA (SEQ ID NO:6) (with the underlined sequence being a target for the psoralen linker) gave significantly reduced product yields (15 to 20% reduced), indicating that the RNA target sequence 20 was important. Next it was determined that the product yield could be increased by the repeated replacement of psoralen linkers that had been inactivated during the course of the reaction. This experiment was carried out as follows. The RNA, linker, and 1Ox buffer were combined and heated to 80*C for 2 minutes. The reaction mixture was then slowly cooled to room 25 temperature and irradiated as described above. Then an additional 1 nmole of linker and I il of buffer were added. The linker annealed to the RNA and was then irradiated. This process was repeated, by adding 2 nmole of linker and 2 - 20 pil of buffer and irradiating. This procedure allowed an increased product yield from 20% to greater than 40% for certain sequences. The performance of the ligation products generated by photochemical crosslinking methods was also evaluated. In experiments with 5 linkers of different lengths (psoralen plus 15 base pairs of target hybridization domain plus dACCPu [where n=7, 12, 17, or 22]) the following observations were made. Long linkers gave the highest RNA-protein fusion yields under high salt conditions (500 mM KCI plus 50 mM MgCl 2 ). In buffers with reduced salt (250 mM KCI plus 10 mM MgCl 2 , or 250 mM KCl), the short 10 linkers produced higher yields than the longer ones, but the overall yields were generally lower. In general, yields seem to be comparable to those obtained with enzymatically ligated RNA templates. In other exemplary methods, various mRNAs and puromycin linkers were synthesized such that the peptide acceptor was positioned at the 3' end of 15 the linker. The linkers were annealed to the target mRNAs and evaluated for their efficiency in forming mRNA-protein fusions molecules through in vitro translation techniques. The effect of the linker length and composition on fusion molecule yield was also determined. Figure 7D illustrates the design of the mRNAs used in the 20 crosslinking reactions. Each of the RNA molecules contained a portion of the tobacco mosaic virus 5'-UTR (TMV) with good initiation codon context (Gallie et al., Nucleic Acids Res., 16:883-93, 1988; and Kozak, Microbiol. Rev., 47:1-45, 1983), and a translation initiation codon (AUG). All mRNAs also carried a 3'-terminal ten nucleotide linker hybridization sequence 25 5'-GCAUCCGCUAUU coding for the amino acid sequence ASA, and ending with the dinucleotide sequence 5'-UA for psoralen photocrosslinking (const.), as described by Sinden and Hagerman (Biochemistry, 23:6299-6303, 198) and -21 - Gamper et al., (Photochem. Photobiol., 40:29-34, 1984). In addition one mRNA (mRNA 1) contained a Flag epitope, DYKDDDDK (Hopp et al., Biotechnology, 6:1205-1210, 1988) followed by a Strep-Tag II sequence, WSHPQFEK (Schmidt et al., J. Mol. Biol., 255:753-66, 1996). Other mRNA 5 molecules (mRNAs 2, 3, and 4) contained a 4-4-20 scFv, anti-fluorescein single chain antibody template, as described by Bedzyk et al. (J. Biol. Chem., 265:18615-20, 1990) and Mallender et al. (J. Biol. Chem., 271:5338-46, 1996). In addition, mRNAs 3 and 4 contained a downstream stop codon (UAA) to induce mRNA release from the ribosome after translation of residual 10 un-crosslinked template (vide infra). A poly-A tail was also attached to mRNA 4 for purification by oligo-dT. The mRNAs used in these studies (Figure 7D) were prepared by T7 RNA polymerase run off-transcription (Megashortscript Transcription Kit, Ambion, TX) of PCR DNA templates, according to the methods of Milligan et 15 al. (Nucleic Acids Res., 15:8783-8798, 1987). After transcription, all RNAs were purified by electrophoresis on 6% TBE-urea polyacrylamide gels (Novex, CA). The product bands were visualized by UV-shadowing, excised, crushed, and soaked overnight in 0.3 M NaOAc. Following ethanol precipitation, the RNAs were resuspended and stored in HO. Radiolabeled RNA was 20 synthesized according to the same procedure by including [a- 32 P] UTP (Amersham, IL) in the transcription buffer. The puromycin-linkers used in these studies (Figure 7E) were prepared using an Expedite Synthesizer Model 8909 (PerSeptive Biosystems, MA), using conventional solid-support phosphoramidite chemistry. 25 Puromycin-CPG, DNA phosphoramidites, 2'-OMe-RNA phosphoramidites, psoralen C6 phosphoramidite, and triethylene glycol (TEG) phosphoramidites (spacer 9) were used according to the recommended protocols (Glen -22- Research). A psoralen moiety was attached through a C6 alkyl chain to the 5'-phosphate of the linker. Flexible triethyleneglycol phosphate (TEG) spacers and polynucleotide sequences of various lengths were used to tether 5'-dCdC-puromycin to the 3'-end (see the table in Figure 7E). The linker 5 hybridization sequence was prepared from 2'-OMe-RNA phosphoramidites to enhance the pairing stability of the stem structure (Inoue et al., Nucleic Acids Res., 15:6131-6148, 1987; and Majlessi et al., Nucleic Acids Res., 26:2224-2229, 1998). Following deprotection in concentrated ammonium hydroxide for 8 hours at 55'C, the linkers were purified by reversed phase 10 HPLC on a C18 Spheri-5 column (Perkin Elmer, CA) with 50 mM triethylammonium acetate in 5% v/v acetonitrile as buffer A, and 50 mM triethylammonium acetate in 70% v/v acetonitrile as buffer B and with a flow rate of 1.5 ml/min. A linear gradient of 15 - 60% buffer B over 45 minutes was used for elution. After drying, the linkers were resuspended and stored in 15 H 2 0. The linkers (5 pM) were annealed to the target mRNAs (2.5 sM) in 25 mM Tris HCl buffer, pH 7, and 100 mM NaCl by heating to 85'C for 30 seconds, followed by cooling to 4'C over a period of 5 minutes. The reaction mixture was irradiated for 15 minutes at room temperature in borosilicate glass 20 vials (Kimble/Kontes, NJ), using a handheld multiwavelength UV lamp model UVGL-25 (UVP, CA) set to a long wave (wavelength X> 300 nm). The product mixture of the photocrosslinking reaction between radiolabeled mTRNA 1 and linker B was analyzed on a denaturing 6% TBE-Urea gel (Novex) and visualized on a phosporimaging system (Molecular Dynamics, 25 CA) (Figure 7E). These photocrosslink product mixtures generally contained <20% unreacted and >80% photocrosslinked mRNA, and were used directly for in vitro translation and subsequent mRNA-protein fusion formation - 23 without further purification. For the longer mRNA substrates 2, 3, and 4 (> 800 nucleotides), the relative size difference between mRNA and crosslinked mRNA became too small to be separated on a gel. In these cases, the crude photo-crosslinking reaction mixtures were directly added to the lysate for 5 fusion formation, without further purification. Translation and fusion formation of the mRNA fusion molecules were first tested using mRNA 2 in the following experiments. In vitro translation reactions were performed using rabbit reticulocyte lysates (Ambion) for 30 minutes at 30*C. The reactions contained 100 pmole photo-crosslinked 10 mRNA (see above), 10 mM creatine phosphate, 150 mM KOAc, 0.5 mM MgCl 2 , 0.1 mM of each amino acid except methionine, 150 sCi of ["S] methionine (Amersham), and 67% v/v of lysate in a total volume of 300 gIl. mRNA-protein fusion formation was promoted by the addition of KCI and MgC 2 to the final concentrations of 590 mM and 50 mM, respectively, in a 15 500 gl volume, according to the methods of Roberts & Szostak and Szostak et al. (supra). Incubation was continued for another 60 minutes at 20*C. Varying concentrations of KCl and MgCl 2 were also tested to explore salt dependence on fusion formation. The in vitro translation products were isolated by diluting the lysate 20 into 10 ml of binding buffer (100 mM Tris HCl, pH 8.0, 10 mM EDTA, 1 M NaCl, 0.25% v/v Triton X-100) and by adding to the mixture 10 mg of oligo-dT cellulose type 7 (Pharmacia, NJ). The samples were rotated for 60 minutes at 4*C. The solid support was then washed with 5 ml of ice-cold binding buffer, followed by elution with 100 IL aliquots of deionized H 2 0. 25 The amount of mRNA-protein fusion isolated was determined by scintillation counting of the incorporated ["S] methionine. The product was analyzed by electrophoresis on 4-12% NuPage gels using MES running buffer (Novex). - 24 - The gels were dried after extensive washing to remove excess ["S] methionine, and bands were visualized on a phosphorimager system (Molecular Dynamics). Gel analysis showed two bands that corresponded to the 5 peptidyl-tRNA and the free peptide (Figure 7F, lane 1). When translation was carried out with photo-crosslinked mRNA 2, a third and slower migrating band appeared, thus indicating successful mRNA-protein fusion formation (Figure 7F, lane 2). Increased yields of free protein and fusion (approximately 20% more) were obtained with mRNA 3, which was identical in coding sequence to 10 mRNA 2, but carried a stop-codon downstream of the photo-crosslink site (Figure 7F, lanes 3 and 4). Relative band intensities indicated that 30% of the total amount of synthesized protein was converted into mRNA-protein fusion (Figure 7F, lane 4). An mRNA-scFv fusion molecule, prepared from mRNA 4 of Figure 15 7E, was purified by binding of the A linker regions to oligo-dT cellulose, followed by washing with binding buffer, according to the methods of Roberts & Szostak and Szostak et al. (supra) (Figure 7E, lane 5). The fusion product could be isolated with a 1.3% yield based on the amount of photo-crosslinked input mRNA. Physical properties (gel mobility, binding to oligo-dT cellulose, 20 selective peptide binding to affinity reagents) of the fusions prepared from photo-crosslinked mRNA were found to be identical to those of the fusion product obtained from enzymatically ligated mRNA templates. In order to confirm the composition of the peptide portion of the fusion molecules, fusions prepared from the mRNA template 1 crosslinked to 25 linker B of Figure 7E, encoding the Flag and Strep-Tag II epitope, were tested for protein binding. A solution of 10 g l of 1 5 S-labeled mRNA-peptide fusion (prepared from mRNA 1 with linker B) was added to 20 gl of Anti-Flag M2 - 25 - Affinity Gel (Sigma, MO) in 300 stl of buffer containing 50 mM Tris HCI, pH 7.4, 1% NP 40, 150 mM NaCl, 1 mM EDTA, 1 mM Na 3 VO4, and 1 mM NaF. A second precipitation experiment was carried out by adding the same fusion product to 20 pl of StrepTactin Sepharose (Genosys, TX) in 300 g1 of buffer 5 containing 100 mM Tris-HCI, pH 7.1, 1 mM EDTA, and 0.5 mg/mil yeast tRNA. Both precipitation mixtures were processed in parallel under identical conditions: The mixtures were rotated for 1 hour at 4*C and then transferred onto an Ultrafree-MC filter unit (0.45 y m; Millipore, MA). The buffer was removed by centrifugation, and the residue washed with 5 x 300 sl of ice-cold 10 buffer. The residues were analyzed by scintillation counting, and fusion binding was determined to be 54% and 62% for the Anti-Flag M2 matrix and the StrepTactin matrix, respectively. A control reaction with protein-A agarose (Sigma) showed no detectable binding to the matrix. To test the effect of linker length and composition on fusion 15 formation in the presence of various salt concentrations, mRNA 3 was photo-crosslinked to linkers A-F of Figure 7E and each template was subsequently tested for fusion formation (Figure 7G). After incubation for 30 minutes at 30'C, the samples were divided into aliquots, to which varying amounts of KCl and MgCl 2 were added. The samples were incubated for 20 another 60 minutes at 20"C; then, the fusion product was analyzed by gel-electrophoresis. The highest mRNA-protein fusion yields were obtained with the long linkers A and B (40 and 35 nucleotides, respectively) under high salt conditions. A lower salt concentration resulted in a significant drop in fusion formation with the linkers A and B. On the other hand, fusion yields for 25 the shorter linkers C to F were less salt dependent. In addition, the fusion molecule yield generally increased with the number of flexible TEG spacers. Analysis of the crude mRNA-protein fusion molecule lysates revealed that up - 26 to 45% of the total protein was present as mRNA-protein fusion. Because linker F lacked the oligo-dA track needed for oligo-dT purification, mRNA 4 was prepared with an As stretch at its 3'-end, which allowed fusion purification on oligo-dT cellulose. 5 In an alternative to the above techniques which involve attachment of psoralen to the linker 5' end, the psoralen moiety may also be incorporated at an internal position of the linker region of the peptide acceptor, as shown in Figure 713, and as described in Pieles et al. (Nucleic Acids Res. 17:8967, 1989), or by incorporation of a branched phosphoramidite (Clontech, Palo 10 Alto, CA) within the linker sequence, followed by addition of a psoralen phosphoramidite (Glen Research, Sterling, VA). For example, the psoralen moiety may be crosslinked to the RNA through a branched linker, as described in U.S.S.N. 09/453,190, incorporated herein by reference. In one exemplary approach, a linker with the sequence 5' cgt agg cga 15 gaa agt gat X AAA AAA AAA AAA AAA AAA AAA AAA AAA CC Pu (where Pu is Puromycin-CPG [Glen Research]; C and A are standard 3' amidites [Glen Research]; a, t, c, and g are 5'-phosphoramidites [Glen Research]; and X is an asymmetric branching amidite [Clontech]) has been synthesized according to standard manufacturer protocols, followed by 20 selective deprotection of the branching point X (according to the instructions of Clontech) and subsequent coupling of a psoralen C6 amidite (Glen Research). This linker was then photocrosslinked to an RNA with the target sequence 5' ...GCA UCC GCU CUU UCA CUA UA using the photocrosslinking techniques described above. This RNA-linker construct was 25 then successfully used for the synthesis of RNA-protein fusions. - 27 - In yet another alternative, a linker containing a psoralen attached to the 3' end of the target hybridization domain may also be constructed (Figure 7C). In this method, the 5' end of the psoralen-containing linker is extended by another linker, that after reversal of strand orientation, terminates with a 3' 5 puromycin. A subsequent digestion of the RNA target domain with RNase H is optional and will increase the flexibility of the linker construct. In addition, this approach allows the attachment of the linker at internal positions of much longer RNA molecules, and the untranslated region downstream from the linker site is then clipped off before translation and RNA-protein fusion 10 formation. In one exemplary approach to this method, a linker with the sequence 5' Pso atg cga gaa agt gat aaa aaa aaa aaa CC Pu (where Pu is Puromycin-CPG [Glen Research]; C is a standard 3-amidite [Glen Research]; a, t, c, and g are 5'-phosphoramidites [Glen Research]; and Pso is a psoralen 15 C6 phosphoramidite [Gen Research]) was constructed according to standard manufacturer protocols. The linker was then photocrosslinked to an RNA with the target sequence 5' ...GUA UAC GCU CUU UCA CUA using the photocrosslinking techniques described above. This RNA-linker construct (with and without prior treatment with RNase H) was then successfully used 20 for the synthesis of RNA-protein fusions. One advantage of the photochemical methods for attaching a peptide acceptor to an RNA is that these methods do not require chemical modification of the RNA prior to ligation. This makes the process very robust and selective, and allows the use of an RNA from a crude T7 transcription reaction as the 25 substrate for the chemical ligation. - 28 - Example 6: Photocleavable Biotin-based RNA Purification and Ligation If desired, an affinity-based RNA purification step may be combined with a photochemical ligation procedure described above (Figure 7H). A suitable linker molecule is modified at its puromycin terminus with a 5 photocleavable biotin moiety (e.g., EZ-LinkTM NHS-PC-LC-Biotin, Pierce, Rockford, IL). The target RNA (obtained, for example, from a crude transcription reaction) is then hybridized to a defined amount of linker, and the resulting duplex is captured on a solid support, such as streptavidin (or a related) resin. The excess RNA, as well as the components of the transcription 10 reaction are then removed by extensive washing. Irradiation with long-wave UV light simultaneously leads to photocrosslink formation and product release from the resin. The ligated RNA can then be directly fed into a translation and fusion formation reaction, without further purification. This method is advantageous over previous RNA purification 15 schemes. For example, after transcription, the amount of RNA merely has to be estimated to exceed the amount of linker used. When subjected to the described procedure, its amount is automatically reduced to a quantity not more than the amount of linker used. This, in turn, allows one to proceed to the next step without further quantization of the RNA (or ligated RNA) by, for 20 example, taking A 2 60 UV readings, and the RNA amounts do not have to be adjusted otherwise. The nucleic-acid protein fusion molecule preparation process is therefore more suitable for automation. In one exemplary technique, this biotin-based RNA purification and ligation protocol may be carried out as follows. In this photocleavable biotin 25 based RNA purification and ligation procedure, the linker is first biotinylated. The linker C6-psoralen-2-OMe[U AGC GGA UGC) dA 18

TEG

2 dCdC-puromycin is biotinylated by combining 100 A of 100 p M linker (10 - 29 nmol total), 50 p1 of 1 pmole EZ-Linkm PC-LC-Biotin in DMSO, 20 pl of lOx PBS, pH 7.4, and 30 pl H 2 0. The mixture is incubated at room temperature for 2 hours, and results in a quantitative yield. The mixture is then precipitated twice with ethanol, with an expected recovery of >90% of the 5 PC-biotinylated linker, and resuspended in 200 gl H20. RNA is next transcribed using, for example, the T7 Megashortscript Kit (Ambion). Transcription is carried out using 10 pmole of a DNA template containing the sequence GCA UCC GCU AUU UAA An at the 3'-terminus for a 250 pl reaction. This transcription reaction should yield approximately 2-5 10 nmol of crude RNA product. No purification (phenol extraction, NAP-5 column, or RNeasy column) is required prior to proceeding to the next step. In the next step of this purification and ligation technique, the biotinylated linker (10 pl of 50 pmol/pl biotinylated linker; 500 pmol total) is annealed to 62.5 to 250 pl of the transcription mixture (containing an estimated 15 minimum of 500 pmole of RNA) using, for example, a PCR machine (heating for 30 seconds to 80'C, then cooling to 4'C at 0.3 0 C/sec) with 15 gl of 5 M NaCl (to a final concentration of 0.25 M for a 300 gI reaction) and H20, to a final reaction volume of 300 pl. The RNA of the reaction mixture is next immobilized onto 100 g1 of Neutravidin beads (Pierce, Rockford, IL) by gently 20 rocking the reaction mixture at 4'C for 30 minutes. The beads are then washed and resuspended in 300 sl H20. The beads are next spun down and washed 3 times with 100 pl of buffer (25 mM Tris pH 7.0 and 0.25 M NaCl). Then the beads are UV irradiated for 15 minutes at room temperature (using a hand-held UV lamp 25 UVGL-25; a microcentrifuge tube containing the beads is put directly on the lamp) to chemically ligate the linker to the RNA and photo-release the ligated molecule from the beads. It is expected that 250 pmol of ligated RNA is -30photo-released. Seventy-five pl of H 2 0 is added to the tubes, the tubes are vortexed for 30 seconds, and the beads are spun down. The 75 pl supernatant, containing the ligated RNA/linker, is used for translation and formation of nucleic acid-protein fusion molecules. 5 The nucleic acid-protein fusion molecules are formed by combining the 75 pl of supernatant from the previous step, and combining it with 225 pl of the buffer components and lysate of the Rabbit Reticulocyte Lysate Kit (Ambion), and incubating the mixture for 30 minutes at 30*C. KCl and MgCl 2 are next added to final concentrations of 500 mM and 50 mM, 10 respectively, and the reaction continues to incubate for 60 minutes at room temperature to produce the nucleic acid-protein fusion molecules. Example 7: Peptide Acceptor Attachment to an RNA Through Strong Non Covalent Bonds As an alternative to covalent bond formation for the attachment of a 15 peptide acceptor to an RNA molecule, methods relying solely on strong complex formation are also possible and are part of the present invention. One method involves the use of peptide nucleic acids (PNAs) for RNA recognition and binding, as shown in Figure 8A. PNAs are DNA mimics comprising a backbone composed of achiral and uncharged N-(2- aminoethyl)glycine units 20 (Knudsen and Nielsen, Nucleic Acids Res. 24:494,1996). PNAs have been shown to hybridize with sequence specificity and high affinity to complementary single-stranded DNA and to RNA. In particular, triple-helix-forming constructs comprising two PNA molecules binding to RNA, thereby forming a clamp, can provide an efficient means for strong 25 binding of RNA, since such constructs can be extremely resistant to thermal denaturation and conditions used for in vitro translation (Figure 8A) (Hanvey -31 et al., Science, 258:1481, 1992). The use of pseudoisocytosine bases further enhances stability at neutral and basic pH (Egholm et al,, Nucleic Acids Res. 23:217, 1995). It has been demonstrated that such PNA-clamps remain associated with mRNA 5 under in vitro translation conditions and cannot be displaced by the ribosome (Knudsen and Nielsen, Nucleic Acids Res. 24:494, 1996). This property maximizes the stability of the corresponding RNA-protein fusion constructs. The preparation of nucleic acid linker-PNA constructs may be accomplished by solid phase synthesis starting with puromycin-CPG, as shown 10 in Figure 8B and as described in Uhlmann et al. (Angew. Chem. Int. Ed. Engl. 35:2633, 1996). After assembly of the desired nucleic acid portion (or PEG-spacer, if so desired) using standard automated synthesis, the solid phase synthesis is continued by attaching the PNA domain with the appropriate reagents (for example, as described in Uhlman et al. (supra)). Alternatively, 15 the PNA can be pre-synthesized as a separate moiety (PE Biosystems, Foster City, CA), followed by chemical coupling to the desired linker portion. In one particular example, a puromycin-DNA linker may be modified with a 5' terminal amino group, which can be further converted into a chemically activated ester (e.g., an NHS-ester through reaction with 20 disuccinimidyl glutarate or related reagents; this technique is described, for example, in Cox et al. (J. Immunol. 145:1719, 1990); and http://www.piercenet.con/Products/ linksearch.cfm; Pierce, Rockford, IL). Subsequent reaction with the PNA moiety (having either an unprotected amino-terminus or a carboxy-terminal lysine) covalently links the domains. 25 This process may be carried out in a homogenous solution containing the final DNA-linker product, or with the DNA bearing protecting groups and remaining attached to the solid resin. - 32 - Example 8: Optimization of Linker Length and Composition For all of the strategies described above, it is preferable to optimize the linker construct. Factors to be considered, for example, are the potential inclusion of template/target recognition elements and the steric accessibility of 5 attached functional groups. Particularly when template or target recognition through nucleic acid hybridization is involved, factors including the target sequence and the chemical nature of the linker are preferably optimized. For example, RNA hybridization strength and consequently ligation efficiency are known to be increased by the use of 2-OMe RNA or propyne-modified 10 nucleobases, rather than DNA (as described, for example, in Inoue et al., Nucleic Acids Res. 15:6131, 1987); Kibler-Herzog et al., Nucleic Acids Res. 19:2979, 1991; and Wagner et al., Science 260:1510, 1993). The linkers may also be optimized for their effectiveness in the RNA-protein fusion reaction. This will generally involve varying the length of 15 the linker, but may also involve the use of different building blocks for RNA protein joining. In one particular example, the deoxynucleotides of the linker may be replaced with PEG-spacers or 2-OMe-RNA units (both from Glen Research, Sterling, VA). Other embodiments are within the claims. 20 All patents and patent applications are hereby incorporated by reference. The entire disclosure in the complete specification of our Australian Patent Application No. 2005211548 is by this cross-reference incorporated into the present specification. -33-

Claims (13)

1. An RNA molecule crosslinked to a peptide acceptor through a psoralen moiety.

2. The RNA molecule of claim 1, wherein said psoralen moiety is attached to said s peptide acceptor through a linker.

3. The RNA molecule of claim 2, wherein said psoralen moiety is positioned at the 5' terminus of said liner.

4. The RNA molecule of claim 2, wherein said psoralen moiety is positioned at the 3' terminus of said linker. 0

5. The RNA molecule of claim 2, wherein said psoralen moiety is positioned within said linker.

6. The RNA molecule of claim 2, wherein said linker comprises a C6 alkyl chain.

7. The RNA molecule of any one of claims 2 to 6, wherein the linker is a branched linker. 5

8. The RNA molecule of any one of claims 2 to 6, wherein the linker is psoralen amidite, 2'-OMe-RNA phosphoramidite, psoralen C6 phosphoramidite, triethylene glycol or phosphoramidite.

9. The RNA molecule of any one of claims 2 to 6, wherein the linker is branched phosphoramidite. 20

10. The RNA molecule of any one of claims 2 to 6, wherein the linker is extended by a second linker, wherein said second linker terminates with a 3' puromycin.

11. An RNA molecule chemically ligated to a peptide acceptor, wherein said RNA molecule is crosslinked to said peptide acceptor and said RNA molecule contains a psoralen photocrosslink target site. 25

12. The RNA molecule of claim 11, wherein said RNA contains a stop codon positioned at its photocrosslink target site.

13. The RNA molecule of claim I or claim 11, substantially as herein described with reference to any one or more of the examples and/or figures. - 34 3159724_1 (GHManers) P44319.AU.2 21-Feb-12