WO1989010978A1

WO1989010978A1 - Method of making nucleic acid dimers

Info

Publication number: WO1989010978A1
Application number: PCT/US1989/001966
Authority: WO
Inventors: Barbara Chen Fei Chu; Leslie Eleazer Orgel
Original assignee: The Salk Institute For Biological Studies
Priority date: 1988-05-09
Filing date: 1989-05-08
Publication date: 1989-11-16
Also published as: PT90511A; ZA893390B; AU3764989A; PT90511B

Abstract

An improved method is provided for making a dimer of a first nucleic acid, terminally derivatized with a moiety of formula -(OPO2)NH(R11)SS(R12)NH2, and a second nucleic acid, terminally derivatized with a moiety of formula -(OPO2)NH(R21)SS(R22)NH2, wherein R11, R12, R21 and R22 are the same or different and each is an alkylene group of 2 to 20 carbon atoms or a cycloalkylene group of 3 to 20 carbon atoms. The improvement entails reducing the disulfide bonds of the two nucleic acids and, prior to oxidation of the resulting thiol adducts to form dimers, dialyzing from the nucleic acid solution the compounds of formulae HS(R12)NH2 and HS(R22)NH2, made upon reduction of the disulfides, while maintaining the concentration of reducing agent in the solution sufficiently high to maintain the concentrations of the nucleic acid thiol adducts. In a preferred embodiment, after removal of most of the compounds of formulae HS(R12)NH2 and HS(R22)NH2 from the nucleic acid solution, reducing agent is removed from said solution by dialysis and the nucleic acid thiol adducts are dimerized by oxidation with atmospheric oxygen. The nucleic acids are preferably derivatized at the 5'-carbons of their 5'-termini. The preferred groups R11, R12, R21 and R22 are -(CH2)2-. The preferred reducing agent is dithiothreitol. In a preferred application of the invention, one of the derivatized nucleic acids is a replicative RNA and the other is an oligonucleotide with the sequence of an affinity molecule for a nucleic acid analyte to be detected in a nucleic acid probe hybridization assay.

Description

METHOD OF MAKING NUCLEIC ACID DIMERS

The United States Government has certain rights in the invention disclosed and claimed herein as a result of funding of work related to said invention under a grant from the U. S. National Institutes of Health.

TECHNICAL FIELD

The present invention generally relates to linking together two nucleic acids. More particularly, the invention relates to a method of making a dimer of two nucleic acids, covalently joined through a linker which includes a disulfide bond and which is bonded to a ribose carbon of a terminal nucleotide of each of the nucleic acids.

BACKGROUND OF THE INVENTION

New procedures for attaching discrete biopolymers to one another are being actively pursued in the fields of molecular biology and biotechnology. One approach that has been widely used involves the incorporation of a suitably protected linker group-containing nucleotide analogue in place of one of the standard monomers in a solid-phase synthetic procedure for making a synthetic oligodeoxynucleotide and then, after isolation and deprotection of the oligonucleotide, reaction of the linking moiety on the linker group with a suitable linking moiety joined to another nucleic acid or other type of biopolymer. Such procedures that introduce a thiol group at the 5'-terminus or the 3'-terminus of an oligonucleotide have been reported, and in one case the oligonucleotide product has been coupled to a protein, micrococcal nuclease. These synthetic procedures however are not applicable when it is necessary to attach a ligand to an unprotected oligonucleotide or nucleic acid that has been synthesized enzymatically or isolated from natural sources. Therefore, it would be advantageous to have an alternative methodology for linking an unprotected oligonucleotide or polynucleotide to any of a wide range of ligands, including other nucleic acids.

Chu et al., Nucleic Acids. Res. 11, 6513-6529 (1983); Chu and Orgel, Proc. Nat'l. Acad. Sci. USA 82, 963-967 (1985); and Chu and Orgel, DNA 4, 327-331

(1985) disclose methods for derivatizing short oligonucleotides, through a terminal phosphate, with compounds with a primary amino group, which forms a phosphoramidate group with the phosphate. In Chu et al., Nucleic Acids Res. 14, 5591 - 5603 (1986), and PCT International Application Publication No. WO 87/06270, derivatizing oligonucleotides and replicative RNAs with cystamine, (NH₂(CH₂)₂SS(CH₂)₂NH₂), through a 5'-terminal phosphate of replicative RNA or a 5'-terminal or 3 '-terminal phosphate of oligonucleotide, is described, as is the linking of a replicative RNA and an oligonucleotide, so derivatized, through a linking moiety of formula

-O(PO₂)NH(CH₂)₂SS(CH₂)₂NH(PO₂)O-. Chu et al., Nucleic Acids Res. 14, 5591 - 5603

(1986) and PCT International Application Publication No. WO 87/06270 also disclose the use of replicative RNAs, such as MDV-1 RNA, covalently joined to an oligonucleotide affinity molecule as a probe in a nucleic acid probe hybridization assay for an analyte nucleic acid, to which the affinity molecule binds specifically through complementary base-pairing. In one such assay, in which the linker between replicative RNA and affinity molecule includes a disulfide group, replicative RNA is separated from affinity molecule, after hybridization of replicative RNA-affinity molecule adduct to analyte in a sample and washing from the assay system replicative RNA-affinity molecule adduct that has not hybridized to analyte, by reduction of the disulfide and then the released replicative RNA is autocatalytically replicated with a suitable replicase and detected.

Consequently, a facile and efficient method for making adducts of single-stranded nucleic acids with oligonucleotide (or polynucleotide) affinity molecules, wherein the linker between the nucleic acids of an adduct includes a disulfide group, and particularly such adducts between replicative RNAs and oligonucleotide (or polynucleotide) affinity molecules, would be advantageous.

The present invention provides such a facile and efficient method which, additionally advantageously, does not require protecting groups on nucleic acid reagents, can be applied with nucleic acids that have been synthesized enzymatically or made in vivo, and does not require the use of biotin-avidin or other proteinaceous substances in the linker between nucleic acids.

SUMMARY OF THE INVENTION

The invention is an improved method for synthesizing an adduct of two nucleic acids which are covalently joined by a linker which includes a disulfide and which is joined through one linking moiety to the 5'-carbon of the 5'-nucleotide or the 3'-carbon of the 3'-nucleotide of one of the nucleic acids and through the other linking moiety to the 5'-carbon of the 5'-nucleotide or the 3'-carbon of the 3'-nucloeotide of the other of the nucleic acids. In the improved method, the two nucleic acids to be dimerized, each terminally derivatized (i.e., through a linking moiety bound to the 5'-carbon of the 5'-nucleotide or the 3'-carbon of the 3'-nucleotide) with a disulfide-containing group, are combined in an aqeous solution with a reducing agent at a disulfidereducing effective concentration and then, while maintaining the solution under reducing conditions, the low molecular weight, sulhydryl-group-containing compounds, formed by the reduction of the disulfides joined to the nucleic acids, are dialyzed from the solution. Finally, the solution is subjected to oxidizing conditions,, optionally while reducing agent is dialyzed from the solution, and the desired nucleic acid dimer, with nucleic acids joined through a disulfide, forms. The desired dimer can be isolated by standard chromatographic techniques.

DETAILED DESCRIPTION OF THE INVENTION

The present invention is an improvement in a method of ligating a first nucleic acid, which is terminally derivatized with a moiety of Formula I -O(PO₂)NH(R₁₁)SS(R₁₂)NH₂ I and a second nucleic acid, which is terminally derivatized with a moiety of Formula II

-O(PO₂)NH(R₂₁)SS(R₂₂)NH₂ II wherein R₁₁, R₁₂, R₂₁ and R₂₂ are the same or different and each is an alkylene group of 2 to 20 carbon atoms or a cycloalkylene group of 3 to 20 carbon atoms, to make a dimer of said two nucleic acids joined by a linker of Formula III

-O(PO₂)NH(R₁₁)SS(R₂₁)NH(PO₂)O- III which method comprises combining said first and second nucleic acids in a first aqueous solution with a disulfide reducing-effective concentration of a reducing agent selected from the group consisting of dithiothreitol, dithioerythritol, M₁(BH₄) and M₁(BH₃CN), wherein M₁ is selected from the group consisting of Li⁺, Na⁺ and K⁺, and then subjecting said first solution to oxidizing conditions, whereby said dimer of said two nucleic acids is formed.

The improvement of the invention comprises dialyzing said first solution against a second aqueous solution, which has a composition such that (i) the concentration in said first solution of a reducing agent selected from the group consisting of dithiothreitol, dithioerythritol, M₁(BH₄) and M₁(BH₃CN), wherein M₁ is selected from the group consisting of Li⁺, Na⁺ and K⁺, is maintained at a disulfide reducing-effective level and (ii) the ratio of the molar concentration in said first solution of the compound of Formula IV

HS(R₁₂)NH_{2 IV} to the molar concentration in said first solution of all derivatives of said first nucleic acid (i.e., of said first nucleic acid in all of its forms) and the ratio of the molar concentration in said first solution of the compound of Formula V HS(R₂₂)NH₂ V to the molar concentration in said first solution of all derivatives of said second nucleic acid decrease.

The skilled will understand that moieties, other than those defined above for R₁₁, R₁₂, R₂₁ and R₂₂, can be employed to covalently join the phosphoamidate group to the sulfhydryl or disulfide (in the case of R₁₁ and R₂₁) and the disulfide to the amino group (in the case of R₁₂ and R₂₂) in the nucleic acid adducts employed in, or made in accordance with, the invention. While the alkylene and cycloalkylene groups described above are suitably stable, unreactive with functional groups in nucleic acids, and, particularly in the case of n-alkylenes with between about 2 and 8 carbon atoms, conveniently available, other groups which are stable to the reaction conditions employed in carrying out the invention, which do not react with functional groups in nucleic acids, and which do not reduce the solubility in aqueous solution of the derivatized nucleic acids to levels so low that the methods of the invention could not be carried out on a practically useful time scale, could be employed. Thus, for example, each of R^, R₁₂, R₂₁ and R₂₂ could be of formula -R₃₁-R₃₂-R₃₃-, wherein R₃₁ is the same as or different from R₃₃ and each of R₃₁ and R₃₃ is selected from alkylene, preferably of 1 to 4 carbon atoms, and R₃₂ is selected from the group consisting of phenylene or -(NH)(CO)-.

By "disulfide reducing-effective" concentration or level is meant that the ratio of the concentration of reducing agent (in all forms, including fully reduced, partially oxidized (if there is more than one oxidation state), and fully oxidized) in a solution to the concentration in the solution of the terminally derivatized nucleic acids (of all derivatives thereof, including monomers, homodimers and heterodimers) to be linked in accordance with the method of the invention is such that, at equilibrium at room temperature, at least half of the molecules of each of said nucleic acids would be in monomeric form, i.e., not linked through a linker with a disulfide group to another nucleic acid molecule. "Disulfide- reducing effective" concentrations will depend on the nature of the reducing agent, the concentration of oxidizing agent(s) if any in the solution, and the concentrations of nucleic acids in the solution to be joined in accordance with the invention. Ascertaining disulfide reducing effective concentrations of reducing agents employed in the method of the invention is simple and straightforward for those of ordinary skill.

For the purpose of driving disulfide reductions to completion rapidly or ensuring that virtually all nucleic acid-linked disulfides in a solution are maintained in the reduced state, concentrations of reducing agent far in excess of "disulfide reducing-effective" concentrations may be employed.

The most preferred reducing agent for use in accordance with the invention is dithiothreitol. As indicated above, however, other reducing agents such as dithioerythritol and salts, particularly the alkali metal salts, of BH₄- and BH₃CN- can also be employed. Generally, any reducing agent can be employed that, in water, is soluble to at least about 1 mM and sufficiently stable (i.e., is not oxidized, decomposed or otherwise rendered ineffective as a reducing agent in water with a t_1/2 of less than about 1 hour) and does not form, with a sulfhydryl group, linked through a group of formula -O(PO₂)NH(R₁₁)- to a 5'-carbon of a 5'-nucleotide or a 3'-carbon of a 3'-nucleotide of a nucleic acid, an adduct that is so stable that, under oxidizing conditions, the formation of disulfides between such sulfhydryls linked to nucleic acids is substantially prevented. Thus, for example, beta-mercaptoethanol is not an acceptable reducing agent for use in the improved method according to the invention.

In the carrying out the method of the invention, it is preferred that the dialysis, while reducing agent is maintained at a disulfide-reducing effective level, be continued until the molar concentration of the compound of Formula IV in the first solution becomes less than about 10 % of the molar concentration of the first nucleic acid, in all of its forms, in that solution and the molar concentration of the compound of Formula V in the nucleic acid solution similarly becomes less than about 10 % of the molar concentration of the second nucleic acid, in all of its forms, in that solution.

It is not necessary that, at the beginning of the dialysis to decrease the concentrations of compounds of Formulae IV and V in said first nucleic acid solution, reducing agent, in the second solution, against which said first solution is dialyzed, consist of the same compound(s) as reducing agent in said first solution. It is only necessary that, whatever the compositions of the respective reducing agents and how the composition of reducing agent in said first solution may change in the course of the dialysis, that the concentrations of the compound(s) of reducing agent in said first solution remain disulfide reducing effective during the dialysis. It is, however, preferred that one of dithiothreitol and dithioerythritol be employed as reducing agent in both said first and said second solutions. As indicated above, most preferred is dithiothreitol.

It is after the dialysis, to decrease the concentrations of compounds of Formulae IV and V, that, in accordance with the invention, the first solution is subjected to oxidizing conditions. Methods of subjecting a solution to oxidizing conditions, to cause the formation of disulfides from thiols, are also well known to the skilled. Simply exposing the solution to air, optionally with agitation, is usuually sufficient. An O₂-containing gas other than air, including pure O₂ and mixtures of

O₂ with N₂ or various inert gases, can also be used and can be introduced by purging the solution with the gas or bubbling the gas through the solution.

In a preferred method of subjecting a solution to oxidizing conditions, after the concentrations of the compounds of Formulae IV and V are reduced to a desired level, oxygen from the air is allowed to enter the solution while, or after, the solution is dialyzed against a third solution which is substantially free of reducing agent capable of reducing disulfide bonds. By "substantially free" is meant that, at the beginning of the dialysis, the concentration of reducing agent is at least a factor of ten below the "disulfide reducing effective" concentration in said nucleic acid solution which is being subjected to oxidizing conditions.

Preferably, at the beginning of said dialysis, there would be essentially no reducing agent capable of reducing disulfide bonds in said third solution.

Methods of preparation of derivatized nucleic acids employed in the improved method of the invention are known. See, e.g., Patent Cooperation Treaty Application International Publication No. WO 87/06270.

It is most preferred that the nucleic acids be terminally derivatized through the 5'-carbon of the 5'-nucleotide.

Further, it is preferred that R₁₁, R₁₂, R₂₁ and R₂₂ each be n-alkylene of 2 - 8 carbon atoms and most preferred that each be ethylene (-(CH₂)₂).

In a preferred application of the invention, an oligonucleotide, that is an affinity molecule for a nucleic acid analyte to be detected in a nucleic acid probe hybridization assay and that preferably consists of 10-100 nucleotides in a sequence which is complementary to that of a segment of the analyte, is joined to a replicative RNA. Methods of using such an oligonucleotide-replicative RNA dimer are described in Patent Cooperation Treaty Application International Publication No. WO 87/06270.

The use in nucleic acid probe hybridization assays of dimers of one oligonucleotide joined to another, wherein at least one of the oligonucleotides is an affinity molecule for a nucleic acid analyte, are described hereinbelow.

The invention will now be illustrated in some detail with the following examples.

EXAMPLE 1 Materials and Methods

Gel elecrophoresis was carried out on 0.5-1.0 mm thick, 6% or 20% polyacrylamide gels cast and run in 90 mM Tris borate, pH 8.0, and 1 mM EDTA, with or without 7 M urea. Autoradiographs of gels were obtained by exposure to Kodak X-Omat AR film at -80ºC with or without a Du Pont Cronex Lightning Plus intensifying screen.

RNAs were precipitated from solution by addition of 2 volumes of ethanol at -80°C in the presence of 100 mM NaCl. All RNA and oligonucleotide adducts were extracted from gels with 500 mM ammonium acetate, pH 7.2, 0.1 mM EDTA, and 0.01% SDS. They were then purified by passage through a Du Pont Nensorb²⁰ nucleic acid purification cartridge (DuPont, Inc., Wilmington, Delaware, USA).

Dialysis tubing (1000 m.w. cutoff) was treated with acetic anhydride for 1 hour and then washed with 2% NaHCO₃ and 1 mM EDTA. Unless otherwise specified, adducts linked via disulfide bond were reduced for 1 hour at room temperature with 10 mM DTT in Tris-EDTA (10 mM Tris, 1 mM EDTA) at pH 7.2. Reduction of disulfide bonds required a concentration of at least 5 mM DTT and could be maintained in the reduced condition with 0.1 mM DTT.

Nucleic acid heterodimers formed by the method of the invention can be separated from unreacted components and homodimers by polyacrylamide gel electrophoresis on gels cast and run in 90 mM Tris borate, 1 mM EDTA, 7 M urea (pH 8.0), as known in the art, and thereafter eluted from the gel and concentrated in a speed vac concentrator.

High performance liquid chromatography (HPLC) of oligonucleotides may be performed on RPC-5 at pH 12, using a perchlorate gradient as described by Bridson and Orgel, J. Mol. Biol. 144, 567-577 (1980).

MDV-1 RNA derivatized with cystamine through a phosphoramidate group bonded to the 5'-carbon of the 5 '-nucleotide was prepared as described in Patent Cooperation Treaty Application International Publication No. WO 87/06270. The 5'-cystamine-derivatized MDV-1 RNA may be separated from unreacted cystamine by gel electrophoresis or by HPLC on a SynChropak GPC 60 gel exclusion column (SynChrom, Inc., Lafayette, Indiana, USA) by elution with 0.1 mM EDTA at pH 7.

The oligodeoxyribonucleotide,

5'-CACAATTCCACACAAC-3' (hereinafter "16mer") is complementary in sequence to residues 6170-6185 of the M13mpl8 DNA(+) strand. The oligodeoxyribonucleotide

5'-TCGTATGTTGTGTGGAATTGTGAGCGGATAACAATTT-3' (hereinafter "37mer") has the same sequence as residues 6164-6200 of the M13mp18 (+) strand. The 16mer and 37mer were synthesized on an Applied Biosystems Synthesizer Model 380. The purification of the two oligomers, and their conversion to 5'-phosphates, is described by B.C.F. Chu, et al., Proc. Nat'l. Acad. Sci. USA, 82, 963-967 (1985). The single stranded M13mp18 DNA was purchased from New England Biolabs (Beverly, Massachusetts, USA).

0.01-0.5 O.D. units (measured at 260 nm) of 5'-P-16mer or 5'-[32P]-16mer was reacted with 0.1 M 1-methylimidazole (pH 7), 0.15 M 1-ethyl-3,3-dimethylaminopropylcarbodiimide (carbodiimide) and 0.5 M cystamine dihydrochloride (pH 7.2) in a volume of 50-250 μl at 50°C for 2 hours. The product was separated from reactants by HPLC on RPC-5. The yield ranged between 75% and 85%. 16mer derivatized at the 5'-carbon of the 5'-nucleotide with

NH2(CH2)2SS(CH2)2NH([32P]O2)O- (referred to as the 5'-cystamine-[32P]-16mer) and 5'-[32P]-16mer were also readily separated by gel electrophoresis on 20% polyacrylamide containing 7 M urea.

The corresponding cystamine adduct of 5'-[32P]-37mer was obtained by analogous procedures.

To show that carbodiimide does not react with the disulfide group, a sample of 5'-cystamine-P-16mer was made by first isolating the stable 5'-imidazolide adduct of 5'-P-16mer (see Chu et al., Nucl. Acids. Res. 11, 6513-6529 (1983)) and then treating the imidazolide adduct with cystamine in the absence of carbodiimide This procedure gave a product that had the same mobilities both in HPLC on SPC-5 and gel electrophoresis as the product made by the one-step procedure with 1-methylimidazole.

The 5'-cystamine-P-16mer is converted to 5'-thio-ethylamino-P-16mer in one hour with 5 mM DTT at pH 7, and can be maintained in the reduced state with 0.1 mM DTT. The 5'-P-16mer, 5'-cystamine-P-16mer and 5'-thio-ethylamino-P-16mer are readily separated by gel electrophoresis.

EXAMPLE 2 Synthesis of 5'-[³²P]-16mer-ss-5'-[³²P]-16mer or 5'-[³²P]-37mer

A 5 μM solution (25-50 μl) of 5'-cystamine- [³²P]-16mer (10⁴-10⁶ cpm), or a 5 μM solution of 5'-cystamine-P-16mer and 10³-10⁴ cpm of 5'-cystamine-[³²P]-37mer, were treated with 6 mM dithiothreitol (DTT) in 10 mM Tris buffer containing 1 mM EDTA (Tris-EDTA buffer) at pH 7 for 1 hour at 25°C. The reaction mixture was then dialyzed against one liter of buffer containing 1 mM Tris, pH 7.2, 1 mM DTT for 30 minutes at 4°C. It was next dialyzed against fresh buffer containing 1 mM Tris, pH 7.2, and 1 mM

EDTA for a further 30 minutes. After concentration in a speed-vac concentrator, the products were separated from reactants by electrophoresis on 20% polyacrylamide under denaturing conditions (i.e., with 7 M urea). Yields of dimerized 16mer ranged from 30% to 40% and of 5'-P-16mer-ss-[³²P]-37mer from 45% to 55%.

EXAMPLE 3

Synthesis of 5'-P-16mer-ss-5'-[³²P]-MDV-1 RNA

10⁴-10⁶ cpm (5-500 ng) of 5'-cystamine- [³²P]-MDV-1 RNA and a 5 μM solution of 5'-cystamine-P-16mer (total volume 25-50 μl) was treated for 1 hour with 6 mM DTT in Tris-EDTA buffer at pH 7. The reaction mixture was then dialyzed against one liter of buffer containing 1 mM Tris, pH 7.2, 1 mM EDTA and 0.1 mM DTT at 4°C for 30 minutes. The solution was next dialyzed against 1 liter of buffer (saturated with air at room temperature) containing I mM Tris, pH 7.2, and 1 mM EDTA for a further 30 minutes at 4°C. The reaction solution was then concentrated in a speed-vac concentrator, and the product separated from reactants on 6% polyacrylamide containing 7 M urea. The yield of desired dimer ranged from 45% to 55%.

EXAMPLE 4 Hybridization of 5'-[³²P]-16mer and 5'-P-16mer-ss-5'-[³²P]-MDV-1 RNA to M13mp18DNA (+) strand

Hybridization procedures described by Meinkoth and Wahl, Anal. Biochem. 138, 267-284 (1984), were used to immobilize single-stranded M13mp18 (+) DNA on nitrocellulose filters. Filters containing 100, 10 and 1 ng of single-stranded M13mp18 DNA and 100 ng lambda DNA were pre-hybridized for 1 hour at 32°C in hybridization buffer (900 mM NaCl, 6 mM EDTA, 90 mM Tris, pH 7.5, 0.1% SDS) containing 200 μg/ml homochromatography mix I (randomly cleaved RNA) (Jay, et al., Nucleic Acids Res. 1, 331-354 (1974). Hybridization was carried out overnight at 32°C with approximately 7,000 cpm of 5'-[³²P]-16mer (0.0045 pmole or 0.025 ng per ml), 5'-P-16mer-ss-5'-[³²P]-MDV-1 RNA (approximately 0.0045 pmole or 0.36 ng RNA per ml), or 5'-[³²P]-cystamine- MDV-1 RNA (0.0045 pmole or

0.34 ng RNA per ml). The filters were then washed several times with 1 × SSPE (180 mM NaCl, 10 mM Na₂HPO₄ and 1 mM EDTA, pH 7.5) containing 0.1% SDS at room temperature. After drying, the filters ware autoradiographed.

No hybridization was detected with the cystamine-derivatized MDV-1 RNA. A minimum of 10 ng of the M13mp18 DNA could be detected with the 16mer alone, and a minimum of 100 ng with the 16mer-MDV-1 RNA dimer. However, in these assays, probe was present at very low concentration and, furthermore, MDV-1 RNA was not released and amplified. These results demonstrate that, in the usual case, in which probe is used in substantial excess over analyte, the presence of the MDV-1 RNA in the probe would interfere by significantly less than a factor of 10 with the sensitivity of a 16-mer probe.

After removal of the filters, the hybridization fluid containing unhybridized material was passed through an Amicon 30 microconcentrator. An aliquot of the recovered materials was analyzed by gel electrophoresis to test the stability of 5'-P-16mer-ss-5'-[³²P]-MDV-1 RNA under hybridization conditions and it was found that the adduct does not decompose under hybridization conditions.

EXAMPLE 5 Hybridization of 5'-P-16mer-ss-5'-[³²P]-MDV-1 RNA to a complementary 37mer

0.03 pmoles of the target 37mer, or 0.03 pmole of a non-complementary 25mer were heated at 95°C-100°C for 1 minute and rapidly chilled before addition of

0.02 pmole of 5'-P-16mer-ss-5'[³²P]-MDV-1 RNA or 5'-cystamine-[³²P]-MDV-1 RNA. Hybridization was carried out for 7 hours at 25°C in buffer (10 μl) containing 50 mM Tris at pH 7.2, 1 mM EDTA, 500 mM NaCl and 20 μg randomly cleaved RNA. Hybrid, if formed, was separated from non-hybridized 5'-P-16mer-ss-5'-[³²P]-MDV-1 RNA by electrophoresis on a 6 % polyacrylamide gel under non-denaturing conditions (i.e., without urea) at 120 v (6ma) for 15 hours. A hybrid was detected with complementary 37mer but not with non-complementary 25mer.

EXAMPLE 6 Synthesis of 5'-cystamine-[³²P]-MDV-1 RNA

The procedure given above for the preparation of cystamine adducts of short oligodeoxynucleotides is not applicable to RNA because high moleculr weight RNA reacts with carbodiimide at 50°C. The synthesis of 5'-cystamine-[-³²P]-MDV-1 RNA has been previously described (Chu et al., Nucl. Acids Res. 14, 5591 - 5603 (1986)) and requires isolation of the 5'-imidazolide of MDV-1 RNA and removal of the carbodiimide before reaction with cystamine at 50°C. The RNA is separated from excess cystamine either by gel electrophoresis or, more quickly (in 15 minutes), by HPLC on a gel filtration column. The product purified by gel electrophoresis contains a mixture of unconverted 5'-[³²P]-MDV-1 RNA and 5'-cystamine- [³²P]-MDV-1 RNA. The product, if purified by HPLC gel filtration, may also contain 5'-[³²P]-RNA degradation products and their cystamine adducts. The entire synthesis, starting from 5'-[³²P]-MDV-1 RNA, can be completed in 4 hours.

EXAMPLE 7

Linkage of 16mer with 16mer, 37mer or MDV-1 RNA via a disulfide bond

Dimerization was achieved by reducing a 5 μM solution of 5'-cystamine-[³²P]-16mer with DTT, removing the 2-thio-ethylamine under reducing conditions and finally removing the DTT by dialysis.

After the removal of DTT, air oxidation occurs in the dialysis bag within a half hour, resulting in disulfide bond formation between two Iδmers. Ligation of 16mer with 37mer or with MDV-1 RNA was carried out in the same way. The oxidation reaction was complete in 30 minutes, and the entire ligation procedure required 2 hours. One advantage of this procedure is that all derivatives can be stored as stable 5'-cystamine adducts, rather than as less stable thiol adducts; the thiol adducts do not need to be isolated before the start of the reaction. When the disulfide-linked adducts of 16mer with 16mer, 37mer or MDV-1 RNA were treated with 10 mM DTT, the S-S links were reduced and the oligonucleotides released as 5'-thio-ethylamino derivatives.

EXAMPLE 8

This example describes an assay for simultaneously detecting the presence of one or both of two nucleic acids sequences thought to possibly be present in a biological sample. The two nucleic acid sequences may be related to each other in that, although they differ in sequence and are from pathogenic organisms which are different, the organisms cause diseases which may have similar symptoms or manifestations. For instance, it is well-known that diseases accompanied by diarrhea may be caused by Eschericia Coli or Shigella dysenteriae. In such a case, an assay of the kind described in this example can be employed to ascertain the actual etiological agent, if it is present in a clinical specimen. A nucleic acid fraction from a specimen obtained from a patient is prepared and immobilized in single- stranded form on a solid substrate. 5'-cystamine-derivatized, ³²P-Iabelled oligonucleotide probes (i.e., affinity molecules, one for each of two nucleic acid analytes, one of which is characteristic of each of the organisms being assayed for) are prepared as in Example 1. The two probes, differing in sequence and nucleotide chain length (so as to be separable on the basis of molecular weight), are dimerized as in Example 2. The oligonucleotide heterodimer is allowed to hybridize for 1 hour with the immobilized nucleic acid sample and then the support is washed extensively with a reducing buffer. As the skilled will understand, only nucleotide probe immobilized by direct hybridization to the immobilized nucleic acid sample will remain associated with the support after the wash. The nucleic acid is then eluted from the support and electrophoresed on a polyacrylamide gel under denaturing conditions. The radiolabeled dimer is run as a control. Because of the difference in relative mobility of the two oligo-nucleotides on the gel, the presence of either or both of the nucleotide probes in the gel, and thus of the respective analyte nucleic acids in the sample, can be discerned by autoradiography of the gel.

The invention has been described herein with some specificity, and those of ordinary skill will recognize numerous modifications and variations from what has been described that are within the spirit of the invention. It is intended that such modifications and variations be within the scope of the invention as described and claimed herein.

Claims

WHAT WE CLAIM IS :

1. In a method of ligating a first nucleic acid, which is terminally derivatized with a moiety of Formula I

-O(PO₂)NH(Rn)SS(R₁₂)NH₂ I and a second nucleic acid, which is terminally derivatized with a moiety of Formula II -O(PO₂)NH(R₂₁)SS(R₂₂)NH₂ II wherein R₁₁, R₁₂, R₂₁ and R₂₂ are the same or different and each is an alkylene group of 2 to 20 carbon atoms or a cycloalkylene group of 3 to 20 carbon atoms, and wherein at least one of said first and said second nucleic acids is an affinity molecule for a nucleic acid analyte detectable in a nucleic acid probe hybridization assay, to make a dimer of said two nucleic acids joined by a linking moiety of Formula III -O(PO₂)NH(R₁₁)SS(R₂₁)NH(PO₂)O- _III which method comprises combining said first and second nucleic acids in a first aqueous solution with a disulfide reducing-effective concentration of a reducing agent selected from the group consisting of dithiothreitol, dithioerythritol, Mι(BH₄) and Mι(BH₃CN), wherein Mi is selected from the group consisting of Li⁺, Na⁺ and K⁺, and then subjecting said first solution to oxidizing conditions, whereby said dimer of said two nucleic acids is formed, the improvement which comprises dialyzing said first solution against a second aqueous solution, which has a composition such that (i) the concentration in said first solution of a reducing agent selected from the group consisting of dithiothreitol, dithioerythritol, Mι(BH₄) and Mι(BH₃CN), wherein M₁ is selected from the group consisting of Li⁺, Na⁺ and K⁺, is maintained at a disulfide reducing-effective level and (ii) the ratio of the molar concentration in said first solution of the compound of Formula IV

HS(R₁₂)NH₂ IV to the molar concentration in said first solution of all derivatives of said first nucleic acid and the ratio of the molar concentration in said first solution of the compound of Formula V

HS(R₂₂)NH₂ v to the molar concentration in said first solution of all derivatives of said second nucleic acid decrease.

2 . The improvement according to Claim 1 wherein each of R₁₁, R₁₂, R₂₁ and R₂₂ is an n-alkylene group of 2 - 8 carbon atoms and wherein the reducing agent present in said first aqueous solution at a disulfide reducing-effective concentration prior to said dialysis is maintained at a disulfide reducing-effective concentration during said dialysis and is selected from the group consisting of dithiothreitol and dithioerythritol.

3. The improvement according to Claim 2 wherein said dialysis lowers the molar concentration of the compound of Formula IV

HS(R₁₂)NH₂ IV in said first solution to less than 10 percent of the molar concentration of all derivatives of said first nucleic acid in said first solution and lowers the molar concentration of the compound of Formula V

HS(R₂₂)NH₂ v in said first solution to less than 10 percent of the molar concentration of all derivatives of said second nucleic acid in said first solution and wherein, subsequent to said dialysis against said second solution, O₂ is introduced into said first solution to oxidize thiols to disulfides therein while said first solution is dialyzed against a third aqueous solution with a composition such that the concentration of reducing agent in said first solution is reduced to less than a disulfide reducing-effective concentration.

4. The improvement according to Claim 3 wherein O₂ introduced into said first solution is introduced from air.

5. The improvement according to Claim 1 wherein said first terminally derivatized nucleic acid is a replicative RNA derivatized at the 5'-carbon of its 5'-nucleotide and said second terminally derivatized nucleic acid is a nucleic acid of 10 - 100 bases which has the sequence of an affinity molecule for a nucleic acid analyte and is derivatized at the 5'-carbon of its 5'-nucleotide.

6. The improvement according to Claim 2 wherein said first terminally derivatized nucleic acid is a replicative RNA derivatized at the 5'-carbon of its 5'-nucleotide and said second terminally derivatized nucleic acid is a nucleic acid of 10 - 100 bases which has the sequence of an affinity molecule for a nucleic acid analyte and is derivatized at the 5'-carbon of its 5'-nucleotide.

7. The improvement according to Claim 3 wherein said first terminally derivatized nucleic acid is a replicative RNA derivatized at the 5'-carbon of its 5'-nucleotide and said second terminally derivatized nucleic acid is a nucleic acid of 10 - 100 bases which has the sequence of an affinity molecule for a nucleic acid analyte and is derivatized at the 5'-carbon of its 5'-nucleotide.

8. The improvement according to Claim 4 wherein said first terminally derivatized nucleic acid is a replicative RNA derivatized at the 5'-carbon of its 5'-nucleotide and said second terminally derivatized nucleic acid is a nucleic acid of 10 - 100 bases which has the sequence of an affinity molecule for a nucleic acid analyte and is derivatized at the 5'-carbon of its 5'-nucleotide.

9. The improvement according to Claim 5 wherein each of R₁₁, R₁₂, R₂₁ and R₂₂ is -(CH₂)₂-.

10. The improvement according to Claim 6 wherein each of R₁₁, R₁₂, R₂₁ and R₂₂ is -(CH₂)₂-.

11. The improvement according to Claim 7 wherein each of R₁₁, R₁₂, R₂₁ and R₂₂ is -(CH₂)₂-.

12. The improvement according to Claim 8 wherein each of R₁₁, R₁₂, R₂₁ and R₂₂ is

-(CH₂)₂-.