Compositions and methods for functionalizing nucleic acids

Abstract

A composition and a method for 5'-labelling polynucleotides undergoing solid phase synthesis wherein a phosphoramidite of an .omega.-hydroxylamine is condensed to a support-bound polynucleotide.

Description

BACKGROUND

The present invention pertains in general to compositions and methods for functionalizing nucleic acids and in particular to compositions and methods for the 5' labelling of polynucleotides attached to a support.

Attachment of a detectable label to a polynucleotide permits detection and quantitation of a nucleic acid. The attachment of detectable labels to nucleic acids is particularly important in hybridization assays wherein a labelled polynucleotide probe is used to search a sample for a target nucleic acid which has a complementary nucleotide sequence and which has been immobilized by hybridization to a support-bound polynucleotide probe. Labelling with a reactive group permits attachment of reporter groups to or immobilization of the polynucleotide.

One approach to labelling a polynucleotide probe for use in hybridization assays involves binding a radioisotope (e.g., .sup.32 P, .sup.3 H, or .sup.125 I) to the probe. However, difficulties inherent in the two methods of detecting radioactive labels limit the usefulness of the technique. Autoradiography is a time-consuming procedure which relies upon reduction of silver ions to form silver grains in a photographic emulsion; and scintillation counting, the other detection technique, requires expensive equipment and a certain amount of delay as well. Furthermore, radioisotopes require special handling for safety reasons. Some radioactive isotopes, such as .sup.125 I, have relatively short shelf-lives, which further limit their usefulness in a clinical diagnostic setting.

In non-radioactive labelling systems, a probe is "labelled" with a reporter group which is associated with a signal to enable detection. A reporter is an agent which is used to associate a signal with a probe for indicating the presence or location of the probe. The signal itself, which is directly perceptible, may be generated by a separate or separable signal molecule. A label is properly a type of reporter which incorporates a signal.

One approach to the attachment of labels to probes is described in Ward, et al., European patent application No. 63,879. Ward discloses the preparation of probes having a biotin reporter molecule covalently attached to a purine or pyrimidine ring. Selected biotinylated purines and pyrimidines are then directly incorporated within the phosphodiester backbone of the nucleic acid of the probe by enzymatic means. However, enzymatic techniques are costly and difficult to perform.

Other approaches link a label to a probe by way of a protein. Single-stranded polio virus RNA is naturally linked to a protein which may be reacted with the N-hydroxysuccinimidyl ester of biotin to obtain an RNA probe having a biotinylated reporter group detectable by specific attachment of avidincoated spheres. Richards, et al., Proc.Natl.Acad.Sci. (USA), 76: 676-680 (1979). Similarly, biotin-labelled cytochrome c may be coupled to RNA by reaction in the presence of formaldehyde and thereafter labelled with avidin-coated spheres. Manning, et al., Chromosoma (Berl.), 53: 107-117 (1975). Nevertheless, because not all nucleic acids desired to be labelled are naturally associated with proteins and because the location and amount of cytochrome c binding to a nucleic acid is not readily predictable, it is desirable to have a chemical synthetic technique for end-labelling.

In one such chemical synthetic technique, nucleic acids are converted to 3'-aldehydes by oxidation and condensed with alkyldiamines or polyamines to provide a reporter group for the attachment of biotin. Broker, et al., Nucleic Acids Res., 5: 363-384 (1978). Similarly, aldehydes generated by the periodate oxidation of nucleic acids may be used to couple fluorescent labels to the nucleic acids. Bauman, et al., J. Histochem.Cytochem., 29: 227-237 (1981). However, it is desirable to have a technique for attaching reporter groups to polynucleotides attached to a support in an automated process for nucleic acid synthesis.

In yet another approach to 5' labelling, biotin is converted to 2-(biotinylamido)ethanol and condensed to a phosphorylated, polymer-supported nucleotide. The condensation of the aminoethanol derivative of biotin to the 5' hydroxyl group of a ribose ring gives a stable phosphodiester bond upon deprotection of the nucleotide. Kempe, et al., Nucleic Acids Res., 13: 45-57 (1985). Nevertheless, specific reporter groups are attached by this approach so that the approach does not permit preparation of an oligonucleotide with a reactive functionality which may later be used to attach a variety of desired reporter groups.

Nucleotides in solution have been amine-functionalized by condensation with protected 6-amino-1-hexanol phosphate. Barker, et al., J.Biol.Chem., 22: 7135-7147 (1972). However, these procedures are difficult to perform and have not been integrated with solid-phase synthesis.

In another approach to binding nucleotides to supports, in the purification of nucleases by affinity chromatography, single nucleotides, 3'-derivatized with p-aminophenol are attached to a gel matrix by a linker. The linker is formed by attaching 3,3' diaminodipropylamine to the matrix using cyanogen bromide and azide. The resulting amine-functionalized gel is treated with succinic anhydride and then coupled to the amine-functionalized nucleotide. Cuatrecasas, J.Biol.Chem., 12: 3059-3065 (1978). Nevertheless, the manufacture of the amine-functionalized nucleotide itself has been performed in solution by tedious procedures. See, e.g., Barker, et al., J.Biol.Chem., 22: 7135-7147 (1972).

In an approach to 5' labelling, a 5-aminoalkyl phosphoramidate derivative of a kinased, unprotected oligonucleotide is prepared in solution. Kination of a deprotected oligonucleotide is accomplished by T4 polynucleotide kinase. The kinased oligonucleotide is reacted with an imidazole and a diamino alkane. The resulting 5'-aminoalkyl phosphoramidite DNA is reacted with biotin-N-hydroxysuccinimidyl ester in potassium N,N-dimethylformamide to produce a biotin-labelled oligodeoxyribonucleotide. Chollet, et al., Nucleic Acids Res., 13: 1529-1541 (1985). Although this method makes use of an oligonucleotide synthesized using the solid-phase phosphoramidite method or the solid-phase phosphotriester method, the oligonucleotide is deprotected, detached from the solid phase, isolated, and purified before functionalization. After functionalization, a second isolation and purification is required. Because each isolation and purification entails losses and is time-consuming, it is desirable to functionalize a polynucleotide during solid-phase synthesis. Furthermore, the kination reaction suffers from the problems of cost and difficulty generally associated with enzymatic procedures.

Therefore, there is a need for a method and composition for the generic attachment of reporter groups to polynucleotides undergoing solid phase synthesis.

SUMMARY OF THE INVENTION

A composition according to the present invention involves an end-label for attachment to a nucleic acid during solid phase synthesis. The end-label is a phosphoramidite of a hydroxylamine.

A method according to the present invention involves attaching an end-label to a nucleic acid during solid phase synthesis. Specifically, a phosphoramidite of a hydroxylamine is condensed to a support-bound deoxyoligonucleoside.

DETAILED DESCRIPTION

According to the present invention, a phosphoramidite of a hydroxylamine, preferably an .omega.-hydroxylamine, is condensed to a oligonucleoside bound to a support. The support may be of any sort useful for the solid phase synthesis of nucleic acids, including, but not limited to, silicates and cellulose. The polynucleoside may be a ribonucleoside or a deoxyribonucleoside.

An .omega.-hydroxylamine according to the present invention may generally be of any length, but is preferably from two to ten carbons in length. The .omega.-hydroxylamine may be alkyl, aryl, cycloalkyl, or generally of any structure which permits condensation of the hydroxyl moiety to an oligonucleoside and also permits reaction of the amine moiety with a reporter group without steric hindrance. Specifically, preferred .omega.-hydroxylamines include ethanolamine, propanolamine, butanolamine, pentanolamine, hexanolamine, heptanolamine, octanolamine, nonanolamine, decanolamine, and p-aminophenol.

Reporter groups for use with the present invention include biotinyl, N-hydroxysuccinimide, and fluorescein isothiocyanate. A particular advantage of the present invention over methods wherein a reporter group is directly attached to the synthetic oligonucleotide, is that labels or reporter groups are not required to withstand the harsh conditions required to deprotect a synthetic polynucleotide. Thus, by the method according to the present invention, an oligonucleotide may be 5' labelled with biotin, hapten, antigen, and fluorescent or chemiluminescent functionalities, among others.

For use in the hybridization procedures of the following examples, a single-stranded phage containing either the (+) plus (coding) strand or the (-) minus (anticoding) strand of the Herpes Simplex Virus Type I (HSV-I) Glycoprotein D (gD) gene was employed as the target sequence. A portion of the double-stranded gene sequence is set out in Table I below, the bottom strand be-ng the anticoding strand. This sequence has been published in Watson, et al., Science, 218: 381-384 (1982). Portions of the plus strand have been employed as probes according to the present invention. These single-stranded probe sequences have been designated on Table I by a lettered line drawn above the coding strand of the gene.

TABLE I__________________________________________________________________________ ##STR1## ##STR2## ##STR3## ##STR4## ##STR5## ##STR6## ##STR7## ##STR8## ##STR9## ##STR10## ##STR11## ##STR12## ##STR13## ##STR14## ##STR15## ##STR16## ##STR17## ##STR18## ##STR19## ##STR20## ##STR21## ##STR22## ##STR23## ##STR24## ##STR25## ##STR26## ##STR27## ##STR28## ##STR29## ##STR30## ##STR31## ##STR32## ##STR33## ##STR34## ##STR35## ##STR36## ##STR37## ##STR38## ##STR39##__________________________________________________________________________

The target used in the examples is a single-stranded phage, phage 2 (.PHI.2), which contains 1,454 bases of the HSV-I glycoprotein D (gD) gene (i.e., bases 67 through 1,287, initiation codon nucleotide number 241 cloned into a plasmid, M13mp18. The minus strand sequence of gD in .PHI.2 is employed as a target complementary to the (+) plus strand probes identified above.

The following examples describe a series of experiments demonstrating various aspects of the present invention.

Example 1 shows the effectiveness of labelling of an antibody-coated polynucleotides according to the present invention. Example 2 illustrates the use of a probe labelled according to the present invention for capturing a hybridization sandwich comprising two probes bound to a target.

EXAMPLE 1

A polynucleotide having the sequence

5'-P-ACC GAA TGC TCC TAC AAC AAG TCT C-3' was labelled with an antigen at its 5' end employing the method according to the present invention.

Specifically, a first probe was oligonucleotide G as described above. Such 5' labelling of oligonucleotide G may be accomplished with fluorescein.

Oligonucleotide G was 5' fluorescein labelled by reacting a 5' amine functionalized oligonucleotide G with fluorescein isothiocyanate. The 5' amine functionalized oligonucleotide G was formed by reacting oligonucleotide G bound by its 3' end to a solid support with a phosphoramidite having the general formula [(CH.sub.3).sub.2 CH ].sub.2 NP(OCH.sub.3)O(CH.sub.2).sub.8 NH(DMT) wherein DMT is a dimethoxytrityl group.

In the synthesis of this phosphoramidite, about 8 ml of diazomethane-ether solution were added to 159.2 mg (1 mmole) of .omega.-aminocaprylic acid (available from Aldrich Chemical, Milwalkee, Wisconsin) in 10 ml of methanol. The methanol was evaporated to yield 174.9 mg of m-aminocaprylic acid methyl ester. Next, 173 mg (1 mmole) of the .omega.-aminocaprylic acid methyl ester, 1 mmole of dimethoxytrityl chloride, and 1 mmole of diisopropylethyl amine were added to 5 ml of anhydrous tetrahydrofuran under an argon atmosphere at 0.degree. C. This mixture was warmed to 25.degree. C. and stirred for 1 hour. The solvent was evaporated and the crude product was diluted with 50 ml of ethyl acetate and washed successively with two portions of water, saturated bicarbonate, and brine. The product was dried over anhydrous magnesium sulfate and evaporated to yield 460 mg of a dimethoxytrityl derivative of the .omega.-aminocaprylic acid methyl ester (ACAM-DMT). To 0.17 mmoles of ACAM-DMT in 1 ml of anhydrous tetrahydrofuran under an argon atmosphere at -78.degree. C. was added 1.24 ml of 1 molar lithium aluminum hydride in tetrahydrofuran. This reaction mixture was stirred for 5 minutes at -78.degree. C. and was then stirred for 30 minutes at 25.degree. C. before being diluted with 10 ml of 5% H2O in tetrahydrofuran, 200 ml of ether, 3 g of cellite, and 0.5 g of anhydrous magnesium sulfate. The resulting mixture was stirred for 30 minutes and filtered to yield an alcohol having the general formula HO(CH.sub.2).sub.8 NH-DMT.

To 0.72 mmoles of HO(CH.sub.2).sub.8 NH-DMT in 10 ml of anhydrous dichloromethane was added 0.76 mmoles of diisopropyl ethyl amine and 0.76 mmoles of chloro-N,N'-diisopropylaminomethoxy phosphene (as available from American Bionuclear, Emeryville, California). This mixture was stirred for 40 minutes at 25.degree. C., and then diluted with 50 ml of ethyl acetate and washed with four portions of brine. The product of this reaction was coupled to a support-bound deoxynucleoside G employing the phosphoramidite synthesis technique for use with deoxynucleoside N,N-diisopropyl amino methoxyphosphines in polymer support deoxyoligonucleotide synthesis according to McBride, et al., Tetrahedron Letters, 24: 245 (1983).

Quantitation of binding of the reporter group according to the present invention to the deoxyoligonucleoside was determined by a demethoxytrityl assay as follows: ##EQU1## wherein A.sub.498 is the absorbence of the solution containing the removed demethoxytrityl components measured at a wavelength of 498 nm. Coupling was determined to be in excess of 90% by this method.

EXAMPLE 2

A second probe, oligonucleotide A, was labelled with .sup.32 P according to the procedure of Maniatis, et al., Cell, 15: 687 (1978). The specific activity of the probe on the date of use was 3.2.times.10.sup.6 cpm/picomole.

Oligonucleotide G without a 5' fluorescein label was used as a first probe control. A second control probe, having the sequence 5' CATGATCTTGCGGTCGGATTCTTC 3', which does not complement any of the target sequence, was also .sup.32 P-labelled and had a specific activity on the date of use of 3.2.times.10.sup.6 cpm/picomole.

The target used was single-stranded .PHI.2. Singlestranded .PHI.2 is complementary to the first and second probes and to the first probe control, but not to the second control probe.

As a support, one-quarter-inch polystyrene beads of the sort available from Pierce Chemical, Rockland, Illinois, were coated with fluorescein antibody (anti-fluorescein). Anti-fluorescein production was induced in rabbits. The antifluorescein was purified by ammonium sulfate precipitation, followed by DEAE cellulose chromatography. In solution, the anti-fluorescein had an affinity of approximately 10.sup.12 and quenched the fluorescence of fluorescein by about 99%.

To prepare an anti-fluorescein-coated bead, the bead is cleaned by ultrasonication for 15 seconds in 10 mM NaHCO.sub.3 buffer at pH 8. After ultrasonication, the beads are washed in deionized water until all fines are removed. Approximately 200 heads are covered by 40 ml of 10 mM NaHCO.sub.3. Next, 7 ml of purified anti-fluorescein at a concentration of 0.57 mg/ml is added. The beads are incubated for approximately 65 hours at room temperature. After incubation, the beads are washed with deionized water and air-dried on a suction filter.

Each of the anti-fluorescein coated beads is capable of binding greater than 1.4 picomole of fluorescein, as demonstrated by incubation of single beads with 1.5 ml of 1 nM fluorescein in TDX buffer (0.1 M NaPO.sub.4, pH 7.5; 0.1% NaN.sub.3 ; 0.1% bovine gammaglobulin). During 20 hours of incubation at 25.degree. C., 97% of the fluorescein was removed from solution. After washing the beads three times in 5 ml of deionized water and blotting the beads dry after each wash, the beads were incubated in 0.1 M NaOH for 10 minutes, in which 60% of the originally applied amount of fluorescein was released into solution. Thus, each bead has approximately 0.9 picomole of fluorescein binding capacity.

(1) A series of capture experiments employing 5'-fluorescein-labelled oligonucleotides, 5'-biotin-labelled oligonucleotides (both 3'-.sup.32 P end-labelled), and kinased .sup.32 P-labelled oligonucleotides and polystyrene beads coated with anti-fluorescein were run under the following conditions.

With 200 .mu.g/ml denatured sheared salmon sperm DNA (Sigma Chemical Company, St. Louis, Missouri) containing 1 picomole of one of the .sup.32 P-labelled oligonucleotides, 100 .mu.l of TDX buffer (0.1 M sodium phosphate, pH 75; 0.1% NaN.sub.3 ; and 0.01% bovine gamma globulin, Sigma Chemical Company, St. Louis, Missouri) was mixed. An anti-fluorescein-coated polystyrene bead was added to this solution. After incubating this system for 18 hours at 25.degree. C., the bead was removed and washed for 5 minutes in 1 ml of TDX buffer at 25.degree. C. The bead was then counted in a scintillation counter.

The stability of the antibody complex on the bead was tested by washing the bead for 5 minutes at increasing temperatures. The capture efficiency and stability of a series of such beads is shown in Table II.

TABLE II______________________________________Percent cpm CaptureComplexesTemper- 5' fluorescein- 5' biotin- 5' .sup.32 P-labelledature labelled complex labelled complex complex______________________________________25 63 4 335 61 1 045 56 0 055 51 0 065 42 0 075 35 0 085 20 0 095 0 0 0______________________________________

As illustrated by Table II, these beads have a high capture efficiency and stability of the sort which is useful in a hybridization capture system. Because little or no biotin or .sup.32 P-labelled oligonucleotide binds to these beads, indicating little non-specific binding to the beads, the background in such a system is very low.

(2) In order to more precisely determine the rate of capture of a fluorescein-labelled oligonucleotide by a fluorescein antibody-coated bead, each of a series of beads was incubated for a different amount of time with 1 picomole of 5'-fluorescein-labelled oligonucleotide A which had been 3' end-labelled with .sup.32 P. The percent of capture was determined for each bead and the results are shown below in Table III.

TABLE III______________________________________ Percent OligonucleotideTime Capture______________________________________0 015 minutes 2030 minutes 451 hour 482 hours 753 hours 914 hours 905 hours 886 hours 867 hours 858 hours 82..20 hours 68______________________________________ As illustrated in Table III, 90% of the 5' fluorescein-labelled oligonucleotide is captured by the bead in 2 to 3 hours. The slow decline in the amount of radiolabel over time on the bead most likely represents a small amount of leakage of the antibody from the bead.

(3) Experiment 1. The capture-efficiency of the anti-fluorescein-coated beads being established, 1 picomole of the first probe (5'-fluorescein-labelled oligonucleotide G), 1 picomole of the second probe (.sup.32 P-labelled oligonucleotide A), specific activity on date of use 3.2.times.10.sup.6 cpm/picomole), and 1 picomole of the target (.PHI.2 SS, complementary to both the first and second probes) were diluted to 50 .mu.l with 5.times.SSPE diluted from 20.times.SSPE (3.6 M NaCl; 0.23 M NaH.sub.2 PO.sub.4, pH 7.5; and 20 mM EDTA). This hybridization solution was incubated for 3 hours at 50.degree. C. This hybridization solution was diluted with 100 .mu.l of TDX buffer and one anti-fluorescein-coated bead was added. After incubation for 3 hours at 25.degree. C., the bead was washed with 1 ml of TDX buffer for 5 minutes at 37.degree. C. and was re-washed with 1 ml of TDX buffer for 5 minutes at 37.degree. C. before counting in a scintillation counter.

Control Experiments. Three control experiments were run according to the same protocol but with the following modifications. In a first control experiment (Control 1), 5' fluorescein-labelled oligonucleotide G, as a first probe, and 5' .sup.32 P-labelled oligonucleotide A, as a second probe, were incubated with the anti-fluorescein-coated bead in the absence of any target. A second control experiment (Control 2) involved the use of 1 picomole of unlabelled oligonucleotide G as a first probe for the fluorescein-labelled oligonucleotide G of experiment 1. Finally, a third control experiment (Control 3) was performed with 1 picomole of 5'-fluorescein-labelled oligonucleotide G, as a first probe, 1 picomole of a .sup.32 P-labelled oligonucleotide designated 32-B.sub.2 (the sequence of which is complementary to .PHI.2 SS), as a second probe, and 1 picomole of .PHI.2 SS as a target.

The results of these experiments are summarized in Table IV.

TABLE IV______________________________________ % .sup.32 P OligonucleotideExperiment Bound to the Bead______________________________________Experiment 4.2Control 1 0.002Control 2 0.07Control 3 0.22______________________________________

A comparison of Experiment and Control 1 indicates that the hybrid comprising fluorescein-labelled oligonucleotide G, .PHI.2 SS, and .sup.32 P-labelled oligonucleotide A may be selectively captured by an anti-fluorescein-coated solid support. Controls 2 and 3 demonstrate that in the absence of the correct antigenlabelled first probe or in the absence of the correct target complementary second probe, a hybrid is not effectively generated or captured.

It is expected that numerous modifications and variations will occur to those skilled in the art upon consideration of the present invention. For example, although .omega.-hydroxylamines are exemplified herein, any molecule having hydroxyl and amine groups available for respectively forming a bond with a support-bound nucleoside and a reporter group may be employed. Consequently, it is intended that the present invention be given the full scope of the appended claims.

Claims

1. An end-label for attachment to a nucleic acid during solid phase synthesis comprising a phosphoramidite of a hydroxylamine condensed to a polynucleoside.

2. The label according to claim 1 wherein said hydroxylamine contains from 1 to 10 carbon atoms.

3. The label according to claim 2 wherein said hydroxylamine is selected from the group comprising ethanolamine, propanolamine, butanolamine, pentanolamine, hexanolamine, heptanolamine, octanolamine, nonanolamine, decanolamine, and p-aminophenol.

4. A method of attaching an end-label to a nucleic acid during solid phase synthesis comprising the steps of:

preparing a phosphoramidite of a hydroxylamine; and

condensing the phosphoramidite of a hydroxylamine to a support-bound polynucleoside.