WO2000077250A2 - Method for circularizing oligonucleotides around a double stranded nucleic acid, resulting structures and uses thereof - Google Patents

Abstract

The invention concerns a method for circularizing an oligonucleotide which consists in: incubating a double stranded nucleic acid comprising a target sequence, with a single stranded oligonucleotide comprising a central part capable of being fixed on said target sequence, bound by spacer segments to 5' and 3' terminal sequences, said sequences being: either a) complementary to the sequence of an oligonucleotide matrix; or x) partly complementary to each other, so as to be capable of being hybridized to each other forming a double strand while preserving a single stranded end, which is capable of being hybridized with the terminal sequence of another oligonucleotide, called loop-oligonucleotide, which has a hairpin structure, with one single stranded part and a folded part forming a double strand; contacting, in the first case a) the complex formed with said oligonucleotide matrix, in conditions enabling the hybridization of the 3' and 5' ends, adjacently, with the matrix, and in the case x), contacting the complex formed with said loop-oligonucleotide, in conditions enabling, through hybridization of the free ends of the single stranded oligonucleotides and of the loop-oligonucleotide, the juxtaposition of the 3' and 5' ends of said oligonucleotides; ligating the hybridized ends so as to circularize the oligonucleotide around the double stranded nucleic acid and, optionally, isolating the circularized structure. The invention also concerns the applications and the resulting structures obtained for diagnostic and therapeutic purposes.

Description

 PROCESS FOR CIRCU ARIZATION OF OLIGONUCLEOTIDES AROUND A DOUBLE-STRANDED NUCLEIC ACID, THE STRUCTURES OBTAINED AND THEIR

APPLICATIONS

The subject of the invention is a method of circularization of ¹ oligonucleotides around a double-stranded nucleic acid, the structures obtained and their applications.

Short chain oligonucleotides have previously been described which can specifically bind to double helix DNA to form a triple helix or triplex structure. The fixation takes place at the level of the large groove of the DNA double helix with establishment of non-covalent interactions.

Such oligonucleotides have been developed with the aim of artificially controlling the expression of specific genes at the level of transcription, but can be used for other applications, such as for example the purification of plasmids.

It has also been proposed to chemically modify the triplex-forming oligonucleotides to induce irreversible modifications of the target nucleic acids. Target cleavage can be achieved with an oligonucleotide covalently linked to an Fe-EDTA or Cu-phenanthroline complex, while psoralen-oligonucleotide conjugates can be crosslinked to their target after UV irradiation.

Interest has been focused in recent years on circular oligonucleotides due to their topological properties and recognition of nucleic acids. Compared to linear oligonucleotides, circular oligonucleotides make it possible to design improved genetic analysis tools, in particular thanks to their efficient amplification by loop replication.

Binding of a preformed circular oligonucleotide to a short double stranded DNA fragment can be accomplished in two topologically distinct ways.

In one of these pathways, the oligonucleotide binds in the large DNA groove while remaining outside this duplex. In the other way, the circular oligonucleotide circles the duplex after one of the ends of it has slipped into the oligonucleotide circle. This second structure can be inserted into a longer circular double-stranded DNA fragment, but this procedure requires the intervention of ligation steps whose yield and efficiency are limited.

The invention relates to a new method allowing the circularization of an oligonucleotide directly around a double-stranded nucleic acid, whatever its length and its topological state and, without the need to modify the target d enzymatically or chemically. 'nucleic acid. This process is based on the development of a three-strand complex, comprising an oligonucleotide which can be wound around the double strand of nucleic acid, and which can be circularized.

The circularization process according to the invention is characterized in that it comprises - the incubation of a double-stranded nucleic acid comprising a target sequence, with a single-stranded oligonucleotide comprising a central part capable of binding to said target sequence, linked by spacer sequences to 5 'and 3' terminal sequences, these sequences being either a) complementary to the sequence of an oligonucleotide matrix,

- or x) partially complementary to one another, so as to be able to hybridize with one another by forming a double strand while retaining a single-stranded end, which is capable of hybridizing to the terminal sequence of another oligonucleotide, called a loop oligonucleotide, which has a hairpin structure, with a single-strand part and a folded part which forms a double strand, said incubation step being carried out under conditions allowing fixing the single-stranded oligonucleotide at the target sequence by formation of reversible bonds, which leads to the winding of the oligonucleotide around its target, and in case x), after a heating step, in order allow the opening of the oligonucleotide and its winding around the target DNA before rehybridization of its ends,

the bringing into contact in the first case a) of the complex formed with said oligonucleotide matrix, under conditions allowing the hybridization of the 5 ′ and 3 ′ ends, adjacent to the matrix, and in case x), the bringing the complex formed into contact with said loop oligonucleotide, under conditions allowing, by the hybridization of the free ends of the single-strand oligonucleotide and of the loop oligonucleotide, the juxtaposition of the 5 'and 3' ends of these oligonucleotides , ligation of the hybridized ends, so as to circularize the single-stranded oligonucleotide (case a) or

The single-stranded oligonucleotide and the looped oligonucleotide (case x), around the double-stranded nucleic acid and, if desired,

- Isolation of the circularized structure, where the target nucleic acid is padlocked by the circular oligonucleotide, followed where appropriate by its purification.

It will be observed that, advantageously, this circularization technique can be carried out without chemical or enzymatic modification of the target nucleic acid and only involves a short sequence on the target as a recognition site.

The double-stranded nucleic acid used according to the invention consists, according to one embodiment, of a DNA in double helix.

It should be noted that such double helix DNA can be a linear or circular plasmid, and in particular a supercoiled circular plasmid.

As regards the single-stranded oligonucleotide, which must be circularized, in accordance with the invention, around the nucleic acid target, this oligonucleotide must be of sufficient length to, on the one hand, bind to the target and , on the other hand,

- case a) bind to the oligonucleotide matrix, case x) allow the hybridization of its ends and the hybridization of the loop oligonucleotide. It can thus comprise more than 20 nucleotides, in particular from 20 to several hundred nucleotides. An oligonucleotide comprising 20 to 300 nucleotides, in particular 30 to 100, is advantageously used with, as central part, intended to be fixed on the nucleic acid target, a sequence, of 10 to 25 nucleotides.

According to a provision of the invention, one can use single-stranded oligonucleotides in which the chain comprising nucleic bases is not constituted only of natural internucleoside bonds of phosphodiester type, but can comprise for example bonds of phosphorothiate type, phosphora idate, methylphosphonate or peptide nucleic acid, or mixtures of these bonds.

It is also possible to use single-stranded oligonucleotides in which the nucleotides are chemically modified at the sugar level, in particular at the 2 ′ position, for example by amino or methoxy groups, or at the base level, for example 5-propinyl-pyrimidines.

The single-stranded oligonucleotide used may comprise one or more non-nucleoside links, for example of the hexaethylene glycol type, or be covalently linked to one or more molecules stabilizing the triple helices, such as for example the benzo-indolo-quinolines, benzo-quino-quinoxalines, benzo-pyrido-indoles and benzo-pyrido-quinoxalines.

According to another arrangement of the invention, the single-stranded oligonucleotide (case a) or the looped oligonucleotide (case x) can be chemically modified so as to be covalently linked to one or more low molecular weight ligands.

They may, for example, be fluorescent molecules, markers such as biotin, or peptides. Binding of a peptide can be carried out directly on a base, for example using a thymine carrying an amino group in position 5 and a bifunctional binding agent, such as the N-hydroxysuccinimide ester of 4- (N-maleimidomethyl acid). ) - cyclohexane-1-carboxylic (SMCC).

In general, the single-stranded oligonucleotide binds to the target nucleic acid sequence and wraps around the double-strand, if necessary, constituted, as indicated above, by DNA in double helix.

It will be observed that the oligonucleotide can bind simultaneously to two double-stranded sequences carried by two different nucleic acids, in particular two different DNAs, which can be in double helix, and for example plasids.

In one embodiment of the invention, the attachment of the central part to a target DNA sequence in double helix leads to the formation of a triple helix.

The target sequence on the double-stranded DNA is then advantageously an oligopurine oligopyrimidine sequence.

The binding of the oligonucleotide to the double-stranded nucleic acid, in particular the formation of a triple helix, can be induced by adding to the reaction medium a ligand of low molecular weight. By decreasing the amount of ligand, or even removing the ligand, it is possible to detach the ring from its target and control its movement along the target.

In a preferred embodiment of the invention, the oligonucleotide binds to an oligopurine sequence of the target.

The oligonucleotide can be linked in parallel to the oligopurine strand of the oligopurine sequence. oligopyrimidine or, as a variant, antiparallel with respect to the oligopurine strand of said sequence according to the triplets engaged.

The oligonucleotide matrix on which the ends of the single-stranded oligonucleotide a) will hybridize contains 10 to 36 nucleotides. The ends which hybridize with such a matrix then generally comprise, each 5 to 18 nucleotides.

The loop oligonucleotide on which the free end of the single-stranded oligonucleotide x) will hybridize contains from 16 to 100 nucleotides. The single-stranded ends of these two oligonucleotides hybridize to form a double helix which generally comprises from 4 to 40 bp.

To carry out the incubation, the conditions, in particular pH, temperature, ionic strength, are chosen so as to obtain the desired fixation. Generally, the operation is carried out at a pH of 3 to 9, in particular 4.5 to 8.5, in particular 6 to 8, with or without heating.

The hybridization and ligation steps are carried out after the incubation step, in the same or different buffer than that used for the incubation.

The temperature depends on the ligation agent used.

It can be an enzyme. We will then advantageously operate at a temperature of 16 to 45 ° C with an E. coli ligase, or again from 35 to 65 ° C with a thermostable ligase like that sold under the brand Ampligase.

Alternatively, a chemical agent is used for the circularization of the oligonucleotide. Examples include N- (3-dimethylaminopropyl) -N '-ethylcarbodiimide (EDC), cyanogen bromide, or cyanoimidazole. With such agents, the ligation can be carried out at temperatures below 10 ° C., in particular of the order of 4 ° C.

Alternatively, a 5'-iodo and 3 '-phosphorothioate oligonucleotide is used. The chemical circularization reaction is initiated when these ends are juxtaposed, which is obtained by hybridization on a matrix. This reaction can be carried out at temperatures from 4 to 50 ° C, at pH varying from 3 to 8.

The ionic concentrations can vary from 0 to 20 mM approximately, for ions such as Mg ", and from 0 to 1M for ions such as Na ^* or K ^* . Said oligonucleotide matrix, (case a) or said looped oligonucleotide (case b) are added after the incubation and just before the addition of the ligation agent.

It is observed that in case x), two ligation reactions are carried out, 1 circular oligonucleotide being the result of the chain circularization of two oligonucleotides.

In order to isolate the complex structure, the ligase is inactivated, and advantageously the purification of the complex. This purification step can be carried out by loading the sample on a gel, for example a 1% agarose gel, by chromatography, exclusion column, or using beads, optionally using a receptor for a ligand attached to the single strand oligonucleotide, the loop oligonucleotide or the template oligonucleotide.

The invention relates, as new products, to the intermediate structures and the circularized structures formed.

It thus falls within the scope of the invention to intermediate nucleic acid structures characterized in that they comprise a single-stranded oligonucleotide as defined above, fixed by reversible bonds to the target nucleic acid sequence.

More specifically, the invention relates to such structures in which the nucleic acid target is a double helix DNA, and in particular a plasmid, in particular a supercoiled plasmid. In such structures, the oligonucleotide is attached to the nucleic acid target, wrapped around that target.

In particular, the oligonucleotide is attached to DNA in a double helix and forms a triple helix.

In such a structure, the target sequence of the double-stranded DNA is advantageously an oligopurine sequence. oligopyrimidine.

These structures include in particular a sequence with 10 to 25 triplets, formed by the oligopurine-oligopyrimidine sequence of the target to which the oligonucleotide is attached.

In these triple helices, the oligonucleotide is fixed parallel to the strand containing the oligopurine sequence, the triplets formed being for example T.A x T; C.G x C; C.G x G.

Alternatively, the oligonucleotide is fixed antiparallel with respect to the oligopurine sequence, the triplets formed being T.A x T; T. x A; C.G x G.

In other structures according to the invention,

The oligonucleotide binds simultaneously to two different nucleic acid sequences, carried by two different nucleic acids, in particular two double helix DNAs, in particular two plasmids.

In still other intermediate structures, the 2 ends of the oligonucleotide are also hybridized on the oligonucleotide matrix and advantageously connected enzymatically or chemically. The corresponding structures therefore comprise a triplet sequence involving the target nucleic acid, linked on each side, by spacer sequences, to a doublet sequence involving the oligonucleotide matrix.

Advantageous structures comprise on the one hand a target comprising 10 to 25 triplets and on the other hand a matrix comprising 10 to 36 doublets, preferably from 14 to 20.

In other intermediate structures, the two ends of the oligonucleotide are hybridized to one another and advantageously linked enzymatically or chemically to the two ends of a looped oligonucleotide. The corresponding structures therefore comprise a triplet sequence involving the target nucleic acid, linked on each side, by spacer sequences, to a sequence in rod-loop structure (or hairpin).

The invention relates in particular to the structures as defined above but released from the matrix, comprising a double-stranded nucleic acid encircled by a ring formed by a single-stranded oligonucleotide.

In the circularized structures obtained the ring is fixed noncovalently to the nucleic acid target. Alternatively, it is free to move along the target nucleic acid. As indicated above, the movement of the ring relative to the target can be controlled by adding or removing a ligand specific for the complex formed between the sequence of the oligonucleotide involved in the binding and that of the target. In these various structures, the oligonucleotide may contain bonds other than phosphodiester, or be chemically linked to other molecules, as defined above.

It is also observed that in case x), the same loop oligonucleotide, modified or not, can be used for single-strand oligonucleotides of different sequences, and therefore for different targets.

As indicated above in relation to the method, the target is advantageously DNA in double helix. In particular, the target is a linear or circular plasmid, in particular a supercoiled circular plasmid. The corresponding structure then consists of 2 circular DNA molecules, one in double strand, the other in single strand, which are attached to each other and can be separated, if desired, by breaking of a covalent bond.

In structures where the oligonucleotide is fixed simultaneously on two different nucleic acid sequences, carried for example by two different plasmids, the two plasmids are linked after circularization covalently.

It will be observed that the single-stranded oligonucleotide circularized around the double-stranded nucleic acid allows the labeling of this nucleic acid and thus constitutes a tool of great interest for detection applications or for the purification of nucleic acids.

The invention therefore relates to the application of the circularization process defined above for marking or detection of double-stranded nucleic acids, in particular of plasmids, comprising the use of a single-strand (case a) or loop (case x) oligonucleotide linked covalently or not to a molecule of interest. The non-covalent bond can be obtained for example by the use of a biotinylated oligonucleotide which interacts strongly with an avidin or streptavidin molecule linked to a molecule of interest.

Such labeling, which requires no chemical or enzymatic modification of the target, is particularly advantageous in the case of plasmids and more particularly of supercoiled plasmids.

The labeling can be radioactive or chemical, for example using fluorescent groups.

The intensity of the labeling can be improved by using anti-streptavidin antibodies coupled to biotin, followed by a second layer of streptavidin-fluorescein.

Alternatively, the labeling can be carried out via another nucleic acid molecule capable of hybridizing with the single-stranded oligonucleotide.

Detection can also be carried out by a loop replication mechanism which makes it possible to amplify the sequence of the circular single-stranded oligonucleotide.

Circularized structures produced with such oligonucleotides advantageously make it possible to count the plasmids in the cells, even present in small numbers, which constitutes an advantage compared to the FISH techniques which do not allow the measurement of transfection efficiency only when the plasmid is present with a very high copy number.

They also make it possible to differentiate two differing sequences by only 1 or 2 mutations, by detecting the oligonucleotide attached to one of these two sequences by means of the molecule carried by this oligonucleotide or by a loop replication process.

According to another application, it will be noted that the circularization process according to the invention can be implemented to select, for example from degenerate single-stranded nucleic acid sequences, nucleic acid sequences capable of binding to an acid double-stranded nucleic acids, or sequences capable of promoting the penetration of double-stranded nucleic acids into cells, or the targeting of these nucleic acids towards specific cellular compartments, in particular towards the nucleus.

Using this process, structures are formed where the oligonucleotide is irreversibly attached to its target, which can be isolated, for example, by chromatography, gel purification, exclusion columns. By using oligonucleotides containing a partially random sequence, it is possible to isolate and, if desired, amplify the sequences which give the oligonucleotide the ability to bind to the nucleic acid target by coiling around him. The invention thus relates to the application of the circularization process defined above for the purification of double-stranded nucleic acids, in particular of plasmids, comprising the use of a single-stranded oligonucleotide or of a looped oligonucleotide covalently linked to a molecule which can interact with another molecule attached to a solid support, in particular an affinity column.

Such purification can also be carried out using a single-stranded oligonucleotide circularized around a plasmid and a matrix covalently linked to a molecule which can interact with another molecule attached to a solid support, in particular an affinity column.

The invention also relates to the applications of said structures, in particular in detection and in gene therapy.

In accordance with the invention, the single-stranded oligonucleotide may comprise a residue of chemical compound chosen to improve, for example, plasmid purification protocols, the intracellular delivery of plasmids to a specific cellular compartment (nucleus, mitochondria, for example), or targeting in gene therapy applications.

It is thus possible to vectorize double-stranded nucleic acids, advantageously DNAs, and in particular plasmids, using structures in which the single-stranded oligonucleotide is coupled, covalently or non-covalently, to susceptible molecules. to promote its penetration through the cell and / or nuclear membrane. These include, for nuclear targeting, peptides corresponding to the nuclear localization sequence (NLS) of the product of the large T antigen of the SV40 virus (Pro-Lys-Lys-Lys-Arg-Lys-Val) or synthetic molecules of a nature other than peptide, for example glycoconjugates or nucleic acid fragments (DNA or RNA) which can be integrated into the sequence of the single-stranded oligonucleotide. For cell penetration, peptides having a translocation property may be used, for example a peptide corresponding to residues 48 to 57 of the HIV TAT protein

(Gly-Arg-Lys-Lys-Arg-Arg-Gln-Arg-Arg-Arg), or a peptide corresponding to residues 43 to 58 of the homeodomain of the Antennapedia protein (Arg-Gln-Ile-Lys-Ile- Trp-Phe-Gln-Asn-Arg-

Arg-Met-Lys-Trp-Lys-Lys). For cell targeting, membrane receptor ligands, for example TGF-α, will advantageously be used.

Sequences with these properties can be isolated by selection-amplification experiments from minicircles containing a random sequence.

The formation of triplex structures according to the invention is also of interest for specifically repressing the expression of a gene. This repression can be inducible, in particular by using a specific ligand for the triple helices. It is thus possible to eliminate the expression of the transgene by treating patients with said ligand.

They also make it possible to irreversibly link 2 different molecules of double-stranded circular nucleic acid, and in particular 2 different plasmids, which makes it possible to increase, for example, co- transformation of bacteria or co-transfection of cells, ensuring the concomitant presence of 2 different plasmids.

It will be noted that the formation of triplex structures of the invention can also be used to carry out genetic analyzes, for example to detect mutations in cytological samples, or even to study the distribution of certain sequences.

Other characteristics and advantages of the invention are given in the examples which follow. In the description of these examples, reference is made to FIGS. 1 to 17, which respectively represent:

FIG. 1, a schematic diagram of the method of circularization of an oligonucleotide around a plasmid and around a DNA in double helix,

FIG. 2, a target sequence of the plasmid construction pAR1 with insertion of the promoter of the androgen receptor (- 146, +131), a monobrin oligonucleotide to be circularized and an oligonucleotide matrix,

- Figure 3, the test results showing the formation of a triple helix,

FIG. 4, a photo of an electrophoresis gel under denaturing conditions, with the ligation and cleavage products,

FIG. 5, a photo of a non-denaturing gel showing the dissociation of triple helices in the absence of magnesium, - Figure 6, a target sequence of the plasmid construction pGAl, consisting of the insertion of a synthetic duplex between the EcoR I and Pst I sites of the plasmid pBluescript SK +, marketed by Proméga, the sequence of the central part of the oligonucleotide to be circularized for constructs 2 and 3, and the ligands used for construct 1,

- Figure 7, a photo of a gel in denaturing conditions, with ligation and cleavage products,

- Figure 8, a photo of an agarose electrophoresis gel, with ligation products, viewed either by lighting with a UV lamp in the presence of ethidium bromide (right), or by autoradiography after drying of the gel (left), and

FIG. 9, a photo of an electrophoresis gel under denaturing conditions, with the ligation products,

FIG. 10, a photo of an agarose electrophoresis gel, showing the modified or unmodified plasmid and the products of various enzymatic digestions, viewed by lighting with a UV lamp in the presence of ethidium bromide,

FIG. 11, a photo of an electrophoresis gel under denaturing conditions, with the ligation products,

FIG. 12, a photo of an electrophoresis gel under denaturing conditions, with the ligation products,

- Figure 13, a photo of an electrophoresis gel under non-denaturing conditions, showing that the oligonucleotide circularizes around the plasmid by a comigre chemical reaction with this plasmid, and then dissociates from this plasmid when it is linearized and the triple helix is destabilized,

FIG. 14, a photo of an electrophoresis gel under non-denaturing conditions, showing the presence of streptavidin attached to the circular oligonucleotide after purification and digestion of the plasmid,

FIG. 15, a target sequence of the plasmid construction pBluescript SK +, where the sequence of the single-stranded oligonucleotide, the central part of which fixes by forming a triple helix and the sequence of the mother loop oligonucleotide 32 circularizes together with this oligonucleotide , and

Figure 16, a photograph of a gel electrophoresis under denaturing conditions ^1, with the ligation products.

- Figure 17, a description of the reaction scheme for the synthesis of a peptide-oligonucleotide loop conjugate.

Triple helix constructions 1, 2 and 3

The diagram of the circularization process used in the examples is given in FIG. 1, the examination of which shows the central part (1) of the linear oligonucleotide which binds to the large groove of a specific sequence inside the plasmid ( 2) (left part) or double stranded DNA (3) (right part), to form a triple helix complex. Its 5 'and 3' ends (4) and (5) hybridize next to each other, to an oligonucleotide matrix (6) and are linked by a ligation agent (7) to form a circular oligonucleotide (8) padlocked at the plasmid DNA (left part) or to a DNA fragment (right part)

According to this diagram, 3 different types of triple helixes are described.

.Construction 1, between the BamH I and Xho I sites:

This construction is illustrated by FIG. 2 which represents the plasmid pAR1 constructed by insertion of the promoter of the mouse androgen receptor (-146, +131) between the BamH I and Xho I site of the plasmid pBL-CAT (in the plasmid pAR4, the GC base pairs indicated by an * have been mutated to TA) SEQ ID No. 1 represents the sequence -119, -70 of the promoter and SEQ ID No. 2, the complementary sequence.

SEQ ID No.1:

5 * AGGGGAGAAAAGGAAAGGGGAGGGGAGGGGAGGGGAGAGAGAAAGGAGGT

SEQ ID N ° 2:

5 'ACCTCCTTTCTCTCTCCCCTCCCCTCCCCTCCCCTTTCCTTTTCTCCCCT 3'

The target DNA sequence for the formation of the triple helix is located between positions -102 and -88 on SEQ ID No. 1. It corresponds to SEQ ID No. 4.

Figure 2 also gives the complete sequence of

1 mother oligonucleotide 89 (SEQ ID No. 3) on which the target binding site ("triplex site") and which corresponds to SEQ ID No. 4, and the sequence of the 20 nucleotide matrix (SEQ ID) have been indicated ID No. 5). SEQ ID N ° 3:

5 'CGTACGGTCGACGCTAGCTTTTTTTTTTTTTTTTTTTTTAGAGGGGAGGGGAGGGTTTTTTTTTT TTTTTTTTTCACGTGGAGCTCGGATCC 3'

The mother oligonucleotide 89 comprises

- a central sequence (SEQ ID N ° 4) capable of forming a triple helix:

SEQ ID N ° 4: GAGGGGAGGGGAGAGGG

- two lin ers connecting SEQ ID N ° 4 to terminal sequences of ten bases each, located at the ends, which can respectively form ten base pairs with the matrix of 20 nucleotides (SEQ ID N ° 5),

SEQ ID N ° 5: CGACCGTACGGGATCCGAGC

A phosphate group is added to the 5 'end of

1 oligonucleotide for the next step of enzymatic circularization. Using ³² P-ATPs, the radiolabelled oligonucleotide is obtained.

The position of the restriction site for BsiW 1 within the 20 bp oligonucleotide matrix is also indicated.

Two other matrix sequences were used, namely

SEQ ID N ° 6, 36-mother:

5 'GCTAGCGTCGACCGTACGGGATCCGAGCTCCACGTG 3' SEQ ID N ° 7, 14 - mother: 5 'CCGTACGGGATCCG 3'

.Constructions 2 and 3, between Hind III and Pst I:

FIG. 6 comprises the sequences of the synthetic duplex between the EcoR I and Pst I sites (SEQ ID No. 8 and SEQ ID No. 9) of the plasmid pGAl. SEQ ID N ° 8:

5 'AGCTTCGTACGGCGCCCGAGTTAAGGGAGAAGAGGAAAGAGATTGAGCGAGCCCTAGGGACG TCCCGGGCTAGCGAATTCCTGCA 3'

SEQ ID N ° 9:

5 'GGAATTCGCTAGCCCGGGACGTCCCTAGGGCTCGCTCAATCTCTTTCCTCTTCTCCCTTAAC TCGGGCGCCGTACAG 3'

The target sequence is an oligopurine sequence. 20 bp oligopyrimidine (SEQ ID No. 10).

SEQ ID N ° 10: 5 "AAGGGAGAAGAGGAAAGAGA 3 '

Two types of oligonucleotides have been developed so that they can be circularized around this target sequence.

.Construction 2:

Those of construction 2 have a central part of 18 nucleotides composed of G and T

(SEQ ID No.11). Their total length can vary from 43 to 89 nucleotides.

SEQ ID No ll: 5 'TGTTTGGTGTTGTGGGTT 3' The following corresponding single-stranded oligonucleotides were developed:

5 'CGTACGGTCGACGCTAGCTTTTTTTTTTTTTTTTTTTGTTTGGTGTTGTGGGTTTTTTTTTT TTTTTTCACGTGGAGCTCGGATCC 3' (SEQ ID N ° 12, 89-mother),

CGGATCC3 '(SEQ ID N ° 13, 69-mother),

5 'CGTACGGTCGACGCTAGCTTTTGTTTGGTGTTGTGGGTTTTCACGTGGAGCTCGGATCC3' (SEQ ID N ° 14, 59 mother),

5 'CGTACGGTCGACGTTTTGTTTGGTGTTGTGGGTTTTGGAGCTCGGATCC3' (49-mother) (SEQ ID N ° 15, 49-mother), and

5 'CGTACGGTCGTTTTGTTTGGTGTTGTGGGTTTTGCTCGGATCC3 * (SEQ ID N ° 16, 43 -mother)

.Construction 3:

The oligonucleotides of construction 3 have a central part of 15 nucleotides composed of T and C (SEQ ID No. 17).

SEQ ID N ° 17: 5 'CTCTTCTCCTTTCTCT 3'

The single-stranded oligonucleotide corresponding to the following sequence was produced: 5 'CGTACGGTCGACGCTAGCTTTTCTCTTCTCCTTTCTCTTTTTCACGTGGAGCTCGGATCC 3' (SEQ ID NO: 18). Example 1 Analysis of the formation of a triple helix by co-migration test

The formation of a triple helix is verified in the first construction, between the oligonucleotide 89-mother and its target sequence on the plasmid, by carrying out the following co-migration test, the results of which are given in FIG. 3

- the 89-mer oligonucleotide (10 fmol) is incubated in 10 μl of 50 mM Tris-HCl (pH 7.5), ₂ lOmM MgCl, 10 mM dithiothreitol, 1 mM ATP, 25 μg / ml of albumin bovine serum, in the absence of

- Plasmid (lane 1 in FIG. 3), in the presence of 60 fmol of overwound pAR1 (lane 2), 60 fmol of pARl digested with BamHI and Xho I (lane 3), and 60 fmol of pAR4 (lane 4).

The samples are heated to 75 ° C, then cooled to 37 ° C and loaded onto a 6% polyacrylamide gel.

The migration is carried out at 37 ° C. in a TBE buffer containing 10 mM of MgCl ₂ •

Staining of the gel with ethidium bromide shows that the supercoiled plasmid fully penetrates the gel

(lane 2) and the comparison with the electrophoretic mobility of molecular weight markers indicates that the lane of lane 3 migrates as a 700 bp duplex.

When the oligonucleotide and the plasmid pAR1 are incubated together, the labeled oligonucleotide migrates along with the plasmid. This migration does not occur with the plasmid pAR4, in which the formation of the triple helix does not can occur as a result of 6 G to T mutations in the target sequence (these mutations are shown in Figure 2, and designated by an asterisk).

When pAR1 is digested with both Xho I and BamH I, the labeled oligonucleotide migrates along with the 700 bp fragment which contains the target sequence.

Example 2 Analysis of Ligation and Cleavage Products by Gel Electrophoresis under Denaturing Conditions with Construction 1

a - study of ligation products

The 89-mer labeled oligonucleotide is incubated as described above with reference to FIG. 3 when, as the ligase, the phage T4 DNA ligase is used as ligase. With Ampligase, a Tris-HCl buffer using 20 mM (pH 8.3), 25 mM KC1, lOmM MgCl _2, 0.5 mM NAD, 0.01% Triton XlOO and with E. coli ligase, 50 mM Tris-HCl buffer (pH 7.8), ₂ mO MgCl, 10 mM DTT, 26 μM NAD, bovine serum albumin 25 μg / ml.

The ligation reaction is carried out at 45 ° C, for 1 hour, using 40 U of T4 DNA ligase (New England, Biolabs). Carrying out the reaction at this temperature makes it possible to reduce the formation of molecules of linear dimers which are formed as a result of the ligation of two different 89-mer oligonucleotides which hybridize to the 20-mer matrix. The ligase is inactivated by heat for 15 minutes at 65 ° C before treatment with restriction endonucleases. The results of the various tests are given in FIG. 4. In tracks 3 to 6 and 8, the plasmid is present during the ligation reaction.

In lane 7, the plasmid is added after circularization of the 89-mer oligonucleotide, followed by heat denaturation of the DNA ligase, as indicated by the asterisk. The samples are precipitated by addition of ethanol and resuspended in 80% formamide, 10 mM NaOH, 1 mM EDTA as loading buffer.

The reaction mixture is heated at 90 ° C for 5 minutes, then loaded onto a 6% denaturing polyacrylamide gel.

It can be seen on examining FIG. 4 that in the presence of the matrix of 20 oligonucleotides, the linear oligonucleotide

89-mother is transformed into a circular molecule. The migration speed in a 6% denaturing polyacrylamide gel appears to be lower.

As indicated above, the ligation is also carried out in the presence of the plasmid pAR1, under conditions allowing the formation of the triple helix.

The labeled oligonucleotide is transformed into a species with a very low migration speed, similar to that of the plasmid itself, as revealed by the ethidium bromide staining (Figure 4, lane 4).

The nature of this species has been identified using two different approaches. b - study of the cleavage products

The sample was treated with the restriction endonuclease Xho I, which makes it possible to linearize the plasmid pAR1.

This treatment releases the circular oligonucleotide as shown by the electrophoretic mobility (Figure 4, lane 6).

Alternatively, the sample was treated with

The restriction endonuclease BsiW I. This enzyme does not cleave pAR1, but can cleave the circular oligonucleotide in the presence of the oligonucleotide template.

The cleavage product migrates exactly as

1 linear oligonucleotide from 89 mothers (Figure 4, lane 5).

The species with low migration cannot be detected in the following tests:

- when using, in place of pAR1, pAR4 which does not form a triplex with the oligonucleotide of 89 mothers (lane 8),

- when the plasmid pAR1 is added after the circularization of the 89 mer oligonucleotide and the heat denaturation of the T4 DNA ligase (lane 7), and also

when another linear oligonucleotide which cannot form a triple helix is used for circularization. These experiments show that the oligonucleotide was circularized around the double stranded DNA target and therefore was padlocked to the plasmid.

The two molecules remain associated under highly denaturing conditions and can only be separated by breaking the chemical bonds.

However in this complex, the circular oligonucleotide, capable of forming a triple helix, can diffuse in one dimension along the double helix.

Example 3 Dissociation of the Triple Helices in the Absence of Magnesium in a Non-Denaturing Gel

The samples are prepared as described above in the presence of magnesium and loaded onto a 6% polyacrylamide gel free of magnesium.

The migration is carried out at 37 ° C. in a TBE buffer in the absence of MgCl ₂ .

The results of the tests are given in FIG. 5.

No triple helix formation is detected between

The linear mother oligonucleotide and the plasmid pAR1 when the samples incubated in a buffer containing magnesium are migrated in a non-denaturing gel, without magnesium, (FIG. 5, lane 2). The radiolabelled oligonucleotide remains associated with the plasmid only when it has been circularized in the presence of the plasmid and of magnesium before loading the gel (lane 4).

When the plasmid is digested with Xho I,

The circular oligonucleotide remains associated with the linear plasmid when the gel contains magnesium, but dissociates in the absence of magnesium, although the cleavage site is located 200 bp from the target site of the triple helix.

In lane 6, the plasmid is added after circularization of the 89 mer oligonucleotide, followed by heat denaturation of the DNA ligase.

The absence of magnesium (or any other divalent cation) therefore appears sufficient to prevent the formation of the triple helix and allow the circular oligonucleotide to move around the plasmid.

Other strategies can be used to label the plasmid so as to lose the interaction of the triple helix after ligation, for example by forming a triple helix induced by a ligand or a given pH value.

EXAMPLE 4 Analysis of the Ligation and Cleavage Products by Gel Electrophoresis, in Denaturing Conditions with Construction 2

The results obtained are shown in Figure 7 The oligonucleotide, which can be a 59-mer (tracks 1 to 6), a 69-mother (tracks 7 to 12), or an 89-mother (tracks 13 to 18) (OFMOL) is incubated in the medium. described in connection with

Figure 3 in Example 2, in the absence of plasmid (lanes 1,2,7,8,13,14), in the presence of pGAl (lμg) (lanes 4,5,6,10,11,12 , 16, 17, 18), or in the presence of plasmid pBluescript which does not contain the target for the formation of the triple helix (lμg) (tracks 3.9 and 15) and in some cases in the presence of ligand SD46 (4μM) (tracks 3,5,6,9,11,12,15,17,18). The samples are heated to 70 ° C, then cooled to 37 ° C. The circularization reaction is carried out at 37 ° C. for 1 h, using 40 U of T4 DNA ligase. The matrix was not added in lanes 1,7 and 13. The ligase is inactivated by heat. 10 μl of a formamide solution containing charge dyes are then added directly to the samples. The reaction mixture is heated at 90 ° C for 5 min, then loaded onto a 5% denaturing polyacrylamide gel.

It will be seen on examining FIG. 7 that, in the presence of the matrix of 20 nucleotides, the oligonucleotides of 59, 69 and 89-mer are transformed into circular molecules of lower migration. The appearance of very slow migration products is observed only if the oligonucleotide has been previously incubated with the plasmid pGA1 containing the target sequence and the ligand SD46. This species was treated with the Hind III restriction endonuclease, which makes it possible to linearize the plasmid pGAl. This treatment releases the circular oligonucleotide. These experiments demonstrate that the oligonucleotides were circularized around the plasmid by formation of a triple helix inducible by the ligand SD46. Example 5 Radiolabelling of a Supercoiled Plasmid Using the 89-mer Oligonucleotide

The results are given in FIG. 8.

The oligonucleotide 89-mother is incubated as indicated above, in the presence of plasmid pGAl (lμg) (tracks 1,2,4 and 5) or pBluescript (tracks 3 and 6), in the presence of 2μM of ligand SD27 (tracks 1 to 3) or SD46 (tracks 4 to 6). In tracks 1 and 4, the matrix is omitted. After circularization reaction and inactivation of the ligase, 2 μl of glycerol were added to the samples which were loaded on a 1% agarose gel and migrated for 30 min at 50V in TBE buffer. The gel was viewed under UV lamp. The presence of ethidium bromide in the gel reveals the presence of supercoiled plasmid in all of the wells (on the right) (wells 7 and 8 respectively contain the plasmids pBluescrpit and pGAl which have not undergone any treatment). The gel was then dried and autoradiographed. An important radioactive signal is detected at the plasmid level when the circularization has been carried out in the presence of pGA1 and of ligand SD27 or SD46. This experiment demonstrates that the topological state of the plasmid is not affected by the treatments carried out and that a radiolabeled oligonucleotide locked with a plasmid makes it possible to radiolabel this plasmid.

Example 6 Analysis of the Ligation Products by Gel Electrophoresis in Denaturing Conditions with Construction 3

The results are given in FIG. 9. The 59-mer oligonucleotide (10 fmol) is incubated in 10 μl of 50 mM MES (pH 6.0), 20 mM MgCl ₂ , 3 mM DTT, ATP 1 mM, in the absence of plasmid (lanes 1 and 2), in the presence of pGAl (5 μg) (lane 4) or pBluescript (lane 3). The samples are heated to 70 ° C, then cooled to 4 ° C. After approximately 14 h, the circularization reaction is carried out at room temperature, for 2 h, using 40 U of T4 DNA ligase. The ligase is inactivated by heat. 10 μl of a formamide solution containing charge dyes are then added directly to the samples. The reaction mixture is treated as indicated above with FIG. 4. Under these conditions, the oligonucleotide can be circularized by the DNA ligase of phage T4 (lane 2). It is circularized around the plasmid when it contains the target sequence for the formation of the triple helix (well 4).

Example 7 Circularization of an Oligonucleotide Around a Plasmid with Triple Helix Sequences and Oligonucleotides of Different Length

Other sequences of single-stranded oligonucleotides are reported below designed so that they can be circularized around the target sequence of construction 2, and comprising a central part composed of G and T which may be shorter than that used in the oligonucleotides described above (SEQ ID No. 11, 18 nucleotides). Thus, this central part can be 14 nucleotides (SEQ ID No 19) for the 57-mer oligonucleotide (SEQ ID No 21) or even 12 nucleotides (SEQ ID No 20) for the 54-mother oligonucleotides (SEQ ID N ° 22), 43-mother (SEQ ID N ° 23), 33-mother (SEQ ID N ° 24). Their length varies from 31 to 57 nucleotides, and hexaethylene glycol-type spacers can be included in the sequence. SEQ ID N ° 19: GTTTGGGTGTTGTG

SEQ ID N ° 20: GTGTTGTGGGTT

SEQ ID N ° 21: CGTACGGTCGACGCTAGCTTTTGTTTGGGTGTTGTGTTTTCACGTGGAGCTCGGATCC

SEQ ID N ° 22: CGTACGGTCGACGCTAGCTTTTGTGTTGTGGGTTTTCACGTGGAGCTCGGATCC

SEQ ID N ° 23: CGTACGGTCGTTTTTTGTGTTGTGGGTTTTTTTGCTCGGATCC

SEQ ID N ° 24:

CGTACGGTCGTXGTGTTGTGGGTTXGCTCGGATCC (X is a hexaethylene glycol spacer)

The oligonucleotides SEQ ID Nos. 21, 22, 23 and 24 can be circularized around their target carried by the plasmid pGAl.

Example 8 Dissociation of triple helices when extracting the specific ligand for triple helices with a trap oligonucleotide forming an intramolecular triple helix: analysis of the mobility of the circular oligonucleotide along the plasmid.

In this example, a trap oligonucleotide (SEQ ID No. 25) is used, capable of forming an intramolecular triple helix,

in order to capture the ligand intercalated in the triple helices, and thus to destabilize them. SEQ ID N ° 25 CTTTCCTTCTCTCCTTTTTGGAGAGAAGGAAAGTTTTTGTTTGGTTGTGTGG

For these experiments, the plasmid pGA2 is used which is derived from the plasmid pGAl by the introduction of two mutations which reveal a unique restriction site for the enzyme Bgl II. This site partially covers the binding site of the oligonucleotides forming triple helices, represented by a black line (FIG. 10). The formation of a triple helix inhibits cleavage by this restriction enzyme.

The samples are prepared as follows. The mother oligonucleotide 59 (SEQ ID No. 14) (1 μM) is incubated in the T4 DNA ligase buffer (50 mM tris-HCl, 10 mM

MgCl ₂ , 10 mM DTT, 1 mM ATP, pH = 7.5), in the presence of plasmid pGA2

(1 μg in 10 μl) and a specific ligand for triple helices, BQQ (20 μM). The samples are heated to 75 ° C and then gently cooled to 37 ° C. The circularization reaction is carried out for 1 h at 37 ° C., adding 3 μM of matrix and 20 μl of T4 DNA ligase. The ligase is inactivated by heat. 2 μl of this solution are diluted in 40 μl of reaction buffer for the Bgl II restriction endonuclease (50 mM tris-HCl, 10 mM MgCl ₂ , 100 mM NaCl, 1 mM DTT, pH = 7.9), then 10 u of Bgl II are added. The reaction is stopped after 20 minutes by adding 5 μL of a 1% SDS solution - 0.4 M EDTA. The samples are then precipitated with 1 ethanol and loaded onto a 1% agarose gel migrated in a 0.5 buffer. x TBE.

The results of the tests are given in FIG. 10.

Well 1 shows the supercoiled plasmid pGA2, whose migration profile reveals the supercoiled plasmid (se) with the presence of a small amount of released plasmid (oc). Well 2 shows the linearized plasmid pGA2 (1) after cutting by the enzyme Bgl II. The circularization reaction as described above was carried out in wells 3 to 8. Wells 9 to 14 were subjected to the same treatment, but the matrix and the ligase were not added. Wells 4, 5, 7, 10, 11 and 13 were treated with the enzyme Bgl II (B), wells 8 and 14 with the enzyme Hind III (H). In wells 5, 6, 7, 11, 12 and 13, 1 trap oligonucleotide (TRAP) (SEQ ID No. 25) was added. In wells 6, 7, 12 and 13, ligand SD46 was added after the trap oligonucleotide. It will be seen on examining this figure that when the circularization reaction has been carried out, the cleavage by the restriction enzyme Bgl II is inhibited (well 4), while the cleavage by Hind III is not inhibited (well 8). When a trap oligonucleotide is added, the Bgl II cleavage is again observed (well 5). Finally, if SD46 is added (see FIG. 6), the cut by Bgl II is again inhibited (well 7). Cleavage by the enzyme Bgl II is never inhibited in the absence of circularization (wells 10, 11, 13, 14).

These results demonstrate that the circular oligonucleotide forms a more stable triple helix than the linear oligonucleotide and remains fixed on its target after dilution, which results in the inhibition of the enzymatic cleavage. It is possible to dissociate this complex by adding a trap oligonucleotide, which results in the restoration of the cleavage by the restriction enzyme. After this dissociation, the addition of a new ligand specific for the triple helices makes it possible to reposition the circular oligonucleotide on its target and to inhibit the enzymatic cleavage again. .Construction of triple helix 4 and 5

The target sequence used in the experiments given below is an oligopurine sequence. oligopyrimidine 18

base pairs (SEQ ID No. 26 and its complement SEQ ID No. 27) which is present in the origin of replication of phage fl, and as such in many plasmids commonly used in research laboratories, for example the pBluescript plasmids (sold by Stratagene), pUC family plasmids, pGFP family plasmids (sold by Clontech), pGL family plasmids (sold by Proméga). This sequence is not present in the plasmid pBR322.

SEQ ID N ° 26 AGAAAGGAAGGGAAGAAA SEQ ID N ° 27 TTTCTTCCCTTCCTTTCT

Two types of oligonucleotides have been developed so that they can be circularized around this target sequence.

.Construction 4

The oligonucleotides of construction 4 have a central part composed of G and T (SEQ ID No. 28). They can also carry one or two molecules of biotin grafted in position 5 of a thymine. SEQ ID N ° 28 TTTGTTGGGTTGGTTTGT The following oligonucleotides have been developed:

SEQ ID N ° 29

CGTACGGTCGACGCTAGCTTTTTGTTGGGTTGGTTTGTTTTCACGTGGAGCTCGGATCC

SEQ ID N ° 30

CGTACGGTCGACGCXAGCTTTTTGTTGGGTTGGTTTGTTTTCACGTGGAGCTCGGATCC (X represents a thymine carrying a biotin in position 5).

SEQ ID N ° 31

CGTACGGTCGACGCXAGCTTTTTGTTGGGTTGGTTTGTTTTCACGXGGAGCTCGGATCC (X represents a thymine carrying a biotin in position 5).

Construction 5

The oligonucleotides of construction 5 have a central part composed of C and T (SEQ ID No. 32) SEQ ID No. 32 TCTTTCCTTCCCTTCTTT

The following oligonucleotides have been developed:

SEQID N ° 33

SEQID N ° 34

CGTACGGTCGACGTTTTGTTTGGTGTTGTGGGTTTTCTTTCCTTCCCTTCTTTTTGGAGCTCGG

ATCC

Example 9: Analysis of the ligation products by gel electrophoresis under denaturing conditions, with construction 4.

It is verified that it is possible to circularize the oligonucleotides of construction 4 around the plasmids pBluescript SK + and pEGFPCl. The results are shown in Figure 11.

The oligonucleotides of sequence SEQ ID No. 29, SEQ ID

No. 30 and SEQ ID No. 31 are used respectively in wells 1 to 5, 6 to 10 and 11 to 15. The radiolabeled 5 'oligonucleotides (10 nM) are incubated in the absence of plasmid (wells 1, 2 , 6, 7, 11 and 12), with 1 μg of plasmid pBluescript (wells 3, 8 and 13), 1 μg of plasmid pEGFPCl (wells 4, 9 and 14), or 1 μg of plasmid pBR322 (wells 5, 10 and 15), in a T4 DNA ligase incubation buffer in the presence of 20 μM BQQ. After having heated the samples to 65 ° C. and then cooled to 37 ° C., the 20-mer template oligonucleotide and 4 μl of T4 DNA ligase are added. The reaction is carried out at 45 ° C for 1 h.

It will be seen on examining FIG. 11 that, in the presence of the matrix of 20 nucleotides, the oligonucleotides are transformed into circular molecules of lower migration. The appearance of very slow migration products is observed when the oligonucleotide has been previously incubated with the plasmid pBluescript or pEGFPC1, containing the target sequence. It is not observed when the oligonucleotide is preincubated with the plasmid pBR322 which does not contain the target sequence. These experiments demonstrate that the oligonucleotides of construction 4 can be circularized around plasmids containing the origin of replication of phage f1. Example 10: Analysis of the ligation products by gel electrophoresis under denaturing conditions, with construction 5.

It is verified that it is possible to circularize the oligonucleotides of construction 5 around the plasmid pBluescript SK +. The results are shown in Figure 12.

The 68-mer oligonucleotide (SEQ ID No. 34) phosphorylated in 5 ′ and radiolabelled (3.9 pmol) is incubated with 10 μg of plasmid pBluescript (5 pmol) (wells 3, 5, 6, 7 and 8) or 10 μg of plasmid pBR322 (well 4), in an incubation buffer 20 mM Ammonium acetate pH = 4.5, 20 mM MgC12, 3 mM DTT, 1 mM ATP

(incubation volume 7 μl). After an overnight incubation at room temperature (20 ° C.), the 20-mer matrix oligonucleotide (1 μl) is added, then 1 μl of a 250 mM Tris solution pH = 8.0 and 1 μl (4u) of T4 DNA ligase (wells 2, 3, 4, 5, 7, 8). The reaction is carried out at room temperature for 2 hours. The matrix was not added to wells 1 and 5. In well 7, water was added in place of the Tris solution. In well 8, the incubation was carried out in T4 DNA ligase buffer.

It will be seen on examining FIG. 12 that, when the oligonucleotide has been preincubated at acid pH, the circularization reaction can only be carried out if the solution of Tris pH = 8.0 is added (compare well 2 and 7). The appearance of very slow migration products is observed if the oligonucleotide has been preincubated at acidic pH with the plasmid pBluescript SK +. It is not observed when The oligonucleotide is preincubated with the plasmid pBR322 which does not contain the target sequence.

These experiments demonstrate that an oligonucleotide whose central part forming a triple helix is composed of T and C can be circularized around the plasmid under pH conditions close to neutrality provided that the triple helix has been preformed at acid pH.

Example 11: Another sequence used to circularize an oligonucleotide around a double-stranded DNA.

Construction 6:

The target sequence used is an oligopurine sequence. 23 base pair oligopyrimidine (SEQ ID No. 35 for the oligopurine portion) which is present in the promoter of the murine IGF1 gene. In the experiments reported below, this sequence is carried by the plasmid pl71lb / luc. This plasmid is constructed by cloning the sequence (-1711, +328) of the murine IGF1 gene into the plasmid pGL2 (Proméga).

SEQ ID N ° 35 AGAAGAGGGAGAGAGAGAGAGAAGG

The oligonucleotide of construction 6 has a central part composed of G and T (SEQ ID N ° 36) capable of

form a triple helix of 18 base triplets by attaching to the above target. SEQ ID N ° 36 GGTTGTGTGTGTGTGGGT

The single strand oligonucleotide (SEQ ID NO: 37) used for the circularization experiments is produced from this sequence in the same way as the oligonucleotide SEQ ID No: 14 is designed from the sequence SEQ ID No 11.

SEQ ID N ° 37

CGTACGGTCGACGCTAGCTTTGGTTGTGTGTGTGTGGGTTTCACGTGGAGCTCGGATCC

Experiments have shown that the oligonucleotide SEQ ID No. 37 can be circularized around its target carried by the plasmid pl711b / luc.

.Constructions 7 and 8

The implementation of the chemical reaction requires the use of a 5 '-iodo and 3'-phosphorothioate oligonucleotide. Since oligonucleotides having a 5 'thymine -iodo are currently the only ones available, the sequence of the template has been modified.

SEQ ID N ° 38

GACCGTACGATATCGCGAGC

Two types of oligonucleotides have been developed so that they can be circularized using a chemical method. .Construction 7

The oligonucleotide of construction 7 has a central part composed of G and T (SEQ ID No ll) capable of

fix on the plasmid pGA2 using the target sequence SEQ ID No. 10.

SEQ ID N ° 39

5 '-Iodo- TCGTACGGTCGACGCTAGTTTTGTTTGGTGTTGTGGGTTTTCACGTGGAGCTCGCGATA 3' - phosphorothioate

.Construction 8

The oligonucleotide of construction 8 (SEQ ID No 40) has a central part composed of C and T (SEQ ID No 32) capable of binding to the plasmid pBluescript SK + thanks to the target sequence (SEQ ID No 26 and its complementary SEQ ID N ° 27).

SEQ ID NO "40

5'-Iodo- TCGTACGGTCGACGCTAGTTTTCTTTCCTTCCCTTCTTTTTCACGTGGAGCTCGCGATA 3 '- phosphorothioate

Example 12: Analysis of the chemical ligation and cleavage products by gel electrophoresis under non-denaturing conditions, with construction 7.

The 59-mer oligonucleotide (SEQ ID No. 39) (wells 2 to 10) was incubated with 1 μg of plasmid pGA2 (wells 3, 4, 7 to 10) in a 50 mM tris-HCl incubation buffer, 10 mM MgCl ₂ , 10 mM DTT, 1 mM ATP, pH = 7.5, in the presence of 20 μM BQQ. After having heated the samples to 65 ° C., then cooled to 37 ° C., the matrix oligonucleotide SEQ ID No. 37) is added which is radiolabeled at 5 '. The reaction is carried out at 37 ° C for 1 h, except in wells 4 and 5 (2 minutes). Then digested with the Hind III restriction endonuclease (wells 9 and 10). The samples from wells 8 and 10 are heated in the presence of a trap oligonucleotide SEQ ID No. 24. Well 1 shows the radiolabelled matrix alone.

The results are shown in Figure 13.

It will be seen on examining the figure that the matrix does not hybridize on the 59-mer oligonucleotide until the incubation is sufficiently long. Other ligation experiences

Enzymatic studies have shown that this is due to the fact that the matrix does not comigre with the 59-mother until it has been circularized. The formation of an internucleoside link between the 5 ′ -iodo and 3 ′ -phosphorothioate ends of the oligonucleotide requires a 1 hour incubation with the template. When the circularization is carried out in the presence of the plasmid pGA2, a species which migrates very slowly in the gel appears (wells 3, 7, 8 and 9). When this species is digested with Hind III, comigration of the matrix is still observed (well 9). This reveals that the triple helix formed by the circular oligonucleotide and the linear plasmid does not dissociate in the gel. However, by adding a trap oligonucleotide (SEQ ID No. 25), the BQQ is removed from the complex, which has the effect of destabilizing the triple helix and releasing a circular oligonucleotide (well 10). The same treatment carried out on the undigested plasmid does not allow this release (well 9). These experiences demonstrate that the Chemical circularization of the oligonucleotide around the plasmid pGA2 does indeed result in the formation of a catenate complex.

Likewise, it has been demonstrated that the oligonucleotide of construction 8 (SEQ ID No. 40) can be circularized around the plasmid pBluescript SK + in the acetate buffer at pH 4.5 described in Example 10.

Example 13: Attachment of streptavidin to a plasmid carrying a biotinylated oligonucleotide.

Biotinylated (SEQ ID NO: 30) (well 1 to 9) or non-biotinylated (SEQ ID No: 29) (well 10 to 12) oligonucleotides 59-mer (well 10 to 12), radiolabelled in 5 ′, were circularized as described in the example 9, in the absence (wells 3 and 4) or in the presence (wells 5 to 12) of the plasmid pGA2. Streptavidin (200 nM) was then added to wells 2, 4, 6, 9, 11 and 12. The complexes were then precipitated (wells 7, 8, 9 and 12) in order to remove the oligonucleotides which did not are not circularized around the plasmid, then optionally digested with the Hind III restriction endonuclease (wells 8, 9 and 11).

Figure 14 summarizes the results obtained. O denotes the migration position of the oligonucleotides alone. OS designates the migration position of the oligonucleotide-streptavidin complexes, and P designates the migration position of plasmin. Complexation of streptavidin slows the migration of the linear (well 2) or circular (well 4) oligonucleotide. The precipitation makes it possible to eliminate the oligonucleotides which are not linked to the plasmid (well 7). After digestion with Hind III, it can be observed that the circular biotinylated oligonucleotide is released from the plasmid, alone (well 8) or linked to streptavidin (well 9). If the radiolabelled biotinylated oligonucleotide is added after precipitation of a complex formed with the non-biotinylated 59-mer in the presence of streptavidin, the migration of this oligonucleotide is not delayed (well 12), which demonstrates that the precipitation of the streptavidin is indeed due to its attachment to the biotinylated oligonucleotide attached to the plasmid.

Triple helix constructs for the system using a loop oligonucleotide

The principle of the circularization process used is given in FIG. 15, the examination of which reveals the central part of the linear oligonucleotide (1) which can bind in the large groove of a specific sequence inside the plasmid (2 ) to form a triple helix complex (3), and 5 'and 3' ends capable of hybridizing to each other by forming a short double helix (4). One end protrudes from the short double-stranded sequence by a single-stranded sequence (5). This end can form a double-strand with a complementary single-strand sequence (6) protruding from an oligonucleotide which has the form of a loop or hairpin (7). The formation of this double-strand results in the juxtaposition of the 5 'and 3' ends of the single-strand and loop oligonucleotides. These ends are then linked by a ligation agent to form a circular oligonucleotide locked with plasmid DNA or with a DNA fragment. The looped oligonucleotide can be chemically modified so as to carry, for example, a biotin molecule, or else an amino function which can be used to covalently link a peptide to this oligonucleotide. In accordance with this scheme, several different types of triple helixes have been produced

. Construction 9

This construction is illustrated in FIG. 15 which represents the target sequence located on the plasmid pBluescript SK + (SEQ ID Nos. 26 and 27). The figure also gives the complete sequence of the mother oligonucleotide 63 (SEQ ID No. 41) on which the site binding to the target has been indicated ("triplex site") and which corresponds to SEQ ID No. 28, and the sequence of the 32 nucleotide loop oligonucleotide (SEQ ID NO: 42).

The mother oligonucleotide 63 comprises:

- a central sequence capable of forming a triple helix,

- two linkers connecting this central part to terminal sequences, one (the 3 'end) of 6 nucleotides, the other (the 5' end) of 12 nucleotides, located at the ends, which can form 6 base pairs by hybridizing to each other. The 5 'end protrudes from this short double-stranded sequence as a single-strand of 6 nucleotides. The oligonucleotide is phosphorylated at 5 'to allow enzymatic ligation.

SEQ ID N ° 41 CATCGACCTAGTTCGACGCTAGCTTTTTGTTGGGTTGGTTTGTTTTCACGTGGAGCTACTAGG

The 32-mer oligonucleotide comprises:

- two complementary sequences which form a double-strand of 11 base pairs, linked together by a linker of 4 nucleotides, a 5 ′ end phosphorylated in single strand of 6 nucleotides complementary to that of the oligonucleotide 63 -mother.

SEQ ID N ° 42

TCGATGTCCGGATTGGCTTTTGCCAATCCGGA

A construction of analogous structure was carried out using as target sequence of the formation of a triple helix that of constructions 2 and 3 (SEQ ID No. 10), using an oligonucleotide 63 mother (SEQ ID No. 43) or 73 mother (SEQ ID No. 44) whose central part forming a triple helix contains G and T (SEQ ID No. 11).

SEQ ID N ° 43

CATCGACCTAGTTCGACGCTAGCTTTTGTTTGGTGTTGTGGGTTTTCACGTGGAGCTACTAGG

SEQ ID N ° 44

CATCGACCTAGTTCGACGCTAGCTTTTTTTTTGTTTGGTGTTGTGGGTTTTTTTTTCACGTGGA GCTACTAGG

Another construction of analogous structure was carried out using as target sequence of the formation of a triple helix that of constructions 4 and 5 (SEQ ID No 26 and 27), using a mother oligonucleotide 63 (SEQ ID No 45 ) whose central part contains uracils and thymines carrying a methoxy group in position 2 'of the riboses.

SEQ ID N ° 45 CATCGACCTAGTTCGACGCTAGCTTTUGUUUGGUUGGGUUGUUUTTCACGTGGAGCTACTAGG (the underlined nucleotides are 2'0-methyl nucleotides) Example 14: Analysis of the ligation products by gel electrophoresis under denaturing conditions, with construction 9.

The radiolabeled mother oligonucleotide 63 (SEQ ID No. 41) is incubated in the T4 DNA ligase buffer, the sample is heated to 80 ° C. before the subsequent ligation step using the T4 phage DNA ligase. The ligation reaction is carried out at 37 ° C., for one hour, by adding the loop oligonucleotide and 40 μl of T4 DNA ligase (New England Biolabs). The reaction is then stopped by adding an equivalent volume of a loading solution containing 80% formamide, 10 mM NaOH and 1 mM EDTA. The reaction mixture is heated at 90 ° C for 5 minutes, then loaded onto an 8% denaturing polyacrylamide gel.

The results of the various tests are given in FIG. 16. In lanes 3 and 4, the plasmids pBluescript SK + and pEGFPCl, respectively, are present in the ligation reaction. In lane 1, the loop oligonucleotide is not added.

It can be seen on examining the figure that in the presence of the loop oligonucleotide of 32 nucleotides, the linear oligonucleotide 63-mer is transformed into a circular molecule of 95 nucleotides. The low intensity intermediate band corresponds to a linear 95-mer, obtained when only one of the two ligation reactions is carried out. When the reaction is carried out in the presence of a plasmid containing the target sequence, under conditions allowing the formation of the triple helix, the radiolabelled oligonucleotide is transformed into a species having a very low migration speed, similar to that of the plasmid itself.

Example 15: Synthesis of a loop peptide-oligonucleotide conjugate.

An NLS peptide was conjugated to a loop oligonucleotide by the reaction described using FIG. 17. The oligonucleotide (6 nmol) carrying an amino group grafted in position 5 of a thymine is taken up in 50 μl of a 100 mM Borate pH 8.5 solution. 50 μl of a 30 mM solution of SMCC (Sigma catalog) in DMF are added and the mixture is left to incubate for 1 h 30 min. This results in an oligonucleotide-maleimide conjugate capable of reacting with a cysteine carried by a peptide. The excess SMCC is removed by passge on a nick-spin column (Amersham Pharmacio Biotech) pre-equilibrated with PBS buffer (GibcoBRL). The NLS peptide (100 nmol) is then added and the reaction is allowed to proceed overnight at room temperature with stirring. The reaction product is purified on a 12% denaturing polyacrylamide gel.

Claims

1 / Method for circularizing oligonucleotides around a double-stranded nucleic acid, characterized in that it comprises - the incubation of a double-stranded nucleic acid comprising a target sequence, with a single-stranded oligonucleotide comprising a central part capable of binding to said target sequence, linked by spacer sequences to 5 'and 3' terminal sequences, these sequences being

. either a) complementary to the sequence of an oligonucleotide matrix, or x) partially complementary to each other, so as to be able to hybridize to one another by forming a double strand while retaining one end in single strand, which is capable of hybridizing to the terminal sequence of another oligonucleotide, known as loop oligonucleotide, which has a hairpin structure, with a single strand part and a folded part which forms a double strand , said incubation step being carried out under conditions allowing the fixation of the single-stranded oligonucleotide at the level of the target sequence by formation of reversible bonds, which leads to the winding of the oligonucleotide around its target, and in case x), after a heating step, in order to allow the opening of the oligonucleotide and its winding around the target DNA before rehybridization of its ends,

the bringing into contact in the first case a) of the complex formed with said oligonucleotide matrix, under conditions allowing the hybridization of the 5 ′ and 3 ′ ends, so adjacent to the matrix, and in case x), bringing the complex formed into contact with said loop oligonucleotide, under conditions allowing, by hybridization of the free ends of the single-stranded oligonucleotide and of the oligonucleotide- loop, the juxtaposition of the 5 ′ and 3 ′ ends of these oligonucleotides, the ligation of the hybridized ends, so as to circularize the oligonucleotide, single strand (case a) or the oligonucleotide single strand and the oligonucleotide loop (case x) , around the double-stranded nucleic acid and, if desired,

- Isolation of the circularized structure, where the target nucleic acid is padlocked by the circular oligonucleotide, followed where appropriate by its purification.

2 / A method according to claim 1, characterized in that the nucleic acid used is a double helix DNA.

3 / A method according to claim 2, characterized in that the nucleic acid is a plasmid, in particular a supercoiled plasmid.

4 / A method according to any one of claims 1 to 3, characterized in that the single-stranded oligonucleotide comprises more than 20 nucleotides, in particular from 20 to 300 nucleotides, in particular from 30 to 100, with as central part, intended for bind to the nucleic acid target, a sequence of 10 to 25 nucleotides.

5 / A method according to any one of claims 1 to 4, characterized in that single stranded oligonucleotides are used in which the chain of nucleic bases contains bonds other than phosphodiester bonds, comprising phosphorothiate, phosphoramidate, methylphosphonate or peptide nucleic acid bonds, or mixtures of these bonds or, alternatively, these oligolinucleotides are chemically modified at the level sugars or at the base level, or are covalently linked to one or more ligands of low molecular mass, or contain one or more non-nucleoside bonds.

6 / A method according to any one of the preceding claims, characterized in that the single-stranded oligonucleotide binds to the target nucleic acid sequence and is wound around the double-stranded nucleic acid constituted, where appropriate, by double helix DNA.

7 / A method according to any one of the preceding claims, characterized in that the fixing of one single stranded oligonucleotide is carried out on a target DNA sequence in double helix and leads to the formation of a triple helix.

8 / A method according to any one of the preceding claims, characterized in that the target sequence on the double-stranded DNA is an oligopurine .oligopyrimidine sequence.

9 / A method according to any one of the preceding claims, characterized in that the binding of the oligonucleotide to the double-stranded nucleic acid, in particular the formation of a triple helix, is induced by adding to the reaction medium a ligand of low molecular weight. 10 / A method according to any one of the preceding claims, characterized in that the oligonucleotide matrix on which the ends of the single-stranded oligonucleotide will hybridize (case a) contains 10 to 36 nucleotides or that the oligonucleotide loops on which will hybridize to the free end of the single-stranded oligonucleotide (case x) contains from 16 to 100 nucleotides.

11 / A method according to any one of the preceding claims, characterized in that the step of incubating the nucleic acid with the oligonucleotide is carried out at a pH of 3 to 9, in particular from 4.5 to 8.5 , in particular from 6 to 8, with or without heating.

12 / A method according to any one of the preceding claims, characterized in that the hybridization step and the ligation are carried out after the incubation step, in a buffer identical or different from that used for incubation.

13 / A method according to any one of the preceding claims, characterized in that the ends of

1 oligonucleotide single strand (case a) or single strand and loop (case x) are connected by an enzymatic process.

14 / A method according to any one of claims 1 to 13, characterized in that the ends of one oligonucleotide single strand (case a) or single strand and loop (case x) are connected by a chemical process.

15 / A method according to any one of the preceding claims, characterized in that the isolation and purification of the circularized structure are carried out on gel, by chromatography, exclusion column, or using beads, optionally using a receptor for a ligand attached to the single-stranded oligonucleotide, the loop oligonucleotide or the olignucleotide matrix.

16 / Complex structure of nucleic acids, characterized in that it comprises a single-stranded oligonucleotide as used in the method according to any one of the preceding claims, fixed by reversible bonds to the target nucleic acid sequence, especially on a DNA double helix.

17 / Structure according to claim 16, characterized in that the oligonucleotide is attached to the nucleic acid target, wound around this target.

18 / Structure according to claim 16, characterized in that the oligonucleotide is attached to a DNA in double helix, forming a triple helix.

19 / Structure according to claim 18, characterized in that the target sequence of the double-stranded DNA is an oligopurine sequence. oligopyrimidine.

20 / Structure according to any one of claims 16 to 19, characterized in that it comprises a sequence with 10 to 25 triplets, formed by the oligopurine-oligopyrimidine sequence of the target to which the oligonucleotide is attached.

21 / Structure according to any one of claims 16 to 20, characterized in that the 2 ends of one oligonucleotide are hybridized, case a) on the oligonucleotide matrix, or case x) to each other and also to the looped oligonucleotide, the juxtaposed ends of the oligonucleotides being optionally connected to each other by enzymatic or chemical route.

22 / Complex structure of nucleic acids comprising a double-stranded nucleic acid encircled by a ring formed by a single-stranded oligonucleotide, the ring being optionally free to move along the target nucleic acid.

23 / Structure according to claim 22, characterized in that the movement of the ring is controlled by the addition or the removal of a specific ligand of the complex formed between the sequence of the oligonucleotide and that of the target sequence.

24 / Structure according to claim 22 or 23, characterized in that the nucleic acid is a double helix DNA, in particular a plasmid, in particular a supercoiled plasmid.

25 / Application of the circularization process according to any one of claims 1 to 15, for labeling or detecting double-stranded nucleic acids, in particular plasmids, comprising the use of a single-stranded oligonucleotide (case a) or of a loop oligonucleotide (case x), linked covalently or not to a molecule of interest, the non-covalent bond being able in particular to be obtained via the interaction biotin / streptavidin.

26 / Application of the circularization method according to any one of claims 1 to 15, for the detection of a specific sequence present on the target double-stranded nucleic acid, comprising the use of molecules coupled to the single-stranded oligonucleotide (case a) or loop (case x), or the implementation of a loop replication process.

27 / Application of the circularization method according to any one of claims 1 to 15, to differentiate two sequences differing by only 1 or 2 mutations, by detecting the circular oligonucleotide attached to one of these two sequences thanks to the molecule carried by this oligonucleotide or by a loop replication method.

28 / Application of the circularization method according to any one of claims 1 to 15, to select, for example from degenerate single-stranded nucleic acid sequences, nucleic acid sequences capable of binding to a double-stranded nucleic acid , and in particular sequences capable of promoting the penetration of double-stranded nucleic acids into cells, or the targeting of these nucleic acids towards specific cellular compartments, in particular towards the nucleus.

29 / Application of the circularization method according to any one of claims 1 to 15, for the purification of double-stranded nucleic acids, in particular of plasmids, comprising the use of a single-stranded oligonucleotide or of a linked loop oligonucleotide covalently to a molecule which can interact with another molecule attached to a solid support, in particular an affinity column.

30 / Application of the circularization method according to any one of claims 1 to 15, for the purification of double-stranded nucleic acids, in particular of plasmids, comprising the use of a single-stranded oligonucleotide circularizes around a plasmid and a matrix covalently linked to a molecule which can interact with another molecule attached to a solid support, in particular an affinity column.

31 / Application of the structures according to any one of claims 16 to 24, in gene therapy, for the specific inhibition of a gene carried by a double-stranded nucleic acid, in particular by a plasmid, by attachment of the oligonucleotide circular on the nucleic acid, and control of this fixation by molecules of low molecular mass, in particular by a specific ligand for triple helices.

32 / Application of the structures according to any one of claims 16 to 24, in which the single-stranded oligonucleotide (case a) or the looped oligonucleotide (case x) is covalently or non-covalently coupled to a molecule making it possible to make penetrating a nucleic acid into a cell, or even allowing targeting of the structure towards a cellular and / or nuclear compartment.