`
`PCT
`
`WORLD INTELLECTUAL PROPERTY ORGANIZATION
`International Bureau
`
`
`
`INTERNATIONAL APPLICATION PUBLISHED UNDER THE PATENT COOPERATION TREATY (PCT)
`
`(51) International Patent Classification 6 :
`
`(11) International Publication Number:
`
`WO 9631622
`
`C12Q 1/68
`
`(43) International Publication Date:
`
`10 October 1996 (10.10.96)
`
`(21) International Application Number:
`
`PCT/GB96I00848
`
`(81) Designated States: JP. US, European patent (AT, BE, CH, DE.
`DK, ES, FI, FR, GB. GR, IE, IT, LU. MC. NL. PT. SE).
`
`(22) International Filing Date:
`
`9 April 1996 (09.04.96)
`
`Published
`With international search report.
`Before the expiration of the time limit for amending the
`claims and to be republished in the event of the receipt of
`amendments.
`
`(30) Priority Data:
`95072385
`
`7 April 1995 (07.04.95)
`
`GB
`
`ISIS INNO-
`(71) Applicant (for all designated States except US):
`VATION LIMITED [GB/GB]; 2 South Parks Road. Oxford
`0X1 SUB (GB).
`
`(72) Inventors; and
`SOUTHERN, Edwin.
`(7S) Inventors/Applicants (for US only):
`Mellor [GB/GB]; 12 School Road, Kidlington OX5 ZHB
`(GB). PRITCHARD, Clare, Elizabeth [GB/GB]; 43 Cherry
`Tree Close, Southmoor. Abingdon 0X13 53E (GB). CASE-
`GREEN, Stephen, Charles [GB/GB]; Flat 2, Sylvia House,
`32A Union Street, Oxford 0X4 111’ (GB).
`
`(74) Agent: PENNANT. Pyers; Stevens Hewlett & Perkins,
`Serjeants’ Inn, Fleet Street. London EC4Y ILL (GB).
`
`l
`
`or variable number tandem repeat section may be analysed. Arrays of immobilised oligonucleotides are provided for use in the method.
`
`(54) Title: DETECTING DNA SEQUENCE VARIATIONS
`
`(57) Abstract
`
`A method of analysing a polynucleotide target involves incubating the target with an oligonucleotide probe, generally an array of
`immobilised oligonucleotide probes, to form a duplex, and using ligase or polymerase to extend one chain of the duplex. A point mutation
`
`
`
`FOR THE PURPOSES OF INFORMATION ONLY
`
`Codes used to identify States party to the PCT on the front pages of pamphlets publishing international
`applications under the PCT.
`
`Viet Nam
`
`United Kingdom
`Georgia
`Guinea
`Greece
`"“118“?
`Ireland
`Italy
`Japan
`Kenya
`Kyrgystan
`Democratic People’s Republic
`of Korea
`Republic of Korea
`Kazakhstan
`Liechtenstein
`Sri Lanka
`Liberia
`Lithuania
`Luxembourg
`Latvia
`Monaco
`Republic of Moldova
`Madagascar
`Mali
`Mongolia
`Mauritania
`
`Armenia
`Austria
`Australia
`Barbados
`Belgium
`Burkina Faao
`Bulgaria
`Benin
`Brazil
`Belarus
`Canada
`Central African Republic
`Congo
`Switurland
`COte d‘Ivoire
`Cameroon
`China
`Czechoslovakia
`Czech Republic
`Germany
`Denmark
`Estonia
`Spain
`Finland
`France
`Gabon
`
`GI!
`GE
`GN
`GR
`HU
`IE
`IT
`JP
`KE
`KG
`KP
`
`KR
`KZ
`LI
`LK
`LR
`LT
`LU
`LV
`M C
`MD
`MG
`ML
`MN
`MR
`
`Malawi
`Mexico
`Niger
`Netherlands
`Norway
`New Zealand
`Poland
`Portugal
`Romania
`Russian Federation
`Sudan
`Sweden
`Singapore
`Slovenia
`Slovakia
`Senegal
`Swaziland
`Chad
`Togo
`Tajikistan
`Trinidad and Tobago
`Ukraine
`Uganda
`United States of America
`Uzbekistan
`
`
`
`wo 96/31622
`
`'
`
`PCT/6396100848 .
`
`DETECTING DNA SEQUENCE VARIATIONS
`
`INTRODUCTION
`
`Detection of variation in DNA sequences forms the basis of
`
`many applications in modern genetic analysis: it is used in linkage analysis
`
`to track disease genes in human pedigrees or economically important traits
`
`in animal and plant breeding programmes; it forms the basis of
`
`IO
`
`15
`
`fingerprinting methods used in forensic and paternity testing [Krawczak
`
`and Schmidtke, 1994]; it is used to discover mutations in biologically and
`
`clinically important genes [Cooper and Krawczak. 1989]. The importance
`
`of DNA polymorphism is underlined by the large number of methods that
`
`have been developed to detect and measure it [Cotton, 1993]. Most of
`
`these methods depend on one of two analytical procedures, gel
`
`electrophoresis or molecular reassociation, to detect sequence variation.
`
`Each of these powerful procedures has its drawbacks. Gel electrophoresis
`
`has very high resolving power. and is especially useful for the detection of
`
`variation in the mini- and microsatellite markers that are used in linkage
`
`20
`
`analysis and fingerprinting; it is also the method used to analyse the
`
`variation found in the triplet repeats that cause a number of mutations now
`
`known to be the cause of around ten genetic disorders in humans
`
`[Vlfillems, 1994]. Despite its great success and widespread use. gel
`
`electrophoresis has proved difficult to automate: even the systems which
`
`25
`
`automate data collection require manual gel preparation; and as samples
`
`are loaded by hand. it is easy to confuse samples. The continuous reading
`
`electrophoresis machines are expensive, and manual analysis is
`
`technically demanding, so that its use is confined to specialised
`
`laboratories which have a high throughput. Furthermore, difficulties in
`
`30
`
`measuring fragment size preclude rigorous statistical analysis of the
`
`
`
`WO 96131622
`
`'
`
`-
`
`PCT/6396100848 _
`
`results.
`
`By contrast, oligonucleotide hybridisation lends itself to
`
`automation and to quantitative analysis [Southern et al.. 1992]. but it is not
`
`well suited to the analysis of variation in the number of repeats in the
`
`micro- and minisatellites, as the small fractional change in the number of
`
`repeats produces a barely detectable change in signal strength; and of
`
`course it would not be possible to distinguish two alleles in the same
`
`sample as each would contribute to a single intensity measurement. Thus,
`
`many different combinations of alleles would produce the same signal.
`
`Present hybridisation methods are much better suited to analysing
`
`variation in the DNA due to point mutation - base substitution deletions and
`
`insertions, for which it is possible to design allele specific oligonucleotides
`
`(ASOs) that recognise both the wild type and the mutant sequences
`
`[Conner et al., 1983]. Thus it is possible in principle, in a relatively simple
`
`test.
`
`to detect all possible genotypes. However, a problem that arises in
`
`practice in the use of oligonucleotide hybridisation is that in some cases
`
`the extent of reassociation is barely affected by a mismatched base pair.
`
`THE INVENTION
`
`The invention describes a general approach which can be
`
`applied to all forms of variation commonly used as DNA markers for
`
`genetic analysis.
`
`It combines sequence-specific hybridisation to
`
`oligonucleotides. which in the preferred embodiment are tethered to a solid
`
`support, with enzymatic reactions which enhance the discrimination
`
`between matching and non-matching duplexes, and at the same time
`
`provide a way of attaching a label to indicate when or which reaction has
`
`taken place. Two enzymatic reactions, chain extension by DNA dependent
`
`DNA polymerases and DNA strand-joining by DNA ligases, are dependent
`
`10
`
`15
`
`20
`
`on perfect matching of sequences at or around the point of extension or
`
`30
`
`joining. As we shall show, there are several ways in which these enzymes
`
`
`
`wo 96131622
`
`'
`
`-3-
`
`PCT/GB96/00848 .
`
`can be used with sequence-specific oligonucleotides to detect variation in
`
`target sequences.
`
`In all cases, the sequence to be analysed, the target
`
`sequence, will be available as a nucleic acid molecule, and may be a DNA
`
`molecule produced, for example, by the polymerase chain reaction.
`
`However, the methods are not confined to analysis of DNA produced in
`
`this way.
`
`In all applications, the target sequence is first captured by
`
`hybridisation to oligonucleotides which are preferably tethered to a solid
`
`support; for example, the oligonucleotides may be synthesised in situ as
`
`described [Maskos and Southern, 1992]; or they may be presynthesised
`
`and then coupled to the surface [Khrapko et al. 1991].
`
`In one aspect of the invention the novelty arises from the
`
`exploitation of enzymes in combination with substrates or primers tethered
`
`to solid supports. A further novelty exploits the observation that DNA
`
`ligases and polymerases can be used to distinguish sequence variants
`
`which differ in the number of units of a tandemly repeating sequence. This
`
`observation is surprising, as tandemly repeated sequences can form
`
`duplex in any register, thus in principle, length variants can form duplexes
`
`which match at the ends even when the two strands contain different
`
`numbers of repeat units. Although we demonstrate the application of this
`
`method in conjunction with tethered oligonucleotides, it should be evident
`
`that this reaction could be used to analyse VNTR (variable number tandem
`
`repeat) sequences in the liquid phase followed by some other method of
`
`IO
`
`15
`
`analysis, such as gel electrophoresis.
`
`25
`
`In one aspect the invention provides a method of analysis
`
`which comprises: providing a polynucleotide target including a nucleotide
`
`at a specified position, and an oligonucleotide probe, tethered to a support,
`
`said probe being complementary to the target and terminating at or close
`
`to the said specified position; and performing the steps:
`
`30
`
`a)
`
`incubating the target with the probe to form a duplex,
`
`
`
`WO 96/31622
`
`- 4 -
`
`PCTIGB96/00848 ’
`,
`
`b)
`
`incubating the duplex under ligation conditions with a labelled
`
`oligonucleotide complementary to the target,
`
`c)
`
`and monitoring ligation in b) as an indication of a point
`
`mutation at the specified position in the target.
`
`In another aspect the invention provides a method of analysis
`
`which comprises: providing a polynucleotide target having a variable
`
`number tandem repeat section and a flanking section. and an
`
`oligonucleotide probe having a section complementary to the repeat
`
`section and a flanking section of the target; and performing the steps:
`
`a)
`
`b)
`
`incubating the target with the probe to form a duplex,
`
`incubating the duplex with a labelled oligonucleotide and/or at
`
`least one labelled nucleotide under chain extension conditions,
`
`c)
`
`and monitoring chain extension as an indication of the length
`
`of the variable number repeat section of the target.
`
`A polynucleotide target is provided, in solution when the
`
`probe is tethered to a support, and may be DNA or RNA. This
`
`polynucleotide target is caused to hybridise with an oligonucleotide probe.
`
`The term oligonucleotide is used here, as common terminology for the
`
`' primers and substrates commonly utilised by polymerase and Iigase
`
`enzymes. However, the term is used in a broad sense to cover any
`
`substance that serves as a substrate for the enzymes, including single
`
`stranded chains of short or moderate length composed of the residues of
`
`nucleotides or of nucleotide analogues, and also longer chains that would
`
`ordinarily be referred to as polynucleotides.
`
`The probe may be tethered to a support, preferably by a
`
`covalent linkage and preferably through a 5' or 3' terminal nucleotide
`
`residue. An array of oligonucleotide probes may be tethered at spaced
`
`locations, for example on a derivatised glass surface or the surface of a
`
`silicon microchip, or alternatively on individual beads.
`
`In another aspect the invention provides an array of
`
`10
`
`15
`
`20
`
`25
`
`30
`
`
`
`WO 96131622
`
`'
`
`PCT/GB96/00848
`
`-5-
`
`oligonucleotides, for analysing a polynucleotide target containing a variable
`
`sequence, in which each component oligonucleotide i) comprises a
`
`sequence complementary to the target including an expected variant of the
`
`target. and ii) is tethered to a solid support in a chemical orientation which
`
`a) permits duplex formation with the target, and b)permits chain extension
`
`only when the sequence of the oligonucleotide matches the variable
`
`sequence of the target.
`
`In another aspect the invention provides a set or array of
`
`oligonucleotides, for analysing a polynucleotide target containing a variable
`
`number tandem repeat sequence, in which each component
`
`oligonucleotide i) comprises a sequence complementary to a part of the
`
`target immediately adjacent the repeat sequence,
`
`ii) comprises a
`
`sequence complementary to the repeat sequence of the target and
`
`containing a number of repeats expected in the target, and iii) is configured
`
`in a way that a) permits duplex formation with the target, and b) permits
`
`chain extension only when the number of repeats in the oligonucleotide
`
`equals or is less than the number of repeats in the target.
`
`In another aspect the invention provides an array of
`
`oligonucleotides in which different oligonucleotides occupy different
`
`locations and each oligonucleotide has a 3’ nucleotide residue through
`
`which it is covalently tethered to a support and a 5’ nucleotide residue
`
`which is phosphorylated.
`
`The invention also provides a method of making an array of
`
`different oligonucleotides tethered to different locations of a support, which
`
`method comprises the steps of: providing a first intermediate
`
`oligonucleotide tethered to the support and a second intermediate
`
`oligonucleotide in solution, and a third oligonucleotide that is
`
`complementary to both the first and second intermediate oligonucleotides.
`
`forming a duplex of the third oligonucleotide with the first and second
`
`intermediate oligonucleotides, and ligating the first intermediate
`
`10
`
`15
`
`20
`
`25
`
`30
`
`
`
`WO 96/31622
`
`'
`
`PCTIGB96I00848=
`
`-5-
`
`VI
`
`10
`
`15
`
`oligonucleotide with the second intermediate oligonucleotide; and
`
`repeating the steps with oligonucleotides tethered to different locations of
`
`the support
`
`Reference is directed to the accompanying drawings in which
`
`each of Figures 1 to 6 is a series of diagrams illustrating a method
`
`according to the invention.
`
`Figure 1 shows detection of point mutation by single base
`
`extension.
`
`Figure 2 shows detection of point mutation by hybridisation to
`
`allele specific oligonucleotides and chain extension.
`
`Figure 3A shows detection of point mutation by tag ligation to
`
`allele specific oligonucleotides.
`
`Figure SB shows detection of point mutation by ligation to
`
`library of differentially tagged allele specific oligonucleotides.
`
`Figure 4A shows analysis of variable numbered tandem
`
`repeats by ligation of tag to allelic variants.
`
`Figure 4B shows analysis of variable number tandem repeats
`
`by ligation of tag to allelic variants.
`
`Figure 5 shows measurement of variable number tandem
`
`repeats by labelled chain extension.
`
`Figure 6 shows analysis of variable number tandem repeats
`
`by ligation of tag followed by chain extension.
`
`DETAILED DESCRIPTION
`
`Detection of point mutation
`
`I. Single base-specific extension of tethered primers.
`
`in this application, the tethered oligonucleotide terminates at
`
`a position one base before the variable base in the target sequence
`
`(Fig.1). A nucleotide precursor triphosphate or dideoxyribonucleotide
`
`30
`
`triphosphate, labelled, for example with a fluorescent tag, is added in the
`
`
`
`
`
`wo 96131622
`
`‘
`
`PCP/GB96I00848 .
`
`-7-
`
`presence of a nucleic acid synthesising enzyme which requires a specific
`
`template in order to incorporate the complementary base. In the case of
`
`DNA polymerase. the labelled base will be incorporated from a
`
`deoxyribonucleotide precursor only if the precursor base is complementary
`
`to the base in the target sequence. Thus, mutants will give a negative
`
`result.
`
`ll. Chain extension from tethered ASOs.
`
`In this case, the tethered oligonucleotide terminates in a base
`
`which is complementary to the variable base in the target sequence.
`
`Labelled precursor nucleoside triphosphates and polymerase are added.
`
`Polymerisation takes place only if the last base of the primer is
`
`complementary to the variable base in the target (Fig.2). Thus. mutants
`
`will give a negative result.
`
`Ill. Ligation of tag sequences to tethered ASOs.
`
`In this method, the tethered oligonucleotide may end at the
`
`variable position in the target sequence. or it may end close to this
`
`position.
`
`ln either case, hybridisation of the target to the tethered ASOs
`
`will produce a substrate for ligating a tag oligonucleotide only if the bases
`
`at the join are well matched (Fig. 3a). Thus, mutants which are close
`
`enough to the joining position to prevent ligation will give a negative result.
`
`Alternatively. the tethered oligonucleotide may terminate at the base before
`
`the variable position; in this case. the ligation reaction can be carried out
`
`using a mixture of tag oligonucleotides, one for each of the possible
`
`alternative variants. Each tag would be labelled differently, for example,
`
`with a different fluophore, so that those that ligated could be recognised
`
`identifying the variant base (Fig. 3b).
`
`[0
`
`15
`
`20
`
`30
`
`
`
`WO 96/31622
`
`I
`
`PCT/GB96/00848 .
`
`-3-
`
`Analysis of VNTR Lengths by Ligation to Anchored VNTRS
`
`in this application “tag" oligonucleotides are ligated to sets of
`
`tethered oligonucleotides after hybridisation of the target. which acts as a
`
`template to bring the tags and the tethered oligonucleotides together.
`
`in Figure 4a, the tethered oligonucleotides comprise three
`
`parts, an “anchor" sequence which is common to all members of a VNTR
`
`set, which is attached to the solid support and which hybridises to a
`
`sequence flanking the variable region, a variable number of the repeated
`
`sequence unit, and a distal sequence. Each allele, represented by a
`
`different number of repeats, is located on a different solid support or at a
`
`different location on the same solid surface. Hybridisation of the target
`
`sequence will produce a series of duplexes. the structures of which depend
`
`on the number of units that the target contains.
`
`If the number matches the
`
`number in a tethered oligonucleotide, the target will meet the end of a tag
`
`when the tag is hybridised to the distal sequence of the tethered
`
`oligonucleotide.
`
`If the number is greater or smaller, there will be a gap in
`
`the duplex which reduces or prevents ligation of the tag.
`
`In Figure 4b, the tethered oligonucleotides comprise two
`
`parts: an “anchor" sequence which is common to all members of a VNTR
`
`set and which hybridises to a sequence flanking the repeated region, and a
`
`variable number of the repeated sequence unit. Each allele, represented
`
`by a different number of repeats. is located on a different solid support, or
`
`at a different location on the same solid surface. Hybridisation of the target
`
`sequence will produce a series of duplexes, the structures of which depend
`
`on the number of repeat units that the target contains (Fig.4b).
`
`If the
`
`number matches the number in a tethered oligonucleotide. the latter will
`
`meet the end of the tag when the tag is hybridised to its complement in the
`
`target sequence, and form a substrate for ligation of the tag.
`
`If the number
`
`is greater or smaller, there will be a gap in the duplex which reduces or
`
`10
`
`15
`
`20
`
`25
`
`30
`
`prevents ligation of the tag.
`
`
`
`
`
`W0 96/3 1622
`
`.
`
`PCTlGB96/00848 _
`
`-9-
`
`Analysis of VNTR Lengths by Chain Extension
`
`The number of repeat units in a VNTR may vary within small
`
`limits, and in these circumstances. the method of analysis described
`
`above, using a ligase. will be appropriate; in other cases. for example the
`
`trinucleotide repeats associated with a number of human inherited
`
`disorders. the variation may be too large to analyse in this way. For a
`
`number of triplet repeats. the variation can be from around 10-50 in the
`
`normal chromosome to more than a thousand in the affected individual
`
`(Table 1).
`
`It is probably unrealistic to measure such large numbers of
`
`repeats using the ligase reaction.
`
`In these cases. where the difference
`
`between the normal and the mutant allele is large, an alternative is to
`
`measure approximately the number of repeat units using labelled
`
`precursors with a pclymerising enzyme. The enzyme may be either a
`
`polymerase, such as DNA dependent DNA polymerase, or a ligase. In the
`
`former case, the oligonucleotides have to be tethered at their 5' ends to
`
`satisfy the requirement for enzymatic extension by the polymerase. The
`
`solid support carries an oligonucleotide anchor complementary to a
`
`sequence flanking the repeat unit; for example, the sequence can be that
`
`of one of the primers used to amplify the test sequence by the PCR. After
`
`hybridisation of the test sequence to the anchor, the repeat insert may be
`
`copied by a polymerase or a ligase (Fig. 5) incorporating a labelled
`
`precursor. The amount of label incorporated is proportional to the number
`
`of repeat units.
`
`lncompiete hybridisation of the target to the anchor
`
`sequence would give a deceptively low measure of the repeat number.
`
`This problem can be overcome by standardising the measurement in one
`
`of several possible ways. For example. if the target sequence itself is
`
`labelled as shown in Fig. 5, the final measurement will be a ratio of two
`
`labels: the target and the incorporated precursor.
`
`Alternatively, in the case of a triplet repeat, incorporation will
`
`end at the point in the sequence where the missing precursor base is
`
`l0
`
`15
`
`20
`
`25
`
`
`
`W0 96I3 1622
`
`I
`
`PCTIGB96I00848
`
`-10-
`
`needed for further extension; where a ligase had been used to polymerise
`
`monomers of the basic repeat unit, this will also end at the end of the
`
`VNTR insert. At this point a labelled "capping" sequence can be added by
`
`ligation.
`
`In such cases, the measurement will be the ratio of cap to
`
`polymer labels.
`
`10
`
`15
`
`2O
`
`25
`
`Analysis of VNTR Lengths by Combining Ligation and Chain
`
`Extension
`
`A more powerful way of analysing VNTRs which may vary in
`
`length over a wide range, would be to test first for ligation to a labelled tag
`
`oligonucleotide; this would give the results already described for targets
`
`with different lengths of repeats: a negative result where the VNTR lengths
`
`were longer or shorter in the target than in the tethered oligonucleotides,
`
`and positive results where they were the same. Following ligation, which
`
`we have shown can be made to go to completion, the different length
`
`classes will behave differently as substrates for DNA polymerase. Those
`
`targets in which the repeat number is less than that of the tethered probe
`
`will not act as substrates (Fig. 6c). Targets which have the same number
`
`of repeats as the probes will not be elongated by polymerase, because the
`
`ligated tag will block extension (Fig. 6a). The only cases where extension
`
`will occur are those for which the targets are longer than the probes (Fig.
`
`6b).
`
`If the analysis is done on an array of probes with different numbers of
`
`inserts up to a certain limit, there will be a clear indication of the number of
`
`repeats in the targets from the ligation results provided they are within the
`
`range of sizes represented in the array of probes.
`
`If, within the targets,
`
`there is one which is longer than this range, it will show up in the
`
`polymerase analysis. This test will be especially useful for the triplet
`
`repeats associated with the so-called “dynamic mutations", for example,
`
`that which is found in the fragile X mutation, where the size range varies
`
`3O
`
`from ca. 10-1000.
`
`it would be difficult to accommodate all of these size
`
`
`
`
`
`WO 96/31622
`
`‘
`
`PCTlGB96/00848
`
`- 11 -
`
`classes on a single array.
`
`EXPERIMENTAL SUPPORT FOR THE CLAIMS
`
`Properties of DNA Polymerases and Ligases
`
`Most DNA polymerases. reverse transcriptases. some DNA
`
`dependent RNA polymerases and Iigases can use as substrates one or
`
`more oligonucleotides which are bound to a long DNA strand through
`
`Watson-Crick base pairing.
`
`In the case of polymerases, an oligonucleotide
`
`is used as a primer to which the first base in the growing chain is added.
`
`In the case of Iigases. two oligonucleotides are joined provided that both
`
`are paired to the DNA strand and perfectly matched in the base pairs at or
`
`close to the junction point.
`
`It is these properties that make the enzymes
`
`useful for the detection of DNA sequence variation; in particular, the
`
`requirement for specific base pairing at the site of extension or joining
`
`complements the sequence discrimination that is already provided by the
`
`Watson-Crick pairing between the oligonucleotide and the target sequence
`
`that is needed to form a stable duplex. Thus. it has been found that
`
`discrimination by hybridisation alone is most sensitive if the variant base(s)
`
`is (are) close to the middle of the oligonucleotide. By contrast, for the
`
`enzymes, discrimination is highest if the variant mismatching bases are
`
`close to the end where the extension orjoin takes place. Together,
`
`hybridisation under stringent conditions and enzymatic extension or joining
`
`provide greater discrimination than either alone, and several methods have
`
`been developed to exploit this combination in systems for genetic analysis
`
`[references in cotton, 1993]. The hybridisation and the enzyme reaction
`
`are normally carried out in solution. following which the product is captured
`
`on a solid support, or separated by gel electrophoresis for detection and/or
`
`measurement.
`
`In one embodiment, the invention described here employs
`
`oiigonucleotides coupled to a solid surface, so that the advantages of
`
`I0
`
`15
`
`20
`
`25
`
`3O
`
`
`
`WO 96131622
`
`‘
`
`PCT1G396100848 ‘
`
`-12-
`
`working in mixed phase are brought to all steps: hybridisation, enzymatic
`
`extension or joining, and detection. This provides great sensitivity and
`
`convenience. As many different oligonucleotides can be bound to one
`
`surface in an array, it enables many different sequences to be analysed
`
`together, in a single reaction; this also ensures that all reactions are carried
`
`out under identical conditions, making comparisons more reliable.
`
`l0
`
`IS
`
`20
`
`Support-bound Oligonucleotides
`
`Two different methods have been developed for making
`
`oligonucleotides bound to a solid support: they can be synthesised in situ,
`
`or presynthesised and attached to the support.
`
`In either case. it is possible
`
`to use the support-bound oligonucleotides in a hybridisation reaction with
`
`oligonucleotides in the liquid phase to form duplexes; the excess of
`
`oligonucleotide in solution can then be washed away. Hybridisation can be
`
`carried out under stringent conditions, so that only well-matched duplexes
`
`are stable. When enzymes are to be used, the chemical orientation of the
`
`oligonucleotide is important; polymerases add bases to the 3' end of the
`
`chain; ligases join oligonucleotides which are phosphorylated at the 5' end
`
`to those with a 3'-OH group. Oligonucleotides tethered to the solid
`
`substrate through either end can be made in situ by using the appropriate
`
`phosphoramidite precursors [references in Beaucage and lyer, 1992]; or
`
`presynthesised oligonucleotides can be fixed through appropriate groups
`
`at either end. We will demonstrate that oligonucleotides can be
`
`phosphorylated at the 5’ end in situ using ATP and polynucleotide kinase,
`
`or they may be phosphorylated chemically [Horn and Urdea, 1986].
`
`Tethered Oligonucleotides as Substrates for DNA Modifling Enzymes
`
`The applications envisaged here require that the
`
`oligonucleotides tethered to the solid substrate can take part in reactions
`
`30
`
`catalysed by DNA polymerases and ligases.
`
`
`
`
`
`W0 96/3 1622
`
`‘
`
`PCI‘16396100848 _
`
`-13-
`
`DNA polymerase
`
`The M13 sequencing primer -
`
`SKGTAAAACGACGGCCAGT-B' - attached to aminated polypropylene
`
`through its 5' end was synthesised as described. A solution of M13 DNA
`
`(single-strand, replicative form, 0.1 pl, 200 nglpl) was applied in two small
`
`spots to the surface of the derivatised polypropylene. A solution containing
`
`three non-radioactive deoxyribonucleotide triphosphates, dATP, dGTP,
`
`'lTP (1O umol each), a32P-dCTP (1O pCi), Taq DNA polymerase and
`
`appropriate salts, was applied over a large area of the polypropylene,
`
`including the area where the M13 DNA had been spotted. The
`
`polypropylene was incubated at 37°C for 1 hr in a vapour saturated
`
`chamber.
`
`It was then washed in 1% SDS at 100°C for one minute, and
`
`exposed to a storage phosphor screen for one minute and scanned in a
`
`phosphorimager. The regions where the DNA had been applied showed a
`
`high level of radioactivity, against a low background where no DNA had
`
`been applied. This experiment shows that oligonucleotides tethered to a
`
`solid support can act as primers for DNA-dependent synthesis by DNA
`
`polymerase, as required for applications using this enzyme for mutation
`
`detection.
`
`Experiments described below show that both polynucleotide
`
`kinase and DNA ligase can be used to modify oligonucleotides tethered to
`
`a solid support. There are several ways in which phosphorylated
`
`oligonucleotides and the ligase reaction can be used to detect sequence
`
`variation.
`
`Methods for Making Arrays of Seguence Variants.
`
`1. Allele specific oligonucleotides for point mutations.
`
`For the preferred embodiment. it will be necessary to use
`
`oligonucleotides tethered to a solid support. The support may take the
`
`form of particles, for example, glass spheres, or magnetic beads.
`
`In this
`
`IO
`
`15
`
`20
`
`25
`
`30
`
`
`
`WO 96/31622
`
`.
`
`PCT/GB96/00848 _
`
`-14-
`
`case the reactions could be carried out in tubes. or in the wells of a
`
`microtitre plate. Methods for both synthesising oligonucleotides and for
`
`attaching presynthesised oligonucleotides to these materials are known
`
`[Stahl et al.. 1988]. Methods for making arrays of ASO's representing point
`
`mutations were described in patent application PCT/GBBQ/OO460 and in
`
`Maskos and Southern (1993). We also demonstrated how oligonucleotides
`
`tethered to a solid support in an array could distinguish mutant from wild
`
`type alleles by molecular hybridisation.
`
`For the present invention. the same methods could be used
`
`to create oligonucleotide arrays of A805. but in order that they can be
`
`used as substrates for the enzymes, they need to be modified; for ligation.
`
`it may be necessary to phosphorylate the 5' end; for extension by
`
`polymerase it will be necessary to attach to oligonucleotides to the solid
`
`substrate by their 5' ends.
`
`2. Arrays for scanning regions for mutations.
`
`It is often desirable to scan a relatively short region of a gene
`
`or genome for point mutations: for example, many different sites are
`
`mutated in the CFTR gene to give rise to cystic fibrosis; similarly, the p53
`
`tumour suppressor gene can be mutated at many sites. The large
`
`numbers of oligonucleotides needed to examine all potential sites in the
`
`sequence can be made by efficient combinatorial methods [Southern et al.,
`
`1994]. A modification of the protocol could allow such arrays to be used in
`
`conjunction with enzymes to look for mutations at all sites in the target
`
`10
`
`15
`
`20
`
`sequence.
`
`3. VNTRs.
`
`The most commonly used VNTRs are repeats of very short
`
`units, typically mono to tetranucleotides. However. there is another class,
`
`the minisatellites. in which the repeat unit is somewhat longer. up to 20 or
`
`
`
`
`
`WO 96/31622
`
`I
`
`PCTlGB96/00848 .
`
`-15-
`
`more nucleotides. The short repeats may be made using chemical
`
`synthesis; in the case of inserts with large numbers of repeat units. it would
`
`be more economical to use a synthetic route which used the repeat unit as
`
`a reactant, rather than building them up one base at a time; such methods
`
`have been used to make polynucleotides by chemical synthesis. An
`
`attractive alternative would be to build the repeat units by ligating the
`
`monomer units: they could be added stepwise, one unit at a time provided
`
`a method could be found to block one end to prevent polymerisation; for
`
`example the oligonucleotide building block may be terminated by a
`
`hydroxyl group. which is then phosphorylated after ligation so that the unit
`
`becomes an acceptor for the next one; the monomer may have a
`
`phosphate group protected by a cleavable group, such as a
`
`photocleavable group, which can be removed after ligation to allow a
`
`subsequent ligation [Pillai, 1980]. A second alternative, which would be
`
`especially favourable for longer units such the minisatellites, would be to
`
`attach either cloned or enzymatically amplified molecules to the solid
`
`support. For example. each variant sequence could be amplified by the
`
`PCR, using a biotinylated oligonucleotide for one of the primers. The
`
`strand starting with this group could then be attached to a streptavidin
`
`coated surface, and the other strand removed by melting [Stahl et al.,
`
`1988].
`
`IO
`
`)5
`
`20
`
`EXAMPLE 1
`
`DEMONSTRATION OF THE ANALYSIS OF LENGTH POLYMORPHISM
`
`BY LIGATION TO AN ARRAY OF VNTRs
`
`An array of VNTRs was made as described in Fig. 4b, in
`
`which the anchor sequence was 5'-tgtagtggtgtgatcaaggc-3'. The repeat
`
`unit was 5'-cttt—3’; stripes, ca 3 mm wide, of sequence variants of the form:
`
`30
`
`Anchor-Repeat“, with N = 4-10. were made as stripes on the surface of a
`
`
`
`W0 96/3 1622
`
`‘
`
`PCT[6396100848 .
`
`-15-
`
`sheet of polypropylene. The synthesis was carried out using 3'-
`
`deoxyribophosphoramidites, this chemical orientation produces
`
`oligonucleotides tethered through their 3' ends to the polypropylene, and a
`
`free 5’ hydroxyl group. To create a substrate for ligation, this OH group
`
`was phosphorylated by immersing a strip of the polypropylene (3mm x
`
`18mm), carrying the array of oligonucleotides, in 0.5 ml of a solution
`
`containing 4mM AT