 |
Description  |
|
|
TECHNICAL FIELD
The present invention relates to the analysis of DNA or RNA, particularly
for disease diagnosis.
BACKGROUND OF THE INVENTION Nucleic acid analysis, particularly for DNA, is
becoming important in the diagnosis of infectious as well as genetic
disease (Caskey, C. T. (1987), Science, 236, 1223). The inheritance of a
substantial number of disease traits can be predicted by analysis of
genetically linked markers such as restriction fragment length
polymorphisms (RFLP). Moreover, for an increasing number of genetic
diseases, the genes involved have been identified and mutant alleles
characterized. The more common genetic diseases fall in either of two
broad categories: those that can be attributed to any of several
independent mutational events, typically seen in X-linked recessive
diseases and those that are caused by homozygozity for one of a few
alleles, which may be maintained at an appreciable level in the population
because of an advantage to the heterozygous carriers (Rotter, J. I., et
al. (1987), Nature, 329 287). The characterization of such common disease
causing alleles makes large scale screening for carrier status a
possibility, limited by the cumbersome nature of the available detection
methods.
Recent estimates indicate that about 85% of mutations occurring in the
human genome are point mutations, that is they involve a very short
sequence of nucleotides. Such mutations typically comprise substitutions
of at least one nucleotide for another. This is in contrast to grosser
changes such as deletions, insertions, inversions or duplications of
longer DNA sequences. Current detection procedures capable of detecting
single base substitutions include procedures based on differential
denaturation of mismatched probes, as in allele specific oligonucleotide
hybridization (Wallace, R. B., et al. (1979), Nucl. Acids Res., 6, 3553)
or denaturing gradient gel electrophoresis (Myers, R. M., et al. (1985),
Nature. 313, 495). Alternately, the sequence of interest can be
investigated for abberations using restriction enzymes as in RFLP analysis
(Geever, R. F. (1987), Proc. Natl. Acad. Sci. USA. 78, 508) or using RNAse
A to cleave a mismatched nucleotide of an RNA probe hybridized to a target
molecule (Myers, R. M., et al. (1985), Science, 230, 1242; Winter, E., et
al. (1985), Proc. Natl. Acad. Sci. USA, 82, 7575). While the two
techniques using denaturing gradient gels or RNAse A have the advantage of
surveying long stretches of DNA for mismatched nucleotides, they are
estimated to detect only about half of all mutations involving single
nucleotides. Similarly, it is suggested that only approximately one third
of the nucleotides in the human genome can be analyzed as part of
restriction enzyme recognition sequences. The only existing technique
capable of identifying any single nucleotide differences, short of
sequencing, is allele specific oligonucleotide hybridization. This
technique involves immobilizing separated (Wallace, supra) or
enzymatically amplified fragments of test DNA (Saiki, R. K., et al.
(1986), Nature, 329, 166), hybridizing with oligonucleotide probes, and
washing under carefully controlled conditions to discriminate single
nucleotide mismatches.
Whiteley, N. M. et al., EPO Publication No. 0 185 494, discloses an assay
in which target sequences in nucleic acids are identified using two
oligonucleotide probes of different length (e.g. a 15-mer and a 65-mer),
selected to hybridize to contiguous regions of a target nucleic acid which
has been previously immobilized. One of the probes is labeled. The short
probe includes a potential mutation. Hybridization and/or washing is
stated to be performed under conditions of stringency so specific that one
probe complementary to a particular sequence will not hybridize (or remain
hybridized during high stringency washes) if it is mismatched in one
position. If hybridization of both probes occurs under high stringency
conditions, the two probes are joined by ligation. Hybridization and
ligation is determined by detecting whether label is incorporated in a
ligated product. The presence of the label is stated to be an indication
of the matching, at the selected stringency, of the short probe with the
target nucleic acid sequence. However, this technique is subject to
imprecision in that the stringency conditions of hybridizing are difficult
to adjust so accurately that they can distinguish hybridization
differences based upon a single base pair mismatch.
The references discussed above are provided solely for their disclosure
prior to the filing of the instant case, and nothing herein is to be
construed as an admission that such references are prior art or that the
inventors are not entitled to antedate such disclosure by virtue of prior
invention or priority based on earlier filed applications.
SUMMARY OF THE INVENTION
In accordance with the present invention, an assay is provided for
determining the nucleic acid sequences in a region of a nucleic acid test
substance, typically DNA or RNA, of a type having a known normal sequence
and a known possible mutation at at least one target nucleotide position.
At least two oligonucleotide probes are selected to anneal (hybridize) to
immediately adjacent segments of a substantially complementary test DNA or
RNA molecule. One of the probes (the target probe) has an end region
consisting of the end nucleotide at the probe junction and one or more
nucleotides removed therefrom (the positions collectively called "the
target probe end region"), wherein one of the end region nucleotides is
complementary to and therefore capable of base pairing with either the
normal or abnormal nucleotide at the corresponding target nucleotide
position. A linking agent, preferably a ligase, is added under conditions
such that when the target nucleotide is correctly base paired, the probes
are covalently joined and if not correctly based paired the probes are
incapable of being covalently joined under such conditions. The annealed
probes are separated from the test substance, as by denaturing. Then, the
presence or absence of linking is detected as an indication of the
sequence at the target nucleotide. This procedure is capable of
identifying known mutations in samples of genomic DNA.
The present invention specifically provides a technique for detecting one
or more point mutations in biologically derived DNA or RNA test molecules,
such as genomic DNA or mRNA. The use of a ligase to determine base pairing
in the end region of a probe sets this technique apart from other methods
dependent on differential denaturation of probes which are matched or
mismatched to a test nucleotide sequence.
The technique includes the steps of
(a) annealing an oligonucleotide target probe of predetermined sequence to
a first sequence of a test substance so that said target nucleotide
position is aligned with a nucleotide in an end region of the target
probe,
(b) annealing an adjacent oligonucleotide probe of predetermined sequence
to a second sequence of said test substance contiguous to said first test
substance sequence, so that the end region of said target probe is
directly adjacent to said adjacent probe,
(c) contacting said annealed target and adjacent probes with a linking
agent under conditions such that the directly adjacent ends of said probes
will link to form a linked probe product unless there is nucleotide base
pair mismatch between the target probe and test substance at the target
nucleotide position,
(d) separating said test substance from said annealed probes, and
(e) detecting whether or not linking occurs as an indication of nucleotide
base pair matching or mismatching at said target nucleotide position.
In a preferred embodiment, at least one of the target or adjacent probes
includes a label which is detected. One way to accomplish this is to label
only one of the probes and to immobilize the non-labeled one before or
after linking. The immobilized probe fraction, containing the linked probe
product, is then melted from the test substance and separated from the
remainder of the medium. If the label is present on the immobilized probe
fraction, this indicates that linkage has occurred.
In another embodiment, the reaction mixture of steps (a) through (d) is
placed into a migration substance, such as a gel, in which the probes are
caused to migrate, e.g. by electrophoresis. If the two probes are linked
to form a linked probe product, it travels at a different rate than the
unlinked probes, and so its presence may be detected by observing the
position of the label in the migrating medium.
In another embodiment, a diagnostic kit for the detection of abnormalities
in the DNA or RNA test substance is provided. The kit includes a container
of a first target probe capable of annealing to a first portion of the
test substance. The target probe has a nucleotide end region complementary
to the normal or mutated nucleotide at the mutation position, i.e. the
target nucleotide position. The kit also includes a probe capable of
serving as an adjacent oligonucleotide probe for annealing to a second
portion of the test substance contiguous to the first portion. Preferably,
one of the probes is labeled and the other one is immobilized or includes
a binding moiety capable of being immobilized. A covalent linking agent,
such as a ligase, may also be included.
BRIEF DESCRIPTION OF THE DRAWINGS
FIG. 1 is one diagnostic approach in accordance with the present invention.
FIGS. 2A and 2B illustrate two typical sequences of target oligonucleotide
probes and adjacent probes for determining alleles encoding normal and
sickle-cell globin. One oligonucleotide probe is specific for the globin
.beta..sup.A gene and the other for the globin .beta..sup.S gene.
FIG. 3 illustrates the detection of the .beta..sup.A and .beta..sup.S
globin alleles contained in plasmid DNA, genomic DNA and nucleated cells.
FIG. 4A, 4B and 4C illustrate detection of the .beta..sup.A and
.beta..sup.S alleles in nucleated blood cells and the failure to detect
such sequences in a gene deletion sample control using differentially
labelled target probes.
FIG. 5 illustrates the strategy used to determine the effect of nucleotide
position on ligation.
FIG. 6 is an autoradiogram illustrating the results of Example 3.
FIGS. 7 and 8 illustrate the strategy employed and results obtained in
Example 4.
FIGS. 9, 10A and 10B illustrate the strategy employed and results obtained
in Example 5.
FIG. 11 illustrates the results obtained in Example 6.
DETAILED DESCRIPTION OF PREFERRED EMBODIMENTS The assay of the present
invention is particularly useful for analyzing nucleic acids (a DNA or RNA
test substance) for the diagnosis of infectious and genetic diseases. A
major advantage of the technique is that it uses two complementary
phenomena to provide an exquisitely sensitive overall composite assay for
determining the presence or absence of a point mutation in the test
substance. One technique is the use of oligonucleotide probes which are
complementary to two contiguous predetermined sequences of the test
substance. These are the target and adjacent probes in FIG. 1. If these
probes anneal in a juxtaposed position, there is a reasonable certainty
that the sequence being investigated is the relevant one. In addition, the
nucleotide sequences of the oligonucleotide probes are selected so that
the target nucleotide is positioned to determine if base pairing with the
end region of the target probe occurs. In other words, the target probe
end region is aligned with the known position of the point mutation. The
annealed probes are then exposed to a linking agent which will link the
adjacent ends of the probes only if the nucleotides base pair at the
target nucleotide position. Then, the presence or absence of ligation is
determined by one of a number of techniques to be described below. In FIG.
1, the technique depicted involves denaturation to produce a single
stranded linked probe product or only unligated target and adjacent probes
if no ligation occurs. The 5' end of the linked probe product has biotin
attached thereto ("B") and a label such as .sup.32 P attached to the 3'
end ("*"). The denatured reaction mixture is then contacted with a biotin
binding support such as stretavidin bound to a solid support. This permits
isolation of the target probe if no ligation occurs or the isolating of
the labelled linked probe product if ligation has occurred. This technique
and others disclosed herein provides for the positive identification of
each possible allele thereby permitting the identification of heterozygous
and homozygous states.
A typical DNA test substance is a portion of a gene of known sequence
associated with a known possible genetic disease caused by a single
nucleotide mutation. One such disease is sickle-cell anemia. In this
instance, the normal and abnormal DNA test substance used in the assay are
allelic. The assay may be performed with test DNA derived from purified
genomic DNA or may be performed in situ with the test DNA present in
cells.
The normal nucleotide and amino acid sequence of the relevant portion of
the .beta.-globin gene is shown in FIG. 2A. The same relevant portion of
the nucleotide in the amino acid sequence of the sickle cell gene is shown
in FIG. 2B. As can be seen, a point mutation in the codon for amino acid 6
(GAG) results in the substitution of the normal amino acid glutamine in
position 6 with the amino acid valine. Thus, in sickle cell disease, a
single point mutation results in the substitution of one amino acid for
another.
In accordance with the invention, three probes are shown in FIGS. 2A and 2B
which may be used to detect the presence of the normal .beta.-globin gene
or sickle .beta.-globin gene. These probes are designated P-133, B-131 and
B-132. (The "P" designation indicates that a probe is labeled with .sup.32
P at its 5' terminus whereas the "B" designation indicates the probe has
biotin attached to it at its 5' terminus.) As indicated, the B-131 and
P-133 probes are complementary to the relevant nucleotide sequence in the
normal .beta.-globin gene whereas the two probes B-132 and P-133 are
complementary to the relevant portion of the .beta..sup.S globin gene. The
probes B-131 and B-132 are the target probes which are identical except
for their end region which is contiguous with the adjacent probe P-133. As
indicated, the 3' end of the B-131 probe ends with the nucleotide A which
is complementary to the nucleotide T in the non-coding strand of normal
.beta.-globin. The target probe B-132, on the other hand, has an end
region containing the terminal nucleotide T which is complementary to the
A which is present in the non-coding strand in the relevant portion of the
.beta..sup.S allele. As will be shown, these probes (or probes
representing the complementary sequence) can be readily used to
distinguish DNA encoding normal .beta.-globin from DNA encoding sickle
.beta.-globin.
In addition to directly detecting point mutations within a structural gene,
the assay of the invention may also be used to indirectly detect the
presence of normal or defective structural genes. Thus, for example,
various DNA sequence polymorphisms genetically linked to the structural
gene for known genetic defects are known to be highly correlated with that
particular defect. Thus, in the case of sickle cell disease, a linked HpaI
restriction endonuclease polymorphism has been found to be highly
correlated with the .beta..sup.S defect. (Kan, Y. W., et al. (1978), Proc.
Natl. Acad. Sci., 75, 5631). In this case, appropriate target and adjacent
probes can be designed and synthesized to detect the presence of the HpaI
restriction site thereby permitting detection of a marker linked to the
.beta..sup.S allele. The indirect determination of a gene defect is most
practical when a particular polymorphism is highly correlated with a
defective phenotype. Thus, linked markers are useful, for example, when
the defect is not known (e.g. Huntingtons disease) or when there are many
different defective alleles (e.g. Lesch Nyhans disease). In either case it
may be possible to follow the inheritance of these disorders and the
prenatal diagnosis thereof, if a known polymorphism is associated with the
particular defect which may be detected by the assay of the invention.
The system can also be used to detect RNA sequences, e.g. those present in
populations of mRNA molecules expressed in different tissues or rRNA
molecules characteristic of specific infectious organisms, by use of
complementary DNA or RNA probes. Preferably, inhibitors of the enzyme
ribonuclease should be added to prevent degradation of the RNA. Assays for
RNA are of value for example, in the quantitative analysis of gene
expression in various tissues and in the analysis of infectuous organisms
expressing multiple copies of known RN molecules.
DNA probes are preferred for a DNA or RNA test substance because of their
ease of synthesis and stability. DNA probes are suitably synthesized in an
automated fashion using an automated DNA synthesis instrument such as
Applied Biosystems, Inc. model 380A DNA Synthesizer or may be formed by
other methods known to those skilled in the art.
Conditions for annealing DNA probes to DNA test substances and to RNA test
substances are well known, e.g. as described in Nucleic Acid
Hybridization, A Practical Approach, Eds. B. D. Homes and S. J. Higgins,
IRL Press, Washington, D.C. (1985) and by Wetmur, J. G. and Davidson, N.
(1968), Mol.Biol., 31 349. In general, whether such annealing or
hybridization takes place is influenced by the length of the probes and
the test substances, the pH, the temperature, the concentration of
mono-and divalent cations, the proportion of G and C nucleotides in the
hybridizing region, the viscosity of the medium and the possible presence
of denaturants. Such variables also influence the time required for
hybridization. The preferred conditions will therefore depend upon the
particular application. Such conditions, however, can be routinely
determined without undue experimentation.
A critical feature of the present invention is the careful selection of the
adjacent ends of the two oligonucleotide probes. The target probe when
annealed to the test DNA or RNA sequence is selected so that the probe end
region is positioned over the target nucleotide. In this regard, the end
region of the target probe (which contains a nucleotide complementary to
the target nucleotide in the normal or abnormal test substance) consists
of the end nucleotide (i.e., the 5' or 3' nucleotide immediately adjacent
to the adjacent probe) and one, two, or more nucleotides removed from the
target probe terminal nucleotide. This end region may extend from the end
nucleotide through about the third nucleotide from the juncture point with
the adjacent probe. As indicated in the examples, the end nucleotide in
the target probe and the nucleotide immediately adjacent thereto in the
target probe are nucleotide positions which if not properly base paired
with the first sequence of a test substance results in substantially no
ligation between the test and adjacent probes, whereas if proper base
pairing is present, ligation occurs. However, the base pair
matching/mismatching between the target nucleotide and
complementary/non-complementary nucleotide in the test probe that will
effect ligation is believed to extend beyond the second nucleotide in the
end region such that ligation occurs, or fails to occur, because of base
pair matching or mismatching in the third nucleotide in the target probe.
Thus, the target probe end region can be defined functionally as any of
those nucleotides extending from the 5' or 3' end of the target probe
immediately contiguous with the adjacent probe which when base pair
mismatched with the corresponding target nucleotide in the test substance
results in substantially no ligation between the target and adjacent
probes
The system detects whether or not the two probes are linked in the presence
of the linking agent. Annealing and ligation may be performed
simultaneously under appropriate conditions. Alternatively, it may be
desired to perform the steps sequentially so that pre-selected conditions
adapted for annealing may be utilized and other conditions adapted for
ligation may be used for that step.
1. Homogeneous Annealing and Liqation
In one embodiment of the invention, both probes are initially in solution
phase and the reaction is carried out in solution. For simplicity the
system will be described with respect to DNA probes and DNA test
substances. A DNA test substance containing a selected region which may
contain a point mutation is denatured by conventional means. One of the
probes is labeled. The other one is attached to a first binding moiety
which is capable of being subsequently immobilized by attachment to a
second immobilized binding moiety.
The two probes are selected to be substantially complementary to the
selected region of the DNA test substance and directly adjacent to each
other in a head to tail relationship. It is important that the region
suspected of containing the mutant nucleotide is selected to be in the
target probe end region which may be in either the labeled or unlabeled
probe. It is preferable to select the terminal or end nucleotide because
ligation would be most sensitive to mismatching at that position. However,
one or more of the immediately adjacent nucleotides may be selected to
align with the target nucleotide. This may be preferable when, for
example, the target nucleotide corresponds to one of two adjacent
nucleotides in a codon which may be separately mutated to produce a known
mutant amino acid substitution. Thus, it is possible to simultaneously
assay several target nucleotide positions for complementarity to two
target probes (containing separably detectable labels) each of which is
complementary to one of two possible nucleotide target positions (or one
or more different nucleotides at a single nucleotide target position).
The ultimate size and sequence of the oligonucleotide probes to be used
will be determined by the functions which they perform. For example, the
probes (except for the target probe end region as discussed elsewhere
herein and the corresponding "end region" of the adjacent probe contiguous
with the target probe) are selected to be "substantially" complementary to
the selected regions of hybridization of the test DNA or RNA. For this
purpose, although it is preferable for the probes to match the exact
sequence of the test substance template, this is not necessary if, under
the conditions of annealing, the probes will hybridize with the test
regions. Thus, the stringency conditions of annealing should be selected
so that this occurs when the probes are sufficiently complementary to
permit annealing of the probe to a complementary selected region of the
test substance or to one containing a limited number of internal
variations. In the latter case, it is well known to those skilled in the
art that a certain number of mismatches may be tolerated within an
oligonucleotide probe, vis-a-vis the ability of that probe to hybridize to
a particular DNA sequence under given conditions.
The number of mismatches that can be tolerated is dependent upon the number
of nucleotides in a particular probe. Thus, as a general rule of thumb,
each nucleotide added to a given probe will increase the melting
temperature of its corresponding DNA duplex by about 3.degree. C. However,
an internal mismatch between the oligonucleotide probe and the
substantially complementary DNA strand reduces the melting temperature by
approximately 5.degree. C. In general, the longer the probe, the greater
the number of internal mismatches that may be tolerated under the given
hybridization conditions. Thus, the annealing conditions and/or probe
lengths may be selected to permit annealing to nucleic acids having
non-conserved regions, e.g., DNA or RNA from organisms that are known to
mutate at a high rate (e.g. HIV) if determining the presence of such
organisms. On the other hand, stringency should be sufficient to prevent
the two probes from fortuitously hybridizing in an incorrect region of the
test substance.
Since there is flexibility in the stringency with which annealing may
occur, it is possible to simultaneously investigate several different loci
in a given test sample for different point mutations. In so doing, test
and adjacent probes unique for other point mutations may be employed
simultaneously with other sets of target and adjacent probes (provided
there are means to detect the ligation of each set of target and adjacent
probes). Since each set of probes may have a variable G/C content the
probes annealed to a test substance may have variable stability and
therefore melt (denature) at different temperatures. However, since there
is no need to finely tune the stringency conditions in practicing the
present invention, a simultaneous assay at multiple loci is possible.
As indicated, the length of the oligonucleotide probes affects the
conditions of annealing. If the probes are too short, stable hybridization
may not occur when the probe is internally mismatched in a region away
from the target nucleotide. On the other hand, if the probes are too long,
it may be possible to anneal in a region other than that of the
predetermined target template.
Another factor relating to probe length is the complexity of the DNA test
sample. In the human genome, it is estimated that there is a non-repeating
length of about 3.times.10.sup.9 nucleotides. Thus, to prevent incorrect
annealing and ligation of the probes, the combined length of the target
and adjacent probes should be of the order of 19 nucleotides or longer.
This represents the approximate lower limit in the overall length of the
two probes. Thus, a 9-mer and 10-mer could be used as target and adjacent
probes.
If the test DNA used in the assay is amplified, such as by the method
"polymerase chain reaction" described by Saiki, et al. (1985), Science,
230, 1350-1354, shorter probes can be used. This is because there is less
likelihood for nonspecific annealing due to the fact that the regions
selectively amplified by such techniques comprise a very small portion of,
for example, the human genome thereby reducing the complexity of the
sample being assayed.
Excellent results are obtained whether assaying genomic or amplified DNA
samples with probes that are in the range of 16-20 nucleotides in length.
Longer probes such as 30-mers, or longer, may also be used.
For detection, any known label for proteins, peptides or nucleic acids may
be used in accordance with the present invention. Such labels include
radioactive tags, enzymes, fluorescent tags, and colorimetric tags.
Specifically, one preferred form of labelling is to add phosphorous 32
(.sup.32 P) to the 5' hydroxyl group of the oligonucleotide probe using
the enzyme polynucleotide kinase. Other labels can be used such as iodine
125 (.sup.125 I) although labeling with this radioisotope would be less
convenient. The oligonucleotides can also be labeled with organic
fluorophors such as carboxy-fluorescein or
carboxy-2,7-dimethoxy-4,6-dichlorofluorescein, e.g. by coupling to a 5'
aminothymidine in the probe. Also, chelated rare earth metal ions such as
Eu.sup.3+ and Te.sup.3+ may give a high signal to noise ratio permitting
sensitive detection utilizing time resolved fluorescence techniques
(Soini, et al. (1983), Clin. Chem., 27 65). In addition, enzymes may be
attached to one of the probes and used as a detection label analogous to
an ELISA assay (Jablonski, et al. (1986), Nucl. Acids Res., 14, 6115).
In general, it is desirable to use fluorescent rather than radioactive
probes. Advantages to this approach include safety aspects, stability of
the reagents, and immediate access to the result. In addition, fluorescent
detection permits multicolor analysis whereby the presence of alternate
alleles or a quantitative comparison of two genes can be analyzed in an
internally controlled fashion.
In some instances, it may be desirable to increase the amount of test DNA
before the assay of the invention using the recently described polymerase
chain reaction (Saiki, et al. (1985), Science. 230, 1350-1354). Such
instances occur, for example, when genes are present at less than one copy
per genome equivalent. Examples include rearranged lymphocyte receptor
genes in populations of lymphoid cells or parasitic genomes contaminating
the host cells at low frequency. In addition, amplification is of value
when a limited amount of material is available for analysis. In this
amplification procedure a small segment of DNA can be exponentially
amplified by repeated cycles of copying of a template DNA to produce new
strands from two primers, one with a sequence upstream from, and the other
in the opposite orientation downstream from, the DNA segment of interest.
Amplification is performed by cyclically varying the temperature, each
cycle including a denaturing step, a step for the annealing of primer
oligonucleotides and a step where a DNA polymerase extends a new
complementary strand from the primers by incorporating nucleotides.
After amplification, the target and adjacent probes for detecting a normal
or abnormal sequence in the amplified DNA segment of interest are added
and annealed under the appropriate conditions. Then, during or after
annealing, a linking agent is added which is capable of forming a covalent
bond between the contiguous probes. A linked probe product is formed if
the correct base pairing is present at the juncture of the two probes.
The preferred linking agent is a ligase, preferably T4 DNA ligase, using
well known procedures (Maniatis, T. in Molecular Cloning, Cold Spring
Harbor Laboratory (1982)). Other DNA ligases may also be used. T4 DNA
ligase may also be used when the test substance is RNA (The Enzymes, Vol.
15 (1982) by Engler M. J. and Richardson C. C., p. 16-17. Methods in
Enzymology, Vol. 68 (1979) Higgins N. P. and Cozzarelli N. R. p. 54-56).
With regard to ligation, other ligases, such as those derived from
thermophilic organisms may be used thus permitting ligation at higher
temperatures allowing the use of longer probes (with increased
specificity) which could be annealed and ligated simultaneously under the
higher temperatures normally associated with annealing such probes. The
ligation, however, need not be by an enzyme and, accordingly, the linking
agent may be a chemical agent which will cause the probes to link unless
there is a nucleotide base pair mismatching at the target nucleotide
position. The invention will be described using T4 DNA ligase as the
linking agent. This enzyme requires the presence of a phosphate group on
the 5' end that is to be joined to a 3' OH group on a neighboring
oligonucleotide.
Ligation conditions are adjusted so that ligation will occur if there is a
base pair match at the target nucleotide position and will not occur if
there is a mismatch at that position. Assuming simultaneous annealing and
ligation, the ligation may be performed at a temperature below the melting
temperature of the annealed oligonucleotide probes. A suitable temperature
for this purpose is about 5.degree. C. to 30.degree. C. below the melting
temperature of the hybridized sequences. For ligation after hybridization,
a suitable temperature is about 37.degree. C. for T4 DNA ligase. In the
examples, simultaneous annealing and ligation takes place at about
37.degree. C. Factors that determine whether or not mismatching at the
target nucleotide position can be discrimated, as detected by ligation,
include salt concentration and the amount of enzyme (ligase) used.
Suitable salt concentrations range from 0 to 200mM. Typically, the
oligonucleotide probes are present in large excess of the test substance,
e.g. 10.sup.3 times or more. Suitable ligase concentrations range from
10.sup.-4 to about 1 Weiss unit for such probe concentrations. The
preferred salt and ligase concentrations are readily determined
empirically for a particular application. Thus, as described in Example 5,
150-200 mM NaCl and 0.15-50.0.times.10.sup.-3 Weiss units of ligase are
preferred when T4 DNA ligase is used.
After hybridization and treatment with ligase, ligation of the target and
adjacent probes (to produce the "linked probe product") may be detected by
a number of techniques. In each of these techniques, it is preferable that
the linked probe product (if formed) be separated from the test substance
to ensure that ligation rather than only hybridization is being detected.
Thus, DNA/DNA or DNA/RNA duplexes containing linked probe product may be
denatured by techniques known to those skilled in the art.
The detection of ligation is preferably performed by detecting the label
contained by either the target probe or adjacent probe. Thus, for example,
ligation can be determined by electrophoretic techniques such as
polyacrylamide gel electrophoresis (PAGE) under denaturing conditions. In
this technique, unligated labeled probe of known size is separated from
labeled linked probe product which is of a known larger size. Thus,
ligation may be detected during or after electrophoresis. In the former
case, a label detector may be mounted to detect label as it passes a
particular point in the electrophoretic medium during electrophoresis. In
such a system, unligated label would be detected followed by the linked
probe product (if formed) at a predictable time depending upon the
location of the label detector and the electrophoretic conditions. Devices
capable of detecting fluorescent labels in this manner are known in the
art (Smith, L. M. et al. (1986) Nature 321:672; Prober, J. M., et al.
(1987), Science, 238, 336). Alternatively, the electrophoresis may be
completed and the label detected by standard techniques such as by
autoradiography of radioisotope labelled probes
Another technique for detecting ligation employs the immobilization of the
linked probe product. In this technique, the adjacent probe contains a
first binding moiety which is reactive to a second binding moiety. The
second binding moiety is immobilized on a solid support. The adjacent
probe containing the first binding moiety together with the target probe
containing the label is hybridized with test substance and treated with a
ligase. Then, the linked probe product (if formed) containing the first
binding moiety is bound to its corresponding immobilized second moiety.
One such first binding moiety is biotin which may be covalently attached
to the probe by known techniques. The binding moiety is subsequently bound
to an immobilized complementary second moiety such as streptavidin
immobilized on a solid support such as agarose (Updyke, T., et al. (1984),
J. Immunol. Methods, 73. 83; Syvanen, et al. (1986), Nucl. Acids Res., 14,
5037). Suitable other pairs of first and second binding moieties are: (a)
antigens and antibodies, (b) carbohydrates and lectins, (c) complementary
strands of DNA, (d) mutually reactive chemical groups and the like. This
permits separation of the linked probe product a discussed below.
The reaction products are preferably denatured prior to binding of the
first and second binding moieties to assure that the ligated probes reach
the immobilized second binding moiety, e.g. streptavidin coated beads. If
this is not a problem, the reaction product may be denatured after
immobilizing. In either event, the reaction products are preferably
denatured prior to detection so that only ligated products are detected.
After immobilization, the immobilized probe reaction product or fraction is
separated from the remainder of the fluid system as by filtration. If the
probes anneal and ligate to form the linked probe product, the labeled
probe will be covalently bound to the adjacent probe containing the first
binding moiety to form a detectable reaction product.
A number of separation techniques for the immobilized probe reaction
product may be used. For example, probes bound to beads via a
biotin-streptavidin linkage may be separated from free probe by filtration
over a membrane. Alternatively, the second binding moiety may be present
on a dip stick, introduced into the reaction well at the end of the assay
and then washed prior to detection.
Then, the label contained in the separated immobilized reaction product is
detected by conventional means. For example, if the label is radioactive,
it may be detected by known procedures and instrumentation. The absence of
a signal indicates either that (a) the detecting and adjacent probes did
not both anneal to the target substance or, if annealed, (b) ligation did
not occur.
In essence, annealing places the investigator in the right region of
sequence while ligation determines with | | |
