Patent Grant
7498131
The present invention relates to the field of biotechnology. In particular the present invention provides a method for the high throughput separation and detection of nucleotide sequences, and the use of the method in the discrimination and identification of target sequence such as single nucleotide polymorphisms. The invention further provides for probes that are capable of hybridising to the target sequence of interest, primers for the amplification of ligated probes, use of these probes and primers in the identification and/or detection of nucleotide sequences that are related to a wide variety of genetic traits and genes and kits of primers and/or probes suitable for use in the method according to the invention.
There is a rapidly growing interest in the detection of specific nucleic acid sequences. This interest has not only arisen from the recently disclosed draft nucleotide sequence of the human genome and the presence therein, as well as in the genomes of many other organisms, of an abundant amount of single nucleotide polymorphisms (SNP), but also from marker technologies such as AFLP. The recognition that the presence of single nucleotide substitutions (and other types of genetic polymorphisms such as small insertion/deletions; indels) in genes provide a wide variety of information has also attributed to this increased interest. It is now generally recognised that these single nucleotide substitutions are one of the main causes of a significant number of monogenically and multigenically inherited diseases, for instance in humans, or are otherwise involved in the development of complex phenotypes such as performance traits in plants and livestock species. Thus, single nucleotide substitutions are in many cases also related to or at least indicative of important traits in humans, plants and animal species.
Analysis of these single nucleotide substitutions and indels will result in a wealth of valuable information, which will have widespread implications on medicine and agriculture in the widest possible terms. It is for instance generally envisaged that these developments will result in patient-specific medication. To analyse these genetic polymorphisms, there is a growing need for adequate, reliable and fast methods that enable the handling of large numbers of samples and large numbers of (predominantly) SNPs in a high throughput fashion, without significantly compromising the quality of the data obtained.
Even though a wide diversity of high-throughput detection platforms for SNPs exist at present (such as fluorometers, DNA microarrays, mass-spectrometers and capillary electrophoresis instruments), the major limitation to achieve cost-effective high throughput detection is that a robust and efficient multiplex amplification technique for non-random selection of SNPs is currently lacking to utilise these platforms efficiently, which results in suboptimal use of these powerful detection platforms and/or high costs per datapoint. “Throughput” as used herein, defines a relative parameter indicating the number of samples and target sequences that can be analysed per unit of time.
Specifically, using common amplification techniques such as the PCR technique it is possible to amplify a limited number of target sequences by combining the corresponding primer pairs in a single amplification reaction but the number of target sequences that can be amplified simultaneously is small and extensive optimisation may be required to achieved similar amplification efficiencies of the individual target sequences. One of the solutions to multiplex amplification is to use a single primer pair for the amplification of all target sequences, which requires that all targets must contain the corresponding primer-binding sites. This principle is incorporated in the AFLP technique (EP-A 0 534 858). Using AFLP, the primer-binding sites result from a digestion of the target nucleic acid (i.e. total genomic DNA or cDNA) with one or more restriction enzymes, followed by adapter ligation. AFLP essentially targets a random selection of sequences contained in the target nucleic acid. It has been shown that, using AFLP, a practically unlimited number of target sequences can be amplified in a single reaction, depending on the number of target sequences that contain primer-binding region(s) that are perfectly complementary to the amplification primers. Exploiting the use of single primer-pair for amplification in combination with a non-random method for SNP target selection and efficient use of a high throughput detection platform may therefore substantially increase the efficiency of SNP genotyping, however such technology has not been provided in the art yet.
One of the principal methods used for the analysis of the nucleic acids of a known sequence is based on annealing two probes to a target sequence and, when the probes are hybridised adjacently to the target sequence, ligating the probes. The OLA-principle (Oligonucleotide Ligation Assay) has been described, amongst others, in U.S. Pat. No. 4,988,617 (Landegren et al.). This publication discloses a method for determining the nucleic acid sequence in a region of a known nucleic acid sequence having a known possible mutation. To detect the mutation, oligonucleotides are selected to anneal to immediately adjacent segments of the sequence to be determined. One of the selected oligonucleotide probes has an end region wherein one of the end region nucleotides is complementary to either the normal or to the mutated nucleotide at the corresponding position in the known nucleic acid sequence. A ligase is provided which covalently connects the two probes when they are correctly base paired and are located immediately adjacent to each other. The presence or absence of the linked probes is an indication of the presence of the known sequence and/or mutation.
Abbot et al. in WO 96/15271 developed a method for a multiplex ligation amplification procedure comprising the hybridisation and ligation of adjacent probes. These probes are provided with an additional length segment, the sequence of which, according to Abbot et al, is unimportant. The deliberate introduction of length differences intends to facilitate the discrimination on the basis of fragment length in gel-based techniques.
WO 97/45559 (Barany et al.) describes a method for the detection of nucleic acid sequence differences by using combinations of ligase detection reactions (LDR) and polymerase chain reactions (PCR). Disclosed are methods comprising annealing allele-specific probe sets to a target sequence and subsequent ligation with a thermostable ligase, optionally followed by removal of the unligated primers with an exonuclease. Amplification of the ligated products with fluorescently labelled primers results in a fluorescently labelled amplified product. Detection of the products is based on separation by size or electrophoretic mobility or on an addressable array.
Detection of the amplified probes is performed on a universally addressable array containing capturing oligonucleotides. These capturing oligonucleotides contain a region that is capable of annealing to a pre-determined region in the amplified probe, a so-called zip-region or zip code. Each amplified probe contains a different zip code and each zip code will hybridise to its corresponding capturing oligonucleotide on the array. Detection of the label in combination with the position on the array provides information on the presence of the target sequence in the sample. This method allows for the detection of a number of nucleic acid sequences in a sample. However, the design, validation and routine use of arrays for the detection of amplified probes involves many steps (ligation, amplification, optionally purification of the amplified material, array production, hybridisation, washing, scanning and data quantification), of which some (particularly hybridisation and washing) are difficult to automate. Array-based detection is therefore laborious and costly to analyse a large number of samples for a large number of SNPs.
The LDR oligonucleotide probes in a given set may generate a unique length product and thus may be distinguished from other products based on size. For the amplification a primer set is provided wherein one of the primers contains a label. Different primers can be provided with different labels to allow for the distinction of products.
The method and the various embodiments described by Barany et al. are found to have certain disadvantages. One of the major disadvantages is that the method in principle does not provide for a true high throughput process for the determination of large numbers of target sequences in short periods of time using reliable and robust methods without compromising the quality of the data produced and the efficiency of the process.
More in particular, one of the disadvantages of the means and methods as disclosed by Barany et al. resides in the limited multiplex capacity when discrimination is based inter alia, on the length of the allele specific probe sets. Discrimination between sequences that are distinguishable by only a relatively small length difference is, in general, not straightforward and carefully optimised conditions may be required in order to come to the desired resolving power. Discrimination between sequences that have a larger length differentiation is in general easier to accomplish. This may provide for an increase in the number of sequences that can be analysed in the same sample. However, providing for the necessary longer nucleotide probes is a further hurdle to be taken. In the art, synthetic nucleotide sequences are produced by conventional chemical step-by-step oligonucleotide synthesis with a yield of about 98.5% per added nucleotide. When longer probes are synthesised (longer than ca. 60 nucleotides) the yield generally drops and the reliability and purity of the synthetically produced sequence can become a problem.
These and other disadvantages of the methods disclosed in WO 97/45559 and other publications based on oligonucleotide ligation assays herein lead the present inventors to the conclusion that the methods described therein are less preferable for adaptation in a high throughput protocol that is capable of handling a large number of samples each comprising large numbers of sequences.
The specific problem of providing for longer probes has been solved by Schouten et al. (WO 01/61033). WO 01/61033 discloses the preparation of longer probes for use in ligation-amplification assays. They provided probes that are considerably longer than those that can be obtained by conventional chemical synthesis methods to avoid the problem associated with the length-based discrimination of amplified products using slab-gels or capillary electrophoresis, namely that only a small part of the detection window/resolving capacity of up to 1 kilo base length is used when OLA probes are synthesised by chemical means. With an upper limit in practice of around 100-150 bases for chemically synthesised oligonucleotides according to the current state of technology, this results in amplification products that are less than 300 base pairs long at most, but often much less (see Barany et al.). The difficulty of generating such long probes (more than about 150 nucleotides) with sufficient purity and yield by chemical means has been countered by Schouten et al., using a method in which the probes have been obtained by an in vivo enzymatic template directed polymerisation, for instance by the action of a DNA polymerase in a suitable cell, such as an M13 phage.
However, the production and purification of such biological probes requires a collection of suitable host strains containing M13 phage conferring the desired length variations and the use of multiple short chemically synthesised oligonucleotides in the process, thus their use is very laborious and time-consuming, hence costly and not suitable for high-throughput assay development. Furthermore, the use of relatively long probes and relatively large length differences between the amplifiable target sequences may result in differential amplification efficiencies in favour of the shorter target sequences. This adversely affects the overall data quality, hampering the development of a true high throughput method. Thus the need for a reliable and cost-efficient solution to multiplex amplification and subsequent length-based detection for high throughput application remains.
Other solutions that have been suggested in the art such as the use of circular (padlock) probes in combination with isothermal amplification such as rolling circle amplification (RCA) are regarded as profitable because of the improved hybridisation characteristics of circular probes and the isothermal character of RCA.
Rolling circle amplification is an amplification method wherein a first primer is hybridised to a ligated or connected circular probe. Subsequent primer elongation, using a polymerase with strand displacement activity results in the formation of a long polynucleotide strand which contains multiple representations of the connected circular probe. Such a long strand of concatamers of the connected probe is subsequently detected by the use of hybridisation probes. These probes can be labelled. Exponential amplification of the ligated probe can be achieved by the hybridisation of a second primer that hybridises to the concatameric strand and is subsequently elongated. (Exponential) Rolling Circle Amplification ((E)RCA) is described inter alia in U.S. Pat. No. 5,854,033, U.S. Pat. No. 6,143,495 WO97/19193, Lizardi et al, Nature genetics 19(3):225-232 (1998).
U.S. Pat. No. 5,876,924, WO98/04745 and WO98/04746 by Zhang et al. describe a ligation reaction using two adjacent probes wherein one of the probes is a capture probe with a binding element such as biotin. After ligation, the unligated probes are removed and the ligated captured probe is detected using paramagnetic beads with a ligand (biotin) binding moiety. Zhang also discloses the amplification of circular probes using PCR primers in a rolling circle amplification, using a DNA polymerase with strand displacement activity, thereby generating a long concatamer of the circular probe, starting from extension of the first primer. A second PCR primer subsequently hybridises to the long concatamer and elongation thereof provides a second generation of concatamers and facilitates exponential amplification. Detection is generally based on the hybridisation of labelled probes.
However, these methods have proven to be less desirable in high throughput fashion. One of the reasons is that, for a high throughput method based on length discrimination, the use of (E)RCA results in the formation of long concatamers. These concatamers are problematic, as they are not suitable for high throughput detection.
U.S. Pat. No. 6,221,603 disclosed a circular probe that contains a restriction site. The probe is amplified using (E)RCA and the resulting concatamers are restricted at the restriction site. The restriction fragments are then separated by length and detected. Separation and detection is performed on a capillary electrophoretic platform, such as the MegaBACE equipment available from Molecular Dynamics Amersham-Pharmacia For detection labelled dNTP's may be incorporated into the fragments during amplification, or the fragments may be detected by staining or by labelled detection probes. Partial digestion by the restriction enzyme may however affect the reliability of the method. Furthermore, the methods for labelling of the fragments as disclosed in U.S. Pat. No. 6,221,603, do not allow to fully utilise the MegaBACE's capacity of simultaneous detection of multiple colours.
The present inventors have set out to eliminate or at least diminish the existing problems in the art while at the same time attempting to maintain the advantageous aspects thereof, and to further improve the technology. Other problems in the art and solutions provided thereto by the present invention will become clear throughout the description, the figures and the various embodiments and examples.
The present invention relates to methods for high throughput separation and detection of multiple sequences. The present method resolves many of the problems previously encountered in the art. More in particular the present invention provides for a multiple ligation and amplification assay that allows for the rapid and high throughput analysis of a multiplicity of samples, preferably containing a multiplicity of sequences. The present invention also provides for a method for the high throughput discrimination and detection of a multitude of nucleotide sequences based on a combination of length differences and labels. The present invention combines the advantages of certain methods while at the same time avoids disadvantages associated with the various technique, thereby providing for an improved method for the detection of targets sequences in a reliable and reproducible manner and suitable for a high throughput detection method.
In a first aspect the invention relates to a method for high throughput separation and detection of a multiplicity of target sequences, optionally in a multiplicity of samples comprising subjecting each sample to a ligation-dependent amplification assay.
The method preferably is a method for determining the presence or absence of at least one target sequence (2) in a sample, wherein the method comprises the steps of:
(a) providing to a nucleic acid sample at least one circular probe (26) for each target sequence to be detected in the sample, whereby the probe has a first target specific section at its 5′-end (4) that is complementary to a first part of a target sequence (5) and a second target specific section at its 3′-end (6) that is complementary to a second part of the target sequence (7), whereby the first and second part of the target sequence are located adjacent to each other, and whereby the probe further comprises a tag section (8, 9) that is essentially non-complementary to the target sequence, whereby the tag section may comprise a stuffer sequence (10, 11) and whereby the tag section comprises at least one primer-binding sequence (12, 13);
(b) allowing the first and second target specific sections of the circular probe to anneal to the first and second parts of target sequences whereby the first and second target specific sections of the probe are annealed adjacent on the target sequence;
(c) providing means for connecting the first and second target specific sections annealed adjacently to the target sequence and allowing the first and second target specific sections to be connected, to produce a connected circular probe (28), corresponding to a target sequence in the sample;
(d) providing a primer pair comprising a first primer (16) that is complementary to a first primer-binding sequence (12), a polymerase enzyme and an optional second primer (17) that is complementary to a second primer-binding sequence (13);
(e) amplifying the resulting mixture to produce an amplified sample (19) comprising amplicons (20) that are linear representations of the connected circular probes;
(f) determining the presence or absence of a target sequence in a sample by detecting the presence or absence of the corresponding amplicon.
Probe
The circular oligonucleotide probe used in the present invention is a single linear oligonucleotide probe that is provided in step (a) for each target sequence in a sample. This single linear oligonucleotide probe combines the two target specific section into a single molecule that is circularised in step (c) when the annealed complementarity sections are connected. Thus, in the single linear probe the sections of target complementarity are each present at the extreme ends of the single linear probe. The complementarity sections at the extreme ends are intervened by the sequences that may serve as primer-binding sequences and may further be intervened by stuffer sequences of variable length. An example of such an arrangement of functional groups in the circular probe is: (target-complementarity section 1-stuffer sequence 1, primer-binding sequence 1-primer-binding sequence 2-stuffer sequence 2-target-complementarity section 2). The skilled person will appreciate that the circular probes are synthesised and applied in a linear form and that they will only be circular when the two complementary sections at the extreme ends of the probe are connected (ligated) annealing to the appropriate target sequence. Thus, the term “circular probe” as used herein actually refers to a linear molecule that is circularised by target sequence dependent connection (ligation). Only the term “connected circular probe” as used herein refers to a molecule in true circular form.
The complementary sections of the oligonucleotide probes are designed such that for each target sequence in a sample a probe is provided, preferably a specific probe, whereby the probes each contain a section at both their extreme ends that is complementary to a part of the target sequence and the corresponding complementary parts of the target sequence are located essentially adjacent to each other. Within a circular oligonucleotide probe, the oligonucleotide probe has a section at its 5′-end that is complementary to a first part of a target sequence and a section at its 3′-end that is complementary to a second part of the target sequence. Thus, when the circular probe is annealed to complementary parts of a target sequence the 5′-end of the oligonucleotide probe is essentially adjacent to the 3′-end of the oligonucleotide probe such that the respective ends of the probe may be ligated to form a phosphodiester bond and hence become a circular probe.
Circular probes are advantageous in the ligation step (c) because both target-complementarity sections are contained in the same molecule. Compared with conventional linear probes such as disclosed inter alia by WO97/45559, this means that there are equimolar amounts of the two target specific sections present and in each others vicinity. Such probes are more likely to hybridise to their respective target sequences because hybridisation of the first target-complementarity section to the target facilitates hybridisation of the second one and vice versa. In addition, the use of circular probes reduces the chances of the formation of incorrect ligation products that result from ligation between probes of different target sequences, due to the lower number of possible combinations of ligation products that can be formed when the first and second probes are part of the same circular molecule.
For more details regarding the characteristics, design and construction, use and advantages of padlock probes reference is made, inter alia, to the following documents: M. Nilsson et. al., “Padlock Probes: Circularizing Oligonucleotides for Localized DNA Detection,” Science 265: 2085-88 (1994); Pickering et al. in Nucleic Acids Research, 2002, vol. 30, e60-U.S. Pat. No. 5,854,033; U.S. Pat. No. 5,912,124; WO 02/068683, WO 01/06012, WO 0077260, WO 01/57256 the contents of which are hereby incorporated by reference.
For each target sequence for which the presence or absence in a sample is to be determined, a specific oligonucleotide probe is designed with sections complementary to the adjacent complementary parts of each target sequence. Thus, in the method of the invention, for each target sequence that is present in a sample, a corresponding (specific) amplicon may be obtained in the amplified sample. Preferably, a multiplicity of oligonucleotide probes complementary to a multiplicity of target sequences in a sample is provided. An oligonucleotide probe for a given target sequence in a sample will at least differ in nucleotide sequence from probes for other target sequences, and will preferably also differ in length from probes for other targets, more preferably a probe for a given target will produce a connected probe and/or amplicon that differs in length from connected probes corresponding to other targets in the sample as described below. Alternatively, amplicons corresponding to different targets may have an identical length if they can be otherwise distinguished e.g. by different labels as described below.
Tag & Primer Binding Sites
The oligonucleotide probe further contains a tag that is essentially non-complementary to the target sequence. The tag does not or not significantly hybridise, preferably at least not under the above annealing conditions, to any of the target sequences in a sample, preferably not to any of the sequences in a sample. The tag preferably comprises at least one, preferably two primer-binding sites and may optionally comprises one or more stuffer sequences of variable length and/or a blocking section (see below).
Stuffers
The tag of the oligonucleotide probes may comprise one or more stuffer sequence of a variable length. The length of the stuffer varies from 0 to 500, preferably from 0 to 100, more preferably from 1 to 50. The length of the tag varies from 15 to 540, preferably from 18 to 140, more preferably from 20 to 75. The stuffer may be a unique sequence as is known as a Zip-code sequence as described by Iannone et al. (2000), Cytometry 39: pp. 131-140.
Blocking Section
In an alternative embodiment, the circular probe can contain a blocking section (27). The blocking section blocks primer elongation. The blocking section is preferably located between the two primer binding sites. Preferably the blocking section is located essentially adjacent to the 3′-end of the forward primer and essentially adjacent to the 5′-end of the reverse primer binding site, see also
In an alternative embodiment, the ligated or connected circular probe comprising a blocking section can also be amplified using only one primer, preferably the forward primer. This amplification will result in the linear accumulation of amplicons with each amplification round. The circular probe in this case may contain one or more primer binding sites as long as only one primer is provided.
The blocking section can also be in the form of a recognition site for a restriction endonuclease. After ligation, the connected circular probe can be restricted with a suitable restriction endonuclease to provide linearised connected circular probes. To facilitate restriction a suitable oligonucleotide can be provided to locally generate a double stranded sequence that can be restricted using a restriction endonuclease. The connected circular probe is preferably restricted prior to the amplification step.
The advantage associated with the restriction of the circular probe prior to the amplification step is that, in the case that the polymerase has a strand displacement activity, whether residual or not, the formation of (long) concatamers is prevented. The absence or reduced presence of (long) concatamers is advantageous in length based separation such as preferably used in the detection step of the present invention as the resulting sample to be detected contains oligonucleotides with a length in a pre-determined size range. This increases the high throughput capacity as it reduces the time period between the subsequent analysis of more than one sample. A sample comprising ligated circular probes that is restricted incompletely, i.e. circularised probes remain present in the sample does not significantly, or only to a reduced extent, influence the detection step. More in particular, remaining circular probes in the sample do not or to a lesser extent influence the separation time (23) between consecutively injected samples in the way that a (long) concatamer does when a post-amplification restriction step is performed incompletely. Another advantage is that the amplification of short strands is generally more reliable than the formation of long strands. To amplify a multitude of short strands to generate amplicons is more reliable and quicker in general than the generation of one long concatamer. This attributes to an increased reliability of the method of the invention in a high through put fashion. These linearised connected circular probes can subsequently be amplified, essentially as described hereinbefore, using one or two primers. The restriction endonuclease can be any restriction endonuclease. Preferably it is a simple and cheap endonuclease such as MseI. It is preferred that the sequence of the oligonucleotide probe does not contain any further restriction sites for this endonuclease. An alternative is the incorporation of at least one RNA at the position of the blocking section and subsequent restriction with an RNAse.
Generating linear representations of the connected circular probes, the amplicons, the problem of long concatamers can be overcome, rendering the method suitable for true high throughput electrophoretic technologies. By amplification of only short strands of oligonucleotides, using the blocking section as described hereinabove or by using at least one primer in combination with a polymerase lacking in strand displacement activity, a set of amplicons representing a sample can be obtained wherein the amplicons are of a discrete length within a predetermined range, based on the design of the probes. Subsequent loading on a electrophoretic device will result in the swift separation of the amplicons.
Hybridisation
In step (a) a multiplicity of different target sequences, i.e. at least two different target sequences, is brought into contact with a multiplicity of specific oligonucleotide probes under hybridising conditions. The oligonucleotide probes are subsequently allowed to anneal to the adjacent complementary parts of the multiple target sequences in the sample. Methods and conditions for specific annealing of oligonucleotide probes to complementary target sequences are well known in the art (see e.g. in Sambrook and Russel (2001) “Molecular Cloning: A Laboratory Manual (3rd edition), Cold Spring Harbor Laboratory, Cold Spring Harbor Laboratory Press). Usually, after mixing of the oligonucleotide probes and target sequences the nucleic acids are denatured by incubation (generally at between 94° C. and 96° C.) for a short period of time (e.g. 30 seconds to 5 minutes) in a low salt buffer (e.g. a buffer containing no salts or less salts than the ionic strength equivalent of 10 mM NaCl). The sample containing the denatured probes and target sequences is anthen allowed to cool to an optimal hybridisation temperature for specific annealing of the probes and target sequences, which usually is about 5° C. below the melting temperature of the hybrid between the complementary section of the probe and its complementary sequence (in the target sequence). In order to prevent aspecific or inefficient hybridisation of one of the two probe sections, or in a sample with multiple target sequences, it is preferred that, within one sample, the sections of the probes that are complementary to the target sequences are of a similar, preferably identical melting temperatures between the different target sequences present in the sample. Thus, the complementary sections of the probes preferably differ less than 20, 15, 10, 5, or 2° C. in melting temperature. This is facilitated by using complementary sections of the probes with a similar length and similar G/C content. Thus, the complementary sections preferably differ less than 20, 15, 10, 5, or 2 nucleotides in length and their G/C contents differ by less than 30, 20, 15, 10, or 5%. Complementary as used herein means that a first nucleotide sequence is capable of specifically hybridising to second nucleotide sequence under normal stringency conditions. A nucleotide sequence that is considered complementary to another nucleotide sequence may contain a minor amount, i.e. preferably less than 20, 15, 10, 5 or 2%, of mismatches. Alternatively, it may be necessary to compensate for mismatches e.g. by incorporation of so-called universal nucleotides, such as for instance described in EP-A 974 672, incorporated herein by reference or by the use of suitable locked nucleic acids (LNAs) and peptide nucleic acids (PNAs). Since annealing of probes to target sequences is concentration dependent, annealing is preferably performed in a small volume, i.e. less than 10 μl. Under these hybridisation conditions, annealing of probes to target sequences usually is fast and does not to proceed for more than 5, 10 or 15 minutes, although longer annealing time may be used as long as the hybridisation temperature is maintained to avoid aspecific annealing.
In a preferred embodiment of the invention, excellent results have been obtained by prolonged hybridisation times such as overnight hybridisation or for more than one hour. Prolonged hybridisation times can be advantageous in these assays as the difference in signal due to different hybridisation efficiencies is reduced and it is considered desirable to achieve complete hybridisation and ligation of all probes for which a target sequence is present. Excellent results have been obtained by a combined hybridisation-ligation step using a thermostable ligase described herein. In this embodiment the hybridisation-ligation was performed by allowing the probes to hybridise during 1 hour in the presence of a thermostable ligase, followed by a denaturation step. Repeating these steps for at least 2 times provided good results. Repeating these steps 10 times provided excellent results.
To avoid evaporation during denaturation and annealing, the walls and lids of the reaction chambers (i.e. tubes or microtitre wells) may also be heated to the same temperature as the reaction mixture. In preferred oligonucleotide probes the length of the complementary section is preferably at least 15, 18 or 20 nucleotides and preferably not more than 30, 40, or 50 nucleotides and the probes preferably have a melting temperature of at least 50° C., 55° C. or 60° C.
Non-Hybridised Probes
The probes that are not complementary to a part of the target sequence or that contain too many mismatches will not or only to a reduced extent hybridise to the target sequence when the sample is submitted to hybridisation conditions. Accordingly ligation is less likely to occur. The number of spurious ligation products from these probes in general will therefore not be sufficient and much smaller than the bonafide ligation products such that they are outcompeted during subsequent multiplex amplification. Consequently, they will not be detected or only to a minor extent.
Ligation
The respective 5′- and 3′-ends of the oligonucleotide probe that are annealed essentially adjacent to the complementary parts of a target sequence are connected in step (c) to form a covalent bond by any suitable means known in the art. The ends of the probes may be enzymatically connected in a phosphodiester bond by a ligase, preferably a DNA ligase. DNA ligases are enzymes capable of catalysing the formation of a phosphodiester bond between (the ends of) two polynucleotide strands bound at adjacent sites on a complementary strand. DNA ligases usually require ATP (EC 6.5.1.1) or NAD (EC 6.5.1.2) as a cofactor to seal nicks in double stranded DNA. Suitable DNA ligase for use in the present invention are T4 DNA ligase, E. coli DNA ligase or preferably a thermostable ligase like e.g. Thermus aquaticus (Taq) ligase, Thermus thermophilus DNA ligase, or Pyrococcus DNA ligase. Alternatively, chemical autoligation of modified polynucleotide ends may be used to ligate two oligonucleotide probes annealed at adjacent sites on the complementary parts of a target sequence (Xu and Kool, 1999, Nucleic Acid Res. 27: 875-881).
Both chemical and enzymatic ligation occur much more efficient on perfectly matched probe-target sequence complexes compared to complexes in which one or both of the ends of the probe form a mismatch with the target sequence at, or close to the ligation site (Wu and Wallace, 1989, Gene 76: 245-254; Xu and Kool, supra). In order to increase the ligation specificity, i.e. the relative ligation efficiencies of perfectly matched oligonucleotides compared to mismatched oligonucleotides, the ligation is preferably performed at elevated temperatures. Thus, in a preferred embodiment of the invention, a DNA ligase is employed that remains active at 50-65° C. for prolonged times, but which is easily inactivated at higher temperatures, e.g. used in the denaturation step during a PCR, usually 90-100° C. One such DNA ligase is a NAD requiring DNA ligase from a Gram-positive bacterium (strain MRCH 065) as known from WO 01/61033. This ligase is referred to as “Ligase 65” and is commercially available from MRC Holland, Amsterdam.
Gap Ligation
In an alternative embodiment, for instance directed to the identification of indels, the respective ends may be annealed such that a gap is left. This gap can be filled with a suitable oligonucleotide and ligated. Such methods are known in the art as ‘gap ligation’ and are disclosed inter alia in WO 00/77260. Another possibility to fill this gap is by extension of one end of the probe using a polymerase and a ligase in combination with single nucleotides, optionally preselected from A, T, C, or G, or di-, tri- or other small oligonucleotides.
Primers
The connected probes are amplified using a pair of primers corresponding to the primer-binding sites. In a preferred embodiment at least one of the primers or the same set of primers is used for the amplification of two or more different connected probes in a sample, preferably for the amplification of all connected probes in a sample. Such a primer is sometimes referred to as a universal primer as these primers are capable of priming the amplification of all probes containing the corresponding universal primer binding site and consequently of all ligated probes containing the universal primer binding site. The different primers that are used in the amplification in step (d) are preferably essentially equal in annealing and priming efficiency. Thus, the primers in a sample preferably differ less than 20, 15, 10, 5, or 2° C. in melting temperature. This can be achieved as outlined above for the complementary section of the oligonucleotide probes. Unlike the sequence of the complementary sections, the sequence of the primers is not dictated by the target sequence. Primer sequences may therefore conveniently be designed by assembling the sequence from tetramers of nucleotides wherein each tetramer contains one A, T, C and G or by other ways that ensure that the G/C content and melting temperature of the primers are identical or very similar. The length of the primers (and corresponding primer-binding sites in the tags of the probes) is preferably at least 12, 15 or 17 nucleotides and preferably not more than 25, 30, 40 nucleotides.
In a preferred embodiment, at least two of the oligonucleotide probes that are complementary to at least two different target sequences in a sample comprise a tag sequence that comprises a primer-binding site that is complementary to a single primer sequence. Thus, preferably at least one of the first and second primer in a primer pair is used for the amplification of connected probes corresponding to at least two different target sequences in a sample, more preferably for the amplification of connected probes corresponding to all target sequences in a sample. Preferably only a single first primer is used and in some embodiments only a single first and a single second primer is used for amplification of all connected probes. Using common primers for amplification of multiple different fragments usually is advantageous for the efficiency of the amplification step.
The connected probes obtained from the ligation of the adjacently annealed probe sections are amplified in step (d), using a primer set, preferably consisting of a pair of primers for each of the connected probes in the sample. The primer pair comprises primers that are complementary to primer-binding sequences that are present in the connected probes. A primer pair usually comprises a first and at least a second primer, but may consist of only a single primer that primes in both directions. Excellent results have been obtained using primers that are known in the art as AFLP—primers such as described inter alia in EP534858 and in Vos et al., Nucleic Acid Research, 1995, vol. 23, 4407-44014.
Selective Primers
In a particular preferred embodiment, one or more of the primers used in the amplification step of the present invention is a selective primer. A selective primer is defined herein as a primer that, in addition to its universal sequence which is complementary to a primer binding site in the probe, contains a region that comprises so-called “selective nucleotides”. The region containing the selective nucleotides is located at the 3′-end of the universal primer.
The principle of selective nucleotides is disclosed inter alia in EP534858 and in Vos et al., Nucleic Acid Research, 1995, vol. 23, 4407-44014. The selective nucleotides are complementary to the nucleotides in the (ligated) probes that are located adjacent to the primer sequence. The selective nucleotides generally do not form part of the region in the (ligated) probes that is depicted as the primer sequence. Primers containing selective nucleotide are denoted as +N primers, in which N stands for the number of selective nucleotides present at the 3′-end of the primer. N is preferably selected from amongst A, C, T or G.
N may also be selected from amongst various nucleotide alternatives, i.e. compounds that are capable of mimicking the behavior of ACTG-nucleotides but in addition thereto have other characteristics such as the capability of improved hybridisation compared to the ACTG-nucleotides or the capability to modify the stability of the duplex resulting from the hybridisation. Examples thereof are PNA's, LNA's, inosine etc. When the amplification is performed with more than one primer, such as with PCR using two primers, one or both primers can be equipped with selective nucleotides. The number of selective nucleotides may vary, depending on the species or on other particulars determinable by the skilled man. In general the number of selective nucleotides is not more than 10, but at least 5, preferably 4, more preferably 3, most preferred 2 and especially preferred is 1 selective nucleotide.
A +1 primer thus contains one selective nucleotide, a +2 primer contains 2 selective nucleotides etc. A primer with no selective nucleotides (i.e. a conventional primer) can be depicted as a +0 primer (no selective nucleotides added). When a specific selective nucleotide is added, this is depicted by the notion +A or +C etc.
By amplifying a set of (ligated) probes with a selective primer, a subset of (ligated) probes is obtained, provided that the complementary base is incorporated at the appropriate position in the desired of the probes that are supposed to be selectively amplified using the selective primer. Using a +1 primer, for example, the multiplex factor of the amplified mixture is reduced by a factor 4 compared to the mixture of ligated probes prior to amplification. Higher reductions can be achieved by using primers with multiple selective nucleotides, i.e. 16 fold reduction of the original multiplex ration is obtained with 2 selective nucleotides etc.
When an assay is developed which, after ligation, is to be selectively amplified, it is preferred that the probe contains the complementary nucleotide adjacent to the primer binding sequence. This allows for pre-selection of the ligated probe to be selectively amplified.
The use of selective primers in the present invention has proven to be advantageously when developing ligation based assays with high multiplex ratios of which subsequently only a specific part needs to be analyzed resulting in further cost reduction of the ligation reaction per datapoint. By designing primers together with adjacent selective nucleotides, the specific parts of the sample that are to be amplified separately can be selected beforehand.
One of the examples in which this is useful and advantageous is in case of analysis of samples that contain only minute amounts of DNA and/or for the identification of different (strains of) pathogens. For example, in an assay directed to the detection of various strains of anthrax (Bacillus anthracis), for each of the strains a set of representative probes is designed. The detection of the presence or absence of this set (or a characterizing portion thereof) of ligated probes after the hybridisation and ligation steps of the method of the invention may serve as an identification of the strain concerned. The selective amplification with specifically designed primers (each selective primer is linked to a specific strain) can selectively amplify the various strains, allowing their identification. For instance, amplification with an +A primer selectively amplifies the ligated probes directed to strain X where a +G primer selectively amplifies the ligated probes directed to strain Y. If desired, for instance in the case of small amounts of sample DNA, an optional first amplification with a +0 primer will increase the amount of ligated probes, thereby facilitating the selective amplification.
For example, a universal primer of 20 nucleotides becomes a selective primer by the addition of one selective nucleotide at its 3′ end, the total length of the primer now is 21 nucleotides. See also
One of the further advantages of the selective amplification of ligated probes is that an assay with a high multiplex ratio can be adapted easily for detection with methods or on platforms that prefer a lower multiplex ratio.
One of many examples thereof is the detection based on length differences such as electrophoresis and preferably capillary electrophoresis such as is performed on a MegaBACE or using nano-technology such as Lab-on-a-Chip.
Amplification
In step (d) of the method of the invention, the connected probes are amplified to produce (detectable) amplified connected probes (amplicons) that are linear representations of the connected circular probes by any suitable nucleic acid amplification method known in the art. Nucleic acid amplification methods usually employ two primers, dNTP's, and a (DNA) polymerase. A preferred method for amplification is PCR. “PCR” or “Polymerase Chain Reaction” is a rapid procedure for in vitro enzymatic amplification of a specific DNA segment. The DNA to be amplified is denatured by heating the sample. In the presence of DNA polymerase and excess deoxynucleotide triphosphates, oligonucleotides that hybridise specifically to the target sequence prime new DNA synthesis. It is preferred that the polymerase is a DNA polymerase that does not express strand displacement activity or at least not significantly. Examples thereof are Amplitaq and Amplitaq Gold (supplier Perkin Elmer) and Accuprime (Invitrogen). One round of synthesis results in new strands of determinate length, which, like the parental strands, can hybridise to the primers upon denaturation and annealing. The second cycle of denaturation, annealing and synthesis produces two single-stranded products that together compose a discrete double-stranded product, exactly the length between the primer ends. This discrete product accumulates exponentially with each successive round of amplification. Over the course of about 20 to 30 cycles, many million-fold amplification of the discrete fragment can be achieved. PCR protocols are well known in the art, and are described in standard laboratory textbooks, e.g. Ausubel et al., Current Protocols in Molecular Biology, John Wiley & Sons, Inc. (1995). Suitable conditions for the application of PCR in the method of the invention are described in EP-A 0 534 858 and Vos et al. (1995; Nucleic Acids Res.23: 4407-23:4407-4407-4407-4414), where multiple DNA fragments between 70 and 700 nucleotides and containing identical primer-binding sequences are amplified with near equal efficiency using one primer pair. Other multiplex and/or isothermal amplification methods that may be applied include e.g. LCR, self-sustained sequence replication (3SR), Q-β-replicase mediated RNA amplification, or strand displacement amplification (SDA). In some instances this may require replacing the primer-binding sites in the tags of the probes by a suitable (RNA) polymerase-binding site as long as they lead to linear amplification products as defined herein before, i.e. of discrete lengths and corresponding to the length of the circular probes.
As described herein, linear representations of the connected circular probes can be obtained by exponential amplification of the circular probe with two primers, one forward and one reverse, using a polymerase that does not or not significantly have a strand displacement activity. The first primer elongation in the amplification with the forward primer generates an oligonucleotide product until the 5′end of the forward primer is reached. There the primer elongation is terminated, due to the substantial absence of strand displacement activity of the polymerase used, leaving a elongated primer with substantially the same length as the connected circular probe. The second cycle of denaturation, primer hybridisation and primer elongation will, for the forward primer, produce the identical strand as during the first primer elongation, while the reverse primer will hybridise to the oligonucleotide product from the elongation of the first primer elongation and thereby produce the complementary strand, resulting in the exponential amplification of the circular probe to thereby produce amplicons of discrete length which are representations of the connected circular oligonucleotide probes.
Amplicons
The term ‘amplicon’ as used herein refers to the product of the amplification step of the connected or ligated probe. The term ‘amplicon’ as used herein thus refers to an amplified connected probe. After the ligation step wherein the two target specific section are connected by mean of a ligase, the connected or ligated probe is combined with one or more primers and a polymerase and amplified. The ligated probe, the primers, the polymerase and/or other parameters and variables are such that the amplification results in linear representations of the circular probe. In the present invention the amplicon is a linear oligonucleotide having a length that does not substantially exceed the length of the circular probe. The minimum length of the amplicon is at least the sum of the length of the two target complementary sections. It is preferred that the length of the amplicon corresponds to the length of the circular probe. It is more preferred that the length of the amplicon is indicative of the ligation of the corresponding circular probe. Preferably an amplicon does not contain repetitions of sections of the circular probe, i.e. is not a concatamer or a multimer of the circular probe or a multimeric representation thereof. Preferably an amplicon is a linear and monomeric representation of the connected circular probe.
The advantage obtained by the conversion from circular probes to linear amplicons is that the advantageous characteristics of the circular probe are used (improved kinetics, increased hybridisation to the target strand due to the formation of the ‘padlock’ conformation), while the resulting amplicons are of a discrete length an can be detected subsequently without the need for additional steps such as restriction and labelling.
Detection
Detection of the labelled separated samples is performed by a detector to result in detection data. The detector is of course dependent on the general system on which the separation is carried out (capillary electrophoresis, slab-gel electrophoresis, fixed detector-continuous gel-electrophoresis) but is also depending on the label that is present on the primer, such as a fluorescent or a radioactive label.
The amplicons in a sample are preferably analysed on an electrophoretic device. The electrophoretic device preferably separates the different amplicons in an amplified sample on the basis of length, after which the separated amplicons may be detected as described below. A suitable electrophoretic device may be a gel-electrophoresis device, e.g. for conventional (polyacrylamide) slab gel-electrophoresis, or a capillary electrophoresis device such as exemplified by the MegaBACE equipment available from Molecular Dynamics Amersham-Biosciences. An alternative is the nano-sized capillary electrophoretic devices known as Lab-on-a-Chip. The electrophoretic device preferably is a multichannel device in which multiple samples are electrophoresed in multiple channels in parallel. The electrophoretic device has an application location (per channel) for application (loading) of the amplified sample to be electrophoresed, a separation area over which the fragments in the sample migrate by electrophoresis, and preferably also a detection device located at a detection location distal from the application location. The detection device will usually comprises a photomultiplier for the detection of fluorescence, phosphorescence or chemiluminescence. Alternatively, in the case of gel-electrophoresis, the separated fragments may be detected in the gel e.g. by autoradiography or fluorography.
Length Discrimination
To discriminate between different target sequences in the sample preferably a difference in length of the respective corresponding amplicons is used. By separating the amplicons based on length, the presence of the corresponding target nucleotides sequences in the sample can be determined. Accordingly, in a preferred embodiment of the present invention, the discrimination between amplicons derived from different target sequences in a sample is based on a length difference between the respective amplicons corresponding to different target sequences in a sample or amplified sample.
Preferably, the length difference is provided by the length of the stuffer sequence(s) in the oligonucleotide probes. By including in each oligonucleotide probe a stuffer of a pre-determined length, the length of each amplicon in an amplified sample can be controlled such that an adequate discrimination based on length differences of the amplicon obtained in step (d) is enabled. In a preferred embodiment of a probe according to the invention, the stuffer is located between the probe's section complementary to the target sequence and a primer-binding sequence. As there are two target specific sections at both ends of the probe and two primer binding sites, two stuffer can be incorporated in the probe therein between. As such, the total length of the stuffer is provided by the combination of the length of the first stuffer and second stuffer in the probe. Accordingly, in a preferred embodiment, the oligonucleotide probe comprises two stuffers, preferably in the non target complementary tags. A graphic representation thereof can be found in
The length differentiation between amplicons obtained from target sequences in the sample is preferably chosen such that the amplicons can be distinguished based on their length. This is accomplished by using stuffer sequences or combinations of stuffer sequences which (together) result in clear length differences that may be distinguished on electrophoretic devices. Thus, from the perspective of resolving power, the length differences between the different amplicons, as may be caused by their stuffers, are as large as possible. However, for several other important considerations, as noted before, the length differences between the different amplicon is preferably as small as possible: (1) the upper limit that exists in practice with respect to the length of chemically synthesised probes of about 100-150 bases at most; (2) the less efficient amplification of larger fragments, (3) the increased chances for differential amplification efficiencies of fragments with a large length variation; and (4) the use of multiple injections of detection samples on the detection device which works best with fragments in a narrow length range. Preferably the length differences between the sequences to be determined and provided by the stuffers is at least sufficient to allow discrimination between essentially all amplicons. By definition, based on chemical, enzymatic and biological nucleic acid synthesis procedures, the minimal useable size difference between different amplicon in an amplified sample is one base, and this size difference fits within the resolving power of most electrophoresis devices, especially in the lower size ranges. Thus based on the above it is preferred to use multiplex assays with amplification products with differ in length by a single base(pair). In a preferred embodiment, the length difference between different amplicons in an amplified sample is at least two nucleotides. In a particularly preferred embodiment of the invention the amplicon corresponding to different target sequences in a sample have a length difference of two nucleotides.
Labels
In a preferred embodiment, at least one of the primers complementary to the primer-binding sites of the first and second oligonucleotide probes in the sample comprises a label, preferably the second primer comprises a label. The label can be selected from a large group, amongst others comprising fluorescent and/or phosphorescent moieties such as dyes, chromophores, or enzymes, antigens, heavy metals, magnetic probes, phosphorescent moieties, radioactive labels, chemiluminescent moieties or electrochemical detecting moieties. Preferably the label is a fluorescent or phosphorescent dye, more preferably selected from the group of FAM, HEX, TET, JOE, NED, and (ET-)ROX. Dyes such as FITC, Cy2, Texas Red, TAMRA, Alexa fluor 488™, Bodipy™ FL, Rhodamine 123, R6G, Bodipy 530, Alexafluor™532 and IRDyes™ by Licor as used on the NEN Glober IR2 platform are also suitable for use in the present invention. Preferably the label may be chosen from amongst the fluorescent or phosphorescent dyes in the group consisting of FAM, TET, JOE, NED, HEX, (ET-)ROX, FITC, Cy2, Texas Red, TAMRA, Alexa fluor 488™, Bodipy™ FL, Rhodamine 123, R6G, Bodipy 530, Alexafluor™532 and IRDyes™.
By using a primer set comprising differently labelled primers, the number of connected probes that can be discriminated in a sample and hence the number of target sequences in a sample can be doubled for each additional label. Thus, for each additional label that is used in a sample, the number of target sequences that can be analysed in a sample is doubled. The maximum number of labels that can be used in one sample in a high throughput method is governed mostly by the limitations in the detection capabilities of the available detection platforms. At present, one of the most frequently used platforms (MegaBACE, by Molecular Dynamics—Amersham-Biosciences Ltd. allows the simultaneous detection of up to four fluorescent dyes, being FAM, JOE or HEX, NED and (ET-)ROX. However, alternative capillary electrophoresis instruments are also suitable, which includes ABI310, ABI3100, ABI3700 (Perkin-Elmer Corp.), CEQ2000 XL (Beckman Coulter) and others. Non-limiting examples of slab-gel based electrophoresis devices include ABI377 (Perkin Elmer Corp.) and the global IR2 automated DNA sequencing system, available from LI-COR, Lincoln, Nebr., USA.
Length and Label
Throughput can be increased by the use of multiple labelled primers. One of the problems associated with the use of different labels in one sample is cross talk or residual cross talk. Cross talk or residual cross talk, as used herein, refers to the overlap between the emission spectra of different (fluorescent) labels. For instance when fluorescent dyes are used, each dye has a different emission (and absorption) spectrum. In case of two dyes in one sample, these spectra can overlap and may cause a disturbance of the signal, which contravenes the quality of the data obtained. Particularly when two nucleotide fragments to be detected in a sample are labelled with a different label and one of the fragments is present in an abundant amount whereas the other is present only in minute amounts, residual cross talk can cause that the measured signal of the fragment that is present in only minute amounts is mostly derived from the emission of another label with an overlapping emission spectrum that is abundantly contained in a fragment with identical size of another sample. The reciprocal effect of the other dye may also occur but in this example its effect is probably less because of the abundance differences between the amplicons labelled with the respective dyes.
Chehab et al. (Proc. Natl. Acad. Sci. USA, 86:9178-9182 (1989) have attempted to discriminate between alleles by attaching different fluorescent dyes to competing alleles in a single reaction tube by selecting combinations of labels such that the emission maximum of one dye essentially coincides with the emission minimum of the other dye. However, at a certain wavelength at which one dye expresses an absorption maximum, there is always also some remaining absorption from another dye present in the sample, especially when the sample contains multiple dyes.
This route to multiplex analysis was found to be limited in scale by the relatively few dyes that can be spectrally resolved. One of the major problems with the use of multiple dyes is that the emission spectra of different fluorescent labels often overlap. The resulting raw data signals have to be corrected for the contribution of similar size fragments that are detected simultaneously and are labelled with another fluorescent dye by a process called cross-talk correction. Cross-talk correction is commonly carried out by mathematical means, based on the known theoretical absorption spectra for both dyes, after “raw” data collection from the detection device. Mathematical correction is based on theoretical spectra and ignores that emission spectra of labels are sensitive and often affected by the composition of the detection sample. These sensitivities can affect the brightness and/or the wavelength of the emission. This means that parameters such as pH, temperature, excitation light intensity, non-covalent interactions, salt concentration and ionic strength strongly influence the resulting emission spectrum. In particular, it is known that the presence of residual salts in a sample affects the fluorescence signal emitted by the dye and is a critical factor in case of detection by capillary electrophoresis using electrokinetic injection because it then also affects the injection efficiency. Thus, spectral overlap is a potential source of error that negatively impacts on data quality in case of multiplex detection using different fluorescent dyes.
The present invention provides for a solution to this problem such that two (or more) labels with overlapping spectra can be used in the same sample without significantly affecting data quality. By a predetermined combination of length differences and labels, an increase in the number of target nucleotide sequences that can be detected in sample is obtained while the quality of the data remains at least constant. In a preferred embodiment of the invention, spectral overlap between two differently labelled sequences is reduced by the introduction of a length difference between the two sequences. This label-related length difference can be provided for by the length of the stuffer sequence as described herein. The number of different labels that can be used in the same sample in the present method is at least two, preferably at least three, more preferably at least four. The maximum number of labels is functionally limited by the minimum of spectral overlap that remains acceptable, which for most applications typically amounts to less than 15 percent of the true signal, preferably less than 10 percent, more preferably lees than 5 percent and most preferably less than 1 percent of the true signal.
In order to avoid the potential influence of residual cross-talk on the data quality in case different samples are labelled with multiple fluorescent dyes with overlapping emission spectra and fragments with identical length are detected simultaneously in the same run, in a particular preferred embodiment it is preferred to choose the stuffer sequences such that amplicons differ by at least two base pairs within a multiplex set and differ by a single base pair between multiplex sets labelled with the different dyes that have overlapping spectra. By doing so, the length of the fragments labelled with the respective dyes can be chosen such that the potential influence of residual cross-talk on the quality of the data is circumvented because unique combinations of fragments size and labelling dye are defined (
A particular preferred embodiment of the invention is directed to a method in which a sample comprising amplicons is derived from a multiplicity of target sequences. These amplicons are differently labelled, thereby defining groups of amplicons carrying the same label. Within each group, the stuffer provided for a length difference of at least two, preferably two nucleotides. Between two groups with labels having spectral overlap, the stuffer provides a length difference of one nucleotide, effectively resulting in one group having an even number of nucleotides and one group having an odd number of nucleotides as described above.
In one aspect the present invention pertains to a method for the improved discrimination and detection of target sequences in a sample, comprising providing at least a two or more groups of oligonucleotide probes, wherein the amplicons obtained with different groups of oligonucleotide probes have different labels, wherein substantially each amplified connected probe target sequence within a group has the same label, wherein within a group of identically labelled amplicons a length difference is provided between each identically labelled probe within that group, wherein between the first and second group an additional length difference is provided such that each amplified connected probe in the amplified sample is characterised by a combination of length of the sequence and the label.
In a preferred embodiment of the method of the invention, at least two groups of oligonucleotide probes are provided to a sample, whereby each group of oligonucleotide probes has tag sequences with at least one group specific primer-binding site. The connected probes of each group are amplified from a primer pair wherein at least one of the first and second primers is complementary to the group specific primer-binding site, and whereby at least one of the first and second primers of a group comprises a group specific label. In each group, an amplicon corresponding to a target sequence in the sample differs in length from an amplicon corresponding to a different target sequence in the sample. The group specific labels are preferably such that the detection device can distinguish between the different group specific labels. The length difference is preferably provided by the length of the stuffer sequence. Preferably in this embodiment of the method of the invention, a first part of the groups has amplicons having an even number of nucleotides and a second part of the groups has amplicons having an odd number of nucleotides. Preferably, the groups of amplicons having an even number of nucleotides and the groups amplicons having an odd number of nucleotides are labelled with (fluorescent) labels, which have the least overlap in their emission spectra. Thus, two groups of amplicons, each group having an odd number of nucleotides are labelled with labels, which have the least overlap in their emission spectra. The same holds for two groups of amplicons, each group having an even number of nucleotides. Two groups of amplicons, one group having an odd number of nucleotides and the other group having an even number of nucleotides are labelled with labels that have a larger overlap in their emission spectra. The relative notions as used herein of ‘the least overlap in their emission spectra’ and ‘have a larger overlap in their emission spectra’ refer to a group of labels from which a selection of the labels can be made for use in the present invention. This group of labels may depend on the detection platform used to other factors such as those disclosed herein before. In a particularly preferred embodiment of this method, a first and second groups of amplicons having an even number of nucleotides are produced and a third and fourth group of connected amplified probes having an odd number of nucleotides are produced and whereby the first and second group are labelled with FAM and NED, respectively, and the third and fourth group are labelled with (ET-)ROX and either JOE or HEX, respectively; or vice versa, whereby the first and second group are labelled with (ET-)ROX and either JOE or HEX, respectively, and the third and fourth group are labelled with FAM and NED, respectively. Thus, in these embodiments, the fluorescent labels are chosen such that the groups of amplicons that co-migrate, because they both contain fragments with either even or odd numbers of nucleotides, have labels which have the least overlap in their emission spectra, thereby avoiding as much as possible cross-talk in the detection of amplicons in different groups (see also below).
In a preferred embodiment to avoid cross-talk it is therefore desirable to combine a difference in length with a different label when analysing a set of amplicons in such a way that the influence of spectral overlap on the data quality is avoided by length differences between the amplicons labelled with the dyes that have overlapping emission spectra.
It is preferred that in each sample the connected probes derived from each target sequence differ from any other connected probe in the sample in length, and/or in the label or, preferably in the combination of the length and the label. To provide for an adequate separation of the amplicons of different length it is preferred that the length difference between two different connected probes is at least two nucleotides, preferably two. When detecting polymorphisms it is preferred that the difference in length between two or more (SNP) alleles of the polymorphism is not more than two, thereby ensuring that the efficiency of the amplification is similar between different alleles or forms of the same polymorphism. This implies that preferably both alleles are amplified with the same pair of primers and hence will be labelled with the same dye.
In a preferred embodiment, for example directed to the detection of different alleles of a multiplicity of loci, the distribution between odd/even lengths within a group can be designed in the following way. Two loci L1, L2 are each represented by two alleles A11, A12 for L1 and A21, A22 for L2. The lengths of the various alleles (or ligated and amplified probes representing those alleles) is such that A11>A12>A21>A22; A12-A11=2; A22-A21=2; A12-A21=3. Between groups G1 and G2 carrying labels that may have an overlap in their spectra there can be a length difference of 1 nucleotide. Thus G1(A11)-G2(A11)=1, hence the group starts with either an even or an uneven length.
This distribution has some significant advantages compared to the more densely packed distribution disclosed herein. It is known that due to conformational differences that different sequences of identical length generally differ in their electrophoretic mobility. When there is only a difference in length of one nucleotide, this may cause overlap between the peaks if the sequences are of a very different mobility. For instance the difference in mobility between two alleles of one locus (A11, A12), will be less than the difference in mobility between two alleles from different loci (A12, A21). When there is a significant difference in mobility between A12 and A21, this may lead to unreliable detection. By creating length distributions as herein disclosed this can be avoided. The lower throughput is then weighed against the reliability of the detection.
The problem of the overlap between the spectra of the different labels is then adequately avoided. This is schematically depicted in Table A.
In an embodiment of the present invention there is provided between the amplicons within one group, a length difference of alternating two and three nucleotides, i.e. 0, 2, 5, 7, 10, 12 etc. The other group then has a length difference of 1, 3, 6, 8, 11, 13 etc.
Target Sequences
In its widest definition, the target sequence may be any nucleotide sequence of interest. The target sequence preferably is a nucleotide sequence that contains, represents or is associated with a polymorphism. The term polymorphism herein refers to the occurrence of two or more genetically determined alternative sequences or alleles in a population. A polymorphic marker or site is the locus at which divergence occurs. Preferred markers have at least two alleles, each occurring at frequency of greater than 1%, and more preferably greater than 10% or 20% of a selected population. A polymorphic locus may be as small as one base pair. Polymorphic markers include restriction fragment length polymorphisms, variable number of tandem repeats (VNTR's), hypervariable regions, minisatellites, dinucleotide repeats, trinucleotide repeats, tetranucleotide repeats, simple sequence repeats, and insertion elements such as Alu. The first identified allelic form is arbitrarily designated as the reference form and other allelic forms are designated as alternative or variant alleles. The allelic form occurring most frequently in a selected population is sometimes referred to as the wild type form. Diploid organisms may be homozygous or heterozygous for allelic forms. A diallelic polymorphism has two forms. A triallelic polymorphism has three forms. A single nucleotide polymorphism occurs at a polymorphic site occupied by a single nucleotide, which is the site of variation between allelic sequences. The site is usually preceded by and followed by highly conserved sequences of the allele (e.g., sequences that vary in less than 1/100 or 1/1000 members of the populations). A single nucleotide polymorphism usually arises due to substitution of one nucleotide for another at the polymorphic site. Single nucleotide polymorphisms can also arise from a deletion of a nucleotide or an insertion of a nucleotide relative to a reference allele. Other polymorphisms include small deletions or insertions of several nucleotides, referred to as indels. A preferred target sequence is a target sequence that is associated with an AFLP® marker, i.e. a polymorphism that is detectable with AFLP®.
DNA
In the nucleic acid sample, the nucleic acids comprising the target may be any nucleic acid of interest. Even though the nucleic acids in the sample will usually be in the form of DNA, the nucleotide sequence information contained in the sample may be from any source of nucleic acids, including e.g. RNA, polyA+ RNA, cDNA, genomic DNA, organellar DNA such as mitochondrial or chloroplast DNA, synthetic nucleic acids, DNA libraries, clone banks or any selection or combinations thereof. The DNA in the nucleic acid sample may be double stranded, single stranded and double stranded DNA denatured into single stranded DNA. Denaturation of double stranded sequences yields two single stranded fragments one or both of which can be analysed by probes specific for the respective strands. Preferred nucleic acid samples comprise target sequences on cDNA, genomic DNA, restriction fragments, adapter-ligated restriction fragments, amplified adapter-ligated restriction fragments. AFLP fragments or fragments obtained in an AFLP-template preamplification.
Samples
It is preferred that a sample contains two or more different target sequences, i.e. two or more refers to the identity rather than the quantity of the target sequences in the sample. In particular, the sample comprises at least two different target sequences, in particular at least 10, preferably at least 25, more preferably at least 50, more in particular at least 100, preferably at least 250, more preferably at least 500 and most preferably at least 1000 additional target sequences. In practice, the number of target sequences is limited, among others, by the number of connected circular probes. E.g., too many different oligonucleotide probes in a sample may corrupt the reliability of the multiplex amplification step.
A further limitation is formed e.g. by the number of fragments in a sample that can be resolved by the electrophoretic device in one injection. The number can also be limited by the genome size of the organism or the transcriptome complexity of a particular cell type from which the DNA or cDNA sample, respectively, is derived.
Multiple Injection
In a preferred embodiment of the invention, in order to come to a high throughput method of a multiplicity of samples, a number of samples are treated similar to thereby generate a multiplicity of amplified detection samples which can then be analysed on a multichannel device which is at least capable of detecting the labels and/or length differences. Suitable devices are described herein.
To increase throughput on electrophoretic platforms methods have been developed that are described in this application and are commonly depicted as multiple injection. By injecting multiple samples containing fragments of discrete, pre-determined lengths, in the same electrophoretic matrix and/or in short consecutive runs, throughput can be increased. All detectable fragments preferably have a length within a specific span and only a limited number of fragments can be detected in one sample, hence the advantage of selective amplification for the reduction of the multiplex ratio by the selection of a subset of the connected probes in the amplification step resulting in a subset of amplicons.
Steps (a) to (e) of the method of the invention may be performed on two or more nucleic acid samples, each containing two or more different target nucleic acids, to produce two or more amplified samples in which is presence or absence of amplicons is analysed.
The multiplex analysis of the amplified samples following the method of the invention comprises applying at least part of an amplified sample to an electrophoretic device for subsequent separation and detection. Preferably such an amplified sample contains, or is at least suspected to contain, amplified connected probes, which is an indication that a target sequence has hybridised with the provided oligonucleotide probes and that those probes were annealed adjacently on the complementary target sequence so that they where connected, i.e. ligated. Subsequently, an amplified sample is subjected to a separating step for a selected time period before a next amplified sample is submitted.
In the method of the invention, (parts of) two or more different amplified samples are applied consecutively to the same channel of the electrophoretic device (
In a preferred embodiment the method of the invention further comprises the following steps:
(e1) repeating steps (a) to (e) to generate at least two amplified samples;
(e2) consecutively applying at least part of the amplified samples obtained in steps (e) and (e1), to an application location of a channel of an electrophoretic device, electrophoretically separating the amplicons in the amplified samples and detecting the separated amplicons at a detection location located distal from the application location of the channel; whereby the time period between the consecutively applied amplified samples is such that the slowest migrating amplified connected probe in an amplified sample is detected at the detection location before the fastest migrating amplified connected probe of a subsequently applied amplified sample is detected at the detection location.
The method according to the invention allows for the high throughput analysis of a multiplicity of samples each comprising a multiplicity of different target sequences by the consecutive injection of amplified samples, comprising amplicons corresponding to the target sequences in the samples, in a channel of a multichannel electrophoretic device such as a capillary electrophoresis device. The method according to the invention allows for the analysis of a multiplicity of target sequences in a multiplicity of samples on a multiplicity of channels, thereby significantly increasing the throughput of the number of samples that can be analysed in a given time frame compared to conventional methods for the analysis of nucleotide sequences. This method profits from samples containing amplicons to be detected that are of a discrete size range as thereby the time period (23) between the successive injections can be significantly reduced compared to methods wherein the (remains of) concatamers are present.
The selected time period prevents that consecutively applied samples after separation have an overlap of amplicons at the detection point. The selected time period is influenced by i). the length of the amplicons; ii). the length variation in amplicons; and iii). the detection device and its operating conditions. Applying samples and separating consecutively applied samples in the same channel can be repeatedly performed in one or more channels, preferably simultaneously to allow for consecutive electrophoretic separation of multiple samples in one channel and/or simultaneous analysis of multiple samples over multiple channels and/or simultaneous analysis of multiple samples over multiple channels carried out consecutively. A graphic representation thereof is given in
The period of time between two consecutively loaded amplified samples can be determined experimentally prior to executing the method. This period of time is selected such that, given the characteristics of an amplified sample, especially the difference in length between the shortest and the longest amplicons in an amplified sample, as well as other experimental factors such as gel (matrix) and/or buffer concentrations, ionic strength etc., the fragments in an amplified samples are separated to such extent at the detection location which is located at the opposite end (distal) from the application location where the sample was applied, that the different amplicons in a sample may be individually detected. After applying the last amplified sample, the separation can be continued for an additional period of time to allow the amplicons of the last sample to be separated and detected. The combination of the selected period of time between applying two consecutive samples and the optional additional time period is chosen such that at the detection location the different amplicons in consecutively applied samples are separated such that they may be individually detected, despite the limited length variation that exists between the different amplicons within a single sample. Thus overlapping migration patterns are prevented when samples containing fragments of varying length are consecutively applied (injected) on the electrophoretic device.
Using the method according to the invention, it is in principle possible and preferred to continuously apply, load or inject samples. Preferably the device is able to perform such operation automatically, e.g. controlled by a programmable computer. Preferably the multichannel device is suitable for such operation or is at least equipped for a prolonged operation without maintenance such as replacement of buffers, parts etcetera. However, in practice this will generally not be the case. When a final sample is submitted it is generally needed to continue the separation for an additional time period until the last fragment of the final sample has been detected. In a preferred embodiment of the invention, the stuffers present in the tags of the oligonucleotide probes is are used to provide the length differences (i.e. 0 to 500 nucleotides, bases or base pairs) between the amplified connected probes. The total length of the amplicon and the variation in the length is governed mostly by the techniques by which these fragments are analysed. In the high throughput multiple injection method of the present invention, it is preferred that the range of lengths of amplicons in an amplified sample has a lower limit of 40, 60, 80, or 100 and an upper limit of 120, 140, 160, or 180 nucleotides, bases or base pairs, for conventional (capillary) electrophoresis platforms. It is particularly preferred that the range of lengths of the amplicons varies from 100 to 140 nucleotides. However, these numbers are strongly related to the current limits of the presently known techniques. Based on the knowledge provided by this invention, the skilled artisan is capable of adapting these parameters when other circumstances apply.
The reliability of the multiplex amplification is further improved by limiting the variation in the length of the amplified connected probes. Limitations in the length variation of amplicons is preferred to use multiple injection more efficiently and further results in reduction of the preferential amplification of smaller amplicon in a competitive amplification reaction with larger connected probes. This improves the reliability of the high throughput method of the present invention. Together with the multiple injection protocol as herein disclosed, these measures, alone or in combination provide for a significant increase in throughput in comparison with the art. A further improvement of the high throughput capacity is obtained by limiting the number of different amplicons in a sample. It is regarded as more efficient and economical to limit the multiplex capacity of the ligation/amplification step in combination with the introduction of a multiple injection protocol. One of the most advantageous aspects of the present invention lies in the combination of multiplex ligation, multiplex amplification, preferably with a single primer pair or with multiple primer pairs which each amplify multiple connected probes, repeated injection and multiplex detection of different labels. One of the further advantageous aspects of the present invention resides in the combined application of length differences with different (overlapping) labels such that each connected probe and hence each target sequence within one sample can be characterised by a unique combination of length and label. This allows for a significant improvement of the efficiency of the analysis of target sequences as well as a significant reduction in the costs for each target analysed.
The multiple injection protocol can be performed in a variety of ways. One of these is the multiple loading of two or more samples in the same matrix. This is considered as advantageously as the matrix is re-used by performing consecutive short runs, thereby increasing efficiency and throughput. Another one is the multiple loading of two or more samples in the same matrix in the same run. It is preferred to re-use the matrix by performing short consecutive runs. In this embodiment, a first sample is injected and separated. As soon as the last fragment is detected, the next sample is loaded. Preferably, between these two consecutive short runs the matrix is not replaced so that the runs are performed in the same matrix. This provides for additional efficiency and improved economics as less changes o the matrix need to occur, reducing the amount of consumables of this type of analysis (i.e. buffers etc.), reducing the cost per datapoint. Furthermore time-consuming replacements of the matrix can be avoided to a large extent, further increasing the efficiency of the method.
In itself, certain aspects of multiple loading or multiple injection have been described inter alia in U.S. Pat. No. 6,156,178 and WO 01/04618. The latter publication discloses an apparatus and a method for the increased throughput analysis of small compounds using multiple temporally spaced injections. The publication discloses that samples comprising primers, extended by one nucleotide (single nucleotide primer extension or SnuPE, also known as minisequencing) could be detected using multiple temporally spaced injections on a capillary electrophoresis device. Minisequencing is based on annealing a complementary primer to a previously amplified target sequence. Subsequent extension of the primer with a separately provided labelled nucleotide provides for identification of the nucleotide adjacent to the primer. Principally, the primer extension product is of a constant length. To increase throughput the use of successive injections of extension products of the same length per run is suggested. To further increase the throughput, primers of a different length can be used, varying typically from 15 to 25 nucleotides. In contrast, the present invention contemplates analysing multiplex amplification products themselves directly with a length variation typically between 50 and 150 nucleotides. This is significantly more economical than minisequencing or SnuPE as outlined hereinbefore because multiple target sequences are amplified in a single reaction, whereas with minisequencing or SnuPE amplification is carried out individually for each target sequence. Furthermore, the use of primers of a different length and complementary to the target sequence compromises the efficiency of the subsequent amplification step needed in the method of the present invention. These applications in general do not address the problems associated with high throughput detection of highly multiplexed samples, nor provide solutions thereto.
Exonucleases
A preferred method of the invention further comprises a step for the removal of oligonucleotide probes that are not annealed to target sequences and/or that are non-connected/ligated. Removal of such probes preferably is carried out prior to amplification, and preferably by digestion with exonucleases. By removal/elimination of the oligonucleotide probes that are not connected/ligated a significant reduction of ligation independent (incorrect) target amplification can be achieved, resulting in an increased signal-to-noise ratio. One solution to eliminate one or more of the non-connected/ligated components without removing the information content of the connected probes is to use exonuclease to digest non-connected/ligated oligonucleotide probes.issensitive. sensitive. Blocking groups include use of a thiophosphate group and/or use of 2-O-methyl ribose sugar groups in the backbone. Exonucleases include ExoI (3′-5′ activity), Exo III (3′-5′ activity), and Exo IV (both 5′-3′ and 3′-5′ activity). The circular probes of the present invention are, once ligated, insensitive to the exonuclease, as opposed to the unligated circular probes This is a further advantage of the use of padlock probes in the present invention.
An advantage of using exonucleases, for example a combination of Exo I (single strand specific) and Exo III (double strand specific), is the ability to destroy both the target sequence and the unligated oligonucleotide probes, while leaving the ligation product sequences substantially undigested. By using an exonuclease treatment prior to amplification, the oligonucleotide probes in each set are substantially reduced, and thus hybridisation of the remaining unligated oligonucleotide probes to the original target DNA (which is also substantially reduced by exonuclease treatment) and formation of a ligation product sequence which is a suitable substrate for PCR amplification by the oligonucleotide primer set is substantially reduced, thereby improving the signal to noise ratio.
Size Ladder
The sample can be supplied with a nucleotide fragment size standard comprising one or more nucleotide fragments of known length. Methods of preparing and using nucleotide size standards are well known in the art (see e.g. Sambrook and Russel, 2001, supra). Such a size standard forms the basis for appropriate sizing of the amplicons in the sample, and hence, for the proper identification of the detected fragment. The size standard is preferably supplied with every sample and/or with every injection. A size standard preferably contains a variety of lengths that preferably spans the entire region of lengths to be analysed. In a particular embodiment of the invention, it is considered advantageously to add flanking size standards from which the sizes of the amplicons can be derived by interpolation. A flanking size standard is a size standard that comprises at least two labelled oligonucleotide sequences of which preferably one has a length that is at least one base shorter than the shortest amplified connected probe and preferably one that is a least one base longer than the longest amplified connected probe to allow interpolation and minimise the introduction of further length variation in the sample. A preferred flanking size standard contains one nucleotide that is one nucleotide shorter the shortest amplified connected probe and one that is a least one base longer than the longest amplified connected probe and is labelled with at least one dye that is identical to the label used for labelling the amplicons contained in the sample.
A convenient way to assemble a suitable size standard is by (custom) chemical synthesis of oligonucleotides of the appropriate lengths, which are end-labelled with a suitable label. The size standard is applied with every consecutively applied sample to serve as local size references to size the loaded sample fragments. The size standard may be applied in the same channel or lane of the electrophoretic device as the sample to be analysed, i.e. together with the sample, or may be applied in a parallel channel or lane of a multichannel/lane device. The flanking size standard can be labelled with any of the labels used in the method. If the size standard is applied in the same channel of the device, the fragments of the standard are preferably labelled with a label that can be distinguished from the labels used for the detection of the amplicons in a sample.
Pooling
In a variant of the technology, the starting (DNA) material of multiple individuals are pooled such that less detection samples containing this material are loaded on the detection device, This can be advantageous in the case of Linkage Disequilibrium (LD mapping) when the objective is to identify amplified connected probes (such as those representing SNP alleles) that are specific for a particular pool of starting samples, for example pools of starting material derived from individuals which have different phenotypes for a particular trait.
Application
One aspect of the invention pertains to the use of the method in a variety of applications. Application of the method according to the invention is found in, but not limited to, techniques such as genotyping, transcript profiling, genetic mapping, gene discovery, marker assisted selection, seed quality control, hybrid selection, QTL mapping, bulked segregant analysis, DNA fingerprinting and microsatellite analysis. Another aspect pertains to the simultaneous high throughput detection of the quantitative abundance of target nucleic acids sequences. This approach is commonly known as Bulk Segregant Analysis (BSA).
Detection of Single Nucleotide Polymorphisms
One particular preferred application of the high throughput method according to the invention is found in the detection of single nucleotide polymorphisms (SNPs). A first target complementary part of the circular oligonucleotide probes is preferably located adjacent to the polymorphic site, i.e. the single polymorphic nucleotide. A second target complementary part is designed such that its terminal base is located at the polymorphic site, i.e. is complementary to the single polymorphic nucleotide. If the terminal base is complementary to the nucleotide present at the polymorphic site in a target sequence, it will anneal to the target sequence and will result in the ligation of the two target complementary parts. When the end-nucleotide, i.e. the allele-specific nucleotide does not match, no ligation or only a low level of ligation will occur and the polymorphism will remain undetected.
When one of the target sequences in a sample is derived from or contains a single nucleotide polymorphism (SNP), in addition to the probes specific for that allele, further probes can be provided that not only allow for the identification of that allele, but also for the identification of each of the possible alleles of the SNP (co-dominant scoring). To this end a combination of target complementary parts can be provided: one complementary part is the same for all alleles concerned and one or more of the other complementary parts which is specific for each of the possible alleles. These one or more other type of complementary parts contain the basically the same complementary sequence but differ in that each contains a nucleotide, preferably at the end, that corresponds to the specific allele. The allele specific part can be provided in a number corresponding to the number of different alleles expected. The result is that one SNP can be characterised by the combination of one complementary part with four other (allele-specific) complementary parts, identifying all four theoretically possible alleles (one for A, T, C, and G), by incorporating stuffer sequences of different lengths (preferred) or different labels into the allele specific probes.
In a particular embodiment, preferably directed to the identification of single nucleotide polymorphisms, the first complementary part of the oligonucleotide probe is directed to a part of the target sequence that does not contain the polymorphic site and the second complementary part of the oligonucleotide probe contains, preferably at the end distal from first complementary part, one or more nucleotide(s) complementary to the polymorphic site of interest. After ligation of the adjacent parts, the connected probe is specific for one of the alleles of a single nucleotide polymorphism.
To identify the allele of polymorphic site in the target sequence, a set of oligonucleotide probes can be provided wherein one first complementary part is provided and one or more second complementary parts. Each second complementary part then contains a specific nucleotide at the end of the complementary sequence, preferably the 3′-end, in combination with a known length of the stuffer. For instance, in case of an A/C polymorphism, the second complementary part can contain a specific nucleotide T in combination with a stuffer length of 2 nucleotides and another second complementary part for this polymorphism combines G with a stuffer length of 0. As the primers and the complementary parts of the probes are preferably the same length, this creates a length difference of the resulting amplicons of 2 nucleotides. In case the presence and/or the absence of all four theoretically possible nucleotides of the polymorphic site is desired, the stuffer-specific nucleotide combination can be adapted accordingly. In a sample containing multiple target sequences, amplified with the same pair of amplification-primers (and hence label) or with multiple pairs of amplifications primers with labels that have overlapping emission spectra, the combined stuffer lengths are chosen such that all connected probes are of a unique length. In
Detection of Specific Target Sequence
The target sequence contains a known nucleotide sequence derived from a genome. Such a sequence does not necessarily contain a polymorphism, but is for instance specific for a gene, a promoter, an introgression segment or a transgene or contains information regarding a production trait, disease resistance, yield, hybrid vigour, is indicative of tumours or other diseases and/or gene function in humans, animals and plants. To this end, the first and second complementary parts of the circular probe are designed to correspond to a, preferably unique, target sequence in genome, associated with the desired information. The complementary parts in the target sequence are located adjacent to each other. In case the desired target sequence is present in the sample, the two probes will anneal adjacently and after ligation and amplification can be detected.
Detection of AFLP Markers
AFLP, its application and technology is described in Vos et al., Nucleic Acids Research, vol. 23, (1995), 4407-4414 as well as in EP-A 0 534 858 and U.S. Pat. No. 6,045,994, all incorporated herein by reference. For a further description of AFLP, its advantages, its embodiments, its techniques, enzymes, adapters, primers and further compounds, tools and definitions used, explicit reference is made to the relevant passages of the publications mentioned hereinbefore relating to AFLP. AFLP and its related technology is a powerful DNA fingerprinting technique for the identification of for instance specific genetic markers (so-called AFLP-markers), which can be indicative of the presence of certain genes or genetic traits or can in general be used for comparing DNA, cDNA or RNA samples of known origin or restriction pattern. AFLP-markers are in general associated with the presence of polymorphic sites in a nucleotide sequence to be analysed. Such a polymorphism can be present in the restriction site, in the selective nucleotides, for instance in the form of indels or substitutions or in the rest of the restriction fragment, for instance in the form of indels or substitutions. Once an AFLP marker is identified as such, the polymorphism associated with the AFLP-marker can be identified and probes can be developed for use in the ligation assay of the present invention.
In another aspect the present invention pertains to a circular nucleic acid probe comprising a first and a second part that is capable of hybridising to corresponding parts of a target sequence and further comprising at least one, preferably two primer-binding sequence and a stuffer. Further embodiments of the probe according to the present invention are as described herein above. The invention also pertains to a set of probes comprising two or more probes wherein each probe comprises a first part and a second part that is complementary to part of a target sequence and wherein the complementary first an second parts are located essentially adjacent when hybridised to the target sequence and wherein each probe further comprises a stuffer, which stuffer is located essentially next to the complementary part and at least one, preferably two primer-binding sequence located essentially adjacent to the stuffer.
The invention in a further aspect, pertains to the use of a circular probe or set of probes in the analysis of at least one nucleotide sequence and preferably in the detection of a single nucleotide polymorphism, wherein the set further comprises at least one additional probe that contains a nucleotide that is complementary to the known SNP allele. Preferably the set comprises a probe for each allele of a specific single nucleotide polymorphism. The use of a set of probes is further preferred in a method for the high throughput detection of single nucleotide polymorphisms wherein the length of the stuffer in the probe is specific for a locus and/or allele of a single nucleotide polymorphism
Another aspect of the invention relates to the primers and more in particular to the set of primers comprising a first primer and one or more second primers, wherein each second primer contains a label and which second primer comprises a nucleotide sequence that is specific for said label.
The present invention also finds embodiments in the form of kits. Kits according to the invention are for instance kits comprising probes suitable for use in the method as well as a kit comprising primers, further a combination kit, comprising primers and probes, preferably all suitably equipped with enzymes buffers etcetera, is provided by the present invention.
The efficiency of the present invention can be illustrated as follows. When a capillary electrophoretic device with 96 channels and capable of detecting four labels simultaneously is used, allowing for 12 subsequent injections per run per channel with a empirically optimised minimum selected time period between the injections, a sample containing 20 target sequences of interest allows for the high throughput detection of 96 (channels)*12 (injections)*20 (targets)*4 (labels)=92160 target sequences, using the method of the present invention. In the case of co-dominant SNP-detection, data regarding 46080 SNPs can be detected in a single run.
A target sequence (2) comprising a first (5) and a second (7) part to which parts first and second probes can be hybridised with sections (4) and (6) that are complementary, respectively. The probes contain a tag sequence (8,9) that is not complementary to the target sequence. The tag sequence may comprise a stuffer sequence (10,11) and a primer-binding site (12,13). After probe hybridisation and ligation the connected probe (15) can be amplified using primers (16, 17) capable of hybridising to the corresponding primer-binding sites. At least one of the primers contains a label (L). Amplification results in an amplified sample, comprising amplicons (20)
(a) only one probe contains a stuffer (10) and primer-binding sequences (12,13); and
(b) both probes contain a stuffer (10, 11) and primer-binding sequences (12,13).
B. Partial electropherogram of FAM labelled detection of the Landsberg erecta sample on a capillary electrophoretic device (MEGABace). The same multiplex mixture was injected. Amplified connected probes in a size range 97-134 bp and flanking sizer fragments (designated S) are 94, 95 and 137 bp. Probes and sizers are all labelled with FAM.
B: Mathematical cross-talk correction resulting in a processed, cross-talk corrected trace file. Cross talk is reduced, but remains of the overlapping spectra (FAM, JOE) are present, resulting in false positive (or negative) signals.
C, D, E, F: single label plots illustrate the presence of remnants (D, E) of the mathematical correction, compared to the positive signals (C, F)
B: Representation of the effect of the use of cross-talk correction by length-label combinations. Scored data and expected data are correctly interpreted and false-positive or negative data are eliminated.
All signals of A, B, C and D are obtained after processing by Genetic Profiler version 1 software from Molecular Dynamics. Signal shown in D is corrected for cross talk and hence shows processed signals. The signals in A, B, and C are raw data and are not corrected for cross talk.
A. Selective amplification of set 1 with E01k-Ned and M01k.
B. Selective amplification of set 2 with E03k-5′+T-JOE and M04k.
C. Selective amplification of set 5 with E04k-Fam and M03k.
All channels are visible. It is clear that it is possible to amplify a specific set out of a multiplex ligation product for more sets.
In order to prevent cross-hybridisation between the amplification products, it is preferred that the sequences of the stuffer sequences are different and do not form hair-pins. In the tables 1-5, stuffer sequences are presented which can be used for the development of probes for each fluorescent dye, and have been verified for the absence of hairpins using Primer Designer version 2.0 (copyright 1990, 1991, Scientific and Educational software) The stuffer sequences are assembled from randomly chosen tetramer blocks containing one G, C, T and A, and have therefore by definition a 50% GC content. The stuffer sequence in the forward OLA probe for the two SNP alleles are kept identical to avoid preferential SNP allele amplification.
Recombinant Inbred (RI) lines generated from a cross between the Arabidopsis ecotypes Colombia and Landsberg erecta (Lister and Dean, Plant Journal, 4, pp 745-750, (1993) were used. Seeds from the parental and RI lines were obtained from the Nottingham Arabidopsis Stock Centre.
DNA was isolated from leaf material of individual seedlings using methods known per se, for instance essentially as described in EP-0534858, and stored in 1×TE (10 mM Tris-HCl pH 8.0 containing 1 mM EDTA) solution. Concentrations were determined by UV measurements in a spectrophotometer (MERK) using standard procedures, and adjusted to 100 ng/μl using 1×TE.
The Arabidopsis SNP's that were selected from The Arabidopsis Information Resource (TAIR) website: http://www.arabidopsis.org/SNPs.html:, are summarised in Table 6 in
The oligonucleotide probes (5′-3′ orientation) were selected to discriminate the SNP alleles for each of the twelve SNP loci described in Example 2. PCR binding regions are underlined, stuffer sequences are double underlined. Reverse primers are phosphorylated at the 5′ end:. p indicates phosphorylated. The sequences are summarised in Table 7.
CGCCAGGGTTTTCCCAGTCACGACTTCAGGACTAGTCTATACCTTGAG
CGCCAGGGTTTTCCCAGTCACG
ACGACTTCAGGACTAGTCTATACCTT
CT
CGCCAGGGTTTTCCCAGTCACGACCTGCTCTTTCCTCGCTAGCTTCAGA
CGCCAGGGTTTTCCCAGTCACGAC
GACTGCTCTTTCCTCGCTAGCTTCA
CCGCT
CGCCAGGGTTTTCCCAGTCACGACGAAGAGGAGAGTGGCTACGAACTCT
CGCCAGGGTTTTCCCAGTCACGAC
GAGAAGAGGAGAGTGGCTACGAACT
TTATCCGCT
CGCCAGGGTTTTCCCAGTCACGACAATCGGCCTAAGCAAGCTTGTTTTC
CGCCAGGGTTTTCCCAGTCACGAC
GAAATCGGCCTAAGCAAGCTTGTTT
ATTGTTATCCGCT
CGCCAGGGTTTTCCCAGTCACGACGATCGGAAAGATATCGGAGCTCCTT
CGCCAGGGTTTTCCCAGTCACGACGAGATCGGAAAGATATCGGAGCTCC
TGAAATTGTTATCCGCT
CGCCAGGGTTTTCCCAGTCACGACGAACTGGCATCAATCAGGCCTCCAA
CGCCAGGGTTTTCCCAGTCACGAC
GAGAACTGGCATCAATCAGGCCTCC
TGTGTGAAATTGTTATCCGCT
CGCCAGGGTTTTCCCAGTCACGACGGACTCCAAGGTATTGTTAGGCGCC
CGCCAGGGTTTTCCCAGTCACGAC
GAGGACTCCAAGGTATTGTTAGGCG
T
TCCTGTGTGAAATTGTTATCCGCT
CGCCAGGGTTTTCCCAGTCACGACCATCTCTTGCGCCTTCTCAGTGTTG
CGCCAGGGTTTTCCCAGTCACGAC
GACATCTCTTGCGCCTTCTCAGTGT
TATCG
TCCTGTGTGAAATTGTTATCCGCT
CGCCAGGGTTTTCCCAGTCACGACAGTTTCAAAACCCATGACGCTTCTA
CGCCAGGGTTTTCCCAGTCACGAC
GAAGTTTCAAAACCCATGACGCTTC
GGATCAGCT
TCCTGTGTGAAATTGTTATCCGCT
ATAGCACTGAGTC
TCCTGTGTGAAATTGTTATCCGCT
CGCCAGGGTTTTCCCAGTCACGACGTAGAGGCTCTAAACAGCTGCTTCC
CGCCAGGGTTTTCCCAGTCAC
GACGAGTAGAGGCTCTAAACAGCTGCTT
CT
CGCCAGGGTTTTCCCAGTCACGACTAAGTCAGCTCCTAAGCTTCCATCG
CGCCAGGGTTTTCCCAGTCACGAC
GATAAGTCAGCTCCTAAGCTTCCAT
CCGCT
The sequences of the primer used for PCR amplification were complementary to the PCR primer binding regions incorporated in the ligation probes described in Example 3. The sequences represent the so called M13 forward and M13 reverse primers. Usually the forward primer is labelled with FAM or □33P-dATP depending on the detection platform. The sequence of the primers in 5′-3′ orientation are:
The concentration of these oligo's was adjusted to 50 ng/μl.
The composition of the buffers was: Hybridisation buffer (1×), 20 mM Tris-HCl pH 8.5, 5 mM MgCl2, 100 mM K.Cl, 10 mM DTT, 1 mM NAD+.Ligation buffer (1×) 20 mM Tris-HCl pH 7.6, 25 mM Kac, 10 mM MgAc2, 10 mM DTT, 1 mM NAD+,0.1% Triton-X100.PCR buffer (10×):10×PCR buffer (contains 15 mM MgCl2). (Qiagen, Valencia, United States of America).No additions were used in the PCR
Ligation Reactions:
Ligation reactions were carried out as follows: 100 ng genomic DNA (1 μl of 100 ng/μl) in 5 μl total volume was heat denatured by incubation for 5 minutes at 94° C. and cooled on ice. Next 4 fmol of each OLA forward and reverse probes described in Example 3 (36 oligonucleotides in total) were added, and the mixture was incubated for 16 hours at 60° C. Next, 1 unit of Taq Ligase (NEB) was added and the mixture was incubated for 15 minutes at 60° C.
Next, the ligase was heat-inactivated by incubation for 5× minutes at 94° C. and stored at −20° C. until further use.
PCR Amplification:
PCR reactions mixture contained 10 μl ligation mixture, 1 μl of 50 ng/μl (FAM or 33P) labelled M13 forward and reverse primer (as described in Example 4), 200 μM of each dNTP, 2.5 Units HotStarTaq Polymerase Qiagen, 5 μl 10×PCR buffer in a total volume of 50 μl.
Amplifications were carried out by thermal cycling in a Perkin Elmer 9700 thermo cycler (Perkin Elmer Cetus, Foster City, United States of America), according to the following thermal cycling profile:
Profile 1: Initial denaturation/enzyme activation 15 min at 94° C., followed by 35 cycles of: 30 sec at 94° C., 30 sec at 55° C., 1 min at 72° C., and a final extension of 2 min at 72° C., 4° C., forever.
Profile 2: Initial denaturation/enzyme activation 15 min at 94° C., followed by 35 cycles of: 5 se
In case a 33P end-labelled M13 forward PCR primers was used, the labelling was carried out by kination as described in Vos et al., 1995 (Nucleic Acids Research, vol. 23: no. 21, pp. 4407-4414, 1995 and patent EP0534858).
This Example illustrates that at least six SNP's can be simultaneously ligated and amplified using the multiplex ligation/amplification procedure. This example further illustrates that at least 12 SNPs can be detected in one sample. The results are represented in Table 8
Gel electrophoresis was performed as described in Vos et al., Nucleic Acids research 23(21), (1995), 4407-4414. After exposure of the dried gel to phospho-imaging screens (Fuji Photo Film Co., LTD, Type BAS III) for 16 hours, an image was obtained by scanning using the Fuji scanner (Fuji Photo Film Co., LTD, Fujix BAS 2000) and stored in digital form.
In case of detection using the MegaBACE 1000 capillary sequencing instrument, desalting and purification of the PCR reactions mixtures was carried in 96-well format, using the following procedure:
A. Preparation of the 96-Well Sephadex Purification Plates
Dry Sephadex™ G-50 superfine (Amersham Pharmacia Biotech, Uppsala, Sweden) was loaded into the wells of a 96-well plate MultiScreen®-HV, Millipore Corporation, Bedford, Mass., USA), using the 45 microliter column loader (Millipore Corporation) as follows:
1. Sephadex G-50 superfine was added to the column loader.
2. Excess Sephadex was removed from the top of the column loader with a scraper.
3. The Multiscreen-HV plate was placed upside-down on top of the Column Loader.
4. The Multiscreen-HV plate and the Column Loader were both inverted.
5. The Sephadex G-50 was released by tapping on top or at the side of the Column Loader.
Next, the Sephadex G-50 was swollen en rinsed as follows:
6. 200 μl Milli-Q water was added per well using a multi-channel pipettor.
7. A centrifuge alignment frame was placed on top of a standard 96-well microplate, the Multiscreen-HV plate was place on top and the minicolumns were packed by centrifugation for 5 min at 900 g.
8. The 96-well plate was emptied and placed back.
9. Steps 5-7 were repeated once.
10. 200 μl Milli-Q water was added to each well to swell the Sephadex G-50 and incubated for 2-3 hours. Occasionaly, at this stage the Multiscreen-HV plates with swollen mini-columns of Sephadex G-50 superfine were tightly sealed with parafilm and stored a refrigerator at 4° C. until further use.
11. A centrifuge alignment frame was placed on top of a standard 96-well microplate, the Multiscreen-HV plate was placed on top of the assembly and the minicolumns were packed by centrifugation for 5 min at 900 g.
12. The 96-well microplate was removed.
13. The mixtures containing the amplified connected probes were carefully added to the centre of each well.
14. Using the centrifuge alignment frame, the Multiscreen-HV plate was placed on top of a new standard U-bottom microtitre plate and centrifugation was carried out for 5 min at 900 g.
15. The eluate in the standard 96-well plate (approximately 25 μl per well) contains the purified product.
B. Dilution of the Purified Products
Purified samples were diluted 25-75 fold in Milli-Q water before injection.
Preparation of the Samples:
A 800-fold dilution of ET-900 Rox size standard (Amersham Pharmacia Biotech) was made in water. 8 μl diluted ET-900 Rox was added to 2 μl purified sample. Prior to running, the sample containing the sizing standard was heat denatured by incubation for 1 min at 94° C. and subsequently put on ice.
Detection on the MegaBACE:
MegaBACE capillaries were filled with 1×LPA matrix (Amersham Pharmacia Biotech, Piscataway, N.J., USA) according to the manufacturer's instructions. Parameters for electrokinetic injection of the samples were as follows: 45 sec at 3 kV. The run parameters were 110 min at 10 kV. Post-running, the cross-talk correction, smoothing of the peaks and cross-talk correction was carried out using Genetic Profiler software, version 1.0 build 20001017 (Molecular Dynamics, Sunnyvale, Calif., USA), and electropherograms generated.
The minimum time interval for adequate separation between two consecutively injected samples was determined by injecting the sizer sample as described in Example 8. The resulting time interval was used, with a small additional margin, when injecting the purified amplified connected probes from the oligonucleotide assay. The results are presented in
A. Partial electropherogram of FAM labelled detection of the Colombia sample on a capillary electrophoretic device (MegaBACE). The same multiplex mixture was injected twice. Amplified connected probes (size range 97-134 bp) and flanking sizer fragments (94, 95 and 137 bp) are all labelled with FAM
B. Partial electropherogram of FAM labelled detection of the Landsberg erecta sample on a capillary electrophoretic device (MegaBACE). The same multiplex mixture was injected twice. Amplified connected probes (size range 97-134 bp) and flanking sizer fragments (94, 95 and 137 bp) are all labelled with FAM.
In this experiment the use of different length-label combinations to avoid the negative influence of incomplete cross-talk removal on the quality of a dominantly scored (presence/absence) dataset of SNP markers is demonstrated. Stuffer lengths were chosen such that ET-ROX and JOE-labelled fragments have identical sizes, and that FAM and NED fragments have identical sizes, but differing by 1 basepair from those of ET-ROX and JOE-labelled fragments. The result is that even in case of incomplete cross-talk removal between dyes with overlapping emission spectra, the observed signal will not result in incorrect scoring because the expected sizes of the amplification products are known for every label. Hence length-label combinations define the expectance patterns for genuine signals are signals originating from incomplete cross-talk correction. The results are presented in
The example shows in
A). The effect of incomplete cross talk removal on the data quality in case of a sample that contains a ET-ROX labelled fragment of 120 basepair and a NED labelled fragment of 124 basepairs in a situation where fragments of a particular size can be observed in combination with all labels. In this case, incomplete cross-talk of ET-ROX signal into the FAM Channel at 120 bp removal leads to the incorrect scoring of a FAM fragment of 120 basepairs (in reality an ET-ROX labelled fragment of 120 basepairs). Similarly, incomplete cross-talk correction removal of NED signal into JOE at 124 bp leads to incorrect scoring of a JOE fragment of 124 basepairs (in reality a NED labelled fragment of 124 basepairs), in addition to the correct fragments.
B). The effect of the use of cross-talk-optimised length-label combinations such that ET-ROX- and FAM-labelled fragments of the same length are not avoided by choosing different stuffer lengths, because their emission spectra overlap. Similarly, same-size amplified connected probe fragments labelled with JOE and NED are avoided. In case of a hypothetical sample containing a 120 bp ET-ROX-labelled fragment and a 124 bp NED labelled fragment (identical to the that described above in A), the small but detectable signals (peaks) of FAM at 120 bp and of JOE at 124 bp that remain after incomplete (mathematical) cross-talk correction will not be scored because they are known to originate from cross talk of ET-ROX and NED signals, respectively. Hence, they have no impact on the data quality and both fragments are scored correctly.
The selected SNPs are identified and summarized in Table 9.
The circular oligonucleotide probes (5′-3′ orientation) were selected to discriminate the SNP alleles for each of the SNP loci described in Example 14. PCR binding regions are underlined, stuffer sequences are double underlined. Reverse primers are phosphorylated at the 5′ end:. p indicates phosphorylated. The sequences are summarised in Table 10.
The sequence of one of the primers used for PCR amplification was complementary to the PCR primer binding regions incorporated in the ligation probes described in Example 15. The sequence of the second PCR primer matched the PCR primer binding region of the probe. Usually the forward primer is labelled. The concentration of the oligonucleotides was adjusted to 50 ng/μl. The sequence of the primers in 5′-3′ orientation are depicted in Table 11.
9 samples (samples 1-9) of homozygous tomato lines (Example 14) were subjected to a multiplex oligonucleotide ligation reaction using a mixture of 20 padlock probes (set 4). Conditions used were 1× Taq DNA ligase buffer (NEB), 0.2 U/μl Taq DNA ligase, and 0.05 fmol/μl of each probe in a volume of 10 μl. Ligation was performed in a thermocycler (Perkin Elmer) with the following cycling conditions: 2 minutes at 94° C.+10*(15 seconds at 94° C.+60 minutes at 60° C.)+4° C. continuously. Following ligation, the 10 μl ligation product was diluted with 30 μl 1× Taq DNA ligase buffer. The 40 μl of each reaction was used to perform 4 amplification reactions using 4 different labelled E00k primers each combined with M00k. The E00k primer labelled with ET-ROX and JOE were designed with an extra 1 bp in comparison with E00k labelled with FAM and NED length, to prevent possible crosstalk between fluorescent labels when analysing these products on the MegaBACE. Conditions used were 30 ng labelled E00k primer and 30 ng M00k primer, 1× Accuprime buffer I, 0.4 ul Accuprime polymerase nitrogen) on 10 μl diluted ligation product in a 20 μl PCR reaction. PCR was performed in a thermocycler with the following cycling conditions: 2 minutes at 94° C.+35*(15 seconds at 94° C.+30 seconds at 56° C.+60 seconds at 68° C.)+4° C. continuously. PCR product was purified using Sephadex 50 and diluted 80 times with MQ. Diluted PCR product was analysed on the MegaBACE. The different fluorescent-labelled products were run separately and in different combinations (2, 3 and 4 fluorescent dyes). The results are presented in
The amplification products of set 4 were analysed using consecutive runs without replacement of the LPA matrix between runs. Samples of the amplification products were injected after a run period of 40 minutes without changing the matrix. Results are presented in
This experiment demonstrates the possibility of a higher multiplex of oligonucleotide ligation, in combination with the selective amplification of a subset of the formed ligation products using (AFLP) amplification primers with selective nucleotides.
Using the 4 designed probe sets, primers are based on set 1, 2, 4 and 5 but with additional selective nucleotides located immediately 3′ of the primer binding sites in the probes.
Each set was ligated separately, and in combination with other sets, up to a multiplex of 40 based on the 4 sets together. AFLP+1/+1 amplifications using different labelled E00k primers were performed using the scheme depicted below.
Conditions used were 1× Taq DNA ligase buffer (NEB), 0.2 U/μl Taq DNA ligase, and 0.05 fmol/μl of each probe in a volume of 10 μl. Ligation was performed in a thermocycler (Perkin Elmer) with the following cycling conditions: 2 minutes at 94° C.+10*(15 seconds at 94° C.; 60 minutes at 60° C.)+4° C. continuously. Following ligation, the 10 μl ligation product was diluted with 30 μl 1×Taq DNA ligase buffer. Conditions used were 30 ng labelled E00k primer and 30 ng M00k primer, 1× Accuprime buffer (Invitrogen) I, 0.4 ul Accuprime polymerase (Invitrogen) on 10 μl diluted ligation product in a 20 μl PCR reaction. PCR was performed in a thermocycler with the following cycling conditions: 2 minutes at 94° C.+35*(15 seconds at 94° C.+30 seconds at 56° C.+6 minutes at 68° C.)+4° C. continuously. PCR product was purified using Sephadex 50 and diluted 80 times with MQ. Diluted PCR product was analysed on the Megabace. The different fluorescent-labelled products were run in separate capillaries. The results are presented in
Buffer Compositions:
1×Taq DNA ligase buffer
| Number | Date | Country | Kind |
|---|---|---|---|
| 01204912 | Dec 2001 | EP | regional |
| Filing Document | Filing Date | Country | Kind | 371c Date |
|---|---|---|---|---|
| PCT/NL02/00834 | 12/16/2002 | WO | 00 | 9/13/2005 |
| Publishing Document | Publishing Date | Country | Kind |
|---|---|---|---|
| WO03/052142 | 6/26/2003 | WO | A |
| Number | Name | Date | Kind |
|---|---|---|---|
| 6156178 | Peponnet et al. | Dec 2000 | A |
| 6221603 | Mahtani | Apr 2001 | B1 |
| Number | Date | Country |
|---|---|---|
| WO 95 22623 | Aug 1995 | WO |
| WO 96 15271 | May 1996 | WO |
| WO 01 06012 | Jan 2001 | WO |
| Number | Date | Country | |
|---|---|---|---|
| 20060121458 A1 | Jun 2006 | US |
