 |
Description  |
|
|
FIELD OF THE INVENTION
This invention relates to methods for detecting the presence or absence of, and analysing the sequence of, target nucleic acids in a sample. Such methods may also be applied to target nucleic acid units such as nucleotides and nucleosides and
their analogues. The invention also relates to chemical complexes for use in such methods, to a kit of reagents for use in carrying out the methods and to certain novel compounds of use in the methods.
BACKGROUND TO THE INVENTION
Detection of Nucleic Acids
There are many situations in which it is necessary to detect, either qualitatively or quantitatively, the presence of nucleic acids such as DNA and RNA or their constituent nucleotides. Examples of such situations include medical diagnosis (eg,
the detection of infectious agents like bacteria and viruses, the diagnosis of inherited and acquired genetic diseases and the establishment of tissue type), forensic tests in criminal investigations and paternity disputes and of course the more general
attempt to sequence human and animal genes.
Techniques are already known for detecting nucleic acids and nucleic acid units. Available methods include, for instance:
a) fluorescence spectroscopy--this is technically very demanding if high sensitivities are to be achieved. In biological assays, its use tends to be complicated by autofluorescence of the analytes.
b) radiolabelling--this also requires high levels of technical skill but tends to be less sensitive than fluorescence spectroscopy. It also suffers from the obvious hazards involved in handling radioactive materials.
c) chemiluminescence--although this technique can be relatively quick to carry out, and avoids the problem of autofluorescence and the need to handle toxic substances, it is unfortunately relatively insensitive and yet is still technically
demanding.
A disadvantage common to many known techniques is their need for large amounts of the target analyte, ie, their relatively low sensitivity. Often
in the situations mentioned above the target is simply not available in sufficiently high concentrations. As a result, the available target material has to be amplified before its presence can be accurately detected.
Again, techniques are known for amplifying a nucleic acid. The most common is the well-known "polymerase chain reaction" ("PCR"). Alternatively, the target nucleic acid may be cloned into a biological vector such as a plasmid, a phage or the
like, which is then inserted into a (typically bacterial) host cell. The host is permitted to multiply and the desired vector is "harvested" from the host cell after an appropriate period of time.
Clearly, the need for amplification makes a detection method more complex, costly and time-consuming and introduces greater potential for error and for contamination of the target material.
There is therefore a need for a nucleic acid detection method which is sensitive to relatively low target concentrations, and which can preferably be carried out directly on an unamplified sample. It is this need that the present invention
addresses.
Surface Enhanced Raman Scattering
The invention provides a technique based on the principle of "surface enhanced Raman scattering" (SERS) and on a modification of that principle known as SERRS (surface enhanced resonance Raman scattering). These principles are already known and
well documented, and have been used before in the detection and analysis of various target materials.
Briefly, a Raman spectrum arises because light incident on an analyte is scattered due to excitation of electrons in the analyte. "Raman" scattering occurs when an excited electron returns to an energy level other than that from which it
came--this results in a change in wavelength of the scattered light and gives rise to a series of spectral lines at both higher and lower frequencies than that of the incident light. The scattered light can be detected orthogonally to the incident beam.
Normal Raman lines are relatively weak and Raman spectroscopy is therefore too insensitive, relative to other available detection methods, to be of use in chemical analysis. Raman spectroscopy is also unsuccessful for fluorescent materials, for
which the broad fluorescence emission bands (also detected orthogonally to the incident light) tend to swamp the weaker Raman emissions.
However, a modified form of Raman spectroscopy, based on "surface-enhanced" Raman scattering (SERS), has proved to be more sensitive and hence of more general use. The analyte whose spectrum is being recorded is closely associated with a
roughened metal surface. This leads to a large increase in detection sensitivity, the effect being more marked the closer the analyte sits to the "active" surface (the optimum position is in the first molecular layer around the surface, ie, within about
20 nm of the surface).
The theory of this surface enhancement is not yet fully understood, but it is thought that the higher valence electrons of the analyte associate with pools of electrons (known as "plasmons") in pits on the metal surface. When incident light
excites the analyte electrons, the effect is transferred to the plasmons, which are much larger than the electron cloud surrounding the analyte, and this acts to enhance the output signal, often by a factor of more than 10.sup.6. Fluorescence is also
quenched, giving cleaner Raman spectra and allowing fluorescent dyes to be used as detectable analytes. Generally, the signal enhancement means that a much larger range of analytes may be usefully detected than using normal Raman spectroscopy.
Furthermore, the enhancement means that a less powerful light source is required to excite the analyte molecules.
A further increase in sensitivity can be obtained by operating at the resonance frequency of the analyte (in this case usually a dye attached to the target of interest). Use of a coherent light source, tuned to the absorbance maximum of the dye,
gives rise to a 10.sup.3 -10.sup.5 -fold increase in sensitivity. This is termed "resonance Raman scattering" spectroscopy.
When the surface enhancement effect and the resonance effect are combined, to give "surface enhanced resonance Raman scattering" or SERRS, the resultant increase in sensitivity and robustness is more than additive. Moreover, the sensitivity does
not seem to depend so critically on the angle of orientation of the analyte to the surface, as is the case with SERS alone. A SERRS signal can be more easily discriminated from contamination and background and tends to be less variable with local
conditions (eg, ionic strength or pH when an analysis is carried out in solution). SERRS is thus a surprisingly sensitive detection technique; in many instances it appears to be at least as good as, if not better than, fluorescence spectroscopy (see eg,
C Rodger et al, J. Chem. Soc. Dalton Trans. (1996), pp791-799).
SERRS can also be used selectively to detect several analytes without the need for prior separation as would be necessary for fluorescence spectroscopy (see C H Munro et al in Analyst, April 1995, 120, pp993-1003).
Prior Art Relating to SERS and SERRS
SERS and SERRS have been used in the past for detecting a variety of species. Examples of relevant prior art documents include:
Appl. Spectroscopy (1993), 47, pp80-84 (J C Rubim et al)--preparation of SERS-active brass surfaces and the SERS detection of benzotriazole.
J. Raman Spectroscopy (1994), 25, pp899-901 (H Wilson et al)--SERS detection of benzotriazole deposited onto a silver colloid surface.
J. Phys. Chem. (1995), 99, pp879-885 (C H Munro et al)--use of SERRS to detect an azo dye, Solvent Yellow 14, and an explanation of the mechanisms involved.
Analyst, April 1995, 120, pp993-1003 (C H Munro et al)--SERRS detection of acidic monoazo dyes.
J. Raman Spectroscopy (1991), 22, pp771-775 (J Clarkson et al)--the effects of solvent on SERS detection of organic species on silver colloid surfaces.
U.S. Pat. No. 4,674,878 (Vo-Dinh)--ways of preparing SERS substrates, and example spectra for various organic compounds (though not nucleic acids). Detection sensitivities at nanogram and sub-nanogram levels are reported.
U.S. Pat. No. 5,400,136 (Vo-Dinh)--special coatings for SERS-active surfaces. In the examples, relatively high powered lasers are used as the light source, suggesting a fairly low level of sensitivity. Again, there is no reference to nucleic
acids as target analytes.
Anal. Chem. (1990), 62, p2437-2441 (J M Bello et al)--the use of fibre optic sensors in obtaining SERS spectra. Detection limits of no lower than .about.10.sup.-7 M are quoted for various organic compounds.
Appl. Spectroscopy (1995), 49, No. 6, pp780-784 (K Kneipp et al)--detection of relatively low concentration (.about.10.sup.-16 M) of the dye rhodamine 6G, using SERRS. It should be borne in mind that this dye is likely to interact differently
with a SERRS-active surface than would a Raman-labelled nucleic acid.
Mention has also been made of using SERS and SERRS to detect DNA and RNA. However, the concentrations detected have been relatively high. This suggests that prior art methods have not been sensitive enough to detect unamplified samples.
The following documents are relevant to the use of SER(R)S to detect nucleic acids:
J. Raman Spectroscopy (1991), 22, pp729-742 (T M Cotton et al)--this provides an overview of the applications of SERS and SERRS spectroscopy in biological systems. The detection of DNA is referred to, and potential problems are discussed. There
is no indication of the detection sensitivities achievable in DNA analyses.
U.S. Pat. No. 5,306,403 (Vo-Dinh)--this proposes the detection of DNA by labelling with a dye and adsorbing the resulting complex onto a SERS-active surface. However, there is no enabling disclosure of a technique with sufficient sensitivity
to be used without prior DNA amplification. Most of the examples relate to detection of isolated dyes, rather than of a dye-DNA complex (which, as explained below, would behave very differently under SERS conditions)--in these examples, the minimum dye
concentration in the solutions investigated is 0.05 mg/ml, which probably equates to the detection of between .about.10.sup.7 and 10.sup.11 molecules. Only one example is given of the detection of a (very short) oligonucleotide labelled with
aminoacridine; no concentration data is given in this example at all.
Anal. Chem. (1994), 66, pp3379-3383 (Vo-Dinh et al)--this paper reports the detection of DNA using SERS, but only at relatively high concentrations (10.sup.19 M or greater; whilst it is impossible to make exact calculations, it is unlikely that
fewer than .about.10.sup.5 molecules of target were detected in the examples given). These detection levels, and the reference to the use of PCR in the paper's conclusion, indicate that the technique disclosed would still be unsuitable for detecting
unamplified DNA samples.
U.S. Pat. No. 5,266,498, U.S. Pat. No. 5,376,556 and U.S. Pat. No. 5,445,972 (Tarcha et al)--these describe the detection, using SERS, of an analyte by monitoring an analyte-mediated ligand binding event. A "capture reagent" is prepared by
attaching a SERS-labelled binding member, specific to the target analyte, to a SERS-active surface. Binding of the specific binding member to the analyte, in a test sample, causes a detectable change in the SERS spectrum for the capture reagent.
Nucleotide sequences are briefly mentioned as possible analytes, but the documents give no example of this and no explanation as to how appropriate sensitivities might be achieved, particularly for unamplified nucleotide samples.
J. Molecular Structure (1986), 145, pp173-179 (K Kneipp et al)--SERS detection of DNA on silver sols. The DNA concentration in the experiments is .about..mu.g ml.sup.-1 ; the possibility of detecting nanogram quantities of DNA is also mentioned.
Studia Biophysica (1989), 130, pp45-50 (J Flemming et al)--again, SERS detection of DNA on silver colloid surfaces, at concentrations .about..mu.g ml.sup.-1.
Anal. Chem. (1990), 62, pp1958-1963 (F Ni et al)--investigates the possibility of combining SERS spectroscopy with flow injection analysis, to detect RNA bases at relatively high (.about.10.sup.-4 M) concentrations.
Anal. Chem. (1991), 63, pp437-442 (R Sheng et al)--use of reversed-phase high performance liquid chromatography in combination with SERS, to detect nanomolar quantities of nucleic acid bases. Sensitivity limitations are discussed, as are
possible ways of overcoming them.
J. Molecular Structure (1991), 244, pp183-192 (K Kneipp et al)--SERS detection of various nucleic acids, including DNA and RNA, at concentrations no lower than .about.10 .mu.g ml.sup.-1.
Appl. Spectroscopy (1994), 48, pp951-955 (K Kneipp et al)--near-infrared SERS detection of the DNA base adenine adsorbed onto silver or gold colloidal particles. The lowest base concentration detected is 10.sup.-7 M.
Thus, earlier experiments have in common the fact that they use relatively large quantities of nucleic acid analyte. None has yet demonstrated sensitivities high enough to allow the detection of unamplified nucleic acid samples (ie, the
detection of perhaps 1-100 molecules in a sample).
That SERS and SERRS have never been proposed for use in the detection of unamplified nucleic acids is due at least in part to the obvious difficulties in achieving the appropriate sensitivities. These difficulties are partly due to problems
specific to nucleic acids, problems which are therefore not addressed in the more general SER(R)S literature.
The skilled person seeking to detect nucleic acids or nucleic acid units would thus consider SER(R)S spectroscopy to lack the necessary sensitivity or robustness, certainly without target amplification. The need remains for an alternative
detection method, suitable for use with very low concentrations of target, and this is what the present invention provides.
STATEMENTS OF THE INVENTION
First Aspect
According to its first aspect, the present invention provides a method for detecting the presence of a target nucleic acid or nucleic acid unit in a sample, the method comprising the steps (in any appropriate order) of:
a) forming a primary complex between a SER(R)S-active label and any target present in the sample, optionally via a target binding species containing a nucleic acid or nucleic acid unit;
b) preparing a detection sample in which the primary complex, or a secondary complex containing the label and the target binding species and derived directly from the primary complex, is associated with a SER(R)S-active surface; and
c) detecting the presence of the primary or the secondary complex in the detection sample (and hence of the target in the original sample) by obtaining and analysing a SER(R)S spectrum for the detection sample;
wherein, in the detection sample, the concentration of the target present in the primary complex, or of the nucleic acid or unit contained in the target binding species in the secondary complex, is no higher than 10.sup.-10 moles per liter.
The target, the label, the primary target-label complex, the target binding species, the secondary complex and the SER(R)S-active surface are more specifically defined below. Concentration in the detection sample refer to concentrations in the
sample actually under investigation, ie, in the case of a fluid phase investigation, the sample from which a SER(R)S spectrum is directly taken or, in the case of a solid phase investigation, the sample which is applied to a SER(R)S-active surface in
order to obtain a spectrum.
Clearly, subject to practical constraints, there is no lower limit on the nucleic acid/nucleic acid unit concentration which the present invention may be used to detect. For instance, it might usefully be carried out using detection samples
containing fewer than 100 copies, for example fewer than 50, or perhaps fewer than 20 or fewer than 10 copies or, in particular, fewer than 5 copies of the relevant nucleic acid or nucleic acid unit. Certainly the detection of picomolar (10.sup.-10 to
10.sup.-12 moles per liter) or femtomolar (10.sup.-13 to 10.sup.-15 moles per liter) or lower, possibly much lower, perhaps attomolar (10.sup.-16 to 10.sup.-18 moles per liter) concentrations, or even below, can be envisaged.
The method may of course also be used to establish the absence of the target in the sample, by carrying out the same steps and detecting an absence of the relevant primary or secondary complex in the detection sample.
As explained above, neither SERS nor SERRS has been used in the past to detect nucleic acids at such low concentrations, ie, at concentrations likely to represent unamplified samples of target material. The present invention makes such detection
possible by greatly increasing the sensitivity of conventional SER(R)S techniques, hence providing a completely new and improved method for detecting nucleic acids and their constituent units, a method which can potentially be carried out much more
quickly and cheaply, and with less skill, than existing detection methods.
The increased sensitivity may be achieved, in the present invention, in the manner described below. It involves the use of at least one, preferably more, of three modifying features, all of which help to bring the primary or secondary complex
into closer proximity with the SER(R)S-active surface being used to obtain the SER(R)S spectrum.
Second Aspect
Thus, a preferred, second, aspect of the invention provides a method for detecting the presence or absence of a target nucleic acid or nucleic acid unit in a sample, the method comprising the steps (in any appropriate order) of:
a) forming a primary complex between a SER(R)S-active label and any target present in the sample, optionally via a target binding species containing a nucleic acid or nucleic acid unit;
b) preparing a detection sample in which the primary complex, or a secondary complex containing the label and the target binding species and
derived directly from the primary complex, is associated with a SER(R)S-active surface; and
c) detecting the presence or absence of the primary or the secondary complex in the detection sample (and hence of the target in the original sample) by obtaining and analysing a SER(R)S spectrum for the detection sample;
wherein one or more of the following features is used:
i) the introduction into the detection sample, prior to detection, of a monomeric or polymeric polyamine;
ii) modification, prior to detection, of the target, and/or of the nucleic acid or nucleic acid unit contained in the target binding species, in a manner that promotes or facilitates its chemi-sorption onto the SER(R)S-active surface;
iii) inclusion of a chemi-sorptive functional group in the SER(R)S-active label.
For this method too, the concentration in the detection sample of the target present in the primary complex, or of the nucleic acid or unit contained in the target binding species in the secondary complex, is preferably no higher than 10.sup.-10
moles per liter.
This method makes use of at least one (preferably more than one, more preferably all three) modification to enhance the sensitivity of existing SER(R)S detection methods, thus allowing it to be used, in some cases, to detect extremely low
concentrations of unamplified target.
The effect of each of features (i)-(iii) in increasing sensitivity is believed to be at least additive; two or more features used together can have a synergistic effect on detection sensitivity. Each feature is described in more depth below,
following the explanation of terms used in defining the invention.
Meaning of SER(R)S
Firstly, by SER(R)S is meant either surface enhanced Raman scattering or surface enhanced resonance Raman scattering. The methods of the invention may involve either form of spectroscopy, since the essential principle (the association of a
Raman-active label with a Raman-active surface) is the same in each case. Preferably, the methods of the invention involve SERRS rather than SERS, since operating at the resonant frequency of the label gives increased sensitivity--in this case, the
light source used to generate the Raman spectrum is a coherent light source (eg, a laser) tuned substantially to the maximum absorption frequency of the label being used. (Note that this frequency may shift slightly on association of the label with the
SER(R)S-active surface and the target and/or target binding species, but the skilled person will be well able to tune the light source to accommodate this. Note too that the light source may be tuned to a frequency near to the label's absorption
maximum, or to a frequency at or near that of a secondary peak in the label's absorption spectrum.)
SERRS may alternatively involve operating at the resonant frequency of the plasmons on the active surface, although in the methods of the invention it is believed to be preferable to tune to the resonant frequency of the label.
Type of Target and Sample
The methods of the invention may be used for the quantitative or qualitative detection of target nucleic acids and nucleic acid units, and to detect the absence as well as the presence of a target in a sample. They may form part of an overall
method for determining the sequence of a nucleic acid, by detecting the presence in it of several selected target nucleotides or nucleotide sequences.
The target nucleic acid may be a naturally occurring DNA, RNA, mRNA, rRNA or cDNA, or a synthetic DNA, RNA, PNA or other nucleic acid analogue. Typically, it will be a naturally occurring DNA or RNA. It may be an oligonucleotide or a
polynucleotide. In this document, unless the context requires otherwise, the term "nucleotide" is used to refer to either a deoxyribo- or a ribo-nucleotide or an analogue thereof; "oligonucleotide" to a nucleotide sequence of between 2 and 100 base
units; and "polynucleotide" to a nucleotide sequence of 50 base units or more.
Target oligonucleotide or polynucleotide may be substantially single- or double-stranded. It will be appreciated that an initial target may be subjected, prior to detection, to molecular biological manipulations such as digestion with
restriction enzymes or copying by means of nucleic acid polymerases, thus allowing modifications to be introduced into it.
The target may be a nucleic acid "unit", by which is meant a nucleotide or nucleoside or a modified nucleotide or nucleoside or a nucleotide or nucleoside analogue, or an individual nucleobase. The choice of target will depend on the purpose for
which the detection method is ultimately to be used--eg, for detecting the presence of bacteria or viruses in cells, it is likely that genomic DNA or RNA would be the most suitable target to detect.
The sample may be any suitable preparation in which the target is likely to be found. In the case of medical diagnostic techniques, for instance, the sample may comprise blood (including plasma and platelet fractions), spinal fluid, mucus,
sputum, semen, stool or urine. Particularly suitable samples include, eg, 20-1000 .mu.l of blood or 1-10 ml of mouthwash. Samples may also comprise foodstuffs and beverages, water suspected of contamination, etc . . . These lists are clearly not
exhaustive.
The sample will typically be pre-treated to isolate the target and make it suitable for subsequent SER(R)S analysis. Many methods and kits are available for pre-treating samples of various types.
The detection sample may be in any appropriate form such as a solid, a solution or suspension or a gas, suitably prepared to enable recordal of its SER(R)S spectrum. The detection sample can be at any suitable pH, typically an acidic pH.
The SER(R)S-Active Label
The label can be any suitable material which is SER(R)S-active, ie, which is capable of generating a SERS or SERRS spectrum when appropriately illuminated. It must also be capable of forming a primary complex with the target, either directly or
via a target binding species, in the manner described below.
It is possible in some cases, particularly when using SERS spectroscopy, that the target itself, or the target binding species, could also act as the SER(R)S-active label, capable of generating its own Raman spectrum.
Many SER(R)S-active labels are already known and referred to in SER(R)S literature. They include species containing chromophores and/or fluorophores which can be detected relatively easily using SER(R)S.
Examples of suitable SER(R)S-active species include fluorescein dyes, such as 5- (and 6-) carboxy-4',5'-dichloro-2',7'-dimethoxy fluorescein, 5-carboxy-2',4',5',7'-tetrachlorofluorescein and 5-carboxyfluorescein; rhodamine dyes such as 5- (and
6-) carboxy rhodamine, 6-carboxytetramethyl rhodamine and 6-carboxyrhodamine X; phthalocyanines such as methyl, nitrosyl, sulphonyl and amino phthalocyanines; azo dyes such as those listed in C H Munro et al, Analyst (1995), 120, p993; azomethines;
cyanines and xanthines such as the methyl, nitro, sulphano and amino derivatives; and succinylfluoresceins. Each of these may be substituted in any conventional manner, giving rise to a large number of useful labels. The choice of label in any given
case will depend on factors such as the resonance frequency of the label, the other species present, label availability, etc . . .
Most preferred are labels containing a chemi-adsorptive functional group, described below in connection with feature (iii) of the invention.
Preferred SER(R)S-active labels are, moreover, those which possess appropriate functional groups to allow their easy attachment both to the target (or an appropriate target binding species) and to the SER(R)S-active surface. The label should
clearly not contain groups likely to interfere with the target or the target binding species.
Where the label is to be bound to the target via a target binding species, the label is preferably used in the form of a pre-prepared label-binding species complex. Ways in which labels may be complexed with target binding species are outlined
below.
The Target Binding Species
Typically when carrying out the methods of the invention, it will be necessary to bind the target to the SER(R)S-active label via a target binding species. This may be in the form of, or contain, a nucleic acid or nucleic acid unit which is
substantially complementary to at least part of the target--in other words, the target binding species is typically a form of nucleic acid probe or primer specific to the target. (By "substantially complementary" is meant that the nucleic acid or unit
is capable of selective hybridisation to the target.) Subsequent detection of the SER(R)S-active label attached to the target binding species gives information as to the presence (or absence) of any target to which it is, or has been, bound.
The Primary Target-Label Complex
The primary complex, as described above, generally involves the specific binding of at least a part of the target with a target binding species which is in turn attached to the SER(R)S-active label. The main requirement of the target-label link
is that it forms a complex of the two species so that detection of the label by means of its SER(R)S spectrum would be equivalent, in effect, to detection of the attached target.
The link between the label and the target, or where applicable between the label and the target binding species, may involve any suitable form of attachment. Many methods are known, eg, from fluorescence spectroscopy, for linking dyes and other
labels to nucleic acids and/or nucleic acid units. Some involve chemical modification of the basic label structure. The skilled person would have no difficulty in selecting one appropriate for the particular label, target and target binding species
concerned. The attachment may be direct, via a covalent bond or chelating link, for instance. More preferably it is indirect through a separate linking group--again, appropriate linking groups are known, and these can help separate the label from
attached nucleic acids and nucleic acid units which can potentially (as explained below) interfere with the vital interaction between the label and the SER(R)S-active surface. DNA binding proteins may, for instance, function as "linking groups" in this
context.
The label is preferably attached to the 5' end of the relevant nucleic acid or nucleic acid unit, although attachment to the 3' end or to an intermediate position (eg, to a base or to a backbone sugar group) is also possible.
Two methods by which a SER(R)S-active label may be attached to a nucleic acid or nucleic acid unit include, for instance:
1. The nucleic acid (or unit) is synthesised with a nucleophilic primary amino group, usually at the 5'-terminus. After deprotection it is reacted with an appropriate reactive site (eg, an active ester site) on the label. Purification, usually
by chromatography, yields the desired product. (See eg, J Goodchild, Bioconjugate Chem. (1990), 1, pp165-187.)
2. The label is synthesised with a chemical group (usually an alcohol) capable of undergoing phosphorous functionalisation. The active phosphorous compound is then reacted with the nucleic acid or nucleic acid unit. This reaction can be
accomplished using several types of standard chemistry, as detailed for example in M J Gait, Oligonucleotide Synthesis: A Practical Approach (1984), IRL Press Oxford.
Further examples of suitable attachments, including via linking groups, appear in P Theisen et al, Tetrahedron Letters (1992), 33, No. 35, pp5033-5036; J M Prober et al, Science (1987), 233, pp336-341; D B Shealy et al, Anal. Chem. (1995), 67,
pp247-251; and C Mackellar et al, Nuc. Acids Res. (1992), 20, pp3411-3417.
Still further examples of ways in which a SER(R)S-active label may be bound to a nucleic acid or nucleic acid unit are the following typical known methods. Although 5' labelling of nucleic acids is illustrated, essentially identical
modifications of the 3' end and/or of internal sites are equally feasible.
Modified Dye ##STR1##
(See P Theisen et al, supra.)
Coupled Dye (Coupling May Take Place in Situ) ##STR2##
(X=a suitable group for attack by a primary amine, eg, isothiocyanate or N-succinamide; Y is its derivatised form.)
(See J Goodchild, supra for a specific example.)
Coupled Dye With a Modified Nucleobase ##STR3##
(X, Y=as above; B=any base in the nucleic acid or nucleic acid unit. Where the nucleic acid or unit is a target binding species, the product may then be used as a monomer in the synthesis of a labelled polynucleotide suitable for binding to the
target being detected.)
(See J M Prober et al, supra.)
The Secondary Complex
The "secondary complex" is derived directly from the primary target-label complex, so that its presence depends on the formation of the primary complex. Detection of the secondary complex, even if the original target is no longer present, then
corresponds (both qualitatively and preferably quantitatively) to detection of the target. The secondary complex still contains the SER(R)S-active label, so as to yield a detectable SER(R)S spectrum.
A secondary complex may be formed, for instance, by cleaving away all or part of the original target from the primary target-label complex. There are other similar situations in which a secondary complex may be derived directly from the primary
one by the addition or removal of species--the only requirement is that the presence of the secondary complex in the detection sample reflects the presence of the primary complex in the original sample.
At the time of detection, the primary and secondary "complexes" need not involve direct links, such as covalent bonds, between their constituent species. Also envisaged are situations in which, for instance, the target or the target binding
species is merely associated with the label, ie, it is still present in the detection sample and its presence is still capable of influencing the interaction of the other species present and therefore the SER(R)S spectrum, but it has for example been
cleaved from the label-containing complex.
Order of Combination of Species
The order of combination of the target, the label (with target binding species if used) and the SER(R)S-active surface is not critical in the methods of the invention.
Preferably the primary or secondary complex is formed and then added to the active surface. As an alternative, the first step may be to confirm the presence (or absence) of the target by reaction with a target binding species, which can
subsequently be detected by the addition of an appropriate SER(R)S-active label and surface.
It is also possible to envisage a situation in which the target is brought into contact with an appropriate surface (eg, in the form of a coating on a waveguide) prior to addition of the label and target binding species if necessary.
In each case, the result should be a system in which the label-containing complex is associated with, and ideally sits as close as possible to, the electron pools on the active surface. Features (i)-(iii), described below, all help to optimise
the proximity between the complex and the surface, thus increasing sensitivity.
The SER(R)S-Active Surface
The SER(R)S-active surface may again be any suitable surface, usually metallic, which gives rise to enhancement of the Raman effect, of which many are known from the SER(R)S literature. It may for instance be an etched or otherwise roughened
metallic surface, a metal sol or, more preferably, an aggregation of metal colloid particles. Silver, gold or copper surfaces, especially silver, are particularly preferred for use in the present invention and again, aggregated colloid surfaces are
believed to provide the best SER(R)S effect.
The surface may be a naked metal or may comprise a metal oxide layer on a
metal surface. It may include an organic coating such as of citrate or of a suitable polymer, such as polylysine or polyphenol, to increase its sorptive capacity.
Where the surface is colloidal, the colloid particles are preferably aggregated in a controlled manner so as to be of a uniform and desired size and shape and as stable as possible against self-aggregation. Processes for preparing such
unaggregated colloids are already known. They involve, for instance, the reduction of a metal salt (eg, silver nitrate) with a reducing agent such as citrate, to form a stable microcrystalline suspension (see P C Lee & D Meisel, J. Phys. Chem. (1982),
86, p3391). This "stock" suspension is then aggregated immediately prior to use. Suitable aggregating agents include acids (eg, HNO.sub.3 or ascorbic acid), polyamines (eg, polylysine, spermine, spermidine, 1,4-diaminopiperazine, diethylenetriamine,
N-(2-aminoethyl)-1,3-propanediamine, triethylenetetramine and tetraethylenepentamine) and inorganic activating ions such as Cl.sup.-, I.sup.-, Na.sup.+ or Mg.sup.2+. To increase control over the process, all equipment used should be scrupulously clean,
and reagents should be of a high grade. Since the aggregated colloids are relatively unstable to precipitation, they are ideally formed in situ in the detection sample and the SER(R)S spectrum obtained shortly afterwards (preferably within about 15
minutes of aggregation).
| | |
