NO880010L - ANALYTICAL METHOD FOR DETECTING AND MEASURING SPECIFIC NUCLEIN ACID SEQUENCES. - Google Patents

Description

The present invention relates to a method for determining nucleic acids that contain a particular base sequence. Determination of such nucleic acids, in DNA or RNA that are present in biological samples, is very important in biomedical and genetic studies.

It is currently a clinical necessity to detect specific nucleotide sequences in biological samples, e.g. body fluids, tissue samples such as chorionic villus samples, and swabs. The results from such analyzes can be used to identify and/or detect disorders ranging from genetic abnormalities and suspected cancer to infectious diseases of bacterial, viral and fungal origin. As it stands now, the methods used for such nucleotide detection are not easy to apply in a routine environment, and cannot be performed without specialized staff and laboratory facilities. This is due to the complexity of the procedures, the necessity of hazardous materials (e.g. radioisotopes) and the long incubation times that are necessary to achieve a result.

The desired nucleic acid sequence is conventionally detected by incubation with the complementary nucleotide that is pre-labeled for detection after the reaction. This complementary sequence (or probe) can be easily synthesized using known technology in the field.

The power of the nucleic acid probe technology comes from the specificity of the detection procedure which comes both from the uniqueness of the nucleic acid sequences and the possibility of stringent hybridization conditions that allow discrimination between perfect and not perfect sequence matching. It is possible to discriminate between two sequences present in DNA that differ only by a single nucleotide in the sequence using probes of appropriate sizes and specific hybridization conditions. This allows the detection of the presence of simple point mutations in the test sequence, and where such mutations can be correlated with disease, such as specific cancers and others, these procedures will be useful in the diagnosis of such diseases. Second, the development of automated synthesis systems of specific oligonucleotide sequences has facilitated the construction of probes on base sequences derived from the extensive and rapidly growing nucleic acid databases currently available.

Detection of specific DNA sequences, for example, is a three-step procedure (P. SZABO et al., Trends in Biochemical Sciences (1982), pp. 425-427). First, a hybridization link occurs between the sample, which is usually attached to a solid substrate, and a probe DNA. Second, the non-specifically bound probe DNA is washed away and, third, the specifically bound DNA is determined. Previously, probe DNA has been radiolabeled by incorporating <32>P as an internal label into the sugar-phosphate "backbone", or using <l25>i as an external label incorporated into the aromatic structure of the bases. In this way, autoradiography or isotope counting techniques, the detection systems have achieved a very high sensitivity, on the size of one genome per cell (Hasse E.T. et al. (1982), Virology 119: 339-410). Isotope labeling for quantifying the degree of binding is a well-known technique within immunoassays. However, due to the risks involved in working with radioactive compounds, great efforts have been made during the last ten years to avoid the use of isotopes in immunoassays and, instead, to use enzymes, fluorescent or chemiluminescent labels. Because of the great experience that has been gained in immunoassays, the search for non-isotopic labels for nucleic acids has begun. However, due to the difficult conditions of hybridization tests, it is not possible to use enzyme labels since they would be denatured under the hybridization conditions used today.

Therefore, several methods have been attempted to avoid the current undesirable conditions of radioisotope labels and nucleic acid probes and also the long reaction times and specialized equipment required to visualize <32>P-labeled DNA. Davidson and co-workers have covalently bound biotin to RNA via cytochrome c or polyamine bridges. After hybridization with the sample, the probe is determined by reaction with avidin immobilized on methacrylate spheres. Thus, with the spheres as labels, the hybridization sites can be detected by electron microscopy (J.E. Manning et al., Chromosoma 53 (1975), 107-117; A. Soja et al., Nucleic Acid Research 5 (1978), 385-401). This method has been successfully used in chromosome mapping procedures. In an attempt to create a system of more general applicability, Langer et al (Proceedings of the National Academy of Sciences, USA 78 (1981), 6633-6637) synthesized a series of nucleotide analogs containing biotin covalently bound to the pyrimidine or purine ring. Biotin-labeled derivatives of uridine triphosphate (UTP) and dUTP were prepared using a complex series of chemical reactions. These analogues were incorporated into DNA by in vitro reaction with a number of DNA polymerases. The modified DNA molecule (probe) was then used to bind either avidin or anti-biotin antibodies. Here, both avidin and antibody signal generators were determined colorimetrically using peroxidase-chromogen systems. Although successfully demonstrated, this method has several downsides: first, it involves technically difficult preparations of relatively unstable nucleoside-triphosphate-biotone analogs. Second, the enzymatic incorporation of such analogs into DNA is difficult to control, inefficient, and generally results in a maximum of one to two biothion molecules per probe DNA molecule. The affinity that the biothion antibodies have for biotin (up to 10<8> to 10<10>) is also much less than the affinity for avidin. A further development of such use of nucleotide analogues was demonstrated by Vincent et al., (Nucleic Acid Research (1982) 10, 6787-6796) who, constructing according to a similar method as Langer et al. performed a dinitro-phenol (DNP) nucleotide analog. This analog was successfully incorporated into the DNA at an appropriate ratio of one to two molecules of DNP per probe molecule. Detection of the probe was performed by reaction with an anti-DNP antibody, followed by detection of specifically bound antibody by reaction with an anti-species antibody-peroxidase conjugate. However, it was discovered that the system was not sufficiently sensitive when compared to isotopic techniques. A main reason was that part of the antigen (DNP) molecules were bound within the hybridization complex, and in this way were not available for reaction with antibody.

Recently, improved techniques have been developed for the detection of DNA based on the use of fluorescence signal generators and binding of probe DNA to substrates. FR-A-2,480,943 (WANG et al.; ABBOTT Laboratories) discloses a new class of dyes that fluoresce strongly when in contact with nucleic acids, particularly double-stranded DNA. Without the target nucleic acids, there is no fluorescence in aqueous solutions and relatively little fluorescence with single-stranded polynucleotides (mRNA). Measurements are carried out on solutions using a fluorescence microscope. Concentrations of 1-100 nM DNA can, among other things, be measured using aqueous 3-'y-dimethylaminopropyl-2-p-dimethylaminostyrylbenzothiazodium iodide (2 mg/ml). The fluorescence is linearly proportional to the amount of DNA in the sample.

Document EP-A-57,553 (BIRNBOIM; Atomic Energy of Canada) teaches the use of fluorescent dyes which react with duplex DNA and produce fluorescence, while no fluorescence appears with single-stranded DNA. This technique makes it possible to measure the degree of DNA denaturation due to heat, pH, chemicals, radiation or other denaturing conditions. Suitable colors are ethidium bromide; 4',6-diamino-2-phenylindole dihydrochloride; mitramicin, Hoechst 33258 and others. Fluorescence is measured on the bulk of the solution.

US-A-4,423,153 (D.F. RANNEY et al) relates to a specific DNA fluorochrome that promotes the formation of fluorescence that is directly proportional to the total double-stranded DNA present in a sample. The preferred fluorescence enhancer is mitramycin which binds specifically to duplex DNA. Fluorescence is measured throughout the solution with a fluorimeter. Several suitable commercial fluorimeters are specifically mentioned.

Document US-A-4,257,774 (RICHARDSON et al; Meloy lab.) discloses direct binding of fluorescent intercalators to DNA, e.g. ethidium salts, daunomycin, mepacrine and acridine orange in addition to 4',6-diamidino-2-phenylindole to quantify DNA. Fluorescence polarization is used in tests for the determination of non-fluorescent DNA binders that usually compete with the binding of said dyes, which thus causes a corresponding decrease in fluorescence.

In addition, probes immobilized on supports have been presented. Document EP-A-131830 (DATTAGUPTA et al; Molecular Diagnostics) discloses the labeling of nucleic acids for detection in hybridization for the determination of DNA containing specific base-pair sequences. Where the labeled nucleic acid contains at least one single-stranded base sequence that is substantially complementary to or homologous to the sequence to be detected. The probe can be immobilized on a solid phase such as nitrocellulose, modified nylon and fluorocarbons in the form of filters or sheets. For labeling, enzymes or biothion-avidin systems are provided. Likewise, EP-A-130,523 (DATTAGUPTA et al; Molecular Diagnostics) discloses the immobilization of nucleic acid probes on solid substrates. Substrates with reactive groups such as carboxyl, amino and hydroxy, e.g. cotton, paper, carboxymethyl cellulose, etc. are suitable.

Nucleic acids are bound to the substrates using photochemically reactive nucleic acid-binding ligands, e.g. intercalators such as furocoumarin, angelicin, psoralen and aminophenantridium halides or non-intercalators such as neotropsin, distamicin, Hoechts 33258 and bis-benzlmidazole. Coupling of ligands and substrates is carried out using bridging compounds such as CNBr, dialdehydes and di-glycidyl ethers.

Detection of substrate immobilized fluorescent formers was also investigated. EP-A-70687 (M.J. HELLER; AMOCO CORP.) relates to reactant immobilization in DNA hybridization. An immobilized single-stranded polynucleotide is hybridized with a light-emitting labeled elemental polynucleotide reagent. After separation of the non-hybridized reagent from the immobilized sample, the light signal is excited and the light response detected. The immobilization phases include glass beads, polyacrylamide, agarose, Sephadex, cellulose, etc. The light signals can be chemiluminescent, bioluminescent, phosphorescent or fluorescent. For detection, commercially available detectors can be used, e.g. photomultiplier tubes.

EP-A-117,777 (THANH THUONG NGUYEN et al, CNRS) also discloses specially made oligonucleotides which are covalently attached to an intercalator group such as acridine, furocoumarin and ellipticine derivatives which can be used as nucleic acid sequence detection systems. Measurements based on whole solutions or gel fluorescence are described.

Despite the presentations as mentioned above, further progress was desired, especially with regard to faster and more sensitive tests and avoidance of many of the skilled manipulations associated with the techniques nowadays.

This was achieved according to the method in this invention which essentially makes use of an optical waveguide as probe immobilization phase, where the fluorescent markers that are used to label the polynucleotide probe hybridization products react with the wave energy that passes through said waveguide and give a fluorescent signal that is detected and measured as an outlet of said waveguide. This method is broadly summarized in the attached claim 1. In this claim, determination shall cover detection, identification, quantification, concentration measurements and the like which are carried out on an analyte.

The nucleic acid probe is bound using conventional or new methods to an optically clear phase that acts as a light guide, or waveguide. Single-stranded sample nucleic acid reacts with the probe in the presence of intercalated fluorescent dye. Such dyes intercalate themselves between the strands of double-stranded nucleic acid (i.e. they intercalate). After the intercalation, there is an increase in color fluorescence quantum yield, an increase in fluorescence half-life and a red-shift of the fluorescence emission wavelength that can be optically monitored. The probe is immobilized on the waveguide surface either before or after the hybridization, but only the fluorescent centers of the duplex that are bound to the waveguide provide the fluorescent signal that is used for the determination.

It is worth noting that it will not be possible using conventional transmission photometry and at the same time distinguish between hybridization of molecules attached to the phase surface and hybridization reactions that occur in the solution. In contrast, by internal reflection of the fluorescence excitation light within the waveguide phase, i.e. discriminating the part of the optical signal that arises from the interaction of a short vector with the complex at the waveguide surface-analyte from other optical changes that occur within the solution, the hybridization reaction at the waveguide surface can be measured. Using this method in the preparation of this invention, kinetic measurements of nucleic acid hybridization reactions were performed without the necessity of pre-preparing labeled probes, without the necessity of allowing the reaction to proceed to completion (as monitoring is at the right time) and without the necessity of separating the other substances from the system before the measurement (as only the surface reaction is read).

More details will now be presented with reference to the attached drawings.

In this drawing, fig. 1 schematically shows a hybridization reaction that occurs with polynucleotides that have complementary base sequences, some of which (the probe) are attached to the surface of a phase. Fig. 2 schematically represents a method which is preferable when binding the probe to the phase surface. Fig. 3 illustrates schematically the propagation of a light beam by internal reflection in a waveguide with refractive index ni greater than n2 in the surrounding medium. Fig. 4 i.e. Figs 4A, 4B, 4C and 4D schematically illustrate devices for illuminating a waveguide to be brought into contact with an analyte containing species which forms an optical signal at the interface with the waveguide and for detecting and collecting said signal. Fig. 5 (A & B) illustrates curves which are representative of the measurements made with the devices in fig. 4.

Fig. 6 represents another curve similar to that in fig. 5.

Fig. 7 schematically represents a hollow cylindrical representation of a waveguide. Fig. 8 is a diagram illustrating various embodiments of the method of this invention. Fig. 1 schematically shows a waveguide 1 in contact with an analyte solution 2 containing single-stranded nucleic acid chains 7 and intercalating dye molecules 9. Waveguide 1 is covered with oligonucleotide probes 5 which are immobilized on the waveguide surface using physical or chemical linkers 6 .Nucleotide bases covalently bound to the polynucleotide backbone are numbered 8. Some of the nucleic acid probe strand 7 have a base sequence homologous to probe 5 and they hybridize to form base pairs 5a-7a. Formation of bond between the threads is indicated by number 8a. Other hybrids are also formed within the solution itself as shown in the drawing. When the duplex structures have been formed, the dye molecules 9 intercalate between the chains (9a) and become fluorescent. Normally, a dye molecule intercalates for every fifth base pair in DNA and for every tenth pair for RNA. (J.B. LEPECQ et al., J. Mol. Biol. 27 (1967), 87-106). Excitation light 3 passes through the waveguide by several or individual internal reflections at an angle G greater than the critical angle Øc. This excitation light reacts near the interphase boundary with its short vector component. The degree of penetration of this light component (a fraction of a wavelength) is adjusted to provide the electromagnetic field of maximum strength corresponding to interaction with maximum efficiency with the fluorescent centers of the bound complex; one way to vary this penetration is by changing the angle G as presented in USP 4,608,344. Thus, there are improved sensitivities in terms of surface reactions in thin films when 0 is selected from values close to Gc. Another useful reference in this area is N.J. HARRICK, Internal Reflection Spectroscopy, Wyley Interscience, New-York (1967). The fluorescence resulting from this interaction is then re-injected into the waveguide and can be collected and read as an output thereof in the usual manner. Alternatively or in combination, the fluorescence emitted from the page, e.g. usually (10a) to the waveguide surface can also be collected and measured in some cases, when the waveguide consists of a cuvette wall that holds the reactant solution and the fluorescence 10a is emitted through the wall towards the outside. One of the main features of this invention is that the signal coming from the fluorescence centers bound to the waveguide surface can be distinguished from that coming from the main length of the solution. Therefore, the background effect of the side reactions that occur in the analyte solution independently of the probe that is bound to the waveguide can be effectively deleted. Thus, in this method, only double-stranded material that has been formed at the waveguide surface (i.e., with probe polynucleotide) is detected since the excitation light only penetrates a fraction of a wavelength beyond the solid/liquid interface. The fluorescence response is emitted in all directions (10,10a) but has a preferred angular probability which is close to the incident light reflection critical angle Øc.

Another main factor is the immobilisation of the probe nucleic acid on the surface of the waveguide either before or after the hybridisation. Much technology is available when it comes to the immobilization of biological macromolecules such as nucleic acid on solid phases. These, or modifications thereof, are equally useful in this invention. In the case of an optically clear waveguide certain limitations are imposed where the immobilization technique should not adversely affect the optical properties of the surface. Two main techniques for immobilizing biological compounds to surfaces are physical adsorption and covalent chemical bonding. For an overview see WB Jakoby and M. Wilcheck, (1974). Affinity techniques, Methods in Enzymology, vol. XXXIV, Academic Press, N.Y; Mosbach, K (1976). Immobilized enzymes, Methods in Enzymology, Vol. XLIV, Academic Press, NY; Scouten, WH

(1981). Affinity Chromatography, Wiley Interscience, NY; Meinkoth, J and Wahl, G (1984). Anal. Biochem. 138; 267-284. These techniques have both advantages and disadvantages as they are used today. Physical adsorption to the waveguide surface can result in a large part of the probe molecules not being available for reaction when the nucleotide bases are fixed directly on the waveguide surface. Likewise, when it comes to covalent binding of the probe, current methods rely on a covalent reaction between functional groups on the bases and a reactive group that has been placed on the solid phase. The resulting probe will be attached at multiple sites where many bases are not available for hybridization. An ideal system would attach the probe to the surface where all relevant bases are available for reaction with the sample nucleic acid, and the bulk of the probe should be immobilized within the penetration depth of the exciting light to give a maximum signal after reaction with the sample nucleic acid.

Waveguides with probes attached according to the above considerations are made available in this invention. This is illustrated in fig. 2. Here the terminal nucleotide 12 of probe 5 has been modified (e.g. at its phosphate end) to cause a covalent binding of one end of the probe to an active group 13 on the waveguide surface 3. This group 13 can be NHg in a amino-silane which, after reaction with the probe, leads to a phosphamidate bond. Alternatively, a urethane-phosphamido bond of the formula -CONH-(CH2)n-NH-P is also possible from the reaction of, inter alia, an isocyanato-silane and an alkylenediamine. On the waveguide surface lie the molecules 14 with a positive charge towards the solution 2. Positively charged molecules 14 attract the negatively charged nucleotide phosphate groups 15 in the probe 5. The ratio between the groups 13 and the positively charged sites is adjusted to achieve optimal immobilization of the probe even if the bases 8 are kept free of hybridization, i.e. the equivalent amount of 13 available will roughly correspond to the molality of the probe while the amount of positive centers will largely depend on its average length (average number of bases in the segments). The resulting complex of probe positively charged molecules may involve a probe molecule that is fully or partially immobilized parallel to the waveguide surface with the relevant bases 8 available for reaction with the sample nucleic acids. A presentation of this system will be described in more detail in the experimental section. The unexpected advantage of this method is that by concentrating the DNA probes on such a surface in such a configuration, a fast and sensitive detection method for hybridization analysis has been found. It is also unexpected that this probe:surface molecule complex, once formed, resists displacement by chemical or competitive binding from other nucleic acid molecules. The positively charged centers on the waveguide can be, among other things, quaternary amino groups. Such groups can be obtained, for example, by quaternizing a dialkylamino end group to a dialkyl-amino-alkyl-amine which is attached to the waveguide surface using a quaternizing agent, e.g. an alkyl halide. An example is treatment of the waveguide surface with trihydroxy-"Y-amino-propylsilane and quaternization with methyl bromide.

There is no particular limitation when it comes to the choice of waveguides that can be used in this invention and most types that are now known in this field can be used, such as clear plates, rods, optical fibers and the like. Many of these are described in several publications which also describe the illumination sources required, photodetectors and optical configurations for handling the probes and performing the tests. Fluorescence detection analyzes and quantification of the signals arising according to this method can also be enhanced by the use of photon counting devices (photo multipliers). Differentiation between free and intercalated color can be done by combining the technique of this invention with other known fluorescence techniques, such as fluorescence polarization and/or timed fluorescence. Other techniques that are also useful in this context are "Fluorescence Råman Scatter-ing", Fluorescence correlation spectroscopy, and Fluorescence fluctuation spectroscopy. Significant references in this area are "Modem Fluorescence Spectroscopy (198...) Ed. WEBRY, Heydn, New-York; WO-83011230; USP-3,975,084; 3,436,159; 4,447,548; 3,999,855; 3,604,977.

Although this method is mainly based on the measurement of a fluorescence signal at an outlet of the waveguide, i.e. at one or both ends thereof or laterally thereto, it is not limited to this type of interaction. Reaction of nucleic acids at the interphase in a waveguide can also be detected using other optical phenomena such as signal absorption, scattering and the like, especially since the nucleotide bases have strong optical absorption at some specific frequencies. Therefore, excitation by radiation in the UV, visible and IR ranges is also possible.

Although standard waveguides such as prisms, plates, fibers made of glass, quartz or suitable plastics have been mentioned above, the possibility exists, in this invention, to enhance the optical interactions at the interphase using several techniques which include the use of very thin waveguides called optical cavities which are very thin (one or more) non-conductive layers (of 1-10 microns) placed on another non-conductive material (see the references mentioned before by Harrick, pp. 147-177 and J. DELAY et al, Eur. J. Biochem 12 (1970), 133; V. A. R. HUSS et al., System Appl. Microbiology 4 (1983), 184-192). Hollow waveguides, i.e. capillary and cylindrical waveguides where the analyte is circulated are also possible.

Another potential method for amplifying the signal from such an interphase is to add a thin metal layer (of 500-600 ÅngstrOm) between the immobilized nucleic acid and the waveguide. This technique, with an appropriate choice of optical conditions (e.g. type of metal, thickness of layer, angle of incident light, polarization of incident light rays, wavelength of incident light rays) can generate an electromagnetic field in the optically rarer medium, i.e. the sample, which is up to eighty times stronger than without the metal film, (for references, see H. Raether, (1977). Surface Plasmon Osclllations and Their Applications in Physics of Thin Films, Advances in Research and Development, Eds, G. Haas , and MH Francombe, Academic Press, NY. pp. 145-261; H. Raether (1980). Excitation of Plasmons and Interband Transition by Electrons. Springer-Verlag, FRG; B. Liedberg, C. Nylander and I. Lundstrom, (1983).Sens. Actuators 4,299-304). The surface plasmon resonance technique can be used as a sensitive detector of a change in thickness and the layer at the interface. It may also be a method of enhancing the fluorescence intensity at the surface due to the increased intensity in the excitation light-magnetic fields. Although the above-mentioned publications describe in detail the principles of internal reflection spectroscopic techniques, a brief description will be given as follows with reference to Figs. 3, and obtain a clarity and continuity in this invention.

When a light beam 21 irradiates the interface 22 between two clear media (n^>n2 as shown in Fig. 3A, from the medium with higher refractive index n^23, a total internal reflection occurs when the reflection angle 0 is greater than the critical angle Øc, the value given by :

In this case, the short wave penetrates a distance (dp) of a fraction of a wavelength, behind the reflecting surface and into the dilute medium 26. According to Maxwell's equation, a radiant sinusoidal wave is established, perpendicular to the reflecting surface, in the denser medium (Fig. 3B). Although there is no net energy flow into a non-absorbing, dilute medium, there is a short field in this medium. Due to continuity conditions of the field vectors, the electric field amplitude (E) is largest at the surface interface (EG) and decreases exponentially with distance (Z) from the surface according to:

The depth of penetration (dp), defined by the distance required for the electric field amplitude to drop to exp (-1) of its value at the surface, is given by:

This size decreases with increasing 0 and increases with closer index calculation (i.e. when ng/n^->> 1). Also, because dp is proportional to wavelength, it is larger at longer wavelengths.

Thus, by an appropriate choice of the refractive index n^ of the internal reflection element, of the incident angle, and of the wavelength, one can select a dp to promote optical interaction mainly with compounds close to or attached to the interphase and minimally with the bulk of the solution .

The penetration depth is one of four factors that determine the attenuation produced by absorbing films in internal reflection.

The other factors are the polarization-dependent electric field intensity at the reflective interphase, the sample area which increases with increasing 0 and the matching of the refractive index of the denser medium to the dilute medium which in turn controls the strength of the optical coupling. The appropriate quantity that takes into account all these factors is the effective thickness, de. It represents the actual thickness of the film that would be required to achieve the same absorption in a transmission experiment. To increase sensitivity, multiple reflection elements are often used. The number of reflections (N) is a function of the length (L), thickness (T) of the waveguide and angle of incidence (9).

The longer and thinner the waveguide, the larger N and the more often the short wave reacts with the surface layer of the antibody-antigen complexes. For one reflection, the reflectivity (R) is: where a is the absorption coefficient and de is the effective thickness of a weak absorbing layer, after N reflections the reflection loss is:

i.e. it is increased by a factor N.

The short wave can be advantageously used in this invention to monitor surface reactions using a variety of optical techniques, one of which is described in detail below with reference to Figures 4A, B and C by way of illustration, although this is not limiting.

The apparatus used is schematically represented in fig. 4C which shows as a block diagram the main components; These components include a monochromator 39, a light source 36, a current cell 37 with waveguide 38, a filter 52 and electronics with data acquisition and processing microcomputer including a photomultiplier detector 40, a pre-amplifier 41, a microprocessor light source control system 42, a micro-computer 43, a printer 44, and a "memory" (e.g. floppy disk) 45.

To collect the fluorescence at a waveguide/liquid interface, different signal collection techniques can be used (Figs. 4A and 4B). Fluorescence emitted at an interphase can be detected either conventionally where detector 54 is positioned at right angles to the interphase (Fig. 4A), and/or, like detector 40, in line with the primary light beam (Fig. 4B). Looking at the very small fixed emission angle in in-line detection compared to the emission angle in right-angle geometry detection, it appears that the former would not be very efficient. But, there is an amplifying effect, and the theory assumes that for a fused silica waveguide with water as n2 medium, the in-line fluorescence intensity will be 50 times higher than the fluorescence emitted at right angles to the waveguide. This effect, that the fluorescence is fed back into the waveguide, has been verified both theoretically and experimentally (eg C.K. Carniglia et al., I. Opt. Soc. Amer. 62 (1972); 479-485 ).

In the first step, an incident plane wave forms a short wave that excites the molecule near the surface with a local distribution proportional to the electric field intensity (see equation 2). After a characteristic excited-phase lifetime, these molecules emit fluorescent radiation with a local distribution near the surface that is very similar to the exciting intensity distribution described by equation 2 i.e. to a short wave, but at the fluorescent wavelength. The question of what happens to the fluorescent short wave can be answered using the optical reciprocity principle, which states that this light is coupled back to the waveguide as a plane wave in the same way as the primary process when a plane wave generates a short wave. The theory shows that the fluorescence intensity emits peaks at the critical angle of total internal reflection so that it can be reflected in the interior. To increase sensitivity, the internal reflected elements can be structured to collect fluorescence light from multiple reflections and direct it to the detector.

This is particularly advantageous when an optical fiber is used as an internal reflection element, as fluorescence light emitted at right angles from an elongated fiber cannot easily be collected in a detector. In-line detection avoids the measurement of fluorescence through the bulk of the sample solution surrounding fibers, which can otherwise give a significant interference, depending on the fluorescence dye used.

Light source 36 in this preparation was a Xenon flash lamp (E.G. & G. Salem, MA, USA) and the monochromator was equipped with a concave halographic grating (JOBIN-YVON, Paris) allowing a resolution of 5 nm. Use of the flash lamp was controlled by micro-computer 42. To inject the samples through an inlet 48 into cell 37, a programmable automatic pipette (Microlab-P; Hamilton Bonaduz AG, Bonaduz, Switzerland) was preferably used. The optical component further included two mirrors M1 and M2 and two prisms 46 and 47. A photomultiplier tube for detector 40 (R928; Hamamatsu, Tokyo, Japan) located at the outlet of the waveguide directly monitors the change in light intensity. Alternatively or simultaneously, a detector 54 with filter 53 provided a signal from right-angle fluorescence emulsion (see Fig. 4A). The signals from the photomultiplier tubes were amplified (41), integrated over the flash time (42) and transferred by a standard 12-bit analog/digital converter (not shown) into a digital format. The I-house developed microcomputer 42 performed fixed signal averaging calculations, and all data was adjusted for variations in flash lamp intensity by reference to a photodiode 49 placed in the monochromator. The signals were transmitted to a micro-computer 43, preferably an APPLE II model, for display and storage.

The analytical cell or cuvette illustrated in Figure 4C is based on a microscope slide waveguide system. The illustrated system shows current cell 37 where the bottom is actually the microscope slide 38. Firmness is ensured with a sealing device 50; the slides 38 were placed in direct optical contact using index-matching oil with two quarter-round silica prisms 46 and 47, preferably from Heraeus. Index matching oil, made specially polished, optically flat waveguide surfaces unnecessary. The prisms were stored to allow easy adjustment of the incident light angle 0 (see Figure 3A) and to prevent light contact with the closing seal device 50.

The flow cells, made from aluminum alloy, met the criterion and allow fast, bubble-free laminar flow along the light path. The design also ensured quick and careful disassembly and repositioning. We chose an aluminum alloy, although other metals are also suitable, e.g. brass, due to its good thermal conductivity, lack of reactivity with saline solution, and low optical reflectivity after being anodized matte black to prevent light-glow effects. Sealing device 50 was 0.5 mm thick medical type silicone rubber and watertight at a constant sealing pressure of 2 kg/cm<2>. Including the inlet 58 and outlet 51 channels, the total cell volume was 1.8 ml, the volume directly above the waveguide was 0.66 ml (53 x 25 x 0.5 ) and the volume above the light path was 0.29 ml (36 x 16 x 0.5). The apparatus described above is used as follows: In the first step, waveguide plate 38 is covered with probe DNA using the technique described according to fig. 2 and described in detail hereafter, and then the covered waveguide which forms the bottom of the analytical cuvette 37 is installed and after attaching the current-carrying and detecting elements, an analyte solution (containing nucleic acid fragments and a dye) is added to the cell whereupon hybridization between complementary polynucleotide strands occurs. The color molecules are intercalated between the duplex threads and form fluorescence upon excitation. Fluorescence formed at the waveguide interface is therefore emitted at the outlet of the waveguide and collected by detectors 40 or emitted at right angles to the waveguide and collected by detector 54 after filtering out the remaining excitation component with filter 52 or 53, respectively. The corresponding signal in detector 40 is then processed by the remaining components 41 - 45 of the apparatus to produce the desired results. The degree of increase in fluorescence is a measure of the degree of hybridization and therefore depends on the concentration in the sample of DNA complementary to the probe and its affinity thereto. Fixed time and endpoint measurements may also be performed to provide similar or related results. Standard tests can be run to calibrate the equipment with known samples consisting of complementary DNA and then used for the determination of unknown samples. This equipment can also be used in cases where the probe is attached to the waveguide after hybridization with polynucleotides to be determined, the only main difference being that the reactants are allowed to react before the mixture is introduced into the cell.

Fig. 4D illustrates an alternative apparatus for detecting and monitoring the fluorescent light that is returned through the waveguide and excited therefrom on the input side. The working components of this embodiment are almost the same as in the first embodiment and include a source 57, a filter 58, a wavelength 59, a reference detector 61, a fluorescence emission filter 62 and a signal detector 63. It additionally includes a beam bottom splitter 60 which reflects the incident the wavelength from the source towards the waveguide.

Here, excitation light generated by source 57 and filtered through filter 58 is reflected and light input time 64 to waveguide 57 via optical beam splitter 60. Fluorescence light generated at the interface of the waveguide and the analyte solution (not shown) is fed back to the waveguide and excites from page 64. With the correct selection of the optical design of beam bottom splitter 60, a large part of this fluorescent signal is sent directly through the beam bottom splitter to be detected after filtering (62) by detector 63.

An advantage of this setup compared to that illustrated in fig. 4C is that the excited light is not directly incident on sample detector 63, so that background signal levels are reduced. A suitable beam bottom splitter can be a single quartz slide or a more complex dichroic laminate. The latter is preferable as it selectively reflects the excitation light into waveguide 59 and transmits the fluorescent light to signal detector 63.

In a variant of the preparation in fig. 4D, planar waveguide 59 can be replaced by a hollow waveguide 70 (see fig. 7) where the liquid to be tested is circulated by inlet 71 and outlet type 72. This hollow waveguide with coating 70a can consist of a capillary tube where the inner area can be very large relative to its internal volume where the arrangement provides a high density in terms of interaction sites between the short wave signal and the fluorescence centers of the hybridized material represented conventionally by cover 73. The positions of inlet 71 and outlet channels 72 are optional. Into fig. 7, the ends of the tube, which acts as a test cuvette, are blocked by the plugs 74 and 75; any other means of keeping the liquid in the tube and preventing it from leaking at the ends may be used.

Operation of this waveguide varlant is approximately the same as that illustrated in Fig. 1. 4D. The input light shown by the arrows 76 is propagated through the walls 70 of the waveguide and the short wave component reacts with material 73 and forms a fluorescent signal schematically shown by the dotted arrows 77. This fluorescence is picked up by a detector similar to detector 63 1 fig. 4D. As for the rest, the layout and operation is exactly the same as shown in fig. 4D.

The advantages of the tubular waveguide are that it allows circulation of a relatively large volume of solution (which is held in a reservoir, and not represented in the drawing) or recirculation of the same sample to ensure maximum uptake of all analyte molecules by the probe which is covered on the inside of the pipe wall.

Therefore, even in those cases where the nucleic acid to be detected by hybridization is very dilute, it will progressively attach to the probe as circulation continues and sensitivity will be greatly increased.

The dyes used as intercalators have been characterized and investigated and their use in measuring double-stranded nucleic acids is known within this field. Lists of suitable colorants are provided in EPA 84107624.3 and FR-A-8107928. Of particular interest in this application are phenanthridine dyes (eg, ethidium bromide and ethidium bromide homodimers) and acridine dyes. An overview of some of the most important optical and structural properties is given in RH Sarma, (1980), Nucleic acid geometry and dynamics, Pergamon Press, London. Other key references are J-B LePecq and C. Paoletti, (1967). J. Mol. Biol. 27,87-106; J-B LePecq, M. LeBret, J. Barket, and B. Roques, (1975). Pro. Nati. Acad. Pollock. (United States). 72, 2915-2919; J. Olmsted and DR Kearns,

(1977). Biochem. 16, 3647-3654., M. LeBret and 0. Chalvet

(1977). J. Mol. Struct. 37, 299-319.; YEAR. Morgan, DH. Evans, JS Lee and D. Pulleybank, (1979). Nucleic Acids Res. 7, 571-594.; J. Markovits, B-P Roques and J-B LePecq, (1979). Anal. Biochem. 94, 259-264. The most important factors regarding these dyes are changes in the spectral properties after the dye has intercalated between double-stranded nucleic acids. After the intercalation, there is an increase in the fluorescence lifetime, the fluorescence quantum yield is increased and there is a shift to red in the fluorescence emission wavelength. With an appropriate choice of optical system, the fluorescence from intercalated dyes can be easily differentiated from non-intercalated dyes, e.g. when choosing an appropriate emission filter. The fluorescence signal is directly proportional to the concentration of double-stranded nucleic acid (where the dye is in excess with regard to concentration). These are the main points in the U.S. patent number 4,423,153, a method for the detection of total cellular DNA.

Such dyes have been shown to react with double-stranded DNA from many species (eg, calf thymus DNA, mouse phage T4, E. coli, P. vulgarls, M. lysodelcticus, human DNA, various RNAs, and homopolynucleotides such such as polydArpolydT, polydCrpolydG (from the above-mentioned references)) and is therefore universally usable.

The interaction of certain dyes (eg, Acridine orange) with double-stranded nucleic acid can be inhibited by other intercalators (eg, Actinomycin D). This is the basis of a test system for the detection of compounds that react with nucleic acids described in U.S. Pat. patent number 4,257,774.

Since nucleic acids must be in a double-stranded form for intercalation to occur, breakage of this double-strand, as occurs with low doses of radiation or certain chemicals, can result in a lower fluorescence signal compared to control samples that have not been exposed to harmful substances. These are the main points of EPA 82300376.01 which is a method for the detection of damaged DNA.

None of these methods can be used for the detection of specific nucleotide sequences in a mixture consisting of many nucleic acid molecules. This is the main purpose of this invention.

The use of photochemically activated intercalating compounds for the synthesis of labeled probes has been described in EPA 84107624.3. Here, the photochemically active intercalator is bound to the nucleotide and either the color fluorescence is used to detect the probe after hybridization or the intercalator is used as a "bridge" to bind other substances to be detected as labels, e.g. enzymes, avidin-biotin, antibodies. This method is aimed at conventional hybridization assay systems and is not related to waveguide, no-separation/no-wash/kinetic hybridization assays and the previously mentioned techniques require a pre-labeling of the probes before the reaction, and not an in situ labeling as in this method.

It should be mentioned that it is possible to form double-stranded DNA at an interphase as presented here, and then label the double-stranded DNA with an intercalator which will cross-link the strands. This allows the immobilized material to work without loss of specific signal with a further increase in sensitivity due to minimal background signals. It may be necessary when very high sensitivity is required. One can also imagine a reaction sequence where more than one probe is used. For example, a type of sandwich hybridization assay (M. EADOKI et al., Gene 21 (1983), 77-85) where sample DNA reacts with the immobilized probe and a second (or more) fluorescently labeled probe(s) in solution. The resulting complex can be measured at the interphase using the techniques of this invention. The advantages of such a system are that it allows the use of "generic" solid-phase probes suitable for a number of selected applications while the specific probe can be added to solution, thereby precluding the necessity of fabricating a variety of solid-phase probe waveguides .

This invention is different compared to previous types and has a number of unique properties that make it very useful for performing analyzes of a clinical diagnostic nature, and which do not exist in other current systems. These properties consist in particular of: rapid detection of defined nucleic acid sequences in the kinetic target system; the technical ease of use of a kit/instrumental system upon which this invention is based due to the need not to pre-label reagents; no need to wash the solid phase and in situ detection of the hybridized probe; the sensitivity due to enhanced fluorescence of the intercalated dye and due to the use of internal reflection detection techniques; the specificity due to the nucleic acid probe and the use of internal reflection to detect only the thin surface film and no other extraneous substances.

For on-site use (e.g. clinical laboratories, agricultural test laboratories, food technology, raw material quality control, etc.) there is a kit system in this invention. Such a kit contains the following components: a test device (syringe, swab, etc.); a sample preparation device for deproteinized solutions consisting of single-stranded nucleic acid; the kit also contains a nucleic acid hybridization system according to the method of this invention. Depending on which production is used, the system consists of a simple device with a low production capacity or a large multi-analyte, an analytical equipment for multi-samples, manual or automatic. In both cases the waveguide is part of a cheap disposable cuvette (e.g. molded plastic) probe nucleic acid is attached to parts of the inside of the surface. All reactions take place within the cuvette and the optical measurements involve the covered parts of the cuvette. When it comes to more sophisticated production, it is appropriate to attach the probes to a device that can be used again where the analytical unit is of the current cell type (see fig. 4C) with the waveguide as an integrated part. In the case of DNA, a series of solutions is sent through the cell for measurement of base-line signal, specific binding signal after injection of the sample and dye, then breaking of the double-stranded DNA and upon equilibration of the system which is then ready to receive the next sample . Breaking of duplex DNA can be achieved using common chemical or physical methods. The simplest way to denature duplex DNA is to heat the sample at a temperature that depends on the relative base-pair concentration, the length of the hybridized molecule and the goodness-of-fit of the two fused strands. There is an average denaturation temperature (Tm) at which the two strands separate. The Tm value can be used to separate strands when ready for the second test, but a temperature control device incorporating the features of this invention has a number of advantages.

The relationship between base-pair sequence and Tm is defined with the following equation:

where G and C are the equivalent ^-concentrations of guanine and cytosine, respectively, and X is an ionic strength of the buffer. This formula may need to be modified in the presence of certain polymers, e.g. polyethylene glycol, due to exclusion properties of such polymers (see M. GILLIS et al, European Journal of Biochemistry 12 (1970), 143-153). As a general rule, the maximum efficiency of nucleic acid rehybrldization occurs at reaction conditions at moderately elevated temperatures (> 30°C) and the maximum rate of the reaction between complementary strands occurs at a temperature approximately 25°C below Tm (J. MARMUR eta 1 ., J. Molec, Biology 3 (1961), 585-600). The manner of achieving this method involving temperature controlled hybridization and denaturation can easily be optimized to operate at maximum binding efficiency by selecting appropriate temperatures according to the above relationship.

Another purpose of temperature control is related to the goodness-of-fit between two nucleic acid strands. By using a temperature gradient through the cell, an accurate measurement of Tm can be performed when subjecting the hybridized molecule to non-fusion and kinetic monitoring. Tm changes when "goodness-of-fit" becomes smaller, i.e. when they fit together less well (R.B. WALLACE et al., Nucleic Acid Res. 6

(1979), 3543-3557). The device incorporating the technique of this invention leads to an accurate measurement of Tm, comparison of the observed Tm in a sample of DNA with an expected Tm and thus determining the "goodness-of-fit" of the DNA strands of interest. This is very important in the detection of partial base-pair homology! which may exist related microorganisms, or by the detection of single-point mutations that are related to certain hereditary diseases. Other techniques involving goodness-of-fit to which this invention relates include the following: Ionic strength variations (B.J. McArthy et al., Biochemical Journal 2 (1968), 37-48). Formamide concentration (B.C. Mc CONAUGHY et al., Biochemistry 8 (1969), 3289-3295). Other chaotropic agents such as sodium periodate (K. HAMA-GUCHI et al., J.A.C.S. 84 (1962), 1329-1338); sodium perchlorate, sodium thiocyanate and potassium iodide (D.R. ROBINSON et al., J. Biological Chem. 241 (1966), 4030-4042), dextran sulfate (G.M. WAHL et al., Proe. of the Nat. Acad. Sc. USA 76

(1979), 3683-3687. Recently, a significant hybridization rate increase has been demonstrated using some polyethylene glycol in the reaction buffer (R.M. AMASINO et al, Annals Biochem. 152 (1976), 304-307). The optimal conditions for hybridization buffers will also take into account what is mentioned above.

In the reactions that immobilize nucleic acid (single-stranded probes or duplex structures) on waveguide surfaces (glass or plastic), three main ways are possible: a) immobilization of the probe, i.e. known polynucleotide sequences, on the waveguide and then put the probe on the waveguide

react with an analyte of DNA to be determined.

b) immobilizing the hybridization product with specific binding sequence on the probe and c) immobilizing the DNA mixture to be analyzed on the waveguide and allowing the DNA to react with a solution containing known probe DNA. Techniques (a) and (b) are particularly important in this invention.

To achieve immobilization of DNA on a substrate, the substrate can be nucleophilic, i.e. the substrate will then bind electrophilic nucleic acid centers (electron-deficient sites, e.g. C=0 to activated carboxyl esters, or phosphate esters). The substrate can also be electrophilic (among other things positively charged) and then attracts nucleophilic nucleic acids (e.g. phosphate). This is the option illustrated in fig. 2 where duplex DNA lies flat on the substrate surface after dye intercalation, although the probe is only covalently immobilized at one point (3' or 5' end). Anchoring at just a single point increases reactivity since the chain has the opportunity to twist freely and twist up in solution. Orienting the fluorescence complex parallel to the waveguide interphase increases the interaction of the fluorescence centers with the short wave component and increases test sensitivity.

Attaching the probe to only one end requires the development of special chemistry.

Some techniques have been reported for binding unmodified DNA to cellulose or agarose. The following references are indicative: M.S. Poonian et al.; Biochemistry 10 (1971) 424-27; P.T. Gilham Adv. Exp. Biol. With. 42 (1974), 172-185; P. Venetianer et al.; P.N.A.S. USA 71 (1974) 3892-3895; DJ Arndt-Jovin et al., Eur. J. Bioch. 54 (1975), 411-418; T. Y. Shih et al., Biochemistry 13 (1974), 3411-18; B.E. Noyes et al., Cell 5

(1975), 301-310; M. L. Goldberg et al., Methods in Enzymol. 68

(1979), 206-220; LG Moss et al., J. Biol. Chem. 256, (1981), 12655-58. Poor yields have been reported in the case of solid phases bearing either nucleophilic or electrophilic groups. Good results have been described using strong basic conditions with an electrophilic solid phase (see MOSS above). But multiple conditions cannot be used with labile linkers on discs. Modification of DNA/RNA or at least activation of nucleic acids was also necessary to obtain good results under mild conditions.

The waveguide surfaces were, among other things, silanized with trihydroxy or trihydroxysilanes with reactive groups such as amino, SH, NCS, NCO, epoxy, azide, CHO, halide, carboxy, ester and the like. Silane compounds bearing 3-aminopropyl, 3-thiohydroxyalkyl, glycidoxypropyl can be used in silanation; coupling of the amine with an aryl or alkyl diisothiocyanate then forms NCS-terminated moieties. Polylysine can then be bound to the free NCS to increase the number of nucleophilic amino groups on the waveguide surface.

Probe DNA was synthesized by usual techniques, including methods suitable for the formation of oligonucleotides. See, among other things, S.L. Beaucage et al., Tetrahedron Letters 22

(1981), 1859-1862; M.H. Caruthers, Chemical and Enzymatic Synthesis of Gene Fragments, Ed. H. G. Gasser & A. Lang (Verlag Chemie, Weinheim, FRG) (1981) pp. 71-79.

In one embodiment, an oligo-(dC)-3'-methylglucoside of approximately 25 nucleotides was immobilized, after oxidation with periodate, on a glass plate that had been silanized with an aminopropylsilane in a cyanoborohydride solution.

According to another embodiment, polyribonucleotides can also be immobilized using the same procedure. Polydeoxy-ribonucleotides can be modified enzymatically at the (3') end by the addition of a ribonucleotide using terminal deoxy-nucleotidyl transferase and ribonucleoside triphosphates (see R. ROYCHOUDHURY et al., Methods in enzymology 65 (1980), 43-62 ). Then, as mentioned above, periodate oxidation and reductive condensation lead to efficient immobilization of DNA on a silanated waveguide surface.

Synthesis of oligonucleotides with an amine group at the 3' end was also performed. Thus, oligo(dC)-3')-methylglucoside (25-mer) oxidized with periodate as described previously was purified chromatographically and bubbled with a 1,6-diaminohexane in tricine buffer. Then the product was reduced with cyanoborohydride, purified by H.P.L.C. and immobilized on an isothiocyanate grafted waveguide glass plate.

Modification of nucleic acids at the 5'-terminal end was achieved using techniques involving activation with imidazole or analogs under weakly acidic conditions. This technique was used analogously to the production of B.C.F. CHU et al., Nucleic acid res. II (1983), 6513-29; P.N.A.S. USA 1982, 963-967. It was successfully performed with both immobilized short synthetic DNA chains and long restriction fragments.

Oligonucleotides, e.g. dC (30) prepared using the solid-phase method on a controlled pore glass carrier was phosphorylated with ATP and T4 polynucleotide kinase, where the reaction was followed using a 1/10 aliquot labeled with "γ-(32P) ATP. Phosphorylated oligonucleotide was activated with p<->imidazole in the presence of a carbodiimide, then it was bound to the surface of a waveguide plate that was silanized with an aminopropylsilane at pH 8.5.

Duplex DNA was also applied to a waveguide. Phage X DNA was fragmented with appropriate endonucleases and the ethanol-precipitated fragments were activated with Imidazole and coupled to aminopropyl residues grafted onto a waveguide plate using a carbodiimide. This construction demonstrates the use of the invention in preparation (b) mentioned above when also with regard to optical fluorescence response when adding an intercalator dye. For carrying out the optical measurements according to the invention, the set-up which is illustrated in fig. 4 used. Waveguides that were used consisted of flat glass or injection molded plastic microscope slides. The angle of incidence was controlled using mirrors Ml and M2. The solution to be analyzed was pumped into current cell 37 and the fluorescence response formed at the interphase was detected either at a right angle (53) or at an outlet in the waveguide (40). Filter 52 blocked the remaining emission wavelength. The signal generated at detector 40 (or 53) was then processed by the electronic elements 41-45 and detected according to conventional methods.

In addition to testing the sensitivity by controlling the actual surface density of probe DNA on the waveguide surface, other sensitivity increasing techniques were made available according to prior techniques in DNA detection in the solution range.

Among other things, according to the invention, it is entirely probable that the size of the signal from multiple internal reflectance photometry depends on the concentration of the fluorescence generator sites within the space available for the short wave component. For a given intercalator dye, this response is a function of the total amount of duplex DNA within that region (assuming the linear density of intercalating sites remains constant).

However, duplex DNA present need not necessarily be limited to the hybrid comprising the original probe DNA associated with complementary fused test DNA. There are methods that supplement the previously formed duplex structures, these methods include, among other things, extension of the existing polynucleotide probe chains and hybridization of single-stranded DNA that is complementary to the extended sequence; another approach would be to add double-stranded hybrids.

One method includes supplementing the reaction medium including probe hybridized materials with DNA that is complementary to the test DNA, i.e. to the part of the test DNA sequence that is not homologous to the probe DNA. This leads to an increase in the amount of base pairs available for the fluorescent site formation within the optical region at the waveguide interface.

In a variant, the supplemented DNA can be obtained in prehybridized form which will increase available chain lengths and fluorescence site concentration. Further supplementation can be achieved by network formation by alternating complementary and non-complementary sequences.

Another method is to obtain several fluorescence sites that previously labeled complementary DNA, where the labeling has been carried out with fluorescent complexes, such as FITC; thus, if fluorescent nucleotide substitutions are used in the synthesis of complementary DNA that react as indicated, this will increase the total fluorescence within the optical range of interest.

Regarding the second variant b) above (see page 21), where the probe reacts with analyte DNA in the solution before the duplex is bound to the waveguide, the main advantage is an increase in hybridization kinetics. Here, the probe has its immobilization sequence (among other things part of an immuno-pair complex) attached to the non-hybridizing end and the waveguide is covered with the respective binding partner. Examples of several possible preparations are given in the above-mentioned references and include the following specific cases: i): - Adding a polynucleotide to the (3') end of the probe and covering the waveguide with the complementary polynucleotide. 11): - Connecting one (or more) parts of an immunological binding pair to the probe and covering the waveguide with the immunological binding partner (e.g. biotin probe and anti-biotin antibody on the surface).

ili): - Coupling of one (or more) parts of a natural or modified binding pair to the probe and its respective partner to the waveguide (eg biotin probe and avidin on the surface).

Practical details are now given in the experimental section as follows:

Experimental part.

1. Preparation of waveguide surface for binding DNA.

a) Use of nucleophilic sites, e.g. grafting with "γ-amino propylsilane: Waveguides in the form of glass or plastic molded microscope slides were rinsed in cleaning solution (glasses were etched for 30 minutes at room temperature in F3C-COOH) and under argon placed in a 2% by weight toluene solution consisting of 3-aminopropyltriethoxysilane overnight at 100°C under argon. Then the plates were washed with ethanol and water and dried under vacuum. The seeding density was approximately 10"<10> to 10"<9>mol/sq cm but could also be higher .

The grafted plates were stored under argon before use in acetone solution.

a2) Grafting with -γ-mercaptopropylsilane.

The above-mentioned procedure was used with the exception that amine silane was replaced by -γ-thiohydroxypropyl-triethoxy-silane. The grafting density was approximately as mentioned above. b) Formation of electrophilic sites, e.g. grafting with isothiocyanate.

Plates obtained as under (a) were dipped for 2 hours in a 5% by weight dry DMF solution consisting of 1,4-phenylenediisothiocyanate. The plates were then washed with 1 DMF, dried with ether and stored under argon. c) Covering waveguides with polylysine to increase the amine density on the waveguide.

A waveguide plate grafted with isothiocyanate groups as prepared in (b) was reacted for 3 h in 0.1 M Hepes buffer (pH 7.6) with an excess 10-<3>M polylysine solution (Mrø~~4000 ) at room temperature. The plate was washed with methanol and water, dried and stored under argon before use. The coverage density was approximately 50 picomoles of lysine/cm<2> of waveguide surface.

d) Formation of positively charged groups on the waveguide.

Plates from (a) or (c) above were quaternized with methyl bromide or methyl sulfate according to standard procedures. This quaternization reaction involved approximately 10 to 40% of the available amino groups. Alternatively, binding of quaternary ammonium compounds to the waveguide surface can be achieved by treating the surface with an aqueous solution consisting of a silane of the following formula (MeO)3SI-(CHg )3~NR3 where R is a lower alkyl, e.g. Me or Et. A silane of this type is commercially available. 2. Modifications of nucleic acids for binding to the grafted waveguide.

a) 3'-end modification of oligonucleotide with a terminal sugar or ribonucleoside - synthesis of CPG carrying a

immobilized methyl glucoside group.

Methyl α-glucopyranoside (1.94 g, 10 mmol) was etherified with a dimethoxytrityl protecting group (DMT) in the usual manner using the corresponding chloride (4.2 g, 11 mmol) in pyridine (20 mL) and DMF (30 mL). The resulting methyl 6-O-dimethoxytrityl glucoside (3.3 g, 62$) was reacted in pyridine containing dimethylaminopyridine (1$) with succinic anhydride

(1.8 g, 3 equivalents), and the product was purified by flash chromatography on silica gel (eluted with CHCl 3 /MeOH 80/20 with 1% pyridine). After evaporation of the eluate, the acid was taken up with pyridine and evaporated to eliminate residual methanol; yield 2.5 g. The succinylated sugar was dissolved in dioxane (12 mL) containing pyridine (1.5 mL), DCC (0.88 g, 4 mol) and p-nitrophenol (0.56 g, 4 mmol), and this solution was shaken for two hours at room temperature. Dicyclohexylurea was removed by filtration and the solution was immediately used for coupling onto controlled pore glass (CPG) beads [CPG/long chain alkylamine (LCAA); pore diameter: 500 Angstroms; black 24875 from Pierce Inc.] CPG (5 g) was quickly washed with a solution of diisopropylethylamine (5% in ether) and suspended in dry DMF (30 mL). p-nitrophenylsuccinylated sugar derivative was added and the suspension was left at room temperature with gentle shaking. After washing with DMF, remaining free amine and hydroxyl groups were added to acetic anhydride (1 ml) 1 pyridine (7 ml) containing dimethylaminopyridine (1%). After washing (DMF), the CPG beads were resuspended in dry pyridine (10 mL) containing DCC (0.6 g) and p-nitrophenol (0.4 g) and shaken gently for six hours. Morpholine (0.4 ml) was then added and the suspension was shaken for a further 15 minutes.

After washing and drying, the solid phase was analyzed for trityl cation: 19 pmol of sugar was immobilized per gram of CPG.

Chemical synthesis of oligonucleotides containing a (3')-modified end.

CPG (53 mg; 1 pmol sugar) was placed in the bottom of a 2.5 ml glass syringe with a macroporous polyethylene filter, and the oligonucleotide was synthesized in this device by the phosphoramidite method presented by S.L. Beaucage et al. Tetrah. Letters 1982, 22-1859.

The open syringe of CPG was placed in a test tube and rinsed with dichloroacetic acid (2%) until (5')-end deprotection was complete. After washing with acetonitrile and DMF, the syringe was reassembled and the solid phase was dried with three subsequent washes by suction in dry acetonitrile through a septa using a needle attached to the syringe. CPG was suspended in dry acetonitrile (1 ml) and the solution was applied to a 1.5 ml plastic Eppendorf tube with dry deoxycytosine nucleoside phosphoramidite (16 mg; 20 equivalents). This dissolved building block was then transferred to another 1.5 ml plastic Eppendorf tube containing tetrazole (14 mg; 200 equivalents). Then the dissolved activated phosphorus amldite was sucked into the syringe. And the syringe was rotated on a mechanical rotating device for three minutes. After washing with 2,6-lutidine, the unreacted material was added to a solution (1 ml) of acetic anhydride (5%), lutidine (2056), dimethylaminopyridine (10#) in THF for two minutes.

Then the syringe plug was removed and all remaining oxidation, washing and acid treatment steps were performed with an open syringe attached to a needle in a septa covering a vacuum Erlenmeyer flask.

Iodine solution (2 mL) containing iodine (1.7 g), THF (40 mL), 2,6-lutidine (20 mL) was added to the flask. After 30 to 45 seconds, the CPG was washed with acetonitrile and dichloromethane. After a new treatment with dichloroacetic acid (2$), the solid phase was ready for a new addition of a new building block.

Oligonucleotides were made with methylglucoside- and nucleoside-derived carriers using the described technique. After each cycle, a 20-fold excess of the desired nucleoside phosphoramidity was used, and the cycle was repeated as many times as necessary. These synthetic steps can also be carried out using genetic engineering.

Release and deprotection were performed as described in M.H. Caruthers, Chemical and Enzymatic Synthesis of Gene Fragments, GASSER & LANG (Verlag Chemie, Weinheim, BRD; pp. 71-79 (1982) and in Oligonucleotide Syntheses, A Practical Introduction, M.J. GAIT Editor (IRL Press, Oxford, pp. 35-81 (1984).

(3')enzymatic modification of oligo-d(C).

Oligo d(C)-mer) was prepared with protected -d(C) derived CPG and purified by gel electrophoresis (20$ urea gel) and BPLC on reversed phase C-8 column (25 cm) after complete deprotection.

Oligo d(C) (40 µl), 10x transferase buffer (20 µl), rCTP (2 nmol) and terminal transferase (40 units) in a final volume of 200 µl were incubated for 6 hours at 37 degrees. After treatment with EDTA (0.5M; 20 µl) and extraction with phenol/chloroform as usual, the resulting nucleic acid was desalted on a Sephadex G-50 column (Pharmacia Fine Chemicals) and detected by UV absorption (solution: 1 mM Tricine buffer).

Periodate oxidation of oligo d(C)-(3')-methylglucoside or oligo d(C)-(3')-oligo rC.

Oligo d(C)-(3')-methylglucoside (25-mer) (10 Gpug in 100 µl) or oligo d(C) (30-mer) cytidinated at the (3') end (10 µg in 100 µl) was reacted with sodium metaperiodate (1$ solution; 1 µl) at 4 deg. After 30 minutes NaCl (4M; 100 µl), Na2HPO4 (0.5M; 100 µl) and glycerol (1# solution; 10 µl) were added. After repeated incubation at 4 degrees for 30 minutes, the solution was freeze-dried and the residue desalted on a Sephadex G-50 column as described previously. Finally, the nucleic acid solution was adjusted to 1 ml.

Immobilization of oligo d(C) on aminopropyl glass plate.

An aminepropyl-derived glass slide was covered with periodate oxidized oligonucleotide solution (500 µl; 3-5 µg) in Tricine buffer (pH 7.0). It was incubated for three hours with occasional shaking. Then cyanoborohydride (1% solution; 1 µl) was added, and the reaction was allowed to stand for two more hours. After extensive washing with water, the plate was placed under argon at 4 degrees until use.

The previously mentioned technique was also used to fabricate oligo d(A) covered waveguides.

Immobilization of nucleic acid containing nucleofilamine at its (3') end.

Oxidized oligo d(C)-(3')-methylglucoside (25-mer) (about 10 µg) was desalted as usual, and half the solution was reacted with 1,6-dlaminohexane (1 solution; 1 µl) for two hours at room temperature in Tricine buffer (0.1 M; pH 7.0). Then cyanoborohydride (1% solution; 1 µl) was added and after two hours at room temperature the compound was purified on H.P.L.C. on a C-8 column in ammonium acetate buffer (0.1M; pH 5.5). Immobilization was performed on an isothiocyanate-derived glass slide in borate buffer (0.1M; pH 9.5) for two hours at room temperature.

b) (5')-end modifications of nucleic acid.

The direct condensation of (5 *)-phosphorylated nucleic acids

with the presence of soluble DCC and a "nucleophilic" solid phase is usually poor. However, upon activation with imidazole or analogs at pH around 6, the condensation is more efficient and specific, as reported by Chu B.C.F. and Orgel L.E.; P.N.A.S 1985, 82, 963-67 and Chu B.C.F. , Wahl G.M., Orgel L.E.; Nucleic Acids Research 1983, 11, 6513-29.

This method has been used to immobilize short synthetic DNA chains and long restriction fragments. Activation and immobilization of synthetic oligo d(C) (30-mer) at its (5') end.

Oligo d(C) (30-mer) (40 pg) was phosphorylated with T4 polynucleotide kinase (10 units) in kinase buffer at 37 degrees for two hours in the presence of ATP.

Phosphorylation was performed in a 1/10 aliquot with -v(32P)-ATP. After processing and freeze-drying, the oligonucleotide was desalted on a Sephadex G-50 column.

About 4 µg was dissolved in 300 µl of imidazole. HCl buffer (0.1M; pH 6.1) containing N-Ethyl-N'-(3-dimethylaminopropyl)-carbodiimide hydrochloride (soluble DCC) (0.1M). The solution was left at room temperature for three hours with occasional shaking. The activated nucleotide was purified on reverse phase workup with a triethyl ammonium acetate (0.1M, pH 7.0)/acetonitrile gradient. It was then reacted with a waveguide plate with amino groups in 10% lutidine/water solution at 60°, pH 8.5 for 4 minutes. The plate was then washed with primer and stored at 4 degrees until it was to be used.

c) Immobilization of single and double stranded DNA.

X DNA (3 pg in 50 µl) was treated with restriction enzyme Hpa

II in low salt buffer for two hours at 37 degrees. After processing, the DNA was ethanol precipitated and taken up in imidazole buffer (300 µl; pH 6.2) containing soluble DCC (0.1M) and NaCl (0.5M). The solution was left at room temperature for two hours with occasional shaking. Citrate buffer (0.2M; 300 µl; pH 8.4) containing NaCl (4M) was added to the activated aminopropyl glass plate which had recently been washed with diisopropylethylamine (5%) in ether. The activated DNA was added and mixed on the plate and left at 40 degrees for half an hour. After washing with NaCl (2M), the plate was placed under argon at degrees 1 closed atmosphere.

Alternatively, single-stranded DNA was bound to the glass slide in the absence of NaCl by heating at 70 degrees for 4 minutes. A new solution of cut X DNA was denatured by heating for 3 minutes at 100°C and added to the plate which was covered with NaCl solution (4M; 300 µl) and set at 80°C for 3 minutes. Then the solution was slowly cooled after which hybridization occurred.

d) Extension of the immobilized probe along the hybridized nucleic acid by the fill-in technique using the

four dNTPs and DNA polymerase I (Klenov fragment).

A synthetic 45-mer non-coding chain of pBR-322 (nucleotide no. 3871 to 3916) was immobilized on a waveguide as shown under section 2(b).

EcoR I cut of pBR322 was denatured for 3 min at 100 degrees in 200 µl buffer B solution (40 mM Tris.HCl Ph 7.5 20 mM MgCl2). After a rapid cooling, 1 M NaCl (200 µl) was added, and single-stranded DNA was hybridized to the probe that was on the waveguide (5 min at 80 degrees with slow cooling to room temperature (15')).

After washing with buffer containing NaCl (0.5M) and dithiothreitol (1mM) 40 µl NTP (3 mM of all four deoxyribonucleoside triphosphates) was added to 400 µl buffer B containing NaCl (0.5M) and dithiothreitol (1 nM) at room temperature and then 20 units of Klenov large fragment were added. e) Example of "sandwich hybridization" using a second probe with an oligo dC (3') end and poly dG and poly dC for

formation of a double-stranded complex network.

EcoR cut of plasmid pBR322 (1 pg) was denatured 1 5' at 100 degrees and hybridized to a synthetic 45-mer attached at its (5') phosphate end to the waveguide (Nucleotide no. 3871 to 3916 on the non-coding strand). A second oligonucleotide (0.5 pg) with primary structure as nucleotide No. 70 to 100 on the non-coding strand of pBR322 was elongated at its (3') end with dCTP (5 mM) in cacodylate buffer (according to Meth. of Enzymology No 65, p. 435). After hybridization of the second probe to the immobilized nucleic acid poly dG and poly dC was added to form a double-stranded complex network.

f) Example of "sandwich hybridization" (as in the previous example), where deoxycytidine to poly dC is replaced by

N<4->(6-aminohexyl)deoxycytidine (dah dC), and was reacted with

FITC.

Poly(dC) (5A units; 0.3-0.4 mg) in 1 ml of freshly prepared solution consisting of IM Na2S205/1M 1.6-diaminohexane/10mM hydroquinone (pH 7.3) was heated at 60 degrees for 4 days. The solution was then dialyzed against lOOmM borate buffer (pH 9.0) and run through a Sephadex G-50 column. This modified polymer was reacted with fluorescein isothiocyanate in borate buffer (pH 9.0) to form fluoresceined poly(dah dC), which was used as in the previous example in a "sandwich hybridization" experiment from a poly dG to form a complex network.

g) Example of solution hybridization followed by immobilization and (detection of) the hybridized complex which is

attached in the next section).

Briefly, a simple genetic sequence, for example poly-dG, is bound to the waveguide. Then, a second probe element, a continuous poly-dC having the length of poly-dG, which carries a sequence complementary to the nucleotides, for example 1-50, of the analyte sequence is hybridized in solution with said analyte sequence; then the signal forming duplex hybridized with the bound genetic sequence.

Experimentally, as in e) above, the prepared oligonucleotide (0.5 pg) of 70 to 100 b.p. of the noncoding strand of pBR322 elongated at its (3') with dCTP hybridized to heat-denatured EcoR I cutting of plasmid pBR322 (1 pg) for 5 min. The solution was then reacted with a waveguide covered with polydG for immobilization of the probe.

The sensitivity of the technique mentioned above can be improved (in those cases where the test sequence has, e.g. 200 bp) by adding, before hybridization with probe DNA, a third probe sequence element which is complementary to, e.g. nucleotides 55-200 of the test sequence, where hybridization of this additional element is performed under optimal conditions with an excess of probe. The partial hybrid is then hybridised to the probe which is bound to the waveguide. The signal that forms the potential of the hybrid is therefore amplified.

3. Experiments to measure DNA.

Apparatus that has been used is described above in connection with figures 4A, 4B and 4C. Waveguide 38 is flat injection molded plastic microscope slide. Light from a flash lamp 36 is sent through a monochromator 39 for selection of specific wavelength. The angle of incidence of the excitation light at the waveguide/liquid interface is controlled using two mirrors Ml, M2. Light is coupled in and out of the waveguide using two quarter-round quartz prisms, 46, 47. Fluorescent light is detected either at right angles to the waveguide with a detector 54, otherwise the fluorescent light that is fed back into the waveguide is detected by the waveguide arrangement with a detector 40, including a photomultiplier tube which detects fluorescent light passing through emission filter 52. The system is illustrated to collect returned fluorescent light, but can be easily adjusted to measure fluorescence at right angles by setting the filter/detector as in fig. 4A. The signal is pre-amplified at 41, and sent to microprocessor-controlled flash control/data system 12-45. The digitized signal is then stored on floppy disk 45, or written as hard copy 44. The samples are introduced using a current cell 37, and the volume in contact with the waveguide surface is defined by the rubber sealing clip 50. The samples are injected into the cell using an automatic pipette (eg, Microlab P, Hamilton Industries, Bonaduz, CH) that controls the injection volumes and rates. In general, the probe nucleotides were coupled to the waveguide using the techniques described in Section 2 above. The grafted waveguide was then placed in the optical setup field and the current cell was placed on top as shown in fig. 4C. Then a solution consisting of hybridization buffer (NaCl M) was placed in the cell and, after start-up, a baseline was formed. Then the specific reaction was started by adding a test solution to the cell and read by the change in fluorescence either at the waveguide outlet or under the waveguide.

Two main experiments were conducted:

a) End point (Fig. 5A). This is a control experiment to show the feasibility of the test, i.e. formation of a

sufficiently strong signal by adding an intercalator dye to double-stranded DNA. In a first variant, waveguides produced as described in section 2 (c), i.e. covered with non-denatured X DNA fragments were treated with different amounts of ethidium bromide in the same buffer (NaCl M; 100 pg dye/ml stock solution). Usable responses to ethidium bromide were obtained at concentrations in the cell from 1 to 100 pg/ml. The response was also proportional to the density of DNA on the waveguide.

Fig. 5A illustrates the conditions where the DNA probe has a density of approximately 10-3 - 10-1 pmol XDNA/cm<2> of the probe surface against a concentration of 60 pg/ml ethidium bromide in the cell. In fig. 5A, line B represents the increase in response resulting from addition of the dye to the cell in the presence of duplex DNA on the waveguide. Line A is the control obtained under identical conditions but with no DNA on the waveguide. The small jump in curve A is due to less solution fluorescence and the small fluorescence signal from non-intercalated dye.

In another variant, single-stranded DNA on waveguides was used. The probe plates used are according to procedure no. 2 (c) in section 2 above followed by denaturation and containing single stranded DNA derived from X DNA. Then, a DNA solution containing denatured DNA was added. This solution was prepared by denaturing a mixture of DNA after treatment with Hpa II endonuclease as indicated under Section 2(c), first paragraph above. After addition of single-stranded DNA, hybridization was performed for 20 min. at room temperature whereupon ethidium bromide solution was added as in the first variant above. A response similar to fig. 5A, line B was obtained. For a given amount of ethidium bromide reagent, it was proportional to the complementary DNA in the analyte sample.

In a third variant, the effect of elongation of double-helical polynucleotide after hybridization was tested.

Reference is made to Section 2(d) above. In the preparation in this section, first paragraph, a pBR322 cut denatured hybridized with a synthetic polynucleotide probe. When ethidium bromide was added to the hybridization solution, a response curve such as that illustrated in curve B of Figure 5A was obtained.

When a similarly prepared waveguide was contacted with the solution containing the elongated hybridized product presented in the second paragraph of Section 2(d) above for 15 minutes at room temperature and then ethidium bromide in 0.5M aqueous NaCl was added, a response curve of approximately twice the height of the first signal obtained.

In a fourth variant, solution hybridization and detection of the hybridized complex after formation, by construction of a probe oligonucleotide containing an immobilization sequence, was tested. Here the sequence was a polynucleotide with repeating known bases, but it could just as easily be any repeating polymer that allows immobilization of the complex to the surface via its complementary binding but that will not interfere with the hybridization reaction with the probe (e.g. immunological binding pairs , lectins and sugars, other proteins and hapten binders). To perform solution hybridization followed by immobilization and detection of the hybridized complex, the procedure described under 2(g) was followed, then ethidium bromide in aqueous NaCl (0.5M) was added and fluorescence was measured. The data show two response curves similar to curve B in fig. 5A but with different height. First lower curve, is the response obtained without denatured target DNA pBR322, and the signal is derived from hybridization of only probe-polydC fused to polydG attached to the waveguide.

Another similar curve, one that was higher, was obtained after hybridizing the probe to the target DNA and performing the immobilization reaction; this demonstrated the presence of hybridized probe/pBR322 complex and proved that the probe bound to the surface of the waveguide after hybridization with its complementary DNA. Thus, solution hybridization can be performed and the hybridized product can be measured according to the methods of this invention. Solution hybridization can include special advantages in some cases, but consideration is given to how quickly results are achieved.

b) Kinetic response.

The method of manufacture was essentially the same as for the other

the variant in Preparation 3(a) above except that ethidium bromide (50 µg/ml into the cell) was injected simultaneously with the complementary DNA sample. The results are shown in fig. 5B. Here, base line C has the same meaning as line A in fig. 5A. Line D shows the rate of fluorescence increase proportional to the degree of hybridization of complementary DNA in the analyte with DNA on the probe. The height of line D is proportional to the amount of DNA in said analyte. The results above show that color intercalation occurs immediately relative to the degree of hybridisation.

The general experimental conditions are given below.

The hybridization buffer was always 1 ml/L NaCl in distilled water. Hybridizations carried out before the reaction with the dye were reacted at 100°C. Hybridizations performed in the cell were performed at 22°C. The detailed effects of temperature on in situ readout of hybridizations have not yet been determined. The excitation wavelength was chosen to be 300 nm. A narrow band-pass interference filter was placed over the P.M. tube to measure fluorescence at 600 nm (type 600S10-25, Oriel). The angle of incidence of the excitation light at the interphase was chosen, as close to the critical angle as possible, at 66°C. This was very important to maximize the sensitivity of the system because at this angle there is a maximum number of reflections and therefore the maximum area of the surface is detected, and the field strength at the surface is strongest at angles close to the critical angle which thus increases the possibility of reading the surface reaction . Ethyldium bromide concentration (Fluka Chemicals) was varied, but best between 1 and 100 pg/ml.

c) Effect of excitation wavelength.

To demonstrate the effect of the excitation wavelength on the

signal detection became a waveguide with double-stranded DNA on it

the surface located in the optical device; excitation lambda was scanned from 300 to 600 nm with buffer in the cell and after reaction with ethidium bromide. This was repeated with a clean waveguide without DNA on the surface to act as a control. The results of four wavelengths 300, 360, 500 and 550 nm are given below in arbitrary units;

The above results illustrate that 300 nm is the best lambda to use in this particular system. These results demonstrate that the intercalated dye exerts similar fluorescence properties when the DNA is attached to the surface as when the DNA is free in solution or in a gel.

d) Kinetic derivation of formation of double-stranded DNA.

A plate waveguide with poly-dA attached to it was used (see Fig

the presentation under section 2 (a)). Then, the waveguide was attached to the current cell and a base-line signal was obtained with buffer in the cell. Both ethidium bromide and .25U/ml poly-dT were mixed and injected into the flow cell at the same time and the reaction was read. The results are given in figure 6, line

F. A control experiment is also shown where only ethidium bromide was injected into the cell (line E). These results demonstrate that the binding between complementary DNA strands can be read as occurs in this invention. The reading reaction is effective immediately and only specific intercalation between double-stranded DNA was detected. This could be very useful for the formation of an analyzer used to perform DNA probe analyzes in clinical and diagnostic use. Similar data were found with poly-dC and poly-dG, but the signals were stronger. Another test was performed to illustrate the general utility of using "sandwich" type hybridization combined with formation of a double-stranded complex network with polydC:polydG to provide many intercalation sites. The procedure described in section 2(e) was followed and hybridization was performed in the presence of ethidium bromide. The degree was measured by fluorescence as above and curves similar to those illustrated in Fig. 6 was achieved. The first curve (the lower one) represents the signal obtained before the final step of complex formation. Another curve (similar to F in Fig. 6) was obtained by mixing the two polynucleotides (polydG and polydC) with ethidium bromide and kinetic reading of the reaction. These data illustrate the higher sensitivity of this method relative to non-complex formation.

Fig. 8 illustrates, by a series of diagrams (A, B, C and D), some of the main analytical techniques provided by this invention. The purpose of these diagrams is to show a brief summary of the different methods presented in this specification.

Diagram A shows a waveguide 100 on which a probe polynucleotide 101 has been attached via a linker 102 to a binding site 103 on the waveguide. The diagram also shows a segment of analyte polynucleotide 104 on which the center section is complementary to the probe and which hybridizes to the probe.

Diagram B shows the case where a piece of DNA 104 hybridized with a probe 101 attached to a waveguide 100 on which the single-ended strand 106 has been filled with complementary bases 105 to achieve expansion (see section 2(d)).

Diagram C illustrates the progressive complementation of DNA 107 complementary to the free section 106 of test DNA 104, this complement DNA 107 carries a sequence 108 which proceeds to be later complemented by DNA 109 and 110. These supplementary sequences may consist of synthetic homopolynucleotides (poly-dA, poly-dT, poly-dC, poly-dG). This illustrates that the "sandwich hybridization procedure in section 2(e). This elongation increases the number of fluorescence sites probe unit and increases the sensitivity. It should be noted that the sequences defined by the numbers 109 and 110 although represented as heterogeneous, may consist of the same or different nucleotides, the condition being that part of 109 is complementary to 108 and that at least part of 110 is complementary to at least part of 109. Thus, as an example, 109 could consist of only poly-dG and 110 poly-dC only.

It is also important to note that for the purposes of preparation illustrated in diagram C, probe DNA may consist of a piece of DNA to be attached to the waveguide by hybridization with a complementary DNA already attached to the waveguide by a link 102- 103.

Diagram D illustrates solution hybridization of the probe with test DNA 104 and subsequent attachment of the hybrid to waveguide 100 via the linking parts 102 and 103.

The measurement technique in this invention concerns both reading of fluorescence variation (i.e. when hybridization occurs on the surface of the waveguide or when the duplex resulting from solution hybridization is attached to the waveguide) and to end-point determination (i.e. measurement of fluorescence at the outlet of the waveguide after the reaction is complete or that the hybridized product has adhered to the waveguide).In this case, the outlet signal can be measured while the waveguide is still in contact with the analyte sample solution or after separation of the waveguide from the solution. The advantages of performing this separation before measuring the signal are to increase or eliminate the background noise of the solution.

Claims (1)

1. Method for determining nucleic acids (a) that include one or polynucleotide sequences of interest that are present in a liquid analyte sample mixture consisting of polynucleotides, characterized in that the method includes: i) contacting said sample in a reaction medium with a polynucleotide probe (b) that includes polynucleotides corresponding to said sequence of interest with the presence of a solid phase that has binding affinity for said probe, whereupon hybridization between (a) and (b) occurs with the formation of a two-stranded duplex structure, 11) addition of a labeled compound designed to specifically label said duplex, e.g. intercalate between the strands in said duplex, which thus forms a fluorescent complex signal produces which indicates the formation of said duplex, characterized in that said solid phase is a waveguide on which said probe is bound before or after step i) and which carries an internally reflected wave signal and which reacts with said bound complex at the surface thereof and produces a fluorescence response which is measured at an outlet on said waveguide and which thus produces data representative of said determination. 2. Method according to claim 1, characterized in that the probe is bound to the waveguide before step i, whereupon hybridization at the surface of the waveguide and the increase in fluorescence is representative of the degree of said hybridization, i.e. for the amount of nucleic acid analyte in the sample. 3. Method according to claim 1, characterized in that step i) is started before the probe is bound to the waveguide, whereupon fluorescence increase with time denotes the degree of binding of the duplex structure to the waveguide surface and the total end-point fluorescence relates to the amount of analyte nucleic acid in the sample. 4. Method according to claim 1, characterized in that said waveguide is a rectilinear transparent solid material, whereupon a signal outlet in said waveguide is collected laterally at right angles to said waveguide or axillary at one end of said waveguide or at both. 5 . Method according to claim 1, characterized in that the waveguide is an optical rod or fiber or a plate which forms a wall in an analytical reaction test cuvette. 6. Method according to claim 5, characterized in that an outlet signal in said waveguide is collected laterally at right angles to said interphase whereupon the emitted fluorescent light is picked from a side of the waveguide which is opposite to said interphase. 7. Method according to claim 1, characterized in that said interaction between the reflected wave signal and said fluorescent complex mainly occurs at a fraction of a wavelength away from the waveguide/analyte interface which can clearly distinguish between the fluorescence produced by said complex which is bound to the waveguide from the which occurs within the bulk of the analyte. 8. Method according to claim 2, characterized in that the nucleic acid probe is covalently attached to the waveguide at a terminal nucleotide, whereupon all or most of the bases of said nucleic acid are available for hybridization. 9. Method according to claim 8, characterized by the probe nucleotides aligning themselves in a parallel relationship to the waveguide surface, as this orientation is a result of the presence on the waveguide surface of sites with electrostatic affinity for the phosphate part in said probe polynucleotides. 10. Method according to claim 9, characterized in that said seats result from the presence on the waveguide surface of positively charged groups, e.g. quaternary ammonium groups. 11. Method according to claim 8, characterized in that said covalent bond is achieved by first grafting tri- oxy-silane-bearing reactive groups such as -NHg, -OH, -SH, glycidoxy, -NCO and the like onto the waveguide and then using these groups or derivatives thereof as linkers for the probe polynucleotides. 12. Method according to claim 8, characterized in that the nucleic acid probe consists of oligonucleotides in a range of approximately 10 to 105 nucleotides. 13. Method according to claim 8, characterized in that said covalent bond is achieved by reacting the probe nucleic acids with reactive groups that have previously been incorporated on the waveguide substance. 14. Method according to claim 1, characterized in that the waveguide is part of or consists of the entire molded disposable plastic cuvette. 15. Method according to claim 1, characterized in that the waveguide is cast from high optical quality plastic such as polycarbonate or polymethyl methacrylate. 16. Method according to claim 1, characterized in that determination of the fluorescence signal is carried out by either measuring said fluorescence at equilibrium (end-point determination) or by measuring the degree of change of the fluorescence while the reaction is in progress, or both. 17. Method according to claim 1, characterized in that the reaction is carried out under controlled temperature conditions. 18. Method according to claim 17, characterized in that the temperature is varied continuously, linearly or according to a stepwise gradient. 19. Method according to claim 18, characterized in that the measured results are calculated to obtain the average thermal denaturation temperature Tm of the hybridized product. 20. Method According to claim 1, characterized in that one or more chaotropic agents or polymers selected from dextran sulphate polymers, polyethylene glycol, sodium iodide and perchlorate salts are added to the reaction medium. 21. Kit for carrying out the method according to claim 1, characterized in that it contains the following components: a sampling device for biological cellular material ale; a sample preparation device for extraction, deproteinization and separation of the strands in the selected nucleic acid sample from said cell material; a cuvette waveguide incorporating a nucleus acid probe; a means of introducing the prepared sample and an intercalator dye into the cuvette waveguide; an optical device for illuminating the waveguide with a total reflected signal and timing and determining the fluorescence generated at the waveguide surface upon hybridization of the nucleic acid sample with the probe; a signal processing device and a device which electronically processes the determined fluorescence and which presents early results representative of said hybridization reaction. 22. Kit according to claim 21, characterized in that the cuvette waveguide is either a disposable cheap cast cuvette or a current cell that can be used several times. 23. Method according to claim 1, characterized in that it includes amplification of the fluorescence signal which is of interest by increasing the available fluorescent site concentration in the space which is optically probed by the excitation signal. 24 . Method according to claim 23, characterized in that said concentration increase is achieved by expanding the amount of duplex structure resulting from hybridization at the probe surface. 25. Method according to claim 24, characterized in that said length increase is achieved by adding several nucleic acid homologues to the not yet hybridized sequences of the bound test nucleic acid, after which a further fusion takes place with the achievement of several color intercalator sites. 26. Method according to claim 1, characterized in that a 5'-phosphate-linked nucleic acid probe is extended along the hybridized form using the four deoxynucleoside triphosphates and a DNA polymerase or a reverse transcriptase (fill-in technique), or both. 27. Method according to claim 1, characterized in that the probe is bound to the waveguide by hybridization with complementary DNA which is attached to said waveguide or covered thereon. 28. Method according to claim 1, characterized in that the probe is attached to one partner of a bond pair and whereupon another partner of said pair is attached to the waveguide surface, whereupon probe attachment to the waveguide takes place via formation of the bond pair. 29. Method according to claim 28, characterized in that said first partner is selected from antigens, haptens, glycosides, sugar, biotin, avidin and other binding proteins, whereupon the second partner is complementary to said first partner. 30. Method according to claim 1, characterized in that the waveguide is a hollow tube and the probe is immobilized on the inner or outer surface thereof or on both. 31. Method according to claim 1, characterized in that said measurement is carried out either during the hybridization by determining the fluorescence, or, after the hybridization has been completed, by fluorescence end-point determination. 32. Method according to claim 31, characterized in that it includes fluorescence end-point determination, whereupon the measurement is carried out either in the presence or without liquid analyte solution.