JP2002517183A - Multiparametric fluorescence in-situ hybridization - Google Patents

Abstract

  (57) [Summary] The present invention provides (i) a combinatorially labeled oligonucleotide probe set, wherein each member has (i) a predetermined label that can be distinguished from the label of any other member of the above set. And (ii) those that can physically hybridize to a given chromosome or nucleic acid molecule, and to the use of such molecules alone or in response to nucleic acid amplification methods.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001] The present invention relates to reagents and methods for accomplishing nucleic acid chemistry, in particular, multiplex image analysis of chromosomes and chromosomal fragments. The present invention can be used to diagnose chromosomal abnormalities, infectious agents, and the like. This invention was made, in part, with a government fund. The government has certain rights in the invention.

[0002] This application is a pending application of US Patent No. 08 / 580,717, filed December 29, 1995, which is a continuation-in-part of US Patent Application No. 08 / 577,622, filed December 22, 1995. No. 08 / 640,657, filed May 1, 1996, which is a continuation-in-part application of which is incorporated herein by reference in its entirety. included. BACKGROUND OF THE INVENTION Determining the presence and status of chromosomes and chromosome fragments in biological samples is of immense importance in diagnosing disease. Traditionally, such determinations have been made manually by examining metaphase chromosome specimens that have been treated with special stains to reveal characteristic staining patterns. Unfortunately, the interpretation of such dyeing patterns requires substantial skill and is technically difficult. Therefore, alternative methods of analyzing chromosome presence and alignment have been sought. One alternative approach to the problem of chromosome identification has involved the use of labeled chromosome-specific oligonucleotide probes to label repeats of interphase chromosomes (Cremer, T. et al., Hum. . Genet. 74:
346-352 (1986); Cremer, T. et al., Exper. Cell Res. 176: 119-220 (1988).
). Such methods are known to be useful for prenatal diagnosis of Down syndrome, as well as for detecting chromosomal abnormalities associated with tumor cell lines. Chromosome-specific probes of repetitive DNA confined to distinct subregions of the chromosome, however, are not suitable for the analysis of many types of chromosomal abnormalities (eg, translocations or deletions).

[0003] Ward, DC, et al. (PCT application WO / 05789, incorporated herein by reference) provide for the specific labeling of selected mammalian chromosomes in a manner that allows recognition of chromosomal abnormalities. A method for chromosome in-situ suppression ("CISS") hybridization is disclosed. In that method, the sample DNA is denatured and hybridized with a mixture of a fluorescently labeled chromosome-specific probe of high genetic complexity and an unlabeled non-specific competitor probe. Chromosomal images were obtained as described by Manuelidis, L. et al. (Chromosoma 96: 397-410 (1988); incorporated herein by reference). The method provides a rapid and highly specific assessment of individual mammalian chromosomes. The method allows for the visualization of some or all sub-regions of chromosomes in a specimen with judicious choice of appropriate probes and / or labels. For example, by using more than one probe each specific to a subregion of a target chromosome, the method allows for the simultaneous analysis of several subregions on that chromosome. The number of available fluorophores limits the number of chromosomes or chromosomal subregions that can be visualized simultaneously.

As described in PCT application WO / 05789, a “combination” variant of the CISS method is used.
Can be. In the simplest case, two fluorophores have three different chromosomes or chromosomes
Simultaneous visualization of subregions. In this variant, the hybridization probe mix
The compound consists of two halves, each separately labeled with a different fluorophore.
The array is made from a single set of arrays. Upon hybridization, two fluoro
The fore emits three fluorescent signals that are optically distinguishable from the color of the individual fluorophore.
Occurs. An extension of this approach to a Boolean combination of n fluorophores is 2 ⁿ -Allows labeling of one chromosome.

[0005] Ried, T. et al. (Proc. Natl. Acad. Sci. (USA) 89: 1388-1392 (1992), which is incorporated herein by reference) describes three fluorophores. Describes the use of an epi-fluorescence microscope equipped with a digital video camera and computer software to "pseudo-color" the fluorescence pattern obtained from the simultaneous in-situ hybridization with 7 probes using FT. The use of a wavelength selective filter allows for the isolation and collection of a separate grayscale image of each fluorophore. These images are then merged via appropriate software. CC
The sensitivity and linearity of D-cameras overcome the technical difficulties inherent in color film-based micrographs.

[0006] Such efforts have increased the number of chromosomes that can be simultaneously detected and analyzed using in-situ hybridization techniques, but have enabled the simultaneous detection and analysis of many different chromosomes and chromosomal subregions. It is highly desirable to limit the set of fluorophores having distinct emission spectra. The present invention provides such reagents and methods and devices for their use.

SUMMARY OF THE INVENTION The present invention provides for the combined labeling of nucleic acid probes sufficient to allow visualization and simultaneous identification of all 22 human autosomes and the human X and Y chromosomes or their restricted subregions. And reagents and methods for the same. Such specific labeling of whole chromosomes or their restricted subregions is called "coloring". The present invention further relates to reagents and methods for combinatorial labeling of nucleic acid probes sufficient to allow characterization of bacteria, viruses and / or lower eukaryotes that may be present in clinical or non-clinical specimens. .

In particular, the present invention provides a set of combinatorially labeled oligonucleotide probes consisting of first and second sub-sets of probes, wherein (A) each member of the first sub-set of probes comprises: (I) linked or linked to a predetermined label that is distinguishable from the label of any other member of the first or second subset of probes, and (ii) one of the predetermined autosomal or sex chromosomes of the human karyotype. A plurality of oligonucleotides capable of specifically hybridizing with each other, wherein the first subset of the probe set is sufficient to specifically hybridize each autosomal or sex chromosome of the human karyotype with at least one member. (B) each member of the probes of the second sub-set is linked or conjugated to a predetermined label which is distinguishable from the label of any other member of the first or second sub-set. Ori (Ii) a probe comprising a plurality of oligonucleotides capable of specifically hybridizing to one of the extrachromosomal polynucleotide copies of a predetermined region of a human karyotype autosome or sex chromosome.

[0009] The present invention further provides a set of combinatorially labeled oligonucleotide probes consisting of a first subset of genotype probes and a second subset of phenotypic probes, wherein (A) the first subset of genes Each member of the type probe is linked or conjugated to a predetermined label that is distinguishable from the label of any other member of the first or second subset of probes, and (ii) the preselected bacteria, virus Or a plurality of oligonucleotides capable of specifically hybridizing to a region of a lower eukaryotic nucleic acid, wherein the first sub-set of the probe set comprises a pre-selected bacterium, virus or Have sufficient members to be distinguishable from a virus or lower eukaryote, and (B) each member of the phenotypic probe of the second sub-set comprises (i) any other member of the first or second sub-set. The members of the sign Linked or conjugated to a predetermined label that is distinguishable,
And (ii) the predetermined polymorphism of the chromosome of the preselected bacterium, virus or lower eukaryote to enable determination of whether the preselected bacterium, virus or lower eukaryote exhibits the preselected phenotype. The present invention relates to a probe comprising a plurality of oligonucleotides capable of specifically hybridizing with a nucleotide region or an extrachromosomal copy thereof.

The present invention further provides a method for simultaneously identifying and differentiating individual autosomes and sex chromosomes having a human karyotype, comprising: (I) under conditions sufficient to cause nucleic acid hybridization. A single-stranded form of the chromosome sample is provided by a set of combinatorially labeled oligonucleotide probes consisting of first and second sub-sets of probes, wherein (A) each member of the first sub-set of probes comprises: (I) linked or linked to a predetermined label that is distinguishable from the label of any other member of the first or second subset of probes; and (ii) one of the predetermined autosomal or sex chromosomes of the human karyotype. A plurality of oligonucleotides capable of specifically hybridizing with each other, wherein the first subset of the probe set specifically hybridizes each autosomal or sex chromosome of the human karyotype with at least one member. A predetermined label that has sufficient members to rediscover and (B) each member of the probes of the second subgroup is (i) distinguishable from the label of any other member of the first or second subgroup. And (ii) comprising a plurality of oligonucleotides capable of specifically hybridizing to one of the extrachromosomal polynucleotide copies of the predetermined region of the autosomal or sex chromosome of the human karyotype. (II) for each chromosome of the specimen hybridized to a member of the first sub-set of probes, detecting and identifying the member's intended label, and identifying the identity of the member's label with the member's specificity. Correlating with the identity of the autosomal or sex chromosomes of the human karyotype to specifically hybridize, thereby identifying the chromosome hybridized to its members; Repeating step (II) until each of the autosomal and sex chromosomes of the karyotype is identified in the specimen; and (IV) a probe hybridized with a predetermined extrachromosomal polynucleotide copy of an autosomal or sex chromosomal region. For each member of the second sub-group of members of the group, the expected label of that member is detected and identified, and the identity of the label of that member is compared to the autosomal or sex chromosome of the human karyotype to which that member specifically hybridizes. Identifying the region of the autosomal or sex chromosome that correlated with the identity and thereby hybridized to that member; and (V) repeating step (IV) for each member of the second subset of probes. To a method comprising:

The invention further provides a method of identifying and distinguishing a preselected bacterium, virus or lower eukaryote simultaneously with another bacterium, virus or lower eukaryote, comprising: (I) in- A sample suspected of containing a preselected bacterium, virus or lower eukaryote under conditions sufficient to allow for in situ nucleic acid hybridization, the genotype probe of the first sub-set and the phenotype of the second sub-set A set of combinatorially labeled oligonucleotide probes consisting of a probe, wherein (A) each member of the first set of genotype probes comprises (i) any other member of the first or second set of probes. And (ii) capable of specifically hybridizing to a region of the nucleic acid of the preselected bacterium, virus or lower eukaryote, wherein the plurality is linked to or linked to a predetermined label that is distinguishable from the label of The first subset of the probe set is sufficient to distinguish the preselected bacteria, virus or lower eukaryote from other bacteria, viruses or lower eukaryotes present in the specimen. (B) each member of the second sub-set of probes is linked or conjugated to a predetermined label that is distinguishable from (i) the label of any other member of the first or second sub-set. And (ii) the nucleic acid of the preselected bacterium, virus or lower eukaryote to enable determination of whether the preselected bacterium, virus or lower eukaryote exhibits the preselected phenotype. Contacting with a probe comprising a plurality of oligonucleotides capable of specifically hybridizing to the predetermined polynucleotide region or extrachromosomal copy thereof, (II) staining of preselected bacteria, viruses or lower eukaryotes For each member of the first subset of probes hybridized to a region of the body, the member's intended label is detected and identified, and the member's label identity specifically hybridizes to the member's label identity. Correlating with the identity of the bacterium, virus or lower eukaryote, thereby identifying the bacterium, virus or lower eukaryote hybridized to the member; (III) each of the first sub-set of probes Repeating (II) for members, (IV) each of a second subset of probes hybridized to a predetermined polynucleotide region of a chromosome of a preselected bacterium, virus or lower eukaryote or an extrachromosomal copy thereof. For a member, detecting and identifying the member's intended label, and correlating the member's label identity with the identity of the intended region, thereby pre-selecting bacteria To identify the presence of regions for on a chromosome of a virus or lower eukaryotes, it relates to a method comprising the step of repeating (V) for each member of the second sub-set of probes, step (IV).

The invention is particularly directed to a method wherein said members of said set of probes are detectably labeled with a fluorophore and at least one member of said set is a fluorophore FITC, Cy3, Cy3.5.
, Cy5, Cy5.5, and Cy7 are labeled in combination with one, two, three, four, or five fluorophores selected from the group consisting of: Embodiments that are labeled with one fluorophore are contemplated.

The invention further relates to a set of combinatorially labeled oligonucleotide probes, each member of which (i) has a predetermined label that is distinguishable from the labels of any other member of the set; ii) a set of probes capable of specifically hybridizing to the expected autosomal or sex chromosome telomere region of a human karyotype, wherein each of the human karyotype autosomes or sex chromosomes specifically binds to at least one member. Provide a set with sufficient members to be able to hybridize.

The present invention further provides a method for simultaneously identifying and distinguishing individual autosomes and sex chromosomes of a human karyotype, comprising: (a) under conditions sufficient to cause nucleic acid hybridization, A sample of the single-stranded form of the chromosome is obtained by combining a set of combinatorially labeled oligonucleotide probes, each member of which has (i) a predetermined label that can be distinguished from the label of any other member of the set. And (ii) specifically hybridize to a telomere region of the intended autosomal or sex chromosome of the human karyotype, wherein the set is specific for each autosomal or sex chromosome of the human karyotype to at least one member. Contact with a combinatorially labeled oligonucleotide probe having sufficient members to hybridize specifically, whereby contact of each of the autosomal or sex chromosomes of the specimen with Causing at least one to hybridize to at least one member of the set of probes; (b) detecting, for each chromosome in the specimen, hybridized to a member of the set of probes, a prospective marker for that member And identify, and correlate, the identity of the label of the member with the identity of an autosomal or sex chromosome of the human karyotype to which the member specifically hybridizes, thereby displacing the chromosome hybridized to the member. And (c) until each of the autosomal and sex chromosomes of the human karyotype is identified in the specimen.
Providing a method comprising repeating step (b).

DESCRIPTION OF THE PREFERRED EMBODIMENTS Overview of the Invention Fluorescence in-situ hybridization (FISH) is used in various areas of research and clinical diagnosis (Gray, JW et al., Curr Opin Biotech 3: 623-631 (1992)).
; Xing, Y. et al., In: The Causes and Consequences of Chromosomal Aberrat
ions. IR Kirsch Ed. CRC Press, Boca Raton, page 3-28 (1993)). It is an indispensable tool for studying the chromosome and superchromosome construction of cell nuclei (Cremer, T. et al., In: Cold Spring Harbor Symposia on Quantitative
Biology, Volume LVIII, pp. 777-792, Cold Spring Harbor Laboratory Press,
NY (1994)). Most importantly, FISH provides the ability for multi-parameter identification. This is due to the combination (Nederlof, PM et al., Cytometry 1
0: 20-27 (1989); Nederlof, PM et al., Cytometry 11: 126-131 (1990); Ri
ed, T. et al., Proc Natl Acad Sci (USA) 89: 1388-1392 (1992a); Ried,
T. et al., Hum Mol Genet 1: 307-313 (1992b); Lengauer, C. et al., Hum Mo
l Genet 2: 505-512 (1993); Popp, S. et al., Human Genetics 92: 527-532 (1
993); Wiegant, J. et al., Cytogenet Cell Genet 63: 73-76 (1993)) or ratio labeling (Dauwerse, JG et al., Hum Mol Genet 1: 593-598 (1992); Nede
rlof, PM et al., Cytometry 13: 839-845 (1992); du Manoir, S. et al., H
um Genet 90: 590-610 (1993);), allowing simultaneous visualization of several DNA probes using any of the strategies. Up to 12 DNA probes have been visualized (Dauwerse, JG et al., Hum Mol Genet 1: 593-598 (1992)). As a result, the goal of 24 different colors has long been sought (Ledbetter, DH, H
um Mol Genet 5: 297-299 (1992)). The 24 different colors are important thresholds because they allow the simultaneous visualization of 22 autosomes and both sex chromosomes. In addition to improved karyotyping, the possibility of moving and hybridizing 24 different and distinguishable DNA probes allows for the treatment of a number of important biological problems. However, the multicolor systems published to date lack the versatility for expansion to 24 colors, and only principle lighting experiments have always been published.

The present invention results, in part, from the realization of a multiparametric fluorescence in-situ hybridization method to achieve simultaneous visualization of 24 different gene targets using a combinatorial labeling strategy. This strategy allows discrimination between more target sequences than spectrally distinguishable labels. The simplest way to perform such labeling is to use a simple "Boolean" combination, ie, the fluorophore is completely absent (ie, assigned a value of "0") or has a (Value of 1). There is only one useful combination for a single fluorophore A (A = 1), and for two fluorophores A and B, three useful combinations (A = 1 / B = 0; A = 0 / B =
1; A = 1 / B = 1). Seven combinations of three fluorophores, fifteen combinations of four fluorophores, 31 combinations of five fluorophores, 63 combinations of six fluorophores, and so on (n fluorophores are 2 ⁿ -1 chromosome). Thus, to uniquely identify all 24 chromosome types in the human genome using a chromosome coloring probe set, only five distinguishable fluorophores are needed (31 combinations in total). If each probe set is labeled with one or more of the five spectrally distinct fluorophores in a combinatorial fashion, then using a simple Boolean combination, the spectrum dictated by the fluorophore composition Each DNA probe can be identified by its characteristics (Speicher, MR et al., Nature
Genet. 12: 368-375 (1996), the contents of which are incorporated herein by reference.

B. Terms of the Invention The present invention relates to a set of recombinantly labeled oligonucleotide probes, each member of which has (i) a predetermined label that is distinguishable from the labels of any other members of the set; (Ii) a probe capable of specifically hybridizing to a predetermined autosome or sex chromosome of a human karyotype. In a most preferred embodiment, the set has a sufficient number of members to specifically and distinguishably hybridize each of the human karyotype autosomes or sex chromosomes with at least one member. As used herein, the term "karyotype" refers to the complement of chromosomes found in normal or abnormal cells. In normal cells, the number of chromosomes is 46 and consists of 22 pairs of autosomes and two sex chromosomes (2X chromosomes (for females) or X and Y chromosomes (for males)). A sign is said to be distinguishable if the particular sign (identity of that member) of any member of the set is different from the particular sign and identity of any other member of the set. Each probe member is a unique chromosome (or subchromosomal region)
Because of the ability to specifically hybridize to and the identity of the label and the probe being known in advance, the detection of a particular label associated with an unidentified chromosomal region requires that the probe bearing that label be identified by the unidentified chromosome. It means that it has become hybridized with the region. Since the chromosomes to which the probe hybridizes specifically are known, detection of a distinguishable label allows identification of a chromosomal region.

More particularly, the present invention relates to such a probe, wherein the multiparametric fluorescence in-situ
It can be used to label oligonucleotide probes as can be used for hybridoma formation. As used herein, a "fluorophore" or "fluorophore" is a reagent that can emit a detectable fluorescent signal upon excitation. Most preferably, the fluorophore is directly attached to the pyrimidine or purine ring of the probe nucleotide (Ried, T. et al., Proc. Natl. Acad. Sci. (USA) 89: 1388-1392 (1
992) (the contents of which are incorporated herein by reference); US Pat. No. 4,687.
Nos., 732, 4,711,955, 5,328,824 and 5,449,767, the contents of which are incorporated herein by reference. Alternatively, the fluorophore can be indirectly attached to the nucleotide, for example, by conjugating the fluorophore with a ligand capable of binding to the modified nucleotide residue. The most preferred ligands for this purpose are avidin, streptavidin, biotin-conjugated antibodies and digoxigenin-conjugated antibodies. The method of performing such a conjugation is described in Pinkel, D. e.
Natl. Acad. Sci. (USA) 83: 2934-2938 (1986), the contents of which are incorporated herein by reference.

The term “multiparametric fluorescence” refers to the combined use of multiple fluorophores to simultaneously label the same chromosome or subchromosomal fragment, and their detection and characterization. Chromosomes or sub-chromosomal fragments can co-localize when exposed to more than a single chromosome-specific probe under conditions sufficient to hybridize each chromosome-specific probe separately to its target chromosome. It is said to be labeled. As used herein, therefore, not all such hybridization reactions need to be initiated and completed simultaneously. The co-labeling enabled by the present invention is therefore in contrast to protocols where chromosomes are exposed at one time to only one chromosome-specific probe.

[0020] The simultaneous detection and characterization enabled by the present invention allows for the multiple (and most preferably all) autosomes in a sample without any need to add additional reagents or probes after the detection of the first chromosome. And / or the ability to detect sex chromosomes. In the simplest embodiment, a digital image of the chromosome is obtained for each fluorophore used, thereby providing a series of grayscale fluorescence intensities associated with each fluorophore and each chromosome. By pseudo-coloring the compounded gray scale intensity for each chromosome, the final image is obtained.

The present invention is therefore a method for simultaneously identifying and differentiating individual autosomal and sex chromosomes of the human karyotype, comprising the steps of: And contacting said set of combinatorially labeled oligonucleotides under conditions sufficient to effect nucleic acid hybridization.

Such treatment causes at least one of each autosome or sex chromosome of the specimen to hybridize to at least one member of the set of probes. For each chromosome in the specimen that has hybridized to a member of the set of probes, the human is then detected and identified with the member's intended label, and the member's label identity specifically hybridizes to the member. Correlate with karyotype autosomal or sex chromosome identity. This step identifies chromosomes that hybridize to that member. This final step is repeated until each of the human karyotypes or the desired number of autosomal or sex chromosomes has been identified in the specimen.

The oligonucleotide probes used according to the method of the invention have one of two general characteristics. In one embodiment, such probes are chromosome or subchromosome specific (ie, they hybridize to a particular chromosomal DNA at a lower c ₀ t ^1/2 than DNA of other chromosomes; c ₀ t ^1/2 is the time required for half the initial concentration (c ₀ ) of the probe to hybridize to its complement). Alternatively, such probes are feature-specific (eg, telomeres, centromeres, etc.). Such probes that are proximal to the telomeres can limit and identify translocations that might otherwise be potentially latent near chromosome ends.

[0024] Both types of probes can be used if desired. Such probe sources are available from American
Available from the An Type Culture Collection and similar depositories. Oligonucleotide probes used in accordance with the methods of the present invention are of sufficient size to allow probe entry and to optimize rear annealing hybridization. Generally, labeled DNA fragments of less than 500 nucleotides in length, more preferably probes of about 150-250 nucleotides in length, are used. Probes of such length can be made by synthetic or semi-synthetic means, or obtained from longer polynucleotides using restriction endonucleases or other techniques suitable for fragmenting DNA molecules. Alternatively, longer probes (eg, polynucleotides) can be used.

Most preferably, oligonucleotide probes are synthesized to contain biotinylated or otherwise modified nucleotide residues. Methods for achieving such biotinylation or modification are described in U.S. Pat.Nos. 4,687,732, 4,711,9
Nos. 55, 5,328,824 and 5,449,767, each of which is incorporated herein by reference. Biotinylated nucleotides and probes are obtained from Enzo Biochem, Boehringer Mannheim, Amersham and other companies. In essence, such biotinylated or otherwise modified nucleotides are produced by reacting a nucleoside or nucleotide with a mercuric salt in an environment sufficient to produce a mercurated nucleoside or nucleotide derivative. . The mercurated material is then reacted in the presence of a palladium catalyst with a moiety having a reactive end group and containing three or more carbon atoms (eg, a biotin group). This reaction adds that moiety to the purine or pyrimidine ring of the nucleoside or nucleotide.

In a highly preferred embodiment, such modified probes are described by Ward et al. (WO 90/0
5789) (this description is incorporated herein by reference) in conjunction with competitor DNA in the manner described. Competitor DNA is DNA that acts to suppress hybridization signals from ubiquitous repeats present in human and other mammalian DNA. In the case of human DNA, Ward et al. (WO90 / 05789
Alu or kpn fragments can be used, as described in First, the probe DNA bearing the detectable label and the competitor DNA are combined under conditions sufficient to cause hybridization between molecules having complementary sequences. As used herein, two sequences are said to be capable of hybridizing to each other if they are complementary and thus can form a stable antiparallel double-stranded nucleic acid structure. Conditions for nucleic acid hybridization suitable for forming such a double-stranded structure are described in Maniatis, T. et al. (Molecular Cloning, A Laboratory Manual, Cold Spring
Harbor Laboratories, Cold Spring Harbor, NY (1982)), Haymes, BD
(Nucleic Acid Hybridization, A Practical Approach, IRL Press, Washin
gton, DC (1985)) and Ried, T. et al. (Proc. Natl. Acad. Sci. (US
A.) 89: 1388-1392 (1992)). For the purposes of the present invention,
The sequences need not show strict complementarity, but need only be sufficiently complementary to be able to form a stable double-stranded structure. Thus, deviations from perfect complementarity are permissible unless such deviations are sufficient to completely prevent hybridization and formation of a double-stranded structure.

[0027] The amount of probe DNA that is combined with the competitor DNA is adjusted to reflect the relative DNA content of the chromosomal target. For example, as disclosed by Ward et al. (WO90 / 05789), chromosome 1 contains about 5.3 times the DNA present in chromosome 21. Thus, when using a chromosome 1 specific probe, a relatively higher probe concentration is used.

Next, the resulting hybridization mixture is processed to remove any DN present
Denature A and incubate at about 37 ° C. for a time sufficient to promote partial reannealing. The sample containing the chromosomal DNA to be identified is also heated to facilitate hybridization with the probe. Next, the hybridization mixture and the sample are combined under conditions sufficient for hybridization to occur. Thereafter, detection and analysis of the hybridizing substance is performed by detecting the fluorophore label of the probe in any of the following methods.

In an alternative embodiment, Ried T. et al. (Proc. Natl. Acad. Sci. (USA) 89: 1388
(1392) (1992), a variation of the method of which is incorporated herein by reference. Thus, the probes are labeled directly (eg, with fluorescein) or indirectly (eg, with biotinylated nucleotides or other types of labels) and hybridized to chromosomal DNA. After hybridization, the hybridized complex is incubated in the presence of streptavidin conjugated to one or more fluorophores. Streptavidin binds to the biotinylated probe of the hybridized complex, thereby allowing detection of the complex as described below.

C. Preferred Fluorophores of the Invention Combination and labeling with two or more fluorophores allows for discrimination between more subjects than available fluorophores. The simplest way to perform such labeling is by Boolean combination, ie, the fluorophore is completely absent (0) or has a unit amount (1).
). For a single fluorophore A, there is only one useful combination (A = 1). For the two fluorophores A and B, three useful combinations (A = 1, B = 0
A = 0, B = 1; A = 1, B = 1). For the three fluorophores A, B and C, seven combinations of (A = 1, B = 0, C = 0; A = 0, B = 1, C = 0;
A = 0, B = 0, C = 1; A = 1, B = 1, C = 0; A = 1, B = 0, C = 1;
A = 0, B = 1, C = 1; A = 1, B = 1, C = 1). There are 15 combinations of 4 fluorophores, 31 combinations of 5 fluorophores, 63 combinations of 6 fluorophores, and so on.

To uniquely encode all 24 chromosome types in the human genome, five distinct combinatorial fluorophores are required. For the 5-fluorophore set, 15 chromosomes can be distinguished using four combinations of the five fluorophores. Labeling of the remaining nine chromosomes requires all five fluorophores used in combination. No seven available 5-fluorophore combinations are required. Thus, there is a certain amount of tolerance available to avoid any 5-fluorophore combination that can prove particularly difficult to solve. In particular, quaternary or quinary combinations can be avoided.

One aspect of the present invention relates to the identification of a set of seven fluorophores that can be fully elucidated by excitation-emission contrast (EEC) techniques. As mentioned above, multiple fluorophore combinatorial labeling generally relies on the acquisition and analysis of the spectral properties of each subject, ie, by obtaining the relative metric coefficients of the constituent fluorophores. Since full spectroscopic analysis of mixed fluorophore spectra (eg, by interferometry) has not yet been fully developed, the method of choice was to use epi-fluorescent triplets, ie, excitation filters, dichroic reflectors and emission bandpass filters. It was a conventional wide-field-of-view image with limited bandwidth. The limited spectral bandwidth available for imaging (approximately 380-750 nm) and the broad overlap between the spectra of the organic fluorophores makes spectrophotometric separation of multiple fluorophores during the imaging process a significant technical test.

In order to actually make the software segmentation of the light source image as straight as possible, a target value of <10% crosstalk between any given fluorophore and two adjacent channels was set. The computer model is DAPI + 5 combination fluorophores FITC, Cy3, Cy3.
For 5, Cy5, Cy5.5, this level of contrast showed that no matter how narrow the filter bandwidth, it could not be achieved using excitation or emission selection alone. Thus, both the excitation and emission selections must be triggered simultaneously. This is referred to as excitation-emission contrast (EEC).

A contrast ratio plot was first computed for the fluorophore versus each of its two neighbors. These plots show areas where the pairing contrast is high enough to be useful. A constraint on the particularly achievable contrast is that the high contrast region is generally well below at least one side of the spectrum, i.e. where excitation and / or emission is tightly suboptimal. Furthermore, it is necessary to use a filter with a bandwidth in the range of 5 to 15 nm in order to obtain the required degree of selectivity (compared to about 50 nm for the "standard" filter set). At the same time, they impose a heavy sensitivity penalty. The goal of 10% maximum crosstalk represents an acceptable practical compromise between sensitivity and selectivity.

The basic asymmetry exists between the excitation contrast and the emission contrast. Cooling C
For low noise detection, such as CD, the limitation of the excitation bandwidth is the achievable image S / N
Has little effect on the ratio. The only penalty is a long exposure time. However, emission bandwidth limitations are highly undesirable because all fluorescent photons blocked by the filter exhibit irreversible photochemical bleaching of the fluorophore. For this reason, the highest viable contrast was brought to the excitation side. A suitable emission filter was then found to produce the required EEC ratio. Filter selection is additionally imposed by the fact that each channel must properly reject both adjacent channels at the same time, and one improvement can significantly resolve the other. Good contrast was indeed achievable for all fluorophores except the peripheral Cy5.5 / Cy5 pair. For this reason, Cy7 was later used instead of Cy5.5. Other considerations regarding filter selection are: Commercial narrow band interference filters can have a large amount of wedges, or non-parallel, between the top and bottom surfaces. This results in a large image shift (equivalent to several microns). The shift is the vector characteristic of each filter and its orientation in the epicube. Therefore, automatic compensation for image replacement is a necessary part of the processing software.

[0036] 2. Manufacturing variations of several nm at the peak wavelength and FWHM specifications can have a significant effect on the EEC ratio. Filter error for long wavelengths can be fine-tuned by tilt, but this view is severely reduced in the case of emission filters due to increased image anomalies and worsened pixel shift. There is no equivalent method to compensate for short wavelength errors. 3. The need to keep the infrared light emitted by the arc lamp from reaching the detector.
Silicon CCDs are very sensitive in this area. The filter set for the blue and medium visible fluorophores was found not to require additional IR blocking, but the loss of image contrast due to false IR was significant for the red-infrared-near-IR fluorophores. Turns out to be a problem. Thermal filters routinely used in microscopes (eg, Schott BG-38 glass) are completely inappropriate to alleviate this problem. Therefore, extensive additional blocking was required. However, commercial interference filters for infrared blocking filters transmit poorly even at near UV and therefore cannot be inserted into the excitation path. Instead, it has been found necessary to place an IR blocking filter in the emission path. To minimize the loss of image quality due to the insertion of these filters in the image path, they are placed inside the CCD camera directly in front of the window. Indeed, for use with Cy5, Cy5.5 (Oriel # 58893; 740 nm cutoff)
Two interchangeable filters were selected, one for use with and one for Cy7 (Oriel 58895; 790 nm cutoff).

The first member of the set of fluorophores is a counterstained DAPI, which produces a weak G-like staining pattern. Five of the remaining six fluorophores can be used in combination to color the entire human chromosome set. All are available as avidin conjugates (for secondary detection of biotinylated probe libraries) or directly linked to dUTP (for direct labeling).

Thus, we identified a set of six fluorophores and corresponding optical filters, spaced 350-750 nm apart, which achieves high discrimination between all possible fluorophore pairs. These fluorophores include the preferred fluorophores of the present invention. They are listed below: 4'-6-diamidino 2-phenylindole (DA
PI), fluorescein (FITC) and the new generation cyanine dyes Cy3, Cy3.5, Cy
5, Cy5.5 and Cy7. Of these, Cy3, Cy3.5, Cy5 and Cy7 are particularly preferred. The absorption and emission maxima for each fluorophore are as follows: DAPI (absorption max: 350 nm; emission max: 456 nm), FITC (absorption max: 490 nm; emission max: 520)
nm), Cy3 (absorption maximum: 554 nm; emission maximum: 568 nm), Cy3.5 (absorption maximum: 581 n)
m; emission maximum: 588 nm), Cy5 (extinction maximum: 652 nm; emission maximum: 672 nm) and Cy
7 (extinction maximum: 755 nm; emission maximum: 778 nm). The complete characteristics of the selected fluorescent labeling reagent are W
aggoner, A. (Methods in Enzymology 246: 362-373 (1995)), the contents of which are incorporated herein by reference. In light of the foregoing, it will be readily apparent that other combinations of fluorophores and filters having appropriate spectral resolution can be alternatively used in accordance with the methods of the present invention.

D. Method for detecting fluorescent in-situ hybridization 1. Theory of Fluorescence Detection Of the various methods for contrast generation in site-specific labeling, fluorescence is arguably the most powerful because of its high absolute sensitivity and multi-parameter discrimination ability. Modem electronic cameras and state of the art optical filters used in combination with high numerical aperture microscope objectives can image structures labeled with as few as 10-100 fluorophores per pixel. Thus, fluorophore-tagged single copy DNA sequences as small as several hundred bases can be detected under desirable conditions. The availability of spectrally identifiable fluorophores allows the simultaneous imaging of several different targets in the same analyte, either directly or by combination or analog multiplexing. In principle, multifluorophore identification is based on differential excitation, differential emission,
It can be based on fluorescence lifetime differences, or more complex but still analyzable observables, such as fluorescence anisotropy. This discussion assumes epi-imaging geometry. Table 1 describes the symbols and operators relevant to the theoretical considerations of fluorescence.

[0040]

[Table 1]

[0041]

[Table 2]

A predetermined excitation rate at a pixel location p produces an acceptable fluorescence signal / noise ratio (defined as S / N = [signal average] / [variance due to all noise sources]) in a predetermined integration interval. Whether or not it depends on the number of fluorophore molecules in p, their quantum yield and photochemical stability, and the quantum efficiency and noise performance of the detector. In the limit of weak excitation (Ψ ⁽ λ ⁾ · σF ⁽ λ ⁾ << τ ⁻¹ ), the pixel p in the target space
The excitation rate of N molecules of fluorine F in

[0043]

(Equation 1)

Is as follows. Ψ5 ⁽ λ ⁾ is the spectral distribution of the excitation light passing through the focal plane of the microscope (photon nm ^-1 · cm ^-2 · s ^-1 ). It is Ψ _S ⁽ λ ⁾ · φ _S · f1 (λ) · R (λ)
· Α ² · μ (wherein, alpha = diameter of the collimated light from the diameter / condensing lens of the objective lens entering the pupil) indicated generally by. The integration is based on the non-zero extinction cross section σF (
Treating the entire band (i) with λ), this involves the equation:

[0045]

(Equation 2)

Is related to the molar extinction coefficient εF (λ) by In practice, ∫ _i Ψ ^{5 (} λ ^{, p)} · dλ may be measured with a bolometer detector, eg, a calibration microthermocouple array located at or near the focal plane of the objective lens. For a perfect noise-free detector and a non-bleachable fluorophore, the S / N of each pixel
Increases indefinitely as (detected photons) ^1/2 . The rapid increase in S / N below depends on the excitation intensity, but not the relationship between S / N and dose. The effect of the non-zero bleaching constant is to turn the t ^1/2 function into an asymptotic function that is in the form of a bleaching mechanism. However, since the bleaching rate and signal intensity are linearly related, the asymptote does not depend at all on the excitation rate, but the speed of the approach to the asymptote does. It can be seen that if finite camera noise is added to the photobleaching, the S / N rises to a maximum and then drops when the fluorophore is ejected. Here, both the kinetics and the peak S / N are dependent on the excitation rate; generally, the faster the excitation, the higher the maximum achievable S / N. However, for a co-cooled CCD camera, the dark noise is so slow that it is almost negligible on the bleaching timescale (typically a few minutes). A more important factor in determining the final S / N is the stray light background (particularly from non-specific luminescence and leakage of excitation light).

It should be noted that the microscope objective compresses the excitation beam, but non-confocal microscopes do not focus it at a point and are therefore independent of the fluorescence image resolution (non-interfering emitter). . The most common excitation source for fluorescence microscopy is the high pressure short mercury arc, whose spectrum extends from the UV to the middle region of the visible spectrum superimposed on a weaker thermal continuum (dominant wavelengths of 334.1 nm, 365.6 nm, 404.7 mm , 435.8 nm, 546.1 nm, 577.9 nm)
Pressure-expansion line. Many fluorophores have an excitation spectrum that acceptably overlaps one or the other mercury line. It has nothing else (the best known is FITC), but can be suitably excited by a continuum if a sufficiently wide excitation bandwidth is used. Another common light source is a high pressure xenon short arc, which produces a nearly uniform continuum from about 300 nm to over 900 nm. However, power / nm is almost everywhere below mercury continuum for the same arc wattage. When high intensity light without structure is required (eg, for fluorescence ratio imaging), high CW power or pulsed quartz halogen lamps can be about 45
Above 0 nm, the performance is better than xenon. Certain fluorophores are well suited for laser excitation (eg, Ar + Ａｒ488 nm for FITC, He-Ne＠632.8 nm for Cy5, semiconductor diode-excited YAG ＠ 680 n for Cy5.5).
m).

In single fluorophore imaging, the use of available spectral bandwidth is rarely strict. The excitation filter F1 and the dichroic beam splitter DB1
Usually, it can be chosen to produce an appropriate overlap between the source spectrum and the fluorophore excitation spectrum. If an arc line is available, the F1 band need not be wider than that line. If not, and if a portion of the thermal continuum must be used, the wider the F1 band, the larger the available excitation flux. However, when using low noise integrating detectors, the goal of high excitation efficiency generally derives from the need to exclude excitation light from the emission path. This limits how close the excitation and emission bands can be to each other, thus constraining the excitation bandwidth. For fluorophores with small Stokes shifts, high quality filters with steep skirts are required. The excitation filter must be strictly "blocked" on the long wavelength side and has no pinholes, scratches or light leakage around the edges.

Dichroic beam splitters are generally not considered to be “evolving” over bandpass interference filters, since the slope of their transmission <−> reflection transitions is much greater than the skirt slope of rare notch filters. Small, and if they fluctuate between intermediate states of partial reflectivity and transmittance, it means that large spectral intervals can exist. The primary purpose of a dichroic beam splitter is to improve the combined efficiency of excitation and emission, rather than limiting the wavelength response of the instrument.

The resolvability of overlapping fluorophores in an imaging microscope is critically dependent on the degree of excitation contrast that can be achieved (see C below). The variation with wavelength of the ratio of the extinction coefficients of the two fluorophores is the excitation contrast spectrum. It can be easily calculated from the digitized absorption spectrum. Due to the overlapping of the absorptions, their ratio spectra show different peaks or can grow indefinitely. In any case, usually useful (from a factor of 3-4 times up to 100 or more)
One can select an excitation wavelength that favors one fluorophore over another. Some difficulties in obtaining high contrast multifluorophore images are: a. The excitation wavelength required for high contrast imaging often differs from the absorption peak. Therefore, a high degree of intensity exchange can occur to obtain high signal contrast to other fluorophores.

B. From the above, no standard filter set is used. c. Arc source spectral lines useful for exciting a single fluorophore do not show high contrast discrimination for adjacent fluorophores. d. When using a broadband light source, or a continuous pair spectrum of an arc light source, the need for a narrow excitation bandwidth can reduce the excitation flux to a problematic low level. The constraint on the excitation wavelength can sometimes be relaxed for more effective excitation.

Collecting F fluorescence and imaging it with high efficiency on a detector is a major planning goal for light emitting optics. Operationally, it is even more important to have a more efficient emission path than an efficient excitation path. The reason is that inefficient excitation causes inefficient photobleaching. The only penalty is a long image integration time (assuming a low noise detector). On the other hand, any fluorescence loss on its way to the detector is indicative of photobleaching without a concomitant increase in the information content of the image.

The detector pixel p accumulates the signal (detected photons) at a certain ratio:

[0054]

(Equation 3)

The integration spans the bandwidth (ii) where the detector quantum yield φd (λ)> 0. G represents the efficiency with which the optical element collects the fluorescence and transmits it to the detector; it can be assumed to be a wavelength independent of the first order. The main factor of G is the numerical aperture (NA) of the objective lens, which determines the fraction of isotropically emitted fluorescence collected by the imaging diameter (§ = 1 / π · sin ⁻¹ (θ) = 1 / π · sin ^-1 (NA / n), where θ is
It is half the angle bounded by the objective lens from its focal point. For oil immersion lenses, NA = 1.3; n = 1,515; § = 0.328). The NA further determines the spatial resolution of the microscope as it measures the dimensions of the Fraunhofer diffraction pattern generated in the image plane by a point source in the specimen plane. Several "rules" are being adopted to specify the resolution of a lens, depending on how much overlap of the Airy disks of two adjacent objective lenses constitutes a resolution threshold. A commonly used Rayleigh criterion is r = 1.22λ / 2NA (eg, r = 0.24μ for the above NA = 1.3 lens operating at 500 nm).

For a noise-free detector, the image “noise” at p is determined by the statistical variance of the number S (p) of fluorescent photons detected at the time interval Δt. The only detector property that has any indication in this regard is its quantum efficiency φd (λ). In the absence of photobleaching (establishment of destruction / hit Πb = 0), S (p) = F (p
) · △ t, variance S (P) ^1/2 , that is, S / N = S (P) ^1/2 . Thus, the image quality increases infinitely with Δt, but at a gradual rate.

In the presence of photobleaching (Πb ≠ 0), S (p) is the integral of:

[0058]

(Equation 4)

(Where f (t) is the photobleaching decay function). S / N follows a somewhat complicated path
The asymptotic value corresponding to the total emission of the fluorophore. For single-molecule photobleaching
For this, it is an exponential function, ie, N (p, t) = N₀(1-exp- [k_b/ Σk] · t) (where k_b= Πb^-1Is the photobleaching constant, Δk is the excited state of F
Represents all other steps). The standardized asymptote in the first order system is k_bOnly for / Σk
Note that the intensity is independent of the intensity of the excitation. Therefore
, The extent of bleaching is exponentially related to the accumulated excitation dose, but independent of the pathway.
I am in charge. However, in practice, bleaching of fluorophores in solution is mechanistic and kinetic.
Theoretically, it can be very complex. Common mechanisms are, for example,¹O_TwoOr O_Two ^{Two -} Ring opening after peroxidation of the excited state of the fluorophore. This type of bleaching requires strict
By oxygenation or by oxygen radical scavengers (ie anti-fading agents), for example
Tertiary amines (p-phenylenediamine or DABCO)
Can be slowed down. Still, other (not yet fully characterized) irreversible
It is not excluded, including, but not limited to, reactive steps, such as reaction with impurities.

Non-ideal detectors are involved in many types of noise, and their detailed analysis is difficult to process. The simplest noise component is the so-called "dark current", ie the fluctuation of the flow of the thermally excited carrier in the detector. Assuming this noise is random,
It adds to the photon shot noise in an RMS manner. Thus, if the average light producing signal is Fs ⁻¹ and the average dark count is Ds ⁻¹ , then the S / N after time Δt is F (Δt) ^1/2 / (F + D) ^1/2 in and, S / N is increased still if a (△ t) ^1/2, but slower than the case of no noise detector. However, when light bleaching is present, the situation is quite different. In this case, the S / N is (△ t
At ^1/2 , it rises initially, but at some point reaches a shallow maximum, then begins to fall again when the fluorescence signal decreases, but the thermal noise power remains constant. In this case, it is in principle desirable to continuously monitor the S / N of the image and terminate when the S / N reaches a peak. Most commercial digital imaging systems do not provide for this. Fortunately, the state of the art cooled CCD cameras have very low dark noise per pixel (typically <0.01 electrons / s counterclockwise), and therefore, until the fluorophore is almost completely exhausted. S / N does not start to decrease. In fact, autofluorescence and stray light dominant performance is adequate before the noise threshold of the CCD is achieved.

The main planned goals for a single fluorophore imaging system are as follows: Achieving the appropriate rate of fluorophore (F) excitation. 2. Collect F fluorescence and image it on a detector with high efficiency and the required spatial resolution. 3. Preventing reflected and / or scattered excitation light from reaching the detector.

The design of the emission channel for a single fluorophore is straightforward. The dichroic beam splitter transition wavelength is specified for example in the excitation passband red at 20 nm. this is,
High level rejection of excitation light reflected and / or scattered from the specimen and / or microscope optics is assured. The emission filter cut-on is usually much steeper than the dichroic edge, and thus can actually be placed at the same time. The most effective emission filter is a long-pass element. A preferred filter of this type is Scott glass, which transmits more than 90% of the total fluorescence to its cut-on red, while transmitting very much other light (especially any excitation light passing through a dichroic beam splitter). High order, typically rejects> 10 ⁵ . However, in particular, the emission channel was "widely open" in the near-infrared using a silicon detector that showed high sensitivity there.
It is usually not wise to leave it.

The multiparametric imaging of the present invention not only increases the throughput of information about the system under observation, makes the use of biological materials more efficient,
Spatial and temporal correlations that would otherwise be difficult to establish could be specified.
If a large number of different objects must be visualized, two or more labels can be used in combination that allow discrimination between more object types than with spectrally identifiable labels. Some examples of multifluorophore imaging are as follows: a. Co-distribution of proteins in structures such as microtubule networks can be easily visualized using immunolabels linked to different fluorophores.

B. Multiple genes can be mapped simultaneously to a single metaphase chromosome spread by fluorescence in-situ hybridization (FISH). Such signals are usually indistinguishable on the basis of intensity alone and are usually morphologically identical (diffraction limit). However, they are easily identified by separate or combined multifluorophore labeling.

C. Identification of the same chromosomal translocation is accomplished by coloring using a chromosome-specific DNA probe library linked to a separable fluorophore used alone or in combination.
Most easily done. d. Analysis of a mixed population of morphologically identical bacteria can also be achieved using a species-specific DNA or ribosomal RNA probe conjugated to a separable fluorophore.

The main goal of the multi-fluorophore imager (in addition to that for a single fluorophore imager) is to spectrally decompose the fluorescence at every pixel location into components corresponding to each fluorophore. It is. A method for spectrally decomposing a composite signal in a fluorescence microscope image is briefly described below. There are several ways to spectrally resolve the composite signal, ie, to determine that the fluorophore is responsible for the fluorescence at a given pixel location. The most common method is in principle to spectrally disperse the 2D image along a third axis orthogonal to the x, y axes. This basically means that at each position on the z-axis, a small spectral bandwidth λ + λ
That is, imaging is performed by an optical system having a large amount of chromosomal abnormalities such that images x and y containing only λ exist. Next, an area detector having a very small depth of field (eg, a spatially filtered confocal imager) is incrementally moved along the z-axis to a group of each containing a small spectral interval itself. Get an image. The trace through the image at a given x, y constitutes the emission spectrum for that pixel. Unfortunately, implementing such a plan is technically very difficult.

If a spectrum of only a small number of objects in the image is required (such as individual stars in a telescope image), the solution is to use probes (eg fiber optics) to address each object. Extracting light and dispersing it on a one-dimensional array detector using an imaging spectrograph. In order to be useful in a microscope, such a device must be arbitrarily positionable in the field of view and has an adjustable receiving area.

The most rational method for full spectrum analysis under a microscope is to image through a tunable narrow band filter. Images are recorded at each wavelength. The intensity values at a given pixel location throughout the series represent a metric emission spectrum that can be fitted to a linear combination of known spectra of the constituent fluorophores. The count is the product of their extinction coefficient at the excitation wavelength and their fluorescence quantum yield and the relative molar amount of the fluorophore. Knowing the last two gives the first one from the fitness. In general, it is necessary to obtain several such image sets at several excitation wavelengths to achieve a unique fit. With sufficient repetition, the method yields a 3D surface with intensity values as a function of both excitation and emission wavelength. This encompasses the full spectral properties of the pixel and yields a very highly constrained solution for the relative amounts of its constituent fluorophores. Mapping the molar ratio back onto the x, y coordinates of the image using the appropriate pseudo-color coding gives a "composition map". This general (and quantitative) method has many technical difficulties, but none of them is insurmountable. The first is
It takes multiple images to evaluate a single microscope field. Imaging times are long, and extensive differential photobleaching of the fluorophore makes it impossible to achieve a self-consistent "fit" with the spectral data. Instrument stability is also a problem, especially when using arc light sources whose output spectrum varies over their lifetime. Finally, the amount of computation required to generate a composition map effectively limits the analysis to only small image areas.

For most applications, the fluorophores to be analyzed are known in advance, and do not require any full blown spectral analysis capability. Thus, it is only necessary to be able to identify which fluorophore is which and, for multiple imaging, be able to ascertain the relative amounts of each present. A preferred approach to the problem in this field involves the use of filter sets to achieve a high degree of selective excitation and visualization of each component fluorophore. An ideal system of this kind preferably isolates each fluorophore, i.e. "one image-one fluorophore", and has zero off-diagonal elements in the matrix of intensity counts. However, the fact that the spectrum of a typical fluorophore is 10-30% of the total available bandwidth, and the fact that the emission filter must have a significant bandwidth to allow a usable amount of light, is due to the spectral isolation. Place strict limits on the achievability of Nevertheless, it is not difficult to achieve useful levels of contrast between fluorophores that are properly selected such that residual crosstalk can be numerically eliminated.

The Excitation-Emission Contrast (EEC) approach, in principle, involves a large number of fluorophores exhibiting a fine distribution of molar fractions, subject to limitations in image S / N and differential bleaching rate and light source instability. The present invention can be applied to the analysis of an image (eg, fluorescence ratio imaging). However, it is possible that the fluorophore is strictly Boolean (presence = 1 or non-presence =
The limitation of the combination labeling multiplexed in 0) is particularly simple. In this case, it is only necessary to be able to reliably discriminate between 1 and 0, and an intensity ratio between any fluorophore and its neighbors of 10: 1 is sufficient. However, for many useful fluorophores, this is not achievable on the basis of excitation or emission contrast alone. Simultaneous selection based on both excitation and emission is required.

In some cases, it is necessary to excite and image several fluorophores simultaneously, for example, for direct viewing and / or color video recording. Excitation and emission optics without wavelength selectivity cannot be used because the excitation light scattered in the detector overwhelms the fluorescence by several orders of magnitude. One solution is to use a multi-band filter intended for the particular set of fluorophores used. The excitation filter defines a narrow band that overlaps the fluorophore excitation spectrum. The emission filter penetrates between the excitation bands and defines a similar band that overlaps the fluorophore emission spectrum (the reddest fluorophore may use a long pass filter). Dichroic beam splitters alternate between reflection (overlapping excitation bands) and transmission (overlapping emission bands).

The use of a multi-pass filter set is beneficial even when multi-color visualization is not required, ie, when there is no geometric permutation (pixel shift) between signals passing through several emission bands. . However, when used with a grayscale imaging system, it is necessary to be able to switch the excitation or emission filter to determine which fluorophore the grayscale signal corresponds to. A preferred choice is to switch the excitation filter so that this does not cause image replacement.

A second theoretical solution is to excite at a wavelength that all the fluorophores absorb. This is often possible because a large number of fluorophores can be excited to a higher state than S1 with photons at mid-UV, but because the internal relaxation step results in "normal" fluorescence. For example, many laser dyes can be excited with a nitrogen laser wavelength of 337 nm, far from their visible absorbance of blue. It is directed straight to block such excitation light from the emission path using a long-pass filter (eg, 380 nm) while allowing all the fluorescence to reach the detector simultaneously. Disadvantages to the use of UV excitation include an increased rate of photochemical degradation of the fluorophore and the
V optical elements are expensive. Therefore, the method has not found wide application.

The multi-band method has the limitation that it is very difficult to construct a multi-band device that produces adequate contrast between more than three fluorophores. A generally stronger approach is
Build optimized filter sets for each fluorophore and switch them if necessary. In the case of a single fluorophore, the primary target on the excitation side is a high excitation flux to produce a bright image. However, when integrating multiple fluorophores, this comes down to the goal of achieving high contrast between the fluorophores. The weak overlap restriction results in a fluorescent switch by eventually switching a filter designed using the same criteria as a single fluorophore set, except that longpass emission filters are prohibited for all but the longest wavelength fluorophore. The troupe can be imaged. A few percent crosstalk is usually acceptable and can be compensated numerically if necessary. More generally,
Again, the creation of adequate contrast between the fluorophores requires both highly selective excitation and emission. The optimum wavelength can be found by indicating the spectrum of the adjacent fluorophore as a ratio. These wavelengths are far from the excitation maximum, which implies a significant trade-off between signal brightness and contrast. Standard filter sets cannot be used. The need for tightly-located excitation reduces the likelihood of strong excitation lines being available and reducing the available flux from the continuum source. Broadband light source (
Arc or filament lamp) temporarily to a very high power level,
Or it is necessary to supplement it with a laser light source.

A major technical problem with continuous imaging is image replacement when changing filters, ie, the coordinate system of the individual members of the image series does not exist in exact representation. This problem arises from non-ideality in the emission channel optics. Two replacement components can be identified: i. Unique repeatability offset for each filter set. This component is a fixed vector, mainly due to the imperfect parallelism between the top and bottom of the emission bandpass filter (
I.e. wedging). There is also a small component due to wedging in the dichroic beam splitter, but the effect is small since this element is so thin. Since the wedging vector is constant for each filter set, it can be automatically removed in the computer. The size of the offset can be reduced to very small values (<0.1μ) by choosing the emission filter for a high degree of parallelism, for example by measuring with a laser autocollimator.

Ii. Random components due to wave and hysteresis in the filter switching mechanism. The magnitude of the noise depends on the filter switching mechanism. The worst are manual push-pull sliders, especially when they are operated by the operator using uncompensated forces. Best is a motorized filter cassette in which all mechanical torque acts on the microscope body.

It should be noted that both constant and random components of image offset noise are minimized by using the highest possible magnification objective lens, and the least amount of magnification in the camera projection optics. Image replacement in the case of an optical microscope is only a linear vector. There is no evidence for significant changes in rotation or scale.

The design issues for high contrast imaging of multiple overlapping fluorophores are similar to those already outlined for the excitation channel. By calculating the ratio of the emission spectrum of each fluorophore to the spectrum of its neighbors, it is possible to identify high-contrast spectral regions that can be limited using narrowband filters. High contrast is often achievable at the expense of wasting significant amounts of fluorescence (greater than 90%).
Inefficiencies in the fluorescence channel are more damaging to the image S / N than excitation inefficiencies. Thus, where possible, selective excitation is the preferred method of achieving contrast between multiple fluorophores. However, in difficult cases, both excitation and emission contrast are required.

2. Optical Filters As mentioned above, in a preferred embodiment of the invention, the detection of the fluorophore is performed by Ried, T. et al. (Proc. Natl. Acad. Sci. (USA) 89: 1388-1392 (1992)). The description is effected using optical filters, in a variant of the method described herein, which is incorporated herein by reference.

DAPI Imaging 4 ', 6-diamidino-2-phenylindole (DAPI) is a chromosome rich
Commonly used to preferentially insert into the AT region to produce a weak G-type staining pattern
DNA counterstain. It is a very bright fluorophore (ε = 3.3 x 10 at 347 nm^Four M^{- 1} cm^-1. When inserted into the minor groove of double-stranded DNA, the fluorescence quantum yield is increased by about 20 times).
And is relatively resistant to light bleaching. The following points are the high contra
Related to imaging DAPI with strikes: a. DAPI has a very broad excitation and emission spectrum, and
Has a large Stokes shift. Therefore, the DAPI excitation maximum (347 nm) is
The color of the cascade blue (CB) becomes blue, and the fluorescence of the DAPI peak is
Lou turns red and, in fact, overlaps very strongly with FITC.

B. Normal excitation wavelengths (Hg 366 nm line) for DAPI are DAPI and CB
Is almost absorptive with respect to, and thus produces no excitation contrast,
Not used for this purpose. c. The peak of the DAPI / CB excitation contrast spectrum is 320 nm,
This is too far to UV to pass through the microscope optics. An acceptable compromise is that the microscope is excited with a Hg 334.1 nm line that is transmitted with about 30% efficiency. Ealin
g 35-2989 interference filter has the appropriate band. DAPI / C at 334.1 nm
The excitation contrast ratio for B has an absolute value of 4.0.

D. To further increase the DAPI / CB selectivity, the emission contrast
Must be used in addition to the excitation contrast. Note that DAPI emits a large amount of its fluorescence in the red of the CB. DAPI vs. CB and FI
The wavelength of maximum emission contrast for both TCs is 490 nm. A suitable imaging grade filter is Omega 485DF22. No mercury rays were found in that zone,
Therefore, a good DAPI image with low flare is expected. The calculated emission contrast between DAPI and both Cascade Blue and FITC is 6.8. Therefore, the overall achievable contrast for DAPI versus Cascade Blue is about
27 times. The overall contrast of DAPI vs FITC is much higher than this,
This is because at the DAPI excitation wavelength, the excitation of FITC approaches zero. Omeg
a 450DRLP02 dichroic beam splitter is very well adapted to the proposed excitation and emission filters.

Cascade Blue Imaging Cascade Blue (CB) has a broad 2-peak excitation spectrum that overlaps DAPI extensively, but not its neighbor on the red side (FITC). The Stokes shift for CB is very small, ie there is a very wide overlap between its excitation and emission spectra. These factors are the DAPI in question
To create a visualized CB in the presence of In summary: a. The peak of the CB / DAPI excitation contrast spectrum is at 396 to 404 nm. However, due to the very small Stokes shift of the CB, this
Very close to the emission contrast peak (408 nm). Since it is more important for the image S / N to collect light emission with high efficiency than to excite it with high efficiency, the region of 400 to 410 nm
Reserved for B fluorescence. The best compromise for CB excitation is Omega 38OH15, which overlaps with a sufficient Hg 366 nm line to provide good excitation intensity.

B. In order to achieve any emission contrast, all-over-DAPI, the emission filter must be narrow and placed very carefully. A suitable filter is Omega 405DF10, but the theoretical maximum contrast is only 2.5. Therefore, the imaging CB is expected to be peripheral while eliminating DAPI.

C. Cascade blue shows high excitation contrast to FITC (Omeg
a Contrast ratio by 38OHT15 filter = 6). d. The emission contrast for CB / FITC is very large, less than 490, and is essentially infinite at the location of the 405DF10 filter. The above discussion indicates that it is possible to use DAPI or Cascade Blue, but not both together unless the DAPI counterstain is weak. Cascade Blue is well separable from FITC and the other five combination fluorophores discussed herein.

FITC Imaging The following points relate to high contrast imaging of FITC: a. The excitation spectrum of FITC has a non-significant overlap with the spectrum of Cascade Blue (contrast parameter Rb for FITC / CB is 420
very large beyond nm). This allows excitation of FITC at the high frequency shoulder of its absorbance, thus avoiding apparent excitation of Cy3.

B. No mercury line which is well overlapped with the excitation spectrum by FIT is observed. Therefore, as in the case of single fluorophore imaging, it is necessary to use a continuous pair to excite this fluorophore. The gap between FITC and CB allows the use of very wide excitation filters. Omega 455DF70 is well suited for this function. c. The Omega 455DF70 band also coincidentally corresponds to the maximum in the excitation contrast ratio for FITC / Cy3 (460 nm: absolute excitation ratio _Ra = 8.8).

D. The emission spectrum of FITC overlaps the Cy3 (red side), and the DAPI and CB (blue side) spectra to an appreciable extent. However, FITC
They are irrelevant because the limited conditions for excitation do not excite DAPI or Cascade Blue at all. FITC / Cy3 emission contrast ratio is 540
The value is relatively high below nm. Thus, the Omega 530 DF30 filter produces a very high emission contrast, which is the FITC vs. C
Compatible with high excitation contrast ratio (8.8) for y3. Therefore, it is considered that FITC can be imaged very clearly.

Cy3 Imaging The following points are relevant for high contrast imaging of Cy3: a. The absorption peak of Cy3 is 551 nm, at which wavelength the excitation of FITC is essentially zero (the contrast parameter Rb jumps to very high values in the red direction at 525 nm).

B. The Cy3 excitation peak strongly overlaps with the Hg 546.1 nm line. c. The excitation contrast for Cy3 / Cy3.5 is small everywhere and varies weakly with wavelength. At 551 nm, the absolute value of the excitation contrast for Cy3 vs. Cy3.5 is less than 2, because the Cy3 class altitude is very low and FI
It only rises significantly away from blue when the TC absorbance is high. In fact,
An apparent increase in _Ra of about 460-470 nm may be a low absolute accuracy artifact of the spectrum in that region. From the above discussion, it is clear that the available excitation contrast for Cy3 versus Cy3.5 is too low to be useful.

D. The emission contrast ratio for Cy3 / Cy3.5 increases unexpectedly below 570 nm. The 567 nm peak of Cy3 fluorescence, the absolute value of S _a is about 6. this is,
Combined with a factor of 2 in the excitation contrast parameter, it should almost meet the goal of a 10-fold discrimination between Cy3 and Cy3.5. The Ealin 35-3722 narrow band interference filter is suitable, but it significantly overlaps the Hg 577/579 line. Any stray light from this line coming through the 546DF10 filter, however, is expected to be well blocked by the selected dichroic Omega 560DRLP02.

E. The inability to differentially excite Cy3 versus Cy3.5 implies uneconomical bleaching of Cy3.5 during Cy3 imaging. Cy3.5 Imaging The following points summarize the high contrast imaging with this dye: a. The excitation contrast ratio between Cy3.5 and Cy3 rises to very high values above about 565 nm. This region contains the peak of the Cy3.5 excitation spectrum.

B. The mercury arc line at 577/579 nm almost exactly coincides with the Cy3.5 class peak. At this wavelength, the excitation contrast ratio is about 25, ie very strong selective excitation of Cy3.5 compared to Cy3 is possible. The ideal filter for this purpose is Ealing 35-3763. Note that high contrast is primarily a result of the Hg line position, not the filter band.

C. At the Hg 577/579 excitation wavelength, the excitation contrast ratio for Cy3.5 is also much higher than Cy3 (absolute value about 8.0). d. The emission contrast parameter for Cy3.5 vs. Cy3 is small at all wavelengths, in which case the Cy3.5 emission is usefully strong, ie, the isolation of Cy3.5 from Cy3 must mainly depend on the excitation contrast.

E. The emission contrast for Cy3.5 vs. Cy5 is also large over significant spectral spacing (and rises to very high values below 640 nm). This allows the considerably broader filters used to image Cy3.5. A suitable device is Omega 615DF45. Surface bleeding of Cy3 or Cy5 into the Cy3.5 channel is hardly predicted using the above combination of excitation and emission conditions.

F. Omega 590DRLP02 is a suitable dichroic for this channel. Cy5 Imaging The following points are relevant for imaging with this dye: a. Above 600 nm, the excitation contrast ratio for Cy5 to Cy3.5 is very large, ie a very high excitation contrast compared to Cy3.5 is conceivable.

B. The configuration of the Cy5 excitation is determined by the excitation ratio compared to Cy5.5 rather than Cy3.5. The Cy5 / Cy5.5 excitation contrast parameter is 2.25 at 650 nm.
The peak is reached with the numerical value. Therefore, like the Cy3 / Cy3.5 pair, Cy3
The excitation contrast for 5 / Cy5.5 is insufficient. c. No Hg line is available for excitation of Cy5. Thus, with an arc light source, a continuum must be used, similar to FITC excitation. Cy in this way
A "certified" filter for exciting 5 is the Omega 640DF20, which produces an excitation contrast with Cy5 of about 1.8.

D. A very bright light source for exciting Cy5 is a He-Ne laser (632.
8 nm). However, it does not improve the excitation contrast for Cy5.5. e. The emission contrast for Cy5 versus Cy3.5 peaked at 673 nm,
The red side of the fluorescence intensity peak. The closest available filter is the Omega 660DF3
In this case, the emission contrast ratio is about 3.1. This trades off with the very high excitation contrast for Cy5 versus Cy3.

F. The emission contrast for Cy5 vs. Cy5.5 becomes very large at wavelengths shorter than 675 nm. Omega 660DF32 filters are ideally tuned and receive this full benefit. Unfortunately, it is unpleasantly close to the 640DF20 exciter, so some flare from the reflected / scattered excitation light is expected. The use of He-Ne laser
Eliminate this problem.

G. The best available dichroic beamsplitters for Cy5 coupling are especially H
When e-Ne is used as the excitation light source, it is Omega 645DRLP02. Cy5.5 Imaging Cy5.5 is the penultimate dye in the set and is very well separated from Cy7. Therefore, only its contrast compared to Cy5 needs to be considered in detail: a. The contrast parameter _Rb for the Cy5.5 / Cy5 pair increases to a large value in the red direction at 670 nm. Thus, very high excitation contrast for this pair of fluorophores (as well as Cy3.5 / Cy3) can be achieved.

B. As with Cy5, the Hg arc is an insufficient light source to excite Cy5.5. The best available light source is a 680 nm diode-excitation frequency doubled YAG microlaser (Amoco), which is consistent with the Cy5.5 class peak. 68
At 0 nm, the excitation contrast ratio for Cy5.5 / Cy5 is 5.1. A suitable excitation filter is Ealing 35-4068.

C. The emission contrast ratio between Cy5.5 and Cy5 peaks at 750 nm, approximately 3 nm in the red direction of the Cy5.5 intensity curve. Figures for S _b at this point is 4. If an Omega 700EFLP longpass emission filter is used, approximately 3 (800
A contrast ratio of (average to nm) is expected. This is combined with high excitation contrast. Makes the picture of Cy5.5 very beautiful.

D. If a slightly lower emission contrast (which is not significant) and some loss of intensity is costly, a bandpass filter such as Ealing 35-6345 can be used with an Omega 700E
Can be used instead of FLP. A possible advantage is the reduction of the infrared background, ie the overall improvement of the image contrast. Cy7 Imaging Cy7 is the reddest dye in this set. The excitation and emission spectra are Cy
Well separated from 5.5 and well compatible with Omega 740DF25 / 770DRLPO2 / 780EFLP triplets. Oriel 58895 is a suitable IR blocker for Cy7.

The filters selected for imaging the DAPI, FITC, Cy3, Cy3.5, Cy5, Cy5.5, Cy7 fluorophore set are summarized in Table 2 below. Since narrow band excitation and fluorescence detection are essential to achieve sufficient contrast, none of these filter sets correspond to those provided by conventional fluorescence microscope manufacturers.

[0105]

[Table 3]

The features of the microscope system are described in Ballard SG et al. (J. Histochem. Cytochem. 41: 175
5-1759 (1993)), the contents of which are incorporated herein by reference. Due to its nearly constant spectral power distribution, high pressure 75
W DC xenon arc (XBO) was used as the excitation light source. Cooled CCD camera (P
A Zeiss Axioskop microscope workstation equipped with hotometrics CH250) was used. The objective was a 63x1.25 NA Plan Neofluar, a "plan" with high numerical aperture and apochromat. The selected filter set was able to distinguish between the six fluorophores with maximum contrast (Table 2). To minimize crosstalk between fluorophores, the filter set was selected based on maximum spectral discrimination rather than maximum photon throughput. The photon flux difference and the flux excitation cross section were adjusted by changing the image exposure time. Narrow band excitation and fluorescence detection are essential to achieve sufficient contrast. Using appropriate excitation and emission filter sets, these fluorophores were optically distinguished (Table 2).

The combinatorial labeling strategy relies on accurate measurement of intensity values for each fluorophore. Key features are accurate alignment of the different images, correction of chromosomal abnormalities, and specific quantification of each fluorophore. Since simple manual image manipulation failed to fulfill these requirements, new software was developed in our lab. This program sequentially includes the following steps: (1) correction of geometric image shift; (2)
Calculation of DAPI segmentation masks; (3) calculation and subtraction of background intensity values, calculation of thresholds and creation of segmentation masks for each combined fluorophore; (4) each fluorescence to establish a "Boolean" property of each probe. Use of this segmental mask of the group; (5
) Display of chromosomal material next to the DAPI image for each chromosome; (6) Preparation of composite gray value image. In this case, each labeling object is coded with a unique gray value; (7) Final presentation of the results using a look-up table (LUT) that assigns each gray value to a specific color.

An image analysis package (BDS) running on the Macintosh Quadra 900 ™
The above program was developed on the basis of the image). Waggoner, A. et al. (Methods Ce
ll Biol 30: 449-478 (1989)), and modifications (du Manoir,
S. et al., Cytometry 9: 4-9 (1995); du Manoir, S. et al., Cytometry 9: 2
1-49 (1995)), the contents of which are incorporated herein by reference, due to the alignment of the center of gravity (center of mass) of a single chromosome in each image, due to optical and mechanical defects. The induced image shift was corrected. The morphological boundaries of each chromosome were defined using DAPI images. Precise chromosome segmentation was achieved by pre-filtering the image through a top-hat filter (Smith, TG, et al.,
J. Neurosci Methods 26: 75-81 (1988); du Manoir, S. et al., Cytometry 9
: 4-9 (1995); du Manoir, S. et al., Cytometry 9: 21-49 (1995)).
The method of the gray level histogram of the top hat filtered DAPI image was used as the threshold. For each fluorophore, background was excluded by subtracting the average interchromosomal fluorescence intensity from the image. The average of the chromosomal fluorescence intensities was used to calculate a threshold for each segment mask of each fluorophore. Individual DNA targets were assigned different gray values depending on the "Boolean" characteristics of each probe, ie, the combination of fluorophores used to label this DNA probe. In the final step, a pseudo-color was assigned to each DNA target for display by this gray value using a look-up table.

The switching of the filter sets in the excitation and emission paths, and the dichroic mirror were performed manually, but the computer-controlled electromechanical solution allows the automation of the procedure. 3. Optical Combs In an alternative embodiment, selective excitation and analysis of fluorescence spectral properties of multiple fluorophores can be performed using dispersive optics rather than wavelength-selective transmission filters. Using such an optical element, a filter having any band characteristics, such as short-pass, long-pass, single-band and multi-band functions, can be produced. In this method, a dispersive element (prism or grating) is used with a wavelength selective spatial filter to produce the desired spectral response. The combination is referred to herein as a "comb filter." Comb filters can be used to tailor the spectral distribution of the excitation light for optimal simultaneous excitation of multiple fluorophores. An inverse comb filter can also be used to selectively block from the CCD camera only the wavelengths used for excitation. The remaining wavelength spacing (corresponding to the gap between the teeth of the comb) is available for spectral analysis of the fluorescence emitted by the fluorophore. This analysis constitutes a spectral characteristic.

[0110] 4. Interferometer Instead of using the optical filters described above, an interferometer can be used with an epi-fluorescence microscope. A light source for excitation of the fluorescence, which may be coherent (eg, an argon laser) or incoherent (eg, a mercury arc lamp), may be used. Due to its strong mercury line and wide xenon visible and infrared continuum, mercury-xenon mixed gas arc lamps are preferred.

[0111] Any of a variety of interferometer designs (eg, a Michelson interferometer) can be used, but the use of a Sagnac interferometer is preferred. Sagnac interferometers have larger acceptance angles, larger entendues than similar Michelson interferometers, and are less sensitive to alignment, vibration and temperature fluctuations. Sagnac interferometer is a common path interferometer. An interferometer consists of two or more interfering beams of light. In a common path interferometer, there are two beams, each traveling on the same path, but in opposite directions. The optical path is created, for example, by reflecting light through a beam splitter.

Multi-beam interferometers operate by splitting the optical energy from a light source into two substantially equal beams of light. The two beams of light are merged after one is passed through the sample, and an interference pattern (a change in the intensity of the merged light caused by the interference of the two beams) is detected. In a Sagnac interferometer, the light source is also divided into two substantially equal parts. Changing the angle of the incident light on the beam splitter (by rotation of the interferometer, or rotation of an optical element, such as a galvanometer drive mirror in the interferometer), causes the optical path length to be along one of the optical axes of the interferometer. be changed. This produces a fringe pattern along one axis of the detector, eg, a CCD detector. The other axis of the detector may sample grayscale. As the optical path length is scanned by rotating the interferometer or mirror, the fringe pattern produces an interference image at each pixel. The Fourier transform of this interference image is CC
D produces a spectrum of light falling on that pixel. Thus, an advantage of the Sagnac interferometer is that it produces an optical path difference across the field of view rather than a single point.

Modulation, such as one small shift of the optics, affects both beams in the same way and therefore does not affect the measurements. This mechanical stability also makes the interferometer relatively insensitive to temperature changes. Therefore, another advantage of the common path interferometer is that
Its inherent stability. Sagnac interferometers and their use are well known (e.g., U.S. Patent Nos. 3,924,952, 4,410,275, 4,529,312, 4,637,722,
4,671,658, 4,687,330, 4,836,676 and 5,108,183).

One implementation of the Sagnac interferometer (J. Bruce Rafert et al., “Monolithic Fourier-Tran
In the "sform imaging spectrometer", Applied Optics, November 1995), the acceptance angle of the interferometer is determined as follows:

[0115]

(Equation 5)

Where w / a = tan30 °, w is the interferometer aperture width, a is the length of each foot, and n is the refractive index of the interferometer glass. The incident beam on the interferometer does not need to be collimated. The interference pattern or image can be a 512 × 512 pixel or larger CCD camera (eg, Princeton Instruments Frame Transfer CC).
D camera). Since the interference image in the Sagnac interferometer has an angle dependency, each pixel of the CCD detector measures a small interval of the interference image. The fringe interval of the interference image is set so that pixels on the CCD detector can appropriately sample the interference image. The optical path difference (OPD) that can be measured by a pixel is expressed by the following relationship in order to properly sample the interference image:

[0117]

(Equation 6)

(Where λ min is the shortest wavelength of the spectrum measured by the interferometer). This OPD pixel establishes the theoretical limit on the resolution of the interferometer. Since the OPD is changed by the rotating mirror, the interference image is moved through the CCD detector, and thus the maximum optical path is given by the following relationship:

[0119]

(Equation 7)

(Where N is the linear dimension of the CCD detector at the pixel). Each angular displacement of the light incidence on the interferometer beam splitter then corresponds to one or several OPD pixels. And in the case of a CCD detector, one frame of CCD data is needed to sample this angular displacement. Finally, each pixel contains an interference image containing information about the spectrum of light falling on the pixel, the intensity of light falling on the pixel, and the x and y coordinates of the pixel. The spectrum of the light can be recovered from the interference image by use of a computerized Fourier transform algorithm.

In fact, due to the limited dynamic range of the CCD, typically around 10,000: 1, the light used to excite the fluorescence must be blocked from entering the interferometer. This excitation light is often 10 ⁸ -10 ¹² or more intense than the fluorescence emitted from the sample. If not blocked by the use of optical filtering, this excitation light will saturate the CCD. However, this filter only needs to be fabricated to block the excitation, and can pass all other wavelengths.

In one embodiment, the fluorescent probe is excited using ultraviolet (UV) radiation. UV light is easily blocked by long-pass interference filters that can pass visible and near infrared colors through the interferometer. This embodiment has the advantage that UV excites many of the commonly used fluorescent dyes. This embodiment also has the advantage of allowing good transmission of visible fluorescence of more than 90% to the interferometer. A disadvantage of UV is that it photobleaches the dye faster than visible light.

Both the input and output lenses of the interferometer are preferably extremely high efficiency camera lenses and do not significantly affect the efficiency of imaging. The focus of the image in the interferometer is most preferably adjusted to be constant with respect to the variable power of the zoom eyepiece, and therefore a microscope (eg, Olympus AX70 microscope) with the feature of infinite imaging distance is preferred.

The interferometer has certain advantages over optical filters. One important advantage is that all the light emitted by the fluorescence is theoretically available for detection, while the transmission through the interference filter is reduced. Another advantage is that
Since the filters do not have to be changed, there is no image shift due to the relative parallelism of the filters.

E. Uses of the invention The FIS of the invention for detecting and analyzing chromosomal abnormalities such as translocations, inversions, duplications, etc.
The ability of the H (fluorescence in-situ hybridization) method and reagents can be used for a number of applications. 1. Uses Related to Cytogenetic Diagnosis of Genetic Disorders The main uses of the present invention include cytogenetic diagnosis of genetic disorders, such as prepartum or postpartum diagnosis of the disease, classification of multiple tumor types, There is an analysis of the locus, especially through the use of telomere specific probes. It provides a method for automatic chromosome identification and analysis. Numerous diseases (prepartum diseases, cancer (especially BRCA1 or BRCA2)
Associated breast cancer), leukemia, Down's syndrome, etc.) are characterized by rearrangements and other chromosomal abnormalities that can be recognized using the methods of the present invention.

[0126] Chromosome karyotyping by conventional cytogenetic sorting involves both time and expense and is not easy to automate. The detection of recurrent gene changes in solid tumor tissue by karyotyping is particularly problematic because of the difficulty in routine preparation of metaphase spreads of sufficient quality and quantity, and the complexity of numerous chromosomal changes. Second, it is very difficult to identify marker chromosomes based only on the staining pattern. In fact, attempts to automate karyotyping (eg, pattern matching, proprietary analysis) over the past two decades have shown that robust computer algorithms can reliably decode complex dissemination patterns, especially on a wide variety of rearranged chromosomes. It failed because it was not developed.

The next generation of cytogenetic techniques has been proposed to be much better by using bands that are molecularly limited by hybridization of probes or probe sets, each labeled with a different color. (Nederlof, PM et al., Cytomet
ry 11: 126-131 (1990); Nederlof, PM et al. ,, Cytometry 13: 839-845 (199
2); Lengauer, C. et al., Hum Mol Genet 2: 505-512 (1993)). This provides a high versatility and constitutes a spike in the abundance that is well comparable to the introduction of chromosome segregation and high resolution analysis of chromosomes in the pre-metaphase. The advantage of the FISH karyotype is the immediate identification of the chromosomal origin of the marker chromosome, the DM and homologous staining regions ("HSR").

“Low quality” chromosomal spreads can also be evaluated. If desired, probe sets can be designed for a particular use or for a particular clinical use, for example, for hematological disorders, pre- or postpartum diagnosis. The development of specific probe sets that stain specific regions of the chromosome (eg, telomere regions) for the identification of potential translocations overcomes the limitations of whole chromosome coloring probes. Similarly, such probes can be used to generate multicolor "barcodes" on individual chromosomes, thereby facilitating automated karyotyping.
Probes can also be designed that are specific to a particular arm of the chromosome, thereby permitting the molecular characterization of translocation breakpoints, recombination hotspots, and the like. Multicolor F on human chromosome
Protocols for ISH include, as expected, rapid evolutionary testing where other uses may be tailored for applications in other species.

[0129] 2. Uses for the Detection of Infectious Agents The methods of the present invention are useful for detecting infectious agents in tissues or in blood or blood products.
Treponema pallidum, Rickettsia, Borrelia, Hepatitis virus,
HIV, influenza virus, herpes, group B streptococci, agents causing diarrhea, pathogens causing acute meningitis, etc.). This can be achieved by using labeled probes specific for such agents. In addition, by using serotype-specific probes, the methods of the present invention allow for rapid serotyping of such agents, or for determining whether any such agents carry drug resistance determinants. To The methods of the present invention can be used to assess chromosomal abnormalities caused by exposure to radiation (eg, personnel exposed to nuclear fuel plant radioactivity).

The methods of the present invention can be used to quantify microorganisms that are difficult to grow (eg, anaerobic microorganisms involved in periodontal disease). The method of the present invention provides a means for rapid diagnosis of acute bacterial meningitis. Just as a serotype analysis could be performed with a serotype-specific probe, a probe specific to a particular drug resistance determinant could be used, thereby only the presence and identity of the infectious agent. But its susceptibility or resistance to a particular antibiotic can also be quickly determined.

[0131] 3. Additional Applications The method of the present invention further allows for the simultaneous mapping of many different DNA probes. This technique facilitates the analysis of chromosome numbers and constructs in individual intact cells. Interphase cytogenetics has already been enabled with small region specific probes, such as YAC-clones. The accuracy of such analysis can be increased by three-dimensional analysis using a laser scanning microscope, more preferably a CCD camera system with a Z-axis stepper motor coupled with three-dimensional (3-D) image deconvolution software. . Suitable 3-D video deconvolution software is available from Imstar Corp. (Paris, France) or Scanalytics, Inc. (Bost
on). In addition, the use of such a scanning microscope system can ultimately visualize all chromosome-colored probes in the interphase nucleus, as well as all telomere- or centromere-specific probe sets, and the developmental state, cell cycle or To address issues related to nuclear chromosome organization as a function of the disease state. Finally, different models for chromosomal organization in interphase nuclei can be explored. Conventional laser scanning microscopes do not excite some of the commonly used fluorophores, but more appropriate fluorophores or tools, such as three-dimensional deconvolution imaging systems, can be used.

When extended to non-dividing cells, the methods of the invention allow one to examine nuclear chromosomal constructs or quantify chromosomal content in a single hybridization experiment. Thus, issues related to nuclear chromosome organization as a function of developmental state, cell cycle or disease state may be addressed. In addition, the ability to quantitatively assess the levels of multiple mRNAs or proteins in a single cell, or to determine whether they exhibit different subcellular distributions, is extremely difficult when dealing with a myriad of interesting biological problems. It is found to be useful. The multiparametric imaging of the present invention not only increases information throughput, but also makes more efficient use of biological material. Thus, it can reveal spatial and temporal correlations, as well as tissue mosaics, that are otherwise difficult to establish reliably. The recombination labeling strategy cannot be used in these experiments because the intracellular distribution of mRNA and proteins, as with the nuclear chromosomal domains, is not speculatively known to be spatially different. However, many mRNA and protein light sources can be spectrally resolved and detected using a multiplex format with n fluorophores. Thus, for the first time, the intracellular distribution of a cancer or tumor suppressor protein can be determined simultaneously in the same cell.

F. Automated karyotyping One aspect of the invention relates to automation, preferably computerized karyotyping.
As mentioned above, in one embodiment of the present invention, chromosomes of a particular karyotype are pseudo-colored, thereby promoting chromosome assignment or translocation, deletion, and the like. In one sub-embodiment, digitized images of the chromosomes are stored in a computer-readable storage device (eg, magnetic or optical disk) to facilitate their comparison with other chromosomal images, or their transmission testing. In this regard, the pseudochromosomes or chromosomal elements may be displayed in a chromosome image showing the cytogenetic staining pattern (eg, the cytogenetic staining pattern of a patient's metaphase chromosome) to a researcher. Locus-specific or subchromosomal element or region specific probes may be used. The position and size of each band is preferably digitized and stored so that the chromosome image can be stored on a computer. Similarly, the exact location of any translocation or other karyotypic abnormality may be recognized and preserved.

Thus, the method of the present invention allows karyotyping to be performed later and more accurately than previously possible. Thus, the present invention can be used to systematically correlate karyotypic abnormalities with diseases or conditions. For example, to find a correlation between a patient's karyotype and his or her predisposition to a different disease and condition, the karyotype of a non-symptomatic individual is obtained, followed by any disease (eg, cancer, Alzheimer's disease). Etc.) or symptoms (eg, hypertension, atherosclerosis, etc.). Similarly,
To find a correlation between a particular karyotype and the future course of the disease or condition, the karyotype of the individual with the diagnosed disease or condition can be obtained to determine the degree of any subsequent progression or remission of the disease or condition. It can be evaluated by looking at it.

In yet another sub-embodiment, a computer or other digital signal analyzer may be used to orient and align chromosomal images, and assign and identify chromosomal karyotypes. Thus, a computer or other data processor may specify that a particular chromosome image is an image of the karyotype of chromosome 7 being evaluated, and may therefore use a homologue (e.g., Group the chromosomes designated by the second chromosome of the chromosome, and, via a printer, monitor or other output means, an ordered array of chromosome images (eg, each autosome is paired with its homolog, Sex chromosomes X and Y are paired together).

In one sub-embodiment, the chromosomal image of such an array is a pseudo-color image as described above. Alternatively, such pseudocoloring is intrinsic to the chromosome identity assignment process and is not displayed on the output of the computer or digital signal analyzer. Rather, in this sub-embodiment, the output generated is a light microscopically visible cytogenetic pattern of metaphase chromosomes whose karyotype is being evaluated. In yet another sub-embodiment, the scale (Morgan units or H or other suitable units) is superimposed on the chromosome image.

Having generally described the present invention, the present invention can be more readily understood by reference to the following examples. The examples are given by way of illustration, and the invention is not limited thereto unless otherwise specified. Example 1 Combinatorial Labeling of Chromosomes Probes representing 22 autosomes and 2 sex chromosomes were used to examine the feasibility of generating 24-color chromosome coloring. The DNA probes used were generated by microdissection.

A microdissection probe (National Center for Human Genome Research, Bethesda,
MD) results in a very uniform labeling of the target area. Detailed protocols for microdissection and PCR amplification are described in Telenius et al. (Telenius, H. et al., Genes, Ch.
romosomes & Cancer 4: 257-263 (1992); Telenius, H. et al., Genomics 13: 7
18-725 (1992); Meltzer, PS et al., Nature Genetics 1: 24-28 (1992); G
uan, XY et al., Hum Mol Genet 2: 1117-1121 (1993); Guan, XY et al.,
Genomics 22: 101-107 (1994); Guan, XY et al., Hum Genetics 95: 637-640.
(1995)), the contents of which are incorporated herein by reference. Some chromosomes, ie, 2, 4, 5, 10, 11, 16, 18
For Y and Y, different DNA probes are available for the p- and q-arms. For other chromosomes, a microdissection probe that colored all chromosomes was used.

DAPI, the first member of the preferred fluorophore set, was used as a general DNA counter dye. The remaining fluorophores, fluorescein, Cy3, Cy3.5 (the emission and excitation spectra are between Cy3 and Cy5), Cy5 (Mujumdar, RB et al., Cyt.
10: 11-19 (1989)) and Cy7 (Cy5 emits light in the red direction (Ernst, LA)
et al., Cytometry 10: 3-10 (1989))), and the different probes were recombinantly labeled (Table 3). The distinguishing features of these dyes are high excitation count, quantum yield and photostability. Fluorescein is a xanthene dye with an excitation count of about 70,000 L / mol and a quantum yield in the optimal buffer of about 0.7. Respective values for cyanine are 1-200,000 and 0.3 (Waggoner, A., Methods in Enzymology 2
46: 362-373 (1995)).

After microdissection, the probes were subjected to PCR amplification and labeled by nick translation. Fluorescein (Wiegant, J. et al., Nuc Acids Res 19: 3237-32
41 (1991)), Cy3 and Cy5 were directly ligated to DUTP for direct labeling. Cy3.5 and Cy7 were available as avidin or anti-digoxin conjugates for secondary detection of biotinylated or digoxigenylated probes. They were synthesized using conventional N-succinamide ester coupling chemistry. For each probe, one to three separate nick translation reactions were required, each using a single labeled fluorophore-labeled triphosphate or biotin or digoxigenin (Table 3).

[0141]

[Table 4]

As expected, probes labeled with equal amounts of different fluorophores did not produce equivalent signal strength for each fluorophore, which was due to the choice of filter set to maximize spectral discrimination rather than photon throughput. Reflects the fact that To reduce signal intensity differences, the probe concentration for the hybridization mix had to be carefully determined in a number of control experiments. Hybridization conditions were optimized for these multiplex probes. Therefore, the probes were denatured and hybridized with a metaphase chromosomal spread in a conventional 50% formamide hybridization cocktail for 2-3 nights at 37 ° C. Slides were washed three times at 45 ° C. in 50% formamide / 2 × SSC at 45 ° C., then three times at 0.1 ° SSC at 60 ° C. to remove excess probe. After a 30 minute blocking step at 37 ° C. in 4 × SSC / 3% bovine serum albumin, the biotinylated probe was detected with avidin Cy3.5 and the dig-labeled probe with anti-dig-Cy7. Fluorescein-dUTP, Cy3-dUTP and Cy5-dUTP did not require any immunological detection steps. After a final 3 washes with 4 × SSC / 0.1% Tween 20 at 45 ° C., place the media, apply a coverslip, and image the hybridization signal from each fluorophore using the filter set listed in Table 3. It has become.

FIG. 1 is a diagram illustrating a CCD camera and a microscope used according to the present invention. FIG. 2 shows raw data from karyotyping of chromosomes from a bone marrow patient (BM2486).
Adjacent to each light source image is a chromosome "mask" generated by a software program. In FIG. 2, panels A and B are DAPI images and masks;
Panels C and D are FITC images and masks; Panels E and F are Cy3 images and masks; Panels G and H are Cy3.5 images and masks; Panels I and J are Cy5 images and masks; Panels K and L are Cy7 images and masks.

FIGS. 3A and 3B show the identification of individual chromosomes by spectral characteristics. FIG.
A is the same photograph as FIG. 2, except that it is grayscale pseudo-colored. FIG. 3B shows a karyotype array of chromosomes. The exceptional power of the method of the present invention is illustrated by the ease with which translocations of chromosomes 5 and 8 can be identified in FIGS. 3A and 3B, as compared to conventional non-chromosome-specific karyotyping.

The above experiments demonstrate that five fluorophores are distinguished in the spectrum, resulting in at least 24 different colors. The combinatorial labeling scheme consists of 5 steps for probe labeling.
If one fluorophore is used, only nine colored probes need to be labeled with as many as three fluorophores and need not be a complex as previously thought. A sixth fluorophore for probe labeling allows for up to 63 possible fluorophore combinations. Such high numerical values of different targets are not necessary for most applications, but allow the selection of the combination with the best spectral properties. These aforementioned protocols allow for highly reliable and reproducible multicolor FISH.

Example 2 Rapid Analysis of Multi-Oral Bacteria in Mixed Cell Populations Using Multi-Parametric Fluorescence In-Situ Hybridization (M-FISH) The in-situ hybridization was performed for individual bacteria in the absence of culture. It has been recognized that it has a potential use as a means of analyzing cells. For example, DeLong, EF etc.
Science 243: 1360-1363 (1989)), a primordial bacterium (Methanosarcina acetivorans).
) Was distinguished from the eubacteria Bacillus megaterium and Proteus vulgaris, an oligonucleotide probe complementary to the 16s ribosomal RNA (rRNA) sequence. Van Den Berg, FM, etc. (J.
Clin. Pathol. 42: 995-1000 (1989)) used whole genomic DNA to detect Campylobacter pylori of Campylobacter in gastric tissue.
Amann, R. et al. (Nature 351: 161-164 (1991)) describe the protozoan paramecium Paramec.
Holospora obtusa was detected in the large nucleus of ium caudatum.

However, prior to the present invention, the use of in-situ hybridization methods to discriminate microorganisms in clinical samples using mixed populations, especially in a quantitative manner, was difficult and time consuming. This is especially true for bacteria that have nutritionally troublesome growth requirements and whose culture method can result in the selective enrichment of a small group of organisms present in the specimen.

A. Analysis of Non-Cross-Hybridizing Microbial Species To further illustrate the present invention, the M-FISH method of the present invention is used to isolate bacteria involved in the microbial pathogenesis of major diseases. Several methods have been used in the past to establish the microbial etiology of major diseases. Well over 200 bacteria can be identified in sublingual plaque samples using labor intensive culture methods (Dzink, JL et al., J. Clin. Periodont. 12: 648
-659 (1985); Moore, WEC; J. Periodont. Res. 22: 335-341 (1987); Moor
e, WEC, Infect. Immunity 38: 651-667 (1987)). This difference has prevented the association of certain microorganisms in the pathogenesis of various forms and stages of inflammatory periodontal disease.
Direct assessment of both qualitative and quantitative plaque bacteria eliminates the need for culture and the attendant difficulties. DNA probes (French, CK et al., Oral Micr
obiol. Immunol. 2: 58-62 (1986)) has been widely used in a blot hybridization format as a direct method for bacterial assessment. However, this method does not allow for direct visualization of the bacteria or provide information on the relative abundance of particular bacteria in complex mixtures.

To more simply illustrate the ability of the invention to seed microbial strains, tests for in-situ hybridization specificity were performed using a mixture of morphologically different, non-cross-hybridizing bacteria. carry out. Using such a mixture, one can immediately discriminate positive hybridization from those that show non-specific binding by correlating the correct form with the correct probe.

Accordingly, Fusobacterium nucleatum (25586) of the genus Fusobacterium, Porphyromonas gingivalis (33277) of the genus Porphyromonas, and Eikenel of the genus Eikenella
La corrodens (23834), Prevotella intermedia (25611) of the genus Prevotella, Actinobacillus actinomycetemcomitans (29522) of the genus Actinobacillus to America
n Type Culture Collection (Bethesda, MD) and the capnositofaga C. ochracea (C25) and C. gingivalis (DR2001) from the National Institute of
Obtained from Dental Research, Bethesda, MD.

Using a modification of the procedure by Chassy, BM et al. (Appl. Environ Microbiol. 39: 153-158 (1980)) by Donkersloot, JA et al. (J. Bacteriol. 162: 1075-1078 (1985)) F. nucleatum, P. gingivalis, E. corrodens, P. intermedia, A. a
Total genomic DNA was isolated from ctinomycetemcomitans, C. gingivalis and C. ochracea. Each DNA was labeled with ³² P and tested for cross-hybridization with each of the other DNAs by dot plot hybridization.
In all cases, a whole genomic DNA probe shows specificity for the bacterium from which it was obtained. Hybridization with any of the other bacterial DNAs was not detected. Because the genomic DNA probes were highly specific in the dot plot assay, they were used directly for in-situ identification of these bacteria in the reconstituted mixture.

Biotin-11, as described in Lichter, P. et al., Hum. Genet. 80: 224-234 (1988), which is incorporated herein by reference. -D
Whole genome bacterial DNA is labeled by nick translation using UTP (BIO), digoxigenin-11-dUTP (DIG) or dinitrophenol-11-dUTP (DNP). 10 mM Tris-HCl / 1 mM EDTA / 0.1
Unincorporated nucleotides are removed using a Sephadex G-50 spin column equilibrated with% SDS, pH 8.0. The labeled DNA (2-6 μg) is ethanol precipitated and redissolved in 100% deionized formamide. F. nucleatum DN
A is labeled with BIO-11-dUTP (BIO) and detected using cascade blue avidin. A. actinomycetemcomitans DNA is labeled with DNP-11-UTP (DNP) and detected indirectly using a FITC-conjugated antibody. And E. corrodens D
NA was labeled with DIG-11dUTP (DIG) to give rhodamine-labeled anti-DIG.
Detection is performed using an antibody.

To prepare slides for in-situ hybridization, cultured bacteria
A small aliquot (1-2 ml) is centrifuged at 1200 xg in an HBI centrifuge. cell
The pellet was suspended in phosphate-buffered saline (PBS) to give approximately 10 ^Three Adjust to bacteria. Aliquots of both pure cultures and synthetic mixtures are prepared in Maples, J.
.A. (Amer. J. Clin. Pathol. 83: 356-363 (1985)).
Activated glass slides prepared as described in (included herein)
And covalently bond. Alternatively, spot bacteria directly on positive (+) charged slides
And air dry. Suitable slides are available from Fisher Scientific, Pittsburgh, P
Commercially available from A (catalog number # 12-550-15), all of the above bacterial strains were
Retains high efficiency during the in situ hybridization procedure. Next, slide all Carnois B
Fix with a fixative (ethanol: chloroform: acetic acid, 6: 3: 1) for 5 minutes.

Bacteria are hybridized simultaneously with three differentially labeled genomic DNAs.
50 ng of labeled probe and DNase-treated salmon sperm DNA in 50% (vol / vol) deionized formaldehyde / 2 × SSC (0.3 M sodium chloride / 0.03 M sodium citrate, pH 7.0) / 5% dextran sulfate ( 15 μl containing 15 μg)
Each sample is hybridized using the hybridization solution of (1). The solution is applied to the sample, covered with a coverslip and sealed with rubber cement. Bacterial DNA
And the labeled DNA probe are both denatured by heating in an oven at 80 ° C. for 5-8 minutes. These steps are sufficient to make the bacteria permeable for probe acceptability.

The DNA is reassociated by incubating the slides in a humid chamber at 37 ° C. for 18 hours. Post-hybridization washes, blocking and detection was performed using Li
This was performed as described by chter et al. (7). Briefly, the slides were washed three times in 50% formamide / 2 × SSC at 42 ° C. for 5 minutes and then three times in 0.1 × SSC at 60 ° C. for 5 minutes. The slides were then incubated in 3% bovine serum albumin / 4x SSC (block solution) at 37 ° C for 30 minutes.

The biotinylated probe was fluorescein isothiocyanate- (FITC)
-Avidin DCS (Vector Laboratories, Burlingame, CA, 5 g / ml), Cascade Blue-avidin (Molecular Probe Inc., Eugene, OR, 10 g / ml) or Rhodamine-avidin (Boehringer Mannheim, Indianapolis, IN, 10 g / ml).
Detected using FITC or rhodamine conjugated sheep anti-digoxigenin F
The digoxigenin-labeled probe was detected using the ab fragment (Boehringer Mannheim, 2 μg / ml). Rat anti-DNP antibody (Novagen, Madison, WI, 1: 500
Together with a goat anti-rat FITC-conjugated antibody (Sigma, St. Louis).
, MO, 1 μg / ml) to detect the dinitrophenol-labeled probe. In some experiments, bacterial DNA was counterstained with 4,6-diamidino-2-phenylindole (DAPI) at a concentration of 200 ng / ml.

For detection, fluorochrome-conjugated antibody or avidin was added to 4 × SSC / 1%
Dilute in a solution of BSA and 0.1% Tween 20 (200 μl / slide) and incubate with the sample for 30 minutes at 37 ° C. in the dark. Then slide 4 x SS
After washing three times at 42 ° C. in C / 0.1% Tween 20, observation is made with an epifluorescence microscope.

Epifluorescence microscopy is performed using a Zeiss Axioskop-20 wide-field microscope equipped with a 63x NA 1.25 Plan Neofluar oil immersion objective and a 50 W high pressure mercury arc lamp. An image is projected using a Zeiss SFL-10 optical eyepiece onto a cooled charge coupled device (CCD) camera (Photometrics CH220; 512 × 512 pixel array). The effective magnification is set by changing the microscope-camera distance using a bellows. DAPI
An 8-bit grayscale image is continuously recorded using a FITC and Rhodamine filter set (manufactured by C. Zeiss, Inc., Germany) to minimize image offset. Preferably familiar software (Dr. Marshall Long; Yale Universi
Perform camera control, grayscale image acquisition and image preprocessing using an Apple Macintosh 11x computer running ty). Images are pseudo-colored, aligned, and merged using a Macintosh 11ci computer equipped with an accelerated 24-bit graphics system (SuperMac Spectrum 24PDQ). Preferably, the merged software "
Gene Join Max Pix "(Yale University) is used. The software assigns a user-definable pseudocolor and grayscale value to each of the light source images, and then generates an output ("merged") image based on the strongest light source image at each pixel location. Thus, an object with a high intensity (eg, a probe signal) will override the DAPI counterstain signal at the corresponding image location. Objects displayed in the merged image will retain their assigned colors, and thus appear to be distinct and easily identifiable. The merged image was photographed directly from a computer monitor.

A mixed sample of F. nucleatum, E. corrodens and A. actinomycetemcomitans in a 1: 1: 1 ratio was used as a test sample. As can be seen in FIG. 4, the three types of bacteria are DA
When stained with PI, it is identifiable based on morphology. FIG. 4 shows the differentiation of bacteria by the in-situ hybridization method. Panels A to E
Shows three bacteria (F. nuc) using a mixture of genomic DNA probes for each organism.
1 shows the in-situ hybridization assay results for a laboratory-derived mixture of C. leatum, A. actinomycetemcomitans and E. corrodens). Panel A shows all bacteria present in the microscopic field detected by the common DNA binding fluorophore (DAPI). F. nucleatum was observed using cascade blue detection (Panel B)
), A. actinomycetemcomitans using FITC detection (Panel C), and E. cor.
rodens are observed using rhodamine detection (Panel D). Panel E is a computer-combined composite of Panels BD showing the discrimination of each bacterium in the sample. single
A. actinomycetemcomitans cannot be identified by DAPI staining (Panel A)
Detected in E. corrodens colonies. Panel F is a similar analysis of a mixture of C. gingivalis (rhodamine) and P. intermedia (FITC) when hybridized with a combination of genomic DNA probes for those organisms. Panel G
Figure 4 shows in-situ hybridization assays for combinations of seven different bacteria hybridized with a mixture of seven genomic DNA probes. F. nucleatum (1), E
corrodens (2) and A. actinomycetemcomitans (3) are shown in blue (cascade blue detection). P. gingivalis (4) and C. ochracea (5) are shown in red (rhodamine detection), while P. intermedia (6) and C. gingivalis (7)
) Are observed in yellow (FITC detection). Using phase contrast microscopy, E. corrodens and A. actino were not readily apparent by hybridization-generated signals
Helped determine morphological differences between mycetemcomitans. Panel H shows an in-situ hybridoma formation assay for the presence of A. actinomycetemcomitans (using FITC detection) in plaque samples obtained from patients with localized juvenile periodontitis.

When assayed for Cascade Blue, the F. nucleatum bacteria (FIG. 4, panel B)
Only, and tested for FITC, A. actinomycetemcomitans (
Only FIG. 4, panel C) is visible and assayed for rhodamine.
Only the dens bacteria (FIG. 4, panel D) were positive. Next, the three separate fluorophore images were merged to produce the composite image shown in panel E of FIG. Each morphologically distinct bacterium exhibits proper fluorescent properties.

[0161] Four additional bacterial species are included in subsequent experiments. These bacteria are merged in pairs so that each mixture is again morphologically different. Panel F of FIG. 4 shows P. intermedia (
2 shows a combined analysis of BIO labeling and FITC detection) and C. gingivalis (DIG labeling and rhodamine detection). C. gingivalis and its morphology are P. intermed
Analysis of P. gingivalis similar to ia, and C. ochracea and P. gingivalis whose morphology is similar to P. intermedia and C. gingivalis pairs is also performed. In both cases, the absolute specificity of the probe for the cognate bacterium is demonstrated.

The above experiments show that no bacteria cross-hybridize, and that various types of bacteria, when grouped, can be distinguished by morphology as well as by hybridization through the binding of specific fluorochromes. To show, it was assumed that all seven bacteria could be detected in a single mixed sample. The experimental design was to label DNA from bacteria inside each of three morphologically distinct groups with the same "reporter" (BIO, DIG or DNP). The three groups were first distinguished by the specificity of the hybridization and the fluorescence signal. The bacteria in a group were then distinguished based on morphology.

To demonstrate this ability of the present invention, P. intermedia and C. gingivalisD
NA was labeled with DIG. F. nucleatum, A. actinomycetemcomitans and E. co
rrodens DNA was labeled with BIO, while P. gingivalis and C. ochracea DN
A was labeled with DNP. All seven probes were then combined and used for in-situ hybridoma formation on a mixed bacterial sample containing the whole organism (FIG. 4, panel G). In panel G of FIG. 4, F.nucl was detected using cascade blue (BIO detection).
Indicate eatum, E. corrodens, and A. actinomycetemcomitans groups, numbering their members 1, 2, and 3, respectively. P. gingivalis and C. ochracea groups are observed with rhodamine (DIG detection), and these bacteria are numbered 4 and 5, respectively. C. gingivalis and P. intermedia groups were identified by FITC (DNP detection) and numbered 6 and 7, respectively. All seven bacteria are clearly distinguished using a combination of hybridization fluorescence and morphology. Morphological identification can be easily made based on the size, shape and color of the hybridization signal, while more restrictive morphological characterization is made by examining the sample using phase contrast optics.

[0164] To how little hybridization-positive bacteria in the sample is determined whether may be detected, A. in 1.0μl of E. corrodens of 4.0μl containing 10 ⁵ bacteria / [mu] l
Mix with a serial dilution of actinomycetemcomitans and apply 5 μl mixture to slides. Then, using a BIO-labeled probe for the bacterium, A. actinomycetemco
Inspect the sample for the presence of mitans. After hybridization, all bacteria on the slide are visualized using propidium iodide, and hybridization-positive bacteria are detected using avidin-FITC. The result of this experiment was a single A. act
4 shows that inomycetemcomitans cells can be detected in the presence of an extra amount of at least 10 ⁴ E. corrodens cells. Thus, it is clear that hybridization-positive bacteria can be easily identified, even if they constitute only a small fraction of the mixed cell population.

A clinical sample typically does not contain contaminating debris, such as a laboratory-derived mixture of bacteria. To determine the potential utility of M-FISH for the analysis of clinical specimens, studies (Slots, J., Dent. Res. 84: 1 (1976); Slots, J. et al., Clin Periodontal)
13: 570-577 (1986); Zambon, J. et al., J. Periodontal. 54: 707 (1983).
) We will use plaque samples from patients with localized juvenile periodontitis previously suggested to be involved in A. actinomycetemcomitans as a microbial pathogen. FIG. 4 panel H shows BIO labeling.
FIG. 3 shows the results of in-situ hybridization experiments using patient samples detected by FITC-conjugated avidin using A. actinomycetemcomitans DNA. As can be observed, a large number of A. actinomycetemcomitans or other closely related organisms were detected. Quantitative tests are performed on individual identifiable bacteria observed in a sample.
Indicates that more than 50% hybridized with A. actinomycetemcomitans. Culture tests on this sample indicated the presence of A. actinomycetemcomitans.

All of the above experiments use the standard overnight hybridization time (〜18 hours) to sort out the bacteria. Although such analyzes can often be made faster than traditional methods of culture and biochemical testing, a number of experimental parameters can be adjusted to yield faster assays. It is well known that hybridization time is directly related to probe concentration. Thus, by increasing the probe concentration, the hybridization time can be proportionally reduced. Therefore, F. n
ucleatum and a mixture of A. actinomycetemcomitans were used previously (50 ng
(〜0.5 μg / slide) using 10 times higher concentration of probe. At higher probe concentrations, no specific or non-specific hybridization was observed with the non-denatured probe and bacterial DNA. However, when the probe and bacterial DNA were simultaneously denatured and the wash cycle started immediately, some of the DNA hybridized during the time required for subsequent sample processing.

Before starting the washing cycle, 0.5 μg of the A. actinomycetemcomitans probe was
Specimens containing both actinomycetemcomitans and F. nucleatum bacteria at 37 ° C
In this experiment, which is introduced using a 2.5 minute incubation, the hybridization signal is strong and specific as observed in the experiment hybridizing for 18 hours with 50 ng of probe. In addition, the blocking and antibody detection times
By reducing to 20 minutes, a total assay time of 1.5 hours is achieved. Further reductions in assay time are clearly obtained using probe DNA directly labeled with a fluorophore detector.

The above experiments provide a relatively quick and simple method for identifying individual bacteria in a mixed microbial population using M-FISH. The method provides a quantitative assessment of a particular microorganism without the need for culture. This technique can be applied to samples containing only a limited number of organisms, and a single hybridization positive bacterium
10 can be very easily detected at ^four times excess amount of hybridization-negative bacteria. In addition, duplication and "masking" among groups of E. corrodens cells (FIG. 4, panel E).
As exemplified by A. actinomycetemcomitans cells, specialized bacteria that are not clearly identified by non-specific analysis (ie, DAPI staining of DNA, FIG. 4, panel A) can be detected by hybridization-specific assays. Bacterial hybridization specificity and morphological characteristics can be assessed simultaneously by this approach, and the number of different bacterial species can be further increased by judicious selection of probe labeling and fluorophore detector combinations.

However, because many different bacterial species are morphologically similar and some are polymorphic by culture methods, morphological features are not always a reliable indicator of bacterial identification. Not exclusively. For example, A. actinomycetemcomitans is generally coccoid in clinical specimens, but can be considered rod-like and even filamentous at in vitro passage and in clinical specimens. Therefore, the additional fluorescent colors are preferably used to expand the number of different bacteria that can be analyzed simultaneously and clearly identified.

The intensity of the hybridization signal is very strong when using whole genomic DNA or subtraction (suppression) probes. Data is collected by digital imaging,
However, the hybridization signal is easily detected by the eye and is recorded by conventional photography. However, the sensitivity and photon counting capabilities of CCD camera-based imaging systems offer significant advantages for further improving microbial analysis by in-situ hybridization. Detection of a single copy sequence for a particular toxin or drug resistance gene in an individual bacterium needs to be feasible. Individual human and murine genes have been visualized by FISH on both metaphase chromosomes and interphase nuclei (Lichter, P. et al., Hum. Genet. 80: 224-234 (19
88)). The level of cross-homology between bacterial strains can be easily assessed by reducing the photon output of the hybridization signal from the individual bacteria in the sample. Finally, and most importantly, digital imaging is essential to fully exploit the power of combinatorial fluorescence.

As exemplified herein, whole genomic DNA often provides the appropriate specific probe, and thus can eliminate the usual cloning and screening efforts required for the probe approach. Due to the absence of cross-hybridization between bacterial genomes, suppression hybridization techniques (Lichter, P. et al., Hum. Genet. 80: 224-234 (1988)), although not required in these tests, were used. Can be used to suppress cross-hybridization between organisms that share partial sequence homology. For example, C. orchracea and C. sputigena show 22-23% sequence homology (Za
mbon, J. et al., J. Periodontal. 54: 707 (1983)), and by pre-annealing the C. ochracea probe with extra C. sputigena DNA just prior to the in-situ hybridization reaction. Specific hybridization with C. orchracea is obtained using genomic DNA.

In summary, seven anaerobic and any oral bacteria (Fusobacterium nucleatum, Porphyromonas (Bacteroides) gingivalis, P.
revotella intermedia, Eikenella corrodens, Actinobacillus actinomycetemc
Total genomic DNA from omitans, Capnocytophaga gingivalis and Capnocytophaga ochracea) is non-isotopically labeled and hybridized in-situ to reconstruct the bacterial mixture. Each probe is uniquely hybridized to the bacteria from which it was obtained. The three bacterial strains can be distinguished simultaneously by labeling the DNA with different groups of reporters and visualizing the hybridization signal with reporter-specific detector proteins labeled with different fluorophores. By assessing bacterial morphology together with the hybridization signal, one can identify each bacterium in a mixture of seven organisms. The total test time is as short as 1.5 hours. A dental plaque sample from a patient with localized juvenile periodontitis, a mixed oral microbial infection, has been reported to be A. a putative microbial pathogen for this disease. When probed with DNA from actinomycetemcomitans, a number of hybridization positive bacteria were revealed. This example demonstrates the in-sit of the present invention for the identification of individual non-cross-hybridizing bacteria in a mixed cell population.
Demonstrate the usefulness of the u-hybridization method.

The seven bacterial species analyzed above are any or essential anaerobic bacteria often found in connection with inflammatory periodontal disease. The role of these microorganisms is not clear because there is no quick and reliable way to identify the species. Direct analysis of oral microbial specimens counteracts the bias imposed by culture methods and may allow a better understanding of the role of these bacteria in the pathogenesis of periodontal disease. A. Samples from patients with localized juvenile periodontitis The data relating to the detection of actinomycetemcomitans
It has been shown that n-situ hybridization methods facilitate such studies. It further complements traditional culture methods for the identification of a wide range of microorganisms, especially well-known slow growing bacteria such as anaerobic or Mycobacterium bacteria. For example, swabs taken from an initial culture plate can be linearized on slides and assayed for suspected pathogens by in-situ hybridization using a probe.

The present invention can further increase the number of co-detectable bacteria by using combinatorial fluorescence imaging without increasing the number of available fluorophores (Nederl
of, PM et al., Cytometry 11: 126-131 (1990)). Two (or more)
By labeling a single probe with a different reporter molecule, the resulting hybridized probe is detected by more than one fluorescent detector and the signal appears in more than one separate fluorescent image. When these separate pseudo-color images are merged into a composite image (as in panels E, F and G of FIG. 4), the fluorescent signals appearing at the same site of the two images are "mixed" and distinguished from the original. Produces new pseudo colors that are possible. In this format, three fluorophores were used in combination to simultaneously visualize seven probes, each appearing as a different color (Ried, T. et al., Proc Natl.
Acad Sci (USA) 89: 1388-139219920). By applying this combined fluorescence imaging strategy to bacterial identification, different pseudocolors can be obtained for each of the seven bacteria analyzed above.

The above example describes a multiparametric fluorescence in-situ hybridization (M
-FISH) and digital imaging microscope to demonstrate the feasibility of discriminating multi-bacterial species (eg, those found in periodontal microbial flora). By monitoring both hybridization signal specificity and bacterial morphology, seven different bacteria are easily and simultaneously identified in a mixed population using three different fluorophores. Such methods include other clinical samples (eg, those containing complex mixtures of microbial flora (eg, feces, saliva, throat swabs, sputum, vaginal secretions, etc.), as well as generally (eg, spinal fluid (eg, Meningitis), blood (sepsis, etc.)
Or it is easily applicable to the detection of bacteria (and viruses and / or lower eukaryotes) that may be present in non-clinical samples (food, reservoirs, ecosystems, etc.). Further, by increasing the number of spectrally resolvable fluorophores used in a combinatorial format, the number of different types of bacteria, viruses and / or lower eukaryotes that can be analyzed in a mixed population can be significantly increased.

B. Analysis of Cross-Hybridizing Microorganism Species As described above, using a differentially labeled whole genomic DNA probe
-Situ hybridization and digital imaging fluorescence microscopy can simultaneously identify seven microorganisms in a mixed sample from a laboratory. This technique has single cell detection sensitivity,
As well as the ability to correlate hybridization signal specificity with microbial morphology. The relative amounts of the complex specific organism in a mixture or microorganism, can be readily assessed, and a single hybridization-positive bacteria can be detected in other bacteria excess of 10 ^4. Further, the assay can be completed in less than 1.5 hours. However, none of the genomic DNAs from the seven bacteria used in the above test showed cross-hybridization.

To illustrate the ability of the present invention to identify related (ie, cross-hybridizing) microbial flora, a DF1 strain of the genus Capnositofaga: C. gingivalis (DR2001)
), C. sputigena (D4) and C. ochracea (C25).
Such strains show cross-hybridization (Rubin, SJ, Eur. J.
Clin. Microbiol. 3: 253-257 (1984)). The strain is from the National Institute of Den
Available from tal Research, Bethesda, MD. Bacterial DNA is isolated as described above. Probes were made as follows: Lichter, P. et al., Hum.
Biotin-11-dUT, as described in Genet. 80: 224-234 (1988).
By nick translation using P (BIO), digoxigenin-11-dUTP (DIG) or dinitrophenol-11-dUTP (DNP)
Label whole genome bacterial DNA. 10 mM Tris-HCl / 1 mM EDTA / 0.1% S
Unincorporated nucleotides are removed using a Sephadex G-50 spin column equilibrated with DS, pH 8.0. The labeled DNA (2-6 μg) is ethanol precipitated and redissolved in 100% deionized formamide.

Capnositofaga is an oropharyngeal flora (Holdeman, LV et al. J. Period)
on. Res. 20: 475-483 (1985)), a spindle-shaped, gram-negative,
Includes a group of capnophilic, fermentative, gliding bacteria (Lecbett
er, ER et al., Arch. Microbiol. 122: 17-27 (1979)). C. gingivalis,
C. ochracea and C. sputigena have been reported for gingivitis in children (Moore, WEC, Infect. I
46: 1-6 (1984)) and periodontitis in juvenile diabetic patients and adults (Dzink, JL et al., J. Clin. Periodont. 12: 648-659 (1985); Mashimo, P.
.A. Et al, J. Periodont. 54: 420-430 (1983); Slots, J. et al., Dent. Res.
63: 414-421 (1984)). These three types of bacteria are
of the Desease Control (Speck, H. et al., Zbl. Bakt. Hyg. A. 266: 390-40
2 (1987)) as a member of the first of three groups of "fertile fermenters" (DF-1). DF-2 group comprises C. canimorsus and C. cynodeami
However, these bacteria appear to differ in both phenotype and genotype from the DF-1 family, Capnositofaga (Brenner, DJ et a).
l., J. Clin. Microbiol. 27: 231-235 (1989)). Most of the microorganisms that make up the DF-3 group have not been characterized, but they are believed to be closely related to DF-1 of the genus Capnositofaga (Gill, VJ et al., J. Clin.
Microbiol. 29: 1589-1592 (1991)). Members from all three groups of DF have been identified as etiologic agents found in several cases of sepsis, complex infections and endocarditis (Arlet, G. et al., Ann. Biol. Clin. 44 : 373-379 (1986); Juhl, G. et.
al., J. Infect. Dis. 149: 654 (1984); Matlow, A. et al., J. Infec. Dis.
152: 223-234 (1985); Mosher, CB et al., J. Clin. Microb. 24: 161-162 (
1986); Parenti, DM et al., J. Infect. Dis. 151: 140-147 (1985); Ratne
r, H., Clin Microbiol 6:10 (1984); Rummens, JL et al., J. Clin. Micro
biol. 75: 376 (1985); von Graevenitz, A., Eur. J. Clin. Microbiol. 3: 223.
-224 (1984)). The increased range of oral and oral infections caused by the DF-1 species of the genus Capnositofaga indicates the pathogenic potential of this genus, and differences in the clinical significance of each species in medically important infections and periodontal disease (Von G
raevenitz, A., Eur. J. Clin. Microbiol. 3: 223-224 (1984)). Conventional methods of identifying bacteria from oral sources include first isolating a single bacterial colony from the oral microbial environment and secondly isolating the isolate with cell morphology, gram stain, carbohydrate fermentation and various Requires assignment to genera and species based on hydrolysis. These procedures require considerable time and expertise and take more than two weeks to identify a single species from a clinical sample. Such conventional methods often fail to isolate Capnositofaga from oral samples due to the gradual growth of these bacteria and difficulties with respect to culture conditions, and thus the overgrowth of other oral flora (Mashimo, PA et al. al
, J. Periodont. 54: 420-430 (1983); Rummens, JL et al., J. Clin. Micr.
obiol. 75: 376 (1985)). Furthermore, such methods cannot non-ambiguously distinguish C. ochracea and C. sputigena due to phenotypic similarity (Gill, VJ et al.,
J. Clin. Microbiol. 29: 1589-1592 (1991)).

Enzyme assays (Heltsberg, O. et al., Eur. J. Clin. Microbiol. 3: 236-240 (19
84); Kristinnsen, JE et al., Eur. J. Clin. Microbiol. 3: 236-240 (1984
)), Specific immunological or molecular probes (Murray, PA et al., Oral Microb
iol. Immunol. 6: 34-40 (1991); Smith, GLF et al., Oral Microbiol. Imm
unol. 4: 41-46 (1989)) have provided some successful examples of identifying bacteria in clinical oral samples. Except for enzymatic techniques, these methods have not been applied to the identification of the genus Capnositofaga. Enzyme assay
Still requires isolation of pure bacterial colonies, while immunological and molecular methods are performed directly on oral samples, thereby overcoming the difficulty of isolating pure colonies of difficult-to-vegetative microorganisms I do. In general, molecular methods rely on the use of DNA probes formatted as dot-blot hybridizations for the isolation of specific microorganisms in clinical samples (Parenti, DM et al., J.
Infect. Dis. 151: 140-147 (1985)). The major drawback of the method is that such assays require 10 ³ to 10 ⁶ DNA from the microorganism in question, and therefore do not appear to detect small amounts of bacteria in the mixed population. In addition, cross-hybridization, which is common in closely related species, can prevent identification at the species level.

It has been previously reported that C. gingivalis, C. Ochracea and C. sputigena DNA show up to 22% sequence homology, based on reassociation kinetics (Will
iams, BL et al., Arch. Microbiol. 122: 35-39 (1979)). When using conventional in-situ hybridization techniques, such extensive homology precludes the use of whole genomic DNA probes for capnositofaga speciation and reduces the production and characterization of species-specific subgenomic clones. It seems necessary. However, one of the attributes of the computerized digital imaging camera system of the present invention is the ability to collect hybridization data in the form of a quantitative fluorescent image and process such image data with appropriate computer software.

The intensity (ie, fluorescence photon output) of the hybridization signal from the bacterium (“Bacteria A”) A is greatest for genomic DNA obtained from that bacterium. Bacteria A
Related strains that share some DNA sequences in common with the DNA probe of the genomic DNA (Genomic DNA Probe A) produce low intensity hybridization signals, which directly reflects the degree of sequence homology. Unrelated bacteria that share little or no sequence homology with Bacteria A end up producing no fluorescent signal. Once the relative degree of DNA sequence homology between the stains has been determined, the cross-hybridization noise can be suppressed electronically (preferably by appropriate computer software and video thresholding techniques), thus directly identifying the bacteria. Can lead to

Computer fluorescence thresholding is used to determine the degree of cross-hybridization (ie, sequence homology) between the three capnocytofaga species. In this experiment, a pure sample of each bacterium is separately hybridized with biotinylated DNA probes from all three species. For example, three slides containing only C. ochracea are prepared. The first slide was hybridized in-situ with BIO-labeled C. ochracea genomic DNA, and the second slide was BIO-labeled C. gingivalis.
Hybridize in-situ with genomic DNA and the third slide hybridizes in-situ with BIO-labeled C. sputigena genomic DNA. The slides are then washed and hybridization positive bacteria are identified using FITC-avidin.

C. ochracea and C. gingivalis did not cross-hybridize with each other, however, C. sputigena cross-hybridized with both C. gingivalis and C. ochracea. Computer analysis of the intensity of the fluorescent hybridization signal
C. sputigena had significant sequence homology (と 22%) to C. ochracea, while C. gingiva
alis shows much lower (~ 5%).

To demonstrate how video thresholding can be used for bacterial speciesing, C. spu
Slides containing a mixture of tigena and C. orchracea were placed on a Bio Bio from C. sputigena.
-Hybridize with labeled genomic DNA. To prepare such slides for in-situ hybridization, a small aliquot (1-2 m
l) is centrifuged at 1200 xg in an HBI microcentrifuge. The cell pellet is suspended in phosphate buffered saline (PBS) and adjusted to approximately 10 ³ bacteria / μl. Aliquots of both pure cultures and synthetic mixtures are commercially available (Fisher Scientif
Spot directly on positive (+) charged slides of ic, Pittsburgh, PA, catalog number # 12-550-15) and air dry. Next, all slides are fixed with Carnowa B fixing solution (ethanol: chloroform: acetic acid, 6: 3: 1) for 5 minutes.

50 ng in 50% (vol / vol) deionized formaldehyde / 2 × SSC (0.3 M sodium chloride / 0.03 M sodium citrate, pH 7.0) / 5% dextran sulfate
Containing labeled probe and DNAse-treated salmon sperm DNA (15 μg)
Hybridization is achieved using μl of hybridization solution. The solution is applied to the sample, covered with a coverslip and sealed with rubber cement. Both the bacterial DNA and the labeled DNA probe are denatured by heating in an oven at 80 ° C. for 5-8 minutes. These steps are sufficient to make the bacteria permeable for probe acceptability.

By incubating the slides in a humid chamber at 37 ° C. for 2.5 minutes, the DNA
Reassociate. Washing, blocking and detection after hybridization are performed as described by Lichter, P. et al. (Hum. Genet. 80: 224-234 (1988)). Briefly, slides were placed in 50% formamide / 2 × SSC at 42 ° C. for 5 minutes,
After washing twice, the plate is washed three times in 0.1 × SSC at 60 ° C. for 5 minutes. Then slide 3%
Incubate at 37 ° C. for 30 minutes in a solution of bovine serum albumin / 4 × SSC (blocking solution).

The biotinylated probe was fluorescein isothiocyanate- (FITC)
-Avidin DCS (Vector Laboratories, Burlingame, CA, 5 g / ml), Cascade Blue-avidin (Molecular Probe Inc., Eugene, OR, 10 g / ml) or Rhodamine-avidin (Boehringer Mannheim, Indianapolis, IN, 10 g / ml).
Detected using Digoxigenin-labeled probes were detected using FITC or rhodamine conjugated sheep anti-digoxigenin Fab fragment (Boehring Mannheim, 2 μg / ml). Along with rat anti-DNP antibody (Novagen, Madison, WI, 1: 500 dilution), then goat anti-rat FITC-conjugated antibody (Sigma, St. Louis, MO,
Incubation with 1 μg / ml) detected the dinitrophenol-labeled probe. In some experiments, bacterial DNA was counterstained with 4,6-diamidino-2-phenylindole (DAPI) at a concentration of 200 ng / ml. For detection, fluorochrome-conjugated antibody or avidin was added to 4 x SSC / 1%
Dilute in a solution of BSA and 0.1% Tween 20 (200 μl / slide) and incubate with the sample for 30 minutes at 37 ° C. in the dark. Then slide 4 x SS
After washing three times at 42 ° C. in C / 0.1% Tween 20, observe.

Hybridization is visualized by epifluorescence microscopy using a Zeiss Axioskop-20 wide-field microscope equipped with a 63x NA 1.25 Plan Neofluar oil immersion objective and a 50 W high pressure mercury arc lamp. Cooling charge coupled device (CCD) camera (Ph
An image is projected onto the otometrics CH220 (512x512 pixel array) using a Zeiss SFL-10 optical eyepiece. The effective magnification is set by changing the microscope-camera distance using a bellows. Continuously record 8-bit grayscale images using DAPI, FITC and Rhodamine filter set (manufactured by C. Zeiss, Inc., Germany) to minimize image offset. In the same manner as described above, camera control, grayscale image acquisition, and image preprocessing are performed.

All the bacteria were found to be hybridization positive in this experiment, but some appear to fluoresce more strongly than others. After thresholding the fluorescent image (using Gene Join Max Pix software (Yale University)) to remove image components that show less than 25% of the maximum fluorescent intensity, only the original bacterial subset shows positive hybridization . All of these bacteria are C. sputigena because they only show 100% homology with the probe. Using this general strategy, it has been found that multiple related organisms can be sieved, provided that their level of sequence relatedness can apply appropriate thresholding parameters. Bacteria showing up to 40% sequence homology can be identified with genomic DNA probes using this technique.

In the second method, bacterial typing can be achieved with genomic DNA probes using Gene Join Max Pix software. This embodiment shows that when the separate fluorescent images (eg, fluorescein and rhodamine) are merged to form a composite image, the light source image having the maximum fluorescence intensity (ie, photon output) at each pixel location is color-coded in the merged image. Use the fact that you show dominance. This attribute allows the use of multiple differentially labeled genomic DNA probes to probe single or mixed bacterial populations simultaneously. Since each bacterium in the mixture shows a maximum fluorescence signal with a genomic probe that is 100% homologous, the image merging step results in effective sorting.

This is demonstrated by the following experiment. 1: 1 each labeled with a different reporter:
Using all three genomic DNA probes, a mixture of three capnocytofaga strains is hybridized. C. ochrace for detection by FITC
aDNA labeled with DNP-11dUTP (yellow-green), C. gingivalis DNA labeled with biotin-11-dUTP (blue) for detection by cascade blue,
Then, C. sputigena DNA was converted to DIG-11-dU for detection with rhodamine.
Labeled with TP (red). The image generated by Cascade Blue shows a strong signal for any bacteria present in the sample, probably C. gingivalis cells, and a weak signal for other bacteria. FITC generated by C. ochracea probe
The fluorescent image and the rhodamine fluorescent image generated by the C. sputigena probe both show strong signals for any bacteria and weak signals for others, reflecting interspecies hybridization. However, each bacterium shows a strong signal in some of the three images and a weak (or absent) signal in others (reflecting interspecies hybridization). Therefore,
By using thresholding (ie, using the generated signal to determine which probe produced a more dominant signal for which bacteria), the method allows for the determination of each bacterial species.

Thus, in contrast to the often difficult and time-consuming traditional methods of culturing and assaying biochemical products for bacterial differentiation and / or seeding, the in-situ hybrids of the present invention The formation method provides a rapid and simple method for the seeding of cross-hybridizing microorganisms, such as the genus Capnositophaga in the mouth.
The basis of this method relies on the finding that all microorganisms in different genera and species have unique DNA sequences in their genome. Closely related organisms are:
Having several DNA sequences in common, and the amount of such cross-hybridizing sequences directly affects their ability to discriminate by in-situ hybridization.

In this analysis, the fluorescence image obtained from the in-situ hybridization using each detector (eg, rhodamine, FITC or cascade blue), reflecting the amount of hybridization between the bacteria and the genomic DNA probe, Digitize on a computer. If a bacterium does not have a cross-hybridizing sequence with any other bacterium, it will only hybridize to its homologous DNA, thereby being detected by only a single detector, and its presence to the detector. Only found in the corresponding image. However, if one bacterium has a cross-hybridizing sequence with one of the other bacteria, it is detected by a DNA probe from each bacterium, and its presence is observed in the image corresponding to each probe (detector). You. However, the intensity of the signal generated in each image is directly proportional to the amount of hybridization with the bacterial DNA probe. The corresponding images from each detector are pseudo-colored and superimposed (merged) to create a composite image. Since any given pixel on the monitor is only a single color, only the strongest signal at each pixel is shown and displayed as the corresponding false color.

Two of the genus Capnositofaga in the above experiments (C. gingivalis and C. ochracea
) Does not show cross-hybridization, while the third species (C. sputige)
na) cross-hybridizes with both of the above organisms. in-situ hybridization
Of the intensity of the hybridization signal after hybridization by computer analysis
The amount of cross-hybridization in this case matches well with the values found in the literature.
(22-23% for C. sputigena and C. ochracea and C. sputigena
9-11% for a and C. gingivalis (Heltberg, O. et al., Eur. J. Clin
Microbiol. 3: 236-240 (1984)). 22-23% of C. sputigena and C. ochracea
Cross-hybridization is evident in synthetic images within some C. sputigena
Yes, but not clear in C. ochracea cells. This is a cross hybrid
Sequence may reflect differences in how they are aligned within the genome.

This example further demonstrates that for in-situ hybridization assays, DNA probes need not be entirely specific. Cross-hybridizing sequences must only be reduced below the level required to show specificity. The level of unique sequences required for specificity is likely to be influenced not only by the total number of cross-hybridizing sequences, but also by how they are aligned in the genome, so it should be ascertained for each such bacterium. desirable. However, some microorganisms such as Neisseria meningitis and N. gonorrhoe of the genus Neisseria
a shows greater than 80% sequence homology (Kingsbury, DT, J. Bacteriol. 94: 870
-874 (1967)). Such large amounts of cross-hybridizing sequences can result in ambiguous analysis using in-situ hybridization with whole genomic DNA probes that can be handled by the method of the present invention.

Thus, in essence, using multiparametric fluorescence in-situ hybridization (M-FISH) and computer-assisted digital imaging microscopy, C. gingivalis, C. ochracea and C. sputige with genomic DNA probes.
A rapid separation method for na has been developed. These bacteria show up to 22% sequence homology, but the cross-hybridization signal can be completely suppressed. These three
Species of one Capnositophaga can be achieved in a short time of 30-90 minutes,
This is significantly faster than the 2-3 weeks required using conventional microbiological methods. The general discrimination strategy reported herein is applicable to a wide range of microbial pathogens of unknown degree of sequence homology.

Example 3 Synergistic Use of Nucleic Acid Amplification Techniques in Cooperation with M-FISH Analysis Nucleic acid amplification techniques greatly enhance the ability to address detailed problems with genotypes or transcript phenotypes in small biological samples And has provided momentum for a number of significant advances in biology, especially in the field of genetics.

One aspect of the invention is the detection and characterization of mutations, chromosomal elements, chromosomal rearrangements, and infectious agents (eg, viruses, proviruses, etc.) that may be present in small or small proportions in the sample being analyzed. Multiparametric fluorescent in-situ hybridization (M-FISH) in conjunction with one or more methods of nucleic acid amplification to promote
) Concerning the use of the law. For example, through the use of fluorophore combinations intended for multiparametric color coding, the 6-fluorophore tagging approach allows the simultaneous identification of 63 different types of signals due to their unique spectral properties.

As described above, the M-FISH method of the present invention can identify and identify all human karyotype chromosomes. However, by using one or more labeled probes capable of amplifying a particular gene, allele or chromosomal element and hybridizing specifically to the amplified sequence, the present invention provides that the particular karyotype is It allows the determination of whether the gene, allele or chromosomal element is contained. Illustratively, the use of probes specific for deletions, rearrangements, insertions, or other mutations characteristic of genetic disorders (eg, hemophilia, cystic fibrosis, breast cancer or other cancers, etc.) Can be diagnosed as indicating a genetic disease (eg, diagnosing cystic fibrosis, etc.). Similarly, the present invention allows for the determination of whether a patient's chromosome carries a recessive allele associated with a genetic disease. Similarly, the present invention relates to genetic lesions associated with future disease states (eg, Alzheimer's disease, heart disease, cancer, etc.) (eg,
The presence of a mutation in the apoE, p53, rb or brca1 / 2 genes) allows the determination of whether a patient is predisposed to the disease.

[0200] Similarly, as applied to the study of microorganisms (eg, bacteria, viruses and lower eukaryotes), the M-FISH method of the present invention is directed to the species of microorganisms present in clinical or non-clinical samples. Can be identified and identified. However, by using one or more labeled probes capable of amplifying a particular gene, allele or chromosomal element and specifically hybridizing to the amplified sequence, the present invention provides for the identification of a particular microorganism Of the gene, allele or chromosomal element. Thus, by using probes specific for antibiotic resistance determinants, cell antigens, toxins, etc., the combination of M-FISH and nucleic acid amplification can be used to determine whether a particular microorganism or virus strain is resistant to antibiotics or pathogenic. Sexuality or to express a toxin. Thus, such a combination of methods allows for serotyping and pathogen subtyping without the need for any culturing and / or purification.

The M-FISH method of the present invention particularly relates to probe sets and methods sufficient to characterize both the genotypic and phenotypic characteristics of microorganisms present in a specimen. Genotypic probes are those that can specifically hybridize to phylogenetically related species of microorganism. Thus, hybridization of such a probe with the nucleic acids of a specimen indicates whether a preselected microorganism or its cross-hybridizing (ie, phylogenetic) related microorganisms are present in such a specimen. A phenotypic probe is one that can specifically hybridize to a gene or genetic element associated with a particular phenotype (eg, antibiotic resistance, toxin production, antigen presentation, etc.).

The present invention specifically contemplates the multiparametric use of multifluorophores, and in particular, the detection and / or characterization of 3, 7, 15, 31, 63 or more combinations of genotype or phenotypic elements. It relates to embodiments where 2, 3, 4, 5, 6, or more fluorophores are used to enable. Thus, for example, using five fluorophores, 31 genotype / phenotype elements can be probed. For example, 25 of such devices may be genotypic devices (thus allowing identification of 25 different species of microorganisms), while 6 of such devices may be genotypic ( For example, which of the 25 identified species is resistant to which of the four antibiotics,
Or which of the two toxins is present). Similarly, 10 of such elements are genotype elements (thus allowing identification of 10 different species of microorganisms), while 21 of such elements may be genotypic (
For example, which of the 10 identified species is resistant to which of the nine antibiotics, which of the three toxins, which of the four surface antigens, and which of the five genes It is possible to determine whether to end).

Using such methods, strains of a wide range of bacteria (eg, strains of E. coli and other enteric bacteria (eg, Salmonella), Clostridria, Vibrio, Corynebacterium, Listeria, Bacillus (especially charcoal) B. anthracis), Staphylococcus,
Associated with streptococci (especially β-hemolytic streptococci and S. pneumoniae S. pneumoniae), Borrelia, mycobacterium (especially M. tuberculosi), Neisseria (especially N. gonorrhoeae), treponema, and periodontal disease Bacteria (eg, Fusobacterium, Porphyromonas, Eikenella, Prevotella, Actinobacillus, and Capnocytophaga), viruses (eg, parvovirus, papovirus, herpes virus, togavirus, retrovirus (especially HIV), rhabdovirus, influenza virus, etc.) ) And lower eukaryotes (eg, fungi (eg, dermatophytes, pneumocystis, trypanosomes, etc.), yeasts, helminths, nematodes, etc.) can be detected and characterized.

Importantly, a combination of the nucleic acid amplification method and the multiparametric fluorescence in-situ hybridization method of the present invention can be used to examine cell quiescence or expression status. Thus, by using a probe specific for RNA produced by, for example, proliferating cells, cells expressing tumor antigens or hormones, the method of the present invention not only allows the presence of tumor cells, but also their malignancy. The degree can also be determined. Such mRN
A-profiling can also be performed in environments where standard cDNA hybridization approaches are not sensitive enough to detect small changes in the concentration of genetic material.

The combination of the nucleic acid amplification method and the multiparametric fluorescence in-situ hybridization method of the present invention can be performed using any suitable single or multiple amplification methods (Mullis, K. et al., Col.
d Spring Harbor Symp. Quant. Biol. 51: 263-273 (1986); Erlich H. et al.,
EP 50,424; EP 84,976, EP 258,017, EP
237,362; Mullis, K., EP 201,184; Mullis, K., U.S. Patent 4,683.
Erich, H., U.S. Pat. No. 4,582,788; Saiki, R. et al., U.S. Pat.
No. 683,194 and Higuchi, R. "PCR Technology," Ehrlich, H. (ed.), Stockton.
Press, NY, 1989, pp61-68), Strand Displacement Amplification (Walker,
GTet al, Proc. Natl. Acad. Sci. (USA) 89: 392-396 (1992); Walker et.
al., US Patent No. 5,270,184; Walker, US Patent No. 5,455,166); Ligase C
hain Reaction (Segev. WO90 / 01069; Birkenmeyer, WO93 / 00447) can be used, but a preferred amplification method is a rolling cycle type amplification (RCA) method (Caplan, M. et al.
et al., PCT Patent Application Publication WO 97/19193, the contents of which are incorporated herein by reference.

Rolling circle amplification (RCA) is an amplification strategy in which nucleic acid amplification is driven by a DNA polymerase that can replicate circularized oligonucleotide probes with linear or geometric kinetics under isothermal conditions. In particular, RCA involves incubating at least one rolling circle replication primer (RCRP) with at least one amplification target circle (ATC). ATC is RCR
Includes single-stranded circular DNA molecules containing a region complementary to RCRP so that P can hybridize to ATC and mediate amplification in the presence of DNA polymerase.

Two primers, one that hybridizes to the “+” strand of DNA and a “−”
In the presence of those that hybridize to the strands, a complex pattern of DNA strand displacements follows, which produces 10 ⁹ or more of each sensation in 90 minutes to produce human genomic DNA.
Allows the detection of medium point mutations. This extended cascade of strand displacement and fragment generation events is called "DNA hyperbranching", and the special rolling circle amplification driven by two primers is called "hyperbranched RCA" or "HRCA". With a single primer, RCA generates hundreds of tandem linked copies of a covalently closed ring in minutes.

Large fragments of exo (−) Vent DNA polymerase or Bst DNA polymerase (Aliotta, JM et al., Genet. Anal. 12: 185-195 (1996); Thomas, DC
et al., Clin. Chem. 43: 2219 Ab 38 (1997)), both of which are incorporated herein by reference, and T7 DNA polymerase sequenase.
It is preferred to perform HRCA with the 2.0 variant. Sequenase has been shown to support rolling circle-type amplification of circular oligonucleotides but relatively slowly (Fire, A. et al., Proc. Natl. Acad. Scad.
ci. (USA) 92: 4641-4645 (1995); Liu, D. et al., J. Amer. Chem. Soc. 118
: 1587-1594 (1996)). The efficiency of RCA and HRCA reactions is increased by the addition of proteins that bind single-stranded DNA. E. coli single-chain binding protein (SS
Addition of B) stimulates sequenase-catalyzed HRCA, whereas the phage T4 gene = 32 protein stimulates Vent exo (-)-catalyzed HRCA. In contrast, Bst polymerase does not require such single-stranded binding proteins to maximize activity.

When matrix-related, the DNA product remains attached to the site of synthesis where it is tagged, condensed, and imaged as a point source. A linear oligonucleotide probe covalently attached to a glass surface can generate an RCA signal whose color indicates the allelic status of the target as a result of a specific target-specific ligation event. R
Single molecule counting by CA provides a powerful mutation detection method for studies using oligonucleotide arrays and is particularly amenable to analysis of rare somatic mutations.

In yet another embodiment, the combined M-FISH / nucleic acid amplification method of the invention (particularly RC-
A / HRCA), a cyclizable oligonucleotide called a “padlock probe” linked to a single copy gene in a cytological specimen (Landegren, U. et al.,
Methods 9: 84-90 (1996); Landegren, U. et al., Ann. Med. 29: 585-590 (199
7); Nilsson, M. et al., Science 265: 2085-2088 (1994); Nilsson, M. et a
l., Nat. Genet. 16: 252-255 (1997)), the contents of which are incorporated herein by reference.

[0211] There are many uses for the gap-filling reaction described above, in which the target complementary sequence is incorporated into the ring by copying and covalent ring closure. In principle, all DNA sequences thus captured in circular DNA can be amplified by RCA or HRCA. Therefore, this reaction is useful in situations where it is desirable to interrogate the sequence incorporated into the padlock probe at some point after RCA. Longer sequences, such as microsatellite repeats, also need to be able to be copied to circularizable probes for amplification and analysis. After such a copy process, rolling circle amplification modifies the number of repeats incorporated into the circularization probe (Fire,
Acad et al., Proc. Natl. Acad. Sci. (USA) 92: 4641-4645 (1995)), so direct measurement of repeat size is feasible.

In particular, three alternative HRCA / RCA strategies for allele discrimination using ligation and rolling circle-type replication of circularizable DNA probes— “gap-probe RCA”, “polymerase-mediated gap-filling RCA” and "Ligase-mediated extension of the immobilized probe RCA" may be used in conjunction with the M-FISH method of the invention. For solution testing of complex genomes, a polymerase-mediated "gap-fill" reaction is preferred over a "gap probe" ligation reaction. This is because gap oligonucleotides that are not washed out in the solution assay are preferably used at relatively high concentrations for the ligation step, thus interfering with the HRCA reaction and inducing the formation of amplicon artifacts. is there. In contrast, the gap ligation reaction is
It may be preferable for allele analysis of DNA in cytological specimens. Here, the specificity of ligation needs to be enhanced for gapless probes because there are three different sequence recognition events and two separate ligation events must occur prior to "padlock" closure. is there. A third assay method, ligase-mediated elongation of oligonucleotides linked to a solid surface, reduces individual hybridization / ligation events and provides an overall approach to assess rare somatic mutations. Provide a new approach.

In a very preferred embodiment of HRCA / RCA, the RCA generated D
NA is labeled with a combinatorially labeled fluorescent DNP-oligonucleotide tag that hybridizes at multiple sites with the tandem DNA sequence. "Decorated" DNA labeled with a specific coding tag is then condensed to a small object by crosslinking with a multivalent antibody (e.g., anti-DNP IgM). The wild-type specific primer is a fluorescently labeled DNP
Producing an RCA product capable of hybridizing with the oligonucleotide tag,
On the other hand, the mutant RCA product hybridizes with the Cy3-labeled DNP-oligonucleotide. The method of condensing the amplification rings after hybridization of the encoded tag is called "CACHET". CAC associated with the M-FISH method of the present invention
The use of HET is particularly preferred.

Example 4 Use of M-FISH Analysis to Assess Telomere Integrity Conventional analysis of chromosomes using, for example, Giemsa staining of metaphase chromosomes shows some prominent defects. In particular, karyotypic changes involving fragments of chromosomes with very similar staining patterns cannot be elucidated using this method (Saccone, E. et al.
Natl. Acad. Sci. (USA) 89: 4913-4917 (1992)). Such findings are especially true for the telomere zone, which is ultimately all Giemsa-negative and has the most gene-rich zones in the genome. Accurate identification of the telomere region is very important because telomeres can be involved in breakage and healing events that can result in truncation, gene amplification or cryptic translocation.

In fact, hemoglobin H (Lamb, J et al., Lancet 2: 819-824 (1989)), Cri-
du-Chat (Overhauser, J. et al., Amer. J. Hum. Genet. 45: 296-303 (1989))
), Wolf-Hirschhorn (Altherr, MR et al., Amer. J. Hum. Genet. 49: 1235-
12342 (1991)) and Miller-Dieker encephalopathy syndrome (Kuwano, A. et al., Amer
Increasing numbers of clinical cases have been found to be the result of cryptic translocations, such as .J. Hum. Genet. 49: 707-714 (1991). In addition, cryptic translocations occur in up to 6% of the population with mild to moderate mental retardation that show no detectable chromosomal changes during karyotyping. Many of these genetic changes are not detected using standard cytogenetic analysis.

As described above, the present invention provides a method for hybridizing a complete set of chromosome-specific DNA probes, each labeled with a different combination of dyes,
And the ability to identify chromosomes simultaneously. One aspect of the present invention is the recognition, by the use of telomere-specific probes, that the methods of the present invention can be used to identify translocations that were not identifiable (ie, cryptic) using conventional methods. .
Thus, the present invention first provides a simple screening test to assess telomere integrity using a single hybridization reaction. Such multiplex hybridization assays significantly improve the ability to detect terminal deletions and cryptic chromosomal rearrangements, thereby avoiding detection by conventional cytogenetic staining or multicolor full-chromosome coloring. Promotes identification of statistical abnormalities. Thus, this aspect of the invention relates to karyotyping techniques (eg, Speicher, MR et al., Nature Genet.
12: 368-375 (1996); Speicher, MR et al., Bioimaging 4: 52-64 (1996); S
Chrock, E. et al., Science 273: 494-497 (1996)) and telomere integrity test techniques (see, for example, Ledbetter, DH, Amer. J. Hum. Genet. 51: 451-456 (1992)). Expand traditional efforts in both.

To illustrate this ability of the present invention, a set of YAC clones containing human DNA sequences specific for the subtelomeric region of each chromosomal arm is collected. Because no suitable subtelomeric YAC for the short arm was identified, this probe set used microdissection probes for the short arm of the Y chromosome (Guan, XY, Nature Genet. 12: 10-11 (1996)).
This description is incorporated herein by reference). A YAC for the short arm could be identified. Similarly, due to extensive cross-hybridization, the probe set did not include a probe for the p-arm of acrocentric chromosomes 13, 14, 15, 21 and 22. The probes are combinatorially labeled such that the telomeric regions of these chromosomes are visualized in different pseudocolors based on their unique fluorescent composition.

The p-arm and q-arm telomere proximal probes for each chromosome are labeled with the same “color code”. For example, a chromosome 1 telomere is labeled with color "a", while a chromosome 2 telomere is labeled with color "b", and so on. Thus, cytogenetic cryptic chromosomal translocations that occur in non-acrocentric chromosomes are visualized in metaphase spreads as chromosomes that carry two colors rather than a single color per chromosome. The 24-color telomere integrity assay uses a complete probe mixture, but such probe cocktails can hybridize with high reproducibility. Such hybridization is visualized as above, for example, with an epifluorescence microscope equipped with a suitable filter set and a colored CCD camera. Fluorophores and high contrast filters
Same as above. In short, fluorescein (FLU) for direct labeling,
Cy3 and Cy5 are ligated to dUTP. Cy3.5 and Cy7 are available as avidin or anti-digoxigenin conjugates for secondary detection of biotin- or digoxigenin-labeled probes. One to three separate nick translations are performed for each probe.

The telomere probe set consists of a YAC, Half-Yac (Vocero-Akbani, A. et al., Genomics 36: 492-5), which is selected to contain most telomere gene markers.
06 (1996); Macini, RA et al., Genomic Res. 5: 225-232 (1995)) and C
It consists of a mixture of EPH-YAC (Chumakov, IM et al., Nature 377: 175-297 (1995)). Half-YAC is generated from a specialized YAC library enriched for telomere sequences. "Half-YAC" refers to the fact that one vector telomeric sequence of the molecule is provided by human DNA rather than yeast DNA. Half
-YACs contain subtelomeric repeats that are known to be residues on different chromosomes, so in some cases they hybridize with multiple telomeres. However, the specific M-FISH signal is, for example, a DNA amplification (eg,
(Using PCR with Alu primers to reduce the relative representation of repetitive sequences). Addition of Cot-1 DNA to the hybridization cocktail suppresses hybridization of any residue repeats. Suitable YAC, half-Yac and / or CePH-YAC are Vocero-Ak
bani, A. et al. (Genomics 36: 492-506 (1996)).
of Health and Institute of Molecular Medicine Collaboration (Nature Gen
et 14: 86-89 (1996)) and by Bray-Ward, P. et al. (Genomics 36: 1-1).
4 (1996)) (all of which are incorporated herein by reference).
Is described. The fluorophores used to label the telomere region are shown in Table 4. In Table 4, the following were referenced: (1) Vocero-Akbani, A. et al., Genom.
ics 36: 492-506 (1996); (2) National Intitutes of Health and Institute
of Molecular Medicine Collaboration (Nature Genet 14: 86-89 (1996));
(3) Bray-Ward, P. et al. (Genomics 36: 1-14 (1996)); (4) Bray-Ward,
P. et al. (Genomics 36: 104-111 (1996)); (5) Guan, XY, Nature Gene
t. 12: 10-11 (1996) (all of which are incorporated herein by reference).

[0220]

[Table 5]

[0221]

[Table 6]

[0222]

[Table 7]

[0223]

[Table 8]

To examine the reproducibility of the specific signal with the halfYAC probe, each is carefully examined in a light-colored experiment on normal metaphase spreads from at least 5 different donors. These tests allow assessment of polymorphism, as well as frequency or cross-hybridization with multiple telomere repeats. An appropriate half-YAC performs highly reproducibly. In a few cases where occasional cross-hybridization occurs, the non-specific signal usually has a very low fluorescence intensity and can therefore be easily distinguished from the true signal.

[0225] CEPH-YAC was selected to contain most telomeric gene markers. Because the distance of the chromosome from the true telomere end is unknown, these probes were examined by FISH for their distance band location. Six telomere regions (
For 2p, 2q, 5q, 7q, 8p, 8q), hybridization of a single telomere YAC resulted in low fluorescence intensity. Therefore, to increase the signal, two (
5q, 7q, 8p, 8q) or 3 (2p, 2q) subtelomeric YACs were pooled. Alu
-Using fingerprint analysis to confirm that the YACs overlap.

An improved version of the previously published M-FISH software for whole chromosome coloring probes was used for probe identification (Speicher, MR et al., Bioimaging
4: 52-64 (1996); Schrock, E. et al., Science 273: 494-497 (1996); Yale U
niversity). Briefly, whole chromosome-colored probes (for each fluorophore used in the experiment) are hybridized simultaneously with the pool of telomeric YACs to correct for image shifts caused by optical and mechanical defects. The pixel shift correction of the chromosome coloring probe is performed in the same manner as above, however, the chromosome color is not shown on the final image to facilitate the color identification of the subtelomeric probe.
Instead of calculating a specific threshold for each fluorophore based on the whole chromosome coloring signal, the chromosome end regions are specifically analyzed. Chromosome ends are identified and segmented masks with different thresholds are calculated and checked for regions showing increased fluorescence intensity. This is necessary because simultaneous hybridization of a large number of small region-specific probes often results in signals being at different focal planes. In the absence of Z-axis optical truncation and image merging, this indicates that some Y
This produces an AC probe. If the specific telomere region is not labeled with a particular fluorophore, no increase in fluorescence intensity peak is observed with at least one segmented mask.
Above the calculated fluorophore segmentation mask, the DAPI segmentation mask delineating the chromosome boundaries did not overlap (since some of the telomere FISH signals could be outside the DAPI segmentation mask). Individual YAC clones are first assigned to different gray values depending on the Boolean characteristics of each probe or the combination of fluorophores used to label it (Speicher, MR et al., Bioimaging 4: 52-64 (1996)
Schrock, E. et al., Science 273: 494-497 (1996)). Then, using a look-up table, each DNA target is assigned to a pseudo color by this gray value. Finally, this pseudo-colored image was overlaid on the DAPI-stained chromosome image to which blue light was assigned.

To optimize the experimental parameters for simultaneously hybridizing a large number of such combinatorially labeled YAC clones, multiple telomere probe sets were hybridized with normal metaphase spreads obtained from peripheral blood lymph. Let it. Similar to the above results obtained using the whole chromosome coloring probe, Y labeled with equal amounts of different fluorophores
AC probes do not always show the same signal strength for each fluorophore. To reduce signal intensity differences, control experiments establish the probe concentration for the hybridization formulation and a reliable combinatorial labeling scheme. YACs producing a large fluorescent signal were preferentially labeled with three different fluorophores, and YACs producing a much weaker signal were labeled with only one fluorophore.

Using such probes, a 24-color telomere integrity assay is performed on the karyotypes of 10 normal male and female donors. No indication of polymorphism was detected, suggesting that cryptic translocations are very infrequent in the normal population. A typical metaphase spread is observed (FIG. 5A), and a normal karyotype based on the Boolean properties of our subtelomeric probe is obtained as expected (FIG. 5B).

In the step of examining the 24-color telomere integrity assay, the probe set is hybridized on metaphase spreads from a myeloproliferative disease patient. Karyotyping using the G band
Chromosome 8 trisomy is designated as the only detectable cytogenetic change. This is demonstrated by M-FISH using a chromosome-specific coloring probe. However, the telomere integrity assay produces an interesting hybridization pattern on chromosome 8. Chromosome 8 trisomy is demonstrated in all 10 metaphase spreads analyzed. However, for eight of these ten metaphase spreads, a split telomere signal was observed on two chromosomes 8, indicating an inversion in this region (FIG. 6). This analysis demonstrates the power of this novel technique to detect structural abnormalities that cannot be detected by standard cytogenetic methods or 24-color techniques using chromosome-specific coloring probes.

[0230] The application of the telomere integrity test is of great value to patients with mental retardation,
Assessment of the karyotype of patients with a combination of mental retardation and dysmorphic features, as well as patients with cancer cells known to have high genomic instability. The latter cells would be very susceptible to subtelomeric translocation events. Screening for cryptic translocations cannot be performed efficiently with chromosome-specific coloring probes because they were not intended to detect insidious deletions or rearrangement events.

In yet another embodiment of the above method, PCR-assisted chromatography can be used to improve telomere-specific probe sets (Craig, J. et al., Hum. Gen.
et. 100: 472-476 (1997)), the contents of which are incorporated herein by reference. This tool removes repetitive DNA, including polymorphic repeats from haldYAC or other M-FISH probes, thereby reducing Cot-1 suppressor DNA.
Promotes the generation of probes that specifically hybridize in the absence of

Although the invention has been described in connection with specific embodiments thereof, the invention is capable of further modifications, which adaptations in accordance with the principles of the invention are well known in the art to which the invention pertains. Or any modification of the invention, including any such departure from the disclosure, as a result of a conventional practice and as applicable to the essential features described above and in accordance with the appended claims. , Use or indication and should be understood to be.

[Brief description of the drawings]

FIG. 1 presents a schematic of a CCD camera and microscope used in accordance with the method of the present invention.

FIG. 2 shows raw data from karyotyping of chromosomes from a bone marrow patient (BM2486).
A chromosome "mask" generated by the software program is adjacent to each light source image. In FIG. 2, panels A and B are DAPI images and masks; panels C and D are FITC images and masks; panels E and F are Cy3 images and masks; panels G and H are Cy3.5 images and masks. Mask; panels I and J are Cy5
Images and masks; and panels K and L are Cy7 images and masks.

FIGS. 3A and 3B show the identification of individual chromosomes by spectral characteristics of patient BM2486. FIG. 2 shows that FIG. 3A except that it has been grayscale pseudo-colored.
It is the same photograph as. FIG. 3B shows the karyotype sequence of the chromosome.

FIG. 4 shows the differentiation of bacteria by the in-situ hybridization method. Panels A to E
Shows three bacteria (F. nuc) using a mixture of genomic DNA probes for each organism.
1 shows the in-situ hybridization assay results for a laboratory-derived mixture of C. leatum, A. actinomycetemcomitans and E. corrodens). Panel A shows all bacteria present in the microscopic field detected by the common DNA binding fluorophore (DAPI). Panel B shows F. nucleatum using cascade blue detection. Panel C shows A. actinomycetemcomitans using FITC detection. Panel D shows E. corrodens using rhodamine detection. Panel E is a computer-combined composite of Panels BD showing the discrimination of each bacterium in the sample. Panel F shows C. g when hybridized with a combination of genomic DNA probes for those organisms.
A similar analysis of a mixture of ingivalis (using rhodamine detection) and P. intermedia (using FITC detection). Panel G shows an in-situ hybridization assay for a combination of seven different bacteria hybridized with a mixture of seven genomic DNA probes. In panel G, numbers (1)-(7) identify bacteria. F.
nucleatum (1), E. corrodens (2) and A. actinomycetemcomitans (3)
Is shown in blue (cascade blue detection). P. gingivalis (4) and C. oc
hracea (5) is shown in red (rhodamine detection), while P. intermedia (6)
And C. gingivalis (7) are observed in yellow (FITC detection). Panel H shows A. actinomycetemcomitans in plaque samples obtained from patients with localized juvenile periodontitis.
1 shows an in-situ hybridoma formation assay for the presence of (using FITC detection).

FIG. 5A shows normal male metaphase chromosomes developed after hybridization with a 24-color set of telomere-specific probes, shown as pseudocolor images. FIG. 5B
Indicates the final karyotype generated based on the Boolean spectral properties of the telomere-specific probe.

FIG. 6A shows chromosome 8 subtelomeric YACs for expanded normal metaphase chromosomes.
1 shows the hybridization pattern of a telomere-specific probe). FIG. 6B shows the same probe for metaphase chromosomes developed from myeloproliferative disease patients (FIG. 6A).
(Same as the one used in (1)). Conventional cytogenetic analysis of this patient using G-staining revealed the chromosome 8 trisomy as the only change. The M-FISH telomere-specific probe has two of the three chromosomes 8
The split telomere signal for the short arm of the book is shown, indicating an additional change, an inversion in this region.

[Procedure amendment]

[Submission date] December 12, 2000 (200.12.12)

[Procedure amendment 2]

[Document name to be amended] Drawing

[Correction target item name] Fig. 1

[Correction method] Change

[Correction contents]

FIG.

──────────────────────────────────────────────────続 き Continued on the front page (72) Inventor Ballard, Stephen Gwyn, United States, Connecticut 06517, Hamden, Hartford Turnpike 459 (72) Inventor Wilson, John Tee. 31522, Georgia, United States, Cent Simon Island, Beach Drive 419 F-term (reference) 2G045 AA25 AA28 AA40 DA12 DA13 DA14 FB01 FB02 FB04 FB07 FB12 GC15 4B024 AA11 AA20 CA09 HA14 4B063 QA01 QA13 QA18 QQ02 QQQL Q

Claims (57)

[Claims] 1. A combinatorially labeled oligonucleotide probe set comprising a first and second sub-set of probes, wherein (A) each member of the first sub-set of probes comprises: Is linked or linked to a predetermined label that is distinguishable from the label of any other member of the two-subset probe, and (ii) specifically hybridizes to one of the predetermined autosomal or sex chromosomes of the human karyotype. The plurality of oligonucleotides obtained, wherein said first subset of probe sets has sufficient members to specifically hybridize each autosomal or sex chromosome of the human karyotype with at least one member, and (B) each member of the probes of the second sub-set is linked or conjugated to a label that is distinguishable from (i) the label of any other member of the first or second sub-set; and ii A) a probe set comprising a plurality of oligonucleotides capable of specifically hybridizing to one of the extrachromosomal polynucleotide copies of the predetermined region of the autosomal or sex chromosome of the human karyotype. 2. The combinatorially labeled oligonucleotide probe set of claim 1, wherein said extrachromosomal polynucleotide copy of said human karyotype autosomal or sex chromosomal putative region is an RNA molecule. 3. The combinatorially labeled oligonucleotide probe set of claim 1, wherein said extrachromosomal polynucleotide copy of said human karyotype autosomal or sex chromosome predetermined region is a DNA molecule. 4. The combinatorially labeled oligonucleotide probe set of claim 1, wherein said set contains at least one member probe specific for said human karyotype autosomal or sex chromosome subchromosomal fragment. 5. A combinatorially labeled oligonucleotide probe set comprising a first sub-set of genotype probes and a second sub-set of phenotype probes, comprising: (A) each of the first sub-set of genotype probes. The member is (i) linked or linked to a predetermined label that is distinguishable from the label of any other member of the first or second sub-set of probes, and (ii) a preselected bacterium, virus or lower A plurality of oligonucleotides capable of specifically hybridizing to a region of a eukaryotic nucleic acid, wherein said first sub-set of probe sets comprises a pre-selected bacterium, virus or lower eukaryote to another bacterium, virus or Having sufficient members to be distinguishable from lower eukaryotes, and (B) each member of the phenotypic probe of the second sub-set comprising: (i) any other member of the first or second sub-set; The member signs Linked to or linked to a predetermined label that is distinguishable, and (ii) said pre-selected bacterium, virus or lower eukaryote to be able to determine whether it exhibits a pre-selected phenotype. A probe set comprising a plurality of oligonucleotides capable of specifically hybridizing to a predetermined polynucleotide region of a chromosome of a selected bacterium, virus or lower eukaryote or an extrachromosomal copy thereof. 6. The combinatorially labeled oligonucleotide probe set of claim 5, wherein said extrachromosomal polynucleotide copy of said predetermined region of said chromosome is an RNA molecule. 7. The combinatorially labeled oligonucleotide probe set of claim 5, wherein said extrachromosomal polynucleotide copy of said predetermined region of said chromosome is a DNA molecule. 8. The combinatorially labeled oligonucleotide probe set of claim 5, wherein said predetermined region of said chromosome of said bacterium, virus or lower eukaryote encodes an antibiotic resistance determinant. 9. The combinatorially labeled oligonucleotide probe set of claim 5, wherein said predetermined region of said chromosome of said bacterium, virus or lower eukaryote encodes a toxin. 10. The combination of claim 5, wherein the predetermined region of the chromosome of the bacterium, virus or lower eukaryote encodes a protein that is determinant of the species of the bacterium, virus or lower eukaryote. Labeled oligonucleotide probe set. 11. The predetermined region of the chromosome of the bacterium, virus or lower eukaryote encodes a protein that is a determinant of the species subspecies of the bacterium, virus or lower eukaryote. Item 11. The combinationally labeled oligonucleotide probe set of Item 10. 12. The combinatorially labeled oligonucleotide probe set of any of claims 3 and 7, wherein said DNA molecule is generated by in-situ nucleic acid amplification. 13. The combinatorially labeled oligonucleotide probe set of any of claims 3 and 7, wherein said in-situ nucleic acid amplification is an in-situ polymerase chain reaction. 14. The combinatorially labeled oligonucleotide probe set of any of claims 3 and 7, wherein said in-situ nucleic acid amplification is an in-situ rolling circle amplification reaction. 15. The combinatorially labeled oligonucleotide probe set of any one of claims 3 and 7, wherein said probe contains a nucleotide residue labeled with a biotin component. 16. The combinatorially labeled oligonucleotide probe set of claim 15, wherein said probes are further labeled with a labeled biotin binding ligand. 17. The set of combinatorially labeled oligonucleotide probes of claim 16, wherein at least one of said probes is labeled with a biotin binding ligand that includes one or more fluorophores. 18. The method of claim 18, wherein at least one of said fluorophores is a fluorophore F.
18. The combinatorially labeled oligonucleotide probe set of claim 17, wherein the set is selected from the group consisting of ITC, Cy3, Cy3.5, Cy5, Cy5.5 and Cy7. 19. The set of combinatorially labeled oligonucleotide probes of claim 17, wherein at least one of said probes is labeled with a biotin binding ligand that includes more than one fluorophore. 20. The biotin-binding ligand comprising one or more fluorophores, wherein one of the fluorophores is a fluorophore FITC, Cy3, Cy3.5,
20. The combinatorially labeled oligonucleotide probe set of claim 19, wherein the set is selected from the group of Cy5, Cy5.5 and Cy7. 21. The biotin-binding ligand comprising more than one fluorophore, wherein all of the fluorophores are fluorophores FITC, Cy3, Cy3.5,
21. The combinatorially labeled oligonucleotide probe set of claim 20, wherein the set is selected from the group of Cy5, Cy5.5 and Cy7. 22. The method of claim 20, wherein at least one member of the set is a fluorophore FITC.
, Cy3, Cy3.5, Cy5, Cy5.5 and Cy7
2. The method of claim 1, wherein the member is labeled with more than one fluorophore, and each member of the set is labeled with at least one fluorophore selected from the fluorophore group.
7. Combinatorially labeled oligonucleotide probe set 7. 23. At least one member of said set each comprises a fluorophore F
At least one fluorophore labeled with two fluorophores selected from the group consisting of ITC, Cy3, Cy3.5, Cy5, Cy5.5 and Cy7, and wherein each member of the set is selected from the fluorophore group 3. The label of claim 2,
Two combinatorially labeled oligonucleotide probe sets. 24. At least one member of said set each comprises a fluorophore F
At least one fluorophore labeled with three fluorophores selected from the group consisting of ITC, Cy3, Cy3.5, Cy5, Cy5.5 and Cy7, and wherein each member of the set is selected from the fluorophore group 3. The label of claim 2,
Two combinatorially labeled oligonucleotide probe sets. 25. A method for simultaneously identifying and differentiating individual autosomes and sex chromosomes of a human karyotype, comprising: (I) single-stranded under conditions sufficient to effect nucleic acid hybridization. A sample of a chromosome in the form of a combinatorially labeled oligonucleotide probe set comprising first and second sub-sets of probes, wherein (A) each member of said first sub-set of probes comprises: Is linked or linked to a predetermined label that is distinguishable from the label of any other member of the one or second subset of probes, and (ii) specifically binds to one of the predetermined autosomal or sex chromosomes of the human karyotype. A plurality of oligonucleotides capable of hybridizing, wherein said first subset of probe sets specifically hybridizes each autosomal or sex chromosome of said human karyotype with at least one member. Have enough members to
And (B) each member of the second subset of probes is linked or conjugated to (i) a predetermined label that is distinguishable from the label of any other member of the first or second subset; (Ii) contacting with a probe set including a plurality of oligonucleotides capable of specifically hybridizing with one of the extrachromosomal polynucleotide copies of the predetermined region of the autosomal or sex chromosome of the human karyotype; For each chromosome of the sample that has hybridized to a member of a sub-set of probes, the member's intended label is detected and identified, and the member's label identity is specifically hybridized to the member. Correlating with the identity of the autosomal or sex chromosomes of the human karyotype, thereby identifying a chromosome hybridized to the member; (III) the human karyotype Repeating step (II) until each of the autosomes and sex chromosomes of the probe is identified in the specimen; and (IV) the probe of For each member of the second subgroup, the expected label of that member is detected and identified, and the identity of the label of that member is compared to the autosomal or sex chromosomes of the human karyotype to which the member specifically hybridizes. Identifying the region of the autosome or sex chromosome that has been correlated with the identity and thereby hybridized to the member; (V) repeating step (IV) for each member of the second subset of probes A method consisting of steps. 26. The method of claim 25, wherein said extrachromosomal polynucleotide copy of said predetermined region of said human karyotype autosome or sex chromosome is an RNA molecule. 27. The method of claim 25, wherein said extrachromosomal polynucleotide copy of said predetermined region of said human karyotype autosome or sex chromosome is a DNA molecule. 28. The method of claim 25, wherein said set of oligonucleotide probes contains at least one member probe specific for said human karyotype autosomal or sex chromosome subchromosomal fragment. 29. A method according to claim 29, wherein the preselected bacterium, virus or lower eukaryote is replaced with another bacterium,
A method for simultaneously identifying and differentiating a virus or lower eukaryote, comprising: (I) said pre-selected bacterium, virus or lower under conditions sufficient to cause in-situ nucleic acid hybridization. A combinatorially labeled oligonucleotide probe set comprising a first sub-set of genotype probes and a second sub-set of phenotypic probes, comprising: (A) said first sub-set. Wherein each member of the genotype probe of (i) is linked or conjugated to a predetermined label that is distinguishable from the label of any other member of the first or second subset of probes; and (ii) A plurality of oligonucleotides capable of specifically hybridizing to a region of the nucleic acid of the selected bacterium, virus or lower eukaryote, wherein said first subset of probe sets is (B) having sufficient members to distinguish a constant bacterium, virus or lower eukaryote from other bacteria, viruses or lower eukaryotes present in said specimen; Each member of the probe is (i) linked or linked to a predetermined label that is distinguishable from a label of any other member of the first or second sub-set, and (ii) the preselected bacterium, virus or A predetermined polynucleotide region of said nucleic acid of said preselected bacterium, virus or lower eukaryote or an extrachromosomal copy thereof to enable determination of whether the lower eukaryote exhibits a preselected phenotype. Contacting a probe set comprising a plurality of oligonucleotides capable of specifically hybridizing; (II) hybridizing to a region of a chromosome of a preselected bacterium, virus or lower eukaryote. For each member of the first subgroup of probes, the expected label of that member is detected and identified, and the identity of that member's label is determined by the bacteria, virus or lower eukaryote to which that member specifically hybridizes. Correlating with the identity of the organism,
Identifying a bacterium, virus or lower eukaryote that has hybridized to said member; (III) repeating step (II) for each member of said first sub-set of probes; For each member of the second subset of probes hybridized to the selected polynucleotide region of the chromosome of the selected bacterium, virus or lower eukaryote, or an extrachromosomal copy thereof, detecting and identifying the predetermined marker for that member. And correlating the identity of the label of the member with the identity of the predetermined region, thereby identifying the presence of the predetermined region on the chromosome of the preselected bacterium, virus or lower eukaryote; C.) Repeating step (IV) for each member of the second subset of probes. 30. The method of claim 29, wherein said extrachromosomal polynucleotide copy of said predetermined region of said chromosome of said bacterium, virus or lower eukaryote is an RNA molecule. 31. The method of claim 29, wherein said extrachromosomal polynucleotide copy of said predetermined region of said chromosome of said bacterium, virus or lower eukaryote is a DNA molecule. 32. The method of claim 29, wherein said predetermined region of said chromosome of said bacterium, virus or lower eukaryote encodes an antibiotic resistance determinant. 33. The method of claim 29, wherein said predetermined region of said chromosome of said bacterium, virus or lower eukaryote encodes a toxin. 34. The method of claim 29, wherein said predetermined region of said chromosome of said bacterium, virus or lower eukaryote encodes a protein that is determinant of said bacterial, virus or lower eukaryote species. . 35. The predetermined region of the chromosome of the bacterium, virus or lower eukaryote encodes a protein that is a determinant of the species subspecies of the bacterium, virus or lower eukaryote. Clause 34. The method of clause 34. 36. The method of claim 27, wherein said in-situ nucleic acid amplification is an in-situ polymerase chain reaction. 37. The method of claim 36, wherein said in-situ nucleic acid amplification is an in-situ rolling circle amplification reaction. 38. The method of claim 36, wherein said probe contains a nucleotide residue labeled with a biotin moiety. 39. The method of claim 36, wherein said probe is further labeled with a labeled biotin binding ligand. 40. The method of claim 39, wherein at least one of said probes is labeled with a biotin binding ligand that includes one or more fluorophores. 41. The method according to claim 41, wherein at least one of said fluorophores is a fluorophore F.
41. The method of claim 40, wherein the method is selected from the group consisting of ITC, Cy3, Cy3.5, Cy5, Cy5.5, and Cy7. 42. The method of claim 40, wherein at least one of said probes is labeled with a biotin binding ligand that includes more than one fluorophore. 43. The biotin-binding ligand comprising one or more fluorophores, wherein one of the fluorophores is a fluorophore FITC, Cy3, Cy3.5,
43. The method of claim 42, wherein the method is selected from the group of Cy5, Cy5.5, and Cy7. 44. The biotin-binding ligand comprising more than one fluorophore, wherein all of the fluorophores are fluorophores FITC, Cy3, Cy3.5,
44. The method of claim 43, wherein the method is selected from the group of Cy5, Cy5.5, and Cy7. 45. The method wherein at least one member of the set is a fluorophore FITC
, Cy3, Cy3.5, Cy5, Cy5.5 and Cy7
5. The method of claim 4, wherein the member is labeled with more than one fluorophore, and each member of the set is labeled with at least one fluorophore selected from the fluorophore group.
0 method. 46. The at least one member of said set each comprising a fluorophore F
At least one fluorophore labeled with two fluorophores selected from the group consisting of ITC, Cy3, Cy3.5, Cy5, Cy5.5 and Cy7, and wherein each member of the set is selected from the fluorophore group 5. The method of claim 4, wherein
Method 5. 47. Each of the at least one member of the set is a fluorophore F
At least one fluorophore labeled with three fluorophores selected from the group consisting of ITC, Cy3, Cy3.5, Cy5, Cy5.5 and Cy7, and wherein each member of the set is selected from the fluorophore group 5. The method of claim 4, wherein
Method 5. 48. The method of any of claims 25 and 29, wherein an optical comb is used in step (b) to detect a predetermined label of the hybridized probe member. 49. The method of any of claims 25 and 29, wherein an optical filter is used in step (b) to detect a predetermined label of said hybridized probe member. 50. The method of claim 49, wherein a plurality of optical filters are used sequentially in step (b) to detect a predetermined label of said hybridized probe member. 51. The method wherein at least one member of the set is a fluorophore FITC
, Cy3, Cy3.5, Cy5, Cy5.5 and Cy7
5. The method of claim 4, wherein the member is labeled with more than one fluorophore, and each member of the set is labeled with at least one fluorophore selected from the fluorophore group.
Method 8. 52. Each of the at least one member of the set is a fluorophore F
At least one fluorophore labeled with two fluorophores selected from the group consisting of ITC, Cy3, Cy3.5, Cy5, Cy5.5 and Cy7, and wherein each member of the set is selected from the fluorophore group 5. The method of claim 4, wherein
Method 9. 53. At least one member of said set each comprising a fluorophore F
At least one fluorophore labeled with three fluorophores selected from the group consisting of ITC, Cy3, Cy3.5, Cy5, Cy5.5 and Cy7, and wherein each member of the set is selected from the fluorophore group 5. The method of claim 4, wherein
Method 9. 54. A set of combinatorially labeled oligonucleotide probes, comprising:
Each member of which (i) has a predetermined marker that is distinguishable from that of any other member of the set, and (ii) specifically hybridizes to a telomere region of a single autosomal or sex chromosome of a human karyotype. A set of probes that are capable of soybeans, the set of probes having sufficient members to specifically hybridize each of the human karyotype autosomes or sex chromosomes with at least one member. 55. The method according to claim 55, wherein at least one of the labels is a fluorophore FITC, C
55. The combinatorially labeled oligonucleotide probe set of claim 54, which is a fluorophore selected from the group consisting of y3, Cy3.5, Cy5, Cy5.5 and Cy7. 56. A method for simultaneously identifying and distinguishing individual autosomes and sex chromosomes of a human karyotype, comprising: (a) single-stranded under conditions sufficient to effect nucleic acid hybridization. Forming a sample of said chromosome in a form of a combinatorially labeled oligonucleotide probe set, each member of which (i) has a predetermined label distinguishable from the label of any other member of said set; and (ii) ) Is capable of specifically hybridizing to a telomere region of a predetermined autosomal or sex chromosome of a human karyotype, wherein the set specifically hybridizes each autosomal or sex chromosome of the human karyotype to at least one member. Contacting a set of combinatorially labeled oligonucleotide probes having sufficient members to be capable of soybean, wherein said contacting reduces the number of each autosomal or sex chromosome of said specimen. And (b) detecting, for each chromosome of the specimen that has hybridized to a member of the set of probes, a prospective marker for that member. And identifying and correlating the identity of the label of the member with the identity of the autosomal or sex chromosome of the human karyotype to which the member specifically hybridizes, whereby the chromosome hybridized to the member. And (c) repeating step (b) until each of said human karyotype autosomes and sex chromosomes is identified in said specimen. 57. The method wherein at least one of said labels is a fluorophore FITC, C
57. The method of claim 56, which is a fluorophore selected from the group consisting of y3, Cy3.5, Cy5, Cy5.5, and Cy7.