WO2007030373A2

WO2007030373A2 - Method for in situ hybridization analysis

Info

Publication number: WO2007030373A2
Application number: PCT/US2006/034005
Authority: WO
Inventors: Thomas Curran; Dennis S. Rice; Susan Magdaleno; Patricia Jensen
Original assignee: St. Jude Children's Research Hospital
Priority date: 2005-09-07
Filing date: 2006-08-30
Publication date: 2007-03-15
Also published as: WO2007030373A3

Abstract

The present invention is a method for in situ hybridization analysis of the temporal and/or spatial expression of one or more ribonucleic acid target sequences in a tissue sample. The disclosed method can be adapted for high- throughput analysis of a plurality of target sequences and a plurality of tissue samples and the method is highly sensitive based upon the combination of a radiolabeled ribonucleic acid probe and silver emulsif ication of the tissue sample to detect the temporal and/or spatial expression of the target sequence.

Description

METHOD FOR IN SITU HYBRIDIZATION ANALYSIS

Introduction

This invention was made in the course of research sponsored by the National Institute of Neurological

Disorders and Stroke (Grant Nos. NOl-NS-0-2331 and R37

NS36558-06) . The U.S. government may have certain rights in this invention.

Background of the Invention

The challenge of the post-genomic era is not only to assign functions to individual genes, but also to determine how sets of genes act in concert to control biological processes. This task is daunting when one attempts to understand the complex genetic programs underlying, e.g., nervous system development. More than half of the approximately 25,000 genes in the mouse genome are thought to be involved in development and function of the nervous system (Nekrutenko (2004) MoI. Biol. Evol . 21:1278-82; Waterston, et al . (2002) Nature 420:520-62), but only 30% of genes have any function assigned to them (Kiss-Toth, et al. (2004) Cytokine Growth Factor Rev. 15:97-102). Identifying the temporal and spatial expression patterns of these genes throughout development is a critical initial step in their functional analysis. High throughput in situ analysis of gene expression is hindered by labor intensive tissue sample preparation and analysis methods. U.S. Patent No. 6,696,271 discloses arraying frozen tissues into a frozen recipient block for sectioning. Tissues in accordance with the disclosed method are not fixed prior to embedding, and sections from the array are evaluated without fixation or post-fixed according to the appropriate methodology used to analyze a specific gene at the DNA, RNA, and/or protein levels.

Summary of the Invention The present invention is a method for preparing a tissue sample for microscopic analysis. The method involves embedding at least two fixed tissues, each of a different age classification or plane of orientation, in a single embedding mold and sectioning the embedded, fixed tissue so that a tissue sample is prepared for microscopic analysis. In particular embodiments, at least three or at least four fixed tissues are embedded.

The present invention is also a method for determining the temporal or spatial expression of a ribonucleic acid target sequence in a tissue sample. The method involves contacting a tissue sample with a radiolabeled ribonucleic acid probe which is complementary to a single ribonucleic acid target sequence in the tissue sample so that the radiolabeled ribonucleic acid probe binds to the ribonucleic acid target sequence; exposing the tissue sample to silver emulsion for a selected period of time based upon the intensity of the signal produced by the bound radiolabeled ribonucleic acid probe; and developing and fixing the tissue sample so that the temporal or spatial expression of the ribonucleic acid target sequence in the tissue sample can be determined. In one embodiment, the tissue sample is imaged and analyzed. In another embodiment, a tissue sample image is stored in a database.

Detailed Description of the Invention

Novel methods for sectioning and analyzing multiple tissues have now been developed. It has been found that two or more fixed tissues can be placed in a single embedding mold, embedded in a single block of tissue, and simultaneously sectioned. This method is particularly useful for simultaneously sectioning a fixed tissue in two or three planes of orientation or for sectioning a fixed tissue taken from an organism at different ages of development so that spatial or temporal detection of RNA, DNA, or protein can efficiently be achieved by analyzing one tissue sample. Tissue samples prepared in accordance with the instant method can be readily adapted to high throughput in situ analysis. In particular embodiments, in situ hybridization analysis of tissue samples' is carried out by contacting a tissue sample with a radiolabeled ribonucleic acid probe which is complementary to and binds a single ribonucleic acid target sequence in the tissue sample, exposing the tissue sample to silver emulsion, and developing and fixing the tissue sample so that the temporal and/or spatial expression of the target sequence can be determined. Advantages of the instant methods are adaptability to high throughput gene analysis, cost- effectiveness, and breadth of data that can be generated.

As used herein, a tissue is a group of cells in an organism (e.g., a mammal, plant, or other multicellular organism) that perform the same or related functions. Tissues of an mammal include, e.g., connective tissue such as bone, blood, adipose, ligaments and tendons,- epithelial tissue which covers organ surfaces and includes skin and the inner lining of the digestive tract; muscle tissue including skeletal, cardiac, and smooth muscle; or nervous system tissue such as the brain, spinal cord, and peripheral nervous system. Tissues of a plant include parenchyma, collenchyma, sclerenchyma, xylem, phloem or epidermal tissue. In one embodiment, a tissue can be a particular organ, e.g., heart, lung, kidney, liver, bladder, eye, pancreas, root, stem, leaf, flower, etc. In another embodiment, a tissue can be a whole organism (e.g., an embryo or a seedling) .

Tissues prepared in accordance with the instant method are first fixed with a chemical fixative to preserve the tissue structure and chemical composition. Desirably, the fixative penetrates into the tissue, prevents putrefaction by bacteria, stabilizes and hardens the tissue against the detrimental effects of subsequent processing and staining procedures, and prevents changes and alteration of the tissue. Chemical fixation can employ a coagulant chemical fixative such as mercuric chloride, picric acid, chromium trioxide or ethanol, which change the fine fibrous network or meshwork of proteins of tissues into a coarse network. This does not destroy the structure at the microscopic level, but permits the embedding compound to easily infiltrate the interior of the tissue to form a tissue of proper consistency for sectioning. Another advantage is that some coagulant chemical fixatives strengthen the chemical bonds within and between proteins against breakdown during later histological steps. Alternatively, chemical fixation can employ a noncoagulant chemical fixative such as formaldehyde, potassium dichromate, acetic acid or osmium tetroxide, which do not drastically change the fine network of proteins of tissues thereby providing good histological and cytological details. Exemplary fixatives include, but are not limited to paraformaldehyde or formaldehyde. In general, chemical fixatives are used at a concentration ranging from about 2% to about 80% depending on the fixative and the tissue being processed.

Fixation time is dependent upon the size of the specimen and can vary from 1 hour to 48 hours or more. Fixed tissues are prepared for sectioning by embedding multiple fixed tissues in a single embedding mold to form a block of tissue. A block of tissue can encompass a plurality of the same or different tissues, which are oriented in a plurality of planes and/or represent a plurality of age classifications of the organism (s) being analyzed. In accordance with the instant method, a block of tissue contains at least two age classifications (i.e., age of development) or planes of orientation. In one embodiment, a block of tissue contains at least two, three, four or more age classifications of a tissue, each in the same plane of orientation. In another embodiment, a block of tissue contains at least two or three tissues in different planes of orientation and of the same age. For example, three pieces of a mammalian tissue can be embedded in a block of tissue such that when sectioned a traverse

(X-Y) , coronal (X-Z) and sagittal (Y-Z) section of the tissue is obtained from a single block. Likewise, three pieces of a plant tissue can be embedded in the block of tissue such that when sectioned a cross, radial and tangential section of the tissue is obtained from a single block. In a further embodiment, a block of tissue contains a combination of age classifications and planes of orientation. For example, tissue representing two age classifications and two planes of orientation can be prepared as exemplified herein using an Ell .5 embryo in the sagittal plane in combination with brain from E15.5 embryo placed for sectioning in the coronal and sagittal planes. In some embodiments, a block of tissue contains at least two age classifications and at least two planes of orientation of a tissue. In other embodiments, a block of tissue contains at least three, four or more age classifications of a tissue and at least two planes of orientation of a tissue. In still other embodiments, a block of tissue contains at least three, four or more age classifications of a tissue and at least three planes of orientation of a tissue. In yet further embodiments, a block of tissue contains at least two age classifications and at least three planes of a tissue. A block of tissue can also contain one or more tissues, defined as having different structures or functions. For example, a block of tissue can contain organs such as the heart, lungs, and kidney from at least two of the developmental ages of the organism being analyzed. Further, it is contemplated that a block of tissue can provide for a side-by-side comparison of a tissue from two different organisms, e.g., in two different planes of orientation. Any suitable embedding compound can be used to embed the fixed tissues placed in the single embedding mold. In general, tissues are embedded in a resin, plastic, wax or blends thereof. Examples of embedding compounds include, but are not limited to, paraffin, epoxy resin, butyl/methyl methacrylate, Steedman's wax, glycol methacrylate or tissue freezing medium as exemplified herein.

Sectioning of the block of tissue, e.g., cryosectioning with an ultramicrotome or microtome, generates one or more 0.1 - 20 μm sections of tissue which are referred to herein as the tissue sample. The tissue sample thus prepared can be placed on a surface or slide for subsequent microscopic analysis. In one embodiment, at least one, two or three tissue samples from a block of tissue are placed on a surface or slide. A slide can contain one or more tissue samples from a single block of tissue or contain two or more tissue samples, each from a different block of tissue. In this manner, a slide can contain tissue samples which represent a plurality of tissues having different structures or functions, a plurality of age classifications of tissues, and a plurality of tissue planes. A slide can further contain tissue samples from a plurality of organisms. As such, a maximum amount of information can be obtained from tissues using a minimum number of slides with the added advantage of minimizing sample block-to-sample block sectioning artifacts .

Mounting of tissue samples to a surface or slide and preparation of the same for subsequent microscopic analysis (e.g., histological staining, immunohistochemical analysis, in situ gene expression profiling, etc.) can be carried out according to standard methods. For example, tissue samples can be placed on a microscope slide, treated with agents such as proteinase K and acetic anhydride to reduce background, probed with a radiolabeled riboprobe, and analyzed for temporal and spatial expression of a gene of interest . Histological staining of a tissue sample composed of a plurality of tissues having different structures or functions, a plurality of ages of tissues, and a plurality of tissue planes can be carried out using any selected dye solution to reveal all or some of the structures in the tissue sample. For example, Nissl stain (e.g., cresyl violet or cresyl echt violet, thionine, toluidin blue O, or methylene blue) can be used to stain aggregates of rough endoplasmic reticulum and ribosomes in neuronal cell bodies and dendrites, whereas Hematoxylin & Eosin can be employed for general staining of tissue sections; Mallory's Phosphotungstic Acid Hematoxylin (PTAH) for staining nuclei, centrioles, spindles, mitochondria, fibrin, fibrils of neuroglia; and Mason's Trichrome for staining nuclei, collagen, and cytoplasm. Immunohistochemical analysis can be carried out with, e.g., fluorescently labeled antibodies which bind specific epitopes.

In certain embodiments, a set of reference samples are also obtained from each block of tissue. The reference samples can, similar to the tissue samples to be hybridized with a radiolabeled ribonucleic acid probe, represent a plurality of tissues having different structures or functions, a plurality of ages of tissues, and a plurality of tissue planes. Reference samples are generally stained with a selected dye solution as described herein to reveal all or some of the structures in the tissue sample.

In a particular embodiment, a tissue sample is analyzed by in situ hybridization. In this embodiment, the tissue sample is contacted with at least one radiolabeled ribonucleic acid probe under suitable in situ hybridization conditions so that the radiolabeled ribonucleic acid probe binds to a ribonucleic acid target sequence present in the tissue sample. A radiolabeled ribonucleic acid probe used in accordance with the instant method is complementary (i.e., antisense) to a single ribonucleic acid target sequence in the tissue sample being analyzed. Radiolabeled antisense ribonucleic acid probes can be generated using any established in vitro transcription reaction for generating a radiolabeled antisense ribonucleic acid probe from a DNA template. In general, such reactions involve mixing a select DNA template encoding all or a part of a ribonucleic acid target sequence, a transcription buffer, ribonucleotides (GTP, CTP, and ATP) , a selected RNA polymerase and a radiolabeled ribonucleotide (e.g., UTP) and incubating the mixture for a sufficient amount of time to generate a radiolabeled antisense ribonucleic acid probe (also referred to herein as a radiolabeled antisense riboprobe, radiolabeled riboprobe, antisense riboprobe, or simply riboprobe) .

The ribonucleic acid target sequence and its associated DNA template can be selected randomly or as exemplified herein systematically selected based on specified criteria {e.g., function, association with a particular disease or disorder, etc.) . Generally, the select DNA template is a plasmid or other DNA sequence

(e.g., RT-PCR product, cosmid, phasmid, etc.), which encodes all or a part of a ribonucleic acid target sequence

(e.g., a cDNA) and contains the necessary nucleotide sequences for recognition and binding of a selected RNA polymerase (e.g. T7, S6, or T3 RNA polymerase) . Nucleotide sequences for binding such selected RNA 'polymerases are well-known in the art.

Radiolabeling of the antisense riboprobe can be carried out using any suitable radioisotope including, but not limited to ³²P (half-life = 14.3 days, maximum beta- energy = 2270 keV) , ³³P (half-life = 24.4 days, maximum beta-energy = 250 keV) , or ³⁵S (half-life = 87.4 days, maximum beta-energy = 167 keV) , for example. In certain embodiments, ³⁵S or ³³P are employed for their lower emission energy. In particular embodiments, ³³P is employed to generate high specific-activity ³³P-labeled antisense riboprobes with high sensitivity and relatively low background levels compatible with in situ hybridization analysis. Accordingly, detection of mRNA transcripts that are below the limits of detection by non-isotopic methods can be achieved. In some embodiments, at least one tissue sample is contacted with one radiolabeled antisense riboprobe. In other embodiments, a plurality of tissue samples (e.g., at least two, three, four, or more) is contacted with one radiolabeled antisense riboprobe. In particular embodiments, a plurality of tissue samples (e.g., at least two, three, four, or more) is contacted with each of a plurality of antisense riboprobes (e.g., at least two, three, four, or more riboprobes) . In this embodiment, each antisense riboprobe of the plurality of antisense riboprobes is contacted with a plurality of tissue samples placed in a single hybridization reaction vessel. Accordingly, a plurality of hybridization reaction vessels are used for hybridizing the corresponding number of plurality of. antisense riboprobes. A plurality of antisense riboprobes can be prepared by in vitro transcription reactions carried out in a multi-well format {e.g., 24- well, 96-well, or 384-well plates). In situ hybridization refers to hybridization of a riboprobe to a ribonucleic acid target sequence that exists within a cytological or histological preparation or tissue sample. Under suitable in situ hybridization conditions, the riboprobe binds to the target sequence and a hybrid is produced. Suitable in situ hybridization conditions generally refers to the combination of conditions that are employable in a given stringent hybridization procedure to produce specific hybrids between the riboprobe and the target sequence. Such conditions typically involve controlled temperature, liquid phase, and contact between a riboprobe and a ribonucleic acid target sequence of a tissue sample. Conveniently and preferably, the riboprobe is denatured prior to being contacted with tissue sample. In general, in situ hybridization conditions can encompass the use of a -50:50 volume ratio mixture of a hybridization solution (e.g., as exemplified herein) and formamide, an illustrative temperature in the range of about 40 to about 65⁰C applied for a time that is generally in the range of about 1 to about 18 hours. Exemplary hybridization conditions are disclosed herein; however, other hybridizing conditions can also be employed.

Following hybridization, unbound riboprobe and riboprobe of non-specific hybrids (i.e., between the riboprobe and non-target sequence) can be removed by treating the slides with an RNase and washing the slides in, for example, consecutive high concentration and low concentration salt solutions (e.g., 2X SSC and 0.2X SSC) at an illustrative temperature in the range of about 40 to about 65⁰C.

To facilitate the subsequent step of exposing the tissue sample to silver emulsion, the intensity or strength of the signal (i.e., emission) produced by the bound radiolabeled riboprobe is determined. Probe signal intensity can advantageously be determined by exposing a sheet of X-ray film to a tissue sample for an appropriate amount of time (e.g., overnight) and measuring the relative or absolute intensity of the signal produced by the bound radiolabeled riboprobe. For example, a relative intensity of riboprobe signal can be obtained by visually inspecting the X-ray film and semi-quantitatively assigning dark, medium-to-light, and light-to-undetectable exposures as strong, medium, and low intensity riboprobe signals, respectively. A control or standard sheet of exposed X-ray film can also be used for comparing and assigning relative intensities. Alternatively, an absolute intensity of riboprobe signal can be directly quantitated by phosphorimage analysis, wherein predetermined quantities of signal represent strong, medium, and low intensity riboprobe signals. As yet another alternative, a radioactivity detector such as a Geiger counter can be placed over the tissue sample to assess the relative intensity of signal produced by the bound radiolabeled riboprobe .

Sensitivity of the instant method is provided by exposing the tissue sample to liquid silver emulsion, i.e., silver salt crystals or silver halides. When the silver emulsion is exposed to the beta emissions produced by the radioisotope of the bound riboprobe, the silver crystals break down into black silver. Accordingly, areas of tissue where the target sequence is highly expressed will appear darker than areas with low or no target sequence expression. Optimal exposure time to the silver emulsion is selected based upon the intensity of the signal produced by the bound radiolabeled riboprobe, as determined above. The period of time of silver emulsion exposure can be in the range of, e.g., 1 to 10 days with longer exposure times selected for low intensity riboprobe signals and shorter exposure times selected for high intensity riboprobe signals. By way of illustration, tissue samples with a high intensity signal probe are exposed to silver emulsion for 1 day, whereas tissue samples with medium-to-low or low-to- undetectable probe signal are exposed to silver for 3 or 7 days, respectively.

Visualization of the temporal and/or spatial expression of the ribonucleic acid target sequence in the tissue sample can be carried out using any standard developing and fixing process. Any suitable developing process that enhances the image produced on the tissue sample by breaking down more crystals into silver can be used. Likewise, any chemical fixative, which converts the remaining silver halides into colorless salts so that the tissue sample is no longer sensitive to light, can be used. Exemplary developer and fixative reagents are commercially available from KODAK^®. The tissue samples can be directly viewed under a microscope (e.g., a dark-field microscopes) to determine the temporal and/or spatial expression of the ribonucleic acid target sequence in the tissue sample. Advantageously, the use of silver emulsion in combination with dark-field microscopy visualization provides clear detection of gradients of gene expression in tissue samples. In particular embodiments, an image of the tissue samples can be captured using any traditional photographic system or digital imaging system so that the temporal and/or spatial expression data can be analyzed, stored, and distributed.

Given that a plurality of images can be generated which represent multiple ages, planes, tissue, and organisms, another embodiment of the instant invention provides for an image-centric database for storing and distributing or sharing of image datasets (see, e.g., Bug and Nissanov (2003) Neuroinformatics 1 (4) : 359-77) . In addition to the images which reveal expression patterns of target sequences, the image-centric database can further contain other biological information about each target sequence {e.g., function, nucleotide sequence of the riboprobe template, GENBANK accession number, chromosomal location of the gene, aliases, official and alternate gene symbols, etc. ) . The images and information in the image-centric database can be distributed or shared using an user- friendly, internet-based website. Darkfield images for each gene can be displayed as a set of thumbnails to provide a snapshot of temporal and spatial gene expression patterns across development. Each thumbnail image can then be linked to an intermediate-sized image for convenient comparison with a nearby reference tissue sample, and the original

(e.g., 1392 x 1040 pixels²) image can be downloaded. The website can provide a side-by-side "gene expression viewer" to facilitate direct, side-by-side comparisons between expression patterns of genes in the database. An unlimited number of darkfield images can be displayed together to allow analysis of multiple images simultaneously. The "gene expression viewer" is useful for examining gene family members, genes whose products participate in ligand- receptor interactions, and genes involved in signal transduction pathways within the developing brain. Links to other database websites can provide additional information for each target sequence. For example, information from the National Center for Biotechnology and Informatics (NCBI) databases (e.g., PubMed, UniGene, and Gene) and the Gene Ontology Consortium database can be obtained. Likewise, links to the Online Mendelian Inheritance in Man (OMIM) web site, the GENSAT databases at Rockefeller University and NCBI, the Mutant Mouse Regional Resource Center (MMRRC) , the Microarray Consortium, and the Mouse Genome Informatics database (MGI) can be incorporated for each target sequence to provide additional information.

It is contemplated that the target sequences in the image-centric database can be browsed in alphabetical order or the database can be systematically searched using a variety of gene identifiers such as official gene symbol, gene name, gene alias, genetic location, and/or gene ontology. Biological information from search results can be exported to allow systematic searches in other databases or information sets. Image data can be amassed in specific, saved searches in a gene expression viewer cart that can be revisited, manipulated, and distributed to others thereby facilitating information-sharing amongst collaborators regardless of geographic location. The image-centric database of the instant invention will also allow one of skill in the art to combine gene expression information with other expression data obtained by other experimental parameters. For example, gene-chip studies can result in an overabundance of candidate genes that are upregulated or down-regulated in a single experiment. Knowing the temporal and spatial context of gene expression within, e.g., the developing nervous system can provide key information to categorize results of gene- chip studies into meaningful groups and/or pinpoint the most important genes that should undergo further experimentation. Thus, combining information from various types of experiments with the information about gene expression patterns contained in an image-centric database disclosed herein will enable construction of novel hypotheses .

To illustrate the method of the instant invention, gene expression analysis was conducted on a plurality of target sequences (also referred to as target genes) during nervous system development in mice. Whole embryo from day 11.5 (Ell.5); sagittal and coronal tissue sections of embryonic 15.5 (E15.5), postnatal day 7 (P7) , and adult (P42) mouse brains; transverse tissue sections of E15.5, P7, and P42 spinal cord; and sections through P7 and P42 retina were analyzed for spatial and temporal gene expression. These ages represent critical stages of nervous system development, including periods of rapid proliferation, neuronal migration, and synaptogenesis . Blocks of tissue were prepared by embedding E15.5 and P7 brains and spinal cords for sectioning in coronal and sagittal orientations in a single cryomold in TBS^® tissue freezing media. Ell.5 embryos were embedded individually so that sagittal sections could be obtained. Because of size constraints and quality control issues, adult nervous system tissues that were to undergo sagittal sectioning and those that were to undergo coronal sectioning were embedded in separate blocks. A cryostat was used to systematically generate 16-μm-thick sections of multiple blocks of tissue. Tissue samples included 101 coronal sections of Ell.5 and adult blocks of tissue on 51 slides and 151 sagittal sections of E15.5, P7, and adult blocks of tissue on 51 slides. Slide 0 and every 10^th slide from each block of tissue were Nissl-stained as reference tissue samples and for quality control purposes.

The genes targeted for analysis were selected for their ^■ potential role in neurodevelopment, neurodegeneration, signal transduction, and cognitive development. In addition, cDNA clone collections, including the National Institute of Aging (NIA) 15K mouse clone collection, the Incyte 1.0 Unique mouse clones, and the

Brain Molecular Anatomy Project collection (Bonaldo, et al .

(2004) Genome Res. 14:2053-63; Carter, et al . (2003) C R Biol. 326:931-40) were searched to identify clones either derived from nervous system tissue or involved in biological processes such as neurogenesis, cell migration, differentiation, cytoplasmic remodeling, plasticity, cell proliferation, cell death, and metabolism. Prioritization of target genes to be analyzed was aided by the use of datasets from microarray studies of brain tissues; from nervous system mutagenesis studies of flies (see, flybase with the extension .org at the world-wide web), worms (see wormbase with the extension, org at the world-wide web) , and other organisms; and from the characterization of genes associated with human neurological disorders (see the OMIM database of the National Institutes of Health National Library of Medicine) . The DNA templates used in probe preparation by in vitro transcription were either custom- designed and made by RT-PCR or were derived from cDNA clones ^' from various collections. All templates were verified by nucleotide sequence analysis, and the sequences of templates representing each target gene were stored in an image-centric database.

A full analysis of a single target sequence by in situ hybridization yielded information about expression patterns in at least 55 individual tissue samples representing three planes of orientation at four developmental time points. In this illustrative example, more than 15,000 images were generated which represented expression data obtained by using more than 1400 unique riboprobes in in situ hybridization analysis of more than 63,000 individual sections of nervous system tissue.

Images were captured by using a NIKON^® PHOTOMETRICS^® COOLSNAP^® ES camera mounted on a ZEISS^® stereomicroscope and a Darklight 3" slide edge illuminator (Micro Video Instruments, Inc., Avon, MA) . The brightness and contrast of high-resolution (1392 x 1040 pixels²) images were optimized by using ADOBE^® PHOTOSHOP^® 8.0; the images were reduced to intermediate and thumbnail sizes for distribution and analysis.

Approximately 65% of the genes analyzed exhibited temporally restricted patterns of expression, spatially restricted patterns of expression (i.e., discrete populations of positive cells) or both during brain development. The remaining genes displayed either ubiquitous or low to undetectable signals. One feature of the high throughput process disclosed herein coupled with the target gene selection strategy was that equal weight was given to both characterized and uncharacterized genes, including novel, predicted genes revealed by genome sequencing (Burge and Karlin (1998) Curr. Opϊn. Struct. Biol. 8:346-54) and RIKEN-derived genes (Kawai, et al . (2001) Nature 409:685-90). Thus, the exemplified dataset represents the first step in characterization of these genes. The equal-weighting feature provided, e.g., gene expression patterns within the nervous system for many genes, including keratocan {kera) (Liu, et al. (1998) J. Biol. Chem. 273:22584-8), dermatan sulfate proteoglycan 3 {dspg3) (Johnson, et al . (1999) Dev. Dyn. 216:499-510), heat shock 7OkDa protein 12A (hspal2a) (Han, et al . (2003) Proc. Natl. Acad. Sci . USA 100:1256- 61), and angipoietin 2 (agpt2) (Nico, et al . (2004) Brain Res. 1013:256-9), whose expression had only been previously characterized in non-neuronal tissues. The exemplified dataset also included expression patterns for novel and predicted genes with no previously available expression information. In addition, the exemplified dataset contained expression data for numerous well-characterized genes such as solute carrier 17, member a6 (Slcl7a6) (Hisano, et al . (2000) Brain Res. MoI. Brain Res. 83:34-43; Sakata-Haga, et al . (2001) Brain Res. 902:143-55), neuro-oncological ventral antigen 1 (noval) (Buckanovich, et al . (1996) J. Neurosci. 16:1114-22; Jensen, et al . (2000) Neuron 25:359- 71), DOH4S114 (Studler, et al . (1993) Eur. J. Neurosci. 5:614-23; Wakana, et al . (2000) Brain Res. 857:286-90), and SLIT-ROBO Rho GTPase activating protein (srgapl) (Wong, et al. (2001) Cell 107:209-21). However, in most cases, the published expression data for these genes is available for only a single age or tissue, and/or expression was analyzed by molecular biology techniques. The systematic analysis of gene expression in accordance with the instant method facilitates comparison of expression patterns at multiple ages and in tissues. Advantageously, the in situ hybridization method of the instant invention can be adapted for high-throughput gene analysis as it is based on cost-effective routine laboratory practices that do not require robotics. This method can be readily adapted for high-throughput analysis of gene expression in any tissue or model organism. The system can also be expanded to accommodate complementary technologies such as immunohistochemistry (e.g., using radiolabeled or fluorescently-tagged antibodies) , mass spectrometry, or other imaging modalities.

The invention is described in greater detail by the following non-limiting example.

Example 1: Embryonic Tissue Collection and Preparation Female C57Black6 mice were examined for the presence of vaginal mucus plugs. Female mice with plugs were removed to a holding cage and embryos were labeled as embryonic day 0.5. At the desired embryonic ages, pregnant dams were euthanized by cervical dislocation. Each embryo was carefully removed from the uterus at the placenta. Under microscope, each yolk sac was opened using sharp watchmaker forceps. The placenta and yolk sac were discarded.

Ell.5 embryos were transferred to a 50 mL conical tube containing ice cold 4% paraformaldehyde (PFA) in 0.1 M phosphate-buffered saline (PBS) using a blunt spatula.

Heads of E15.5 embryos were removed using sharp scissors. To facilitate fixation of the brain, the snout was carefully removed using scissors, without puncturing or nicking the brain. The trunks of E15.5 embryos were processed by removing the arms, legs and tail. The gastrointestinal track, heart, lungs, liver and kidneys were removed using blunt forceps and discarded.^' The head and trunk were subsequently placed into a 50 mL conical tube containing ice cold 4% PFA in 0.1 M PBS.

Embryonic tissues were fixed overnight on an orbital shaker in a cold room. Subsequently, the tissues were washed three times, 5 minutes per wash, with 0.1 M PBS.

E15.5 embryos were further dissected by removing the brains from the heads and carefully collecting the retina, brain, and cochlea/vestibular nucleus. Brains were prepared by cutting one brain between the forebrain and the hindbrain at the level of the tectum and one brain down the midline in the sagittal plane so that each block would contain one forebrain and one hindbrain cut in the coronal plane, one of the hemispheres in the sagittal plane, one trunk, one retina and one cochlea/vestibular nucleus. The dissected tissue set (i.e., of the E15.5 embryo) or whole embryo (i.e., the Ell.5 embryo) was placed in a 10 mL scintillation vial with 8 mL of 30% sucrose in 0.1 M PBS for cryoprotection. Tissues were incubated in the sucrose solution on an orbital shaker in a cold room until the tissues sunk to the bottom of the tube (e.g., at least 24 hours) .

One Ell.5 embryo was placed in a cryomold in the sagittal plane with tail curling towards the surface. Similarly, one E15.5 dissected tissue set was placed in one cryomold mold {e.g., hindbrain in upper left quadrant, forebrain in upper right quadrant, sagittal hemisphere in lower left quadrant and retina in the lower right quadrant) . Cryomolds were labeled with the tissue sample description, i.e., left or right sagittal hemisphere and age of mouse. Tissues were embedded in TBS^® tissue freezing medium (Triangle Biomedical Sciences, Durham, NC) and frozen on a block of flat dry ice. Frozen blocks were subsequently transferred to ZIPLOC^® bags and stored at - 8O⁰C until analyzed.

Example 2 : Neonatal Tissue Collection and Preparation Newborn 57Black6 mice (postnatal day 0.5) were allowed to develop to postnatal day 7 (P7) . On day 7, fresh 4% PFA in 0.1 M PBS was prepared by placing 3 raL of 0.1 M PBS in perfusion pump tubing followed by 5-7 mL of fixative. A needle (27 gauge) was inserted onto one end of tubing. P7 pups were anesthetized with 0.2 mL of AVERTIN™ delivered by IP injection and subsequently delivered fixative, via the perfusion pump, by insertion of the 27 gauge needle into the bottom of the left ventricle. Blood was flushed out with 2 mL of PBS followed by 5-7 mL of 4% PFA in 0.1 M PBS. Tissues were examined for stiffening to ensure good fixation.

Perfused mice were dissected with large scissors. The head was removed without disturbing the hindbrain and the skin and snout were subsequently removed to facilitate fixation of the brain. Arms, legs and tail were removed from trunk as were the heart, lungs, liver and gastrointestinal track to expose the back of the mouse.

Heads and trunks of dissected mice were placed into 50 mL conical tubes filled with 4% PFA in 0.1 M PBS and fixed overnight on an orbital shaker in a cold room. Tissues were rinsed in 0.1 M PBS and subsequently the brain, retina and cochlea/vestibular nucleus were removed from the head and the spinal cord removed from the trunk.

Brains were prepared by cutting one brain between the forebrain and the hindbrain between the superior and inferior colliculus and one brain down the midline in the sagittal plane. The spinal cord was dissected into the cranial nerves, cervical, thoracic and lumbar regions. Accordingly, each block contained one forebrain and one hindbrain cut in the coronal plane, one of the hemispheres in the sagittal plane, one retina, one cochlea/vestibular nucleus, and 4 pieces of spinal cord in the transverse plane.

Dissected tissue sets were placed in 10 mL scintillation vials with 8 mL of 30% sucrose in 0.1 M PBS for cryoprotection. Tissues were incubated in the sucrose solution on an orbital shaker in a cold room until the tissues sunk to the bottom of the tube {e.g., at least 24 hours) .

One P7 dissected tissue set was placed in one cryomold mold, which was labeled with the tissue sample description, i.e., left or right sagittal hemisphere and age of mouse. All cut surfaces were embedded toward the bottom of the mold. Tissues were embedded in TBS tissue freezing medium

(Triangle Biomedical Sciences, Durham, NC) and frozen on a block of flat dry ice. Frozen blocks were subsequently transferred to ZIPLOC^® bags and stored at -80⁰C until analyzed.

Adult mice, i.e., 42 days of age (P42) were prepared in a similar fashion.

Example 3: Cryosectioning of Tissues Three ages (i.e., Ell.5 and E15.5 embryos and P7 pups) were placed on a single series of microscope slides (Fisher SUPERFROST™ plus slides) , labeled with a block number and slide number {e.g., 0 through 150), and placed on a slide warmer at 37°C. One block of frozen tissue from E15.5 and Ell.5 embryos and P7 pups was removed from the freezer, transferred to a cryostat chamber, and allowed to equilibrate for at least 15 minutes. Subsequently, one block was mounted onto a stage. Sections (16 μm) were cut and one section per age was placed on a slide with the sections from the first block cut close to the frosted end of the slide. The remaining blocks of tissue were sectioned and mounted juxtaposed to the sections cut in the previous block. Three different blocks were mounted onto one set of slides, each section representing a different age with sagittal hemispheres (i.e., left or right) matching for the E15.5 and P7 tissue. The slides were stored overnight in microscope boxes at room temperature . A similar set of slides were prepared for P42 adult tissue sections.

The 0 slide and every 10^th slide were removed for cresyl violet staining. The remaining slides were placed into two microscope slide boxes and stored at -80⁰C until analyzed.

Example 4: Preparation of Slides for in situ Hybridization

To analyze 45 genes via in situ hybridization, the slides were removed from -80⁰C storage and allowed to equilibrate at room temperature and dry. Selected slides were removed from the microscope boxes and placed vertically on a microscope rack. Using a SHUR/MARK^® pen, 3 slides containing embryo/neonatal tissue and 3 slides containing adult tissue were labeled for each gene with a clone number and a run date. One set of slides containing embryo/neonatal or adult tissue was composed of 3 microscope slides spaced at 800 μm intervals. To illustrate, Clone G00002 was hybridized to block G2 , slides 2, 52 and 102; clone G00003 was hybridized to block G2 , slides 3, 53 and 103; clone G00004 was hybridized to block G2, slides 4, 54, 104; etc.

For each gene, at least 54 individual tissue sections were analyzed, i.e., two sagittal sections of an Ell.5 whole embryo, three sagittal and six coronal sections of E15.5 and P7 brains, three sagittal and eight coronal sections of P42 brains, three transverse sections of an E15.5 spinal cord, at least 9 transverse sections of P7 and P42 spinal cord, and a minimum of one section through P7 and P42 retina.

To track the analysis of each gene (i.e., the slide numbers analyzed, riboprobe preparation, gene expression analysis and image collection) , the block number, slide number, clone number and run date were entered into a

FILEMAKER^® Pro database for each gene.

Example 5 : Riboprobe Preparation

Riboprobes were prepared in a 96 -well format. A 96- well plate layout chart was generated for each experiment and color-coded for different promoters used in combination with the template (e.g., red=SP6, green=T3 , blue=T7) . A 4 μL aliquot of each template was placed into a well of the 96-well plate and the plate was stored overnight at 4°C. As an alternative to the 96-well format, individual tubes were used. In such experiments, 3 set of 1.7 mL microcentrifuge tubes were labeled with the clone number for each probe. A 4 μL aliquot of each template was placed into the first set of tubes. For the in vitro transcription reaction, a master mix was prepared and stored on ice until used. The master mix was composed of 2.6 μL H₂O; 1.4 μL 1OX Transcription buffer

(Ambion, Inc., Austin, TX); 0.5 μL rNTPs (G, A, and C; 2 mM each) (PROMEGA™, Madison, WI); and 0.5 μL T7, T3 , or SP6 RNA POLYMERASE PLUS™ (20 U/μL; Ambion, Inc., Austin, TX) per reaction. If POLYMERASE PLUS™ enzymes were not used, 1 μL of RNase inhibitor was added to the reaction and 1 μL of H₂O was subtracted for each reaction and H₂O treated with diethyl pyrocarbonate (DEPC) was used.

A 5 (jL aliquot of master mix was place into each tube or well containing template. Subsequently, a 5 μL aliquot of P³³-UTP (250 μCi/25 μL; PERKINELMER^® , Boston, MA) was added to each tube or well under a radiation hood. When using a 96-well plate format, the reactions were incubated in a ROBOCYCLER^® at 37⁰C for 1 hour. When using individual tubes, the reactions were incubated under a radiation hood in a 37°C water bath for 1 hour.

During the 1 hour incubation, an RQl DNase I master mix was prepared and stored on ice. This master mix was composed of 30.5 μL H₂O; 3.5 μL RQl DNase 1OX Buffer

(PROMEGA™, Madison, WI); and 1.0 μL RQl RNase-free DNase (PROMEGA™, Madison, WI) . After the 1 hour incubation at 37⁰C, a 35 μL aliquot of RQl DNase I master mix was placed into each tube or well. When using a 96-well plate format, the reactions were incubated in a ROBOCYCLER^® at 37°C for 30 minutes. When using individual tubes, the reactions were incubated under a radiation hood in a 37⁰C water bath for 30 minutes .

During the 30 minute incubation, scintillation racks were filled with the appropriate number of scintillation vials labeled with clone numbers (e.g., clone G00002, clone G00003, clone G00004, etc.) and a 5 mL aliquot of scintillation fluid (LSC-cocktail SCINTISAFE^® 30%; Fisher Scientific, Fairlawn, NJ) was placed into each scintillation vial .

To purify riboprobes in the 96-well plate format, the seal was removed from a PERFORMA^® DTR 96-well plate purification kit (Edge Biosystems, Gaithersburg, MD) , the lid was placed on the top and the flat-bottom 96-well collection plate placed under the columns. The assembly was centrifuged for 3 minutes at 850 x g. The flat-bottom 96- well collection plate containing eluate was discarded and a V-bottom plate was placed under the packed columns. Using a multi-channel pipettor, probe reaction mixtures were slowly- transferred onto the center of each packed column and the columns were centrifuged for 3 minutes at 850 x g to purify each -45 μL aliquot of riboprobe .

To purify riboprobes in the microcentrifuge format, a PROBEQUANT™ G-50 Micro Column (Amersham Biosciences, Piscataway, NJ) for each reaction was briefly vortexed and centrifuged at 3000 rpm for 2 minutes to remove storage buffer. The columns were subsequently placed into the second set of 1.7 mL microcentrifuge tubes labeled with the clone number as described above. Each riboprobe reaction mixture was pipetted onto the center of an individual packed column and the column was centrifuged at 3000 rpm for 2 minutes to purify each ~45 μL aliquot of riboprobe.

To determine riboprobe radiolabeling, 1 μL of purified riboprobe was placed into each labeled scintillation vial and counted using a scintillation counter according to standard methods. The remaining purified riboprobe was stored overnight in a radioactivity-safe box at -20°C.

Probes having ≥ 100,000 cpm scintillation counts were used in hybridizations. At least four slides were analyzed per probe . The appropriate amount of probe to use in the hybridizations was determined using the following formula:

(1,000,000 cpm X number of slides) / (probe cpm count) = μL of probe. To ensure that enough counts were used, the calculation was performed based on one additional slide instead of the actual number of slides that were to be hybridized. For example, if a probe were used in a hybridization with 5 slides, the calculation would be: (1,000,000 cpm X 6 slides) / (probe cpm count) = μL of probe .

Example 6: Riboprobe Hybridization All steps were carried out using WHATMAN™ glass staining dishes and metal slide racks (50 slides/rack) . Each solution used in slide preparation had no more than two racks of slides passed through it. All pre- hybridization solutions were made in carboys except for the triethanolamine (TEA) buffer, which was prepared as needed.

Slides were treated for 10 minutes with 0.00025% proteinase K solution. For 600 mL of proteinase K solution, 150 μL of proteinase K (10 mg/mL; Fisher Scientific, Fairlawn, NJ) was freshly added to a solution containing 60 mL of 1 M Tris-HCl buffer, pH=8.0 (Fisher Scientific, Fairlawn, NJ); 60 mL of 0.5 M EDTA, pH=8.0 (Fisher Scientific, Fairlawn, NJ) ; and 480 mL of DEPC-treated H₂O.

Slides were subsequently rinsed in 600 mL of DEPC- treated H₂O for 3 minutes and in 600 mL of 0.1 M TEA, pH=8.0, for 3 minutes. For 600 mL of 0.1 M TEA, 8 mL of 7.5 M triethanolamine (Sigma, St. Louis, MO) was added to 592 mL of DEPC-treated H₂O with the pH adjusted to 8.0 with glacial acetic acid using a pH meter.

Acetylate of slides was carried out for 10 minutes. For 600 mL acetic anhydride/0.1 M TEA, 1.5 mL of acetic anhydride (Sigma, St. Louis, MO) was mixed with TEA buffer.

Slides were rinsed in 600 mL 2X SSC (60 mL 2OX SSC (Fisher Scientific, Fairlawn, NJ) , 540 mL DEPC-treated H₂O) and dehydrated in a graded series of ethanol (i.e., 50% EtOH, 70% EtOH, 95% EtOH, and 100% EtOH) for 3 minutes each. Slides were allowed to air dry before being incubated with the riboprobe. A probe reaction cocktail was prepared by mixing an equal volume of 2X hybridization buffer and Formamide, and subsequently adding 20% SDS such that the final concentration of Formamide was 50%, the hybridization solution was IX, and SDS was 0.1%. The master cocktail mix was prepared based upon the total number of slides to be hybridized. Per slide, the probe reaction cocktail was composed of 50 μL 2X hybridization buffer; 50 μL Formamide

(Fisher Scientific, Fairlawn, NJ); and 0.5 μL 20% SDS. A 20% SDS stock solution was prepared by mixing 20 grams of sodium lauryl sulfate (Fisher Scientific, Fairlawn, NJ) or sodium dodecyl sulfate (SDS) with 100 mL of DEPC-treated H₂O. A 50 mL stock solution of 2X hybridization buffer was prepared by adding to 14.21 mL of DEPC-treated H₂O, 12 mL 5 M NaCl (Ambion, Inc., Austin, TX); 1 mL 1 M Tris-HCl buffer, pH=8.0 (Fisher Scientific, Fairlawn, NJ); 330 μL 6% FICOLL^® (Amersham Pharmacia Biotech, Piscataway, NJ) ; 330 μL 6% polyvinylpyrrolidone (Sigma, St. Louis, MO); 330 μL 6% bovine serum albumin (Sigma, St. Louis, MO); 200 μL 0.5 M EDTA, pH=8.0 (Fisher Scientific, Fairlawn, NJ); 1 mL 10 mg/mL salmon sperm DNA (Ambion, Inc., Austin, TX); 500 μL 10 mg/mL total yeast RNA (Ambion, Inc., Austin, TX); 100 μL 50 mg/mL yeast tRNA (Sigma, St. Louis, MO); 20 mL 50% dextran sulfate (ISC Bioexpress, Kaysville, UT) . Unused 2X hybridization buffer was stored at -2O⁰C.

In preparation for adding the radiolabeled riboprobe, an appropriate volume of hybridization cocktail was transferred into tubes labeled with each clone number and transferred to a 6O⁰C THERMOMIXER™ . The probes was boiled for 10 minutes at 100⁰C using either a ROBOCYCLER^® or THERMOMIXER™ and subsequently stored on ice until used. When using a 96-well plate format, the riboprobes were first transferred to a standard 96-well plate (e.g., GeneMate^®) with a multichannel pipettor. To each tube of hybridization cocktail was added the appropriate amount of probe to obtain 1 x 10^s cpm/slide. The riboprobe in hybridization cocktail was shaken in a 60°C THERMOMIXER™ until it was applied to the slides.

To each slide was added 100 μL riboprobe cocktail. A coverslip or another slide was placed on the slide containing tissue sections and incubated overnight in a humid chamber at 6O⁰C.

Example 7: Post-Hybridization

All post-hybridization steps were carried out using

600 mL WHATMAN™ glass staining dishes and metal slide racks

(50 slides/rack) . Each solution used in post-hybridization treatment had no more than two racks of slides passed through it. For solutions at 37°C or 60°C, dry incubators were employed; however waterbaths could also have been used to keep solutions at appropriate temperatures.

While slides were still warm, coverslips (or slides) were gently removed and slides containing tissue sections were placed directly into racks submerged in room temperature 2X SSC/50% Formamide. Exposure to air was minimized. Slides were subsequently rinsed in 600 mL of 2X SSC for 15 minutes, transferred to 600 mL of prewarmed RNase A solution (60 mL 5 M NaCl; 6 mL 1 M Tris-HCl, pH=8.0; 1.8 mL 0.5 M EDTA, pH=8.0; 0.48 mL of 25 mg/mL Rnase A (Sigma, St. Louis, MO); and 531.72 mL H₂O) and incubated for 30 minutes at 37⁰C. After RNase A treatment, the slides were rinsed in 600 mL of prewarmed post- hybridization buffer for 30 minutes at 60⁰C. A 600 mL solution of post-hybridization buffer was prepared by mixing 60 mL 5 M NaCl; 6 mL 1 M Tris-HCl, pH=8.0; 1.8 mL 0.5 M EDTA, pH=8.0; 0.48 mL of 25 mg/mL RNase A (Sigma, St. Louis, MO); and 532.2 mL H₂O. A 1 hour rinse in 2X SSC buffer was conducted at 6O⁰C followed by a 1 hour incubation in 0.2X SSC at 6O⁰C. The slides were subsequently dehydrated in a graded series of ethanol (1 wash in 50% EtOH, 1 wash in 70% EtOH, 1 wash in 95% EtOH and 3 washes in 100% EtOH) for 5 minutes each. The ethanol solutions were changed after each slide rack was passed through it. In preparation for autoradiography, slides were allowed to air dry for about 15 minutes.

Example 8 : Autoradiography

BIOMAX™ MR film (KODAK^®, Rochester, NY) labeled with the appropriate clone numbers was exposed overnight to each set of slides per clone. For example, one piece of film could be exposed to 3 columns of 5 slides, wherein each column represented a different clone number. Time of exposure was recorded so that the total exposure time could be calculated. Exposed film was developed according to standard methods and exposure time was recorded. Autoradiographs were stored individually in plastic sleeves. The autoradiographs were used to measure the strength of probe signal. If the signal for a probe was very dark, the corresponding tissue slides were exposed to emulsion for 1 day. If the signal was medium or light, the corresponding tissue slides were exposed to emulsion for 3 days. If the signal was light or undetectable, the corresponding tissue slides were exposed to emulsion for 7 days.

Example 9: Slide Emulsification

Slides were sorted by the number of days of exposure to emulsion (i.e., 1, 3, or 7 days) and placed in appropriately labeled boxes containing desiccant pills. All steps employing emulsion were carried out in the dark. One bottle of KODAK^® NTB2 emulsion (KODAK^®, Rochester, NY) was heated for approximately 45 minutes, or until completely liquefied. A 3:1 dilution of a full bottle of emulsion (118 mL) was prepared by thoroughly mixing the liquefied emulsion with 39.3 mL of 42°C double-distilled H₂O without generating bubbles. The 3:1 emulsion mixture was poured into a dip miser (Electron Microscopy Sciences, Hatfield, PA) submerged in a 42⁰C waterbath to keep the emulsion warm and liquid. Slides were dipped in the emulsion up to the frosty label without generating bubbles. The back of each slide was wiped with dampened sterile gauze to remove emulsion. Each slide was placed upright in a drying rack lined with paper towels to wick away excess emulsion. Slides were allowed to dry at least 2 hours in the dark. Subsequently, the slides were placed in a slide box with desiccant (AGM Container Control, Tucson, AZ) . Each slide box was wrapped with 2 pieces of aluminum foil, labeled with the number of days of exposure to emulsion, and stored for the appropriate amount of time (1 day, 3 days, 7 days, etc.) at 4°C.

Example 10: Slide Development Developer and fixer were prepared prior to use as follows. An entire bag of KODAK^® D-19 Developer (Sigma, St. Louis, MO) was slowly added to 3.8 L H₂O (1 gallon) and thoroughly mixed with a large stir bar on a stir plate while warming on the "low" setting. The developer was filtered into 500 mL amber glass bottles and stored at room temperature until used. An entire bag of KODAK^® Fixer

(Sigma, St. Louis, MO) was slowly added to 3.8 L H₂O (1 gallon) and thoroughly mixed with a large stir bar on a stir plate. The fixer was filtered into 500 mL amber glass bottles and stored at room temperature until used.

On the first, third, and seventh day after emulsification, the appropriate slide boxes were removed from the cold room and equilibrated to room temperature (~2 hours) . The boxes remained closed and wrapped in foil until the slides were developed.

In the dark, slides were removed from the foil-wrapped boxes, placed in slide racks and incubated for 4 minutes in KODAK^® developer. Developer was prepared by cooling and diluting 1:1 with double-distilled water. During the developing and fixing process, the developer and fixer were held between 14-17⁰C. No more than two slide racks, each containing 20 slides, were passed through a 500 mL volume of 1:1 diluted developer.

After the slides had developed, the slide racks were lifted through the developer 1-2 times to remove air bubbles and washed in water for 2 minutes, again lifting 1- 2 times. Slides were placed in KODAK^® fixer for 5 minutes, followed by gentle lifting 1-2 times to remove air bubbles, and rinsed in running water for at least 2 hours or overnight .

Slides were subsequently dehydrated in a graded series of ethanol (1 wash in 50% EtOH, 1 wash in 70% EtOH, 1 wash in 95% EtOH and 3 washes in 100% EtOH) for 5 minutes each and then washed 3 times in CITRISOLV (Fisher Scientific, Pittsburgh, PA) for 5 minutes each wash. Slides remained in CITRISOLV until coverslips were applied. Coverslips were applied to each slide with PERMOUNT^® (Fisher Scientific, Pittsburgh, PA) and placed flat in drying racks in a hood.

Slides were inspected for bubbles, fingerprints or smears and a new coverslip was applied when appropriate. PERMOUNT^® was allowed to dry for 4-5 days before image analysis.

Claims

What is claimed is;

1. A method for preparing a tissue sample for microscopic analysis comprising embedding at least two fixed tissues, each of a different age classification or plane of orientation, in a single embedding mold and sectioning the embedded, fixed tissue so that a tissue sample is prepared for microscopic analysis.

2. The method of claim 1 comprising at least three fixed tissues.

3. The method of claim 1 comprising at least four fixed tissues.

4. A tissue sample prepared according to the method of claim 1, claim 2 or claim 3.

5. A method for determining the temporal or spatial expression of a ribonucleic acid target sequence in a tissue sample comprising contacting the tissue sample of claim 4 with a radiolabeled ribonucleic acid probe which is complementary to a single ribonucleic acid target sequence in the tissue sample so that the radiolabeled ribonucleic acid probe binds to the ribonucleic acid target sequence; exposing the tissue sample to silver emulsion for a selected period of time based upon the intensity of the signal produced by the bound radiolabeled ribonucleic acid probe ; and developing and fixing the tissue sample so that the temporal or spatial expression of the ribonucleic acid target sequence in the tissue sample can be determined.

6. The method of claim 5, further comprising imaging and analyzing the temporal or spatial expression of the ribonucleic acid target sequence in the tissue sample.

7. The method of claim 6, wherein the tissue sample image is stored in a database.