In Situ Hybridization

In situ hybridization visualizes mRNA expression in intact tissue sections thus providing anatomical localization. Because it employs DNA or RNA probes complimentary to specific mRNA sequences, in situ hybridization visualizes only the product of the target gene and not related gene products; it thus can provide greater specificity than immunohistochemistry, which may visualize proteins that have a common epitope but are otherwise unrelated to the target protein.

A significant advantage of doing in situ hybridization on free-floating tissue sections (as opposed to slide-mounted sections) is that sections from different rats may be hybridized at the same time within the same vial, and thus under identical conditions. Tissue sections from different rats are distinguished by labeling the sections during cutting with notches or punctures, so that they may be sorted by individual after hybridization. Because rats from control and experimental groups are processed under identical conditions, quantitative comparisons are more reliable.

Reagents

prices as of 2011-11

AG 501-X8 Resin (Mixed bed resin for deionizing formamide), 100g, BioRad #143-7424, $323

dATP-alphaS35, (250 uCi vial = 20 ul) Perkin Elmer #NEG034H250UC $172

made first Tuesday of every month

Denhardt's Solution 50x, 100 ml, Invitrogen # 750018 $122

(made from PVP (polyvinyl pyrrolidone) 40 & Ficoll Type 400)

DEPC (diethylpyrocarbonate), Sigma #D-5758-50 ml $203

Dextran Sulfate sodium salt, Sigma #D-8906-50 g, $143

DTT (dithiothreitol) (aka Cleland's reagent) Ultrapure, 5 g, Invitrogen #15508013 $102.00

Formamide Ultrapure, 500g bottle, Invitrogen #15515026 $97

For cDNA ISH: High Prime DNA Labeling Kit, 50 reactions, Roche Applied Science, #11585584001 $333

For Oligo ISH: Terminal Transferase Tail-Labeling Kit,

ProbeQuant G-50 micro spin columns, 50 pack, GE Healthcare, # 28-9034-08 $199

Salmon Sperm DNA 10 mg/ml, 10 x 1 ml tubes, Invitrogen #AM9680 $198

SSC Buffer 20x Ultrapure, 1L bottle, Invitrogen #15557-044 $34

Solutions

DEPC-Treated Water

• 1 ml of DEPC per 1 L ddH2O
• Squirt DEPC forcefully into water.
• Cover and shake vigorously.
• (Some protocols recommend sitting overnight.)
• Autoclave.

20x SSC (We now purchase this from Invitrogen).

• 400 ml DEPC-H2O
• 87.65 g NaCl
• 441 g Citric Acid
• Bring to pH 7.0
• Bring volume to 500 ml with DEPC-H2O
• Autoclave. (make sure volume back to 500 ml)

Hybridization of DNA or RNA probes to target RNA is dependent on salt concentration (Na+ neutralizes the PO3- charges on the DNA/RNA backbones.) SSC is an old-time decoagulant that was a common buffer in early biochemical labs. The hybridization conditions are optimized for 2x SSC.

Deionized Formamide:

• Mix 50 ml formamide and 5 g of mixed bed ion exchange resion (Bio-Rad AG 501-X8, 20-50 mesh).
• Stir 30 min at room temperature.
• Filter twice thru Whatman #1 filter paper
• Store at -20C in 15 ml tubes covered with aluminum foil.

Formamide denatures nucleic acids (lowers effective Tm). Because the breakdown products of formamide degrade nucleic acids, for sensitive applications formamide should be deionized by treatment with a mixed-bed ion-exchange resin.

5x TED

• 250 mM Tris, pH 7.4, 25 mM EDTA, 50% Dextran Sulphate
• Weigh 3.05 g Trizma Base
• Weigh 0.93 g EDTA
• Add ~35-40 ml DEPC-H2O
• Bring to pH 7.4 with HCL
• Add 50 g Deteran Sulfate slowly, while heating slowly to boil
• Bring volume to 100 ml with DEPC-H2O
• Store at 4C in 15 ml tubes covered with aluminum foil

Dextran is corn starch; serves as a thickener to increase effective concentration of all the other reagents.

50x Denhardt's Solution (We now purchase this from Invitrogen).

• 1% Ficoll, 1% PVP, 1% BSA
• 1 g Ficoll, 1 g PVP, 1 g BSA in 100 ml DEPC-H2O.
• Filter.
• Store in 1 ml tubes at -20C.

Denhardt's is a mixture of high molecular weight polymers that blocks non-specific binding sites

Herring/Salmon Sperm DNA (We now purchase this from Invitrogen).

• Dissolve 10 mg salmon or herring sperm DNA per 1 ml DEPC-H2O
• Extraction with phenol and phenol-chloroform optional.
• Sonicate for a few minutes until solution is no longer viscous (~3 min)
• An OD260 to determine exact concentration of DNA is also optional.
• Boil solution for 10 min
• Cool on ice
• Aliquot at 800 ul and store at -20C

Random DNA serves as blocker of nonspecific DNA binding

DTT (dithiothreitol) Aka Cleland's reagent; prevents oxidation of thiol groups and breaks down disulfide bonds.

Methods Section

cDNA ISH

Reference: From Rivera et al, 2009, PMID 19168037

Prepare cDNA probes by cDNA Random Priming.

Free-floating tissue sections were collected into 20 ml glass scintillation vials containing ice-cold 2 X SSC (0.3M NaCl, 0.03M sodium citrate) for in situ hybridization. The SSC in each vial was pipetted off, and sections were then suspended in 1 ml of warm prehybridization buffer (50% formamide, 10% dextran sulfate, 2 X SSC, 1X Denhardt’s solution, 50mM dithiothreitol, 0.5 mg/mL denatured salmon sperm DNA) and placed in a 48°C water bath. Two h later, 35S-dCTP-labeled cDNA probes (10 x 10^6 CPM/vial) were added to the vials and hybridized overnight at 48°C. After overnight hybridization, the sections were washed at 15-min intervals in decreasing concentrations of SSC (2X, 2X, 1X, 0.5X, 0.25X, 0.125X, 0.125X) at 48°C. After washes, the tissue sections were stored in 0.1M phosphate buffer at 4°C and then mounted on gelatin-subbed slides, air-dried, and apposed to Kodak Biomax autoradiographic film (Eastman Kodak Co., NY,

Riboprobe ISH

Reference: Kwon, B.S., M. Goltz, and T.A. Houpt. Expression of AP-1 family transcription factors in the amygdala during conditioned taste aversion learning: role for Fra-2. Brain Res. 1207 (2008) 128-41. PMID 18374904. PMCID PMC2756721.

For in situ hybridization, antisense RNA probes of [c-fos, fra-2, c-jun, and junD] cDNAs were made by in vitro transcription. Amplified cDNAs from RT-PCR were inserted into pCRII-TOPO cloning vectors (Invitrogen, Carlsbad, CA). The cloning vectors containing each cDNA were linearized by restriction enzymes. The linearized plasmids were purified by QIAquick spin column (Qiagen, Valencia, CA), and the MAXIscript kit (Ambion, Austin, TX) was used for in vitro transcription. The linearized plasmid template (1 µg), 2µl of 10X transcription buffer, 1µlof ATP, CTP, and GTP (each 10 mM), 5µl of 35S-labeled UTP (20 mci/ml) (Amersham, UK) 2µl of T7 or SP6 RNA polymerase, and RNase-free H20 were mixed in a total reaction volume of 20 µl. The mixture was incubated at 37 ºC; after 30 min, DNase I (1 µl) was added and the incubation continued for another 15 min. After stopping the reaction by the addition of EDTA (1 µl, 500 mM), the probe solution was purified through ProbeQuantTM G-50 spin columns (Amersham, Piscataway, NJ).

Forty micron free floating sections were cut on a freezing, sliding microtome and transferred into 20-ml glass scintillation vials containing 0.15 M NaCl, 0.015 M sodium citrate (2xSSC) buffer. Sections were prehybridization at 55ºC for 2-3 h in 1 ml per vial of hyrbidization buffer (50% formamide, 2xSSC, 10% dextran sulfate, 0.7% Ficoll, 0.7% polyvinylpyrrolidone, 0.7% bovine serum albumin (BSA), 85 mM dithiothreitol (DTT) and 1.4 mg/ml of yeast transfer RNA). Sections were hybridized at 55ºC for 18 h with heat-denatured 35S-labeled or biotinylated antisense RNA probes. Sections were then washed sequentially in 2xSSC, 2xSSC, 1xSSC, 0.5xSSC, 0.25xSSC, 0.125xSSC, 0.125xSSC at 55ºC for 15 min each. To reduce background, tissue sections were incubated with RNase A (30 ug/ml) both before and after mounting on gelatin coated slides. Finally, slides were washed sequentially in 2xSSC/50% formamide/0.1% β-mercaptoethanol at 53ºC for 15 min, 0.1xSSC/1% β-mercaptoethanol at 53ºC for 30 min, 50% ethanol/0.3 M NH4OAC at RT for 3 min, 85% ethanol/0.3 M NH4OAC at RT for 3 min, 100% ethanol at RT for 3 min. Radiolabeled tissue sections on slides were apposed to Biomax MR film. Biotin-labeled slides were processed for immunohistochemistry with anti-biotin antiserum (Roche) and fluorescently-tagged secondary antibodies (Vector) for visualization by fluorescent microscopy.

Oligo ISH

Reference: Swart, I., J.M. Overton, and T.A. Houpt. Hypothalamic NPY, AGRP and POMC mRNA responses to leptin and refeeding in mice. Am. J. Physiol. 283 (2002) R1020-R1026. PMID 12376393.

Prepare probes by Oligo Tail Labeling.

Coronal 40 micron sections were cut on a sliding freezing microtome through the rostral caudal extent of the hypothalamus. Alternate sections were placed into ice-cold 2x SSC (SSC = 0.15M NaCl/0.015M Na Citrate). Free floating tissue sections were prehybridized in glass vials in 1ml of 60% formamide, 0.02 M Tris pH7.4, 1mM EDTA, 10% dextran sulfate, 0.8% Ficoll, 0.8% PVP, 0.8% BSA, 2x SSC, 0.1M dithiothreitol, and 1.6 mg/mL herring sperm DNA for 2h at 37°C. After 2h prehybridization, radiolabeled probe was added (~1.0 x 107 cpm/vial) and incubated for 16-20h at 37°C. Hybridization was performed with either tail labeled prepro NPY (ppNPY) antisense oligonucleotide (bases 59-88), agouti-related protein (AGRP) antisense oligonucleotide (bases 1-45) or proopiomelanocortin (POMC) antisense oligonucleotide (bases 482-517). The oligonucleotides were tail-labeled to approximately equal specific activities (~10^7 cpm/100ng) with 35S-dATP by terminal transferase reaction (Roche), Sections were then sequentially rinsed in 2x SSC, 1x SSC and 0.5x SSC for 10 minutes at 37°C. The tissue sections were mounted from 0.05M sodium phosphate onto gelatin- coated slides, air dried, and exposed to autoradiographic film (β-max, Amersham) for 2-3 days. Tissue sections from different groups were hybridized in parallel and exposed to film together to ensure that in situ hybridization was carried out on representative members of each experimental group at the same time under identical conditions, allowing direct comparison of mRNA expression.

Quantification of ISH

Reference:

For ISH results, pixel density is quantitated from the autoradiographic films using a custom software program (MindsEye, T. Houpt). Films were digitized through a Zeiss Stemi2000C dissecting scope with a Fostec flat fiberoptic light source; light levels are adjusted to standardize gray levels of film background across ISH experiments. Images werw captured in a 10 mm x 7.5 mm frame using a scientific-grade CCD camera (Dage MTI DC330E) [UPDATE CAMERA] and frame-capture board (Scion CG-7). The region of interest (i.e. the arcuate nucleus) is delineated automatically by an algorithm that finds clusters of 10 or more contiguous pixels with an intensity 2 standard deviations above the average tissue background (mRNA-positive pixels). This algorithm works particularly well on structures with relatively discrete patterns of gene expression. Arbitrary units of mRNA expression are then derived by summing the intensities of all the mRNA-positive pixels, and subtracting the average tissue background value. For each rat and probe, average pixel densities were obtained from 3-5 brain sections. Individual mean values for each region were then averaged across rats within experimental groups.

Interpretation of ISH results

Changes in area vs. density of autoradiographic in situ hybridization signal

In autoradiographic ISH, hybridization is measured as the optical density of film exposed to tissue hybridized with a radioactive probe. Both area and density (i.e. darkness of the exposed film) of the hybridization signal are presumed to reflect mRNA levels in the tissue.

Autoradiographic ISH assumes that there is a threshold of detection for measuring a hybridization signal above background. (Typically, we chose 2 standard deviations above the average background optical density of the film and nonhybridizated tissue as an objective threshold.). Thus, a tissue must express a certain minimum amount of mRNA in order for a hybridization signal to be visible on the film. Furthermore, if above the detection threshold, the signal  is assumed to be roughly linear with increasing amounts of mRNA above threshold.

Given these assumptions, we interpret the variables as follows: "Area" represents the tissue area (i.e. number of cells) that express mRNA above the threshold of detection, within the specificed region. “Density” represents the amount of mRNA per unit area (i.e. mRNA per cell) within the specificed region. The total amount of RNA expressed within the specified region is the product of the two values (area X density).

Changes in either area or density represent changes in mRNA levels. Differential results for area and density are a consequence of ISH methodology. An increase in signal area after treatment would imply that some cells in the specified area express the mRNA below the threshold for detection in the control condition, and that the treatment increased mRNA levels in those cells above the detection threshold.  A concomittant increase in signal density would imply a higher level of mRNA expression throughout the region, including those cells that were already above threshold in the control condition. If the signal area is increased but signal density is not increased, this would imply that the cells which expressed detectable levels of mRNA in the control condition did not increase their expression after treatment. The failure to increase mRNA levels in cells that already express the gene could be due either to insensitivity to the treatment or to a “ceiling” of maximal possible expression.

Note that the ceiling or maximal level of expression may vary across regions, even for the same gene product. For example, the MR has a lower and sparser expression of 5HTT and Pet-1 mRNA than the DR under baseline conditions. Thus the dynamic range of MR cells may be lower than cells in the DR, and an increase in expression within the MR may be detected as an increase in area rather than in density. The physiological bases for the heterogenity in baseline and stimulated levels of gene expression within the raphe nuclei is a profound topic, to which this paper makes a small contribution with respect to estrogen effects.