Detection of radiolabeled probes entails apposition of the samples to x-ray film or phosphorimaging plates and subsequent development. Higher, cellular resolution entails coating of the sample with a nuclear emulsion (described here). These steps are performed after detection of the digoxigenin-labeled probes if the two types of probes are used simultaneously.

Materials

Prepare the nuclear emulsion for coating the sections
1. Under safelight conditions, scoop out 40 ml of emulsion with spatula into a coplin jar containing 1.6 ml of 7.5M ammonium acetate (final concentration of 300mM).
2. Place the coplin in a 40° C water bath for 20-30 mins to allow air bubbles to rise. Mix gently and look for bubbles on a clean slide after dipping it into the emulsion.
3. Dip slides into the emulsion, and stand up for several hours to dry.

We place 5 slides in red plastic slide grips and dip them 5 at a time into the emulsion. We then hang them from a custom-made plexiglass holder.

4. The emulsion-coated slides are placed in black slide boxes with desiccant capsules. Tape the edges of the box with black photography tape and store the boxes at 4° C in the dark.

Develop the emulsions and stain the tissue sections
5. Put the slides in racks and pass through the solutions as follows at 17° C. (with agitation every 30s): D-19 for 2mins; running tap water 15s with slight agitation; and Kodak Rapid Fix (without hardener) for 2mins.

The room lights may be turned on after all slides are fixed.

6. Rinse in running tap water for 8 minutes. Counterstain, if desired, for 30s in 0.4% toluidine blue, 2µg/ml ethidium bromide, hematoxylin/eosin, or stain of choice, and rinse again briefly to remove excess stain.

Some stains may obscure or destroy colorimetric detection of the digoxigenin probe or silver grains (e.g., periodic acid Schiff, cresyl violet).

7. Dip very briefly into deionized water, then 70% ethanol and place on slide warmer to thoroughly dry.
8. Coverslip slides with Cytoseal 60 (or, similar organic-based) mountant.

Return to top
REAGENT AND SOLUTIONS

Use DEPC-treated H₂O to make solutions for solutions for pretreatment and hybridization. De-ionized H₂O may be used for the subsequent wash and digoxigenin development steps.

Slides for mounting tissue sections:

1. Subbed slides
   
   • Place slides in racks and soak in soap solution for 1 hr.
     
   • Rinse in deionized water. Change the water several times to be sure that all of the soap is removed.
     
   • Dissolve 1.88 g of gelatin (300 bloom swine) in 750 ml hot H₂O (do not allow to boil). Cool and dissolve 0.188 g CrK(SO₄)₂*12H₂O in the solution.
     
   • Dip the slides into the subbing solution, drain the slides onto a paper towel and allow to air dry for 1 hr.
     
   • Dip the slides into the subbing solution again. Drain and cover loosely with plastic wrap or bench paper.
     
   • When thoroughly dry, store the slides in slide boxes.

2. Silanized slides:
   
   • Clean slides with a lint free cloth manually (lots of work, but needed) and put them in racks.
     
   • Dip slides in 2% aminoalkalynsilane (Sigma A-3648) in dry acetone for 10 sec.
     
   • Rinse in deionized water 3 times.
     
   • Air dry overnight and store in boxes protected from dust.

3. Positively charged slides:
   
   • We buy these slides: Superfrost Plus microscope slides (4951+, Erie Scientific, Portsmouth, NH 03801) or you can make them as follows:
     
   • Clean slides with a lint free cloth manually (lots of work, but needed) and put them in racks
     Dip slides in 50?g/ml poly-L-lysine
     
   • Air dry overnight and store in boxes protected from dust.

10X Phosphate-buffered saline, pH 7.4 (PBS)

3.7% Formaldehyde solution

0.25% acetic anhydride, pH 8.0

20X SSPE

Ribonucleic acid solution

50X Denhardts solution

Hybridization buffer

Hybridization solution

RNase A solution

We have had RNase A lots that are more potent than others, so when new RNase A is obtained, it should be tried over a range of concentrations to determine the best signal-to-noise ratio.

Northern hybridization solution

Return to top
COMMENTARY
Background Information
Hybridization histochemistry provides a method to detect specific mRNAs in tissue sections. Furthermore, as mRNA levels may change from one state to another (e.g., during development or after physiological manipulations), hybridization histochemistry can provide snapshots through the course of a dynamic situation. This unit reprises our protocols to examine the expression of genes within tissue sections at a light microscopic resolution (Young et al., 1986; Young et al., 1990; Bradley et al., 1992; Young, 1992). There are a number of excellent sources of information for the reader interested in localizing transcripts in whole mount tissues, to chromosomes or at the electron microscopic level (Wilkinson, 1992; Rosen and Beddington, 1993; Albertson et al., 1995; Morey, A.L., 1995; Swiger, and Tucker, 1996).
Hybridization histochemistry is generally amenable to combining with other techniques, such as immunohistochemistry, tract-tracing (Burgunder, and Young, 1988) and in vitro receptor autoradiography (Westlake et al., 1994). The combination with immunohistochemistry and/or tract-tracing may necessitate perfusion fixation of the animal (in order to preserve immunoreactivity and/or tracer deposition) prior to freezing the specimen and sectioning it. These sections, however, generally have a reduced signal-to-noise ratio for the hybridization histochemistry. The immunohistochemical steps are usually performed after the the hybridization histochemical ones to avoid loss of mRNA from exposure to RNases present in the antibody and development solutions. To use the same tissue for in vitro receptor autoradiography and hybridization histochemistry, alternate fresh-frozen sections are used.

Critical Parameters and Troubleshooting
Successful hybridization histochemistry seeks a balance between preserving tissue morphology and permeabilizing the tissue to allow access of the probe to the transcripts. While a number of protocols utilize HCl and/or proteases to permeabilize the tissue sections, our approach avoids these harsh treatments through the use of chloroform to de-fat the sections. However, paraffin-embedded tissue sections require the use of protease.
Our experience has shown that the longer riboprobes that, obviously target longer stretches of the transcripts, than single oligodeoxynucleotide probes, offer greater sensitivity. However, the use of multiple oligodeoxynucleotide probes targeted against the same transcript can significantly improve sensitivity compared to a single oligodeoxynucleotide probe. Theoretically, and in practice, riboprobes are more sensitive than the equivalent stretch of bases presented by labeled, double-stranded cDNA probes. Some researchers also employ alkaline hydrolysis of their riboprobes to increase the ease of tissue penetration. Again, we have not found this to help with our protocol and may be inconsistent in the probe sizes produced. References to other approaches and further discussion are presented at the end of the unit (Valentino et al., 1987; Wilkinson, 1992)
We label our probes for hybridization histochemistry with ³⁵S for a number of reasons. It provides greater resolution and higher efficiency of grain production than either ³²P or ³³P. Also, it has a half-life of 87 days, compared with 14 and 25 days for ³²P or ³³P, respectively. These reasons more than compensate for ³⁵S's lower specific activity. Although ³H provides greater resolution and has a much longer half-life, its specific activity is so low, that it is not practical to label probes targeted against transcripts of relatively low abundance. We generally use the digoxigenin-labeled probes to enable double simultaneous detection of two different transcripts within the same tissue sections (and within the same cells). The radiolabeled probes permit more accurate quantitation of transcript levels and are still more sensitive.
Controls for specificity are, of course, the essence of any experiment. Unfortunately, there is no single, absolute control for hybridization histochemistry. Instead, the researcher relies on as many different checks as possible. The ones we prefer, in a roughly descending order of usefulness, are as follows: 1. Same distribution of signal with probe directed against a different portion of the same transcript. 2. Blockage of signal by prior hybridization with unlabeled probe. 3. Correlation of signal with immunocytochemical results. 4. Different distribution of signal with probes against unrelated transcripts, including sense probes (be aware, however, that occasionally the sense probe detects mRNA transcribed from the opposite DNA strand. However, a signal with the sense probe does not necessarily invalidate the findings obtained with the antisense probe). 5. Northern analysis using the probe under the same degrees of stringency shows band(s) of expected size(s). Two commonly used controls are not recommended: the use of RNase prior to hybridization is analogous to using a protease prior to immunohistochemistry and the dilution of of labeled probe with unlabeled probe only serves to reduce the specific activity of the probe. Both of these procedures are essentially worthless.
One should be aware of the potential artifacts that arise from autoradiography and/or color techniques. Positive and negative chemography, the spurious creation and destruction of grains, respectively, are constant concerns with autoradiography. Positive chemography is probably more common and is best assessed using sections that were not hybridized, or hybridized with a sense probe. Grains are especially susceptible to loss during staining or after coverslipping if moisture still remains in the tissue sections. These and other aspects of autoradiography are expertly discussed by Rogers (1979). Color development artifacts with alkaline phosphatase may be due to endogenous peripheral-type enzyme and may be blocked with levamisole (intestinal alkaline phosphatase is more refractory and needs treatment with 0.1M HCl for 10 min. at room temperature; Kiyama and Emson, 1991). Also, DTT that is present during the enzymatic development can impart a strong purplish color. The use of non-hybridized sections should reveal whether adventitious color formation is occurring. Loss of alkaline phosphatase staining occurs with exposure to ethanol.

Anticipated Results
Hybridization histochemistry should enable the researcher to determine whether a given gene is expressed in particular cells. Figure 1 shows the simultaneous detection of two different transcripts detected through the use of radiolabeled and digoxigenin-labeled probes. Riboprobes are more sensitive than oligodeoxynucleotide probes, enabling one to see 5 or fewer transcripts per cell, and traditionally, ³⁵S-labeled probes are more sensitive than colorimetrically detected ones. However, as we gain more experience with amplification techniques, this superiority of radiolabeled probes may vanish. Furthermore, the recent introduction of the tyramide amplification system (TSA, also known as catalyzed reporter deposition or CARD) (Bobrow et al., 1989) offers a number of branch points for varying degrees of amplification and different reaction products (Kerstens et al., 1995; Hunyady et al., 1996).

Quantitative Analysis of Autoradiograms
X-ray or tritium sensitive films may provide the easiest means to quantitation if the signal is sufficient and the cells are closely grouped. In these cases, the film optical densities can be converted to copies of probe hybridized through the use of simultaneously exposed brain paste standards that incorporate known amounts of the radioisotope. Phosphorimaging devices (e.g., those of Fuji Medical Systems or Molecular Dynamics) offer 2 advantages over films: their sensitivity is up to 40-fold greater and the signals are directly proportional to the amount of hybridized radiolabeled probe. We generally examine our sections with the phosphorimaging system prior to dipping them into nuclear emulsion. Detailed protocols for quantitative analysis of autoradiograms are available (Gerfen, 1989; Young, 1992).

Time Considerations
Hybridization histochemistry may be viewed as composed of three steps: preparation of the tissue sections, hybridization and washing of the sections, and detection of the hybridization signal. Preparation of the tissue sections, after collection of the tissue specimens, essentially consists of cutting the sections and, of course, depends upon the numbers of sections needed and the size of the region(s) studied. This may take hours to weeks.
The hybridization and washing steps take either 2 or 3 consecutive days, depending on whether radiolabeled or digoxigenin-labeled probes are used, respectively. Detection of the digoxigenin-labeled probes is then complete at the end of the third day. Depending on the signal strength and degree of resolution needed, radiolabeled probe deposition can be determined over the course of minutes using film or phosphorimaging plates to months after coating with nuclear emulsion.

Literature Cited

1. Albertson, D.G., Fishpool, R.M. and Birchall, P.S. 1995. Fluorescence in situ hybridization for the detection of DNA and RNA.Methods Cell Biol 48:339-364.
   
2. Bobrow, M.N., Harris, T.D., Shaughnessy, K.J. and Litt, G.J. 1989. Catalyzed reporter deposition, a novel method of signal amplification.J Immun. Meth. 125:279-285.
   
3. Bradley, D.J., Towle, H.C. and Young, W.S. III. 1992. Spatial and temporal expression of alpha and beta thyroid hormone receptor mRNAs, including the b₂ subtype, in the developing mammalian nervous system.J. Neurosci. 12:2288-2302.
   
4. Burgunder, J.-M. and Young, W.S. III. 1988. The distribution of thalamic projection neurons containing cholecystokinin messenger RNA, using in situ hybridization histochemistry and retrograde labeling.Mol. Brain Res. 4:179-189.
   
5. Gerfen, C.R. 1989. Quantification of in situ hybridization histochemistry for analysis of brain function. In Methods in Neuroscience (Conn, P.M., ed) pp. 79-97. Academic Press, New York
   
6. Hunyady, B., Krempels, K., Harta, G. and Mezey, É . 1996. Immunohistochemical signal amplification by catalyzed reporter deposition and its application in double immunostainings.J. Histochem. Cytochem.12:1353-1362
   
7. Kerstens, H.M.J., Poddihe, P.J. and Hanselaar, A.G.J.M. 1995. A novel in situ hybridization signal amplification method based on the deposition of biotinylated tyramine.J. Histochem. Cytochem. 43:347-352.
   
8. Kiyama, H. and Emson, P.C. 1991 An in situ hybridization histochemistry method for the use of alkaline phosphatase-labeled oligonucleotide probes in small intestine. J. Histochem. Cytochem. 39: 1377-1384.
   
9. Morey, A.L. 1995. Non-isotopic in situ hybridization at the ultrastructural level.J. Pathol. 176:113-121.
   
10. Rogers, A.W. 1979. Techniques of Autoradiography. Elsevier, New York.
    
11. Rosen, B. and Beddington, R.S. 1993. Whole-mount in situ hybridization in the mouse embryo: gene expression in three dimensions.Trends Genet. 9:162-167.
    
12. Swiger, R.R. and Tucker, J.D. 1996. Fluorescence in situ hybridization: a brief review.Environment. Mol. Mutagen. 27:245-254.
    
13. Valentino, K.L., Eberwine, J.H. and Barchas, J.D. (eds.). 1987. In Situ Hybridization. Applications to Neurobiology. Oxford University Press, New York.
    
14. Westlake, T.M., Howlett, A.C., Bonner, T.I., Matsuda, L.A. and Herkenham, M. 1994. Cannabinoid receptor binding and messenger RNA expression in human brain: an in vitro receptor autoradiography and in situ hybridization histochemistry study of normal aged and Alzheimer's brains.Neuroscience 63:637-652.
    
15. Wilkinson, D.G. (ed). 1992. In Situ Hybridization. A Practical Approach. Oxford University Press, New York.
    
16. Young, W.S. III. 1992. In situ hybridization with oligodeoxyribonucleotide probes. In In Situ Hybridization. A Practical Approach (Wilkinson, D.G., ed) pp. 33-44. Oxford University Press, New York
    
17. Young, W.S. III., Mezey, É . and Siegel, R.E. 1986. Vasopressin and oxytocin mRNAs in adrenalectomized and Brattleboro rats: analysis by quantitative in situ hybridization histochemistry.Mol. Brain Res. 1:231-241.
    
18. Young, W.S. III., Reynolds, K., Shepard, E.A., Gainer, H. and Castel, M. 1990. Cell-specific expression of the rat oxytocin gene in transgenic mice. J. Neuroendocrinol.2:917-925.

Photomicrograph shows oxytocin and vasopressin neurons within the human supraoptic nucleus labeled with 35S-labeled and digoxigenin-labeled probes, respectively. Neurons containing the digoxigenin-labeled probe show a dark stain from development of the alkaline phosphatase on the anti-digoxigenin antibodies. The deposition of the radiolabeled probe is indicated by the green-colored silver grains. Click on the image to see an enlarged version.