Detection of mRNA molecules coding for neuropeptide hormones of the pond snail Lymnaea stagnalis by radioactive and non-radioactive in situ hybridization: a model study for mRNA detection

To develop and optimize non-radioactive in situ hybridization techniques for mRNA detection, we used the neuropeptidergic system of the pond snail Lymnaea stagnalis as a biological model system. First, we investigated the in situ hybridization procedure using radioactive-labeled cDNA and synthetic oligonucleotide probes specific for egg-laying hormone (ELH) mRNA and molluscan insulin-like peptide (MIP) mRNA. The results show an intense grain deposit above the caudodorsal cells and light-green cells expressing, respectively, ELH mRNA and MIP mRNA. Good results with relation to signal strength and tissue morphology were obtained with freeze-dry paraformaldehyde vapor fixation. The necessity to perform tissue pre-treatment appeared to be dependent on the cell type of interest. The optimized in situ hybridization protocol proved to be applicable using probes that are either sulfonated/transaminated or labeled with acetylaminofluorene (AAF). In situ hybridization of such haptenized probes led to intense and specific staining of the cytoplasm of the caudodorsal cells. Egg-laying hormone mRNA appeared not to be homogeneously distributed in the cytoplasm but showed a "patch-like" pattern. Nuclear and axoplasmic staining for mRNA was also observed.     

Ultrastructural non-radioactive in situ hybridization of GH mRNA in rat pituitary gland: pre-embedding vs ultra-thin frozen sections vs post-embedding.

In situ hybridization at the ultrastructural level can be carried out using three different methods: on vibratome sections before embedding in epoxy resin, on ultra-thin frozen sections, or on ultra-thin sections of tissues embedded in hydrophilic resin such as Lowicryl. With the purpose of comparing the sensitivity, resolution, and ultrastructural preservation of these three methods, we examined the expression of the growth hormone (GH) gene in anterior pituitary cells by in situ hybridization at the ultrastructural level, using a synthetic oligonucleotide complementary to the codons of the mRNA from Gln 45 to Ser 54 labeled at the 3' end of biotin-21dUTP. All these methods gave similar results: mRNA was located on the lamellar endoplasmic reticulum of somatotrophs. The pre-embedding method gave the best ultrastructural preservation, with low resolution with the enzymatic detection system and an intermediate sensitivity. A probe concentration of 10 pmol/ml was sufficient to obtain a signal. With this method gold particles could not be used without pre-treatment. The frozen section method gave the best sensitivity (a signal was observed with 4 pmol/ml of probe) but the lowest ultrastructural preservation. On ultra-thin Lowicryl sections, resolution was as high as with the frozen-section method, ultrastructural conservation was intermediate, and sensitivity was low. These results indicate that the last method seems to be a good compromise between sensitivity and ultrastructural preservation.     

3'-end fluorochromized and haptenized oligonucleotides as in situ hybridization probes for multiple, simultaneous RNA detection

We have used fluorescein-, digoxigenin- and biotin-(di)deoxyXTPs and terminal deoxynucleotidyl transferase for small scale labeling of synthetic oligonucleotide probes and here we show the applicability of such probes for the in situ detection of multiple RNA sequences. The enzymatic 3'-end-labeling methods proved to be good alternatives for the chemical fluorochrome and hapten labeling of 5'-end alkylamino-derivatized oligonucleotides. By combining 3'-end fluorescein-, biotin-, and digoxigenin-labeled oligonucleotides, double and triple hybridizations are feasible. For example, we demonstrated simultaneously mRNAs coding for caudodorsal cell hormone, a molluscan insulin-related peptide, and 28 S ribosomal RNA in cryostat sections of the cerebral ganglia of the pond snail Lymnaea stagnalis.     

Demonstration of anionic sites on the luminal and abluminal fronts of endothelial cells with poly-L-lysine-gold complex

An attempt was made to demonstrate the anionic sites on the endothelial cell (EC) surfaces of mouse brain micro-blood vessels (MBVs) after embedding of tissue samples in hydrophilic media: Lowicryl K4M, LR White, and Polyamph-10. As a cationic probe, poly-L-lysine-gold complex (PLG), prepared according to the procedure of Skutelsky and Roth (J Histochem Cytochem 34:693, 1986), was used. In ultra-thin sections of brain samples embedded in Lowicryl K4M and LR White, the anionic sites were demonstrated in the entire cross-section of the vessel wall. After embedding in Polyamph-10, however, the anionic sites could not be detected. Brain capillaries, representing blood-brain barrier type MBVs, showed polar distribution of anionic sites, evidenced by more intense labeling of luminal than of abluminal plasma membrane of the EC. Some differences in labeling of ECs and of basement membrane in arterioles and venules were also noted. The use of cationic gold and the ultra-thin sections of tissue samples embedded in hydrophilic media (Lowicryl K4M and LR White) seems to be a promising new method for detection of anionic constituents located on both luminal and abluminal surfaces of the EC, in the basement membrane, and in other components of the vessel wall.     

Ultrastructural localization of fibronectin mRNA in chick embryo by in situ hybridization using 35S or biotin labeled cDNA probes

We have studied the expression of the fibronectin gene in 7 day-old chick embryo (stage 32) by in situ hybridization at the light and electron microscope levels, using a 397 base-pairs chicken cDNA, labeled by radioisotope or biotin-11dUTP. Cryostat sections of whole chick embryos displayed a selective label on the upper layer of the dermis, fibrous sclera and mesenchymal cells but not on cartilagenous sclera cells. These results show that the expression of the fibronectin gene varies in relation to the morphogenetic events. Hybridization at the ultrastructural level on thin sections of sclera embedded in Lowicryl K4M showed a selective labelling on various cell compartments. Biotin-11dUTP and radiolabeled probes were compared. The labeling was found precisely on the membrane of the rough endoplasmic reticulum and on the nuclear envelope. A few silver grains were located on the nucleus and in the perinucleolar region. This study shows that the postembedding in situ hybridization is a powerful procedure to study the expression of the extracellular protein genes and gives further information on the localization of mRNA.     

The use of silver-enhanced 1-nm gold probes for light and electron microscopic localization of intra- and extracellular antigens in skin.

We used colloidal gold (1-nm diameter) with silver enhancement, in conjunction with a low-temperature post-embedding immunolabeling technique, to localize several antigens in normal skin at both the light and the electron microscopic level within the same tissue blocks. Normal skin subjected to cyrofixation and cryosubstitution and embedded in Lowicryl K11M was used as a substrate. Semi-thin sections (1 micron) were incubated in primary antibody (against epidermal basement membrane zone associated antigens and two keratin sub-types), biotinylated secondary antibodies, and then in 1-nm gold-conjugated streptavidin. Finally, the 1-nm gold label was enhanced using silver staining. Labeling of both basement membrane and keratin antigens was well demonstrated, and the area in the semi-thin sections showing the best structural preservation and the greatest intensity of immunolabeling was used to identify the part of the block to be used for ultra-thin sectioning. Ultra-thin sections were treated using a similar procedure to that employed for semi-thin sections. The labeling with silver-enhanced 1-nm gold probes was intense and readily visible by electron microscopy, even at low magnification. We have found this technique to have a high degree of specificity and sensitivity for labeling both intra- and extracellular antigens in skin, with the added advantage of providing the means for studies at both light microscopic and electron microscopic level.     

Usefulness of the immunogold technique in quantitation of a soluble protein in ultra-thin sections

We used a model system to study whether measurements of absolute local antigen concentrations at the electron microscopic level are feasible by counting immunogold labeling density in ultra-thin sections. The model system consisted of a matrix of a variable concentration of gelatin, which was mixed with given concentrations of rat pancreas amylase and fixed according to various fixation protocols. With a relatively mild fixation, there was no clear proportionality between anti-amylase gold labeling and amylase concentration in ultra-thin cryosections. This was presumably due to uncontrolled loss of amylase from the sections. After stronger fixation with 2% glutaraldehyde for 4 hr, labeling density reflected the amylase concentration very well. We observed that matrix (gelatin) density influenced labeling density. A low gelatin concentration of 5% allowed penetration of immunoreagents into the cryosection, resulting in a high and variable labeling density. In gelatin concentrations of 10% and 20%, labeling density was lower but proportional to amylase concentration. To establish an equal (minimal) penetration of immunoreagents, we embedded model blocks with different matrix densities in polyacrylamide (PAA). In ultra-thin cryosections of these PAA-embedded blocks, anti-amylase labeling was proportional to amylase concentration even at a low (5%) gelatin concentration. Anti-amylase labeling in ultra-thin sections from Lowicryl K4M low temperature-embedded blocks was higher than in PAA sections, but the results were less consistent and depended to some extent on matrix density. These results, together with the earlier observation that acrylamide completely penetrates intracellular compartments (Slot JW, Geuze HJ: Biol Cell 44:325, 1982), demonstrate that it is possible to measure true intracellular concentrations of soluble proteins in situ using ultra-thin cryosections of PAA-embedded tissue.     

Electron microscopic autoradiographic localization of prolactin mRNA in rat pituitary

Recent immunoelectron microscopic studies have shown that immunoreactive prolactin (PRL) in rat pituitary can be detected not only in typical PRL cells, characterized by large secretory granules, but also in another type of cell, which contains small secretory granules. To determine whether or not these two cell types are involved in PRL biosynthesis, we developed a procedure to investigate PRL gene expression by using in situ hybridization at the ultrastructural level. Rat pituitary was fixed and vibratome sections were incubated with a PRL [35S]-cDNA probe and subsequently flat-embedded in Araldite. Semi-thin and ultra-thin sections were processed for autoradiography. The results indicate that only the two PRL cell types were labeled. When immunolabeling for PRL was applied to ultra-thin sections, only immunopositive cells were seen to contain silver grains. In these cells the silver grains were associated with the rough endoplasmic reticulum and nucleus. When a growth hormone (GH) [35S]-cDNA probe was used as a control, only GH-secreting cells were labeled. This study confirms that the two PRL cell types are involved in biosynthesis of PRL. Moreover, this simple in situ hybridization technique provides a new approach to accurately localize mRNA in complex tissue and to investigate the subcellular distribution of mRNA under differing experimental conditions.     

Effect of particle size on labeling density for catalase in protein A-gold immunocytochemistry

Effect of particle size on labeling intensity in protein A-gold immunocytochemistry was studied. Catalase labeling of rat liver peroxisomes was used as a labeling model. Ultra-thin sections of Lowicryl K4M-embedded rat liver were stained for catalase with protein A-gold (pAg) probes. Five different sizes of colloidal gold probes, from 5 nm to 38 nm in diameter, were prepared. Labeling intensity decreased as the particle size of the pAg probes increased. The highest labeling was obtained by the 5-nm pAg probe and the lowest by the 38-nm pAg probe. Quantitative analysis also showed that labeling density was inversely proportional to the size of gold particles. The results suggest that the pAg probe with small gold particles has high sensitivity.     

Adaptation of a non-radioactive in situ hybridization method to electron microscopy: detection of tenascin mRNAs in mouse cerebellum with digoxigenin-labelled probes and gold-labelled antibodies

In this study we describe a method for the detection of mRNAs at the ultrastructural level using a non-radioactive in situ hybridization method based on digoxigenin-labelled cRNA probes and gold-labelled digoxigenin-specific antibodies. We applied this protocol to an analysis of the expression of the extracellular matrix protein tenascin in the developing cerebellar cortex of the mouse. To gain an impression of the sensitivity attainable with digoxigenin-labelled probes, we first established at the light microscopic level that the hybridization signal obtained with the non-radioactive probe is as sensitive as that obtained with a ³⁵S-labelled probe. The non-radioactive hybridization protocol was then combined with electron microscopic post-embedding and immunogold detection techniques. Tenascinspecific, digoxigenin-labelled cRNA probes were hybridized to ultrathin sections of Lowicryl K4M-embedded tissue and the probe/target mRNA hybrids were detected using gold-labelled antibodies to digoxigenin. In agreement with the observations from in situ hybridization at the light microscopic level, specific labelling was observed in Golgi epithelial cells in the region of the Purkinje cell layer and cells in the internal granular layer, which could be identified as astrocytes by ultrastructural criteria. Labelling was detectable in association with free ribosomes and ribosomes of the rough endoplasmic reticulum. In addition, focal hybridization signals were occasionally found in the nucleus. No signal was observed in Golgi epithelial cells or astrocytes using sense or in any other cerebellar cell type using either sense or anti-sense probes. The described in situ hybridization technique uses ultrastructural criteria to associate the presence of a given mRNA species with a particular cell type. Additionally, it provides information about the target mRNA's subcellular distribution, thus offering the possibility to study intracellular transport of particular mRNAs.     

Ultrastructural distribution of myosin heavy chain mRNA in cardiac tissue: a comparison of frozen and LR White embedment

Electron microscopy (EM) in situ hybridization provides the higher resolution necessary to determine the spatial relationship between a specific mRNA and the organelle containing the protein encoded by that message. EM in situ hybridization was used to determine the subcellular myosin heavy chain (MHC) mRNA distribution with respect to the myofibril in normal cardiac tissue. Sections of frozen or acrylic-embedded tissue were compared for ultrastructural integrity and content of endogenous mRNA. Papillary muscles dissected from hearts of normal rabbits were aldehyde-fixed and either frozen or embedded in LR White. EM in situ hybridization with no riboprobe, vector sequence, same-sense, and anti-sense biotinylated riboprobes was detected by indirect immunocytochemistry. Labeling density using an antisense probe was highest over the intermyofibrillar space, with an average signal five times that of background. Background labeling by nonspecific sense probe was consistently low but not random, also having the highest density of gold clusters over the intermyofibrillar space. Ultracryomicrotomy yielded a higher absolute number of gold clusters, but sections were fragmented and disrupted striated muscle morphology. LR White embedment maintained ultrastructural integrity but gave a lower absolute signal. Fortunately, MHC mRNA is an abundant message and can tolerate the decreased sensitivity of LR White.     

Liquid phase SPR imaging experiments for biosensors applications

Surface plasmon resonance (SPR) has recently gained attention as a label-free method for the detection of biological molecules binding onto functionalised surfaces. It is one of the most sensitive detection method for monitor variations in the thickness and refractive index in ultra-thin films. Here, the adsorption processes of oligonucleotides onto gold substrates have been investigated in aqueous buffer solution using SPR imaging measurements. The hybridization of a thiol-modified, single stranded oligonucleotide anchored to a gold surface via thiol group, with its complementary sequence has been observed and characterised monitoring the hybridization process by SPR equipment. In situ investigation of smallest changes in SPR imaging measurements dynamically performed in liquid phase in the presence of DNA complementary probes was performed. Infrared spectroscopy and scanning electron microscopy characterisation of the functionalised gold surfaces of the biosensor were compared with the images obtained by SPR experimental apparatus.     

Suitability of Different Protein A-Gold Markers for Immunogold-Silver Staining in Paraffin Sections

An incubation protocol to immunolabel Lowicryl semithin sections was applied to paraffin probes. To improve the labeling density, colloidal gold complexes of different preparations and sizes were compared. The type of colloidal gold preparation used was found to affect the specificity of the immunostaining. Gold colloid of 5 nm diameter particle size prepared with white phosphorus minimized nonspecific background labeling of β-casein in paraffin embedded sections of the mammary epithelium of pregnant mice. Gold colloids of 5 nm and 9 nm diameter particle size prepared in varying concentrations of tannic acid generated significant nonspecific staining in similar tissue preparations.     

Suitability of Different Protein A-Gold Markers for Immunogold-Silver Staining in Paraffin Sections

An incubation protocol to immunolabel Lowicryl semithin sections was applied to paraffin probes. To improve the labeling density, colloidal gold complexes of different preparations and sizes were compared. The type of colloidal gold preparation used was found to affect the specificity of the immunostaining. Gold colloid of 5 nm diameter particle size prepared with white phosphorus minimized nonspecific background labeling of β-casein in paraffin embedded sections of the mammary epithelium of pregnant mice. Gold colloids of 5 nm and 9 nm diameter particle size prepared in varying concentrations of tannic acid generated significant nonspecific staining in similar tissue preparations.     

In situ hybridization with non-radioactive digoxigenin-11-UTP-labeled cRNA probes: localization of developmentally regulated mouse tenascin mRNAs.

An improved method for in situ hybridization was developed in order to identify the tissue-specific expression of messenger RNA (mRNA) for the novel extracellular matrix glycoprotein, tenascin, during mouse development. Non-radioactive RNA probes were generated by incorporating digoxigenin-11-UTP instead of conventional isotopic labels. Hybridization of anti-sense probes to complementary mRNAs was detected by a chromogenic staining reaction catalyzed by an anti-digoxigenin antibody-alkaline phosphatase conjugate. Markedly improved enhancement of staining was achieved by expanding the complexity of probes and strictly controlling the degree of proteolytic digestion of paraformaldehyde-fixed tissue sections. Six different complementary RNA (cRNA) probes representing most of the tenascin mRNA sequence were prepared. Very weak signals were obtained after single applications of each probe, but strong specific signals were present when all six probes were mixed together. In either case, no signal was found without prior proteolytic digestion of tissue sections with proteinase K. Treatment with increasing concentrations of proteinase K initially resulted in increased sensitivity of signal detection, but extensive digestion resulted in histological sections of poor quality for light microscopy. Optimal conditions varied according to the tissue type examined. In lung, in situ hybridization detected tenascin mRNA in the relatively large cells lining alveolar walls adjacent to type I pneumocytes. In cerebellum, glial cells of the Purkinje cell layer contained tenascin mRNA, but Purkinje cells did not. In both cases, hybridization signals were confined to the cytoplasm of cells, and no extracellular staining was observed. This method provides a promising new tool for analysis of spatio-temporal regulation of tenascin gene expression during embryogenesis and oncogenesis.     

In situ hybridization at the electron microscope level: localization of transcripts on ultrathin sections of Lowicryl K4M-embedded tissue using biotinylated probes and protein A-gold complexes

A technique has been developed for localizing hybrids formed in situ on semi-thin and ultrathin sections of Lowicryl K4M-embedded tissue. Biotinylated dUTP (Bio-11-dUTP and/or Bio-16-dUTP) was incorporated into mitochondrial rDNA and small nuclear U1 probes by nick-translation. The probes were hybridized to sections of Drosophila ovaries and subsequently detected with an anti-biotin antibody and protein A-gold complex. On semi-thin sections, probe detection was achieved by amplification steps with anti-protein A antibody and protein A-gold with subsequent silver enhancement. At the electron microscope level, specific labeling was obtained over structures known to be the site of expression of the appropriate genes (i.e., either over mitochondria or over nuclei). The labeling pattern at the light microscope level (semi-thin sections) was consistent with that obtained at the electron microscope level. The described nonradioactive procedures for hybrid detection on Lowicryl K4M-embedded tissue sections offer several advantages: rapid signal detection: superior morphological preservation and spatial resolution; and signal-to-noise ratios equivalent to radiolabeling.     

Immunocytochemical localization of acid phosphatase in rat liver

Localization of acid phosphatase (ACPase) in rat liver was investigated by immunocytochemical techniques. Rat liver was fixed by perfusion and cut into thick tissue slices, which were embedded in Epon or Lowicryl K4M. For light microscopy (LM), semithin Epon sections were stained for the enzyme ACPase by an indirect immunoenzyme technique. For electron microscopy (EM), ultra-thin Lowicryl K4M sections were stained by a protein A-gold technique. By means of LM, granular reaction deposits were observed in hepatocytes and sinus-lining cells. Stained granules were present in the juxtanuclear cytoplasm, but they did not correspond to a typical staining pattern for the Golgi complex. EM revealed that gold particles indicating ACPase antigens were present on lysosomes and on some vesicles locating in the trans Golgi region. Endosomelike vesicles were strongly positive for the labeling. Golgi cisterna were mostly negative, but weak signals were noted in dilated sacules. The plasma membranes on the sinusoidal and bile canalicular sides were labeled by a few gold particles. The results indicate that ACPase is present in endosomes and in a restricted area of plasma membrane, as well as in the lysosomal system.     

Immunoelectron microscopic localization of sarcoplasmic reticulum proteins in cryofixed, freeze-dried, and low temperature-embedded tissue

Immunoelectron microscopic labeling of calsequestrin on ultra-thin sections of rat ventricular muscle prepared by quick-freezing, freeze-drying, and direct embedding in Lowicryl K4M was compared to that observed on ultra-thin sections prepared by chemical fixation, dehydration in ethanol, and embedding in Lowicryl K4M. Brightfield electron microscopic imaging of cryofixed, freeze-dried, osmicated, and Spurr-embedded rat ventricular tissue showed that the sarcoplasmic reticulum was very well preserved by cryofixation and freeze-drying. Therefore, the four structurally distinct regions of the sarcoplasmic reticulum (i.e., the network SR, the junctional SR, the corbular SR, and the cisternal SR) were easily identified even when myofibrils were less than optimally preserved. As previously shown by immunoelectron microscopic labeling of ultra-thin frozen sections of chemically fixed tissue, calsequestrin was confined to the lumen of the junctional SR and of a specialized non-junctional (corbular) SR, and was absent from the lumen of network SR in cryofixed, freeze-dried, Lowicryl-embedded myocardial tissue. In addition, a considerable amount of calsequestrin was also present in the lumen of a different specialized region of the non-junctional SR, called the cisternal sarcoplasmic reticulum. By contrast, relocation of calsequestrin to the lumen of the network SR was observed to a variable degree in chemically fixed, ethanol-dehydrated, and Lowicryl-embedded tissue. We conclude that tissue preparation by cryofixation, freeze-drying, and direct embedding in Lowicryl K4M for immunoelectron microscopic localization of diffusible proteins, such as calsequestrin, is far superior to that obtained by chemical fixation, ethanol dehydration, and embedding in Lowicryl K4M.