Histochem Cell Biol (1998) 110:431–437
© Springer-Verlag 1998
O R I G I N A L PA P E R
&roles:Mariëtte P.C. van de Corput · Roeland W. Dirks Rob P.M. van Gijlswijk · Frans M. van de Rijke Anton K. Raap
Fluorescence in situ hybridization using horseradish peroxidase-labeled oligodeoxynucleotides and tyramide signal amplification for sensitive DNA and mRNA detection &misc:Accepted: 15 April 1998
&p.1:Abstract We have used horseradish peroxidase-labeled 40 mer oligodeoxynucleotides (HRP-ODNs) specific for the human cytomegalovirus immediate early gene (HCMV-IE) and a novel dinitrophenol-tyramide signal amplification reagent (DNP-TSA plus) to evaluate their utility in fluorescence in situ hybridization (FISH). For DNA FISH, single or cocktails of HRP-ODNs were hybridized to metaphase chromosomes of rat 9G cells which, as determined by DNA fiber FISH, carry an integrated tandem repeat of 50–60 copies of the HCMV-IE gene. With one layer of DNP-TSA plus deposition and subsequent detection with a fluorochrome-conjugated antibody, four HRP-ODNs were needed to detect the HCMV-IE integration site. When employing two TSA amplification rounds, one HRP-ODN was sufficient for obtaining a strong signal of the integrated gene cluster, indicating that 50–60 HRP molecules can be detected with ease. In addition to DNA FISH, we report here the first use of HRP-ODN probes for mRNA detection by FISH. A single HRP-ODN and one DNP-TSA plus step resulted in clear visualization of the HCMV-IE gene transcripts in rat 9G cells induced for HCMV-IE expression by cycloheximide. Two TSA detection steps enhanced signal intensities even further. Parallel experiments with hapten-labeled ODN and cDNA probes and conventional detection methods illustrated the superiority of the HRP-ODN/TSA approach in DNA and RNA FISH.&bdy:
Introduction Fluorescence in situ hybridization (FISH) has become a valuable tool for gene localization and gene expression M.P.C. van de Corput · R.W. Dirks · R.P.M. van Gijlswijk F.M. van de Rijke · A.K. Raap (✉) Laboratory for Cytochemistry and Cytometry, Department of Molecular Cell Biology, Leiden University Medical Centre, Wassenaarseweg 72, NL-2333 AL Leiden, e-mail:
[email protected] Tel.: +31-71-5276187, Fax: +31-71-5276180&/fn-block:
studies. Optimization of the various steps of this technique has resulted in a widespread use of FISH for detection of specific DNA and mRNA sequences in a histoand cytological context (Dirks 1996). Both fluorochrome- and hapten-labeled probes are used for that purpose. Fluorochromized probes have the advantage of direct visualization, but have a limited sensitivity. Hence, the use of haptens as nucleic acid probe labels is often favored over fluorochromes because multiple immunological layers can be applied for signal enhancement. However, use of more than three detection layers invariably leads to increasing numbers of probe and/or immunological reagent-related background signals. For FISH to low abundance mRNAs or low copy number and unique DNA targets, reduction of background signals is therefore as important as increasing the detection sensitivity. By reducing the probe complexity by switching from commonly used cDNAs or cRNAs to oligodeoxynucleotides (ODNs), background signals emanating from nonspecifically bound probe sequences can in principle be diminished. This will, however, be at the cost of detection sensitivity, often to the point at which autofluorescence levels are not superseded. The peroxidase-driven tyramide signal amplification (TSA) detection system has recently proven to considerably enhance the detection sensitivity of various in situ techniques (Kerstens et al. 1995; Raap et al. 1995; De Haas et al. 1996; Hunyady et al. 1996; Van Gijlswijk et al. 1997; Van Heusden et al. 1997). We reasoned that, by conjugating the peroxidase enzyme directly to ODNs, the TSA system would generate sufficient signal intensity for microscopic visualization of ISH applications (Van Gijlswijk et al. 1996b). Furthermore, primary horseradish peroxidase (HRP)conjugated immunological detection layers, which cause prominent noise signals, can be omitted. Here we present experiments that confirm this reasoning by showing that DNA and RNA FISH with HRP-ODNs in combination with TSA detection strategies is superior in sensitivity to conventional methods using haptenized, large DNA probes and immunological detection systems. Hence, the
432
HRP-ODN/TSA approach may be useful for detection of low abundance mRNAs by FISH.
Materials and methods Cells Rat 9G cells have a stable integration of a tandem repeat of approximately 50–60 copies of plasmid pES in which a 7.0-kb immediate early gene fragment of the human cytomegalovirus (HCMV-IE) was cloned (Boom et al. 1986; Fig. 1A). For fiber and RNA FISH, rat 9G cells were cultured on uncoated microscope glass slides in Dulbecco’s modified Eagle medium without phenol red containing 4.5 mg/ml glucose and 110 µg/ml pyruvate supplemented with 10% fetal calf serum, 0.03% glutamine, 1000 U/ml penicillin/streptomycin in a 5% CO2 atmosphere. After attachment to the glass slides, expression of the IE gene was induced by adding 50 µg/ml cycloheximide (Sigma, St. Louis, Mo., USA) to the culture medium, resulting in S phase-dependent HCMV-IE expression in approximately 30% of the cells. After 4–6 h of incubation, cells were washed with PBS and fixed in 4% formaldehyde, 5% acetic acid in PBS for 20 min at room temperature. After fixation, the cells were washed twice with PBS and stored in 70% ethanol at 4°C until further use. DNA fiber FISH The number of integrated HCMV-IE DNA copies in rat 9G cells was determined using fiber FISH technology as described by Wiegant et al. (1992) and Florijn et al. (1995). The stretched DNA fibers were simultaneously hybridized with a digoxigenin-labeled probe (pSS) specific for the IE gene fragment of the integrated pES plasmid and a biotinylated probe (pBR) specific for the vector sequence (see Fig. 1A). After hybridization, the digoxigenin-labeled probe was immunocytochemically detected using mouse antidigoxin (Sigma) followed by rabbit anti-mouse-fluorescein (Sigma) and goat anti-rabbit-fluorescein (Sigma). The biotin-labeled probe was detected using streptavidin-Texas Red (Boehringer-Mannheim, Mannheim, Germany) followed by biotinylated anti-streptavidin (Vector, Burlingame, Calif., USA) and streptavidin-Texas Red.
2 of HCMV was used (Boom et al. 1986). pSS was labeled with digoxigenin-11-dUTP (Boehringer) using a standard nick-translation protocol. The pSS plasmid probe was diluted in hybridization mixture to a final concentration of 5 ng/µl. The plasmid probe hybridization mixture contained 60% deionized formamide, 2×SSC, 50 mM sodium phosphate buffer pH 7.0, 5 ng/ml salmon sperm DNA, and 10% (w/v) dextran sulfate. Before hybridization, the plasmid probe was denatured for 5 min at 100°C and chilled on ice. Sense and anti-sense 5′-aminohexyl-ODN sequences with a GC content of about 40% were synthesized (Eurogentec, Belgium; for nucleotide sequence and positions, see Table 1). They were either labeled with digoxigenin-11-dUTP (Boehringer Mannheim) using a terminal deoxynucleotidyl transferase labeling kit (Promega Madison, WI, USA) according to the manual or conjugated to HRP (Pierce Rockford, IL, USA), using bi-functional cross-linkers (Van Gijlswijk et al. 1996b). HRP-ODNs were purified by ion-exchange chromatography using the Pharmacia Smart FPLC system equipped with a Mono Q PC 1.6/5 column and a 0.3–0.7 M NaCl gradient in 10 mM TRIS-HCl, 1 mM EDTA pH 8.0. Purified HRP-ODN probes were stored at 4°C. The digoxigenin-labeled ODN probes were diluted to a final probe concentration of 100 ng/ml in hybridization mixture containing 40% deionized formamide, 2×SSC, 10% dextran sulfate, and 2× Denhardt’s solution. Just prior to hybridization, HRPODNs were diluted in the same hybridization mixture to a final concentration of 100 ng/ml. The specificity of the HRP- and digoxigenin-labeled ODN probes was verified by hybridizing them to Southern blots of 1-kb PCR fragments generated from the pSS plasmid and various heterologous PCR products. After hybridization, the hybrids were detected using the DAB/Ni reaction (Van Gijlswijk et al. 1996a). Detection was performed directly for HRP-ODN and indirectly with sheep anti-digoxigeninHRP (Boehringer Mannheim; 1:1000) for digoxigenin-labeled ODN. In situ hybridization
For control hybridization, a genomic plasmid probe, pSS, containing a 5.0-kb SphI-SalI fragment coding region for IE regions 1 and
In general, DNA FISH procedures were according to Wiegant et al. (1991) and the RNA FISH procedure was according to Dirks et al. (1993) with minor modifications. Briefly, prior to ODN in situ hybridization, metaphase spreads were treated with 100 µg/ml RNase A in 2×SSC for 30 min at 37°C followed by 3×5-min 2×SSC washes. Next, slides were pretreated with 0.005% (w/v) pepsin (Sigma) in 10 mM HCl, for 10 min at 37°C. The slides were washed for 2×5 min in PBS followed by postfixation with 1% (w/v) formaldehyde (in PBS) for 10 min at room temperature. Slides were washed with PBS for 2×5 min to remove fixative, dehydrated in 70%, 90%, and 100% ethanol, and air dried. For HCMV-IE mRNA detection, rat 9G cells were pretreated with 0.1% (w/v) pepsin pH 2.0 for 1 min at 37°C followed by two quick washes with PBS. Cells were dehydrated in 70%, 90%, and 100% ethanol and air dried. For detecting HCMV integration sites on metaphase spreads and mRNA in rat 9G cells, the targets were denatured prior to hy-
Table 1 Nucleotide sequence and GC content of digoxigenin- and horseradish peroxidase-labeled oligodeoxynucleotide probes used for in situ hybridization studies. The human cytomegaloviros immediate early (HCMV-IE) gene nucleotide sequence numbering is
according to Akrigg et al. (1985). All oligonucleotide sequences are positioned on exon 4 and have a 40% GC content with the exception of CMVas3, which has a 45% GC content. (CMV Cytomegalovirus, ss sense strand, as anti-sense strand, bp base pair)&/tbl.c:&
Metaphase spreads Metaphase spreads of rat 9G cells were prepared according to routine procedures (Wiegant et al. 1991) and stored at –20°C until further use. Probes
Probe
Position (bp)
Nucleotide sequence
% GC
CMVss1 CMVas1 CMVas2 CMVas3 CMVas4 CMVas5 CMVas6
1691–1730 1691–1730 2075–2114 2514–2553 1587–1626 1879–1918 1972–2011
5′-ctaactatgcagagcatgtatgagaactacattgtacctg-3′ 5′-caggtacaatgtagttctcatacatgctctgcatagttag-3′ 5′-ccttgtactcattacacattgtttccacacatgtagtgag-3′ 5′-tcttcctcatcactctgctcactttcttcctgatcactgt-3′ 5′-aaacatcctcccatcatattaaaggcgccagtgaatttct-3′ 5′-cttagggaaggctgagttcttggtaaagaactctatattc-3′ 5′-aaatattttctgggcataagccataatctcatcaggggag-3′
40 40 40 45 40 40 40
&/tbl.:
433 Table 2 The various labels and detection systems used for the detection of the HCVM-IE DNA and mRNA in rat 9G cells. (Dig Digoxigenin, DNP dinitrophenol, TSA tyramide signal amplifica-
tion, HRP horseradish peroxidase, CMV cytomegalovirus, as antisense, ss sense)&/tbl.c:&
Probe
Label/method of labeling
Detection scheme
pSS plasmid
Dig/nick translation
Conventional: TSA:
Dilution mouse anti-dig–fluorescein (Sigma) rabbit anti-mouse–fluorescein (Sigma) sheep anti-dig–HRP (Boehringer) fluorescein–TSA direct (NEN Life Science)
1:250 1:500 1:500 1:1000
CMVas 1–6 CMVss1
Dig/terminal deoxy nucleotidyl transferase (3′ end labeling)
Conventional:
mouse anti-dig–fluorescein (Sigma) rabbit anti-mouse–fluorescein (Sigma) One TSA round: sheep anti-dig–HRP (Boehringer) DNP-TSA plus (NEN Life Science) rat anti-DNP–rhodamine red (home-made)
1:250 1:500 1:500 1:250 1:250
CMVas 1–6 CMVss 1
HRP/chemical conjugation
One TSA round: DNP-TSA plus (NEN Life Science) rat anti-DNP–rhodamine red (home-made) Two TSA rounds: DNP-TSA plus (NEN Life Science) rat anti-DNP–HRP (home-made) fluorescein–TSA direct (NEN Life Science) rhodamine–TSA direct (NEN Life Science)
1:250 1:250 1:250 1:250 1:1000 1:5000
&/tbl.: bridization. Denaturation mix containing 70% formamide and 2×SSC pH 7.0 was applied to the slide and covered with a 24×60mm coverslip. Targets were denatured on a hot plate at 80°C for 2.5 min. Slides were washed in ice-cold 2×SSC to remove the coverslips followed by a 5-min wash in 70% ethanol at –20°C. Slides were dehydrated with 90% and 100% ethanol at room temperature and air dried. Next, 10 µl of probe hybridization mixture was applied to the slide and covered with a 18×18-mm coverslip. Hybridization was allowed for 30 min at 37°C in a moist chamber. After hybridization, all slides were washed briefly in 2×SSC at 37°C to remove the coverslips. Slides hybridized with plasmid probes were washed 3×10 min with 60% formamide, 2×SSC pH 7.0 at 37°C. Slides hybridized with ODN probes were washed with 40% formamide, 2×SSC pH 7.0 at 37°C. Next, all slides were washed 2×3 min with 2×SSC at room temperature to remove the formamide and rinsed briefly in TBS (0.1 M TRIS-HCl pH 7.5, 0.15 M NaCl) containing 0.05% Tween-20 (TNT). Detection of in situ-bound digoxigenin and HRP-labeled ODNs An overview of the detection systems used in DNA and RNA FISH is presented in Table 2. We note that in the RNA FISH part of this study, a home-made rhodamine red anti-DNP antibody was used but that in the course of this study other fluorochrome antiDNP antibodies (NEN Life Sciences, Boston, MA, USA) were successfully used as well. For detection, the various antibodies were diluted in TBS containing 0.5% Boehringer Blocking reagent and incubated for 45 min in a moist chamber at 37°C. After each antibody incubation, slides were washed 3×5 min in TNT. Tyramide amplification was performed by pipetting 1 ml of tyramide solution, diluted in the amplification diluent provided with the TSA kit, directly onto the slide. After an incubation of 30 min at room temperature, slides were washed 3×5 min in TNT. The second tyramide reaction was done using either fluoresceintyramide (1:1000) or the rhodamine-TSA direct reagent system (1:5000). After the final incubation, slides were washed 3×5 min with TNT, dehydrated, air dried, and embedded in Vectashield (Vector) containing 40 ng/ml 4,6-diamidino-2-phenylindol.2HCl for nuclear counterstaining. Microscopy and photography Slides were examined with a Leica DM microscope equipped with single bandpass filter for FITC and rhodamine red and 40×, 63×, and 100× oil objectives with a 1.3 numerical aperture. Photo-
graphs were taken with automatic exposure time settings using Scotch 3 M 640 ASA color slide films or were digitally recorded using a KAF1400 CCD camera. Digital fiber FISH image analysis was performed according to Vrolijk et al. (1996).
Results Assessment of HCMV-IE DNA copy number in rat 9G cells The number of HCMV-IE copies integrated into the rat 9G genome was originally estimated by Southern blot analysis to be ten copies (Boom et al. 1986). In order to determine the number of integrated pES plasmids more precisely, we used fiber FISH to DNA fibers released from rat 9G cells (Wiegant et al. 1991; Florijn et al. 1995). Fibers with three fluorescent tracks of approximately 150 kb long, spaced by 40-kb gaps were observed. Each of the 150-kb fluorescent tracks contained 15–20 hybridization signals of the pSS and pBR probes. The number of the integrated pES plasmids was therefore estimated to be 50–60 copies (Fig. 1B). Detection of the HCMV-IE integration site in metaphase spreads of rat 9G cells Single as well as cocktails of two, four, and six HCMVIE HRP-ODN probes were hybridized to metaphase spreads of rat 9G cells (DNA FISH results are summarized in Table 3). When one DNP-TSA plus step followed by anti-DNP-rhodamine red detection was used, a cocktail of at least four HRP-ODN probes was needed to visualize the HCMV-IE integration site through the fluorescence microscope, but it was difficult to image by conventional microphotography (see Fig. 1C for a digital image acquired with a cooled integrating CCD camera). When two tyramide detection layers were used (i.e.,
434
DNP-TSA plus followed by anti-DNP-HRP incubation and fluorescein-TSA detection), a single HRP-ODN probe was sufficient to detect the HCMV-IE integration site (Fig. 1D). Hybridization with a cocktail of six HRPODN probes and two-layer TSA detection resulted in an increase of signal intensity and still had good resolution. No leaking of signals to neighboring chromosomes or loss of positional information was observed. In these experiments it was noticed that one of the six different HRP-ODNs (CMVas3 in Table 1) gave rise to background spots on all chromosomes. This CMVas3 probe was the HRP-ODN probe in this study with the highest GC content (45%). It could be that non-specific hybridization of HRP-ODN probes may be related to their GC content. In some of the metaphases hybridized with a cocktail of six HRP-ODNs, the amount of deposited fluorescein-tyramide was so high that in the middle of the specific signal, quenching of fluorescence occurred. Detection of HCMV-IE mRNA in rat 9G cells Cycloheximide-treated rat 9G cells have an S phase-dependent HCMV-IE RNA expression, resulting in HCMVIE transcription in about 30% of the cell population. Generally, cells are thermally denatured in our RNA FISH protocols. This implies that with anti-sense ODNs, both cytoplasmic and nuclear RNA as well as nuclear DNA are targeted, while with sense ODNs only the nuclear DNA is potentially detected. After hybridization of one anti-sense HRP-labeled ODN, the hybrids were detected using a DNP-TSA plus layer followed by a rat anti-DNP-rhodamine red detection. Strong cytoplasmic signals were observed in about 30% of the cell population. The expressing cells also showed intense nuclear signals due to the presence of many nuclear transcripts at the HCMV-IE gene cluster (Fig. 2A; RNA FISH results are summarized in Table 4). In the non-expressing cells, small nuclear spots of the HCMV-IE integration site could be observed. Application of the double TSA detection system on cells hybridized with one anti-sense HRP-ODN resulted in very intense fluorescent signals of the cytoplasmic mRNA. The fluorescent mRNA signals of very abundantly expressing cells were so intense that the fluorescent signal appeared to spill to the surrounding glass surface. Actively transcribing cells also showed intense fluorescent nuclear spots. Following 3′ end digoxigenin labeling of the HCMVIE-specific ODNs and conventional immunological detection (mouse anti-digoxigenin followed by rabbit antimouse-fluorescein), a cocktail of six anti-sense ODNs was needed to obtain RNA FISH signals. However, the number of mRNA-positive cells detected was low compared to the FISH results obtained with a HRP-ODN. Only in very actively transcribing cells could a faint nuclear spot and cytoplasmic mRNA be detected (Fig. 2B). When, however, digoxigenin-labeled ODNs were visualized using the TSA strategy, one digoxigenin-ODN was
Fig. 1A–D Two-color fluorescence in situ hybridization (FISH) of DNA fibers extracted from rat 9G cells cultured on glass slides. The pES plasmid is stably integrated into the rat 9G genome and is located on the end of a long chromosome. The DNA fibers of rat 9G cells were hybridized with biotinylated pBR probe and digoxigenin-labeled pSS probe which were subsequently detected with Texas Red and fluorescein, respectively. A The localization of the probes on the integrated pES plasmids. Analysis of the FISH signals consistently showed three fluorescent hybridization tracks of approximately 150 kb on a single DNA fiber. The gaps between the three tracks cover a distance of approximately 40 kb (B, bottom). Each of these tracks consists of 15–20 alternating hybridization signals of the pBR probe in red fluorescence and hybridization signals of the pSS probe in green fluorescence (B, middle and top). The copy number of the integrated pES plasmids in rat 9G cells was estimated to be 50–60. Metaphase spreads of rat 9G cells hybridized with a cocktail of six horseradish perioxidase-labeled oligodeoxynucleotides (HRP-ODNs) detected with DNP-tyramide followed by an incubation with rat anti-DNP-rhodamine red. The human cytomegalovirus immediate early gene (HCMV-IE) integration site appeared at the end of a long chromosome as two small fluorescent spots (C, CCD image). When signals were enhanced with a second tyramide layer, integration sites could be clearly distinguished when only a single HRP-ODN probe (either CMVas1, 2, 4, 5 or 6) was used for hybridization (D). Bars 7 µm&ig.c:/f
Fig. 2A–C Detection of HCMV-IE DNA and mRNA in cycloheximide-treated rat 9G cells. Hybridization of one HRP-ODN followed by one round of tyramide signal amplification (TSA; DNPtyramide plus subsequently detected by rat anti-DNP-rhodamine red) resulted in intense cytoplasmic staining of about 30% of the cell population. Furthermore, integration sites of the HCMV-IE constructs were visible as a fluorescent spot in each cell nucleus (A). When a cocktail of digoxigenin-labeled ODN probes was used for hybridization and subsequently detected by conventional immunological staining, the specific signals were low in intensity (B). When the digoxigenin-labeled ODN probes were detected using one TSA layer, the intensity of the specific signals increased but some background signals were observed (C). Bars 10 µm&ig.c:/f Fig. 3A–E HCMV-IE mRNA detection of cycloheximide-treated rat 9G cells. When the rat 9G cells were hybridized with digoxigenin-labeled pSS plasmid probe and subsequently detected using conventional immunocytochemistry, the HCMV-IE integration sites were visible in each nucleus as a bright fluorescent spot, while mRNA fluorescent signals were present in the cytoplasm of about 30% of the cells (A). After hybridization with the pSS-digoxigenin probe followed by one TSA layer, the mRNA signals in the cytoplasm and nucleus had increased in fluorescence intensity compared to conventional detection methods. Furthermore, some background staining was observed (B). Double hybridization of pSS plasmid conventionally detected (green fluorescence, C) and a single HRP-ODN detected by one DNP-tyramide and rat antiDNP-rhodamine red (red fluorescence, D) revealed co-localization of the two different probes (double exposure, E). Bars A, B 10 µm, C–E 3 µm&ig.c:/f
436 Table 3 Detection of the HCMV-IE integration site in metaphase spreads of rat 9G cells. (HRP Horseradish peroxidase, ODN oligodeoxynucleotide, DNP-TSA plus dinitrophenol-tyramide signal amplification reagent system, Q! quenching of fluorescent signal)&/tbl.c:& Probe
Detection system
Specific Backsignal ground
One HRP–ODN Cocktail of two Cocktail of four Cocktail of six One HRP–ODN Cocktail of two Cocktail of four Cocktail of six
DNP-TSA plus – Rat anti-DNP–rhodamine red ± + + DNP-TSA plus ++ Rat anti-DNP–HRP ++ Fluorescein–TSA direct +++ ++++ Q!
– – – – – – – +
Discussion
&/tbl.:
sufficient to obtain cytoplasmic and nuclear signals (Fig. 2C). RNA FISH with control (sense or non-specific) ODNs, labeled either with HRP or digoxigenin, resulted in low background signals in all cells, which were similar in number and intensity to the non-expressing cells seen in the specific hybridizations. Using digoxigenin-labeled ODNs, background signals tended to be somewhat higher than with HRP-labeled ODN hybridizations. RNA FISH experiments were also conducted with the digoxigenin-labeled pSS plasmid probe (Boom et al. 1986). After hybridization, HCMV-IE mRNA was detected using conventional immunology (Fig. 3A) as well as single-step TSA detection (Fig. 3B). Conventional detection of the in situ hybridized digoxigenin-labeled pSS probe resulted in mRNA signals that were less intense in fluorescence than those obtained when rat 9G cells were hybridized with one anti-sense HRP-ODN and detected with one direct TSA layer (compare Figs. 2A and 3A). Detection of in situ hybridized digoxigenin-labeled pSS probe with one DNP-TSA plus layer led to intense cytoplasmic signals in about 30% of the cell population, but also a slight increase in background signals was observed. In all cells, nuclear spots of the HCMV-IE integration site could be detected (Fig. 3B). Double hybridization experiments were performed in which first one anti-sense cytomegalovirus-specific Table 4 A semi-quantitative summary of fluorescence in situ hybridization results obtained with different probes, various labels and detection systems for the detection of HCMV-IE DNA and mRNA in cycloheximide-treated rat 9G cells. (dig Digoxigenin, HRP horseradish peroxidase, ODN oligodeoxynucleotide, Q! quenching of fluorescent signal)&/tbl.c:&
HRP-ODN was hybridized and detected with the TSA system (Fig. 3D), followed by a hybridization with a digoxigenin-labeled pSS probe which was conventionally detected (Fig. 3C). These double hybridization experiments show exact co-localization and proved the specificity and superiority of the new HRP-ODN/TSA approach over standardized conventional RNA FISH methods, in that, with one ODN similar or even better signal intensities were obtained than when large cDNA probes were used for hybridization.
For RNA FISH the use of synthetic ODNs as probes is an obvious choice because of their single strandedness, small size, specificity, and easy synthesis. However, the number of hapten or fluorochrome labels that can be coupled to ODNs is restricted and thus they have limited detection capabilities. Therefore, cDNA or cRNA probes are often favored over hapten- or fluorochrome-labeled ODNs in RNA FISH studies. As shown in immunohistochemical and FISH studies, the HRP moiety of HRP-conjugated antibodies or HRPlabeled ODNs can deposit many hapten- or fluorochrome-labeled tyramides at the hybridization site (Kersten et al. 1995; De Haas et al. 1996; Hunyady et al. 1996; van Gijlswijk et al. 1996b). The use of HRP-labeled ODN probes for highly repetitive sequences in DNA FISH showed intense fluorescent signals upon hybridization and these signals were as good as, and often better than, those obtained by conventional, indirect immunofluorescence detection of plasmid probes (Van Gijlswijk et al. 1996b). Thus, hapten- or fluorochrome-tyramide deposition can compensate for the limited sensitivity inherent in the use of ODN probes and may have potential for RNA FISH studies as well. In a recent study, Shonhuber et al. (1997) showed improved detection of ribosomal RNA in prokaryotic cells using specific HRP-ODNs and TSA detection. Here, we report the first use of the HRP-ODN/TSA approach for mRNA detection in eukaryotic cells. Hybridization of
Label
Detection method
Integration site
Cytoplasmic mRNA
Background
pSS plasmid One sense ODN One anti-sense ODN Cocktail of six ODNs
dig
Conventional
++ ± ± +
+++ – – +
– – – –
pSS plasmid One senses ODN One anti-sense ODN Cocktail of six ODNs
dig
One tyramide layer
+++ + ++ ++++
++++ – +++ +++++
+ + + +++
One sense ODN One anti-sense ODN Cocktail of six ODNs
HRP
One tyramide layer
++ +++ ++++++
– +++++ +++++++
– – +
One sense ODN One anti-sense ODN
HRP
Two tyramide layers
+++ +++++++ Q!
– +++++++
+ ++
&/tbl.:
Probes
437
HRP-ODNs to rat 9G cells resulted in intense nuclear and cytoplasmic HCMV-IE mRNA signals, which were more intense in fluorescence than those obtained with large cDNA probes conventionally detected (see Table 4 for summary and, e.g., Fig. 3C–E). In addition, the specificity and sensitivity of the HRPODN/TSA approach has been illustrated by DNA FISH to the HCMV-IE integration site in metaphase chromosomes of rat 9G cells. With a single 40 mer HRP-ODN probe, the 50–60 copies of the integrated tandem repeat of the 7.0-kb pES fragment could be easily visualized. This result shows that a cluster of maximally 50–60 HRP molecules can be readily detected with TSA. The conclusion from this work is that the HRPODN/TSA approach provides the methodological basis for low abundance mRNA detection by FISH because the probe-related background is profoundly reduced and the TSA system is sensitive enough for the detection of hybridized ODN probes. We note, however, that in rat 9G cells, HCMV-IE mRNA is abundantly expressed and that further studies to less abundantly expressed mRNAs will have to be conducted to validate this conclusion. Such studies are currently in progress. &p.2:Acknowledgements This study was supported by the Dutch Science Organization, Area Medical sciences (MW-NWO) project 900–543–109, and by NEN Life Science Products, Boston, USA.
References Akrigg A, Wilkinson GWG, Oram JD (1985) The structure of the major immediate early gene of human cytomegalovirus. Virus Res 2:107–121 Boom R, Geelen JL, Sol CJ, Raap AK, Minnaar RP, Klaver BP, Van der Noordaa J (1986) Establishment of a rat cell line inducible for the expression of human cytomegalovirus immediate-early gene products by protein synthesis inhibitors. J Virol 58:851–859 De Haas R, Verwoerd NP, Van der Corput MPC, Van Gijlswijk RPM, Siitari H, Tanke HJ (1996) The use of peroxidase-mediated deposition of biotin-tyramide in combination with timeresolved fluorescence imaging of europium chelate label in immunohistochemistry and in situ hybridization. J Histochem Cytochem 44:1091–1099 Dirks RW (1996) RNA molecules lighting up under the microscope. Histochem Cell Biol 106:151–166
Dirks RW, Rijke FM van de, Fujishita S, Ploeg M van der, Raap AK (1993) Methodologies for specific intron and exon localization in cultured cells by haptenized and fluorochromized probes. J Cell Sci 104:1187–1197 Florijn RJ, Bonden LAJ, Vrolijk H, Wiegant J, Vaandrager J-W, Baas F, Den Dunnen JT, Tanke HJ, Van Ommen G-JB, Raap AK (1995) High-resolution DNA fiber-FISH for genomic DNA mapping and colour bar-coding of large genes. Hum Mol Genet 4:831–836 Hunyady B, Krempels K, Harta G, Mezey É (1996) Immunohistochemical signal amplification by catalyzed reporter deposition and its application in double immunostaining. J Histochem Cytochem 44:1353–1362 Kerstens HMJ, Poddighe PJ, Hanselaar AGJM (1995) A novel in situ hybridization signal amplification method based on the deposition of biotinylated tyramine. J Histochem Cytochem 43:347–350 Raap AK, Van de Corput MPC, Vervenne RAW, Van Gijlswijk RPM, Tanke HJ, Wiegant J (1995) Ultra-sensitive FISH using peroxidase-mediated deposition of biotin- or fluorochrome-labeled tyramides. Hum Mol Genet 4:529–534 Shonhuber W, Fuchs B, Juretschko S, Amann R (1997) Improved sensitivity of whole-cell hybridization by the combination of horseradish peroxidase-labeled oligonucleotides and tyramide signal amplification. Appl Environ Microbiol 63: 3268–3273 Van Gijlswijk RPM, Gijlswijk-Janssen DJ van, Raap AK, Daha MR, Tanke HJ (1996a) Enzyme-labelled antibody-avidin conjugates: new flexible and sensitive immunochemical reagents. J Immunol Methods 189:117–127 Van Gijlswijk RPM, Wiegant J, Vervenne R, Lasan R, Tanke HJ, Raap AK (1996b) Horseradish peroxidase-labeled oligonucleotides and fluorescent tyramides for rapid detection of chromosome-specific repeat sequences. Cytogene Cell Genet 75:258–262 Van Gijlswijk RPM, Zijlmans JMAA, Wiegant J, Bobrow MN, Erickson TJ, Adler KE, Tanke HJ, Raap AK (1997) Fluorochrome-labeled tyramides: use in immunocytochemistry and fluorescence in situ hybridization. J Histochem Cytochem 45:375–382 Van Heusden J, De Jong P, Ramakers F, Bruwiere H, Borgers M, Smets G (1997) Fluorescein-labeled tyramide strongly enhances the detection of low bromodeoxyuridine incorporation levels. J Histochem Cytochem 45:315–319 Vrolijk H, Florijn RJ, Van de Rijke FM, Van Ommen G-J B, Den Dunnen JT, Raap AK, Tanke HJ (1996) Microscopy and image analysis of fibre-FISH. Bioimaging 4:73–83 Wiegant J, Galjart NJ, Raap AK, D’Azzo A (1991) The gene encoding human protective protein (PPGB) is on chromosome 20. Genomics 10:345–349 Wiegant J, Kalle W, Mullenders L, Brookes S, Hoovers, JMN, Dauwerse JG, Van Ommen G-JB, Raap AK (1992) High-resolution in situ hybridization using DNA halo preparations. Hum Mol Genet 1:587–591
