Single-copy Gene Detection Using Branched DNA (bDNA) In Situ Hybridization

Abstract

We have developed a branched DNA in situ hybridization (bDNA ISH) method for detection of human papillomavirus (HPV) DNA in whole cells. Using human cervical cancer cell lines with known copies of HPV DNA, we show that the bDNA ISH method is highly sensitive, detecting as few as one or two copies of HPV DNA per cell. By modifying sample pretreatment, viral mRNA or DNA sequences can be detected using the same set of oligonucleotide probes. In experiments performed on mixed populations of cells, the bDNA ISH method is highly specific and can distinguish cells with HPV-16 from cells with HPV-18 DNA. Furthermore, we demonstrate that the bDNA ISH method provides precise localization, yielding positive signals retained within the subcellular compartments in which the target nucleic acid sequences are localized. As an effective and convenient means for nucleic acid detection, the bDNA ISH method is applicable to the detection of cancers and infectious agents.

(J Histochem Cytochem 49:603–611, 2001)

Keywords

branched DNA (bDNA) signal amplification in situ hybridization (ISH)cervical cancer cell lines human papillomavirus (HPV)

In situ hybridization (ISH) is a valuable tool for detection of specific nucleic acid sequences in morphologically intact cells or tissues. Since the introduction of ISH by Pardue and Gall (1969), there have been a number of advances in ISH methodology. Although the use of radioisotopic labels has long served as a standard for ISH (Angerer et al. 1987; Simmons et al. 1989), the use of non-isotopic probes has eliminated some of the problems inherent in radioactive ISH, such as long turn-around times, risk for exposure to radioactivity, and waste disposal (Burns et al. 1987; Nuovo and Richart 1989; Furuta et al. 1990; Herrington et al. 1991; Holm et al. 1992; Wiedorn et al. 1999). In addition, the sensitivity of ISH has been greatly improved by the application of PCR-based target amplification and catalyzed signal amplification (CSA) methods (Bobrow et al. 1992; Patterson et al. 1993; Komminoth and Werner 1997; Schmidt et al. 1997; Baunoch et al. 1999).

Branched DNA (bDNA) signal amplification technology has been used extensively in a microwell format to detect and quantify specific nucleic acid sequences. Inherently quantitative and highly reproducible, bDNA technology utilizes nonradioactive synthetic oligonucleotide probes and can readily be applied to the detection of any nucleic acid target for which a sequence is known. A number of bDNA assays have been developed for quantification of viral nucleic acids, including human immunodeficiency virus type 1 (HIV-1) RNA (Kern et al. 1996), simian immunodeficiency virus (SIV) RNA (Sodora et al. 1998), hepatitis B virus (HBV) DNA (Hendricks et al. 1995), hepatitis C virus (HCV) RNA (Detmer et al. 1996), hepatitis G virus (HGV) RNA (Brandhagen et al. 1999), and cytomegalovirus (CMV) DNA (Chernoff et al. 1997; Pellegrin et al. 2000). More recently, bDNA technology has been used to detect and measure the expression of cellular mRNAs, including cytokines (Breen et al. 1997; Shen et al. 1998), progesterone and estrogen receptors (Nargessi et al. 1998a,b), insulin (Wang et al. 1997), glucokinase (Cabrera–Valladares et al. 1999), c-fos (Shyamala et al. 1999), aP2 (Burris et al. 1999), cytochrome P450 (Hartley and Klaassen 2000), and uncoupling proteins (Zhou et al. 2000). Although these bDNA assays were developed to measure nucleic acids in serum, plasma, or cell lysates, other studies have shown that it is possible to adapt bDNA technology to an ISH format for detection of mRNA (Cao et al. 1998; Antao et al. 2000).

Here we report the experimental results obtained after development of a bDNA ISH method for the detection of human papillomavirus (HPV) DNA and mRNA in whole cells. The bDNA ISH method utilizes a series of non-isotopic oligodeoxyribonucleotide probes, hybridized sequentially, to generate chromogenic and fluorescent signals. By modifying conditions of cell pretreatment, either mRNA or DNA can be detected with the same set of probes. Using human cervical cancer cell lines that are well characterized with regard to HPV genotype and number of integrated HPV genomes, we evaluate the sensitivity and specificity of the bDNA ISH method. Furthermore, we assess whether signals generated by the bDNA ISH method are retained in the subcellular compartments in which the target nucleic acid sequences are localized.

Materials and Methods

Cell Culture

Human cervical cancer cell lines were obtained from the American Type Cell Culture Collection (ATCC; Manassas, VA). Cell lines positive for HPV-16 or HPV-18 included HeLa, CaSki, and SiHa, which are well characterized with regard to genotype and number of integrated HPV genomes (Yee et al. 1985; Baker et al. 1987; Mincheva et al. 1987; Siadat-Pajouh et al. 1994; Plummer et al. 1998; Meissner 1999). Controls included the HPV-39-positive ME180 and the HPV-negative C33A and HT3 cell lines. Cells were grown either in flasks or on chamber slides coated with 10 μg/ml poly-l-lysine using conditions provided by the ATCC.

Oligonucleotide Probes

A total of 26 HPV-16-specific and 32 HPV-18-specific DNA degenerate oligonucleotide target probes were designed to cover 90% of the E6 and E7 regions of the HPV genome. These probes were generated using the ProbeDesigner Software (Bayer Diagnostics; Emeryville, CA) (Bushnell et al. 1999) with a constant melting temperature of 63 ± 2C. Final probes were selected after screening for possible interactions with 39 other HPV genotypes as well as human genomic DNA sequences by using HybSimulator software (Advanced Gene Computing Technologies; Irvine, CA) and Blast2 software (National Center for Biotechnology Information; National Library of Medicine, NIH, Bethesda, MD). Other DNA oligonucleotide probes, which are part of the signal amplification system and include preamplifier, amplifier, and alkaline phosphatase (AP)-conjugated label probes, have been described in detail (Collins et al. 1997). To reduce potential nonspecific hybridization, non-natural nucleotides 5-methyl-2′-deoxyisocytidine (isoC) and 2′-deoxyisoguanosine (isoG) were included in the target, preamplifier, amplifier, and AP-conjugated label probes. Although non-natural nucleotides have recently been incorporated into bDNA technology, specific and sensitive detection of target nucleic acid sequences is routinely achieved using probes without non-natural nucleotides (Urdea and Wuestehube 2000).

bDNA ISH for DNA Detection

Harvested cells were fixed with 4% formaldehyde in PBS (0.01 M phosphate buffer, pH 7.5) for 30 min on ice, pipetted into double-spotted cytospin funnels (Shandon; Pittsburgh, PA), and centrifuged for 6 min at 1500 rpm in a cytospin centrifuge. The cells were dehydrated on slides through a graded ethanol series, air-dried at RT for 10 min, and stored at −80C for up to 2 weeks. For pretreatment, cells were rehydrated on slides through a graded ethanol series and washed in PBS. Slides were incubated at 37C for 1 hr in 40 μg/ml RNase in 2 × SSC (1 × SSC is 0.15 M NaCl, 0.015 M Na-citrate), and washed in PBS, before incubation in 7.5–10 μg/ml proteinase K in PBS at 37C for 10 min. Cells were then dehydrated and air-dried at RT for 10 min.

After pretreatment, cells were incubated with a prehybridization solution, hybridized with a set of oligonucleotide target probes (described above), and then hybridized with a series of oligonucleotide probes for signal amplification (Figure 1). As a prehybridization step, each spot on the slide was incubated with 150 μl hybridization Solution A [3 × SSC, 50% formamide, 10% dextran sulfate (MW 500,000), 0.2% casein, 10 μg/ml poly A, and 100 μg/ml denatured salmon sperm DNA] at RT for 30 min. Cells were denatured in a humidity chamber (Hybaid Omnislide instrument; Phenix Research Products, Hayward, CA) at 92C for 10 min and then cooled at RT for 5 min. Cells were incubated in 100 μl of hybridization Solution A containing 0.6 pmole HPV-specific target probes at 40C for 3 hr in a humidified chamber. Slides were washed at RT with a decreasing series of SSC buffers containing 0.0025% Brij-35 detergent (Surfact-Amps 35; Pierce, Rockford, IL) for 1–2 min each in 2 × SSC, 0.2 × SSC, 0.1 × SSC, and 2 × SSC. Cells were then incubated with 100 μl hybridization Solution B (5 × SSC, 0.1–0.3% SDS, 10% dextran sulfate, 1 mM ZnCl₂, and 10 mM MgCl₂) containing 90 fmoles preamplifier at 55C for 25 min in the humidity chamber. Slides were washed twice in 0.1 × SSC, 1 mM EDTA for 1 min and 4 min, respectively, and then incubated in 100 μl hybridization Solution B containing 90 fmoles amplifier at 55C for 25 min in the humidity chamber. Slides were washed twice in 0.1 × SSC, 1 mM EDTA for 1 min and 4 min, respectively, and then incubated in 100 μl hybridization Solution B containing 90 fmoles AP-conjugated label probe at 55C for 15 min in the humidity chamber. After washing in 100 mM Tris, pH 8.0, containing 0.1% Brij-35, 1 mM ZnCl₂, and 10 mM MgCl₂ at RT for 5 min, cells were incubated in 50 μl buffered AP substrate (Fast Red, #K597; DAKO, Carpinteria, CA) at RT for 10 min. Slides were counterstained with either Gills hematoxylin or 0.0001% bisbenzimide. Slides were mounted with Ultramount (DAKO), Permount, or 75% glycerol and stored at RT. Slides were viewed using a Nikon E800 fluorescence microscope with an FITC or triple bandpass filter or a ×60 brightfield objective, and images were captured using an Optronics cooled 3-chip color CCD camera. Fluorescent or chromogenic images were captured using Image Probe Software (Media Cybernetics; Silver Springs, MD). Figures were generated using Adobe Photoshop 3.0 and printed using a Sony Digital Printer.

Figure 1.

Schematic depiction of the bDNA ISH method.

For the cell mixing experiments, HeLa cells were labeled with the fluorescent cell tracer carboxyfluorescein diacetate, succinimidyl ester (CFDA SE; Molecular Probes, Eugene, OR) at 37C for 15 min in PBS containing 5 μM CFDA SE. After washing away excess CFDA SE, the cells were incubated at 37C for an additional 30 min in PBS. Cells were harvested, fixed in 4% formaldehyde in PBS, mixed with unlabeled formaldehyde-fixed CaSki cells, and then cell mixtures were spun onto cytospin slides and assessed for DNA targets, as noted above.

bDNA ISH for RNA Detection

Cells grown on chamber slides were fixed with 4% formaldehyde in PBS for 30 min at RT, treated with 10 μg/ml proteinase K in PBS for 10 min at RT, and then washed twice for 5 min in PBS. Samples were incubated at 40C for 3 hr with 1 pmole HPV-specific target probes in target probe buffer (6 × SSC, 25% formamide, 0.2% Brij-35, and 0.2% casein) and then washed with the same decreasing series of SSC buffers as was used in the bDNA ISH protocol for DNA detection. Similarly, the preamplifier, amplifier, and AP-conjugated probe hybridization and wash conditions were the same as that for the bDNA ISH protocol for DNA detection. AP substrate was added and slides were incubated at RT for 4 min. After stopping the reaction by washing in PBS, samples were postfixed in 4% formaldehyde in PBS for 5 min at RT and then counterstained for 40 sec with either hematoxylin or bisbenzimide. Slides were viewed and images were generated and printed as described above.

Results

Sensitivity and Specificity of bDNA ISH for HPV DNA Detection

The bDNA ISH method is illustrated in Figure 1. After fixation and permeabilization of cells, the target DNA or RNA is hybridized to a series of synthetic oligonucleotide probes. First, a set of target probes is hybridized to the target nucleic acid molecules in the cell. Preamplifier molecules are hybridized to the target probes, providing a bridge for the hybridization of amplifier molecules. Amplification of the signal is accomplished by the binding of up to 14 amplifier molecules to each preamplifier and of up to 196 AP-conjugated label probes to each target probe. Addition of Fast Red AP substrate results in the deposition of a red reaction product in the vicinity of the target nucleic acid that can be visualized with standard optical light or fluorescent microscopy.

The results shown in Figure 2 illustrate the effectiveness of the bDNA ISH method for detection of viral DNA and gene expression. The HPV-16 genome was detected in CaSki cells (Figure 2A) and SiHa cells (Figure 2B), which contain 400–600 and 1–2 copies of HPV-16 DNA, respectively. No signal was detected with HPV-16 probes in cells lacking HPV-16 DNA, including HeLa (Figure 2C) and C33a cells (Figure 2D). On hybridization with HPV-18 target probes, positive signal detection was observed in HeLa cells (Figure 2G), which contain 10–50 copies of HPV-18 DNA. No signal was detected with HPV-18 probes in cells lacking HPV-18 DNA, including CaSki (Figure 2E), SiHa (Figure 2F), and C33a cells (Figure 2H). No signal was detected with either HPV-16 or HPV-18 target probes in the HPV-negative HT3 cell line or the ME180 cell line, which harbors a DNA sequence similar to that of HPV-39 (not shown). These results show that one to two copies of HPV-16 DNA are detected in SiHa cells using the bDNA ISH method. Furthermore, the target probes used for bDNA ISH can distinguish between HPV-16 and HPV-18 genomic sequences. Although in some experiments 100% of cells were positive for HPV-16 DNA (CaSki and SiHa) or HPV-18 DNA (HeLa) detection, 70–90% of HPV-positive cells per slide is a conservative estimate of what was observed on a routine basis.

Figure 2

Sensitivity of bDNA ISH for detection of HPV DNA in human cervical carcinoma cell lines. Hybridization of HPV-16 probes results in deposition of Fast Red–AP reaction product in CaSki cells containing 400–600 copies HPV-16 DNA per cell (A) and SiHa cells containing 1–2 copies HPV-16 DNA per cell (B). No reaction product is detected with HPV-16 probes in cells lacking HPV-16 DNA, including HeLa cells (C) and C33a cells (D). Hybridization of HPV-18 probes results in deposition of Fast Red–AP reaction product in HeLa cells containing 10–50 copies HPV-18 DNA (G). No reaction product is detected with HPV-18 probes in cells lacking HPV-18 DNA, including CaSki cells (E), SiHa cells (F) and C33a cells (H). All cells were counterstained with Gills 1 hematoxylin.

A number of additional controls were performed to determine whether the signals observed were specific for HPV DNA targets (not shown). No signal was observed when nonspecific target probes were used or when HPV-16 or HPV-18 target probes, amplifier, or AP-conjugated label probes were omitted from the bDNA ISH method. Omission of the proteinase K digestion or DNA denaturation steps or treatment of cells with DNase also resulted in a loss of signal. As a further control for the DNase digestion experiments, the RNase treatment step was omitted to confirm that this loss in signal was not due to DNase degradation of the DNA oligonucleotide probes. Under these conditions, signal was again detected, indicating that the DNA oligonucleotide probes were not degraded and thus were capable of binding to HPV RNA targets.

Genotype-specific Detection of HPV DNA by bDNA ISH

Mixed cell populations were used to further assess the specificity of the bDNA ISH method for HPV DNA detection. Mixed cell samples were composed of unlabeled CaSki cells (containing 400–600 copies of HPV-16 DNA) and CFDA-labeled HeLa cells (containing 10–50 copies of HPV-18 DNA). As shown in Figure 3, hybridization with HPV-16 target probes yielded signal detection only in HPV-16-infected CaSki cells and not in HeLa cells (Figure 3A and 3C). HeLa cells surrounded by positively stained CaSki cells sometimes appeared to be partly stained, but this likely is an artifact from overlapping CaSki and HeLa cells (indicated by arrow in Figure 3A). Likewise, hybridization with HPV-18 target probes yielded signal detection only in HPV-18-infected HeLa cells and not in CaSki cells (Figures 3B and 3D). Although the vast majority of HeLa cells were well labeled with the CFDA SE cell tracer, a few cells were faintly labeled (Figure 3D, cell at middle bottom). As expected, greater signal intensity (i.e., larger number and size of spots) was observed for HPV-16 DNA detection in CaSki cells compared to HPV-18 DNA detection in HeLa cells. This difference in signal intensity is likely a reflection of the higher HPV DNA copy number present in CaSki cells (400–600 HPV-16 DNA copies/cell) compared to HeLa cells (10–50 HPV-18 DNA copies/cell). These results demonstrate that, in a mixed population of cells, the bDNA ISH method can distinguish cells with HPV-16 DNA from cells with HPV-18 DNA. Because signals are retained within the appropriate cell types, there is no diffusion of the AP–Fast Red reaction product (Speel et al. 1992) from one positive cell type to another negative cell type within the same sample. Moreover, even low-abundance viral targets can be readily detected with the bDNA ISH method without interference from nonspecific hybridization to related viral sequences or other sequences.

Subcellular Co-localization of Signal and Target Nucleic Acids with bDNA ISH

Because subcellular compartments are more easily visualized in cells cultured on chamber slides, HeLa cells grown on chamber slides were used to assess whether the bDNA ISH method could detect the subcellular distribution of HPV mRNA and HPV DNA. As shown in Figure 4, hybridization of HeLa cells (prepared for RNA detection) with HPV-18 target probes resulted in detection of HPV-18 mRNA mainly in the cytoplasm (Figure 4A). Although mRNA detected in the cytosol appears to overlap with cell nuclei when whole cells are viewed through a microscope, this is consistent with the cell nucleus being naturally surrounded by cytosol within cells. No signal was observed in these same cells incubated with HPV-16 target probes (Figure 4B). In contrast, hybridization of HeLa cells (prepared for DNA detection) with HPV-18 target probes resulted in the detection of HPV-18 DNA in HeLa cell nuclei (Figure 4C). No signal was observed in these same cells incubated with HPV-16 target probes (Figure 4D). These results demonstrate that the signal observed with bDNA ISH is retained in the subcellular compartment known to contain the nucleic acid target: viral mRNA is detected in cytosol, whereas viral DNA is detected in cell nuclei. In other words, the target and signal are co-localized with the bDNA ISH method.

Discussion

The bDNA ISH method described here provides a rapid, sensitive, and reproducible means for detection of specific DNA and mRNA sequences in various cell types. Our results show that the sensitivity of the bDNA ISH method is sufficient to detect relatively low-abundance targets, as few as one or two copies of HPV-16 DNA in SiHa cells. Our results also demonstrate that the bDNA ISH method is highly specific and provides subcellular localization of the target sequence. In mixed cell population experiments, signals were detected only in the appropriate cell types, indicating that there was no diffusion of AP–Fast Red reaction product from one cell type to another. Moreover, signals obtained with bDNA ISH are retained within the cellular compartment in which the target sequence is localized. Subcellular localization experiments with HPV-18-infected HeLa cells showed that viral mRNA was predominantly detected in the cytoplasm, whereas viral DNA was detected in cell nuclei.

Figure 3.

Genotype-specific detection of HPV DNA by bDNA ISH in mixed samples of HPV-18-positive HeLa cells and HPV-16-positive CaSki cells. For ready identification of cell types, HeLa cells were labeled with CFDA SE cell tracer. In all images, HeLa cells are indicated by arrows and CaSki cells are indicated by arrowheads. In fluorescent images (C,D), emission of Fast Red AP reaction product is shown in red, nuclear staining of bisBenzimide in blue, and CFDA SE cell tracer that marks HeLa cells in green. (A) Hybridization of HPV-16 probes results in deposition of Fast Red AP reaction product in selected cells. (C) Merge of fluorescent images of cells in A showing that HPV-16 probes hybridize to HPV-16 DNA in non-CFDA SE-labeled CaSki cells but not to HPV-18 DNA in CFDA SE-labeled HeLa cells. (B) Hybridization of HPV-18 probes results in deposition of Fast Red AP reaction product in selected cells. (D) Merge of fluorescent images of cells in B showing that HPV-18 probes hybridize to HPV-18 DNA in CFDA SE-labeled HeLa cells but not to HPV-16 DNA in non-CFDA SE-labeled CaSki cells.

The bDNA ISH method described here works well, with very little modification, when different target probe sets and different cell and tissue types are used. We have reproducibly detected single-copy HIV-1 DNA targets in HIV-positive 8E5 cells and two copies of the heterogeneous nuclear ribonuclear protein gene in CaSki and 8E5 cell lines (unpublished work). Although in the present study we have limited our focus to HPV detection in whole cells, in other experiments we have detected HPV in paraffin-embedded and frozen cervical tissue sections (unpublished work). These results suggest that the bDNA ISH method may be an effective means for nucleic acid detection in a variety of specimen types.

Figure 4

Localization of HPV-18 mRNA predominantly in cytoplasm and HPV-18 DNA in nuclei of HeLa cells by bDNA ISH. (A) Hybridization of HPV-18 probes to HPV-18 mRNA results in deposition of Fast Red AP reaction product predominantly in HeLa cell cytoplasm (arrowhead) but not in HeLa cell nuclei (arrow). (B) Control showing that HPV-16 probes do not hybridize with HPV-18 RNA in HeLa cells. (C) Hybridization of HPV-18 probes to HPV-18 DNA results in deposition of Fast Red AP reaction product in HeLa cell nuclei (arrow) but not in HeLa cell cytoplasm (arrowhead). (D) Control showing that HPV-16 probes do not hybridize with HPV-18 DNA in HeLa cells. For RNA detection, cells were counterstained with Gills 1 hematoxylin.

Another signal amplification system that has been applied to ISH is the catalyzed reporter deposition tyramide signal amplification (CSA) ISH method, which utilizes biotinylated probes, biotinyl tyramide, and streptavidin-conjugated horseradish peroxidase (Schmidt et al. 1997). Our results with the bDNA ISH method agree with other studies that have detected HPV-16 DNA in SiHa cells using CSA (Adler et al. 1997). In our lab, we compared CSA ISH using the GenPoint System (DAKO) with bDNA ISH and found that the two methods have similar sensitivities for detecting HPV-16 DNA in SiHa cells. We found that the HPV-16-specific signal generated with either the bDNA ISH or CSA ISH method was visualized as one or two dots in SiHa cells, and possibly up to four dots in replicating cells. Studies have shown that there is principally one HPV chromosomal integration site in SiHa cells (Baker et al. 1987; Mincheva et al. 1987). It has been suggested that the dots detected probably reflect HPV copies per cell (Siadat–Pajouh et al. 1994).

A couple of features distinguish the bDNA ISH method from CSA ISH. One significant difference is in the probe design. Whereas the HPV-16-specific probes for the bDNA ISH method consist of a set of synthetic oligonucleotides less than 30 bases in length that specifically recognize the E6 and E7 viral genes and transcripts, the probes for the DAKO CSA ISH kit are restriction fragments of a 7-kb biotin-labeled cDNA probe that recognize the E1, E2, E3, E4, E5, E6, and E7 open reading frames (GenPoint System; DAKO Corporation reagent data sheet). The smaller probes generated for bDNA ISH are easier to synthesize and can detect smaller target sequences, thereby providing greater flexibility than longer probes. In principle, the degree of signal amplification can be controlled by adjusting the number of target probes utilized. We have noted that fewer probes are necessary for visualization of HPV in cells with higher HPV copies. Although we did not determine the fewest number of target probes needed for HPV detection with the bDNA ISH method, we found in the course of method development that 15 target probes were sufficient for detection of HPV-16 in SiHa cells (containing one to five HPV copies) and two target probes were sufficient for detection of HPV-16 in CaSki cells (containing 500 HPV copies). Smaller probes also provide more flexibility in obtaining the desired specificity. For example, degenerate target probe sets have been designed for equal detection of several different viral genotypes in the microwell format for HBV, HCV, and HIV (Hendricks et al. 1995; Detmer et al. 1996; Kern et al. 1996). By comparison, as was done for this study, target probe sets can be designed so that detection is genotype-specific. Genotype-specific detection is clinically important for HPV DNA testing because only a subset of HPV genotypes is associated with cervical cancers and high-grade precursor lesions (Manos et al. 1999). Another feature that distinguishes bDNA ISH from CSA ISH is that bDNA ISH does not use an avidin–biotin signal amplification system and hence is not affected by binding of avidin-conjugated reporter molecules to endogenous biotin. By providing greater flexibility in probe design and avoiding potential interference by endogenous biotin, the bDNA ISH method represents a significant advance in low-copy nucleic acid detection.

Compared to PCR-based methods, the bDNA signal amplification system offers a number of advantages for ISH. Because bDNA ISH is based on the sequential hybridization of synthetic DNA probes, it does not require any DNA or RNA polymerase activity and hence is not affected by the presence of polymerase inhibitors in specimens. The potential for diffusion of amplification products away from the target site, which is a concern with PCR-based ISH methods (Nuovo et al. 1991; Wiedorn et al. 1999), also is not a concern with bDNA ISH. Because the AP-conjugated reporter molecule that provides the signal is tethered to the spatially fixed nucleic acid target in the bDNA ISH method, the AP–Fast Red reaction product is retained at the target site. Another advantage is that the bDNA ISH method does not require repeated cycling through elevated temperatures. Because repeated incubation at high temperatures can damage delicate cell morphology, avoiding high-temperature incubations is important for applications in which preservation of intricate cell morphology is important.

In summary, we have developed a bDNA ISH method for sensitive detection of DNA target sequences that overcomes many of the challenges facing ISH techniques today. Based on bDNA signal amplification technology, the bDNA ISH method is highly sensitive (can detect one or two copies of DNA target per cell), specific, and provides subcellular localization of the target sequence. By modifying a few steps in the bDNA ISH procedure, detection of mRNA or DNA targets can be achieved using the same set of DNA oligonucleotide probes. The bDNA ISH method is nonisotopic, rapid (yields results within a day), can be adapted to generate chromogenic and/or fluorescent signals, and should be amenable to automation and quantification. Given its ease of use and reliability, the bDNA ISH method is an attractive alternative for sensitive detection of nucleic acid sequences in a well-preserved morphological context.

Footnotes

Acknowledgements

Acknowledgments

We wish to thank Peter Dailey for helpful discussions about assay design and for review and comments on the manuscript, and Linda Wuestehube for writing and editorial assistance.

References

Adler

Erickson

Bobrow

(1997) High sensitivity detection of HPV-16 in SiHa and CaSki cells utilizing FISH enhanced by TSA. Histochem Cell Biol 108:321–324

Angerer

Stoler

Angerer

(1987) In situ hybridization with RNA probes—an annotated recipe. In Valentino

Eberwine

Barchas

, eds. In Situ Hybridization—Applications to Neurobiology. Oxford, Oxford University Press, 71–96

Antao

Player

Kolberg

(2000) In situ hybridization using the bDNA technology. In Patterson

, ed. Techniques in Quantification and Localization of Gene Expression. Boston, Birkhauser Press, 81–93

Baker

Phelps

Lindgren

Braun

Gonda

Howley

(1987) Structural and transcriptional analysis of human papillomavirus type 16 sequences in cervical carcinoma cell lines. J Virol 61:962–971

Baunoch

Lobell

Kane

Fredericks

Thompson

(1999) The development of an automated in situ assay for the detection of human cytomegalovirus in peripheral blood leukocytes. J Virol Methods 81:31–37

Bobrow

Litt

Shaughnessy

Mayer

Conlon

(1992) The use of catalyzed reporter deposition as a means of signal amplification in a variety of formats. J Immunol Methods 150:145–149

Brandhagen

Gross

Jr Poterucha

Charlton

Detmer

Kolberg

Gossard

Batts

Kim

Germer

Wiesner

Persing

(1999) The clinical significance of simultaneous infection with hepatitis G virus in patients with chronic hepatitis C. Am J Gastroenterol 94:1000–1005

Breen

Salazar–Gonzalez

Shen

Kolberg

Urdea

Martinez–Maza

Fahey

(1997) Circulating CD8 T cells show increased interferon-gamma mRNA expression in HIV infection. Cell Immunol 178:91–98

Burns

Graham

Frank

Fleming

Evans

McGee

(1987) Detection of low copy human papilloma virus DNA and mRNA in routine paraffin sections of cervix by non-isotopic in situ hybridisation. J Clin Pathol 40:858–864

10.

Burris

Pelton

Zhou

Osborne

Cryan

Demarest

(1999) A novel method for analysis of nuclear receptor function at natural promoters: peroxisome proliferator-activated receptor gamma agonist actions on aP2 gene expression detected using branched DNA messenger RNA quantitation. Mol Endocrinol 13:410–417

11.

Bushnell

Budde

Catino

Cole

Derti

Kelso

Collins

Molino

Sheridan

Monahan

Urdea

(1999) ProbeDesigner: for the design of probesets for branched DNA (bDNA) signal amplification assays. Bioinformatics 15:348–355

12.

Cabrera–Valladares

German

Matschinsky

Wang

Fernandez–Mejia

(1999) Effect of retinoic acid on glucokinase activity and gene expression and on insulin secretion in primary cultures of pancreatic islets. Endocrinology 140:3091–3096

13.

Cao

Connolly

Zagala

Beard

Hirsch

Kolberg

Teramoto

(1998) A sensitive, rapid and non-isotopic in situ bDNA assay for detection of hnRNP A2 mRNA. Proc Am Assoc Cancer Res 89th Annu Meet, New Orleans, LA, Section 2287

14.

Chernoff

Miner

Hoo

Shen

Kelso

Jekic–McMullen

Lalezari

Chou

Drew

Kolberg

(1997) Quantification of cytomegalovirus DNA in peripheral blood leukocytes by a branched-DNA signal amplification assay. J Clin Microbiol 35:2740–2744

15.

Collins

Irvine

Tyner

Fine

Zayati

Chang

Horn

Ahle

Detmer

Shen

Kolberg

Bushnell

Urdea

(1997) A branched DNA signal amplification assay for quantification of nucleic acid targets below 100 molecules/ml. Nucleic Acids Res 25:2979–2984

16.

Detmer

Lagier

Flynn

Zayati

Kolberg

Collins

Urdea

Sanchez–Pescador

(1996) Accurate quantification of hepatitis C virus (HCV) RNA from all HCV genotypes by using branched-DNA technology. J Clin Microbiol 34:901–907

17.

Furuta

Shinohara

Sano

Meguro

Nagashima

(1990) In situ hybridisation with digoxigenin-labelled DNA probes for detection of viral genomes. J Clin Pathol 43:806–809

18.

Hartley

Klaassen

(2000) Detection of chemical-induced differential expression of rat hepatic cytochrome P450 mRNA transcripts using branched DNA signal amplification technology. Drug Metab Disp 28:608–616

19.

Hendricks

Stowe

Hoo

Kolberg

Irvine

Neuwald

Urdea

Perrillo

(1995) Quantitation of HBV DNA in human serum using a branched DNA (bDNA) signal amplification assay. Am J Clin Pathol 104:537–546

20.

Herrington

Graham

McGee

(1991) Interphase cytogenetics using biotin and digoxigenin labelled probes: III. Increased sensitivity and flexibility for detecting HPV in cervical biopsy specimens and cell lines. J Clin Pathol 44:33–38

21.

Holm

Karlsen

Nesland

(1992) In situ hybridization with nonisotopic probes using different detection systems. Mod Pathol 5:315–319

22.

Kern

Collins

Fultz

Detmer

Hamren

Peterkin

Sheridan

Urdea

White

Yeghiazarian

Todd

(1996) An enhanced-sensitivity branched-DNA assay for quantification of human immunodeficiency virus type 1 RNA in plasma. J Clin Microbiol 34:3196–3202

23.

Komminoth

Werner

(1997) Target and signal amplification: approaches to increase the sensitivity of in situ hybridization. Histochem Cell Biol 108:325–333

24.

Manos

Kinney

Hurley

Sherman

Shieh–Ngai

Kurman

Ransley

Fetterman

Hartinger

McIntosh

Pawlick

Hiatt

(1999) Identifying women with cervical neoplasia: using human papillomavirus DNA testing for equivocal Papanicolaou results. JAMA 281:1605–1610

25.

Meissner

(1999) Nucleotide sequences and further characterization of human papillomavirus DNA present in the CaSki, SiHa and HeLa cervical carcinoma cell lines. J Gen Virol 80:1725–1733

26.

Mincheva

Gissmann

zur Hausen

(1987) Chromosomal integration sites of human papillomavirus DNA in three cervical cancer cell lines mapped by in situ hybridization. Med Microbiol Immunol 176:245–256

27.

Nargessi

Khabbaz

Zamroud

Kolberg

Collins

(1998a) Quantitation of estrogen receptor mRNA in breast carcinoma by branched DNA assay. Breast Cancer Res Treat 50:47–55

28.

Nargessi

Shimizu

Connolly

Zamroud

Collins

Kolberg

(1998b) Quantitation of progesterone receptor mRNA in breast carcinoma by branched DNA assay. Breast Cancer Res Treat 50:57–62

29.

Nuovo

MacConnell

Forde

Delvenne

(1991) Detection of human papillomavirus DNA in formalin-fixed tissues by in situ hybridization after amplification by polymerase chain reaction. Am J Pathol 139:847–854

30.

Nuovo

Richart

(1989) A comparison of biotin- and 35S-based in situ hybridization methodologies for detection of human papillomavirus DNA. Lab Invest 61:471–476

31.

Pardue

Gall

(1969) Molecular hybridization of radioactive DNA to the DNA of cytological preparations. Proc Natl Acad Sci USA 64:600–604

32.

Patterson

Till

Otto

Goolsby

Furtado

McBride

Wolinsky

(1993) Detection of HIV-1 DNA and messenger RNA in individual cells by PCR-driven in situ hybridization and flow cytometry. Science 260:976–979

33.

Pellegrin

Garrigue

Ekouevi

Couzi

Merville

Merel

Chene

Schrive

Trimoulet

Lafon

Fleury

(2000) New molecular assays to predict occurrence of cytomegalovirus disease in renal transplant recipients. J Infect Dis 182:36–42

34.

Plummer

Sperry

Lloyd

(1998) In situ hybridization detection of low copy nucleic acid sequences using catalyzed reporter deposition and its usefulness in clinical human papillomavirus typing. Diagn Mol Pathol 7:76–84

35.

Schmidt

Chao

Zhu

DeBiasio

Fisher

(1997) Signal amplification in the detection of single-copy DNA and RNA by enzyme-catalyzed deposition (CARD) of the novel fluorescent reporter substrate Cy3.29-tyramide. J Histochem Cytochem 45:365–373

36.

Shen

Sheridan

Cao

Dailey

Salazar–Gonzalez

Breen

Fahey

Urdea

Kolberg

(1998) Quantification of cytokine mRNA in peripheral blood mononuclear cells using branched DNA (bDNA) technology. J Immunol Methods 215:123–134

37.

Shyamala

Khoja

Anderson

Wang

Cen

Kavanaugh

(1999) High-throughput screening for ligand-induced c-fos mRNA expression by branched DNA assay in Chinese hamster ovary cells. Anal Biochem 266:140–147

38.

Siadat–Pajouh

Ayscue

Periasamy

Herman

(1994) Introduction of a fast and sensitive fluorescent in situ hybridization method for single-copy detection of human papillomavirus (HPV) genome. J Histochem Cytochem 42:1503–1512

39.

Simmons

Arriza

Swanson

(1989) A complete protocol for in situ hybridization of messenger RNAs in brain and other tissues with radiolabeled single-stranded RNA probes. J Histotechnol 12:169–181

40.

Sodora

Lee

Dailey

Marx

(1998) A genetic and viral load analysis of the simian immunodeficiency virus during the acute phase in macaques inoculated by the vaginal route. AIDS Res Hum Retrovir 14:171–181

41.

Speel

Schutte

Wiegant

Ramaekers

Hopman

(1992) A novel fluorescence detection method for in situ hybridization, based on the alkaline phosphatase-fast red reaction. J Histochem Cytochem 40:1299–1308

42.

Urdea

Wuestehube

(2000) Branched DNA (bDNA) technology. In Kessler

, ed. Nonradioactive Analysis of Biomolecules. New York, Springer-Verlag, 388–395

43.

Wang

Shen

Najafi

Kolberg

Matschinsky

Urdea

German

(1997) Regulation of insulin preRNA splicing by glucose. Proc Natl Acad Sci USA 94:4360–4365

44.

Wiedorn

Kuhl

Galle

Caselitz

Vollmer

(1999) Comparison of in-situ hybridization, direct and indirect in-situ PCR as well as tyramide signal amplification for the detection of HPV. Histochem Cell Biol 111:89–95

45.

Yee

Krishnan–Hewlett

Baker

Schlegel

Howley

(1985) Presence and expression of human papillomavirus sequences in human cervical carcinoma cell lines. Am J Pathol 119:361–366

46.

Zhou

Cryan

Minor

Gunnet

Demarest

(2000) A branched DNA signal amplification assay to quantitate messenger RNA of human uncoupling proteins 1, 2, and 3. Anal Biochem 282:46–53