In situ hybridization (ISH) is a laboratory technique that detects specific DNA or RNA sequences directly inside cells and tissues, showing exactly where those sequences are located. Unlike methods that extract genetic material and analyze it in a test tube, ISH works on intact tissue samples, preserving the spatial context so scientists and clinicians can see which cells contain a particular gene, virus, or genetic abnormality. It’s one of the most widely used tools in cancer diagnostics, infectious disease testing, and gene expression research.
How It Works
The core principle is simple: DNA and RNA are made of base pairs that naturally bind to their complementary sequences. If you design a short piece of nucleic acid (called a probe) that matches the sequence you’re looking for, and attach a detectable label to it, that probe will seek out and bind to its target inside the tissue. The label then produces a visible signal, revealing exactly which cells contain the target sequence and where within the tissue they sit.
The process has three main stages: tissue preparation, hybridization, and washing. During preparation, the tissue is chemically fixed (usually with formaldehyde) to preserve its structure and then permeabilized so the probe can physically reach the target sequences inside cells. Fixation times vary enormously depending on the sample. Cultured cells might need only 10 minutes, while a whole mouse brain can require up to 24 hours. Getting this balance right matters: too little fixation and the tissue falls apart, too much and the targets become chemically locked away.
During hybridization, the probe is applied to the tissue and allowed to bind its target. Samples are typically heated to around 75°C for about 10 minutes to separate the two strands of DNA or unfold RNA, making the target sequences accessible. Then the temperature is lowered and the probe incubates with the tissue for 12 to 24 hours, giving it time to find and lock onto complementary sequences. After hybridization, the tissue is washed to remove any probe that didn’t bind specifically, reducing background noise and leaving only the true signal behind.
Types of Probes
Four main probe types are used in ISH, each with different strengths. Double-stranded DNA probes are versatile and easy to produce. Single-stranded DNA probes bind only one strand of the target, which can reduce background noise. RNA probes (called riboprobes) are especially useful for studying gene expression because they bind very tightly to messenger RNA and can reveal not just where a gene is active but how strongly it’s being expressed. Synthetic oligonucleotides, including modified versions like PNA and LNA, are short, chemically stable probes that can be precisely designed for difficult targets.
Each probe carries a label that makes it detectable. Early ISH relied on radioactive isotopes, which produced high-quality signals but required long exposure times and special safety handling. Modern methods overwhelmingly use non-radioactive labels: fluorescent dyes that glow under specific wavelengths of light, or molecules like biotin and digoxigenin that can be detected through color-producing chemical reactions viewed under a standard microscope.
FISH vs. Chromogenic Detection
The two dominant detection strategies in clinical labs are fluorescence in situ hybridization (FISH) and chromogenic in situ hybridization (CISH). FISH uses fluorescent labels in red, orange, or green that light up under a fluorescence microscope. CISH instead produces colored deposits visible under a regular bright-field microscope, the same kind used for routine pathology slides.
FISH has long been considered the gold standard for genetic testing, particularly in cancer diagnostics. But it has practical drawbacks. Fluorescent signals fade over time, the slides can’t be easily archived, and automated scanning is slow. In one comparison study, digitally scanning a FISH slide took about 764 seconds per square millimeter, while a CISH slide took just 29 seconds. FISH also had a higher failure rate during automated scanning, with issues like autofluorescence and focusing errors.
CISH’s main advantage is convenience. Results can be viewed on a conventional microscope, compared side by side with standard tissue stains, and stored indefinitely. Its limitation is that densely amplified gene signals can sometimes blur together, making colors harder to distinguish. For high-volume testing, CISH is generally faster and more practical. For smaller-scale, highly specialized analysis, FISH remains a strong choice.
Clinical Uses in Cancer and Infection
ISH plays a critical role in cancer diagnosis by identifying genetic changes that guide treatment decisions. The most prominent example is HER2 testing in breast cancer. HER2 is a growth-promoting protein, and tumors with extra copies of the HER2 gene tend to grow more aggressively but respond well to targeted therapies. FISH testing counts the number of HER2 gene copies in tumor cells and compares them to a reference point on the same chromosome. Current guidelines from the American Society of Clinical Oncology and the College of American Pathologists classify results into five groups based on the ratio between HER2 copies and the reference signal, along with the average number of HER2 copies per cell. A ratio of 2.0 or higher with an average of 4.0 or more copies per cell is the clearest positive result.
Beyond breast cancer, ISH is used to detect gene rearrangements in lung cancer, chromosomal abnormalities in leukemias and lymphomas, and other genetic markers that determine whether a patient qualifies for specific therapies.
Detecting infectious agents is another major clinical application. Pathology labs routinely use ISH to identify human papillomavirus (HPV) in cervical and tonsil tissues, and Epstein-Barr virus in certain lymphomas and nasopharyngeal tumors. It can also detect cytomegalovirus, hepatitis B and C, herpes simplex, and HIV within tissue samples. ISH is particularly valuable when standard antibody-based detection methods produce too much background staining to be reliable.
Tissue Preparation: Fresh Frozen vs. Preserved
Most clinical ISH is performed on formalin-fixed, paraffin-embedded (FFPE) tissue, the standard format for pathology samples. These samples are durable, easy to store, and can be re-examined years later. The tradeoff is that formalin creates chemical cross-links between proteins that can block probes from reaching their targets. Antigen retrieval steps, such as enzyme digestion with proteinase K or heat treatment, are used to reverse this masking effect and open up access.
Fresh frozen tissue avoids the cross-linking problem entirely, which can improve signal quality. But frozen sections are thicker, harder to interpret morphologically, and can’t be stored long-term as easily. In practice, most diagnostic labs work with FFPE tissue and adjust their protocols accordingly.
Newer Technologies: Single-Molecule Detection
Traditional ISH had real limitations. Background noise was high, low-abundance targets were hard to detect, turnaround times were long, and only one gene could typically be examined per tissue section. Newer platforms like RNAscope (developed by Advanced Cell Diagnostics) have addressed these problems through a clever signal amplification design.
RNAscope uses pairs of short probes that must both bind to adjacent sequences on the target RNA before any signal is generated. If only one probe of a pair binds to a non-specific sequence, no amplification occurs, which dramatically reduces false signals. When both probes bind correctly, a cascade of amplification molecules attaches to them, theoretically boosting the signal up to 8,000-fold per target. Under optimal conditions, a single dot of signal corresponds to a single RNA molecule, giving researchers the ability to literally count individual transcripts inside a cell.
Even more recent advances have pushed ISH into the territory of spatial transcriptomics, where hundreds or thousands of different RNA species are detected simultaneously in a single tissue section. These methods use combinatorial barcoding: each gene is assigned a unique pattern of “on” and “off” signals across multiple rounds of imaging. By cycling through these rounds and decoding the patterns, researchers can map the activity of vast numbers of genes while preserving the exact tissue architecture. This is transforming how scientists study complex tissues like the brain, tumors, and developing organs, revealing not just which genes are active but precisely where in the tissue each cell type lives and what it’s doing.