FISH Technology Ready to Leap from Lab to the Clinic

Spectacular views of human chromosomes are coming from the laboratory of David C. Ward, Ph.D., acting chair of genetics at the Yale University School of Medicine (New Haven, CT). Using a tour-de-force of hybridization experiments coupled with innovative optical and software touches, the Yale team has achieved the long-sought goal of applying FISH (fluorescence in situ hybridization) simultaneously to all 24 types of human chromosomes.

"Whatís unique about David Wardís work is heís used combinatorial and multicolored fluorescent dyes to look at the entire genome," says Brian Ward, Ph.D., laboratory director at Integrated Genetics (Framingham, MA), one of several companies using FISH technology in kits for research-only chromosome analysis. (Brian Ward is no relation to David Ward.)

Although FISH has yet to win FDAís blessing to move from the research lab to the clinic, the potential market is in the multi-billion dollar range, according to Larry Fox, vice-president for technology and business development at Vysis (Downerís Grove, IL).

The field of cytogenetics began in 1923, when Theophilus Painter, Ph.D., sketched what he thought were the 48 human chromosomes, an erroneous count that was accepted for 30 years simply because of the difficulty of obtaining a microscope slide with the chromosomes spread apart enough to distinguish. In the 1950s, Albert Levan, Ph.D., and Joe-Hin Tjio, Ph.D., perfected methods of preparing white blood cells for chromosome analysis, establishing the correct number of 46 in 1956. In 1959, the first abnormal chromosome chart, or karyotype, depicted the extra chromosome 21 of Down syndrome.

But these early karyotypes were crude, staining chromosomes a uniform color, so that they were grouped into size classes, rather than numbered from largest to smallest, as they are today. In 1971 Torbjorn Caspersson, Ph.D., and Lore Zech, Ph.D., contributed the technique of combining stains to yield high-resolution chromosome bands. A few years later, Jorge Yunis, Ph.D., developed a method to synchronize cells in culture and arrest them when their chromosomes are the easiest to visualize. Then came FISH, which binds fluorescent dyes to DNA probes that then home in on their chromosomal homes. This added a specificity not seen with the more generalized existing chromosome stains.

FISH was invented by Joseph W. Gray, Ph.D., and Daniel Pinkel, Ph.D., in the mid 1980s when they were staff scientists at Lawrence Livermore National Laboratory. The approach grew out of their work in separating chromosomes for flow cytometry and in building chromosome-specific gene libraries. Drs. Gray and Pinkel are now director and co-director of the University of California at San Francisco Laboratory for Molecular Cytogenetics.

In January, 1986, Livermore and the University of California filed a patent for a technique to block repetitive DNA sequences, key to making bound FISH probes stand out, and exclusively licensed the technology to privately-held Vysis in 1987. The patent covers "methods and compositions for chromosome-specific staining" and was granted on September 5, 1995.

Issuance of the patent has caused some confusion because FISH has become an accepted technique in research. Oncor Inc., a Gaithersburg, MD-based, publicly-held firm marketing FISH-based kits, is currently challenging the patent. Stephen Turner, chairman and CEO of Oncor, claims that "this patent does not cover in situ hybridization products nor ... the general technique of FISH."

Beyond the bickering, though, lies the fact that FISH hasnít become a standard clinical cytogenetics tool quite as quickly as some had anticipated. "Five years ago, we predicted FISH would take over, but that hasnít happened, for a variety of reasons," says Michelle Le Beau, associated professor of medicine at the University of Chicago Medical Center.

Using FISH to distinguish all the chromosomes at once was an obvious extension of the technology, but was technically challenging. A milestone in 1992 by Dr. David Ward and then-postdoctoral associate Thomas Ried was combinatorial labeling with 3 fluorophores with digital imaging microscopy to reveal 7 DNA probes at a time.

FISHING for all the chromosomes requires a minimum of 5 fluors (fluorescent markers), because the number of combinations (2^ND) must exceed 24 (22 autosomes, the X and Y). And thatís precisely what Dr. Wardís team did, although by the time the paper appeared, they were already onto their sixth fluor. They use fluorescein and several different cyanine dyes tagged to chromosome-specific probes, 2 probes for the long and short arms of chromosomes 3, 5 and l1, and one for each of the other chromosomes. For example, chromosome 13 is marked with fluorescein, Cy3 and Cy7; chromosome 2 with only Cy7: and the Y chromosome with fluorescein and Cy3.

In the past, FISH has been limited by difficulty in distinguishing between fluors, both because of the short fluorescence band width, and the fact that the emission spectra from different organic fluorophores overlap. But Dr. Ward and his colleagues used computer modeling of digitized excitation and emission spectra to select filters and laser excitation sources that would allow each fluorophore to emit a distinct wavelength range.

"The key for us was being able to identify filter sets to allow imaging of 6 fluors, so that we could identify each chromosome uniquely. Once we did that, we could use a combinatorial scheme to mix and match fluors," says Dr. Ward. He calls the technique multiplex-FISH, or M-FISH.

Next, researchers use software to calculate a "spectral signature" for each multi-labeled probe, convert that to a gray value, and then assign each gray value a pseudocolor. The researchers first used M-FISH to construct normal karyotypes from peripheral blood lymphocytes, then analyzed several blinded chromosome abnormalities already described by conventional banding. These included translocations, deletions and trisomies. They were able to accurately detect the anomalies.

The next step in their work is to use conventional banding to detect complex chromosomal rearrangements, such as inversions and small translocations, deletions and duplications. "So far, we can paint each chromosome a different color, but that is not the same as banding. Our plan is to use the sixth fluor to give a banding profile by hybridizing to an Alu repeat. Thatís a repetitive sequence that gives a banding pattern," Dr. Ward says.

M-FISH promises to be a welcome new tool in cancer diagnosis and evaluation of treatments, particularly in the leukemias and lymphomas, where the chromosomes can be incredibly "jumbled," as Dr. Ward says. "M-FISH will help clinicians interested in looking at chromosomes in cancer patients to identify recurrent chromosome changes and rearrangements."

Detecting chromosome anomalies can help determine cancer type. For example, a translocation between chromosomes 15 and 17 indicates acute promyelocytic leukemia, whereas a translocation between chromosomes 9 and 22 means chronic myelogenous leukemia. Chromosomal clues also aid prognosis. Acute lymphoblastic leukemia, for example, has a good prognosis if cells have extra chromosomes 4 and 10, but a poor outcome if an extra bit of chromosome 22 is present. FISH has even revealed a cancer-associated translocation that conventional banding couldnít detect, because the sizes and banding patterns of the swapped regions were so alike. This translocation between chromosomes 12 and 21 affects 25% of patients with B cell acute lymphoblastic leukemia.

FISH may take longer to crack the prenatal testing market, where amniocentesis followed by conventional karyotyping is the gold standard. A few labs, such as Integrated Genetics and Oncor, offer FISH analysis of fetal chromosomes on a research basis for the most common chromosome anomaliesñtrisomies of chromosomes 13, 18 or 21, or extra or missing X and Y chromosomes. M-FISH could add coverage of the other chromosomes, but given the current cost-cutting climate, it may be awhile before it replaces the standard technique.

"FISH is not a stand alone technology because it must be followed up with amniocentesis. People wonít pay for two tests," says Dr. Le Beau.

Even though M-FISH will require more probes to approach the discrimination capability of conventional banding, in some ways it is already more powerful. A team at the Max-Planck Institute of Molecular Genetics in Berlin led by Thomas Haaf and working with the Yale group report using FISH to detect expanding trinucleotide repeats, a type of mutation that causes fragile X syndrome, myotonic dystrophy, Huntington disease, and several other neurological disorders. They labeled an expanding triplet repeat in a patient with paranoid schizophrenia, and were able to detect it in a microscope, unaided by image analysis. This work is reported in the February Nature Genetics.

One area where FISH excels is in speed. Dr. Ward estimates M-FISH can bring down amniocentesis time from 7 days to 4. It would still take 3 to 4 days to culture cells, but from there, prep time greatly accelerates. "What manually takes 2 to 4 hours in conventional cytogenetics becomes a computer-assisted 5 minutes with our system," he says.

An even faster approach is on the horizon. A technology called SpectraCube being developed by Applied Spectral Imaging Ltd. (Migdal Haíemek, Israel, and Carlsbad, CA) and Thomas Ried, Ph.D., now a researcher at NIH, takes one minute to Dr. Wardís five. Their SD-200 Spectral Bio-Imaging System uses an interferometer to determine the wavelength of light coming from each pixel of a digital image of a chromosome prep taken with a CCD camera. Software provides a spectra1 signature for each pixel, then integrates it into an image.

"This increases the resolution, but it also increases the complexity," says Dr. Ward. "The cost of the interferometer adds $80,000 to the cost of the system. Our filters do the same thing, and add only $2000 to $3000," he adds.

Whatever the speed and cost of M-FISH, it will take FDA approval to catapult the technology from a research-only tool to a clinical testing method. "Our job is to demonstrate at least equivalency to manually prepared conventional cytogenetic approaches. It will happen, but not for a few years. It will be the year 2000 before conventional cytogenetics changes," concludes Dr. Ward.

feedback form | permissions | international | locate your campus rep | request a review copy

digital solutions | publish with us | customer service | mhhe home

---

Copyright ©2001 The McGraw-Hill Companies.
Any use is subject to the Terms of Use and Privacy Policy.
McGraw-Hill Higher Education is one of the many fine businesses of the The McGraw-Hill Companies.