The research is featured in the latest issue of Analytical Chemistry.
Combining our expertise in biology and microfluidics in a collaborative project with Technion – Israel Institute of Technology, our research group at IBM Research-Zurich developed a new method for designing fluorescence in situ hybridization (FISH) assays capable of accurately characterizing the hybridization kinetics for any given set of probes or target sequences. What’s so great about this? Well, we’ve made it possible to monitor FISH signals in real time without the need for special probes. This means that biomedical scientists will be able to design new, faster, and more efficient FISH assays as well as optimize existing ones.
To the best of our knowledge this is the first method of its kind. And today, our work is being featured on the cover of the peer-reviewed journal Analytical Chemistry.
FISH and its limitations
Commonly known as FISH, fluorescence in situ hybridization is a molecular cytogenetic technique developed by biomedical researchers in the early 1980s that uses fluorescent probes to identify and localize the presence or absence of specific DNA and RNA sequences on chromosomes inside cells, tissues, and cell blocks. As this technique has the ability to provide a detailed spatial analysis of gene expression as well as distinguish cell types and detect many types of chromosomal abnormalities, it is extensively applied in a wide range of research and medical diagnostic applications.
Schematic illustration of the liquid switching scheme and real-time imaging of the FISH signal development.
FISH has its limits, however. For one, the technique relies on DNA hybridization inside a cell, a procedure that can take anywhere between 8 and 94 hours per test. This certainly leaves a great margin for error. Various conditions such as the presence of salt, fluctuations in temperature, and the concentration of different reagents can have critical effects on hybridization results. The procedure also becomes more complex, not to mention labor-intensive and time-consuming when testing multiple effects of reagents. What’s more, hybridization rates obtained in solution or through surface-based methods do not account for factors such as intracellular and intranuclear molecular crowding or spatial arrangement and target accessibility. This makes it difficult to acquire the appropriate reaction rates of the in situ reaction.
Essence of real-time fluorescence measurement
Our main goal from the start of this research project was to respond to the growing interest in more precise evaluations of in situ hybridization rates by developing a new method that achieves probe designs with improved sensitivity and specificity as well as reduce typically long assay times. While there have been several attempts to further develop FISH assays, there are currently no techniques for the quantification of the hybridization rates that can be readily applied to any probe or target. Why does this matter? Without quantification, the design procedures of FISH probes and assays remain mainly empirical. Moreover, hybridization conditions for a given probe are estimated experimentally using an end-point analysis of the signal, which when multiple effects and conditions are being tested, is a highly complex, labor-intensive and time-consuming task.
We concluded that an accurate quantification of the hybridization rates in FISH reactions requires monitoring the process in real time. But our biggest obstacle was getting around the background signals of a high concentration of probes present on the surface at all times throughout the experiment. To solve the problem, we developed a method that enables the direct monitoring of intracellular hybridization through rapid alternation between the delivery of the fluorescent solution containing probes, which bind to the chromosome inside the cell, and a non-fluorescent imaging buffer. Because the FISH signal is only recorded during the imaging-buffer phase, the background signal associated with unhybridized probes is completely removed. Thus, the signal of the hybridized DNA in the cell can be imaged directly and accurate kinetic measurements can be obtained in real-time.
Putting it to the test
The research team: (from left) Moran Bercovici, Govind Kaigala and Nadya Ostromohov
Our research team—Deborah Huber (PhD student) and Prof. Moran Bercovici from Technion, Govind V. Kaigala and I—decided to test the method using breast cancer cell blocks. The results showed that our method is capable of assaying multiple conditions in a short time. What’s more, it can be applied using a fluorescent microscope, which is the standard equipment used in today’s research and medical laboratories.
We believe our method has the potential to open the door to new insights for a better understanding of in situ reaction processes, which previously could not be quantified due to the limitation of FISH techniques. It will also facilitate the design of new assays and protocols and pave the way for more accurate theoretical or computational models describing in situ hybridization.
At the 18th European Conference on Computational Biology and the 27th Conference on Intelligent Systems for Molecular Biology, IBM will present significant, novel research that led to the implementation of three machine learning solutions aimed at accelerating and guiding cancer research.