YouTube Facebook LinkedIn Google+ Twitter Xingrss  


By Robert M. Frederickson
By Robert M. Frederickson

February 18, 2004 | The ability to measure the expression of an organism's entire genetic complement in parallel is of intense interest to both the basic research and the drug development communities. While most attention has focused on oligonucleotide and cDNA arrays, some fascinating alternative platforms are available.

Microarrays can be as comprehensive as one wants, provided that the genes of interest have been identified and validated. Microarray use is also restricted to those genes present above a detection limit of approximately 1 in 10,000 transcripts — many important genes involved in disease pathology fall below this limit. And comparison of microarray data between labs, and even experiments, remains a difficult task.

 Measuring Up: In massively parallel signature sequencing, each of 2 million Lynx Therapeutics beads is simultaneously interrogated by a series of decoder probe hybridizations to identify the DNA sequence for positions 1-20. The graph shows the relative intensity of the G, A, T, and C signals for each of 20 bases for one bead. 
Many of these limitations are addressed, however, by a technique called massively parallel signature sequencing (MPSS), which has been commercialized by Lynx Therapeutics. Sequencing a short bit of every cDNA as a proxy for the mRNAs in a cell, MPSS essentially counts individual mRNA molecules through recognition of specific 20-base signatures. Importantly, the method does not depend upon prior knowledge of the genome or of the expressed genes, so it can be used for gene discovery as well as expression analysis.

A typical MPSS data set contains about a million sequence signatures, a measure of the relative expression of the genes in the data set. "MPSS data," says Tom Vasicek, Lynx's vice president of business development, "provide a complete, accurate, and permanent digital record for the expression of every gene in the cell. So the results can be compared to any other data collected at any time in any laboratory."

The Lynx approach begins by cloning a tagged cDNA library onto the surface of microbeads. Amplified 20-base signature sequences from every cDNA attach to unique 32-mer "address" oligonucleotide "combitags." These tags then find their complementary "capture" oligos synthesized onto the surface of the beads. "There are some 16.7 million different combitags, so no two transcripts end up associated with the same one," Vasicek says.

Each individual bead therefore represents a unique expressed sequence tag (EST), and is coated with about 100,000 copies of the amplified signature from one mRNA transcript. The number of beads corresponds to the prevalence of the mRNA molecules.

In order to "count" the messages present in this microbead-based cDNA library, the method essentially takes a "sequence sample" from every bead. The beads are laid out onto a 2-D surface packed within a microfluidic flow cell. A restriction enzyme digests the de facto microarray, repeatedly cleaving a few bases off the end of each signature, creating four-base overhangs that can be read using a set of 16 fluorescent adapter oligonucleotides — every possible combination of four nucleotides. This process is continuously imaged using a CCD camera, with the results decoded and the sequence recorded. The decoded sequence is then identified by comparing it to genomic and EST sequence databases. (A detailed description of this ingenious technique, invented by Nobel laureate Sydney Brenner, can be found at www.lynxgen.com/wt/tert.php3?page_name=mpss.)

Although microarrays are generally simpler to use for routine expression analysis, MPSS is finding wide application in discovery research. Lynx has also inked agreements with pharmaceutical developers who will use the technology to screen expression differences in cells from diseased and normal patients. In collaboration with IBM and Seattle's Institute for Systems Biology, for example, the technology will be used to study how the immune system responds to disease.



Robert M. Frederickson is a biotech writer based in Seattle. E-mail: rfreder@yahoo.com. 









For reprints and/or copyright permission, please contact  Jay Mulhern, (781) 972-1359, jmulhern@healthtech.com.