A new technique that spots very faint signals in gene activity may one day help doctors better pinpoint the earliest signs of some diseases. Scientists at Harvard Medical School say they are able to detect low levels of gene expression with unprecedented sensitivity–an advantage that may allow geneticists to identify the activity of disease-related genes that until now have been obscured.
Picking out weak signals amid the normal din of genetic activity is no easy task. A single human cell contains more than 30,000 genes. At any given time, these genes are switched on or off depending on environmental factors, or in response to other genetic activity. As a gene is activated, it makes a copy, or multiple copies, of its DNA, and it passes these “instructions,” in the form of messenger RNA, to areas of the cell involved in protein synthesis. The cell then produces proteins based on the RNA blueprint.
Until now, researchers have been able to detect gene expression by imaging high levels of messenger RNA with techniques such as serial analysis of gene expression (SAGE) and microarray analysis. However, Jon Seidman, professor of genetics at Harvard Medical School, says that these technologies only provide half the picture. “When we label RNA with dye, the strong signals are obvious, and that’s good news; it has worked well with a number of diseases,” says Seidman. “The problem is, it captures some of the information but not all of it.”
Seidman and his colleagues have been keen on obtaining a more complete genetic profile of a rare heart condition, called hypertrophic cardiomyopathy. To do so, Seidman needed a method to detect messenger RNA at very low levels. So he and his colleagues developed polony-multiplex analysis of gene expression, or PMAGE, to increase the signal-to-noise ratio, bringing low-level gene activity to the foreground.
In a study published in Science, Seidman’s team took heart-tissue samples from two sources: a normal mouse, and a mouse with gene mutation to the heart disease. The researchers isolated messenger RNAs from each sample and converted each messenger RNA into complementary DNA sequences. They then split each double-stranded DNA molecule into a smaller fragment and attached molecular primers to either end of the fragment, creating a tag. Seidman mixed the genetic tags with magnetic nano beads. Through a polymerase chain reaction, the team cloned millions of the same DNA sequence tags on each bead. The result was an amplification of an otherwise weak genetic signal.
In their experiment, Seidman and his colleagues discovered more than 700 genetic discrepancies between their two samples, compared with about 50 found using traditional screening methods. “We’re pretty sure it’s a big deal,” says Seidman. “These mutations lead to a thickening in the heart wall, and also cause muscle cells to die, and also cause cardiac arrhythmias. And we had thought that probably one followed the other. But now we think there are probably three different pathways activated for each process. And we think this method would be very useful for distinguishing those.”
However, some scientists, like Silke Sperling, a cardiovascular geneticist at the Max Planck Institute for Molecular Genetics, warn against overinterpreting such results. “To date, the importance of genes with very low RNA levels is unclear in general,” says Sperling. “[Also], the individual variability of gene-expression profiles even between litter mates in mice is quite high, and to make a general causative biological assumption, the analysis of several mice is required.”
Seidman acknowledges this risk. “It’s absolutely true that if you see an RNA go up or down, you don’t know if RNA is the cause of the disease or the result,” he admits. But, he says, further studies could help clarify the roles of the RNA.
Down the road, Seidman also sees applications for the new technique in better diagnosis of other diseases such as different types of cancer. Scientists have already shown that “some changes in RNA expression were diagnostic of a particular type of tumor. If we could measure all RNAs, that might allow for further stratification of tumor types.”