Test Pulls Gene Regulators from DNA Package

The sequencing of ancient scraps of DNA from fossils of long-extinct animals has given rise to hopes that fantastical creatures like Tyrannosaurus rex and the woolly mammoth may once again roam the planet—or at least the carefully guarded precincts of a Jurassic theme park. But it could be a while before such scenarios make the leap from the silver screen.

“In Jurassic Park, they start out by saying they found the DNA of a dinosaur in amber and now they could create a dinosaur. But you can’t create an organism just out of the DNA,” said Robert Kingston, HMS professor of genetics in the Department of Molecular Biology at Massachusetts General Hospital. “A lot of the information that makes up an organism is determined by the proteins that are covering the DNA.”

Scientists have known for decades that DNA is only part of the story of how genes work—the bare bones, so to speak. Each strand is wound, interlaced, and layered by myriad proteins and molecules that work together to turn the genetic code into actual features—limbs, claws, teeth, and tusks. Researchers have been avidly studying this dense interplay of DNA and other molecules—the chromatin—hoping to isolate the proteins that regulate individual genes.

In a feat that until recently belonged to the realm of science fiction, Jerome Dejardin, a postdoctoral fellow working in Kingston’s lab, has netted what appears to be a nearly complete array of proteins associated with a specific locus, the region responsible for forming the chromosome-capping telomeres. He did so using a technique that he and Kingston hope may someday be used to capture the entire set of proteins associated with genes on other regions of the chromosome.

“Nobody yet knows every protein that works on each gene, which is exactly what we’re developing this technique to find out,” said Kingston, adding that such knowledge could lead to new ways to control genes in the lab and clinic.

The method, which appears in the Jan. 9 Cell, is dubbed proteomics of isolated chromatin segments, or PICh—and it works like a kind of reverse version of the well-known chromatin-capturing technique, chromatin immunoprecipitation (ChIP). “With ChIP, you pull down a protein and ask, ‘What DNA is there?’” said Dejardin, HMS research fellow in genetics at MGH. “Here we’re pulling down the DNA and asking, ‘What proteins are there?” At the heart of the new technique is a nucleic acid probe capable of binding a specific segment of DNA and stabilizing its retinue of proteins so tightly that few fall off during capture.

Using the PICh probe, Dejardin was able to isolate and identify almost all of the proteins previously associated with telomeres—and then some. He found eight novel telomere-associated proteins and, using ChIP analysis, confirmed that they are, indeed, associated with telomeric DNA, though it is not entirely clear how they function.

“Even on this very studied region, we found a lot of proteins that nobody had found before,” said Kingston. “People have been trying to do this for 30 years—it’s a very hard thing to do, to pull out a locus with all the proteins,” he said.

Knowing which of the thousands of DNA-binding proteins are actually present at a given gene might someday help to bring extinct creatures to life, but Dejardin’s method could have more immediate and no less exciting implications for extant human beings. By tweaking these gene-associated proteins, it might be possible to alleviate disease, for example, by boosting underperforming genes or silencing harmful mutants.

Dejardin knew it would be a difficult road when he set out. Trained as a cytogeneticist in his native France, he had developed an abiding fascination with chromosomes—and a deep frustration. First spotted in the mid-19th century, chromosomes remained, in his eyes, an exasperating blur of protein and DNA, due in large part to an inability to pinpoint exactly which proteins and molecules bind which segments of the chromosome.

Aware that others had tried to purify gene-specific factors, Dejardin made a strategic decision at the outset to focus on telomeres. One of the problems of previous approaches was that they yielded too little protein to be analyzed by protein-detecting methods such as mass spectrography. Telomeric DNA is located at both chromosomal tips, so it is twice as plentiful as single genes. In addition, it is single stranded, and it is made up of repeating sequences, both of which could make capturing the DNA easier.

As for the method of capture, Dejardin thought that some variant of the well-known DNA-baiting method called fluorescence in situ hybridization (FISH) would work. But his initial attempts using a special bait, a locked nucleic acid (LNA) probe, which is known to bind DNA with very high affinity, met with miserably low protein yields and unacceptably high levels of contaminants.

He continued tinkering with the LNA probes, substituting biotin for one of the nucleic acids, adjusting the length of an element known as the spacer, and making other adjustments. “The key was getting biotin,” he said. He got the probe to work, creating a stable hybrid between the LNA and the DNA and its attendant proteins. Using a series of contamination-ridding detergents, he was able to purify enough protein to satisfy the rigors of mass spectrography.

Telomeric DNA is famously maintained by the enzyme telomerase, but there is another less well–understood route, the alternate lengthening of telomeres (ALT) pathway. Dejardin wanted to see if his method could distinguish between them. He performed the PICh analysis on cells containing each of the two kinds of telomeres, uncovering about 200 proteins for each. Half of these were found in both classes. “We expected this,” said Dejardin. Many were ubiquitous proteins, found all along the chromosome. But remarkably, Dejardin was able to isolate almost all of the previously known telomere-binding proteins, 33 out of about 40.

He also identified eight previously unknown ALT-related factors, five of which were, to his surprise, orphan receptors. “We were not expecting this type of protein to be bound here because these proteins are typically transcription factors and are located at regulatory regions of genes, such as the promoter,” said Kingston. To see how they might be functioning, Dejardin knocked down one of the most abundant of the receptors. Other than a slight shortening of telomeres in some cases, the results were not particularly striking. “Nothing truly amazing happened,” said Kingston. “But we still think they are important.”

Dejardin plans to follow up on the orphan receptor findings and also to try out his approach on other regions of the chromosome.

“We don’t know how easy it will be to use it on non-telomeres. Telomeres are a very specific kind of locus. Jerome and I will feel more comfortable when it has worked well on another locus,” said Kingston. “In a perfect world, we want a technique that can be used on every single gene.”

Students may contact Robert Kingston at kingston@molbio.mgh.harvard.edu for more information.

Conflict Disclosure: The authors report no conflicts of interest.

Funding Sources: The National Institutes of Health (R.K.) and European Molecular Biology Organization (EMBO) postdoctoral fellowship (J.D.)