Genetics, Molecular Biology
Fred Hutchinson Cancer Research Center
Dr. Henikoff is also a member of the Basic Sciences Division at the Fred Hutchinson Cancer Research Center, Seattle, and an affiliate professor of genome sciences at the University of Washington, Seattle.
Steven Henikoff performs research on chromatin dynamics, transcriptional regulation, and centromere inheritance and develops experimental and computational tools for studying these processes. Recent methods map nucleosome turnover, transcription factors, chromatin remodelers, RNA polymerases, nucleosomes, and DNA torsion genome-wide at high resolution. Application of these tools has elucidated the relationship between transcription, torsion, and nucleosome turnover; mapped chromatin proteins at base pair resolution; identified the nucleosome barrier to transcription; and determined the molecular organization of centromeric nucleosomes.
Histones are homely proteins that make human life possible. They form the spools on which our 3 billion DNA base pairs are wound for safekeeping and selectively unwound when genes need to be expressed. DNA comes in an elegant double helix, while the four core histone proteins double up into an eight-armed, utilitarian clump. Until recently, the unglamorous histones were easy to underestimate. But appearances can be deceiving, says Steven Henikoff, a pioneer in unraveling the secret life of histones.
Henikoff's latest work on histones kicks a major dogma of cancer biology in the shins. At issue is the relationship between histones and DNA methylation, the insertion of "methyl" molecular tags onto DNA bases that can silence a gene. DNA methylation is a central mechanism of epigenetics, the cell's "other" system of heredity, which passes on information from generation to generation in dividing cells without changing a single gene.
DNA methylation helps prevent unscheduled expression and keeps genomic parasites in check. But in cancer, inappropriate DNA methylation can wreak havoc, allowing cancer cells to race off down the pathway to cause destruction. A wrongly placed DNA methyl tag can switch off a gene that monitors cell growth by attracting other proteins to shut it down—or so goes the dogma. Not so fast, says Henikoff.
"People have their minds set on this paradigm that's been around for 10 years—that methyls will recruit proteins that will shut the gene off," Henikoff explains. "I didn't find the evidence for that to be compelling." In a November 2008 paper in Nature, Henikoff's group presented their own evidence that DNA methylation shuts off genes by preventing incorporation of a histone variant, H2A.Z.
"H2A.Z in the histone spool promotes gene expression," says Henikoff. "Methylation works by excluding H2A.Z."
If Henikoff is right, this is no minor molecular dustup. Cancer biology researchers are focused on plotting DNA methylation patterns in cancer and figuring out therapies to disrupt or reverse them. In demonstrating antagonism between H2A.Z and DNA methylation, Henikoff may have opened a new approach to epigenetic cancer drugs.
Changing paradigms is something of a favorite Henikoff activity. In a 2007 paper in PLoS Biology, Henikoff and colleagues broke the canonical pattern for the histone spool. Histone spools have always come in a clump of eighttwo proteins each of the four core histones, H2A, H2B, H3, and H4. Together the eight proteins form an octamer, an eight-member macromolecule. All histone reels are octamers, or so said the established model of chromosome construction—that is, until Henikoff's group discovered a four-protein histone spool, a tetramer, composed of a single protein from each of the four core histones.
Henikoff and his colleagues retrieved their heretical tetramer from the centromere, the chromosome region that Henikoff cheerfully calls the "black hole" of genomics. "If you look at the human genome project, supposedly we sequenced the genome in 2001 and have updated it since. But there is still not a single centromere (on any of the 46 human chromosomes) that has been sequenced. Centromeric DNA is just so repetitive that it's impossible with multibase arrays to assemble one and say this is the sequence." Pulling these half-sized histone spools out of the genomic black hole challenges assumptions about how histone spools evolved, he says.
Henikoff has been a scientific iconoclast almost from the beginning of his research career at the Fred Hutchinson Cancer Research Center in Seattle. In 1986, he announced the discovery in fruit fly DNA of a "gene within a gene." This was heresy to the doctrine of the day that all DNA information was linear—that is, a gene could encode only one set of instructions at a time. Henikoff's hidden gene was in the intron, the supposedly meaningless sequences that bracket the coding sequences of otherwise perfectly respectable genes. Finding genes within genes is commonplace today. Finding new paradigms and new methods is still standard procedure at the Henikoff lab.
The discovery of hidden genes grew out of Henikoff's early work at the "Hutch" on one of the first bioinformatics programs. Working with his wife Jorja, a trained mathematician, Henikoff designed a protein motif comparison program to find homologous genes in widely separated species. The Henikoffs, who are fond of acronyms, called their first bioinformatics program BLOCKS, which begot BLOSUM, which lives on at the NIH National Center for Biological Information's BLAST protein and gene database. The Henikoff lab continues to invent new bioinformatics tools with funny acronyms such as TILLING, which winnows out novel mutations, and SIFT, which strains proteins for mutational effects.
The whimsy of the names matches the playful curiosity at the heart of Henikoff's science. "I always tell people that I got into science as a hobby," Henikoff explains. As a teenager in Chicago, he took up photography, which led him to darkroom chemistry, which led to chemistry, then biochemistry, then genetics, and so on to histones, centromeres, and epigenetic medicine. "It's still a hobby," says Henikoff.