Pervasive Network Discovered Driving Protein Production And Placement In Cells

In a study to be published on Oct. 27 in the journal Public Library of Science-Biology, investigators led by Pat Brown, PhD, professor of biochemistry and a Howard Hughes Medical Institute investigator, and Dan Hogan, a graduate student in Brown's lab, have accumulated persuasive evidence that certain proteins guide the fates of the molecules conveying genetic instructions from cells' nuclei to the diverse intracellular destinations where these instructions are to be carried out. The research was done on yeast, a simple model system for studying protein/genome interplay.

Proteins - which are the main structural components of living cells and do most of the work - must be produced in an orderly way. Different cells need different proteins, in different amounts, at different times and different places. But virtually all of an organism's cells contain the same genes, so how is this accomplished?

"If you gaze into the genome, you see lots of regions that don't code for proteins but are conserved in nature - they remain unchanged from organism to organism over eons of evolutionary time," said Hogan. "This suggests that these regions may be playing some important regulatory role."

It's been known for some time that specialized regulatory proteins called transcription factors are able to enter the nucleus and recognize and bind to specific barcode-like sequences of nucleotides found on DNA. By occupying these sites, transcription factors can make it more or less likely that RNA copies of a specific gene will be produced. Because many genes scattered throughout the genome have identical nucleotide barcodes, transcription factors can simultaneously modulate the expression of entire batteries of widely spaced genes. Clusters of proteins with interdependent functions can be produced in a coordinate manner.

The Stanford investigators have discovered an analogous system of regulation instigated by a class of proteins that are distinguished by their binding not to DNA but to RNA. Scientists have long been aware of the existence of numerous proteins that bind RNA - an obvious example is the ubiquitous, protein-building molecular machines called ribosomes, which are themselves made largely of proteins. They also know that in isolated cases, RNA-binding proteins, or RBPs, seem to influence the fates of the RNA molecules to which they bind.

But few had suspected the global nature of such interactions, which the Stanford team has now revealed to guide the fate of most, if not all, protein-coding RNA molecules. The study showed that virtually every protein-coding RNA molecule encoded in the yeast genome appears to be bound by specific combinations of RBPs, which have been directed to their target RNAs by short, barcode-like sequences on those RNA molecules.

Equally significant was the group's discovery that, in many cases, individual RBPs associated with groups of RNAs that coded for proteins related either in function (they need to operate in close coordination with one another) or location (they do their jobs in the same part of the cell.) This is consistent with the notion that many RBPs guide their bound RNA molecules to particular locations in the cell, preventing intervention by ribosomes along the way and ensuring that a protein gets made in the right place at the right time.

"Our work suggests that what had seemed like a relatively few specialized RBPs that are involved in some specific regulatory processes instead constitute a pervasive system of biological regulation, in many ways paralleling that of transcription factors," said Brown. "Just as transcription factors are recruited to the specific sets of genes they regulate by recognizing specific DNA sequences, these RBPs are recruited to a specific set of RNAs they regulate by recognizing specific sequences in those RNA molecules."

More than 500 different RBPs have already been found in yeast. Using a sample of just 40 yeast RBPs, the investigators observed interactions between the great majority of protein-coding RNA molecules and at least one RBP. On average, each RNA molecule interacted with three different RBPs in the course of its life. (Brown said he suspects that if all 500 RBPs were tested, the fraction of RNAs to which at least one RBP binds would approach 100 percent, and the number of different RBPs to which the average RNA molecule binds would increase as well.)

This multiplicity of RBP-specifying sequences - which can turn up on either side of an RNA molecule's protein-coding region, or right smack in the middle - may help explain what the huge sections of all protein-coding RNA molecules that don't code for protein assembly are really up to.

There is no reason to think RBPs' regulatory activity is confined to yeast, said Brown. Even more of them are believed to exist in multicellular organisms, including humans. "Here's this whole system - which has gone virtually unnoticed - that looks like it has a pervasive role in regulation," he said. "And we still know almost nothing about it. It looks like a great, unexplored area, which is just what you love if you're a scientist."

Other authors of the study, which was funded by HHMI and the National Cancer Center, included Dan Riordan, a graduate student working in Brown's lab; Andre Gerber, PhD, a former postdoctoral scholar in the Brown lab now on the faculty at ETH Zurich in Switzerland; and Daniel Herschlag, PhD, professor of biochemistry.