By John Russell
July 14, 2004 | Natural biological systems perform a seemingly endless variety of functions. Imagine being able to modify or create entirely new “bio” systems to manufacture products, process information, or perform otherwise useful tasks. Welcome to the nascent world of synthetic biology -- a far cry from the long-practiced art of fermenters seeking better beer and wine.
At last month’s sold-out First International Meeting on Synthetic Biology* at MIT, engineers and researchers discussed exciting progress in creating a “library” of biological parts that can be combined in novel ways to create new machines.
“Think Legos,” said Thomas Knight, a researcher at MIT’s laboratory for artificial intelligence and the department of electrical engineering and computer science. Knight is a key player in efforts to build a library of “BioBricks” comprising the MIT Registry of Standard Biological Parts (parts.mit.edu). BioBricks are functional pathways -- not just DNA segments -- and typically code for operators, protein-coding regions, and transcriptional terminators. Some of the parts have been combined into logic gates.
Currently there are about 100 BioBricks. MIT has even coined a new unit of measure -- TIPS (transcripts per second) -- to describe signal output. Bricks are combined into “full circuits” in plasmids and expressed (usually in Escherichia coli) using traditional microbiology techniques.
Knight argued that synthetic biology requires an approach fundamentally different than traditional biology. His can-do attitude was evident in new studies designed “to engineer a simpler chassis and power supply” by stripping out redundant genes from the genome of bacterium Mesoplasma florum.
This work to build a “reduced” or minimal genome, which provides room to insert desirable genes by deleting unnecessary genes, builds on studies in 1999 conducted by J. Craig Venter and colleagues from The Institute for Genomic Research that showed that no more than 350 protein-coding genes of the bacterium Mycoplasma genitalium are required to sustain life.
Fred Blattner, a pioneer in E. coli sequencing from the University of Wisconsin, good-naturedly jabbed at Knight. “This is fascinating, though, if I might say, a little overly enthusiastic,” Blattner said. “We’re going in the same line … in attempting to get simplicity, but we’re also trying use an engineering principle of practicality.”
Starting with the K12 strain of E. coli, Blattner has engineered several “multiple deletion strains” (MDS). His approach retains core genes required for growth on minimal media, while removing intervening sequences and “potential virulence genes, adherence genes, toxic genes, cryptic operons, and to do it all without leaving any remnants.”
The current focus of Blattner’s work is strain MDS72, which will have roughly 20 percent less “stuff” in its genome. “3,700 genes is what nature is telling us you need to be a good E. coli,” Blattner said, adding that since many of the deletions represent islands of pathogenicity, the stripped-down E. coli will also be safer to use.
Among other highlights at the meeting, Carlos Barbas of The Scripps Research Institute discussed modifying protein scaffolds to create novel pathways that act like software programs. Drew Endy, a conference co-organizer from MIT, led a discussion of intellectual property issues. Another speaker suggested it might be possible to build a counter that incremented each time a cell divided.
* First International Meeting on Synthetic Biology; MIT, June 10-12.