Universal genetic code may not be so universal

Cells take a number of complicated steps to translate their sequence of basic DNA building blocks into proteins, which then act as workhorses to carry out the vital functions of life. Since many different proteins are encoded on a single DNA strand, the cell uses markers to know when to start and stop making a protein.

Many biology textbooks say that the start marker, called a start codon, always encodes for a compound called methionine. Yet William Duax, a structural biologist at the State University of New York at Buffalo, says new research by his team suggests the textbooks could be wrong. He will present the research at the 66th annual meeting of the American Crystallographic Association, help July 22-26 in Denver, Colorado.

"We have ample evidence that hundreds of the oldest ribosomal proteins still start with a valine or a leucine code and do not have the codon for methionine in the DNA," Duax said, referring to proteins found in basic cell components called ribosomes. "We have found unequivocal evidence that the earliest species on earth are still using a primitive form of the genetic code consisting of only half of the standard 64 codons," he said.

The results are contradictory to a widely held belief among biologists. "There are significant errors in text books. The universal code is not universal and all species now on earth do not use a code "frozen in time" as claimed by Watson and Crick," Duax said. "Some basic assumptions about evolution are incorrect." Duax also noted that the results raise questions about some aspects of a hypothesis on the origins of life, called the RNA world, which posits that RNA, which is similar to DNA and is still used in cells, was the first genetic material.

Duax and his team obtained their results by combing through a database that contains the sequences of more than 90 million genes. The genes encode proteins and the researchers used new techniques to accurately identify all members of each family of proteins and distinguish them from all other families that have remained unchanged for 3 billion years.

The research team developed programs to expedite the complete capture and perfect alignment of families of proteins having 25,000 members and encompassing all species for which genomes are reported. From those perfect alignments researchers could identify the precise location and function of the most conserved residues in the alignment, meaning the proteins that have stayed the same for the longest period of time. From these primordial proteins the researchers found evidence that the oldest proteins do not start in the standard way or use many of the other parts of the standard codes for making proteins.

Perhaps as surprising as the research and its findings is the way that Duax helped fund his research. He developed a three-week summer school in molecular bioinformatics and evolution for highly motivated high school students. In the past six summers he has trained more than 220 students to trace the origin and evolution of the protein composition and folding, of all cellular species and of the genetic code.

In addition to changing the way we look at genetic coding and rewriting textbooks, Duax's work has applications in genetic therapies that exploit structural details of bacteria to develop therapies that are selective and have fewer side effects.

The next step for the research team is to publish the results of their work and receive feedback from other researchers.

"Some of my students have been in the program for three years and are already equipped to prepare manuscripts for submission to journals in molecular evolution and structural science," Duax said. However, the team is just beginning.