X-rays reveal the structure of proteins
Biochemist Irina Dementieva and biophysicist and crystallographer Youngchang Kim work with the first robot of its type in the nation to automate protein purification. The robot, which is housed in a refrigerator, is an integral part of the Midwest Structural Genomics Centerís plan to automate the protein crystallography process.
Biologists are using the newest biological detective devices to determine the structures of proteins and provide insight into the details of life from cell communication to gene activation.
Argonne researchers are piecing together the puzzle of life from all angles by:
- Operating the world's fastest facility for determining 3-D structures of proteins and their functions,
- Developing an assembly line to speed the delicate laboratory work, and
- Using computers to focus future research.
"Structural biology is about where the human genome project was 15 years ago," said Structural Biology Center Director Andrzej Joachimiak. "We have an amazing task before us, but we have exciting challenges we are confident we can meet."
Argonne's Structural Biology Center (SBC) provides atomic-scale structural data faster than any other biological research facility. In 2001, more than 340 SBC users produced more than 100 new structures and 57 publications in peer-reviewed journals including Science, Nature and Cell.
The structural information comes from shining X-rays from the Advanced Photon Source--the nation's most brilliant source of X-rays--onto tiny, frozen protein crystals. The X-ray images are captured by a quick, electronic camera. Advanced software converts the data into three-dimensional images that biologists use to infer how these proteins work and interact with other molecules. One day, the data may help medical doctors treat or cure diseases.
Biologists tease information from biomolecules by reading the letters of life--the As, Gs, Cs and Ts that represent the nucleotides that make up DNA--or by determining structure using 3-D images.
"Understanding the structure of the proteome--the entire protein and RNA components encoded by the genome--from the As, Gs, Cs and Ts of the human-genome language is like building a car from a book written in forgotten language," Joachimiak explains. "But when you take a 3-D car and see how it works, then you can build one. This is what structural genomics does."
The Midwest Center for Structural Genomics
While the SBC provides structural data quickly, the process leading up to the X-ray imaging is delicate, time-consuming and expensive. With funding from the National Institute of General Medical Sciences, part of the National Institutes of Health, Argonne is leading a group of research teams working to slash the average cost of analyzing a protein from $200,000 to $20,000, and the average time from months and years to hours and days.
The Midwest Center for Structural Genomics includes Argonne biologists and university teams from the United States, Canada and Great Britain.
To speed the structural biology visualization process, Argonne researchers are automating laboratory work, developing and using improved computer analysis programs and working with computer scientists to identify potential targets for comparative analysis.
Argonne biologists are becoming robotics specialists to cut the time-consuming lab work of turning proteins into crystals for study. In 2002, Argonne began using the first robot of its type in the nation for one of the many protein crystallography steps--protein purification. The average laboratory can purify four proteins a week manually; Argonne purifies 16 a week robotically.
A protein-cloning robot installed in 2001 is now capable of creating 400 clones a week. An Agilent lab-on-a-chip can analyze proteins in 30 minutes instead of two hours. Automating the crystallization process is the next step. "An around-the-clock production line is essential if we are to succeed in this genome-scale project," Joachimiak said.
The emerging science of bioinformatics
To streamline the process even further, Argonne's computational biologists are helping create the new field of bioinformatics. Instead of working in a traditional biology wet lab, these biologists use supercomputers to mine massive databases of metabolic, genetic and physiological information.
The field of bioinformatics is based on homology--collections of similar genes from humans and other living beings are assumed to have similar functions and relationships. This approach allows computational biologists to predict a protein's function and helps structural biologists to focus on verifying the predicted functions.
Computational biologists gather their data from WIT2, an Argonne-created interactive database of information, and perform their analyses with high-performance computing resources in Argonne's Mathematics and Computer Science Division.
The Midwest Center for Structural Genomics is NIH's most productive structural genomic center. It has provided more structures--40--to the Protein Data Bank than the other centers have.
Cells reveal communication secrets
In 2002, Argonne's Rong-guang Zhang and SBC collaborators, including Joachimiak, revealed a method cells may use to communicate. Bacteria communicate by releasing and sensing chemical signals known as pheromones--a phenomenon called quorum sensing. Biologists may one day manipulate this communication to thwart harmful bacteria or aid helpful bacteria.
Biologists determined the structure of TraR protein of the well-known Agrobacterium tumefaciens--an agricultural pathogen that causes tumors in plants.
"The structure," Zhang said, "is the most asymmetrical we have seen for protein-DNA complex. It is shaped liked a butterfly with its wings bent back."
Pheromones lie fully embedded within the protein. To activate the pheromones, several amino acid residues critical to RNA polymerase activation, or gene copying, make contact with the "butterfly body" of TraR.
Argonne researchers collaborated with Cornell University and Monsanto Company. This research was published in Nature.