Everything's Coming up Combinatorial at ASBMB

Contents
Introduction
Drug Discovery
Going Global: Looking Beyond a Single Gene
Combinatorial Transcription
And That's Not All Folks

Introduction (Back to Top)
Some 2,000 plus scientists met in San Francisco at the annual meeting of the American Society for Biochemistry and Molecular Biology (ASBMB). In four days of general sessions and two days of satellite meetings, the full spectrum of modern biological research was covered, from genes and genomes to the RNA World (a satellite) to protein design and enzymology. In plenary sessions, such notables as Joseph Goldstein and Michael Brown of the University of Texas Southwestern Medical Center (Dallas), Peter Schultz of UC Berkeley (Berkeley, CA), and Eric Landers of MIT's Whitehead Institute (Cambridge, MA) spoke, and that was just in the first day.

While combinatorial chemistry has become a household word among laboratories working in drug discovery, this concept has not been overlooked in nature, as several talks detailed schemes for building transcription complexes using this concept. And combinatorial chemists have turned their sights toward RNA, as five out of six presentations in the combinatorial chemistry session were devoted to RNA-directed strategies.

Drug Discovery (Back to Top)
Talks describing the development of new technologies for designing and isolating biologically active compounds came from all sectors of the scientific spectra—academic laboratories, research and development firms, and large pharmaceutical companies were all represented in these discussions.

Gregory Verdine of Harvard (Cambridge, MA) described an innovative approach he terms "peptide morphing." Rather than mimicking peptides, which, he contends, are poor models for drugs—prisoners of their own chemistry—he morphs them in several different ways. One way is to replace the amide bond with something functionally similar, but lacking some of the unwanted properties of the polar amide bond. A second approach involves changing the conformational properties by stretching the molecule. Using this peptide morphing approach, a small molecule with a stable alpha-helix was described that competes with p53 for binding to the MDM2 oncogene, an interaction which in nature targets p53 for degradation, possibly releasing cells into uncontrolled growth. Verdine believes that releasing the constraints inherent in peptido-mimetics as it is normally conducted could lead to the isolation of a whole new class of molecules with enhanced biological activity and greater utility.

Stephen Fesik of Abbott Laboratories (Abbott Park, IL) discussed the use of NMR as a tool in drug research, in particular a method he calls SAR or structure activity relationships by NMR for discovering ligands with high affinities to biological targets. Using this method, small organic molecules that bind to proximal sites of a protein are identified, their binding optimized and linked together. One of the disadvantages of NMR as a screening tool has been its relative lack of sensitivity, which limits the number of compounds that can be screened in a sample. Fesik showed results using a high-resolution probe, Cryoprobe, which increases the sensitivity of NMR, thereby increasing by at least an order of magnitude the number of compounds that can be tested in a sample. Using this technology, a peptide mimetic for phosphotyrosine was discovered.

Peter Dervan, professor of chemistry at CalTech (Pasadena, CA), described his group's efforts at developing synthetic DNA binding ligands. Using a set of three aromatic ring amino acids, structures were described that distinguish the four Watson-Crick base pairs and bind DNA with high affinities, comparable with DNA binding proteins. These ligands exhibit a high degree of specificity, able to discriminate single base pair mismatches. As an example, a molecule that was targeted to the HIV promoter was found to have biological activity in both in vitro and in vivo assays. Such an approach could lead to the discovery of cell-permeable molecules with gene-specific effects on regulation.

Mark Gallop of Affymax (Palo Alto, CA) described the variety of strategies used by researchers at his company for constructing and screening combinatorial libraries of small molecules for drug discovery—among them screens for soluble compound pools and solid-phase systems. The approach that shows the most promise, in terms of the numbers of compounds that can be screened, uses encoded, bead-based libraries of compounds in an automated system. Going well beyond the industry standard, 96-well plate, the system devised at Affymax currently can screen 125,000 compounds against 8 targets a week per robot, or roughly a million compounds a week.

Going Global: Looking Beyond a Single Gene (Back to Top)
Strategies for looking at large regions of genomes, or sampling the expression of large numbers of genes, were also described. While some of these studies focus on simple experimental model systems, such as yeast and Drosophila, they are using some of the most sophisticated tools available today to get as complete a picture as possible of genetic regulatory mechanisms.

David Botstein of Stanford University (Stanford, CA) described comparative studies of genes from yeast and worms, the genomes of which are now both available in toto. Such an analysis has led him to conclude that the genes that carry out core functions are conserved throughout species. On the other hand, proteins carrying out functions found only in the worm are novel, though they sometimes employ domains found in conserved regions of genes.

Botstein described another kind of experiment—chemical evolution—for unveiling some of the genetic strategies available to organisms under stress. Using a chemostat to control growth, Botstein, after several rounds of selection under conditions limiting growth, found hundreds of alterations in gene expression in yeast. He believes that mutation in a regulatory gene or genes may underlie this massive change in the genetic read-out, rather than individual mutations in the myriad of genes whose expression is modulated, suggesting that such simple selections could provide insight into complex regulatory mechanisms.

Gerald Rubin of UC Berkeley spoke on his group's efforts at laying out the underlying genetic scaffolding of the fruit fly, Drosophila melanogaster. Chosen for its relative simplicity, having only somewhere between 12 to 15,000 genes, the fly is well endowed with information on both its genetics and developmental biology. Rubin described the work that he and his collaborators are conducting using P1 transposable elements to saturate the genetic map with lethal mutations that can be fished out of the genome and characterized. Combining P1 mutagenesis with DNA chip technology, some 300 previously unknown lethals have been found to have homologs among the 80,000 expressed sequence tags that exist in the Drosophila Sequence Tagged Sites (STS) database. Sequence comparison studies have allowed definition of the function of some of these genes.

C.M. Kao, a student of Pat Brown at Stanford, described how yeast expression arrays, developed in the Brown laboratory, are being used to dissect out regulatory pathways and to identify novel genes with important physiological functions. Using glucose starvation as an example, she showed the expression profiles of proteins known to function in this pathway (glycolysis, gluconeogenesis, etc.), and showed how genes of unknown function that show interesting responses, either singly or part of a cohort, can be identified using this technology. Other metabolic shifts will be studied in order to build genome-side models of metabolism in yeast.

Combinatorial Transcription (Back to Top)
The transcription complex is aptly named, as this little cellular machine for turning on specific genes is quite complex, comprised of a small arsenal of proteins—activators and enhancers—which in turn exist in complex families of proteins. It is thus not surprising that determining what exact features of transcription complexes provide the exquisite specificity required for normal development has garnered a great deal of attention. A few of the systems presented at ASBMB follow.

Gerald Crabtree of Stanford described his work dissecting the machinery that transmits a membrane signal to chromatin to effect chromatin decondensation in lymphocytes. No less than eleven proteins are involved with the BAF activation complex as he isolates it from T-lympohcytes, many of them with correlates in transcriptional complexes in yeast. One difference between mammals and yeast, however, is the complexity of the gene family encoding members of this complex. In yeast, each subunit is uniquely encoded in a gene, while in mammals, each subunit is from a family, the members of which can be put together in a combinatorial way. This, Crabtree believes, is the basis of the specificity of transcriptional activation. Crabtree was able to show the effect of a cell-type specific BAF associated factor on T-cell development in knock-out mice—in mice lacking a certain factor (BAF 57) recombination of the T-cell receptor gene and subsequent presentation of the T-cell receptor on the surface were blocked.

Eric Olson of the University of Texas Southwestern Medical School described a complex set of transcription factors that control the development of the three types of muscle cell types in vertebrates, including two cleverly named factors—tinman and tinwoman—that when mutated lead to disruptions in heart development. A combinatorial code of transcription factors was proposed to explain how the various lineages that give rise to a four chambered heart are established during embryogenesis. Evidence was presented that these same factors active during development are reactivated during conditions that lead to cardiac hypertrophy, causing the heart to revert to a fetal pattern of expression.

And That's Not All Folks (Back to Top)
This is just the tip of the iceberg. Literally hundreds of other scientists and laboratories were represented in plenary sessions symposia and poster sessions, many of which were full to over-flowing. Other plenary sessions addressed emerging technologies, protein folding and design, gene therapy to mention but a few. Next year, the society will meet from June 4–8, in Boston.

For more information: American Society for Biochemistry and Molecular Biology, 9650 Rockville Pike, Bethesda, MD 20814-3996. Tel: 301-530-7145. Fax: 301-571-1824. Email: asbmb@asbmb.faseb.org.

By Laura DeFrancesco