Alisa Goldstein received her Ph.D. in genetic epidemiology from the University of California at Los Angeles in 1988 and came to NIH that summer as an IRTA fellow in the Environmental Epidemiology Branch of the NCI Division of Cancer Epidemiology and Genetics. She is now a senior investigator in the Genetic Epidemiology Branch and section chief of the newly formed Population and Statistical Genetics Section.

My interests are the genetic and environmental determinants of cancer. My research at NCI has focused on family and genetic epidemiologic studies of several cancers, including cutaneous malignant melanoma/dysplastic nevi (CMM/DN) and the nevoid basal cell carcinoma syndrome (NBCCS), and on the development and assessment of genetic epidemiologic methods. My studies emphasize the integration of epidemiologic, clinical, and molecular approaches.

The Genetic Epidemiology Branch has a long-standing interest in the study of familial melanoma. I have taken the lead role in the genetic epidemiologic analyses of familial melanoma since 1990. My laboratory colleagues and I have identified and evaluated two melanoma-susceptibility genes (p16/CDKN2A and CDK4). CDKN2A appears to account for one-third of familial melanoma kindreds, whereas CDK4 mutations have been detected in only two kindreds to date. Although the tumor suppressor p16 and proto-oncogene CDK4 are hypothesized to have different functions, we have shown that there are no differences in the ages at CMM diagnosis, number of CMM tumors, or clinical course of disease between kindreds with p16 mutations and those with CDK4 mutations.

Recently, we incorporated genetic data on p16 into clinical and epidemiologic analyses of the CMM kindreds previously screened for CDKN2A mutations, comparing kindreds with and without mutations that impair the function of p16. We showed that the risk of pancreatic cancer was significantly increased only in kindreds with p16 mutations. Genetic factors, such as the kind of mutations found in p16, may explain the inconsistent occurrence of other cancers in melanoma-prone kindreds. To follow up on these findings, we are currently examining the relationship between factors such as sun exposure, clinical features, and p16 and/or CDK4 mutations. My current and future work involves searching for additional melanoma genes, assessing risks of different tumors associated with the various genes, evaluating environmental and clinical risk factors, and examining gene-gene and gene-environment interactions in this complex, heterogeneous, and potentially fatal form of skin cancer that will claim an estimated 7,300 lives this year in the United States.

Another major focus of my research has been on NBCCS, a multisystem disorder with variable expression. My colleagues and I helped localize the NBCCS gene to a small region on chromosome 9q and, last year, identified a candidate gene, PTCH, the human homologue of the drosophila patched gene. We previously examined the relationship between sun exposure and the development of basal cell carcinomas (BCCs) in patients with NBCCS; although sun exposure did not appear necessary for the development of BCCs, it exacerbated BCC development. Also, clinical evaluation of African-American families with NBCCS (in whom the development of BCCs was rare) provided corroborating evidence that sun exposure and skin pigmentation contribute to the expression of BCCs in NBCCS gene carriers. My future work will include examining genotype-phenotype correlations, assessing the interaction of sun exposure, X-irradiation, and skin type, and searching for modifying genes and environmental risk factors. We have recently started a study of patients with medulloblastoma in collaboration with other researchers from NIH and the Children's National Medical Center. We are clinically evaluating patients with medulloblastoma, assessing risks of cancer in family members, examining tumors for mutations in the PTCH, APC, (adenomatous polyposis coli), or other candidate genes and evaluating the relationship between molecular genetic alterations, tumor characteristics, response to treatment, and survival in this cohort.

I am also involved in two new studies. First, colleagues from NCI, NIDR, and Taiwan have started a genetic epidemiologic study of nasopharyngeal carcinoma (NPC) in Taiwan. The goals of the study are to map gene(s) that predispose individuals to NPC, to examine gene-environment interactions, and to evaluate risk-factor differences between multiple-case and single-case families. Second, collaborating researchers from NCI, NHGRI, and Shanxi Province, China, have recently started a genetic epidemiologic study to examine the role and interaction of genetic and environmental factors in the etiology and prevention of cancers of the esopha-gus and stomach.

My group is interested in the fundamental mechanisms cells use to transmit signals across membranes. Our approach is to use X-ray crystallography and other structural methods to understand in atomic detail how signaling proteins work as miniature machines. We choose projects we hope will provide information relevant to large families of homologous proteins, and we always aim for structural data in the presence of activators or substrates in order to learn as much as possible about mechanism. In particular, we are focusing on the regulated production and mode of action of the second messengers diacylglycerol, calcium, and cyclic AMP. My work concerns three of the key protein players in these second-messenger pathways: phospholipase C, protein kinase C, and adenylyl cyclase.

Protein kinase Cs are a family of about a dozen calcium and lipid-activated enzymes with a ubiquitous role in signaling. In what was a technically innovative strategy - entirely determining structure by using phase information from zinc ions - we determined the first crystal structure of a protein kinase C C1 domain bound to its potent activator, the tumor promoter phorbol ester. The structure explains how phorbol ester activates protein kinase C by stabilizing the membrane-inserted form of the C1 domain. Phorbol ester caps a polar groove on the otherwise highly hydrophobic tip of the C1 domain, creating a nearly ideal membrane interaction surface. The structure is a template for understanding more than 50 other C1 domains from protein kinase Cs and many other signaling proteins.

Phospholipase Cs are a family of enzymes that play a key role in signaling downstream of many receptors by generating the second messengers IP3 and diacylglycerol. We determined a structure of phospholipase C-delta 1 that showed how the catalytic and C2 domains work together to dock phospholipase C onto membranes. C2 domains are of great interest because they occur in over 60 proteins besides phospholipase Cs. By analyzing the structure bound to a calcium analogue, we directly characterized a calcium-induced conformational change in a C2 domain for the first time.

Taken together, our findings on protein kinase C and phospholipase C show how two ubiquitous lipid-activator binding domains, C1 and C2, interact with phospholipid bilayers in strikingly different ways. We propose that C1 inserts deeply into the bilayer in pursuit of its hydrophobic ligand and probably serves as a high-affinity membrane anchor. In contrast, we think that C2 binds polar anionic headgroups at the bilayer surface. The rigidity of C2's interactions with other protein domains suggests that C2 could have an important role in governing the stereochemistry of multidomain protein-membrane complexes.

Adenylyl cyclases are the classic effector of receptor-mediated signal transduction, producing cyclic AMP in response to hormone stimulation. Mammalian adenylyl cyclases are complex integral membrane proteins, but their catalytic domains are soluble. We solved the first structure of a soluble catalytic domain from adenylyl cyclase and showed that it is a dimer that looks like a Christmas wreath. The structure was solved in the presence of a potent small molecule activator, forskolin, which binds at the dimer interface and helps "glue" the wreath together. The active site is in the center of the wreath, while regulatory sites that bind proteins are on the outside of the wreath. This permits many different regulatory mechanisms to operate at the same time and allows "mix-and-match" regulation of different adenylyl cyclase isoforms.

Using site-directed mutagenesis and biophysical techniques, my lab is currently exploring the role of the C2 domain in phospholipase C activation; we're using computer modeling to understand the array of different adenylyl and guanylyl cyclases based on the structure we've solved. We are also working on structures of several other proteins in phosphoinositide and cyclic nucleotide signaling - and I'm always interested to hear from NIH colleagues who work on signaling proteins that present new structural questions.

Klaus Strebel received his Ph.D. in microbiology from the University of Heidelberg, Germany, in 1985. He joined the NIAID Laboratory of Molecular Microbiology in 1986 and is currently a microbiologist in that lab.

From a molecular virologist's point of view, viruses represent comparatively simple model systems in which to study fundamental biological and biochemical mechanisms. Despite the relative simplicity of their genomes, virus replication is a complex process that depends heavily on the activity of host cellular factors. Using HIV as a model system, my group is interested in studying the function of viral proteins and their interaction with host cell machinery. My work to date has focused on two HIV accessory proteins, Vif and Vpu, which function in early and late stages of virus replication. The experiments in my lab involve a wide variety of biochemical, virological, and immunocytochemical techniques, and we hope our studies will not only contribute to the general understanding of retrovirus function, but also provide a basis for the assessment of viral proteins as potential antiviral targets.

When I joined the LMM in 1986, my initial work involved the functional characterization of the HIV-1 Vif protein. I was one of the first to show that Vif has a crucial function in regulating viral infectivity. However, despite the dramatic impact of Vif on virus replication in human lymphocytes and macrophages, the precise biochemical function of Vif is still unclear. One of our main efforts is aimed at the identification of viral or cellular targets for Vif. We recently demonstrated the stable association of Vif with the core of HIV particles, a finding that suggests a role for Vif as a virion component. In addition, we observed a striking association of Vif with intermediate filaments, in particular vimentin, in virus-producing cells. The association of Vif with vimentin results in a reversible, microtubule-dependent perinuclear aggregation of intermediate filament networks. We are currently investigating the possible role of vimentin association of Vif with its role in regulating viral infectivity. In particular, we are investigating the possibility that the interaction of core-associated Vif with the intermediate filament network assists in nuclear targeting of preintegration complexes following virus entry into a target cell.

In a separate line of research, we have recently identified a phenotypically dominant (transdominant) Vif mutant that interferes with HIV replication in permissive cells. Permissive cells normally do not require Vif function, most likely because of the presence of a cellular Vif-like factor. We postulate that transdominant Vif is capable of interfering not only with the function of wild-type Vif but, in addition, with the activity of a putative cellular Vif-like factor in permissive cells as well.

Interestingly, transdominant Vif protein has a reduced affinity for vimentin but is incorporated into virions at increased levels. We are currently investigating the molecular basis of the transdominant effect of the mutant Vif protein and its potential as an antiviral compound.

In 1987, I identified a novel protein, Vpu, encoded exclusively by HIV-1. Since then, we have been characterizing this protein to understand its function. We know now that Vpu is an integral membrane protein that regulates the release of virus from the cell membrane and, independently, causes degradation of the HIV receptor, CD4. While HIV-2 lacks a Vpu gene, we recently found that this virus nonetheless expresses a similar Vpu-like activity, encoded by its Env glycoprotein, to regulate virus release. The ability to induce CD4 degradation, however, is unique to Vpu. CD4 degradation is a multistep process that occurs in the endoplasmic reticulum (ER), is energy-dependent, and involves the physical interaction between Vpu and CD4. Amino acids critical for this function of Vpu are located in its cytoplasmic domain. In contrast, regulation of virus re
lease by Vpu requires the Vpu transmembrane domain, is regulated from a post-ER compartment, and correlates with an ion-channel activity of Vpu. No such information is currently available for HIV-2 Env; however, because Vpu and HIV-2 Env can functionally complement each other, it is likely that regulation of virus release is based on similar mechanisms for both viruses.

Future efforts will focus on the detailed characterization of both the mechanisms of CD4 degradation and the regulation of virus release. From those studies, we expect to gain insights into general principles of protein degradation in the ER and the mechanisms involved in the late stages of virus production.

Robert Tycko received his Ph.D. in chemistry from the University of California at Berkeley in 1984. After postdoctoral research at the University of Pennsylvania, he became a member of the technical staff at AT&T Bell Laboratories in 1986. He joined the Laboratory of Chemical Physics of NIDDK as an investigator in 1994.

I have worked in the area of solid-state nuclear magnetic resonance (NMR) spectroscopy since 1980. Before I moved to NIH in 1994, my research was primarily in physical chemistry and condensed-matter physics. Among other things, my group at AT&T Bell Laboratories discovered and studied unusual molecular rotational motions in the molecular forms of carbon called "buckyballs" (Science magazine's Molecule of the Year in 1991), measured electronic properties of superconductors derived from buckyballs that are important to our understanding of the physical basis of superconductivity, and used a novel technique called "optically pumped NMR" to obtain the first experimental evidence for peculiar states of electrons, called "skyrmions," in thin films of semiconductors at very low temperatures. We also developed a method, based on complicated radio-frequency pulse sequences, for obtaining high-field NMR spectra of solid samples that look as if the spectra were taken in zero magnetic field. Such "zero-field" spectra exhibit much higher resolution than do ordinary solid-state NMR spectra and permit the direct measurement of interatomic distances in disordered solids.

When I moved to NIH, I changed the focus of my research to problems in biophysics and structural biology. Biomolecular NMR spectroscopy is a huge and mature field, in part due to many important contributions from my colleagues in the Laboratory of Chemical Physics. However, most activity in this field is liquid-state NMR, meaning that it is restricted to biopolymers in isotropic solutions. Biomolecular solid-state NMR is currently a small field at a relatively early stage of development, with less than 20 active groups worldwide. Current solid-state NMR techniques yield local structural information, such as interatomic distances or bond angles at specific isotopically labeled sites in a biopolymer, rather than complete, global-structure information. But solid-state NMR measurements can be carried out on biopolymers in noncrystalline solid or fibrous form, in frozen solutions, in membrane-bound form, and in disordered or unfolded states. These measurements can therefore provide atomic-level structural information about systems that cannot be characterized by existing liquid-state NMR or diffraction techniques.

In recent work, we developed a new approach to the determination of peptide and protein backbone conformations, or local secondary structure, that uses two-dimensional (2D) spectroscopy in combination with a solid-state NMR technique called "magic-angle spinning" (MAS). Our 2D MAS methods provide structural information in the form of the relative orientations of isotopically labeled chemical groups, for example ¹³C-labeled carbonyl groups in a peptide backbone, rather than in the more traditional form of internuclear distances. In general, these methods will be useful as a means of testing models for protein structures and as a means of characterizing the conformations of partially structured proteins and protein-folding intermediates. We are currently pursuing several specific applications. One is in a structural study of a peptide derived from the V3 loop of the HIV-1 envelope glycoprotein gp120, bound to an anti-gp120 antibody. Our 2D MAS measurements so far indicate that the conserved GPGR motif in the middle of the V3 loop does not adopt a type II b-turn conformation that had been predicted by other researchers. A second application involves an investigation of the conformational distributions of a 17-residue helix-forming peptide, originally designed by the group of R.L. Baldwin at Stanford, with the goal of addressing a current controversy regarding the precise nature of the helical structures that such peptides adopt. Our solid-state NMR measurements on frozen solutions show that when the helix content is high (i.e., when helical conformations predominate over random-coil conformations), a-helical conformations are strongly preferred over 3₁₀ helical conformations but that 3₁₀ helical conformations may become more common when the helix content is reduced by the addition of a denaturant. These results contradict suggestions in the literature that were based on electron-spin resonance measurements by others, but they are generally in line with recent theoretical work.

Our recent work has begun to demonstrate and exploit the potential of solid-state NMR in biophysics and structural biology. I plan to continue our research in these areas, and I look forward to developing new collaborations and interactions with other intramural scientists in the months and years to come.

 

DNA-Sequencing Center Soon To Materialize Plans for a shared intramural DNA-sequencing center have gotten the go-ahead from the Board of Scientific Directors and the co-sponsorship thus far of 14 institutes with NHGRI as the lead institute. The center will use automated equipment and specially trained staff for large-scale sequencing. It's designed to meet acute needs for significant sequencing throughput. NHGRI will shoulder the largest share of the costs of building the center. Participating institutes will contribute to the center's start-up over the next two years and will be charged on a fee-for-service basis for sequencing projects. Only investigators in participating ICDs will have access. NCI will be constructing an adjacent sequencing facility for its research. In addition to sequencing services, the new center will provide informatics expertise for exploiting the data that come pouring out, including state-of-the-art computational tools for DNA-sequence analysis. The fee for service - for both sequencing and informatics - should fall below that for commercial sequencing alone. The new center will not be responsible for small or routine sequencing projects that can be handled by individual labs or contractors. A governing board of scientific directors or their designates will oversee the center, and a panel of working intramural scientists will prioritize project proposals. Eric Green (<egreen@nhgri.nih.gov>) will direct the facility, which should be operational by late summer or early fall.