Biography of Philip A. Beachy

From an early interest in science and medicine through a subsequent period of enthusiasm for the cello, Philip A. Beachy ultimately developed a passion for music of a different sort—the rhapsody of molecular events that direct the early development of all animals. Beachy, a professor of Molecular Biology and Genetics at Johns Hopkins University School of Medicine (Baltimore, MD) and an investigator of the Howard Hughes Medical Institute, was elected to the National Academy of Sciences in 2002 for his pioneering work with genes expressed during embryonic development. A few years prior, Beachy received the National Academy of Sciences Award in Molecular Biology for “studies of a developmental morphogen, its processing and structure, and its covalent attachment to cholesterol.”

Focused primarily on the Hedgehog (Hh) signaling pathway, Beachy's work has uncovered surprising aspects of Hh signaling, including modification of the Hh protein by cholesterol (1); involvement of Hh in the birth defect holoprosencephaly (2); and Hh involvement in a number of cancer types (3–6). In his Inaugural Article in this issue of PNAS, Beachy and colleagues (7) provide further details about the molecular mechanisms that regulate Hh signaling. In Drosophila, they found that the transmembrane protein Smoothened (Smo) is extensively phosphorylated and activated upon Hh stimulation. This phosphorylation may subsequently play a modulatory role in Smo activation, Hh pathway activity, and phosphorylation of other pathway components in Drosophila.

From Puerto Rico to the American Midwest

Beachy spent 8 of his early formative years in the hills of central Puerto Rico, where his father pastored a rural church. He attended a Spanish-language school during the day and learned to read and write English in the evening at home. When he was 9 years old, Beachy's family returned to their home base in the small town of Goshen, IN, where Beachy attended public schools to complete his secondary education. At 16, staying close to home, Beachy entered Goshen College (Goshen, IN), a small private liberal arts college in his hometown. “Unlike many people who knew they were going to be scientists from a very early age, I didn't decide that I would try to become a scientist until fairly late on in college,” he says.

At Goshen, Beachy's interest in music blossomed. Having taken up the cello in high school, he studied throughout college even though he majored in natural sciences and not music. He played in local chamber groups and orchestras and eventually performed a senior recital and a concerto with the school orchestra. Beachy was briefly attracted to the idea of a career in music performance, but the possibility of a scientific career in research also began to emerge. “What I loved most was playing chamber music, but the path to a professional career in performance just was not clear,” he says. “As I began to get interested in research, music began to seem more like a great avocation than a profession.” Beachy spent a summer working on fungal growth in the biology department at Goshen College. This was followed by an additional year of research after graduation, during which he completed several chemistry courses offered at the nearby South Bend campus of Indiana University.

Figure 1.

Philip A. Beachy

One decisive factor in Beachy's decision to pursue a life of research came in the pages of The New Yorker magazine. Upon reading excerpts from The Eighth Day of Creation, a book by Horace Freeland Judson (8) that recounted the birth of molecular biology, Beachy became hooked on science. “It was a scholarly account but very well written and entertaining,” he says. Beachy also read Scientific American articles on recombinant DNA and transposable genetic elements (9, 10) written by Stanley Cohen in the Department of Genetics at Stanford University (Stanford, CA), and this further propelled him into a career in science. Says Beachy, “That got me very interested in modern genetic research.” This interest led Beachy to apply to several graduate programs, including Stanford's Department of Genetics. Lured in equal parts by the academic environment, temperate climate, and progressive culture of the San Francisco Bay Area, he chose Stanford. “It was quite a change from the Midwest and northern Indiana.”

Leaving Genetics for Biochemistry

Although happy with his choice of attending Stanford, Beachy soon realized that he was more interested in the work going on in the Department of Biochemistry than in his Department of Genetics. “I was captivated by lectures of David Hogness in one of the biochemistry courses,” Beachy says. Hogness, a pioneer in recombinant DNA technology, was working on the molecular biology of Drosophila development. Hogness and colleagues were studying the genetic mutations that caused errors in segmental development in the fly, errors that led to extra sets of wings or legs. These phenotypes were caused by mutations in a region of the fly chromosome called the bithorax complex, which houses genes important for specifying segment identity.

To hear more about Hogness' work, Beachy—an impoverished graduate student at the time who could not afford to pay conference registration fees—snuck into a scientific meeting in San Francisco where Hogness was presenting his work. “To hear about these mutant phenotypes and that the DNA encoding these genes was being isolated—that was just fabulous,” he says. “The prospect of actually being able to study the products of genes with such striking mutant phenotypes was exhilarating. I just had to do that!”

After 9 months in the Department of Genetics, Beachy transferred to Hogness' laboratory in the Department of Biochemistry and began to work on homeotic genes, principally Ultrabithorax (Ubx), a component of the bithorax complex. By making fusion proteins and raising antibodies to these proteins, Beachy was able to show that the proteins encoded by Ubx genes were localized to the nucleus (11). Years earlier, Ed Lewis, who shared the 1995 Nobel Prize for his studies of the genes involved in embryonic development, had suggested that the genes of the bithorax complex were master regulators that controlled the expression of other genes. “To see that these proteins localized in the nucleus in the segments where the genetics said they ought to be was very exciting and confirmed an important feature of Lewis' model,” says Beachy. “That was my earliest real `Eureka!' moment.”

Because the hypothesis at the time proposed that these bithorax complex genes encoded regulatory proteins, the next challenge was to figure out whether these proteins bound DNA and to what sequences they bound specifically. Although Beachy demonstrated that Ubx proteins bound DNA specifically (12), the article detailing this would not be published until well after he graduated in 1986. “David [Hogness] didn't like to pay attention to the realities of the outside world. Publishing wasn't his motivation,” explains Beachy.

Although publication urgency was not a quality that Hogness conveyed, Beachy did find his intense focus on the science influential. “It was the inherent interest in the work that kept things going,” says Beachy. “So that's something I think he communicated to many people, and it's something I'm very grateful for.” Hogness also encouraged autonomy among his students, a value that has echoed throughout Beachy's independent research career. Says Beachy, “Because of the way David ran his lab, there was a lot of independence of thought and action.”

Hunting Hedgehog

That search for independence led Beachy to accept a junior research position in the Department of Embryology at the Carnegie Institution of Washington (Baltimore, MD). At Carnegie, Beachy continued working on homeotic genes, searching for target genes of Ubx. A colleague at Carnegie, Allan Spradling, had developed methods for using β-galactosidase activity as a marker to screen Drosophila genes based on their tissue expression patterns. These enhancer-trap Drosophila lines carried insertions of a modified transposable element, each line having a different and independent site of insertion. While sifting through Spradling's rejected lines in 1987, Beachy found a number of potentially interesting genes whose segmentally modulated expression patterns suggested regulation by Ubx or other homeotic genes. In addition, Beachy found one particular strain, a “striper,” that expressed β-galactosidase in a strip of cells in every segment. Although this was of great interest because of potential function in Drosophila segmentation, Beachy did not have the laboratory manpower to study this area and put the striper studies on hold, concentrating instead on regulatory targets of homeotic genes.

Fortunately, his striper studies did not have to wait long. A few days before he was to give a seminar at Johns Hopkins University, his hosts at Johns Hopkins' Department of Molecular Biology and Genetics called to ask whether he would mind if the seminar were considered a job interview. “That was a surprise, coming scarcely a year after leaving graduate school, and I wasn't quite sure I was ready for a full faculty job,” admits Beachy. However, he agreed to the interview and was soon offered a faculty position in the department. At Johns Hopkins, Beachy's laboratory grew in size and resources, and, in 1990, he was able to begin looking more seriously at the striper line.

“We began to look at this enhancer-trap line and figured out that the P element was located near the Hedgehog gene,” says Beachy. The Hedgehog gene (hh) is important in segmentation, and by using the P element to clone DNA from the region and excising it to make additional mutations, Beachy's group cloned and characterized the Drosophila hh gene. They demonstrated that hh encoded a secreted protein (13). Beachy also found that the protein product Hh underwent autoproteolysis to yield two separate fragments and determined that the carboxyl-terminal sequences were responsible for carrying out the cleavage (14). In a subsequent article, Beachy showed that the amino terminal fragment, which largely remains bound to the cells that produce it, was the active species in signaling (15).

This finding marked a turning point in Beachy's research pursuits. He says, “As Hh waxed, homeotics waned in my lab, until over the years, everyone in my lab was working on Hh and related questions.” In parallel with the Drosophila studies, Beachy and others had cloned homologues of the Hh gene from vertebrates, where the genes constitute a multigene family, and found similar biochemistry. Making the move into vertebrate development and utilizing the insights initially derived from studies of Hh processing in Drosophila, Beachy and colleague Thomas Jessell at Columbia University (New York, NY) found that exposing naïve, undifferentiated neural plate explants from chick embryos to varying concentrations of a recombinant Hh protein [the homologue Sonic hedgehog (Shh)] produced different outcomes depending on the concentrations. The highest concentrations produced the cell types that were closest to the midline, whereas lower concentrations produced other cell fates more distant from the midline (16).

In a separate experiment, Beachy collaborated with John Fallon at the University of Wisconsin (Madison, WI) to show that recombinant Shh protein also was capable of reprogramming the pattern of the developing limb (17). “This looked like Shh was functioning as a morphogen, that is, as a graded signal produced by a discrete source that elicits concentration-dependent responses in cells of the surrounding tissues,” he says. “This, it turns out, is how Shh is working in the developing neural tube, the developing limb, and in a number of other tissues.”

One thing still puzzled Beachy about the Hh protein: if the amino terminus does all the signaling, then why make it as a precursor that has to be cleaved to produce an active signal? The reason, it turned out, was that something else was happening to the Hh protein in addition to being cleaved. Beachy and colleagues (1) found that, in concert with the cleavage reaction, cholesterol was added to the amino terminus. “To find that a secreted protein actually had a cholesterol present and that this was added through an autocatalytic reaction was really a huge surprise to us and others,” he says. “The covalent addition of cholesterol, and of a second lipid moiety, palmitate (18), greatly influences the potency and distribution of the Hh protein within tissues and is therefore critical in regulating Hh patterning activity.”

Vertebrate Studies

Soon Beachy began to study vertebrate models of development. In collaboration with Jeffry Corden at Johns Hopkins and Heiner Westphal at the National Institutes of Health (Bethesda, MD), Beachy's laboratory produced a mutation of the mouse Shh gene by targeted recombination. The Shh knockout caused a phenotype with several profound patterning defects (2). Says Beachy, “This is a spectacular mutant because it got pretty far in development, far enough so you could see the pattern defects, and the pattern defects are quite remarkable.” The most striking of these patterning defects was cyclopia, in which a single eye develops in the center of the face. In humans, this rare condition is called holoprosencephaly. These mutant mice also showed defects in patterning of the brain, spinal cord, internal organs, and limbs.

One of the most surprising connections revealed by the Shh mutant phenotype was to an outbreak of cyclopia that occurred in sheep in the 1950s. Pregnant ewes that had grazed on a plant, the corn lily (Veratrum californicum), gave birth to lambs with cyclopia. The closely related plant compounds cyclopamine and jervine were found to be the cause of the defect. “When we saw this great similarity between our Shh knockout mouse and the effects of cyclopamine, we thought surely cyclopamine was affecting Hh signaling,” says Beachy. He proposed that these plant compounds might exert their effect by interfering with processing and cholesterol modification. He exposed chick embryos to these compounds and observed the same defects. However, the chemicals had no effect on sterol modification but inhibited the ability of target tissues to respond to Shh (19).

Having ruled out production of the Hh signal, Beachy and colleagues considered the cellular components that mediate response to the Hh signal, including proteins encoded by the Smoothened (Smo) and Patched (Ptc) genes. In the absence of the Hh signal, Ptc regulates Smo activity and keeps the pathway in a state of quiescence (20). Hh protein, when present, binds to Ptc and blocks its activity, which in turn allows Smo to become activated. Activated Smo then relays the signal from the cell membrane to other components within the cell. Beachy and colleagues (21, 22) found that cyclopamine disrupts this chain of events by binding directly to Smo and preventing its activation.

Cancer Connections

Around the time Beachy was working on the Shh knockout, a connection emerged between the Hh signaling pathway and cancer. Several groups demonstrated that a human condition, basal cell nevus syndrome, was associated with heterozygous Ptc mutations. People with this syndrome were known to have a predisposition to certain kinds of cancers (basal cell carcinoma, medulloblastoma, and rhabdomyosarcoma). Upon loss of the remaining intact Ptc allele, Smo activity would be continually activated, triggering cell division and facilitating tumor formation in certain cell types.

Beachy's finding that cyclopamine directly inhibited Smo activity suggested that cyclopamine might be useful in treating certain kinds of cancers because it could block pathway activity. Indeed, in subsequent studies, Beachy and colleagues (3) used cyclopamine to demonstrate a role for pathway activity in growth of cancer types such as medulloblastoma. In addition, cyclopamine treatment revealed a requirement for growth of an expanding number of cancer types not typically associated with basal cell nevus syndrome that include small cell lung cancer (4), gastrointestinal cancers (5), and prostate cancer (6).

Beachy and his colleagues had made a nonobvious connection: the same or similar compounds that causes severe birth defects such as cyclopia may actually be useful in treating various cancers.

Beachy's current research studies in cancer focus on the relationship between abnormal Hh pathway activity in cancer growth and normal activity of the pathway in the regeneration and repair of injured tissues (23, 24). “Many types of cancer occur at much higher frequency in the setting of chronic injury and inflammation, and cancer in many ways resembles a state of continuous tissue repair,” he says. “One of these similarities is continuous activity of signaling pathways like Hh.”

Smoothened Return to Drosophila

In his PNAS Inaugural Article (7), Beachy returns to the question of how cells sense the Hh signal in Drosophila and how this signal is transmitted from the membrane into the cell. He and his team found that Hh stimulation leads to phosphorylation of Smo. “There are 26 residues that are phosphorylated on the cytoplasmic tail of Smo. This degree of phosphorylation is, to my knowledge, unprecedented in membrane-associated signaling proteins,” he says. Beachy's group systematically mutagenized all 26 residues and looked for effects on activity of the pathway. In a subset of these residues collectively changed to glutamate (which mimics phosphorylation), Smo was constitutively activated without the need for Hh stimulation. In contrast, when the residues were changed to alanine (which prevents phosphorylation), Smo was incapable of being activated.

“In analyzing this pattern of phosphorylation, we've come across a switch which appears to regulate the activity state of Smo,” explains Beachy. “What we don't yet know, which will be the subject of future investigation, is the mechanism by which that occurs. That is, how does this phosphorylation affect Smo activity?” In predicting future findings, Beachy is “betting it's a conformational change in the cytoplasmic tail of Smo, which may then influence how it interacts with downstream components of the pathway.”

Currently, Beachy and his colleagues are working to identify phosphoresidues in a number of other downstream components, and to understand how phosphorylation of these components is involved in regulation of their activity and interactions with other pathway components. This information will shed light on the molecular mechanisms involved in transduction of the Hh signal across the membrane and may help in devising novel approaches to manipulation of the pathway for therapeutic purposes.

When Beachy reflects on finding himself returning to the realm of medicine through the link between development and cancer, he states, “For me it's been a great set of unexpected connections—from profound effects during development and surprising molecular findings like autoprocessing, the involvement of cholesterol, and the action and mechanism of cyclopamine, then ultimately to disease, malformation, and cancers. And none of it predictable from the starting point.”