Today’s graduate students, like those showcased in our Winter 2015 issue, often become tomorrow’s scientific leaders. The careers of France Córdova (PhD ’79), Arati Prabhakar (MS ’80, PhD ’85), and Ellen Williams (PhD ’81), offer dramatic examples of how true that can be.

These women now lead three of the nation’s top science, technology, and research agencies: the National Science Foundation (NSF), the Defense Advanced Research Projects Agency (DARPA), and the Advanced Research Projects Agency-Energy (ARPA-E).

France Córdova

France Córdova
France Córdova

Since 2014, Córdova has led the NSF, a $7 billion-a-year federal agency that supports fundamental research and education in all the nonmedical fields of science and engineering.

Córdova studied physics as a graduate student at Caltech, working on X-ray astronomy. It was a time she remembered in a recent interview in the Caltech Alumni Association’s publication, Techer, as “rigorous, collaborative, and fun. … As graduate students, you were able to learn from and work right alongside all of these incredible minds, like theoretical physicists Murray Gell-Mann and Richard Feynman.”

After Caltech, Córdova built an impressive resume that included working for a decade at Los Alamos National Laboratory; leading the department of astronomy and astrophysics at Pennsylvania State University; and becoming the youngest person and first woman to hold the position of NASA chief scientist. Over those and subsequent years, the positions she held shifted from those focused primarily on research to more administrative roles, eventually including vice chancellor for research at UC Santa Barbara, chancellor of UC Riverside, president of Purdue University, and chair of the Board of Regents of the Smithsonian Institution before being named as the NSF’s director.

Arati Prabhakar

Arati Prabhakar
Arati Prabhakar

Prabhakar serves as director of DARPA—an agency of the U.S. Department of Defense that develops emerging technologies for use by the military and whose achievements include the creation of ARPANET, the precursor to the Internet.

Prabhakar first joined DARPA in 1986 after receiving her doctorate in applied physics from Caltech. Her initial job with the agency was to manage programs in advanced semiconductor technology and flexible manufacturing, and to manage demonstration projects to insert new semiconductor technologies into military systems.

She discussed how Caltech prepared her for that role in a 2011 interview with ENGenious, a publication of the Division of Engineering and Applied Science. In that interview, she said that “having a very solid technical foundation really helped with judgments I had to make in my career. … I was investing in people that I thought were going to make big leaps forward in technology. I wasn’t in the lab doing the work, but I was trying to exercise good judgment about where real breakthroughs might come from. That wouldn’t have been possible without the solid technical foundation I received at Caltech.”

In 1993, President Bill Clinton named Prabhakar director of the National Institute of Standards and Technology, a post she held until 1997, when she stepped down to pursue entrepreneurial interests in the Silicon Valley, funding and managing engineers and scientists to create new technologies and businesses. She returned to DARPA, this time as its director, in 2012.

Prabhakar appears in a 2015 video describing DARPA’s mission here.

Ellen Williams

Ellen Williams
Ellen Williams

Since 2014, Williams has served as director of the Advanced Research Projects Agency-Energy (ARPA-E), a federal agency modeled after DARPA and tasked with promoting and funding research and development of advanced energy technologies.

In 2014, as part of the kickoff to President Thomas F. Rosenbaum’s inauguration, she participated in a panel discussion at Caltech on “Science and the University-Government Partnership,” in which she described ARPA-E’s job as similar to that of a stockbroker, putting money into investments—in this case technologies—that will perform solidly but also rounding out the portfolio with riskier investments that nonetheless “have the potential to really win big.”

She said, “We have to take some risks [because] traditionally something like 20 percent of the initial investment of a technology portfolio will give 80 percent of the benefits—you just don’t know which are the 20 percent.”

Prior to joining ARPA-E, Williams served as the senior adviser to the United States Secretary of Energy and as the chief scientist for BP, where she was responsible for the company’s long-range scientific plans and activities as well as its major university research programs around the world.

Before working in industry and for the federal government, Williams built a 30-year career in academia, conducting research in nanoscience. She joined the faculty at the University of Maryland shortly after receiving her doctorate in chemistry from Caltech in 1981 and is currently on leave from her position of Distinguished University Professor in the Department of Physics and the Institute of Physical Science and Technology there.

Taking Dinosaur Temperatures with Eggshells (+)

Caltech geochemist John Eiler has a knack for finding novel scientific niches to investigate, as we reported in our Fall 2015 article “Ready, Set, Explore.”

Now, he and Rob Eagle, a former Caltech postdoctoral scholar now at UCLA, have measured the body temperatures of a wide range of dinosaurs, providing insight into how the animals may have regulated their internal heat.

In the October 13, 2015, issue of the journal Nature Communications, the pair described how they used the analysis of isotopic ratios to reveal the way in which sauropods—a group that includes some of the biggest dinosaurs ever to live—performed the basic task of balancing their energy intake and output.

Read on to learn more about how they examined dinosaurs’ metabolisms and uncovered one of the dinosaurs’ biggest secrets.

Featured image: A large clutch of titanosaur eggs that has been cleaned for research.
Credit: Luis Chiappe, LA County Natural History Museum

Freezing a Bullet (+)

X-Ray Vision, an article in our Spring 2015 issue, examined the central role Caltech has played in developing a powerful technique for revealing the molecular machinery of life. In May, chemist André Hoelz, who was featured in the article, published a new paper describing how he used the technique to reveal how protein-synthesizing cellular machines are built.

Crystal structure of the assembly chaperone of ribosomal protein L4 (Acl4) that picks up a newly synthesized ribosomal protein when it emerges from the ribosome in the cytoplasm, protects it from the degradation machinery, and delivers it to the assembly site of new ribosomes in the nucleus.   Credit: Ferdinand Huber/Caltech
Crystal structure of the assembly chaperone of ribosomal protein L4 (Acl4) that picks up a newly synthesized ribosomal protein when it emerges from the ribosome in the cytoplasm, protects it from the degradation machinery, and delivers it to the assembly site of new ribosomes in the nucleus.
Credit: Ferdinand Huber/Caltech

Ribosomes are vital to the function of all living cells. Using the genetic information from RNA, these large molecular complexes build proteins by linking amino acids together in a specific order. Scientists have known for more than half a century that these cellular machines are themselves made up of about 80 different proteins, called ribosomal proteins, along with several RNA molecules and that these components are added in a particular sequence to construct new ribosomes, but no one has known the mechanism that controls that process.

Now researchers from Caltech and Heidelberg University have combined their expertise to track a ribosomal protein in yeast all the way from its synthesis in the cytoplasm, the cellular compartment surrounding the nucleus of a cell, to its incorporation into a developing ribosome within the nucleus. In so doing, they have identified a new chaperone protein, known as Acl4, that ushers a specific ribosomal protein through the construction process and a new regulatory mechanism that likely occurs in all eukaryotic cells.

The results, described in a paper that appears online in the journal Molecular Cell, also suggest an approach for making new antifungal agents.

The work was completed in the labs of André Hoelz, assistant professor of chemistry at Caltech, and Ed Hurt, director of the Heidelberg University Biochemistry Center (BZH).

“We now understand how this chaperone, Acl4, works with its ribosomal protein with great precision,” says Hoelz. “Seeing that is kind of like being able to freeze a bullet whizzing through the air and turn it around and analyze it in all dimensions to see exactly what it looks like.”

That is because the entire ribosome assembly process—including the synthesis of new ribosomal proteins by ribosomes in the cytoplasm, the transfer of those proteins into the nucleus, their incorporation into a developing ribosome, and the completed ribosome’s export back out of the nucleus into the cytoplasm—happens in the tens of minutes timescale. So quickly that more than a million ribosomes are produced per day in mammalian cells to allow for turnover and cell division. Therefore, being able to follow a ribosomal protein through that process is not a simple task.

Hurt and his team in Germany have developed a new technique to capture the state of a ribosomal protein shortly after it is synthesized. When they “stopped” this particular flying bullet, an important ribosomal protein known as L4, they found that its was bound to Acl4.

Hoelz’s group at Caltech then used X-ray crystallography to obtain an atomic snapshot of Acl4 and further biochemical interaction studies to establish how Acl4 recognizes and protects L4. They found that Acl4 attaches to L4 (having a high affinity for only that ribosomal protein) as it emerges from the ribosome that produced it, akin to a hand gripping a baseball. Thereby the chaperone ensures that the ribosomal protein is protected from machinery in the cell that would otherwise destroy it and ushers the L4 molecule through the sole gateway between the nucleus and cytoplasm, called the nuclear pore complex, to the site in the nucleus where new ribosomes are constructed.

“The ribosomal protein together with its chaperone basically travel through the nucleus and screen their surroundings until they find an assembling ribosome that is at exactly the right stage for the ribosomal protein to be incorporated,” explains Ferdinand Huber, a graduate student in Hoelz’s group and one of the first authors on the paper. “Once found, the chaperone lets the ribosomal protein go and gets recycled to go pick up another protein.”

The researchers say that Acl4 is just one example from a whole family of chaperone proteins that likely work in this same fashion.

Hoelz adds that if this process does not work properly, ribosomes and proteins cannot be made. Some diseases (including aggressive leukemia subtypes) are associated with malfunctions in this process.

“It is likely that human cells also contain a dedicated assembly chaperone for L4. However, we are certain that it has a distinct atomic structure, which might allow us to develop new antifungal agents,” Hoelz says. “By preventing the chaperone from interacting with its partner, you could keep the cell from making new ribosomes. You could potentially weaken the organism to the point where the immune system could then clear the infection. This is a completely new approach.”

Co-first authors on the paper, “Coordinated Ribosomal L4 Protein Assembly into the Pre-Ribosome Is Regulated by Its Eukaryote-Specific Extension,” are Huber and Philipp Stelter of Heidelberg University. Additional authors include Ruth Kunze and Dirk Flemming also from Heidelberg University. The work was supported by the Boehringer Ingelheim Fonds, the V Foundation for Cancer Research, the Edward Mallinckrodt, Jr. Foundation, the Sidney Kimmel Foundation for Cancer Research, and the German Research Foundation (DFG).

—Kimm Fesenmaier

The Quiet Man Who Helped Make the Noyes (+)

A painting and plaque honoring Arthur Amos Noyes hangs in the entrance of the building that bears his name.
A painting and plaque honoring Arthur Amos Noyes hang in the entrance of the building that bears his name.


Our Summer 2015 issue’s examination of all things noise omitted one homophonic subject also of interest to the Institute—the Arthur Amos Noyes Laboratory for Chemical Physics, built in honor of and named for the chair of the Division of Chemistry and Chemical Engineering from 1926 to 1936.

Although named for Noyes, the building was actually built with funds provided by Caltech alum Chester F. Carlson. Carlson graduated in 1930 with a degree in physics and went on to invent an electrophotographic process that came to be called “xerography”—Greek for “dry writing.” (Carlson described the xerographic process in a 1940 issue of E&S.)

The self-effacing inventor and philanthropist, who gave away more than $100 million to foundations and universities, requested that his contribution to the building remain anonymous until after his death.

Some 21 years after his 1968 death, Carlson was remembered by the U.S. Postal Service by a stamp issued in his honor to commemorate the 50th anniversary of his invention. (See third page of the linked PDF.)

The 92,000-square-foot Noyes Laboratory, opened in 1967, bears plaques honoring Carlson and Noyes.

The Noyes building honors Chester F. Carlson, who was instrumental in its creation, with this plaque bearing his likeness.

Caltech Biochemist Sheds Light on Structure of Key Cellular ‘Gatekeeper’ (+)

Credit: Lance Hayashida/Caltech Office of Strategic Communications and the Hoelz Laboratory/Caltech
Click image to enlarge
Credit: Lance Hayashida/Caltech Office of Strategic Communications and the Hoelz Laboratory/Caltech

Facing a challenge akin to solving a 1,000-piece jigsaw puzzle while blindfolded—and without touching the pieces—many structural biochemists thought it would be impossible to determine the atomic structure of a massive cellular machine called the nuclear pore complex (NPC), which is vital for cell survival.

But after 10 years of attacking the problem, a team led by André Hoelz, assistant professor of chemistry, recently solved almost a third of the puzzle. The approach his team developed to do so also promises to speed completion of the remainder.

In an article published online February 12 by Science Express, Hoelz and his colleagues describe the structure of a significant portion of the NPC, which is made up of many copies of about 34 different proteins, perhaps 1,000 proteins in all and a total of 10 million atoms. In eukaryotic cells (those with a membrane-bound nucleus), the NPC forms a transport channel in the nuclear membrane. The NPC serves as a gatekeeper, essentially deciding which proteins and other molecules are permitted to pass into and out of the nucleus. The survival of cells is dependent upon the accuracy of these decisions.

Understanding the structure of the NPC could lead to new classes of cancer drugs as well as antiviral medicines. “The NPC is a huge target of viruses,” Hoelz says. Indeed, pathogens such as HIV and Ebola subvert the NPC as a way to take control of cells, rendering them incapable of functioning normally. Figuring out just how the NPC works might enable the design of new drugs to block such intruders.

“This is an incredibly important structure to study,” he says, “but because it is so large and complex, people thought it was crazy to work on it. But 10 years ago, we hypothesized that we could solve the atomic structure with a divide-and-conquer approach—basically breaking the task into manageable parts—and we’ve shown that for a major section of the NPC, this actually worked.”

To map the structure of the NPC, Hoelz relied primarily on X-ray crystallography, which involves shining X-rays on a crystallized sample and using detectors to analyze the pattern of rays reflected off the atoms in the crystal.

It is particularly challenging to obtain X-ray diffraction images of the intact NPC for several reasons, including that the NPC is both enormous (about 30 times larger than the ribosome, a large cellular component whose structure wasn’t solved until the year 2000) and complex (with as many as 1,000 individual pieces, each composed of several smaller sections). In addition, the NPC is flexible, with many moving parts, making it difficult to capture in individual snapshots at the atomic level, as X-ray crystallography aims to do. Finally, despite being enormous compared to other cellular components, the NPC is still vanishingly small (only 120 nanometers wide, or about 1/900th the thickness of a dollar bill), and its highly flexible nature prohibits structure determination with current X-ray crystallography methods.

To overcome those obstacles, Hoelz and his team chose to determine the structure of the coat nucleoporin complex (CNC)—one of the two main complexes that make up the NPC—rather than tackling the whole structure at once (in total the NPC is composed of six subcomplexes, two major ones and four smaller ones, see illustration). He enlisted the support of study coauthor Anthony Kossiakoff of the University of Chicago, who helped to develop the engineered antibodies needed to essentially “superglue” the samples into place to form an ordered crystalline lattice so they could be properly imaged. The X-ray diffraction data used for structure determination was collected at the General Medical Sciences and National Cancer Institutes Structural Biology Beamline at the Argonne National Laboratory.

With the help of Caltech’s Molecular Observatory—a facility, developed with support from the Gordon and Betty Moore Foundation, that includes a completely automated X-ray beamline at the Stanford Synchrotron Radiation Laboratory that can be controlled remotely from Caltech—Hoelz’s team refined the antibody adhesives required to generate the best crystalline samples. This process alone took two years to get exactly right.

Hoelz and his team were able to determine the precise size, shape, and the position of all atoms of the CNC, and also its location within the entire NPC.

The CNC is not the first component of the NPC to be fully characterized, but it is by far the largest. Hoelz says that once the other major component—known as the adaptor–channel nucleoporin complex—and the four smaller subcomplexes are mapped, the NPC’s structure will be fully understood.

The CNC that Hoelz and his team evaluated comes from baker’s yeast—a commonly used research organism—but the CNC structure is the right size and shape to dock with the NPC of a human cell. “It fits inside like a hand in a glove,” Hoelz says. “That’s significant because it is a very strong indication that the architecture of the NPC in both are probably the same and that the machinery is so important that evolution has not changed it in a billion years.”

Being able to successfully determine the structure of the CNC makes mapping the remainder of the NPC an easier proposition. “It’s like climbing Mount Everest. Knowing you can do it lowers the bar, so you know you can now climb K2 and all these other mountains,” says Hoelz, who is convinced that the entire NPC will be characterized soon. “It will happen. I don’t know if it will be in five or 10 or 20 years, but I’m sure it will happen in my lifetime. We will have an atomic model of the entire nuclear pore.”

Still, he adds, “My dream actually goes much farther. I don’t really want to have a static image of the pore. What I really would like—and this is where people look at me with a bit of a smile on their face, like they’re laughing a little bit—is to get an image of how the pore is moving, how the machine actually works. The pore is not a static hole, it can open up like the iris of a camera to let something through that’s much bigger. How does it do it?”

To understand that machine in motion, he adds, “you don’t just need one snapshot, you need multiple snapshots. But once you have one, you can infer the other ones much quicker, so that’s the ultimate goal. That’s the dream.”

Along with Hoelz, additional Caltech authors on the paper, “Architecture of the Nuclear Pore Complex Coat,” include postdoctoral scholars Tobias Stuwe and Ana R. Correia, and graduate student Daniel H. Lin. Coauthors from the University of Chicago Department of Biochemistry and Molecular Biology include Anthony Kossiakoff, Marcin Paduch and Vincent Lu. The work was supported by Caltech startup funds, the Albert Wyrick V Scholar Award of the V Foundation for Cancer Research, the 54th Mallinckrodt Scholar Award of the Edward Mallinckrodt, Jr. Foundation, and a Kimmel Scholar Award of the Sidney Kimmel Foundation for Cancer Research.

Written by Jon Nalick

Leading by Example (+)

Harry Gray
Photo by Verity Smith

Harry Gray, the Arnold O. Beckman Professor of Chemistry, has a knack for grooming academic leaders. At last count, six protégés who have passed through his lab over the past 50 years have gone on to lead universities, and 125 are professors of chemistry at institutions worldwide.

Those colleagues who went on to lead universities are:

  • Dave Dooley (PhD ’79), president at University of Rhode Island (2009-present)
  • Greg Geoffroy (PhD ’74), president of Iowa State University
  • Holden Thorp (PhD ’89), chancellor of the University of North Carolina at Chapel Hill
    (2008-2013) (now provost at Washington University in St. Louis)
  • T. Manoharan, vice chancellor* of the Indian Institute of Technology Madras (1997-1999)
  • K. Poon, president of Hong Kong Polytechnic (1990-2008)
  • Mark Wrighton (PhD ’72), chancellor of Washington University in St. Louis (1995-present)

In addition, many of Gray’s former graduate students and postdocs are on the faculties of Harvard, Penn State, Cornell, University of Chicago, Northwestern, University of Illinois, Purdue, Berkeley, Stanford, and UC San Diego. Steve Mayo, Caltech’s Bren Professor of Biology and Chemistry and the William K. Bowes Jr. Leadership Chair of the Division of Biology and Biological Engineering, was also a student of Gray’s, as were Caltech faculty members Nate Lewis (BS ’77, MS ’77), the George L. Argyros Professor and Professor of Chemistry, George Rossman (PhD ’71), the Eleanor and John R. McMillan Professor of Mineralogy, and Jay Winkler (PhD ’84).

* Madras’s vice chancellor position is equivalent to that of a university president.
‡ Manoharan was a graduate student of Gray’s while both were at Columbia University

Snake-Bite Science

Snake-Bite Science

Steven Sogo (MS ’89) and Samantha Piszkiewicz (BS ’14)

Steven Sogo, a science teacher at Laguna Beach High School, had become frustrated by his chemistry curriculum. On paper, his students performed well in science placement exams, but still, he was troubled. “The type of students who scored high knew how to memorize facts and take tests, but they weren’t necessarily good scientists,” Sogo says. “I wanted to teach a class that rewarded curiosity, experimentation, and the risk of failure.”

So in 2007, Sogo partnered with Ken Shea, a professor of chemistry at nearby UC Irvine. Shea had developed new processes for molecular imprinting—a technique used to create nanoparticles capable of latching onto organic molecules. “We call them ‘plastic antibodies,’” Shea explains. One of the first applications was a synthetic antidote to bee venom.

Sogo enlisted Shea’s help to establish a similar lab at Laguna High, but the high schoolers needed a target. One of Sogo’s students, Samantha Piszkiewicz, voiced her fascination with the Mozambique spitting cobra, which (as its name suggests) spits its venom, a noxious cocktail of protein toxins that break down the lining of cell walls. Starting in the fall of 2008, Sogo led Piszkiewicz and her fellow students in adapting and applying Shea’s techniques for molecular imprinting to the snake venom. The following spring, they had successfully synthesized an antibody. “The first test result we got was so beautiful and encouraging,” Piszkiewicz said. “We saw 85 to 95 percent inhibition of cell destruction.”

In 2009, the students presented their research at a science competition held at Caltech. “That felt like a home- coming of sorts,” Sogo says. “It was a chance to show off the research, and also to introduce my students to a place that made such an impression on me.”
It also made an impression on Piszkiewicz—who went on to enroll at Caltech, graduating this past spring with a bachelor’s degree in chemistry.

Sogo, meanwhile, continued to work on the snake venom project. New classes of Laguna students carefully refined and documented their procedures, and in 2013 their work was published in Chemical Communications, considered one of the field’s leading journals.
“They weren’t published just because they were high school students,” Shea says. “They made a valuable contribution and their work serves as a model for other high schools.”

“It’s hard to imagine that my first published project is for work I did when I was 16,” says Piszkiewicz, who was listed as lead author. Now pursuing her PhD in biophysics at the University of North Carolina, Chapel Hill, she dreams one day of leading her own research lab. “I wouldn’t be the researcher—or the person—that I am today without that class.” And perhaps that’s where Sogo’s real success lies. In addition to creating an antivenom, he is helping scientists like Piszkiewicz to discover themselves.

Alumni stories provided by the Caltech Alumni Association. For more about these stories and to read about other alumni in the news, visit

Agents of Change

In the lab of chemist Jim Heath, Caltech’s Elizabeth W. Gilloon Professor and professor of chemistry, researchers are working to develop new capture agents for cancer—chemicals that could bind to a particular cancer biomarker, allowing the protein to be identified and studied more easily. The goal is to replace antibodies, the current gold standard for capture agents, with something cheaper and more stable.

The biomarkers that the researchers want to target are hundreds of amino acids long. Yet it is often the case that a single mutation within that sequence is enough to cause cancer. So graduate student Kaycie Butler Deyle (PhD ’14) and her colleagues have been trying to zoom in on just the chunk of protein where a mutation is known to occur. For example, Deyle focused on a point mutation on the protein AKT1, where the amino acid E at position 17 is known to change to amino acid K, allowing the protein to stay attached to a cell membrane four times longer than usual—a signal that tells the cell to continue to grow, triggering cancer. In the lab, she first synthesized the chunk of AKT1 that holds the mutation. Then she needed to come up with a chemical that could grab and hold onto that five-amino-acid-long chunk.

To figure out what that chemical might be, she used something called click chemistry, which relies on the ability of two types of molecules, or click handles, to click together when near one another. Typically this requires the incorporation of a copper catalyst, but the Heath lab came up with a new approach. Deyle first inserted one of the click handles two amino acids away from the mutation in her synthesized chunk of AKT1. Then she screened a million-member library to find a short sequence of amino acids with the other click handle attached that would bind to the AKT1 and click together with the first handle. That sequence of amino acids makes up a new capture agent for AKT1. “Essentially, we use the cancer protein to catalyze the formation of its own capture agent,” Deyle says.

Next, Deyle attached a cell- penetrating peptide to her capture agent, and she used a dye to spy on its progress, making sure that the agent was getting through. It was. “Even in cells, our capture agent is still really selective for the mutation,” Deyle notes. With that work in hand, the researchers began trying to block the action of the mutant protein completely. What they’ve found is that an expanded version of their capture agent can successfully stop the mutant protein from binding to the cell membrane. Thus far, the work has only been done on the benchtop. The next step will be to try it in cells. “We hope this is a route to a unique therapeutic for cancer,” Deyle says.

—Kimm Fesenmaier