Xiaowei Zhuang makes snuff films. First, she isolates her victims. Then she forces them into a closed chamber, surrounds them with known killers, and lets her camera run.
A couple of years ago, she won a MacArthur "genius" award for her grisly work. At 33, she's a beacon in her field, winner of more than a dozen prizes worldwide. And, no, she didn't go to film school.
Zhuang is a biophysicist. Her movie studio is a state-of-the art laboratory at Harvard, where she works as an assistant professor. Her crew is composed of 15 postdocs and grad students. And her cast? The victims are live monkey cells. The killers are influenza viruses.
Zhuang's direct-to-video releases may not be especially entertaining - they all end the same way - but to anyone interested in potential treatments for diseases ranging from HIV to cystic fibrosis, they're more revealing than a Michael Moore documentary. Most virologists have concentrated on before-and-after stills of viral attacks. As a result, they didn't know, for example, whether viruses moved through the cell to the nucleus through diffusion or active transport. But Zhuang has developed a technique to capture the process as it unfolds inside a single cell. These movies are crucial to scientists searching for opportunities to block viruses in transit. Equally important, researchers may learn from Zhuang's films how to mimic viruses, which could help them engineer drugs that penetrate cells and treat genetic disorders from within.
"I like to be able to see what I'm doing," Zhuang says in her soft voice, strolling past a lab bench where grad students are preparing monkey cells for their impending demise. A small woman dressed in the trim fashion of an international executive, Zhuang expresses herself in terms equally simple and polished. "I believe that you can learn something new about any system if you really look at it. You just have to be careful to follow every particle."
She walks into a room dominated by a microscope tricked out with a pair of color-specific digital cameras and a couple of laser beams. Zhuang designed the apparatus, but its lineage can be traced directly to another pioneer in direct visualization - 19th-century photographer Eadweard Muybridge, who sought to discover whether a galloping horse ever has all four hooves off the ground. While others bickered over how the animal's great velocity could overcome its enormous weight, Muybridge devised a photographic system that captured motion in a series of rapid snapshots. The result: proof that the creature gets airborne and a visual record of the entire process.
Muybridge's stop-action photographs laid the groundwork for motion pictures. Hollywood is one of his progeny. Zhuang is another.
Zhuang's father was a physicist. She was so eager to become one herself, and so quick a study, that she skipped several years of high school and college, never bothering to graduate formally from either. This enabled her to evade emigration restrictions, bypassing public service obligations she'd have had to the Chinese government if she actually held a diploma. In 1991, she enrolled in UC Berkeley's physics department, which granted her first diploma - a master's. She had her PhD by the time she was 24.
Zhuang focused on optics from the beginning. And when she was awarded a postdoc at Stanford, she teamed up with Nobel Prize-winning physics professor Steve Chu because she admired the visual approach he used for his experiments in polymer dynamics. The polymer that Chu used was DNA, a complex molecule that's easy to replicate. Looking for a problem of her own, Zhuang started studying RNA, DNA's working-class cousin. She found that there was considerable confusion as to how certain types of RNA folded, contorting to build proteins from amino acids. A biological question, to be sure, yet one that she thought optics might help answer.
The approach of other researchers was to force a large sample of RNA to go through the folding process - generally by adding magnesium - taking measurements along the way. With this information, the folding sequence can be surmised, much as we might assume that a shirt we get back from the cleaners was folded by first bending the arms back and then creasing the torso. The trouble is that our assumption could be incorrect. Each shirt might get folded differently, one with the left arm bent back first, the other with the right. In other words, before-and-after assessment will characterize how shirts might get folded, but not necessarily how a particular shirt is folded in practice. The same goes for folding RNA molecules.
This is a model case for direct visualization, watching one particle at a time. By filming individual molecules in action, Zhuang was able to see how they behaved. And she was able to show that they were less like robots than like dancers, idiosyncratic performers in an elaborate ballet.
Success led her to extend the technique to proteins, including one integral to a flu virus. Soon Zhuang realized that she could use her microscopic movie setup to look at the whole infection process, which was plagued with the same sort of ambiguities as RNA folding. By the time she got to Harvard, she was preparing to make her first snuff.
A graduate student, Melike Lakadamyali, sets a plastic petri dish under a microscope, while fellow grad student Michael Rust switches on red and green lasers that shine from below. An ultrathin glass slide lets through the maximum amount of light with minimum distortion. The dish contains several live monkey cells that have been genetically engineered to glow fluorescent yellow.
At Rust's signal, Lakadamyali deposits several thousand viruses onto the dish with a micropipette. They've spent the last hour bathed in red fluorescent dye so they blaze like fireflies on one side of a split-screen computer monitor. The other side shows the ghostly glow of a cell membrane, a thousand times larger.
The assault has begun. The viruses swarm the cells from every direction. Within a couple of minutes, five or six of them have attached to a cell, which mistakes them for nutrients and encloses them in membrane pockets. A pocket passes through the cell wall and pinches free on the inside, where it takes a few minutes to carry the virus to the region surrounding the nucleus. Several more seconds pass before the virus starts to leak out, depositing its genome in the host nucleus, which will replicate the viral RNA thousands of times over the next few days.
Only the first part of that process - the virus binding to the cell wall - is captured in this particular experiment, and even then most of the action can be seen only in replay, when the left and right channels are overlaid and the viruses that don't bind - the vast majority - are digitally filtered out. "It's a little anticlimactic in real time," Rust confesses. But, Lakadamyali says, "you have the opportunity to ask quantitative questions about things people have known about for a long time but have never really characterized."
Indeed, though influenza has long been studied, Zhuang and her students were the first to reveal, in a 2003 article in Proceedings of the National Academy of Sciences, previously undescribed levels of detail in the three stages of virus transport. In the final step, the virus package travels back and forth in the perinuclear region before bursting through its membrane pocket. That pattern was particularly unexpected and is now undergoing closer scrutiny in labs around the world.
Knowing the specifics of the intermediate states of infection, and seeing, for example, that a virus might take one of several pathways to the nucleus, is crucial. If the interaction between the virus and the cell could be modified slightly, the whole viral mechanism might be rendered ineffective. Thus far, every detected virus-cell interaction exploits a function necessary for cell survival. "The virus is the best opportunist that nature has ever created," Zhuang explains. "It does almost nothing by itself." Block cells from taking in viruses and you'll starve them of nutrients, too. But there's a good possibility that the virus also depends on some small maneuver not used in ordinary cellular function, an evolutionary artifact, perhaps - and therefore a perfect drug target.
That's one way Zhuang's work might lead to a medical breakthrough. Another might occur if researchers learn to harness the cleverness of viruses. Gene therapies for diseases like cystic fibrosis and Parkinson's repair cells by replacing faulty DNA. Viruses can be genetically engineered to carry the replacement DNA to the nucleus, but they are difficult to control. As a result, synthetic carriers, built to order in the lab out of modified viruses, have become increasingly popular, but they're still woefully inefficient. By filming them, Zhuang has found a possible reason: They don't take the same fast-track pathways as the wild viruses she's studied. Whether synthetic carriers might work better if they're rerouted remains to be determined, but before Zhuang came along, researchers in her field didn't even know to ask the question.
Questions are infectious. When Muybridge's stop-action contraption revealed how horses gallop, he soon found himself wondering how all animals moved, including humans. Muybridge made the study of comparative anatomy dynamic.
Similarly, Zhuang is using the most advanced motion-visualization technology of our day - and her own keen desire to see - to create a body of research that runs across the traditional disciplines of physics, biology, and chemistry. In collaboration with researchers at Harvard and MIT, she has recently begun looking at other viruses, such as polio and polyoma. Zhuang is onto something big; it's the actors that got small.
Zhuang uses lasers, a microscope, and pair of hi-res digicams to capture viral infection in action. Here's how it works.
The Setup
1. Red and green lasers travel along a single path to the back of the microscope, where they are reflected upward.
2. Monkey cells that glow under green laser light and viruses that react to red laser light are placed on the microscope stage.
3. Two cameras - one sensitive to red light, one to green light - feed the action to a split-screen monitor.
The Results
1. Superimposed images show the virus (red) attaching to the outer membrane of the cell, which surrounds it and pinches off to form a pocket containing the virus particles.
2. The virus pocket makes a beeline for the nucleus. It travels along a microtubule conveyer belt, exploiting the machinery of the cell to select the most efficient route.
3. In the region surrounding the nucleus, molecular motors pull the virus pocket back and forth. The pH drops, triggering the pocket to release its viral cargo into the cell's nucleus.
Jonathon Keats (jonathon_keats@yahoo.com), a novelist and conceptual artist, wrote about email hoaxes in issue 12.07.
credit John Midgley
Xiaowei Zhuang
Light show: Xiaowei Zhuangés filming techniques allow her to see a bright flash when a virus is released into a nucleus.
credit Bryan Christie
The Setup, from left to right: 1) Red and green lasers travel along a single path to the back of the microscope, where they are reflected upward; 2) Monkey cells that glow under green laser light and viruses that react to red laser light are placed on the microscope stage; 3) Two cameras-one sensitive to red light, one to green light-feed the action to a split-screen monitor.
credit Bryan Christie
The Results, from left to right: 1) Superimposed images show the virus (red) attaching to the outer membrane of the cell, which surrounds it and pinches off to form a pocket containing the virus particles; 2) The virus pocket makes a beeline for the nucleus. It travels along a microtubule conveyer belt, exploiting the machinery of the cell to select the most efficient route; 3) In the region surrounding the nucleus, molecular motors pull the virus pocket back and forth. The pH drops, triggering the pocket to release its viral cargo into the cellés nucleus.
Feature:
The Deadly Art of Viral Cinema
Plus:
