Trends in Cell Biology
Without an inner skeleton, our cells would be limp little bags that drift aimlessly, like jellyfish through the sea. Of course, nothing about cells is aimless or limp: cells assemble proteins inside them into a structured skeleton and then harness this exquisite order to become surprisingly strong and agile.
Indeed, this order allows cells to be, at various times, everything: rigid or soft, still or mobile. The cytoskeleton reconstructs itself whenever the cell needs to change its shape and behavior in response to environmental demands. For example, a cell that sits stiffly one moment can within minutes "morph" into a squishy, crawling bleb that squeezes through tiny openings.
Lewis Cantley (left) is trying to pin down a slippery
class of molecules
thought to
pull the strings on the actin cytoskeleton: fatty acids that
reside in the cell's outer membrane. The modest
yeast cell, a creature
highly
amenable to scientific manipulation, allows Rong Li (below)
to make inroads into the molecular thicket
of the cytoskeleton.
For years, the cytoskeleton has captivated scientists but also eluded them. It has revealed just enough secrets to hint that it somehow figures into most fundamental biological processes, including movement and growth. It also shapes the functional identity of cell types, be it a platelet's power to stop bleeding, a yeast cell's ability to bud off daughter cells, or an embryonic neuron's capacity to explore the territory it is entering. Yet the cytoskeleton is so complicated that it has defied attempts at a complete understanding of it.
Recently, however, scientists have begun closing in on their moving target. Taking different approaches, they have assembled a mosaic of bits and pieces of information that may soon fall into one big picture. Meanwhile, this field has become "the major new direction of research in the Cell Biology Department," says Lewis Cantley, a professor in that department who works at the Harvard Institutes of Medicine through Beth Israel Deaconess. "Our recent faculty appointments have focused on people with an interest in that area."
Many researchers across the University are working in this emerging field. Focus cannot cover them all, but plans in a future issue to describe current work by several major contributors, including Marc Kirschner, chairman of cell biology, and Timothy Mitchison, a professor there and codirector of the new Institute for Chemistry and Cell Biology. While the basic research intensifies, it is already beginning to pay practical dividends (see sidebar next page).
Movers and Shakers
Part of the research has focused on the proteins that actually perform the dynamic changes of the cytoskeleton. The most abundant protein of the cytoskeleton is actin, which constitutes up to 20 percent of a cell's protein content. At any given moment, about half of a cell's actin occurs as monomers; the other half is strung into twisted, polymeric filaments. Some of these strands cross-link to form a continuous mesh running just under the cell's outer membrane, whereas others line up to form bundles that span the middle of the cell.
Of the roughly 100 different proteins thought to bind to actin in living cells, a handful serve to disassemble filaments in one area of the cell and rebuild them elsewhere, says Paul Janmey, associate professor of medicine (biochemistry) at Brigham and Women's Hospital. Over the past twenty years, researchers have learned much about them.
Its outer membrane washed away with a detergent, a human platelet reveals its inner life to the electron microscopist. At the bottom, actin filaments interweave into a meshwork. Individual filaments pointing toward the top indicate actin polymerization in the course of the platelet's activation, a process stopped in its tracks to yield this micrograph.
For example, proteins called thymosins maintain a pool of monomeric actin that can polymerize as needed. One type of thymosin may be involved in prostate cancer (see sidebar). The protein profilin keeps monomeric actin from polymerizing inappropriately. The protein ABP280, discovered by Thomas Stossel and John Hartwig, stitches existing actin filaments into a continuous meshwork.
The protein gelsolin severs existing filaments. It attacks a filament simultaneously at many spots along its length, breaking it into little pieces. That way gelsolin assures that the filaments dissolve quickly. When inappropriately cleaved, this protein causes amyloid deposits leading to a form of neuropathy. Finally, the protein cofilin speeds up the shrinking of filaments, in effect revving up the dynamics of the whole system, says Janmey.
All the other proteins found to associate with actin do not reorganize the cytoskeleton itself. Instead, they benefit from one of the myriad functions of this pervasive organelle, says Donald Ingber, associate professor of pathology at Children's Hospital. The actin cytoskeleton offers a platform for all sorts of enzymes--for example, those involved in transmitting growth signals--to meet and get close enough to their reaction partners to carry out their biochemical function.
Missing Links Between Signals and Actin
Orchestrating the dynamic changes of the actin cytoskeleton are signal transduction cascades that relay incoming messages. The dialog that links cross-talking signal transduction pathways to the proteins making and breaking actin filaments is still poorly understood.
A hypothetical example illustrates this interplay. When a cell settles down in one location, it forms highly specific attachments. These so-called focal adhesion points are spots on the membrane where proteins gather, tying the actin cytoskeleton across the membrane to extracellular matrix molecules. During this docking process--which involves many signaling events--monomeric actin polymerizes into filaments in a local burst just under that spot. That way, a cell straps itself to a surface like a tent with the pegs firmly in the ground. Much of its actin forms struts that run across the cell like tent poles and tie into focal adhesion points, making the cell quite stiff. But adding a growth factor that triggers certain signaling cascades will cause this cell to stir and become more fluid. Signaling events at the membrane, involving enzymes called small GTP-binding proteins, somehow instruct the actin-binding proteins to take apart the tent poles and reform actin filaments at the cell's edges. As a result, the membrane begins to "ruffle" and, viewed through a microscope, seems to wave at the observer.
When a yeast mother cell buds a daughter
cell (a), distinct patches of actin networks (bright dots) help push out
the bulging membrane of the bud. In yeast lacking the protein Bee1, this
coordinated event is disturbed (b), leaving the cell's actin in aberrant
shapes.
Image courtesy of Rong Li.
One way scientists imagine directed motion could happen, says Janmey, is that the membrane flutters quite randomly. But at those ripples where growth factors have collected, these factors generate signals underneath the membrane that trigger actin polymerization. This would build a wall of actin, shoring up selected bulges and making them into semipermanent feet.
When the cell crawls in a certain direction, its cytoskeleton acts like a continuous disassembly-assembly line: the actin-binding proteins chew up filaments at the back of the cell and rebuild them at its leading edge, thus constantly recycling the cell's supply of actin.
Many of the observations that have yielded this simplified scenario were gleaned from studies of cells crawling on two-dimensional culture dishes or isolated proteins studied in test tubes. But one of the knottiest problems in understanding what the proteins do in a living organism is that cells act in a three-dimensional environment. It is technically difficult to design experiments that capture a real-life scenario yet allow for the type of manipulation necessary for the scientist to tease apart complex protein-protein interactions. John Hartwig (see sidebar) has validated many in vitro findings in human platelets, but the system that most readily yields in vivo information is the lowly yeast cell.
The Buzz of Wasps and Bees
This is where Rong Li, assistant professor of cell biology, has carved out a niche. Yeast deploys its actin cytoskeleton to perform a biological process that allows her to study the responsible proteins in their natural environment. When a mother yeast cell sprouts a bud, patches of webbed actin filaments appear underneath the bulging bud's membrane (see figure at right).
Li's approach combines the power of yeast genetics with biochemistry to speed up greatly the process of identifying genes and studying the function of the accompanying proteins. She has developed a biochemical assay that allows her to test and manipulate the function of actin-related proteins. She pokes holes in yeast cells to block the formation of actin patches in these cells' buds. Then she slips in specific proteins to see which ones can rekindle this process.
In the July 14 Journal of Cell Biology, Li and graduate student Terry Lechler describe two proteins necessary to form the patches in vivo. One of these proteins Li had discovered earlier and dubbed Bee1, after its human relation "WASp." The WASp gene is mutated in a hereditary disease, and its protein may turn out to be one of the sought-after links between various signal transduction pathways and the cytoskeleton, much like a multisocket adapter. Though studying the complex human protein in its natural environment is difficult, Li's work is beginning to identify just what role its humble cousin plays in assembling the nimble networks of actin.
Li and other researchers, at Harvard and beyond, are quickly zeroing in on the molecular intersections between signal transduction and actin reorganization. Several local groups are working to uncover the precise role of a still-shadowy but probably crucial intermediary: a lipid component of the cell membrane. Understanding its function ultimately will help illuminate the elaborate interplay between a cell and its "outside world," in health and disease.
--Gabrielle Strobel
Basic research into how the actin cytoskeleton manages to remodel itself almost continuously is beginning to spin off practical applications.
Flagging the Aggressor: Take the case of thymosin beta15. Known to bind actin, this protein recently rose to become a candidate diagnostic marker for prostate cancer, the most common cancer among men. A research team led by Bruce Zetter, professor of surgery at Children's Hospital, reported last December that this gene is prominently expressed in prostate tumor cells that will go on to become metastatic, but not in normal cells or benign tumor cells. Somehow thymosin beta15 gears up the cell's ability to break free of its ties to neighboring cells, making it prone to course through the body and start tumors elsewhere. The researchers are now conducting a long-term clinical study to investigate whether this gene can help identify those prostate cancer patients whose tumors need surgery among the majority of patients who do not.
Biologist John Hartwig (middle) and biophysicist Paul Janmey (right) are members of a group, led by clinician Thomas Stossel (left), that studies the cytoskeleton of blood cells. Part of that work is already moving into the clinic.
Platelets Keep Their Cool: A potential application of actin research grew out of John Hartwig's work on the cytoskeleton of human platelets. Platelets used for blood donations cannot be stored well because refrigeration triggers actin polymerization, which causes the platelets to aggregate into clumps (see image pg. 9). Researchers are now developing a way to cover the ends of the actin filaments with a drug that would thwart actin polymerization. Being able to store platelets as easily as serum would have a great clinical impact, says Hartwig, associate professor of anatomy and cellular biology at BWH.
Breathing Freely: Research into how actin filaments alter the viscous and elastic properties of a fluid may help clear the airway congestion that afflicts people with cystic fibrosis. The sticky sputum of people with this common genetic disease was long known to contain DNA. Researchers led by Thomas Stossel, professor of medicine at BWH, found it contains actin as well; moreover, the DNA is intertwined with actin fibers, making the sputum highly viscous. The actin gets spilled into the airways by dying neutrophil cells, which fight bacterial infections there. In an ongoing clinical trial in Canada conducted by the biotechnology company Biogen, patients inhale aerosols containing gelsolin, a protein that breaks actin filaments, presumably making the sputum more fluid so patients can clear it more easily.
In the long run, researchers hope to mine the actin cytoskeleton field for treatments against conditions involving unwanted cell mobility, including inflammation, metastasis, and certain infectious diseases.