By David N. Leff

"What makes her so different, the girl with the three blue eyes?"

Baby boomers hummed that song in the 1950s or 60s. In a way, it could apply to the visible difference between normal and cancerous cells.

Comparing these under a microscope, tumor oncologists and clinical pathologists notice one dissimilarity, almost as striking as a third blue eye: Normal human cells (except for blood cells) live and grow, adhering to a solid surface. Malignant cells spurn this anchorage dependency.

Cell biologist Donald Ingber, a researcher at Harvard Medical School's Children's Hospital, has been working for 20 years on how anchorage regulates cell growth. "It does so," he told BioWorld Today, "by two mechanisms: One's chemical; the other, mechanical."

The chemical mechanism, Ingber explained, "involves a molecule called integrin, which binds to the extracellular matrix to which the cell sticks. Integrins are a big topic for drug discovery now," he observed, "as they turn on all kinds of signal transduction pathways."

For years, Ingber has been promoting a mechanical pathway, also through integrin molecules, which he calls "bridges between the cell's outer skeleton and inner scaffold. This mechanical signaling system," he said, "is activated by changes in the extracellular matrix -- the anchorage surface -- that alters its ability to resist the tension that the cell generates in its cytoskeleton.

"Cells generate tension through actomyosin interactions," Ingber said, "just like muscle cells do. They all pull on their anchoring points by transmitting force over integrins." Then he added a salient point, embodied in his paper in today's Science, dated May 30, 1997, and titled: "Geometric control of cell life and death."

"Changing the ability of the anchorage to resist that pull alters the shape of the cell," Ingber said, "not just whether it's square or round, but the degree of mechanical distortion in the cytoskeleton and/or the cell nucleus."

Ingber continued: "It's known that normal cells need to flatten out and spread in order to grow and form tissues, whereas tumor cells can proliferate whether round or flat. To get real tumor formation, you need to uncouple, or derail, both the chemical and mechanical signaling pathways."

He elaborated his concept: "Because, after all, when a tumor starts to form, it's really a normal cell that normally would not grow, owing to what I call crowd control. It feels its neighbors packing it in, and so it says, 'Oh, I'm confluent; I'd better stop growing.' But somehow, tumor cells can't sense the crowd pressure, and continue to proliferate. And that sensing mechanism is largely shape-controlled."

Cell Shape: A Matter Of Life And Death

In today's Science paper, Ingber and his co-authors show that changes in the shape of human capillary endothelial cells "govern whether these cells live or die. The process is critical to angiogenesis and tumor growth."

To test this hypothesis, they constructed the biological equivalent of an electronic silicon chip. "It began," Ingber recounted, "by designing on a computer screen round, square and other shapes about five microns in diameter. Then, by photolithography and other steps, we created a kind of rubber stamp-like device that imprinted these micro-shapes onto gold-covered microscope slides coated with adhesive molecules such as fibronectin or collagen."

When capillary cell culture was applied, the cells spread out to neatly fill the tops of the shaped adhesive islands studding the mini-anchorage site.

"In terms of biotechnology," Ingber said, "[co-author] George Whitesides and I have patents on the use of this microchip technology for positioning cells and controlling their function. And there's interest in it, as a way, for example, of automating cell screening with live-cell-based diagnostics and drug screening."

He offered a concrete example: "Let's say you want to screen for drugs that affect apoptosis but not growth, or differentiation without altering growth, or growth without altering apoptosis. You could keep the cells on the same microchip in different functional states at a real-time read-out, with one flow-through for one drug.

"We've showed," he went on, "that when a cell is partially retracted, it differentiates. And when it's completely spherical, but still adherent, it dies. So cells can be switched between growth, differentiation and death through the mechanical side, even though they have the same chemical inputs -- such as G protein and tyrosine kinase signal transduction, for example -- coming in at each state of the cell. You get one concerted response."

Defining these multiple simultaneous inputs in terms of his own experimental capillary system, Ingber observed: "One capillary cell might spread out and grow to form new capillaries, while right next to it, only microns away, the next cell might be undergoing apoptosis, while a third cell is becoming quiescent and differentiating, even though you have growth factors everywhere."

Capillaries, of course, are the initial building blocks of angiogenesis, by which tumors acquire a nutrient-rich blood supply to fuel their growth and metastasis.

His cell-shaping mechanism, Ingber said, "opens up a different angle on angiogenesis inhibition, rather than just trying to inhibit the integrin ligation, which is where people are making real advances today. One may be able to subtly disrupt this mechanical signaling system in the cytoskeleton.

"The cytoskeleton is going to be the big new target for drug development in the next 10 to 15 years," he said. "Angiogenesis inhibition is not just cancer therapy. The drug we have in in vivo trials in animal models is also the best thing around for inhibiting rheumatoid arthritis, because the breakdown of cartilage is due to capillary ingrowth and release of enzymes. It's also good for diabetic retinopathy and blindness."

Ingber concluded: "Understanding the fundamental mechanism of capillary growth control and apoptosis, which really looks like a way in which angiogenesis is physiologically regulated, could have enormous implications."

He is the founder of a new start-up company, Molecular Geodesics Inc. (MGI), of Cambridge, Mass. It has just obtained a $6.4 million contract from the Defense Advanced Research Projects Agency, Ingber said, "to develop bioskins that would be mechanically strong but permeable, and also kill biological warfare agents in situ, before they reach the soldier."

MGI's research and development program, he added, is based on "new biomaterials that exhibit the mechanical responsiveness and biochemical processes of living cells and tissues, which involves the translation of mechanical design principles that we discovered in the cytoskeleton." *