Identifying source of disease: Faulty proteins account for most of the world’s sickness

Virtually all the biological processes that keep us alive are controlled by proteins in our bodies. Therefore, most, if not all, of our diseases can be traced to faulty proteins. In a major leap toward learning the basics of human biology and what makes it go awry, researchers have built the prototype of a high-tech chip that rapidly identifies proteins and their functions. Such chips may ultimately help to determine which proteins are responsible for which diseases.

“We don’t yet know how many different proteins make up a human body,” admits Gavin MacBeath of Harvard University’s Center for Genomics Research. “We think it’s somewhere between 25,000 and 120,000. It may not be possible to make every one in a laboratory, but I believe that many of them can be made. Then they could be put on small glass chips that would allow us to determine what proteins potential drugs would bind to, or help us search for new drugs more efficiently.”

“These chips could be used to profile proteins, a capability that would be invaluable in distinguishing proteins in normal cells from early-stage cancer cells, and from malignant, metastatic (spreading) cancer cells that are the real killers,” says Stuart Schreiber, Morris Loeb Professor of Chemistry.

MacBeath and Schreiber employed a high-precision robot capable of putting more than 10,000 droplets of purified proteins on a glass slide about 1 inch wide and 2 inches long. Passing drug molecules over the slide reveals which proteins they interact with and can provide information needed to design the most effective drugs. Such highly selective drugs, it is believed, can reduce side effects that often make chemotherapy a sickening ordeal for cancer patients.

Chipping genes

Progress in sequencing the complete set of human genes promises to eventually identify all proteins because each gene carries the code for making an individual protein. The gene produces messenger RNA (ribonucleic acid), which carries out the actual assembly of proteins in cells. So-called gene, or DNA, chips can reveal the activity of messenger RNA and, thus, which genes are most active in a normal or diseased cell. But to understand what really goes on, researchers need to know how proteins are involved.

“We use DNA chips to study the activity of genes, but they don’t tell us what the proteins are doing,” explains MacBeath. “Nor do they tell us what happens to a protein after it’s made.”

Many biological processes are triggered when two or more proteins bind to each other. Such binding can turn proteins on or off, that is, make them active or inactive. For example, proteins known as kinases interact with other proteins to regulate metabolism, make it possible for a fetus to develop in a womb, or boost growth of tumors. DNA chips don’t reveal such interactions.

Using readily available materials, MacBeath and Schreiber built a system capable of disclosing which proteins bind to each other, to kinases and other catalysts, and to drug molecules. “Everything we used can be purchased off the shelf,” MacBeath comments. “There was no fancy engineering; no sophisticated machines had to be custom-built.”

Purified proteins are given to a robot with mechanical “fingers” so steady it places thousands of protein-containing droplets, the diameter of a human hair or two, on the small slide. The slide is treated with chemicals so the droplets attach to the surface and don’t roll off. Other chemicals prevent the proteins from evaporating or deteriorating.

Researchers then add fluorescent molecules to other proteins they wish to study and expose them to proteins on the slide. The slide then is washed and wherever fluorescent spots appear, proteins have bound to each other. From such bindings, the function of each protein is discerned.

Testing drugs

Testing drugs works much the same way. To determine which proteins interact with a specific drug, the latter would be combined with a fluorescent molecule and exposed to a chip. To find a drug that kills a protein essential for the life of a cancer cell or a cell infected by the AIDS or West Nile viruses, different drug molecules could be passed over various proteins arrays.

In experiments done to date, the thousands of droplets on each chip consisted of only two proteins. Now that the technology is proven, MacBeath looks forward to putting 10,000, even 20,000 different proteins on the same chip. If human bodies contain 40,000 proteins, two chips might hold them all. Even 120,000 proteins might be accommodated with a half-dozen chips.

“Now that we’ve demonstrated that our system works, the next challenge is to make large sets of purified proteins,” MacBeath points out. “Once that’s done, the robot can make hundreds of copies of each slide so we can test hundreds of target proteins or drugs at the same time.”

Chips of the future will reveal not only the function of proteins but their abundance and how they change after they are made in a cell. Abundance would be measured by fluorescence; the more abundant a protein, the brighter spot it would create on the chip.

Normal cells from a tissue sample might be labeled with red and cancer cells with green. After adding the mix to a slide, investigators would look for how much red versus green exists in a cancer cell relative to a normal cell. That would form the basis for designing drugs to attack the green proteins.

MacBeath notes that this hasn’t been done yet. “There are still a few problems to be solved, and we are working on them,” he says.

Shooting at specific targets

Another project on the near horizon involves putting arrays of antibodies on chips. Antibodies, proteins produced by your body’s immune system, bind tightly to other proteins such as those of an invading virus or a cancer tumor. MacBeath plans to make antibody chips to detect the amount, location, and modified states of proteins in samples of normal and diseased tissue. That would enable researchers, for example, to do in-depth studies of what goes wrong with proteins in a cancer cell relative to a normal one. “I hope that will lead to a better understanding of cancer as well as aid in the discovery of new anti-cancer drugs with fewer side effects,” he says.

The potential applications of protein chips are generating intense excitement among researchers as well as among drug and biotech companies. In a step toward commercializing this technology, MacBeath has formed a new company with colleagues from M.I.T. and the University of California, San Francisco, called Merrimack Pharmaceuticals.

With the rush to treat each patient individually, according to his or her specific genes and proteins, MacBeath and Merrimack will soon have lots of company.