Antioxidant Enzymes Show Unexpected Behavior in Leukemia Cells Treated with Imatinib

Division of Cancer Treatment and Diagnosis

FDA approval of the targeted drug imatinib in 2001 revolutionized the treatment of chronic myelogenous leukemia (CML), changing it from a difficult-to-treat cancer with a poor prognosis to a chronic, manageable condition. A long-term, follow-up study found that CML patients treated with imatinib, who have been in remission for at least two years, have the same risk of death as their peers in the general population.

Imatinib targets a specific genetic mutation that causes almost all cases of CML. In this mutation – a type known as translocation – pieces of two different chromosomes become stuck together and produce a protein called BCR-ABL. This protein is like a switch stuck in the “on” position: it constantly signals cancer cells to divide. Imatinib stops cancer cell growth by blocking BCR-ABL.

Unfortunately, some patients eventually become resistant to imatinib, requiring a change in treatment. Alan Diamond, Ph.D., professor of pathology at the University of Illinois at Chicago, is researching whether antioxidants – substances that protect cells from the damage caused by free radicals (also called reactive oxygen species, ROS) – contribute to imatinib resistance.

With funding from NCI*, his laboratory has looked at how levels of an antioxidant enzyme, glutathione peroxidase-1 (GPx-1), change during treatment with imatinib. GPx-1 requires the trace element selenium, found naturally in many foods, to work.

Studies from Dr. Diamond’s and other laboratories have found an association between lack of selenium in the diet, reduced GPx-1 activity in the body, and an increased risk of cancer. When he read a paper in 2003 that the ABL protein (part of BCR-ABL that drives CML) activates GPx-1, “my thought was that [GPx-1] is a beneficial protein, and people who have their ABL deactivated by therapy [with imatinib] might be suffering,” he said.

Dr. Diamond and colleagues hypothesized that CML patients taking imatinib would have decreased GPx-1 activity, and tested this idea in blood samples taken before and during treatment from seven patients**. Instead of a decrease, “we saw the complete opposite,” Dr. Diamond recalled. “The clear trend was that levels of GPx-1 were actually higher in the patients being treated.”

To look closer at this unexpected finding, Dr. Diamond’s research team performed a series of experiments in cultured cancer cells. They found that not only do GPx-1 levels increase in the presence of the combination of imatinib and the BCR-ABL protein, but levels of another important antioxidant enzyme, MnSOD, also rise. While GPx-1 levels approximately double in CML cells given imatinib, levels of MnSOD rise about five-fold.

Dr. Diamond is currently planning several research projects to understand how these newly discovered mechanisms regulate important antioxidant genes. The idea that more antioxidants are not always better may seem counterintuitive, but makes sense, he explained. “Years ago, we thought that antioxidants were always a good thing. [But] human epidemiology is now revealing that sometimes they’re a good thing and sometimes they’re a bad thing.”

“The unproven model is that in the course of your lifetime antioxidants can be a very good thing,” Dr. Diamond added. “They reduce the chance of carcinogenic mutations. So from birth to a certain time in your life, they’re very protective. However, once you reach a certain point and cancer is starting to develop, at that point these same mechanisms can protect the cancer cells.”

An added layer of complexity being studied by Dr. Diamond comes from the fact that naturally occurring genetic variations – called polymorphisms – can also alter the levels of antioxidant enzymes in cells. Extra amounts of these enzymes are not necessarily a good thing, even in normal cells. The compounds produced when these enzymes break down free radicals are not stable. If a cell doesn’t have enough other antioxidant molecules from the diet to finish the break-down process, the compounds can covert back to the damaging free radicals, Dr. Diamond said.

“The increase in MnSOD antioxidant enzyme that’s occurring in CML may actually be more significant than the increase in GPx-1, because a high amount of MnSOD could cause oxidative stress,” he explained. Leukemia cells experiencing oxidative stress can end up with additional mutations, some of which could contribute to resistance to drugs targeting BCR-ABL, Dr. Diamond said.

William D. Merritt, Ph.D., program director at NCI’s Cancer Therapy and Evaluation Program, commented: “It is very gratifying to see how funding for this small two year discovery grant provided a novel link between resistance to a widely-used therapeutic for CML, imatinib, and antioxidants, and points to the need for further studies to explore changes in antioxidant enzymes in patients receiving this and similar therapies.”