Healthcare

Could Mitochondria Numbers Be the Key to Solving Cancer Drug Resistance?

Share this post:

Mitochondria inside cells

Sister cells display distinct morphology and varying amounts of mitochondria. Mitochondria diversity imparts a selective advantage to tumors, benefiting the survival of cells with higher amount of mitochondria. Cells derived from malignant glioma (brain cancer). Magenta, mitochondria. Green, actin fibers. Blue, nucleus. (Photo: Luis Santos.)

In the late 1940s, the first treatment for childhood leukemia was developed by Sidney Farber. The spectacular remissions it caused created a great deal of initial excitement, which rapidly disappeared as these turned out to be only temporary, and resistance to the treatment emerged1.

We now know that cancer cells can have identical genomes and still respond differently to therapy. The question of how this can happen remains a major challenge in oncology today. But this diversity of behavior is one of the mechanisms that underlies the partial response, resistance to treatment, and evasion of remission that characterizes some cancers. Therefore, it stands to reason that suppressing this diversity may help to curtail resistance to treatment and potentially eliminate the disease.

We worked with researchers in the laboratories of Jerry Chipuk at the Icahn School of Medicine at Mount Sinai in New York and Marc Birtwistle (now at Clemson University) to try to find the non-genetic sources of this variability in cells. The results of this research are now published in Nature Communications and reveal—for the first time—that the number of mitochondria in a cell is partly responsible for the drug dose at which each cell dies.

Cells can regulate how many mitochondria they have, which results in a variable number of mitochondria in each cell. Mitochondria are best known to generate energy in the form of ATP molecules2. They also act as an emergency red button, where signals converge to activate the cell death program, also called apoptosis3. This internal program of cell suicide can be triggered in very particular situations, such as when exposed to certain drugs. This fact inspired our team to explore the hypothesis that genetically identical cancer cells with different number of mitochondria may have different propensity to die if exposed to the same drugs.

Indeed, we discovered that cells with more mitochondria are more resilient to apoptosis. Cells with fewer mitochondria, on the other hand, are more sensitive to apoptotic signals, and this could make them less resistant to treatment. This could have a far-reaching impact in cancer treatment, since scientists could potentially use a combination of drugs to lower the number of mitochondria in cancer cells to induce cell suicide.

In early experiments, we observed up to a 5-fold difference in the number of mitochondria between cells grown in cell cultures, where all cells have the same genome, even when controlling for cell-size effects. In subsequent experiments, Mount Sinai post-doctoral researcher Luis Santos exposed several types of cells to up to six different concentrations of a pro-apoptotic drug and measured the number of mitochondria in the surviving cells at each dose. The results clearly showed that surviving cells had a tendency to have more mitochondria than the initial population of cells and that the higher the drug dose, the more mitochondria were in the surviving cells.

These experiments generated massive datasets with information on the abundance of mitochondria in tens of thousands of single cells at several doses in a variety of cell lines, which begged for a rigorous statistical analysis. To handle this data complexity, researchers at IBM Research developed a mathematical framework called DEPICTIVE (the acronym of “Determining Parameter Influence on Cell-to-cell variability Through the Inference of Variance Explained”) to extract the shape of the number of surviving cells as a function of the drug dose (what is known as the dose-response curve). This information is necessary to quantify the diversity in the live/dead response of individual cells due to mitochondrial number.

The simple idea behind the analysis is that the shallower the dose-response curve—or lower the change in the rate of cell death with increasing dose—the more diversity exists in responses of individual cells in the population to the drug. Conversely, the steeper the curve, the more similar are the responses of individual cells in the population.  When the contribution of variations in mitochondrial number among cells was mathematically extracted out, the dose-response curve became much steeper (less diverse). Surprisingly, only mitochondrial number could explain up to 30 percent of the variability in the response to the drug, a giant feat for this single feature in cells that has never before been associated with this variability.

Understanding the correlation between mitochondrial number and response to drugs could potentially lead to more successful cancer treatments and impact future drug development. This should be a relatively easy step, as researchers used standard chemicals and laboratory equipment and are currently measuring mitochondria numbers in patient-derived cells at Icahn School of Medicine at Mount Sinai. The idea is to find drug combinations that reduce treatment resistance by lowering the diversity in the cellular response controlled by mitochondria.

Furthermore, the number of mitochondria also varies widely across cell types; for example, erythrocytes (red blood cells) do not contain any mitochondria, whereas liver cells and muscle cells such as heart cells are filled with them. It is interesting to note that heart cancer is extremely rare4. So researchers think that this could also help target drugs more specifically to cell types given their differences in mitochondrial number, such that certain concentrations and combination of drugs will kill only a specific cell type depending on its mitochondrial number.

Finally, the DEPICTIVE mathematical framework developed by IBM scientists to identify the sources behind each cell’s degree of response can be applied to other situations beyond cell death and mitochondria. DEPICTIVE can help scientists understand the cell-to-cell sources of variability in the response to a drug by measuring in each individual cell the abundance of other cellular components that are thought to cause variability, such as the concentration of any protein, enzyme, or RN; the size of the nucleus of the cell; or how much lipids, ions, or vitamins it contains.

  1. https://www.nejm.org/doi/full/10.1056/NEJM194806032382301
  2. https://doi.org/10.1083/jcb.91.3.227s
  3. https://doi.org/10.1146/annurev-genet-102108-134850
  4. https://www.ahajournals.org/doi/10.1161/CIRCULATIONAHA.115.016418

IBM Research

Abram Falk

IBM Research

More Healthcare stories

AI Models Predict Breast Cancer with Radiologist-Level Accuracy

Our team of IBM researchers published research in Radiology around a new AI model that can predict the development of malignant breast cancer in patients within the year, at rates comparable to human radiologists.

Continue reading

AI Offers Hope for Earlier Screening for Type 1 Diabetes

IBM researchers launch the first in-depth, large scale study of time courses when different antibodies appear, and their correlations to type 1 diabetes.

Continue reading

IBM and Johns Hopkins University School of Medicine Discover Unique, Pathogenic Autoimmune Cells in Type 1 Diabetes

Researchers at JHU and IBM identify new cells involved in the autoimmune attack on insulin-producing cells in type 1 diabetes.

Continue reading