Alumna Ellen Kendall is a computational scientist, but part of her work during her fellowship year at the National Heart, Lung and Blood Institute involves processing blood in a leukemia lab at the NIH Clinical Center. graduate, Ellen Kendall. Kendall is a computational scientist and will be returning to CRWU for medical school. Photo by Daniel Sone.
For undergraduates at Case Western Reserve, completing a capstone research project is a rite of passage, like tossing their graduation caps in the air. It’s also a requirement. Students choose a topic and work with a faculty advisor over the course of one or two semesters. When their project is complete, many of them share their results at Intersections, a university-wide event organized by SOURCE (Support of Undergraduate Research and Creative Endeavors) and held at the end of each fall and spring semester.
This story highlights three recent alumni of the College of Arts and Sciences who were honored for their capstone projects during the 2017–18 academic year. Ellen Kendall (CWR ’18) and Nicholas Barendregt (CWR ’18) tied for first place in the Intersections poster competition, Natural Sciences category, last fall. Mark Kaminski (CWR ’18) won first place in the oral presentation competition, Biological Sciences category, last spring.
Before she started a summer 2017 fellowship at the Jackson Laboratory for Mammalian Genetics in Bar Harbor, Maine, Ellen Kendall had never heard of liposarcoma. Not many people have. It’s a rare cancer of the body’s fatty tissue, affecting only about 2,000 people in the United States annually. But it’s also deadly: The five-year survival rate is less than 50 percent, and most patients don’t respond well to chemotherapy.
During her first week working in the facility’s computational science lab, Kendall’s supervisor handed her an intriguing research paper: the results of a study examining metabolism in mice. The authors had stumbled upon something curious. When they removed two particular genes from the mice, the animals developed late-stage liposarcoma in the backs of their necks within 12 months. Kendall’s supervisor wanted her to examine the data from the study and try to suss out what was happening.
“When I was handed the data set, the researchers still didn’t have a good idea of what was going on,” recalls Kendall, who graduated with a degree in biology last May. “They had decided to publish the model in case someone would come along and take an interest. I was lucky that I got to be that someone.”
At the time, Kendall didn’t know that the assignment would spark an interest in genetic research and that liposarcoma progression would become the subject of her capstone project.
Kendall began her analysis by trying to understand what made this set of mice different from the mice in a control group and from mice that had only one gene knocked out. She broke down the data into different time periods—three, seven, and 12 months—and used genetic expression data to see whether she could determine at what point—and why—the mice began to develop tumors.
What emerged at the genetic level were some differences in metabolic pathways, or how the mice were processing food. Normal cells have the ability to break down fat into other nutrients, a process called lipolysis. But when Kendall looked at data from the mice that developed the tumors, she saw the cells were not breaking down fat, but sugars instead, a process called glycolysis.
“That was really interesting, because where the tumor was growing was in the fat,” she says. Something about the absence of these two genes caused the mice’s metabolic systems to change, possibly bringing on the cancer. Could a similar genetic malfunction also be causing the onset of liposarcoma in humans?
After finishing her 10 weeks at the lab, Kendall returned to CWRU and continued her research. This time, she wanted to see whether she could correlate the mouse findings with information gathered from people diagnosed with liposarcoma.
Using an open genetic database, Kendall compared information from human patients with the results from the mouse study. Sure enough, after weeks of crunching data and running computer models, she discovered that mice and humans with liposarcoma have similar biomarkers—enzymes, proteins or other substances in the body that indicate the presence of disease.
“When you write computer scripts and press Enter, you never know what the results are going to look like,” Kendall says. “You’re always hoping for the best, and most of the time you get nothing like you’d imagined. But I remember that day when I hit Enter on the program. It was a very aha moment.”
“Ellen dug into this project doggedly and found some very interesting patterns,” says Sarah Bagby, assistant professor in the Department of Biology and Kendall’s capstone advisor. “She was able to analyze a data set from a previous study that didn’t really tell us very much and then say, ‘No, wait, there’s something there.’ It was really neat for her to be the one to discover it.”
Kendall, who plans to attend the CWRU School of Medicine next fall, is currently a postbaccalaureate fellow at the National Heart, Lung and Blood Institute, one of the National Institutes of Health (NIH), in Bethesda, Md. In a lab at the NIH Clinical Center, she’s researching chronic lymphocytic leukemia, trying to figure out why some patients don’t respond—or develop resistance—to traditional therapies. Kendall’s methodology is similar to the one she employed in her liposarcoma work, which she’d like to return to someday. Ultimately, she’s hopeful that further research could help make a rare cancer even rarer.
Like Kendall, Nicholas Barendregt devoted his capstone project to a subject with which he was previously unfamiliar. His advisor, Peter Thomas, professor in the Department of Mathematics, Applied Mathematics and Statistics, is a computational neuroscientist, and it was Thomas who introduced Barendregt to Aplysia california, also known as the California sea slug. An 8-inch-long marine animal with a voracious appetite for seaweed, the sea slug inhabits the waters off the Pacific coast. It’s a favorite research subject of neurologists because it possesses just 20,000 easily studied nerve cells, compared to the billions that exist in humans.
Nicholas Barendregt’s capstone project introduced him to the kinds of mathematical constructs he is now using in his research as a graduate student at the University of Colorado Boulder. Photo by Patrick Campbell.
Thomas had been collaborating with Hillel Chiel, professor in the Department of Biology, who was studying the animal’s brain-body connections. Using MRI data that Chiel had collected, Thomas had created mathematical equations to model the firing of neurons as the sea slug chewed its food.
When it comes to eating, Barendregt explains, the sea slug utilizes three types of nerve cell groups, each carrying out a specific duty. When one group fires, the sea slug extends its mouth and jaws. When the next one fires, it clenches its jaws, and when the final group fires, it retracts its mouth so that it can swallow.
The way in which these groups of cells operate calls to mind a phenomenon called “winnerless competition,” originally discovered by mathematicians Robert May and Warren Leonard in the mid-1970s. That duo created mathematical models demonstrating that when there are three animal populations competing for the same thing—territory, for instance—each enjoys a period of dominance followed by a period of impotence, in a recurring cycle.
“It’s like playing an endless game of rock, paper, scissors,” Barendregt says. Rock breaks scissors; scissors cut paper; paper covers rock. And so it goes.
Recently, researchers have adapted this principle to model the firing of neurons, which is what Thomas did in the case of the sea slug. But Barendregt, for his capstone project in applied mathematics, took Thomas’ work one step further.
As Barendregt explains, even though the three groups of nerve cells take turns dominating as a sea slug eats, there’s a great deal of variability in how long each group of nerves fires. Plus, not every individual nerve cell within a group is activated at the exact same time.
Barendregt was able to create a mathematical model accounting for the variables and randomness involved and to estimate how long each cycle takes on average—something not easily accomplished.
“When you start including these random effects, the math gets a lot more complicated,” Thomas says. “Most undergraduates don’t deal with those sorts of equations. But Nick built a computational model and took several steps toward analyzing it and showing, on average, how long each cycle goes around.”
Barendregt is currently a graduate student and research assistant in the applied mathematics department at the University of Colorado Boulder, where he’s building mathematical models of how the brain accumulates and weighs evidence when making decisions. Although his current subject matter has nothing to do with sea slugs, he’s working with similar mathematical constructs. “I didn’t even know about this field two years ago,” he says. “Now I really enjoy it. It’s been a very happy discovery.”
Mark Kaminski’s capstone project covered more personal ground. Since the age of 16, he has been volunteering as an emergency medical technician, most recently for CWRU’s student-led Emergency Medical Service. As an EMT, Kaminski performed cardiopulmonary resuscitation (CPR) on three people who had suffered cardiac arrest. Unfortunately, of those three, just one survived: a 25-year-old alumnus who had collapsed while playing basketball at the Veale Center.
Mark Kaminski’s project was inspired by his experience as an emergency medical technician, most recently with the student-led emergency response team at Case Western Reserve. Photo by Mike Sands.
Kaminski’s 33 percent “save rate” in performing CPR outside of a hospital was actually higher than the average, which ranges from 4 to 31 percent, depending on the circumstances. He wondered why survival rates were so low. Could something be done to improve a patient’s odds?
His journey started at an emergency medicine conference, where he heard a researcher give a talk about people who experience consciousness while undergoing CPR. The majority of people who need resuscitation aren’t responsive, yet some show facial expressions or other signs of life, despite the fact they lack a pulse, and some remember afterward exactly what occurred. “It was the first time I had ever heard of such a thing,” Kaminski recalls. A dual biology and psychology major, he decided to investigate the topic for his capstone.
As he studied the clinical literature, Kaminski learned that the traditional way of doing CPR—a ratio of 30 chest compressions to two artificial breaths—might not be the best technique for all patients. In some cases, researchers recommend cardiocerebral resuscitation (CCR), which eliminates mouth-to-mouth ventilation and relies exclusively on chest compressions. “Recent data and evidence show that many people who collapse due to a cardiac event already have enough air in their lungs and circulatory system, so when you stop to give them a breath, it doesn’t help,” Kaminski explains. The compressions-only approach enables the responder to focus on the most urgent task: restarting the heart and moving oxygenated blood through the system.
In his capstone presentation, Kaminski argued that measuring neural activity (and thus consciousness) could help EMTs determine which approach to adopt. That’s because the results would indicate the amount of oxygen in the brain.
For instance, a patient showing no or few signs of consciousness is likely to be oxygen-deprived. In such cases, Kaminski says, EMTs should perform traditional CPR and include breaths in addition to compressions in order to introduce much-needed air into the patient’s system. On the other hand, patients whose level of consciousness indicates that the neurons in their brains are sufficiently oxygenated—at least for the moment—would benefit most from CCR.
“The brain only has enough oxygen to function for maybe 5–10 minutes, so if we don’t get the heart started before then, it will be deprived very soon,” Kaminski explains. “Our goal is to restart the heart while maintaining enough oxygen in the brain to prevent cell death and allow for its continuous, nonstop functioning and exhibition of some form of consciousness.”
But how would EMTs assess a patient’s level of consciousness? Instead of expecting them to make subjective decisions, Kaminski would equip them with EEG imaging systems in order to monitor brain activity. “I think that would greatly increase the number of saves,” he says, “because you’d be making targeted decisions based on the oxygenated state of the patient’s brain.”
“It was a privilege to mentor someone like Mark,” says Dianne Kube, a lecturer in the Department of Biology and Kaminski’s advisor on the project. “He was always prepared. He worked hard, and he showed such intellectual excitement. Whenever we’d meet in my office, I would always learn a lot and be impressed by what he brought to the table. This was really one of the best capstones I’ve ever advised on.”
Kaminski plans to pursue a medical career, perhaps concentrating on emergency medicine or neurology. And if he ever trains emergency responders, he hopes to provide them with new tools and protocols to improve the odds.