Science, Medicine, and a Gift of Gab

Whether it’s in the lab, in the operating room, or on television, this Caltech MD/PhD student is getting favorable ratings

By Michael Rogers

Houman Hemmati (left) in his Caltech office looks at a photo (right) that he took in 2002, when he assisted an all-volunteer surgical team in Guatemala. The group of mothers gathered to thank the U.S. surgeons, nurses, and assistants for treating their children.

Interviewing Houman Hemmati is not unlike being locked in a room with a stockbroker. Every other minute or so, the wall phone in his Caltech office rings. His cell phone sounds off periodically. And then his computer bleeps to life every 15 minutes as the e-mail messages pour in. With all these urgent communications, you’d assume he holds some type of high-powered position, but, in fact, he’s just a student. One of a handful of Caltech’s multitalented, hyperdriven MD/PhD students, to be sure, but nevertheless, a student.

At only 28 years old, Hemmati already has the poise of a seasoned neurosurgeon. It’s true that to be an MD/PhD student, you have to be pretty self-assured. After all, it’s hard enough to earn just one of those degrees. But how many students can walk out of an operating room into a live Today Show interview about a breaking news story and handle it like they’ve been fielding questions from the press for years?

Hemmati found himself on national TV in August 2002, right after UCLA doctors had completed a 22-hour operation to separate one-year-old Guatemalan twin girls joined at the head. Hemmati, a medical student at UCLA, had been in the operating room as an observer, and after the operation he happened to walk outside with the lead surgeon, who was about to do an interview for the Today Show. But just before the interview began, the surgeon was called back into the operating room. Hemmati found himself alone, with the camera rolling, as cohost Matt Lauer fired off questions from the Today studios in New York City. No problem—Hemmati seemed to know just what to say.

“Everyone had goose bumps at the end of the procedure,” Hemmati told Lauer. “People were cheering, people were clapping, people were crying.” And then, warming to his subject, he added, “It was more than optimistic. It was overjoyed and we can’t wait until we see these kids playing, laughing, crying like normal baby children.” Hemmati went on to do interviews that day with CNN, Good Morning America, and other programs. In later press reports about the operation, he was identified as a doctor on the case.

Meeting Hemmati, you can understand why someone, spotting him in hospital scrubs, might peg him as a high-powered MD. At six feet tall, with a ready smile and confident, articulate manner, he easily embodies the qualities that casting agents seem to look for when they need actors to play doctors on film and TV. Although he won’t officially become an MD until at least 2006, Hemmati seems to have been preparing for that role since he was a kid.

Hemmati was three years old when his family fled Iran in 1979 during the Islamic revolution and settled in the Los Angeles suburb of Encino. Hemmati’s father is a general surgeon, and when Houman was a teenager, he would often accompany his dad on weekend rounds at Northridge Hospital. Always interested in science, he worked part time for two years during high school in a biology lab at Occidental College. In 1993, Hemmati was accepted into Caltech, paid his deposit, and was planning to attend, when he received an acceptance letter from Stanford and changed his mind.

“I wanted to get a broader education as an undergrad,” he recalls. “I spoke with Caltech’s dean of admissions, and she made me promise that I’d come here for grad school. I figured I never would, since Caltech doesn’t have a medical school and I wanted to be a doctor, but I promised anyway.”

But while he was at Stanford, he says that he realized that just becoming an MD would never be enough. As an undergraduate, he assisted in the lab of biologist Irving Weissman, known for identifying and isolating blood-forming stem cells. Hemmati says that the experience with Weissman taught him that conducting research in a lab could be just as gratifying as working in an operating room.

After graduating, Hemmati says that he was burned out from his studies. His idea of a break was to get a job at UCLA, working for cancer researcher Owen Witte, who studies the relationship between stem cells and prostate cancer. In 1997, he was ready to resume his education. Still hooked on both medicine and research, he started the MD/PhD program at UCLA.

“There’s very little opportunity to innovate as an MD,” Hemmati says. “For obvious reasons, you can’t experiment on your patients. I love to innovate and experiment and I need an outlet for that. Science lets me innovate in a productive way.”

But he says that he quickly realized that he needed a smaller, more rigorous and individualized basic-science training environment. He applied to the nascent Caltech/UCLA program and was accepted in the fall of 1998, fulfilling his earlier half-hearted promise to return to the Institute. He started the PhD portion of his joint degree at Caltech in 1999.

Following his work with Weissman and Witte, Hemmati wanted to continue research into stem cells, which give rise in embryos to multiple specialized cell types that make up the heart, lungs, skin, and other tissues. More recently, as anyone who has followed the controversy over stem-cell research is aware, it has been found that in adults, embryonic stem cells can replace cells that are lost through injury, disease, or aging. Hemmati joined the lab of Marianne Bronner-Fraser, Ruddock Professor of Biology, who studies neural crest cells, a specialized form of stem cell that develops into a diverse range of cell types in vertebrate embryos.

But as a future doctor, Hemmati was also interested in studying the developing brains of humans. “I was studying stem cells in chickens and mice, and I really wanted to see human stem cells in neural development in action in real time,” he says. “And the best way to do that is to observe the nervous system during neurosurgery on children, whose brains are still developing.”

Hemmati says that unlike Caltech, where professors and graduate students freely interact and are often on a first-name basis, student–professor relationships at UCLA tend to be more formal, and he found it hard to connect with a UCLA investigator. But at some point during his first year at Caltech, he met fellow graduate student Kathy Sakamoto, who was already a doctor as well as an associate professor at UCLA (see accompanying article). When she heard that he was interested in studying developing human brains, she suggested that he meet with Jorge Lazareff, director of pediatric neurosurgery at UCLA.

His way eased by Sakamoto’s introduction, Hemmati convinced Lazareff to let him observe some of his operations. A world-renowned surgeon, Lazareff conducts a wide range of operations on children, but Hemmati was particularly eager to observe the brain-tumor surgeries. After witnessing several operations, he began to notice that pediatric brain malignancies always seemed to appear in the midline of the brain, which is also the area where stem cells develop. In an adult brain, tumors can show up anywhere.

In June 2001, during an operation, Hemmati brought up the subject with Lazareff. “I said to him, ‘Stem cells reside in the middle of the brain. Do you think there’s a connection between stem cells and pediatric brain tumors?’

“Lazareff said, ‘That’s a good question. Someone should look into it.’”

In fact, says Lazareff, other researchers were already looking into it, or at least speculating about a possible connection between stem cells and pediatric brain tumors. Stem-cell research is a fairly new discipline, but links had already been made between some forms of cancer, including breast cancer, and stem cells. With the help of UCLA research technician Michael Masterman-Smith, Hemmati started a research program to investigate the source of the tumors. Lazareff agreed to provide specimens of children’s tumors, with approval from patients’ families; Hemmati and Masterman-Smith got a $30,000 seed grant from UCLA; and they hired UCLA undergraduates to do some of the basic work. Back at Caltech, Hemmati’s PhD advisor, Bronner-Fraser, freed up Hemmati so that he could go off in his own direction.

“Discoveries often come from left field,” says Bronner-Fraser. “I believe that if you give people independence, they’ll find something unique. Students will devote themselves much more to something that they really want to do.”

Hemmati says that he owes much to Bronner-Fraser’s liberal outlook. “I’m grateful to Marianne for giving me the freedom and support to follow up on my intellectual initiatives. Her eagerness to let me explore my own interests scientifically is precisely why I chose to come to Caltech. People like Marianne are rare at any other institution.”

Pediatric brain tumors are also quite rare, occurring in only three to five children per 100,000. Although other forms of cancer, like leukemia, are more common in children, brain cancer is more deadly. Treatment options include surgery, radiation, and chemotherapy, but these don’t always work, and even successful outcomes have a high rate of serious side effects. “While two-thirds of children with brain tumors survive after five years, most of those children end up having psychological, physical, and developmental problems” due to the invasive nature of the treatments, Hemmati says.

It might seem surprising that stem cells could be related to tumors, since most of the recent publicity about them has trumpeted their potential therapeutic benefits. The explanation lies in the cells’ innate plasticity. Stem cells have the ability to renew themselves and make undifferentiated cells, but they can also give rise to many highly differentiated cell types. Says Bronner-Fraser, “The hope is that you could use stem cells for curing different diseases where a particular cell type is missing.” If Hemmati could show that stem cells are related to brain tumors, it would mean that neurosurgeons might consider getting rid of stem cells as well as tumor cells during surgery—a radical concept.

Other scientists studying this problem had hypothesized that regular neural stem cells can mutate into cancerous stem cells, which can then self-renew, producing more cancer stem cells and also giving rise to the different cells found in a pediatric brain tumor. Hemmati’s team and a separate group of scientists from Canada were the first to show experimentally that there are transformed stem cells in brain tumors.

While it may resemble an exotic underwater creature, the object in the image above is single neurosphere—a floating ball of cells—derived from neural stem cells. Houman Hemmati cultured the neurosphere from a common human pediatric brain tumor that he found contained a population of cancerous stem cells. The neurosphere has been stained with flourescent antibodies against neurons (red) and gila (green).

In the course of their investigation, Hemmati and his colleagues analyzed 20 tumors over two years. Using different analytical techniques, they were able to determine that the brain tumors contained a small yet significant fraction of cells that behaved like stem cells. “So now we had something from brain tumors that looked and acted like a stem cell, but what we really wanted was to show that stem cells associated with brain tumors can regenerate in a laboratory dish or in an animal a tumor that is nearly identical to the tumor excised from the child,” Hemmati says.

Hemmati’s team, including Masterman-Smith, UCLA professor Harley Kornblum, the UCLA undergraduates, and several graduate students and postdocs in the Bronner-Fraser lab, conducted a series of experiments showing that stem cells in the tumor differentiated into neurons and glia—the two main types of cells found in the brain—“in the same percentages,” says Hemmati, “as found in the original tumor from which they were derived. In effect, they recapitulated their ‘parental’ tumor in a lab dish.” They transplanted human tumor-derived stem cells into the brains of neonatal rats, and after one month they discovered that the cells had begun to migrate and differentiate, with the healthy stem cells producing healthy neurons and glia and the malignant cells giving rise to malignant neurons and glia. They also found that brain-tumor-derived stem cells had very high levels of a gene shown to mutate blood-forming bone marrow stem cells into leukemia stem cells, and that this gene was absent from the stem cells found in the brains of children who did not have leukemia.

“So we showed that neural stem cells have the potential to mutate into a cancer stem cell, eventually giving rise to a pediatric brain tumor,” Hemmati says. “We showed that cancer-derived stem cells can self-renew more often than healthy stem cells. We showed that they can create something that resembles a tumor in vitro, and we have a proposed mechanism as to what might be changing a normal stem cell to a cancer stem cell.”

Hemmati’s research was published late last year in the peer-reviewed journal Proceedings of the National Academy of Sciences. In a Caltech lecture about the project held earlier this year, Bronner-Fraser said, “This study makes an important advance by demonstrating a previously unrecognized connection between stem cells and pediatric brain-tumor-derived cells. By viewing tumors as a type of [undifferentiated] cell gone awry, this opens up new possibilities for diagnosis and treatment.”

Hemmati says that more work needs to be done to show that stem cells can recreate a growth as large as a tumor. “After that, we need to find the process that transforms normal stem cells to cancerous stem cells. What is it generally that causes it to be different? If we find the genetic difference we might be able to control it.”

Hemmati had to reduce his involvement in the project when he returned this past summer to medical school at UCLA, where he has two more years to go before he gets his MD. “I’ll be working 60 to 100 hours a week seeing patients. I’ll be studying internal medicine, surgery, gynecology, psychology, and other areas.”

While there’s some frustration in giving up an active role in research just as it is generating results, Hemmati clearly relishes interacting with patients. At Caltech, he took two weeks off in March 2002 to accompany Lazareff and an all-volunteer surgical team to Guatemala to treat children with neurological diseases.

The group flew to Guatemala City, spent a day repairing surgical equipment in the hospital and cleaning the operating room, and then conducted more than 30 operations on children who had a variety of neurological disorders, including spina bifida and cerebral infections. Hemmati assisted Lazareff, doing tasks like handing over scalpels or putting in shunts.

In some cases, the patient’s condition was so bad that the medical team could not operate. Since Hemmati was one of the few team members who spoke Spanish, he often had to break the bad news to the patients’ families. “It was very emotional,” he says. “The closest I had previously come to life-and-death situations was reading about them. Lazareff told me that without us, none of these children would have survived. You can’t save the world but you can make a difference. If we can do that, that’s worthwhile.”

When their work was completed, the visiting doctors and assistants were called into a hospital ward, where the mothers had gathered with their children to thank them. “They were holding their kids for the first time after the operations,” Hemmati says. “They broke into tears and then we broke into tears. And then we all hugged. It was beautiful. Some of them only spoke Indian dialects, but we understood that what they were saying was that while they couldn’t pay us, God would thank us. I still get chills thinking of that.”

While Hemmati’s abundant self-confidence has not escaped his mentors’ notice, they tend to approach it with wry humor. They also note that he works hard. Says Lazareff, “Houman is an extremely valuable team player. He’s also a great guy and a pleasure to be around. He has great enthusiasm and I sometimes have to remind him that he’s just a medical student. If we let him loose, he will be in charge of the Guatemala economy.”

Adds Bronner-Fraser, “He’s very poised and has a special talent. I can definitely see him on the Today Show as a science specialist. His coadvisor at UCLA and I often joke that we’ll be working for Houman some day.”

Before he joins the talk-show circuit, Hemmati must decide about his future in medicine and science. “For a long time I focused on being a neurosurgeon,” he says. “But that takes 60 to 80 hours of clinical work a week, leaving little time for research. There are other fields where you can do exciting clinical work and do high-quality science, such as neurology or retinal surgery.

“I like multitasking. I love talking to people and taking care of people. My ideal is to have a career in medicine, seeing patients for a certain amount of time, and to have a laboratory where I can try to answer some of the questions in clinical work that might otherwise go unanswered.”