In the December 2015 issue of the JCI, Fazeli et al. explore the role of FGF21 in mice versus humans and demonstrate a marked difference in response to fasting. FGF21 rapidly increases within a 6-hour time period in mice, and its induction is known to be associated with decreased thermogenesis and adiponectin. In contrast, their study reports that the FGF21 surge in humans occurred after 7–10 days of fasting and after ketone induction. This research demonstrates that the previous model suggesting that FGF21 is necessary for starvation-mediated ketogenesis does not translate to humans, and its role may instead be in the later stages of adaptive starvation. This month, we had the opportunity to speak with both the first and senior authors, Pouneh Fazeli and Matthew Steinhauser, about their collaboration in research as faculty members and in life as a married couple pursuing careers as physician-scientists.
Tell us about yourselves and how you both got to where you are now at this stage.
Matthew Steinhauser: I started thinking about a research career as an undergraduate at the University of Michigan. I did a summer internship after my sophomore year and landed in a lab that was really inspirational. Even after 20 years, I am still in contact with my mentors and regularly seek out their advice and help. I really got excited about research at that point! I stayed at University of Michigan for medical school and stayed in the same lab to continue my research projects. I went to Columbia for an internal medicine residency, and largely left research behind for a number of years. I did a year as chief resident and then came to the Brigham for a cardiology fellowship. By the time I got back in the lab, I had not picked up a pipette for six years, but I knew I wanted to get back into it. I joined and had a great experience in the lab of Dr. Richard Lee for my postdoc. I plunged back into it, shook off the rust, and got into cardiac regeneration, studying regenerative medicine and stem cell biology. I focused on getting good basic science training: learning how to test a hypothesis and be rigorous, and learning important techniques that had changed over the 6 years since I had been in the lab. Three years ago, I joined the genetics division at the Brigham to start my own lab focusing on adipose biology and metabolism. I spend most of my time doing research, still attend on the cardiology service, and have an outpatient clinic.
Pouneh Fazeli: My trajectory was not quite as straightforward, because I had not really done much research until my fellowship training. As a sophomore in college, I took a class in reproductive endocrinology, which made me interested in studying chronic undernutrition and women with anorexia nervosa and understanding why they developed amenorrhea. I wrote a paper for a seminar class that focused on this very topic citing a lot of the work from Dr. Anne Klibanski’s lab, who, many years later, became my research mentor. It came full circle! That one undergrad class drove me to go to medical school, and from that point, I knew I wanted to study endocrinology. I was a med student at the University of Pennsylvania and a resident at Columbia, which is where I met Matt. I did a chief residency and fortuitously ended up at Massachusetts General Hospital for my endocrine fellowship, where I worked with Dr. Klibanski studying hormonal changes in states of chronic undernutrition. This research was more of an offshoot, because it is looking at a state of acute undernutrition, but most of my research has been spent looking at hormonal changes in states of chronic undernutrition, like anorexia. While I spend most of my time doing clinical and translational research, I see patients a half day a week at the Neuroendocrine Clinical Center, and I spend a couple of weeks attending on the endocrine consult service each year as well.
Tell us how this research idea was originally generated.
Fazeli: For full disclosure, I should tell you that Matt and I are married. A lot of this research project came out of dinner conversations. I became really interested in FGF21 on the second day of my research fellowship. A Cell Metabolism paper had just come out looking at the role of FGF21 in growth hormone resistance in mouse models (1). This was very exciting, because anorexia nervosa, a state of chronic undernutrition, is also a state of growth hormone resistance. Right off the bat, I became interested in studying this hormone, because a lot of the phenotypic characteristics that we see in anorexia are also seen in the mouse model of FGF21 overexpression. My first paper involving FGF21 found an association between growth hormone resistance and FGF21 in adolescent girls with anorexia nervosa (2). I was constantly talking about this hormone, and Matt caught the bug, too. He became very excited, because it is also an adipocyte-derived hormone and is interesting from an obesity and diabetes standpoint as well; although the hormone is released during states of starvation, it also decreases insulin resistance and causes weight loss in mice. I became interested in looking into what happens to FGF21 over a 10-day fast, since there were conflicting reports in the literature. One study had looked at patients with rheumatoid arthritis during a 7-day fast, but only two time points were measured, and we were not able to see what happened on a day-by-day basis. They did find a modest 74% increase in FGF21 levels at day 7, but a number of shorter-term fasting studies showed that FGF21 did not really increase during fasting.
Steinhauser: The mouse phenotype was so robust: you fast a mouse, and within a few hours the circulating FGF21 goes up. The mouse work led to this notion that FGF21 was the prototypical fasting hormone. After much discussion, we thought there were a few possibilities: one is that the rheumatoid arthritis paper is not applicable to general healthy fasting physiology; another is that FGF21 is not a fasting hormone in humans; and third, there was just a difference between the mouse and human response, and the human response was much more protracted. By joining forces, we had the possibility of definitively resolving that question by doing work in mice, but also leveraging Pouneh’s experience with human-based research to do a more definitive fasting protocol: not only 10 days, but also serial measurements.
How do you view the interspecies difference in FGF21 between the mouse and human?
Steinhauser: I think, undoubtedly, there are some aspects that are going to be the same. One possibility for the interspecies difference is that it is just a time-scale difference — that the metabolic rate of the mouse is much faster than that of the human, at least nearly an order of magnitude faster. Therefore, what you see in a day in a mouse is equivalent to what you see in 10 days in a human. But I think that is an overly simplified explanation. If you look at some of the data in the paper, for example, surrounding the question of the FGF21 role in the ketotic response, which occurs at an early phase of starvation, you see that the FGF21 increase in mice precedes the full onset of ketosis, consistent with the notion that it plays at least some role in the ketotic response.
But in humans, we become ketotic within a day or two of starvation. This simple comparison, in and of itself, shows that the ketotic response in humans occurs before the rise in FGF21 — it actually occurs at a time when most of our study subjects had a subtle decline in FGF21. This suggests that there are some aspects of FGF21 function, at least physiologic function, that are different. There probably are some functions that are similar, such as FGF21’s role in glucose uptake. There are a lot of questions that remain to be answered.
How did you find the balance between research and clinics? How did you find the right way to marry the two in your current careers?
Fazeli: Finding the balance can be really challenging. The clinical world is very different from the research world, and when you are in one of the worlds, you feel like you should be in the other, and vice versa. For me, seeing patients clinically has always been really important. When I am seeing patients, I try to focus on just that. I try to separate the clinical world from the research world, because otherwise, you are always going to feel like you should be in the other place.
Steinhauser: In the clinic, I focus 100%, because there are always research issues to follow up on when you are in the clinic, or patient issues to follow up on when you are doing research. This is touching on something that for many is a challenge, and I am not sure what the right answer is; there is no easy answer. It also depends on whether you are doing basic science or clinical research. Certain areas of clinical research have high overlap with the clinical specialty, so the transition can be seamless. It is increasingly difficult, in this day and age, to be the triple threat that people always talked about, where you were the master clinician, researcher, and educator, especially to do so at the level at which some of the luminaries in the past did. My approach was to say, “I am going to be a great general cardiologist, but my specialty is research.” On the other hand, Pouneh has taken the opposite track by hyperspecializing her practice in pituitary disease. She has became a clinical expert in a narrow area and balances that with research. Everyone has to find their own way, but people get into trouble when they try to do too much of everything. You need focus and depth, especially during your training!
What has been the most rewarding part of pursuing a physician-scientist career?
Steinhauser: For me, it is the idea that we get paid to sit around and do experiments and think about science and solve problems. There are some days where I just pinch myself and I cannot believe that I can make a living doing that. I feel the same way about taking care of patients: patients invite you into their lives, put their trust in you, and put their lives in your hands. It is really an unbelievable privilege if you stop and think about it. It is always important to remind ourselves that it is easy to become jaded, but if you step back and think about what we have the privilege of doing on a daily basis, it is really quite amazing.
Fazeli: I agree. From the research standpoint, it’s really exciting to be able to, like Matt said, sit around and think about these very interesting areas and ideas and to try to figure out how you are going to solve a piece of that puzzle. After three years of work, there is a new little piece of this puzzle that has been solved. It is extremely rewarding! From the clinical standpoint, I think there is nothing more rewarding than taking care of patients: the trust that they put in you and how they allow you to become a part of their lives. That is a different type of reward, but equally rewarding.
References
(1) Inagaki T et al. Inhibition of growth hormone signaling by the fasting-induced hormone FGF21. Cell Metab. 2008;8(1):77–83.
(2) Fazeli PK, Misra M, Goldstein M, Miller KK, Klibanski A. Fibroblast growth factor-21 may mediate growth hormone resistance in anorexia nervosa. J Clin Endocrinol Metab. 2010;95(1):369–374.
About the First and Senior Authors
Pouneh K. Fazeli, MD, received her medical degree from the University of Pennsylvania School of Medicine and completed her residency at Columbia University Medical Center. She is currently an Assistant Professor of Medicine at Harvard Medical School and an attending physician in the Neuroendocrine Unit at the Massachusetts General Hospital.
Matthew Steinhauser, MD, received his medical degree from the University of Michigan Medical School and completed his residency at Columbia University Medical Center. He is now an Assistant Professor of Medicine at Harvard Medical School and an attending cardiologist at Brigham and Women’s Hospital.
About the interviewers
Freddy T. Nguyen is an M.D./Ph.D. candidate at the University of Illinois at Urbana-Champaign. He is the founder of the American Physician Scientists Association and served on the Associate Member Council of the American Association for Cancer Research. His research interests currently lie at the intersection of biomedical optics and cancer research. He received his BS in chemistry and BA in mathematics from Rice University.
Brittany Weber, MD, PhD, is a senior resident in the internal medicine residency program at Brigham and Women’s Hospital and a clinical fellow in medicine at Harvard Medical School. She plans to pursue a cardiovascular disease fellowship, and her scientific research interests include roles for the immune system in heart failure pathogenesis. Dr. Weber is a graduate of the Medical Scientist Training Program (MSTP) at the Perelman School of Medicine at the University of Pennsylvania, where she received a MD and PhD in immunology.
In mice, FGF21 is rapidly induced by fasting, mediates critical aspects of the adaptive starvation response, and displays a number of positive metabolic properties when administered pharmacologically. In humans, however, fasting does not consistently increase FGF21, suggesting a possible evolutionary divergence in FGF21 function. Moreover, many key aspects of FGF21 function in mice have been identified in the context of transgenic overexpression or administration of supraphysiologic doses, rather than in a physiologic setting. Here, we explored the dynamics and function of FGF21 in human volunteers during a 10-day fast. Unlike mice, which show an increase in circulating FGF21 after only 6 hours, human subjects did not have a notable surge in FGF21 until 7 to 10 days of fasting. Moreover, we determined that FGF21 induction was associated with decreased thermogenesis and adiponectin, an observation that directly contrasts with previous reports based on supraphysiologic dosing. Additionally, FGF21 levels increased after ketone induction, demonstrating that endogenous FGF21 does not drive starvation-mediated ketogenesis in humans. Instead, a longitudinal analysis of biologically relevant variables identified serum transaminases — markers of tissue breakdown — as predictors of FGF21. These data establish FGF21 as a fasting-induced hormone in humans and indicate that FGF21 contributes to the late stages of adaptive starvation, when it may regulate the utilization of fuel derived from tissue breakdown.
Pouneh K. Fazeli, Mingyue Lun, Soo M. Kim, Miriam A. Bredella, Spenser Wright, Yang Zhang, Hang Lee, Ciprian Catana, Anne Klibanski, Parth Patwari, Matthew L. Steinhauser