Comment Published in Nature

I recently read an opinion piece by Gary Taubes in the scientific journal Nature, titled "Treat Obesity as Physiology, not Physics", in which he promoted NuSI and repeated the statement that obesity research is a "house of cards" because it focuses on calories in/out, at the expense of studying the "hormonal regulatory disorders" underlying obesity (1).  I wrote a letter to the editor in response to Taubes's commentary, which has been published in Nature (2).

I'm used to seeing these kinds of claims in the popular press at this point, but to see it published in a scientific journal is galling (even if it's in the opinion section).  This is the equivalent of a person who has never held an ax telling a group of lumberjacks they need to focus on cutting trees.  It's part of a disturbing trend of popular writers in the low-carb and Paleo world attacking researchers, and even entire fields of research, they have little understanding of.  Of course this only applies to a minority of the community, but this argumentation style smells of desperation and reflects poorly on the community as a whole.

Our lab, along with many others around the world, studies the "hormonal regulatory disorders" underlying obesity.  We use energy balance (calories in/out) as a research tool to get at the underlying biology, which is the same approach NuSI itself is proposing.  The hormonal basis of obesity has always been a key focus of obesity research, and efforts in this direction have only intensified in recent decades.

The editors asked me to greatly shorten my letter for space reasons, and they also watered it down somewhat (journals don't like to publish things that undermine previously published material).  In any case I can't publish it in full here due to copyright concerns, but those who have access can read it (2).  However, I can do even better: below, I'm copying the original version of the letter, before it was cut by 75 percent:

      In his December 12 commentary in Nature, Gary Taubes suggested that obesity research may be a “house of cards” due to its focus on energy balance, at the expense of studying the “hormonal regulatory disorders” underlying obesity (1).  Yet history shows that obesity research has been studying the “hormonal regulatory disorders” underlying obesity for more than 170 years, and efforts in this direction have only intensified in recent decades.
      In 1840, the German physician Bernard Mohr published the first known account of human obesity associated with abnormalities of the basal hypothalamus (B. Mohr. Wschr Heilkd, 6:565–574. 1840).  Initially ascribed to pituitary dysfunction, lesion studies in the 1920s-1950s established that the critical area was indeed the hypothalamus (3).  The 20th century saw the identification of multiple monogenic obesity models, such as ob/ob mice, db/db mice, and zucker fatty rats, exhibiting phenotypes that some initially attributed to alterations in adipose metabolism and/or insulin signaling (4).  Experiments in the 1950s-1990s demonstrated convincingly that these animals lacked a circulating factor (or its receptor) that regulates body fatness by acting in the hypothalamus—and that this factor was not insulin (5).  In 1994, the veil was lifted with the discovery of leptin—a hormone secreted by adipose tissue that acts primarily in the hypothalamus to regulate body fatness (6, 7).  The alterations in adipose tissue metabolism of monogenic obesity in rodents (and humans) were definitively attributed to primary genetic defects in the leptin-brain axis (6, 8).  It was subsequently shown that obesity is associated with leptin resistance in the brain, and the physiological and psychological “starvation response” to fat loss is mediated in large part by a reduced leptin signal in humans (9, 10, 11).  PubMed currently lists nearly 10,000 scientific publications containing the words “leptin” and “obesity”, all of which have been published in the last 19 years.
      Researchers never stopped studying the “hormonal regulatory disorders” underlying obesity, but their findings have not supported Mr. Taubes’s carbohydrate-insulin hypothesis.