A number of factors can promote the release of fat from fat cells, including:
Epinephrine, norepinephrine, adrenocorticotropic hormone (ACTH), glucagon, thyroid-stimulating hormone, melanocyte-stimulating hormone, vasopressin, and growth hormoneBut only two promote fat storage:
Insulin, and acylation-stimulating protein (ASP)*Therefore if we want to understand body fat accumulation, we should focus on the latter category, because that's what puts fat inside fat cells. Simple, right?
Can you spot the logical error in this argument?
To illustrate the problem with this argument, I'll use an analogy. When you eat food, your brain has to have a way of knowing how much has entered the body-- a feedback mechanism to keep you from overeating. The gut secretes a variety of substances that perform this task. These are called "satiety peptides" because they're secreted when you eat food, and they make you feel full.
Important processes like this tend to be redundant; in other words, the body does not rely on one signal to perform important tasks because if something goes wrong with that signal, you've got a problem. There are a number of known or suspected substances that contribute to satiety, including CCK, GLP-1, amylin, PYY, glucagon, enterostatin, and others (1). But there's one single peptide that stands out from all the others: ghrelin. Ghrelin is the only known gut peptide that promotes food intake instead of limiting it. When you administer ghrelin to animals or humans, they eat more and eventually gain fat** (2, 3).
But the interesting thing is that if you consider ghrelin in the proper biological context, it performs the same function as the satiety peptides: it constrains food intake***. How is that possible? Simple: it's regulated in a reciprocal manner to the others. After you eat a meal, satiety peptides go up, while ghrelin plummets. Both of these act to limit food intake. So these two types of signals have similar effects on food intake, but they accomplish it in a reciprocal manner.
The main point I want to make here is that factors that accelerate the removal of fat from fat cells can still promote fat accumulation if they decrease, and vice versa. All of the factors I listed at the beginning of this post can either promote or oppose fat accumulation by fat cells, depending on how they're regulated. When you think about it that way, the picture of fatty acid trafficking in and out of fat cells suddenly becomes a lot more complicated. You'd almost think we were complex biological systems evolved to regulate fat mass in a sophisticated and redundant manner!
One of the main control points for fatty acid trafficking is nerve terminals that enter fat tissue and release norepinephrine (nor = nerve, epinephrine = adrenaline). Depending on the receptors expressed by fat cells, this either causes them to release or store fatty acids (most often release). Norepinephrine is one of the dominant factors in fatty acid trafficking in/out of fat cells, and this has been universally recognized in the research community for more than half a century.
The brain is the main physiological control center of the body, and it communicates in both directions with almost every organ. It regulates the pulse rate of the heart, breathing rate via the diaphragm, blood pressure via the blood vessel walls and kidneys, regulates temperature by controlling sweat glands, hair follicles and capillaries in the skin, regulates various aspects of digestion, bone metabolism, glucose production by the liver, insulin production by the pancreas, and many other functions. So it's not much of a surprise that it also controls fatty acids moving into and out of fat tissue. Nerve terminals that release norepinephrine onto fat cells are indirectly hooked up to the brain (and ultimately the hypothalamus), and it's clear at this point that the brain exerts a powerful influence on fatty acid release and storage in fat cells via these nerves (4, 5, 6). Cutting the nerves to a specific fat depot increases its size (7). Dr. Timothy Bartness has done quite a bit of research and writing on this.
The second main point I want to make here is that the brain not only controls energy intake and energy expenditure-- factors that are obviously important determinants of fat mass-- it also influences how much fat is moving into and out of fat tissue from the circulation by acting directly on fat cells. Viewed from this perspective, it's no wonder that the brain has consistently been an important focus of obesity research over the last 150 years, and has almost universally been recognized as the central regulator of body fat mass since the 1980s. It's also no surprise that genetic studies have consistently turned up obesity risk factors in genes related to brain function, and the leptin signaling pathway in particular (8, 9). And that most if not all obesity drugs act in the brain (10).
If we want to understand the accumulation of fat in fat cells, first we have to acknowledge the complexity of the system we're dealing with. Then, we have to look beyond the proximal factors that influence fatty acid trafficking in/out of fat cells, and look for the ultimate factors that regulate these proximal factors (i.e., what originally set the ball in motion). Researchers understand this and have consequently been studying these ultimate factors for at least 150 years, and by far the most productive line of investigation to date has been the role of the brain. The role of the brain in obesity is my research specialty, and I chose this field very deliberately because I recognized how important it was. I hope to be able to convey some of this research on my blog, because not only is it fascinating, it will inoculate people against some of the odd claims circulating in the popular media.
So as for the question I posed in the title, the answer is "a lot of things". If it were simple, there wouldn't be thousands of people studying it full time. Under normal conditions****, you can't just measure one factor and predict what will happen to fat cells in an intact living organism.
* Typically ASP is ignored or downplayed in these arguments, but I'm not going to open that can of worms right now.
** Ghrelin also acts in the hypothalamus.
*** Although one could make a good argument that it's important, ghrelin's role in satiety is actually not firmly established in my opinion. One of the main reasons is that the ghrelin receptor knockout mouse has a normal meal structure. This may be because 1) the satiety system is so redundant that knocking out one element has no effect (this phenomenon is commonly observed in knockout animals), or 2) ghrelin really doesn't play an important role in meal termination. I favor explanation #1, but the jury is still out.
**** With the exception of extreme cases. For example, giving someone a shot of epinephrine, a type 1 diabetic who secretes very little insulin, a nerve to fat tissue being cut, or injecting a concentrated dose of insulin into the same fat depot for 10 years.