Collagen sits on a throne of lies

The Question

I know you don’t write about supplements that much. But what do you think of collagen? Is it any good? What’s the difference between collagen that has protein and, say, protein powder? Thanks love your newsletter! -Emily

The Answer

I wish I could remember where I first saw the collagen diet that turned out to be the suspected cause of dozens of deaths in the 70s. I couldn’t believe I’d never heard about it before. A diet product was suspected of killing a lot of people, and it’s not something we hear about constantly? Yes and yes, it turns out. But I am getting ahead of myself.

Osteopath Dr. Robert Linn’s The Last Chance Diet was a book published on January 1, 1976, and it describes a “protein-sparing fasting” diet. Linn wrote about a study that showed that, once bodies run through all their glucose for energy, the body started to break down fats into free fatty acids for fuel, and ketones, which “the brain can use… just as it does glucose.”

“As long as your body can use ketones and free fatty acids as ‘food,’ significant amounts of protein won’t be destroyed for glucose production,” Linn wrote. “Your body will use the ketones rather than protein because they are a quite more economical energy supply.” To keep the body from using protein, dieters had to subsist on a supplement that Linn said was “composed of all the amino acids needed to form a protein molecule” that are “present in the usable proportions necessary for this manufacture,” and with the supplement, any muscle broken down would be “replaced by the protein from the formula.”

Linn claimed dieters would lose seven to 15 pounds in just the first week, a total of 20-25 pounds the first month, the same amount the second month, and 16-18 pounds per month thereafter. “Even people who don’t start off needing to lose many pounds can also expect that spectacular drop in the first month,” he wrote. “Some have hailed it as a breakthrough in the battle against obesity,” wrote the Times.

The protein-replacing supplement was called Prolinn, in Linn’s book, though it quickly spawned imitators. Prolinn was, per Linn, "With no carbohydrates or fats and fifteen grams of protein, a two-tablespoon portion of Prolinn contains sixty calories,” Linn wrote. The Times described these supplements as “enzyme hydrolized [sic] collagen,” made from “fibrous protein collagen that is found in a variety of animal tissues and then converted in laboratories to a vaguely gelatinous liquid.” The idea was that dieters would consume only this supplement, and it would let them lose all their body fat, while the Prolinn would keep them from breaking down their muscle.

By the end of 1977, ten women following a “liquid protein diet” had died, their deaths attributed to heart irregularities; 16 more deaths were suspected to be connected. A spokesman for the FDA called it “a top regulatory priority.” Dr. Robert Linn hit back a few weeks later, calling the accusations “premature.” The director of the CDC testified that he suspected some of the deaths were due to starvation, because the diet provided them with only 300 calories per day. Sure enough, a 1980 study published in the New England Journal of Medicine found that liquid protein diets were associated with cardiac arrythmias. A 1981 study in the American Journal of Medicine found that the liquid protein diet caused a significant loss of minerals in the body, mostly in the kidneys, such that it wouldn’t necessarily be detectable through blood tests.

Soon enough, though, the controversy was forgotten. By 1989, the Times was writing about the liquid protein diet again, calling it “the diet of the stars,” even as it mentioned that 60 deaths ended up attributed to it in the '70s. Collagen had been unleashed upon the world, and its wobbling mass wasn't going back into the bottle.

If the liquid protein was all it was cracked up to be in Linn’s book, no one should have been hurt. Why did people get sick and die? Part of the issue is that no one could survive on one macronutrient for very long. But it was also because some of what Linn was saying mischaracterized how bodies work. Some of it was also outright lies about what collagen is, beginning a tradition of obfuscation about collagen that continues to this day.

The supplement mentioned, the “enzyme hydrolized collagen,” was and is still made from connective tissue (tendons and ligaments), skin, and bones—basically all the parts leftover once the animal’s been stripped of its meat. But what is collagen, really? It is true that collagen is composed of amino acids. But protein and amino acids and collagen are not all interchangeable, and it’s not as simple as “gather up enough amino acids”, yada yada, “ta-da, protein.”

There are 500 kinds of amino acids in the world, 21 of which human bodies use. Of those 21, we need to get nine of them by our diet: histidine, isoleucine, leucine, lysine, methionine, phenylalanine, threonine, tryptophan, and valine.

To understand what “getting our dietary amino acids can look like,” let’s take a 100-gram serving of chicken breast. How much of each amino acid does it have? Going down that list of the nine essential aminos, 100 grams of chicken breast has:

• 1.22g histidine
• 1.64g isoleucine
• 2.7g leucine
• 3.14g lysine
• 0.85g methionine
• 1.32g phenylalanine
• 1.47g threonine
• 1.18g tryptophan
• 1.7g valine

Chicken has more amino acids we don’t need in our diet, like tyrosine, cysteine, glycine, arginine, and proline. Other meats—pork, beef, etc.—will have slightly different amino acid makeups. Other animal protein sources—eggs, yogurt milk—will be even more different. Vegetable sources—beans, peanuts, almonds—will be different again, but we’ll come back to that in a minute. But chicken is a very good source of essential aminos; it has a lot of a lot of them, in the right proportions, so we can make use of them and turn them into muscle very effectively.

We know collagen is also an animal source of amino acids. But again we ask: What is collagen? Collagen is not as specific as “chicken breast”; depending on whether it’s from beef, pork, chicken, a hoof, a bone, skin, or a ligament, its amino acid profiles will be a little different. Most collagens are mainly composed of the amino acids glycine and proline, or glycine and hydroxyproline. The first thing we are going to notice here is none of these is an essential amino acid.

So how much of the essential amino acids are in collagen, if any? What I find funny is that few collagen supplements provide this information on their nutrition panels. Again, it varies depending on the source.[^1] According to one paper that cites numbers for collagen peptides from a “porcine” (pig) source, each 100g consisted of:

• 0.83g histidine
• 1.58g isoleucine
• 2.46g leucine
• 4.22g lysine
• 1.92g threonine
• 0g tryptophan
• 3.6g valine

(In the study's table, the methionine content is combined with non-essential cysteine, but the total amount is only 0.71g. The phenylalanine content is combined with non-essential tyrosine, and the total is 2.91g. I don’t know why this is, but chalk it up to “collagen researchers,” which we will return to in a minute).

What is the other 86 grams? It’s other amino acids we don’t need by diet, like glycine, proline, hydroxyproline, alanine, glutamic acid; plus some vitamins and minerals needed to hold the structure of the collagen together.

But not all collagen is made equal. The essential amino acid profile of collagen made just from mammal skin looks like this:

• 0.005g histidine
• .01g isoleucine
• 0.02g leucine
• 0.03g lysine
• 0.01g methionine
• 0.01g phenylalanine
• 0.02g threonine
• 0g tryptophan
• 0.02g valine

That all adds up to less than one gram of essential amino acids that we need, per 100g of collagen, still with no tryptophan. Which type of collagen are you getting in a little jar of collagen supplement, the mammal skin or the pork ligaments? Who can say.

But anyway, both of these versions of collagen share two big problems.