That’s … a stronger effect than I expected. Whoa.
Benquo
Karma: 12,489
It’s 33% of the amino acids in collagen counting amino acids. It’s less if you count them by weight/mass as glycine is lighter than other amino acids.
This was a mistake on my part and I’ll correct the article.
Masterjohn’s lower bound of 10g matches Meléndez-Hevia et al. 2009, which explicitly accounts for glycine recycling, while 60g is the highest dose used in schizophrenia treatment. He didn’t cite sources in the linked piece, though, so I’ll add a footnote with the sourcing I could find.
I haven’t been able to find evidence that typical oral supplement doses of NAC meaningfully reduce hypertrophy. I did find one paper reporting that an NAC infusion can blunt some ROS signaling after exercise, and a 2017 meta-analysis found no benefits from NAC supplementation on exercise performance, but I can’t find evidence of harm at oral doses.
I still endorse the very limited recommendation I made that people with specific reason to think they have elevated need or meaningfully limited supply of cysteine “may benefit” from the supplement.
The 1971 GRAS rescission was a precautionary regulatory action during a broad FDA review of GRAS substances. The regulation (21 CFR 170.50) cites unspecified animal studies at “high levels” and concern about increasing industrial use of glycine as a food additive.
I did not rely on glycine’s historical GRAS status in the first place, and I don’t think it’s reasonable to ask me to put words in the FDA’s mouth about why glycine was shifted from one category of permissible food ingredients to another category of permissible food ingredients in order to argue with them.
I have not been able to identify which pre-1971 studies the FDA was referring to, and neither apparently have subsequent reviewers. The rescission did not amount to an across-the-board ban; glycine is currently permitted for certain food uses under later regulations (21 CFR 172.320).
The most plausibly concerning animal finding I’ve found is a 1994 carcinogenicity study (Kitahori et al.) that found renal papillae necrosis in Fischer 344 rats given 2.5% or 5.0% glycine in drinking water for 108 weeks. Those concentrations work out to roughly 6-12 g/kg/day in rats, which scales to roughly 70-150g/day in a human. Fischer 344 rats are known for high rates of spontaneous chronic progressive nephropathy, a renal disease with no strict human counterpart, which complicates interpretation of renal findings in this strain. A follow-up study by the same group (Kitamura et al. 1996) found that glycine at 5% did not promote chemically initiated urinary lesions, while sodium aspartate did, suggesting the original renal findings may have been about chronic osmotic or pH stress from very concentrated solutions rather than glycine toxicity per se. The authors of a later 2013 study (Shibui et al.) reached the same interpretation.
That 2013 study, specifically designed to establish a toxicity threshold, found no adverse effects at the maximum tested dose of 2 g/kg/day (scaling to ~23g/day in humans). Human schizophrenia trials have used 30-60g/day for weeks to months; the main reported side effects are gastrointestinal.
While you’re sorting out the peptide formulation, here are two experiments you could try in the interim:
-
Figure out what dosage of sleep deprivation sends a clear statistical signal over what time period in what metrics. Without this, it doesn’t matter how good the rest of your protocol is.
-
Test whether glycine + NAC reduces rebound sleep need after your deprivation nights. Same design (except for whatever changes are implied by the results to 1), but instead of orexin, take 15g glycine (5g with breakfast, 5g with dinner, 5g before bed) + 600mg NAC with dinner.
The rationale for 2: A key function of sleep appears to be clearing reactive oxygen species that accumulate during wakefulness, and glycine is rate-limiting for glutathione, the main antioxidant that does that clearing. Most people on modern diets are substantially glycine-deficient. If the bottleneck on recovery from sleep deprivation is partly substrate availability for ROS clearance, then supplying the substrates should make shortened sleep more efficient, which is the thing you actually want orexin to do. It would then be interesting to see how this compares with orexin alone, or all three together.
I wrote this up in more detail here. Glycine powder is ~$0.03/g with no stability or delivery concerns, and NAC is cheap too, so the cost is basically your time.
-
It’s always gratifying to learn that the information I offered helped someone orient, and especially gratifying to get that validation so quickly. Please let me know what you find out.
Is fever a symptom of glycine deficiency?
I looked into the dipeptide signaling more carefully. Prolyl-hydroxyproline (Pro-Hyp) and hydroxyprolyl-glycine activate fibroblasts on collagen gel in vitro, but dietary collagen seems unlikely to be enough to matter systemically. Fibroblast activation in vitro was reported at about 200 nmol/mL for linear Pro-Hyp, and a more potent cyclic form at 7 nmol/mL. After taking collagen orally, human plasma reaches about 6-33 nmol/mL of hydroxyproline peptides, most of which is some form of Pro-Hyp, while cyclic Pro-Hyp alone reaches only about 0.2 nmol/mL.
Gut effects are more mechanistically plausible than systemic fibroblast signaling, since collagen peptides hit the gut first before being diluted into the full blood volume.
One human exercise-GI study was null on permeability and injury markers, though it reported a suggestive attenuation of post-exercise LPS for the collagen-treated group.
Animal and cell studies are mixed:
Caco-2 cell monolayers: collagen peptides (especially low molecular weight fraction) attenuated TNF-alpha-induced barrier dysfunction by protecting tight junction proteins ZO-1 and occludin
Acetic acid-induced colitis in rats: bovine collagen peptides reduced disease activity and histologic damage, restored ZO-1 expression, and improved colon mucosal architecture, consistent with improved barrier integrity
DSS-induced colitis in mice: walleye pollock collagen peptides worsened colitis, kept tight junction proteins low, increased pro-inflammatory M1 macrophage polarization, and shifted gut microbiota toward inflammation-promoting bacteria
The main good argument for collagen over pure glycine is a heuristic argument that favors whole foods over extracts since just because we don’t know about a cofactor doesn’t mean it isn’t helpful.
I’m not able to follow the citation hyperlinks in that summary, can you provide them?
I agree strongly that wording like “helps stabilize blood sugar” often reflects a totally backwards view of glucose management that implies that energy is a burden to be managed.
The evidence I’ve been able to find on those dipeptides is only that they serve a signaling function, not that we are rate limited on their components. So unless consuming collagen directly generates a signal that isn’t screened off by digestion, or we’re rate limited but haven’t demonstrated it yet, I don’t see the mechanism for dietary proline to matter on the margin like glycine does.
Milk is optimized for rapid growth, which is generally not what adults are doing.
They seem to have any central policy at all, which is good if you think the AI technologies currently under development will produce unsafe AGI unless they are intelligently centrally managed.
This reminds me of the Tucker Carlson interview with Russian President Vladimir Putin. Putin’s discussion of Russian and Ukrainian history seemed bizarre and unmotivated to some of my friends, but it seemed to me like the obvious intent was to explain how he understands Russia and Russia’s legitimate interests, in order to draw an intelligible distinction between aggressive and defensive acts along lines that might not otherwise make sense to foreigners with different assumptions, so that we wouldn’t be forced to construe Russia’s campaign in Ukraine as aggressive.
He was trying to explain how not to have a domino theory about Russia in Ukraine.
Other commenters have complained about your “only ever attacked people sharing a border” threshold. The map of the world doesn’t really help the US. Not only would we be pretty disturbed if China conquered all of Eurasia going border-by-border, but a small country in the Americas participating in the Belt and Road initiative could invite China to establish some sort of protectorate, after which China would only have to traverse a series of borders to get to the US. Likewise, saying that the Uighurs in Xinjiang are within China’s borders doesn’t help; the Tibetans didn’t always used to be and now they are, except for the refugees in India, which now shares a border with China.
But of course it’s equally no good to assume that a state will behave aggressively just because it could. What you would need to develop to be persuasive here is the same class of task Putin attempted when speaking with Carlson: some idea of how Chinese state decisionmakers understand themselves, and their national interests, that credibly constrains anticipations so that we can have some sense of what sorts of actions they are and aren’t likely to take.
For instance, many Americans would likely assume that “no attacks without shared borders” would preclude an invasion of Taiwan by the People’s Republic of China, but in fact both governments notionally agree that they’re in the same country, so to the Chinese, it would be intraborder, not even cross-border.
How do we decide whether the person with short sleeper genes that sleeps four hours per day is more or less “broken” than the person who sleeps eight hours?
ROS clearance, mitochondrial health, and more generally whether recovery tasks that rely on sleep are happening or not.
My impression is that they’re not similarly bottlenecks in people who consume enough protein from animal sources; if someone not on anabolic steroids is consuming at least 0.8g of protein per pound of ideal bodyweight, mostly from animal sources, most amino acids will not be bottlenecks, and glycine is unusual for being an exception, as is cysteine specifically when sick.
Interesting, I did have this misconception. Both Claude and the help-reporting-a-problem chatbot seemed to believe the opposite.
I looked into it a little and the Greek word rhema that the Romans translated as verbum originally seems to have been used simply to mean predicate, something you say about something else, in contrast with onoma, which while literally “name” seems to refer gramatically to the subject. This lines up well with your characterization of Croft’s proposal, and with Aristotelian logic.
Around 100 BC, in his book Art of Grammar, Dionysius Thrax seems to have redefined rhema to mean something more like the modern colloquial sense of “verb”: a part of speech without case inflection, but inflected for tense, person, and number, signifying an activity or process performed or undergone. This is the distinction the Roman grammarians inherited, and defines the semantic context in which Spinoza is writing.
This probably needs to be edited in somewhere as it’s an important link in the chain.
Orexin is a signal about energy metabolism. Unless the signaling system itself is broken (e.g. narcolepsy type 1, caused by autoimmune destruction of orexin-producing neurons), it’s better to fix the underlying reality the signals point to than to falsify the signals.
My sleep got noticeably more efficient when I started supplementing glycine. A key function of sleep appears to be clearing reactive oxygen species that accumulate during wakefulness, and glycine is rate-limiting for glutathione, the main antioxidant that does that clearing.
Most people on modern diets don’t get enough glycine; we can make ~3g/day but can use 10g+, because in the ancestral environment we ate much more connective tissue or broth therefrom (see Balancing Methionine and Glycine in Foods). Glycine is both important for repair processes and triggers NMDA receptors to drop core temperature, which smooths the path to sleep.
I learned much of the info here from Amber O’Hearn’s 2024 paper, Signals of energy availability in sleep: consequences of a fat-based metabolism.
ETA: I wrote this up in more detail here.
I don’t think nonconcatenative morphology ipso facto implies a different ontology.
Greek and Latin have whole different sets of morphological modifications for verbs vs. nouns, some words take the verb modification, others take the noun modifications; there are two different data structures for words. English doesn’t have that; you can say “I went on a run” and be understood with no double-takes. Some words carry “irregular” modifications proper to the source language, especially ones that come from Old English, but in a much more scattered way that doesn’t reflect a coherent categorization across the whole language.
On concrete examples of spurious irregularities, here are two from the Compendium:
The infinitive. Spinoza argues in Chapter 5 that the Hebrew infinitive “is a pure unadulterated noun” that “knows nothing about the present, nor past, nor any time whatever.” In Latin, the infinitive carries tense. Amare is present (“to love”), amavisse is perfect (“to have loved”). So grammarians working from Latin categories classified the Hebrew infinitive as a verb form, and then had to treat it as deficient because it lacks the tense marking a “verb” should have. But Spinoza says it doesn’t carry tense because it isn’t a verb. It’s a name for an action. (The Hebrew metalinguistic term for “infinitive” is shem hapo’al, literally “name of the action.”) The “irregularity” disappears when you stop expecting it to behave like a Latin infinitive.
The passive reflexive. In Chapter 21, Spinoza identifies a verb form he says “seems to be unknown to all the grammarians whom I know.” The specific example is Ezekiel 23:48, where the form וְנִוַּסְּרוּ (weniwwasseru, “that they may be taught”) appears. This is what modern grammarians now call the Nithpael, a passive reflexive combining features of the Niphal and Hithpael.
Here’s how one gets there. We start with a root י-ס-ר (yasar) indicating the idea of applying discipline. Render it reflexive and it’s hithyasser, meaning “applies self-discipline”. Then apply the passive and it means “made to be self-disciplining.”
From a Greco-Latin perspective you encounter this word and the first question you ask is, “which schema do I apply? Is it a verb? Is it a noun?” and it’s hard to answer. But it’s a whole compound predicate. It’s unusual to do that by just stacking two operators on a root even in Hebrew, but it’s not as weird as it would be if you had to first categorize a word as a verb or a noun, and then apply the appropriate modification within that data structure.
The standard framework of Spinoza’s era recognized seven binyanim (standard modifications of a root) and this form didn’t fit any of them, so grammarians either emended it as a scribal error or classified it as anomalous. Spinoza argues it’s a regular pattern the grammarians couldn’t see because their framework didn’t have a slot for it. Modern Hebrew grammar now reifies the Nithpael as a rare but real stem formation.
I think this belongs in the post so I’m going to edit it in.
Many nodes being a single logical agent is ideally compatible with them taking the sorts of adaptive actions over time consistent with being different causal (forwards-in-time) agents.
Shorter sleep periods and longer wake periods with no new health, mood, energy, or cognitive problems, or improved recovery from stressors like exercise, infection, or injury with no longer sleep periods.