Misconceptions Surrounding DNA Methylation

[Travis]

DNA Methylation: The Confusion over 5-Methylcytosine

This post shows why vitamin B₁₂ and glutamate are the only things capable of significantly changing DNA methylation levels, why methionine and folate are ineffective, and points-out the absurdity of the modern paradigm.​

The methylation of DNA is an undeniable phenomenon, and found in most living organisms. The enzymes responsible for this have been extracted, purified, assayed—with their kinetic rates determined, giving indisputable proof that DNA methylation can be an enzymatic event. The DNA encoding for such enzymes have been genetically sequenced, synthesized, and recombined to form fully-functional constructs . . . and the X-structures have been elucidated.

But there exists three forms of methylated DNA: oxygen-linked, as in the case of 6-O-methylguanidine; nitrogen-linked as in the case of N-6-methyladenosine, N-3-methylthymine, and N-4-methylcytosine; and surprisingly, carbon-linked as in the single case of C-5-methylcytosine. All of these modified nucleotides exist and can be found in DNA, and one of them has dubious origins—a biolochemical unicorn.

While this may seem trivial, it is not; as you will see. Misunderstandings surrounding the origin of 5-methylcytosine obfuscates epidemiological correlations, hides important cellular control mechanisms, and can perhaps can even lead to further misunderstandings if compounded—snowballed around the original misconception.

I. The Modern Paradigm

"Thereby, methylation adds extra information to the DNA that is not encoded in the sequence, and the methylated bases can be considered the 5th, 6th, and 7th letters of the genetic alphabet." —Albert Jeltsch
​

Most people have heard of DNA methylation, and had probably heard some speculations towards its purpose. There exists hundreds, if not thousands, of studies attempting to correlate DNA methylation with such things as: longevity, cancer, and diabetes—all with basically null results. Much paper has been wasted in this manner, with a small forest killed by attempting to explain a few classic rat–methionine studies in this way.

But a clear trend has emerged: Methylated DNA strands appear to be more stable; they are slighly-downregulated genes. The methylation of DNA is the protoypical epigenetic mechanism, and the only one acknowledged by many biologists.

"In higher eukaryotes DNA methylation is the only known covalent modification of the DNA." —Albert Jeltsch​

These methylations are also thought to protect cellular DNA from its own enzymes—enzymes which could exist to destroy foreign DNA, or as simply a recycling measure. There are many other speculations on the function and purpose of DNA methylation, but a few things should perhaps be considered before entertaining some of the more abstract and tenable hypotheses out there.

Jeltsch, Albert. "Beyond Watson and Crick: DNA methylation and molecular enzymology of DNA methyltransferases." Chembiochem (2002)

II. Why Carbon:Carbon Bonds are Different

The formation of the carbon:carbon bond is a relatively rare event in biochemistry—almost unheard-of without a carbonyl or phosphate involved. The electronegative oxygen pulls electrons from the carbonyl carbon, allowing backside attack from electron-rich nucleophiles such as a thiols and amides. The condensation reactions in steroid synthesis make use of phosphate groups on the terminal carbons, 'activating' them with electronegativity.

The DNA methyltransferase reaction—which catalyzes the carbon:carbon bond formation of 5-methylcytosine—involves adding a methyl group to a resonant ring. You'd be hard-pressed to find another instance of such, or even another biochemical carbon:carbon bond formation besides maybe the rare Diels–Alder.

The #5 bond of methylcytosine is the only variable carbon:carbon bond in DNA. All other methylations are to either to nitrogen or oxygen, additions which are in no way unusual: Such examples can be found in familiar molecules such as melatonin, methoxytyramine, and epinephrine. So when talking about 'DNA methylation' it becomes necessary to make the distinction between what atoms have been methylated, or assumed to have been.. .

"So far, in metazoa only cytosine-C5 methylation has been found in DNA; this methylation mainly occurs at CG sequences, about 60 ± 90% of which are modified in mammals (corresponding to 3 ± 8% of all cytosine residues)." —Albert Jeltsch
​

The methylated DNA base called 5-methylcytosine gets the most attention, as it's known to be the most effective at silencing genes—affecting both double-helix stability and replication more than the N-linked and O-linked varieties.

It was also the first one discovered.

Jeltsch, Albert. "Beyond Watson and Crick: DNA methylation and molecular enzymology of DNA methyltransferases." Chembiochem (2002).

III. Help from DNA Methyltransferase

"Although AdoMet is a very effective donor for methyl groups, methylation of cytosine residues at position 5 is not a trivial reaction." —Albert Jeltsch
​

The enzyme DNA methyltransferase is thought to do what no other can. It breaks the resonance of the pyrimidine ring with a protonated glutamyl, while forming a thioether with carbon four of cytosine. You might be tempted to wonder why this is so, but you wouldn't want to since it doesn't happen.

"DNA (cytosine-5) methyltransferases introduce a methyl group onto carbon 5 of the target cytosine through a covalent intermediate between the protein and the target cytosine. During this process, the cytosine is flipped 180° out of the DNA backbone into an active site pocket of the enzyme." ―Sriharsa Pradham
​

Carbon five—the site of the imaginary reaction—actually faces the DNA backbone. The enzyme is imagined to work on single-stranded DNA: It is also thought to perform miracles.

"Therefore, the catalytic mechanism of C-MTases can explain why DNA MTases developed a base-flipping mechanism." ―Jeltsch
​

In step four, an unidentified base materializes out of nowhere to abstract a proton from carbon five.

"The nature of the proton abstracting base is not known with certainty," —Albert Jeltsch​

This is a base later found to not exist at 2.8 Å resolution.(Reinisch, 1995).

Of course all of this is too small to see, so they can just sprinkle fairy dust to obscure any irregularities. Step four is technically the molecular equivalent of Bruce Lee in "The Way of the Dragon."

"Although these studies were carried out with different enzymes and none of them finally proves that a base rotating mechanism is operative." —Albert Jeltsch
​

Jeltsch, Albert. "Beyond Watson and Crick: DNA methylation and molecular enzymology of DNA methyltransferases." Chembiochem (2002)

​

IV. Failure of S-Adenosylmethionine to Increase Methylation

"Although MET supplementation significantly decreased the [SAM]:[SAH] ratio in liver and brain, no significant dietary effects on genome-wide DNA methylation were found." —Robert A. Waterland​

All so-called DNA methyltransferase enzymes use S-adenosylmethionine as cofactor. One would then expect this cofactor to increase total DNA methylation by ingestion or by injection. It does, after all, the source of the methyl groups in question.

"Methionine supplementation increased the renal concentration of SAH without changing the SAM/SAH ratio. This unchanged profile was also observed for DNA methylation at the promoter region of the p53 gene." —Cátia Lira DoAmaral​

The failure of studies to demonstrate changes in 'DNA methylation'—as defined as 5-methylcytosine—by methionine supplementation shouldn't go unnoticed. It's unfortunate that many researchers fail to differentiate between the possible and the impossible methylation events in titles and abstracts, as this only adds confusion to an already confused paradigm.

Waterland, Robert A. "Assessing the effects of high methionine intake on DNA methylation." The Journal of nutrition (2006)
Do Amaral, Cátia Lira. "The effects of dietary supplementation of methionine on genomic stability and p53 gene promoter methylation in rats." Mutation Research/Genetic Toxicology and Environmental Mutagenesis (2011)

V. Failure of Folate to do Likewise

"...however, there was not a consistent association between global DNA methylation and folate status across studies." —Krista S. Crider​

Although not considered a cofactor for the DNA methyltransferase reactions, folate is dragged into this imply because it regenerates S-adenosylmethionine. But if S-adenosylmethionine itself cannot even influence DNA methylation, how can you folate capable of this?

When measuring the effects of folate on hundreds of people, the results were negligible—basically null.

"Women with higher (vs. lower) RBC folate had higher mean DNA methylation (5.12 vs. 4.99%) in the pre-fortification period, but lower (4.95 vs. 5.16%) DNA methylation in the post-fortification period." —Sajin Bae​

This was defined with respect to 5-methylcytosine, or course. Increases in nitrogen-methylated DNA is certainly possible from supplemental methionine, but N-methylation is rarely studied.

"Global DNA methylation was assessed by liquid chromatography-tandem mass spectrometry and expressed as a percentage of total cytosine." —Sajin Bae​

Should anyone think that a .03% change is something to consider, take a look at what vitamin B₁₂ can do. It can cause great changes in spite of the fact that it's not a cofactor for DNA methyltransferse.

Crider, Krista S. "Folate and DNA methylation: a review of molecular mechanisms and the evidence for folate’s role." Advances in Nutrition: An International Review Journal (2012)
Bae, Sajin, et al. "Impact of folic acid fortification on global DNA methylation and one-carbon biomarkers in the Women's Health Initiative Observational Study cohort." Epigenetics (2014)

VI. Vitamin B₁₂ Increases Methylation—and Nothing Else

"After 10 wk, the colonic DNA of the deficient rats displayed a 35% decrease in genomic methylation..." —Sang-Woon Choi
​

The only thing shown to reliably change DNA methylation has been vitamin B₁₂. Nobody seems to mention this fact, a fact which conflicts with the commonly held notion that the S-adenosylmethionine—and anything that re-methylates it—will do this. Folate and choline could perhaps be imagined as almost forcing methyl groups onto DNA by positive pressure in the methyl pool: two hydraulic analogies with no place in the genesis of 5-methylcytosine—although perhaps appropriate when talking about catecholamines and homocysteine.

"Identification of cytosine and 5-methylcytosine was obtained by MS analysis of chromatographic peaks." —Sang-Woon Choi​

By unambiguous methods: Vitamin B₁₂ had been shown to increase cytosine methylation; something which cannot be achieved reliably by either folic acid or S-adenosylmethionine—the definitive methyl donor, making this especially enigmatic since S-adenosylmethionine is the very cofactor for the enzyme thought responsible.

Choi, Sang-Woon, et al. "Vitamin B-12 deficiency induces anomalies of base substitution and methylation in the DNA of rat colonic epithelium." The Journal of nutrition (2004)

VII. How 5-Methylcytosine is Actually Formed

Regular cytosine is formed from the amino acid aspartate. This is a straghtforward process, and had been known since the '50s.

A small metabolite called carbamoyl phosphate is attacked by the amide (shown below as an amine but the hydrogen is removed by the enzyme) of aspartate, held in position and acted-upon by the enzyme carbamoyl transferase. This is not a carbon:carbon addition, and neither is the final condensation stage. Such interactions are known to occur all the time.

But when β-methylaspartate is acted upon by these enzymes, you would expect the 5-methylcytosine precursor to be formed: 5-methyldihydroororate, or 5-methyldihydroororic acid (depending on pH.)

Which would then go on to become 5-methylcytidine, and then incorporated into DNA. The substrate β-methylaspartate is no stranger to biology: This molecule is formed through the vitamin B₁₂-catalyzed enzyme glutamate mutase—named so because it actually isomerizes glutamate into β-methylaspartate, similar to B₁₂'s ability to isomerize methylmalonic acid.

Vitamin B₁₂ is only thing capable of isomerizing glutamate into β-methylaspartate.

Now this is the formation of a carbon:carbon bond, to be sure, and could appear no less difficult than the one in question: the methlyation of cytosine on single-stranded DNA. But keep-in-mind that vitamin B₁₂, or cobalamin, is a giant heme-like ring structure with a cobalt atom in the centre. This gives it high electric potential, something that the DNA methyltransferase or S-adenosylmethionine simply doesn't have.

When the glutamate/aspartate ratio is high, you might expect more glutamate to be be converted to β-methylaspartate through the action of glutamate mutase. This is the precursor the 5-methylcytosine, and has nothing do with methionine.

Yates, Richard A. "Pyrimidine biosynthesis in Escherichia coli." Journal of Biological Chemistry (1956)

VIII. Confirmation from Neurotransmitter Studies

"Our data suggest that DNA methylation status controls transcription-dependent regulation of glutamatergic synaptic homeostasis." —Jarrod P. Meadows​

You might also expect this to happen anywhere . . . even in the brain, the place most suitable for such an investigation. The brain has a highly-variable glutamate/aspartate ratio, and thus is a prime candidate for such analysis. These two amino acids are considered to be neurotransmitters involved in learning, motivation, and memory.

"For this reason, many years ago, Francis Crick proposed that a self-perpetuating biochemical autoconversion of methylated DNA might serve as a memory mechanism at the molecular level." —Jarrod P. Meadows​

Following in the footsteps of Francis Crick, Jarrod Meadows had decided to investigate the effect of glutamate on DNA methylation in the brain. He used tetrodotoxin, as this is the standard molecule used to inhibit glutamate release in the brain. The low brain glutamate was confirmed, and he had found that this had resulted in DNA methylation changes greater than those resulting from inhibiting DNA methyltransferase. Tetrodotoxin, ostensibly through glutamate, did this in spite of contributing no methyl groups whatsoever.

"Inhibiting neuronal activity with tetrodotoxin decreased the cytosine methylation of and increased the expression of genes encoding glutamate receptors and trafficking proteins," —Jarrod P. Meadows​

Decreasing glutamate would inhibit the formation of β-methylaspartate, the precursor for 5-methylcytosine, through glutamate mutase. This would lead to lower DNA methylation, exactly as was found.

Other studies have confirmed such a relationship, between asparate levels and DNA methylation (Punzo, 2016).

In the Meadow's study: The genes hypomethylated from low glutamate were, and perhaps not surprisingly, involved in aspartate signalling:

"An additional series of studies found that the Bdnf gene locus is also subject to memory-associated changes in DNA methylation and that this effect is regulated by the N-methyl-D-aspartate (NMDA) receptor," —Jarrod P. Meadows
​

And also in glutamate signalling:

"One particularly interesting possibility is that regulation of methylation of the genes for AMPA-subtype glutamate receptors might be involved, as well as methylation of those genes regulating their trafficking. For example, Jayanthi et al. have observed activity-induced alterations in AMPA receptor gene methylation that correlate with glutamate receptor expression in vivo. Previous studies as well as our own data have demonstrated that TTX-induced synaptic scaling is associated with altered glutamate receptor gene transcription as well as that of arc, an AMPA receptor trafficking regulator." —Jarrod P. Meadows​

This strengthens the idea that the glutamate/aspartate ratio is controlling DNA 'methylation' on the very genes responsible for its regulation—having nothing to do with methionine. This might seem paradoxical, if not for the facts outlined above. With these in mind, this all makes perfect sense.

Meadows, Jarrod P., et al. "DNA methylation regulates neuronal glutamatergic synaptic scaling." Science signaling (2015)
Punzo, Daniela, et al. "Age-related changes in D-aspartate oxidase promoter methylation control..." Journal of Neuroscience (2016)

IX. Hints from Genetics

The methylation of DNA—as defined by 5-methylcytosine—is routinely measured on select genes. This is because: Some regions are particularly-rich in cytosine, leading to more accurate measurements in these places. One of the genes commonly tested for methylation happens to be the one encoding for D-aspartate oxidase:

"The choice of Ddo as a model gene was mainly based on the facts that this gene is developmentally regulated in brain by DNA methylation changes. Age-Related changes in D-aspartate oxidase promoter methylation control extracellular D-aspartate levels and prevent precocious cell death during brain aging. The average methylation levels range between 60 and 30% in different stages." —Ermanno Florio​

You would then think that highly-'methylated' areas would be found in the genes responsible for controlling glutamate:aspartate metabolism, including—but not limited to—aspartate transaminase, γ-glutamyl transferase, and glutamyl mutase.

Florio, Ermanno, et al. "Tracking the evolution of epialleles during neural differentiation and brain development: D-Aspartate oxidase as a model gene." Epigenetics (2017)

X. Five-Methycytosine

"In Fig. 2 a minor constituent designated “epicytosine” is indicated, having a migration rate somewhat greater than that of cytosine." ―Rollin D Hotchkiss​

Was the sixth nucleotide discovered, and the first assumed to be a post-translational modification. Originally discovererd in 1925, it's existence hadn't been confirmed until 1948. Two years later its existence was headlined in the high-visibility journal Nature.

"...nevertheless, the fraction is distinct from cytosine and is clearly not uracil." ―Rollin D Hotchkiss​

Five-methylcytosine was on the minds of researchers then, and the nitrogen-methylated nucleotides weren't known to exist until later.

Johnson & Coghill. "Researches on pyrimidines. The discovery of 5-methyl-cytosine in tuberculinic acid..." Journal of the American Chemical Society (1925)
Hotchkiss, Rollin D. "The quantitative separation of purines, pyrimidines, and nucleosides by paper chromatography." Journal of Biological Chemistry (1948)
Wyatt, G. R. "Occurrence of 5-methyl-cytosine in nucleic acids." Nature (1950)

XI. DNA Methyltransferase

"The rate of methylation observed in this study (31 pmoles/mg DNA/hour), although 85-fold higher than the rate using DNA from mid-log phase cells, is low compared to the calculated pmoles of methyl group needed for saturation of DNA completely devoid of methyl groups which is calculated to be on the order of 15,000 pmoles/ mg DNA." —T.W. Sneider​

With 5-methylcytosine the only known 'DNA methylation' product, any transfer of methyl groups measured from S-adenosylmethionine to DNA would simply been assumed to be those forming 5-methylcytosine. The variable nature of 5-methylcytosine content found in DNA further would further confound such measurements, turning the negligible enzymatic rates into something even less.

Moreover, Sneider was likely measuring the sum of all of the DNA methytransferase rates combined, even the real ones. This is a possibility that he'd even mentioned himself, a likelihood which would further marginalize the already snail-like catalytic rates. He also didn't prove that the methyl groups were actually transferred to cytosine at carbon five; Sneider had simply measured the total radioactivity, using a scintillation counter, of a fraction that wasn't even pure cytosine—a fraction simply purified by extraction and centrifugal techniques.

Strange as it may sound, I don't think this transfer has even been proven. The more modern kinetic studies simply measure the sum of ·CH₃ transferred, assuming that they're going to carbon five as though a foregone conclusion. They never attempt to differentiate between 5-methylcytosine and N-methylcytosine, neither analytically or even conceptually.

"The measured methylation rate constant kchem = 0.26 s⁻¹ for WT M.HhaI agrees well with that recently reported." —Giedrius Vilkaitis​

In every single case, it appears, biochemists simply measure the total ¹⁴CH₃ transferred to DNA. There is no reason to assume, besides the enzyme's formal name, that these methyl groups are adding to cytosine's carbon number five. As if caught in a semantic web, the inappropriate name given to this enzyme forces them to make one big assumption: The enzyme called cytosine-5-methyltransferase selectively transfers methyl groups to cytosine's carbon five.

"The turnover number for the enzyme varied considerably among the DNA templates, from ∼1 to 50 h⁻¹..." —Albino Bacolla​

Rates of less than one methyl group transferred per second were measured.

"With oligonucleotide substrates, the catalytic activity of Dnmt3a is similar to that of Dnmt1: the Km values for the unmethylated and hemimethylated oligonucleotide substrates are 2.5 μM, and the kcat values are 0.05 h⁻¹ and 0.07 h⁻¹, respectively." —Albert Jeltsch​

This is actually slower that Na⁺/K⁺-ATPase, almost like it's not even trying.

"However, the distortion that occurs is as surprising as it is elegant: the m5C-MTases cleanly extend the target cytosine out of the helix and into the catalytic site, without seriously disturbing the rest of the DNA helix." —Kumar, Sanjay​

By contrast, an enzyme known to methylate adenosine on the nitrogen is known operate over a million times faster (Reich, 1992).

Sneider, T. W., W. M. Teague, and L. M. Rogachevsky. "S-adenosylmethionine: DNA-cytosine 5-methyltransferase from a Novikoff rat hepatoma cell line." Nucleic acids research (1975)
Vilkaitis, Giedrius, et al. "The mechanism of dna cytosine-5 methylation kinetic and mutational dissection of hhai methyltransferase." Journal of Biological Chemistry (2001)
Bacolla, Albino, et al. "Recombinant human DNA (cytosine-5) methyltransferase II. Steady-state kinetics reveal allosteric activation by methylated DNA." Journal of Biological Chemistry (1999)
Gowher, Humaira, and Albert Jeltsch. "Enzymatic properties of recombinant Dnmt3a DNA methyltransferase from mouse: the enzyme modifies DNA in a non-processive manner and also methylates non-CpA sites." Journal of molecular biology 309.5 (2001): 1201-1208.
Reich, N. O., et al. "In vitro specificity of EcoRI DNA methyltransferase." Journal of Biological Chemistry (1992)
Kumar, Sanjay, et al. "The DNA (cytosine-5) methyltransferases." Nucleic acids research(1994)

XII. The CpG repeat

Carbon five-'methylated' cytosine (C) is found more often than next to a guanosine (G) on the DNA backbone—separated by a phosphate (p). So much so, in fact, that CpG islands are considered indicative of genes controlled by 'methylation,' and they are often taken as representative of DNA 'methylation' potential. This cannot be denied and needs to be stressed; this is of of prime importance.

The prevalence of CpG islands in DNA sequences should tell us whether or not said gene is controlled by methylation. These repeat is unidirectional; it's not synonymous with its reverse (CpG ≠ GpC). Randomly expected prevalence of such repeats is estimated at 4.41% of the genome, though you probably would have guessed 6.25% (¹⁄₄ × ¹⁄₄ = ¹⁄₁₆). [This is because cytosine and guanosine are found at a prevalence of ~21%, not 25%.] However, the actual prevalence of CpG sequences—in humans—is found to be ~1%.

"...the frequency of CpG is only 20% of the predicted value." ―Nigel Atkinson​

Under the scheme that I am proposing, you would expect the genes that control aspartate–glutamate metabolism to be "CpG islands." Let us take a look at one common definition of the "CpG island."

"Q: How is it defined? A: 200-3000 bp in length, and greater than 60% CpG [relative to expected value]." ―Nigel Atkinson ​

The aforementioned island, the one for D-aspartate oxidase, certainly does fall under this definition. The first 1000·bp of this gene has a found/expected ratio of 29.4—the Australia of CpG islands (using 1% as the denominator). In contrast, the first 1,000·bp of the gene encoding fatty acid synthase has only five CpG repeats, giving it a CpG(f)/CpG(e) ratio of exactly one—the expected prevalence for CpG repeats.

Atkinson, Nigel. "Biology 327: Epigenetics." University of Texas—Austin
Anonymous Sequencer. "D-aspartate oxidase, exon 1-3." European Nucleotide Archive
Anonymous Sequencer. "Chicken fatty acid synthase gene." European Nucleotide Archive

XIII. Confirming the Hypothesis

The gene which encodes aspartate transaminase in Ralstonia solanacearum is listed as 1185·bp long. Of the first one thousand nucleotides, there are 135 CpG repeats. The gives this gene a found over expected ratio of 27. Since the gene which encodes D-aspartate oxidase is an established CpG island, this one must also be considered a CpG island.

The gene encoding γ-glutamyl transferase has a ratio of 29 for its first 1000·bp, higher than D-aspartate oxidase. This represents a 2900% enrichment over what you'd expect to find from a random sequence.

As the enzyme which creates β-methylaspartate, and thus 5-methylcytosine, you'd expect gluatamate mutase to enriched. This enzyme occupies a central hub in aspartate–glutamate metabolism. For this reason, I had decided to count the CpG repeats in the entire 4042·bp gene. I found 374 such repeats, for an absolute prevalence of 18.5%. The expected prevalence of 1% can even be considered generous in light of findings.

"Thus in human DNA, where the fraction of (G+C) is 0.4, we would expect CpG to occur with a frequency of 0.2 x 0.2 = 0.04, whereas the observed frequency is about 0.008. [.8%]" ―Adrian P.Bird​

Anonymous Sequencer. "Ralstonia solanacearum Aspartate transaminase." European Nucleotide Archive
Anonymous Sequencer. "Streptomyces malaysiensis putative gamma-glutamyl transferase." European Nucleotide Archive
Anonymous Sequencer. "Citrobacter amalonaticus DNA for glutamate mutase." European Nucleotide Archive
Bird, Adrian P. "DNA methylation and the frequency of CpG in animal DNA." Nucleic acids research (1980)

XIV. The Methylaspartate Cycle

"Haloarchaea (class Halobacteria) live in extremely halophilic conditions and evolved many unique metabolic features, which help them to adapt to their environment. [...] Aerobic haloarchaea gained two anaplerotic acetate assimilation pathways, the glyoxylate cycle and the methylaspartate cycle." ―Farshad Borjian
​

Halobacteria have a methylasparate cycle. This is a metabolic cycle; a branch-off of the Citric Acid Cycle where glutamate is isomerized by glutamate mutase, deaminated, and used for energy.

"The methylaspartate cycle branches off the tricarboxylic acid cycle on the level of 2-oxoglutarate. [...] It allows the separation of the flows of the methylaspartate and tricarboxylic acid cycles, thus preventing the competition between these two cycles for intermediates." ―Farshad Borjian
​

Holobacteria have much higher β-methylasparate than mammals. From this, you would expect it to have more 5-methylcytosine in its DNA.

This certainly appears to be the case. The nucleic cytosine of halobacteria is 100% 5-methylcytosine.

"The genome of ΦN consists of linear double-stranded DNA, 56 kb in size, whose dCMP is totally replaced by 5-methyl-dCMP. This is the second case of a fully cytosine-methylated genome..." ―Heike Vogelsang-Wenke
​

More indication that DNA 'methylation' is simply a function of β-methylasparate levels (C-methylation).

"Therefore, the methylaspartate cycle appears to be a glutamate overflow mechanism: it functions only at glutamate concentrations, which exceed a certain threshold signalizing the availability of nitrogen and energy to perform anabolic reactions." ―Farshad Borjian​

Borjian, Farshad, et al. "The methylaspartate cycle in haloarchaea and its possible role in carbon metabolism." The ISME journal (2016)
Vogelsang-Wenke, Heike. "Isolation of a halobacterial phage with a fully cytosine-methylated genome." Molecular and General Genetics (1988)


XV. Anatomy of a Unicorn

The enzyme usually assumed to create all the natural 5-methylcytosine on the planet has been purified, crystallized, and imaged. This was done despite it's extremely low catalytic rates, conversions which would likely approach that of S-adenosylmethionine and DNA—'enzyme' excluded. Nonetheless, this has been done. The structure of this enzyme has been characterized down to 2.8 angstroms, and even more bizarre reaction schemes had to be created to account for this increased resolution.

"The basic moiety that abstracts the C5-H proton is not obvious from either complex." ―Karin M. Reinisch​

The magic base, usually just assumed to make an apparition at the right moment, was nowhere to be found.

"Alternatively, a water molecule could act as the catalytic base. If so, there is no evidence for it in the M. Haelll electron density maps, but this may simply be a consequence of limited resolution (2.6 Å)." ―Karin M. Reinisch​

So conspicuous was its absence that water was even hypothesized to assume this role—to actually be capable of extracting a hydrogen from an sp³-hybridized carbon. But thankfully, there was no evidence of this; so we don't have to think too hard about dissolving in rainstorms.

"a carboxyl oxygen of Glu-109 is hydrogen-bonded to the cytosine N3 at near neutral pH, the carboxyl pKa must be shlfted relative to the solution value of approximately 4.5. This shift, assumed in the proposed catalytic mechanism, is plausible, because Glu-109 is relatively solvent inaccessible." ―Karin M. Reinisch​

Glutamate needs to become glutamic acid for this scheme, which is difficult to imagine at bodily pH. This doesn't seem to matter however, since the interior of any enzyme can always be imagined as 'inaccessible.' But this contradicts their very scheme, since they need water to abstract the proton from carbon five.

"the base pairing rearrangement required by M. Haelll is not energetically feasible." ―Karin M. Reinisch​

This enzyme is presumed to work on double-stranded DNA by these authors, a fact which forces the reaction scheme from the impossible towards the absurd:

"M. Haelll provides an even more extreme example of DNA distortion." ―Karin M. Reinisch
​

The initial step, the incorporation of the target cytosine, is energetically impossible: A 'breathing' mechanism is actually proposed to account for this.

"The lifetime of a GC pair in B-DNA is consistent with the relatively slow turnover rate of methyltransferases (e.g., 0.02/s for M. Hhal). Thus, the enzyme could capture the DNA as it breathes. However, the additional base rearrangements in M. Haelll are almost certain to be energetically costly, since the change in base pairing requires unstacking the DNA both opposite Gl 1’ and between pairs 9-9’ and 1 l-10’." ―Karin M. Reinisch
​

Is there anything this enzyme can't do?

"Second, it is more plausible that the substrate cytosine is flipped out as the DNA breathes." ―Karin M. Reinisch
​

A more plausible mechanism for the creation of 5-methylcytosine is through β-methylaspartate, before it's incorporated into DNA.

Reinisch, Karin M., et al. "The crystal structure of Haelll methyltransferase covalently complexed to DNA: An extrahelical cytosine and rearranged base pairing." Cell (1995)

XV. Lack of Evidence for Carbon–Carbon Methylation

There is a conspicuous lack of evidence that this enzyme can do as stated—that it can actually transfer a methyl group to the five carbon of cytosine.

XVI. Implications

The assumption that this 'enzyme' controls DNA methylation influences the conclusions made in hundreds of scientific articles. This assumption appears to have started in the years following its 'discovery,' and was highly influenced by Sneider having it named "DNA-cytosine 5-methyltransferase." He jumped the gun, made assumptions, and thought he'd found a way to explain 5-methylcytosine's existence in DNA. But any radioactive methyl groups were most likely simply transferred to the 4-nitrogen during those studies . . . and at snail-like pace.

There is no indication that the nuclear 'enzyme' called cytosine 5-methyltransferase is responsible for 5-methylcytosine's existence in DNA, and every indication that β-methylaspartate is. Consuming more glutamate than asparate would be expected to increase increase β-methylaspartate levels, 5-methylcytosine levels, and total DNA methylation—which is especially concentrated on the DNA regulating glutamate and aspartate metabolism, as an evolutionary control mechanism. The only strong correlation that will ever be found examining DNA methylation will probably be dietary glutamate/aspartate ratios and vitamin B₁₂—necessary for transforming glutamate into β-methylaspartate: the precursor for 5-methylcytosine, the presence of which determines DNA methylation as it's most often defined.

 

[Koveras]

"Women with higher (vs. lower) RBC folate had higher mean DNA methylation (5.12 vs. 4.99%) in the pre-fortification period, but lower (4.95 vs. 5.16%) DNA methylation in the post-fortification period."

Decrease maybe partly due to the fact that folic acid can actually reduce intracellular levels of active folate

Total folate and unmetabolized folic acid in the breast milk of a cross-section of Canadian women. - PubMed - NCBI
Background: Folate requirements increase during pregnancy and lactation. It is recommended that women who could become pregnant, are pregnant, or are lactating consume a folic acid (FA)-containing supplement.Objectives: We sought to determine breast-milk total folate and unmetabolized folic acid (UMFA) contents and their relation with FA-supplement use and doses in a cohort of Canadian mothers who were enrolled in the MIREC (Maternal-Infant Research on Environmental Chemicals) study.
Design: Breast-milk tetrahydrofolate (THF), 5-methyl-THF, 5-formyl-THF, 5,10-methenyl-THF, and UMFA were measured with the use of liquid chromatography-tandem mass spectrometry (n = 561). Total daily supplemental FA intake was based on self-reported FA-supplement use.
Results: UMFA was detectable in the milk of 96.1% of the women. Total daily FA intake from supplements was associated with breast folate concentration and species. Breast-milk total folate was 18% higher (P < 0.001) in supplement users (n = 401) than in nonusers (n = 160), a difference driven by women consuming >400 μg FA/d (P ≤ 0.004). 5-Methyl-THF was 19% lower (P < 0.001) and UMFA was 126% higher (P < 0.001) in supplement users than in nonusers. Women who consumed >400 μg FA/d had proportionally lower 5-methyl-THF and higher UMFA than did women who consumed ≤400 μg FA/d.
Conclusions: FA-supplement use was associated with modestly higher breast-milk total folate. Detectable breast-milk UMFA was nearly ubiquitous, including in women who did not consume an FA supplement. Breast-milk UMFA was proportionally higher than 5-methyl-THF in women who consumed >400 μg FA/d, thereby suggesting that higher doses exceed the physiologic capacity to metabolize FA and result in the preferential uptake of FA in breast milk. Therefore, FA-supplement doses >400 μg may not be warranted, especially in populations for whom FA fortification is mandatory.

 

[Travis]

Folate is pretty amazing; it reminds me of thiamine in the way it works. I'll have to take a closer look at the enzyme serine hydroxymethyltransferase.

The enzyme which is most similar to cytosine-5-methyltransferase could very well be tryptophan-2-methyltransferase. This enzyme is a B₁₂-dependent enzyme, like most any other carbon:carbon bond-forming enzymes. The other ones that can do this rely on an iron cluster:

"Radical SAM enzymes use SAM and a four-iron, four-sulfur cluster to catalyze complex chemical transformations." ―Kylie D. Allen​

Kylie Allen was investigating an enzyme he called Fos3, or fosfomycin synthase (step 3). This enzyme carries out a methyl addition, but has 'binding sequences' suggestive for both vitamin B₁₂ and an iron cluster. I don't think he even considered that this reaction could be done without either vitamin B₁₂ or an iron cluster.

"Radical SAM enzymes use this motif together with a molecule of SAM to bind and generate a reduced cluster that donates an electron to reductively, homolytically cleave SAM." ―Kylie D. Allen​

This article here, on tryptophan-2-methyltransferase, shows nicely the sort of techniques needed to prove a methyl-transfer reaction.

Benjdia, Alhosna, et al. "The thiostrepton A tryptophan methyltransferase TsrM catalyses a cobalamin-dependent methyl transfer reaction." Nature communications 6 (2015).​

Nothing of this sort has been done on cytosine-5-methyltransferase, perhaps because it would take weeks to get enough reactant to actually measure. Cytosine-'5-methyltransferase' has the lowest catalytic rate of any enzyme.

It also appears to be the only enzyme thought to catalyze a carbon:carbon methyl addition, without either vitamin B₁₂ or an iron cluster.

But I found more studies which suggest that β-methylaspartate is the precursor for 5-methylcytosine, and quite by accident. I was reading about methylation and came across another CpG island in the brain called GAD₆₇ which, when hypermethylated, becomes the most common finding in schizophrenic brains. This is the the most common physical and undeniably real finding.

Guidotti, Alessandro. "Decrease in reelin and glutamic acid decarboxylase #67 (GAD67) expression in schizophrenia and bipolar disorder: a postmortem brain study." Archives of general psychiatry (2000)​

This gene encodes an enzyme called glutamic acid decarboxylase (number sixty seven), another CpG island involved in glutamate metabolism. All CpG islands that I've seen so far are directly related to glutamate metabolism. Also a common finding in schizophrenia appears to be low glutamate levels, making the hypermethylation of GAD₆₇ seem like an adaptive mechanism to increase glutamate concentrations.

Goff, Donald C. "The emerging role of glutamate in the pathophysiology and treatment of schizophrenia." American Journal of Psychiatry (2001)​

Wherever glutamate:β-methylaspartate dysfunctions are found, also are found DNA methylation changes; conveniently on CpG island genes which just-so-happen to regulate glutamate:β-methylaspartate metabolism. Yet, everyone simply assumes an impossible enzyme is responsible for all DNA methylation changes, despite the fact that that S-adenosylmethionine seems to do little.

​

 

[Travis]

XV. Cytosine‐5‐Methyltransferase Continues to Dissapoint

"Unmethylated trinucleotide repeats had a very low kcat (1.2 h⁻¹)." ―Sriharsa Pradhan​

At a rate of 1.2 methylations per hour, this protein shouldn't even be considered an enzyme. If it really is an enzyme, then it substrate can hardly be considered DNA. In fact, its best putative substrate—double stranded polyinosine alternating cytosine—has also been found to inhibit it:

"We conclude that poly(dI-dC)·poly(dI-dC) acts both as a substrate and as an inhibitor of the methyl transfer reaction with human DNMT." ―Sriharsa Pradhan
​

A real S‐adenosylmethionine methyltransferase—the enzyme catechol‐O‐methyltransferase—was measured to have a Kcat of 16.4 . . . per minute. This is 984 methylations per hour, 820% times faster that the so‐called 'cytosine‐5‐methyltransferase,' a putative enzyme which doesn't even follow any normal kinetics.

"Overall, these data indicate that the methyl transfer reaction with poly(dI-dC)·poly(dI-dC) by human DNMT1 does not follow simple kinetics." ―Sriharsa Pradhan​

If the low Kcat value doesn't raise eyebrows, then the putative reaction mechanics certainly will:

"During this process, the cytosine is flipped 180° out of the DNA backbone into an active site pocket of the enzyme." ―Sriharsa Pradhan
​

Sriharsa Pradham didn't, of course, rigorously investigate the location of the DNA methyl‐transfer events. He had simply measured the total radioactivity emanating from the tritiated methyl groups on the entire DNA strand. Due to the impossible nature of the carbon‐5 methylation, these tritiated methyl groups were most likely added to cytosine's nitrogen #4 and guanine's oxygen #6.

"The filters were dried; 3 ml of Opti-fluor (Packard) was added to each, and tritium incorporation was measured." ―Sriharsa Pradhan
​

This putative enzyme is also non‐selective, contradicting prior claims of specificity.

"Hence we conclude from these results that human DNMT1 also methylates non-CG sequences." ―Sriharsa Pradhan
​

These are the usual kcat values; all other measurements of similar magnitude. The 'cytosine‐5‐methyltransferases' from all species have been measured to have similar snail‐paced rates—a value which approaches the rate of nonenzymatic methyl transfer, or simply what you'd find in the absence of the putative 'enzyme.' The terminal ⋯C–S–CH₃ bond of S‐adnenosylmethionine has a bond energy of only 247 kilojoules per mole, lower than that of the O‐methylguinine bond that has been found to exist in CpG repeats.

You'd highly expect this reaction to proceed nonenzymatically, and at rates similar in magnitude gained through experiment and ascribed to the putative methyltransferase.

​

Dean, J. A. "Properties of atoms, radicals, and bonds." Lange’s handbook of chemistry (1999)
Pradhan, Sriharsa, et al. "Recombinant human DNA (cytosine-5) methyltransferase I. Expression, purification, and comparison of de novo and maintenance methylation." Journal of Biological Chemistry (1999)
Bonifácio, Maria João, et al. "Kinetics of rat brain and liver solubilized membrane-bound catechol-O-methyltransferase." Archives of biochemistry and biophysics (2000)

​

XVI. If it's not a methyltransferase, then . . . what is it?

This protein is most likely simply a transcription factor—discovered before transcription factors were known to exist: A protein which seemed suitable at the time to explain the presence of 5‐methycytosine in DNA . ..in a fit of over‐optimism confirmed by lack‐of‐rigour. It's affinity for CpG repeats suggests its function; one most likely of suppressing transcription of said repeats—especially at the high-density areas named CpG islands. All such islands investigated so far encode for enzymes or receptors involved in glutamate metabolism, the very thing responsible for creating the hypermethylated CpG islands themselves—through its transformation to β-methylasparate, then to 5-methylorotate, and finally to 5-methylcytosine.

 

[Amazoniac]

Vitamin B: new research - Charlyn M. Elliot
ESPN-13: 978-1-60692-697-0
"Firstly, patients suffering from any of the four neurological disorders just described [Parkinson's, encephalitis lethargica, multiple sclerosis, and amyotrophic lateral sclerosis] should display evidence of excessive oxidative stress. There is a significant literature to support this reality [25-26]. Secondly, high doses of natural methyl acceptors should slow the development of these neurological disorders. Thirdly, in untreated patients, one might expect serious deficiencies of natural methyl acceptors, such as thiamine (vitamin B-1), riboflavin (vitamin B-2), niacin (vitamin B-3) and ubiquinone (coenzyme Q10)."

Have you considered posting kitten videos here? Perhaps they would be more engaging..

 

[Travis]

 

[Regina]

 

[Such_Saturation]

Travisord with the planet destroying superlaser thread @Amazoniac

 

[Travis]

This could be one of the biggest unicorns yet. Essentially all DNA methylation studies are underpinned by the assumption that cytosine‐5‐methyltransferase is catalyzing these difficult-to-create carbon:carbon bonds—and through snail‐like kinetic rates! . . . and absurd Jackie-Chan‐like chemical mechanisms‼

That idea that 5‐methylcytosine is made from β-methylaspartate isn't totally-new, there was some talk of is being actual source over 50 years ago, but there have been essentially zero metabolic studies examining this: I think there should be.

Whether this is a real effect or simply the product of a coffee‐induced schizophrenia remains to be determined; but I strongly suspect the former.

This could have implications for schizophrenia itself, since the most consistent postmortem finding is the hypermethylation of the genes called reelin (don' think about Steely Dan) and GAD₆₇: An acronym for glutamic acid decarboxylase. This is another example of a CpG island involved in glutamate metabolism—an observation which I had just stumbled‐upon without looking for it. Moreover, glutamate imbalance is a consistent enough finding in schizophrenia to have had theories centered around it:

Kim, J. S.. "Low cerebrospinal fluid glutamate in schizophrenic patients and a new hypothesis on schizophrenia." Neuroscience letters (1980)
Olney, John W. "Glutamate receptor dysfunction and schizophrenia." Archives of general psychiatry (1995)
Coyle, Joseph T. "Glutamate and schizophrenia: beyond the dopamine hypothesis." Cellular and molecular neurobiology (2006)
Guidotti, Alessandro "Decrease in reelin and glutamic acid decarboxylase (GAD₆₇) expression in schizophrenia and bipolar disorder: a postmortem brain study." Archives of general psychiatry (2000)​

And even a Wikipedia page: Glutamate hypothesis of schizophrenia

Which under my scheme would, tentatively, be translated into something like: Glutamate is funneled through the β-methylaspartate (precursor of 5-methylcytosine) pathway leading to hypermethylated DNA on all glutamate genes (CpG islands) including the one encoding GAD₆₇—an enzyme that turns glutamate into GABA, a process which could be seen as a previously-unrecognized negative-feedback mechanism attempting to regulate a neurotransmitter by sensing a metabolic product of same—a control function set inside the DNA nucleotides themselves, as a carbon:carbon bond to cytosine #5.

But I haven't really looked at this particular example in any great detail. That was just an example of how glutamate levels could control its own metabolism—throttle its own destruction—through DNA, but I'm not really sure how glutamate relates to schizophrenia yet.

If that wasn't enough: Fragile X syndrome is characterized by a hypermethylated CpG island, as well as glutamate receptor anomalies.

Wherever you look, even when not even trying, you'll find enzymes and receptors involved in glutamate metabolism encoded within CpG islands. Why? Because it's glutamate itself which causes 'methylated DNA,' through the simple pathway outlined above. So-called CpG islands could be a primitive and fundamental control mechanism for regulating glutamate—not only a protein amino acid, but an important neurotransmitter as well; this fact puts glutamate in the class of the more important amino acids, on the level of tryptophan and tyrosine (and they can't even enter the citric acid cycle like glutamate can).

The CpG islands only represent between 1–2% of the genome; this makes the probability of all six glutamate enzymes/receptors I've checked also landing on CpG islands about 1 : (75)⁶ . . . or 1 : (177,978,515,625).

In words, these odds can be stated as about . . . one to. . one hundred billion. This seems high, I know, but check the math and you'll soon agree. For confirmation that CpG islands really do represent only between 1–2% of total DNA,* then check here:

Antequera, Francisco. "Number of CpG islands and genes in human and mouse." Proceedings of the National Academy of Sciences (1993)
​

*Not to be confused with the percentage of CpG repeats, which is also near 1%. The CpG repeats are the individual 2·bp units; the CpG islands are large clusters of such repeats.

 

[Travis]

XVII. Glutamate cyclase

So I decided to examine other genes relevant in glutamate metabolism. The first that came to mind was glutamate cyclase. Believe it or not, there's a cyclic form of glutamate called pyroglutamate; it is also called 5-oxoproline.

This is most certainly a CpG island. The entire 151,543·bp length is nearly saturated with CpG repeats. The first 1,000·bp consists of 81 CpG repeats, making the percentage 16.2%; the first 1,000·bp on the 5'‐end has CpG repeats at a prevalence 16.2× times normal.

XVII. Glutamate cyclase

The next that came to mind was gamma-glutamyl carboxylase, the enzyme which transfers that extra carboxyl group to γ-carboxyglutamate. This is a Ca²⁺-chelating domain, and can be found on blood and bone proteins. This enzyme is vitamin K-dependent.

This is most certainly a CpG island, as the entire length indicates. The first 1,000·bp consists of 85 CpG repeats, making the percentage 17.0%. The first 1,000·bp on the 5'‐end has CpG repeats at a prevalence 17× times normal.

 

[haidut]

There is no indication that the nuclear 'enzyme' called cytosine 5-methyltransferase is responsible for 5-methylcytosine's existence in DNA, and every indication that β-methylaspartate is. Consuming more glutamate than asparate would be expected to increase increase β-methylaspartate levels, 5-methylcytosine levels, and total DNA methylation—which is especially concentrated on the DNA regulating glutamate and aspartate metabolism, as an evolutionary control mechanism. The only strong correlation that will ever be found examining DNA methylation will probably be dietary glutamate/aspartate ratios and vitamin B₁₂—necessary for transforming glutamate into β-methylaspartate: the precursor for 5-methylcytosine, the presence of which determines DNA methylation as it's most often defined.

Thanks for the informative post. So, things that lower glutamate such as biotin, vitamin K, GABA, progesterone, CO2, niacinamide etc should all lead to decreased methylation levels, right? On the flip side, wouldn't lower glutamate/aspartate ratio lead to other issues? Aspartic acid is excitotoxic in high doses and is a NMDA agonist, albeit weaker than glutamate. It is also implicated in hyperammonemia, which incidentally has been shown to lead to toxicity through activation of NMDA.

 

[Travis]

Thanks for the informative post. So, things that lower glutamate such as biotin, vitamin K, GABA, progesterone, CO2, niacinamide etc should all lead to decreased methylation levels, right? On the flip side, wouldn't lower glutamate/aspartate ratio lead to other issues? Aspartic acid is excitotoxic in high doses and is a NMDA agonist, albeit weaker than glutamate. It is also implicated in hyperammonemia, which incidentally has been shown to lead to toxicity through activation of NMDA.

I didn't want to give the impression that DNA 'methylation" was either good or bad; but of it's a DNA control mechanism for the glutamate/asparate ratio like it appears to be, I wouldn't worry much about trying to change it.

This could just one thing to keep in mind when reading studies on DNA 'methylation,' which barely show any strong correlations anyway. The N‐linked methylation could have some relevance, but I'm convinced that the C‐linked 'methylation' occurs simply from the endogenous biosynthesis of cytosine from the precursor methylaspartate—instead of aspartate as usual.

I really think this might be a better way to interpret the findings of some DNA 'methylation' studies, and highlights how some enzymes—like Na⁺/K⁺‐ATPase, which can't do what in vitro what they're assumed to do in vivo could simply be unicorns—real proteins but forced to occupy a role that they don't out of impatience, or in trying to explain everything under the contemporaneus paradigms. Looking back on the history of science, its easy to find many cases of terribly‐wrong explanations: I think this may be one such wrong explanation.

 

[jb116]

Might be of interest: http://learntruehealth.com/methylation/

 

[ecstatichamster]

Niacinamide has been used in schizophrenia in megadoses by complementary medicine.

 

[Amazoniac]

Travisord, do you know what's up with supplemental folate being associated with increased cancer risk? Is it a matter of the amounts that people consume and the wrong form? Would you expect that 400mcg of Metafolin(your R here) is harmful in the long-run? I have difficulty believing so.

Maypond Tear for example mentioned this:
Your Own Health And Fitness: Heart, Brain, Cancer, And Hormones

About 40 years ago, three different doctors diagnosed, who told me that I should have a biopsy on three different things and since I had already been studying what doctors do and know about cancer, I ignored them and increased my thyroid and used a little nutritional addition like extra vitamin A and folic acid, and I put some progesterone and DHEA on the area.

http://perfecthealthdiet.com/2012/03/food-fortification-a-risky-experiment/ (massive folic acid supplementation)
I don't agree with him on the niacin part. I was reading about niacin absorption and it's one of the easiest and fastest nutrients to be absorbed, the absorption is nearly complete. Which means that if it's feeding an infection, it must be mainly intracellular, so to starve them you would need to starve yourself first, and even then it wouldn't work because there would be ways to divert the energy. The success of niacin and vit C in disease tells you how safe they are, even during sickness.
The problem with niacin is that people never suppress their appetite with empty calories. Thiamine was suggested for people that have no appetite on another thread, but niacin is a major one. Its fortification was a.. kind gesture to prevent poor quality food from becoming unappetizing.

Ray mentioned that it takes a while (up to a day) to be able to notice the effects of vit B6 supplementation. If you suspect that you're not getting enough folate, and with that in mind, here's something to try if you want to isolate the variables:
Folate 400 - Vitamins & Minerals - Category - Products
@contaminantstrangulator

 

[Travis]

Travisord, do you know what's up with supplemental folate being associated with increased cancer risk? Is it a matter of the amounts that people consume and the wrong form? Would you expect that 400mcg of Metafolin(your R here) is harmful in the long-run? I have difficulty believing so.

I think the 'natural vs synthetic' folate meme is mostly the product of inaccurate assumptions and borderline‐sensational journalism:

Take Figure 2, for instance:

You can plainly see the population with the lowest quartile folate intake has also detectable 'unmetabolized' folic acid in their blood, the chemical species that many supplement salesmen and bloggers will malign by giving the impression that non‐reduced folate is perhaps carcinogenic—somehow.

The difference if four hydrogens: three N‐linked hydrogenes and one C‐linked hydrogen—reductions which don't seem entirely specific. Tetrahydro‐(H₄)‐folate is known to convert back to dihydro‐(H₂)‐folate within the body during the synthesis of thymidylate. Kris Kresser (or one of his ghost writers) gives the impression that all folate must be electronically reduced in the intestine or liver or it never will; I have seen no real indication of this. What 'unmetabolized folic acid' in the blood really appears to be is a symptom of altered redox status, or lower steady‐state NAD⁺–NADH levels.

The enzymes which reduce folate into dihydrofolate, and then tetrahydrofolate, are of course NADH‐cofactor enzymes. From the above data—and the association of 'unmetabolized [nonreduced] folic acid' with cancer—one is led to the obvious conclusion: Nonreduced folic acid in the bloodstream is simply consequent of low NADH levels—or altered redox status. Since cancer is characterized by lower NADH/NAD⁺ ratios, you would actually expect cancer patients to exhibit lower serum tetrahydrofolate/folate ratios (or higher folate/tetrahydrofolate ratios). Exactly the reason why non‐reduced folate in the bloodstream is considered 'unmetabolized' seems a bit odd: these reductions are very minor chemical reactions and likely occur spontaneously; they can be found in people with low folate intakes; non‐reduced folate has never actually been shown to be harmful. Herbivores consume such massive amounts of folate that they'd certainly be expected to have higher absolute amounts of non‐reduced ('unmetabolized') folate in their bloodstream than humans, yet we certainly don't see higher cancer rates among herbivores; we don't see much cancer in wild animals at all. Cancer seems only to occur, practically, when unnatural foods are eaten or prepared in unnatural ways: heat denatured, fermented, and/or adulterated.

Non‐reduced folic acid is very likely a symptom of cancer—not a 'cause.' I think you could find many more instances of non‐reduced chemical species, in general, prevailing in cancer patients over the respective reduced analogues . . . as well as a lower NADH/NAD⁺ ratio.


Wahba, Albert J. "Direct spectrophotometric evidence for the oxidation of tetrahydrofolate during the enzymatic synthesis of thymidylate." J. biol. Chem (1961)
Bailey, Regan. "Unmetabolized serum folic acid and its relation to folic acid intake from diet and supplements in a nationally representative sample of adults aged≥ 60 y in the United States." The American journal of clinical nutrition (2010)

 

[Amazoniac]

Unless there is a deficiency I would not supplement with it [B12] or folic acid. Both have been implicated in cancer. High B12 levels are seen in patients with severe SIBO.

I read conflicting opinions on this as well because B12 deficiency isn't uncommon because, if it's a conditions that tends to occur when the metabolism is weak, you would expect that B12 absorption becomes impaired. Or maybe the allowance for overgrowth is compensatory as it was mentioned here:
Vitamin B2 (riboflavin) Positively Modulates The Microbiome

But let's say it's always associated with an excess. Is it a reason to provide more of everything else to balance? Vits B9, B6, B2, etc.

__
Travisord, speaking of folate and cooking, you consume raw kale from time to time, right?

Read this, it discusses how cooking inactives the enzyme responsible for activating the goitrogenic compounds, yet bacteria posses the ability to turn them active regardless of cooking. The enzyme is released from mechanical damage as well, but I wonder if your body can get rid of the toxins fast, you're at least guarded from bacterial action since the antimicrobials are still sharp when raw. It can vary.

Bioavailability of Glucosinolates and Their Breakdown Products: Impact of Processing

"It has been shown that the incubation of human feces in presence of pure glucosinolates or cruciferous vegetable juices, in which myrosinase was inactivated by heating, leads to the formation of ITCs (94, 98). These metabolites are also formed in the germ-free colon of rats, following the colonization with human intestinal bacteria, and feeding with a pure glucosinolate (99)."

"The formation of other breakdown products from glucosinolates by intestinal microbiota is very likely, but still poorly documented. The formation of amines from the secondary degradation of ITCs has been demonstrated after the incubation of human feces with glucosinolates (100)."

"Brassica plants shredded finely demonstrated significant decrease in the contents of glucosinolates up to 75% over 6 h. On the other hand, thermal treatment by steam cooking, microwaving, and stir-frying did not induce significant changes in the contents of glucosinolates. However, boiling was more effective in reducing the levels of glucosinolates (approximately by 90%), by leaching into cooking water. The authors concluded that avoiding boiling of vegetables could increase the bioavailability of ITCs."

"In another work, the contents of glucosinolates in chopped raw Brassica vegetables was investigated (124). Results demonstrated that aliphatic glucosinolates were partially breakdown in cabbage, whereas high level of indolyl glucosinolates was observed for chopped cabbage and broccoli stored at room temperature. After 48 h storage, chopped white cabbage showed higher contents (15 times more) in 4-methoxy- and 1-methoxy-3-indolylmethyl glucosinolates. Most of the glucosinolate contents (with the exception of 4-hydroxy- and 4-methoxy-3-indolylmethyl glucosinolates) were significantly reduced after chopping and storage. This reduction is mainly mediated by the action of myrosinase, which hydrolyzes the glucosinolates as reported above."​

 

[Travis]

Read this, it discusses how cooking inactives the enzyme responsible for activating the goitrogenic compounds, yet bacteria posses the ability to turn them active regardless of cooking. The enzyme is released from mechanical damage as well, but I wonder if your body can get rid of the toxins fast, you're at least guarded from bacterial action since the antimicrobials are still sharp when raw. It can vary.

I did read a good deal about this roughly one year ago. Did you catch the term 'indoyl glucosinolate'—a tryptophan–glucose conjugate? Perhaps plants have a tad bit more tryptophan than realized? as these tryptophans wouldn't show‐up in the standard assay—a process of amino acid quantification resulting from analyzing exclusively the protein fraction.

It all comes down to the thiocyanate ion (S=C=N⁻)—which, has a similar affinity for the thyroid as I⁻ has. It has the same charge and has a similar size—believe it or not; iodine atoms are huge—and is thought to occupy the thyroid in a manner which absolutely precludes both the thiocyanate ion and the iodide ion from both existing in the same space simultaneously. Thus: one displaces the other—a phenomenon so highly‐suggested by experiment and epidemiology that you could rightly consider as fact. In line with this mechanism, these goitrogens—thiocyanate‐releasing glucose conjugates found in Brassica species, divers other genera, and elsewhere—are goitrogenic only to the extent that one is iodine‐deficient: The thiocyanate ion has very little goitrogenic potential when consumed by a person with relatively high iodine intake (as in the kelp‐consuming coastal populations of the Orient.) Thus, the iodide ion is the competing ion—and the antidote—for the thiocyanate ion . . . perhaps analogous somewhat to the well‐known competition between the Na⁺:K⁺ and Ca²:Mg²⁺ ionic pairs.

The thiocyanate ion (S=C=N⁻) is not exclusively a product of Brassica species, and can actually be found in the bloodstream of all people. Smokers have a higher concentration this ion than non‐, and it can also be released consequent of cassava root ingestion.

I have kelp tablets with a relatively high concentration of the iodide ion, and take them whenever it occurs to me that I ought to displace a bit of any intrathyroidal thiocyanate with I⁻ ions. I would imagine that the thiocyanate ion is then displaced, filtered by the kidneys, and then becoming something for the fish—or divers other aquatic species—to consider (I hope they have Sci‐Hub so they too can learn how to deal with it).

 

[CarbAppreciator]

I didn't want to give the impression that DNA 'methylation" was either good or bad; but of it's a DNA control mechanism for the glutamate/asparate ratio like it appears to be, I wouldn't worry much about trying to change it.

This could just one thing to keep in mind when reading studies on DNA 'methylation,' which barely show any strong correlations anyway. The N‐linked methylation could have some relevance, but I'm convinced that the C‐linked 'methylation' occurs simply from the endogenous biosynthesis of cytosine from the precursor methylaspartate—instead of aspartate as usual.

I really think this might be a better way to interpret the findings of some DNA 'methylation' studies, and highlights how some enzymes—like Na⁺/K⁺‐ATPase, which can't do what in vitro what they're assumed to do in vivo could simply be unicorns—real proteins but forced to occupy a role that they don't out of impatience, or in trying to explain everything under the contemporaneus paradigms. Looking back on the history of science, its easy to find many cases of terribly‐wrong explanations: I think this may be one such wrong explanation.

Is this a general condemnation of scientific assumptions/dogma or does this change your understanding of methylation in terms of its general biological meaning?

 

[Travis]

Is this a general condemnation of scientific assumptions/dogma or does this change your understanding of methylation in terms of its general biological meaning?

It certainly attributes it a meaning, but doesn't of course change the observations that 'methylation' in general acts to suppress a gene. The main point that I try to make is there is a fundamental difference between C‐linked and N‐linked methylation. While N‐linked methylation could be considered a post‐translational modification, C‐linked 'methylation' is non‐labile and had actually existed the moment the DNA strand was synthesized. The reason I say this must be the case is: The enzyme assumed responsible for this has dismal kinetic rates, impossible reaction mechanisms, and appears to have never actually been demonstrated as such. Any radioactive methyl groups found transferred—at a rate of about one per hour [sic]—through experiment could very well have been the result of a low‐energy nitrogen methylation, a completely realistic event. The enzymatic formation of a carbon–carbon bond usually requires a transition element in the enzyme's catalytic domain, something cytosine‐5‐methyltransferase does not have; some biochemists appear to think they can just wave a magic wand to explain these irregularities, but they don't even have to. The methyl group has its origin before the DNA nucleotide cytosine had even been synthesized, formed the moment glutamate was isomerized to β-methylasparatate. Regular cytosine is synthesized from aspartate, and methylcytosine is synthesized by β-methylasparatate. All than needs to be done to explain the C‐linked methylation is to take the consensus scheme for cytosine biosynthesis and instead run β-methylasparatate through that very scheme. The result of this, of course, is 5-methylcytosine; I'm convinced that this is how it's formed. This provides a more grounded and reasonable explanation than the impossible-sounding feats and low kinetic rates of the putative enzyme cytosine-5-methyltransferase.

Did I mention the fact that S-adenosylmethionine—the very cofactor of cytosine-5-methyltransferase—does not accelerate the rate of DNA methylation like it would be expected to, could this enzyme actually do as supposed?

Accepting this leads to the idea that the ratio of β-methylasparatate/asparate alone is fundamentally responsible, and drives so-called C‐linked 'methylation.' And since β-methylasparatate is isomerized from glutamate, this is also dependent on the β-methylasparatate/glutamate ratio. You could even then algebraically substitute one of these in the other to deduce the suspicion that the glutamate/asparate ratio is responsible.

Confirmation of this idea comes in the observation that genes which control enzymes involved in glutamate and aspartate metabolism are CpG islands—areas of abnormally high C-linked DNA methylation. Accepting all of this leads to the natural conclusion that the glutamate/asparate ratio itself regulates the very genes which control its metabolism through the β-methylasparatate/asparate ratio, the 5-methylcytosine/cytosine ratio, DNA methylation itself, and finally: the suppression of genes in a manner which brings the glutamate/asparate ratio back into equilibrium.

This can certainly be tested: Try to find instances of DNA methylation on genes involved in glutamate or aspartate metabolism. One can also check if certain diseases characterized by glutamate disturbances are correlated with abnormal DNA methylation.