Xenonucleation: Oral-Pathogen Outer Membrane Vesicles as Transmissible Matrix-Vesicle Equivalents Bridging Dental Calculus and Coronary Plaque Calcification

Author: S. Thaddeus Connelly, DDS, MD, PhD, FACS · Published 2026-07-15 · Version 1.0.0 · Coins: Xenonucleation · Status: Hypothesis / synthesis

Here's the thought I can't put down. What if the mouth and the coronary artery aren't just two sick neighbors on the same bad street — what if one is quietly mailing packages to the other? A seed to start the crystal, the phosphate to feed it, and a little knife to cut the guard rope on the way in.

Abstract

Your blood is a liquid that quietly wants to turn to stone. It carries calcium and phosphate right at the brink of crystallizing, held back only by a thin crew of bodyguard proteins — pyrophosphate, matrix Gla protein, fetuin-A, and, in the mouth, statherin. Tartar on a tooth and chalk in a coronary artery turn out to be that same chemistry in two disguises: the same final crystal (hydroxyapatite), the same soft scaffold, the same trigger — a bubble about 100 nm across called a matrix vesicle. This piece makes one central bet: that the little bubbles (outer membrane vesicles, OMVs) shed by gum-disease bacteria — chiefly Porphyromonas gingivalis — are portable copies of that host bubble, a phenomenon I'm calling xenonucleation. They carry the whole crystal-starting kit — an enzyme to make phosphate, a surface to grow on, and protein-cutting knives to clear the bodyguards — out of the gum and into the bloodstream, where they talk artery-wall muscle cells into turning bone-like. The bacterium never leaves the mouth; it just ships its talent. A fluoride footnote falls out of this: fluoride has actually been measured in dental tartar, yet almost nobody has chemically measured it in coronary plaque — in arteries, "fluoride" nearly always means the radioactive tracer we inject to photograph mineral that's actively growing. And I'm not alone at this door: separate teams have shown the body's own vesicles seed artery calcium, and a 2026 model independently routes oral-biofilm mineral into heart disease — each holding a piece, none naming the join. The population numbers back the broad link, while the one thing this idea specifically predicts — calcification itself — is the least-tested, and the genetics warn that much of the association may be coincidence. So the falsifiable bet is sharp: the more of this bug a person carries, the more their coronary calcium should be actively growing — brighter on an 18F-NaF scan — even after you subtract ordinary inflammation.

The coined term

Xenonucleation. The seeding of host ectopic mineralization by a transmissible bacterial extracellular-vesicle nucleation program. An OMV acts as a matrix-vesicle equivalent — a cell-free ~100 nm phospholipid nanocarrier exporting phosphatase activity, an anionic phosphate-rich template, calcium-binding cargo, and inhibitor-degrading proteases from its site of origin to a distant host tissue, where it triggers or directly nucleates calcium-phosphate deposition.

That's a working name — I'll swap it in a heartbeat if something better lands. It sits under my biomineralization hypothesis (the RUNX2/ALP thesis), sharpens the bEV Cascade, and hands the OVN Nexus the little engine it's been missing at the mouth end.

1 · The shared substrate: one control system, two sites

Let's slow down and look at the machine both diseases actually share — because it really is one machine. Strip the tartar and the plaque down to their chemistry and you get the same short list every time: the same final crystal (hydroxyapatite), the same soft scaffold for it to grow on, and the same starter — a membrane bubble about 100 nm wide called a matrix vesicle. Two diseases that look nothing alike, running off one shared starter motor.

Now here's the part I love. That matrix vesicle isn't some disease gadget — it's the body's own, on-purpose, rather beautiful tool. It's how bone, growth-plate cartilage, dentin, and the cementum on your tooth roots are supposed to turn to mineral: the bubble stuffs itself with phosphate (using an enzyme that, in the very same motion, shreds the inhibitor that would have stopped it), grabs calcium, and gives the first crystal a place to sit. The trouble was never the tool. The trouble is the tool going off in the wrong room. Vascular calcification is just the bone-building program firing where no bone belongs. And that is exactly why a bacterial bubble could hijack it so easily — it doesn't have to invent mineralization, it just borrows a machine the body already keeps oiled and running.

2 · The structural claim: an OMV is a bacterial matrix vesicle

So look at what the gum bug actually puts in the mail. A P. gingivalis OMV is a membrane bubble, 20–250 nm across, and when you turn out its pockets you find: an enzyme that makes phosphate, a surface crawling with phosphate groups and bacterial DNA (which happens to be a lovely spot for a crystal to take hold), and a set of protein-cutting knives called gingipains. Lay that next to the host's own matrix vesicle and squint — it's the same toolkit. Make the substrate, lay a template, handle the calcium, cut the guards: that's a crystal-starting machine, no matter who built it. The bacterium never has to leave the gum pocket. It just packs its talent into a bubble and ships it.

3 · What the literature actually supports

A good idea deserves a hard poke, so let's be honest about what's actually nailed down versus what I'm still hoping is true. The strong evidence says the OMV pulls the trigger — it talks the host cell into calcifying. Whether the bubble can grow crystal all by itself, in a dish with no cells anywhere near it, nobody has shown yet.

Claim	Evidence	Strength
Pg OMVs drive vascular smooth muscle calcification	OMVs induce VSMC osteogenic differentiation and calcification, concentration-dependent, via ERK1/2–RUNX2 (Yang 2016)	Direct, experimental
Pg couples to the host matrix-vesicle apparatus	Pg infection raises phosphate-induced VSMC calcification via phenotype switch, apoptosis, and host matrix-vesicle release; accelerates aortic calcification ex vivo (Park 2020)	Direct, experimental
OMVs disseminate systemically with functional cargo	Pg OMVs translocate to the liver and deliver active gingipains, altering distant metabolism (Seyama 2020)	Direct, in vivo
OMVs are defined cargo carriers	Proteomically distinct protein/sphingolipid/gingipain cargo; modulate host response (Rocha 2021); Type IX secretion system sorts cargo and drives OMV blebbing (Wang 2025)	Direct, mechanistic
"Organism sparse, toxin florid" at distant sites	OMV/T9SS cargo distributes far beyond bacterial load — the gingipains hypothesis (Nara 2021)	Argued, association
Pg virulence factors calcify vessels & periodontal link	Pg-LPS induces VSMC ALP/RUNX2/BSP and calcium-nodule formation; periodontal scaling proposed to reduce it (Li 2021)	Direct, in vitro

Straight talk: what's proven is the modest version — the bacterial bubble talks the host into it, coaxing artery-wall muscle cells to turn bone-like and to release their own matrix vesicles. The bolder version — the bubble seeding crystal directly — is still an open question. And honestly the coaxing story might be the prettier one: the bacterial vesicle doesn't do the deed itself, it recruits the host's vesicle to do it. One bubble whispering to another. (That's Park 2020, and it's the hinge the whole idea swings on.)

4 · The gingipain move: bringing the enzyme that dissolves the guard

You've probably already asked the sharp question: if blood is so eager to turn to stone, why doesn't it? Because blood is guarded — proteins like fetuin-A and matrix Gla protein are bouncers that grab stray crystal before it can grow up. Which is where the knives earn their keep. Gingipains are all-purpose protein scissors, and the OMV carries them clear across the body (Seyama 2020). A bubble that brings both a place to grow crystal and a tool to cut down the bouncers isn't just a seed — it clears its own way in. It's the bacterial echo of the very trick the body's own enzyme pulls when it chews up pyrophosphate. Fair warning: that specific step — gingipains shredding fetuin-A or MGP — is a prediction I'm making, not a result anyone has shown. It's near the top of the to-do list below.

5 · The fluoride corollary — and a genuine measurement gap

Fluoride is a one-trick genius. Everywhere it goes it does the same thing: it slips into the crystal where a hydroxyl group used to sit and pulls the whole lattice tighter, harder to dissolve. But notice when it acts: late, on a crystal that already exists. It's a bodyguard for finished stone, not a lock on the door where nucleation happens. So fluoride can't stop an OMV from planting a seed. At best it hardens whatever that seed grows into.

Site	Native fluoride measured?	How
Dental calculus	Yes — present but highly variable; up to ~10% w/w in one specimen; rises to 3–28% w/w after topical NaF (Pawlaczyk-Kamieńska 2006; Borysewicz-Lewicka 1989)	Direct chemical / X-ray microanalysis
Coronary / aortic plaque	Essentially not — native fluoride content of plaque apatite is close to unmeasured	Almost exclusively the injected 18F-NaF PET tracer, which marks active microcalcification surface, not endogenous F content (Joshi 2014; Syed 2022; Hsu 2018)

Here's a gap you could drive a truck through. Every time the literature says "fluoride" in an artery, it means the tracer — radioactive fluorine we inject to photograph where mineral is actively growing. Nobody seems to have asked the plain chemist's question: how much fluoride is actually built into the stone of a coronary plaque — and how does that stack up against the tartar in the same person's mouth? Take calculus and plaque from the same donors, measure the fluoride in each, lay the numbers side by side. That experiment is just sitting there, waiting for someone to run it.

6 · Testable predictions that discriminate the hypothesis

An idea is only worth something if it's willing to be wrong. So here are the experiments that could kill this one — or make it.

Abiotic nucleation assay (the decisive one). Isolated oral-pathogen OMVs in cell-free supersaturated CaP medium. Apatite nucleation with no cells present supports direct nucleation; no nucleation places the mechanism in host-program induction.
OMV cargo characterization. Proteomics/lipidomics for phosphatase activity, calcium-binding cargo, anionic template (phosphatidylserine-equivalent, LPS-phosphate), and packaged eDNA.
Inhibitor degradation. Do OMV-borne gingipains cleave fetuin-A / MGP in plasma, lowering systemic anti-calcific tone?
In-human link. Periodontal therapy should lower 18F-NaF vascular uptake (active microcalcification) if OMV flux drives it. The money experiment.
Co-localization. Vascular microcalcifications co-staining with bacterial-vesicle / Pg lipid signatures.
Paired fluoride micro-analysis. Native F in calculus vs. plaque apatite from the same donors (closes the gap in §5).

7 · Independent convergence — the field is arriving at the same door

Something here ought to give you more confidence, not less: I'm not the only one wandering toward this door. Three different groups, coming from three different directions, each got an honest grip on a piece of the same animal and stopped just short of the others. When strangers keep triangulating the same spot on the map, it's usually because there's really something there.

Framework	Nucleating agent	Scope
Canonical matrix-vesicle biology	Host matrix vesicle	Bone / dentin (physiological)
Kapustin & Shanahan 2013	Host extracellular vesicle	Vascular (host-derived)
Hegde et al. 2026 (mineral encapsulation)	Mineral-encapsulated microbe / particle	Oral → vascular
Xenonucleation (this work)	Bacterial OMV as transmissible matrix-vesicle equivalent	Oral calculus ↔ coronary plaque

It's the old parable of the blind men and the elephant — each has a true, honest hold on a trunk or a leg, and nobody has the whole beast. What none of them names, and what xenonucleation puts on the table, is the single idea that ties it together: the bacterial OMV is a working stand-in for the host's own matrix vesicle. That one move makes the seed portable — able to travel from mouth to artery — and it hooks the story back to dental calculus, the one place where everybody already agrees little bubbles start the stone.

8 · Population-level support — strong for the axis, open at the keystone

Now, do the numbers from real people back any of this up? They do — and in a way that's almost too on-the-nose. The population data are rock-solid exactly where the story is vague, and go quiet exactly where the story gets sharp. That's not bad luck. That's a treasure map with an X on the one spot nobody has dug.

Signal	Effect size	Base
Periodontal disease → MACE / CHD / MI / stroke / cardiac death / all-cause mortality	RR 1.24 / 1.20 / 1.14 / 1.26 / 1.42 / 1.31 (all significant)	Meta-analysis, 39 cohorts, 4.39 million (Guo 2023)
Periodontitis → hypertension	OR 1.49 (severe); +4.5 mmHg SBP	Meta-analysis, 40 studies (Muñoz Aguilera 2020)
P. gingivalis seropositivity → incident stroke (15 yr)	OR 1.63 (men IgA), 2.30 (women IgG), 3.31 (never-smoker men)	Nested case-control, n=8,911 (Pussinen 2006)
Tooth loss (dose-response) → CVD mortality	HR up to 3.24 (highest burden)	NHANES cohort, n=8,207 (Wang 2024)

But now I have to hold my own feet to the fire. The one thing my idea specifically predicts — calcification — is the exact thing the population data have tested least, and where they have looked, the answer has mostly been a shrug. The one direct study — periodontitis vs. coronary artery calcium in the ARIC cohort — was null and underpowered (OR 1.78 for CAC ≥ 100, 95% CI 0.65–4.86; Nakib 2004). Periodontal therapy lowers CRP, IL-6, and systolic BP across 21 RCTs, but not lipids or flow-mediated dilation (Meng 2024). And two-sample Mendelian randomization finds no causal periodontitis→coronary-atherosclerosis effect, attributing the observational signal to shared risk factors (Zhou 2021; Li 2024). The American Heart Association's standing position is likewise that causation is unproven.

The discriminating prediction. The lazy version of this whole field — "bad gums cause inflammation, inflammation is bad for hearts" — explains everything and therefore predicts nothing. So let me stake out a claim that can actually be wrong: the more P. gingivalis and its bubbles a person carries, the more their coronary calcium should be actively growing — brighter on an 18F-NaF scan — and it should still hold after you subtract out ordinary inflammation (CRP). That last clause is the whole ballgame. If the signal rides on calcification activity and not on generic inflammation, it's pointing at a seed, not a fever — and a seed leaves fingerprints an ordinary confounder can't wash away, which is exactly why the gene studies keep coming up empty-handed.

9 · Why it matters

So why should anyone care? Because for a hundred years we've said "people with bad gums have more heart attacks" and then mostly shrugged at why. Xenonucleation trades that shrug for a machine you can chase: a physical bubble carrying a physical seed across the body. It also tidies up the fluoride puzzle we started with. Fluoride is a hero in the mouth, where the enemy is the tooth dissolving. But in the artery, where the enemy is young, crumbly mineral cracking a plaque open, fluoride is beside the point: it hardens the stone and does nothing about the bubble that planted it. Which leaves a genuinely strange, hopeful thought to end on — the best place to reach in and lower somebody's risk of a heart attack might not be the heart at all. It might be the mouth.

Honest status & limitations

This is a hypothesis-and-synthesis artifact, not a clinical claim. The dental fluoride data are well established; the vascular xenonucleation mechanism is supported for host-program induction (RUNX2/ERK, host matrix-vesicle release) but not for direct abiotic OMV nucleation, and the gingipain–inhibitor and fluoride-measurement links are stated as predictions. Nothing here constitutes guidance about fluoride and cardiovascular disease. The constituent mechanisms are independently published (see §7); the contribution claimed here is the synthesis — the bacterial OMV as a transmissible matrix-vesicle equivalent, the dental-calculus↔coronary-plaque unification, and the fluoride corollary — not the individual links, which are credited to their authors. The term Xenonucleation is a proposed mint.

References

Based on articles retrieved from PubMed (U.S. National Library of Medicine).

Yang WW, et al. P. gingivalis-derived OMVs promote calcification of vascular smooth muscle cells through ERK1/2-RUNX2. FEBS Open Bio, 2016. 10.1002/2211-5463.12151
Park HJ, et al. Infection of P. gingivalis increases phosphate-induced calcification of vascular smooth muscle cells. Cells, 2020. 10.3390/cells9122694
Seyama M, et al. OMVs of P. gingivalis attenuate insulin sensitivity by delivering gingipains to the liver. BBA Mol Basis Dis, 2020. 10.1016/j.bbadis.2020.165731
Rocha FG, et al. Sphingolipid-containing OMVs limit macrophage immune response to P. gingivalis. Infect Immun, 2021. 10.1128/IAI.00614-20
Nara PL, et al. P. gingivalis OMVs as the major driver of neuropathogenesis in Alzheimer's disease. J Alzheimers Dis, 2021. 10.3233/JAD-210448
Wang S, et al. The lipid A 1-phosphatase LpxE requires a functional type IX secretion system. J Oral Microbiol, 2025. 10.1080/20002297.2025.2600179
Li J, et al. Effect of Pg-LPS on calcification of human umbilical artery smooth muscle cells. Exp Ther Med, 2021. 10.3892/etm.2021.10087
Joshi NV, et al. Will 18F-sodium fluoride PET-CT identify vulnerable coronary plaques? Curr Cardiol Rep, 2014. 10.1007/s11886-014-0521-4
Syed MBJ, et al. 18F-sodium fluoride PET and CT in acute aortic syndrome. JACC Cardiovasc Imaging, 2022. 10.1016/j.jcmg.2022.01.003
Hsu JJ, et al. Effects of teriparatide on morphology of aortic calcification in aged hyperlipidemic mice. Am J Physiol Heart Circ Physiol, 2018. 10.1152/ajpheart.00718.2017
Kapustin AN, Shanahan CM. Role of extracellular vesicles in de novo mineralization. Arterioscler Thromb Vasc Biol, 2013. 10.1161/ATVBAHA.112.300128
Hegde MN, et al. Mineral encapsulation of microorganisms in calcified oral biofilms. Front Dent Med, 2026. 10.3389/fdmed.2026.1849108
Guo X, et al. Periodontal disease and subsequent risk of cardiovascular outcome and all-cause mortality (39 cohorts, 4.39M). PLoS One, 2023. 10.1371/journal.pone.0290545
Muñoz Aguilera E, et al. Periodontitis is associated with hypertension. Cardiovasc Res, 2020. 10.1093/cvr/cvz201
Pussinen PJ, et al. Systemic exposure to P. gingivalis predicts incident stroke. Atherosclerosis, 2006. 10.1016/j.atherosclerosis.2006.06.027
Wang F, et al. Associations of tooth loss with all-cause and cause-specific mortality among US adults with diabetes. J Dent, 2024. 10.1016/j.jdent.2024.105304
Nakib SA, et al. Periodontitis and coronary artery calcification: the ARIC study. J Periodontol, 2004. 10.1902/jop.2004.75.4.505
Meng R, et al. Effect of non-surgical periodontal therapy on risk markers of CVD. BMC Oral Health, 2024. 10.1186/s12903-024-04433-0
Zhou M, et al. Causal association between periodontal diseases and cardiovascular diseases. Genes, 2021. 10.3390/genes13010013
Li Z, et al. A Mendelian randomisation analysis reveals no relationship between periodontitis and coronary atherosclerosis. Int Dent J, 2024. 10.1016/j.identj.2024.01.027
Pawlaczyk-Kamieńska T, et al. Fluoride content in supragingival dental calculus measured with X-ray microanalysis. Ann Acad Med Stetin, 2006. PMID 17469508.
Borysewicz-Lewicka M. Effect of local NaF on calcium, phosphorus and fluorine in young dental calculus. Protet Stomatol, 1989. PMID 2641163.