THE INQUIRY: What if the bacterial colony quietly establishing itself in the gums of a woman in her forties — the one her dentist mentions in passing as "early periodontal disease, nothing to worry about yet" — is not, in fact, contained to the gums, and is at this moment travelling pathways that will determine the architecture of her brain at seventy?
THE SYNTHESIS
For most of the twentieth century, Alzheimer's was understood as a disease that began in the brain and ended in the brain — a slow architectural collapse of neurons, traceable through the protein debris that accumulated in the tissue, but never quite explained at its origin. The gum line was the gum line. The brain was the brain. They were not, in the medical imagination, in particular conversation with each other.
A different answer has been gathering, slowly, in the peer-reviewed literature. It runs through the mouth.
In 2019, a research group led by Dominy and colleagues published in Science Advances the result of a tissue analysis on postmortem human brains. Brains from people who had died with Alzheimer's, compared to brains from people who had not. They were looking for a specific bacterial signature: gingipains, the toxic proteases produced by Porphyromonas gingivalis, the keystone pathogen in chronic periodontal disease. Gingipains were detected in 96% of Alzheimer's brain samples for one variant, 91% for the other, with gingipain load correlating with both tau and ubiquitin pathology — the principal architectural signatures of Alzheimer's neurodegeneration. The same gingipains were also found in some neurologically healthy control brains, at lower frequency, suggesting that the relationship is gradient rather than absolute. But the signal was unmistakable, and the implication — that an oral bacterium may not, in fact, stay in the mouth — became suddenly difficult to ignore.
The mechanism has since been mapped with increasing precision. P. gingivalis can invade neurons directly, persist intracellularly, produce active gingipains over time, drive tau phosphorylation, disrupt cellular architecture, and cause synapse loss — the precise neuropathological hallmarks of Alzheimer's. Outer membrane vesicles released by the bacterium contain gingipains and have been shown to breach the blood-brain barrier by degrading the tight junction proteins that normally hold it intact. The bacterium does not need to enter the brain whole; it sends its proteases ahead.
What this body of work proposes, taken together, is a quiet revision in how the most consequential neurodegenerative disease of midlife may begin. Not in the brain. In the gums. Decades before any cognitive symptom is noticed, in the precise years women are told their bleeding gums are a hygiene oversight rather than a clinical signal.
The translation for women specifically is worth naming. The Mosconi work covered in Vol. 001 of this Archive established that the female brain undergoes a distinct biological transition in midlife — beta-amyloid begins to deposit twenty years before clinical presentation, brain glucose metabolism declines, the regions essential to memory begin to thin. None of these women have noticed anything yet. All of them are, by clinical standards, fine. The Dominy work, layered onto the Mosconi work, suggests one of the routes by which the depositing begins: chronic, low-grade, asymptomatic colonization of the gum tissue, sending its inflammatory signal across the blood-brain barrier for decades before anyone thinks to look. The dentist appointment, in this frame, is not a hygiene appointment. It is the earliest practitioner most women will see who is actively reading the brain at sixty-five.
THE CONSIDERED RESPONSE
The research community has not yet declared P. gingivalis a cause of Alzheimer's. The honest framing is that the evidence is associational, mechanistic in animal models, and increasingly difficult to dismiss as coincidence — but causation in humans has not been formally established. A clinical trial of the gingipain-inhibitor drug atuzaginstat showed promise in subgroups but encountered safety signals and was halted. The pharmaceutical answer has not yet arrived.
The clinical answer that has arrived, for any reader inclined to act on this, is older and quieter than a drug. The thing that gives gingipains access to the brain is the same thing that gives them access to the bloodstream: chronic, untreated periodontal disease. The receding gum line, the bleeding when the gums are brushed, the pocketing that develops in midlife and goes unaddressed because dentistry, in most women's lives, has been treated as cosmetic rather than systemic.
A reader inclined to take this research seriously has, available to her, a set of interventions that are evidence-supported, low-cost, and wholly within her control. They are not novel. They are not biotech. They are the unglamorous daily and quarterly work of treating the mouth as continuous with the brain — because, it turns out, it is.
LE PROTOCOLE: Turning the Research into Intelligence
The Dominy body of work, taken alongside the broader literature on inflammation and neurodegeneration, organises around three disciplines: the periodontal, the inflammatory, and the systemic. We read them as a single architecture.
The Periodontal Floor: A periodontist appointment, treated as serious medicine rather than a six-month hygiene reminder. Pocket depth, attachment loss, and bleeding on probing are quantitative measures of an active inflammatory process at the gum line. They can be tracked, year over year. They can be improved. Daily flossing — the kind that actually happens, not the kind that is intended — is among the most consequential preventive interventions for which the evidence base is genuinely strong.
The Inflammatory Signal: Periodontal disease is, in the longevity frame, a chronic low-grade infection generating systemic inflammatory load. That load shows up in blood biomarkers — high-sensitivity C-reactive protein, IL-6, fibrinogen — long before any cognitive symptom registers. Reading these markers serially, year over year, is one of the more direct ways of monitoring the trajectory the Dominy work describes. Inflammation is the bridge between the gum line and the hippocampus.
The AION Atelier Baseline: The inflammatory, metabolic, and cardiovascular biomarkers that map onto this trajectory — read together, interpreted against women's reference ranges, in the context of where a reader sits relative to menopause — are readable now. The AION Atelier Baseline reads them not as isolated numbers but as a single architecture, which is the context that makes the data actionable rather than abstract.
— The Archive Editors AION Atelier
To receive The Archive, twice weekly — subscribe here.
We do not provide medical advice. We provide the intelligence to ask better questions.
THE SOURCE: Dominy et al., Science Advances (2019) — landmark study identifying gingipains in 91–96% of postmortem Alzheimer's brains, with neuropathology correlation; Haditsch et al., Journal of Alzheimer's Disease (2020) — neuronal infection model demonstrating intraneuronal P. gingivalis and tau dysregulation; subsequent literature on outer membrane vesicles and blood-brain barrier permeability. The atuzaginstat clinical program (Cortexyme/Quince Therapeutics) is on regulatory hold; the foundational research continues to be cited and replicated.
The Archive — a publication of AION Atelier. Longevity, with intention.