Saliva

What saliva actually is

Saliva is about 99 percent water. Everything interesting is in the remaining one percent: electrolytes including calcium, phosphate, bicarbonate, potassium and fluoride, and several hundred proteins and peptides, among them mucins, amylase, statherin, proline-rich proteins, histatins, cystatins, lysozyme, lactoferrin, peroxidase and secretory immunoglobulin A.¹

It comes from three pairs of major glands plus several hundred minor glands scattered through the lips, cheeks and palate. At rest, the submandibular glands supply roughly 65 percent of whole saliva, the parotids about 20 percent, the sublinguals 7 to 8 percent, and the minor glands under 10 percent.¹ The mix shifts substantially under stimulation, when the parotids, which produce thin serous fluid rich in bicarbonate and amylase, take over. The minor glands contribute a disproportionate share of the mucins that make saliva feel like saliva rather than water.

Flow rates

The sleep figure matters more than it looks. Salivary flow follows a circadian rhythm and falls to near zero overnight, which removes buffering, clearance and antimicrobial supply for seven or eight hours at a stretch. This is the physiological reason that what happens in the mouth before bed carries more weight than what happens at any other time of day.

How saliva remineralizes enamel

Tooth enamel is mostly hydroxyapatite, a calcium phosphate mineral. Whether it dissolves or rebuilds is not a biological decision but a chemical equilibrium, and the direction depends on how the surrounding fluid compares to the mineral's solubility.

Supersaturation is the whole trick

Saliva is supersaturated with respect to hydroxyapatite. It carries more calcium and phosphate in solution than would be stable in equilibrium with tooth mineral. At neutral pH the thermodynamic gradient therefore points toward deposition rather than dissolution, and mineral lost from the enamel surface is continually replaced.^1,3

That state is precarious. A supersaturated solution wants to precipitate, and if salivary calcium phosphate crashed out spontaneously it would coat every surface in the mouth as mineral scale. Two families of salivary proteins prevent that. Statherin, a 43-residue phosphoprotein, is the principal inhibitor of spontaneous calcium phosphate precipitation from saliva, and acidic proline-rich proteins do similar work, binding to nascent crystal faces and stalling growth.^3,4,5 Both were shown in seeded-growth experiments to sharply reduce hydroxyapatite crystal growth rates in supersaturated solution.³

So saliva holds a high mineral load in reserve, chemically ready to deposit, and releases it onto tooth surfaces rather than into the bulk fluid. Statherin also happens to be one of the first proteins to adsorb onto enamel, which places the inhibitor and the reservoir exactly where they need to be.

Critical pH and the Stephan curve

Supersaturation is pH-dependent. As pH falls, the solubility of hydroxyapatite rises, and at some point the fluid stops being supersaturated and becomes undersaturated. That crossover is the critical pH, conventionally about 5.5 for enamel, and higher, roughly 6.2 to 6.7, for the softer dentin and cementum exposed at receded gum lines.⁶ Above critical pH, mineral is being added. Below it, mineral is being lost.

In 1944 Robert Stephan measured plaque pH after a sugar rinse and produced the curve that still carries his name: a sharp drop within two to five minutes as bacteria ferment the sugar to acid, a trough below critical pH, and a slow recovery over roughly 30 to 60 minutes as saliva buffers and clears the acid.⁶ Caries is what happens when the area under the curve below critical pH, summed over months and years, exceeds what remineralization can replace. This is why frequency of sugar exposure is a stronger predictor than total quantity. Six small exposures produce six excursions below critical pH. One large exposure produces one.

The acquired pellicle

Within seconds of a tooth being cleaned, salivary proteins begin adsorbing onto it. The resulting bacteria-free organic film is the acquired enamel pellicle. It approaches saturation somewhere between 30 minutes and two hours after brushing.⁷ The first arrivals are the same phosphoproteins doing the mineral work above: proline-rich proteins, statherin and histatins, followed by mucins, amylase, cystatins, lysozyme and lactoferrin.⁷

The pellicle acts as a selectively permeable diffusion barrier. It slows the arrival of acid at the enamel surface and regulates the outward diffusion of calcium and phosphate, which reduces the effective dissolution rate.⁷ In an in situ erosion experiment, the presence of a pellicle changed the character of acid attack outright: enamel with a pellicle softened rather than dissolved, while bare enamel lost surface material.⁸ That is a qualitative difference in failure mode, not a marginal improvement.

The pellicle is also the surface bacteria colonize. It presents specific binding sites that determine which organisms attach first, which means the same film that protects the tooth chemically also sets the initial conditions for the biofilm that grows on it.

Buffering: why flow rate is the variable that matters

Saliva has three buffer systems. Bicarbonate is dominant in stimulated saliva and does most of the work. Phosphate contributes more at low flow rates, when bicarbonate is scarce. Proteins, and urea via ammonia production, add a smaller residual capacity.

The important property is that bicarbonate concentration is flow-dependent. It rises steeply as flow rate increases, so buffering capacity is not a fixed personal attribute but a function of how fast saliva is moving at that moment.⁹ Resting saliva at 0.3 mL/min has modest buffering. Stimulated saliva at 2 mL/min has several times more, and it arrives faster. In practical terms the mouth's defense against an acid challenge is largely determined by whether anything is stimulating flow at the time.

This also explains why low-flow individuals are at compounding risk. They lose buffering capacity and clearance rate together, and both losses act on the same Stephan curve, deepening the trough and lengthening the recovery.

Clearance: how fast the mouth flushes itself

Buffering neutralizes acid. Clearance removes the substrate that produces it. Colin Dawes modeled the mouth as an incomplete siphon: saliva accumulates to a volume of about 1.1 mL, swallowing removes most but not all of it, and roughly 0.8 mL of residual film remains.^10,11 Anything dissolved in that residue is diluted by incoming saliva rather than removed outright, which makes clearance exponential rather than linear.

At an unstimulated flow rate of 0.3 mL/min, the model gives a clearance half-time of about 2.2 minutes for a substance such as glucose.¹⁰ Halve the flow rate and the half-time roughly doubles, so sugar lingers, acid production continues, and the Stephan trough extends. Dawes identified unstimulated flow rate and the pre- and post-swallowing volumes as the dominant parameters, which is another way of saying that clearance is mostly a plumbing problem.

How saliva controls bacterial growth

Saliva does not sterilize the mouth, and it is not trying to. The oral cavity holds several hundred bacterial species, most of them commensal and some of them useful. What saliva maintains is an ecology: a set of overlapping constraints that keep populations bounded and make it harder for any one organism to dominate. The components work through genuinely different mechanisms, which is why the system is difficult for bacteria to defeat.^12,13

Two features of this system are worth drawing out. First, much of it is anti-adhesive rather than bactericidal. Mucins and secretory IgA mostly prevent attachment and hand the problem to swallowing, which is a lower-cost strategy than killing and leaves the commensal population intact. Second, the components interact. Lysozyme, lactoferrin, peroxidase and secretory IgA appear to function as a coordinated system rather than as independent agents.¹⁴

The nitrate pathway: saliva working for the rest of the body

One salivary function has nothing to do with teeth. The salivary glands actively concentrate nitrate from the bloodstream and secrete it into the mouth, where certain commensal bacteria, chiefly Veillonella, Actinomyces, Haemophilus and Neisseria, reduce it to nitrite. Swallowed nitrite is converted to nitric oxide, a vasodilator. This loop is called the enterosalivary circulation of nitrate, and roughly a quarter of plasma nitrate passes through it.¹⁵

Humans cannot perform the first reduction step. Mammalian tissue lacks an effective nitrate reductase, so the pathway is entirely dependent on oral bacteria. Higher urinary nitrate excretion has been associated with lower blood pressure in cohort data, with systolic pressure about 3.4 mmHg lower and diastolic about 1.9 mmHg lower in the higher-excretion group.¹⁶ Interventions that suppress the oral microbiome, including antiseptic mouthrinse and aggressive tongue cleaning, measurably blunt the nitrate pathway and the blood pressure response to dietary nitrate.^15,17

Treat the magnitude with appropriate caution. These are associations and short interventional studies, not long-term outcome trials, and the effect sizes are small. The conceptual point is the durable one: the oral bacterial population performs a metabolic function the host cannot, and "fewer bacteria" is not automatically a better state.

What happens when saliva fails

The clearest evidence for what saliva does comes from people who do not have enough of it. Xerostomia is the subjective sensation of dry mouth. Hyposalivation is the measured reduction in flow. They overlap but are not the same, and either can occur without the other.²

The three major causes of severe hyposalivation are medication effects, Sjögren's syndrome, and radiotherapy to the head and neck.^2,18 Drug-induced dry mouth is by far the most common. Hundreds of medications list it, with anticholinergics the most direct offenders because they block the cholinergic signaling that drives secretion, and polypharmacy compounds the effect. Radiation-induced xerostomia occurs in 60 to 90 percent of patients treated for head and neck cancer.² Dehydration, mouth breathing, uncontrolled diabetes and anxiety states all reduce flow as well.

The consequences follow directly from the mechanisms above. Losing salivary flow means losing the mineral reservoir, the buffer, the clearance mechanism, the pellicle supply and the antimicrobial protein load, all at once. The clinical picture is rampant caries, often at root surfaces where the critical pH is higher to begin with, oral candidiasis, mucosal soreness and ulceration, altered taste, and difficulty with speech and swallowing.^2,18 Severe hyposalivation is one of the strongest caries risk factors in clinical dentistry, and it operates independently of how carefully someone brushes.

What raises and lowers salivary flow

Flow is driven by autonomic reflexes, primarily mechanical and gustatory. Chewing raises it, which is the mechanism behind the chewing-gum literature and is largely independent of what is being chewed. Sour and bitter tastes raise it strongly. Smell contributes modestly. Flow rises through the day, peaks in mid-afternoon, and falls to near zero during sleep.^1,11

Flow is lowered by anticholinergic medications, and by many antidepressants, antihypertensives and antihistamines. Dehydration, habitual mouth breathing, tobacco, alcohol including the alcohol in mouthrinses, radiotherapy and autoimmune damage to the glands all reduce it as well.^2,18

Age is a common assumption and a weak one. Healthy unmedicated older adults retain most of their salivary function. The strong association between age and dry mouth is largely mediated by medication burden and disease, not by aging of the glands themselves.²

Where this evidence is strong and where it is not

Strong. The physical chemistry of supersaturation, critical pH and the Stephan curve is old, well replicated and not seriously contested. The inhibitory function of statherin and acidic proline-rich proteins is established from direct crystal-growth experiments. The clinical consequences of severe hyposalivation are unambiguous and well documented.

Weaker. Salivary flow rate and buffering capacity are only modest predictors of caries risk at the individual level, and studies relating salivary parameters to caries experience report inconsistent results.⁹ Population-level mechanisms do not translate cleanly into individual prediction. The nitrate and blood pressure literature is associational and short-term. And the numbers throughout this page are reference ranges around wide biological variation, not personal targets.

Not addressed here. Whether any specific intervention meaningfully changes salivary function or caries outcomes is a separate question that this page does not attempt to answer.