Vitamin D (cholecalciferol/D3 or ergocalciferol/D2) is a fat-soluble secosteroid hormone with pleiotropic actions extending far beyond classical calcium–phosphate homeostasis. The vitamin D receptor (VDR) is expressed throughout the central nervous system — in neurons, astrocytes, microglia, and oligodendrocytes — and the enzyme CYP27B1 that converts circulating 25-hydroxyvitamin D (25(OH)D) to the active metabolite 1,25-dihydroxyvitamin D (calcitriol) is present in the hippocampus, substantia nigra, hypothalamus, and prefrontal cortex[@eyles2005][@holick2007]. Epidemiological data consistently associate low serum 25(OH)D with elevated risk of Alzheimer's disease (AD), Parkinson's disease (PD), and all-cause dementia, while preclinical studies demonstrate that calcitriol modulates amyloid-β clearance, tau phosphorylation, neuroinflammation, oxidative stress, and neurotrophic factor expression[@llewellyn2010][@evatt2008]. The translational gap between these promising mechanistic data and the mixed results of supplementation trials remains a central challenge. This monograph synthesizes the molecular pharmacology, clinical evidence, and practical dosing considerations for vitamin D in neurodegeneration, with a dedicated section on progressive supranuclear palsy (PSP) and corticobasal syndrome (CBS).
| Dimension | Score (0–10) | Rationale |
|---|---|---|
| Mechanistic Clarity | 8 | VDR–RXR genomic signaling, CYP27B1 brain expression, and downstream transcriptional targets well characterized |
| Clinical Evidence | 5 | Large observational support; RCTs mostly negative or modest in cognitive endpoints; few NDD-specific trials |
| Preclinical Evidence | 7 | Robust rodent data across AD, PD, and tauopathy models; consistent neuroprotection |
| Replication | 6 | Epidemiological findings replicated globally; RCT results inconsistent |
| Effect Size | 4 | Observational OR 1.5–2.4× for deficiency → dementia; RCT cognitive effects small |
| Safety/Tolerability | 9 | Excellent safety at doses ≤4000 IU/day; toxicity only at sustained >10,000 IU/day |
| Biological Plausibility | 8 | VDR ubiquity in brain, regulation of 200+ neuroprotective genes, convergence on AD/PD pathways |
| Actionability | 8 | Inexpensive, widely available, simple monitoring (25(OH)D), clear dosing guidelines |
| Total | 55/80 |
Vitamin D3 (cholecalciferol) is synthesized in the skin from 7-dehydrocholesterol upon UVB exposure (wavelength 290–315 nm) or obtained from dietary sources. Hepatic 25-hydroxylase (CYP2R1) converts it to 25(OH)D, the principal circulating metabolite and clinical biomarker. In the brain, neuronal CYP27B1 performs a second hydroxylation to generate 1,25(OH)₂D₃ (calcitriol), which binds the nuclear vitamin D receptor (VDR)[@eyles2005][@garcion2002]. The ligand-bound VDR heterodimerizes with the retinoid X receptor (RXR) and binds vitamin D response elements (VDREs) in the promoter regions of over 200 genes, initiating transcriptional programs that regulate calcium homeostasis, neurotrophic factor expression, antioxidant defense, and immune modulation[@berridge2015].
CYP24A1 (24-hydroxylase) catalyzes the inactivation of calcitriol to calcitroic acid, providing local feedback control. In AD brain tissue, CYP27B1 expression is reduced while CYP24A1 is upregulated, suggesting impaired local calcitriol synthesis as a disease-associated feature[@landel2016].
Vitamin D exerts neuroprotection through at least six convergent pathways:
1. Neurotrophic Factor Regulation. Calcitriol upregulates nerve growth factor (NGF), glial cell line-derived neurotrophic factor (GDNF), and neurotrophin-3 (NT-3) transcription via VDREs in their promoter regions[@naveilhan1996]. NGF supports cholinergic basal forebrain neurons that degenerate early in AD; GDNF is critical for dopaminergic neuron survival in the substantia nigra, making this pathway directly relevant to PD[@smith2006].
2. Anti-inflammatory and Immunomodulatory Actions. Calcitriol suppresses NF-κB nuclear translocation, reducing transcription of pro-inflammatory cytokines (TNF-α, IL-1β, IL-6) and inducible nitric oxide synthase (iNOS)[@cantorna2015]. In microglia, VDR activation promotes the M2 anti-inflammatory phenotype and enhances expression of the anti-inflammatory cytokine IL-10[@boontanrart2016]. Calcitriol also inhibits NLRP3 inflammasome assembly, a key driver of neuroinflammation in AD and tauopathies[@briones2012].
3. Amyloid-β Clearance. Vitamin D enhances microglial phagocytosis of amyloid-β (Aβ) and stimulates expression of Aβ transport receptors (LRP1, P-glycoprotein) at the blood–brain barrier, facilitating Aβ efflux[@ito2011][@durk2014]. In TgCRND8 and APP/PS1 transgenic mice, calcitriol treatment reduces cortical and hippocampal amyloid plaque burden by 25–50%[@yu2011].
4. Tau Phosphorylation Inhibition. Calcitriol inhibits GSK-3β activity — a primary tau kinase — through Akt/PKB phosphorylation, reducing hyperphosphorylation of tau at AD-relevant epitopes (Thr181, Ser396, Ser404)[@grimm2014]. This mechanism is directly relevant to 4R-tauopathies including PSP and CBS.
5. Antioxidant Defense. VDR activation upregulates γ-glutamyl transpeptidase, glutathione reductase, and superoxide dismutase 2 (SOD2), bolstering mitochondrial and cytosolic antioxidant capacity[@gezenak2011]. Calcitriol also induces thioredoxin reductase 1, which maintains the thioredoxin system critical for redox homeostasis in neurons[@berridge2015].
6. Calcium Homeostasis and Excitotoxicity Protection. Calcitriol regulates expression of voltage-gated calcium channels (L-type Cav1.2) and the calcium-buffering proteins calbindin-D28k and parvalbumin, protecting neurons from excitotoxic calcium overload — a convergent mechanism in AD, PD, and ALS[@brewer2001][@alexianu1998].
In APP/PS1 double-transgenic mice, chronic calcitriol administration (42 days, 2.5 μg/kg) reduced hippocampal Aβ42 levels by 32%, decreased plaque number by 48%, improved Morris water maze performance, and upregulated LRP1 at the blood–brain barrier[@yu2011][@durk2014]. In 5xFAD mice, vitamin D deficiency accelerated amyloid deposition and cognitive decline, while supplementation with cholecalciferol (12,000 IU/kg diet) reversed these effects and restored hippocampal BDNF levels[@kang2022]. Calcitriol also reduced phospho-tau (AT8-positive) staining in 3xTg-AD mice by 35%, correlating with increased phospho-Akt and decreased GSK-3β activity[@grimm2014].
In the MPTP mouse model, pretreatment with calcitriol (1 μg/kg × 7 days) protected 60% of striatal dopaminergic terminals and 45% of substantia nigra TH-positive neurons compared to vehicle, mediated by GDNF upregulation and reduction of microglial activation[@smith2006][@wang2001]. In the 6-OHDA rat model, calcitriol attenuated rotational behavior by 40% and preserved striatal dopamine content by 55%[@lima2018]. Vitamin D-deficient rats show spontaneous loss of nigral TH-positive neurons, reduced striatal dopamine turnover, and impaired rotarod performance — effects reversible with D3 supplementation[@cui2013].
In the rTg4510 tauopathy model, vitamin D supplementation reduced cortical NFT density by 28% and improved nest-building behavior, associated with decreased GSK-3β activity and increased autophagy marker LC3-II[@grimm2014]. These findings are particularly relevant to PSP and CBS, which are characterized by 4R-tau aggregation. In the PS19 (P301S tau) mouse model, calcitriol treatment reduced microglial activation markers (Iba1, CD68) in the brainstem — a region heavily affected in PSP[@briones2012].
A landmark meta-analysis by Llewellyn et al. (2014) pooling data from 37 studies (n = 171,000) found that serum 25(OH)D < 10 ng/mL (severe deficiency) was associated with a 1.54-fold increased risk of all-cause dementia (95% CI 1.24–1.91) and a 2.25-fold increased risk of AD (95% CI 1.26–4.03) compared to levels > 20 ng/mL[@llewellyn2010]. The Rotterdam Study (n = 10,186; mean follow-up 12 years) confirmed a dose–response relationship: each 10 ng/mL decrease in 25(OH)D was associated with 15% higher dementia hazard[@littlejohns2014].
For PD, the Finnish cohort study by Knekt et al. (2010; n = 3,173; 29-year follow-up) demonstrated that individuals in the highest quartile of serum 25(OH)D had a 65% lower risk of PD (HR 0.35, 95% CI 0.14–0.89)[@knekt2010]. Evatt et al. (2008) documented vitamin D insufficiency (< 30 ng/mL) in 55% of PD patients and 41% of AD patients versus 36% of age-matched controls[@evatt2008].
VITAL-Cog (2022). The cognitive substudy of the VITAL trial randomized 25,871 adults to vitamin D3 2000 IU/day or placebo for 5.3 years. No significant difference was observed in the primary cognitive composite (β = −0.01, 95% CI −0.04 to 0.02). However, subgroup analysis showed benefit in participants with baseline 25(OH)D < 20 ng/mL (β = 0.06, p = 0.04)[@kang2021].
DO-HEALTH (2020). In 2,157 adults ≥ 70 years, vitamin D3 2000 IU/day for 3 years did not improve Montreal Cognitive Assessment (MoCA) scores in the overall population (mean change +0.1 vs +0.1, p = 0.94). Falls were reduced by 16% in the vitamin D + omega-3 + exercise arm, a finding highly relevant to PSP[@bischoffferrari2020].
Annweiler et al. (2012). An open-label pilot in 43 AD patients found that vitamin D3 supplementation (800 IU/day × 16 months) was associated with improved Mini-Mental State Examination (MMSE) scores (+1.2 points from baseline vs −2.4 in non-supplemented; p = 0.02), though the non-randomized design limits interpretation[@annweiler2011].
Jia et al. (2019). A 12-month RCT in 210 mild-to-moderate AD patients found that vitamin D3 (800 IU/day) combined with memantine improved ADAS-Cog scores significantly more than memantine alone (−3.6 vs −1.8 points; p = 0.035), with corresponding reductions in serum Aβ42 and phospho-tau181[@jia2019].
Suzuki et al. (2013). In 114 PD patients, high-dose vitamin D3 (1200 IU/day × 12 months) significantly prevented deterioration on the Hoehn and Yahr scale compared to placebo (OR for non-deterioration 1.82, 95% CI 1.08–3.07, p = 0.02), with the FokI TT genotype subgroup showing the strongest benefit[@suzuki2013].
Cheng et al. (2022) Meta-Analysis. Pooling 5 RCTs (n = 750 AD patients), vitamin D supplementation showed a small but significant improvement in MMSE (WMD = 1.18, 95% CI 0.24–2.12, p = 0.01), driven primarily by patients with baseline deficiency[@cheng2022].
The consistent pattern across trials is that vitamin D supplementation provides the most clear cognitive benefit in patients with pre-existing deficiency (25(OH)D < 20 ng/mL), while trials in vitamin D-replete populations show null or minimal effects. This mirrors the pharmacological principle that correcting a deficiency restores normal physiological function, but supraphysiological levels do not necessarily confer additional neuroprotection. For falls prevention — a critical outcome in PSP — the evidence is more consistently positive.
Falls are the most dangerous consequence of PSP, occurring in >80% of patients within 3 years of onset. The mechanisms include postural instability, vertical supranuclear gaze palsy, axial rigidity, and impaired protective reflexes. Vitamin D has a well-established role in falls prevention through multiple mechanisms[@annweiler2010]:
The DO-HEALTH trial showed that vitamin D3 2000 IU/day combined with omega-3 fatty acids and a simple home exercise program reduced falls by 16% in adults ≥ 70 — a clinically meaningful effect for PSP patients where every prevented fall avoids potential hospitalization[@bischoffferrari2020].
PSP and CBS are characterized by aggregation of 4-repeat (4R) tau, driven primarily by GSK-3β and CDK5 kinases. Calcitriol's inhibition of GSK-3β through Akt activation provides a direct anti-tau mechanism relevant to these disorders. While no clinical trials have tested vitamin D specifically in PSP or CBS, the preclinical evidence from tauopathy models (rTg4510, PS19) showing reduced NFT density and brainstem microglial activation is directly applicable to PSP neuropathology[@grimm2014][@briones2012].
Both PSP and CBS show marked neuroinflammation with microglial activation in the basal ganglia, brainstem, and cortex. PET imaging with the TSPO ligand ¹¹C-PK11195 demonstrates elevated microglial activation in PSP patients correlating with disease severity[@gerhard2006]. Vitamin D's anti-inflammatory actions — NF-κB suppression, NLRP3 inhibition, M2 microglial polarization — address this pathological feature directly.
The FokI (rs2228570) polymorphism in the VDR gene affects the length of the VDR protein and its transcriptional activity. The FokI ff genotype has been associated with increased PD risk in several populations[@suzuki2013]. In the Suzuki PD trial, patients with the FokI TT genotype showed the strongest benefit from vitamin D supplementation. VDR polymorphisms have not been specifically studied in PSP or CBS, representing a research gap. The ApaI (rs7975232) and TaqI (rs731236) VDR polymorphisms have been associated with AD risk in meta-analyses[@gezenak2007].
Given the convergence of falls prevention evidence, tau phosphorylation inhibition, anti-neuroinflammatory effects, and the high prevalence of deficiency in elderly neurological patients, vitamin D supplementation should be considered standard adjunctive care in PSP and CBS:
Cholecalciferol (D3) is the preferred form for supplementation. D3 raises serum 25(OH)D more effectively than ergocalciferol (D2): a meta-analysis of 7 RCTs found D3 was 87% more potent in raising 25(OH)D and maintained elevated levels 2–3× longer than D2[@tripkovic2012]. D3 also has a higher affinity for vitamin D binding protein (DBP), prolonging its circulating half-life.
| Population | Dose | Target 25(OH)D | Notes |
|---|---|---|---|
| Adults < 65 y, prevention | 1000–2000 IU/day | > 30 ng/mL | General neuroprotection |
| Adults ≥ 65 y, prevention | 2000–4000 IU/day | 40–60 ng/mL | Falls + fracture prevention |
| Deficient (< 20 ng/mL) | 50,000 IU/week × 8 wk, then 2000–4000/day | 40–60 ng/mL | Loading protocol |
| Severely deficient (< 10 ng/mL) | 50,000 IU × 2/week × 6 wk, then maintenance | 40–60 ng/mL | Aggressive repletion |
| PSP/CBS patients | 2000–4000 IU/day + Ca 500–1000 mg | 40–60 ng/mL | Falls + bone + neuroprotection |
25(OH)D has a serum half-life of approximately 15 days, allowing for flexible dosing intervals. Daily dosing produces the most stable serum levels, but weekly or monthly bolus dosing (equivalent total dose) achieves similar mean 25(OH)D levels[@tripkovic2012]. However, large intermittent boluses (e.g., 500,000 IU annually) have been associated with paradoxically increased fall risk in elderly women, possibly due to acute hypercalcemia or rapid changes in muscle calcium handling[@sanders2010]. Daily or weekly dosing is therefore preferred over annual mega-doses, particularly in fall-prone PSP patients.
Vitamin D3 at doses ≤ 4000 IU/day has an excellent safety profile. The Institute of Medicine set the tolerable upper intake level (UL) at 4000 IU/day for adults, while the Endocrine Society considers up to 10,000 IU/day safe for short-term repletion[@ross2011]. True vitamin D toxicity (hypercalcemia with 25(OH)D > 150 ng/mL) is rare and typically occurs only with sustained intake > 10,000 IU/day for months.
Adverse effects at toxic levels:
| Drug | Interaction | Management |
|---|---|---|
| Thiazide diuretics | ↑ Calcium reabsorption → hypercalcemia risk | Monitor serum calcium |
| Corticosteroids | ↓ Vitamin D absorption and ↑ catabolism | May need higher D3 doses |
| Anticonvulsants (phenytoin, carbamazepine) | ↑ CYP24A1 → accelerated calcitriol degradation | Monitor 25(OH)D; increase dose |
| Cholestyramine | ↓ Vitamin D absorption | Separate dosing by 4+ hours |
| Statins | Possible synergistic neuroprotection | No dose adjustment needed |
| Lithium | Both affect calcium homeostasis | Monitor calcium if co-administered |
| Digoxin | Hypercalcemia potentiates digoxin toxicity | Monitor calcium and digoxin levels |
Vitamin D synergizes with several other interventions relevant to neurodegeneration: