Wang Z, Baker TR, Malin AJ - "Mechanisms of neurotoxicity of fluoride or aluminum: implications for neurodegenerative disease risk" Curr Res Toxicol 11:100313 (2026) https://doi.org/10.1016/j.crtox.2026.100313
ABSTRACT
Fluoride and aluminum are two naturally abundant elements with widespread industrial uses. Fluoride is also added to community water supplies as a public health intervention for dental cavity prevention. However, findings from animal studies show potential links of fluoride and aluminum exposure with neurodegenerative disease risk, particularly at high exposure levels. This review uniquely examines neurochemical and neurobiological impacts of fluoride and aluminum exposure as well as whether these processes may increase the risk of common and rare neurodegenerative diseases, including dementia, Parkinson's Disease, and motor neuron disease.
Fluoride and aluminum can cross the blood-brain barrier and accumulate in neural tissue, where they can interact to produce neurotoxic effects. Chronic exposure to fluoride and aluminum can cause oxidative stress, mitochondrial dysfunction, brain inflammation, and disruption of essential ions. These effects can contribute to impaired nerve signaling, cell damage, and protein aggregation - key factors in neurodegeneration. Co-exposure to aluminum-fluoride complexes may worsen these effects by increasing amyloid buildup and causing nerve cell death, although more research on aluminum-fluoride interactions is needed.
Additionally, many animal studies include relatively high fluoride or aluminum exposure levels, and epidemiological human data are scarce, particularly for less common neurodegenerative diseases. Moreover, these studies often rely on ecological or occupational exposure measures rather than individual biomarkers. Findings of this narrative review underscore the need for methodologically rigorous longitudinal human studies on fluoride, aluminum and neurodegenerative disease risk, particularly given the mechanistic basis for these potential associations.
This review by Zhilin Wang, Tracie Baker, and Ashley Malin is neither comprehensive nor technically adequate. The authors report only a rudimentary search of PubMed and Google Scholar using a narrow set of general terms. They provide no search dates, complete search strings, eligibility criteria, screening procedure, study-selection record, quality assessment, or risk-of-bias evaluation, and explicitly acknowledge that no formal review protocol was applied. Nevertheless, they repeatedly characterize the paper as a "comprehensive narrative review". Without a reproducible method, the reader cannot determine how studies were identified, why particular papers were selected, why others were excluded, or whether the resulting evidence base reflects the available literature rather than the authors' arbitrary choices.
The search strategy appears grossly inadequate for the subject under review. The reported terms do not include "aluminum fluoride", "aluminofluoride", "fluoroaluminate", "AlFx", "cryolite", "sodium aluminum fluoride", "mixture", "co-exposure", "interaction", "G protein", "signal transduction", "pesticide", or "insecticide". This omission is particularly indefensible because research on aluminum-fluoride compounds, including the use and toxicity of cryolite pesticides, extends back to the 1920s. The review therefore approaches a long-established field as though fluoride-aluminum interaction were a recently recognized and sparsely studied possibility. A compilation of the relevant literature contains dozens of experimental, mechanistic, toxicokinetic, developmental, neurological, and signalling studies, yet only a small fraction is represented in the review. Omitted work includes studies of hippocampal pathology, spatial learning and memory, neurotransmitters, developmental exposure, brain accumulation, apoptosis, BDNF-TrkB signalling, microRNAs, MAPK pathways, neuronal cytoskeletal disruption, aluminum absorption, and dose-ratio-dependent interactions.
The most serious deficiency in the review's mechanistic analysis is its near-total omission of G-protein signalling. The authors repeatedly invoke aluminum-fluoride complexes (AlFx) and even present combined fluoride and aluminum exposure as the distinctive feature of the review, yet they do not explain that AlFx is widely used experimentally both as a broad, receptor-independent activator of heterotrimeric G proteins (Nobel Prize 1994). By mimicking the gamma phosphate of GTP and stabilizing GDP-bound Galpha subunits in an activated conformation, AlFx bypasses receptor activation and directly engages G-protein-dependent signalling. The human TSH receptor - like AlFx - can also engage all four heterotrimeric G-protein families and therefore functions as a natural broad-spectrum G-protein activator. AlFx can perturb these same G-protein families, potentially mimicking, distorting, or bypassing components of TSHR signalling wherever functional TSH receptors are expressed. The review nevertheless omits G proteins, adenylyl cyclase, phospholipase C, calcium signalling, protein kinase C, Rho and Rac regulation, tyrosine phosphorylation, and related MAPK pathways. The available literature specifically describes AlFx effects on Gs, Gi/o, and Gq/11-linked signalling, calcium sensitization, ERK signalling, FAK and paxillin phosphorylation, and other upstream signalling events. By excluding this literature, the review bypasses the most obvious candidate molecular initiating mechanism of combined fluoride and aluminum exposure and substitutes a generic list of downstream outcomes such as oxidative stress, mitochondrial dysfunction, inflammation, apoptosis, and protein aggregation.
The treatment of fluoride-aluminum interaction is also scientifically superficial. The evidence does not support a single assumption that co-exposure simply "worsens" toxicity. Different studies report synergistic, additive, antagonistic, biphasic, and endpoint-specific interactions, with the direction and magnitude depending on the fluoride-to-aluminum ratio, chemical species, concentration, exposure route, developmental period, tissue, and measured outcome. Aluminum may enhance some fluoride effects, fluoride may enhance aluminum absorption or toxicity, and complex formation may reduce the freely available concentration of one or both ions. Some experiments report greater neuronal injury under combined exposure, whereas others report reduced fluoride accumulation, attenuated fluoride toxicity, or antagonistic effects on cell viability and embryonic development. A competent review should have organized this evidence by speciation, dose ratio, internal dosimetry, biological endpoint, and direction of interaction. Instead, the authors use vague language about "shared and interacting biological pathways" without adequately examining the conditions that determine whether the interaction is synergistic, antagonistic, or absent.
The review provides no consistent framework for comparing exposures across studies. It moves between fluoride concentrations in drinking water, administered doses of sodium fluoride, exposures to aluminum chloride or aluminum sulfate, measured aluminum concentrations in biological samples, and aluminum-fluoride complexes without clearly distinguishing external exposure, administered compound, internal dose, and chemical species. These quantities cannot be interpreted on the same basis without compound-to-element conversion, route-specific toxicokinetics, absorption, tissue distribution, and chemical speciation. This lack of distinction undermines comparisons of dose, biological relevance, and fluoride-aluminum interactions across the included studies.
There are also numerous factual and internal inconsistencies. Varner et al. is identified as a rat study in the table but described as a mouse study in the narrative. Rodella et al. is reported as using 2.5% aluminum sulfate in the table but 500 mg/L in the text. Capriello et al. is listed as using 11 mg/L in the table but 3.6 mg/L aluminum chloride in the narrative, without explaining whether one value represents elemental aluminum and the other the compound. Salib and Hillier is dated 2018 in the table but 1996 in the text and reference list. Table 1 describes "untreated" APP/PS1 mice as impaired after receiving fluoride treatment. The Taylor systematic review is duplicated as separate 2025a and 2025b references. These errors indicate that study details were not consistently checked against the original publications or even against other sections of the review.
The citation practices raise additional concerns. A paper on fluoride-induced oxidative stress in oral bacteria is cited as evidence concerning oxidative stress in neural tissue, while a study of aluminum toxicity in rice seedlings is used to support statements about neural oxidative stress. General reviews of oxidative stress, apoptosis, and inflammation are repeatedly used to bridge gaps between exposure-specific observations and neurodegenerative disease. Such citations may support broad biological background, but they do not demonstrate that the proposed sequence occurs in neurons at relevant fluoride or aluminum exposures. This creates the appearance of mechanistic depth without establishing exposure-specific causation.
The manuscript is also exceptionally poorly edited. Errors include "biproduct," "represents and import gap," "other study have found," "electrophosphoic backbone," "fenton transformations," and the statement that damaged mitochondria are "unable to slow down ROS." Other sentences contain fragments, incorrect plurals, misplaced modifiers, inconsistent capitalization, inappropriate terminology, and confused descriptions of biological processes. These are not harmless stylistic defects. In a mechanistic review, imprecise language can alter the scientific meaning, and the density of basic grammatical and technical mistakes reinforces the impression that neither the manuscript nor its tables and references underwent adequate expert review.
Overall, this paper is a poorly searched, selectively assembled, internally inconsistent, and inadequately edited account of a much larger literature. It omits major experimental and mechanistic domains, including the central G-protein literature, fails to address the extensive and historically established research on aluminum-fluoride compounds and cryolite, does not distinguish among chemical species or interaction patterns, and repeatedly replaces mechanistic analysis with generic downstream toxicological terminology. Its claim to provide a comprehensive review is therefore untenable. The paper would require a complete literature search, reconstruction of the evidence tables, verification of every exposure and study description, systematic treatment of synergistic and antagonistic findings, incorporation of G-protein signalling and historical aluminum-fluoride research, and thorough scientific and grammatical editing before it could be considered a dependable review.