Should evidence of an autolysosomal de-acidification defect in Alzheimer and Parkinson diseases call for caution in prescribing chronic PPI and DMARD?

ABSTRACT

Nearly fifty million older people suffer from neurodegenerative diseases, including Alzheimer (AD) and Parkinson (PD) disease, a global burden expected to triple by 2050. Such an imminent “neurological pandemic” urges the identification of environmental risk factors that are hopefully avoided to fight the disease. In 2022, strong evidence in mouse models incriminated defective lysosomal acidification and impairment of the autophagy pathway as modifiable risk factors for dementia. To date, the most prescribed lysosomotropic drugs are proton pump inhibitors (PPIs), chloroquine (CQ), and the related hydroxychloroquine (HCQ), which belong to the group of disease-modifying antirheumatic drugs (DMARDs). This commentary aims to open the discussion on the possible mechanisms connecting the long-term prescribing of these drugs to the elderly and the incidence of neurodegenerative diseases.

Abbreviations: AD: Alzheimer disease; APP-βCTF: amyloid beta precursor protein-C-terminal fragment; BACE1: beta-secretase 1; BBB: brain blood barrier; CHX: Ca²⁺/H⁺ exchanger; CMI: cognitive mild impairment; CQ: chloroquine; DMARD: disease-modifying antirheumatic drugs; GBA1: glucosylceramidase beta 1; HCQ: hydroxychloroquine; HPLC: high-performance liquid chromatography; LAMP: lysosomal associated membrane protein; MAPK/JNK: mitogen-activated protein kinase; MAPT: microtubule associated protein tau; MCOLN1/TRPML1: mucolipin TRP cation channel 1; NFE2L2/NRF2: NFE2 like bZIP transcription factor 2; NRBF2: nuclear receptor binding factor 2; PANTHOS: poisonous flower; PD: Parkinson disease; PIK3C3: phosphatIdylinositol 3-kinase catalytic subunit type 3; PPI: proton pump inhibitor; PSEN1: presenilin 1, RUBCN: rubicon autophagy regulator; RUBCNL: rubicon like autophagy enhancer; SQSTM1: sequestosome 1; TMEM175: transmembrane protein 175; TPCN2: two pore segment channel 2; VATPase: vacuolar-type H⁺-translocating ATPase; VPS13C: vacuolar protein sorting ortholog 13 homolog C; VPS35: VPS35 retromer complex component; WDFY3: WD repeat and FYVE domain containing 3; ZFYVE1: zinc finger FYVE-type containing 1.

KEYWORDS:

• Autophagy
• disease-modifying antirheumatic drugs
• lysosome
• neurodegenerative disease
• proton pump inhibitor

During the last few decades, we have witnessed tremendous advances in understanding the roles of autophagy in human health and diseases. One Nobel prize and 70,000 publications later, however, the translation of these benefits into therapeutic promises has been much slower than expected. Facing this challenge, the activities of all approved drugs are being reexamined. As first candidates, the drugs that inhibit lysosomal activity and, by inference, the autophagy pathway, were repurposed as a “miracle” cure to boost the effectiveness of cancer therapy, our immune defenses, and our longevity.

So far, the most successful examples of lysosomotropic drugs that have been used clinically are the proton pump inhibitors (PPIs, such as esomeprazole®, omeprazole®, pantoprazole®, and rabeprazole®) [1]. Since they were clinically introduced in the 1980s, they have been among the top 10 most commonly used medications worldwide. Of the potential indications, PPIs are prescribed for various acid-related disorders: they are effective for managing gastroesophageal reflux disease, ulcer disease, and indigestion. Likewise, they are also used in combination with antibiotics for eradicating Helicobacter pylori, a bacterium that may infect as much as 50% of the world’s population. PPIs are also often co-prescribed for life-long treatment with antiplatelet therapy, such as aspirin, to prevent gastrointestinal bleeding following a stroke, myocardial infarction, or stenting. Considering their broad array of indications, 15% to 40% of patients will use these medications during their lifetime, driving the value of the PPI market to $3 billion in 2020 (https://www.mordorintelligence.com/industry-reports/proton-pump-inhibitors-market).

The other current efforts in the clinic to inhibit autophagy are focused on inhibiting lysosomes using chloroquine (CQ) or the related hydroxychloroquine (HCQ). Far from the recent media madness around COVID-19, chloroquine began as an antimalarial drug in the late 1940s until the parasite grew resistant. Both CQ and HCQ have immunomodulatory and anti-inflammatory effects that are dependent [2] and independent [3–7] of their lysosomotropic actions. As such, they aid in treating autoimmune diseases, including systemic lupus erythematosus, inflammatory bowel disease, and rheumatoid arthritis. Accordingly, CQ and HCQ are part of the group of disease-modifying antirheumatic drugs (DMARDs) [8–10].

FROM BENEFITS TO RISKS. As often happens from this wide use, 50% of PPI prescriptions have no appropriate indication [11]. The treatment of gastroesophageal reflux disease with PPIs, which requires only 4–8 weeks, is well tolerated, safe, and effective. But their chronic use is frequent, without benefit, and is associated with side effects [12,13]. The cost of this inappropriate chronic use of PPIs is alarming, particularly for the elderly. Up to 40% of patients aged 70 years or older receive PPI for at least one year [14–17]. A growing concern is that PPI use may be a risk factor for PD, dementia, and AD [18–22]. Along this line, HCQ is inexpensive and remarkably well-tolerated by most patients. However, an association between dementia and HCQ has also been described [8]. Over 1 billion people, nearly one-sixth of the world’s population, suffer from neurological disorders costing $1 trillion per year. Cost and deaths are expected to double by 2050, driving a neurological pandemic [23,24]. Given this global emergency, the causal relationship between PPI or DMARDs overuse and neurological diseases must be evaluated.

POSSIBLE MECHANISMS. When administered systemically, PPIs are lipophilic weak bases, so they can enter the bloodstream and cross the blood-brain barrier (BBB), as detected in rat brains by HPLC [25–27]. After one single intravenous injection (10 mg/kg), the highest brain-to-plasma ratio of omeprazole was only 0.15, a level likely insufficient to affect cognitive function after acute dose. Yet, recent data suggest that chronic use of PPIs may disrupt BBB tight junctions [28] and increase their brain uptake, exerting adverse effects on memory [29,30]. Similarly, oral CQ and HCQ, which are entirely absorbed, can cross the BBB, reaching concentrations 10–20 times higher than plasma concentrations [31]. However, PPIs and HCQ act differently. PPIs are indeed small-molecule drugs that inhibit the V-ATPase but are not specific to gastric cells. Conversely, HCQ is a lysosomotropic amine that accumulates and increases intralysosomal pH. So how can PPI and DMARD treatments be associated with PD, dementia, and AD?

PD and de-acidification of the lysosome: An established link

Over the past two decades, compelling molecular, clinical, and genetic studies highlighted the central role of the lysosomal pathway in the pathogenesis of neurodegenerative diseases [32,33]. While the accumulating proteins are distinct in each disease, autophagy is the sole process able to clear large protein aggregates [32,33]. For Parkinson disease, a particular form of autophagy called chaperone-mediated autophagy emerges as the main SNCA/ degradative pathway. In line with this finding, the genetic basis for PD converges on autophagy-lysosome dysfunction [34,35]. To name a few, PD risk loci include lysosomal trafficking components (VPS35, VPS13C, RAB29/RAB7L1, and RAB39), lysosomal membrane proteins (TMEM175 and LAMP3), the proton pump ATP6V0A1, which acidifies lysosome, and lysosomal acidic cathepsins (CTSB and CTSD), which degrade SNCA/α-synuclein [36]. It is noteworthy that the leading cause of PD, i.e., the mutations in GBA1 (glucosylceramidase beta 1), result in intralysosomal alkalization [37]. Not surprisingly, many environmental neurotoxic agents such as methamphetamine, rotenone, and MPP⁺ promote PD-like symptoms by impairing lysosomal acidification and activity [38–40]. Altogether, these findings strengthen the relevance of lysosomal genetic risk in PD etiology.

AD and de-acidification of the lysosome: An emerging factor in the pathogenesis

In line with this scenario, a link between AD and an autophagy-lysosomal defect has also been suggested [29,30]. The protein aggregates, i.e., the MAPT/tau tangles and extracellular amyloid-β (Aβ) plaques that contribute to neuronal cell death in AD, are mainly degraded by the autophagy-lysosomal pathway. Along these lines, several autophagy-deficient mouse models recapitulate the formation of MAPT/tau aggregates (deletion of Sqstm1, Nfe2l2/Nrf2, or Lamp2) and Aβ plaques (Atg5, Nrbf2, or Rubcn) [32]. Several risk variants that affect all steps of the autophagy pathway, from autophagosome formation, maturation, and substrate sequestration, have been associated with AD (PIK3C3, RUBCNL, ZFYVE1, ATG10, and WDFY3) [33]. Likewise, the AD patient brains exhibit an expansion of autophagic compartments and accumulation of the autophagy substrate SQSTM1, suggestive of impaired autophagic flux [41,42].

In this landscape, recent compelling evidence links V-ATPase deficiency to the pathogenesis of early-onset AD. Two studies published in April and June 2022 in Science Advances and Nature Neurosciences [43,44] challenge our fundamental understanding of how AD develops. The prevailing model highlights the extracellular accumulation of Aβ plaques as the first step toward AD brain damage. Instead, the new findings conducted in mice by the Nixon team identify the root of brain damage inside the neuron within lysosomes. Deacidifying degradative organelles, as seen after Psen1 loss of function, drives lysosome mistrafficking before extracellular amyloid plaque formation [20]. Reacidifying lysosomes with acidic nanoparticles reverses autophagy dysfunction [45]. Notably, in Psen1* mice brains, similar to AD patients [20], a unique flower-like clustering of autophagy vesicles, called PANTHOS (“poisonous anthos”), accumulates fibrillar Aβ intralumenally. Lysosomal membrane permeabilization and cell death are the principal source of senile plaques [46] (Figure 1). Therefore, these data suggest that the deregulation of lysosomal acidification may be a shared underlying mechanism in AD and PD pathogenesis.

Figure 1. Revisiting the amyloid cascade hypothesis around lysosomal de-acidification. This schematic representation presents the new suggested sequence of events initiated by lysosomal deacidification leading to amyloid deposition. (1) Lysosomal de-acidification. Unlike PD, there are no mutations in the AD risk genes directly involved in lysosomal function. But PSEN1 (presenilin 1), a part of γ-secretase that cleaves APP (amyloid beta precursor protein), is critical for folding the ATP6V0A1 subunit of the vacuolar-type ATPase (V-ATPase) [47]. Loss-of-function PSEN1 mutants (PSEN1*), the leading cause of familial AD, thus impair V-ATPase function and lysosome acidification. (A) Independently of ATP6V0A1, PSEN1* can increase the pH of lysosomes by mobilizing the Ca²⁺ channel TPCN2 and the putative Ca²⁺/H⁺ exchanger CHX [48]. In support of this hypothesis, five AD mouse models demonstrate impaired lysosomal dysfunction [49–51]. (B) Likewise, the APP-derived fragment APP-βCTF (C terminal fragment) that accumulates in the brain of sporadic AD patients, binds V-ATPase and interferes with its assembly and activity [46,52]. As a result, the de-acidification of lysosomes markedly impairs autophagy and endosomal degradation [47,53]. (2) Lysosomal calcium efflux. Lysosomes are the second-largest intracellular stores of calcium. Their Ca²⁺ levels control lysosomal biogenesis, fusion, and exocytosis [43]. Two cation channels, MCOLN1/TRPML1 and TPCN2, reside on late endosome membranes and provide a pore for lysosomal calcium efflux. With AD development, the high lysosomal pH opens the pH-sensitive MCOLN1/TRPML1 channel [45], and the PSEN1* mutants directly activate TPCN2. (3) Impaired retrograde transport. The subsequent rise in the cytosolic Ca²⁺ activates the MAPK/JNK kinases that phosphorylate and reduce dynein-driven transport of vesicles along microtubules from axons to the cell body [44]. Additionally, the expression of the R406W-mutated MAPT was also reported to block dynein-DCTN (dynactin)-mediated axonal vesicle transfer [54]. Without access to the lysosomes, the degradative flux is blocked: the autophagic vesicles, late endosomes, and amphisomes accumulate and fail to clear their cargo. (4) PANTHOS formation. Deacidified immature amphisomes are then the reservoir where amyloid-β (Aβ) peptide is processed and forms plaques. Two processing enzymes, BACE1 and PSEN2, which cleave APP into Aβ, accumulate in the amphisomes, generating increased amounts of Aβ [55]. The limited proteolysis of APP into Aβ accelerates its self-aggregation. The enlarged and defective AVs containing APP and its harmful fragments (APP-βCTF and Aβ peptide) accumulate gradually around the nuclei of neurons, pushing out the cell membrane into blebs and forming unique flower-like rosettes, called PANTHOS (for “poisonous flower”). (5) Secretion of Aβ plaques. In an emergency, disrupting AV maturation robustly upregulates unconventional secretory autophagy to release the toxic aggregates [56]. However, the sustained injury of large amphisomes by the aggregates leads to membrane rupture and subsequent lysosomal cell death. Nixon and colleagues proposed that the PANTHOS neurons are the primary source of Aβ plaques that become extracellular upon cell death [53]

Prerequisites for the adoption of an autophagy moonshot approach for alzheimer and parkinson DISEASES.

While speculative, these findings suggest that chronic use of lysosomotropic drugs, such as HCQ or PPIs, may increase the risk of AD and PD. In support, a study published in 2016 in JAMA Neurology (73,679 patients aged over 75 years from 2004 to 2011) documented an increased risk of dementia in people who regularly received PPIs (omeprazole®, pantoprazole®, lansoprazole®, esomeprazole® or rabeprazole®) [22], in agreement with other published reports [8,21,57–59]. Along this line, several large-scale clinical and case studies have reported increased incidence of PD [34,58,60,61] and early cognitive mild impairment (CMI, such as migraine, anxiety, and visual memory impairments) in PPI users [62–73] (see Table S1). Following this concern, gastroenterology organizations, prescribers, and patients now urge limited PPI use by publishing new guidelines for de-prescribing PPIs [74]. Nonetheless, these disease associations remain intriguing in light of millions of people treated with no side effects and the observational studies that have not linked dementia, AD, or cognitive decline to PPI use in people older than 65 years [10,19,38,75–81].

It is also controversial whether HCQ use is associated with an increased risk of dementia. While no effect of HCQ on the progression of dementia in early Alzheimer disease has been reported [9,10], another case-control study suggested that patients taking HCQ for up to 305 days have an increased risk of dementia compared to unexposed patients [8]. We consistently observed key AD behavioral disturbances, such as depression, anxiety, psychosis, agitation, and aggression in CQ-treated mice (40 mg/kg/day, two weeks of treatment; unpublished results). Coupled with previous evidence (atg5, atg7 knockout mice) [82–84], these new results and our observations raise the critical notion that a defective autophagy pathway could be sufficient to cause AD.

These inconsistencies may stem from the considerable variability in the study design: i) Not all studies included a group control. ii) Nor considered the age and the gender of the enrolled patients iii) Most studies that failed to demonstrate a PPI-AD relationship have inadequate treatment duration and lag time, assessing AD/dementia symptoms only days and months after PPI treatment. iv) Even though AD is remarkably heterogeneous, there was no standardized method for diagnostics among the studies. v) The dementia subtypes were not accurately specified. Such a small number of eligible clinical studies thus makes it impossible to draw firm clinical conclusions on the association between PPI prescription and the incidence of AD/dementia. These significant flows also diminish the credibility of the lack of a PPI-AD relationship (Table S1).

To clarify this controversy, future experimental studies are urgently needed to unequivocally confirm the causal relationship between lysosomotropic treatment and AD or PD. Among the possible causes that can explain this controversy, we can raise the suggestion that the etiologies of AD and PD are complex and involve both genetic and environmental factors. One of the major challenges in tracking the etiology of neurological diseases is the lack of proper modeling systems. While the current AD and PD mice models have provided valuable insights into pathogenesis, they display early and aggressive disease. Thus, the question arises whether defective lysosomal acidification is observed in sporadic late-onset diseases. In this regard, the observation that the APP-derived fragment APP-βCTF that accumulates in the brains of sporadic AD patients similarly inhibits V-ATPase activity [46,52] is of particular interest. In the future, it will be essential to determine whether the presence of polymorphisms of the autophagic machinery [33] would make a population at risk more sensitive to chronic PPI and HCQ treatment. Because there is no cure for AD and considering the worldwide use of PPIs, and to a lesser extent HCQ, we face a major public health problem. Until this predisposed at-risk population is identified, we caution against the long-term prescription of lysosomotropic drugs such as PPIs and HCQ in elderly patients.

Acknowledgements

The authors would like to thank Dr. L Van Elslande and the geriatric department of Nice CHU hospital.

Disclosure statement

No potential conflict of interest was reported by the authors.

Supplementary material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/15548627.2023.2214960

Correction Statement

This article has been corrected with minor changes. These changes do not impact the academic content of the article.

Additional information

Funding

The research was funded by the French Government (Agence Nationale de Recherche, ANR) through the 'Investments for the Future' LABEX SIGNALIFE [ANR-11-LABX-0028-01] and ANR JCJC [ANR-21-CE14-0008-01]; Association pour la Recherche sur le Cancer, ARC [ARCPGA2022010004426]; INSERM; CNRS; IHU RespirERA; CANC'AIR Genexposomic project; Canceropole PACA; Région Sud; INCA Plan Cancer; ITMO Cancer; INSERM cancer; NIH (DJK:NIGMS GM131919); and Roche France.