Weak base drug-induced endolysosome iron dyshomeostasis controls the generation of reactive oxygen species, mitochondrial depolarization, and cytotoxicity

Abstract Objectives Approximately 75 % of marketed drugs have the physicochemical property of being weak bases. Weak-base drugs with relatively high pKa values enter acidic organelles including endosomes and lysosomes (endolysosomes), reside in and de-acidify endolysosomes, and induce cytotoxicity. Divalent cations within endolysosomes, including iron, are released upon endolysosome de-acidification. Endolysosomes are “master regulators of iron homeostasis”, and neurodegeneration is linked to ferrous iron (Fe2+)-induced reactive oxygen species (ROS) generation via Fenton chemistry. Because endolysosome de-acidification-induced lysosome-stress responses release endolysosome Fe2+, it was crucial to determine the mechanisms by which a functionally and structurally diverse group of weak base drugs including atropine, azithromycin, fluoxetine, metoprolol, and tamoxifen influence endolysosomes and cause cell death. Methods Using U87MG astrocytoma and SH-SY5Y neuroblastoma cells, we conducted concentration-response relationships for 5 weak-base drugs to determine EC50 values. From these curves, we chose pharmacologically and therapeutically relevant concentrations to determine if weak-base drugs induced lysosome-stress responses by de-acidifying endolysosomes, releasing endolysosome Fe2+ in sufficient levels to increase cytosolic and mitochondria Fe2+ and ROS levels and cell death. Results Atropine (anticholinergic), azithromycin (antibiotic), fluoxetine (antidepressant), metoprolol (beta-adrenergic), and tamoxifen (anti-estrogen) at pharmacologically and therapeutically relevant concentrations (1) de-acidified endolysosomes, (2) decreased Fe2+ levels in endolysosomes, (3) increased Fe2+ and ROS levels in cytosol and mitochondria, (4) induced mitochondrial membrane potential depolarization, and (5) increased cell death; effects prevented by the endocytosed iron-chelator deferoxamine. Conclusions Weak-base pharmaceuticals induce lysosome-stress responses that may affect their safety profiles; a better understanding of weak-base drugs on Fe2+ interorganellar signaling may improve pharmacotherapeutics.


Introduction
Most pharmaceutical drugs have the physicochemical property of being weak bases with relatively high pK a values [1].Weak base pharmaceuticals have favorable pharmacokinetic properties for absorption, distribution, metabolism, excretion and toxicity (ADMET) [2,3].However, weak base drugs can also enter acidic organelles including endosomes and lysosomes (endolysosomes) wherein they cause lysosome stress responses including de-acidification [4,5].Although pH-driven ion trapping of weak base drugs in endolysosomes is known [6,7], the extent to which and mechanisms by which weak base drugs affect inter-organellar signaling and induce cytotoxicity remain unclear.
Endolysosomes contain high concentrations of readilyreleasable ferrous iron (Fe 2+ ) and are central to iron trafficking and redox signaling [8,9].Drug-induced cytotoxicity and the pathogenesis of diverse disease states are linked to increased iron levels and iron dysregulation [10,11]; Fe 2+ via Fenton-like chemistry generates reactive oxygen species (ROS) and induces cytotoxicity [12].Furthermore, endolysosome Fe 2+ release induced by endolysosome deacidification causes iron dysregulation [13][14][15].We reported recently in numerous cell types that de-acidification of endolysosomes decreased Fe 2+ levels in endolysosomes, increased Fe 2+ and ROS levels in the cytosol and mitochondria, as well as induced cell death [14][15][16].Because endolysosome sequestration of weak base drugs can result in endolysosome de-acidification [17], we tested the hypothesis that ion trapping of weak base drugs in endolysosomes might be a mechanism common to changes in interorganellar signaling and increases in cytotoxicity caused by weak base drugs.
Here, using U87MG astrocytoma and SH-SY5Y neuroblastoma cells, we demonstrated that pharmacologically and therapeutically relevant doses of atropine, azithromycin, fluoxetine, metoprolol, and tamoxifen de-acidified endolysosomes, decreased Fe 2+ levels in endolysosomes, increased Fe 2+ and ROS levels in the cytosol and mitochondria, induced mitochondrial membrane potential depolarization, and increased cell death; effects all prevented by the endocytosed iron-chelator deferoxamine.Thus, the weak base drug-induced increases in Fe 2+ and ROS levels in the cytosol and mitochondria as well as cell death appear downstream of Fe 2+ released from endolysosomes induced by endolysosome de-acidification.Increased understanding of weak base drug-induced alterations of Fe 2+ signaling between organelles may lead to improved pharmacotherapeutics and novel insights into adverse effects of commonly used drugs and disease pathogenesis.

Materials and methods
The experiments performed in this manuscript were not preregistered.No blinding procedures were performed.No statistical method was employed to pre-determine the sample size of the experiments.No randomization methods were used.

Cell cultures
Human neuroblastoma cells (SH-SY5Y) and astrocytoma cells (U87MG) were cultured in Dulbecco's Modified Eagle Medium (Invitrogen, cat.no.11995, Carlsbad, USA) containing 10 % fetal bovine serum (Ther-moFisher, Waltham, USA) and 1 % penicillin/streptomycin (Invitrogen, cat.no.15140122, Carlsbad, USA).SH-SY5Y cells and U87MG cells were grown in T75 flasks and sub-cultured as needed to 70-80 % confluency.Cells were passaged every 3-4 days using 0.025 % trypsin (Invitrogen, Carlsbad, USA) and maintained in an incubator set at 37 • C and 5 % CO 2 .Cells were not used past their 10th passage.The cells used in this manuscript are not listed as a commonly misidentified cell line by the International Cell Line Authentication Committee and no authentication experiments were conducted.

Endolysosome pH
Endolysosome pH was determined using LysoSensor Yellow/Blue DND-160 (ThermoFisher, cat.no.L7545, Waltham, USA), a ratiometric dualexcitation dye.SH-SY5Y cells were grown in 35 mm 2 dishes overnight and then incubated for 5 min at 37 • C with media containing 10 μM DND-160.After incubation, the media was replaced with new media and then images were taken.During imaging, various drug treatments were added gently by pipetting.Light emission at a wavelength of 520 nm in response to excitation for 2 ms at 340/380 nm was measured every 30 s using a filter-based microscope imaging system (Zeiss, Germany).Using a standard curve, we measured the pH as previously described [34] and we calculated changes in proton concentrations within endolysosomes using the formula pH=−log[H + ].

Endolysosome iron
Labile ferrous iron (Fe 2+ ) in endolysosomes was detected using the fluorescence probe FeRhoNox-1 (Goryo Chemicals, cat no.GC901, Darmstadt, Germany).We have previously demonstrated that the FeRhoNox-1 dye does not significantly colocalize in Golgi, but rather significantly colocalizes in endolysosomes with a high Pearson's correlation coefficient and also that FeRhoNox-1 is a specific marker for Fe 2+ [9].SH-SY5Y cells were added to 35 mm 2 dishes at a density of 7 × 10 4 cells/dish and incubated overnight prior to being taken for experimentation.
Cells were incubated with 10 μM FeRhoNox-1 at 37 • C for 1 h, stained with LysoTracker™ Green DND-26 (Invitrogen, cat.no.L7526, Carlsbad, USA) and 1 μg/ml Hoechst 33342 for 10 min, washed three times with warm 1X PBS, and then imaged at an excitation of 532 nm and an emission of 570 nm using our confocal scanning microscope (Zeiss, LSM800).Imaging settings remained the same throughout each set of experiments; images were acquired just prior to (0 h) and 30 min after treatments were applied.Imaris software version 9.7 (Oxford Instruments, Concord, USA) was used to reconstruct the images and data were represented as mean fluorescence intensity (MFI).Endolysosome iron levels were also measured in dissociated SH-SY5Y and U87MG cells using flow cytometry.After treatment and staining, cells were washed three times with warm 1X PBS, dissociated from each well in a 24-well plate by gentle trituration, and then placed in Eppendorf tubes for flow cytometry measurements.FeRhoNox-1 fluorescence measurements were made at an excitation of 532 nm and an emission of 570 nm using our flow cytometer (Attune NxT, ThermoFisher, Waltham, USA).

Cytosolic iron
Cytosolic Fe 2+ levels were determined with Phen Green™ SK, diacetate (PGSK) (Invitrogen, cat.no.P14313, Carlsbad, USA).Although PGSK reacts with a variety of metals including Fe 2+ (Invitrogen), our use of the endolysosome iron-chelator deferoxamine (DFO) minimizes this concern because DFO binds with a high affinity to ferric iron and not those other metals.Therefore, although it is likely that other cations are released from endolysosomes, it appears clear that weak base drug-induced release of iron from endolysosomes quenches PGSK fluorescence.In addition, we have been working (and publishing) with deferoxamine (DFO) for years.We have used this drug in multiple cell lines and primary cultured cells including rat, human, and mouse neurons [9,16,35].We have tested multiple concentrations of DFO and used the chosen concentration of DFO that was effective yet didn't kill the cells.SH-SY5Y cells and U87MG cells cultured in 24-well plates were first stained with 10 μM PGSK for 20 min in complete media.
After 20 min, cells were washed two times with PBS and fresh media was added to the wells.Then cells were incubated with weak base drugs for 1 h.Deferoxamine (50 μM, DFO) was added to the cells 1 h before the addition of weak base drugs.After 1 h incubation, cells were then washed three times with warm 1X PBS to remove unloaded dye, were dissociated from their culture plates by gentle trituration, and placed in Eppendorf tubes where the fluorescence was measured at an excitation of 488 nm and an emission of 530 nm using a flow cytometer (Attune NxT, ThermoFisher, Waltham, USA).Unstained cells were used as controls for background fluorescence.Mean fluorescence intensity (MFI) data were collected from a minimum of 10,000 cells per condition.Because PGSK is a quenching dye, the fluorescence data were transformed and illustrated as the reciprocal of mean fluorescence intensity (1/MFI) to demonstrate the increase in cytosolic iron.

Cytosolic ROS
Levels of cytosolic ROS were measured using approximately 6 × 10 5 SH-SY5Y cells/well following preincubation with weak base drugs in the absence or presence of DFO (50 μM) for 1 h followed by 1 h incubations with various treatments.Following incubations, cells were washed three-times with PBS and 10 μM CM-H 2 DCFDA (5-(and-6)-chloromethyl-2 ′ ,7 ′ -dichlorodihydrofluorescein diacetate) (Invitrogen, cat.no.C6827, Carlsbad, USA), which is an effective dye for hydroxyl radicals that are a product of the Fe 2+ + H 2 O 2 reactants in Fenton's reaction [36] was added into wells containing PBS.Cells were then incubated for 20 min at 37 • C and 5 % CO 2 ; cells were then washed three times with PBS to remove extracellular stain.Cells were then dissociated from their culture plates by gentle trituration, and placed in Eppendorf tubes where the fluorescence was measured at an excitation of 488 nm and an emission of 530 nm using a flow cytometer (Attune NxT, ThermoFisher, Waltham, USA).Unstained cells were used as negative controls to establish background fluorescence.Mean fluorescence intensity (MFI) was acquired on a minimum of 10,000 cells per condition using an Attune NxT flow cytometer (ThermoFisher, Waltham, USA).

Mitochondrial iron
Mitochondrial Fe 2+ levels were measured using rhodamine B 4-[(2,2 ′bipyridin-4-yl) aminocarbonyl] benzyl ester (RDA) (Guidechem, cat.no.952228-30-7, Milwaukee, USA).SH-SY5Y cells and U87MG cells were added at a density of about 6 × 10 5 cells/well to 24-well plates and incubated at 37 • C and 5 % CO 2 overnight prior to preincubation with weak base drugs for 1 h in the absence and presence of DFO (50 μM) for 1 h.Following treatments, cells were washed three times with warm 1X PBS and incubated with 100 nM RDA for 20 min in PBS in a 37 • C and 5 % CO 2 incubator .Cells were washed three times with PBS to remove extracellular dye.Cells were then dissociated from their culture plates by gentle trituration and then placed in Eppendorf tubes where the fluorescence was measured using a flow cytometer (Attune NxT, ThermoFisher, Waltham, USA).A minimum of 10,000 cells were collected and RDA fluorescence was analyzed at an excitation of 562 nm and an emission of 598 nm.Unstained cells were used to establish background fluorescence and stained cells (without any treatment) were used as buffer controls.Mean fluorescence intensity (MFI) was determined by flow cytometry (Attune NxT, ThermoFisher, Waltham, USA).Because RDA is a quenching dye, the fluorescence data were transformed and illustrated as the reciprocal of mean fluorescence intensity (1/MFI) to demonstrate the increase in mitochondrial iron.

Mitochondrial ROS
SH-SY5Y cells (6 × 10 5 cells/well) were added into 12-well plates and incubated at 37 • C and 5 % CO 2 overnight prior to being taken for experimentation.Cells were preincubated with weak base drugs in the absence or presence of DFO (50 μM) for 1 h, washed gently three-times with PBS, and incubated with 2.5 μM of MitoSox (Invitrogen, cat.no.M36008, Carlsbad, USA), which detects superoxide (O 2 collected, and MitoSox fluorescence was analyzed in cells dissociated from the culture plates at an excitation of 510 nm and an emission of 580 nm by flow cytometry (Attune NxT, ThermoFisher, Waltham, USA).

Mitochondrial membrane potential
SH-SY5Y cells were added to 24-well plates at a density of 6 × 10 5 cells/well and were incubated overnight at 37 • C and 5 % CO 2 .Cells were preincubated with weak base drugs in the absence or presence of DFO (50 μM), washed gently three-times, and treated with 100 nM MitoTracker™ Red CMXROS (Invitrogen, cat.no.M7512, Carlsbad, USA) for 30 min in a 37 • C and 5 % CO 2 incubator.Cells were washed three times with warm 1X PBS, re-suspended in 1 ml PBS and dissociated from the culture plates by trituration and placed in Eppendorf tubes and the mean fluorescence intensity was determined using a flow cytometer (Attune NxT, ThermoFisher, Waltham, USA) at an excitation of 579 nm and an emission of 599 nm.For confocal microscopy, SH-SY5Y cells were added to 35 mm 2 dishes and incubated overnight in an incubator kept at 37 • C and 5 % CO 2 .Cells were treated with respective drugs for 1 h, washed gently three-times with PBS, and incubated with 100 nM Mito-Tracker Red CMXROS and 200 nM MitoTracker™ Green FM (Invitrogen, cat.no.M7514, Carlsbad, USA) for 30 min in a 37 • C and 5 % CO 2 incubator.Cells were washed three times with PBS, stained with 1 μg/ml Hoechst 33342 for 10 min, washed with PBS three times, and imaged using confocal microscopy (Zeiss 8000, Jena, Germany) at an excitation of 579 nm and an emission of 599 nm.

Cell death
Cell viability was determined using propidium iodide (PI) (BD Biosciences, cat.no.556463, Franklin Lakes, USA) and flow cytometry.SH-SY5Y cells and U87MG cells were added to 24-well plates at a density of about 6 × 10 5 cells/well, incubated overnight at 37 • C and 5 % CO 2 , and cells were treated with respective drugs for 24 h.DFO was added to the cells 1 h prior to adding the weak base drugs.After 24 h, cells were washed three times with warm 1X PBS and stained with PI (3 μM) at room temperature for 15 min.Cells were then dissociated from their culture plates by gentle trituration and then placed in Eppendorf tubes where the fluorescence was measured at an excitation of 532 nm and an emission of 570 nm using a flow cytometer (Attune NxT, ThermoFisher, Waltham, USA).

Reagents
Chemical reagents were obtained from ThermoFisher Scientific except as noted above.

Data and statistical analyses
All data were reported as means ± standard deviation (SD).Statistical significance between controls and treatments (two groups) were analyzed with a two-tailed Student's t-test, and statistical significance between 3 groups or more (multiple groups) were analyzed with oneway ANOVA plus a Tukey's multiple comparison post-hoc test.A p-value less than 0.05 was considered statistically significant.GraphPad Prism 9.3.1 software was used to perform all statistical analyses.No predetermined exclusion criteria were performed.Normal distribution of data was assessed for statistical analysis by using the Shapiro-Wilk test (p<0.05was considered as not to conform to a normal distribution).The Kruskal-Wallis non-parametric test was used to compare data that had a skewed distribution.The number of cells used for this study was determined based on previous studies [38][39][40].No outliers were detected using ROUT method.Results were based on experiments conducted at least three separate times; each time in triplicate.These technical replicates were used to ensure the reliability of single values.

Ethical statement
Institutional ethical approval was not required for this study.

The effects of weak base drugs on endolysosome pH
To confirm previous findings and because endolysosome deacidification decreases [Fe 2+ ] el [9,14,[18][19][20][21], we next determined the effects of atropine, azithromycin, fluoxetine, metoprolol, and tamoxifen on endolysosome pH.Atropine, azithromycin, fluoxetine, metoprolol, and tamoxifen significantly de-acidified endolysosomes; illustrated in realtime (Figure 3A-E) as well as the peak responses in pH over a duration of 30 min (Figure 3F and G).Endolysosome de-acidification by atropine (Figure 3A), azithromycin (Figure 3B), fluoxetine (Figure 3C), metoprolol (Figure 3D), and tamoxifen (Figure 3E) increased slowly to peak effects over the 30 min incubation period (Figure 3F and G).These pH changes equate to significant decreases in the proton concentrations in endolysosomes of 19 % with atropine (200 nM), 14 % with azithromycin, 19 % with fluoxetine, 17 % with metoprolol, and 20 % with tamoxifen (Figure 3G).The MFI of 3000 endolysosomes was measured per cell-culture preparation, which was done independently five separate times for each group (n=15,000).Each one of the five-separate cell-culture preparations per group represents a data point.For each group plotted, n=total number of endolysosomes.With 15,000 endolysosomes used for experimentation per group, cells were randomly chosen in the field of view in the microscope.

Weak base drug-induced depolarization of mitochondrial membrane potential was prevented by the endolysosome-specific iron chelator deferoxamine
Endolysosome Fe 2+ stores, when released, are sufficiently large to depolarize mitochondrial membrane potentials [15].Accordingly, we next determined the effects of weak base drugs on mitochondrial membrane potential (∆ m ).A 1 h treatment of SH-SY5Y cells with atropine, azithromycin, fluoxetine, metoprolol, and tamoxifen significantly decreased CMXROS fluorescence staining; findings indicative of ∆ m depolarization (Figure 6).DFO pre-treatment for 1 h significantly blocked atropine-, azithromycin-, fluoxetine-, metoprolol-, and tamoxifen-induced ∆ m depolarization, and the ∆ m was significantly polarized (increased) by DFO alone (Figure 6).

Weak base drug-induced cell death was prevented by the endolysosome-specific iron chelator deferoxamine
Endolysosome Fe 2+ stores, when released, are sufficiently large to cause cell death [15,16].Accordingly, we next tested the effects of weak base drugs on cell viability using propidium iodide 24 h post-drug treatments.Atropine, azithromycin, fluoxetine, metoprolol, and tamoxifen significantly increased cell death (Figure 7).DFO alone significantly decreased cell death, and DFO pre-treatment for 1 h  significantly blocked atropine-, azithromycin-, fluoxetine-, metoprolol-, and tamoxifen-induced cell death (Figure 7).Similar effects were observed in U87MG astrocytoma cells in which DFO pre-treatment for 1 h significantly blocked atropine-, azithromycin-, fluoxetine-, metoprolol-, and tamoxifen-induced cell death (Supplemental Figure 3).

Discussion and conclusions
It has been estimated that up to 75 % of modern pharmacotherapeutics are weak base drugs that exhibit relatively high pK a values [1,50].Although weak base drugs have favorable pharmacokinetic properties and are widely used clinically, they enter endolysosomes where they reside because of ion trapping for extended periods of time, they affect endolysosome structure and function, and they can induce cytotoxicity [1,6,7].Endolysosome sequestration of weak base drugs results in endolysosome de-acidification, and endolysosomes contain high concentrations of readilyreleasable Fe 2+ whereupon inhibiting endolysosome acidity (de-acidification) induces iron dysregulation and endolysosome iron release; endolysosome de-acidification and iron release from endolysosomes have been linked directly to neurotoxicity [9,14,51].Here, using pharmacologically and therapeutically relevant doses of five very different weak base drugs [41][42][43][44][45], we found that these drugs de-acidified endolysosomes, decreased Fe 2+ levels in endolysosomes, increased Fe 2+ and ROS levels in the cytosol and mitochondria, induced mitochondrial membrane potential depolarization, and increased cell death; effects all prevented by the endocytosed iron-chelator deferoxamine (DFO) (Figure 8).Iron homeostasis is linked to endolysosome acidity, whereupon iron dysregulation is induced by endolysosome de-acidification [13][14][15][16].Not surprisingly, disruption of endolysosome pH continues to be implicated in neurotoxicity and in a broad range of conditions including cancer, aging, and neurodegenerative diseases [10,11,52].However, despite decades of study, the underlying mechanisms are poorly understood, especially in regards to endolysosomes [11].Thus, our studies were conducted for time intervals ranging from 30 to 60 min because this is when peak endolysosome responses were observed.These intervals were chosen because we were determining early mechanisms that lead to cell death and not mechanisms affected by cell death.
Atropine (pK a =10.3) is a clinically relevant anti-cholinergic drug that induces mitochondrial swelling as well as cell death [29].Azithromycin (pK a =8.5) is one of the most commonly prescribed antibiotics and has a known risk for increased cardiovascular cell death [30].Fluoxetine (pK a =9.8) is a type of anti-depressant known as a selective serotonin reuptake inhibitor (SSRI) that increases levels of ROS and induces cell death [31].Metoprolol (pK a =9.7) is a beta-adrenergic blocker that can induce myocardial ischaemia [32].Tamoxifen (pK a =8.8) is an anti-estrogen drug that induces autophagy and cell death [33].Because these five weak base pharmaceuticals are commonly taken and can provide a therapeutic benefit as well as induce adverse effects, it was important for us to demonstrate the extent to which and mechanisms by which a wide spectrum of weak base drugs affect the pools of endolysosome Fe 2+ .
Previously, we reported that endolysosome de-acidification by the weak base drug chloroquine, bafilomycin which inhibits v-ATPase, the HIV-1 gp120 coat protein, and mu opioid receptor agonists induced endolysosome Fe 2+ release, increased cytosolic and mitochondrial Fe 2+ and ROS levels, and increased cell death [14][15][16]53].Because endolysosomes are important regulators of iron-related neurotoxicity and neurodegeneration, weak base drugs can be neurotoxic because they accumulate in and de-acidify endolysosomes [19,21,22].It was important to determine mechanisms by which a diverse group of weak base drugs including atropine, azithromycin, fluoxetine, metoprolol and tamoxifen induce cytotoxicity and cell death.As demonstrated, these weak base drug-induced increases in Fe 2+ and ROS levels in the cytosol and mitochondria as well as cell death appear to be downstream events of weak base drug-induced de-acidification of endolysosomes and endolysosome Fe 2+ release.
The acidic lumen of endolysosomes generates a pH gradient with the cytosol [54].At physiological pH, the neutral fraction of weak base drugs facilitates drug diffusion through membranes into endolysosome's acidic lumen whereupon ionization occurs and drug egress into the cytosol is restricted; a process known as ion trapping or lysosomotropism [6].Notably, 75 percent of drugs are estimated to have weak base pK a values, and the World Drug Index (WDI) determined that 63 % of drugs were ionizable between pH 2 and 12 [1,55].Furthermore, many cardiovascular and central nervous system drugs are lipophilic amines with amine groups that are ionizable (pK a >6); hence, they are lysosomotropic and can accumulate in endolysosomes at concentrations several hundred times higher than the cytosol [56].Although lipophilic drugs can induce neurotoxicity [57], the entering of weak base drugs via pH partitioning into endolysosome's acidic lumen initiates cytotoxic responses.
Central to neurotoxicity and neurodegeneration is mitochondrial dysfunction, and many weak base drugs induce mitochondrial dysfunction [58,59].In particular, mitochondrial membrane potential (∆ m ) depolarization appears to be an early mitochondrial event leading to cell death [60].Correspondingly, many weak base drugs are known to depolarize the mitochondrial membrane potential [61,62] and our findings clearly showed that the weak base drug-induced efflux of Fe 2+ from endolysosomes caused mitochondrial depolarization as well as cell death.
These five weak base drugs readily and effectively cross the blood brain barrier and reach concentrations much higher than those used in our studies.Fluoxetine treatment over 6-8 months in 22 human subjects reached peak brain levels of 10.7 μg/mL; a concentration of 30.9 μM [23].
Another study of 13 human subjects demonstrated that 5 weeks of fluoxetine treatment (20 mg/day) resulted in mean brain tissue concentrations of 25.5 μM and 120 mg/day resulted in mean brain tissue concentrations of 41.4 μM [24].
Tamoxifen treatment (30 mg/day) in human subjects for 7 days resulted in mean concentrations of 1.8 μM in normal brain tissue and 2.7 μM in brain cancerous tissue [25].Furthermore, a single oral dose of azithromycin (500 mg) in 20 human subjects 48 h prior to surgery resulted in mean concentrations of 5 μM in brain tissue [26].Human subjects who received metoprolol (200 mg/day) for 22 days had mean brain tissue concentrations of 3.5 μM and human subjects who received metoprolol (200 mg/day) for only 3 days had mean brain tissue concentrations of 2.9 μM [27].In a mouse model, a single dose (10 mg/kg) of atropine resulted in mean brain tissue concentrations of 144 nM [28].Thus, the concentrations of these five weak base drugs used and reported in our study are relevant to levels reached with their therapeutic use.
In addition to the U87MG astrocytoma and SH-SY5Y neuroblastoma cell lines used in this study, we have studied these same mechanisms in primary cells including rat cortical neurons [9], human Schwann cells [63], human cortical neurons [64], and mouse hepatocytes (unpublished).In all cases, weak-base drugs de-acidified endolysosomes.These drugs circulate throughout the body and enter the brain at concentrations like those used in this study.Thus, the findings are of pharmacotherapeutic relevance and potential clinical significance.Although we have not studied cardiac myocytes, endothelial cells, or breast cancer cells it is extremely likely that these same effects on pH, iron and ROS would be present given how many different cell types we have studied.Furthermore, we have routinely used positive controls including the weak-base chloroquine and the v-ATPase inhibitor bafilomycin A1 [16,35] to show comparable levels of toxicity with multiple different insults.The observed levels of toxicity were similar to those reported here for the five weak base drugs.Lastly, we confirmed our findings in both SH-SY5Y neuroblastoma cells and U87MG astrocytoma cells to broaden the impact of our studies.Key experiments were conducted using the U87MG cells as mentioned in the Results section (page [16][17][18] and the data were included in the Supplemental Material section.
Overall, our study demonstrates that a functionally and structurally diverse group of weak base drugs induce Fe 2+ release from endolysosomes that is sufficient to significantly increase Fe 2+ levels in the cytosol and mitochondria as well as induce oxidative stress and cell death.The mechanisms underlying this cytotoxicity described here may lead to a better understanding of weak base pharmacotherapeutics and their potential adverse effects.

Figure 2 :
Figure 2: Atropine-, azithromycin-, fluoxetine-, metoprolol-, and tamoxifen-induced decreases in endolysosome Fe 2 levels.(A) Qualitatively, atropine, azithromycin, fluoxetine, metoprolol, and tamoxifen decreased the fluorescence staining of FeRhoNox-1 for endolysosome Fe 2+ .Images were acquired with a confocal spinning-disc microscope of SH-SY5Y cells that were treated for a 30 min period with water (CTL), atropine (200 nM), azithromycin (100 nM), fluoxetine (100 nM), metoprolol (200 nM), and tamoxifen (300 nM).The images were reconstructed with the software Imaris 3D and illustrate FeRhoNox-1 staining (red) of Fe 2+ inside of endolysosomes.(B) Quantitatively, endolysosome Fe 2+ levels as indicated by mean fluorescence intensity (MFI) for the staining of FeRhoNox-1 were significantly (p<0.0001)decreased by atropine (200 nM), azithromycin (100 nM), fluoxetine (100 nM), metoprolol (200 nM), and tamoxifen (300 nM).For analysis, an ANOVA with Tukey's post hoc multiple comparisons test was used.The MFI of 3000 endolysosomes was measured per cell-culture preparation, which was done independently five separate times for each group (n=15,000).Each one of the five-separate cell-culture preparations per group represents a data point.For each group plotted, n=total number of endolysosomes.With 15,000 endolysosomes used for experimentation per group, cells were randomly chosen in the field of view in the microscope.

Figure 3 :
Figure 3: Endolysosome de-acidification by the weak base drugs atropine, azithromycin, fluoxetine, metoprolol, and tamoxifen.(A-E) Representative images of SH-SY5Y cells treated for 30 min with atropine (200 nM), azithromycin (100 nM), fluoxetine (100 nM), metoprolol (200 nM), and tamoxifen (300 nM) showed increased endolysosome pH.(F) Peak levels of endolysosome de-acidification within the 30 min incubation interval were significantly increased by atropine (p<0.0001),azithromycin (p<0.001),fluoxetine (p<0.0001),metoprolol (p<0.0001) and tamoxifen (p<0.0001).(G) Peak levels of endolysosome hydrogen ions within the 30 min incubation interval were significantly (p<0.0001)decreased by atropine, azithromycin, fluoxetine, metoprolol, and tamoxifen.(F and G) Individual data points as well as mean ± SD values were included on each bar.For analysis, an ANOVA with Tukey's post hoc multiple comparisons test was used.The mean pH measurement of 50 endolysosomes across five cells was obtained independently from five cell-culture preparations for each group (n=250) and represents each data point.For each group plotted, n=total number of endolysosomes.With 20 cells used for experimentation per group, cells were randomly chosen in the field of view in the microscope.No cells were intentionally excluded.* p<0.05.

Figure 6 :
Figure 6: Weak base drug-induced depolarization of mitochondrial membrane potential was prevented by the endolysosome iron chelator deferoxamine.The dye CMXROS was used to measure mitochondrial membrane potential (∆ m ) where the intensity of the dye decreases when the ∆ m is depolarized.A 1 h pre-treatment with the endolysosome iron chelator deferoxamine (DFO, 50 μM) significantly (p<0.0001)polarized the mitochondrial membrane potential and significantly (p<0.0001)blocked atropine-(200 nM), azithromycin-(100 nM), fluoxetine-(100 nM), metoprolol-(200 nM), and tamoxifen-(300 nM) induced mitochondrial membrane potential depolarization.For analysis, an ANOVA with Tukey's post hoc multiple comparisons test was used.The MFI of 10,000 cells were obtained independently from five cell-culture preparations for each group (n=50,000) and represents each data point.For each group plotted, n=total number of cells.With 50,000 cells used for experimentation per group, no cells were intentionally excluded.* p<0.05.

Figure 7 :
Figure 7: Weak base drug-induced increases in cell death was prevented by the endolysosome iron chelator deferoxamine.Cell death of SH-SY5Y cells was measured using propidium iodide 24 h post-drug treatments.A 1 h pre-treatment with the endolysosome iron chelator deferoxamine (DFO, 50 μM) significantly (p<0.0001)decreased the percentage of cell death and significantly (p<0.0001)blocked atropine-(200 nM), azithromycin-(100 nM), fluoxetine-(100 nM), metoprolol-(200 nM), and tamoxifen-(300 nM) induced increases in the percentage of cell death.For analysis, an ANOVA with Tukey's post hoc multiple comparisons test was used.The MFI of 10,000 cells were obtained independently from five cell-culture preparations for each group (n=50,000) and represents each data point.For each group plotted, n=total number of cells.With 50,000 cells used for experimentation per group, no cells were intentionally excluded.*p<0.05.

Figure 8 :
Figure 8: Atropine-, azithromycin-, fluoxetine-, metoprolol-, and tamoxifen-induced endolysosome de-acidification caused an efflux of ferrous iron from endolysosomes.The iron released from endolysosomes was sufficient to increase cytosolic and mitochondria iron and ROS levels, decreased mitochondrial membrane potential (MMP), and increased cell death.Deferoxamine (DFO) blocked the effects of the weak base drugs tested.