Prodromal Parkinson's disease and the catecholaldehyde hypothesis: Insight from olfactory bulb organotypic cultures

Parkinson's disease (PD) is a progressive, neurodegenerative disorder with an increasing incidence, unknown etiology, and is currently incurable. Advances in understanding the pathological mechanisms at a molecular level have been slow, with little attention focused on the early prodromal phase of the disease. Consequently, the development of early‐acting disease‐modifying therapies has been hindered. The olfactory bulb (OB), the brain region responsible for initial processing of olfactory information, is particularly affected early in PD at both functional and molecular levels but there is little information on how the cells in this region are affected by disease. Organotypic and primary OB cultures were developed and characterized. These platforms were then used to assess the effects of 3,4‐dihydroxyphenylacetylaldehyde (DOPAL), a metabolite of dopamine present in increased levels in post‐mortem PD tissue and which is thought to contribute to PD pathogenesis. Our findings showed that DOPAL exposure can recapitulate many aspects of PD pathology. Oxidative stress, depolarization of mitochondrial membranes, and neurodegeneration were all induced by DOPAL addition, as were measured transcriptomic changes consistent with those reported in PD clinical studies. These olfactory models of prodromal disease lend credence to the catecholaldehyde hypothesis of PD and provide insight into the mechanisms by which the OB may be involved in disease progression.

Parkinson's disease (PD) is an incurable neurodegenerative disorder affecting more than 10 million people worldwide, a figure expected to double within the next generation. 1pecific causes of the disease remain elusive, rendering the development of treatments and preventive strategies a significant challenge.Furthermore, the diagnosis is based on the clinical presentation, in patients, of cardinal motor symptoms, which arise after years of underlying progressive pathology.][4][5] Olfactory bulb (OB) dysfunction has been associated with PD for decades 6,7 and the OB is involved in many other neurodegenerative disorders, including Alzheimer's and Huntington's disease, together with other synucleinopathies, [8][9][10] pointing to an, as yet undiscovered, common pathological mechanism.Being the first part of the brain affected by PD, symptoms associated with the OB, such as hyposmia, appear decades before the clinical onset of the disease during the so-called prodromal phase 11 and have been used to diagnose and monitor disease progression. 12,13Thus, it is crucial to understand why the OB is affected at such an early stage and if there is any potential to stop pathology spreading from there to the nigrostriatal circuitry.While in recent years some in vivo models have been proposed to study the prion-like spreading of the pathology from the OB, [14][15][16] these animal studies are technically demanding and costly; for this reason, we have developed ex vivo and in vitro models of prodromal PD using OB organotypic slices and mixed primary cultures.
Organotypic slices exploit in vitro cell culture systems' main advantages while keeping the three-dimensional structures and cytoarchitectures of living tissues, thus bridging the gap between in vivo and in vitro. 17,18They are easy to obtain, accessible, convenient and, at the same time, are reliable and relatable to the complex brain milieu.Thus, they provide a useful platform for studying neurodegenerative disorders, while reducing the number of animals used in research. 19Various ex vivo PD models are already present in literature, but they mainly focus on the midbrain and the nigrostriatal system, mimicking a more advanced disease stage.The pathology is often induced by the selective degeneration of dopaminergic neurons, using non-physiologically relevant toxins, like 6-hydroxydopamine (6-OHDA), 20 rotenone 21,22 or it is based on axotomy of nigrostriatal connections. 23ere, we describe the induction of ex vivo prodromal PD-like pathology using 3,4-dihydroxyphenylacetaldehyde (DOPAL), a metabolite of dopamine which is present within the OB under normal physiological conditions.][35] As the published literature on OB primary cells and slices is very limited, we first characterized OB cultures and then explored the ability of tissue exposure to DOPAL to mimic prodromal PD pathology, with particular focus on neurodegeneration, oxidative stress, mitochondrial dysfunction, and transcriptome changes.

| Animals
Sprague-Dawley rats were purchased from Charles-River Laboratories or bred in-house.Animals were housed under standard conditions, 12-h light:12-h dark cycle, and had access to water and food ad libitum.

| Preparation of organotypic slices
OB organotypic slices were prepared from postnatal day 10 (P10) Sprague-Dawley rats and cultured following the Stoppini method. 36Pups were completely anesthetized clinical studies.These olfactory models of prodromal disease lend credence to the catecholaldehyde hypothesis of PD and provide insight into the mechanisms by which the OB may be involved in disease progression.

K E Y W O R D S
catecholaldehyde hypothesis, DOPAL, olfactory bulb, organotypic slices, Parkinson's disease with 5% isoflurane and culled by decapitation.The whole OBs were exposed and placed on a sterile silicon platform on the stage of McIlwain tissue chopper and oriented to obtain coronal slices.Slices were then transferred onto a cell-culture insert (Millipore) and cultured for 7 days in vitro (DIV) before any experimental procedure was applied.The culture media were changed the day after the slicing process (DIV1) to remove debris and dead cells, then every 2-3 days.The medium used contained 50% MEM (Gibco 51200-087), 25% MEME (Sigma M2279), 22% HBSS (Sigma H9269), 1% N-acetyl-cysteine (w/v, Sigma M2279), 1% ascorbic acid (w/v, Sigma A8960), 1% L-glutamine (w/v, Sigma G7513), 5 mg/mL glucose (Sigma), and antibiotics/antimycotics (Invivogen).The donor age and media composition were optimized to maximize the viability of OB organotypic slices (Figure S1).

| Preparation of primary OB mixed cells
After exposing, OBs were minced and digested in a papain solution (Worthington) for 1 h at 37°C.After trituration, cells were passed through a 70-μm cell strainer and counted.Cells were seeded on poly-l-lysine (PLL)-coated plates or coverslips and cultured for at least 7 DIV before the experimental treatment.The culture medium, consisting of Dulbecco's Modified Eagle's Medium/Nutrient Mixture F-12 Ham (Sigma D6421), supplemented with 1% L-glutamine (Sigma), 2% B27 (Gibco 17504-044), 1% FBS (Sigma F7524), 6 mg/mL glucose (Sigma), and antibiotics/ antimycotics (Invivogen), is suitable for postnatal cultures and has been already successfully used in previously published studies. 37-39

| DOPAL treatment
Organotypic slices and cells were cultured for 7 DIV before any experimental treatment.This period allows recovery from the traumatic slicing process and provides sufficient time for primary cell attachment and growth.Particularly in slices, treatments were randomly assigned to each well, eliminating any bias caused by visually inspecting the slices.Different doses of DOPAL were prepared by serial dilution in the culture medium.Since the stock of DOPAL is made up of methanol, vehicle control was always included in each experiment, excluding the MitoStress test where only an untreated control was assessed.This is due to a strict limitation in the number of groups that can be simultaneously tested.The vehicle was made up of the same amount of methanol as the highest dose of DOPAL included in each experiment.DOPAL was obtained from Cayman Chemicals (18448).

| Cell viability assays
OB cultures' viability, both organotypic slices and the primary mixed cell were measured using three different assays.Firstly, the alamarBlue™ assay kit (Invitrogen DAL1100) was used, as per the manufacturer's instruction, to investigate the effect of DOPAL on cells' metabolic activity.Briefly, cultures were incubated with a 10% alamarBlue™ solution in a culture medium for 4 h at 37°C and 5% CO 2 .One-hundred microliters of supernatant was then transferred to a clear, flat-bottomed 96-well plate (in triplicate) and the fluorescence at an excitation wavelength of 570 nm and emission wavelength of 585 nm was measured.Results were expressed as a percentage of alamarBlue™ reduction compared to the vehicle control (methanol) and each well baseline at DIV7.The raw baseline measures before DOPAL treatment are shown in Figure S3.As an indication of cell death, the amount of glucose-6-phosphate dehydrogenase (G6PDH), a ubiquitous cytoplasmic enzyme that is part of the pentose phosphate pathway, was measured in the supernatant.G6PDH was detected using the Vybrant Cytotoxicity Assay Kit (Molecular Probes (V-23111)), a two-step enzymatic process that reduces resazurin into red-fluorescent resorufin, that can be spectrophotometrically quantified using an excitation wavelength of 563 nm and an emission wavelength of 587 nm (Figure S3).Another marker of cell viability was the DNA content of OB cultures, as measured using the Quant-iT™ PicoGreen® dsDNA assay kit (Invitrogen P11496).As per the manufacturer's instructions, cells were lysed to release the DNA by three freeze-thaw cycles before performing the assay.Based on the binding and increased fluorescence of the PicoGreen® reagent with double-strand DNA, the assay was performed in triplicate, in a black opaque 96-well plate (Sarstedt).The fluorescence was read using an excitation wavelength of 480 nm and an emission wavelength of 520 nm.Concentration values were extrapolated from dsDNA standards and normalized to the value of the control group.The Quant-iT™ PicoGreen® dsDNA assay kit was also used to normalize the Seahorse MitoStress test results according to ng of DNA.

| Nitrite and oxidative stress assays
The Griess reagent kit (Invitrogen G7921) and the Fluorometric Hydrogen Peroxide Assay Kit (Sigma Aldrich MAK165) were used to assess the levels of nitrites and reactive oxygen species in the culture medium of organotypic slices.For the nitrite quantification, 150 μL of samples or nitrite standards was added to each well of a clear 96-well plate and incubated for 30 min at room temperature, in the dark, with the Griess reagents.The absorbance was read at 548 nm and the nitrite concentration in μM extrapolated from the standard curve.The Fluorometric Hydrogen Peroxide Assay is based on a red peroxidase substrate, which reacts with H 2 O 2 in the supernatant to produce a fluorescent end-product, in a reaction catalyzed by horseradish peroxide.Samples were incubated for 30 min at room temperature, protected from light in a black opaque 96-well plate.The fluorescence was read at the respective excitation and emission wavelengths of 540 and 590 nm.Data concentrations were expressed in nM after extrapolation with an H 2 O 2 standard curve.To further detect ROS, the fluorogenic probe CellROX Green was used (Invitrogen C10444), which on oxidation binds to DNA, resulting in emission of green fluorescence.As per manufacturer's instructions, the probe was added directly to the culture medium, at a final concentration of 5 μM, for approximately 30 min.After washes, cells were immediately imaged or otherwise fixed and co-stained with a different cell marker.In both cases, DAPI was used to visualize nuclei, and the cells were imaged using confocal microscopy within 24 h.

| Mitochondrial membrane depolarization and flow cytometry
The mitochondrial membrane potential (Ψm) was monitored using the ratiometric dye 5,5,6,6-tetrachloro-1,1,3,3tetraethylbenzimidazolylcarbocyanine iodide (JC-1).The dye accumulates in mitochondria, in either a monomeric or an oligomeric form, depending on the Ψm.The monomer, which has green fluorescence (emission ∼530 nm), is predominant in depolarized mitochondria.Alternatively, the oligomer is formed only when the AΨm is lower than −140 mV and possesses a red fluorescence (emission ∼590 nm). 41,42After a 4-h treatment with DOPAL, cells were incubated with the JC-1 dye (Enzo Life Sciences ENZ-52304) for 30 min, washed twice in PBS, and then analyzed by confocal microscopy, flow cytometry, or spectrophotometrically.Upon incubation with JC-1, cells in flasks were washed with PBS and trypsinized.A solution of 1 μM TO-PRO-3 (Invitrogen T3605) was added to the cell suspension and incubated for 15 min to identify dead cells.Following three washes in PBS, cells were analyzed using the following lasers and filters: 488 nm laser with 530/30 filter for the green JC-1 emission (monomer, depolarized mitochondria), 488 nm laser with 670 low-pass filter for the red JC-1 emission (aggregates, polarized mitochondria), 633 nm laser with 660/20 filter for the viability dye TO-PRO-3.The results were analyzed in FlowJo (version X.07, FlowJo) and displayed as histograms in the red and green channels and scatter plots of polarized versus depolarized mitochondria.The gates used are shown in Figure S4.Cells on coverslips were stained with JC-1 as described.After three washes in PBS, coverslips were quickly mounted on glass slides with a PBS drop to maintain the cells alive for the imaging period.Images were taken using a confocal laser scanning microscope, with both channels excited using the 473 nm lasers and then acquired with a 490-590 nm filter and 575-675 nm filter for the emitted green and red signals, respectively.At least three z-stacks per sample were imaged, and the experiment was repeated in two independent experimental replicates.Cells seeded on a 24-well plate were, after treatment with DOPAL, stained for 30 min with the JC-1 dye.After three washes in PBS, the fluorescence emission was measured in a Varioskan Flash spectral scanning multimode reader (ThermoFisher).The excitation was set at 488 nm, while the emission was detected at 530 nm for the green and 600 nm for the red channel.The red: green ratio, independent of mitochondrial shape, density, or size, and depending only on the membrane potential was calculated and further normalized by the vehicle control before statistical analysis was performed. 42-45

| Extracellular flux analysis
Primary OB cells were seeded into eight-well Seahorse XFp Cell Culture Miniplate (Agilent 103022-100) pre-coated with PLL (Sigma Aldrich) at 80000 cells/well (optimization is shown in Figure S5).After 7 DIV, cells were subjected to the MitoStress test assay (Seahorse Bioscience, Agilent).Two different treatments with DOPAL were applied: an acute exposure, with DOPAL, injected in the medium during the assay, and a 4-h pre-treatment before the MitoStress test.The raw data of oxygen consumption rate (OCR) and extracellular acidification rate (ECAR), for each well, were normalized by the DNA content, as assessed by PicoGreen® assay (Invitrogen P11496) in order to reduce variability caused by different cell numbers.The normalized data of OCR and ECAR were subdivided into component rates and analyzed following the calculation presented by Mookerjee and colleagues. 46,47The bioenergetic health index (BHI) was calculated as by Chacko and co-workers. 48

| Gene expression analysis
Gene expression in DOPAL treated and control OB slices was assessed by qRT-PCR analysis using a commercially available array, specific for Parkinson's disease and for mitochondrial related genes (PARN-124ZA and PARN-087Z Qiagen GmBH, Hilden, Germany).Briefly, total RNA was isolated from organotypic slices exposed to methanol or 100 μM DOPAL, using the phase separation method.The quality, purity, and concentration of the extracted RNA were measured using the NanoDrop 2000c spectrophotometer (ThermoFisher).The RNA obtained from each of five independent experiments was pooled together (450 ng from each sample), the genomic DNA removed and the RNA reverse-transcribed into cDNA using the RT 2 firststrand kit (Qiagen GmBH 330401), as per manufacturer's instructions.A 10 μL mix of cDNA, SYBR Green, and nucleotide-free water were pipetted in each of the 384 wells of the array, the qRT-PCR was then performed, as per instructions, with a LightCycler 480 (Roche).The results were quantified by the comparative C t method, and the values were normalized to the expression of five housekeeping genes, all present in the gene array: β-actin (ACTB), β-2 microglobulin (B2M), hypoxanthine phosphoribosyltransferase 1 (HPRT1), lactate dehydrogenase A (LDHA), and ribosomal protein, large, P1 (RPLP1).

| Protein extraction and quantification
To extract protein from organotypic OB cultures, slices were first separated from the well insert PTFE membrane using a spatula and placed in a 1.5 mL Eppendorf containing 100 μL of RIPA buffer (Sigma-Aldrich RO278) supplemented with proteinase and phosphatase inhibitors (cOmplete™, EDTA-free Protease Inhibitor Cocktail 11873580001 and PhosSTOP™ 4906845001, Roche).After a 5-min incubation at 4°C, the tissue was mechanically dissociated using a pestle and vortexed.This step was repeated twice to ensure complete elution of the protein.Upon centrifugation at 17 000g for 15 min at 4°C, the supernatant was collected, aliquoted, and stored at −80°C until used.The total protein concentration in each sample was measured using the bicinchoninic acid (BCA) assay, developed by Smith and colleagues. 49

| Catalase activity assay
Catalase activity was assessed using the catalase assay kit (Cayman Chemical, 707002-480) based on the enzyme's peroxidase function, which reacts with methanol in the presence of an optimal concentration of H 2 O 2 .The formaldehyde produced is measured colorimetrically with Purpald (4-amino-3-hydrazino-5-mercapto-1,2,4-triazole).According to the manufacturer's instructions, 20 μL of a protein sample, diluted 1:2 in sample buffer, was added to each well of a 96-well plate, together with 100-μL assay buffer and 30-μL methanol.The reaction was initiated by adding 20 μL of hydrogen peroxide in all the used wells.Upon incubation at room temperature for 20 min, the reaction was stopped with 30 μL of potassium hydroxide.The same volume of Purpald was added to each well and after 10 min, 10 μL of potassium periodate was pipetted in each well to oxidize the chromogen.The absorbance was then measured at 540 nm, and the catalase activity (in nmol/ min/mL) was calculated from the formaldehyde standards.The result obtained was further normalized by each well's protein content as assessed with the BCA assay.

| Calcium imaging
Primary OB cells were seeded on PLL-coated coverslips (15 mm ⌀) at a density of 3 × 10 5 cells/coverslip.Cells were cultured for at least 2 weeks before the calcium imaging was performed to ensure the maturity of the culture and formation of synaptic connections.On the day of the assay, cultures were incubated for 4 h with DOPAL before the imaging.Cell cultures were then washed with KREBS buffer twice before being incubated with 2 μM Fluo-4 AM (ThermoFisher, F14201) in KREBS buffer for 20 min at 37°C.Following dye loading, the cultures were returned to normal medium at 37°C for 20 min and imaged in warm KREBS buffer in an imaging chamber (Warner Instruments, RC-26GLP) on a Zeiss Axiovert 200 microscope with a 10× magnification.Videos were captured with a Hamamatsu ORCA284 at 2 Hz and stored as uncompressed image sequences.ATP was added to the cultures after 20 s of recording.Videos were analyzed using FluoroSNNAP in MATLAB (MathWorks, Inc.).Cells with >5% variations in fluorescence intensity during recording were identified by time-lapse analysis and cell defined using batch segmentation.The time-lapse fluorescence trace was calculated, transient onset identified, and background noise (ΔF/F < 0.05) determined and analyzed in Igor Pro V8.0 (Wavemetrics).

| Statistical analysis
Data were analyzed and plotted using GraphPad Prism 7 (GraphPad software) and are expressed as mean ± SEM of at least three independent experiments unless otherwise stated.Two-tailed Student's t test, one-way analysis of variance with Tukey's multiple comparisons, or the Kruskal-Wallis test with Dunn's multiple comparison test were used appropriately.A significant difference was set at p < .05.Statistical difference versus the control (0 μM DOPAL) is denoted with *, while the # is used to compare versus the vehicle control (methanol).While data collection and analysis were not performed blind, treatment group was randomly assigned to the samples by generating random numbers in Excel.

| Generation of OB organotypic slices
In order to obtain the most viable and reproducible slices, cultures were obtained from donors at different developmental stages.Coronal olfactory bulb slices from P10 Sprague-Dawley rats were demonstrated to be more viable and resilient than those obtained from younger (P7) or older donors (P14-P16) (Figure S1A-C).OB organotypic slices could be maintained for long periods in culture while preserving the cytoarchitecture and 3D structures physiologically present in the OB in vivo (Figure 1B).In particular, all six of the layers comprising the OB are well conserved, with only a partial loss of the most inner one, the rostral migratory stream (RMS), where, physiologically, stem cells produced in the subventricular zone migrate in order to integrate into the olfactory circuitry. 50uring the initial optimization of the slice culture conditions, the serum's deleterious effect was initially observed in the medium.Slices exposed to serum underwent progressive flattening in the first week of culture, losing their structural integrity (Figure S1D), an effect that has been already reported in the literature for hippocampal and whole brain slices. 23,51A serum-free culture medium was, therefore, used for all further experiments.

| Generation and characterization of OB mixed cultures
To complement the newly developed 3D organotypic slices, primary OB cultures were employed and characterized and, as previously optimized, generated using P10 Sprague-Dawley animals.Cells were cultured for 7 or 14 days in vitro (DIV) before being fixed, stained, and imaged (Figure 2A).While the neuronal and microglial populations (Figure 2C,D) were stable for the culture period of 2 weeks, oligodendrocytes were more abundant at DIV14, when compared to the earlier timepoint at DIV7 (Figure 2E, top, p = .006).Similarly, astrocytes had a larger average size (Figure 2B, middle, p = .004)and covered a more significant area when cultured a week longer (Figure S2A).Approximately 41% of total cells at DIV7 and 28% at DIV14 did not belong to any of the cell population mentioned above and were believed to be a mixture of radial glia (RGs), neuroepithelial cells (NECs), and olfactory ensheathing cells (OECs) (Figure S2F-H).

| DOPAL toxicity in OB cultures
We then hypothesized that the monoamine oxidase metabolite of dopamine, DOPAL, could induce cell death in primary and ex vivo OB cultures and lead to an overall neurodegeneration process.Firstly, we monitored the health of the organotypic slices over time using the alamarBlue™ assay.The first assay was performed at DIV7 before the 3-day DOPAL treatment was initiated (Figure S3D).The values obtained from each well at this time point were used as a baseline, thus eliminating any effect caused by variation in slice dimensions and allowing detection of changes caused only by exposure to the dopamine metabolite.After a 3-day treatment with the catecholaldehyde, the slices' metabolic activity was reduced in a concentration-dependent fashion (Figure 3A).In particular, with concentrations higher than 250 μM, slices displayed significantly lower viability (p = .01vs. 0 μM and vehicle) with an LC50, calculated from the alamarBlue™ data, of 487.9 μM (Figure S3B).
Following a 3-day treatment, DOPAL was withdrawn, and the metabolic activity of slices was measured 4 days later.There were no signs of recovery; conversely, slice health worsened (Figure 3B).A significant reduction in the alamarBlue™ signal at DIV14 was measured at concentrations as low as 50 μM (p = .012vs. 0 μM, p = .049vs vehicle).This effect was more pronounced for cultures exposed to higher concentrations of the catecholaldehyde, with the three highest concentrations showing a metabolic activity that was at most 30% of that detected in vehicle-treated slices (Figure 3B).As a consequence of slice health deterioration, the LC50 at DIV14 was reduced to 100.6 μM (Figure S3C).Additionally, a decrease in the slices' protein content exposed between DIV7 and DIV10 to DOPAL was detected at DIV14, significant for 1000 μM group (Figure 3C, p = .025vs. 0 μM and p = .010vs vehicle).Similarly, when primary OB cells were exposed for The cellular composition of OB cells was studied after 7 or 14 DIV.Cells were stained for GFAP, Iba1, β-tubIII, and Olig2 to detect astrocytes, microglia, neurons, and oligodendrocytes, respectively.The OB is rich in α-syn and the protein is mainly localized with nuclei and neurons.Astrocytes increase slightly, but not significantly, in number, passing from 15.46% ± 3.31% at DIV7 to account for 18.85% ± 7.01% of total cells at DIV14 (B, top).The mean area covered by each GFAP + cell is higher at the later time point (B, bottom, ** p = .004).Both microglia and neuronal population appeared to be stable for the duration of the culture, indeed their numbers (C, D, top) and area covered (C, D, bottom) were not different at DIV7 and DIV14.Oligodendrocytes, like astrocyte, are proliferative cells, and they accounted for 16.09% ± 0.94% and 21.56% ± 0.44% (E, top, ** p = .006) of total cells at DIV7 and DIV14, respectively, but their average area did not statistically change when cultured a week longer (D, bottom).The composition of the cultures at the two time points is shown in (F), around 41% of total cells at DIV7 and 28% at DIV14 do not belong to the astrocytic, microglial, neuronal, or oligodendrocytic population and are believed to be radial glia (RGs), neuroepithelial cells (NECs), and olfactory ensheathing cells (OECs), see Figure S2.Data are expressed as mean ± SEM of three independent experiments, Student's t test.Scale bar = 100 μm.
4 h to the compound, treated cells displayed a reduced DNA content (Figure 3D).Upon DOPAL exposure, a dose-dependent reduction in the number of neurons and their neurites was visible using β-tubIII staining and confocal microscopy (Figure 3E).Furthermore, particularly in the 500 μM group, β-tubIII + neurons displayed an altered morphology, an effect monitored in real time using time-lapse imaging (Figure S3A).

| DOPAL causes oxidative stress
To further investigate the effect of DOPAL on the expression of reactive oxygen species, we used intracellular and extracellular markers of oxidative and nitrative stress.Specifically, using the probe CellRoX, a fluorescent dye that is detectable only under oxidative conditions, it was possible to detect increased ROS levels in cells treated with 10 μM and 100 μM DOPAL.No increase was detected for the highest concentration, possibly due to extensive cell death and the probe's fluorescent saturation at 500 μM (Figure 4A, green).Intracellular ROS appeared to be widespread and not limited to the neuronal population, stained with β-tubulin III (Figure 4A, red).However, a reduction in the number of β-tubulin III + cells was notable and compatible with the concentration-dependent toxicity of DOPAL previously detected (see Figure 3).
In organotypic slices, a significant and dose-dependent increase in both nitrites (Figure 4C) and hydrogen peroxide (Figure 4D) was measured in the culture supernatant.Additionally, an effect of DOPAL was also discernible by the naked eye: the progressive browning of the tissue, noticeable already at close-to-physiological levels of DOPAL, but more pronounced for higher concentrations (Figure 4B).This phenomenon was detectable within the insoluble fraction remaining after protein extraction (Figure 4F).It is consistent with the oxidation of catechols and the generation of quinone, plus the polymerization of oxidized dopamine, a known step in the black pigment neuromelanin.
The incremental increase in nitrites was linear and significantly different from a concentration of 250 μM DOPAL, where nitrites were eight times more abundant when compared to control levels (p = .0274)or vehicle (p = .0273).Similarly, hydrogen peroxide content dramatically increased after DOPAL exposure, showing a significant 127-fold increase for the highest concentration tested (1000 μM, p = .0002vs both 0 μM and vehicle).Moreover, while the levels of H 2 O 2 were comparable to those of the control for the 10, 50, and 100 μM groups (Figure 4D), this increased level became more prominent with DOPAL doses higher than 250 μM.Finally, when exposed to DOPAL, the antioxidant enzyme catalase activity was reduced, particularly for the three highest concentrations of catecholaldehyde (Figure 4E).

| DOPAL depolarizes the mitochondrial membrane
The persistent oxidative state in the cultures had a detrimental effect on mitochondrial function and performances.The mitochondrial membrane potential, Ψm, one of the first indicators of mitochondrial dysfunction, was monitored using the ratiometric dye JC-1, which generates a green signal in depolarized mitochondria and a red signal in polarized healthy ones.Depolarization of mitochondrial membrane was measured by confocal microscopy, flow cytometry, and spectrophotometrically (Figure 5A).
Upon DOPAL exposure, a progressive loss in the number of polarized mitochondria was visible microscopically.At a physiological concentration of the catecholaldehyde, polarized mitochondria appeared less abundant and showed some signs of fragmentation (Figure 5D, higher magnification pictures).A quantitative measure of the reduction of Ψm was achieved from a spectrophotometric analysis of the cells and a reduction in signal intensity was measured at 590 nm after DOPAL exposure (Figure 5B).The reduced fluorescence (red) was statistically altered for the 500 μM group (p = .006vs. 0 μM and p = .002vs. vehicle); however, no significant changes were observed for the JC-1 green signal.As the decrease in polarized mitochondria could be simply due to a decreased total number of mitochondria, we calculated the red/green ratio influenced only by Ψm.The ratio, normalized by the vehicle control, demonstrated progressive depolarization of mitochondrial membrane with increasing concentrations of DOPAL (Figure 5C).
A more comprehensive and quantitative measure of the Ψm was performed using flow cytometry, as displayed in Figure 5E-G.The spectra of the red signals were comparable across groups (Figure 5E).However, the number of mitochondria with a green signal over the gated threshold was remarkably increased upon DOPAL exposure, particularly for the highest concentration of 500 μM (Figure 5F).Finally, from the scatter dot plot of the two channels, it was possible to detect a reduction in the percentage of polarized mitochondria (Figure 5G, inside gate), reducing from 70.8% in untreated cells to 62.4% for physiological concentrations of DOPAL, and further decreasing to 57.9% and 41.1% for 100 and 500 μM, respectively.Although it should be pointed out that, in methanol-treated cells, only around 58% of the mitochondria were polarized, an effect not seen in the other experiments where vehicle and untreated controls had very similar results.

| DOPAL decreases mitochondrial performance within minutes of exposure
At a functional level, the mitochondrial performance was hindered by DOPAL, as assessed using the Seahorse extracellular flux analysis and the MitoStress test, the mechanism of which is summarized in Figure 6A and Figure S6A,B.
A 4-h exposure to the catecholaldehyde induced a decrease in the bioenergetic health index (Figure 6B), an indicator of mitochondrial health, 48 significantly lower after exposure to the highest concentration of DOPAL (p = .0018vs. 0 μM).More signs of mitochondrial damage were detected in the Electron Transport Chain (ETC).Indeed, the coupling efficiency, a measure of the performance of the ETC, was significantly lower in treated cells (Figure 6D), dropping by 5% and 18% after treatment with 100 μM (p = .013vs. 0 μM) and 500 μM (p < .0001 vs. 0 μM), respectively.The proton leak was 2-fold higher in cells treated with 500 μM (p = .048vs. 0 μM) along with a minor (non-significant) increase of approximately 30% which was also detected for the lower concentration (Figure 6E).Untreated cells were able to nearly double their OCR when needed, increasing their O2 consumption rate by 71.5% when compared to basal levels (Figure S6J).A similar outcome was observed in the 100 μM group, but, with the most severe treatment, this spare respiratory capacity was partially, although not significantly, impaired (138.8 ± 11.67%, p = .103vs. 0 μM).
The exposure to DOPAL appeared to be progressively inhibiting the OxPhos branch of ATP production, forcing cells to increase their glycolytic output.Indeed, both the percentages and absolute values of ATP fluxes produced from glycolysis (J ATPglyc ), displayed in Figure S6D,E reaching statistical significance.The 10% increment further confirmed this GI trend upon treatment with 500 μM DOPAL (Figure 6C).
When DOPAL was instead injected acutely before the assay, it caused an immediate decrease in the oxygen consumption rate (OCR) (Figure 6F), significantly different from untreated cells for the most severe exposure (Figure 6H).Interestingly, after an initial slight increase, observed within the first minutes after injection, the extracellular acidification rate (ECAR), an indicator of glycolysis, also appeared to be partially impaired with lower values in both treated groups, and was statistically different for the highest concentration (Figure 6G).
Despite the reduction in OCR mentioned above, no significant changes were detectable in the levels of basal (Figure S7A), maximal respiration (Figure S7B), or the ATP-production coupled to oxygen consumption (Figure S7C), which were constant across the three experimental groups.However, signs of impaired ETC were detected quickly after DOPAL injection; the mitochondrial membranes become gradually more damaged, with a trend towards increased protons leaking across the membrane in the 100 and 500 μM groups (28% and 58%, respectively) (Figure S7E).Damaged mitochondria become less efficient, as confirmed by a significant reduction in coupling efficiency (Figure S7F) and in the spare respiratory capacity, which decreased linearly with increasing concentrations of DOPAL (Figure S7D).
The dopamine metabolite affected oxidative phosphorylation and other cellular processes consuming oxygen as confirmed by the reduction in the non-mitochondrial OCR, displaying a linear decrease, proportional to DOPAL concentration (Figure 6I), an effect that was not observed for the more prolonged 4-h exposure.Similarly, while glycolysis was slightly higher in the first set of experiments, an acute injection appeared to initially inhibit this energy-producing pathway, with many indicators decreased in the treated groups (Figure S6H-L).

| DOPAL mediated changes in OB culture transcriptome
Alterations in the transcriptome were detected in OB organotypic cultures exposed to 100 μM DOPAL between DIV7 and DIV10 (Figure 7A).The concentration of 100 μM was selected as the lowest to induce measurable effects in the cultures.Pooled cDNA samples were applied to a rat Parkinson's disease gene array containing 84 genes linked to disease pathogenesis and progression.The expression of many genes was altered as shown in the heatmap of the log 2 -fold increase (Figure 7B) and in the scatter dot plot of the log 10 (C t ) for both vehicle-and DOPAL-treated OB slices (Figure 7C); single genes over and under the selected threshold are highlighted (fold increase >3 or fold decrease <3).The genes that had an increase >3 (corresponding to a log fold-increase of 1.58) compared to the vehicle control are reported in Figure 7D.Additionally, the top 15 most upregulated and downregulated genes and their relevance to PD are reported in Tables S1 and  S2, respectively.
While nearly absent in methanol-treated slices, the amyloid-beta precursor protein (APP) was highly expressed following DOPAL treatment, with a log 2 fold increase of 14.34.Similarly, the brain-abundant membrane attached signal protein 1 (BASP1) was also upregulated in DOPAL-exposed slices.Caspase 1 (CASP1) and caspase 7 (CASP7) were expressed at a higher level in treated slices with increments of 10.25 and 5.29, respectively.Two genes of the ataxin family, involved in protein homeostasis and deubiquitination, were also more abundant upon catecholaldehyde exposure, specifically ataxin 3 (ATXN3), which had a log2 increase F I G U R E 5 DOPAL causes depolarization of mitochondrial membrane.(A) Mitochondrial membrane potential was assessed microscopically, spectrophotometrically, and by flow cytometry using the ratiometric dye JC-1.(B) A reduction in the red signal from polarized mitochondria was quantified spectrophotometrically, leading to a significant decrease for the 500 μM group (p = .0064vs. 0 μM p = .0025vs. vehicle).(C) The ratio of the green/red fluorescence gradually decreased as DOPAL concentration was augmented, from a value of 1.097 ± 0.171 for untreated cells to 0.932 ± 0.057 for 10 μM, 0.669 ± 0.139 for 100 μM, and 0.128 ± 0.020 for the 500 μM group (p = .0004vs. 0 μM and p = .0009vs. vehicle).(D) A similar trend was observed microscopically, with a progressive decrease in the red staining.The heterogeneous network formed by the organelles in vehicle and control samples is progressively lost with increasing concentrations of DOPAL (higher magnification images, D).The histograms for the red and green channels are shown, respectively, in (E, F).The signal originating from the monomeric dye, present in depolarized mitochondria, increased with DOPAL concentration, while the red signal appeared more constant, possibly due to a spillover from the green channel.The percentage of polarized mitochondria (inside triangle gate in G) is reduced upon exposure to the catecholaldehyde; however, the vehicle control showed a reduction in the number of healthy mitochondria.Scale bars = 50 μm and 10 mm for the higher magnifications inserts (D).Mean ± SEM of N = 3 independent experiments, n = 3 biological replicates (B, C); Mean only, N = 2 and n = 3 for (E-G).Statistical significance was tested using ordinary two-way ANOVA with Tukey's multiple comparison test (B) or ordinary one-way ANOVA with Tukey's multiple comparison test (C).When both * and # are present, # is vs. vehicle, * vs. control (0 μM).

F I G U E 7
Transcriptomic alterations in ex vivo OB slices exposed to DOPAL.(A) RNA was extracted from slices exposed to DOPAL or methanol (vehicle), cDNA synthesized and PCR was performed using the Parkinson's disease and mitochondria gene array on pooled samples, and the gene expression in vehicle-treated slices was compared to those exposed to 100 μM DOPAL.(B) Heatmap of the log 2fold increase/decrease of treated vs vehicle, showing dysregulation of many genes.(C) Scatter plot of the log 10 (Ct) of the treated group vs vehicle control for each gene tested.The three parallel lines indicate, respectively, a 3-fold increase, no change, and 3-fold decrease.In red and blue, the genes with a fold increase or decrease higher than the threshold are indicated.The fold increase of the most upregulated and downregulated genes is shown in (D, E), respectively, with the line representing the cut-off of a 3-fold change.N = 5 independent experiments, n = 3 biological replicates, pooled.

F I G U R E 6
Respirometry analysis of OB cells after exposure to DOPAL.(A) Schematic of the MitoStress Test.After a 4-h treatment with DOPAL before performing the assay, primary OB cells show signs of mitochondrial dysfunction.(B) The BHI, a measure of the general health of the mitochondria, is significantly reduced, decreasing from 0.765 ± 0.089 in the 0 μM group to 0.642 ± 0.105 for the 100 μM condition, and dropping to −0.179 ± 0.246 for the 500 μM group (p = .0018).As a consequence of impaired OxPhos, the percentage of ATP produced via glycolysis, measured with the glycolytic index (C), shows a growth trend with increasing concentration of DOPAL.A dose-dependent increase was also observed for the proton leak (E), significantly higher for the 500 μM group (*p = .0478).Similarly, but with an opposite trend, the coupling efficiency (D) is decreased with increasing DOPAL concentrations, with a value of 79.50 ± 0.88% in untreated cells, reducing to 74.34 ± 1.54% (p = .013vs 0 μM) after treatment with 100 μM DOPAL, and further, to 61.37 ± 0.98% for the highest concentration (p < .0001vs 0 μM).When cells were instead acutely exposed to the dopamine metabolite, a significant decrease was observed for both the OCR (F) and ECAR profiles (G).The response to DOPAL appeared within minutes after the injection (H), with a significant decrease in the OCR (*p = .0147,0 vs 500 μM).In untreated cells, the non-mitochondrial OCR (K) had a value of 1.057 ± 0.192 pmol/ min/ng DNA, dropping to 0.628 ± 0.083 pmol/min/ng DNA upon injection of 100 μM DOPAL, decreasing further in the 500 μM group, to 0.054 ± 0.023 pmol/min/ng DNA (p = .002vs. 0 μM).The reduction in this parameter was not reported after longer exposures to the catecholaldehyde (Figure S6K).Data are expressed as mean ± SEM or as box-plot of at least five independent experiments with two biological replicates each (each dot in the graphs), ordinary two-way ANOVA with Tukey's multiple comparison test (F, G) or one-way ANOVA with Dunnett's multiple comparison test or Kruskal-Wallis with Dunn's multiple comparison test. of 8.0, and 2 (ATXN2) with 2.40.Intriguingly, upregulation was observed in genes involved in calcium homeostasis, specifically, ATPase Ca2 + transporting plasma membrane 2 (ATP2B2) and Ca2 + -dependent secretion activator (CADPS), which had log increases of 7.2 and 1.60, respectively.Also, the levels of mRNA for the translocase of inner mitochondrial membrane 9 (TIMM9) were nearly six times higher in experimental slices.Interestingly, the gene encoding brain-derived neurotrophic factor (BDNF), was overexpressed in DOPAL-exposed organotypic slices, with a log-fold increase of 3.98, possibly pointing to a neuroprotective response to DOPAL.A few genes, localized in the mitochondria and known to orchestrate apoptosis through BCL2, had also increased expression, such as BCL2 interacting protein 3 (BNIP3, 8.6), BCL2 like 1 (BCL2L1, 5.23), BH3 interacting domain death agonist (BID, 3.46), and BCL2 antagonist/killer 1 (BAK1, 3.37).Two genes encoding lipid-interacting proteins, that is, caveolin 2 and CLN8 had also a differential expression upon exposure to DOPAL, with a log-fold change of 6.89 and 4.57, respectively.
Conversely, many genes were downregulated upon exposure to 100 μM DOPAL as shown in Figure 7E and Supplementary Tables S1 and S2.Slit homolog 1 (SLIT1), a gene encoding a molecular guidance protein, was poorly expressed in treated slices, with a log 2 -fold decrease of −4.44.Similarly, the expression of another gene involved in axonal guidance and neurite extension, neurofascin (NFASC) was largely curtailed in experimental tissue, with a log 2 -fold decrease of −2.02.A decrease in the RNA encoding for some dopaminergic markers was also detected.Specifically, the dopamine transporter's log-fold expression (known as solute carrier family 6 member 3, SLC6A3) and the AADC (referred to as DDC, dopa decarboxylase) had values of −3.91 and −3.38, respectively.Not only dopamine but also GRIA3, a gene encoding a subunit of the glutamate receptor, had a decrease in expression of 4-fold after treatment with the catecholaldehyde.
Synaptotagmin 11 (SYT11), a member of a family of proteins involved in vesicular and membrane trafficking and in autophagy and lysosome regulation, was also less abundant in DOPAL-exposed slices, with a log decrease factor of −1.87.The expression of two Park genes, PARK2, encoding the protein Parkin, and PARK7, encoding DJ-1, were both depleted in experimental slices, with a 3-fold lower expression compared to vehicle slices.Results were similar for mitogen-activated protein kinase 9 (MAPK9) and heat shock protein 4 (HSPA4).Other mitochondrial protein-encoding genes, the transporter SLC25A21 and the uncoupling protein 1 (UCP1) had a log decreased expression of −2.59 and −2.34.
Many pathways were dysregulated in DOPAL-treated OB organotypic cultures (Figures S8-S12).Several genes involved in mitochondrial function and health were differently expressed between DOPAL-and vehicle-exposed tissues.Phosphatase and tensin homolog (PTEN) and solute carrier family 25 member 4 (SLC25A4) were downregulated.The expression of PINK1, also known as PARK6, a protein involved with Parkin in mitochondrial quality control, 52 was more than halved in the treated tissue.Mitochondrial membrane polarization and protein and ion transporters across the organelle membranes were disrupted.
Similarly, genes involved in ion channels and transporters were also altered.Mainly, DOPAL appeared to disrupt calcium homeostasis, upregulating CADPS, ATP2B2, both Ca2 + transporters, but downregulating, among others, S100β a calcium-binding protein and SYT11.However, other than a small increase in spontaneous activity, no significant changes were observed upon live-cell calcium imaging (Figure S13A-D).
Cell adhesion and cytoskeleton regulatory genes demonstrated an altered expression profile in DOPALexposed slices.APP and BASP1 were both considerably upregulated, with a log-fold increase of more than 10.Conversely, NFASC, PARK2, cell division cycle 42 (CDC42), microtubule-associated protein tau (MAPT), and adenomatous polyposis coli (APC) expression were 50% lower in treated OB slices.Interestingly, evidence of cytoskeletal changes could be in primary OB cells exposed to DOPAL, in both neurons and astrocytes (Figure S14 and Figure 3E), pointing at a direct effect of DOPAL on specific structural proteins.Ingenuity pathway analysis also confirms cell movement and migration pathway heavily impacted by the treatment (Figure S11).Additionally, for the lower concentration of 10 μM, there was rare positive staining of phosphorylated α-synuclein (pSer129), the main form present in Lewy bodies (Figure S14).

| DISCUSSION
Parkinson's disease has long been considered a motor disorder and the non-motor symptoms associated with the disease have often been neglected.However, many nonmotor features are known to be present in the so-called 'prodromal' phase, years before the motor symptoms appear, and PD diagnosis is formulated. 11In particular, increasing evidence points to the olfactory bulb and the enteric nervous system as the first regions affected by the pathology.Olfactory dysfunction is a well-established symptom of PD, firstly reported more than 40 years ago 6 and present in 90%-96% of patients, 53,54 making it more prevalent than the characteristic motor symptoms and prompting some to refer to the disease as mainly an olfactory disorder. 55he lack of appropriate models constrains olfactory pathology study to post-mortem Parkinsonian samples or prohibitively expensive, and often inappropriate, animal studies.This work aimed to assess the general effects of the catecholaldehyde metabolite of dopamine, DOPAL, on olfactory bulb organotypic and primary cultures to recapitulate aspects of OB pathology in PD.
3D constructs to model PD provide an exciting and more reliable platform to study the disorder compared to 2D cell culture systems, or too costly animal studies. 17fter optimizing culture conditions for postnatal OB slices, we were able to detect neurodegeneration and cell death in tissue treated with as little as 50 μM of DOPAL, a concentration higher than the reported physiological values of 2-3 μM, but possibly not too far from the levels of DOPAL in Parkinsonian brains, where the exposure is sustained for years and not days. 56The dopamine metabolite's toxicity has already been proven in various in vitro models 27,[56][57][58][59][60] and even assessed in vivo, 29,61 but the focus to date has mainly been on neuronal cells, particularly the midbrain dopaminergic population.Therefore, ours is the first study of the effect of DOPAL in olfactory bulb cultures, and its relevance lies in the following considerations.The proximity to the potentially harmful world makes the OB a possible entry site for many environmental toxins and pathogens that can make their way to deeper brain regions. 50Intriguingly, some of these compounds, such as the fungicide benomyl, are known to induce Parkinsonian symptoms in people chronically exposed to them.Even more compelling is that some of the chemicals have been shown to inhibit ALDH potently, a critical enzyme in DOPAL detoxification, thus, linking the environmental risk factors for PD with the catecholaldehyde hypothesis.
Not only "nature" and "nurture" seem to be connected with DOPAL, as the metabolite could influence the transcriptome of treated slices.The gene encoding for the amyloid-beta precursor protein was prominently expressed upon treatment.The protein, involved in the pathogenesis of Alzheimer's disease, has been recently proved to be a target of phosphorylation of the leucine-rich repeat kinase 2 (LRRK2).Mutations in the LRRK2 gene (or PARK8) are the main contributing factor in the genetic development of PD, and the enhanced kinase activity on APP was proven to increase toxicity in animal models, proof of the connection between apparently different diseases. 62][65] Particularly impaired by catecholaldehyde exposure was the ubiquitination pathway, fundamental in controlling proteostasis.[27][28]66,67 However, the evidence of α-synuclein aggregation in our model was far from being convincing, possibly due to species-specific differences between rodent and human, already reported in the case of dopamine-modified synuclein, detectable only in human and not rodent model of PD. 68 While it has been proven that DOPAL-modified α-synuclein oligomers are unable to form fibrils and aggregate in Lewy bodies, 25 the downregulation of the ubiquitination pathway induced by the catecholaldehyde could instead play a role in the formation of protein inclusions.In contrast, the gene encoding the deubiquitinating enzyme ataxin 3 was highly upregulated, pointing again at the disruption of the physiological proteasome and cellular homeostasis.When subjected to polyglutamine expansion, the protein causes the progressive ataxic disorder known as Machado-Joseph disease, which has been shown to deubiquitinate the E3 ubiquitin-ligase enzyme Parkin (PARK2). 69Interestingly, the expanded form of ataxin 3 increases Parkin's turnover, leading to a decreased expression of the ubiquitin-ligase enzyme, whose loss of function mutations cause early-onset forms of PD. 65 Intriguingly, Parkin's expression, together with those of two other PARK genes, pivotal players in mitochondrial quality control and oxidative stress regulation systems, (PARK6) and DJ-1 (PARK7), was heavily downregulated following DOPAL exposure.The reactive metabolite, prone to ROS generation, combined with a defective antioxidant defense pathway, caused a dramatic increase in oxidative and nitrative stress levels in the cultures.Exceptionally remarkable was the oxidation of lipids and proteins causing a progressive browning of the tissue slices, similar to what was reported by Jinsmaa and co-workers in their study of DOPALquinone formation. 70Oxidative stress is profoundly related to mitochondrial dysfunction, a hallmark of neurodegeneration and Parkinson's disease.DOPAL also has a close association with the organelle since it is produced via the oxidative deamination of dopamine by the MAO enzyme, located on the outer mitochondrial membrane. 71For these reasons, an in-depth analysis was devoted to the study of mitochondrial and bioenergetic function in cells exposed to DOPAL.Collectively, the results show a profound decrease in mitochondrial health, appearing within minutes after catecholaldehyde exposure.In particular, the electron transport chain, responsible for maintaining a proton gradient across the membrane, showed a reduced efficiency, in line with the deficiency of complex I reported in patients, and not only confined to the SN. 72,73The disruption of several pathways fundamental for mitochondrial homeostasis was also detected with gene expression analysis, confirming what was reported using other techniques.Additionally, the organelles are pivotal players in regulating cell death and survival, and the apoptotic signaling was also importantly different in control vs experimental slices.
In summary, DOPAL accumulation and reactivity can link many aspects of PD; thus, our data strongly corroborate and consolidate the catecholaldehyde hypothesis, providing at the same time, a useful platform to study prodromal pathology and test novel therapeutic approaches in Parkinson's disease.

| CONCLUSIONS
After more than 200 years from the first clinical description of Parkinson's disease, the cause(s) of this progressive debilitating disorder remain still unclear.Additionally, the lack of appropriate models to study the etiology and pathogenesis of the disease hinders the development of new disease-modifying treatments, while the ones available to patients are merely symptomatic and did not drastically change over the course of half a century.
Here, we propose a novel organotypic model of Parkinson's disease pathology in the olfactory bulb, one of the areas affected at the earliest stages of the disease.To induce the pathology, 3D slices were treated with 3,4-dihydroxyphenylacetaldehyde (DOPAL), the MAO metabolite of dopamine.Our data confirm the ability of DOPAL to recapitulate many aspects of the disease, and, for the first time, report its effect on gene expression.Cell death, oxidative and nitrative stress, mitochondrial depolarization, and dysfunction were all reported in our model, further supporting the catecholaldehyde hypothesis and the involvement of dopamine and its metabolism in the progression of Parkinson's disease.
The ex vivo model created can provide a useful tool to test potential treatments, maintaining the complex multidimensional structure of the olfactory bulb but in an accessible and controlled environment.
In conclusion, this work corroborates previous evidence linking DOPAL and Parkinson's disease, a link that was first explored due to the connection between pesticides, potent inhibitors of ALDH, and the neurodegenerative disorder.Additionally, this work provides new insight into the effect of the aldehyde on the expression of many mitochondria-and Parkinson's-related genes.

F I G U R E 1
Experimental paradigm followed.(A) Organotypic slices or OB cells were obtained by slicing or enzymatic dissociation of olfactory bulb from P10 animals.(B) OB slices maintain the typical six-layered architecture in culture for long periods.The zoomed section shows an organotypic slice at DIV1 (left) and H&E staining of a P10 OB (right).All the layers are preserved in the organotypic slice.Scale bar = 1000 μm, GL, glomerular layer; EPL, external plexiform layer; MCL, mitral cell layer; IPL, internal plexiform layer; GCL, granule cell layer; RMS, rostral migratory stream.

F I G U R E 3
Neurodegeneration in OB slices and cells exposed to DOPAL.(A) After 3 days of DOPAL treatment, an important drop in the metabolic activity of OB slices was reported by alamarBlue® assay.(B) No recovery was observed in slices cultured for other 4 DIV without DOPAL.(C) Loss of protein content from DOPAL-treated slices after 14 DIV.(D) Loss of DNA in OB cells treated for 4 h with DOPAL, as assessed by Picogreen assay.(G) Representative immunohistochemistry images of neurons, identified by β-tubIII staining, after a 4-h exposure to different concentrations of DOPAL.In the second row, a higher magnification image of the boxed regions.Data are expressed as mean ± SEM of at least three independent experiments, one-way ANOVA with a Tukey's multiple comparison test or Kruskal-Wallis with Dunn's multiple comparison test (*p < .05,**p < .01,***p < .001,****p < .0001)# vs. 0 μm, * vs. vehicle.Scale bar = 100 μm.F I G U R E 4 Oxidative and nitrative stress in OB cultures following DOPAL exposure.DOPAL treatment raised the levels of reactive oxygen and nitrative species in both organotypic and primary cultures.(A) In cells, oxidative stress was evaluated using the CellROX dye, a probe that become fluorescent under oxidative conditions.A clear increase in the CellROX fluorescence was detected after treatment with 10 μM and 100 μM DOPAL.Co-staining with the neuronal marker β-tubulin III confirms a widespread oxidative state, not limited to neurons.(B) In organotypic slices, a browning of the tissue, compatible with oxidation of protein and lipids, was detected in DOPAL-treated samples.The color change was concentration dependent, as clearly demonstrated in bright field images of the slices and of the lipid deposits remaining after protein extraction (F).(C) The levels of nitrites and (D) hydrogen peroxide in the culture supernatant were found to increase with DOPAL concentration.While not statistically different, increased nitrites at lower doses of DOPAL became significant at doses of 250 μM and above, with fold-increases ranging from 8.43 to 23.82.Levels of H 2 O 2 secreted by slices remained approximately constant until slices were exposed to DOPAL concentrations greater than 250 μM.This effect can be explained by a progressive reduction in the antioxidant enzyme catalase activity, that is, this activity appeared to start decreasing at 250 μM DOPAL, becoming significantly lower for the highest concentration of 1000 μM (e).Scale bar = 100 μm in (a), 5 mm and 1000 μm in (b) for images in LH or RH columns, respectively.Mean ± SEM of N = 3 independent experiments, n = 3 biological replicates.Statistical significance was assessed using one-way ANOVA with Tukey's multiple comparison test.When both * and # are present, # is vs vehicle, * vs control (0 μM).