Dopamine oxidation products as mitochondrial endotoxins, a potential

Jun 15, 2018 - Alice Biosa , Irene Arduini , Maria Eugenia Soriano , Valentina Giorgio , Paolo Bernardi , Marco Bisaglia , and Luigi Bubacco. ACS Chem...
0 downloads 0 Views 3MB Size
Subscriber access provided by STEPHEN F AUSTIN STATE UNIV

Article

Dopamine oxidation products as mitochondrial endotoxins, a potential molecular mechanism for preferential neurodegeneration in Parkinson Disease Alice Biosa, Irene Arduini, Maria Eugenia Soriano, Valentina Giorgio, Paolo Bernardi, Marco Bisaglia, and Luigi Bubacco ACS Chem. Neurosci., Just Accepted Manuscript • DOI: 10.1021/acschemneuro.8b00276 • Publication Date (Web): 15 Jun 2018 Downloaded from http://pubs.acs.org on June 16, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Neuroscience

Dopamine oxidation products as mitochondrial endotoxins, a potential molecular mechanism for preferential neurodegeneration in Parkinson Disease

Alice Biosa†,┴, Irene Arduini†,┴,§, Maria Eugenia Soriano†,‡, Valentina Giorgio‡, Paolo Bernardi‡, Marco Bisaglia†,* and Luigi Bubacco†,*

From the



Department of Biology, University of Padova, Italy,



Department of

Biomedical Sciences, University of Padova, Italy

┴ §

*

These authors contributed equally to this work

Current address: Neurobiology Department, University of Pittsburgh, Pittsburgh, PA To whom correspondence should be addressed:

Marco Bisaglia, Department of Biology, University of Padova, 35121 Padova, Italy. Tel: +39 0498276329; e-mail: [email protected]; Luigi Bubacco, Department of Biology, University of Padova, 35121 Padova, Italy. Tel: +390498276346; e-mail: [email protected]

1

ACS Paragon Plus Environment

ACS Chemical Neuroscience 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Abstract The preferential degeneration of dopaminergic neurons in the substantia nigra pars compacta is responsible for the motor impairment associated with Parkinson’s disease. Dopamine is a highly reactive molecule, which is usually stored inside synaptic vesicles where it is stabilized by the ambient low pH. However, free cytosolic dopamine can autooxidize, generating reactive oxygen species, and lead to the formation of toxic quinones. In the present work, we have analyzed the mechanisms through which the dysfunction of dopamine homeostasis could induce cell toxicity, by focusing in particular on the damage induced by dopamine oxidation products at the mitochondrial level. Our results indicate that dopamine derivatives affect mitochondrial morphology and induce mitochondrial membrane depolarization, leading to a reduction of ATP synthesis. Moreover, our results suggest that opening of the mitochondrial transition pore induced by dopamine-derived quinones may contribute to the specific Parkinson’s disease-associated vulnerability of dopamine containing neurons.

Keywords: ATP synthase, dopamine, dopamine-derived quinones, mitochondria, Parkinson’s disease, permeability transition pore

2

ACS Paragon Plus Environment

Page 2 of 30

Page 3 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Neuroscience

Introduction In spite of their relatively low number 1, dopaminergic neurons play a critical role in several mental and physical functions. They are organized in four main pathways named mesolimbic, mesocortical, nigrostriatal and tuberoinfundibular, respectively. They take part in the control of cognition, learning, motivation, movement and hormone release 2. Considering the numerous processes that involve dopaminergic neurons, it is not surprising that degeneration of this class of neuronal cells is implicated in several neurological disorders. The preferential degeneration of dopaminergic neurons in the substantia nigra pars compacta (SN) represents one of the pathological hallmarks of Parkinson’s disease (PD), a chronic and progressive neurodegenerative disorder clinically characterized by motor symptoms such as tremor at rest, rigidity, bradykinesia, and postural instability 3. SN neurons reside in the midbrain and project axons rostrally to the forebrain where they release dopamine (DA) into the striatum. In the striatum, released DA is critical for the initiation and coordination of movements. The reason why SN neurons preferentially degenerate during PD is still unclear. However, several hallmarks of the disease provide clues to the underlying pathogenesis. Among them, mitochondrial dysfunction has been reported long ago, and consistently found in the SN of PD patients 4

. A possible explanation for the high vulnerability of SN dopaminergic neurons

observed in PD is the very presence of DA 2. DA is a highly reactive molecule, normally stored in synaptic vesicles where the acidic environment prevents oxidation. However, DA is synthesized and metabolized in the cytosol where it can undergo spontaneous oxidation. DA oxidation can be strongly accelerated by metal ions such as Fe3+, while it can be inhibited by several antioxidant molecules, such as glutathione. In the frame of PD, it is worth mentioning that in the SN of patients affected by the disease significant iron accumulation with specific decrease in glutathione levels has been reported 5-10.. DA oxidation leads to formation of both reactive oxygen species (ROS) and dopaminequinones (DAQs), namely 3,4-dopamine quinone, aminochrome and indole-5,6-quinone 11

. ROS can damage cellular components such as lipids, proteins, and DNA

12

. The

electron-deficient quinones can also react with, and covalently modify cellular

3

ACS Paragon Plus Environment

ACS Chemical Neuroscience 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

nucleophiles, such as cysteine residues, leading to further cytotoxicity 13, 14. Many proteins have been described as targets of DAQs, including tyrosine hydroxylase, the rate-limiting enzyme in DA synthesis; dopamine transporter, which is involved in DA re-uptake after its release in the synaptic cleft; and parkin, a protein involved in a familial form of PD 15-17. Within this framework, we have characterized the structural modifications and functional effects induced by DAQs on DJ-1, a protein involved in another familial form of PD 18; on α-synuclein, one the major component of the proteinaceous inclusions found in PD

19-21

; and on SOD2, an antioxidant enzyme

which catalyzes the conversion of superoxide anions into molecular oxygen and hydrogen peroxide

22

. Interestingly, in isolated mitochondria it has been shown that DA

oxidation products can interfere with mitochondrial function, inducing for example the swelling of the organelles and decreasing electron transport chain activity 23-26. In the present work, we investigated the functional consequences of DAQs in isolated mitochondria and in the human dopaminergic neuroblastoma SH-SY5Y cell line, with the aim of understanding the molecular pathways connecting DAQs accumulation and mitochondrial dysfunction.

4

ACS Paragon Plus Environment

Page 4 of 30

Page 5 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Neuroscience

Results DAQs accumulation induces cell toxicity. The first set of experiments was aimed at assessing whether the accumulation of DAQs affects cell viability. To this end human neuroblastoma SH-SY5Y cells were incubated for 24 hours with different amounts of DAQs and apoptotic nuclei were then counted through fluorescence microscopy. DAQs were generated from DA by the action of the enzyme tyrosinase (Ty), which does not generate radical species in solution

27

. As represented in Figure 1A-B, the presence of

DAQs induced nuclear fragmentation in a dose-dependent manner. The results were then independently confirmed through flow cytometry, using propidium iodide (PI) to evaluate cell death. PI is a membrane-impermeant dye that is generally excluded from viable cells while it can easily penetrate the damaged, permeable membranes of nonviable cells. When cells were treated with DAQs the cellular population having a higher PI fluorescent signal increased in comparison to untreated cells (Figure 1C) confirming that the accumulation of DAQs promotes cell death. We also verified that the treatment with Ty alone did not induce any toxicity (data not shown). As DAQs were produced in the extracellular medium, we then assessed whether they were able to enter cells and to interact with intracellular proteins. Cell lysates from untreated, DA- or DAQs-treaded cells were transferred to nitrocellulose membrane and stained with Ponceau red and nitro blue tetrazolium, a quino-specific dye

28

. Cytoplasmic proteins from untreated or DA-

treated cells were stained only by Ponceau red. On the other hand, cells incubated for 1 hour in the presence of DAQs displayed bluish-purple protein bands, indicating the formation of DAQs-modified proteins (Figure 1D), thus confirming the intracellular presence of DAQs. It is worth mentioning that neither DAQs alone, nor DAQs formed by the reaction of DA with Ty in the absence of cells (1 hour reaction) and then mixed with cell lysates (Cntr#1 and Cntr#2, respectively, in Figure 1D) gave rise to purple bands indicating that such a coloration in the lane corresponding to cells treated with DAQs is not due to a background color caused by the DAQs itself. DAQs affect mitochondrial morphology. As most of the experiments involving DAQs and mitochondria previously described in the literature were carried out in isolated mitochondria, we explored whether mitochondria could be targets of DAQs within the

5

ACS Paragon Plus Environment

ACS Chemical Neuroscience 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 30

cell. As experimental evidence now exists that mitochondrial morphology plays a crucial role in cell physiology, and changes in mitochondrial shape have been linked to neurodegeneration, calcium signaling, lifespan and cell death

29

, we focused on

mitochondrial morphology. While mitochondrial fragmentation is generally associated with metabolic dysfunction and disease, a hyperfused network is thought to counteract metabolic insults and contribute to the preservation of cellular integrity 30. As reported in Figure 2A-B, when SH-SY5Ycells were treated for 1 hour with DAQs the elongated tubular morphology observed in the control became less prominent, with the appearance of fragmented and doughnut-shaped mitochondria, an effect that was not seen after treatment with DA alone, and that does not depend on the presence of Ty (data not shown). Quantification through ImageJ plug-in described by Bondi et al

31

allowed to

measure mitochondrial maximum length, perimeter, area and circularity (which takes into account the compactness of the particles), and the results indicate that, in contrast to the treatment with DA, in the presence of DAQs a prominent increase in the percentage of smaller mitochondria took place (Figure 2C). DAQs impair mitochondrial ATP synthesis. To test whether the reaction with DAQs affects mitochondrial function, we analyzed, by nuclear magnetic resonance (NMR), ATP synthesis in isolated respiring mouse liver mitochondrial suspensions. The opportunity to directly observe ADP and ATP signals instead of oxygen consumption is of particular interest in the context of the present work because it does not require the elimination of Ty (which uses molecular oxygen as co-substrate to generate quinones). To increase oxygen availability and extend the duration of the experiment pure molecular oxygen was bubbled inside the NMR tube before the addition of mitochondria, and the pseudo-two-dimensional spectrum of a 2.5 mM ADP solution was recorded after the addition of 3 mg of mitochondria (Figure 3A). Concomitant to the disappearance of the peaks relative to ADP, three new peaks became visible, corresponding to the newly synthesized ATP. In the experimental conditions used, the ADP to ATP conversion was completed after approximately 30 min. A similar behaviour was observed in the presence of 150 µM DA (Figure 3B). In contrast, when DAQs were produced in solution using 50 µM DA plus Ty, the reactivity became quite different (Figure 3C). ATP formation was still evident, but ADP was not fully phosphorylated even after 1 hour, suggesting an

6

ACS Paragon Plus Environment

Page 7 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Neuroscience

inhibitory effect of DAQs in some steps of oxidative phosphorylation. In agreement with this hypothesis, increasing DA to 150 µM (still in the presence of Ty) completely inhibited the phosphorylation of ADP to ATP (Figure 3D). DAQs induce the permeability transition. A potential way through which DAQs can induce cell death is by inducing the permeability transition (PT), as previously reported

23

. This phenomenon consists in the opening of a proteinaceous pore (the

mitochondrial permeability transition pore, mPTP) that forms under certain conditions (oxidative stress is one of them) in the inner mitochondrial membrane making mitochondria permeable to solutes of MW up to about 1500 Da

32

. The opening of the

pore and the release of the proapoptotic factors that reside in the intermembrane space are considered one of the possible pathways responsible for triggering cell death

32

. To test

this prediction in a cellular context, we investigated the effects of DA oxidation products on the mitochondrial membrane potential in the human dopaminergic SH-SY5Y cell line. Mitochondrial membrane potential was measured using tetramethylrhodamine methyl ester (TMRM) a fluorescent cationic dye that is accumulated into the mitochondrial matrix proportionally to the membrane potential. A significant depolarization of mitochondria was observed after incubation in the presence of DAQs (Figure 4A) demonstrating, once again, that they interfere with mitochondrial function. To test whether opening of mPTP contributed to the depolarizing effect, cells were treated with cyclosporine A (CsA), a powerful inhibitor of the PT

32

, before exposure to DAQs.

Consistent with mPTP opening as the underlying cause of depolarization, CsA was able to protect mitochondria from depolarization. The ability of DAQs to promote the PT under the conditions of our study was further confirmed by swelling experiments with isolated mouse liver mitochondria (Figure 4B). Treatment with DAQs, but not with DA alone, induced mitochondrial swelling in a dose-dependent manner. This effect was fully abolished by CsA, demonstrating that mPTP opening is consequence of the DAQs treatment. These results support the notion that the oxidation products of DA can induce opening of mPTP. DAQs permeate mitochondria and covalently modify mitochondrial proteins. The last set of experiments was aimed at testing the ability of DAQs to permeate mitochondria. Rat brain mitochondria were incubated with radiolabeled

7

ACS Paragon Plus Environment

14

C-DA in the

ACS Chemical Neuroscience 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

presence or absence of Ty. We observed that only when DAQs were formed radioactivity was retained inside the mitochondria, most probably because of the reactivity of DAQs toward mitochondrial functional groups (Figure 5A). To test this point, we analyzed the reaction between mitochondrial proteins and DAQs. Purified mitochondria were exposed to radiolabeled

14

C-DA or

14

C-DA plus Ty to generate DAQs. Mitochondrial proteins

were then separated by gel electrophoresis and the lanes analyzed by scintillation after slicing. While radioactivity was not observed in DA-treated mitochondria, it was detected in the sample exposed to DAQs (Figure 5B). The reaction of DAQs with mitochondrial targets was further investigated by autoradiography after SDS-PAGE separation. As indicated in Figure 5C, DAQs did bind several mitochondrial proteins, and two bands slightly above the 30 kDa showed the strongest labeling. To identify the corresponding proteins, bands were excised and analyzed by MALDI-TOF mass spectrometry after ingel trypsin digestion. While the upper band could not be identified unambiguously, the lower band was identified as the γ-subunit of the F1 catalytic domain of the ATP synthase complex by the presence of its ATLKDITRXL peptide. In conclusion, our results indicate that DAQs are able to enter mitochondria and to react with several proteins. Among them, the γ-subunit of ATP synthase has been specifically identified. It is worth mentioning that this set of data on ATP synthase modification by DAQs confirms a previously published work 33.

8

ACS Paragon Plus Environment

Page 8 of 30

Page 9 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Neuroscience

Discussion In the present work we focused on the cellular and mitochondrial effects induced by the cytosolic accumulation of dopamine-derived quinones. The cellular toxicity related to DA oxidation has been already emphasized elsewhere 21, 31, 34, 35. Nevertheless, in the vast majority of these previous studies the authors did not discriminate between the effects due to the formation of ROS and those derived from DAQs accumulation. Even though we acknowledge that the production of ROS can contribute to render dopaminergic cells more vulnerable than other neuronal populations to oxidative insults, in this study we focused our analysis on DA-related toxicity, which specifically depends on the formation of DAQs. The underlying rationale is that DAQs could contribute to the preferential loss of dopaminergic neurons observed in PD. Actually, while ROS production is a general process that may occur in every type of cell, DAQs formation is a unique feature of dopaminergic neurons. To investigate their selective untoward effects, DAQs were produced using the enzyme Ty. The use of Ty to rapidly produce DAQs offers the advantage of preventing the formation of radical species in solution. Indeed, Ty uses molecular oxygen to catalyze the hydroxylation of monophenols to diphenols (monophenolase activity) and the oxidation of diphenols to quinones (diphenolase activity). Both reactions take place without the production of radicals, using molecular oxygen as co-substrate

27

. Using this strategy, we were able to focus exclusively on

DAQs-related toxicity. Our data indicate that the accumulation of DAQs induces cell death. Moreover, the presence of DAQs also affects mitochondrial morphology. Specifically, we observed that, following the treatment with DAQs, there is a decrease in the number of tubular elongated mitochondria. As the mitochondria shape can be directly associated with their function, to further correlate the observed morphological modifications with mitochondrial functional impairment, we verified whether ATP synthesis is affected following DAQs exposure. To this aim,

31

P-NMR spectroscopy has been considered the

technique of choice because it can provide nondestructive information about the physical states of nucleotides in complex mixtures. The opportunity to directly observe ADP and ATP signals instead of oxygen consumption is of particular interest in the context of the

9

ACS Paragon Plus Environment

ACS Chemical Neuroscience 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 30

present work because it does not require the elimination of Ty, a necessary component to single out the effects of DAQs. Our results, coherently with what stated above, indicate a strong effect of DAQs in the modulation of the ADP to ATP conversion. The impairment of the ATP synthesis can depend on several factors such as inhibition in some steps of oxidative phosphorylation. Inhibition of complex I and IV has actually been reported after the reaction with DAQs

23, 26

. An additional pathway

contributing to a decline in ATP production comes from the ability of DAQs to induce the PT, as already reported

23, 25

. It is worth mentioning that, since PT renders the inner

mitochondrial membrane permeable to solutes smaller than 1500 Da, it leads to a dissipation of the proton gradient across the inner membrane, thus explaining the impairment in ATP synthesis. While the effects of DAQs on PT were previously described in isolated mitochondria

13a, 13c

, in this study they have been confirmed and

extended to a cellular model. Both sets of data support the notion that pore opening induced by DAQs can participate in the specific PD-associated vulnerability of DAcontaining neurons. In this frame, results obtained in the present study are particularly interesting. Indeed, we found that the oxidized products of DA, can enter the mitochondria and react with several proteins. DAQs have many targets within mitochondria 33, but our analysis has identified the more reactive ones by decreasing the amount of DA to a level allowing detection of a small number of bands by autoradiography. Through this analysis and mass spectrometry one band was assigned to the γ subunit of the F1 catalytic domain of the ATP synthase complex. The mitochondrial ATP synthase consists of two main functional domains joined by central and peripheral stalks. The membrane-embedded Fo domain uses the energy associated to the proton electrochemical gradient across the mitochondrial inner membrane to generate mechanical rotation, which is transmitted through the central stalk to the matrix-exposed F1 catalytic domain. Together with the δ and ε subunits, the γ subunit is part of the stalk, which occupies the central axis of the more globular region of the F1 domain from where it protrudes approximately 30 Å and binds firmly the Fo domain. The analysis of the ATP synthase structure reveals that the γ subunit presents only one well-conserved Cys residue facing the mitochondrial matrix (Cys78) in close proximity to the catalytic α and β subunits. Based on the well-known 10

ACS Paragon Plus Environment

Page 11 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Neuroscience

reactivity of DAQs toward free thiols in exposed Cys residues, it is reasonable to consider Cys78 as the most likely candidate for the DAQ-modification of the γ subunit of the ATP synthase complex, which would be consistent with inhibition of ATP synthesis. It has recently been proposed that ATP synthase constitutes a central component of the mPTP

36-38

. The potential pore-forming sites of the mitochondrial F1Fo ATP

synthase are still under debate and a conclusive picture is still lacking

39, 40

. Two main

hypotheses have been proposed. In the first, the pore forms at the monomer-monomer interface of ATP synthase dimers

36

, while in the second one it is constituted by the

membrane embedded c-ring portion of the Fo subunit 37, 38. In the light of recent studies 41, 42

we think that reaction of DAQs with Cys78 may be involved more in inhibition of ATP

synthesis than with a direct effect on the PTP, which could be due to oxidation of less reactive cysteine residues on the peripheral stalk, an issue that is the focus of active research in our laboratories. It should be mentioned that formation of an inter-subunit disulfide bridge between Cys78 of subunit γ and Cys251 of subunit α has been observed in canine dyssynchronous heart failure

43

, suggesting that modification of this residue

may indeed directly impinge on enzyme catalysis. In summary, we propose that in dopaminergic neurons any event leading to an increased intercellular concentration of DA may result in a proportional increase in DAQs production. These can then be transferred to the mitochondrial matrix, and concur to move the threshold for mPTP opening closer to the resting potential. These mechanisms may reduce ATP production and increase the production of ROS, and contribute to the preferential vulnerability of dopaminergic neurons in PD.

11

ACS Paragon Plus Environment

ACS Chemical Neuroscience 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Methods Cell culture. Human neuroblastoma SH-SY5Y cells (IST, Genova, Italy) were cultured in a 1:1 mixture of Ham's F12 and Dulbecco Modified Eagle Medium (Life Technologies) supplemented with 10% fetal bovine serum, in a 5% CO2 humidified incubator at 37 °C. Hoechst 33342 staining. SH-SY5Y cells were plated on fibronectin-coated coverslips in 24-well plates (50,000). When cells reached ~60% of confluence, they were treated with 0-150 µM DA and Ty (100 units) for 24 hours. Cells were then washed, fixed with 4% paraformaldehyde and stained with Hoechst 33342. Images were recorded with a Leica 5000B epifluorescent microscope with the 100X oil objective. Flow cytometry. SH-SY5Y cells were cultured on 6-well plates and then treated with 50, 100 and 150 µM DA and Ty in Hanks' balanced salt solution (HBSS) without bicarbonate for 1 hour at 37 °C. Then HBSS medium was removed and complete media was replaced in each well. After 24 hours, cells were detached by using papain protease (Worthington), centrifuged at 140 x g for 5 min together with eventual cell debris present in the media. After two washes in FACS buffer (phosphate buffer saline with 1% BSA), cells were resuspended in 190 µl of binding buffer (10 mM HEPES pH 7.4, 140 mM NaCl, 2.5 mM CaCl2 and 1 g/L glucose). Cell suspensions were then transferred into 5 ml round-bottom tubes and 10 µl of propidium iodide (20 µg/ml) were added and incubated for a few seconds at room temperature. Samples were analyzed on Canto II flow cytometer (BD Bioscience) and 5,000 gated events were acquired. Nitro blue tetrazolium/glycinate redox cycling staining. SH-SY5Y cells were plated almost confluent in 6-well plates and treated with either 150 µM DA and Ty (600 units) or Ty or DA alone for 1h at room temperature in 700 µl of HBSS. After 2 washes in phosphate-buffered saline (PBS), cells were detached and pelleted at 600 g. Cells were lysed in 50 µl of lysis buffer (9 M urea, 1% triton X-100, 30 mM Tris, 150 mM NaCl, pH 7.5, with phosphatases and proteases inhibitors) for 30 min on ice. Insoluble material was resuspended by 2 cycles of sonication at 20 kHz for 5 seconds. Insoluble fraction and cell debris were pelleted at 14000 g for 30 min at 4 °C. The supernatant was collected and used to determine protein concentration by bicinchoninic acid (BCA) assay. 50 µg of proteins were loaded into SDS-polyacrylamide gels and proteins transferred into the

12

ACS Paragon Plus Environment

Page 12 of 30

Page 13 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Neuroscience

polyvinylidene difluoride (PVDF) membrane. The membrane was first stained with Ponceau S (0,1% Ponceau S in 5% acetic acid) and then DAQ-modified proteins were detected by incubating the PVDF membrane in nitro blue tetrazolium staining solution (0.24 mM nitro blue tetrazolium, 2 M potassium glycinate, pH 10) for 45 min in the dark. The reaction was stopped with 0.16 M sodium borate buffer, pH 10. Mitochondrial morphology analysis. SH-SY5Y cells were plated on fibronectin-coated coverslips in 24-well plates (90,000 cells/well) and 24 hours later transfected with 0.25 µg of mito-RFP vector using Lipofectamine (Life Technologies) as transfection reagent according to manufacturer’s instructions. The day after, cells were incubated in HBSS without bicarbonate in presence of 150 µM DA and 100 units of Ty for 1 hour at room temperature. Cells were then washed, fixed with 4% paraformaldehyde and nuclei were counterstained using Hoechst. Mitochondrial morphology was assayed by means of Leica 5000B epifluorescent microscope with the 100X oil objective. Data analysis was performed on mitoRFP positive cells, in a blind manner and reported as percentage of cells with tubular, intermediate or fragmented morphology. Eighty cells per replicate were analyzed and 4 independent experiments were performed. Data were analyzed by means of paired t-test. Mitochondrial morphology was also analyzed using the ImageJ macro developed by Bondi et al. 31. Images were acquired using a ZeissLSM700 confocal microscope with 100X oil objective at 740x740 pixel resolution. For the assay, fields with one mitochondria per image were analyzed. Organelles at the borders of the image were excluded from the analysis. Twelve images were taken from 4 independent replicates. Data analysis was performed by paired t-test. Mouse liver mitochondria isolation. Mitochondria were isolated from CD1 mice by a standard procedure as described in Costantini et al.

44

with some modifications. Briefly,

the liver was minced in a buffer containing 250 mM sucrose, 10 mM Tris-HCl, pH 7.4 and 0.1 mM EGTA. The minced liver was homogenized in the same buffer and centrifuged at 700 x g for 6 min. The supernatant was centrifuged at 7000 x g for 6 min. The resulting mitochondrial pellet was resuspended and recentrifuged at 7000 x g for 6 min. The entire procedure was carried out at 4 °C. Mitochondrial protein concentration was determined by the biuret reaction with BSA as a standard.

13

ACS Paragon Plus Environment

ACS Chemical Neuroscience 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

NMR experiments.

31

P-NMR experiments were performed on a Bruker DRX400

spectrometer operating at 161.93 MHz and equipped with a 10-mm diameter probe. Spectra were accumulated at 21 °C in 5 min blocks using a recovery time of 3 s and a spectral width of 9800 Hz. Mouse liver mitochondria (~100 µg/µl) were stored on ice prior to experimentation. Before starting the experiments 2 ml of reaction medium containing 10 % D2O, 10 mM MOPS pH 7.4, 250 mM sucrose, 5 mM Pi and 10 µM EGTA, 20 mM succinate were transferred into the NMR tube and the medium was bubbled 15 min with pure O2 to prolong respiratory activity. In the reference experiments, after recording the spectrum of 2.5 mM ADP alone, 3 mg of mitochondria were added to the solution and the reaction was followed by recording a series of onedimensional spectra. An initial delay of 2.5 min was necessary for probe tuning and field shimming. When DAQs effects on ADP consumption were analyzed, 50-150 μM of DA were first added to the reaction medium in the presence of 100-375 units of Tyand the reaction was carried out for 5 min before the addition of 3 mg of mitochondria. After 1 min, 2.5 mM ADP was added to the solution and the reaction was followed as described above. An experiment in the presence of 375 units of Tywas recorded as control. Mitochondrial swelling. Mitochondria swelling was followed as the change of light scattering of the mitochondrial suspension at 620 nm on a personal computer-interfaced diode array Agilent 8453 UV-visible spectrophotometer, equipped with magnetic stirring and thermostatic control. Mitochondria from mouse liver (0.5 mg/ml) were incubated in 2 ml of medium containing 250 mM sucrose, 10 mM Tris-MOPS, 1 mM Pi, 10 µM EGTA, 5 mM glutamate/2.5 mM malate or 5 mM succinate plus 2 µM rotenone. An amount of Ca2+ under the threshold necessary to trigger permeability transition (PT) was added to accelerate PT by subsequent treatments. A minute after, 10-100 μM DA and 120 units Tywere added to generate DAQs. When indicated CsA (0.8 µM) was added in the incubation medium before addition of mitochondria. TMRM analysis. SH-SY5Y cells (100,000) were seeded onto 24-mm diameter round glass coverslips in 6-well plates and grown for 1 day. The coverslips were then transferred onto the stage of a Zeiss Axiovert 100TV inverted microscope equipped with a HBO mercury lamp (100 watts), and epifluorescence was detected with a 12-bit digital

14

ACS Paragon Plus Environment

Page 14 of 30

Page 15 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Neuroscience

cooled CCD camera (Micromax, Princeton Instruments). Cells were incubated in HBSS without bicarbonate in presence of 150 µM DA and Ty(100 units). After 10 min, cells were washed and incubated in HBSS in presence of 20 nM TMRM + 1.8 µM CsH for 30 min followed by equilibration in the dark for 10 min. When indicated, 1.8 µM of CsA was added 30 min before any treatment. Oligomycin, which blocks the flow of protons through the Fo subunit of ATP-synthase, was added in every experiment to unmask respiratory chain dysfunction as previously described

45

. Fluorescence images were

acquired with a 560 nm dichroic mirror using a 40x/1.3 oil immersion objective (Zeiss), with excitation at 546 ± 5 nm and emission at 580 ± 15 nm. Exposure time was 80 ms, and data were acquired and analyzed with the MetaMorph Metafluour Imaging Software. Clusters of several mitochondria were identified as regions of interest, whereas background was taken from fields not containing cells. Sequential digital images were acquired every 2 min for 60 min, and the average fluorescence intensity of all the regions of interest and of the background was recorded and stored for subsequent analysis. Mitochondrial fluorescence intensities minus background were normalized to the initial fluorescence for comparative purposes. Rat brain mitochondria isolation. Cerebral cortices of two 6–7 week-old male SpragueDawley rats were rapidly removed into 20 ml of ice-cold isolation medium (320 mM sucrose, 5 mM MOPS, and 0.05 mM EGTA, pH 7.4) and homogenized. The homogenate was centrifuged at 900 x g for 5 min at 4 °C. The supernatant was centrifuged at 8500 x g for 10 min, and the resulting pellet was re-suspended in 1 ml of isolation medium. This was layered on a discontinuous gradient consisting of 4 ml of 6% Ficoll, 1.5 ml of 9% Ficoll, and 4 ml of 12% Ficoll (all prepared in isolation medium) and centrifuged at 75,000 x g for 30 min. The synaptosome-free mitochondrial pellet was re-suspended in 250 mM sucrose and 10 mM K-MOPS, pH 7.2, and centrifuged at 8500 x g for 15 min before being re-suspended in the same medium at a final concentration of 10-20 mg of protein/ml as determined by the biuret reaction with BSA as a standard. DAQs modification of mitochondrial proteins. Purified mitochondria from rat brain (0.5 mg) were incubated with 10 µM

14

C-DA (1 µCi) in the presence or absence of

tyrosinase (100 U) for 10 minutes. After the treatment, mitochondria were filtered through silicone and washed three times with fresh buffer to remove unspecifically bound

15

ACS Paragon Plus Environment

ACS Chemical Neuroscience 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

DA. Mitochondria were directly lysed in loading buffer to have a final concentration of 5 µg/µl. Samples were loaded in a 12% Acrylamide/Bis-Acrylamide gel and run in a mini-vertical electrophoresis unit (SE250, Pharmacia). Statistical analysis Data were analyzed using GraphPad Prism 5 software. One-way ANOVA followed by Dunnett’s post hoc test was used to determine whether groups were statistically different. P