The Role of Protein Damage Inflicted by Dopamine Metabolites in Parkinson’s Disease: Evidence, Tools, and Outlook

Abstract

Dopamine is an important neurotransmitter that plays a critical role in motivational salience and motor coordination. However, dysregulated dopamine metabolism can result in the formation of reactive electrophilic metabolites which generate covalent adducts with proteins. Such protein damage can impair native protein function and lead to neurotoxicity, ultimately contributing to Parkinson’s disease etiology. Within this Review, the role of dopamine induced protein damage in Parkinson’s disease is discussed, highlighting the novel chemical tools utilized to drive this effort forward. Continued innovation of methodologies which enable detection, quantification, and functional response elucidation of dopamine-derived protein adducts is critical for advancing this field. Work in this area improves foundational knowledge of the molecular mechanisms that contribute to dopamine-mediated Parkinson’s disease progression, potentially assisting with future development of therapeutic interventions.

Graphical Abstract

INTRODUCTION

Parkinson’s disease (PD) is a progressive neurodegenerative disorder that afflicts over five million individuals globally.¹ PD symptomology manifests as a continual decline in motor function, resulting in rigidity, bradykinesia, tremor at rest, and postural instability.² The cause of motor dysfunction in PD is largely attributed to the depletion of striatal dopamine (DA) due to the death of dopaminergic neurons, particularly in the substantia nigra pars compacta (SNc) (Figure 1).³ However, the events that initiate dopaminergic collapse in the SNc are thought to take place up to a decade prior to diagnosis, as it is estimated that 50–90% of dopaminergic neurons have already been lost by that time.⁴ Currently, there are no approved therapeutics that halt or slow PD. Thus, improving understanding of the molecular underpinnings that initiate PD remains an important area in developing novel treatments.

Figure 1.

Loss of dopaminergic neurons in Parkinson’s disease. Pictural representation of autopsied midbrain sections from healthy and PD diagnosed individuals (right and left respectively). Impairment and death of dopaminergic neurons is a hallmark of PD.

Herein, we discuss the role that dysregulated DA plays in driving PD pathology,⁵ highlighting the contribution of covalent protein damage induced by reactive DA metabolites. Additionally, novel methodologies that enable studying protein damage inflicted by DA metabolites are presented. We conclude with an outlook on this area of PD research and a perspective on how further work investigating DA-protein damage in the brain will aid in elucidating PD etiology.

DOPAMINE METABOLISM & FORMATION OF REACTIVE DOPAMINE METABOLITES

DA was first reported to be a neurotransmitter by Arvid Carlsson in 1957.⁶ Since, DA has been found to modulate a multitude of physiological processes, ranging from motivational salience to motor coordination, and roles outside the central nervous system.^{7, 8} The link between DA and PD was shown by Oleh Hornykiewicz, who reported that DA concentration is drastically decreased in the midbrains of PD patients compared to healthy controls.⁹ This landmark study launched the use of the DA metabolic precursor, L-3,4-dihydroxyphenylalanine (L-DOPA), as a treatment for PD.¹⁰ Despite the symptomatic relief L-DOPA can provide in PD, the underlying neurodegenerative processes involved in the disease are not affected, which has sparked some debate over its continued use.^11–14 Regardless, DA remains a central player in PD etiology, and emerging evidence points to reactive DA metabolites being contributors to the pathology.¹⁵

DA metabolism is a carefully orchestrated process overseen by various enzymatic and sequestering systems.¹⁶ It has been hypothesized that disruption of these systems can lead to production of reactive DA metabolites, buildup of reactive oxygen species (ROS), and impair carbonyl metabolism, all of which can impart neurotoxicity.¹⁷ A summary of DA metabolism is shown in Figure 2. DA generation commences with the hydroxylation of L-Tyr by tyrosine hydroxylase (TH) to give L-DOPA.¹⁸ Next, L-DOPA is decarboxylated via aromatic acid decarboxylase (AADC) through a tetrahydrobiopterin (H4- biopterin) and iron dependent manner to furnish DA.¹⁹ Once DA is created, it can be converted to norepinephrine (NE) by a copper containing monooxygenase, dopamine β-hydroxylase (DβH), or catabolized by one of two major and minor pathways.²⁰ In the major pathway, DA is deaminated to 3,4-dihydroxyphenylacetaldehyde (DOPAL) by monoamine oxidase (MAO). MAO is a flavin adenine dinucleotide (FAD) oxidase that produces hydrogen peroxide and ammonia as additional byproducts in this reaction.²¹ MOA is also anchored to the outer mitochondrial membrane and can shuttle electrons from DA metabolism to the mitochondrial electron transport chain.²² DOPAL is further oxidized to its corresponding acid, 3,4-dihydroxyphenylacetic acid (DOPAC), by an aldehyde dehydrogenase (ALDH).²³ DOPAC is subsequently methylated to form homovanillic acid (HVA) through catechol-O-methyltransferase (COMT).^{24, 25} HVA is an inert byproduct capable of being excreted. In the minor pathway, one of the hydroxyls of DA is first capped by a methyl group via COMT to afford 3-methoxytyramine (MTY), which is consequently deaminated by MAO to give 3-methoxy-4-hydroxyphenyl acetaldehyde (MOPAL) and oxidized by ADH to HVA.²⁶ Additionally, the aldehyde functionality within DOPAL can be reduced, primarily by cytosolic aldehyde reductases (AR) or by alcohol dehydrogenases (ADH), to produce 3,4-dihydroxyphenylethanol (DOPET) as a minor metabolite.^27–29 Together, these enzymatic pathways regulate the biosynthesis and metabolism of DA in presynaptic dopaminergic neurons.

Figure 2.

Primary pathways for dopamine biosynthesis and metabolism. DA is generated from L-Tyr by TH and AADC. DA can be converted by DβH to NE or broken down by two major and minor pathways consisting of transformations catalyzed by MAO, ADH, ALDH, AR and COMT.

Beyond enzymatic regulation, DA concentrations in neurons are controlled by subcellular compartmentalization. An overview of these processes is depicted in Figure 3. Within DA presynaptic neurons, DA is primarily housed within synaptic vesicles. Vesicular monoamine transporter 2 (VMAT2) is the transporter responsible for this shuttling of DA.³⁰ The two enzymes that biosynthesize DA from L-Tyr, TH and AACD are also physically associated with VMAT2 to assist in the confinement of DA.^{31, 32} The H⁺ ATPase localized in the vesicular membrane pumps protons inside the vesicle to maintain an acidic environment in order to stabilize the confined DA.³³ DA concentrations are reported to reach up to 1 M in vesicles, whereas cytosolic concentrations are measured in the low μM range.^{34, 35} DA released into the synaptic cleft undergoes reuptake by dopamine transporter (DAT) for vesicular repacking or is broken down by astrocytes in close proximity.³⁶ Residual cytosolic DA in presynaptic neurons is catabolized to HVA or sequestered into granules of neuromelanin.³⁷ Proper regulation of DA handling systems is critical for dopaminergic health.

Figure 3.

Dopamine handling systems in neurons. DA is created by TH and AADC which is then sequestered into synaptic vesicles by VMAT2. Excess cytosolic DA is converted to HVA by MAO, ADH, and COMT or enveloped into neuromelanin. Residual DA in the synaptic cleft following release by the presynaptic neuron undergoes reuptake by DAT and is repackaged in synaptic vesicles or is broken down by neighboring astrocytes.

Because of its catechol structure, DA has a propensity to spontaneously form reactive metabolites (Figure 4A). When not confined to the acidic environment of synaptic vesicles, DA is susceptible to oxidation by a variety of processes to give dopamine semiquinone (DASQ) or dopaquinone (DAQ).³⁸ DASQ can be generated in vivo enzymatically through H₂O₂ activation by peroxidases and prostaglandin H synthase.^{39, 40} DAQ is formed through auto oxidation at cytosolic pH; this process can be accelerated by metals like Cu and Fe.⁴¹ DAQ rapidly undergoes intramolecular cyclization to afford leukodopaminochrome (LDAC).⁴² However, in some cases, DAQ can be converted to 6-OHDA (6-hydroxydopamine) and 6-OHDQ (6-hydroxydopaquinone) by ROS and Fe.^{43, 44} LDAC is also prone to oxidation, akin to DA, to form dopaminochrome (DAC).⁴⁵ DAC undergoes a tautomeric rearrangement to 5,6-dihydroxyindole (DHI), which further oxidizes to 5,6-dihydroxyindolequinone (DHIQ) and ultimately polymerizes to neuromelanin.⁴⁶ As shown in Figure 4B, the catechol in DOPAL can also be oxidized to a quinone to give DOPAL quinone (DPQAL), another reactive DA derived metabolite.⁴⁷ Collectively, DA oxidation in vivo leads to a complex array of potentially toxic electrophilic metabolites that can react with proteins, necessitating tight regulation over neuronal DA metabolism.

Figure 4.

Dopamine derived reactive metabolites. A) Oxidation products of DA. DA can undergo a 1 e^- oxidation to DASQ or a 2 e^- oxidation to DAQ. DAQ cyclizes to LDAC or reacts with ROS to form 6-OHDA. 6-OHDA can further oxidize to 6-OHDQ. LDAC can oxidize to DAC which can rearrange to DHI. DHI can oxidize to DHIQ which polymerizes due to instability. B) Oxidization of DOPAL to DPQAL.

There is accumulating evidence that DA handling systems are dysregulated in the context of PD, contributing to disease progression.^{48, 49} Specifically, vesicular uptake of DA by VMAT2 is reduced by 87–90% in autopsied striatum sections of PD patients as compared to healthy controls.⁵⁰ Impaired vesicular DA uptake leads to accumulation of cytosolic DA, which could have deleterious consequences on neuronal health.^{51, 52} Accordingly, genetically enhanced vesicular function of DA provides neuroprotection against the dopaminergic toxin 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) in murine models of PD.⁵³ Furthermore, several promoter haplotypes that increase the expression of VMAT2 are associated with a decreased incidence of PD.⁵⁴ In other investigations of PD etiology, accumulation of oxidized DA has been longitudinally linked to a toxic cascade of neurological dysfunction correlated with idiopathic and familial PD cases.⁵⁵ Together, these studies point to the importance of DA regulation in PD.

Beyond improper DA storage in PD, other pathological factors of PD can exacerbate DA inflicted stress.⁵⁶ One of these factors is higher Fe and other metal content in the brains of PD patients as compared to healthy controls.^57–59 The presence of redox active metals can induce oxidative stress and promote DA oxidation.^{60, 61} Additionally, the concentrations of ROS scavenging molecules like glutathione (GSH) are decreased in PD, perpetuating the toxicity of cytosolic DA.^{62, 63} Collectively, there are multiple pathways that can lead to DA dysregulation and faulty DA toxification which can contribute to PD progression. Next, we will explore specific reactions of DA metabolites, along with the ensuing DA damage protein and its relevance to PD.

PROTEIN DAMAGE CAUSED BY DOPAMINE METABOLITES

DA has long been recognized as a driver of oxidative stress in the brain.^{64, 65} Intrastriatal injections of DA in rats caused neuronal degeneration, providing an early demonstration of DA toxicity in the SNc.⁶⁶ Importantly, DA toxicity could be alleviated by co-administration of a reductant such as ascorbic acid or GSH, suggesting that oxidized DA metabolites were responsible for inducing damage. Understanding the molecular mechanisms responsible for DA toxicity in the brain remains an active area of research.⁶⁷

DA metabolites can undergo a variety of chemical reactions that can impart cellular damage.⁶⁸ An overview of these reactions is shown in Figure 5. Some DA metabolites are less stable than others, DASQ being an example. As shown in Figure 5A, DASQ can couple with other radicals, scavenge cellular thiols like GSH, undergo a one electron oxidation by O₂ to generate DAQ and O₂^-, or disproportion to DA and DAQ, all of which can inflict toxicity.^{69, 70} DAQ and other quinone species like DAC, DHIQ, and 6-OHDQ can react with various nucleophilic amino acid residues, which is depicted in Figure 5B.⁷¹ Among amino acids sidechains, Cys has the highest reactivity towards DA-derived quinones, as thiols have a rate constant 10,000-fold higher than amines for nucleophilic attack at the quinone core.^72–74 It has been postulated that the reaction between thiols and ortho-quinones may proceed through a radical mechanism which accounts for the rapid reaction rate and high regioselectivity of thiol addition to the 5 position in the ring.⁷⁵ In addition to Cys, Lys and His sidechains can react with DA metabolites.^76–78 The metabolites DOPAL and DPQAL also harbor reactive groups susceptible to modification and thus have been reported as capable of inducing toxicity.^{17, 79, 80} As shown in Figure 5C, the aldehyde groups of DOPAL and DPQAL can form a Schiff base with amine groups of proteins and peptides.⁸¹ Following Schiff base formation DOPAL can oxidize to its DPQAL form and undergo addition to give rise to crosslinked products.^{28, 82, 83} It is likely that free DPQAL alone is a minor contributor to protein crosslinking due to its inherit instability. However, the chemistry of DOPAL is complex, therefore studies investigating the reactivity of this metabolite need to carefully consider experimental variables such as pH and use of deoxygenated media. Such experimental conditions may bias observed DOPAL reactivity profiles as these factors influence the rates of Schiff base formation and autooxidation processes and thus may not fully recapitulate in vivo conditions. In some cases, two molecules of DOPAL can react with a Lys residue to generate a dicatechol pyrrole adduct.⁸⁴ Additionally, once formed, DA protein adducts can redox cycle and produce additional ROS.⁸⁵ Collectively, the reactivity of DA metabolites is the basis for the damage they can impart on proteins.

Figure 5.

Chemical reactions of dopamine metabolites. A) Reactions by DASQ; radical cross coupling, thiol oxidation, superoxide generation, and disproportionation. Shown from top to bottom. B) Reactions by DAQ: intramolecular cyclization, thiol addition, amine addition (shown from top to bottom). Note that other DA-derived quinones can undergo similar additions. C) Reactions by DOPAL; imine condensation, oxidation to DPQAL, and protein crosslink formation to due to nucleophile addition into the ring.

DA metabolites have been reported to react with a diverse array of proteins. The resulting adducts can impair protein function or cause structural alterations to the native confirmation (Figure 6). These protein modifications can have deleterious effects on many biochemical pathways relevant to PD. Below we discuss various reported DA adducted proteins which are also summarized in Table 1.

Figure 6.

Protein damage inflicted by reactive DA metabolites can drive neurotoxicity through enzymatic inhibition and conformational changes.

Table 1.

Reported dopamine adducted proteins, functional ramifications, and methods of detection.

Protein	Functional consequences of DA modification	Method of adduct detection	Reference(s)
Tyrosine Hydroxylase	Inhibition of enzymatic activity	Radiolabel, NBT staining	86, 87
Dopamine Transporter	Inhibition of transporter activity	Mutagenesis	88
Nuclear Receptor Related-1 Protein	Stimulation of Nurr1 transcriptional activity	X-ray	91
Superoxide Dismutase 2	Inhibition of enzymatic activity	Radiolabel, NBT staining, Mutagenesis	100
Glutathione Peroxidase 4	Inhibition of enzymatic activity	Radiolabel	103
ATP Synthase	Impairment of mitochondrial ATP synthesis	Radiolabel, MS	105
Glutathione-S-Transferase	Inhibition of enzymatic activity	NBT staining	106
Alpha Synuclein	Alteration of oligomer and fibril formation	Radiolabel, Boronate enrichment	123–130
Parkin	Inhibition of E3 ligase function	Radiolabel, Boronate enrichment, MS	134, 135
Glucocerebrosidase	Inhibition of enzymatic activity	nIRF, MS	55
DJ-1	Alteration of protein structure	Radiolabel, NBT staining, Mutagenesis	139
Protein Disulfide Isomerase	Inhibition of enzymatic activity	Bioorthogonal probe, MS	117, 118

DA metabolites have been shown to damage numerous regulatory proteins involved in DA homeostasis. One of the first proteins reported to react with DA metabolites was tyrosine hydroxylase (TH).^{86, 87} Through the use of radiolabeled DA, TH was shown to be covalently modified by oxidized DA metabolites and to become functionally inhibited as a result.⁸⁶ DAT, another DA handling enzyme, was also found to be adducted by DA metabolites at Cys residues which resulted in blocked DA uptake.⁸⁸ Additionally, dihydropteridine reductase (DHPR) activity is inhibited by DA derived quinones, however it is not known if this occurs through covalent modification.⁸⁹ DHPR regenerates tetrahydrobiopterin, which is a cofactor required for hydroxylation reactions and biosynthesis of DA precursors L-Tyr and L-DOPA.⁹⁰ The metabolite DHI has been shown to covalently bind a regulatory Cys in the nuclear receptor related-1 protein (Nurr1) which stimulates Nurr1 activity in cell culture and zebrafish.⁹¹ Nurr1 is a transcription factor that regulates expression of critical maintenance and survival genes of dopaminergic neurons, including DA handling systems like TH, VMAT2 and DAT.^92–97 In fact, all of these aforementioned proteins are important players in DA regulation, thus their inhibition by aberrant DA metabolites fuels a vicious cycle of DA dysregulation relevant to PD.

Oxidized DA metabolites have been linked to mitochondrial dysfunction and disruption of mitochondrial electron transport.^{98, 99} Consequently, various mitochondrial proteins are reported to be impaired by reactive DA species. For example, superoxide dismutase 2 (SOD2), aggregates in the presence of DAQ due to covalent Cys adducts in vitro, which results in diminished catalytic activity.¹⁰⁰ SOD2 is responsible for converting superoxide into H₂O₂ and O₂.¹⁰¹ Inhibition of SOD2 activity can result in mitochondrial dysfunction in vivo.¹⁰² The selenoprotein glutathione peroxidase 4 is also functionally inhibited by DAQ, forming aggregates in vitro and in rat neuronal PC12 cells.¹⁰³ Glutathione peroxidase 4 is a mitochondrial antioxidant enzyme that reduces a variety of cellular peroxides which is combative against oxidative stress and thus hypothesized to be protective against neurodegeneration.¹⁰⁴ In isolated rat brain mitochondria treated with radiolabeled DA, ATP synthase was found to be a target of DA metabolites.¹⁰⁵ Particularly at the γ-subunit of the F 1 catalytic domain as determined by mass spectrometry. This modification resulted in impaired ATP production. In the same study, DA derived quinones also affected mitochondrial morphology. As a whole, mounting evidence suggests that oxidized DA metabolites can elicit mitochondrial dysfunction through protein adduct formation and damage.

Excess reactive DA species have been shown to damage antioxidant pathways. DOPAL inhibits glutathione-S-transferase (GST) in vitro through covalent modification of this protein.¹⁰⁶ GST catalyzes the addition of GSH to reactive metabolites of DA, which serves as a general cellular detoxification mechanism.¹⁰⁷ Overexpression of GST was found to be protective in a Drosophila model of PD.¹⁰⁸ Additionally, high levels of DA administered to astroglial cells resulted in GSH depletion.¹⁰⁹ Thus inhibition of GST and GSH reduction by dysregulated DA may promote oxidative damage in PD.

DA metabolites have been found to modify endoplasmic reticulum (ER) proteins. The ER plays a critical role in maintaining protein homeostasis. ER stress results in accumulation of unfolded proteins and activation of the unfolded protein response (UPR).¹¹⁰ ER stress and UPR is a driver of neurodegenerative processes and PD progression.^111–114 Protein disulfide isomerase (PDI) is a prominent ER chaperon protein responsible for properly folding proteins by catalyzing disulfide bond formation.¹¹⁵ Disruption of PDI activity can lead to ER stress and UPR activation.¹¹⁶ Oxidized DA metabolites have been reported to modify PDI in cell models and inhibit PDI activity in vitro.^{117, 118} Additionally, ER stress is induced by 6-OHDA in cell models.¹¹⁴ Thus it is possible that dysregulated DA metabolism may influence ER stress experienced in PD.

Various proteins with genetic ties to PD are known to be damaged by DA metabolites. Among these is the vesicular trafficking protein, alpha synuclein (AS). AS is closely linked with PD as mutations in this protein results in PD development.^{119, 120} AS is also a main component of Lewy bodies, which are insoluble intracellular protein aggregates observed in the brains of PD patients.^{121, 122} Numerous studies, ranging from in vitro analyses to animal models of PD, have shown that DA metabolites interact with AS, promote AS aggregation, disrupt AS linked synaptic vesicle function, impede chaperone-mediated autophagy processes, and induce degeneration.^{82, 123–130} Additionally, DA modified AS has been found in plasma of PD patients, further strengthening the importance of this adduct.¹³¹ Parkin is another protein where mutations are linked to early onset PD.¹³² Parkin is a E3 ubiquitin ligase that helps clear damaged mitochondria via autophagy and proteasomal mechanisms.¹³³ DA metabolites have been shown to inhibit parkin activity in cell culture, and parkin-DA adducts have been found in autopsied PD brains.^{134, 135} DJ-1 is a redox sensor protein associated with familial PD.¹³⁶ The physiological function of DJ-1 is still being elucidated, but it is well established that DJ-1 confers neuroprotection against oxidative stress.^{137, 138} DJ-1 has been found to be covalently modified by DA derived quinones in vitro, leading to structural alteration of DJ-1.¹³⁹ Mutations in glucocerebrosidase (GCase) are correlated with PD development.¹⁴⁰ GCase hydrolyzes the glycosidic bond in glucocerebroside glycolipids and is a key component in proper lysosome function.¹⁴¹ Reactive DA metabolites have been found to adduct a Cys residue in the active site of GCase in neurons obtained from patients with idiopathic and familial PD, resulting in impaired lysosomal function.⁵⁵ Given that DA metabolites target and inhibit proteins whose dysfunction is already linked to PD, a compelling case can be made for mishandled DA being involved in idiopathic occurrences of PD.

Histones, the proteins involved in DNA packaging within the nucleus, have been reported to be modified by DA via transglutaminase 2 (TGM2).^{142, 143} TGM2 catalyzes amide bond formation between the carboxylic acid sidechain of Glu and the amine of DA to yield a DA-histone linkage. The levels of this modification in rat brains were modulated by cocaine and heroin administration. Consequently, the degree of histone DA modification changed levels of gene expression and drug seeking behavior. Although this histone modification does not involve reactive metabolites of DA, dysregulated DA concentrations in PD may influence levels of this histone mark and lead to pathological alterations in gene expression.¹⁴⁴ Whether histone modifications are relevant in PD pathology remains to be seen, however it is an intriguing area to explore in the future.

Beyond protein dysfunction induced by DA modification, antibodies reactive to DA adducted ovalbumin have been found in PD patient serum. The presence of such DA damage recognizing antibodies creates a possible link to the immune system being involved in PD as DA damage recognizing antibodies could generate or amplify inflammatory responses in PD.¹⁴⁵ Additionally, DA derived neuromelanin is capable of activating innate immune cells, microglia, which induce degeneration of DA neurons.¹⁴⁶ Growing evidence points to immune system dysfunction playing a role in PD progression.^147–149 The studies mentioned above suggest that DA and DA induced protein damage may contribute to aberrant immune responses in PD.

Collectively, covalent protein modifications by reactive DA metabolites can drive neuronal toxicity.^{150, 151} The studies described above provide evidence that dysregulation of DA is a critical factor in PD progression. Further elucidation of the molecular underpinnings of DA dependent damage is key in fully understanding PD etiology. Many efforts in this area utilize chemical tools to drive this research forward. Below, various strategies for studying DA dysregulation using chemical biology and analytical chemistry tools are presented.

TOOLS TO STUDY DOPAMINE DYSFUNCTION

The chemical toolbox used to interrogate the consequences of dysregulated DA is growing. Some of the existing approaches utilize DA mimetic probes, whereas others rely on the chemical structure and reactivity of DA itself. Analytical methodologies are also critical for studying DA dynamics and dysfunction. In the case of identifying DA modified proteins, certain methods provide a higher degree of unambiguous evidence for DA adduction than others. For instance, MS/MS analysis of an adducted peptide provides concrete verification that a particular protein is indeed susceptible to DA modification, whereas other approaches (e.g., radiolabeling, immunoprecipitation, redox detection, etc.) provide only presumptive evidence for a DA adduct within a particular protein. Below, we will discuss examples of these classes, highlight how different tools provide valuable insights into DA’s role in PD, and underscore advantages and disadvantages of given technologies.

DA derivatives come in many modalities (Figure 7). For example, radiolabeled DA is an effective probe for studying DA driven protein damage. In pioneering work, ¹⁴C-dopamine in combination with autoradiography was used to identify proteins modified by DA metabolites.^{152, 153} These studies provided an initial set of DA adducted proteins to be further investigated. While introduction of ¹⁴C in DA is advantageous in terms of imposing minimal structural changes to the DA scaffold, the toxicity and regulatory hurdles associated with ¹⁴C hamper the use of radioactive isotopes.¹⁵⁴ Alternative probe-based strategies to identify proteins damaged by DA derived species have leveraged bioorthogonal chemistries to enrich DA adducted proteins prior to identification by mass spectrometry.¹⁵⁵ Enrichment of structurally modified protein species is important for their identification due to signal suppression by unmodified peptides during mass spectrometry analyses. For example, alkynylated derivatives of 6-OHDA and DA, dubbed 6-OHDA-PEG3-yne and DA^yne respectively, have been reported (Figure 7).^{117, 118} These researchers have successfully demonstrated the utility of bioorthogonal functionalization of DA as it enabled detection and proteomic identification of proteins modified by DA metabolites when 6-OHDA-PEG3-yne was administered to cellular lysates or when DA^yne was incubated in cell culture. It should be noted that care must be taken in the design of structurally altered DA mimetic probes as to not drastically alter their physicochemical properties from those of DA. As a testament to this, 6-OHDA-PEG3-yne is unable to traverse membranes in living cells whereas DA^yne can. As a whole, functionalized DA derivatives make powerful tools for broad scale proteomic studies of DA adducted proteins.

Figure 7.

Classes and structures of dopamine mimetic probes.

Additional DA based probes have been designed for tracking DA dynamics and signaling pathways. These efforts have been recently reviewed.¹⁵⁶ Briefly, ¹⁸F fluorodopamine and ¹⁹F-L-DOPA probes were developed to be used for positron emission tomography (PET) and magnetic resonance imaging (MRI) (Figure 7).^{157, 158} These tools can report on DA localization in vivo. Another class of tools are fluorescent false neurotransmitters (FFNs) like FFN102, FFN200, and FFN206 which are substrates for DA handling enzymes and used in confocal microscopy applications (Figure 7).^159–161 While these molecules do not directly report on protein damage caused by dysregulated DA, they are critical for deciphering DA dynamics and trafficking in vitro and in vivo.

The second category of tools for studying DA disposition in the brain take advantage of the molecule’s redox reactivity (Figure 8). Early efforts in detecting proteins covalently modified by DA metabolites relied on the redox sensitive catechol functionality possessed by DA metabolites.¹⁶² Following immobilization of quinone modified proteins on a membrane, a basic nitroblue tetrazolium (NBT) solution, which redox cycles with the membrane bound adducted proteins to produce nitroblue formazan (NBF), can be applied. NBF in turn deposits on and stains the membrane purple, enabling visual detection of protein bound quinones. This technique was utilized in a comparative proteomic study aimed at profiling DA modified proteins in 2- and 15-month rats.¹⁶³ These researchers found that levels of DA adducted proteins increase with age. This methodology is useful for comparing relative amounts of quinone modified proteins between samples, however it is unable to distinguish different species of quinones from one another. Other redox cycling quinone protein detection strategies have been reported using a dithiothreitol (DTT) and luminol system to generate quantifiable chemiluminescence.¹⁶⁴ Such techniques may have utility for DA metabolites as well. A similar approach for detecting immobilized protein bound ortho quinones or free ortho quinones alone, utilizes the spectroscopic properties of these molecules.¹⁶⁵ Here, a near-infrared fluorescence (nIRF) imager is used to excite bound quinones at 685 nm and the resultant emission from the ortho quinone is measured at 700 nm. This is selective for ortho quinones/catechols over meta and para substituted analogs. This method enables the detection of oxidized DA at nmol levels. However, similar to detection via redox cycling, this procedure is unable to differentiate species of protein bound quinones. Despite limitations, these tools are robust methodologies for quantifying global levels of protein damage induced by DA metabolites.

Figure 8.

Methods to detect dopamine-protein adducts. Top: Schematic of NBT redox cycle with a DA protein adduct to yield NBF, enabling visualization of DA protein adducts. Bottom: Depiction of the nIRF properties of ortho quinones which enable the detection of DA protein adducts.

Other tools that utilize the chemical structure of DA to study DA protein adducts enable enrichment and capture of modified proteins (Figure 9). One example is the use of boronate agarose resin, which chelates the cis diols present on DA at alkaline pH and releases them at acidic pH.¹⁶⁶ Experimental workflows based on boronate resins has enabled the capture of DA modified parkin in human brain samples and DA metabolites modified by AS in vitro.^{123, 134} It should be noted that this approach is useful for enriching endogenously DA damaged proteins by any metabolites that retains catechol functionality upon adduction, but is prone to capturing other biological cis diols such as glycosylated proteins as well. Enrichment strategies and other applications of boronate chemistry can be found in the following review.¹⁶⁷ Another reported strategy to detect proteins susceptible to DA modification used DA functionalized quantum dots (QDs).¹⁶⁸ Covalent attachment of a protein to the DA anchored on the CdTe/ZnS QDs causes a measurable change in the fluorescence intensity of the QDs, which can report levels of adducted protein captured by the nanoparticles. These DA functionalized QDs are cell penetrant and thus can provide a relative measure of the amount of proteins vulnerable to DA modification in a particular cell. However, this approach cannot determine the identity of the adducted proteins.

Figure 9.

Technologies to capture dopamine protein adducts. Top: Schematic of boronate functionalized resin utilized to immobilize and release proteins containing a DA adduct. Bottom: Visualization of DA decorated nanoparticles enabling detection of proteins susceptible to DA modification.

Beyond DA protein adduct detection, there are additional tools to monitor endogenous DA to study DA dynamics in a variety of capacities. Biosensor and chemical sensor tools for DA detection have been recently reviewed.^{169, 170} These methods rely on the intrinsic chemical reactivity of DA for signal generation. Genetically encoded DA sensors that produce fluorescence upon DA binding have enabled recording DA synaptic dynamics in mice.^{171, 172} Small molecule chemical sensors have been designed to covalently trap DA which results in measurable fluorescence changes in the sensor molecule which can be used to track DA location in cells. However distinguishing between DA and NE by this approach remains challenging.¹⁷³ Similar reactivity based DA sensing strategies have also been applied to nanoparticle approaches.^174–177 These tools are valuable constituents of the toolbox needed to probe DA dysfunction in PD as they can be applied in animal and cell models of PD.

Analytical chemistry techniques are also invaluable additions to the chemical toolbox focused on studying DA’s involvement in PD. Fast scan cyclic voltammetry is a powerful way to quantify DA concentration and dynamics.¹⁷⁸ This electrochemical approach detects the amounts of DA adsorbed onto an electrode by increasing and decreasing the voltage applied and simultaneously measuring the resulting current generated by the oxidation and reduction of the catechol in DA; further details can be found in the following review.¹⁷⁹ This methodology has enabled measurement of sub second DA release in vivo and provided valuable insights in how ROS, synaptic vesicle dysfunction, and aging impact PD progression in animal models, but has a low throughput and requires a skilled operator.^180–186

Liquid chromatography mass spectrometry (LC-MS) is a powerful analytical tool for studying dysregulated DA in PD. LC-MS is the foundation for any proteomic based study aimed at identifying protein targets of reactive DA metabolites.¹⁸⁷ Typically, protocols will enrich DA-adducted proteins from cellular lysates through a variety of methods discussed above or generate DA-protein adducts in vitro by incubating a particular protein of interest with reactive DA metabolites prior to LC-MS analysis. Enrichment of adducted proteins is a key component of such analysis. Often adducts are in low abundance in comparison to unmodified proteins which necessitates enrichment to enable their detection by MS. The enriched pool of DA-protein adducts is commonly digested with a trypsin protease to afford peptide fragments.¹⁸⁸ These peptides are separated by a LC system with an in line orbitrap mass analyzer. Inside this type of mass spectrometer, the peptides are fragmented to give a series of Y and B ions, which are deterministic of the parent peptide sequences.¹⁸⁹ Sequenced peptides are then searched in various databases to identify their protein of origin.¹⁹⁰ An overview of this process is shown in Figure 10. Our laboratory has successfully identified 38 DA-protein adducts in SH-SY5Y cells through such an approach, which led us to discover that reactive DA metabolites modify and functionally inhibit PDI.¹¹⁷ Studies like this improve understanding of how reactive DA metabolites inflict damage across the proteome.

Figure 10.

Identification of dopamine modified proteins by mass spectrometry. Low abundant DA protein adducts are enriched via various strategies (e.g. bioorthogonal chemistry and boronate affinity). This enriched set of proteins is digested to peptides and subjected to LC-MS/MS analysis. The MS² spectra can then be searched against a database to enable protein identification.

The specific sites of DA-protein modification can also be elucidated by LC-MS. A notable example of this is highlighted by Burbulla et al, where a specific DA modification was located on a catalytic Cys residue within the active site of GCase.⁵⁵ Mapping sites of modification provides valuable information into structural motifs that are susceptible to DA adduction and gives insight into the molecular mechanisms by which reactive DA metabolites impair protein function. However, there are still challenges in comprehensively mapping the specific sites of DA adducts across the proteome, namely ensuring the peptide containing the DA adduct ionizes well and has a m/z within the analysis range of the spectrometer. In such cases, locations of DA modification can be inferred through mutagenesis experiments. In these workflows, the hypothesized adducted amino acid is mutated to a nonreactive sidechain and the mutant protein is assayed to determine if DA adduction still occurs.

Additionally, LC-MS methodology allows for detection of potential PD biomarkers linked to DA dysfunction.^{191, 192} 5-S-cysteinyl-dopamine, a byproduct of reactive DA metabolites coupling with Cys or GSH, has been reported to be elevated with age and in the brains of PD patients.^193–195 Thus these DA derived metabolic adducts may have utility as potential PD biomarkers.¹⁹⁶

Collectively, there is a large array of chemical tools to study DA. These tools can be leveraged to better understand the roles DA plays in PD progression. Each tool possesses unique advantages and disadvantages that must be kept in mind when designing experiments. The continued expansion and innovation of the DA chemical toolbox will diversify the methodologies researchers have at their disposal to interrogate DA dynamics and will further advance the understanding of functional consequences of DA induced protein damage.

OUTLOOK

PD is a complex and heterogeneous disease. The exact cause of PD remains unknow and there are many competing hypotheses concerning the origins and molecular events that drive PD pathology.¹⁹⁷ It is hypothesized that intracellular accumulation of misfolded AS is the main driver of neuronal toxicity and PD development.^198–200 There is evidence that misfolded AS originating in the gut can propagate through the gut-brain axis and induce neurodegeneration.^{201, 202} Other postulated origins for PD are the age dependent buildup of oxidative stress, mitochondrial dysfunction, and lysosomal impairment.^203–206 There are also genetic risk factors that predispose individuals to PD.^{207, 208} However a majority of PD cases are idiopathic in origin and thus a combination of genetic and environmental factors likely work in concert to influence PD development.^{202, 209} Additionally, there are particular features that SNc dopaminergic neurons possess which confer vulnerability and predispose them to degradation in PD, such as their long length, low levels of myelination, and high energy consumption.^{210, 211} Another risk factor for degeneration of these dopaminergic neurons is the presence of DA itself.

The propensity for DA to degrade into reactive metabolites when improperly displaced from vesicular confinement is why DA dysregulation is a potential player in PD. Rogue DA derived electrophilic metabolites react with a host of proteins, which in turn can inhibit the activity of the adducted proteins, and lead to neurotoxicity. Understanding the nuances of these molecular events is important as it provides insights to the multifaceted etiology of PD, which can aid biomarker discovery for early detection and drive therapeutic development.²¹²

Tremendous progress has been made to characterize protein damage caused by reactive DA metabolites. However, there are still many unanswered questions pertaining to DA’s contribution to PD. For instance, what percentage of a particular protein within a cell needs to be modified by DA metabolites to drive a phenotypic response? Answering questions like this are especially important in cases like Nurr1 DA adducts, which can influence gene expression.⁹¹ Implementing activity-based protein profiling (ABPP) strategies capable of reporting percent occupancy of an electrophile bound to a particular protein, along with the specific site of modification, would be useful towards this end.^213–218 Additionally, controlled and targeted release of DA metabolites to particular proteins and locations in cellular systems would assist in clarifying signaling pathways impacted by DA protein adducts.^{219, 220} These types of precise chemical tools have been developed to study the consequences of protein damage caused by lipid derived electrophiles like 4-hydroxynonenal (HNE) and reactive dicarbonyl metabolites like methylglyoxal (MGO).^221–225 Similar DA tools could be built from reported photoactivatable DA precursors.²²⁶

Other remaining questions include: how does the abundance of DA-protein adducts change over time and what DA metabolites are most influential in driving neuronal dysfunction? Longitudinal studies tracking the formation and identity of proteins modified by DA metabolites over an organism’s life correlated to PD incidence would provide insight in how DA induced damage contributes to PD. Continued development of tools enabling the enrichment and characterization of DA protein adducts would be valuable assets in this area. Further optimization of LC-MS methodology that comprehensively identifies DA-protein adducts would aid this effort. Harnessing the unique reactivity of ortho quinones may prove useful in enriching DA protein adducts in biological samples through cycloadditions with strained cyclic alkenes.²²⁷ Additionally, antibodies raised against particular DA modified proteins are currently lacking and would also be a useful enrichment and detection tool. To determine the pathological contribution of various DA derived metabolites, probes akin to the reported DA^yne and 6-OHDA-PEG3-yne could be designed to mimic specific oxidation states of DA metabolites.^{117, 118} A comprehensive suite of DA derived probe molecules would enable comparative studies between the complex array of different DA metabolites and illuminate their contribution to DA induced protein damage.

To conclude, multiple lines of evidence point to DA dysregulation being a contributor to PD. Reactive DA metabolites can elicit neurotoxicity when they are not properly broken down or when their levels overwhelm natural detoxification machinery. Chemical tools have greatly improved our understanding of DA induced dysfunction. Continued research effort in the area of DA induced protein damage in PD and further expansion of the chemical toolbox needed to study these processes will further elucidate the complex disease etiology of PD.

ABBREVIATIONS

AADC

aromatic acid decarboxylase

ABPP

activity-based protein profiling

ADH

alcohol dehydrogenase

ALDH

aldehyde dehydrogenase

AR

aldehyde reductases

AS

alpha synuclein

COMT

catechol-O-methyltransferase

DA

dopamine

DAC

dopaminochrome

DAT

dopamine transporter

DAQ

dopamine quinone

DASQ

dopamine semiquinone

DβH

dopamine β-hydroxylase

DHI

5,6-dihydroxyindole

DHIQ

5,6-dihydroxyindolequinone

DHPR

dihydropteridine reductase

DOPAC

3,4-dihydroxyphenylacetic acid

DOPAL

3,4-dihydroxyphenylacetaldehyde

DOPET

3,4-dihydroxyphenylethanol

DPQAL

DOPAL quinone

DTT

dithiothreitol

ER

endoplasmic reticulum

FAD

flavin adenine

GCase

glucocerebrosidase

GSH

glutathione; dinucleotide

GST

glutathione-S-transferase

HVA

homovanillic acid

HNE

4-hydroxynonenal

LC-MS

liquid chromatography mass spectrometry

LDAC

leukodopaminochrome

L-DOPA

L-3,4-dihydroxyphenylalanine

MOA

monoamine oxidase (MAO)

MGO

methylglyoxal

MOPAL

3-methoxy-4-hydroxyphenyl acetaldehyde

MPTP

1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine

MRI

magnetic resonance imaging

MTY

3-methoxytyramine

NBF

nitroblue formazan

NBT

nitroblue tetrazolium

NE

norepinephrine

nIRF

near-infrared fluorescence

Nurr1

nuclear receptor related-1

PET

positron emission tomography

PD

Parkinson’s disease

PDI

protein disulfide isomerase

QDs

quantum dots

ROS

reactive oxygen species

SNc

substantia nigra pars compacta

SOD2

superoxide dismutase 2

TGM2

transglutaminase 2

TH

tyrosine hydroxylase

UPR

unfolded protein response

VMAT2

Vesicular monoamine transporter 2

Biographies

Alexander K. Hurben is currently a Ph.D. candidate in the Medicinal Chemistry program at the University of Minnesota under the supervision of Professor Natalia Y. Tretyakova. He received his B.S. in Chemistry from the University of Minnesota where he synthesized chemical probes to study protein prenylation in Professor Mark Distefano’s group. His current research focuses on the development of chemical tools to study non-enzymatic posttranslational modifications caused by endogenous electrophilic metabolites to better understand their role in disease and cell signaling.

Natalia Y. Tretyakova is a Distinguished McKnight University professor of Medicinal Chemistry at the College of Pharmacy and the Masonic Cancer Center at the University of Minnesota- Twin Cities. She is also the Founding Director of the University of Minnesota Epigenetics Consortium and Chair of the ACS Division of Chemical Toxicology. The Tretyakova research program is on the interface of nucleic acid chemistry, chemical carcinogenesis, and bioanalytical chemistry. She is employing the tools of chemical biology, analytical chemistry, and biochemistry to investigate the chemistry and biology of DNA and protein damage by exogenous and endogenous chemicals, to determine the modes of action of anti-tumor drugs, and to develop novel nucleoside analogs as molecular probes and biologically active compounds.