Comprehensive protein expression profiling provides crucial information on abundance, dynamics, and basal state of proteins in DA neurons. Recent studies have mapped the transcriptome and the proteome of DA neurons, which is essential for distinguishing subtype-specific features of SNpc versus VTA DA neurons (Aguila et al., 2021; Azcorra et al., 2023; Zhao et al., 2025).

A recent study by Zhao et al. used proteomics to map the DA neuronal proteome across brain regions. It showed the proteomic landscape of NAc, SNpc, and VTA in mice at three distinct developmental stages: 7, 30, and 60 days old, corresponding to infancy, juvenile, and adult stages. They identified 443 proteins with significant temporal expression changes and reported region-specific clusters of differentially expressed proteins, revealing unique molecular features in each region at different stages. For example, proteins enriched in NAc had a role in functions such as cilium assembly, intracellular transport, and microtubule-based movement. In contrast, terms such as protein-containing complex assembly, integrin-mediated signaling pathways, and vesicle-mediated transport in synapse were enriched in the VTA region (Zhao et al., 2025).

In addition to developmental regulation, proteomic profiling has shed light on how DA neurons respond to environmental stimuli (e.g., oxidative stress, neurotoxins), pharmacological agents (e.g., L-DOPA, antipsychotics), and disease-associated genetic mutations (e.g., LRRK2, SNCA). Comparative studies across different brain regions and disease models have further enriched our understanding of the region- and context-specific molecular signatures of DA neurons (Blumenreich et al., 2024; Teng et al., 2023), which will be discussed in detail later.

Although neurons are post-mitotic, their proteins are in a constant state of synthesis and degradation, a dynamic process called protein turnover, defined by protein half-life. One widely used method for studying protein turnover is stable isotope labeling by amino acids in cell culture (SILAC), combined with mass spectrometry (MS)-based shotgun proteomics (Ong et al., 2002; Figure 2A). In this approach, cells are first grown in media containing “light” amino acids and switched to media containing “heavy” isotopic versions. As proteins are being newly synthesized, they will incorporate the heavy amino acids, which will increase the protein mass. This mass difference can be detected using mass spectrometry and increase of heavy peak intensities can be monitored over time. This technique, termed “dynamic” or “pulsed” SILAC, provides valuable insights into protein synthesis rates, degradation, and overall proteome dynamics (Doherty et al., 2009; Li, 2010).

Li et al. applied in vivo dynamic SILAC labeling in mice, using labeled food, to study protein turnover across 16 tissues and brain regions, including the striatum and the SN. Notably, proteins of the mitochondrial respiratory chain complex I exhibited both high abundance and long half-lives in the brain compared to other tissues and other mitochondrial proteins, emphasizing the importance of oxidative phosphorylation in the brain. Their study also revealed that phosphorylation can influence protein stability and turnover (Li et al., 2025).

Cavarischia-Rega et al. used pSILAC to measure the half-lives of approximately 4,300 proteins in iPSCs derived human DA neurons, reporting a median half-life of 97 h. Interestingly, mitochondrial ribosome proteins had significantly shorter half-lives than the cytosolic ribosomes. Moreover, large and small subunits of the cytosolic ribosome had significantly different half-lives. Additionally, long-lived proteins were enriched in pathways related to cytoskeletal regulation by Rho GTPases, pyruvate metabolism, and de novo purine and cholesterol biosynthesis. In contrast, synaptic vesicle trafficking, Wnt signaling, and Alzheimer’s disease-amyloid and presenilin pathways were overrepresented among the most short-lived proteins (Cavarischia-Rega et al., 2024).

Proteomic profiling helps in the characterization of the differences between subtypes of DA neurons and their response to diverse stimuli, while studying protein turnover allows for a dynamic picture of the proteome in DA neurons. However, this information is only present at the cell level in a general population.

Spatial proteomics is an emerging field that is focused on identifying the tissue-specific and subcellular localization of proteins and understanding how this spatial localization regulates signaling and function. Synapses are highly specialized key compartments of the neurons, which explains the high relevance of their study (Figure 1B). One notable study by Plum et al. investigated the synaptosomal proteome from postmortem human SN samples of PD patients and healthy controls. They identified 362 proteins within the synaptosomal core proteome, out of which 14 proteins were underrepresented in PD, several of which are related to mitochondrial translation. This highlights mitochondrial dysfunction at the synaptic level in PD (Plum et al., 2020).

Recently, Cavarischia-Rega et al. analyzed the enriched axonal proteins in iPSCs derived human DA neurons with the help two-well microfluidic devices, identifying 127 axon-enriched proteins. They also performed differential dynamic SILAC labeling to monitor protein trafficking, which resulted in the detection of translocation of 269 proteins between axons and the soma in the time frame of the analysis (120 h). Importantly, they provided evidence for local synthesis of 154 proteins in the axon and their retrograde transport to the soma, including several proteins involved in RNA biology, such as adenosine deaminase ADAR and RNA helicase DHX30 (Cavarischia-Rega et al., 2024).

Further emphasizing the role of mitochondria, Stauch et al. studied synaptic mitochondria in the striatum of 3-month-old PINK1 knockout and wild-type rat brains. They identified 811 proteins, of which 69 were differentially expressed. The most enriched protein functions were related to the electron transport chain and oxidation-reduction (Stauch et al., 2016). These findings again underline the mitochondrial signature in DA neuron synapses, in particular the respiratory chain. Cavarischia-Rega et al. analyzed the turnover rate of protein subunits of the respiratory chain in dopaminergic neurons and compared this to similar data from human cancer cells (Morgenstern et al., 2021), which revealed that the turnover of mitochondrially encoded subunits remains stable, whereas nuclear encoded subunits differ in their turnover rate between these two very different cell types. Such a comparison of freely available proteomic datasets for specific cell types is a useful resource and offers mechanistic insight for future research.

An important development to study synapses was fluorescence-activated synaptosomes sorting (FASS), which relies on fluorescent tagging of synaptic markers followed by fluorescence-activated cell sorting, enabling high-purity isolation of specific synaptic populations (Luquet et al., 2017; Figure 2B). This technique has been utilized in several proteomics studies to elucidate the complexity of DA synapses. Paget-Blanc et al. combined FASS with label-free quantitative (LFQ) data-dependent acquisition (DDA) proteomics to look into the synapses of the striatum in midbrain mice, identifying 57 highly enriched proteins at the DA synapses. These proteins were related to release sites, vesicles, mitochondria, and cytoskeletal elements. They also described “dopamine hub synapses,” showing that DA synapses associate with cholinergic (14%), GABAergic (26%), and glutamatergic (42%) partners, highlighting the complex microenvironment of DA terminals and underscores the relevance of using organoids over 2D cultures (Paget-Blanc et al., 2022). Van Oostrum et al. also used FASS to investigate synaptosomal differences across different neuronal types in mice, including DA neurons. They focused on the synaptic proteome of striatal DA terminals, where they described 267 proteins significantly enriched at DA terminals. They also revealed novel synapse-enriched proteins specifically for this neuron type, such as MAPK3, ATP6V1G1 or KIF5C. Remarkably, OXR1 (Oxidation resistance protein 1), a protein involved in oxidative stress protection, was absent in DA synaptosomes but present in others, which could potentially explain the vulnerability of DA neurons to oxidative stress (van Oostrum et al., 2023).

Another major advancement in proteomics was the use of enzymatic proximity labeling, which allows the study of protein-protein interactions and subcellular localization. Briefly, the protein of interest or “bait” is tagged with an enzyme, typically a biotin ligase, which, upon the addition of a reagent, gets activated. Upon activation, the enzyme biotinylates proteins in its instant proximity, typically within a 10–20 nm radius. Cells are lysed, and the biotinylated proteins are enriched with streptavidin beads and further analyzed with LC-MS/MS (Figure 2B). The two most commonly used enzymes are the mutant biotin ligase (BioID) (Roux et al., 2012) and the ascorbate peroxidase (APEX) (Rhee et al., 2013). BioID is slower (acts within hours) and has a labeling radius of ∼10 nm, while APEX is much faster (acts within minutes) and has a radius of ∼20 nm (Xu et al., 2021).

Using APEX2, Hobson et al. conducted a comprehensive mapping of somatodendritic and axonal proteomes of DA neurons from mouse midbrain slice. Their analysis revealed that 373 proteins were unique to the ventral midbrain (VM) and 708 to the striatum. In the somatodendritic samples, protein functions associated with protein synthesis and degradation machinery were enriched. Conversely, the axonal/striatum samples were enriched for autophagy, endocytosis, and glycolysis of transport. They also described the unexpected axonal localization of the potassium channels Kv4.3 and GIRK2, which were usually thought to only be present in the soma (Hobson et al., 2022). Finally, Kershberg et al. used BioID to investigate the proteomic composition of dopamine release sites in the mouse striatum. They identified 527 enriched proteins, which were mainly involved in secretory machinery, calcium regulation, and synaptic vesicle function (Kershberg et al., 2022). Notably, their study had a 65% overlap of axonal proteins with those identified by Hobson et al. (2022). Taken together, spatial proteomics has given new insights into the dynamic localization and interaction of DA neurons’ proteins.

Post-translational modifications (PTMs), such as phosphorylation, acetylation, and ubiquitination, play a crucial role in cell signaling. PTMs critically modulate protein function and are often perturbed in disease (Luo et al., 2023; Wu et al., 2023). Phosphorylation can control receptor activation, ion channel function, and neurotransmitter release (Comenencia-Ortiz et al., 2014; Foster and Vaughan, 2017). However, this field is quite understudied, and there are only a few studies in DA neurons. One major limitation of PTMs study is the need for an enrichment step that requires a relatively high initial protein amount (Figure 2B). The previously discussed research by Li et al. explored the phosphoproteome and the influence of specific phosphorylation events on the protein lifetime in the mouse brain (Li et al., 2025). Another example is the study by Meyer et al., which investigated phosphoproteomic profiles in brain organoids derived from patients with BD (Meyer et al., 2025).

Bogerofte et al performed a comprehensive PTM analysis (phosphorylation, sialylation, and cysteine modifications) on iPSCs-derived neurons from four patients carrying a GBA-N370S mutation and four control iPSCs lines. In total, they identified 7,988 phosphorylation sites (on 3,092 proteins), 2,862 N-linked glycosites (on 1,055 proteins), and 11,148 reversible cysteine modifications (on 4,456 proteins). Among the phospho-proteins, functions related to RNA biology, such as RNA splicing or regulation of mRNA metabolic process were enriched whereas for the reversible cysteine-modified PTMs, RNA localization and ribonucleoprotein complex biogenesis were some of the significantly regulated protein functions (Bogetofte et al., 2023).

Lee et al. investigated the role of O-GlcNAcylation, a dynamic glycosylation modification, in the mouse brain. They identified 926 proteins in the SNpc-VTA region and 919 in the striatum. Proteins linked to pathways related to neuronal and synaptic structure and function, including dopamine receptor-mediated signaling, show abundant O-GlcNAcylation. Interestingly, the known marker proteins Tyrosine Hydroxylase (TH) and Dopamine Transporter (DAT) were found to be potentially O-GlcNAcylated (Lee et al., 2020). In summary, these studies show the relevance of PTMs, as they are shown to influence the regulation of multiple pathways including, synaptic function, RNA-related processes, and also protein turnover.

Proteomics has highlighted the involvement of mitochondrial impairment in PD, where DA neurons have a critical role. Functional proteomics of samples with altered expression of LRRK2, GBA1, or α-synuclein (αSyn) has revealed how these proteins disrupt cellular pathways when mutated or misfolded. For example, Blumenreich et al. performed proteomics on five brain regions, including the SN and the striatum, from control, idiopathic PD (IPD), and PD with GBA1 mutation (PD-GBA) patients. This allowed the comparison of protein changes across different brain regions within the same patients. When comparing the IPD and PD-GBA, only two proteins were significantly altered in most brain regions: GCase and ADl1. Mitochondrial oxidative phosphorylation was impaired in PD brains, but more so in PD-GBA cases, reinforcing the link between mitochondrial dysfunction and PD (Blumenreich et al., 2024).

In a proteomic analysis of human SN from 15 PD and 15 healthy control patients, Jang et al. used TMT to identify over 10,000 proteins. Among these, 1,140 were differentially expressed in the SN. They discovered downregulation of mitoribosome proteins in PD, while RNA splicing and complement proteins were upregulated (Jang et al., 2023).

As previously mentioned, the use of patient-derived iPSCs opens the possibility for discovering new therapeutics. This was the case in the study performed by Antoniou et al., where iPSCs derived from a patient with the p.A53T αSyn mutation (G209A in the SNCA gene) were used to study the effect of the multi-kinase inhibitor BX795. To identify compounds with possible neuroprotective properties, the author’s performed a screening of 273 small-molecule kinase inhibitors. Only BX795, an aminopyrimidine compound, showed a consistent dose-response effect. It significantly restored proteins associated with RNA metabolism, protein synthesis, modification, and clearance, and stress response (Antoniou et al., 2022).

Knab et al. analyzed both the whole proteome and the extracellular vesicles of iPSCs from a patient carrying the LRRK2 G2019S mutation and an isogenic gene-corrected control. In the extracellular vesicles, they detected 484 significantly differentially regulated proteins, whereas in cell lysates, they detected 1,833 proteins. In the 123 proteins detected in both datasets, they reported 10 potentially novel biomarkers. Some examples were SHH (sonic hedgehog signaling molecule), RHOA (Ras homolog family member A) and CD44, which were upregulated, or GALC (galactosylceramidase), ANXA1 (annexin A1) and AGO1 (argonaute RISC component 1), which were downregulated (Knab et al., 2024). These findings illustrate the power of combining proteomics with disease-relevant cell models for biomarker discovery.

Beyond PD, proteomics has been employed to uncover the molecular underpinnings of other neuropsychiatric conditions. In alcohol use disorder (AUD), Teng et al. used TMT to analyze multiple different brain regions, including the NAc and VTA, from the brains of 16 male AUD patients and 11 healthy individuals. They found 34 proteins to be differently expressed in the NAc, with lysine degradation and urea cycle pathways enriched. In the VTA, there were 29 proteins differently expressed, with acute phase response signaling being enriched, and six proteins associated with this pathway were elevated in AUD (Teng et al., 2023).

Similarly, Puig et al. studied bulk tissue and synaptosome preparations from the dorsolateral prefrontal cortex (DLPFC) and the NAc from human brains with opioid use disorder (OUD). In the NAc, they identified four upregulated and 29 downregulated proteins in the total proteome and 17 upregulated and 39 downregulated in the synaptosomes. Overall, they showed significant alterations in circadian rhythm, GABAergic, and glutamatergic synaptic processes in both regions of the OUD patients. When looking specifically into the regulated pathways in the NAc synaptosomes, they found mitochondrial and endoplasmic reticulum functions, along with second messenger signaling cascades, to be enriched. They also investigated rhythmic proteins (proteins that fluctuate in sync with circadian rhythms) and their changes between unaffected and OUD subjects in synaptosomes. In OUD, 23 proteins lost rhythmicity in the NAc, while seven did so in the DLPFC. Conversely, 11 proteins gained rhythmicity in DLPFC and none on NAc in the OUD samples (Puig et al., 2023).

As mentioned previously, Meyer et al. developed midbrain organoids derived from iPSCs from patients with BD and healthy controls, with a population of 60-70% of DA neurons. Using TMT multiplexing, they found 234 significantly differentially expressed proteins in BD organoids compared to the controls. Upregulated proteins had enriched functions such as carboxylic acid binding, lysophospholipase activity, and spine morphogenesis. On the other hand, downregulated proteins were mainly related to signaling pathways, including adenylate cyclase activity, NADPH binding, and protein autophosphorylation. They also performed phosphoproteomics, where they identified 196 differentially phosphorylated proteins. Interestingly, synapse and postsynaptic specialization processes, RNA-silencing, ribonucleoprotein granules, and P-bodies processes were observed among the cluster of hypo-phosphorylated proteins. In contrast, hyper-phosphorylated proteins were involved in phosphatidylinositol (PI) signaling system, a known target of the mood-stabilizing drug lithium. These findings show the potential link between molecular alterations and therapeutic mechanisms (Meyer et al., 2025).

Across neurodegenerative and psychiatric disorders, proteomics (especially when coupled with models like patient-derived iPSCs and advanced technologies such as TMT labeling) has proven to be particularly useful in revealing mechanistic insights into disease biology, identifying biomarkers, and facilitating therapeutic discovery. Shared proteomic signatures are emerging across neurodegenerative and psychiatric disorders. Mitochondrial and synaptic dysfunctions were observed not only in PD but also in AUD and OUD, indicating common vulnerability pathways. Interestingly, defects of mitochondrial function and synaptic connectivity are associated with Schizophrenia, a condition affecting both dopaminergic and cortical pathways (Breitmeyer et al., 2023).

Nevertheless, each condition has distinct changes; for example, PD uniquely involves dysregulated lysosomal function, alpha-synuclein pathology, and dopaminergic neuron degeneration, which are not typical features of AUD, BD, or OUD. Furthermore, the primary subtype of DA neuron affected varies, with PD mainly affecting SNpc DA neurons, while VTA DA neurons are more involved in psychiatric disorders. These findings suggest some overlap in DA-related pathophysiology but highlight the need for disorder-specific proteomic analysis. In this regard, further cell type specific and disease specific (genetic models, patient derived or treatment) studies will offer a comprehensive picture of the proteomic landscape of the human brain in health and disease.