The plasma membrane proteome is a rich resource of functional and therapeutically relevant protein targets. Distinguished by high hydrophobicity, heavy glycosylation, disulfide-rich sequences, and low overall abundance, the cell surface proteome remains undersampled in established proteomic pipelines, including our own cysteine chemoproteomics platforms. Here we paired cell surface glycoprotein capture with cysteine chemoproteomics to establish a two-stage enrichment method that enables chemoproteomic profiling of cell Surface Cysteinome. Our <CysSurf= platform captures >2,800 total membrane protein cysteines in 1,046 proteins, including 1,907 residues not previously captured by bulk proteomic analysis. By pairing Cys-Surf with an isotopic chemoproteomic readout, we uncovered 821 total ligandable cysteines, including known and novel sites. Cys-Surf also robustly delineates redox-sensitive cysteines, including cysteines prone to activation-dependent changes to cysteine oxidation state and residues sensitive to addition of exogenous reductants. Exemplifying the capacity of Cys-Surf to delineate functionally important cysteines, we identified a redox sensitive cysteine in the low-density lipoprotein receptor (LDLR) that impacts both the protein localization and uptake of LDL particles. Taken together, the Cys-Surf platform, distinguished by its two-stage enrichment paradigm, represents a tailored approach to delineate the functional and therapeutic potential of the plasma membrane cysteinome.

The cell surface proteome regulates nearly all aspects of cellular function, spanning, cellcell communication133, signal transduction436, and changes to cell state7, including activation8310, proliferation11,12, and death13. Consequently nearly 50% of modern drug targets are found in the membrane proteome, which is also rich in biomarkers indicative of disease14318. Notably, over 200 cell surface proteins have been reported as overexpressed in human cancers19. Despite their prominence as high value therapeutic targets, the hydrophobicity and low abundance of most cell surface proteins complicates functional characterization and therapeutic targeting of this important fraction of the proteome20323.

A number of high value biochemical and mass spectrometry based proteomic techniques have been developed to improve capture and identification of the cell surface proteome, most notably cell surface biotinylation24329 and cell surface capture (CSC)30337. As the hydroxysuccinimidobiotin (NHS-biotin) reagent used in cell surface biotinylation has been shown to label intracellular proteins38,39, CSC has emerged as a favored method for specifically enriching membrane bound glycoproteins. CSC leverages the heavy glycosylation of nearly all membrane proteins490% of the proteins on the cell surface have been reported to be glycosylated404to afford high specificity capture of the cell surface proteome. In the CSC method, cis-diols found on all glycans are first oxidized to aldehyde groups in the presence of sodium periodate.

Subsequently, these aldehyde moieties are trapped with aminooxy- or hydrazide-biotin reagents, which affords robust biotinylation of cell surface proteins, via formation of a stable oxime or hydrozone. Biotinylated protein groups are then identified via established enrichment-based proteomic workflows. Showcasing the widespread utility and adoption of the CSC, recent studies have implemented this chemistry to identify and quantify surfaceome alterations in cancer cell lines18,41344, such as overexpression of CD166 in head and neck squamous cell carcinoma45 and carbonic anhydrase 12 in colon cancer46.

Cysteine residues are a ubiquitous, functionally important, and therapeutically relevant feature of cell surface proteins47. Membrane proteins are particularly rich in structural disulfides and redox-active cysteines are widely known to regulate membrane protein function48352. Multiple cell surface cysteines are reported to be redox active, including C53 of aquaporin-8 AQP8, persulfidation of which is known to gate H2O2 to regulate cell stress53 and C23, C45, and C106 of high mobility group box 1 protein HMGB1, reduction of which are known to be crucial for antiinflammatory processes54,55. Activation-induced changes to cysteine oxidation have been linked to enhanced T cell activitation56358 and cysteines on CD4 and GP120 have been implicated in the entry of HIV in T cells and in the regulation of HIV infection of T cells59,60. Cell surface cysteines have also been targeted by drugs and drug-like molecules, as showcased by Afatinib and Ibrutinib, blockbuster anti-cancer agents, that function by selectively and irreversibly labeling conserved non-catalytic cysteines found within kinase active sites61,62. Consequently, identifying redox sensitive and potentially druggable cysteine residues in membrane proteins is of high value for expanding the scope of the druggable proteome.

Mass spectrometry-based cysteine chemoproteomics is one such method that is well positioned to pinpoint such residues in membrane proteins. As shown by previous reports63367, including our own, the human proteome harbors thousands of cysteines amenable to modification by druglike molecules4each cysteine-ligand pair represents the potential starting point for a drug development campaign. Similar to small molecule screening platforms, redox-directed chemoproteomic methods, including biotin-switch68, OxICAT69, SP3-Rox70, QTRP71, and Oximouse72, robustly report the relative and absolute oxidation states of cysteines analyzed from bulk proteomes. By pairing these methods with proximity labeling, recent methods including our own Cys-LOx have begun to shed light on subcellular compartment-specific cysteine oxidation states, which can vary markedly depending on protein localization73,74. Whether analogous approaches can be extended to the cell surface proteome remains to be seen.

Here, we establish cell Surface Cysteine enrichment (Cys-Surf), which pairs cell surface capture (CSC) with cysteine chemoproteomics to reveal the redox and ligandable cell surface cysteinome. Cys-Surf analysis of Jurkat and primary T cells identified 2,836 reactive cell surface cysteines on 1,046 proteins, which represents a >4-fold improvement of cell surface specificity as compared to bulk proteome analysis. Of these residues, 821 were not previously identified by cysteine chemoproteomics as reported by CysDB63, and we find that 211 proteins are ligandable by cysteine reactive compounds. By pairing Cys-Surf with our previously reported cysteine redox proteomic method SP3-Rox70, we quantify the absolute oxidation states of 1,246 cell surface cysteines on 489 proteins, reveal marked cysteine reduction during T cell activation, and stratify membrane cysteines sensitive to addition of exogenous oxidants and reductants. Guided by this Cys-Surf analysis we pinpoint a heretofore uncharacterized reduction-sensitive cysteine in an EGF repeat of the LDLR receptor, which impacts receptor localization and lipoprotein uptake.

To establish our cell Surface Cysteine enrichment platform (Cys-Surf) (Figure 1A), we envisioned pairing cell surface capture (CSC)30,31 with cysteine peptide enrichment. Guided by our prior study, which paired TurboID proximity labeling with similar cysteine peptide enrichment to achieve subcellular redox proteomics73, we expected that efficient protein recovery throughout the Cys-Surf workflow would be essential to achieve high cysteine coverage.

Therefore, we first prioritized establishing and enhancing the CSC portion of our method. We observed saturable labeling for aminooxy-biotin, with maximal labeling achieved with 1 mM treatment conditions (Figure S1A). As we had previously found that the single-pot, solid-phaseenhanced sample preparation (SP3) method75 afforded enhanced recovery of biotinylated samples64,76, we opted to incorporate SP3 into the CSC workflow. We established an SP3-based protein cleanup method in which the SP3 resin is employed to decontaminate samples prior to protein enrichment on streptavidin resin (Figure S1B). By comparing wash and protein elution conditions, we found that EtOH outperformed MeCN as the washing solvent and that addition of detergent (e.g. 0.2% SDS, which is not required for peptide elution64,77) is essential for protein recovery from SP3 resin (Figure S1C and S1D). Addition of SP3 to the CSC method afforded a modest yet significant improvement in peptide coverage (Figure S1E). As further confirmation that our implementation of CSC was robustly capturing cell surface glycoproteins, we additionally subjected enriched peptides to PNGase F digest (Figure S2A). We used the unique mass shift (0.98 Da) that remains on glycosylated asparagine residues after digest, to identify the glycosylation sites. ~53% of detected peptides feature putative glycosylation sites and, of these, ~93% belong to proteins annotated as localized to the cell surface (Figure S2B).

We next paired our established CSC platform with cysteine labeling to establish our Cys-Surf workflow shown in Figure 1A. We expected that the sodium periodate used to oxidize glycans in the CSC workflow likely would also oxidize exposed cysteine thiols. Confirming this likelihood, we observed a decrease in cysteine reactivity towards iodoacetamide rhodamine (IA-Rho) following treatment of with sodium periodate; in contrast the IA-Rho signal remained unchanged when sodium periodate was applied after cysteine capping (Figure 1B). To circumvent this oxidation, we opted to first cap all cysteines in situ with the pan-cysteine reactive probe iodoacetamide alkyne (IAA;1).

The subsequent two-stage

Motivated by the robust capture of cell surface cysteines, we next set out to enable quantitative measures of subcellular cysteine ligandability by incorporating isotopic labeling into the Cys-Surf workflow. We envisioned that our previously reported isotopically labeled light- and heavy-isopropyl iodoacetamide alkyne (IPIAA-L 4 and IPIAA-H 5), which had enabled high coverage bulk redox proteomics70, would prove compatible with competitive electrophilic small molecule screening. In a manner highly analogous to the widely utilized competitive chemoproteomic profiling platform - isotopic Tandem Orthogonal Proteolysis3Activity-Based Protein Profiling (isoTOP-ABPP)66,82, we treated samples in a pairwise manner with either scout fragment 3,5-bis(trifluoromethyl)aniline chloroacetamide KB03 or vehicle (DMSO) followed by either light- or heavy-IPIAA (Figure 2A). After CSC and peptide level cysteine enrichment, FragPipe search and IonQuant83 was utilized to calculate the competition ration (ratio (R) of IPIAAH over IPIAA-L).

Out of the 2,194 cysteines quantified, 530 cysteines belonging to 452 proteins were found in peptides that showed elevated MS1 ratios (Log2(H/L) > 1), indicating covalent labeling by KB03 (Figure 2B). Out of the 530 ligandable cysteines, 211 cysteines are annotated as from proteins on cell surface or glycosylated. Cys-Surf identified unique ligandable cell surface proteins that differ from those already targeted by FDA approved drugs (Figure 2C). More than 350 ligandable cell surface cysteines identified by our study are unique and have not been previously identified in CysDB63 (Figure 3D). While we do not identify C797 in EGFR4this is the cysteine labeled by targeted tyrosine kinase (TKIs) inhibitors (e.g. afatinib)614we do observe an elevated ratio (Log2(H/L) = 2.74) for C1058 in EGFR. This cysteine has been previously identified as a palmitoylation site located in a disordered loop, which has been implicated in the palmitoylationdependent regulation of EGFR activity84,85. Gene Ontology (GO)-analysis86 of proteins that harbor liganded cysteines demonstrated an enrichment of proteins involved in processes including integrin-mediated signaling pathway, membrane organization and biosynthesis (Figure S4). We expect that many of these ligandable cysteines, including C6 of CD8187 and C132 of CD32088, represent intriguing potential targets for immunomodulation34.

Among the integrin-related ligandable cysteines, C363 from

As the HLA-A Cys363 cysteine had been previously reported as redox active94,95, we next asked whether Cys-Surf would extend to quantification of site-specific cysteine oxidation. Building upon our previously reported SP3-Rox70 and Cys-LOx73 redox proteomics platforms, we modified our Cys-Surf platform to capture the cell surface redox proteome (Figure 3A). Similar to many redox proteomic methods68,69,96,97, we calculated the cysteine percentage oxidation state based on the relative cysteine labeling by our isotopically enriched IPIAA reagents, with IPIAA-L labeling reduced cysteines and IPIAA-H labeling oxidized cysteines following global reduction with DTT. In support of the robustness of our method, we also generated datasets in which the cysteine capping reagents were swapped, with IPIAA-H capping reduced cysteines and IPIAA-L capping the oxidatively modified cysteines after reduction. A high concordance (R^2 = 0.8942, n = 1,538 unique cysteines identified in both datasets) was revealed for the precursor ion intensity ratios calculated by FragPipe IonQuant83 (Figure S6).

Demonstrating the uniquely oxidizing environment of the cell surface cysteinome, 1,173 out of the 1,246 cysteines (89% of cysteines) with cell surface annotations quantified from 489 proteins were found to have elevated ratios (log2(H/L) >1), corresponding to >50% percent oxidation (Figure 3B). In aggregate, cysteines in cell surface proteins showed an average oxidation state of 83%; this high degree of overall oxidation is in stark contrast to the 10% mean cysteine oxidation state previously reported for whole cell proteome70,72,82,983101, which supports Cys-Surf9s capacity to assay the subcellular oxidation state of cysteines in the plasma membrane (Figure 3C).

Given the markedly oxidized nature of most cysteines identified, we next asked whether these residues were known to be involved in formation of disulfide bonds. We find that 495 of the identified cysteines from 192 total proteins, were annotated as disulfide sites by UniProtKB. This coverage (40% of residues identified) represents a marked enrichment for cysteines involved in disulfides when compared to bulk proteome analysis4in CysDB 17% (n = 1,077) of identified cysteines are annotated as involved in disulfides. Exemplary cysteines involved in disulfide bonds include Cys177 of NPC intracellular cholesterol transporter 1 (NPC1), Cys296 of Cluster of Differentiation (CD) 38 antigen, and Cys113 of CD166 antigen ALCAM (Figure 3B, 3D).

Looking beyond these anecdotal examples, we find that the cysteines with annotated disulfide involvement showed a 93%

mean oxidation, which reflects increased oxidation compared with all identified cell surface cysteines and to those lacking disulfide annotations (83% and 78% mean oxidation, respectively; Figure 3E). Notably, 453 additional cysteines, which lacked UniProtKB disulfide annotations, were also detected with >90% percent oxidization (Table S3). We expect that a number of these sites are likely cysteines involved in heretofore unannotated disulfides.

Given the predominance of highly oxidized cysteines in our dataset, we next asked whether Cys-Surf would have sufficient sensitivity to identify domain specific differences in cysteine oxidation. Using UniProtKB topological domain feature annotations, we identified that our dataset contained 175 cysteines located within annotated intracellular protein domains and 589 cysteines in annotated extracellular domains. Consistent with a high degree of oxidation of cysteines in the plasma membrane, we observe mean oxidation states of 92% for extracellular and versus 58% for intracellular domain cysteines (Figure 3F). For the aforementioned ligandable cysteine Cys363 in HLA-A, which is located in an annotated intracellular topological domain, we observe substantially reduced oxidation 61% when compared with the three HLA-A extracellular cysteines (Cys188, Cys227, and Cys283; Figure 3G). A) The schematic workflow of quantification of oxidation states of cell surface cysteines with CysSurf. Reduced cysteines were first labeled using our custom isotopically light isopropyl iodoacetamide alkyne (IPIAA-L) probe. Subsequently, the samples were subjected to reduction and labeled using our isotopically differentiated heavy (IPIAA-H) probe. After CSC and peptide level cysteine enrichment, Fragpipe IonQuant reported the ratio (R) of IPIAA-H over IPIAA-L labeled cysteine peptides, and B) Percentage oxidation of cysteines quantified with Cys-Surf calculated using the formula (R/(1+R))*100 as reported previously70,73. Average percentage oxidation of 4 biological replicates is reported C) Percentage oxidation of cysteines quantified with Cys-Surf compared with that quantified with whole proteome. D) Percentage oxidation of representative cysteines quantified with Cys-Surf. E) Percentage oxidation of cysteines quantified with Cys-Surf with or without disulfide annotation. F) Percentage oxidation of cysteines quantified with Cys-Surf with domain analysis of extracellular or intracellular. G) Percentage oxidation of 4 cysteines in HLA-A quantified with Cys-Surf. For panel E and F, statistical significance was calculated with unpaired student's t-tests, *** p<0.001. For panel C, E, and F, the mean is represented by <x=. MS experiments were conducted in 4 biological replicates in Jurkats celsl. All

During activation, T cells are exposed to reactive oxygen species (ROS) and reactive nitrogen species (RNS), derived from both intracellular mitochondrial ROS production and from extracellular neighboring phagocytes. Intriguingly, this oxidative environment has been connected with both T cell hypo-responsiveness as well as T cell activation, including increased IL2 and NFAT expression1023104. These seemingly antithetical activities of ROS as both pro- and antiproliferative have been ascribed to both dose-dependent activity (low ROS is pro-proliferative and higher ROS-negatively impacting growth) and compartmentalized redox signaling. Intriguingly, while bulk proteome redox analysis revealed increased cysteine oxidation during T cell activation for a number of immunomodulatory targets70,105,106, an increase in cell surface cysteine free thiols is strongly associated with immunological stimuli56,57,1073110.

Inspired by these intriguing observations, we deployed Cys-Surf to pinpoint T cell activation-dependent changes to plasma membrane-associated cysteine oxidation state. Healthy donor T cells (bulk CD4+ and CD8+) were subjected to the Cys-Surf workflow (Figure 4A) preand post- stimulation (anti-CD3 and anti-CD28). In aggregate, the oxidation state of 2,402 cysteines on 1,190 proteins were quantified, in which 1,466 cysteines from 638 proteins are annotated as on the cell surface or are glycosylated (Table S4). Consistent with the aforementioned reports of activation-dependent increases in plasma membrane free thiol content, T cell stimulation afforded a modest yet significant decrease in the measured mean percentage cysteine oxidation state (Figure 4B). Out of 821 cysteines from 471 proteins quantified in both unstimulated and stimulated T cell proteomes, 100 cysteines were significantly reduced (difference < -1, p-value < 0.5) upon T cell stimulation (Figure 4C). GO-analysis86 of genes that harbor significantly reduced cysteines revealed an enrichment of proteins involved in processes including regulation of T cell tolerance induction, lymphocyte differentiation and integrin-mediated signaling, which are all closely related to T cell activation (Figure 4D). Exemplary reduced cell surface cysteines involved in adaptive immune response include a number of disulfide-bonded cysteines (Figure 4E, S7), including C41 from the T cell surface glycoprotein (CD4), which functions as a coreceptor for MHC class II molecule:peptide complex57,111, C70, C83 and C170 from the T cell differentiation antigen (CD6)112,113, and C385, C472 and C551 from the dipeptidyl peptidase 4 (DPP4), which is a known positive regulator of T cell activation1143117. For the receptortype tyrosine-protein phosphatase C (PTPRC or CD45), which is a DPP4 interactor that also acts as a positive regulator of T cell coactivation, active site C853 that forms a phosphocysteine intermediate118 was markedly reduced. Notably, the disulfide-rich extracellular domains of CD45 have been implicated in increased domain rigidity and formation of heterodimers119,120. For CD4121,122 (Cys41) and Tapasin (TAPBP)123 (Cys91), the reduced disulfides are potential 3 RHStaple bonds4these are allosteric disulfides that link adjacent strands in the same β sheet124,125, which aligns with the comparatively labile nature of these bonds126,127. Exemplifying the value of subcellular redox measurements, for C551 of DPP4 and C170 of CD3, which were identified by both our previous bulk proteomic70- and our cys-Surf analysis of T cell activation, we observe markedly divergent state-dependent changes to thiol oxidation (Figure 4E). during T cell activation.

A) The schematic workflow to identify redox sensitive cell surface cysteines during T cell activation with Cys-Surf. B) Redox states of cell surface cysteines quantified with Cys-Surf in naive and activated T cells. The mean is represented by <x=. Statistical significance was calculated with unpaired student's t-tests, *** p<0.001. C) Difference of redox states for cysteines quantified with Cys-Surf during T cell activation. D) Top 10 GO biological processes enriched with proteins harboring reduced cell surface cysteines during T cell activation. E) Difference of redox states of representative cysteines quantified with Cys-Surf or bulk proteomics in unactivated or activated T cells. MS experiments were conducted in 3 replicates in human T cells. All MS data can be found in Table S4.

Inspired by the seeming ubiquity of labile disulfides, we opted to broaden the scope of redox sensitive cell surface cysteines by pairing Cys-Surf with cell treatments with exogenous oxidoreductants. For oxidants, we chose to the exemplary oxidizing agents H2O2 and SNitrosoglutathione (GSNO), with the goal of pinpointing cell surface cysteines sensitive to ROS and RNS-mediated cysteine oxidative modifications. For reductants, we selected both the pancysteine reactive Tris (2-carboxyethyl) phosphine (TCEP) reducing agents as well as Nacetylcysteine (NAC), which is a clinically administered reductant that is also widely taken as a dietary supplement128,129.

We subjected Jurkat cells to in-gel fluorescence analysis using the cell-impermeable cysteine-reactive Alexa Fluor™ 594 C5 Maleimide (Maleimide-Alexa594)130,131 to visualize treatment-dependent changes in cell surface cysteine reactivity (Figure 5A, top). For the H2O2and GSNO-treated cells, we observed a modest decrease in the reactivity of free thiols towards Maleimide-Alexa594 (Figure S8A). In contrast, lysates derived from TCEP- and NAC-treated cells, exhibited increased Maleimide-Alexa594 labeling, consistent with reduction-dependent increased availability of free thiols. When compared to NAC, samples treated with TCEP exhibited a more substantial increase in Maleimide-Alexa594 reactivity, consistent with TCEP9s greater reduction potential (Figure 5B).

As our gel-based analyses confirmed treatment-dependent changes to cysteine oxidation, we next extended these analyses to Cys-Surf characterization of the oxidation- and reduction sensitive cell surface proteomes. Following the workflow shown in Figure 5A (bottom), we identified treatment-dependent changes to cysteine oxidation state based on the difference in MS1 intensity ratios comparing additive- and vehicle-treated samples. Consistent with the comparatively small decrease in Maleimide-Alexa594 labeling observed for oxidant treated samples (Figure S8A), Cys-Surf analysis of H2O2- and GSNO-treated cells revealed no significant treatment-dependent changes in the aggregated mean percentage oxidation of the cysteines identified (Figure S8B). These findings are consistent with our aforementioned observation (Figure 2B, 2C) that, at basal state, most cell surface cysteines are already highly (mean >80%) oxidized and thus insensitive to additional oxidant. This observed insensitivity is in stark contrast to that observed for whole proteome70, where we found that >51.0% of cysteines were significantly oxidized with GSNO. Out of 1,310 cysteines identified in both samples treated with vehicle and H2O2, 16 cysteines showed significant ratio changes (difference >1, p-value <0.05) sensitive to H2O2, including C590 of integrin beta-2 ITGB2, which has been reported as ligandable in human T cells105 (Figure S8C). Out of 305 cell surface cysteines identified in both samples with control and GSNO treatment, 3 cysteines showed significant GSNO induced oxidation (difference >1, pvalue <0.05), including C109 of platelet endothelial cell adhesion molecule PECAM1, a protein that is required for leukocyte transendothelial migration (TEM) under most inflammatory conditions132,133 (Figure S8D).

Guided by the observed comparative insensitivity to exogenous oxidations and high degree of baseline oxidation state of cell surface cysteines, we next asked whether a more substantial fraction of the cell surface cysteinome would be sensitive to exogenous reductants TCEP and NAC. Consistent our in-gel fluorescence analysis (Figure 5B), Cys-Surf analysis revealed a modest yet significant decrease in the measured bulk mean cysteine oxidation state after TCEP and NAC treatment, decreasing from 93.4% to 90.0% and 91.0%, respectively (Figure 5C). Out of the 916 cysteines from 618 proteins identified in both samples with control and NAC treatment, 15 cysteines showed significant NAC induced reduction (difference <-1, p-value <0.05), including C173 of signaling threshold-regulating transmembrane adapter 1 SIT1, which has been reported as immune-enriched and hyperreactive in activated human T cells105 (Figure S9). Out of the 1,189 cysteines from 698 proteins identified in both samples treated with control and TCEP, 39 cysteines were sensitive to TCEP-induced to reduction (difference <-1, p-value <0.05), of which 32 has cell surface annotation. We initially were surprised by the low overlap between the NAC and TCEP datasets, with 534 shared total cysteines. Together with the difference in reducing potentials for TCEP and NAC, we expect that the comparatively modest coverage (1,573 total cysteines) together with the stochastic nature of data dependent acquisition (DDA) likely rationalizes the modest overlap between datasets.

Several intriguing reduction labile cysteines were identified only in the TCEP datasets. Annotated disulfides in multiple CD antigens were identified as reducible, including C145 of CD55, which has been identified as significantly reduced during T cell activation in our previous bulk proteome dataset70 (Figure 5D, S10). C311 of the epidermal growth factor receptor kinase (EGFR) was found to be markedly sensitive to TCEP (difference = -2.55). This cysteine is localized to a disulfide bridge (C311-C326) in a cysteine-rich region of EGFR that is required for kinase activation134. Notably, disulfide breaking mutations at C326 were recently reported to be gain-offunction for Lhermitte-Duclos Disease135.

Similarly, a large number of mutations at disulfide bonded extracellular cysteines in the low-density lipoprotein receptor (LDLR) have been implicated in Familial Hypercholesterolemia (FH), one of the most common genetic disorders136,137. In our TCEP dataset, we identified two cysteines from LDLR, highly reduced C75 (Difference = -1.5) and comparatively reduction insensitive C34 (Difference = -0.3). While variants at these two cysteines have been reported in ClinVar138, with clinical significance as likely pathogenic and associations with familial hypercholesteremia, the impact of these variants on LDL uptake is, to our knowledge, unknown. Notably, mutation at the proximal C46 has been demonstrated to reduce LDL uptake139.

Using fluorescently labeled LDL with GFP tagged on the C-term of LDL, we next assayed the impact of TCEP and C75 mutations on LDL uptake and LDLR activity. TCEP treatment attenuated both basal HEK-293T cellular LDL uptake and LDLR-dependent uptake with heterologously overexpressed LDLR (Figure 5E, 5F, S11). Guided by these findings, we then generated point mutations in our LDLR-GFP construct, including at C75 (LDLRC75W) and the double mutant (LDLRC75W/C34W), with the latter chosen to assess whether C75 mutations would synergize with the reduction-insensitive C34 residue. 50 % to 95 % at 250 nL/min buffer B in buffer A (Table S7). Data was collected with charge exclusion (1, 8, >8). Data was acquired using a Data-Dependent Acquisition (DDA) method consisting of a full MS1 scan (Resolution = 120,000) followed by sequential MS2 scans (Resolution = 15,000) to utilize the remainder of the 1 second cycle time. Precursor isolation window was set as 1.6 and normalized collision energy was set as 30%. The MS data have been deposited to the ProteomeXchange Consortium via the PRIDE partner repository149,150 with the dataset identifier PXD042403. File details can be found in Table S8.

Protein, peptide, and cysteine identification. Raw data collected by LC-MS/MS were converted to mzML and searched with MSFragger (v3.3) and FragPipe (v19.0)151. The proteomic workflow and its collection of tools was set as default and PTMprophet was enabled152. Precursor and fragment mass tolerance was set as 20 ppm. Missed cleavages were allowed up to 1. Peptide length was set 7 - 50 and peptide mass range was set 500 - 5000. For Cys-Surf identification, cysteine residues were searched with differential modification C+463.2366. For Cys-Surf oxidation and ligandability quantification, MS1 labeling quant was enabled with Light set as C+463.2366 and Heavy set as C+467.2529. MS1 intensity ratio of heavy and light labeled cysteine peptides were reported with Ionquant (v1.8.9). Calibrated and deisotoped spectrum files produced by FragPipe were retained and reused for this analysis. The MS search data have been deposited to the ProteomeXchange Consortium via the PRIDE partner repository with the dataset identifier PXD042403. File details can be found in Table S8. Custom python scripts were implemented to compile labeled peptide datasets. Unique proteins, unique cysteines, and unique peptides were quantified for each dataset. Unique proteins were established based on UniProt protein IDs. Unique peptides were found based on sequences containing a modified cysteine residue. Unique cysteines were classified by an identifier consisting of a UniProt protein ID and the amino acid number of the modified cysteine (ProteinID_C#); residue numbers were found by aligning the peptide sequence to the corresponding UniProt protein sequence. When there are multiple cysteines in one peptide, all the modified cysteine residue numbers will be reported as

Data analysis. For the cell surface annotation, our customized localization database was used to cross referenced with the proteins or cysteines identified. For isotopical quantification, the medium of heavy over light ratios for the same cysteine residue from cysteine peptides of different charges and miss cleavages in the same sample was calculated. Means of reported logged ratio values for each condition (+/- H2O2 or +/- GSNO or +/- TCEP or +/-NAC or activated/naive T cells) were calculated for all replicates per condition. Percentage oxidation for a cysteine was calculation based on heavy to light ratio via the following formula: (R/(1+R))*100, using unlogged ratios. When calculating oxidation difference, relative oxidation changes between two cellular conditions were reported by calculating the change of heavy to light ratios between treated and untreated samples.

UniProtKB annotations for disulfide bonds and topological domain annotations were calculated based on matches between the position of the identified residue and UniProtKB functional region. Further parsing of UniProtKB disulfide bond site annotations were extracted to obtain specific residues and amino acid numbers. Exact amino acid positions of UniProtKB cysteines involved in disulfide bonds or within regions of annotated domains were cross-referenced with our cysteine identifiers. Cysteines with 8extracellular9 topological domain annotations were classified as cysteines in extracellular domains, while cysteines with 8cytoplasmic9, 8mitochondrial intermembrane9, 8mitochondrial matrix9, and 8perinuclear space9 were classified as cysteines in intracellular domains.

reported as indicated. Statistical significance was calculated with unpaired Student's t-tests using R stats (v 3.6.2) if applicable. For bar plots in Figure 1D, 5F, 5H, S1E, S2B and S3. Error bars were calculated using standard deviation. Statistical significance was calculated with unpaired Student's t-tests using R stats (v 3.6.2) if applicable. For volcano plots in Figure 4C, 5D, S8C, S8D and S9A, variances were calculated for each sample-condition pairing and a corresponding two-sample t-test was performed using R stats (v 3.6.2) to generate p-values. p-values were adjusted for multiple comparisons using Benjamini-Hochberg procedure. Li, T.; Luginbuhl, D. J.; Wu, B.; McLaughlin, C. N.; Xie, A.; Kaewsapsak, P.; Quake, S. R.; Carr, S. A.; Ting, A. Y.; Luo, L. Cell-Surface Proteomic Profiling in the Fly Brain Uncovers Wiring Regulators. Cell 2020, 180 (2), 373-386.e15. da Cunha, J. P. C.; Galante, P. A. F.; de Souza, J. E.; de Souza, R. F.; Carvalho, P. M.; Ohara, D. T.; Moura, R. P.; Oba-Shinja, S. M.; Marie, S. K. N.; Silva, W. A.; Perez, R. O.; Stransky, B.; Pieprzyk, M.; Moore, J.; Caballero, O.; Gama-Rodrigues, J.; Habr-Gama, A.; Kuo, W. P.; Simpson, A. J.; Camargo, A. A.; Old, L. J.; de Souza, S. J. Bioinformatics Construction of the Human Cell Surfaceome. Proc. Natl. Acad. Sci. 2009, 106 (39), 167523 16757.

Shuster, S. A.; Li, J.; Chon, Ur.; Sinantha-Hu, M. C.; Luginbuhl, D. J.; Udeshi, N. D.; Carey, D. K.; Takeo, Y. H.; Xie, Q.; Xu, C.; Mani, D. R.; Han, S.; Ting, A. Y.; Carr, S. A.; Luo, L. In Situ Cell-Type-Specific Cell-Surface Proteomic Profiling in Mice. Neuron 2022, 110 (23), (4)

Wojdyla, K.; Collier, A. J.; Fabian, C.; Nisi, P. S.; Biggins, L.; Oxley, D.; Rugg-Gunn, P. J. Cell-Surface Proteomics Identifies Differences in Signaling and Adhesion Protein Expression between Naive and Primed Human Pluripotent Stem Cells. Stem Cell Reports

2020, 14 (5), 9723988.

Gaultier, A.; Simon, G.; Niessen, S.; Dix, M.; Takimoto, S.; Cravatt, B. F.; Gonias, S. L.

LDL Receptor-Related Protein 1 Regulates the Abundance of Diverse Cell-Signaling Schumann, J. Molecular Mechanism of Cellular Membranes for Signal Transduction. Membranes (Basel). 2022, 12 (8), 748.

Queiroz, R. M. L.; Charneau, S.; Bastos, I. M. D.; Santana, J. M.; Sousa, M. V.; Roepstorff, P.; Ricart, C. A. O. Cell Surface Proteome Analysis of Human-Hosted Trypanosoma Cruzi Life Stages. J. Proteome Res. 2014, 13 (8), 353033541.

Ravenhill, B. J.; Soday, L.; Houghton, J.; Antrobus, R.; Weekes, M. P. Comprehensive Cell Surface Proteomics Defines Markers of Classical, Intermediate and Non-Classical Monocytes. Sci. Rep. 2020, 10 (1), 4560.

Soday, L.; Potts, M.; Hunter, L. M.; Ravenhill, B. J.; Houghton, J. W.; Williamson, J. C.; Antrobus, R.; Wills, M. R.; Matheson, N. J.; Weekes, M. P. Comparative Cell Surface Proteomic Analysis of the Primary Human T Cell and Monocyte Responses to Type I Interferon. Front. Immunol. 2021. https://doi.org/10.3389/fimmu.2021.600056. (10) Schiess, R.; Mueller, L. N.; Schmidt, A.; Mueller, M.; Wollscheid, B.; Aebersold, R. Analysis of Cell Surface Proteome Changes via Label-Free, Quantitative Mass Spectrometry. Mol.

Cell. Proteomics 2009, 8 (4), 6243638. (11) Karcini, A.; Lazar, I. M. The SKBR3 Cell-Membrane Proteome Reveals Telltales of Aberrant Cancer Cell Proliferation and Targets for Precision Medicine Applications. Sci.

Rep. 2022, 12 (1), 10847. (12) Özlü, N.; Qureshi, M. H.; Toyoda, Y.; Renard, B. Y.; Mollaoglu, G.; Özkan, N. E.; Bulbul, S.; Poser, I.; Timm, W.; Hyman, A. A.; Mitchison, T. J.; Steen, J. A. Quantitative Comparison of a Human Cancer Cell Surface Proteome between Interphase and Mitosis. EMBO J. 2015, 34 (2), 2513265.

S.; Miyano, S.; Print, C. Understanding Endothelial Cell Apoptosis: What Can the Transcriptome, Glycome and Proteome Reveal? Philos. Trans. R. Soc. B Biol. Sci. 2007, 362 (1484), 146931487.

Rev. Drug Discov. 2006, 5 (12), 9933996. (15) Overington, J. P.; Al-Lazikani, B.; Hopkins, A. L. How Many Drug Targets Are There? Nat. (16) Collier, A. J.; Panula, S. P.; Schell, J. P.; Chovanec, P.; Plaza Reyes, A.; Petropoulos, S.; Corcoran, A. E.; Walker, R.; Douagi, I.; Lanner, F.; Rugg-Gunn, P. J. Comprehensive Cell Surface Protein Profiling Identifies Specific Markers of Human Naive and Primed Pluripotent States. Cell Stem Cell 2017, 20 (6), 874-890.e7. (17) Gundry, R. L.; Boheler, K. R.; Van Eyk, J. E.; Wollscheid, B. A Novel Role for Proteomics in the Discovery of Cell-Surface Markers on Stem Cells: Scratching the Surface.

PROTEOMICS – Clin. Appl. 2008, 2 (6), 8923903. (18) Kuhlmann, L.; Cummins, E.; Samudio, I.; Kislinger, T. Cell-Surface Proteomics for the Identification of Novel Therapeutic Targets in Cancer. Expert Rev. Proteomics 2018, 15 (3), (19) McEnaney, P. J.; Parker, C. G.; Zhang, A. X.; Spiegel, D. A. Antibody-Recruiting Molecules: An Emerging Paradigm for Engaging Immune Function in Treating Human Disease. ACS Chem. Biol. 2012, 7 (7), 113931151. (20) Elschenbroich, S.; Kim, Y.; Medin, J. A.; Kislinger, T. Isolation of Cell Surface Proteins for

Mass Spectrometry-Based Proteomics. Expert Review of Proteomics. 2010, pp 1413154. (21) Macher, B. A.; Yen, T.-Y. Proteins at Membrane Surfaces4a Review of Approaches. Mol.

Biosyst. 2007, 3 (10), 705. (22) Wu, C. C.; Yates, J. R. The Application of Mass Spectrometry to Membrane Proteomics.

Nat. Biotechnol. 2003, 21 (3), 2623267. (23) Vuckovic, D.; Dagley, L. F.; Purcell, A. W.; Emili, A. Membrane Proteomics by High Performance Liquid Chromatography-Tandem Mass Spectrometry: Analytical Approaches and Challenges. Proteomics. 2013, pp 4043423. (24) Elia, G. Biotinylation Reagents for the Study of Cell Surface Proteins. Proteomics 2008, 8 (19), 401234024. (25) Rybak, J.-N.; Ettorre, A.; Kaissling, B.; Giavazzi, R.; Neri, D.; Elia, G. In Vivo Protein Biotinylation for Identification of Organ-Specific Antigens Accessible from the Vasculature.

Nat. Methods 2005, 2 (4), 2913298. (26) Li, Y.; Wang, Y.; Mao, J.; Yao, Y.; Wang, K.; Qiao, Q.; Fang, Z.; Ye, M. Sensitive Profiling of Cell Surface Proteome by Using an Optimized Biotinylation Method. J. Proteomics 2019, 196, 33341. (27) Pischedda, F.; Szczurkowska, J.; Cirnaru, M. D.; Giesert, F.; Vezzoli, E.; Ueffing, M.; Sala, C.; Francolini, M.; Hauck, S. M.; Cancedda, L.; Piccoli, G. A Cell Surface Biotinylation Assay to Reveal Membrane-Associated Neuronal Cues: Negr1 Regulates Dendritic Arborization. Mol. Cell. Proteomics 2014, 13 (3), 7333748. (28) Nishimura, N.; Sasaki, T. Cell-Surface Biotinylation to Study Endocytosis and Recycling of