The Choice of Search Engine Affects Sequencing Depth and HLA Class I Allele-Specific Peptide Repertoires.

Standardization of immunopeptidomics experiments across laboratories is a pressing issue within the field, and currently a variety of different methods for sample preparation and data analysis tools are applied. Here, we compared different software packages to interrogate immunopeptidomics datasets and found that Peaks reproducibly reports substantially more peptide sequences (~30-70%) compared with Maxquant, Comet, and MS-GF+ at a global false discovery rate (FDR) of <1%. We noted that these differences are driven by search space and spectral ranking. Furthermore, we observed differences in the proportion of peptides binding the human leukocyte antigen (HLA) alleles present in the samples, indicating that sequence-related differences affected the performance of each tested engine. Utilizing data from single HLA allele expressing cell lines, we observed significant differences in amino acid frequency among the peptides reported, with a broadly higher representation of hydrophobic amino acids L, I, P, and V reported by Peaks. We validated these results using data generated with a synthetic library of 2000 HLA-associated peptides from four common HLA alleles with distinct anchor residues. Our investigation highlights that search engines create a bias in peptide sequence depth and peptide amino acid composition, and resulting data should be interpreted with caution.

Prosit-TMT: Deep Learning Boosts Identification of TMT-Labeled Peptides.

The prediction of fragment ion intensities and retention time of peptides has gained significant attention over the past few years. However, the progress shown in the accurate prediction of such properties focused primarily on unlabeled peptides. Tandem mass tags (TMT) are chemical peptide labels that are coupled to free amine groups usually after protein digestion to enable the multiplexed analysis of multiple samples in bottom-up mass spectrometry. It is a standard workflow in proteomics ranging from single-cell to high-throughput proteomics. Particularly for TMT, increasing the number of confidently identified spectra is highly desirable as it provides identification and quantification information with every spectrum. Here, we report on the generation of an extensive resource of synthetic TMT-labeled peptides as part of the ProteomeTools project and present the extension of the deep learning model Prosit to accurately predict the retention time and fragment ion intensities of TMT-labeled peptides with high accuracy. Prosit-TMT supports CID and HCD fragmentation and ion trap and Orbitrap mass analyzers in a single model. Reanalysis of published TMT data sets show that this single model extracts substantial additional information. Applying Prosit-TMT, we discovered that the expression of many proteins in human breast milk follows a distinct daily cycle which may prime the newborn for nutritional or environmental cues.

Prosit: proteome-wide prediction of peptide tandem mass spectra by deep learning.

In mass-spectrometry-based proteomics, the identification and quantification of peptides and proteins heavily rely on sequence database searching or spectral library matching. The lack of accurate predictive models for fragment ion intensities impairs the realization of the full potential of these approaches. Here, we extended the ProteomeTools synthetic peptide library to 550,000 tryptic peptides and 21 million high-quality tandem mass spectra. We trained a deep neural network, termed Prosit, resulting in chromatographic retention time and fragment ion intensity predictions that exceed the quality of the experimental data. Integrating Prosit into database search pipelines led to more identifications at >10× lower false discovery rates. We show the general applicability of Prosit by predicting spectra for proteases other than trypsin, generating spectral libraries for data-independent acquisition and improving the analysis of metaproteomes. Prosit is integrated into ProteomicsDB, allowing search result re-scoring and custom spectral library generation for any organism on the basis of peptide sequence alone.

Data, Reagents, Assays and Merits of Proteomics for SARS-CoV-2 Research and Testing.

As the COVID-19 pandemic continues to spread, thousands of scientists around the globe have changed research direction to understand better how the virus works and to find out how it may be tackled. The number of manuscripts on preprint servers is soaring and peer-reviewed publications using MS-based proteomics are beginning to emerge. To facilitate proteomic research on SARS-CoV-2, the virus that causes COVID-19, this report presents deep-scale proteomes (10,000 proteins; >130,000 peptides) of common cell line models, notably Vero E6, Calu-3, Caco-2, and ACE2-A549 that characterize their protein expression profiles including viral entry factors such as ACE2 or TMPRSS2. Using the 9 kDa protein SRP9 and the breast cancer oncogene BRCA1 as examples, we show how the proteome expression data can be used to refine the annotation of protein-coding regions of the African green monkey and the Vero cell line genomes. Monitoring changes of the proteome on viral infection revealed widespread expression changes including transcriptional regulators, protease inhibitors, and proteins involved in innate immunity. Based on a library of 98 stable-isotope labeled synthetic peptides representing 11 SARS-CoV-2 proteins, we developed PRM (parallel reaction monitoring) assays for nano-flow and micro-flow LC-MS/MS. We assessed the merits of these PRM assays using supernatants of virus-infected Vero E6 cells and challenged the assays by analyzing two diagnostic cohorts of 24 (+30) SARS-CoV-2 positive and 28 (+9) negative cases. In light of the results obtained and including recent publications or manuscripts on preprint servers, we critically discuss the merits of MS-based proteomics for SARS-CoV-2 research and testing.

New Approaches for Absolute Quantification of Stable-Isotope-Labeled Peptide Standards for Targeted Proteomics Based on a UV Active Tag.

Targeted proteomics depends on the availability of stable isotope labeled (SIL) peptide standards, which for absolute protein quantification need to be absolutely quantified. In the present study, three new approaches for absolute quantification of SIL peptides are developed. All approaches rely on a quantification tag (Qtag) with a specific UV absorption. The Qtag is attached to the peptide during synthesis and is removed by tryptic digestion under standard proteomics workflow conditions. While one quantification method (method A) is designed to allow the fast and economic production of absolutely quantified SIL peptides, two other methods (methods B and C) are developed to enable the straightforward re-quantification of SIL peptides after reconstitution to control and monitor known problems related to peptide solubility, precipitation, and adhesion to vials. All methods yield consistent results when compared to each other and when compared to quantification by amino acid analysis. The precise quantitation methods are used to characterize the in vivo specificity of the H3 specific histone methyltransferase EZH2.

Building ProteomeTools based on a complete synthetic human proteome.

We describe ProteomeTools, a project building molecular and digital tools from the human proteome to facilitate biomedical research. Here we report the generation and multimodal liquid chromatography-tandem mass spectrometry analysis of >330,000 synthetic tryptic peptides representing essentially all canonical human gene products, and we exemplify the utility of these data in several applications. The resource (available at http://www.proteometools.org) will be extended to >1 million peptides, and all data will be shared with the community via ProteomicsDB and ProteomeXchange.

ProteomeTools: Systematic Characterization of 21 Post-translational Protein Modifications by Liquid Chromatography Tandem Mass Spectrometry (LC-MS/MS) Using Synthetic Peptides.

The analysis of the post-translational modification (PTM) state of proteins using mass spectrometry-based bottom-up proteomic workflows has evolved into a powerful tool for the study of cellular regulatory events that are not directly encoded at the genome level. Besides frequently detected modifications such as phosphorylation, acetylation and ubiquitination, many low abundant or less frequently detected PTMs are known or postulated to serve important regulatory functions. To more broadly understand the LC-MS/MS characteristics of PTMs, we synthesized and analyzed ∼5,000 peptides representing 21 different naturally occurring modifications of lysine, arginine, proline and tyrosine side chains and their unmodified counterparts. The analysis identified changes in retention times, shifts of precursor charge states and differences in search engine scores between modifications. PTM-dependent changes in the fragmentation behavior were evaluated using eleven different fragmentation modes or collision energies. We also systematically investigated the formation of diagnostic ions or neutral losses for all PTMs, confirming 10 known and identifying 5 novel diagnostic ions for lysine modifications. To demonstrate the value of including diagnostic ions in database searching, we reprocessed a public data set of lysine crotonylation and showed that considering the diagnostic ions increases confidence in the identification of the modified peptides. To our knowledge, this constitutes the first broad and systematic analysis of the LC-MS/MS properties of common and rare PTMs using synthetic peptides, leading to direct applicable utility for bottom-up proteomic experiments.

Nuclear PKC-θ facilitates rapid transcriptional responses in human memory CD4+ T cells through p65 and H2B phosphorylation.

Memory T cells are characterized by their rapid transcriptional programs upon re-stimulation. This transcriptional memory response is facilitated by permissive chromatin, but exactly how the permissive epigenetic landscape in memory T cells integrates incoming stimulatory signals remains poorly understood. By genome-wide ChIP-sequencing ex vivo human CD4(+) T cells, here, we show that the signaling enzyme, protein kinase C theta (PKC-θ) directly relays stimulatory signals to chromatin by binding to transcriptional-memory-responsive genes to induce transcriptional activation. Flanked by permissive histone modifications, these PKC-enriched regions are significantly enriched with NF-κB motifs in ex vivo bulk and vaccinia-responsive human memory CD4(+) T cells. Within the nucleus, PKC-θ catalytic activity maintains the Ser536 phosphorylation on the p65 subunit of NF-κB (also known as RelA) and can directly influence chromatin accessibility at transcriptional memory genes by regulating H2B deposition through Ser32 phosphorylation. Furthermore, using a cytoplasm-restricted PKC-θ mutant, we highlight that chromatin-anchored PKC-θ integrates activating signals at the chromatin template to elicit transcriptional memory responses in human memory T cells.

PROCAL: A Set of 40 Peptide Standards for Retention Time Indexing, Column Performance Monitoring, and Collision Energy Calibration.

Beyond specific applications, such as the relative or absolute quantification of peptides in targeted proteomic experiments, synthetic spike-in peptides are not yet systematically used as internal standards in bottom-up proteomics. A number of retention time standards have been reported that enable chromatographic aligning of multiple LC-MS/MS experiments. However, only few peptides are typically included in such sets limiting the analytical parameters that can be monitored. Here, we describe PROCAL (ProteomeTools Calibration Standard), a set of 40 synthetic peptides that span the entire hydrophobicity range of tryptic digests, enabling not only accurate determination of retention time indices but also monitoring of chromatographic separation performance over time. The fragmentation characteristics of the peptides can also be used to calibrate and compare collision energies between mass spectrometers. The sequences of all selected peptides do not occur in any natural protein, thus eliminating the need for stable isotope labeling. We anticipate that this set of peptides will be useful for multiple purposes in individual laboratories but also aiding the transfer of data acquisition and analysis methods between laboratories, notably the use of spectral libraries.

Targeted identification of SUMOylation sites in human proteins using affinity enrichment and paralog-specific reporter ions.

Protein modification by small ubiquitin-like modifier (SUMO) modulates the activities of numerous proteins involved in different cellular functions such as gene transcription, cell cycle, and DNA repair. Comprehensive identification of SUMOylated sites is a prerequisite to determine how SUMOylation regulates protein function. However, mapping SUMOylated Lys residues by mass spectrometry (MS) is challenging because of the dynamic nature of this modification, the existence of three functionally distinct human SUMO paralogs, and the large SUMO chain remnant that remains attached to tryptic peptides. To overcome these problems, we created HEK293 cell lines that stably express functional SUMO paralogs with an N-terminal His6-tag and an Arg residue near the C terminus that leave a short five amino acid SUMO remnant upon tryptic digestion. We determined the fragmentation patterns of our short SUMO remnant peptides by collisional activation and electron transfer dissociation using synthetic peptide libraries. Activation using higher energy collisional dissociation on the LTQ-Orbitrap Elite identified SUMO paralog-specific fragment ions and neutral losses of the SUMO remnant with high mass accuracy (< 5 ppm). We exploited these features to detect SUMO modified tryptic peptides in complex cell extracts by correlating mass measurements of precursor and fragment ions using a data independent acquisition method. We also generated bioinformatics tools to retrieve MS/MS spectra containing characteristic fragment ions to the identification of SUMOylated peptide by conventional Mascot database searches. In HEK293 cell extracts, this MS approach uncovered low abundance SUMOylated peptides and 37 SUMO3-modified Lys residues in target proteins, most of which were previously unknown. Interestingly, we identified mixed SUMO-ubiquitin chains with ubiquitylated SUMO proteins (K20 and K32) and SUMOylated ubiquitin (K63), suggesting a complex crosstalk between these two modifications.

Splice variants of the dual specificity tyrosine phosphorylation-regulated kinase 4 (DYRK4) differ in their subcellular localization and catalytic activity.

Dual specificity tyrosine phosphorylation-regulated kinases, DYRKs, are a family of conserved protein kinases that play key roles in the regulation of cell differentiation, proliferation, and survival. Of the five mammalian DYRKs, DYRK4 is the least studied family member. Here, we show that several splice variants of DYRK4 are expressed in tissue-specific patterns and that these variants have distinct functional capacities. One of these variants contains a nuclear localization signal in its extended N terminus that mediates its interaction with importin α3 and α5 and that is capable of targeting a heterologous protein to the nucleus. Consequently, the nucleocytoplasmic mobility of this variant differs from that of a shorter isoform in live cell imaging experiments. Other splicing events affect the catalytic domain, including a three-amino acid deletion within subdomain XI that markedly reduces the enzymatic activity of DYRK4. We also show that autophosphorylation of a tyrosine residue within the activation loop is necessary for full DYRK4 kinase activity, a defining feature of the DYRK family. Finally, by comparing the phosphorylation of an array of 720 peptides, we show that DYRK1A, DYRK2, and DYRK4 differ in their target recognition sequence and that preference for an arginine residue at position P -3 is a feature of DYRK1A but not of DYRK2 and DYRK4. Therefore, we highlight the use of subcellular localization as an important regulatory mechanism for DYRK proteins, and we propose that substrate specificity could be a source of functional diversity among DYRKs.

Deep learning boosts sensitivity of mass spectrometry-based immunopeptidomics.

Characterizing the human leukocyte antigen (HLA) bound ligandome by mass spectrometry (MS) holds great promise for developing vaccines and drugs for immune-oncology. Still, the identification of non-tryptic peptides presents substantial computational challenges. To address these, we synthesized and analyzed >300,000 peptides by multi-modal LC-MS/MS within the ProteomeTools project representing HLA class I & II ligands and products of the proteases AspN and LysN. The resulting data enabled training of a single model using the deep learning framework Prosit, allowing the accurate prediction of fragment ion spectra for tryptic and non-tryptic peptides. Applying Prosit demonstrates that the identification of HLA peptides can be improved up to 7-fold, that 87% of the proposed proteasomally spliced HLA peptides may be incorrect and that dozens of additional immunogenic neo-epitopes can be identified from patient tumors in published data. Together, the provided peptides, spectra and computational tools substantially expand the analytical depth of immunopeptidomics workflows.

Peptide microarrays for profiling of modification state-specific antibodies.

The reversible phosphorylation of serine, threonine, and tyrosine residues is one of the most important intracellular post-translational modifications regulating enzymatic activities and protein/protein interaction in eukaryotic cells. Tools for determining phosphorylation status of proteins and peptides play a prominent role in signal transduction research and proteomics. Pan-specific antibodies claimed to recognize modified amino acid residues independent on the nature of surrounding residues in peptides and proteins are widely used. We used high-content phosphopeptide microarrays and microarrays displaying acetyllysine-containing peptides for comprehensive characterization of commercially available generic anti-phosphopeptide and anti-acetyllysine antibodies. We were able to demonstrate distinct subsite specificity and cross-reactivity for such antibodies.

Probing the epitope signatures of IgG antibodies in human serum from patients with autoimmune disease.

High density peptide microarray technologies can be applied in experimental medicine in general and in clinical immunology in particular. Laboratory diagnostics of autoimmune diseases strongly rely on screening human sera for antibodies against known autoantigens. These assays are still difficult to standardize and quantify. Typically, the results are presented as antibody titers within an assay system. Most assays use recombinant or purified autoantigens that are difficult to obtain and require great efforts of quality control. Here we describe a method to obtain patterns of epitope signatures with peptide microarrays from patients suffering from autoimmune diseases in comparison with healthy controls. One of the final aims will be to define subsets of peptides indicative for marker autoantibodies of autoimmune diseases. Finally, informative epitopes can be used for immunopurifying epitope-specific autoantibodies. Eventually, these antibodies can be further characterized on peptide microarrays displaying mutated epitopes obtained by scanning mutagenesis. Any disease or physiological status that affect humoral immune responses such as autoantibodies in oncology or responses to infections or vaccinations can be monitored.

A non-functional neoepitope specific CD8+ T-cell response induced by tumor derived antigen exposure in vivo.

Cancer-associated mutations, mostly single nucleotide variations, can act as neoepitopes and prime targets for effective anti-cancer T-cell immunity. T cells recognizing cancer mutations are critical for the clinical activity of immune checkpoint blockade (ICB) and they are potent vaccine antigens. High frequencies of mutation-specific T cells are rarely spontaneously induced. Hence, therapies that broaden the tumor specific T-cell response are of interest. Here, we analyzed neoepitope-specific CD8+ T-cell responses mounted either spontaneously or after immunotherapy regimens, which induce local tumor inflammation and cell death, in mice bearing tumors of the widely used colon carcinoma cell line CT26. A comprehensive immune reactivity screening of 2474 peptides covering 628 transcribed CT26 point mutations was conducted. All tested treatment regimens were found to induce a single significant CD8+ T-cell response against a non-synonymous D733A point mutation in the Smc3 gene. Surprisingly, even though Smc3 D733A turned out to be the immune-dominant neoepitope in CT26 tumor bearing mice, neither T cells specific for this neoepitope nor their T cell receptors (TCRs) were able to recognize or lyse tumor cells. Moreover, vaccination with the D733A neoepitope did not result in anti-tumoral activity despite induction of specific T cells. This is to our knowledge the first report that neoepitope specific CD8+ T cells primed by tumor-released antigen exposure in vivo can be functionally irrelevant.

Envelope-Specific Recognition Patterns of HIV Vaccine-Induced IgG Antibodies Are Linked to Immunogen Structure and Sequence.

Background: A better understanding of the parameters influencing vaccine-induced IgG recognition of individual antigenic regions and their variants within the HIV Envelope protein (Env) can help to improve design of preventive HIV vaccines. Methods: Env-specific IgG responses were mapped in samples of the UKHVC003 Standard Group (UK003SG, n = 11 from UK) and TaMoVac01 (TMV01, n = 17 from Tanzania) HIV vaccine trials. Both trials consisted of three immunizations with DNA, followed by two boosts with recombinant Modified Vaccinia Virus Ankara (MVA), either mediating secretion of gp120 (UK003SG) or the presentation of cell membrane bound gp150 envelopes (TMV01) from infected cells, and an additional two boosts with 5 µg of CN54gp140 protein adjuvanted with glucopyranosyl lipid adjuvant (GLA). Env immunogen sequences in UK003SG were solely based on the clade C isolate CN54, whereas in TMV01 these were based on clades A, C, B, and CRF01AE. The peptide microarray included 8 globally representative Env sequences, CN54gp140 and the MVA-encoded Env immunogens from both trials, as well as additional peptide variants for hot spots of immune recognition. Results: After the second MVA boost, UK003SG vaccinees almost exclusively targeted linear, non-glycosylated antigenic regions located in the inter-gp120 interface. In contrast, TMV01 recipients most strongly targeted the V2 region and an immunodominant region in gp41. The V3 region was frequently targeted in both trials, with a higher recognition magnitude for diverse antigenic variants observed in the UK003SG (p < 0.0001). After boosting with CN54gp140/GLA, the overall response magnitude increased with a more comparable recognition pattern of antigenic regions and variants between the two trials. Recognition of most immunodominant regions within gp120 remained significantly stronger in UK003SG, whereas V2-region recognition was not boosted in either group. Conclusions: IgG recognition of linear antigenic Env regions differed between the two trials particularly after the second MVA boost. Structural features of the MVA-encoded immunogens, such as secreted, monomeric gp120 vs. membrane-anchored, functional gp150, and differences in prime-boost immunogen sequence variability most probably contributed to these differences. Prime-boosting with multivalent Env immunogens during TMV01 did not improve variant cross-recognition of immunodominant peptide variants in the V3 region.

Validation of peptide epitope microarray experiments and extraction of quality data.

INTRODUCTION: Within the last decade, the development of antigen microarray slides has enabled the simultaneous measurement of serum reactivity to hundreds of peptides in a single biological sample. Despite this considerable scientific progress, many issues remain regarding the quality, analysis and interpretation of the data these slides produce. There is currently no accepted approach to guide data analysis, and researchers use a wide variety of statistical methods and software tools. We designed and implemented a laboratory experiment to assess the reliability and range of measurement of peptide microarray data, and present graphical and statistical procedures for pre-processing so that quality data can be extracted for addressing biological hypotheses. METHODS: Synthetic peptides spanning the proteins Ag85A, Ag85B, CFP10, MPT51/MPB51, TB10.4 and ESAT-6 were chosen as a paradigm to screen for serum reactivity to Mycobacteria tuberculosis (MTB). We explored various quantitative and graphical methods for presenting the responses from a slide. We replicated assays of samples from five TB-positive individuals to examine reproducibility, and used linear mixed models to investigate the various sources of variability, and to assess the range of measurement. We use our methods to extract data from the five TB-positive individuals and five healthy controls, and analyse the "normalized" responses using the freely available SAM package. RESULTS: The ratio of foreground to background signal (on a log scale) provides an appropriate response index. A two-dimensional graphical display clearly illustrates the responses from the control and peptide features on a slide. Mixed model analysis of the replicated slides found a high reproducibility of the assay between operators, days and experiments. The range of measurement was also satisfactory. Our analysis of the normalized responses from the five TB-positive patients and five healthy controls suggested that 10 of the 363 peptides assessed had significantly higher responses in the TB-positive group. CONCLUSIONS: Carefully designed laboratory experiments and rigorous statistical analysis can enable the removal of technical artefacts to produce quality peptide array data for addressing biological hypotheses. These instruments, which enable valid comparisons across slides and/or batches of slides, will escort future comparative analyses targeting high content serum reactivity profiling against a broad array of B-cell epitopes.

Serological profiling of the EBV immune response in Chronic Fatigue Syndrome using a peptide microarray.

BACKGROUND: Epstein-Barr-Virus (EBV) plays an important role as trigger or cofactor for various autoimmune diseases. In a subset of patients with Chronic Fatigue Syndrome (CFS) disease starts with infectious mononucleosis as late primary EBV-infection, whereby altered levels of EBV-specific antibodies can be observed in another subset of patients. METHODS: We performed a comprehensive mapping of the IgG response against EBV comparing 50 healthy controls with 92 CFS patients using a microarray platform. Patients with multiple sclerosis (MS), systemic lupus erythematosus (SLE) and cancer-related fatigue served as controls. 3054 overlapping peptides were synthesised as 15-mers from 14 different EBV proteins. Array data was validated by ELISA for selected peptides. Prevalence of EBV serotypes was determined by qPCR from throat washing samples. RESULTS: EBV type 1 infections were found in patients and controls. EBV seroarray profiles between healthy controls and CFS were less divergent than that observed for MS or SLE. We found significantly enhanced IgG responses to several EBNA-6 peptides containing a repeat sequence in CFS patients compared to controls. EBNA-6 peptide IgG responses correlated well with EBNA-6 protein responses. The EBNA-6 repeat region showed sequence homologies to various human proteins. CONCLUSION: Patients with CFS had a quite similar EBV IgG antibody response pattern as healthy controls. Enhanced IgG reactivity against an EBNA-6 repeat sequence and against EBNA-6 protein is found in CFS patients. Homologous sequences of various human proteins with this EBNA-6 repeat sequence might be potential targets for antigenic mimicry.

High-Throughput Microarray Incubations Using Multi-Well Chambers.

Peptide microarrays are ideal tools for a variety of applications ranging from epitope mapping to immune monitoring. Here we present a method for high-throughput screening of biological samples using only standard microtiter plate equipment. Parallel incubation of a large number of samples with a small library of peptides is enabled by printing multiple identical mini-arrays on one microarray slide and further combining four slides to yield an incubation frame possessing the dimensions of a 96-well microtiter plate. Applying conventional lab equipment such as ELISA washers, hundreds of samples can be processed in 1 day yielding approx. 200 data points in triplicates per sample.

Antibody Responses After Analytic Treatment Interruption in Human Immunodeficiency Virus-1-Infected Individuals on Early Initiated Antiretroviral Therapy.

The examination of antibody responses in human immunodeficiency virus (HIV)-1-infected individuals in the setting of antiretroviral treatment (ART) interruption can provide insight into the evolution of antibody responses during viral rebound. In this study, we assessed antibody responses in 20 subjects in AIDS Clinical Trials Group A5187, wherein subjects were treated with antiretroviral therapy during acute/early HIV-1 infection, underwent analytic treatment interruption, and subsequently demonstrated viral rebound. Our data suggest that early initiation of ART arrests the maturation of HIV-1-specific antibody responses, preventing epitope diversification of antibody binding and the development of functional neutralizing capacity. Antibody responses do not appear permanently blunted, however, because viral rebound triggered the resumption of antibody maturation in our study. We also found that antibody responses measured by these assays did not predict imminent viral rebound. These data have important implications for the HIV-1 vaccine and eradication fields.