3. Results and discussion
3.1. The approach
To explore and determine the variety of post-translational modifications (PTMs) present on proteins in human dry blood spots, a comprehensive analysis was conducted. The dataset consisted of a large number of scans (95516) and features (262910) obtained from the samples. Table 2 provides an overview of the data statistics, showcasing the abundance of information captured in the analysis.
The results show that a total of approximately 3,374 unique peptides were identified in the study. Of the 3,374 peptides identified, they corresponded to approximately 354 different protein groups corresponding to 151 unique proteins ( Sup Table 2).The list of 151 proteins includes various molecules involved in important biological processes such as neutrophil aggregation, fibrinolysis, and complement activation via the alternative pathway. Neutrophil aggregation refers to the process by which neutrophils, a type of white blood cell, come together and form aggregates. This process plays a crucial role in the immune response, particularly in the context of inflammation and host defense against pathogens. The proteins in this list likely contribute to the molecular mechanisms underlying neutrophil aggregation, facilitating their proper function in immune responses.
Fibrinolysis is the process of breaking down fibrin, a protein involved in blood clot formation, to prevent excessive clotting and promote the dissolution of existing clots. The proteins in this list are likely involved in regulating the fibrinolysis process, which is essential for maintaining proper hemostasis and preventing thrombotic disorders. Complement activation via the alternative pathway is one of the pathways through which the complement system, a part of the immune system, can be activated. The alternative pathway provides an immediate response to invading pathogens and contributes to the elimination of microbial infections. The proteins in this list likely participate in the activation and regulation of the alternative pathway, enabling effective immune responses against pathogens.
In total, 976 modified peptides were identified across the samples. Each PTM was further manually validated through PTM-specific neutral loss and a, b and/or y fragmentation ions series. Notably, Carbamidomethylation emerged as the most prevalent PTM, which is expected given the sample preparation method involving derivatization for trypsin digestion. The identification of numerous Carbamidomethylation peptides reinforces the robustness of the analysis. Interestingly, it was found that half of the proteins exhibited more than one modification, indicating the complexity and diversity of PTMs in the human dry blood spot proteome (Sup Table 3). Figure 2 illustrates that approximately 46% of the proteins showed additional modifications, including formylation or deamidation. These findings highlight the presence of multiple layers of protein modifications in the analyzed samples, suggesting intricate regulatory mechanisms at play.
The characterization of PTMs in the human dry blood spot proteome not only provides valuable insights into the complexity of protein modifications but also offers potential avenues for understanding their functional implications and their association with various physiological and pathological processes. Further investigations are warranted to elucidate the specific roles and clinical relevance of these PTMs, paving the way for their potential application as biomarkers and therapeutic targets in the future.
3.2. PTMs in the samples
The analysis of PTMs in the samples revealed a diverse array of modifications present on proteins. Among the identified PTMs, several were found to be highly representative across all samples (Sup Table). These PTMs contribute to the functional diversity and complexity of the proteome in the human dry blood spot. One of the prominent PTMs observed was Carbamidomethylation, which was consistently detected in the majority of proteins. This modification involves the addition of a carbamidomethyl group to cysteine residues, thereby stabilizing the protein structure and preventing oxidation. Carbamidomethylation in conjunction with other PTMs such as Acetylation (K), Deamidation (NQ), and Oxidation (M) was also observed, indicating the presence of complex modification patterns on the proteins. Another notable PTM observed was Formylation, which involves the addition of a formyl group to certain amino acid residues. This modification has been associated with various biological processes, including protein degradation and immune responses. Additionally, Deamidation (NQ) was found to be prevalent, which results in the conversion of asparagine (N) and glutamine (Q) residues to aspartic acid (D) and glutamic acid (E), respectively. Other PTMs identified in the analysis included Methylation (KR), Acetylation (N-term), Pyro-glu from E and Q, Oxidation (HW and M), Tryptophan oxidation to kynurenin, and various other modifications such as Acetaldehyde, Carboxyethyl, Propionaldehyde, Sulphone, and Quinone (Sup Table). These PTMs contribute to the structural and functional diversity of proteins, potentially influencing their stability, enzymatic activity, and molecular interactions.
The presence of multiple modifications on proteins, as well as the variety of PTMs observed, underscores the intricate nature of protein regulation and signaling in the human dry blood spot proteome. Further exploration of these modifications and their functional implications will deepen our understanding of their roles in physiological and pathological processes, and may lead to the discovery of novel biomarkers and therapeutic targets.
Citrullination, for example, is a PTM known as deamidation, which is observed in ~11% of cases and involves the conversion of the amino acid arginine to citrulline by the hydrolysis of nitrogen atoms in one of its side chains. In this irreversible PTM, the guanidium group of the arginine residue is converted to a ureido group.
The reaction leads to the production of the non-standard amino acid citrulline, ammonia release and the loss of a positive charge with a monoisotopic mass contrast of +0.984016, lowering the overall charge of the protein each time the reaction takes place [17]. This irreversible modification leads to the loss of a positive charge and the production of a non-standard amino acid, citrulline. Tandem analysis MS (MS /MS) was used to verify the presence of deamidation in the identified peptides. By examining the fragmentation pattern of the peptide ions, characteristic fragments indicative of deamidation can be identified. Manual inspection also revealed that the Cit effect was the most prominent fragmentation pathway when Cit was followed by a His or Pro residue, which is due to an additive His and Pro effect leading to suppressed fragment ion intensity for other ions [18]. The enzymatic reaction responsible for citrullination is catalyzed by a group of enzymes called peptidylarginine deiminases (PADs), which are calcium-dependent and include PAD 1-4 and PAD 6 in the human body [19]. The conversion of arginine to citrulline during citrullination has significant implications for protein properties. Citrullination reduces the overall charge of the protein and affects its isoelectric point, hydrogen bonding capacity, and charge distribution. Additionally, citrullination can increase the local and global hydrophobicity of the modified peptide, potentially influencing protein folding, polarity, and susceptibility to enzymatic degradation [20]. Physiologically, citrullination is involved in various processes such as gene regulation, apoptosis, epidermal terminal differentiation, immune system functions, and skin physiology. It has been implicated in several diseases, including multiple sclerosis, rheumatoid arthritis, Alzheimer’s disease, cancer, central nervous system disorders, and cardiovascular disease. Citrullinated proteins have been identified in cellular proteins like histones, vimentin, and filaggrin, as well as extracellular matrix proteins such as collagens. The association between citrullination and these diseases highlights the role of this PTM in disease pathogenesis [21, 22]. Furthermore, emerging research suggests that citrullination may have implications in epigenetic regulation, as it can directly modify histones or interact with citrullinated histone-modifying enzymes and coactivators [23]. The physiological significance and involvement of citrullination in both normal and pathological conditions underscore its importance in cellular processes and disease mechanisms. Ongoing research continues to explore the biological roles of citrullinated proteins, the pathologies associated with citrullination, and the methodologies employed for their enrichment and detection [19, 24-27].
In this study, we aimed to demonstrate the effectiveness of our strategy by analyzing samples from patients with coronary arterial disease (CAD). By focusing on scientific research, we identified several genes, namely TGM2, PON1, APEH, ANK1, HSPA8, APOC1, YWHAZ, NME1, and FBXO7, that have been implicated in various aspects of CAD. TGM2 is associated with CAD due to its involvement in the formation of advanced glycation end products (AGEs) and the remodeling of the extracellular matrix. These processes are linked to atherosclerosis and plaque stability[28].
In addition, HSPA8, also known as Hsc70, is a heat shock protein involved in protein folding. Its association with CAD stems from its involvement in cellular stress responses, as well as its potential roles in inflammation and plaque formation. The presence of HSPA8 in CAD suggests that it may play a role in the pathogenesis of the disease.
Furthermore, FBXO7 is another gene associated with CAD. It is linked to CAD risk factors such as blood pressure regulation and inflammation, indicating its potential involvement in the development and progression of CAD[29]. Moreover, APOC1, a component of lipoproteins, has also been linked to CAD. Its involvement in lipid metabolism, inflammation, and regulation of cholesterol transport suggests that it may contribute to the development of atherosclerosis and the progression of CAD. These genes, including HSPA8, FBXO7, and APOC1, highlight the multifactorial nature of CAD and emphasize the complex interplay of various biological processes in the pathogenesis of the disease. Further research is needed to fully elucidate the specific mechanisms by which these genes contribute to CAD and to explore their potential as therapeutic targets[30],[31].Additionally, PON1, an enzyme associated with HDL particles, has been linked to CAD due to its crucial role in the metabolism of oxidized lipids. This metabolic process contributes to the development of atherosclerosis, a key factor in CAD progression.
Moreover, APEH (Acylpeptide hydrolase) has been suggested to play a significant role in CAD. It exerts its influence by influencing peptide metabolism and modifying the stability and accumulation of atherosclerotic plaques[32]. APEH’s involvement in these processes highlights its potential contribution to the pathogenesis of CAD.
The inclusion of PON1[33] and APEH in the list of implicated genes emphasizes the diverse molecular mechanisms underlying CAD development. These findings underscore the complex interplay of lipid metabolism, peptide metabolism, and plaque formation in the pathophysiology of CAD. Further research is necessary to deepen our understanding of the precise roles of PON1 and APEH and their potential as therapeutic targets for CAD management.
Furthermore, variants of ANK1 have been linked to CAD risk factors, including elevated levels of LDL cholesterol and triglycerides. The alterations in ANK1 can also have an impact on red blood cell morphology, which has the potential to affect blood flow. These associations highlight the significance of ANK1 in the pathogenesis of CAD, shedding light on the influence of lipid metabolism and blood cell characteristics in the development of the disease[34]. In addition, YWHAZ has been implicated in CAD due to its involvement in signaling pathways associated with vascular smooth muscle cell proliferation and migration. These cellular processes play a crucial role in the development of atherosclerosis, a hallmark of CAD. The participation of YWHAZ in these signaling pathways underscores its potential contribution to the pathogenesis of CAD[35].
The inclusion of ANK1 and YWHAZ among the genes implicated in CAD emphasizes the multifaceted nature of the disease, involving lipid metabolism, blood cell characteristics, and cellular signaling. Understanding the precise mechanisms by which ANK1 and YWHAZ contribute to CAD progression can provide valuable insights for the development of targeted therapies and interventions. Further research is warranted to uncover the full extent of their roles in CAD and their potential as therapeutic targets. It is important to note that while these genes have demonstrated associations with CAD, the specific mechanisms through which they contribute to the disease are still under investigation. Further research is necessary to fully comprehend their roles and potential therapeutic implications in CAD.
Glycosylation, as a major posttranslational modification (PTM), plays a critical role in shaping protein function, stability, and interactions and is therefore an interesting area for disease-related studies. Previous successes in identifying different glycosylation patterns, such as monosaccharides linked to different sugars (topoisomers, e.g., triantennary GlcNAc or bisecting GlcNAc), and variations in linkage positions (linkage isomers, e.g., α2-3-sialic acid versus α2-6-sialic acid), have highlighted their potential importance in proteomic studies[36]. However, despite efforts to look for such glycosylation patterns in this study, the results did not yield positive results. It is possible that the use of a relatively small volume of DBS (10 μl) may have contributed to the lack of glycosylation results[37].
To investigate the potential effects of PTMs and their relevance to CAD, volcano plot analysis was performed, as shown in Table 3 and Figure 4. The aim of volcano plot analysis was to identify proteins that showed changes in abundance in the DBS samples associated with CAD. These scatter plots were generated by plotting the log2 fold change in protein abundance against the negative logarithm (base 10) of the p value, which allowed visualization of statistically significant changes in a large proteomic dataset (Figure 4). Considering fold-change variations with a p-value of less than 0.05, a total of 42 peptides were identified that differed significantly in abundance. These peptides, along with their corresponding proteins, are detailed in Table 3. Volcano plot analysis provided valuable insights into the differential expression of proteins in DBS samples with CAD, which may shed light on the involvement of PTMs in disease development. The statistically significant changes observed in these proteins highlight their potential as candidates for further investigation as biomarkers or mechanistic players associated with CAD. These results highlight the power of volcano diagram analysis in proteomic studies, particularly with large data sets, and underscore the importance of understanding PTMs in the context of disease processes. Further studies and validation efforts are warranted to confirm and explore the functional implications of these identified PTMs in the context of CAD and their potential clinical relevance.
In our study, we observed the presence of deamidation in the peptide serotransferrin at the peptide K.IMN(+.98)GEADAMSLDGGFVYIAGK.C. This finding is consistent with the phenomenon of asparagine deamidation known to occur during protein aging and/or oxidative stress. During asparagine deamidation, asparagine residues are converted to aspartic acid or isoaspartic acid, resulting in changes in the overall charge states of proteins [38]. In particular, deamidation of specific residues, such as Asn-943 in protein Cp (ceruloplasmin), has been associated with oxidative stress-induced changes that may affect its ferroxidase activity. This highlights the potential functional implications of deamidation events in proteins [39].
While the clinical and diagnostic application of mass spectrometry (MS) for post-translational modification (PTM) detection has experienced significant advancements [40], it is important to acknowledge the existing caveats and recognize the opportunities for further progress in mass spectrometry-based clinical investigations. One major caveat is the challenge posed by sample preparation, which remains a labor-intensive process prone to variations in concentration and matrix background. These variations introduce complexity and potential bias into the analysis, limiting the reliability and reproducibility of results. Additionally, a considerable amount of operating time for mass spectrometers is dedicated to maintenance tasks such as cleaning and regenerating high-performance liquid chromatography (HPLC) columns. This maintenance requirement reduces instrument productivity and hampers the throughput of mass spectrometry-based clinical investigations. Moreover, the data analysis of mass spectrometric spectra is a highly complex and time-consuming task that demands the expertise of experienced operators. Current biostatistical tools available for PTM-centric mass spectrometric data analysis are primarily designed for scientific applications and lack universal applicability, reproducibility, automation, and high throughput [41]. This limitation impedes the clinical translation of mass spectrometry-based diagnostic approaches, as it hinders the efficient and standardized analysis of data. However, these caveats also present opportunities for improvement and advancement in mass spectrometry-based clinical investigations. Significant efforts should be directed towards the automation of sample preparation processes, reducing labor intensity and enhancing the reproducibility of results. Furthermore, the development of standardized protocols and regulatory adaptations can optimize mass spectrometer instrument usage, minimizing idle time and increasing overall productivity. To address the challenges associated with data analysis, there is a need for the development of advanced biostatistical tools specifically tailored for PTM-centric mass spectrometric data analysis in clinical settings. These tools should prioritize universal applicability, reproducibility, automation, and high throughput capabilities. By streamlining and standardizing the data analysis process, the clinical translation of mass spectrometry-based diagnostic approaches can be facilitated.