Mass spectrometry based protein biomarkers and drug target discovery and clinical diagnosis in Age-Related progressing neurodegenerative diseases

Drug Metabolism Reviews ISSN: (Print) (Online) Journal homepage: https://www.tandfonline.com/loi/idmr20

Mass spectrometry based protein biomarkers and drug target discovery and clinical diagnosis in AgeRelated progressing neurodegenerative diseases Ishita Agrawal, Pallavi Tripathi & Shyamasri Biswas To cite this article: Ishita Agrawal, Pallavi Tripathi & Shyamasri Biswas (2022) Mass spectrometry based protein biomarkers and drug target discovery and clinical diagnosis in Age-Related progressing neurodegenerative diseases, Drug Metabolism Reviews, 54:1, 22-36, DOI: 10.1080/03602532.2022.2029475 To link to this article: https://doi.org/10.1080/03602532.2022.2029475

Full Terms & Conditions of access and use can be found at https://www.tandfonline.com/action/journalInformation?journalCode=idmr20 DRUG METABOLISM REVIEWS 2022, VOL. 54, NO. 1, 22–36 https://doi.org/10.1080/03602532.2022.2029475

REVIEW ARTICLE

Mass spectrometry based protein biomarkers and drug target discovery and clinical diagnosis in Age-Related progressing neurodegenerative diseases Ishita Agrawala, Pallavi Tripathib and Shyamasri Biswasc a Molecular and Human Genetics, Institute of Science, Banaras Hindu University, Varanasi, India; bManchester Institute of Biotechnology, The University of Manchester, Manchester, UK; cUSA Prime Biotech LLC, Gainesville, FL, USA

ABSTRACT ARTICLE HISTORY

Neurodegenerative diseases correspond to overly complex health disorders that are driven by intersecting pathophysiology that are often trapped in vicious cycles of degeneration and cognitive decline. The usual diagnostic route of these diseases is based on postmortem examination that involves identifying pathology that is specific to the disease in the brain. However, in such cases, accurate diagnosis of the specific disease is limited because clinical disease presentations are often complex that do not easily allow to discriminate patient’s cognitive, behavioral, and functional impairment profiles. Additionally, an early identification and therapeutic intervention of these diseases is pivotal to slow the progression of neurodegeneration and extend healthy life span. Mass spectrometry-based techniques have proven to be hugely promising in biological sample analysis and discovery of biomarkers including protein and peptide biomarkers for potential drug target discovery. Recent studies on these biomarkers have demonstrated their potential for applications in early diagnostics and identifying therapeutic targets to battle against neurodegenerative diseases. In this review, we have presented principles of mass spectrometry (MS) and the associated workflows in analyzing and imaging biological samples for discovery of biomarkers. We have especially focused on age-related progressing neurodegenerative diseases such as Alzheimer’s (AD) and Parkinson’s disease (PD), Amyotrophic Lateral Sclerosis (ALS) and Frontotemporal dementia (FTD) and the related MS-based biomarker developments for these diseases. Finally, we present a future perspective discussing the potential research directions ahead.

Introduction Lately, a significant amount of research attention has been devoted to the discovery of new biomarkers. This research is primarily driven with the goal to meet the ever-growing demand for the development of clinical diagnostic tests to combat complex diseases (Crutchfield et al. 2016). These clinical diagnostic tests are considered important from healthcare point of view since these tests can provide early intervention in complex medical conditions such as neurodegeneration, cancer or cardiovascular diseases. It is believed that such an early intervention especially with laboratory medicine can potentially be immensely helpful in clinical decision making that can result in therapeutic interventions and improved clinical outcomes for patients €zery (Conrads et al. 2006; Dolci and Panteghini 2006; Fu et al. 2013; Li and Chan 2014). Further, in the clinical and pharmaceutical sectors, the success of the clinical outcomes in addition to the biomarkers largely depends on estimating the efficacy CONTACT Shyamasri Biswas

ß 2022 Informa UK Limited, trading as Taylor & Francis Group KEYWORDS

Neurodegeneration; cerebrospinal fluid; plasma; proteomics; biomarkers; drug targets; LC-MS/MS; MALDI-TOF

and safety of a drug compound (Buchberger et al. 2018). To accomplish this task, it is imperative to establish knowledge about the specific location of the drug compound as well as the concentration that is present at these points within the target tissue. This knowledge gives the ability to detect not only the level of exposure, but also the spatial distribution within the target tissue. The knowledge about complex biochemical interactions between a drug and target tissue eventually helps to achieve the goal of effective therapeutic interventions (Aikawa et al. 2016; Yoon and Lee 2018). New analytical capabilities based on mass spectrometry (MS) have shown tremendous promise for the development of biomarkers. Such analytical capabilities are pivotal to study chemical composition of molecules as well as spatial distribution of analytes on complex surfaces (Crutchfield et al. 2016; Buchberger et al. 2018). Previous studies focused on contemporary MS technology including metabolomics and proteomics for the biological studies of human metabolites and proteins that resulted in exciting insights into human

Figure 1. A conceptual depiction of the application of mass spectrometry in combating the challenges of age-related neurodegenerative diseases.

biology. The current efforts are more focused on employing advanced mass spectrometers with broader technological capabilities for translating biomarkers into clinical tests (Crutchfield et al. 2016). To this end, advanced MS has been employed for the development of highly sensitive and accurate novel protein biomarkers for their dual roles in diagnosis and also as therapeutic targets that has impacted the field of proteomics. Figure 1 presents conceptual illustration of the application of MS in age-related neurodegenerative diseases. However, the challenges of biomarker discovery depend on the underlying proteome sampling and bioinformatics processing strategies in addition to the applications of MS to achieve the goals (Conrads et al. 2006). There have also been studies on mass spectrometry imaging (MSI) apart from the applications of MS in biomarkers discovery. MSI has been employed to reveal information on the spatial distribution for the compound in the target tissue. MSI can rapidly localize, detect and identify molecules from the most complex, biological sample surfaces (Chughtai and Heeren 2010). Studies have shown that using MSI, it is possible to do label-free imaging of thousands of molecules, such as metabolites, lipids, peptides, proteins, DNA and glycans in a single experiment. This also includes simultaneously done multivariable analysis of various compounds. Additionally, there are other advantages of using MSI that include spatial resolution of high order, sensitivity to the sample surface and the ability to do statistical analysis. The combined ability of MSI to map a wide range of small molecules with high spatial resolution as well as to quantify them at once can be leveraged for the identification of additional biomarkers that are linked to pathophysiology involving neurodegenerative disorders. These types of results can be eventually translated into blood, urine or cerebrospinal fluid ‘CSF’ based biomarkers. MSI has, therefore, proven to be very

useful in characterization and analysis of metabolites and drug compounds toward biomarkers discovery for applications in neurodegenerative disorders (Buchberger et al. 2018; Yoon and Lee 2018). More recently, matrix-assisted laser desorption/ionization mass spectrometry imaging (MALDI MSI) has emerged as a key technology for label-free bio-imaging and analysis of the spatial distribution of complex biological samples that include biomolecules, pharmaceuticals, and other xenobiotics in tissue sections (Schulz et al. 2019). It is of consensus that the applications of the MALDI-MSI technology could impact biomarker research that could potentially take drug disposition analysis to the next level leading to new therapeutics and drug discovery to battle against complex diseases (Schulz et al. 2019; Richard et al. 2020). In this review, we have emphasized on major agerelated neurodegenerative diseases such as Alzheimer’s disease (AD), Parkinson’s diseases (PD) and Amyotrophic Lateral Sclerosis (ALS) and the biomarkers and imaging techniques based on mass spectrometry. We have described the important workflows of mass spectrometry along with some of the recently demonstrated notable applications of MS based biomarkers targeted at interventions in neurodegenerative disorders. Finally, we have briefly discussed a future perspective.

Analysis of Clinically-Grade biological samples Mass spectrometry (MS) is a powerful, multi-faceted technique to analyze a range of clinically-important biological and pharmaceutical samples. Since its first introduction in analytical chemistry, MS has evolved with multiple technical adjustments over the years that include gaining superior capabilities to ionize, scan, focus, fragment and detect chemical structures, to name a few. All these technical improvements have made MS more advanced and modern that can analyze proteins, peptides, lipids, biofluids and tissue and other pharmaceuticals and their metabolites in biological matrices. It has been shown that high-resolution mass spectrometry (HRMS) can identify clinically important analytes based on exactly accurate masses that can provide more sensitive, selective, robust, and repeatable analyses and also quantification (Loos et al. 2016; Baghel et al. 2017).

Protein biomarker discovery and analysis The main research focus over the last decade has been on the search for protein biomarkers. The search for clinically suitable protein biomarkers involves multiple technologies and research disciplines. An overview of

24 I. AGRAWAL ET AL. workflows of some common MS-based biomarker-discovery is presented in Figure 1. The workflows generally involve various experimental steps. These include collection and fractionation of the sample followed by recording mass spectra. Subsequently, the process involves data analysis and filtering by employing bioinformatics and finally validation of the results/data is done. With respect to collection of biological samples, healthy or diseased individuals donate blood plasma, serum, cerebral spinal fluid or urine that are looked at by the mass spectrometer (Hawkridge and Muddiman 2009). Biological samples are processed to remove abundant proteins. This is done to isolate certain protein classes that include molecular weight, hydrophobicity, and posttranslational modifications. Alternatively, this is also done to make the sample within appropriate size that is compatible with the machine and experimental requirements (Hawkridge and Muddiman 2009). However, it is also to be noted that processing and removal of high abundance proteins is not always necessary. Some more advanced multi-dimensional chromatography approaches allow for minimal processing and robust protein identification. While the workflows can be applied to all the biological fluids, researchers have paid greater attention to plasma due to its major significance in biomarker discovery (Muddiman 2006). For protein identification and relative quantification, the post-processed plasma samples are often directed to one or more MS systems. It is to be noted that for discovery-based proteomics, appropriate instruments are required in order to accurately identify masses that are intact as well as the masses that are fragmented (Hawkridge and Muddiman 2009). Further, it is possible to quantify proteins that are differentially expressed in the diseased samples and compared with the healthy samples. This also allows to evaluate to check the disease relevance of the proteins. Targeted assays can then be prepared of several protein samples for investigations using enzyme linked immunosorbent assay (ELISA) and/or selected reaction monitoring (SRM) (Hawkridge and Muddiman 2009). In general, there are two categories of workflows that include bottom-up and top-down processes. The major differences in equipment required for top-down and bottom-up proteomics are mostly instrument accessibility and size, complexities and cost of the instruments. For example, top-down proteomics often employs much more complicated and expensive magnets and ion-cyclotron instruments, while bottom-up proteomics can be achieved with more accessible instruments (e.g. ion/orbi traps and MALDI etc.). Many workflows come in the bottom-up category that deals

with the plasma proteins that are proteolytically digested into peptides and are more suitable to contemporary separation and MS technologies (Zhang et al. 1994; Wilm et al. 1996; Washburn et al. 2001). On the other hand, the less employed top-down approach has also been shown by the researchers to analyze intact plasma proteins. However, this approach is significantly more challenging since it provides difficulty of separating and mass spectrally analyzing intact proteins (Henry et al. 1989; Loo et al. 1990; Kelleher 2004). Therefore, the top-down approach has not been used by the researchers frequently. Further, it is to be noted that while bottom-up approach-based workflows are more commonly used than the top-down based approach, both approaches have been shown to be equally good. This is proven from the studies on validating MS-enabled biomarkers discovery (Hawkridge and Muddiman 2009). As there are several steps and complexities involved in the workflows and instruments for biomarker discovery (Figure 2), it is usually expected that the analytical variability, especially in the case of single-analyte assays comprising a large number of proteins obtained from a single sample. It is suggested that the measurement that deals with the overall variable analytics is considered. Identification of up or downregulated proteins in diseased versus healthy patients is the final objective (Hawkridge and Muddiman 2009).

Application of mass spectrometry in neurodegenerative diseases It is well understood that neurodegenerative diseases are prevalent and devastating and there is no known cure available to most of the diseases such as Alzheimer’s diseases (AD), Parkinson’s disease (PD), Amyotrophic Lateral Sclerosis (ALS) or Frontotemporal dementia (FTD). Even though there has been extensive research done over the past decades, there is still lack of comprehensive knowledge and understanding of the causes of neurodegeneration along with the therapeutic path to prevent it or slow the progression or reverse it. Researchers have employed different approaches including proteomics and systems biology approaches to investigate the interactions between genome, metabolome, and the environment to understand neurodegenerative pathogenesis. Among new technologies, mass spectrometry has emerged to facilitate such studies especially in diagnostic and therapeutic intervention arenas (Chen et al. 2019; Li et al. 2019; Zhou et al. 2020).

Figure 2. Procedures of standard workflows for mass spectrometry–based protein biomarker discovery and validation are shown (FTMS: Fourier transform mass spectrometry; HPLC: high-performance liquid chromatography; ICAT: isotope-coded affinity tag; IMAC: immobilized metal affinity chromatography; ITRAQ: isotopic tag for relative and absolute quantification; LTQ: linear trap quadrupole; MS/MS: tandem mass spectrometry; RP: reverse-phase liquid chromatography; SCX: strong cation-exchange liquid chromatography; TOF: time of flight. A formula for a small sampling of individual contributions to analytical variability is shown in the box at the bottom [With permission from Annu Rev Anal Chem (2009)].

Proteomics approaches for physiology, biomarker and drug target discovery in ALS and FTD Fast progressive neurodegenerative disorders such as ALS and FTD currently lack targeted therapeutics and have no cures. Studies have shown that neuronal cells become rich with protein inclusions due to non-clearance of proteins and this has been attributed one of the causes of these diseases. Despite notable research advances made in this area, the pathological

mechanisms behind these diseases are still not understood fully. Especially, the current knowledge is incomplete regarding the protein inclusions in the neuronal cells, and if these inclusions cause cellular degeneration (Hedl et al. 2019). It has been shown that in ALS, the motor neurons of the primary motor cortex and spinal cord are affected by the disease pathology (Foster and Salajegheh 2019). Whereas, the primary targets of degeneration in FTD have been shown to be neurons

26 I. AGRAWAL ET AL. of the frontal and temporal lobe (Olney et al. 2017). Researchers have conducted pathological characterizations of both ALS and FTD. These characterizations involved studying neurons having inclusion formation and aberrant protein aggregation. This led to establishing the link with neurodegeneration. Studies have also shown that the main cause of this disease comes from the aggregation of a number of important proteins that include superoxide dismutase 1 (SOD1), TAR DNA binding protein of 43 kDa (TDP-43), and protein tau (Hedl et al. 2019). An early detection and intervention of ALS and FTD is considered vital to slow the disease progression, especially in ALS to allow extending healthy life span. To this end, proteomics methods are considered promising. These methods are helping to shed light on the mechanisms related to the disease. Understanding the disease mechanisms is critically important to identify accurate biomarkers using clinical samples. Proteomicsbased approaches have proven to be very effective in investigating important proteins and obtaining highthroughput quantitative results that are of clinical interests (Hedl et al. 2019). Several studies have focused on understanding different functions of proteins, structures of proteins and the roles behind the appearance of specific proteins in the proteomics methods. Some notable progresses in proteomics technologies have paved the way for breakthroughs in gaining new insights into the proteins that are relevant from pathophysiology point of view as well as other proteins to reveal their roles in disease mechanisms. It is usually considered that body fluids that include blood, wound fluid, urine, and saliva are noninvasive and easy to collect and process compared to tissue biopsies. Therefore, proteomics approaches that employ body fluids are gaining considerable research interests. To this end, different types of clinical samples including blood, urine, tissue, saliva, and also tissue interstitial fluid were collected from each patient and were studied based on proteomics approaches to identify reliable biomarkers. Extraction and separation of proteins usually followed a range of proteomics techniques including 2-DE, LCM, and 2 DDIGE. Further, studies have been conducted to address the challenges that are usually faced with identification of low-abundance proteins in body fluid samples. Strategies that have been suggested include sample pre-fractionation, immune-affinity depletion of highly abundant proteins, and enrichment of posttranslational modifications (PTM) for proteome simplification of body fluids. It has been shown that proteases catalyze an irreversible PTM upon cleaving target proteins. This observation suggests that uncontrolled proteolysis is

associated with many diseases. This implies that proteolytic events are powerful indicators for disease progression. Therefore, it has been suggested that their specific identification in body fluids can be leveraged for establishment of novel biomarkers. Some other studies have suggested proteomics and multiplex immunoassay approaches for peripheral blood biomarker exploration (Patel et al. 2016; Sabino et al. 2017; Comes et al. 2018; Alharbi 2020). Thus, proteomic-based platforms employ overly complex protein screening technology to obtain vital information and understanding of biological systems. This information is then used in combination with genomic data to get insights into the fundamental biological mechanisms underlying neurodegenerative diseases (Hedl et al. 2019). There have been several studies in the past couple of decades that have led to the emergence of newer and sophisticated MS technology offering higher resolution and faster scan rates. Especially, improvements in technology from ion traps to orbitraps and also the application of bioinformatics data processing have broadly changed the field in last decades that has witnessed transitioning of the technology from peptide mass fingerprinting (PMF) to sequencing peptides. This superior technology that has evolved over the years has enabled smoother and quicker identification of complex proteomes with shorter analysis periods (Chapman et al. 2014). In the proteomics analysis, sample preparation procedures involve peptides that are obtained by digesting proteins. This is followed using trypsin and/or Lys-C, which is a protease. A reverse phase C18 liquid chromatographic separation along with analysis by mass spectrometry (LC-MS/MS) is subsequently done (Tholey and Becker 2017). Figure 3 shows the typical procedures for label-free quantitative proteomics for biomarkers for ALS and FTD. It shows the C18 column that is employed for peptides to be eluted. The subsequent process involves ionization and fragmentation for analysis and peptide identification under mass spectrometry. Further, algorithms can be employed to search the data generated by MS by incorporating protein databases to generate lists of identified proteins. To this end, researchers have demonstrated several quantitative proteomics methods. These include stable isotope labeling by amino acids in cell culture (SILAC) and tandem mass tagging (TMT), label-free quantitation (LFQ) and sequential windowed acquisition of all theoretical fragment ion mass spectra (SWATH-MS). These are considered important methods in studies of ALS and FTD (Hedl et al. 2019).

Figure 3. Standard protocol procedures for proteomics for label-free and labeling quantitation of proteins from complex mixtures relevant to gain insights into physiological mechanisms of ALS and FTD diseases. Model systems as well as human tissue can be employed to get protein samples. These proteins can then be labeled for targeted proteomics and analyzed for broader detection (DDA: data-dependent acquisition; DIA: data-independent acquisition; and SILAC: stable isotope labeling by amino acids in cell culture) [With permission from Front Neurosci. (2019)].

Previous studies have shown the advantages of LFQ that can use either data-dependent or data-independent acquisition (DIA) analyses. In the case of datadependent acquisition, quantitation has been performed on either spectral counts or spectrometric signal intensity of product ions. These ions are generated from selected precursors by the mass spectrometer (Yu et al. 2016). Using these principles, the dysfunctional similarities between ALS and FTD diseases were highlighted by the identification of hundreds of proteins obtained from both human ALS and FTD brain and spinal cord samples overlapping mitochondrial and

metabolic pathway alterations (Iridoy et al. 2018). Similar analysis of brain tissues performed in another study showed subtypes of ALS and FTD that distinguished between the diseases by highlighting differences in levels of numerous proteins. This also included differences in protein aggregate assembly, distribution, and morphology (Laferriere et al. 2019). This study involved purification of aggregated proteins that were insoluble in detergents. It employed a sequential biochemical extraction technique called ‘SarkoSpin’ that resulted in identification of proteins that had low solubility and associated with TDP-43 pathology. This also

resulted in the enrichment of phosphorylated TDP-43. SarkoSpin, method is developed to work as a purification assay for the isolation of pathological TDP-43 from patient brain tissue. This purification process is applicable to cell culture material and the process allows the study of biochemical properties of exclusively pathological TDP-43. Further, it is possible to combine SarkoSpin with mass spectrometry to investigate proteins beyond TDP-43 that are abnormally insoluble in a disease subtype–specific manner. Researchers performed subsequent MS of these enriched proteins that revealed distinct patterns of enrichment amongst ALS (23 proteins) and subtypes of FTLD (FTLD-A: 8 proteins; FTLD-C: 10 proteins). This provided critical insights into potential causes of disease heterogeneity (Laferriere et al. 2019). Studies on therapeutic interventions and discovering drug targets have shown successful combination of transcriptomic and proteomics findings that have provided important insights into mechanistic that include finding of lysosomal changes in Grn-/- mice that are modulated by the FTD-risk protein TMEM106B. In this study, researchers investigated early global alterations occurring in 2-month-old Grn / brain at both transcriptional and protein levels by employing genomewide RNA sequencing (RNASeq) and LFQ-LCMS. Subsequently, 364 mouse lysosomal genes on ‘The Mouse Lysosomal Gene Database’ were compared that resulted in identification of 29 and 19 differentially expressed lysosomal genes in RNASeq and LFQ-LCMS, respectively. The results obtained from transcriptomic and proteomic analyses suggest lysosomal dysregulation as one of the earliest global changes occurring in Grn / brain at both transcriptional and protein levels. This leads to the interpretation that the extent of neurodegeneration is determined by the interaction of Grn and Tmem106b genes that have opposite effects on lysosomal enzyme levels (Klein et al. 2017). Some other studies have shown transcriptomic findings with underlying TDP-43 pathology burden in ALS spinal cord lasercapture dissected motor neurons (Krach et al. 2018). As for discovery of biomarkers in motor neuron disease (Turner et al. 2013; Verber et al. 2019), there have been several notable small and large-scale biomarker studies that have been performed on human ALS samples by LFQ (Barschke et al. 2017). To this end, researchers analyzed cerebrospinal fluid ‘CSF’ from ALS and primary lateral sclerosis (PLS) by employing LFQ for novel biomarker discovery (Thompson et al. 2018). This study revealed increase in chitotriosidase, chitinase-3like protein 1 and chitinase-3-like protein 2 that were studied in a cohort of 43 patients with ALS. It was

attributed to neuroinflammation because of increased microglial activation (Thompson et al. 2018). In a recent study, researchers employed chip-based capillary zone electrophoresis combined with high-resolution mass spectrometry to profile blood plasma metabolome of patients diagnosed with sporadic ALS. The study revealed predictive biomarkers and the absence of b-Methyl Amino-l-alanine in blood plasma. Researchers observed creatinine as the most significant marker (p < 0.01) that was found 49% elevated in ALS patients. Whereas, creatinine and methyl histidine showed a decrease by 20 and 24%, respectively, in ALS patients (Bereman et al. 2020). These results indicate that profiling of biofluids could be very promising and readily available and expeditious approach that could be employed to deliver diagnostic and prognostic biomarkers in neurodegenerative diseases. Despite some notable advances made in the field of biomarker discovery and therapeutic interventions in ALS and FTD, it is still emerging and relatively a new area of research. These advances along with other ongoing studies are expected to impact research in these diseases involving neurodegenerative disorders. This is anticipated to pave the way for future breakthroughs that could lead to the discovery of highly effective therapeutic targets and noninvasive biomarkers for ALS and FTD (Hedl et al. 2019).

Biomarker discovery and validation for alzheimer’s and parkinson’s diseases Alzheimer and Parkinson are devastating age-related neurodegenerative diseases that cause serious medication conditions and considered one of the largest health concerns in the world today. Alzheimer’s disease (AD) is known to be a progressive neurodegenerative disease that causes gradual loss of memory and other cognitive abilities along with emotional and behavioral deficits (Botas et al. 2015). Studies on the complex pathogenesis of AD have confirmed the involvement of aberrant processing of amyloid-b proteins and the hyper phosphorylation of tau proteins. These proteins have been shown to aggregate to form amyloid-b plaques, amyloid angiopathy, and tau protein neurofibrillary tangles. Subsequently, these aggregations lead to neural degeneration in the hippocampus and entorhinal cortex that correspond to the centers of learning and memory in the brain (Chen et al. 2019). Parkinson’s disease (PD) is also a progressive neurodegenerative disease. PD is characterized by the death of dopaminergic neurons located within the substantia nigra (Dauer and Przedborski 2003). Unfortunately, there are no existing quantitative biomarkers that can

Figure 4. Projected number of Americans in different age groups with Alzheimer’s disease (AD) from 2020–2050 [With permission from Metabolites (2019)].

enable early diagnosis and therapeutic interventions of these diseases. This results in the progression of neurodegeneration that eventually leads to death (Chen et al. 2019). According to an estimate, the 4.7 million Americans were thought to have AD in 2010. This number has been projected to increase to 13.8 million by 2050 (Figure 4) (Hebert et al. 2013; Chen et al. 2019). Further, it is estimated that about 35.6 million people worldwide are living with dementia that is anticipated to further increase to much higher rate with lowincome countries becoming more industrially developed (Wimo et al. 2013). On the other hand, PD, which is considered the second most common neurodegenerative disease has shown comparative increases in prevalence. For example, the 680,000 Americans aged 45 and older who were diagnosed with PD in 2010 is now expected to increase to more than 1 million Americans by 2030 (Marras et al. 2018). In view of this alarming rise in the projected future cases of AD and PD, it is imperative to make breakthroughs in early identification and interventions to battle against these neurodegenerative diseases (Chen et al. 2019). Thus, current research in neurodegeneration is focused on developing accurate, reliable, and objective biomarkers for AD, PD, and related age-associated neurodegenerative disorders. Such biomarkers are especially needed to help in both diagnoses, particularly at

early stages, and monitoring of disease progression (Zhou et al. 2017). Advances have been made in protein detection platforms that have resulted a number of biomarker candidates for both AD and PD. However, the main bottleneck lies in transitioning of these biomarkers from discovery phase to the clinical applications (Cilento et al. 2019). Figure 5 shows pictorial depiction of proposed approach and the challenges. The technique of selected reaction monitoring (SRM) is considered promising for targeted mass spectrometry to overcome the challenges involving clinical applications. SRM is a fragmentation technique that filters out the background and undesired noise. This makes SRM a highly efficient technique. It is therefore, considered a very useful tool for ion signature analysis from biological samples. SRM can be quickly utilized in the analysis of biofluids as well as tissues and can be very helpful and advantageous in the biomarkers research and developments. An alternative technique is parallel reaction monitoring (PRM) that is also proposed for targeted mass spectrometry and quantification. It is a similar method that allows multiple parallel ions to be monitored. The PRM assay can be performed in a high resolution along with high mass accuracy mode on a MS system. The application of SRM or PRM is considered an important step going forward because it provides a strategy to verify of large panels of protein

Figure 5. The need for the development of accurate, reliable, and objective biomarkers for early diagnosis and tracking of disease progression for Alzheimer’s disease (AD) and Parkinson’s disease (PD). Targeted mass spectrometry is considered promising in achieving the goals. The challenges, however, remain for the biomarkers to transition to clinically useful biomarkers.

biomarker candidates prior to their costly validation testing. Studies in this regard have shown the coupling of discovery-based proteomics with modern targeted MS-based approaches (e.g. SRM) in future workflows to expedite biomarker development and validation for AD and PD. Researchers have shown the possibility of using an SRM pipeline that could potentially improve the efficiency of the overall biomarker development process. This could pave the way to expedite the development of accurate biomarkers that would be highly discriminatory for AD and PD. This is anticipated to ultimately lead to detection at early stages of disease and potential slowing the disease progression (Zhou et al. 2017; Cilento et al. 2019). Currently, the only clinically validated biomarkers for AD are developed from amyloid beta (Ab) and tau proteins/peptides within the CSF. The ongoing studies are extensively based on MS including MS methods. In this regard, researchers have mostly incorporated MS into large scale biomarker studies based on bottom-up shotgun proteomic approaches (Figure 6) (Cilento et al. 2019). In these studies, proteins are usually obtained from a sample. These proteins can then be cleaved by proteolytical approaches to produce smaller peptides. Subsequently, these are further fragmented within the MS and mass spectra are generated. The data base is subsequently compared with the generated mass spectra (Figure 6) (Cilento et al. 2019). Further, researchers have shown that peptide fragmentation patterns that are generated from tandem MS can be leveraged to establish accuracy in identification of protein/peptide and their structural characterization. Also, it has been shown that a combination of MS with the liquid

chromatography (LC) can be leveraged for the proteomics-based analysis of mixtures that are complex. It is to be noted that LC is known for fractionation and spatial separation of biomolecules of interest. Multidimensional fractionation allows for in-depth analysis of even the most complex proteomes. Hence, the implementation of LC methods in tandem with the MS is considered a very promising approach. This combined LC system with MS can be helpful to separate proteins with significantly reduced time for sample preparation and purification. In addition, the method of MS-based proteomic analysis can be employed for quantitative studies. This can be immensely helpful in the studies involving biological samples in diseased state (Cilento et al. 2019). The research focus is also on the application of exploratory MS-techniques, which has resulted in the increased quantity of reported candidate biomarkers for AD, PD and many other diseases. But, to establish a distinction between the discovery and validation phases of biomarker development still needs to be achieved. While biomarkers in CSF have been used in clinical diagnosis of AD, biomarkers for pre-clinical diagnosis of earlystage AD and disease progression are not available. Additionally, CSF based processes are considered highly invasive and not cost effective. To overcome this challenge, blood-based biomarkers are currently considered because of the advantages that include accessible in primary care in a minimally invasive way, convenient to patients, with little cost and suitable for repeated sampling for longitudinal assessments. Mass spectrometry, which can offer higher selectivity than alternative methods has been considered an established analytical tool for measuring markers in blood (Oeckl and Otto 2019). Especially, MS-based blood biomarkers for AD have been shown to be very promising by the determination of Ab peptides by IP- MS (Nakamura et al. 2018; Schindler et al. 2019) and the ‘‘genotyping’’ of ApoE by MRM (Hirtz et al. 2016). On the other hand, in a recent study on protein biomarkers, integrative proteomics were applied to identify CSF biomarkers representing a wide spectrum of AD pathophysiology (Higginbotham et al. 2020). As for biomarkers discovery in PD using CSF, the extended granin family was observed to be reduced in disease that suggested a potential common mechanism for the biological reduction in monoamine neurotransmission in PD patients (Rotunno et al. 2020). Further, MS-based metabolomics have been suggested for biomarkers and therapeutics studies on PD (Shao and Le 2019). In another approach, lewy body-enriched a-Synuclein in PD was analyzed by MALDI-TOF MS for potential biomarkers for abnormal a-synuclein

Figure 6. Top-down ‘shotgun’ proteomics strategy for biomarker identification A: Typical process showing extracting peptides from proteins within complex biological mixtures. Proteins are extracted from biological sources, and then they are separated. Subsequently, proteins are cleaved into peptide components by using enzymatic digestion. B: The process of peptide identification by LC-MS that involves digestion and subsequent separation of peptides by liquid chromatography, ionized, and detected by the MS. This is followed by peptides selected for fragmentation within the MS, and the individual components result in a fragmentation spectrum identified by the detector (MS2). C: The identification process of protein is shown that involves obtaining spectra of the fragmentation. Available data libraries of known peptide spectra can then be used to compare the obtained spectra. This results in determining the identity of the peptide. The use of bioinformatics is then sought to assess the validity of the peptides and also proteins that are revealed by the study [Source: J Neurochem (2019)].

accumulation (Bhattacharjee et al. 2019). In a related study, potential diagnostic and prognostic value of CSF and blood biomarkers were suggested that closely reflected the pathophysiology of Parkinson’s disease, such as a-synuclein species, lysosomal enzymes, markers of amyloid and tau pathology, and neurofilament light chain. To this end, researchers showed a strategy based on a combination of multiple CSF biomarkers that was considered a viable diagnostic and prognostic model. Since an early diagnosis of PD is considered extremely vital for a timely therapeutic intervention, the measurement of CSF a-synuclein aggregates provided encouraging preliminary results toward developing highly accurate early biomarkers for PD. Further, researchers also investigated blood a-synuclein species and neurofilament light chain to develop body fluid based noninvasive biomarkers. These results could pave the way for the development of a

noninvasive diagnostic tool that could be employed both for early and differential diagnosis of Parkinson’s disease that would differentiate from atypical parkinsonian disorders, and also for a systematic monitoring of the disease (Parnetti et al. 2019).

Conclusion & future perspective We have described mass spectrometry (MS)-based techniques to analyze biological samples for biomarkers for age-related neurodegenerative diseases. We have also discussed mass spectrometry imaging (MSI) that can be a truly complimenting technique to MS based biomarkers discovery. The major strength of these techniques is their high specificity for the target analyte and against matrix effects. The high complexity of the matrix and the low abundance of the proteins pose serious challenges in the current research efforts to

measure the core AD and PD biomarkers, Ab42, total tau and phospho-tau, and also in blood with immunoassays. Recent research advances have shown tremendous promise of MS based techniques and also their combination with immunological methods to overcome these challenges. It is believed that one of the most promising applications of MS in the current biomarker development would be achieving improved diagnostic performance of tau determination in blood and measurement of a synaptic blood marker for AD. Regarding discovering biomarkers for early detection and therapeutic intervention in PD, early data has shown promise for an overall fingerprint of metabolite alterations in multiple biofluids and tissues. Studies on such metabolomics have indicated significant potentials for a myriad of biomarkers and therapeutic targets for intervention in PD. Also, the field of advanced MSbased proteomics has emerged as a powerful route to the discovery of biomarkers and drugs and therapeutics for early detection and intervention of ALS and FTD. Such efforts are noteworthy as the success of these research and developmental efforts could potentially lead to slowing the progression of deadly ALS disease and thereby could extend healthy life span. While, notable advances have been made in MS based biomarker discoveries, further improvements are sought in both the sensitivity and specificity of current biomarkers for AD and PD. Research need to focus more on the application of exploratory MS-techniques, which can result in the increased quantity of reported candidate biomarkers for AD, PD and many other diseases. However, establishing a distinction between the discovery and validation phases of biomarker development still needs to be achieved. Future research directions could also focus on standardization of sample collection protocols, the specificity of the antibodies and the technologies used along with a larger number of subjects and controls. This would allow for establishing a large data base including important variables such as a range of age groups that are typically associated with clinical investigations. Especially, some authors have suggested that future studies require further validation in large independent cohorts to adopt CSF and blood biomarkers for improving PD diagnostic and prognostic accuracy (Parnetti et al. 2019). Independent cohort validation in biomarker diagnostics is the crux of the entire endeavor. A proof-of-concept level, cross-validation study using an independent cohort and relatively small sample size (150 patients) was recently conducted that provided evidence of the potential use of a multitiered blood-based proteomic screening method for detecting individuals with

neurodegenerative disease and then distinguishing PD from other neurodegenerative diseases (O’Bryant et al. 2019). However, such studies must be validated in independent cohorts and large sample sizes. With respect to AD, the field of proteomics is believed to advance the biological pathways that are connected to post-translational protein modification, cell migration and axonogenesis. These important properties are thought to synergistically enhance the protein–protein interactions that are involved in synaptic degeneration and brain dysfunction (Chandramouli and Qian 2009; Meyer and Schilling 2017; Portelius et al. 2017; Ludwig et al. 2018; Park et al. 2020). To this end, sequential window acquisition of all theoretical fragment ion spectra (SWATH)-based mass spectrometry (MS) seems to be a promising proteomic tool for future large-scale, high-quality quantification that could be useful biomarkers for AD (Jylha et al. 2018; Park et al. 2020). However, the invasive nature of CSF based processes is a serious concern that needs to be addressed. It is therefore, anticipated that future research would put emphasis on developing noninvasive, user-friendly, and in-expensive diagnostic biomarkers for AD and PD. As regards ALS and FTD, future studies should give greater focus on comparisons between new and existing proteomics datasets (Carlyle et al. 2018; Umoh et al. 2018; Leoni et al. 2019; Barschke et al. 2020; Oeckl et al. 2020). This is especially important as there are growing number of studies that are released using a variety of available techniques. It is also suggested in future studies to use proteomic techniques investigate biochemical changes longitudinally in ALS and FTD diseases. This is expected to reveal the earliest pathogenic changes in these diseases, which may be suitable for therapeutic intervention and clinical tests (Hedl et al. 2019). Another approach may involve comparisons of ALS and FTD datasets and other data sets that are obtained from neurodegenerative diseases different than ALS and FTD. This could be useful to obtain information on how disease-specific alterations arise and also it could shed light on the susceptibility of certain cell populations or CNS regions. These information would be interesting as it is likely that there are similar mechanisms that may be broadly involved in neurodegeneration. A new approach based on investigating defects in autophagy could be promising in future studies focusing on the knockout or expression of ALS/FTD mutant UBQLN2 proteins in cells and mice (Wu et al. 2020). Future studies could provide new insights into ALS disease mechanisms and genetic association signals

Figure 7. A conceptual futuristic therapeutic pathway based on in-expensive and noninvasive blood biomarkers for drug target discovery in age-related, progressing neurodegenerative diseases.

derived from peripheral blood analysis. Recent findings showed consistency with low-grade neutrophilia and hypoxia as ALS phenotypes. This was observed with heterogeneity among patients partly driven by differences in myeloid and lymphoid cell abundance, which could be extended in future studies to develop bloodbased biomarkers for ALS/FTD (Swindell et al. 2019). Another area is metabolites, where similar to discovery proteomics, LC/MS can be performed on biofluids to identify metabolites that differ in quantity in ALS (Verber et al. 2019). Recent studies have shown significant promise of one class of biofluid-based molecules that are short non-coding RNA species (ncRNA) for potential biomarkers for ALS. However, these biomarkers are not yet ready to be applied for prognosis or phenotypic characterization. Further work can focus on including more ncRNA candidates for deeper analysis. This can also include larger and more diverse cohorts with longitudinal samples that could pave the way to develop this biomarker signature for clinical applications (Matamala et al. 2018; Joilin et al. 2019, 2020). Finally, we anticipate that a stronger focus on noninvasive and inexpensive diagnostics will emerge in the near future in order to address the unmet need and grand medical challenge of early diagnostics and therapeutic intervention in age-related progressing neurodegenerative diseases, To this end, we envision a therapeutic pathway based on blood based biomarkers that could potentially be game changing in slowing the disease progression and extending healthy life span.

We also note that it would be worth to extend the studies to other body fluid types such as urine or CSF analyses. Figure 7 schematically illustrates a conceptual therapeutic and diagnostic scheme/pathway for the drug target discovery platform based on novel protein biomarkers for better management of neurodegenerative diseases in future.