Mass Spectrometry Based-Proteomic Analysis of Anisakis spp.: A Preliminary Study towards a New Diagnostic Tool

Anisakiasis is nowadays a well-known infection, mainly caused by the accidental ingestion of Anisakis larvae, following the consumption of raw or undercooked fishes and cephalopods. Due to the similarity of symptoms with those of common gastrointestinal disorders, this infection is often underestimated, and the need for new specific diagnostic tools is becoming crucial. Given the remarkable impact that MALDI–TOF MS biotyping had in the last decade in clinical routine practice for the recognition of bacterial and fungi strains, a similar scenario could be foreseen for the identification of parasites, such as nematodes. In this work, a MALDI–TOF MS profiling of Anisakis proteome was pursued with a view to constructing a first spectral library for the diagnosis of Anisakis infections. At the same time, a shotgun proteomics approach by LC–ESI–MS/MS was performed on the two main fractions obtained from protein extraction, to evaluate the protein species enriched by the protocol. A set of MALDI–TOF MS signals associated with proteins originating in the ribosomal fraction of the nematode extract was selected as a potential diagnostic tool for the identification of Anisakis spp.


Introduction
Among food biological hazards, parasites are particularly dangerous for human health. Globalization has markedly increased the change in eating habits, including the widespread consumption of raw, marinated, or smoked fish. Moreover, a quota of food allergies of unknown origin among the general population may be due to sensitization to Anisakis spp. (roundworms), representing a public health issue, whose clinical manifestations are characterized by digestive disorders, asthma, dermatitis, and even anaphylaxis. Nematodes of the Anisakidae family are fish parasites that can be found all over the world. The larvae live in the gut, visceral peritoneum, and flesh of many marine fish and cephalopod species and can colonize through different trophic bridges, ensuring and widening the parasite life cycle. Infection by several live parasite larvae isolated from salmon portions, bought in a local supermarket in Rome in late 2019, was assessed by macroscopic observation. Additionally, microscopic analysis was performed using an Axiovert 25 microscope (Zeiss, Jena, Germany). Specimens were identified as Anisakis spp. larvae.

Protein Extraction
Five specimens were washed several times in 0.9% NaCl solution, and each larva was stored in a sterile 0.9% NaCl solution in a ratio of 1:3 (parasite weight/volume of the isotonic solution) at −80 • C. Mechanical homogenization of the frozen material was carried out with a steel pestle, followed by an ultrasonic treatment of the biomass at 100% power (BactoSonic, BANDELIN electronic GmbH & Co KG, Berlin, Germany) in five cycles of 30 s. Two-hundred µL of Lysis buffer from the MALDI Sepsityper Kit (Bruker Daltonics GmbH, Bremen, Germany) was added to the resulting sample and mixed by vortexing for 10 s. After centrifugation, the pellet was suspended with 1 mL of Washing Buffer (MALDI Sepsityper Kit) and, after another centrifugation, the supernatant ("sample A") was used for LC-ESI-MS/MS analysis.
Last, the obtained pellet was resuspended in 1.2 mL of a solution water/ethanol 1:3, and 2/3 of it was centrifuged, obtaining a second pellet ("B"), which was resuspended in 35% formic acid (FA), 15% water, and 50% acetonitrile (ACN) for MALDI-TOF MS analysis ("sample B M ", M as the abbreviation for MALDI-TOF MS).

Matrix-Assisted Laser Desorption/Ionization-Time of Flight Mass Spectrometry (MALDI-TOF MS)
One microliter of sample BM was placed on an MSP 96 polished steel target (Bruker Daltonics, GmbH, Bremen, Germany), air-dried, and overlaid with 1 µL of the matrix, consisting in a solution of 5 mg/mL of α-cyano-4-hydroxycinnamic acid (Bruker Daltonics GmbH, Bremen, Germany) in 50% ACN, 47.5% water, 2.5% trifluoroacetic acid (TFA). Each sample was spotted onto eight target spots of the MALDI target plate, and spectral measurements were performed with a Microflex LT mass spectrometer (Biotyper, Bruker Daltonics GmbH, Bremen, Germany), equipped with the FlexControl software package, (version 3.4, Bruker Daltonics GmbH, Bremen, Germany), operating in the positive linear mode (laser frequency 20 Hz; ion source 1 voltage, 20 kV; ion source 2 voltage, 18.4 kV; lens voltage, 6 kV; mass range, 2000 to 20,000 m/z). Three independent mass spectra with 240 shots (from different positions of the target spot) for each spectrum were acquired from each spot, to obtain 24 spectra replicas, externally calibrated by using the Bacterial Test Standard (Bruker Daltonics GmbH,

Matrix-Assisted Laser Desorption/Ionization-Time of Flight Mass Spectrometry (MALDI-TOF MS)
One microliter of sample B M was placed on an MSP 96 polished steel target (Bruker Daltonics, GmbH, Bremen, Germany), air-dried, and overlaid with 1 µL of the matrix, consisting in a solution of 5 mg/mL of α-cyano-4-hydroxycinnamic acid (Bruker Daltonics GmbH, Bremen, Germany) in 50% ACN, 47.5% water, 2.5% trifluoroacetic acid (TFA). Each sample was spotted onto eight target spots of the MALDI target plate, and spectral measurements were performed with a Microflex LT mass spectrometer (Biotyper, Bruker Daltonics GmbH, Bremen, Germany), equipped with the FlexControl software package, (version 3.4, Bruker Daltonics GmbH, Bremen, Germany), operating in the positive linear mode (laser frequency 20 Hz; ion source 1 voltage, 20 kV; ion source 2 voltage, 18.4 kV; lens voltage, 6 kV; mass range, 2000 to 20,000 m/z). Three independent mass spectra with 240 shots (from different positions of the target spot) for each spectrum were acquired from each spot, to obtain 24 spectra replicas, externally calibrated by using the Bacterial Test Standard (Bruker Daltonics GmbH, Bremen, Germany). Subsequently, spectra datasets were imported into ClinPro Tools software (version 3.0, Bruker Daltonics, GmbH, Bremen, Germany) for data mining after peak picking on the calculated total average spectrum, setting the signal-to-noise (S/N) threshold at 3, baseline subtraction (Top Hat), and peak intensity calculation.

In-Solution Protein Digestion
The protein extracts (sample A and B LC ) were subjected to reduction, alkylation, and trypsin digestion according to the filter-aided sample preparation (FASP) protocol [11]. Briefly, the protein extracts were loaded on the Microcon-10kDa Centrifugal Filter Unit with an Ultracel-10 membrane (Merck, Burlington, MA, USA) in the presence of 8 M urea and 100 mM Tris-HCl, pH 8.5; disulfide bonds were reduced for 15 min at 37 • C with 8 mM DTT, then the samples were incubated with 50 mM iodoacetamide for 15 min and subsequently with DTT and digested with 1 µg of sequencing-grade trypsin (Promega, Milan, Italy) at 37 • C in 50 mM ammonium bicarbonate buffer pH 8.0, overnight (16 h). Peptides were eluted from the Microcon, speedvac-dried, and resuspended in a water solution with 2% ACN and 0.1% FA. Total peptide content was determined by NanoDrop 2000 (Thermo Fisher Scientific, Waltham, MA, USA) analysis, with a standard curve of MassPrep Escherichia coli digestion (Waters, Milford, MA, USA).

Liquid Chromatography-Electrospray Ionization-Tandem MS (LC-ESI-MS/MS)
LC-ESI-MS/MS experiments were performed on an UltiMate3000 RSLCnano System directly coupled to an Orbitrap Fusion Tribrid mass spectrometer, operating in positive ionization mode, equipped with a nanoESI source (EASY-Spray NG) (Thermo Fisher Scientific, Waltham, MA, USA). The digested proteins (1.25 µg) were first trapped and desalted onto a µ-precolumn cartridge C18 PepMap100 (5 µm particle size, 100 Å pore size, 300 µm i.d. × 5 mm length, Thermo Fisher Scientific, Waltham, MA, USA) for 3 min at 10 µL/min, with an aqueous solution of 2% ACN and 0.1% TFA, and then separated by reverse-phase chromatography performed on an EASY-Spray PepMap RSLC C18 column (2 µm particle size, 100 Å pore size, 75 µm i.d. × 50 cm length, Thermo Fisher Scientific, Waltham, MA, USA) at a flow rate of 250 nL/min and a temperature of 35 • C, by a one-step linear gradient from 95% eluent A (0.1% FA in water) to 25% eluent B (99.9% ACN, 0.1% FA) in 113 min and total LC run of 160 min. Precursor (MS1) survey scans were recorded in the Orbitrap, at resolving powers of 120 K (at m/z 200). Data-dependent MS/MS (MS2) analysis was performed in top speed mode with a 3 s cycle time, during which most abundant multiple-charged (2+-7+) precursor ions detected within the range of 375-1500 m/z were selected for activation in order of abundance and detected in ion trap at rapid scan rate. Quadrupole isolation with a 1.6 m/z isolation window was used, and dynamic exclusion was enabled for 60 s after a single scan. Automatic gain control targets were 4.0 × 10 5 for MS1 and 2.0 × 10 3 for MS2, with 50 and 300 ms maximum injection times, respectively. For MS2, the signal intensity threshold was 5.0 ×10 3 , and the option "Injection Ions for All Available Parallelizable Time" was set. High-energy collisional dissociation (HCD) was performed using 30% normalized collision energy. Lock mass was set as an internal calibration using polydimethylcyclosiloxane (445.12003 m/z).

Database Searching and Protein Identification
Protein IDs were obtained with the embedded search engine (Sequest HT) of the Proteome Discoverer software (PD, version 2.4, Thermo Fisher Scientific, Waltham, MA, USA) after searching a custom-made database containing the complete UniProtKB/Swiss-Prot sequence entries catalogue (561,568 proteins, release: 2019_11) to which Anisakis UniProtKB/TrEMBL (25,874 proteins, release: 2019_11) and "Salmon" UniProtKB/TrEMBL (233,298 proteins, release: 2020_01) sequence entries were appended. The search parameters included trypsin as the proteolytic enzyme with a maximum of 2 missed cleavages per peptide allowed and oxidation of methionine as a variable modification, whereas carbamidomethylation of cysteine was set as static modification. Precursor and fragment mass tolerance were set to 10 ppm and 0.6 Da, respectively. False discovery rate (FDR) was calculated by the Percolator algorithm, and a cut-off of 0.01 was used for the identifications (i.e., the expected fraction of incorrect protein match in the entire data set was set to less than 1%, calculated on a decoy database). At least two peptides were considered for protein ID.

Functional Analysis
The mapping of orthologous genes from protein lists to Caenorhabditis elegans was carried out by the g:Orth tool of the g:Profiler web server based on data collected into the Ensembl database [12,13] and by STRING (version 11.0, database 11_0) [14,15]. Bioinformatic analyses were performed by the g:GOSt function of the g:Profiler software in order to perform statistical enrichment analysis to find over-representation of information from Gene Ontology (GO) terms and Kyoto Encyclopedia of Genes and Genomes (KEGG) biological pathways. The most significant categories associated with the uploaded datasets were identified by calculating the related significance (p-value) when comparing the protein list to the whole C. elegans proteome. The p-value measures the likelihood that the association between the genes/proteins in the datasets and each GO and KEGG terms is not due to random chance alone, identifying a significant over-representation of molecules in association with a given process. We applied an experiment-wide p-value threshold of 0.05, limiting the FDR (i.e., the expected fraction of false positives among significant terms) to less than 5%; g:GOSt uses multiple testing correction and applies the tailor-made algorithm g:SCS for reducing significance scores.
Protein-protein interaction networks analysis was performed using the STRING application. The highest confidence of 0.9 was chosen as the minimum required interaction score threshold, such that only interactions above this score were included in the predicted networks. Networks were clustered by the Markov clustering (MCL) algorithm with inflation parameter set as 3 (indirectly related to the precision of the clustering, i.e., the higher the inflation, the more abundant the clusters).

Experimental Pipeline
After mechanical homogenization and sonication, the samples obtained from five larvae were treated according with the MALDI Sepsityper Kit protocol (Bruker Daltonics, GmbH, Bremen, Germany). To exploit the advantages of different proteomic strategies, a combined approach based on two mass spectrometry platforms was undertaken in the current study ( Figure 1).

Protein Profiling by MALDI-TOF MS Analysis
After protein extraction, spotting, and MS analysis, high-intensity peaks in the range of 2000-12,000 Da (m/z) were highlighted from the five larvae (sample B M ), with the highest density in the region comprised between 2500 and 8000 Da, with clusters of signals in ranges corresponding to 2700-2900, 5400-5700, and 7100-7500 Da ( Figure 2; Figure 3).
The phenotypic variability of the larvae, inspected by the clustering analysis performed on the complete dataset, was underlined: mass spectra of larva 1 and 4 clustered together and formed one clade with larva 5; larva 2 and 3 were more similar and identified a different cluster ( Figure 4). Genes 2020, 11, x FOR PEER REVIEW 6 of 18   The phenotypic variability of the larvae, inspected by the clustering analysis performed on the complete dataset, was underlined: mass spectra of larva 1 and 4 clustered together and formed one clade with larva 5; larva 2 and 3 were more similar and identified a different cluster ( Figure 4). The mass spectra dataset of each larva were grouped in a different class, and a cut-off ≤8 on the difference between the maximum and the minimum average peak intensities of all classes (DAve), a  The phenotypic variability of the larvae, inspected by the clustering analysis performed on the complete dataset, was underlined: mass spectra of larva 1 and 4 clustered together and formed one clade with larva 5; larva 2 and 3 were more similar and identified a different cluster (Figure 4). The mass spectra dataset of each larva were grouped in a different class, and a cut-off ≤8 on the difference between the maximum and the minimum average peak intensities of all classes (DAve), a The mass spectra dataset of each larva were grouped in a different class, and a cut-off ≤8 on the difference between the maximum and the minimum average peak intensities of all classes (DAve), a minimum peak intensity average of 2.5 (Ave), and 20% of the coefficient of variation (CV) for each class were applied. Regardless of the variability addressed by dendrogram analysis, 19 signals (among a total of 179) for all samples were furthermore identified, representing a collection of fingerprinting classifiers of Anisakis spp. (Table 1).

Protein Profiling by Tandem Mass Spectrometry
After enzymatic digestion and peptide purification, we identified the total protein content of two extraction steps of the adopted protocol by LC-ESI-MS/MS on a high-resolution platform. From sample A of larvae 1, 2, 3, 4, and 5 we identified an overall number of 2179 different proteins, of which 561 were identified as shared; in sample B LC , we identified a total of 3091 diverse proteins, of which 210 were common to the five larvae (Supplementary material S1). We focused our functional analysis only on identified proteins belonging to Anisakis species and to organisms selected according to phylogenetic similarity (Rhabditida order [16], 1732 and 1543 total different proteins in sample A and B, respectively) ( Figure 5, Supplementary Material S2).
Lists of identified proteins belonging to the Rhabditida order were filtered by a three-step process ( Figure 6): 1.
only proteins identified in at least four out of five larvae (Supplementary material S3) were considered; 2.
proteins with different UniProtKB accession code but the same name were collapsed into one single hit; 3.
proteins defined as unknown were deleted (Supplementary material S4). Genes 2020, 11, x FOR PEER REVIEW 9 of 18 Lists of identified proteins belonging to the Rhabditida order were filtered by a three-step process ( Figure 6): 1. only proteins identified in at least four out of five larvae (Supplementary material S3) were considered; 2. proteins with different UniProtKB accession code but the same name were collapsed into one single hit; 3. proteins defined as unknown were deleted (Supplementary material S4).

Functional Analysis
The filtered protein lists were examined for their known GO terms, retrieved by the ProteinCenter application of PD software, and grouped in the respective categories as percentage with respect to the total terms of each sample. The most represented biological process, cellular component, and molecular function were, as expected, similar between sample A and BLC and were linked to metabolic process, membrane, and catalytic activity (Figure 7).

Functional Analysis
The filtered protein lists were examined for their known GO terms, retrieved by the ProteinCenter application of PD software, and grouped in the respective categories as percentage with respect to the total terms of each sample. The most represented biological process, cellular component, and molecular function were, as expected, similar between sample A and B LC and were linked to metabolic process, membrane, and catalytic activity (Figure 7).
In order to highlight an over-representation of information from GO terms and biological pathways, as well as protein-protein interaction networks, related to our dataset, we converted the two protein lists into orthologous genes of C. elegans ( Figure 6 and Supplementary Material S5), as this model organism is the most investigated one for biomedical research and available in enrichment analysis web applications, contrary to Anisakis spp. Orthologous genes are likely conserved through evolution from a common ancestor, may carry out similar function, and are therefore relevant in functional analysis. In order to highlight an over-representation of information from GO terms and biological pathways, as well as protein-protein interaction networks, related to our dataset, we converted the two protein lists into orthologous genes of C. elegans ( Figure 6 and Supplementary material S5), as this model organism is the most investigated one for biomedical research and available in enrichment analysis web applications, contrary to Anisakis spp. Orthologous genes are likely conserved through evolution from a common ancestor, may carry out similar function, and are therefore relevant in functional analysis.
Our protein extraction pipeline followed by tandem mass spectrometry analysis led us to a great enrichment of proteins from ribosome and related to carbon metabolism pathway ( Figure 8 and Table  2) as well as many other GO terms and biological pathways, such as glycolysis/gluconeogenesis, glyoxylate and dicarboxylate metabolism, metabolic pathways (Supplementary material S6). Our protein extraction pipeline followed by tandem mass spectrometry analysis led us to a great enrichment of proteins from ribosome and related to carbon metabolism pathway ( Figure 8 and Table 2) as well as many other GO terms and biological pathways, such as glycolysis/gluconeogenesis, glyoxylate and dicarboxylate metabolism, metabolic pathways (Supplementary Material S6).   Term circles with p-values less than 1.0 × 10 −16 (highly significant) in at least one sample are highlighted and numbered. The circle sizes are in accordance with the corresponding term sizes (i.e., larger terms, which means more hits of the specific term retrieved from the dataset, correspond to larger circles). Consequently, network analysis evidenced a main cluster of ribosomal proteins in sample A ( Figure 9) and sample B LC (Figure 10). In fact, a group of proteins at least partially biologically connected present more interactions than a random set of proteins of similar size drawn from the genome; it indicated an enrichment of the ribosome-associated structural constituents and translation.
Genes 2020, 11, x FOR PEER REVIEW 13 of 18 Consequently, network analysis evidenced a main cluster of ribosomal proteins in sample A ( Figure 9) and sample BLC ( Figure 10). In fact, a group of proteins at least partially biologically connected present more interactions than a random set of proteins of similar size drawn from the genome; it indicated an enrichment of the ribosome-associated structural constituents and translation. Figure 9. Network analysis of sample A. The network nodes are proteins, and the edges represent the predicted functional associations. The highlighted red cluster is related to ribosomal proteins.

Discussion
Clinical laboratories take advantage of the MALDI-TOF MS technology to identify pathogenic microorganisms, including bacteria, yeasts, and filamentous fungi, thanks to the ease of use, speed of analysis, cost-effectiveness, and accuracy of IDs. MALDI-TOF MS analysis allows the detection of different types of biomolecules in a range of concentrations close to sub-femtomoles; in clinical diagnosis, peptides and proteins are fingerprinting classifiers [17]. In particular, one of the main advantages of this particular approach, i.e., rapidity, is grounded on the availability of mass spectra collections unambiguously identifying an organism by matching these spectral libraries with signals obtained from the samples.
Moreover, the applications comply with the European In Vitro Diagnostic Devices Directive (98/79/EC), which means that MALDI-TOF MS is a CE Marking for all in vitro diagnostic (IVD) devices, may be legally commercialized in the EU as a diagnostic tool, and the associated processes are similar to those of medical devices. In fact, the Bruker Biotyper and the VITEK MS system received first CE-IVD status in 2009 and 2011, respectively, and FDA clearance (for both) in 2013 [17]. The technique and the expertise have been growing over time; the implementation of associated software and reference databases, as well as sample preparation kits commercialized by the vendor have allowed researchers and laboratory specialists to both increase the type of microorganisms that can be identified and potentially treat/process "direct samples" (e.g., blood cultures), facilitating the management of infected patients. As an example, Bruker has developed the MBT Sepsityper IVD Kit for the identification of microorganisms from blood cultures using a MALDI-TOF MS platform [18-

Discussion
Clinical laboratories take advantage of the MALDI-TOF MS technology to identify pathogenic microorganisms, including bacteria, yeasts, and filamentous fungi, thanks to the ease of use, speed of analysis, cost-effectiveness, and accuracy of IDs. MALDI-TOF MS analysis allows the detection of different types of biomolecules in a range of concentrations close to sub-femtomoles; in clinical diagnosis, peptides and proteins are fingerprinting classifiers [17]. In particular, one of the main advantages of this particular approach, i.e., rapidity, is grounded on the availability of mass spectra collections unambiguously identifying an organism by matching these spectral libraries with signals obtained from the samples.
Moreover, the applications comply with the European In Vitro Diagnostic Devices Directive (98/79/EC), which means that MALDI-TOF MS is a CE Marking for all in vitro diagnostic (IVD) devices, may be legally commercialized in the EU as a diagnostic tool, and the associated processes are similar to those of medical devices. In fact, the Bruker Biotyper and the VITEK MS system received first CE-IVD status in 2009 and 2011, respectively, and FDA clearance (for both) in 2013 [17]. The technique and the expertise have been growing over time; the implementation of associated software and reference databases, as well as sample preparation kits commercialized by the vendor have allowed researchers and laboratory specialists to both increase the type of microorganisms that can be identified and potentially treat/process "direct samples" (e.g., blood cultures), facilitating the management of infected patients. As an example, Bruker has developed the MBT Sepsityper IVD Kit for the identification of microorganisms from blood cultures using a MALDI-TOF MS platform [18][19][20]. Interestingly, a recent article reported the use of the MBT Sepsityper Kit in order to profile four nematodes (Dirofilaria repens, Dirofilaria immitis, Ascaris suum, and Ascaris lumbricoides) [10], which intrigued and prompted us to test this method on Anisakis spp. In fact, our future goal, that has driven the preliminary study here presented, is to develop a pipeline for a proteomic profiling of Anisakis spp. for diagnostic purposes; the existence of an IVD kit will facilitate this future application of our method.
Although widely used as a diagnostic tool for accurate identification of bacteria, very few studies have tried to translate the advantages of the MALDI-TOF MS technique to clinical parasitology [8,9], and therefore, as far as we know, there are no related commercially MALDI-TOF MS ID spectral library databases.
The aim of our study was to assess the proof of concept that MBT Sepsityper Kit-based protein extraction from Anisakis larvae and subsequent MALDI-TOF MS Biotyper-based collection of spectra may provide diagnostic signals of biomedical interest related to this Anisakis life stage. The proposed method will work for the identification of already extracted nematode larvae from patients to improve the current diagnostic approaches. Although a variability in the MS peaks was ascertained, a few detected signals were statistically significant in representing all the larvae samples. This evidence is promising for the development of new diagnostic tools for Anisakis. We are aware that a definite assessment of the herein hypothesized diagnostic pipeline will require a collection of more Anisakis larvae samples, also with different geographical origin, as well as control samples from different nematodes in order to obtain a precise taxonomic typing. Hopefully, in the near future, efforts towards this development will be made in a consortium of research and clinical laboratories.
An interesting outcome of our study was the confirmation that the adopted protein extraction protocol and the MALDI-TOF MS analysis resulted in fingerprinting spectra attributable to peptides and proteins belonging to the plethora of ribosome molecular species, as already known [21]. The molecular weight of ribosomal proteins is within/around the mass range of the linear MALDI-TOF MS acquisition set-up, and our shotgun profiling experiments, by tandem MS on an higher resolution platform, highlighted the enrichment of ribosome proteins in two experimental steps.
Moreover, among the 448 proteins identified in four out of five larvae (Supplementary material S4), we identified also proteins related to infection molecular mechanisms, which have not yet been fully understood [22]. Ingested Anisakis has to survive in the highly acidic human stomach, penetrate the gastrointestinal wall, through degradation of the mucosa and submucosa, and migrate to the final location, causing tissue damage and inflammation. At the end, anisakiasis arises from both the tissue damage and the interplay between the host immune system and substances secreted by or contained within the larvae [2]. We identified some harmful antigens (proteinase inhibitor, somatic paramyosin, tropomyosin, and heat shock proteins), including Ani s 2 (UniProtKB accession number: L7V1I9), Ani s 4 (Q14QT4), Ani s 5 (A1IKL2), and Ani s 8 allergens (two isoforms: 1 and 10, A7M6S9 and A1IKL2, respectively), which are present in the Allergome database [23] and registered as Anisakis allergenic proteins by WHO/IUIS [24]. Ani s 2 is a cytoskeletal paramyosin protein, showing highly conserved sequences with respect to paramyosins of different origin (other nematodes, insects, shellfish), which account for the cross-reactivity in IgE binding. The excretory-secretory (ES) Ani s 4 allergen is a cysteine-type endopeptidase with inhibitor activity, present both in excretory glands and below the cuticle. Ani s 5 and Ani s 8 are other heat-stable ES allergens and members of the nematode SXP/RAL-2 protein family. In particular, Ani s 5 is assumed to be secreted in the human gastrointestinal tract from the third-stage larvae of Anisakis simplex; its putative magnesium ion transporter functional feature may be inferred by its magnesium ion binding capability and structural similarity to Calmodulin [25][26][27]. ES molecules are responsible for a multitude of functions during infection, such as penetration of host tissues and evasion of host immune responses, and at the same time are known to elicit immune responses. Therefore, ES proteins are hypothesized to be the major contributors in clinical manifestation of the disease in humans [28].
The identified Synthase trehalose-6-phosphate (A0A0M3JTQ2) may be important for the infection mechanism because it can be associated with a plethora of physiological and biochemical adaptive mechanisms that parasitic nematodes put in place under adverse environmental conditions, such as the unfavorable pH that Anisakis finds in the human gastrointestinal milieu, in order to survive [29].
Proteolytic enzymes, such as peptidases, are responsible for Anisakis pathogenicity because of their role in biological pathways linked to fundamental host-pathogen interactions. Among the several peptidases present in our dataset, we identified Metalloendopeptidase (A0A3G5BC99) belonging to the Astacin peptidase family M12A (Pfam code: Pf01400, secreted or membrane-anchored proteases that requires zinc for catalysis) and Carboxypeptidase (A0A158PN74), part of the Peptidase_S10 family (Pf00450). Interestingly, both protein families were recently identified amongst the upregulated transcripts in the pharynx of A. simplex and Anisakis pegreffii [22]. Moreover, the Metalloendopeptidase mRNA expression level was found to be higher in A. simplex third-stage larvae compared to the fourth-stage larvae, enforcing the role of this enzyme as a significant player in host tissues invasion [30].

Conclusions
A set of MALDI-TOF MS signals was identified as potential consensus "biomarkers" peak list, characterized by specific averaged m/z and intensity, for the identification of Anisakis spp. from nematode larvae present in patients tissue to improve the current diagnostic approaches. In fact, due to the similarity of symptoms with those of common gastrointestinal disorders and lack of highly skilled microscopists in biomedical laboratories, Anisakis infection is often underestimated and alternative diagnostic tools are enviable.
Additionally, the shotgun bottom-up analysis of Anisakis proteins, obtained by the extraction method based on the MBT Sepsityper Kit and performed by LC-ESI-MS/MS on a high-resolution platform, underlined the presence, in the nematode extract, of both an enrichment of ribosome proteins and specific proteins potentially associated with molecular mechanisms that accompany infection.