Proteomic Profiling and In Silico Characterization of the Secretome of Anisakis simplex Sensu Stricto L3 Larvae

Anisakis simplex sensu stricto (s.s.) L3 larvae are one of the major etiological factors of human anisakiasis, which is one of the most important foodborne parasitic diseases. Nevertheless, to date, Anisakis secretome proteins, with important functions in nematode pathogenicity and host-parasite interactions, have not been extensively explored. Therefore, the aim of this study was to identify and characterize the excretory-secretory (ES) proteins of A. simplex L3 larvae. ES proteins of A. simplex were subjected to liquid chromatography-tandem mass spectrometry (LC-MS/MS) analysis, and the identified proteins were then analyzed using bioinformatics tools. A total of 158 proteins were detected. Detailed bioinformatic characterization of ES proteins was performed, including Gene Ontology (GO) analysis, identification of enzymes, Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways analysis, protein family classification, secretory pathway prediction, and detection of essential proteins. Furthermore, of all detected ES proteins, 1 was identified as an allergen, which was Ani s 4, and 18 were potential allergens, most of which were homologs of nematode and arthropod allergens. Nine potential pathogenicity-related proteins were predicted, which were predominantly homologs of chaperones. In addition, predicted host-parasite interactions between the Anisakis ES proteins and both human and fish proteins were identified. In conclusion, this study represents the first global analysis of Anisakis ES proteins. The findings provide a better understanding of survival and invasion strategies of A. simplex L3 larvae.


Introduction
Anisakis simplex sensu stricto (s.s.), a nematode species belonging to the family Anisakidae, is among the most important foodborne parasites capable of causing a disease in humans called anisakiasis. This roundworm has an indirect lifecycle involving several hosts. Marine mammals are definitive hosts, fish, and cephalopods are intermediate or paratenic hosts, while crustaceans are intermediate hosts [1,2]. Humans, which are accidental hosts of A. simplex, are infected by third-stage (L3) larvae of this nematode, the source of which are infected marine fish or cephalopods [1,3]. A. simplex (s.s.) is mainly distributed in the northern Atlantic and Pacific Oceans [1]. However, other areas of occurrence (e.g., western Mediterranean Sea) were also reported [1,4,5].
The global incidence of anisakiasis is 0.32 cases/100,000 inhabitants [6]. However, according to recent studies, the prevalence of the disease is estimated to be much higher [7]; this discrepancy is linked to the nonspecificity of symptoms and the limitations of diagnostic tools. Furthermore, anisakiasis has become increasingly more important as human health risk, especially in regions where the consumption of raw or only lightly processed fish and seafood is frequent [1,8,9]. Therefore, the majority of anisakiasis cases are noted in proteins identification within individual biological replicates are presented in Supplementary File S1.2.

Comparison of A. simplex Secretome Proteins with Secretome Proteins of Other Nematodes
The secretome proteins of A. simplex (s.s.) were compared with those of selected nematodes to determine similarities and differences in their profiles. The following secretomes had the largest number of highly similar proteins (≥70% similarity) to those of the A. simplex secretome: adult Ascaris suum (32 proteins), adult Ancylostoma caninum (27 proteins), and L3 larvae of Spirocerca lupi (26 proteins). Fewer highly similar proteins were detected in the secretomes of Toxocara canis larvae (17 proteins) and Ascaris suum L3 larvae (three proteins). Among the highly similar proteins, homologs of the following A. simplex proteins were the most common: peptidyl-prolyl cis-trans isomerase (A0A0M3JT42), which has homologs in adult A. caninum, adult A. suum, L3 larvae of A. suum, and L3 larvae of S. lupi, and an uncharacterized protein (A0A0M3JWS2) and putative actin (A0A0M3J0M4), both of which have homologs in adult A. caninum, adult A. suum, L3 larvae of S. lupi, and T. canis larvae. Details of the comparison of secretome proteomic profiles are presented in Figure 1 and Supplementary File S1.3-8.

Protein Family Classification
The proteins in the A. simplex (s.s.) secretome belong to different families, among which the following were most frequently represented: immunoglobulin-like fold (15 proteins), immunoglobulin-like domain superfamily (12 proteins), annexin superfamily (six proteins), and thioredoxin-like superfamily (five proteins). Most protein families (87 protein families) were represented by only one protein. A total of 143 protein families were identified (see Supplementary File S1.9).

Protein Family Classification
The proteins in the A. simplex (s.s.) secretome belong to different families, among which the following were most frequently represented: immunoglobulin-like fold (15 proteins), immunoglobulin-like domain superfamily (12 proteins), annexin superfamily (six proteins), and thioredoxin-like superfamily (five proteins). Most protein families (87 protein families) were represented by only one protein. A total of 143 protein families were identified (see Supplementary File S1.9).

Secretory Pathway Prediction
Based on bioinformatics prediction, 21 proteins with identified signal peptides were classified into the conventional secretory pathway, and 77 proteins were assigned to unconventional protein secretion. Furthermore, among ES proteins the following proteins known to be associated with extracellular vesicles (EVs) released from Anisakis [27] were found: putative actin (A0A0M3J0M4), heat shock 70 kDa protein cognate 1 (A0A0M3K9V2), glutamate dehydrogenase (NAD(P)(+)) (A0A0M3K4H2), uncharacterized protein (A0A0M3KAB8), superoxide dismutase [Cu-Zn] (A0A0M3J718), and pepsin-I3 domain-containing protein (A0A0M3JAH0). Additionally, 24 proteins probably EV-associated were found in the Anisakis secretome. These protein are homologs of EV-associated protein which were identified in the secretomes of the following nematodes: A. suum [28], Brugia malayi [29], Pathogens 2022, 11, 246 4 of 27 and Nippostrongylus brasiliensis [30]. Among the potential EV-associated proteins, the best matches were as follows: proteasome subunit alpha type (A0A0M3K144), proteasome subunit alpha type-3 (A0A0M3JSH7), and triosephosphate isomerase (A0A0M3JVA5). The top 10 best matches from identification of proteins potentially associated with EV are shown in Table 1, and all the results from the secretory pathway and EV-associated protein predictions are presented in Supplementary Files S1.10-12.

Gene Ontology (GO) Annotation and Enrichment Analysis
GO annotations of identified A. simplex (s.s.) proteins were grouped into three categories: biological process, molecular function, and cellular component. Around 90% of the proteins were annotated with GO terms. A total of 1242 GO annotations were identified. Only 10 proteins were assigned a single GO annotation, and the remaining proteins were annotated with 2-60 GO terms. In the biological process category, the most frequent GO terms were cellular component organization (39 proteins), organonitrogen compound metabolic process (35 proteins), and system development (31 proteins). In the molecular function category, the following GO terms were the most abundant: cation binding (26 proteins), anion binding (16 proteins), nucleotide binding (14 proteins), and nucleoside phosphate binding (14 proteins). The most abundant terms in the cellular component category were as follows: intracellular organelle (62 proteins), nonmembrane-bounded organelle (43 proteins), and membrane-bounded organelle (30 proteins). The top 15 GO terms in the three ontology categories are shown in Figure 2, and all GO annotations for individual proteins are listed in Supplementary File S1.13.  Enrichment analysis allowed for the mapping of over-and underrepresented GO terms of A. simplex (s.s.) secretome proteins. This analysis was performed by comparing the representation of GO terms for the detected secretome proteins with that for the whole A. simplex proteome. A total of 174 overrepresented and 11 underrepresented GO terms were identified. According to the calculated p-values, the following GO terms were the most overrepresented: structural constituent of cuticle (p-value = 3.88 × 10 −10 ), medial layer of collagen and cuticulin-based cuticle extracellular matrix (p-value = 2.02 × 10 −9 ), and desmosome (p-value = 1.81 × 10 −8 ). Conversely, the following GO annotations were the Enrichment analysis allowed for the mapping of over-and underrepresented GO terms of A. simplex (s.s.) secretome proteins. This analysis was performed by comparing the representation of GO terms for the detected secretome proteins with that for the whole A. simplex proteome. A total of 174 overrepresented and 11 underrepresented GO terms were identified. According to the calculated p-values, the following GO terms were the most overrepresented: structural constituent of cuticle (p-value = 3.88 × 10 −10 ), medial layer of collagen and cuticulin-based cuticle extracellular matrix (p-value = 2.02 × 10 −9 ), Pathogens 2022, 11, 246 6 of 27 and desmosome (p-value = 1.81 × 10 −8 ). Conversely, the following GO annotations were the most underrepresented: integral component of membrane (p-value = 6.57 × 10 −6 ), regulation of signal transduction (p-value = 0.001), and ion transmembrane transporter activity (p-value = 0.007). The top 20 overrepresented and underrepresented GO terms are shown in Figure 3A,B, respectively. The detailed results of the GO enrichment analysis are shown in Supplementary File S1.14.
Pathogens 2022, 11, x FOR PEER REVIEW 6 of 29 most underrepresented: integral component of membrane (p-value = 6.57 × 10 −6 ), regulation of signal transduction (p-value = 0.001), and ion transmembrane transporter activity (p-value = 0.007). The top 20 overrepresented and underrepresented GO terms are shown in Figure 3A,B, respectively. The detailed results of the GO enrichment analysis are shown in Supplementary File S1.14.

Enzyme Identification and Enrichment Analysis
Bioinformatics analysis of A. simplex (s.s.) secretome proteins was used to identify enzymes and proteins involved in metabolic pathways. Seventy-two proteins were assigned to six enzyme classes. The most abundant class of enzymes was hydrolases (25 proteins), and the less represented classes were isomerases, transferases, oxidoreductases, translocases, and lyases. No proteins belonging to the ligase class were detected. The distribution of the number of proteins in each enzyme class is shown in Figure 4A.
As determined using the OmicsBox software, 38 enzymes (including enzyme classes/subclasses) were overrepresented in the secretome compared to the whole A. simplex proteome. Based on the calculated P-values, the most overrepresented enzymes were isomerases (p-value = 3.08 × 10 −7 ), methylmalonyl-CoA epimerase (p-value = 4.31 × 10 −7 ), acting on superoxide as acceptor (p-value = 4.26 × 10 −6 ), and superoxide dismutase (P-value = 4.26 × 10 −6 ). No underrepresented enzymes were found in the Anisakis secretome. Figure 4B shows the 20 overrepresented enzymes in descending order of their abundance in the secretome. Details of the enzyme enrichment analysis are presented in Supplementary File S1.15.

Enzyme Identification and Enrichment Analysis
Bioinformatics analysis of A. simplex (s.s.) secretome proteins was used to identify enzymes and proteins involved in metabolic pathways. Seventy-two proteins were assigned to six enzyme classes. The most abundant class of enzymes was hydrolases (25 proteins), and the less represented classes were isomerases, transferases, oxidoreductases, translocases, and lyases. No proteins belonging to the ligase class were detected. The distribution of the number of proteins in each enzyme class is shown in Figure 4A.
As determined using the OmicsBox software, 38 enzymes (including enzyme classes/ subclasses) were overrepresented in the secretome compared to the whole A. simplex proteome. Based on the calculated P-values, the most overrepresented enzymes were isomerases (p-value = 3.08 × 10 −7 ), methylmalonyl-CoA epimerase (p-value = 4.31 × 10 −7 ), acting on superoxide as acceptor (p-value = 4.26 × 10 −6 ), and superoxide dismutase (p-value = 4.26 × 10 −6 ). No underrepresented enzymes were found in the Anisakis secretome. Figure 4B shows the 20 overrepresented enzymes in descending order of their abundance in the secretome. Details of the enzyme enrichment analysis are presented in Supplementary File S1.15.  In addition, proteases and protease inhibitors were identified using the M database. According to the used BLAST search cutoff, 36 of these enzymes were f which 27 were proteases and nine were protease inhibitors. Eighteen proteases/ inhibitors showed 100% similarity to A. simplex proteins reported in the M database. A large group (eight proteins) also includes proteins that show sim In addition, proteases and protease inhibitors were identified using the MEROPS database. According to the used BLAST search cutoff, 36 of these enzymes were found, of which 27 were proteases and nine were protease inhibitors. Eighteen proteases/protease inhibitors showed 100% similarity to A. simplex proteins reported in the MEROPS database. A large group (eight proteins) also includes proteins that show similarity to proteases/protease inhibitors of other parasitic helminths, such as A. suum, Trichuris suis, Trichinella nativa, Onchocerca volvulus, Hymenolepis nana, and Hymenolepis diminuta. Table 2 shows the top 10 secretome protein matches of proteases/protease inhibitors, and all results are presented in Supplementary File S1.16.

Kyoto Encyclopedia of Genes and Genomes (KEGG) Pathway Identification and Enrichment Analysis
The KEGG pathway profiling of Anisakis secretome proteins revealed proteins involved in 44 pathways. Among them, the most frequent were metabolic pathways (21 proteins), followed by carbon metabolism (15 proteins), and glyoxylate and dicarboxylate metabolism (seven proteins). The 15 most abundant KEGG pathways are shown in Figure 5A. The other 29 KEGG pathways were represented by 1-2 proteins. All identified KEGG pathways are listed in Supplementary File S1.17.
KEGG pathway enrichment analysis revealed seven overrepresented pathways in the secretome compared to the whole A. simplex proteome ( Figure 5B). In ascending order of the calculated p-value, the most overrepresented KEGG pathways were as follows: carbon metabolism (p-value = 1.39 × 10 −6 ), glyoxylate and dicarboxylate metabolism (p-value = 0.00097), and propanoate metabolism (p-value = 0.00253). No underrepresented KEGG pathways were found. Detailed results of the KEGG pathway enrichment analysis are presented in Supplementary File S1.17.

Identification of Essential Proteins
Among the identified secretome proteins of A. simplex (s.s.), 33 essential proteins were predicted (see Supplementary File S1.18) using the DEG database. Essential proteins are those indispensable for the survival of an organism. These proteins belong to various protein families. The best three matches against the sequences from the database were as follows: putative actin (A0A0M3J0M4), calmodulin (A0A0M3K916), and an uncharacter- ized protein (A0A0M3K916) that shows homology to alpha-actinin-4. Table 3 presents the 10 Anisakis proteins with the highest similarity to known essential proteins.

Identification of Essential Proteins
Among the identified secretome proteins of A. simplex (s.s.), 33 essential proteins were predicted (see Supplementary File S1.18) using the DEG database. Essential proteins are those indispensable for the survival of an organism. These proteins belong to various protein families. The best three matches against the sequences from the database were as follows: putative actin (A0A0M3J0M4), calmodulin (A0A0M3K916), and an uncharacterized protein (A0A0M3K916) that shows homology to alpha-actinin-4. Table 3

Identification of Potential Pathogenicity-Related Proteins
Using three databases, nine putative pathogenicity-related proteins were identified in the A. simplex (s.s.) secretome. Four proteins were found in multiple databases, and five proteins were detected in a single database. The highest number of potential pathogenicity-related proteins (eight proteins) was identified using the VICTORS database. Hits with the highest similarity to confirmed pathogenicity-related proteins were as follows: heat shock 70 kDa protein cognate 1 (A0A0M3K9V2), 78 kDa glucose-regulated protein (A0A0M3K5H6), and an uncharacterized protein (A0A0M3K4G1). The 3D structures reveal similarities between these proteins and their homologs with confirmed pathogenic properties (see Figure 6). A relatively high number of detected potential pathogenicityrelated proteins (A0A0M3K9V2, A0A0M3K5H6, A0A0M3K4G1, and A0A0M3K0Q9) have homologs in the heat shock protein (HSP) family. Furthermore, the majority of the putative pathogenicity-related proteins in the Anisakis secretome (five proteins) are homologs of bacterial virulence proteins. Three A. simplex proteins show similarity to Cryptococcus neoformans virulence proteins, and two are homologs of Toxoplasma gondii HSP. All of the pathogenicity-related proteins identified in the study are shown in Table 4.

Allergen and Potential Allergen Identification
Of all identified proteins, only one (Ani s 4) is listed by the World Health Organization and the International Union of Immunological Societies (WHO/IUIS) Allergen Nomenclature Sub-Committee. By contrast, using the FARRP database, 18 potential allergens were identified. The three proteins with the best identification against the FARRP database were as follows: SXP/RAL-2 family protein 2 isoform 1 (A0A0M3KA05) and two globin-like proteins (A0A0M3KIW7, A0A0M3JEL6). The 3D structures of these potential allergens in comparison with their homologous allergens are shown in Figure 7. The AllerCatPro server confirmed that all proteins identified using the FARRP database have possible allergenic potential. The AllerCatPro tool determined the allergenic properties of 11 proteins with high confidence and 7 proteins with low confidence. The five detected potential allergens showed high similarity (>92%) to A. simplex allergens, and four others are highly similar to mite allergens. The other detected putative allergens showed similarity to allergens of A. suum, mosquito, fish, freshwater crayfish, and fungus. All potential allergens found in the A. simplex (s.s.) secretome are presented in Table 5.

Predicted Protein-Protein Interactions in A. simplex (s.s.) Secretome
The protein interaction network was established using STRING and showed predicted interactions between ES proteins of A. simplex (s.s.). Fifty-three proteins involved in the interaction network were revealed with high prediction confidence. Twenty-five of these proteins were associated with KEGG metabolic pathways, particularly carbon metabolism, in which ten proteins were involved. Furthermore, proteins involved in the interaction network were associated with the following groups: essential proteins (21 proteins), proteases/protease inhibitors (16 proteins), potential allergens (11 proteins), and potential pathogenicity-related proteins (seven proteins). Twenty-nine proteins were assigned to only one of the groups listed above, and eleven were categorized into several groups simultaneously. Nineteen proteins from the interaction network were not assigned to any of the explored groups. The detailed analysis of the protein interaction network is shown in Figure 8.
the FARRP database were as follows: SXP/RAL-2 family protein 2 isoform 1 (A0A0M3KA05) and two globin-like proteins (A0A0M3KIW7, A0A0M3JEL6). The 3D structures of these potential allergens in comparison with their homologous allergens are shown in Figure 7. The AllerCatPro server confirmed that all proteins identified using the FARRP database have possible allergenic potential. The AllerCatPro tool determined the allergenic properties of 11 proteins with high confidence and 7 proteins with low confidence. The five detected potential allergens showed high similarity (> 92%) to A. simplex allergens, and four others are highly similar to mite allergens. The other detected putative allergens showed similarity to allergens of A. suum, mosquito, fish, freshwater crayfish, and fungus. All potential allergens found in the A. simplex (s.s.) secretome are presented in Table 5.

Predicted Host-Parasite Protein Interactions
The HPIDB 3.0 server was used for the prediction of host-parasite interactions between the A. simplex (s.s.) ES proteins and both human and fish (Atlantic herring) proteins.
Eighteen proteins of Anisakis and 87 human proteins were identified in the hostparasite interaction network (see Figure 9A). The following groups of proteins were detected among Anisakis secretome proteins involved in interactions with human proteins: essential proteins (17 proteins), proteases/protease inhibitors (12 proteins), KEGG pathway proteins (12 proteins), potential allergens (seven proteins), and potential pathogenicity-related proteins (seven proteins). Most secretome proteins (16 proteins) were classified into two or more of these groups. Furthermore, it should be noted that the following Anisakis ES proteins showed the highest number of potential interactions with human proteins: transaldolase (A0A0M3KAE3; 22 interactions), proteasome subunit alpha type (A0A0M3JT99; 13 interactions), putative actin (A0A0M3J0M4; 12 interactions), heat shock 70 kDa protein cognate 1 (A0A0M3K9V2; 11 interactions), elongation factor 2 (A0A0M3K613; 11 interactions), and Rab GDP dissociation inhibitor (A0A0M3JZR1; 10 interactions). Human proteins that were predicted to be involved in the host-parasite interaction network belong to many different families, and the most highly represented of them were the following: laminins (17 proteins), methyltransferase proteins (13 proteins), Ras-related proteins (nine proteins), and Hsp70-binding proteins (seven proteins). Furthermore, among human proteins, polyubiquitin-C (P0CG48) showed potential interactions with the highest number of Anisakis proteins (13 interactions).

Predicted Host-Parasite Protein Interactions
The HPIDB 3.0 server was used for the prediction of host-parasite interactions between the A. simplex (s.s.) ES proteins and both human and fish (Atlantic herring) proteins.
Eighteen proteins of Anisakis and 87 human proteins were identified in the hostparasite interaction network (see Figure 9A). The following groups of proteins were detected among Anisakis secretome proteins involved in interactions with human proteins: essential proteins (17 proteins), proteases/protease inhibitors (12 proteins), KEGG pathway proteins (12 proteins), potential allergens (seven proteins), and potential pathogenicityrelated proteins (seven proteins). Most secretome proteins (16 proteins) were classified into two or more of these groups. Furthermore, it should be noted that the following Anisakis ES proteins showed the highest number of potential interactions with human proteins: transaldolase (A0A0M3KAE3; 22 interactions), proteasome subunit alpha type (A0A0M3JT99; 13 interactions), putative actin (A0A0M3J0M4; 12 interactions), heat shock 70 kDa protein cognate 1 (A0A0M3K9V2; 11 interactions), elongation factor 2 (A0A0M3K613; 11 interactions), and Rab GDP dissociation inhibitor (A0A0M3JZR1; 10 interactions). Human proteins that were predicted to be involved in the host-parasite interaction network belong to many different families, and the most highly represented of them were the following: laminins (17 proteins), methyltransferase proteins (13 proteins), Ras-related proteins (nine proteins), and Hsp70-binding proteins (seven proteins). Furthermore, among human proteins, polyubiquitin-C (P0CG48) showed potential interactions with the highest number of Anisakis proteins (13 interactions).

Discussion
This study is the first global proteomic analysis of the A. simplex (s.s.) L3 larval secretome. Previous proteomic investigations of Anisakis did not include profiling of the secretome proteins [13,[31][32][33][34][35][36][37][38]. Therefore, knowledge on A. simplex ES proteins is very fragmented, although important aspects related to metabolism, pathogenicity, and hostparasite interactions are known to be associated with ES proteins. In this study, LC-MS/MS and bioinformatics analyses were applied to provide insights into these issues. Five proteins of Anisakis secretome and 19 proteins of Atlantic herring were predicted in the fish-parasite interaction network (see Figure 9B). All Anisakis ES proteins identified in this interactome were identified also in the human-parasite interaction network. The following groups of proteins were detected among Anisakis ES proteins involved in interactions with fish proteins: essential proteins (five proteins), proteases/protease inhibitors (3 proteins), KEGG pathway proteins (three proteins), and potential allergen (one protein). Rab GDP dissociation inhibitor (A0A0M3JZR1), and calmodulin (A0A0M3KFJ2) showed potential interactions with the highest number of fish proteins (10 and four interactions, respectively). The most highly represented fish proteins involved in the interactome were Ras superfamily proteins (10 proteins), mainly members of the Rab family. Phosphodiesterases (four proteins) were the second largest group of the fish-parasite interactome. The detailed results of identification of proteins involved in potential host-parasite interactions are presented in Supplementary Files S1.19-20.

Discussion
This study is the first global proteomic analysis of the A. simplex (s.s.) L3 larval secretome. Previous proteomic investigations of Anisakis did not include profiling of the secretome proteins [13,[31][32][33][34][35][36][37][38]. Therefore, knowledge on A. simplex ES proteins is very fragmented, although important aspects related to metabolism, pathogenicity, and hostparasite interactions are known to be associated with ES proteins. In this study, LC-MS/MS and bioinformatics analyses were applied to provide insights into these issues.
Prior to identifying Anisakis ES proteins by mass spectrometry, their SDS-PAGE profile was analyzed. Electrophoretic analysis confirmed the distribution of protein bands over a wide range of molecular weights, characteristic of the ES proteins of A. simplex L3 (see Supplementary Figure S1). Subsequently, LC-MS/MS analysis allows identification of 158 proteins in the Anisakis secretome, which is currently the largest proteomic dataset of A. simplex ES proteins. The number of ES proteins identified in this study corresponds to approximately 0.8% of the genes encoding A. simplex proteins. Comparing the number of identified ES proteins of A. simplex to the secretomes of other closely related pathogenic nematodes with similar genome size, such as A. suum and T. canis (see Figure 1), reveals that there is a relatively high number of ES proteins in Anisakis. By contrast, the lower number of identified Anisakis ES proteins compared to the A. caninum secretome is presumably due to the much larger genome size of A. caninum. Furthermore, only about one-quarter of the identified Anisakis ES proteins showed high similarity to the proteins of the Toxocara or Ascaris secretomes. This relatively low similarity is probably due to differences in hosts and life cycles of Anisakis nematodes and those of Toxocara and Ascaris.
The majority of detected Anisakis ES proteins were assigned to an unconventional secretory pathway (approximately 49% of proteins). This prediction is consistent with the secretory pathway analysis of ES proteins of other nematodes, such as Dirofilaria immitis [39] or Strongyloides venezuelensis [40]. Most secretome proteins of these nematodes were also classified into the unconventional secretory pathway. Furthermore, 15% of Anisakis proteins were classified as potentially EV-associated proteins. Prediction of these proteins was based on similarity to EV-associated proteins secreted by other nematodes. This is a particularly important identification because, among other considerations, the EV released by parasites play an important role in delivering molecules that can modulate the host immune response or the transfer of pathogenicity-related factors [28]. In this study, six EV-associated proteins were found which were previously identified by Boysen et al. [27] (see Section 2.4). There are currently no other published studies on the identification of Anisakis EV-associated proteins, and because of their important functions, this topic requires further exploration.
The secretome proteins detected in Anisakis were characterized by high diversity, as evidenced by their classification into 143 protein families. Among the most frequently identified protein families in the Anisakis secretome, attention should be paid to the annexin superfamily. Annexins have multiple functions, such as in cellular anti-inflammation, signal transmission, anticoagulation, ion channel regulation, membrane repair, and membrane transport, and likely participate in cell proliferation, differentiation, and apoptosis [41,42]. Furthermore, parasite annexins are considered potential drug and vaccine targets [43]. The thioredoxin-like superfamily is another of the protein families most frequently detected in the Anisakis secretome that is also important. Thioredoxins, inter alia, regulate thiolbased redox control and prevent the aggregation of cytosolic proteins in the cell [44,45]. The extracellular activities of thioredoxins include anti-inflammatory and antiapoptotic activities and, thus, cytoprotective effects [44,46].
In general, the identified ES proteins of A. simplex (s.s.) have multiple functions, as demonstrated by the GO analysis. On average, nine GO terms were detected for all 142 proteins that were assigned GO annotations. A large variety of GO terms among nematode secretome proteins is quite typical [47]. Many of the detected secretome proteins, such as thioredoxins [44], annexins [48], and HSPs [49], are moonlighting proteins that form a subset of multifunctional proteins in which one polypeptide chain exhibits more than one physiologically relevant biochemical or biophysical function [50].
GO annotation enrichment analysis provided interesting data. Of the many enriched annotations, the most abundant and most enriched GO terms were, in general, related to the glycolytic process, larval development, antioxidants, and cuticle. These annotations cover functions that are important to parasite metabolism, lifestyle, and survival, and they are also found to be enriched in the annotated secretomes of other nematodes [40,47]. Among the enriched GO terms mentioned above, those related to the cuticle may seem to be unassociated with the secretome. However, it should be noted that ES proteins are also released from the surface of the cuticle, in addition to specialized excretory-secretory organs and parasite intestine [24]. Proteins produced and presented at the parasite-host interface during invasion play a critical role in the induction and development of immune responses [24]. Furthermore, secretome proteins could also play essential roles in ensuring cuticle integrity [51].
Possible enzymes were detected among secretome proteins, which is in line with previous studies that confirmed the enzymatic properties of the A. simplex secretome [52]. Furthermore, Kim et al. [53] found that protease related genes are highly expressed in the transcriptome of A. simplex L3 larvae. In the present study, proteases were highly represented in the secretome (17% of ES Anisakis proteins). These enzymes are known to be especially important in the pathogenesis of anisakiasis and other parasitoses [54,55]. Proteases play an important part in host-parasite interactions, such as invasion of the host, migration through host tissues, protection of the parasite against the host immune system, and activation of the inflammatory response [56,57]. Proteases also participate in important biological processes in parasitic nematodes, as they are directly involved in their growth and survival, embryonic development, digestion of protein for nutrients, molting, and numerous metabolic processes [58,59]. Another important enzyme group detected in the secretome is antioxidant enzymes, such as thioredoxin-dependent peroxiredoxin and superoxide dismutase. These enzymes are found to be enriched in the secretome, and they function to protect against the toxic contents released by immune effector cells as a first-line host defense mechanism [22]. Acetylcholinesterase (AChE) was also detected as an enriched enzyme in the secretome. The enrichment of AChE in the Anisakis secretome is in agreement with previous investigations [60]. AChE secretion by Anisakis larvae is presumed to be an adaptive mechanism, and its secretion increases in response to a direct and/or indirect effect of neurotoxic compounds released by the host [60]. Furthermore, AChE has recently received attention as a potential anthelmintic drug and vaccine target in nematodes [61].
Among the KEGG pathways, an important group that was enriched comprises the following carbohydrate metabolism pathways: glyoxylate and dicarboxylate metabolism, propanoate metabolism, glycolysis/gluconeogenesis, and pentose phosphate pathway. Indeed, carbohydrate metabolism is an essential energy source for Anisakis larvae [62]. Carbohydrates play important roles in many basic processes, including development, morphogenesis, immunity, and host-pathogen interactions [63]. Two sugars, trehalose and glycogen, were detected in A. simplex L3 larvae [64]. Łopieńska-Biernat et al. found [65] that trehalose plays a key role in providing energy during thermotolerance and starvation processes. It is also worth mentioning that the secretome was enriched in numerous members of the longevity-regulating pathway, which is associated with the regulation of numerous processes, such as oxidative stress, autophagy, glycogen accumulation, and fat accumulation. In addition, the secretome was found to be enriched in proteins belonging to the valine, leucine, and isoleucine degradation KEGG pathway. This KEGG pathway is important since valine, leucine, and isoleucine are likely to be essential amino acids in Anisakis, as is the case in C. elegans [66].
Approximately 21% of ES proteins were predicted to be essential for life. These are proteins that are critical to the survival of the cell or organism under certain conditions [67]. Among the top matches of secretome essential proteins such proteins were found as putative actin (A0A0M3J0M4), calmodulin (A0A0M3KFJ2), elongation factor 2 (A0A0M3K613), and HSPs (A0A0M3K5H6, A0A0M3K9V2, A0A0M3K4G1). Actin is a family of globular multi-functional proteins that form microfilaments [68]. These proteins participates in many important cellular processes, including muscle contraction, cell motility, cell division and cytokinesis, vesicle and organelle movement, cell signaling, and the establishment and maintenance of cell junctions and cell shape [68]. Calmodulin is a highly conserved protein ubiquitously and abundantly expressed in eukaryotic cells [69]. The functions of calmodulin include Ca 2+ binding and alteration of calcium signal transduction pathway to control a variety of biological processes, such as cytoskeletal assembly/reorganization, activation of phosphorylase kinase, abiotic stress responses, neurotransmission, smooth muscle contraction, metabolism, and cell motility [70]. Elongation factor 2 catalyzes the guanosine triphosphate-dependent ribosomal translocation step during translation elongation. HSPs are important molecular chaperones for maintaining cellular functions to prevent proteins from misfolding and aggregating in crowded surroundings. HSP expression levels increase when the organism is exposed to stress conditions, such as heat shock, alkaline treatment, and some chemical reagents, in order to help pathogens survive unfavorable conditions in the host [71][72][73]. Furthermore, essential for life proteins may be promising therapeutic targets for drugs and vaccines [74]. Such proteins include phosphoglycerate mutase (A0A0M3KAY8), which in this study was predicted to be an essential protein and shows homology with cofactor-independent phosphoglycerate mutase (iPGM) of C. elegans. This enzyme is involved in glycolytic and gluconeogenic pathways, and inhibition of iPGM activity has been shown to have a lethal effect on C. elegans [75]. Therefore, iPGM is considered a potential drug target or vaccine candidate in several nematodes, such as Wuchereria bancrofti [76], B. malayi [77], and Leishmania donovani [78].
Nine potential pathogenicity-related proteins in the A. simplex (s.s.) secretome were identified using database searching. The majority were HSPs which are the first line of attack and help in fortifying pathogen virulence [79,80]. The HSPs of Anisakis are poorly characterized; to date, only the expression patterns of HSP90 and HSP70 in Anisakis have been analyzed [81,82], whereas the contribution of Anisakis HSPs to pathogenicity has not been investigated. By contrast, HSP70 has a better-known role in the pathogenicity of Toxoplasma gondii, which is based on the modulation of nitric oxide production by macrophages [83]. HSP70 of T. gondii was found to be homologous with two top-matched potential pathogenicity-related proteins in the A. simplex secretome. Another two predicted pathogenicity-related proteins of A. simplex (s.s.) were homologous with GroEL of Bartonella, which is a chaperonin and exhibits pathogenicity via apoptosis inhibition and mitogenic stimulation of host cells [84]. The next two potential pathogenicity-related proteins in the Anisakis secretome were found to be homologous with thiol-specific antioxidant protein 1 of Cryptococcus, which is essential for resistance to oxidative, nitrosative, and temperature stress [85]. A homolog of hypothetical protein CNAG_05449 of Cryptococcus was also detected in the Anisakis secretome. This protein is in the metallothionein family, members of which play a crucial role in the pathogenicity and resistance of Cryptococcus against the host immune response, since they are directly involved in the detoxification of high concentrations of copper produced by macrophages fighting the infection [86]. Another protein detected in the Anisakis secretome is a homolog of the glycine cleavage system H protein of Francisella tularensis, which contributes to the intracellular replication of the pathogen in serine-limiting environments [87]. Another predicted pathogenicity-related protein of the A. simplex (s.s.) secretome is homologous with glucose-6-phosphate isomerase, which is required for the extracellular polysaccharide biosynthesis of Haemophilus influenzae [88].
Of the seven A. simplex ES allergens listed by WHO/IUIS [89], only Ani s 4 was detected in the present study. Ani s 4 is significant because of its heat-and pepsin-resistant properties and its ability to cause anaphylaxis [13,90,91]. Other Anisakis allergens were presumably not expressed in the in vitro culture conditions, or their concentration was below the limit of detection of LC-MS/MS. In addition to known allergens, potential allergens were also identified in this study. In order to increase the specificity of this analysis, hits detected using the FARRP database were confirmed by the AllerCatPro server [92], combining various bioinformatics approaches. The majority of the potential allergens detected in the Anisakis secretome are homologs of nematode and arthropod allergens, which is in line with the cross-reactions between Anisakis antigens and these organisms described by other authors [93][94][95]. Of the 18 potential allergens identified in this study, the following 6 were detected in our previous investigations of extracts from A. simplex L3 larvae: A0A158PP35, A0A0M3K5H6, A0A0M3KA05, A0A0M3JU57, A0A0M3K9V2, and A0A0M3K8L6 [13,33]. The first five are also proteins with potential thermostability [13]. Furthermore, many of the possible allergens detected in the Anisakis secretome show similarity to the potential allergens identified by other authors in the whole proteome and transcriptome of Anisakis larvae [32,96,97]. Faeste et al. [37] found in the A. simplex larvae the following proteins, including potential allergens which were similar to putative allergens identified in the present study: haemoglobin (P26914), troponin-like protein (Q9U3U5), HSP 70 (A8Q5Z6), triosephosphate isomerase (P91919), fructose-bisphosphate aldolase 1 (A8P3E5), and calmodulin (O16305). In particular, many sequences similar to the potential allergens of A. simplex detected in this study can be found at the transcriptome level in the ANISAKIS DB database (http://anisakis.mncn.csic.es/public/, accessed on 17 January 2021) [97]. Furthermore, a comparison of the secretome proteins identified in this study with the immunoreactive proteins of A. simplex esophageal gland cells provides interesting data [38]. Of the 13 immunoreactive proteins detected in esophageal gland cells, we also detected the following 4 in the secretome: uncharacterized protein (A0A0M3K6E2), uncharacterized protein (A0A0M3JQQ1), metalloendopeptidase (A0A0M3K299), and SCP domain-containing protein (A0A0M3K1U4). Thus, these ES proteins may also have allergic and/or diagnostic potential.
Interactome analysis was performed to identify and characterize proteins involved in interactions between A. simplex (s.s.) ES proteins as well as proteins involved in hostpathogen interactions. As was expected, such proteins primarily fall into the following groups: proteins of KEGG pathways, essential proteins, and proteases/protease inhibitors, followed by potential allergens and potential pathogenicity-related proteins. These groups of Anisakis proteins are characterized in detail above, and such profile composition is in accordance with the main functions of helminth ES proteins, i.e., penetration, colonization, survival in host tissues, incorporation of host metabolites, and modulation of the host immune response [98,99].
Among the human proteins involved in the interaction with Anisakis ES proteins, laminins were found to be the most highly represented. Laminins, which are the major component of the basal lamina, are enzymatically degraded by parasites such us Anisakis pegreffii during invasion [100], which facilitates internal migration of parasites [101]. Methyltransferase proteins were also highly abundant human proteins which were predicted in the host-parasite interactome. These proteins are known to contribute in deregulation of host expression profile which lead to host cell transformation, or escape of Apicomplexa parasites from the host immune system [102]. It is worth noting that among all human-parasite protein interactions, human polyubiquitin-C is the main target identified as interacting with Anisakis ES proteins. Indeed, ubiquitin is known to modulate host-pathogen interactions, with a particular focus on host innate immune defenses and pathogen immune evasion [103]. Among Anisakis secretome proteins, transaldolase (A0A0M3KAE3) was found to interact with the largest number of human proteins. This protein is an enzyme of pentose phosphate pathway, and a potential allergen. Another A. simplex (s.s.) protein which was predicted to interact with a large number of human proteins is proteasome sub-unit alpha type (A0A0M3JT99) which is characterized by its proteolytic activity. Another important Anisakis protein of predicted human-parasite interactome is heat shock 70 kDa protein cognate 1 (A0A0M3K9V2). This protein was predicted to be a potential protease, a potential allergen, a protein essential for life, and a possible pathogenicity-related protein.
It is known that heat shock 70 kDa cognate plays an important role in the interactions of many parasites with the host organism, since it is highly immunogenic and a target of B and T cells [104,105].
The number of proteins involved in the human-parasite interaction network is about four times the number of proteins in the fish-parasite interactome. This result could be mainly due to the fact that the available database of the non-human host-pathogen interaction is much more limited than for human-pathogen interactions. Among the fish proteins involved in the interaction with Anisakis secretome proteins, Rab-family proteins were found to be the most abundant. These proteins regulate virtually all membrane trafficking events in eukaryotic cells [106,107]. Other abundant proteins in fish-parasite interactome were phosphodiesterases (PDEs). PDEs are metallohydrolases that control the concentration of second messengers cyclic adenosine monophosphate and cyclic guanosine monophosphate. Among Anisakis secretome proteins, Rab GDP dissociation inhibitor (GDI) (A0A0M3JZR1) was found to interact with the largest number of fish proteins. This protein is involved in regulation of GDP-GTP exchange between Rab-family proteins. Rab GDI is an immunoreactive protein of Trichinella britovi [108], but its role in host-parasite interactions is poorly known. Similarly, calmodulin (A0A0M3KFJ2) of Anisakis secretome was predicted to interact with fish proteins. Furthermore, calmodulin it is known immunogenic protein of Fasciola hepatica secretome [109].

ES Proteins Preparation
ES proteins were prepared by incubating 100 viable L3 A. simplex (s.s.) larvae in 5 mL of sterile PBS. After 24 h of incubation at 36 • C, medium containing ES proteins was collected and clarified by centrifugation at 20,000× g for 15 min at 4 • C. Supernatants were concentrated by ultrafiltration at 4 • C (3 kDa cutoff membrane; Thermo Fisher Scientific, Rockford, IL, USA). Subsequently, protein concentration was measured using a spectrophotometer (NanoPhotometer P330, Implen, München, Germany) and secretome samples were stored at −80 • C for further experiments. According to this procedure, three independent biological replicates of ES proteins were prepared.

Sample Processing and LC-MS/MS Analysis
Three batches of A. simplex (s.s.) ES proteins were subjected to in-solution digestion and LC-MS/MS analysis, as described in our previous publication [13]. Briefly, secretome samples were analyzed using a Q Exactive mass spectrometer (Thermo Electron Corp., San Jose, CA, USA) coupled with nano-high-performance liquid chromatography (nano-HPLC) RP-18 column (internal diameter 75 µm; Waters, Milford, MA, USA). Proteins were identified with Mascot search engine server (ver. 2.5; Matrix Science, London, UK; http: //www.matrixscience.com/server.html, accessed on 25 October 2019) using the A. simplex proteome (proteome ID: UP000036680; 20,789 sequences) obtained from the Universal Protein Resource (UniProt, http://www.uniprot.org/, accessed on 25 October 2019) [116]. The following Mascot search parameters were applied: trypsin digestion allowing one missed cleavage, parent ions was set to 5 parts per million (ppm), and fragment ions was set to 0.01 dalton (Da). The ion type was set as monoisotopic, and protein mass was set as unrestricted. Beta-methylthiolation of cysteine was used as a fixed modification, whereas oxidation of methionine was set as a variable modification. Peptides were accepted at False Discovery Rate (FDR) ≤ 0.98%, ion score ≥ 38, and significant threshold of p ≤ 0.00026. Only proteins detected with at least one unique peptide and proteins identified in all three biological replicates were accepted for further analysis. A detailed procedure of samples processing and LC-MS/MS analysis is presented in Supplementary File S2.
Enriched GO annotations and enzymes of identified secretome proteins were compared with those of the whole proteome of A. simplex (proteome ID: UP000036680) by two-tailed Fisher's exact test using OmicsBox software. These tests were performed using a p-value cutoff of 0.05 to indicate significance. Proteins involved in KEGG pathways were detected using the KOBAS 3.0 server (http://kobas.cbi.pku.edu.cn/, accessed on 23 April 2021) [120] based on the C. elegans proteome as a reference. Enriched KEGG pathways were identified by Fisher's exact test using KOBAS 3.0 with the C. elegans proteome as a reference and the whole A. simplex proteome as background. In this case, a p-value less than 0.05 with Benjamini and Hochberg correction was considered to indicate significance.
All BLASTP analyses in this study were performed using the OmicsBox software. Only hits with a BLAST e-value ≤ 1.0 × 10 −5 and similarity ≥70% were considered. The databases, software tools, and servers used for bioinformatics analyses in this study are listed in Supplementary Table S1.

Conclusions
Proteomic analysis was performed for the first broad-scale identification and characterization of ES proteins of A. simplex (s.s.) L3 larvae. A total of 158 proteins, belonging to 143 different proteins families, were identified in Anisakis secretome using mass spectrometry technique. Comparison of Anisakis secretome proteins with ES proteins of closely related nematodes revealed that the A. simplex secretome contains a relatively high number of proteins with a low level of overall similarity to ES proteins of related parasites. Prediction of secretory pathways allowed the classification of the majority of proteins (approximately 49% of ES proteins) to the unconventional route. In addition, six Anisakis proteins previously known to be associated with EVs were detected and 24 new possibly EV-associated proteins were predicted. GO annotations, KEGG pathways, and enzymes were assigned to ES proteins and enrichment analysis of these terms was performed by comparison with whole A. simplex proteome. The most enriched GO annotations were terms related to the glycolytic process, larval development, antioxidants, and cuticle, while among the KEGG pathways the main enriched group was associated with carbohydrate metabolism. Furthermore, proteases were found to be highly represented enzymes in the secretome (17% of ES proteins). Another finding was identification of essential proteins (approximately 21% of ES proteins) that are indispensable for the survival of an organism. Important findings were identification of pathogenicity-related proteins, allergens, and potential allergens. Nine potential pathogenicity-related proteins were predicted, which were mostly homologs of chaperones. Of all secretome proteins, one was identified as an allergen, which was Ani s 4, and 18 were putative allergens, most of which were homologs of nematode and arthropod allergens. Another finding was prediction of proteins possible involved in interactions between A. simplex ES proteins as well as proteins involved in interactions between hosts and parasite.
As summarized above detected ES proteins play an important role in many biological processes and provide a better understanding of A. simplex survival, development, and invasion strategy. In addition, the identified secretome proteins could be used as targets for new drugs, vaccines, and diagnostic assays. Nevertheless, it should be noted that functional analysis of ES protein was performed using a bioinformatics approach. Therefore, future in vitro and in vivo studies are needed to confirm our findings regarding the role of detected proteins.  .  Table S1: The databases, software tools, and servers used for bioinformatics analyses in the present study. Figure S1: Secretome protein analysis of A. simplex (s.s.) L3 larvae using SDS-PAGE stained with SYPRO Ruby. Molecular weight (MW) estimations are presented in kilodaltons (kDa). Figure  S2: Restriction fragment length polymorphism patterns of the ITS region of the rDNA of Anisakis L3 larvae using the Hinf I (A) and HhaI (B) restriction enzymes.