Tissue-specific and sex-biased glycoproteomic landscape of Schistosoma mansoni

Abstract

Schistosomiasis is a major global health burden, with praziquantel as the sole treatment. Its inability to prevent reinfection highlights the need for new interventions such as vaccines. Protein glycosylation is essential for parasite biology and immune evasion, yet remains poorly characterized in schistosomes. Here, we conduct N- and O-glycoproteomic and intact glycopeptides analyses of adult male and female Schistosoma mansoni, integrating single-cell transcriptomics. We uncover tissue-specific and sex-biased glycosylation patterns, with greater glycan complexity in the parenchyma and gut and reduced diversity in muscles and neurons. Female- and male-biased glycoproteins are linked to key sex-specific functions. We establish a glycan database for S. mansoni and identify unclassified glycans and HexA modifications. Disruption of glycosylation via RNAi targeting four glycosyltransferases significantly impairs parasite viability. We further show that several vaccine candidates are glycoproteins and characterize their glycosylation. This work provides a foundational glycoproteomic resource supporting the development of glycan and glycoprotein-based strategies for schistosomiasis control.

This study defines tissue- and sex-specific N- and O-glycoproteomes of adult Schistosoma mansoni, uncovers novel and HexA-modified glycans, shows glycosylation is essential for parasite survival, and reveals vaccine candidates as glycoproteins.

Introduction

Schistosomiasis, a neglected tropical disease caused by parasites of the genus Schistosoma, poses a significant global public health and economic burden. It is estimated that ~240 million people are infected worldwide, resulting in over 250,000 deaths annually¹. Schistosomes have a complex life cycle, relying on aquatic snails as intermediate hosts and mammals as definitive hosts. Among these, Schistosoma mansoni is the most widely distributed species, transmitted through intermediate hosts such as Biomphalaria glabrata, and causes severe hepatic diseases in Africa, the Arabian Peninsula, and Latin America². Currently, the treatment of schistosomiasis primarily relies on praziquantel, the only available drug³. While praziquantel is effective against adult worms, it has low effect on larval stages, cannot prevent reinfection, and faces potential drug resistance issues^4,5. Therefore, developing novel control strategies, particularly vaccines, has become an urgent priority. Four potential vaccine candidates are currently in various stages of clinical development, including S. mansoni calpain (Sm-p80/GLA-SE)⁶, S. mansoni tetraspanin (Sm-TSP-2)⁷, S. mansoni 14-kDa fatty acid-binding protein (Sm14/GLA-SE)⁸, and S. haematobium 28-kDa glutathione S-transferase (Sh28GST/Alhydrogel)⁹. However, none of them has been successfully developed as a vaccine.

Protein glycosylation is a crucial post-translational modification conserved across all domains of life, encompassing primarily asparagine (Asn)-linked N-glycosylation and serine (Ser)/threonine (Thr)-linked O-glycosylation. This process typically begins with the transfer of glycans in the endoplasmic reticulum (ER), followed by further maturation in the Golgi apparatus, where various glycosyltransferases and glycosidases modify the glycan structures. The diversity of glycan structures and their functions vary widely across species¹⁰. Mass spectrometry-based analyses have identified several typical glycan structures and motifs in schistosomes, including core β2-xylose, core α3-fucose, Galβ1–4(Fucα1–3)GlcNAc (Lewis X), GalNAcβ1-4GlcNAc (LDN), and GalNAcβ1–4(Fucα1–3)GlcNAc (LDN-F)^11–13. Glycomics studies have systematically characterized the glycan compositions of S. mansoni eggs, cercarial secretions, and various life cycle stages, revealing gender-specific expression and differential localization of certain glycan motifs^11–13.

Protein glycosylation plays diverse biological roles, including protein folding, stability, cell-cell interactions and immunity¹⁰. In pathogens, protein glycosylation has been recognized as a crucial contributor to the immunomodulatory properties of antigens, which can significantly influence their immunogenicity. For instance, in a vaccination setting, the native gut-derived glycoprotein H11 from Haemonchus contortus induces strong protective immunity, whereas the recombinant H11 protein shows lower protection¹⁴. A possible key difference lies in the absence of specific glycans in the recombinant form, which are present in the natural product¹⁴. In schistosomes, protein glycosylation has been linked to the modulation of host immune responses. The glycan antigens are widely expressed across the schistosome life cycle, continuously exposed to the host immune system. Specifically, the egg-derived glycoprotein omega-1 plays a key role in promoting Th2-mediated immune regulation¹⁵. These insights underscore the significance of glycosylation in parasite-host interactions and immune responses.

Despite the importance of understanding glycobiology in parasite development and host interactions, challenges in detection technologies and analytical strategies persist. Current glycomics studies typically provide broad glycan compositions derived from glycoproteins and lipids but fall short of revealing specific associations with proteins or site-specific glycosylation patterns. As a result, schistosome glycoproteomics remains in its early stages, with only a few egg glycoproteins (e.g., IPSE/alpha-1, kappa-5, and omega-1) having fully characterized glycosylation structures^15–17. On the other hand, Liquid chromatography-tandem mass spectrometry (LC-MS/MS) analysis of intact glycopeptides, which simultaneously resolves peptide sequences, glycosites, and glycan compositions, is a cornerstone of glycoproteomics research¹⁸. However, most tools for intact glycopeptide identification, such as pGlyco3¹⁹ and MSFragger-Glyco²⁰, rely on predefined glycan libraries, limiting their ability to detect rare glycans not included in these databases. Recently, the development of the pGlycoNovo tool, which uses a comprehensive, free oligosaccharide search algorithm, allows for open glycan search and glycopeptide identification without relying on predefined libraries²¹. This innovation opens new avenues for exploring the full diversity of glycosylation in biology.

In this study, we systematically characterize the N- and O-glycoproteomes, as well as intact glycopeptidomes, of adult male and female S. mansoni. By integrating single-cell transcriptomic data²², we uncovered tissue-specific heterogeneity in glycosylation and identified sex-specific differences in glycoproteins and site-specific glycans, revealing distinct sex-biased glycosylation patterns in schistosomes. With this newly constructed comprehensive database of protein and site-specific glycosylation for S. mansoni, we were able to identify de novo-derived glycans and HexA modifications. Additionally, we meticulously characterized the glycan patterns on host-parasite interaction interfaces that may serve as potential vaccine candidates. Through RNA interference targeting various glycosyltransferases, we demonstrated that glycosylation is essential for maintaining schistosome viability and development. These findings provide a comprehensive resource for schistosome glycosylation research and offer a theoretical foundation for the development of novel glycosylation-based control strategies.

Results

Mapping of the S. mansoni glycoproteome

To map the N- and O-glycoproteome of S. mansoni, we harvested adult parasites at 42 days post-infection (dpi) and separated them by sex. As shown in Fig. 1a, male and female worms were processed using distinct approaches to generate N- and O-glycopeptides separately, followed by detection via LC-MS/MS. Then, we employed both the database-dependent pGlyco3¹⁹ and the glycan-first, full-range Y-ion de novo search strategy of pGlycoNovo²¹ to comprehensively identify N-glycoproteins and glycopeptides, with site-specific glycopeptides filtered by GlyScore >20. To enhance the pGlyco3 strategy, we first expanded the glycan database by integrating previously reported glycan compositions from S. mansoni eggs, cercariae (and their secretions), adult worms, and other life-cycle stages^{11–13,23–27} (Supplementary Data 1). For O-glycosylation analysis, due to its high complexity, only pGlycoNovo was employed. Quantitative analysis using pGlycoQuant²⁸ enabled the construction of a large-scale intact N- and O-glycoproteomic library for S. mansoni.

Fig. 1. Comprehensive N-glycoproteomic profiles of S. mansoni.

a Overall workflow of N- and O- intact glycopeptide profiling in S. mansoni. b Distribution of sequence motif of N-glycosites in S. mansoni. (NXS/NXT/NXC, where N is asparagine, X is any amino acid except proline, S is serine, T is threonine, C is cysteine.) c Distribution of singly and multiply glycosylated proteins in S. mansoni. d Distribution of the number of glycans at each site in S. mansoni. e Distribution of glycan size in S. mansoni. f Distribution of glycan types in S. mansoni. g Proportion of specific monosaccharides in S. mansoni. Fuc for fucose, Xyl for xylose, HexA for hexuronic acid, NeuAc for N-acetylneuraminic acid, NeuGc for N-glycolylneuraminic acid. h Glycan composition similarity between S. mansoni and five model organisms. i Molecular function enrichment of N-glycoproteins in S. mansoni. j Protein domain enrichment of N-glycoproteins in S. mansoni. k KEGG pathway enrichment of N-glycoproteins in S. mansoni.

Characterization of site-specific N-glycoproteome in S. mansoni

The N-glycoproteomic dataset included 3274 site-specific N-glycans on 399 glycoproteins and 14,974 glycopeptide-spectrum matches (GPSMs), all reliably identified at a 1% false discovery rate (FDR) at the intact glycopeptide level (Supplementary Data 2). To explore the diversity and species-specificity of N-glycosylation, we performed statistical analyses and compared our data with recently reported site-specific N-glycosylation data from five model organisms (Mus musculus, Arabidopsis thaliana, Drosophila melanogaster, Danio rerio, and Caenorhabditis elegans)²¹ (Supplementary Data 3). We identified canonical glycosylation motifs in S. mansoni, with threonine motifs (53.39%) being more frequently glycosylated than serine motifs (44.65%) (Fig. 1b). Most N-glycoproteins (59.24%) in S. mansoni primarily contained a single glycosite (Fig. 1c), and over half of the glycosites carried multiple glycans (Fig. 1d). In terms of glycan size, S. mansoni exhibited a distribution similar to that of C. elegans and D. melanogaster²¹ (Supplementary Data 3), with uniformly distributed chains of 6–12 monosaccharides and approximately 7.51% small glycans (Fig. 1e).

Based on their distribution patterns, the site-specific glycans in S. mansoni can be classified into five categories, as shown in Fig. 1f. Mature schistosomes exhibit nearly equal proportions of complex and high-mannose glycans, aligning with previous findings¹². Compared to the model organisms, S. mansoni displayed the highest proportion of truncated glycans (5.80%) and a similar proportion of unclassified glycans to that identified in A. thaliana and M. musculus²¹ (Fig. 1f and Supplementary Data 3). These results indicate that the glycans in S. mansoni are remarkably diverse and exhibit unique features, which may be associated with its parasitic biology.

By analyzing site-specific glycans based on modification types, we found that fucosylated glycans accounted for 63.44% of the total (Fig. 1g), exceeding the levels observed in five model organisms and even surpassing A. thaliana (61.43%), previously reported to have the highest²¹ (Supplementary Data 3). Additionally, xylosylated glycans accounted for 7.67% (Fig. 1g), significantly higher than in the four analyzed animal species (all <1%)²¹ (Supplementary Data 3). These fucosylated glycans (including structures such as LeX and LDN-F) and xylosylated glycans have been widely reported to be highly prevalent in S. mansoni and are known to play important functional roles^29,30.

We also identified a high proportion of hexuronic acid (HexA), while NeuAc and NeuGc were nearly absent (Fig. 1g). HexA modification has been reported in the circulating cathodic antigen (CAA) of adult schistosomes and the glycosphingolipids (GSLs) of schistosome eggs^31,32. Previous studies have reported that schistosomes lack endogenous sialic acid modifications and suggested that any detected sialylated glycoproteins in worm preparations are likely host-derived^33–36. The sialic acid detected in our dataset aligns with these observations and is most plausibly attributable to host contamination.

Further comparative analysis of the S. mansoni N-glycan library with recently reported model organisms revealed the highest similarity with Mus musculus and the lowest with C. elegans²¹ (Fig. 1h and Supplementary Data 4).

To analyze glycoproteins, we mapped 399 glycoprotein entries to 415 potential glycoproteins, as some peptides matched multiple proteins (Supplementary Data 2 and 5). By examining the number of unique glycans per protein relative to the number of glycosylation sites, we identified several glycoproteins with high glycan heterogeneity, such as α−2-macroglobulin, Hepatopoietin-A, Serpin, and Cathepsin B1.1 (B1.2) (Supplementary Fig. 1). Molecular function enrichment analysis using DAVID database³⁷ revealed that 415 N-glycoproteins were predominantly associated with hydrolases, receptors, ion channel and glycosylation-related enzymes (Fig. 1i), suggesting their critical roles in metabolic regulation and signal transduction. Enriched protein domains included Immunoglobulin domain (IG), Epidermal Growth Factor-like domain (EGF), and Fibronectin type III domain (FN3) (Fig. 1j), which are involved in immune responses, development, and cell communication. Gene Ontology (GO) enrichment showed that N-glycoproteins were primarily localized to biological membranes and extracellular regions, with some involved in cell adhesion and extracellular matrix organization (Supplementary Fig. 2), further supporting their roles in cell-cell interactions, tissue maintenance, and potential functions at the host-parasite interface. KEGG pathway analysis indicated that these glycoproteins are involved in metabolic processes, lysosomal function, and N-glycan biosynthesis (Fig. 1k). These findings suggest that N-glycoproteins play crucial roles in the development, metabolism, and host interactions of S. mansoni.

Tissue-specific heterogeneity of N-glycosylation in S. mansoni

To further investigate the heterogeneity of N-glycosylation across different tissues of S. mansoni, we integrated published single-cell transcriptomic data²² with our N-glycoproteomic database to explore tissue-specific expression patterns of glycoproteins and glycopeptides (Supplementary Fig. 3-4 and Supplementary Data 6). Our analysis revealed that 28% of the glycoproteins in the database were tissue-specific glycoproteins, corresponding to 37% of tissue-specific glycopeptides (Supplementary Fig. 5). These glycosylated molecules were widely distributed across various tissues of S. mansoni, with particularly high proportions in the parenchyma, muscles, neurons, tegument, and gut (Supplementary Fig. 5).

Notably, we observed significant tissue-specific differences in glycosylation patterns when calculating the average number of glycans per glycoprotein and glycosite (Fig. 2a-b). Glycoproteins in the gut exhibited markedly higher glycosylation complexity at both the protein and glycosite level. In contrast, muscles and neurons displayed more streamlined glycosylation patterns with lower degrees of modification (Fig. 2a, b).

Fig. 2. Tissue-specific N-glycoproteomic profiles of S. mansoni.

a Number of site-specific glycans per glycoprotein in each tissue. b Number of site-specific glycans per glycosite in each tissue. c Distribution of singly and multiply glycosylated proteins in each tissue. d Distribution of glycan size in each tissue. e Distribution of the number of glycans at each glycoprotein in each tissue. f Distribution of glycan types in each tissue. g Proportion of specific monosaccharides in each tissue. Fuc for fucose, Xyl for xylose, HexA for hexuronic acid.

Although most tissue-specific N-glycoproteins contained only a single glycosite, glycoproteins in the parenchyma and muscles exhibited a broader distribution of glycosite numbers per protein (Fig. 2c). In contrast, glycoproteins in the gut and tegument rarely contained more than two glycosites (Fig. 2c). Regarding glycan size, large glycans (8–15 monosaccharides) were predominantly found in the vitelline gland, tegument, parenchyma, gut, and flame cells (Fig. 2d). Conversely, neurons and muscles lacked glycans larger than 13 monosaccharides, with a predominance of smaller glycans (1–7 monosaccharides) (Fig. 2d). Additionally, glycoproteins in the gut and parenchyma exhibited the highest glycan abundance, whereas neurons and muscles had a more limited range (Fig. 2e). Among the five glycan types, tissue-specific glycoproteins displayed distinct expression profiles (Fig. 2f). Notably, high-mannose glycans were absent in esophageal gland-specific proteins, while hybrid glycans were undetected in neurons, muscles, and Mehlis’ gland (Fig. 2f). Regarding specific monosaccharides, approximately 80% of the glycans in the tegument, vitelline gland, and flame cells were fucosylated, while xylosylations were predominantly observed in the vitelline gland and parenchyma (Fig. 2g). Additionally, HexA were more abundant in the gut and Mehlis’ gland (Fig. 2g). These findings highlight the significant heterogeneity and diversity of N-glycosylation in mature S. mansoni.

Sex-specific differences and biases in N-glycosylation of S. mansoni

Schistosomes exhibit pronounced sexual dimorphism, with males and females differing both morphologically and biologically. To investigate sex-specific differences in N-glycosylation, we analyzed quantitative N-glycoproteomic profiles derived from pGlycoQuant (Supplementary Data 7, 8). This analysis identified differentially expressed glycoproteins between the sexes (Supplementary Data 9). To ensure a systematic comparison, we normalized intact glycopeptide datasets using glycoprotein quantification values, allowing for standardized analysis of site-specific glycan sex biases (Supplementary Fig. 6, 7 and Supplementary Data 10). Quantitative analysis revealed that 52 glycoprotein entries were highly expressed in females and 133 in males, corresponding to 59 and 147 glycoproteins, respectively (Supplementary Fig. 8a and Supplementary Data 9). Furthermore, 473 distinct glycopeptide combinations exhibited a significant female-specific abundance bias, while 525 showed a significant male bias (Supplementary Fig. 8a and Supplementary Data 10). These differentially expressed N-glycoproteins and glycopeptide combinations with significant sex-specific variations are highlighted in Fig. 3a, b.

Fig. 3. Sex-specific N-glycoproteomic profiles of S. mansoni.

a Volcano plot of differentially expressed N-glycoproteins between male and female worms. b Volcano plot of differentially expressed N-glycopeptide combinations between male and female parasites. Statistical significance was assessed using limma (two-sided moderated t-test), with Benjamini–Hochberg correction applied for multiple comparisons. The labels in the figure are formatted as “[Protein] [Site] [Glycan Composition]”. c Number of tissue-specific glycoproteins highly expressed in male and female S. mansoni. d Distribution of glycan types and tissue specificity of sex-biased site-specific N-glycans in S. mansoni. e Number of sex-biased site-specific N-glycans with specific monosaccharides in S. mansoni. Fuc for fucose, Xyl for xylose, HexA for hexuronic acid. f Heatmap showing sex-biased glycans of α-2-macroglobulin (Smp_089670) in male and female S. mansoni.

We then conducted a sex-related, tissue-specific analysis of glycoproteins to visualize their distribution (Supplementary Fig. 9). Glycoproteins highly expressed in males were predominantly localized in muscles, neurons, parenchyma, tegument, and esophageal gland (Fig. 3c), potentially associated with motility, pairing and host interactions. In contrast, glycoproteins highly expressed in females were concentrated in the vitelline gland, gut, and Mehlis’ gland (Fig. 3c), likely linked to digestion and oviposition.

The distribution of sex-specific glycans across the five major glycan types was similar between the males and females, with comparable proportions observed in each tissue (Fig. 3d). The numbers of glycans containing each specific monosaccharide were broadly comparable between males and females (Fig. 3e), and the overall glycan size distribution was also similar across sexes (Supplementary Fig. 8b). These observations indicate that sex-specific N-glycosylation differences primarily arise from compositional variations, such as differences in the numbers of specific monosaccharides within glycans. Throughout a detailed exploration, we found that glycans with 3–4 hexoses (Hex) and less than 5 N-acetylhexosamines (HexNAc) were more abundant in males, while females exhibited a bias for glycans with 5–7 Hex and more than 5 HexNAc (Supplementary Fig. 8c, d). Additionally, males exhibited a higher proportion of mono- and tri-fucosylated glycans, whereas females showed an enrichment of di-fucosylated glycans (Supplementary Fig. 8e). These compositional differences may reflect distinct glycan motifs unique to male and female S. mansoni.

To further investigate glycosylation heterogeneity between the sexes at the individual protein level, we analyzed three selected proteins—α−2-macroglobulin (Smp_089670), beta-hexosaminidase (Smp_053900), and prominin protein (Smp_179660)—in detail (Fig. 3f and Supplementary Fig. 10). For example, α−2-macroglobulin exhibited 64 different types of site-specific glycans across its two glycosites, with 18 glycans significantly upregulated in males and 5 glycans more prevalent in females (Fig. 3f). Similarly, beta-hexosaminidase and prominin protein also displayed significant sex-specific glycosylation differences (Supplementary Fig. 10). Although these proteins showed no significant differences in expression levels between the sexes, their site-specific glycan modification patterns exhibited clear sex-specific biases. This observation raises the possibility that differential glycosylation could contribute to sex-specific functional modulation of these proteins.

Identification and validation of S. mansoni N-glycans

Comparison of pGlyco3 and pGlycoNovo identifications revealed that 90.3% of the site-specific glycans identified by pGlycoNovo were also detected by pGlyco3, supporting its accuracy. A total of 8% of site-specific glycans (262 entries) were unique to pGlycoNovo (Fig. 4a and Supplementary Data 6), including 178 unclassified, 20 truncated, 35 high-mannose, 1 hybrid, and 28 complex glycans. In Fig. 4b, we presented 38 site-specific glycans with GlyScore >120, whose high scores indicate strong reliability. Representative annotated spectra of glycopeptides bearing two, three, and four fucose residues were presented in Fig. 4c and Supplementary Fig. 11, respectively. A series of matched Y ions validated the accuracy of their assignments.

Fig. 4. Characterization of unclassified glycans and HexA residues in S. mansoni.

a Venn diagram showing the number of site-specific N-glycans identified by pGlyco3 and pGlycoNovo. b Distribution of GlyScore for glycans identified by pGlycoNovo. Only glycans with a GlyScore >120 are shown. c Annotated spectrum of H(3)N(6)F(3)X(1) on the glycopeptide ‘APFNEJK’ of Smp_025830. d Annotated spectrum of H(7)N(2)F(1)HA(1) on the glycopeptide ‘TPVVQMGPIHJR’ of Smp_149370. The peptide sequence includes “J” indicating the N-glycosite. Glycan symbols: green circle for Hex, blue square for HexNAc, red triangle for fucose, yellow star for xylose, and color block diamond for HexA.

Furthermore, our data provided evidence for the presence of hexuronic acid. This identification was strongly supported by multiple matched b- and y-ions in the mass spectra, minimal mass deviations, and abundant diagnostic glycan fragment ions (Fig. 4b, c and Supplementary Fig. 12).

At the glycan-composition level, 69% (120/174) of previously reported S. mansoni compositions were recovered in our 6-week adult worm dataset, including 106 identified by both pGlycoNovo and pGlyco3. In addition, 84 compositions were uniquely identified by pGlycoNovo, 15 were unique to pGlyco3, and 25 were detected by both tools but have not been reported previously (Supplementary Fig. 13 and Supplementary Data 11).

We further compared our identified N-glycan compositions with previously published datasets (Supplementary Data 1). Among the 117 compositions described across life stages by Smit et al.¹², 99 (84.6%) were recovered in our dataset, including all 59 (100%) compositions reported for 6-week adult worms. Likewise, we detected 45 of 49 (91.8%) egg secretions-derived glycans reported by Lee et al.¹¹, 33 of 34 (97%) egg glycans described by Khoo et al.²³, and all 33 (100%) cercarial N-glycans reported by Khoo et al.²⁴.

Together, these results show that our dataset captures the majority of previously reported N-glycans while also identifying reliable new glycopeptides and compositions. These findings expand the glycan composition library of S. mansoni and provide new insights into the diversity of schistosome N-glycans.

Characterization of site-specific O-glycoproteome in S. mansoni

By removing N-glycosylation from the O-glycoproteomic data obtained from pGlycoNovo and pGlycoQuant (Supplementary Data 12-14), we successfully constructed an O-glycoproteomic dataset comprising 89 O-glycoprotein entries (corresponding to 99 proteins) and 258 site-specific glycans (Supplementary Data 15). The key glycosylation features of S. mansoni O-glycoproteins were as follows: approximately 90% of O-glycoproteins contained 1 to 2 glycosites (Fig. 5a), with 74.5% of these sites modified by a single glycan (Fig. 5b). O-glycans were primarily composed of hexoses (Hex) and N-acetylhexosamines (HexNAc), with a minority of glycans lacking Hex (Fig. 5c). Additionally, 35.7% of the glycans were fucosylated, and the proportion of HexA modifications was significantly higher than in N-glycosylation. Notably, S. mansoni exhibited a uniquely high proportion of xylosylation (8.1%) (Fig. 5d). These distinct characteristics highlight fundamental differences between O-glycosylation and N-glycosylation modification patterns in S. mansoni.

Fig. 5. Comprehensive analysis of protein O-glycosylation in S. mansoni.

a Distribution of singly and multiply glycosylated proteins in S. mansoni. b Distribution of the number of glycans at each site in S. mansoni. c Distribution of glycans with and without Hex in S. mansoni. d Proportion of specific monosaccharides in S. mansoni. Fuc for fucose, HexA for hexuronic acid, Xyl for xylose. e Venn diagram of N-glycoproteins and O-glycoproteins. f Tissue specificity, glycosites, and glycan numbers of unique O-glycoproteins. The numbers on the right indicate the O-glycosites and O-glycans numbers for each gene. g Proportion (pie chart) and quantity (bar chart) of tissue-specific O-glycoproteins (left) and site-specific O-glycans (right). h Tissue-specific distribution of sex-biased and non-significant O-glycans, and residue modification proportions in gut, non-specific tissue, esophageal gland, and parenchyma. Fuc for fucose, HexA for hexuronic acid, Xyl for xylose. i Annotation of O-glycosites and site-specific O-glycans in MEG-14 (Smp_124000.1). O-glycosites are indicated in red.

Further comparative analysis of O-glycoproteins and N-glycoproteins revealed that 44 proteins exhibited both N- and O-glycosylation, while 55 proteins were uniquely O-glycosylated (Fig. 5e and Supplementary Data 16). Functional annotation indicated that these unique O-glycoproteins were associated with diverse biological functions, including enzymatic activity, membrane protein receptors, and potential transcriptional regulation (Supplementary Data 16).

Tissue and sex heterogeneity of O-glycosylation in S. mansoni

According to the single-cell transcriptomic atlas of S. mansoni²², 74.5% (41/55) of the unique O-glycoproteins were predominantly non-tissue-specific (Fig. 5f). These proteins exhibited relatively simple O-glycosylation patterns, with most being modified by single glycans at individual sites (Fig. 5f). Notably, the esophageal gland-specific protein MEG-14 (Smp_124000) stood out with the highest number of glycosylation sites and glycans (Fig. 5f), a feature likely linked to its specialized function within this gland.

Among all 99 O-glycosylated proteins, 31 glycoproteins and 104 site-specific glycans exhibited significant tissue-specific distribution patterns (Fig. 5g, Supplementary Fig. 14 and Supplementary Data 15). The gut, esophageal gland, and parenchyma contained the highest number of tissue-specific O-glycoproteins and exhibited the most diverse glycan profiles. In contrast, the vitelline gland, muscles, neurons, and tegument displayed simpler glycosylation patterns (Fig. 5g). Only a small number of proteins and glycopeptide combinations exhibited sex bias (Supplementary Fig. 15, 16 and Supplementary Data 17, 18). Notably, gut-specific glycans displayed a significant male bias (Fig. 5h), suggesting that certain gut O-glycans may play specialized roles in males. In contrast, O-glycans in other tissues did not show significant sex-specific distribution patterns observed in N-glycans (Fig. 5h). Interestingly, the esophageal gland—despite O-glycan abundance—shows the simplest modification patterns (Fig. 5g, h).

MEG-14 (Smp_124000) is an esophageal gland-specific protein that involved in host immune regulation by interacting with neutrophil inflammatory protein S100³⁸. Despite being a small protein composed of only 143 amino acids, it contained the highest number of glycosylation sites and glycans. As shown in Fig. 5i, MEG-14 exhibited 18 distinct glycan types across 9 serine or threonine glycosites.

Furthermore, leveraging our newly constructed N/O- glycosylation database, we characterized the overall glycosylation features of each glycoprotein. For example, Smp_194050, a schistosome-specific, uncharacterized glycoprotein with high glycosylation complexity and higher expression in the tegument and parenchyma, was found to contain 20 O-glycosites and 3 N-glycosites. These sites were modified by 62 site-specific O-glycans and 67 N-glycans, respectively (Supplementary Fig. 17).

RNAi of glycosyltransferases reveals the functional importance of glycosylation in S. mansoni

Considering the complex N- and O-glycosylation in the proteins of S. mansoni, we reasoned that protein glycosylation played critical roles in this parasite. Thus, we identified four key glycosyltransferases in schistosomes and performed in vitro RNAi on them (Fig. 6a). These included two O-glycosylation-related enzymes—OGT (O-linked N-acetylglucosamine transferase, Smp_046930) and C1GALT1 (Glycoprotein-N-acetylgalactosamine 3-beta-galactosyltransferase 1, Smp_051930)—and two N-glycosylation-related enzymes—ALG1 (Chitobiosyldiphosphodolichol beta-mannosyltransferase, Smp_005010) and ALG11 (GDP-Man:Man(3)GlcNAc(2)-PP-Dol alpha-1,2-mannosyltransferase, Smp_052330), which showed high sequence conservation to their invertebrate homologs (Supplementary Figs. 18, 19).

Fig. 6. Key enzymes in glycosylation pathways of S. mansoni.

a In vitro RNA interference workflow for paired mature S. mansoni. b Biosynthetic pathways of glycosylation catalyzed by OGT, C1GALT1, ALG1, and ALG11. Arrows indicate the monosaccharide residues added by the respective enzymes. S is serine, T is threonine. c Light microscopy images showing the parasites following RNAi of Ogt, C1galt1, Alg1, and Alg11. Scale bars, 1000 μm. d qPCR showing the relative mRNA levels of Ogt, C1galt1, Alg1, and Alg11 in control (Ctrl) RNAi or gene-specific RNAi groups. Data are presented as mean ± SD from three biological replicates, two-sided t-test. Exact P values, ordered from top to bottom, are 0.0488, 0.0005, 0.0003, and 0.0114 (*, p < 0.05; ***, p < 0.001).

OGT is a key enzyme that catalyzes the O-GlcNAcylation of substrate proteins³⁹, and C1GALT1 is a crucial enzyme in the O-glycosylation pathway, generating the core 1 O-glycan structure. It catalyzes the transfer of galactose (Gal) from UDP-Gal to Tn antigen to form the T antigen⁴⁰ (Fig. 6b). Both enzymes act at an early stage of their respective glycosylation pathways. Following 30-day knockdown on either Ogt or C1galt1 led to severe damage in the adult worms. Specifically, silencing Ogt led to obvious tegument damage and the death of both male and female worms. While C1galt1-RNAi caused protrusions on the parasite head, as well as intestinal swelling (Fig. 6c, d).

ALG1 and ALG11 are key enzymes in the N-glycosylation pathway, involved in the assembly of lipid-linked oligosaccharide (LLO) precursors⁴¹. ALG1 catalyzes the addition of the first mannose residue to the dolichol-linked oligosaccharide chain on the cytoplasmic side of the endoplasmic reticulum, while ALG11 adds the fourth and fifth mannose residues during the early ER-associated phase of the pathway⁴¹ (Fig. 6b). Alg1 knockdown caused partial separation of paired worms, swelling or darkening of the worm body (Fig. 6c, d). Alg11 knockdown resulted in severe swelling in intestine and significantly reduced activity (Fig. 6c, d). These findings demonstrate that key genes involved in the early stages of N/O-glycosylation of substrate proteins are essential for the survival of schistosomes, as expected for any multicellular eukaryotic organism, emphasizing the importance of protein glycosylation in the parasite.

Glycosylation profiling of glycoproteins at the host–parasite interface in S. mansoni

Given the critical role of glycosylation in mediating host–parasite interactions and its potential as immunogenic epitopes, we hypothesized that glycosylated proteins at the host interface could serve as strong candidates for immune targeting. By integrating the glycoproteomic signatures with previously reported surface, gut, esophageal gland, and parenchyma proteins of the parasites^42–49, we highlighted a set of interface-exposed glycoproteins. These proteins commonly exhibited diverse glycan types, high glycan abundance, and complex glycan structures (Fig. 7a, Supplementary Fig. 20 and Supplementary Data 19). By analyzing the site-specific glycan types of these proteins, we found that most are complex or high-mannose glycans (Supplementary Fig. 21a). Fucosylated and high-mannose glycans predominated among the 20 most frequently observed glycans (Supplementary Fig. 21b), suggesting that these glycan types are more likely to be involved in host-parasite interactions.

Fig. 7. Glycoproteins at the host interface and their glycosylation features in S. mansoni.

a Glycoproteins expressed in the parasite’s host–parasite interface. Proteins with fewer than 3 N-glycans were indicated on the right of the dashed line. b Structures of the N-glycans linked to glycosite N74 of Sm25. c Compositions of the N-glycans linked to glycosite N115 of Sm29. d Compositions of the N-glycans linked to glycosites N183 and N298 of Cathepsin B1.1 (B1.2). N-glycosites are indicated in pink.

Notably, our screening identified several well-studied vaccine candidate targets, including Sm23⁵⁰, Sm25⁵¹, Sm29⁵², Sm200⁵³, Cathepsin B⁵⁴ and Cathepsin D⁵⁵. Additionally, proteins with preliminary experimental support for antigenic protective potential, such as LAMP⁵⁶ and Cystatin³, were also included in our results, supporting the potential of the glycoprotein repertoire for vaccine target prediction. Additionally, our data provided detailed glycosite information for several previously predicted or confirmed glycoproteins. For example, Sm29 was predicted to have a high probability of N-glycosylation at N58 and N115, although these sites lacked experimental validation⁵⁷. Cathepsin B was reported to contain a predicted N-linked glycosylation site (Asn183–His–Thr) in schistosomes⁵⁸. In this study, we not only confirmed N115 of Sm29 as an authentic N-glycosylation site but also identified glycosylation of Cathepsin B at both N183 and N298.

These previously reported vaccine candidates exhibited distinct glycosylation patterns: Sm25 primarily featured mono-fucosylation (Fig. 7b), Sm29 showed minimal fucosylation and xylosylation (Fig. 7c), while Cathepsin B displayed the most complex glycosylation, including both xylosylation and multi-fucosylation (Fig. 7d). Additionally, we comprehensively analyzed site-specific N-glycans, detailing site occupancy, glycan types, specific monosaccharides, and sex-biased patterns. The results reveal biases for different glycans at the same site, identifying not only high-occupancy glycans but also detecting low-occupancy glycans (<1%) in Sm25 and Cathepsin B1.1(B1.2). Distinct glycosylation complexity was observed across sites: N183 of Cathepsin B1.1(B1.2) exhibited more glycans than N298, while N298 displayed 100% multi-fucosylation. These findings provide deeper molecular insights into their glycosylation patterns (Supplementary Fig. 22−24 and Supplementary Data 20).

Additionally, we selected representative proteins with high glycan heterogeneity for detailed analysis, including parenchyma-specific alpha−2-macroglobulin (Smp_089670) and serine protease inhibitor (Serpin, Smp_090080). Alpha−2-macroglobulin harbored more sites, while Serpin displayed more glycans per site. Both proteins were dominated by complex glycans, although the former exhibited high fucosylation rates at almost all sites, unlike the latter. Tegument-specific Sm200 (Smp_017730) and Smp_331290 were also analyzed. While Sm200 harbored more glycosites, both proteins were predominantly modified with complex glycans. Notably, several glycosites on Sm200 carried no sex-biased glycans, a feature absent in Smp_331290 (Supplementary Figs. 25, 26).

Overall, the database established in this study enables comprehensive characterization of glycosylation features for individual S. mansoni glycoproteins, addressing a key gap in the field and laying a molecular foundation for the rational design of next-generation glycosylation-based vaccines.

Discussion

Glycomics research on S. mansoni has continuously advanced over the years, significantly expanding our understanding of glycan types in this parasitic flatworm. Despite a large amount of glycomics data, a comprehensive glycoproteomic analysis is lacking for S. mansoni, and the lack of appropriate tools to decipher intact glycopeptides beyond established glycan libraries, our knowledge of S. mansoni glycoproteins, glycosites, and site-specific glycans has remained largely incomplete. Utilizing pGlyco3 and pGlycoNovo for integrated identification, this study identified 3274 site-specific N-glycans across 399 N-glycoproteins and 258 site-specific O-glycans on 89 O-glycoproteins in adult male and female S. mansoni. By integrating these glycoproteins with the single-cell RNA-sequencing atlas, our study has further provided a detailed profile of these N- and O-glycoproteins on a tissue and cell level in S. mansoni.

Previous glycomics studies have shown contradictions in the glycosylation compositions in S. mansoni¹². For instance, it was reported that core β2-xylosylation disappear completely in adult worms^13,24, and O-glycans were reported undetectable in the adult stage¹². However, Western blot analysis has indicated the presence of core xylose in adult S. mansoni⁵⁹. With the advancement of sample preparation method, mass spectrometry instrumentation, and analytical tools in our study, we have detected the presence of xyloses, fucoses and hexuronic acids (HexA) in adult worms from our newly generated glycan library that includes 244 N-glycan compositions and 98 O-glycan compositions (Supplementary Data 4). Our study identified hexuronic acids. Previous studies have shown that adult S. mansoni expresses GlcA as part of the O-glycan of the circulating anodic antigen (CAA), suggesting that S. mansoni can synthesize GlcA conjugates³¹. In parasitic helminths, the presence of hexuronic acid residues has been reported in glycomics studies of Dirofilaria immitis, Brugia malayi and Cardicola forsteri^34,60,61. We hypothesize that in S. mansoni, the formation of this glycan residue may be related to glucuronyltransferases (such as Smp_083130), although the biological functions and immune regulatory roles involved remain unknown. These discovery holds significant implications and deserves in-depth exploration.

A small proportion of sialic acid was also detected in our study. However, schistosomes lack the canonical sialic acid biosynthesis pathway, and previous studies have consistently reported an absence of sialylation⁶². Instances of sialic acids detected in helminth preparations have generally been attributed to host contamination^33–36. We therefore consider the small amount of sialic acid in our dataset to be most likely host-derived. This interpretation is further supported by the fact that schistosomes ingest host blood, and host proteins have previously been identified in worm vomitus⁴³.

The N-glycan repertoire encompasses canonical structures, including core fucosylation, core xylosylation, and terminal glycan motifs (e.g., Lewis X and LacdiNAc), along with structural variants such as tri- and tetra-fucosylated configurations. Notably, the high prevalence of truncated and uncharacterized glycan types, as well as fucosylated and xylosylated glycans, further underscoring the complexity and uniqueness of schistosomal glycosylation patterns.

Comparative analysis of the S. mansoni N-glycan library with recently reported model organisms revealed the highest similarity with Mus musculus and the lowest with C. elegans. This may be closely related to the parasitic lifestyle of S. mansoni: As an obligate parasite of mammals, schistosomes require complex interactions with host cells and the immune system, and their glycan structures may have evolved through convergent evolution to resemble those of the host, facilitating immune evasion or infection. Although S. mansoni N-glycans appear broadly similar to those of Mus musculus, they differ in several major aspects. Notably, schistosomes lack sialylated N-glycans and instead display abundant xylosylation and extensive multifucosylation—features that are largely absent from mammalian glycomes. In contrast, although C. elegans is evolutionarily closer to schistosomes, its glycan repertoire is more complex and diverse as a free-living nematode, with a high proportion of rare glycans (18.73%)²¹ (Supplementary Data 3), which may explain the low similarity in N-glycan profiles.

Previous studies on glycan localization have predominantly relied on immunofluorescence techniques, limiting detection resolution to the tissue or cell type level²⁹. Here, we integrated single-cell transcriptomic data with glycomics analysis, enabling a systematic exploration of glycosylation patterns at both tissue and sex levels in S. mansoni. Parenchyma-specific N-glycoproteins harbor abundant glycans at multiple glycosylation sites, which inherently drives the diversity of their glycosylation patterns. Muscles and neurons exhibited streamlined glycosylation patterns, whereas the tegument showed moderate glycosylation complexity. Gut-specific glycoproteins, despite having fewer glycosites, exhibited higher glycan complexity in terms of glycan size and diversity (Fig. 2a-e and Supplementary Fig. 5), suggesting more dynamic and varied glycan modifications at individual sites. Such tissue-specific glycosylation patterns likely correlate closely with their distinct physiological functions. Notably, O-glycosylation showed unique enrichment in the esophageal gland. Lee et al. found that schistosomes lacking esophageal gland die in immunocompetent hosts but survive in B cell/antibody-deficient hosts, suggesting this gland mediate immune evasion⁶³. The gland-specific protein MEG-14, highly O-glycosylated, has been proved to interact with the host protein³⁸, and may participate in lysing red blood cells and immune cells.

Our study has revealed significant differences in glycosylation patterns between mature male and female S. mansoni. Glycoproteins highly expressed in females are likely directly related to nutrient uptake and egg production, while those upregulated in males may be closely associated with motility, pairing behavior, and host interactions (Fig. 3c). The vitelline gland and Mehlis’ gland contain glycans with significant proportions of fucoses, xyloses, and hexuronic acids. These glycans may be integrated into egg antigens or eggshell during oviposition, serving as potent immune inducers that promote granuloma formation and mediate pathology. Evidence shows that core fucosylated and xylosylated N-glycans have been detected in S. mansoni eggs⁶⁴. Additionally, these glycan modifications may play functional roles in the reproductive development of female worms.

Post-translational glycosylation is a critical modulator of protein immunogenicity, with significant implications for vaccine design. For instance, glycosylation of the hepatitis C virus (HCV) E2 protein and the Ebola virus GP2 protein has been shown to substantially enhance vaccine efficacy^65,66. In parasitology, the immunoprotective effects of the native H11 molecule from Haemonchus contortus are closely linked to its N-glycosylation¹⁴. However, in schistosomiasis, research on glycosylated antigen-based vaccines remains limited. Although vaccine candidates such as Sm25, Sm29, and Cathepsin B have been extensively studied^51,52,54, their glycosylation profiles have not been investigated in the context of vaccine development. Furthermore, none of these candidates have shown promising efficacy in clinical trials. In the present study, we demonstrate that these proteins possess complex glycosylation patterns, which merit further evaluation for their potential immunoprotective roles as glycosylated antigens. Importantly, our database provides a comprehensive glycoprotein resource, particularly highlighting surface-exposed proteins such as those in the tegument that interact directly with the host immune system. We propose that the characterization and utilization of naturally occurring glycoproteins represent a promising strategy for schistosome vaccine development.

In conclusion, this study established a comprehensive glycoproteomic profile of adult S. mansoni, uncovering tissue- and sex-specific glycosylation patterns, identifying atypical glycans and HexA modifications, and highlighting the critical role of protein glycosylation in schistosomes. Additionally, glycosylation patterns of several important proteins at the host interface were characterized. These findings provide valuable insights for studies on schistosome glycosylation and provide a resource for the development of schistosome glycoprotein vaccine candidates.

Methods

Sample preparation

The mice used in this study were BALB/c strain (Mus musculus), purchased from Shanghai Jeste Experimental Animal Co. Ltd (Shanghai, China). Mice were housed under standard laboratory conditions, with a 12-h light/12-h dark cycle, an ambient temperature of 22 ± 1 °C, and relative humidity of 55 ± 10%. Six-week-old female mice were infected with 400 S. mansoni cercariae per mouse. Mature worms were harvested at 42 days post-infection by portal vein perfusion. Male and female worms were collected separately and washed three times with PBS. The worms were then resuspended in T-PER^TM Tissue Protein Extraction Reagent (Thermo Fisher Scientific) at a ratio of 20 μL reagent per mg of tissue. The resuspended mixture was transferred to a Dounce homogenizer and thoroughly homogenized on ice. The homogenized samples were centrifuged at 10,000 × g for 30 min at 4 °C, and the supernatant containing the extracted proteins was collected. The protein concentration was determined by BCA method.

Protein digestion

Protein samples requiring N-glycopeptide enrichment were reduced in 10 mM dithiothreitol (DTT) at 37 °C for 60 minutes, followed by alkylation in the dark with 20 mM iodoacetamide (IAA) at room temperature for 30 min. To precipitate proteins, a sixfold volume of acetone was added to the samples, which were then incubated at −20 °C overnight and centrifuged at 10,000 × g. The resulting pellet was dissolved in 25 mM NH₄HCO₃ and digested with trypsin at an enzyme-to-substrate ratio of 1:50 at 37 °C overnight. The digestion was terminated by adding 0.5% trifluoroacetic acid (TFA). The digested samples were centrifuged at 16,000 × g for 10 min, and the supernatant was desalted using a C18 column (Waters). The desalted peptides were dried by vacuum centrifugation and subsequently used for glycopeptide enrichment.

For samples requiring O-glycopeptide enrichment, the initial steps, including reduction, alkylation, and protein precipitation, were performed as described above. The precipitated proteins were dissolved in 25 mM NH₄HCO₃ and treated with PNGase F at an enzyme-to-protein ratio of 1:1000 to remove N-glycans, followed by incubation at 37 °C overnight. Subsequently, trypsin was added at an enzyme-to-substrate ratio of 1:50 for overnight digestion at 37 °C. O-glycopeptides were further released by adding O-glycosidase. Finally, the digested samples were desalted, dried, and prepared for glycopeptide enrichment.

Glycopeptide enrichment

Glycopeptide enrichment was performed using zwitterionic hydrophilic interaction liquid chromatography (ZIC-HILIC). Briefly, 1 mg of desalted peptides was reconstituted in 300 μL of loading buffer (80% acetonitrile, 1% trifluoroacetic acid) and loaded into the column. To maximize glycopeptide binding, the flow-through was collected and reloaded onto the column for four additional cycles. The column was then washed four times with loading buffer to remove non-specifically bound peptides. Subsequently, the specific enriched glycopeptides were eluted with 600 μL of 0.1% trifluoroacetic acid and dried by vacuum centrifugation for further analysis.

LC-MS/MS for glycopeptide analysis

The processed samples were reconstituted in 0.1% formic acid (FA) solution for LC-MS/MS analysis. Online liquid chromatography (LC) was performed using a ThermoFisher Easy-nanoLC II system. Solvent A consisted of 0.1% FA in water, and solvent B consisted of 0.1% FA in acetonitrile (ACN). For tryptic N-glycopeptide analysis, samples were directly loaded onto a C18 column (Acclaim PepMap 100, 75 μm × 50 cm, nanoViper, C18, 3 μm, 100 Å) with a flow rate of 300 nL/min. For intact N-glycopeptide analysis, a 180-min gradient was applied: solvent B increased from 2% to 40% over 165 minutes, then to 51% in 7 minu, followed by a rapid increase to 90% in 1 min, and maintained for 2 min. For intact O-glycopeptide analysis, a 120-minute gradient was used: solvent B increased from 4% to 35% over 105 min, then to 50% in 7 min, followed by a rapid increase to 100% in 1 min, and maintained for 2 min.

Mass spectrometry detection was performed using an Orbitrap Eclipse™ Tribrid™ Mass Spectrometer. The spray voltage was set to 2.3 kV, the ion transfer tube temperature was maintained at 320 °C, and the ion funnel RF levels were set to −35 V/−45 V. Data were acquired in data-dependent acquisition (DDA) mode. For intact N-glycopeptide analysis: (1) Full MS scans were performed over a range of m/z 350–2000 with a resolution of 120,000, an automatic gain control (AGC) target of 500,000, a maximum injection time of 50 ms, and charge states of +2 to +6 were selected for fragmentation. Dynamic exclusion was enabled with a duration of 20 s after one detection. (2) Each selected precursor ion was subjected to MS/MS analysis with the following parameters: isolation window of 4 m/z, Orbitrap detection with a resolution of 15,000, AGC target of 250,000, maximum injection time of 250 ms, normalized collision energy (NCE) of 30%, and stepped NCE mode enabled with a range of ±10%. For intact O-glycopeptide analysis, the full MS resolution was set to 30,000, and MS/MS fragmentation was performed using a combined HCD/EThcD mode, where HCD utilized stepped collision energies (20%, 30%, 40%) and EThcD was set to 27%. All other parameters remained consistent. Each sample group was analyzed in triplicate using LC-MS/MS.

Intact glycopeptide identification and analysis

Raw mass spectrometry data were processed using the pGlycoNovo tool²¹ within the pGlyco 3.1 software for intact glycopeptide identification. The search parameters were configured as follows: mass tolerances for precursor and fragment ions were set to ±2 ppm and ±20 ppm, respectively. The protein database was obtained from Wormbase Parasite (version WBPS18). Enzymatic digestion was specified as full-trypsin with a maximum of 3 missed cleavages. Fixed modifications included carbamidomethylation on all cysteine residues (C + 57.022 Da), while variable modifications included oxidation on methionine (M + 15.995 Da) and acetylation on protein N-termini (+42.011 Da). The maximal numbers of glycan modifications allowed were: HexNAc = 15, Hex = 20, Fuc = 4, NeuAc = 4, NeuGc = 4, HexA = 1, and Xyl = 1. The N-glycoproteomic data were also searched in pGlyco3.1¹⁹ using the ‘mouse’ and ‘plant’ databases, together with the published S. mansoni glycan datasets. The maximum glycan gap was set to 2. Glycopeptide identification was filtered at a false discovery rate (FDR) of 0.01. Detected site-specific O-glycans were compared with N-glycans, and identical O-glycans located within 20 amino acids of an N-glycosite were identified as N-glycans and removed.

Quantitative analysis of intact glycopeptides

Quantitative analysis of intact glycopeptides was performed using pGlycoQuant²⁸ (version 202407) with the raw data and identification results generated by pGlycoNovo. Default software parameters were applied for all analyses. For intact N-glycopeptide quantification, protein-level quantitative results were obtained from the “protein.list” (Supplementary Data 7), while site-specific glycan-level quantitative results were extracted from the “site.list” (Supplementary Data 8). To eliminate the influence of protein abundance, site-specific glycan quantification was normalized by calculating the ratio of the quantitative value of site-specific glycans to the corresponding protein quantitative value for each protein.

For intact O-glycopeptide quantification, which is more complex, protein-level quantitative results were obtained from the “protein.list” (Supplementary Data 13), and site-specific glycan-level quantitative results were extracted from the “modification.list” (Supplementary Data 14). The quantitative values of identical site-specific glycans at the same protein were summed, followed by normalization using the same approach as for N-glycopeptides.

The quantitative data of glycoproteins and processed site-specific glycans were log₂-transformed and filtered in Perseus. Only glycopeptides with valid values in at least 3 out of 6 replicates were retained. Missing values were imputed based on the total observed protein intensities, using a width of 0.3 σ and a downshift of 1.8 σ.

Bioinformatics analysis

For N-glycoproteins, functional enrichment analysis was performed using the DAVID Functional Annotation Tool³⁷. The analysis included the following categories: (1) UP_KW_MOLECULAR_FUNCTION, (2) SMART protein domains, (3) KEGG_PATHWAY, and (4) Gene Ontology (GO) terms encompassing biological processes (BP), molecular functions (MF), and cellular components (CC). N-glycans were classified into five types based on their composition, following the criteria established by Zeng et al.²¹. Any composition that did not match the 1,461 non-redundant N-glycan compositions in the glycan database constructed by Zeng et al.²¹ was designated as “unclassified” in this study.

The single-cell data of S. mansoni was obtained from Wendt et al.²². The Seurat data format was converted to ‘h5ad’ format using the sceasy v0.0.7 package in R (v4.3.2), followed by analysis and visualization in Python (v3.10) using scanpy v1.7.2.

The glycosyltransferases of S. mansoni and their homologous proteins in mouse and human underwent amino acid multiple sequence alignment using ClustalW in MEGA 11 software, followed by visualization in EScript 3.0 (https://espript.ibcp.fr/ESPript/ESPript).

RNA interference

Primers targeting Ogt (Smp_046930), C1galt1 (Smp_051930), Alg1 (Smp_005010), and Alg11 (Smp_052330) were designed as listed in Supplementary Data 21, with the T7 promoter sequence (TAATACGACTCACTATAGGGAG) added to the 5’ end of each primer. PCR primer synthesis was performed by Beijing Tsingke Biotech Co., Ltd. (Tianjin, China). Double-stranded RNA (dsRNA) was synthesized using the T7 High Yield RNA Synthesis Kit (YEASEN, China). The dsRNA was purified by precipitation at room temperature for 10 min with ethanol and 7.5 M ammonium acetate, followed by centrifugation at 10,000 × g for 5 min, brief air-drying, and resuspension in RNase-free water. The purified dsRNA was annealed by successive incubations at 95 °C, 75 °C, and 55 °C (3 min each), followed by cooling at room temperature. For RNAi treatments, ~5 pairs of adult parasites were cultured in 3 mL of Basch 169 medium in a 12-well plate. The parasites were treated with dsRNA (50 μg/mL) at days 0, 2, 4, 6, 10, 14, and 18. Phenotypic observations were conducted on day 30 post-treatment. After that, worms were collected and washed in PBS before stored at −80 °C. Light microscopy imaging of the interfered worms was performed using an Olympus SZX16 stereomicroscope.

RNA Extraction and cDNA synthesis

Schistosome samples were homogenized in 300 μL TRIzol reagent, followed by the addition of 700 μL TRIzol and incubation for 10 min on ice. Subsequently, 200 μL chloroform was added, and the mixture was incubated for 5 min before centrifugation at 15,871 × g for 15 min at 4 °C. The aqueous phase was collected, mixed with an equal volume of isopropanol, and incubated for 15 min to precipitate RNA. The RNA pellet was then obtained by centrifugation under the same conditions, washed once with 75% ethanol, and dissolved in RNase-free water. Following the manufacturer’s protocol, cDNA was synthesized using the EVO M-MLV RT Kit with gDNA Clean for qPCR (ACCURATE BIOTECHNOLOGY (HUNAN) CO., LTD, ChangSha, China).

Quantitative real-time PCR

The reactions were carried out on a LightCycler® 96 instrument (Roche, Switzerland) with 2× SYBR Green qRT-PCR Master Mix (YEASEN, China), following the manufacturer’s protocol. Each 20 μL reaction mixture included 2 μL of cDNA (diluted 1:4), 10 μL of 2× SYBR Green Master Mix, 0.8 μL (5 μM) of each primer, and 6.4 μL of ddH2O. The thermal cycling protocol consisted of an initial denaturation step at 95 °C for 3 min, followed by 40 cycles of 95 °C for 15 s and 60 °C for 30 s. To verify amplification specificity, a melt curve analysis was performed, with temperatures ranging from 60 °C to 95 °C. The gene psmb4 (Proteasome subunit beta 4) was utilized as an internal control. All qPCR primers are detailed in Supplementary Data 21. Primers were synthesized by Beijing Tsingke Biotech Co., Ltd.

Quantification and statistical analysis

Statistical analysis was conducted using the R environment (v4.3.2). The limma package (v3.56.2) from Bioconductor was employed to identify sex-specific differentially expressed glycoproteins and site-specific glycans using a linear model and moderated t-statistics. The p-values were adjusted for multiple testing using the Benjamini–Hochberg (BH) procedure. Glycoproteins and glycopeptide combinations with an adjusted p-value < 0.05 and an absolute log₂-fold change ≥1 were considered statistically significant.

GraphPad Prism 8 software processed and presented the data as the mean with SD (Standard Deviation). Statistical significance was calculated by unpaired two-tailed parametric t test. P-values < 0.05 are considered significant (ns, not significant; *, p < 0.05, **, p < 0.01, ***, p < 0.001, ****, p < 0.0001). Error bars represent SD.

Ethics

All experiments related to animals were conducted in accordance with the guidelines for the Care and Use of Laboratory Animals of the Ministry of Science and Technology of the People’s Republic of China (2006398) and were approved by the Animal Care and Use Committee of Fudan University (Fudan IACUC 2021JS0078) to ensure ethical and responsible treatment of the animals.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Author contributions

J.P.W. and X.C. conducted this project. J.P.W. and W.H. secured funding and supervised the overall research. X.C. performed the bioinformatics analysis and all experiments. Y.M.Y. provided experimental assistance in sample collection, RNAi and qPCR. W.X.L. provided assistance in dsRNA synthesis. S.L., C.Y., M.J.G., W.J.C. G.W.C., Y.P.W. and E.L.T. assisted in sample collection and preparation. L.M.W. and H.J.L. contributed to the MS experiments and assisted with the glycoproteomic analysis. X.C. and J.P.W. drafted the manuscript. All authors critically reviewed and approved the final manuscript.

Peer review

Peer review information

Nature Communications thanks Richard Cummings, and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. A peer review file is available.

Data availability

The mass spectrometry raw data, as well as the pGlycoNovo, pGlyco3 and pGlycoQuant output files generated in this study have been deposited in the ProteomeXchange Consortium via MassIVE, under accession code MSV000097695. (https://massive.ucsd.edu/ProteoSAFe/dataset.jsp?task=f3968fe064f94bb1ba462c0ddc5888dc).

Competing interests

The authors declare no competing interests.

Supplementary information

The online version contains supplementary material available at 10.1038/s41467-026-68400-9.