**Background**
Seminal plasma contains a heterogeneous population of extracellular vesicles (sEVs) that are released by cells of the testis, epididymis, and accessory sex glands. These vesicles are involved in male and female reproductive processes, including sperm function, capacitation, and modulation of the female immune environment. However, the specific roles of different sEV subsets remain poorly understood, partly due to the lack of standardized isolation methods. This study aimed to isolate and characterize large (L-EVs) and small (S-EVs) sEV subsets from pig seminal plasma, decode their proteomic profiles, and infer their potential biological functions.
**Methods**
Ejaculates from nine mature Pietrain boars were pooled into three samples. Seminal plasma was obtained by double centrifugation at 1500×g. After further centrifugation at 20,000×g, pellets (enriched in large EVs) and supernatants (enriched in small EVs) were separately processed. Both fractions were subjected to size exclusion chromatography (SEC) using Sepharose-CL2B columns. Fractions 7–9 (EV-enriched) and 18–20 (non-EV-enriched) were collected. The resulting samples (L-EVs, S-EVs, and non-EV-enriched) were characterized by total protein concentration, nanoparticle tracking analysis (NTA), dynamic light scattering (DLS), transmission electron microscopy (TEM), and flow cytometry for EV markers (CD63, HSP90β) and purity (albumin). Proteomic analysis was performed using liquid chromatography–tandem mass spectrometry (LC-MS/MS) for spectral library generation and sequential window acquisition of all theoretical mass spectra (SWATH-MS) for quantification. Differentially abundant proteins were identified using a fold change >2 or <−2 with p-value <0.05. Gene Ontology (GO) enrichment and pathway analyses were conducted.
**Key Results**
Total protein concentration was significantly higher in S-EVs (216 ± 61 μg/ml) than in L-EVs (24.2 ± 15.3 μg/ml) (p < 0.001). NTA showed particle concentration was higher in S-EVs (13.6 × 10^11 ± 3.65 × 10^11 particles/ml) than in L-EVs (3.7 × 10^11 ± 9.06 × 10^10 particles/ml) (p < 0.01). S-EVs were smaller (mode 151.47 ± 19.87 nm) than L-EVs (196.87 ± 21.60 nm) (p < 0.05). DLS confirmed S-EVs had an average diameter of ~100–125 nm and L-EVs ~230–290 nm. TEM showed S-EVs were spherical (30–100 nm) and L-EVs were heterogeneous (100–350 nm). Flow cytometry revealed both subsets expressed CD63 (S-EVs: 77.62 ± 6.89%; L-EVs: 78.51 ± 6.57%) and HSP90β (S-EVs: 90.74 ± 4.02%; L-EVs: 78.06 ± 14.82%), with low albumin contamination (S-EVs: 2.84 ± 1.38%; L-EVs: 3.28 ± 1.07%).
Proteomic analysis identified 1034 proteins, of which 737 were quantified by SWATH. Principal component analysis explained 69.3% of variance, clearly separating L-EVs, S-EVs, and non-EV-enriched samples. Differential expression analysis revealed 197 differentially abundant proteins between L-EVs and S-EVs (168 more abundant in L-EVs, 29 more abundant in S-EVs). Compared to non-EV-enriched samples, L-EVs had 140 more abundant and 59 less abundant proteins; S-EVs had only 2 more abundant and 35 less abundant proteins. Only two proteins (TMBIM1 and BASP1) were more abundant in both EV subsets versus non-EV-enriched samples.
GO analysis showed L-EV-enriched proteins were involved in metabolic processes, detoxification, and localization, including small GTPases (RAB10, RAB11B, RAB25, RALB, KRAS, RAP2C, RAP1B, RALA, RAC1, RAN), ESCRT-III-associated VPS4B, and SNARE proteins VAMP3 and STX3. Proteins related to sperm capacitation (SLC26A3, ELSPBP1, PEBP1) and antioxidant enzymes (GPX4, SOD1, CYBRD1, LDHA, LDHB, GSTp1, ATOX1, PRDX1, PRDX2, PRDX5, ALDH9A1) were also enriched in L-EVs. In contrast, S-EV-enriched proteins were linked to lysosome function and immune processes, including cathepsins (CTSA, CTSC, CTSH, CTSF), GBA, PRCP, LGMN, FUCA1, FUCA2, LIPA, and B4GALT1. Pathway analysis linked L-EV proteins to glycolysis, gap junction, and metabolism, while S-EV proteins were associated with lysosome, glycan degradation, and glycosaminoglycan biosynthesis.
**Clinical Implications**
This study provides a reliable method to isolate pure subsets of seminal EVs and demonstrates that L-EVs and S-EVs have distinct proteomic profiles, suggesting different cellular origins and biological functions. L-EVs appear to be secreted via conventional multivesicular body fusion or plasma membrane budding and are likely involved in sperm capacitation and protection against oxidative stress. S-EVs may be released via an apocrine pathway and play a role in modulating the immune environment of the female reproductive tract, potentially influencing sperm survival and embryo development. These findings advance the understanding of seminal EV biology and may inform future therapeutic strategies for male infertility or assisted reproductive technologies.