Male Wistar rats, aged four weeks, were purchased from Kenya Institute of Primate Research (KIPRE). The animals were group-housed in spacious cages to allow sufficient space for exercise, social interaction, and natural behaviors, such as climbing and burrowing. Cage floors were covered with wood chips to provide comfort, hygiene, and opportunities for foraging and digging. The animal room was maintained at a temperature range of 20–24°C with a relative humidity of 40–60% under a 12-hour light/dark cycle. The animals were acclimatized for 14 days prior to the experimental procedures to reduce stress and physiological variability. Routine husbandry included regular cage cleaning to maintain hygiene without causing undue disturbance and providing unlimited food and water. The humane endpoints were adapted from Silva-Reis and colleagues (2023). ²⁵ They included body weight, hair/tail appearance and grooming, skin, walking, mental status, and response to stimuli. These were assigned a score of 0-3 where a total score of 4 or 3 for some parameters was an indicator of euthanisia.

In this study, blood samples were collected from three of five rats per group based on pre-established inclusion criteria designed to minimize animal stress and blood volume loss. The rats were randomly chosen within the group after diabetes induction, prior to the experiment, to adequately represent the group. The decision to limit sampling to three rats was made a priori to comply with animal welfare guidelines restricting total blood volume removal per animal and to reduce potential confounding effects from repeated blood draws, such as stress-induced hyperglycemia. No animals were excluded from the study, and all data from the sampled rats were included in the analysis.

The sample size of five rats per group was decided based on several factors, including the use of the resource equation approach, ²⁶ including an attrition rate of 20–50%, ^{27, 28} previous similar studies in diabetic rat models, ²⁹ resource availability, and ethical guidelines.

Diabetes was induced in animals by administering streptozotocin (STZ), obtained from Sigma-Aldrich Chemical Co. (St. Louis, MO, USA), following a 6-8 hour fasting period. STZ was freshly dissolved in 50 mM citrate buffer (pH 4.5) immediately before use and injected intraperitoneally into Wistar rats at a single dose of 40 mg/kg ³⁰ ( Figure 2). To prevent hypoglycemic shock after STZ administration, 10% glucose solution was provided for 24 h to the treated rats. ³¹ Rats with blood glucose levels > 5.2 mmol/L were selected for this study.

S. mansoni glycoproteins (IPSE and Omega-1) were searched and retrieved from the NCBI ( https://www.ncbi.nlm.nih.gov/) and UniProt ( https://www.uniprot.org/) databases using the names of proteins and GenBank accession numbers in the literature. The bioinformatic tool BepiPred 3.0 ( https://services.healthtech.dtu.dk/services/BepiPred-3.0/ ) was used to predict the linear B-cell epitopes on S. mansoni IPSE and Omega-1 protein sequences. The software leverages numerical embeddings from the protein language model ESM-2 to enhance the prediction accuracy for both linear and conformational B-cell epitopes. ³² while antigenicity was assessed using Vaxijen ( https://www.ddg-pharmfac.net/vaxijen/VaxiJen/VaxiJen.html). This was performed to determine the interaction sites of the proteins that are involved in triggering the Th2 response. Expasy’s ProtParam suite was used to determine the theoretical pI and molecular weight of the predicted epitopes from IPSE and Omega-1. ³³

Our goal was to synthesize an active segment solely responsible for directly inducing the Th2 response. Therefore, after predicting the epitopes, we assessed their antigenicity and analyzed the predicted derivatives. Molecular docking was performed using PyRx ( https://pyrx.sourceforge.io/) to predict how these sequences would interact with the target molecule for each protein (IPSE with IgE and Omega-1 with Mannose receptor). The 3D structures of the sequences were predicted using the RPBS portal ( https://mobyle.rpbs.univ-paris-diderot.fr/cgi-bin/portal.py#jobs::overview), and protein structure preparation for molecular docking was performed using ChimeraX ( https://www.cgl.ucsf.edu/chimerax/).

The final peptide sequences were based on the binding affinity and energy from molecular docking. The sequences were then sent to GenScript (Piscataway, NJ, USA) for synthesis.

Stratified randomization was used for allocation based on the blood glucose levels of the animals after diabetes induction. Each peptide (SEA, Omega-1 ₃, IPSE ₂ and IPSE ₃) was assigned to a group of five diabetic rats at two different doses (50 and 100 μg/rat). The peptides were administered intraperitoneally daily for five days, ^{34– 36} and the control groups, insulin and PBS, were each assigned to groups of five diabetic rats. After a treatment period of five days, the rats were monitored for an additional five days. Blood samples were collected at baseline, after diabetes induction, post-treatment, and post-monitoring ( Figure 1). Following this, the rats were euthanized by introducing 100% CO2 at a rate of 30%-70% of the chamber volume per minute, at a flow rate of 1.5 L/min for 2-3 minutes, and their organs were harvested for histological examination.

Rats were provided unlimited access to food and water, and body weight measurements were performed at four time points. Behavioral changes were also monitored. ³⁷

Blood glucose levels were measured weekly using a glucometer (On Call Plus) with blood collected from the tail bleed. Blood samples collected at different time points were processed to obtain serum and used for immunological examinations. The two main cytokines of interest in this study were IFN-γ and IL-4 (representing Th1 and Th2 cytokine profiles). ELISA kits (U-CyTech Biosciences) were used to analyze the cytokine profiles of the experimental and control groups.

The assay used a sandwich ELISA format, in which plates were coated with cytokine-specific capture antibodies, blocked with bovine serum albumin (BSA), and incubated with serum samples and recombinant standards. Biotinylated detection antibodies were then added, followed by streptavidin-horseradish peroxidase (HRP) polymer conjugates and 3,3’,5,5’-tetramethylbenzidine (TMB) substrate to provide a colorimetric signal proportional to the cytokine concentration. The reaction was terminated with a sto p solution, and the absorbance was measured at 450 nm.

After sacrificing all rats by introducing 100% CO2 at a rate of 30%-70% of the chamber volume per minute, at a flow rate of 1.5 L/min for 2-3 minutes, the pancreas was harvested and stored in 10% formalin for 14 days, and examination was performed according to previously described techniques. ^{16, 37} Following fixation, the tissues were dehydrated using graded alcohol, cleared in xylene, infiltrated with paraffin wax, and embedded in paraffin blocks for thin sectioning with a microtome. Following sectioning, the tissues were mounted on slides and stained with hematoxylin and eosin (H&E) for microscopic analysis. Sections of the islets were examined, and the degree of cellular infiltration was evaluated as no infiltration (0), mild infiltration (1), moderate infiltration (2), and severe infiltration. ³⁸

GraphPad Prism (Version 8.02) was used to conduct all analyses. Data are shown as means ± SEM. Repeated measure ANOVA was used to analyze blood glucose, weight and cytokine concentration of the groups across the different time points followed by a Tukey’s comparison test. p- values < 0.05 were considered significant and calculated by GraphPad Prism.

An In-silico approach encompassing epitope prediction, antigenicity analysis, and molecular docking was used to investigate the potential immunogenicity and target binding properties of IPSE and Omega-1 to synthesize peptides from the complete sequence of glycoproteins. Protein sequences of IPSE were obtained using Protein Data Bank accession number 4AKA ³⁹ or Uniprot using Q869D4 and Omega-1 from the Protein database accession number ABB73003.1 (NCBI). ²⁴ These sequences were subjected to epitope prediction, and the antigenicity of the predicted epitopes was analyzed. One of the IPSE’s epitopes was found in the literature with derivatives ³⁹ which we incorporated in this study ( Table 1). The antigenic epitopes were then docked. Thus, the final peptides that were synthesized and used in this study were IPSE ₂, IPSE ₃ and Omega-1 ₃ ( Table 2), where IPSE ₂ is a derivative of IPSE ₃ (the predicted epitope found in the literature).

The BepiPred 3.0 scores for IPSE ( Figure 1A) and Omega-1 ( Figure 1B) were above the threshold of 0.15, indicating a high probability that the predicted IPSE and Omega-1 proteins are likely to be part of an epitope. Multiple predicted antigenic epitopes for IPSE (IPSE ₃ epitope, IPSE ₃ literature, and IPSE ₂ derivative) and Omega-1 (Omega-1 ₃ epitope) ( Table 1) were observed with a score range of 0.6520–1.3383. The predicted molecular weight was between 1261.40 to 2034.30 daltons while theoretical pI ranged between 8.23 to 9.70 ( Table 1). To determine their ability to bind to target molecules, IPSE derivatives (IPSE ₂ and IPSE ₃) and Omega-1 ₃ epitopes were docked. The predicted epitopes exhibited strong binding to the target molecules (IgE for IPSE derivatives and Mannose receptor for Omega-1) with IPSE ₃ and IPSE ₂ having binding energies of -11.7 kal/mol and -11.9 kal/mol respectively while Omega-1 ₃ had a binding energy of -7.9 kal/mol ( Table 2). These findings imply that specific regions within IPSE and Omega-1 possess inherent immunogenic characteristics and validated specific molecular interactions, suggesting their potential role in modulating immune response.

Omega-1 ₃ indicates the epitope predicted from the complete Omega-1 sequence, IPSE ₃ epitope indicates the epitope predicted from the complete IPSE sequence, IPSE ₃ literature indicates the part of the complete sequence of IPSE found in literature as the interaction site for IgE binding ³⁹ and IPSE ₂ derivative indicates the sequence derived from the interaction site (IPSE ₃) identified in the literature. ³⁹

Blood glucose levels in treatment and control groups

To assess the efficacy of SEA, Omega-1, and IPSE variants in reducing hyperglycemia, diabetes was induced in rats using streptozotocin (STZ), and blood glucose levels were monitored after five days of treatment ( Figure 1). STZ induction increased blood glucose levels from baseline (4.0-4.6 mmol/L) to diabetes induction (5.0-26.0 mmol/L) ( Figure 3A-J). Following treatment, SEA (50 μg, 100 μg) and Omega-1 ₃ (50 μg, 100 μg) significantly (p < 0.05) reduced blood glucose levels compared to the PBS control ( Figure 3A-D,J). IPSE ₃ at 100μg showed a significant blood glucose reduction ( Figure 3H) while IPSE ₃ at 50 μg showed no reduction ( Figure 3G). IPSE ₂ at 50 μg and 100 μg showed a non-significant reduction in blood glucose levels ( Figure 3 E-F). Insulin, as a positive control, also significantly reduced blood glucose ( Figure 3I). These findings imply that SEA, Omega-1, and IPSE _2-3 (at higher doses) possess therapeutic potential for managing hyperglycemia in this STZ-induced diabetes model.

Mean weight

The weight of each group showed a progressive increase after diabetes induction and exponential weight gain post- treatment ( Figure 4).

Cytokine levels were assessed following treatment to examine the effects of SEA, Omega-1, and IPSE variants on Th1 (IFN-γ) and Th2 (IL-4) cytokines, major cytokines involved in Th1 and Th2 immune responses, and balance in the STZ-induced diabetes model. Induction of diabetes resulted in a significant (p < 0.05) increase in IFN-γ levels. Treatment with 50 μg and 100 μg SEA significantly suppressed IFN-γ elevation ( Figure 5A-B). Conversely, Omega-1 ₃ ( Figure 5C-D) and IPSE _2-3 ( Figure 5E-F) treatments increased IL-4 levels, but this was insufficient to suppress IFN-γ levels. Insulin treatment also increased IL-4 and reduced IFN-γ levels without causing a major cytokine shift ( Figure 5I), whereas the untreated control showed continuously elevated IFN-γ levels ( Figure 5J). These findings imply that treatments with SEA, and to some extent IPSE/Omega-1, modulate the Th1/Th2 cytokine balance in the diabetic state, with SEA notably suppressing diabetes-induced Th1 elevation.

To assess the relative change (%) in mean blood glucose, IFN-γ, and IL-4 levels from diabetes induction to a day post-treatment ( Figure 6), we used the following formula: Relative change ( % ) = ( T 2 value − T 1 value ) ÷ T 1 value ) × 100 where T2 is time point 2 (1 d post-treatment) and T1 is time point 1(diabetes induction). We observed that compared to the 27.7% decrease in blood glucose levels reported with insulin treatment, SEA at both doses displayed superior glucose-lowering effects, attaining 4.7% and 24.1% greater reductions, respectively, while the higher dose of Omega-1 ₃ also showed a greater decrease of 20.1% ( Figure 6a). IPSE ₂ at 50 μg and 100 μg lowered blood glucose by 22.2% and 21.3%, respectively ( Figure 6a), while the lower dose of Omega-1 ₃ reduced blood glucose by 26.6% ( Figure 6a). In contrast, IPSE ₃ at 100 μg provided only a modest 6.8% reduction ( Figure 6a), whilst the 50 μg dose caused a significant 98.5% increase in blood glucose levels (p < 0.05) ( Figure 6a). The PBS control group exhibited an increase of 110.5% (p < 0.01) ( Figure 6a), confirming disease progression. At the cytokine level, insulin reduced pro-inflammatory IFN-γ by 48.5% ( Figure 6b.1), whereas SEA at 50 μg caused a 12.6% reduction ( Figure 6b.1). The PBS group showed a notable 236.8% increase in IFN-γ ( Figure 6b.1), with other treatments resulting in a less than 85% increase ( Figure 6b.1). Notably, SEA caused the largest anti-inflammatory IL-4 increase of 364% and 276.8% ( Figure 6b.2), which was more than double the levels at diabetes induction. Insulin, Omega-1 ₃, and IPSE ₂ treatments increased IL-4 by over 50% ( Figure 6b.2), while IPSE ₃ at 50 μg increased IL-4 by more than 50% ( Figure 6b.2), but at 100 μg made by more than 50% ( Figure 6b.2). These findings imply that SEA and its peptides, particularly Omega-1 ₃ and IPSE ₂, effectively lower blood glucose levels while modulating immune responses by reducing pro-inflammatory IFN-γ and enhancing protective IL-4, suggesting a dual metabolic and immunomodulatory mechanism that may offer advantages over insulin alone. Conversely, IPSE ₃’s dose-dependent and sometimes paradoxical effects highlight the need for careful optimization.

Histological examination showing leukocytic infiltration in harvested pancreas is a hallmark finding in T1D.

To assess islet pancreatic infiltration in the treated groups and the controls, the animals were euthanized, the pancreas was harvested, and histological examinations were performed after staining. Mild to moderate infiltration was observed in the treated groups compared to that in the negative control group ( Figure 7). Further observations were as follows: in a non-diabetic rat, no pancreatic islet infiltration ( Figure 7 I). In contrast, in the insulin group ( Figure 7 II), similar images were observed in STZ-diabetes-induced rats treated with insulin. Mild/reduced mononuclear leukocyte infiltration into pancreatic islets. The untreated group showed severe inflammatory cell infiltration (mononucleated leukocytes and macrophages) into the pancreatic islets and disruption of islet architecture with increased cellularity ( Figure 7 III). For Omega-1 _3, peri-islet macrophage infiltration and moderate islet infiltration were observed ( Figure 7 IV-V). IPSE ₂ showed moderate mononucleated leukocyte infiltration into the pancreatic islets and mild degeneration and mononucleated leukocyte infiltration into the pancreatic islets ( Figure 7 VI-VII) . IPSE ₃ showed mild pancreatic infiltration (with mononucleated and multilobed leukocytes) within pancreatic islets ( Figure 7 VIII). IPSE ₃ showed vacuolation of peri-islet cells and severe infiltration ( Figure 7 IX). SEA showed mild pancreatic infiltration and vacuolation of peri-islet cells ( Figure 7 X-XI). These findings imply that treatment with SEA, IPSE, and Omega-1 slows down the pathogenesis of the disease.

Pancreatic infiltration in the treated groups and control group was graded according to how much infiltration was observed in the islet cells, and these were assigned as mild, moderate, and severe to indicate the extent of the infiltration in an increasing order and scores 0-3 ( Table 3). The treated groups were mild/moderately infiltrated, while the control group (PBS) showed severe infiltration, as previously stated ( Table 3, Figure 7).