Nanoparticles were prepared on the same day as treatment. They were composed exclusively of soy-phosphatidylcholine (Soy L-α-PC) with no additional lipids or surface modifications. The L-α-PC stock solution was prepared by dissolving 600 mg of Soy L-α-PC in 40 ml of 100% ethanol (15 µg/µl). For each 1 ml preparation, 200 µl of Soy L-α-PC stock (15 µg/µl; 3 mg total) was combined with 1 ml of INSPARIN protein solution (0.5 µg/µl; 500 µg total; Proteintech, Ag19996; GST-tagged), corresponding to a lipid : protein ratio of ~6:1 (w/w). Following ethanol evaporation of Soy L-α-PC using a nitrogen evaporator for 8–10 min, either 1 ml of INSPARIN or 1 ml of deionized water was added. The mixture was extruded using an Avanti MiniExtruder sequentially through 200 nm and then 50 nm polycarbonate filters (11 passes each), concentrated by centrifugal filtration (100 kDa cutoff; centrifuged for 25–30 min at 10,000 × g), and resuspended in 1× PBS (for in vitro studies) or saline (for in vivo studies). The encapsulation efficiency was measured using high-performance liquid chromatography (HPLC) with a C4 column (ACCUCORE 150-C4, 150 × 2.1 mm, 2.6 µm). Encapsulation efficiency was determined once at the beginning of the study and those values were subsequently used to estimate encapsulation efficiency for all preparations. Particle size and Polydispersity Index (PDI) were determined each time using BI-MAS particle size analyzer (Brookhaven Instruments, New Hampshire, USA). The encapsulation efficiency of INSPARIN NP was 74.1%, and nanoparticles typically measured 93–113 nm with a PDI of 0.06–0.23. For 1x dose, 74.1 µg of INSPARIN was injected into each mouse.

3T3-L1 preadipocytes (ATCC, #CCL-173) were purchased from ATCC; vendor authenticated and free of mycoplasma at purchase. No additional testing performed in our lab. Cells were maintained in high glucose (25 mM) Dulbecco’s Modified Eagle Medium (DMEM) (ThermoFisher, #11995065), with 10% Bovine calf serum (Hyclone, #SH30072.03), and 1% antibiotic- antimycotic agent Pen-Strep (Sigma Aldrich, #A5955). C2C12 and HepG2 cells were kept in maintenance media containing high glucose (25 mM) Dulbecco’s Modified Eagle Medium (DMEM) (ThermoFisher, #11995065), 10% Bovine calf serum (Hyclone, #SH30072.03,), and 1% Pen-Strept (Sigma Aldrich, #A5955). At around 70–80% confluency, cells from each well of a six-well plate were treated with 30–50 µl of either void NP or INSPARIN NP (11.11-18.52 µg of INSPARIN) and incubated for 72 h at 37 °C. Following 72 h incubation, the cells were used to perform a glucose uptake assay as described below.

NP-treated cells in 6-well plates were washed twice with PBS before adding 900 µL KRP (136 mM NaCl, 4.7 mM KCl, 10 mM NaPO4, 0.9 mM CaCl2, 0.9 mM MgSO4, 0.9 mM CaCl2) with or without 100 nM insulin (Sigma Aldrich, #15500) for 20 min. One well was treated with KRP plus 100 nM cytochalasin (Sigma Aldrich, #C6762) for 20 min to subtract nonspecific glucose uptake. Fifty µL of 10x isotope solution was then added to each well for a final concentration of 100 nM cold-deoxyglucose and 0.5 µCi/mL [3H]-2- Deoxyglucose (PerkinElmer, #NEC720A250UC) for 5 min. Immediately after incubation, cells were washed with ice-cold PBS and lysed with 1 mL of 0.05% SDS at 37 °C for 30–45 min. To measure glucose uptake, 950 µL of cell lysate was added to a scintillation vial containing 3 mL of scintillation liquid and read on a PerkinElmer scintillation counter (Model No. Tri-Carb 4810TR, now part of Revvity Inc.), while the remaining 50 µL used for protein estimation via BCA assay. Radioactivity Counts Per Minute (CPM) were normalized to protein content for each well.

Unless otherwise noted, all animals were housed in the same room ≤5 mice/cage on a 12-h light/12-h dark cycle and were placed on ad-libitum HFD (45% Kcal from fat; Lab Diets) and water. The experimental procedures were approved by Institutional Animal Care and Use Committee (IACUC) at Texas Tech University, Lubbock, TX. Mice were weighed according to the study design using a digital scale (OHAUS Model CL 201). The body composition of mice was measured using the EchoMRI machine (Echo MRI LLC, Houston, TX).

For the 24 h post INSPARIN NP injection experiment, 7–9 weeks old male C57BL/6J mice (n = 15) (Jackson Laboratory, Bar Harbor, ME) were fed a 45% HFD for 5 weeks and received a single subcutaneous (SC) dose of void or INSPARIN NP (1×;74.1 µg). An OGTT was performed 24 h post-injection. Next, we repeated the same experiment by including higher doses (2×;148.2 µg, or 4×;296.4 µg) and determined glucose clearance at 24 and 48 h post-injection. The doses of 2X and 4X were administered by proportionally increasing the injection volume of the standard INSPARIN nanoparticle formulation.

For long-term daily injections, 14–16 weeks old C57BL/6J mice (n = 10) fed a 45% HFD for 8 weeks were divided into a void and a treatment INSPARIN group, matched for body weights and HbA1c. The groups received daily SC injections of void-NP or 2x of INSPARIN NP (148.2 µg), for 7-weeks (6 days/week; 43 doses). OGTT, corresponding insulin, and HbA1c were determined. 2- or 6- months after the last injection, whole-body histopathology (IDEXX BioAnalytics, Missouri, USA) was performed to determine tissue specific adverse effects if any, due to chronic NP treatment.

For long-term alternate day injections, C57BL/6J mice fed a 45% HFD for 20-weeks were divided into void NP or treatment INSPARIN groups, matched for body weight and HbA1c. The groups received injections of void NP or INSPARIN NP (2×;148.2 µg), respectively, on alternate days for 8-wks, following which, glucose clearance, and HbA1c were determined.

Investigators were not blinded to group allocation during experiments or outcome assessment. Sample sizes were determined based on prior published studies using similar mouse models and expected effect sizes; no formal statistical method was used to predetermine sample size. No animals or data points were excluded from the analysis. In all the experiments, the void group received control nanoparticles composed exclusively of soy-phosphatidylcholine (soy L-α-PC) prepared identically to INSPARIN-loaded nanoparticles, except that deionized water was substituted for the protein.

The OGTT was performed using an oral gavage of glucose (2.5 g/kg body weight, except for one experiment where 2 g/kg body weight was used) following a 4-h fast. During the OGTT, blood glucose levels were measured via tail bleed at baseline (0 min) and at 15, 30, 60, and 120 min after the glucose load. Blood glucose level (mg/dL) during OGTT was measured using the AlphaTrak2 glucose meter. Blood (20 μL) was also collected at these time points in EDTA-coated microvettes (Sarstedt, #16.444.100) for insulin measurements.

Blood sera collected during OGTT were centrifuged at 10,000 rpm for 7 mins at 4 °C and stored at −80 °C. Insulin (ng/mL) was measured using the ELISA kit (EMD Millipore, EZRMI-13 K) or Mercodia mouse insulin ELISA kit (Mercodia, 10-1247-01). The manufacturer’s instructions were followed throughout the experiment. For standard curve preparation, 4-parameter logistic curve fit was used using GraphPad Prism software (version 8).

DiD-labeled void NP were prepared by adding DiD’ solid; DiIC18 [5] solid (1,1’- Dioctadecyl-3,3,3’,3’-Tetramethylindodicarbocyanine, 4-Chlorobenzenesulfonate Salt) (Thermo Fisher, # D7757) in the amount of 2 mol% of Soy L-α-PC and following the same preparation method explained above. Eight mice were injected with saline (control group) or DiD-labeled NP for this experiment. For the SC injection group, 2 mice received the injection on the right inguinal fat tissue, 2 mice on the left inguinal fat tissue, and 1 mouse received an SC injection of saline. For the intravenous (IV) injection group, 2 mice were injected with DiD labeled NP, and 1 mouse received saline via the tail vein. Images were obtained 48- and 72 h post-injection. Mice were anesthetized in the isoflurane chamber before obtaining the images with an IVIS® Lumina XR imaging system.

HbA1c was measured by using A1CNow+, Hba1c Blood Monitor w/Sampler kit by following manufacture’s recommendation (PTS Diagnostics, #3021). Though this kit is for human use, we have tested it on mice. The HbA1c values in mice strongly correlate with their fasting blood glucose. For our experiments involving HbA1c measurement in mice, we have measured the HbA1c at baseline, midpoint, and post-treatment.

Appropriate statistical tests were used, including Student’s t tests for two-group comparisons and one-way ANOVA with Bonferroni correction for multiple group comparisons. Data are expressed as mean ± SEM. For two-group comparisons, Welch’s correction was applied where appropriate, accounting for unequal variances. Normality was not formally tested, but data distributions appeared consistent with assumptions. A p value of <0.05 was considered statistically significant.