Materials
Ester-terminated poly(lactic-co-glycolic)acid (PLGA 85:15, viscosity 0.55–0.75) was obtained from Durect corp (Cupertino, CA). Dichloromethane (DCM), acetonitrile (MeCN), dimethyl sulfoxide (DMSO), poly(vinyl alcohol) (PVA, MW 31,000–50,000, 87–89% hydrolyzed), dithiothreitol, trypsin-chymotrypsin inhibitor and phenylmethylsulfonyl fluoride (PMSF), as well as all other reagents, unless noted otherwise, were purchased from Millipore Sigma (St. Louis, MO). 1,1'-dioctadecyl-3,3,3',3'-tetramethylindodicarbocyanine, 4-chlorobenzenesulfonate salt (DiD) and CellMask™ Orange Plasma Membrane Stain were purchased from Thermo Fisher Scientific (Waltham, MA). The bath sonication system was composed of Fisherbrand Model 505 Sonic Dismembrator, a cup horn (Qsonica LLC, Newtown, CT), and a recirculating chiller (Qsonica LLC) (Fig. S1A). The cup horn is equipped with a tube holder that can hold eight 1.5 ml tubes (Fig. S1B).
The following primary antibodies were obtained commercially and used at the indicated dilutions: 1:1,000 rabbit anti-COX IV and 1:1,000 rabbit anti-GAPDH conjugated with horseradish peroxidase (Cell Signaling Technology, Danvers, MA); 1:1,000 rabbit anti-CD45 (Thermo Fisher Scientific); 1:1,000 rabbit anti-CD11a, 1:1,000 rabbit anti-nucleoporin p62, and 1:10,000 rabbit anti-sodium potassium ATPase (Abcam, Cambridge, MA). Horseradish peroxidase-conjugated AffiniPure goat anti-rabbit IgG (H + L) secondary antibody (1:10,000) was purchased from Jackson ImmunoResearch Laboratories (West Grove, PA).
Cell Culture
Jurkat T lymphocyte cell line was purchased from American Type Tissue Collection (Manassas, VA) and maintained in suspension in RPMI 1640 (ATCC) supplemented with 10% fetal bovine serum (FBS, Thermo Fisher Scientific), 100 U/ml penicillin-streptomycin, and 50 ug/ml amphotericin B at a density of 104 to 105 cells/ml.
Isolation of Subcellular Membrane Fractions
Plasma membranes were extracted from Jurkat cells. Briefly, 2.8x108 cells were collected by centrifugation at 700 g for 5 minutes and resuspended in homogenization buffer containing sucrose (250 mM), Tris hydrochloride (10 mM), magnesium chloride (1 mM), potassium chloride (1 mM), PMSF (2 mM), trypsin-chymotrypsin inhibitor (200 µg/mL), deoxyribonuclease (10 µg/mL), and ribonuclease (10 µg/mL) for 30 minutes. The cell suspension was gently homogenized using a Dounce homogenizer and laid on top of a sucrose density gradient (30%, 40%, 55% w/v). Subcellular fractions were separated by ultracentrifugation with a Beckman SW-28 rotor at 28,000g for 45 minutes at 4°C. Three lipid-rich fractions at the sucrose interfaces were collected, diluted 1:1 with 150 mM saline, and pelleted by centrifugation at 28,000g for 60 minutes. The pellets were lyophilized and stored at -80°C until use.
Western Blot
Lyophilized fractions were rehydrated with diH2O supplemented with protease inhibitor cocktail (Roche, Basel, Switzerland), and protein concentration was assessed using the Pierce Rapid Gold BCA Protein Assay Kit (Thermo Fisher Scientific) according to the manufacturer’s protocol. Samples of membrane lysates (25 µg protein) were prepared with 4x Laemmli sample buffer (Bio-Rad, Hercules CA) and dithiothreitol (100 mM), heated to 95°C for 5 min, and resolved on a Mini-PROTEAN® TGX™ Precast Gel (Bio-Rad) by sodium dodecyl sulfate-polyacrylamide gel electrophoresis. Proteins were transferred onto nitrocellulose membranes, blocked with 5% non-fat dry milk in tris-buffered saline containing 0.1% Tween-20 (30 min., 37°C), probed with the indicated primary and corresponding secondary antibodies, and detected using SuperSignal West Pico PLUS Chemiluminescent Substrate (Thermo Fisher Scientific).
Nanoparticle Synthesis
Nanoparticles (NPs) were prepared by oil-in-water single emulsion.49 Poly (lactic-co-glycolic) acid (PLGA) (100 mg, 85:15, inherent viscosity 0.55–0.75, Durect Corp.) was dissolved in 1 ml organic solvent (DCM and MeCN, 1:1 (v/v)) and added dropwise to 6 ml of 6% PVA in diH2O, w/v) under vigorous vortexing. The mixture was sonicated using an ultrasonic processor with a micro-tip (Cole-Parmer, Vernon Hills, IL) for 10 rounds of 30 seconds on and 30 seconds off while in an ice bath. After sonication, the emulsion was added to a beaker containing 100 ml diH2O and stirred overnight at 300 rpm to allow the organic solvent to evaporate. The NPs were collected by centrifugation at 40,000 g for 30 minutes and washed three times with diH2O. After the final wash, NPs were resuspended in diH2O containing 50 mg sucrose, frozen at -80°C, lyophilized overnight, and stored at -20°C in a desiccator. For fluorescent labeling, DiD was added to the organic phase (0.3% w/w with PLGA).
Nanoparticle Characterization
Size, polydispersity index (PDI), and zeta potential of NPs were assessed with dynamic light scattering using a Zetasizer Nano ZS90 (Malvern Panalytical, Westborough, MA). A suspension of lyophilized NPs (0.1 mg/ml) was prepared in 0.2 µm-filtered diH2O and transferred to a disposable polystyrene cuvette or capillary cell. For size measurement, the dispersant was set to diH2O (η = 0.8872 cP, n = 1.33) at 25°C. NPs were equilibrated for 3 min. in the cuvette and measured at a 90° angle. Each measurement was an average of four separate, consecutive measurements performed by the Zetasizer that met 12 indicators for result quality criteria. The same dispersant specifications were utilized for zeta potential measurement as a function of electrophoretic mobility, plus a dielectric constant of 78.5. The zeta measurements were based on F(ka) model using Henry’s function. NPs were equilibrated for 1 minute in the cuvette and measured as described above.
Membrane Coating of PLGA NP Cores
Prior to use as a coating material, reconstituted membrane vesicles were processed by a series of centrifugation and sonication steps. Lyophilized membrane fractions were reconstituted in 1 ml of chilled diH2O, vortexed, and centrifuged (16,000g, 30 min., 4°C). The pellet was resuspended in diH2O as above and sonicated at 100% amplitude (5 min., 4°C). A recirculating chiller controlled the temperature of the bath (Fig. S1). The sample was subject to another round of centrifugation and sonication as above to remove residual salt and reduce vesicle size. The protein concentration of the final vesicle suspension was measured using Pierce Rapid Gold BCA Protein Assay Kit. Size and PDI were assessed with dynamic light scattering as described above on a 0.1 mg/ml by protein weight suspension of membrane vesicles in diH2O.
Processed plasma membrane vesicles that met size and PDI criteria were coated onto prepared PLGA NP cores by sonication. The general protocol for coating was to combine a predetermined amount of membrane vesicles (protein weight) and PLGA NP cores (dry weight) in a volume of diH2O in a 1.5 ml polypropylene tube and sonicate the tube at a predetermined amplitude (%), temperature (°C), and duration (min.). In some experiments, size and PDI of the suspension were measured immediately after coating. In other experiments, the suspension was diluted to a final concentration of 0.1 mg/ml NPs in 1x PBS and incubated at room temperature under constant agitation in the form of 360° inversion on a HulaMixer (Thermo Fisher Scientific) at 8 rpm to induce aggregation prior to measuring size and PDI.
Experimental Workflow and Design of Experiments
Initial experiments investigated the effects of water bath temperature (5°C-35°C) on the size and PDI of each of the two components, NPs and membrane vesicles, sonicated alone in suspension (100% amplitude, 5 min.). Next, we validated the quantitative outcomes of our experimental design- increases in size and PDI- as indicators of instability and incomplete coating. Per the general protocol for coating membrane vesicles onto NP cores described above, NPs (0.5 mg/ml) were mixed with membrane vesicles (0.05 mg/ml) in diH2O and sonicated at 60% amplitude, 15°C, for 5 minutes. After sonication, samples were diluted to a final NP concentration of 0.1 mg/ml in 1x PBS and incubated at room temperature under constant agitation. Parallel control samples were either not sonicated or not agitated.
We then used the general protocol and sonicated different membrane to NP weight ratios (ranging from NP only to 1.0) of the two components in co-suspension at a fixed amplitude (20%), temperature (15°C), and duration (5 min.). Using the optimal ratio identified in these experiments, we assessed size and PDI of co-suspensions that were sonicated at different volumes (50–400 µl) and NP densities (0.5-3 mg/ml).
Finally, to identify effects of and interactions between the settings on the sonicator, we performed statistical design of experiments (DOE) with Design-Expert software (Stat-Ease, Version 13, Minneapolis, MN) with a 5-level surface central composite design algorithm. Three parameters (variables) were included in the statistical analysis (Table 1): amplitude (%, A), time (min., B), and temperature (°C, C). These variables were incorporated into the development of a Central Composite design- a Response Surface Method that evaluates interactions and quadratic effects by analyzing the variables at five levels. The 32 run combinations used to test the model are listed in Supplementary Table S1. PLGA NPs (1 mg/ml final concentration) were combined with membrane vesicles (0.1 mg/ml final protein concentration) in a volume of 150 µl in 1.5 mL polypropylene tubes. Co-suspensions were sonicated according to the conditions outlined in Supplementary Table S1. After sonication, samples were diluted in PBS and incubated as described above before quantitative assessment of size (nm) and PDI. The evaluated responses (raw values in Supplementary Table S1) were analyzed by ANOVA, allowing us to assess the effect of each variable and their interactions.
Imaging
We used transmission electron microscopy (TEM) and confocal microscopy to visualize morphology and colocalization, respectively, of membrane vesicle and NP components after sonicating different ratios of the two as described above. For TEM imaging, carbon-coated 200 mesh copper grids (Electron Microscopy Sciences, Hatfield, PA) pretreated with 0.002% Alcian blue in 0.03% acetic acid were floated on top of 30 µl drops of NP suspensions (30 min, room temperature). After washing with diH2O, the samples were negatively stained by floating the grid on a 50 µl drop of filtered 2% uranyl acetate (pH = 7.0, 5 min, room temperature). Samples were dried for 12h in a grid storage box before imaging with a Philips CM120 transmission electron microscope (TSS Microscopy, Beaverton, OR).
For confocal microscopy, NPs loaded with lipophilic tracer DiD (0.3% w/w) were suspended with varying amounts of membrane vesicles labeled with CellMask™ orange plasma membrane stain and sonicated as described above. Samples were mixed with an aqueous mounting medium and mounted on glass slides with 12 mm glass coverslips (#1.5H thickness, ThorLabs, Newton, NJ). Slides were imaged using 561 nm wavelength (membrane vesicles, green) and 633 nm wavelength (NP, red) using Leica HCS A confocal microscope.
Statistics
One-way or two-way repeated-measures ANOVA, followed with Tukey’s or Dunnett’s post hoc analysis, were used for comparison of data groups, where appropriate and as indicated in figure legends. In all cases, p < 0.05 was considered statistically significant.