Materials and reagents

The phospholipids, 1,2-dipalmitoyl-sn-glycero-3-phosphocholine (DPPC), 1,2-distearoyl-sn-glycero-3-phosphoethanolamine-N-[methoxy(polyethylene glycol)-2000] (DSPE-mPEG2000), and 1,2-distearoyl-sn-glycero-3-phosphoethanolamine-N-[succinyl(polyethylene glycol)-2000] (DSPE-PEG(2000)-succinyl) were purchased from Avanti Polar Lipids, Inc. (Alabaster, AL). Cholesterol (Chol), N-(3-Dimethylaminopropyl)-N′-ethylcarbodiimide hydrochloride (EDC), and N-Hydroxysuccinimide (NHS) were purchased from Sigma-Aldrich (St. Louis, MO, USA). Calcein violet and lipophilic tracers (DiO and DiD) were purchased from Invitrogen (Eugene, OR). BTZ and Y27632 were purchased from MedKoo Biosciences (Morrisville, NC). Recombinant PSGL-1 protein was purchased from Novoprotein (Summit, NJ). Monoclonal antibodies (mAb) used for immunoblots were purchased from Cell Signaling Technology (Danvers, MA). Phospho-Akt (pAKT; #4060), phospho-Erk1/2 (pERK; #4370), phospho-Rb (pRB; #9308), p21 (#2947), cleaved Caspase3 (cCasp3; #9664), cleaved Caspase 9 (cCasp9; #7237), cleaved PARP (cPARP; #5625), phospho-FAK (pFAK; #3284), phospho-SRC (pSRC; #6943), and phospho-S6 ribosomal protein (pS6R; #4858) were all used at a dilution of 1:1000. α-Tubulin (#2125) was used as a loading control at a dilution of 1:3000. The immunoblots were detected using an ECL Plus chemiluminescent system (PerkinElmer, Waltham, MA). The mAbs used for flow cytometry were purchased from Miltenyi Biotec (Bergisch Gladbach, Germany) unless otherwise noted.

Cell culture

The MM cell lines, MM.1S and H929, were purchased from the American Type Culture Collection (ATCC, Rockville, MD), OPM-2 and green fluorescent protein-labeled and luciferase-transfected MM.1S (MM.1S-GFP-Luc) were a kind gift from Dr. Irene Ghobrial (Dana-Farber Cancer Institute, Harvard Medical School, Boston, MA). Human umbilical vein endothelial cells (HUVECs) were purchased from Angio-Proteomie (Boston, MA). Human samples for this study were collected under informed consent, in concordance with Washington University Institutional Review Board (IRB) approval (IRB protocol number 201102270). MM cell lines were cultured in RPMI-1640 media (Corning, Tewksbury, MA) supplemented with 10% fetal bovine serum (FBS; Gibco, Life Technologies, Grand Island, NY), 2 mmol/L L-glutamine, 100 μg/mL penicillin, and 100 μg/mL streptomycin (Corning CellGro). HUVECs were cultured in Endothelial Growth Medium (EGM, Angio-Proteomie, Boston, MA) supplemented with endothelial growth supplements (including 10% FBS, recombinant growth factors, and 1% penicillin and streptomycin). All cells were cultured at 37 °C and in 5% CO2 in a NuAire water jacket incubator (Plymouth, MN).

Expression of P-selectin in MM-associated ECs in human samples

BMMNCs from healthy (n = 6) and MM patients (n = 6) were washed with PBS and stained with anti-human CD45 (REA747), CD31 (REA730), VEGFR-2 (REA1046), and CD62P (REA389) mAbs or their respective isotype controls for 1 h. Cells were washed and analyzed by flow cytometry. First, we gated CD45-negative cells to exclude the leukocytes. Then from that population, we gated the ECs as CD31+/VEGFR-2+ double-positive cells. ECs were analyzed then for the expression of P-selectin (CD62P) as the ratio of the mean fluorescence intensity (MFI) of CD62P divided by the isotype control (RMFI). Then the RMFI values of P-selectin expression on ECs in MM patients were divided by the values acquired from the healthy donor samples27. Flow cytometry files were analyzed using MACSQuantify and FlowJo.

Expression of P-selectin in ECs in vivo

All animal work presented in this study was approved by the Institutional Animal Care and Use Committee at Washington University School of Medicine in St. Louis. The mice used were NCG male 50–56-day-old mice (Charles River, Wilmington, MA) unless otherwise stated. MM.1S cells (2 × 106 cells/mouse) were injected intravenously into five mice and tumor progression was confirmed using bioluminescent imaging (BLI) at 4 weeks post-injection. Five healthy mice were used as well to assess P-selectin expression. Mice were sacrificed and femurs were crushed and washed with PBS for the collection of BMMNCs. The cells were stained with CD45 (REA737), CD31 (390), VEGFR-2 (REA1116), and CD62P (REA344) murine mAbs or their respective isotype controls for 1 h. Cells were washed and analyzed by flow cytometry. ECs were gated as CD31 and VEGFR-2 positive cells. Then the RMFI values of P-selectin expression on ECs in MM-bearing mice were divided by the values acquired from the healthy mice.

For immunofluorescence, MM.1 S cells (2 × 106 cells/mouse) were injected intravenously into three mice and tumor progression was confirmed using BLI at 4 weeks post-injection. Three healthy mice were used as well to assess VEGFR-2 and P-selectin expression. Mice were sacrificed; femurs were processed and embedded in paraffin28. Briefly, the processed slides were heated at 55 °C overnight to remove the paraffin and incubated in xylene twice for 5 min. To rehydrate the tissue, the slides were put into 100% ethanol twice for 3 min. Then 95% ethanol twice for 3 min and 80% ethanol twice for 3 min. The slides were then washed in PBS for 5 min and soaked in double-distilled water for 30 min. Antigen retrieval process was started by inserting 800 μL of unmasking solution (Vector Labs, Burlingame, CA) into 80 mL of double-distilled water. The slides were put into this solution, put into a pressure cooker, and microwaved on high for 8 min and medium for 3 min. The slides were set to cool in an unmasking solution for 20 min and washed in PBS for 5 min. The slides were then blocked with 120 mL of PBS supplemented with 0.3% Triton-x, 5% goat serum, and 5% FBS for 2 h at room temperature. Following blocking, we washed with PBS for 5 min and incubated the slides in the primary antibodies (VEGFR-2 (REA1116), and CD62P (RB40.34; BD Biosciences, San Jose, CA)) diluted in PBS supplemented with 3% bovine serum albumin. Each antibody concentration was ten micrograms per milliliter. The slides were incubated with the antibodies overnight in 4 °C. PBS was used to wash the slides three times for 5 min. A coverslip was then added with mounting medium for imaging. For imaging, we used an Olympus BX51 fluorescent microscope at Ex/em 365/420 and 620/670 nm.

Tube-like formation of ECs on 2D Matrigel and 3DTEBM

On the first day, DiO-labeled MM cells (3 × 104 cells/well) were cultured in regular 96-well plates (2D) or in the 3DTEBM15. To make the 3DTEBM, 40 µL of BM plasma was added to 60 µL RPMI-1640 complete medium with a final concentration of 1 and 4 mg/mL calcium chloride and tranexamic acid, respectively. On the next day, Matrigel was added on top of the 3DTEBM, or to the regular 96-well, as a control (2D). HUVECs were stained with DiD and plated on top of the Matrigel for all well conditions (2D and 3DTEBM; 3 × 104 cells/well), in RPMI-1640 media without serum. The samples were imaged via ZEN 2009 using a fluorescent microscope 4 h post-addition of the HUVECs. Tube length was measured using ImageJ.

Expression of P-selectin in ECs cultured with cell lines in vitro

The expression of P-selectin in vitro was evaluated in HUVECs using two models. (i) In the 2D tissue culture model, HUVECs (1 × 104 cells pre-labeled with DiO) and MSP-1 stromal cells (1 × 104 cells) were co-cultured with or without MM cell lines (MM.1S, H929, or OPM-2; 3 × 104 cells) per well in a 96-well plate. (ii) In the 3DTEBM model, MSP-1 stromal cells (1 × 104 cells) with or without MM cell lines (MM.1S, H929, or OPM-2; 3 × 104 cells) were suspended in BM plasma and set to solidify into a 3DTEBM matrix. After 2 h, Matrigel (Corning, Tewksbury, MA) was added on top of the scaffold, and HUVECs (1 × 104 cells pre-labeled with DiO) were added on top of the Matrigel with non-supplemented EGM media. HUVECs and stromal cells with and without MM cells were cultured in 2D and 3DTEBM models for 24 h. Then, the cultures were digested with collagenase, and cells were retrieved for flow cytometry analysis. The experiment was independently repeated three times with five replicates each.

To determine that the increase in P-selectin expression on ECs was specifically associated with the presence of MM, we co-cultured ECs with MM cells or normal peripheral blood mononuclear cells (PBMCs; 3 × 104 cells/well) extracted from healthy donors, in the 3DTEBM following the procedure described above. For the effect of Y27632 on the expression of P-selectin, HUVECs were plated on the 3DTEBM containing MM and treated with free Y27632 (25 μM) for 24 and 48 h and analyzed for P-selectin expression via flow cytometry.

Confocal imaging of 3DTEBM cultures of HUVECs

MSP-1 stromal cells (1 × 104 cells pre-labeled with DiD) and MM.1S cells (3 × 104 cells pre-labeled with DiO) were suspended in BM plasma to form a 3D scaffold in a Nunc Lab-Tek II Chamber Slide System (Thermo Fisher, Waltham, MA). After 2 h, Matrigel was added on top of the scaffold and HUVECs (1 × 104 cells pre-labeled with calcein violet) were subsequently added on top of the Matrigel. The HUVECs, stromal cells, and MM cells were cultured in the 3DTEBM for 24 h. Samples were then imaged using a FV1000 confocal microscope with an XLUMPLFLN 20XW/1.0 immersion objective lens (Olympus, Central Valley, PA) (excitation/emission: 405/450 ± 20 (calcein violet), 488/ 520 ± 20 (DiO), and 633/650 + (DiD) nm).

Preparation and characterization of liposomes

Liposomes were prepared using the thin layer evaporation method29. Briefly, lipids (DPPC, Chol, DSPE-mPEG2000, and DSPE-PEG(2000)-succinyl at a molar ratio of 6:3:0.5:0.5) were dissolved in a chloroform/methanol mixture (3:1, v/v) and the solvent was then evaporated through a rotary evaporator (Heidolph, Schwabach, Germany) to form a thin lipid film. The film was then hydrated with PBS and extruded with an extruder set (Avanti Polar Lipids). Fluorescent liposomes were prepared by dissolving DiD in the organic solvent with the lipids (before film formation).

Conjugation of PSGL-1 to the surface of liposomes was performed by using carbodiimide chemistry. Briefly, the liposomes were suspended in a solution of 0.25 M EDC and 0.25 M NHS (in water) and incubated for 10 min at room temperature. Then, PSGL-1 was added to the mixture and the colloidal suspension was incubated at 4 °C overnight in a light-protected environment with gentle-stirring. Unbound protein was removed using Amicon Ultra Centrifugal Filter Units (100 kDa MWCO). The mean sizes, polydispersity index (PDI) and zeta-potential (ZP) were analyzed by dynamic light scattering (DLS) analysis using a Zetasizer Nano ZS (Malvern, Malvern, UK).

Cryogenic transmission electron microscopy imaging of liposomes

Samples were prepared on Quantifoil holey carbon grids (R2/2 300 mesh copper) and plunge frozen using a Vitrobot Mark IV (Thermo Fisher) which was set to 4 °C and 100% humidity. Three microliters of sample was applied to the Quantifoil grids, blotted to remove excess fluid, and plunge frozen into liquid ethane. Grids were then stored in liquid nitrogen until imaged. Vitrified grids were imaged using a JEM-1400 TEM (JEOL, Peabody, MA) operating at 120 kV, equipped with a 626 single tilt liquid nitrogen cryo-transfer holder (Gatan, Pleasanton, CA) and a XR111 CCD camera (Advanced Microscopy Techniques, Woburn, MA). The sample holder was kept at −175 °C during imaging to prevent devitrification. All images were acquired using a nominal magnification of ×25,000 corresponding to a pixel size of 7.38 Å/pixel.

Measurement of the affinity of PSGL-1-targeted liposomes for P-selectin by surface plasmon resonance

The affinity of PSGL-1-targeted liposomes for P-selectin was measured by the biosensor-based surface plasmon resonance (SPR) technique using an automatic apparatus BIAcore T200 (GE Healthcare, Chicago, IL). Recombinant P-selectin protein was immobilized using carbodiimide chemistry on the CM4 sensor chip surface (ligand), and PSGL-1-targeted and non-targeted liposomes were used as the analyte.

Binding of PSGL-1-targeted liposomes to ECs in vitro

HUVECs (pre-labeled with DiO) were grown on top of the 3DTEBM as described above. DiD-labeled non-targeted or PSGL-1-targeted liposomes were cultured with the HUVECs for 2 h. The 3D cultures were then digested, washed, and analyzed via flow cytometry. The experiment was independently repeated three times with five replicates each.

Binding of PSGL-1-targeted liposomes to ECs in vivo

MM.1 S cells (2 × 106 cells/mouse) were injected intravenously into ten mice and tumor progression was confirmed using BLI at 4 weeks post-injection. Mice were then injected intravenously with DiD-labeled non-targeted or PSGL-1-targeted liposomes (2 mg/mL of lipids; 5 mice per group). Mice were sacrificed and femurs were flushed with PBS and crushed. The cells were stained with CD45 (REA737), CD31 (390), and VEGFR-2 (REA1116) murine mAbs or their respective isotype controls for 1 h. Cells were washed and analyzed by flow cytometry. ECs were gated as CD31 and VEGFR-2 positive cells. Then the MFI values of the PSGL-1-targeted liposomes bound to ECs in mice were divided by the values obtained from the non-targeted samples.

Fluorescent imaging of tumor in vivo and organs ex vivo

Four NSG mice (shaven) were injected intravenously with non-targeted and PSGL-1-targeted liposomes that were stained with DiO and DiD, respectively. These mice were imaged at t = 0.1, 1, 4, and 24 h using the In-Vivo MS FX PRO fluorescent imaging system (Bruker, Billerica, MA) at Ex/em of 460/535 and 620/670 nm. Following 24 h, the tissues were extracted from each mouse and imaged in the In-Vivo MS FX PRO fluorescent Bruker imaging system.

Histological assessments of tissue

Following ex vivo imaging, tissues were frozen in OCT (Tissue Tek, Torrance, CA), sliced using a Cryocut 1800 (Leica, Wetzlar, Germany), and imaged using an Olympus BX51 fluorescent microscope at Ex/em 460/535 and 620/670 nm. No staining was used due to the fluorescence of the liposomes.

High-performance liquid chromatography for detection of BTZ and Y27632

BTZ and Y27632 were analyzed using high-performance liquid chromatography (HPLC, Agilent 1100 series, Santa Clara, CA) with a reverse phase C-18 column (Agilent Zorbax Eclipse XDB-C18, 4.6 mm × 150 mm).

For detection of BTZ, a 50% acetonitrile solution in water containing 0.1% trifluoroacetic acid (TFA) was used as the mobile phase at a flow rate of 1 mL/min, as reported previously14. A calibration curve was obtained by plotting the area under the curve (AUC) of the BTZ HPLC peak (at retention time = 2.2 min, λ = 270 nm) for a concentration range of 0–200 μg/mL.

For detection of Y27632, a gradient of acetonitrile/water containing 0.1% TFA was used as the mobile phase at a flow rate of 1 mL/min30. The percent of acetonitrile in the mobile phase was 0% (at 0–3 min), then decreased gradually to 33% water (3–3.5 min), and decreased gradually back to 0% (3.5–7 min). A calibration curve was obtained by plotting the AUC of the Y27632 HPLC peak (at retention time = 4 min, λ = 260 nm) for a concentration range of 0–200 μg/mL. All HPLC data were analyzed via OpenLab.

Drug release and evaluation of drug entrapment efficiency

Each drug was encapsulated in a separate liposome formulation followed by centrifugation (130,000 rcf at 4 °C for 1 h) to remove any free drug and re-suspended in fresh PBS. BTZ liposomes were put into dialysis bags (3.5 kD; Spectrum Labs, Rancho Dominguez, CA); the supernatant was taken for each time point (t = 1, 3, 6, 12, 24, 48) and analyzed using the HPLC methods mentioned above.

To evaluate loading efficiency, liposomes were centrifuged at 130,000 rcf at 4 °C for 1 h using a Beckman Optima™ XPN ultracentrifuge equipped with a SW 50.1 fixed angle rotor (Beckman Coulter Inc., Fullerton, CA, USA). The amount of BTZ and Y27632 in the supernatant was evaluated by HPLC. The entrapment efficiency (EE) was calculated according to the following equation:

$${\mathrm{EE}} = D_{\mathrm{T}} – D_{\mathrm{U}}/D_{\mathrm{T}} \times 100$$

(1)

where DT is the total amount of drug added to the formulation during the preparation, and DU is the amount of unincorporated drug found in the supernatant.

Effect of free and liposomal Y27632 on trans-endothelial migration of MM cells in vitro

Trans-endothelial migration5 was performed by incubating HUVECs (5 × 103 cells) overnight in the upper chamber of a Boyden chamber (Corning) prior to adhesion assay. MM.1S cells were pre-treated with (or without) free Y27632 (25 μM) or liposomal Y27632 (25 μM equivalent) for 3 h. Cells were then placed in the upper migration chamber in the presence or absence of 30 nM SDF-1 in the lower chamber. After 3 h of incubation, cells that migrated to the lower chambers were counted by flow cytometry.

Effect of free and liposomal Y27632 on mobilization of MM cells to the circulation in vivo

MM.1S-GFP-Luc cells (2 × 106 cells/mouse) were injected intravenously into nine mice, and tumor progression was confirmed using BLI at 4 weeks post-injection. Mice were treated with intravenous injections of (i) free Y27632 (2.5 mg/kg, n = 3); (ii) Y27632-loaded non-targeted liposomes (2.5 mg/kg equivalent, n = 3); and (iii) Y27632-loaded PSGL-1-targeted liposomes (2.5 mg/kg, n = 3). Blood was collected from the tail vein of each mouse at 0 (before), 2, 4, and 24 h after injection. Blood samples were lysed with 1X red blood cell lysis buffer (BioLegend, San Diego, CA) using the manufacturer’s instructions and PBMCs were analyzed via flow cytometry.

Effect of free and liposomal drugs on cell signaling in MM cells and ECs

HUVECs were treated with vehicle (control), free Y27632 (25 μM), empty liposomes, and liposomal Y27632 (25 μM) overnight and then cultured with MM.1S for 6 h in serum-free media. The MM cells were then separated from the HUVECs by gently tapping the culture plate multiple times and washing the HUVECs with PBS to separate the floating MM cells from the adherent HUVECs. Proteins were then extracted from each cell type and subjected to immunoblotting for pSRC, pFAK, and α-Tubulin (60, 125, and 52 kDa, respectively).

MM.1 S cells were cultured and treated with vehicle (control), free BTZ (5 nM), empty liposomes, and liposomal BTZ (5 nM) for 24 h. Proteins were then extracted and subjected to immunoblotting for p21, pRB, cPARP, cCasp3, cCasp9, pAKT, pS6R, pERK, and α-Tubulin (21, 110, 89, 17, 37, 60, 32, 140, and 52 kDa, respectively).

For immunoblotting, we cut the membranes according to the target protein size before antibody staining. Cells were lysed with 1X lysis buffer (Cell Signaling, #9803). The protein concentration was determined by Bradford assay (Bio-Rad, Hercules, CA), and 50 μg of protein was loaded per lane. Proteins were separated by electrophoresis using NuPAGE 4–12% Bis-Tris gels (Novex, Life Technologies, Grand Island, NY) and transferred to a nitrocellulose membrane using an iBlot system (Invitrogen). Membranes were blocked with 5% non-fat milk in Tris-buffered saline/Tween-20 (TBST) buffer and incubated with primary antibodies overnight at 4 °C. The membranes were then washed with TBST for 30 min, incubated for 1 h at room temperature with horseradish peroxidase (HRP)-conjugated secondary antibody, washed, and developed using a Novex ECL Plus Chemiluminescent Kit (Thermo Fisher). Blots were imaged on a ChemiDoc XRS imaging system via Bio-Rad Image Lab (Bio-Rad).

Effect of free and liposomal BTZ on MM and EC viability in vitro

DiD-labeled HUVECs and DiO-labeled MM.1S cells were co-cultured overnight and treated with vehicle (control), free BTZ (0–50 nM), empty liposomes, or liposomal Y27632 (0–50 nM equivalent) for 24 h, and the survival of MM and HUVECs was analyzed via flow cytometry. MM cells were gated as DiO+ cells and HUVECs were gated as DiD+ cells, and each cell population was counted and normalized against counting beads (Invitrogen). Survival was calculated as the percent of vehicle-treated controls.

Efficacy of BTZ and Y27632-loaded PSGL-1-targeted liposomes on MM tumor progression in vivo

MM.1S-GFP-Luc cells (2 × 106 cells/mouse) were injected intravenously into 84 mice, and tumor progression was confirmed using BLI at 3 weeks post-injection (average photon flux of 2.5E07 photons per second). Mice were randomized into 12 groups of 7 mice each, which received weekly intravenous injections of (i) saline, (ii) Y27632 as a free drug (2.5 mg/kg), (iii) BTZ as a free drug (1 mg/kg), (iv) a combination of free BTZ and free Y27632, (v) empty non-targeted liposomes, (vi) non-targeted liposomal Y27632 (2.5 mg/kg equivalent), (vii) non-targeted liposomal BTZ (1 mg/kg), (viii) non-targeted liposomal combination of BTZ and Y27632 in the same liposome (2.5 mg/kg and 1 mg/kg, respectively), (ix) empty PSGL-1-targeted liposomes, (x) PSGL-1-targeted liposomal Y27632 (2.5 mg/kg), (xi) PSGL-1-targeted liposomal BTZ (1 mg/kg), and (xii) PSGL-1-targeted liposomal combination of BTZ and Y27632 in the same liposome (2.5 mg/kg and 1 mg/kg, respectively). Tumor progression was assessed weekly by BLI. Weight was recorded twice per week, and the survival and general health of the mice were recorded daily.

Statistical analyses

All data from in vitro and in vivo experiments were expressed as means ± standard deviation. Statistical significance was analyzed using a Student’s t-test or two-way analysis of variance (ANOVA). Log-rank test was used to compare the Kaplan–Meier curves. P values <0.05 were used to indicate statistically significant differences. All graphs were plotted on Microsoft Excel and GraphPad Prism.

Reporting summary

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

Source