Materials
CAT was purchased from Beyotime Biotechnology (Shanghai, China). 1,2-Dioleoyl-sn-glycero-3-phosphoethanolamine-conjugated diphenyl-poly(ethylene glycol)2k monomethyl ether-cyclooctyne (DSPE-PEG2k-DBCO) was purchased from Ruixi Biological Technology Co., LTD. (Xi’an, China). α-Helical polypeptide bearing guanidine groups on the side chains (PG, polymerization degree = 102) was synthesized according to our previous report [54]. Lysotracker Red and Trizol reagent were purchased from Thermo Fisher Scientific (Massachusetts, USA). BCA protein assay kit, 3-(4,5-dimethyl-2-thiazolyl)-2,5-diphenyl tetrazolium bromide (MTT), and 4’,6-diamidino-2-phenylindole (DAPI) were purchased from Alfa Aesar (Shanghai, China). PrimerScript real-time reagent kit and SYBR Premix Ex Taq kit were purchased from Takara Bio (Qingdao, China). Azide-modified sSDF-1α peptide (N3-KSKPVVLSYR, N3-sSDF-1α) and Cy3-labeled N3-sSDF-1α were purchased from Bankpeptide Biotechnology Co., LTD. (Hefei, China). Primary antibodies (anti-Col I, anti-ALP, anti-Runx2, anti-OCN, and anti-OPG) were purchased from Solarbio Life Sciences Co., LTD. (Beijing, China) or Beyotime Biotechnology (Shanghai, China). Secondary antibodies of horseradish peroxidase-conjugated goat anti-mouse IgG and goat anti-rabbit IgG were purchased from Absin Bioscience Co., LTD. (Shanghai, China). APC-labeled anti-CD105, PE-labeled anti-CD73, and Percy/Cy5.5-labeled anti-CD45 were purchased from Elabscience Biotechnology Co., LTD (Wuhan, China). siCkip-1 and negative control siRNA with scrambled sequence (siScr) were purchased from GenePharma Co., LTD. (Shanghai, China), and their sequences were listed in Supplementary: Table S1. All primers were purchased from Sangong Biotech (Shanghai, China), and their sequences were listed in Supplementary: Table S2. All the other reagents were purchased from Energy Chemical (Shanghai, China) or Aladdin (Shanghai, China) and used as received.
Animals and cells
Female C57BL/6 mice (6–8 weeks, 18–20 g) were purchased from Slaccas Experimental Animal Co., LTD. (Shanghai, China) and housed in a specific pathogen-free (SPF) animal lab.
RAW 264.7 cells (mouse monocyte macrophage) were purchased from the American Type Culture Collection (Rockville, MD, USA) and cultured in DMEM containing 10% fetal bovine serum (FBS). Mouse MSCs (bone marrow mesenchymal stem cells) were purchased from Cyagen Biosciences Co., LTD. (Guangzhou, China) and cultured in α-MEM containing 10% FBS.
Mouse femur fracture model
The femur fracture model was established according to a previous report [43]. Particularly, mice were anesthetized, and their right hips, thighs, and knees were sterilized with povidone-iodine solution. A 2-cm medial parapatellar incision was created and the patella was dislocated to expose the femoral condyles. A hole was drilled into the femoral intramedullary canal at the intercondylar notch using a 25-gauge needle to stabilize the impending fracture. Immediately after the needle implantation, blunt dissection of muscle was performed to expose the midshaft of the femur, and a transverse femoral shaft fracture was then created in the right femur of each mouse using a rotary Dremel with a diamond blade attachment. The patella was then reduced, and the incision was closed using 4 − 0 synthetic suture. Mice were allowed to move freely after recovery from anesthesia. On day 5 post fracture, animals were imaged with an X-ray imaging system (Faxitron MX-20, Tucson, AZ) to verify that the mid-diaphyseal fracture in femur had been produced.
Cell membrane isolation
RAW 264.7 cell membrane as the MM was isolated according to a previous report [62]. Briefly, RAW 264.7 cells on the culture dish (60 mm in diameter) were harvested and suspended in the homogenization buffer containing Tris·HCl (pH 7.5, 20 mM), potassium chloride (10 mM), sucrose (75 mM), magnesium chloride (2 mM), and protease/phosphatase inhibitors. The suspension was disrupted with a probe ultrasonic disruptor (JY 92-IIN, Ningbo Scientz, 100 W, sonicate for 5 s and pause for 5 s, 10 min, 4 ºC) and then centrifuged (20,000 g, 25 min). The supernatant was centrifuged (100,000 g, 35 min) again, and the pellet was collected as the RAW 264.7 cell membrane and stored at -80 ºC until use. The protein content of MM was determined using the BCA protein assay kit. For fluorescence microscopy imaging and fluorescence resonance energy transfer (FRET) analysis, DiO-stained MM and DiI-stained MM were prepared by mixing the cell membrane with DiO or DiI at the membrane protein/dye weight ratio of 1000/1 [59].
Preparation and characterization of NCs
PG was prepared according to our previous report [54]. The chemical structure and the secondary structure of PG were determined by 1H NMR and CD, respectively. Then, PG solution (1 mg/mL in DEPC water) was mixed with siCkip-1 solution (0.1 mg/mL in DEPC water) at various PG/siCkip-1 weight ratios (2.5, 5, 10, 15, 20, and 25). The mixture was vortexed for 5 s and incubated at room temperature (RT) for 20 min to form PG/siCkip-1 (PsC) NCs. Then, CAT solution (4 mg/mL) was added to PsC NCs at various CAT/siCkip-1 weight ratios (2.5, 5, 10, and 15), vortexed for 5 s, and incubated for 20 min to obtain the CAT-adsorbed PsC NCs (CPsC NCs). Subsequently, MM-coated CPsC NCs (M@CPsC NCs) were fabricated using the sonication method as reported previously [58]. MM solution (5 mg/mL) was added to CPsC NCs at various membrane protein/siCkip-1 weight ratios (5, 10, 15, and 20), followed by sonication (2 min) to allow membrane coating. The freshly prepared NCs were subjected to electrophoresis (90 V, 20 min) in agarose gel (2%) to observe the siRNA migration. The zeta potential and hydrodynamic diameter of the freshly prepared NCs were recorded on a Zetasizer (Nano ZS 90, Malvern). The morphology of NCs was observed by transmission electron microscopy (TEM) following negative staining with phosphotungstic acid (1%, w/v). The stability of NCs in PBS (pH 7.4) containing 10% FBS was evaluated by measuring the particle size following incubation at RT for various time. To determine the membrane coating efficiency, DiD-stained MM was coated onto CPsC NCs as described above. Then, the obtained DiDM@CPsC NCs were centrifuged (10,000 g, 10 min), and the amount of un-coated DiD-stained MM in the supernatant was determined by spectrofluorimetry (λex = 644 nm, λem = 665 nm). The fluorescence intensity of the freshly prepared DiDM@CPsC NCs before centrifugation was determined and set as 100%.
To prepare sSDF-1α-immobilized M@CPsC NCs (DSM@CPsC NCs), DSPE-PEG2k-conjugated sSDF-1α (DS) was firstly prepared via the click reaction between DSPE-PEG2k-DBCO and N3-sSDF-1α. Briefly, DSPE-PEG2k-DBCO solution (10 mg/mL in PBS, pH = 7.4, 224 µL) was mixed with N3-sSDF-1α solution (5 mg/mL in PBS, pH = 7.4, 200 µL) and stirred at 37 °C for 2 h. DS was obtained after purification by ultrafiltration (MWCO = 3 kDa). The purified solution was collected and subjected to high performance liquid chromatography (HPLC, Thermofisher) analysis equipped with a UV-vis detector (λabs = 214 nm) to determine the sSDF-1α concentration in the final DS solution. A mixture of acetonitrile and water (4:1, v/v) containing 0.1% trifluoroacetic acid was used as the mobile phase. The freshly prepared DS was mixed with M@CPsC NCs at various membrane protein/sSDF-1α weight ratios, vortexed for 5 s, and incubated for 10 min to obtain DSM@CPsC NCs. The bovine serum albumin (BSA)-containing NCs (DSM@BPsC NCs) were similarly prepared, wherein BSA was used instead of CAT. The abbreviations of various NCs were listed in Table S3. FRET assay was conducted to confirm the insertion of DS into the cell membrane. Particularly, DSM@CPsC NCs were constructed from DiO-labeled MM and Cy3-labeled DS at various membrane protein/sSDF-1α weight ratios. As a control, M@CPsC NCs (containing DiO-labeled MM) were mixed with Cy3-labeled N3-sSDF-1α (without DSPE as the membrane-anchoring domain) instead of Cy3-labeled DS. The fluorescence emission spectrum of each sample was recorded between 520 and 600 nm at the excitation wavelength of 480 nm. Macrophage-specific surface markers (MAC-1 and CD68) on MM and DSM@CPsC NCs were examined by Western blot.
In vitro oxygen generation and gas-driven membrane shedding
Free CAT, DSM@CPsC NCs, and DSM@BPsC NCs (0.1 mg CAT or BSA/mL) were incubated with H2O2 (50 mM) at 37 °C for 1 h. The generation of oxygen bubbles was recorded by a digital camera.
To monitor the membrane shedding, freshly prepared DSM@CPsC NCs and DSM@BPsC NCs were treated with H2O2 (100 µM) for different time, followed by measurement of the size and zeta potential. Then, the FRET assay was also conducted. Briefly, DSM@CPsC NCs and DSM@BPsC NCs comprised of DiI-labeled MM and FAM-siCkip-1 were incubated with H2O2 (100 µM) for 4 h. The fluorescence emission spectra of NCs before and after H2O2 treatment were recorded between 500 and 650 nm at the excitation wavelength of 494 nm. The fluorescence recovery of the donor (FAM) at 530 nm was used to represent the membrane shedding from NCs. Finally, confocal laser scanning microscopy (CLSM, Zeiss LSM 800) was used to observe the membrane shedding. The freshly prepared DSM@CPsC NCs comprised of FAM-siCkip-1 and DiI-labeled MM were treated with H2O2 (100 µM) for 4 h followed by CLSM observation.
In vitro MSCs migration
The transwell culture system was adopted to evaluate the sSDF-1α-mediated MSCs migration. Briefly, MSCs were seeded onto the apical side of the inserts (pore size of 8.0 μm, Corning, NY, 1 × 106 cell/mL) and cultured for 24 h. Then, the cell culture medium at the basolateral side was replaced with fresh α-MEM containing sSDF-1α, H2O2-treated (100 μm, 4 h) DSM@CPsS NCs, or untreated DSM@CPsS NCs (1 µg siScr/mL, 1 µg sSDF-1α/mL). After incubation for 48 h, the culture medium at both the apical and basolateral sides was replaced with neutral formalin (10%) and incubated for 10 min. Then, cells at the basolateral side of the transwell membrane were stained with crystal violet (0.1%, 30 min), washed with PBS for three times, and observed by an optical microscope. Six fields at 20× magnification were randomly selected to count the number of migrated MSCs.
Cellular uptake and intracellular distribution of NCs in MSC
MSCs were seeded on 6-well plates (3 × 105 cells/well) and cultured for 24 h. Then, various FAM-siCkip-1-containing NCs, including CPsC NCs, DSM@CPsC NCs, H2O2-treated (100 µM, 4 h) DSM@CPsC NCs, and H2O2-treated (100 µM, 4 h) DSM@BPsSFAM NCs, were added at the final concentration of 1 µg FAM-siCkip-1/mL. After 4-h incubation, cells were washed with PBS for three times, re-suspended in PBS (0.3 mL), and subjected to flow cytometric (FCM, FACS Calibur, BD, USA) analysis. Data were analyzed using the Flowjo software.
The endo/lysosomal escape of NCs was observed by CLSM. MSCs were seeded on glass-bottomed dishes (2 × 104 cells/dish, 20 mm in diameter) and cultured for 24 h. Cells were then incubated with H2O2-treated (100 µM, 4 h), FAM-siCkip-1-containing DSM@CPsC NCs at 1 µg FAM-siCkip-1/mL for 4 h. After washing with PBS containing sodium heparin (20 U /mL) for three times, cells were stained with Lysotracker Red (200 nM, 1.5 h) and Hoechst 33342 (10 µg/mL, 20 min) followed by CLSM observation. The co-localization ratios were analyzed using the Image J software.
In vitro cytotoxicity of NCs
MSCs were seeded on 96-well plates (8 × 103 cells/well) and cultured for 24 h. Various NCs were added at 1 µg siCkip-1/mL and incubated with cells for 24 h. The cell viability was then determined by the MTT assay. Cells treated with PBS were used as the control to represent 100% viability.
In vitro Ckip-1 silencing and osteogenesis
MSCs were seeded on 6-well plates (1 × 105 cells/well) and cultured for 24 h. After replacement with fresh medium, DSM@CPsC NCs, H2O2-treated (100 µM, 4 h) DSM@CPsS NCs, or H2O2-treated (100 µM, 4 h) DSM@CPsC NCs were added at 1 µg siRNA/mL and incubated with cells for 24 h. The Ckip-1 mRNA level in cells was determined by real-time PCR. To evaluate the osteogenic differentiation of MSCs, the mRNA levels of Smad 1/5 and osteogenesis-associated genes (Runx2, Col I, ALP, and OCN) were determined by real-time PCR. Moreover, Ckip-1 and osteogenesis-associated proteins (Runx2, Col I, ALP, and OCN) levels were also determined by Western blot. The concentrations of primary antibody and second antibody were both 1/1000. GAPDH was used as the internal control.
To further explore the NCs-mediated osteogenic differentiation of MSCs, MSCs were seeded on 6-well plates (1 × 105 cells/well) and cultured for 24 h. After replacement with osteo-induction medium, H2O2-treated (100 µM, 4 h) DSM@CPsC NCs or DSM@CPsS NCs were added at 1 µg siRNA/mL. The medium containing various NCs was refreshed every 2 d. After 14-d incubation, cells were washed with PBS for three times, fixed with neutral formalin (10%, 10 min), and stained by Alizarin red S (ARS, 5 mg/mL, 15 min) to show calcium deposition. Cells were washed with PBS for three times and imaged using an inverted microscope (Leica TSR2). Cells were then incubated with cetylpyridinium chloride (10%, pH = 7.0) for 15 min, and subjected to determination of absorbance at 562 nm using a microplate reader (Bio-Tek, Synergy H1).
Pharmacokinetics, biodistribution, and fracture-targeting of NCs in vivo
C57BL/6 mice were i.v. injected with Cy5-siCkip-1-containing CPsC NCs or DSM@CPsC NCs (1 mg Cy5-siCkip-1/kg, 1 mg sSDF-1α/kg). At predetermined time points, blood (70 µL) was collected from the orbit and mixed with the passive lysis buffer (100 µL, supplemented with 1% Triton X-100). Dimethyl sulfoxide (DMSO, 200 µL) was added into the mixture and incubated overnight at RT. After centrifugation (14,800 rpm, 30 min), the concentration of Cy5-siCkip-1 in the supernatant was determined by spectrofluorimetry (λex = 633 nm, λem = 678 nm).
For the evaluation of the in vivo targeting of fractured femur, femur-fractured C57BL/6 mice were i.v. injected with Cy5-siCkip-1-containing DSM@CPsC NCs or CPsC NCs (1 mg Cy5-siCkip-1/kg, 1 mg sSDF-1α/kg) at 24 h post fracture. At predetermined time intervals (1, 3, 6, 9, 12, and 24 h), mice were imaged using the Maestro In Vivo Imaging System. In a parallel study, mice were sacrificed at 24 h post injection. The major organs (heart, liver, spleen, lung, and kidney) and the whole femurs were harvested and imaged (λex = 633 nm, λem = 678 nm).
In vivo photoacoustic (PA) imaging
Femur-fractured C57BL/6 mice were i.v. injected with PBS, DSM@BPsC NCs, or DSM@CPsC NCs (1 mg siCkip-1/kg, 1 mg sSDF-1α/kg) at 24 h post fracture. The echo signal from O2 in the fractured femur was recorded using the PA imaging system (FujiFilm VisualSonics Inc.) with the PA mode (Oxy-hem mode, 750 and 850 nm) at 3 h post injection.
In vivo membrane shedding
The in vivo membrane shedding from NCs was monitored by the FRET assay. Femur-fractured C57BL/6 mice were i.v. injected with DSM@CPsC NCs or DSM@BPsC NCs comprised of Cy5-siCkip-1 and DiI-labeled MM (1 mg Cy5-siCkip-1/kg). At 6 h post injection, mice were sacrificed, and the fractured femurs were harvested and imaged using the Maestro In Vivo Imaging System (λex = 550 nm, λem = 670 nm). The disappearance of the acceptor (Cy5) signal was used to represent the separation of MM and siCkip-1.
In vivo uptake of NCs by MSCs
Femur-fractured C57BL/6 mice were i.v. injected with FAM-siCkip-1-containing NCs (DSM@CPsC NCs and DSM@BPsC NCs) at 1 mg FAM-siCkip-1/kg. At 24 h post injection, mice were sacrificed, and the fractured femurs were harvested, flushed with sterile PBS, grinded with a mortar, and digested with enzymes (2 mg/mL collagenase, 100 µg/mL DNase, and 20 µg/mL RNase) for 1 h to prepare mono-dispersed cell suspensions. Cells were collected via centrifugation (500 g, 3 min), stained with antibodies (PerCP/Cy5.5-labeled anti-CD45, PE-labeled anti-CD73, and APC-labeled anti-CD105) for 0.5 h, washed with PBS for three times, resuspended in the FACS buffer (PBS containing 1% FBS, 0.2 mL), and subjected to FCM analysis. Cells were first gated using FSC/SSC, followed by CD45, CD73, and CD105 gating to identify CD45−CD73+CD105+ populations (MSCs) for the determination of MSCs that had taken up FAM-siCkip-1-containing NCs.
In vivo MSCs recruitment, Ckip-1 silencing, and osteogenesis
Femur-fractured C57BL/6 mice were i.v. injected with PBS, DSM@CPsS NCs, DSM@CPsC NCs, or M@CPsC NCs (1 mg siRNA/kg, 1 mg sSDF-1α/kg) on day 1, 3, and 5 post fracture. Five millimeter-length bones spanning the fracture sites were harvested on day 7, flushed with sterile PBS, grinded with a mortar, and digested with enzymes (2 mg/mL collagenase, 100 µg/mL DNase, and 20 µg/mL RNase) for 1 h to prepare mono-dispersed cell suspensions. Cells were then washed with PBS, stained with antibodies (PerCP/Cy5.5-labeled anti-CD45, PE-labeled anti-CD73, and APC-labeled anti-CD105), washed with PBS, and subjected to FCM analysis. In a parallel study, femurs were harvested on day 7 or 28 post fracture. The mRNA levels of Ckip-1 and osteogenesis-associated genes (Runx2, Col I, ALP, and OCN) in femurs on day 7 were determined by real-time PCR following reported protocols [59]. The Ckip-1 protein level on day 7 was further determined by Western blot as described before [54]. The expression levels of osteogenesis-associated proteins (Runx2, Col I, ALP, OCN, and OPG) on day 28 were determined by immunofluorescence (IFC) or immunohistochemical (IHC) staining and quantified using the ImageJ software. Data were calculated as the mean fluorescence intensity (for IFC) or mean gray value (for IHC) of the positive cells.
X-ray and micro-CT imaging
Femur-fractured C57BL/6 mice were i.v. injected with DSM@CPsS NCs, DSM@CPsC NCs, or M@CPsC NCs (1 mg siRNA/kg, 1 mg sSDF-1α/kg) on day 1, 3, and 5 post fracture. Mice were imaged with an X-ray imaging system (Faxitron MX-20, Tucson, AZ) on day 14 and 28. The radiographs on day 28 were scored for callus opacity, cortical remodeling/bridging, and periosteal/endosteal reaction by three independent assessors blinded to grouping and total scores were calculated [63]. Callus opacity and periosteal/endosteal reaction were scored by the apparent radiographic density and the significance level of periosteal/endosteal reaction, respectively. Scoring range for both callus opacity and periosteal/endosteal reaction went from 0 (none) to 3 (marked). Cortical remodeling/bridging was scored by visibility and mineralization of cortical edges, and scores ranged from 0 (none) to 4 (complete union with well demarcated medullary canal).
In a parallel study, femurs were harvested on day 28 post fracture and scanned using micro-CT (Skyscan 1176, Belgium) as described previously [62]. High resolution scanograms (9–20 mm) were obtained (resolution: 8.8 mm, source voltage: 50 kV, source current: 500 mA, rotation step: 0.7 unit). The data set was reconstructed using the CT analyzer software (Skyscan) to obtain the 3D images of femur and to measure morphometric parameters. The region of interest (ROI) in the fractured femur was chosen for the determination of bone mineral density (BMD), trabecular number (Tb.N), and trabecular separation/spacing (Tb.Sp).
Histological analysis
Femurs were harvested on day 28 post fracture as described above. Excess soft tissues and skin were removed. Femurs were fixed in 4% neutral formalin for 3 d, decalcified in 10% ethylenediaminetetraacetic acid solution for one month at RT, embedded in paraffin, and sliced at 6 μm in thickness. The femur sections were stained with Sirius red, Masson’s trichrome (MT), haematoxylin and eosin (H&E), or Safranin O/fast green to evaluate the Col I expression, total collagen deposition, new bone formation, and bone mineralization, respectively, as important indexes for osteogenesis. The contents of Col I (bright red/yellow collagen fiber), deposited total collagen (dark blue), newly formed bone (including lamellar bone and cartilage), and mineralized bone (green area) were determined using the ImageJ software. Data were denoted as the percentage of stain-positive area to the total area in each section.
Biosafety assessment of DSM@CPsC NCs
Healthy C57BL/6 mice were i.v. injected with PBS or DSM@CPsC NCs (1 mg siCkip-1/kg) for three times with 24 h spaced between each injection. Blood was collected at 24 h post the last injection followed by hematological and biochemical analyses. Major organs were also harvested and subjected to histological assessment using H&E staining.
Statistical analysis
All data were presented as means ± standard deviations, and statistical analysis was performed using One-way ANOVA. The differences between two experimental groups were judged to be significant at *p < 0.05 and very significant at **p < 0.01 and ***p < 0.001.
Add Comment