Local delivery of bone morphogenetic protein-2 from near infrared-responsive hydrogels for bone tissue regeneration

Abstract

Achievement of spatiotemporal control of growth factors production remains a main goal in tissue engineering. In the present work, we combined inducible transgene expression and near infrared (NIR)-responsive hydrogels technologies to develop a therapeutic platform for bone regeneration. A heat-activated and dimerizer-dependent transgene expression system was incorporated into mesenchymal stem cells to conditionally control the production of bone morphogenetic protein 2 (BMP-2). Genetically engineered cells were entrapped in hydrogels based on fibrin and plasmonic gold nanoparticles that transduced incident energy of an NIR laser into heat. In the presence of dimerizer, photoinduced mild hyperthermia induced the release of bioactive BMP-2 from NIR-responsive cell constructs. A critical size bone defect, created in calvaria of immunocompetent mice, was filled with NIR-responsive hydrogels entrapping cells that expressed BMP-2 under the control of the heat-activated and dimerizer-dependent gene circuit. In animals that were treated with dimerizer, NIR irradiation of implants induced BMP-2 production in the bone lesion. Induction of NIR-responsive cell constructs conditionally expressing BMP-2 in bone defects resulted in the formation of new mineralized tissue, thus indicating the therapeutic potential of the technological platform.

Keywords: Bone morphogenetic protein 2; Bone regeneration; Gene therapy; Gold nanoparticles; Hydrogel; Near-infrared.

Figure 1.

Activation of NIR-BMP-2-HG triggered by NIR light. (A) NIR-BMP-2-HG, polymerized with the indicated concentration of HGNP, were cultured for 1 day and then irradiated with NIR laser for the indicated times. IR thermographies (left). The graph shows the mean + SD values of the maximum temperature rises detected during NIR irradiation (right), n = 3. Scale bar = 1 mm. (B) BMP-2 concentration in media conditioned by NIR-BMP-2-HG polymerized with 30 μg mL⁻¹ HGNP. One day after polymerization, NIR-BMP-2-HG were NIR-irradiated for 10 min or not in the presence (+Rm) or absence (-Rm) of 10 nM rapamycin and then cultured for 1 day. The data are relative to the BMP-2 levels (39.6 ± 2.6 pg mL⁻¹) detected in the conditioned media of unirradiated hydrogels in the absence of rapamycin which were given the arbitrary value of 100, n = 3 (C) NIR-BMP-2-HG, polymerized with 30 μg mL⁻¹ HGNP, were NIR-irradiated in the presence of 10 nM rapamycin (Rm) or 100 nM rapalog AP21967 (Rl). Timeline scheme of NIR-BMP-2-HG preparation, NIR irradiation of hydrogel (NIR), culture in the absence (-Rm/Rl) or presence (+Rm/Rl) of rapamycin or rapalog and analytical assays. At days 3, 6 and 8, medium was collected and replaced with fresh medium lacking dimerizer. The histogram shows BMP-2 concentration at days 3, 6 and 8 in media conditioned by NIR-BMP-2-HG that were NIR-irradiated for the indicated times in the presence of Rm. The data are relative to the BMP-2 levels (46.6 ± 10.3 pg mL⁻¹) detected in the conditioned media of unirradiated hydrogels at day 3, which were given the arbitrary value of 100, n = 3. (D) Cell viability in NIR-BMP-2-HG at days 7 and 12. Left: the histogram shows the metabolic activity of cells contained in NIR-BMP-2-HG that were NIR-irradiated for the indicated times in the presence of Rm. The data are relative to the values detected in unirradiated hydrogels and were given the arbitrary value of 100, n = 3. Middle: images show cells stained with calcein-AM (green) and EthD-1 (red) at the area of laser incidence. Scale bar = 50 μm. Right: the histogram shows the percentage of viable cells at the area of laser incidence. (E) BMP-2 concentration at days 3 and 8 in media conditioned by NIR-BMP-2-HG that were NIR-irradiated for 10 min in the presence of Rl. The data are relative to the BMP-2 levels (48.2 ± 27.5 pg mL⁻¹) in conditioned media of unirradiated hydrogels collected at day 3, which were given an arbitrary value of 100, n = 3. *: p < 0.05 compared to non-irradiated samples at the corresponding day.

Figure 2.

Transcriptome analysis of activated NIR-BMP-2-HG. NIR-BMP-2-HG, polymerized with 30 μg mL⁻¹ HGNP, were NIR-irradiated for 10 min in the presence of 10 nM rapamycin (Rm) or 100 nM rapalog AP21967 (Rl). (A) Timeline scheme of NIR-BMP-2-HG preparation, NIR irradiation of hydrogel (NIR), culture in the absence (-Rm/Rl) or presence (+Rm/Rl) of dimerizer and differential gene expression analyses. At days 3 and 8, medium was replaced with fresh medium lacking dimerizer. (B) Left: heatmap showing the fold induction of the 50 most highly induced genes in cells residing in NIR-BMP-2-HG that were NIR-irradiated in the presence of Rm (first column, Rm+NIR+) or Rl (second column, Rl+NIR+) as compared to cells residing in non-irradiated NIR-BMP-2-HG hydrogels incubated with the corresponding dimerizer (Rm+NIR- and Rl+NIR-, respectively). The third column shows the fold induction of same genes in NIR-BMP-2-HG hydrogels that were NIR-irradiated in the absence of dimerizer (NIR+) as compared to cells residing in non-irradiated NIR-BMP-2-HG hydrogels (NIR-). Middle and right: heatmaps showing the fold induction of the 50 most highly induced (middle) or repressed (right) genes in cells residing in NIR-BMP-2-HG that were NIR-irradiated in the absence of dimerizer (third columns, NIR+) as compared to cells residing in untreated and non-irradiated NIR-BMP-2-HG hydrogels (NIR-). The first and second columns show the fold induction of same genes in cells residing in NIR-BMP-2-HG hydrogels that were NIR-irradiated in the presence of Rm (first column, Rm+NIR+) or Rl (second column, Rl+NIR+) as compared to cells residing in non-irradiated NIR-BMP-2-HG hydrogels incubated with the corresponding dimerizer (Rm+NIR- and Rl+NIR-, respectively) (C) Gene Ontology enrichment analysis of biological processes for the 160 induced genes in cells residing in NIR-BMP-2-HG hydrogels that were NIR-irradiated in the presence of Rm or Rl as compared to cells residing in non-irradiated NIR-BMP-2-HG hydrogels incubated with the corresponding dimerizer. The total number of genes in each category is shown at the end of each bar.

Figure 3.

Chondrogenic differentiation induced by activated NIR-BMP-2-HG. (A-B) Influence of dimerizers on BMP-2 bioactivity. (A) Timeline scheme of C3H/10T1/2 micromasses culture seeding, treatments with recombinant human BMP-2 (rBMP-2) in the presence (+) or absence (−) of dimerizer, and chondrogenic differentiation assay. (B) Photographs of micromasses treated with the indicated doses of rBMP-2, rapamycin (Rm) or rapalog AP21967 (Rl) and stained with alcian blue. The histogram shows the quantitative analysis of alcian blue staining intensity of micromasses. The data are relative to the values measured in untreated micromasses, which were given the arbitrary value of 100. *: p < 0.05 compared to untreated samples. (C) Outline of the co-culture system of C3H/10T1/2 micromasses and NIR-BMP-2-HG polymerized with 30 μg mL⁻¹ HGNP, and timeline of the co-culture setting in the presence (+Rl) or absence (-Rl) of 100 nM AP21967, NIR-irradiation of hydrogel for 10 min (NIR) and chondrogenic differentiation assay. (D) Left: photographs and micrographs of alcian blue-stained micromasses that were co-cultured with NIR-BMP-2-HG irradiated (NIR+) or not (NIR-) in the presence (+Rl) or absence (-Rl) of AP21967. Right: photographs and micrographs of alcian blue stained micromasses treated with 100 nM AP21967 (+Rl) and the indicated doses of human recombinant BMP-2 (rBMP-2), following same experimental scheme shown in A. The histogram shows the quantitative analysis of alcian blue staining intensity of micromasses. The data are relative to the values measured in micromasses treated with 100 nM AP21967 alone, which were given an arbitrary value of 100, n = 3. *: p < 0.05 compared to samples treated with 100 nM AP21967 alone. ^#: p < 0.05 compared to samples treated with 100 nM AP21967 and 10 ng mL⁻¹ rBMP-2. Scale bars = 5 mm (photographs), 100 μm (micrographs).

Figure 4.

In vivo control of transgene expression induced by activated NIR-responsive hydrogels implanted in critical-size bone defects. NIR-fLuc-HG or NIR-BMP-2-HG, polymerized with 30 μg mL⁻¹ HGNP, were injected subcutaneously in a critical-size bone defect created in mouse calvaria. One and eight days later, mice were administered rapamycin and implantation region was NIR-irradiated at 11–17 mW mm⁻² for 10 min. (A) Timeline scheme of hydrogel implantation, rapamycin (Rm) administration, NIR irradiation (NIR) and bioluminescence or BMP-2 quantification assays. (B) IR thermographies of animals implanted with NIR-fLuc-HG during NIR-irradiation, at the indicated times. The graph shows the mean + SD of the maximum temperature rises detected at the implantation site during NIR irradiations, n = 5. (C) Bioluminescence imaging of animals implanted with NIR-fLuc-HG that were administered with Rm and exposed (+) or not (−) to the NIR laser. The histogram shows the average luminescence radiance levels detected at the implantation site, n = 5. (D) Mean + SD of the maximum temperature rises detected at the implantation site during NIR irradiations of animals implanted with NIR-BMP-2-HG, at the indicated times, n = 5. (E) BMP-2 levels in NIR-BMP-2-HG constructs retrieved from mice that were administered with Rm and exposed (+) or not (−) to NIR irradiation. The data are relative to the BMP-2 values detected in unirradiated implants at day 3 (0.16 ± 0.03 pg of BMP-2 per mg of retrieved implant) which were given an arbitrary value of 100, n = 4. *: p < 0.05 compared to unirradiated animals. ^#: p < 0.05 compared to NIR-irradiated samples at day 3. Scale bars = 1 cm.

Figure 5.

Control of bone regeneration by activated NIR-BMP-2-HG. (A) Timeline scheme of implantation of NIR-BMP-2-HG that polymerized with 30 μg mL⁻¹ HGNP, rapamycin (Rm) administration, NIR irradiation for 10 min (NIR) and imaging analyses. (B) X-ray imaging of implanted mice that were administered Rm and exposed (NIR+) or not (NIR-) to NIR irradiation. The histogram shows the average diameter of radiolucent areas within critical-size bone defects filled with NIR-BMP-2-HG implants that were exposed or not to NIR laser, n = 5. Scale bars = 5 mm (1 mm for insets). (C) Masson’s trichrome stain of histological cross sections of critical-size bone defects in mice treated as in (B). The histogram shows the histomorphometric quantification, assessed by the scoring method described in Section 2.11, of newly formed bone in the critical-size bone defects. The asterisks mark the positions of the original margins of the bone defects. n = 5. Scale bars = 250 μm. *: p < 0.05 compared to unirradiated animals