Mesenchymal stem cells enhance targeted bone growth from injectable hydrogels with BMP‐2 peptides

Osteoporosis is the most common chronic metabolic bone disease, and the prevalence of osteoporotic fractures is rapidly increasing with the aging population. While bisphosphonates can reduce bone loss and risk of fracture, these drugs are systemic, rely on long‐term use, and patient compliance is low. Recombinant human bone morphogenetic protein‐2 (BMP‐2) is an FDA‐approved protein that can offer a more targeted therapeutic than systemic treatments. DWIVA is a peptide sequence corresponding to the wrist epitope of BMP‐2, and DWIVA‐functionalized hydrogels feature osteoinductive propertiesin vitro and in vivo. This study reports that self‐forming DWIVA‐functionalized hydrogels injected into the intramedullary canal of rat femurs induce a local increase in trabecular bone in as little as 2 weeks. Increases in bone volume, trabecular thickness, and trabeculae count from DWIVA‐laden hydrogels persist for at least 4 weeks, and the inclusion of mesenchymal stem cells (MSCs) significantly enhances the development of mineralized bone. Histological analysis of decalcified femurs also shows that hydrogel injections containing DWIVA peptide and MSCs stimulate unmineralized bone tissue formation and induce an increased count of osteoblasts and osteoclasts at the injection site after 4 weeks. Overall, the MSC‐laden DWIVA peptide‐functionalized hydrogels presented rapidly induce targeted bone formation and have the potential to form nascent bone within bones in jeopardy of an osteoporotic fracture such as the femur.

affected femur breaks, resulting in disability, loss of independence, and a 25% chance of death within a year postinjury. 7Dual-energy X-ray absorptiometry is a commonly used, noninvasive test to measure bone mineral density and diagnose osteoporosis. 8Prophylactic treatments to prevent fracture include exercise, hip protectors, and adequate calcium and vitamin D intake. 9For higher risk patients, drugs that inhibit bone resorption (e.g., bisphosphonates and estrogen) or hormone therapy that increases bone formation (i.e., Teriparatide) are prescribed. 9Major limitations of these preventative treatments is that they are nontargeted (systemic), rely on long-term use, and patient compliance is low.Targeted interventions that rapidly promote increased bone mineral density in high-risk fracture sites such as the femoral head/neck would address these limitations but unfortunately do not exist.
Bone homeostasis relies on a dynamic equilibrium between bone formation and resorption, which is maintained by osteoclasts, osteoblasts, and osteocytes regulated via a myriad of physical and biochemical signals. 10Over 50 years ago it was first reported that unmineralized bone extracellular matrix (ECM) contains factors that stimulate bone formation, 11 and since then these factors have been classified as bone morphogenetic proteins (BMPs). 12While at least 20 different human BMPs have been identified and isolated, 12 BMP-2 and BMP-7 are the two most studied for bone formation, and the only FDA-approved BMP-based product is the INFUSE ® Bone Graft by Medtronic.This bone grafting procedure has been in the market since 2002, and it consists of lyophilized recombinant human BMP-2 (rhBMP-2) that is reconstituted and added to an absorbable collagen spongefortreating tibial non-unions and assisting in spinal fusionsurgeries. 13,14Despite its long clinical history, retrospective studies comparing vertebrae fusion rates with and without rhBMP-2 are inconclusive, 15 and its side effect profile includes postoperative inflammation, ectopic bone formation, and hyperactive osteoclastmediated bone resorption. 16These unwanted effects could be due to rhBMP-2 leakage outside of the implant site, off-target signaling, and supraphysiological rhBMP-2 dosing. 14,17,18drogels are a class of soft biomaterials that can be customized with specific properties for applications in tissue engineering and regenerative medicine. 19,20For instance, hydrogels can be synthesized with bioactive peptides that increase cell adhesion and promote stem cell differentiation. 21,22Peptidefunctionalized hydrogels also feature a high degree of control over peptide concentration and location, and hydrogels formed with rhBMP-2 mimetic peptides can maximize the therapeutic properties of rhBMP-2 while minimizing off-target effects.4][25] The DWIVA peptide sequence derived from the knuckle epitope of rhBMP-2 also has high binding affinity towards BMPR-I, and several studies have demonstrated that DWIVA-modified hydrogels induce osteogenic differentiation in vitro and in vivo. 22,26r group recently developed injectable hydrogels based on a Diels Alder reaction between norbornene and tetrazine modified hyaluronic acid macromers. 27These hydrogels can be used to inject cells through clinically used syringe needles without impacting viability, and peptides can be covalently bound to the hydrogel backbone without altering mechanical properties.Using this platform, we synthesized DWIVA-functionalized injectable hydrogels and found that encapsulated human mesenchymal stem cells (MSCs) exhibited increased alkaline phosphatase (ALP) expression in vitro, and acellular injections induced local mineralized bone formation in vivo after 4 weeks. 26Building from these findings, this study hypothesizes that the inclusion of MSCs within injectable DWIVA hydrogels will result in a more rapid and robust formation of bone ECM and will enhance bone homeostasis.To test this hypothesis, we injected different hydrogel formulations into the intramedullary canal of rat femurs and evaluated the effects of DWIVA and MSCs on nascent bone formation and homeostasis via bone morphometry and histomorphometry of explanted femurs 2-and 4-weeks postinjection.
To synthesize HANor-Me, HANor was dissolved (1% w/v) in distilled water at 4°C, and hydroxyl groups in HANor were modified with methacrylates via an esterification reaction with methacrylic anhydride (MA) added in 15-fold molar excess dropwise while maintaining pH between 8.5 and 9.0 using 5 N sodium hydroxide.After all the MA was added, the solution was left stirring overnight at room temperature, dialyzed (SpectraPor, 6−8 kDa molecular weight cutoff), frozen, and lyophilized, resulting in freeze dried HANor-Me.

| HATet synthesis
To synthesize HATet, carboxyl groups in HA-TBA were modified with tetrazine by dissolving HA-TBA (1% w/v) in 100 mM MES

| Femoral intramedullary canal injection model
To study the effects of DWIVA peptides and the combined effects of DWIVA peptides and MSCs on nascent bone formation, hydrogel injections into rat femurs were used due to the prevalence of femur fractures and similarities in bone anatomy between rats and humans. 30Before starting, the animal studies were reviewed and approved by the Cooper University Health CareIACUC.To carry out the studies, 30 femurs in 8-week-old male Lewis rats (Charles River Laboratories, average weight: 275−300 g) were injected with eitherhydrogel (G), hydrogel with peptide (GP), or hydrogel with peptide and MSCs (GPC) to make up six cohorts (n = 5 per group at two timepoints).To prepare for hydrogel injections, rats were anesthetized, and the intramedullary canal of test femurs were cleared of bone marrow and trabecular bone by drilling (1 mm drill bit) through the intercondylar notch.For each cohort, 300 μL of norbornene-containing and tetrazine-containing hydrogel solutions (2 wt%) were loaded onto separate syringes.Test femurs were completely filled with 300 μL of hydrogel solution which was prepared by mixing norbornene and tetrazine macromer solutions using a Luer Lock coupler and injecting into the intercondylar notch with a 25-gauge syringe needle.Postinjection, the drilled opening was covered with bone wax, and closure of the arthrotomy and skin was completed using Vicryl sutures.The contralateral femurs of randomly chosen rats were used as a baseline (B) control, and the test femur of additional rats underwent bone marrow clearing as described above and served as a drill-only (D) negative control.
After 2 or 4 weeks for G, GP, and GPC groups, and after 4 weeks for B and D groups, rats were euthanized, femurs were extracted, fixed with 10% neutral-buffered formalin for 24 h, washed with distilled water, and stored in 70% ethanol at 4°C.Femurs were then assigned random numbers (1−30) and their designation (i.e., what group and time-point they belonged to) was unknown during bone morphometry and histomorphometry analysis to minimize bias.
Sample number was set after discussions with the IACUC to minimize unnecessary use of animals, especially due to the unknown effect of the material.The "resource equation" method was used post hoc.The "E" value was 24, which suggests that adding more animals would not have increased the chance of finding a significant value. 31

| Bone histomorphometry analysis
After microCT data was acquired, the extracted femurs were decalcified by submersing in 150 mM EDTA (ethylenediaminetetraacetic acid) solution for 3 weeks with solution changes every 48 h.Decalcified femurs were then paraffin-embedded, sectioned longitudinally (5 μm per slice), mounted on histology slides, deparaffinized, and rehydrated using xylene and graded ethanol.Samples were then LOVE ET AL.
| 1601 stained for H&E (hematoxylin and eosin), Masson's Trichrome, TRAP (tartrate resistance acid phosphatase), or ALP according to manufacturer specifications.TRAP and ALP were counterstained with hematoxylin for 2 min, and all samples were imaged with a Zeiss Axio Scan.Z1 microscope.Utilizing ImageJ, cell quantification was performed on individual images, wherein 500 × 500 μm² square areas were designated for analysis.

| Statistical analysis
Statistical analysis was performed using one-way analysis of variance (analysis of variance) with Tukey's post hoc test.Differences in mean values between groups are stated as *p < 0.05, **p < 0.01, ***p < 0.001, and ns when differences between groups are not statistically significant.

| Hydrogel injections localize to the intramedullary canal of rat femurs
To assess targeted bone formation in rat femurs, the knee joints of 8-week-old male Lewis rats were surgically opened under sterile conditions.After proper identification of the intercondylar notch, the intramedullary space was cleared of native trabecular bone and bone marrow and then filled with one of three hydrogel solutions (300 μL per femur): hydrogel alone (G group), hydrogel with 2.0 mM DWIVA peptide (GP group), or hydrogel with 2.0 mM DWIVA peptide co-delivered with MSCs (GPC group) (Figure 1A).Rats were imaged with a GE OEC 9900 Elite C-Arm 1-week postinjection of hydrogels containing a contrasting agent, confirming the presence of the hydrogel at the site of injection (Figure 1B).

| Injectable DWIVA hydrogels induce local mineralized trabecular bone formation
Trabecular morphometry analysis was performed at 2-and 4-weeks postinjection using micro-CT scan renders.Axial (A) views were rendered at the center of the distal shaft as seen in Figure 2A.
Representative axial views of hydrogel (G) and hydrogel with DWIVA peptide (GP) visually show that the GP group induces more trabecular bone growth in comparison to the peptide-free G group (Figure 2B).Analysis oftrabecular bone volume fraction (BV/TV) shows a57.8% increase between G and GP groups at 2 weeks, and a 41.1% increase between the groups at 4 weeks (Figure 2C).almost doubled between 2 and 4 weeks (Supporting Information: Figure S4b).Between the 2-and 4-week GP groups there was a 20.4% increase on average trabecular thickness (Supporting Information: Figure S4c), and the average trabecular spacing exhibited a reduction of over twofold (Supporting Information: Figure S4d).
Since ectopic bone growth is an undesirable side effect of rhBMP-2 therapies, a cortical morphometry analysis was performed to determine whether there was any bone growth outside of the injection site.Evaluation of average cortical bone thickness (Ct.Th), cortical surface area (Ct.Ar), and cortical area of the periosteal envelope (Tt.Ar) revealed no changes in cortical bone between G and GP groups after 2 and 4 weeks post-trabecular injections (Supporting Information: Figure S5a-c).There were also no observable differences between the cortical bone area fraction (Ct.Ar/Tt.Ar) for any of the test groups (Supporting Information: Figure S5d), confirming that trabecular hydrogel injections did not induce bone growth in the cortical space.

| Inclusion of MSCs within injectable DWIVA hydrogels enhances mineralized bone formation
MSC-laden DWIVA peptide hydrogel (GPC) injectionsled to greater overall trabecular bone growth in comparison to acellular DWIVA peptide (GP) groups.Representative axial views of GP and GPC hydrogel groups visually show that the GPC group has more trabecular bone relative to the GP group (Figure 3A).In comparison to GP, BV/TV in the GPC group is 51.1% and 74.2% higherat 2-and 4-weeks postinjections, respectively (Figure 3B).Interestingly, after only 2 weeks postinjection, the GPC group also demonstrated a significantly similar bone fractionvalue compared to the native (B) group.Tb.N was highest in the GPC group, and the increased difference between GP and GPC groups was most pronounced at the 4-week time point (Figure 3C).There was also a trend observed in increasing Tb.Th, however, this was more subtle between the groups and time-points (Figure 3D).A decrease in Tb.Sp was also observed between the GP and GPC groups, with 41.8% and 40.9% reductions at 2-and 4-weeks postinjection, respectively (Figure 3E).Trabeculae count and spacing remained statistically different from the native (B) group, showing higher numbers and less spacing in the native samples.However, after 4-weeks postinjection, TB.Th was statistically similar between GPC and the control.
A direct comparison of the GPC group as a function of time shows a 32.43% increase in BV/TV, an increase from 0.75 to 1.11 trabeculae per millimeter, and a significant increase in trabeculae thickness between 2-and 4-weeks postinjection (Supporting Information: Figure S6a−c).In just 14 days, average intertrabecular spacing also decreased twofold (Supporting Information: Figure S6d).Nascent bone growth also remained in the trabecular injection site with no changes in cortical bone growth for at least 4 weeks postinjection (Supporting Information: Figure S7).

| DWIVA peptide and MSC injections induce increased unmineralized bone ECM
To evaluate the formation of unmineralized bone ECM, micro-CT imaged femurs were decalcified and histomorphometry was performed.To assess the presence of collagen at the trabecular region injection site, G, GP, and GPC groups were stained with Masson's Trichrome (Figure 4).As expected, cortical bone (CB) along the perimeter of the femur is rich in collagen (dark blue, Figure 4A).Representative histology images show increasing presence of collagen (red dotted regions) between the G and GP groups, and the most amount of collagen at the injection site is seen in the GPC group which includes DWIVA peptides and MSCs (Figure 4B).H&E staining of the three groups also showed more overall protein (bright pink) in the trabecular space in the GPC group when compared to the G and GP groups (Supporting Information: Figure S8).

| DWIVA peptides and MSCs induce an increase in osteoblasts and osteoclasts at the injection site
Osteoclasts and osteoblasts are two types of cells that play crucial roles in bone remodeling, which is the process of continuous renewal and maintenance of bone tissue.Decalcified femurs were stained with ALP or TRAP and counterstained with Hematoxylin to count the number of osteoblasts and osteoclasts, respectively.No significant difference in osteoblast count was observed between the G and GP groups.However, there was an over 31.4% increase in osteoblasts at the injection site for the GPC group (Figure 5C).Similar findings were observed for osteoclast count, with an increase in osteoclasts per unit area in the GPC group (Figure 5D).These changes were only seen after 4 weeks postinjection, as no changes were observed in osteoblast or osteoclast counts between G, GP, and GPC groups at the 2-week time point.Interestingly, our evaluation reveals that a mere 2 weeks of treatment is sufficient to yield significant increases in bone volume, trabeculae number, thickness, and a decrease in intertrabecular spacing.Furthermore, notable changes in osteoclast and osteoblast formation are observed specifically in the GPC group after 4 weeks.A balance between these two cell types is crucial for maintaining healthy bone homeostasis, as the process of regeneration requires a specific amount of time.Together, these findings support the hypothesis that MSCs enhancetherapeutic efficacy of DWIVA peptide-functionalized hydrogels.Moreover, it is worth noting that the timing and extent of bone formation can be influenced by several factors, including injury severity, age, and overall health of the subject.Therefore, the design of this study provided a co-analysis of qualitative and quantitative data, which is important for a more comprehensive understanding of the mechanisms underlying the observed effects.Importantly, unwanted ectopic formation of cortical bone, which is denser and more compact than trabecular bone, was not significantly different in any of the groups.

| DISCUSSION
Future studies could include additional bone analysis metrics, such as dynamic histomorphometry, to quantify bone micro architecture and remodeling of the treated areas over time.For example, bone formation rates and resorption parameters can assess aspects related to the speed of bone growth and resorption, providing data on the progression of healthy bone formation.Furthermore, trabecular and cortical parameters, such as trabeculae thickness, trabecular separation, and cortical thickness, would provide complimentary data to our micro-CT scan analyses.In summary, our bone morphometry and histomorphometry analyses provide compelling evidence that peptidefunctionalized hydrogels, when co-delivered with MSCs, promote greater bone regeneration at the injection site.These findings have significant implications for the development of targeted and efficacious therapies for bone regeneration, highlighting the potential of MSCs to enhance the therapeutic efficacy of biomaterial-based approaches.
Cortical and trabecular mineralized bone in extracted femurs was analyzed using micro computed tomography (microCT 45, Scanco-Medical AG) at 10.4 μm isotropic resolution, with 55 kVp energy, and 400 ms integration time.At the center of the distal shaft, 1031-slice-thick and 200-slice-thick VOIs (volumes of interest) were identified, Gaussian filtered (α = 1.2, support = 2), and bone was detected by applying a global threshold of 320 mg hydroxyapatite per cm 3 .Manufacturer-provided software for 3D standard microstructural analysis was used to generate 3D coronal views of the 1031-slice-thick distal shaft.3D axial views of 200-slice-thick VOIs were used for trabecular and cortical bone morphometry analysis.Outcome measures for trabecular bone morphometry were BV/TV (trabecular bone volume fraction), Tb.N (trabeculae count), Tb.Th (trabeculae thickness), and Tb.Sp (intertrabecular spacing).Outcome measures for cortical bone morphometry were Ct.Ar (cortical surface area), Ct.Th (cortical bone thickness), Tt.Ar (cortical area of the periosteal envelope), and Ct.Ar/Tt.Ar (cortical bone area fraction) (Supporting Information: Figure S1).
Trabeculae count (Tb.N) and thickness (Tb.Th) analyses show significantly more (p < 0.001) trabeculae with thicker morphologies in the GP group compared to the G group after just 2 weeks of treatment (Figure2D−E).DWIVA peptide-functionalized hydrogel injections peptides also result in decreased intertrabecular spacing (Tb.Sp) after 2 and 4 weeks (Figure2F).Native, baseline (B) femurs still held greater trabeculae count, thickness, and less trabeculae spacing relative to all hydrogel formulations; however, BV/TV of the GPC group at 4-weeks postinjection was statistically similar to the native (B) control.The drilled sample exhibited a modest increase in BV/TV between Weeks 2 and 4; nevertheless, trabecular growth remained relatively absent, with minimal changes observed over the 4-week period.The supporting information document includes trabecular analysis for both the native (B) and drill-only (D) controls (Supporting Information: FigureS2), along with representative axial views of microCT renders (Supporting Information: FigureS3).Due to increased trabecular bone growth from GP group when compared to peptide-free G group, there was interest in evaluating DWIVA peptide-mediated nascent bone growth as a function of time.There was a63.8% increase in BV/TV at the injection site for the GP group (Supporting Information: FigureS4a), and the trabecular count F I G U R E 1 Experimental design.(A) Three hydrogel variants: hydrogel (G), hydrogel with DWIVA peptide (GP), and hydrogel with DWIVA peptide and MSCs (GPC) were injected into hollowed rat femurs through the intercondylar notch.(B) Representative x-ray micrographsshowthe hydrogel remains in the injection site (trabecular space within the femur shaft) for at least 1-week postinjection.

F
I G U R E 4 Histomorphometric assessment of collagen depositionbetween hydrogel (G), hydrogel with DWIVA peptide (GP), and hydrogel with DWIVA peptide and encapsulated MSCs (GPC).Representative image of (A) rat femur from GPC group stained with Masson's Trichrome and (B) images highlighting areas of collagen tissue (red dotted line) for each respective group surrounded by cortical bone (CB).Scale bars: a,2 mm; b,200 μm.F I G U R E 3 Trabecular morphometry comparison between hydrogel with DWIVA peptide (GP) and hydrogel with DWIVA peptide and encapsulated MSCs (GPC).Representative (A) axial micro-CT images of rat femurs at 2-weeks (orange) and 4-weeks (green) postinjection.Quantification of (B) BV/TV, (C) Tb.N, (D) Tb.Th, and (E) Tb.Sp at 2-and 4-weeks postinjection.The baseline averages (B) with standard deviations (solid lines) for each metric are denoted as dark red dashed lines.Bar graphs represent the mean and error bars represent standard deviation, *p < 0.05, **p < 0.01, ***p < 0.001.Scale bar: a, 1 mm.The GP data provided is identical to the data represented in Figure 2 and is included for the purpose of comparison.
The present study aimed to investigate the combined effects of rhBMP-2 mimetic peptide DWIVA and MSCs on inducing bone growth in the intramedullary canal of rat femursafter 2-and 4-weeks of treatment.Throughbone morphometry and histomorphometry analyses, we observed that codelivery of MSCs within injectable DWIVA hydrogels resulted in a significant increase in mineralized bone, osteoid tissue, and number of bone producing and resorbing osteoblasts and osteoclasts, respectively.These findings propose that incorporating MSCs into a DWIVA peptide-functionalized hydrogel scaffold enhances the regenerative potential of the hydrogel, leading to improved outcomes in bone repair.The quantitative data obtained from this study provides further support for the efficacy of our injectable formulation, as evidenced by the increased trabecula volume, number, and thickness in the GPC group which includes the DWIVA peptide and MSCs.There are two competing theories for this phenomenon: (i) injected MSCs could differentiate into mature osteoblasts and osteoclasts at the site of injection, or (ii) injected MSCs may induce a priming effect on rat MSCs for bone cell differentiation.

F I G U R E 5
Osteoblast and osteoclast counts in trabecular bone between hydrogel (G), hydrogel with DWIVA peptide (GP), and hydrogel with DWIVA peptide and MSCs (GPC) groups.Quantification of (A) osteoblast count and (B) osteoclast count at the trabecular bone injection site at Week-2 postinjection.Quantification of (C) osteoblast count and (D) osteoclast count at the trabecular bone injection site at Week-4 postinjection.Bar graphs represent the mean and error bars represent standard deviation, ***p < 0.001.ns, no significance.