Achieving a solid fusion mass is crucial for long-term success following a number of spine surgery procedures.1–3 Surgeons can employ several different strategies to help achieve this, including local and iliac crest autograft, allograft bone, bone graft extenders such as synthetic ceramics, and recombinant bone morphogenetic protein-2 (BMP-2).4–6 Although allograft bone and bone graft extenders provide an osteoconductive environment, there are few osteoinductive options available to surgeons.
Several years ago, investigators showed that statins (HMG CoA-reductase inhibitors) could stimulate osteoblast differentiation, and demonstrate bone-anabolic properties.7 The anabolic effect of statins is believed to be mediated by the same downstream pathways as BMP-2.7,8 Because BMP-2 associated complications are a dose-dependent phenomenon,9,10 a moderate upregulation of this pathway via statin administration may represent an ideal therapeutic approach. In addition to their action through BMP-2, statins have also been shown to prevent osteoblast apoptosis by activation of the transforming growth factor-β/Smad3 pathway.11
These in vitro data have been corroborated in a number of animal models for fracture healing and the reduction of periprosthetic bone loss.11–17 There is, however, relatively little literature examining the impact of statins in promoting bone formation after spine surgery. Previous attempts have focused on oral administration,17,18 a strategy that is fundamentally limited given the drug's high first-pass metabolism in the liver and low bioavailability in the bone.12,13 Previous investigators have shown that nanoparticle encapsulation of simvastatin can enhance bioavailability by enabling sustained local delivery.19,20
We sought to build on this existing work and develop a vehicle to enable local, prolonged delivery of simvastatin to the spine, and evaluate its effect on spinal fusion in a rat model. We hypothesized that we could successfully deliver simvastatin to the spine using a poly(lactic-co-glycolic acid) (PLGA) nanoparticle and that the use of these nanoparticles would result in increased bone formation and fusion rates.
MATERIALS AND METHODS
We adapted previously described methods15 to design PLGA nanoparticles. In brief, 120 mg of PLGA was dissolved in 2.4 mL chloroform, with or without 12 mg SIM (Sigma-Aldrich, St. Louis, MO) for the fabrication of SimNP. This solution was sonicated for 2 minutes at 30W, immediately added to 24 mL of a 2% polyvinyl alcohol (PVA) solution in water and sonicated again at 30W for 30 seconds yielding an opaque, white suspension. Chloroform was evaporated under vacuum and the resulting particles were centrifuged at 3000 RPM for 5 minutes, and supernatant discarded. The nanoparticle containing pellet was then lyophilized. Through this process, two types of nanoparticles were made: blank nanoparticles (BlankNP) containing only PLGA and simvastatin-loaded nanoparticles (SimNP) containing both simvastatin and PLGA. All nanoparticles were stored at -80°C before use.
Characterization of the Nanoparticles
Nanoparticles were dispersed in methanol; the mixture was then pipetted onto scanning electron microscopy (SEM) sample stubs and the methanol was allowed to evaporate before being coated with carbon. Nanoparticles were characterized using a Zeiss (Carl Zeiss AG, Oberkochen, Germany) EVO 50 scanning electron microscope operating in low-vacuum mode at an acceleration voltage of 15.0 keV using the backscatter detector. Nanoparticle diameters were measured using ImageJ software (https://imagej.net; Madison, Wisconsin); 100 measurements were taken of individual particles and reported as the average.
Sustained release of simvastatin was measured over 2 weeks using an ultraviolet spectrophotometer at 247.2 nm (Figure 1A, http://links.lww.com/BRS/B418). The wavelength was chosen because PLGA, phosphate buffered saline (PBS), and ethanol did not interfere with the absorbance at 247.2 nm. Simvastatin in ethanol and PBS was used as a standard ranging from 5 to 100 μM (Figure 1B, http://links.lww.com/BRS/B418). SimNP, ranging from 217 to 1667 μg/mL, was placed in 15 mL of PBS at 37°C with agitation. Aliquots (0.5 mL) were removed at each timepoint to measure simvastatin concentration.
In Vitro Tests
MC3T3-E1, Subclone 4 osteoblast precursor cells were purchased from ATCC (Manassas, VA) and cultured according to the manufacturer's specifications in complete (COMP) medium. COMP consisted of αMEM (with nucleosides, without ascorbic acid; ThermoFisher Scientific, Waltham, MA) supplemented with 10% fetal bovine serum (FBS; Atlanta Biologicals) and 1% Penicillin/Streptomycin (ThermoFisher Scientific, Waltham, MA). Osteoblastic differentiation of MC3T3-E1 cells was performed over the course of 10 days in mineralizing medium (MIN), consisting of COMP supplemented with 5.0 mM beta-glycerophosphate, 50 μM ascorbic acid, and 100 nM dexamethasone. Cells were cultured in 12 or 6 well tissue culture plates, and seeded at a density of 6250 cells/cm2. After allowing 24 hours for cell attachment, MC3T3-E1 cells (P = 2) were treated in triplicate with one of the following conditions: (1) COMP medium, (2) MIN medium, (3) MIN + 10−6 M SIM, (4) MIN + 16 μg/mL SimNP, or (5) MIN + 16 μg/mL BlankNP. Medium was first replaced after 4 days in treatment conditions, and half-medium changes performed every 3 days for subsequent medium changes. Cells in SIM, SimNP, and BlankNP were maintained in MIN medium after the first medium change and did not receive additional treatment.
Cytotoxicity of SIM, SimNP, and Blank NP was determined using a Thiazolyl Blue Tetrazolium Bromide (MTT; Sigma-Aldrich) cell viability assay. MC3T3-E1 cells (P = 2) were seeded into a 96-well plate at a density of 5000 cells/cm2 and allowed to attach for 24 hours. After 24 hours, cells were treated with increasing doses of SIM, SimNP, or BlankNP in MIN medium. SIM doses ranged from 10−7 to 10−4 M, and SimNP/BlankNP doses ranged from 4 to 200 μg/mL. At 48 hours post-treatment, MTT (5 mg/mL) was added 1:5 directly to the culture medium and placed back at 37°C for 2 hours to allow for the conversion of blue MTT formazan salt. After 2 hours, culture medium was removed leaving behind a purple precipitate, which was solubilized using 0.1 M HCl in 100% isopropanol. Cell viability was then quantified by measuring a570 nm, with background correction at a660 nm.
Alkaline Phosphatase Activity
Alkaline phosphatase activity was measured at 7 days using the Sensolyte pNPP Colorimetric Alkaline Phosphatase kit (Anaspec, Fremont, CA). Cell monolayers were washed once with ice-cold PBS, followed by two washes in ice-cold 1x Assay Buffer. Cells were then lysed for 1 hour in 500 μL 1x Assay Buffer + 1.0 μL Triton-X-100 under rocking conditions at 4°C. Debris was removed by centrifugation, and cell lysate was collected. ALP kinetic assay was carried out in triplicate by adding 50 μL cell lysate + 50 μL pNPP substrate, and measuring a405 nm every 5 minutes for 1 hour.
RNA Isolation and qPCR
Quantitative reverse transcription polymerase chain reaction (qRT-PCR) was performed to quantify mRNA markers of osteoblast differentiation. Cells were cultured for 7 days in one of the following conditions: (1) COMP, (2) MIN, (3) MIN and 10−6 M SIM, (4) MIN and 16 μg/mL SimNP, and (5) MIN and 16 μg/mL BlankNP. Total cellular mRNA was isolated using the TRI-Reagent (Sigma-Aldrich), and purified using the PureLink RNA Mini Kit (Ambion; ThermoFisher Scientific, Waltham, MA), according to the manufacturer's specifications. Concentration of DNase I treated total RNA was measured using a NanoDrop One spectrophotometer (Thermo Fisher Scientific). cDNA was prepared with 1 μg total RNA using the iScript Reverse Transcription Master Mix (Bio-Rad, Hercules, CA), and amplified using SsoAdvanced Universal SYBR Green (Bio-Rad). Gene expression was measured using a CFX-96 Real-Time PCR Detection System (Bio-Rad), and analyzed via the ΔΔCT method. Primer sequences used for this study are as follows: GAPDH- (forward) GGTGCTGAGTATGTCGTGGA, (reverse) GTGGTTCACACCCATCACAA; RPL13a- (forward) AGCGGATGAATACCAACCCC, (reverse) GAGGGATCCCATCCAACACC; Osteocalcin- (forward) ACCTAGCAGACACCATGAG, (reverse) GTTCACTACCTTATTGCCCTC; Osteopontin- (forward) CCGAGGTGATAGCTTGGCTT, (reverse) AGGGATGACATCGAGGGACT; RUNX2- (forward) TCCCCGGGAACCAAGAAGGCA, (reverse) AGGGAGGGCCGTGGGTTCTG; ALPL- (forward) ATGAGAAGGCCAGGGGTACA, (reverse) GGCTCAAAGAGACCTAAGAGG.
In Vivo Tests
A model of posterior spinal fusion was utilized in 40 male 12-week-old outbred Wistar rats. Briefly, a midline posterior skin incision was followed by two separate paramedian fascial incisions to expose the transverse processes (TPs) of L4 and L5. The TPs were decorticated. Corticocancellous bone was harvested from each iliac wing, morselized and implanted bilaterally with one of three treatments (BlankNP, SimNP, or SIM only) to bridge the L4 and L5 TP. Treatments were assigned randomly using a random number generator. The groups were as follows: Blank NP (15 rats), SimNP (15 rats), or SIM drug (10 rats). Dosing of each treatment arm was as follows: 100 to 200 mg/kg BlankNP, 100 to 200 mg/kg SimNP, and 10 to 20 mg/kg SIM drug. Dosing for SIM drug was based on the prior literature20–24 and the SimNP dosing was calculated based on our in vitro release profiles so that we would achieve roughly the same total dose of SIM release.
X-rays to assess for bone formation were obtained at 4 weeks and 9 weeks after the date of surgery. X-rays were scored by three blinded observers using a previously described six-point scale (0: no bone formation, 1: <25% bone filling, 2: 25 to <50% bone filling, 3: 50 to <75% bone filling, 4: 75% to 99% bone filling, 5: clear evidence of successful fusion).25 The average of the three observers’ scores was used for analysis. Interobserver reliability was calculated using intraclass correlation (ICC).26
Animals were sacrificed at 9 weeks. Spines were then explanted and a manual assessment of fusion was performed by three blinded observers checking for motion at L4-5 compared with the levels above and below. Spines were considered fused if two of three observers considered the spine fused. ICC for the MAF was also performed.
Sacrificed spines were fixed with 10% buffered formalin and sent for microcomputed tomographic (uCT) analysis. L3-L4-L5-L6 segments of rat spines were used for fusion analysis on Scanco μCT 35 (Scanco Medical, Brüttisellen, Switzerland) system. Fifteen micrometer voxel size, 55KVp, 0.36° rotation step (180° angular range), and a 400 ms exposure per view were used for the scans that were performed in DPBS 1X. The Scanco μCT software (HP, DECwindows Motif 1.6) was used for 3D reconstruction and viewing of images. After 3D reconstruction, fusion VOIs were segmented using a global threshold of 0.4 g/c. Bone volume (BV) was calculated for the fusion newly formed bone.
This animal study was approved by the institutional IUCAC at Hospital for Special Surgery and the Weill Cornell College of Medicine.
Under electron microscopy, SimNP averaged 208 ± 128.7 nm in diameter ranging in size from 29.4 to 549 nm (Figure 2A, B, http://links.lww.com/BRS/B418). When SimNP were placed in solution with PBS, we saw time-dependent release over a 2-week period (Figure 3, http://links.lww.com/BRS/B418). Approximately 50% of the total elution occurred within the first day followed by a slower release over the remainder of the 2-week interval. Total release efficiency averaged 74.1%. As the concentration of SimNP in PBS was increased, the total drug released also increased in a roughly linear fashion.
In Vitro Response
MC3T3 cells showed no reduction in cell viability at SIM concentrations up to 10−6 M and SimNP concentration upto 16 μg/mL (Figure 4, http://links.lww.com/BRS/B418). After culture for 1 week in differentiation medium, cells treated with SimNP showed significantly greater alkaline phosphatase activity than those without. SimNP also demonstrated greater alkaline phosphatase activity than BlankNP conditions, although this difference was not significant (Figure 5A, http://links.lww.com/BRS/B418). PCR showed significantly greater expression of OPN and OCN mRNA at 7 days in MC3T3 cells treated with Simvastatin and with SimNP (Figure 5B, C, http://links.lww.com/BRS/B418). Cells treated with SimNP also demonstrated greater ALP and RUNX2 mRNA at 7 days, although these differences were not significant (Figure 5D, E, http://links.lww.com/BRS/B418). Although SIM-treated groups showed increased expression of OCN and OPN, there did not appear to be a clear dose-dependent relationship with regards to expression.
In Vivo Response
Three animals were sacrificed postoperatively, two for nerve injury (one each in SimNP and BlankNP) and one due to deep infection (SimNP). In addition, 11 animals required oral antibiotics in their feed for superficial infections (five in the blank NP group, four in the SimNP group, and two in the SIM drug group). The risk of infection did not differ by group (BlankNP 33.3%, SimNP 26.7%, SIM drug 30%).
There was substantial to almost perfect agreement between raters for 4-week and 9-week x-rays (ICC: 0.8 and 0.75, respectively).26 Compared with Blank NP, SIM NP treated rats had significantly higher XR scores at 4 weeks (3.0 vs. 1.9, P = 0.010) and 9 weeks (3.6 vs. 1.8, P < 0.001). There was no difference between the SIM NP and SIM drug rats at 4 weeks but SIM NP rats had a higher XR score at 9 weeks (3.6 vs. 2.1, P = 0.005) (Figure 1).
Manual Assessment of Fusion
There was substantial agreement between raters for manual assessment of fusion (ICC = 0.73). Spines that were fused had a significantly higher XR score than spines that were considered unfused (4.7 ± 0.8 vs. 2.1 ± 0.9, P < 0.001). SimNP rats had the highest fusion rate (42.9%) and this rate was significantly higher than BlankNP (0%, P = 0.006) (Table 1). There was no difference between the fusion rates for SIM drug and BlankNP (22.2% vs. 0%, P = 0.065).
Quantitative analysis of uCT images showed no differences in total volume (TV) between groups. However, SimNP-treated animals had higher BV and percentage of BV (BV/TV) than animals treated with BlankNP (Figure 2A–D). Although the SimNP animals had higher BV and BV/TV than SIM drug animals, these differences did not reach statistical significance. The TV bone mass was calculated by multiplying TV by the apparent mean density throughout the volume. In this analysis, the SimNP group again had the highest TV bone mass and this was significantly higher than the blank controls (SimNP: 169 ± 43 mg vs. BlankNP: 111 ± 25 mg, P < 0.001). Analysis of variance (ANOVA) showed significant differences between groups in BV/TV (P < 0.001) and TV bone mass (P = 0.002). Examples of animals from each group are shown in Figure 2E, F.
Statins have long been used as potent inhibitors of cholesterol biosynthesis and have a well-established role in lowering the risk of coronary artery disease.27 More recently, statins were found to have a bone anabolic effect, in which researchers found an increase in new bone formation similar to that seen after treatment with FGF-1 and BMP-2.7 These findings have spurred significant research interest, and subsequent work has elucidated the mechanism by which statins may assist in new bone formation.7,8,11–13,28 The current thought is that statins function both by increasing osteoblast differentiation and interfering with osteoclast function. The anabolic effect of statins is believed to be mediated primarily by activation of MAPK and upregulation of the BMP-2 and Runx2 pathway.7 Statins may also prevent osteoclastogenesis by inhibiting the action of farnesyl diphosphate synthase and thereby preventing prenylation and activation of GTPase.13 Second, work in mouse bone cell cultures has found the statins increase OPG mRNA expression and decrease RANKL mRNA expression while increasing estrogen receptor expression, though the mechanisms behind these actions are not clearly understood.28
These characteristics have been shown to increase bone healing in animal models or fracture healing.22,29,30 A fracture model study in ovariectomized rats showed the benefits of local administration of simvastatin in fracture healing.22 Similarly, a rat femur fracture model showed an increase in mechanical strength after local administration of simvastatin to the fracture site.29 These results have been replicated in a number of different animal studies.14–16,20,31
Despite success with fracture healing, data in the spine are limited. The two previous studies investigating the use of statins in the spine were limited to oral administration of the drug, and thus had significant limitations.17,18 The earlier of these two studies, which administered a lower dose of simvastatin (6.5 mg/kg) orally, found no difference in the quantity or quality of the fusion mass after administration of the drug.18 The results of supratherapeutic simvastatin dosing (120 mg/kg) were more encouraging, however, showing an increase in fusion rate and grade as well as increased mechanical strength of the fused levels in the treatment group.17 Such high doses of simvastatin are necessary because orally administered statins have low bioavailability and have extremely high affinity for liver cells.27
In this study, we have demonstrated that locally applied statins encapsulated in PLGA-nanoparticles may increase the bioavailability and improve fusion rates in the lumbar spine. We were able to successfully validate the sustained release of SIM and showed that SimNP could induce an increase in mineralization and in markers of bone formation (OCN and OPN). There was no clear dose-response to SIM observed in vitro. This, however, is not inconsistent with the existing literature and might relate to the fact that high local concentrations of simvastatin can be cytotoxic.15,19 Determining the optimal dosing of SIM and SimNP, including the minimum cytotoxic dose, is an important area of future investigation. Our dosing of SIM drug in vivo was guided by previous research on fracture healing20–24 and the SimNP dose was calculated to roughly match this dose. We hypothesize that the sustained release of SIM by the nanoparticles result in lower local concentrations that provide the bone-anabolic effects of SIM while avoiding the cytotoxicity. In addition, sustained release might be beneficial because bolused local delivery of the drug alone (as in the SIM drug group) may be washed out before any meaningful bone healing can occur.
The bone formation observed in our group was quite robust and we were able to confirm the efficacy of SimNP using various methodologies. The SimNP-treated animals had higher XR fusion scores, were more likely to fuse on manual assessment, and had significantly greater bone formation than BlankNP controls using quantitative μCT analysis. These results all support the ability of simvastatin to help achieve spinal fusion. Given the cost and complications associated with BMP-2 use,32–34 simvastatin may represent an attractive, cost-effective alternative.
We believe our results argue strongly for further investigation into the utility of statins as an adjunct to help achieve spinal fusion. Future work involves replication of our results in a large animal model.
To our knowledge, this is the first study to demonstrate that the local delivery of simvastatin using a PLGA nanoparticle can assist in achieving spinal fusion in an animal model. Rats treated with SimNP had significantly more bone formation on XR and μCT analysis and were significantly more likely to achieve fusion judged by MAF than control animals (BlankNP).
1. Kornblum MB, Fischgrund JS, Herkowitz HN, et al. Degenerative lumbar spondylolisthesis with spinal stenosis: a prospective long-term study comparing fusion and pseudarthrosis. Spine (Phila Pa 1976)
2. Fischer CR, Ducoffe AR, Errico TJ. Posterior lumbar fusion: choice of approach and adjunct techniques. J Am Acad Orthop Surg
3. Jr JAC, Dimick JB, Wainess R, et al. Changes in the utilization of spinal fusion
in the United States. Neurosurgery
4. Grabowski G, Cornett CA. Bone graft and bone graft substitutes in spine surgery: current concepts and controversies. J Am Acad Orthop Surg
5. Buser Z, Brodke DS, Youssef JA, et al. Synthetic bone graft versus autograft or allograft for spinal fusion
: a systematic review. J Neurosurg Spine
6. Kaiser MG, Groff MW, Watters WC, et al. Guideline update for the performance of fusion procedures for degenerative disease of the lumbar spine. Part 16: bone graft extenders and substitutes as an adjunct for lumbar fusion. J Neurosurg Spine
7. Mundy G, Garrett R, Harris S, et al. Stimulation of bone formation in vitro and in rodents by statins. Science
8. Ghosh-Choudhury N, Mandal CC, Choudhury GG. Statin
-induced Ras activation integrates the phosphatidylinositol 3-kinase signal to Akt and MAPK for bone morphogenetic protein-2 expression in osteoblast differentiation. J Biol Chem
9. Tannoury CA, An HS. Complications with the use of bone morphogenetic protein 2 (BMP-2) in spine surgery. Spine J
10. Carragee EJ, Mitsunaga KA, Hurwitz EL, et al. Retrograde ejaculation after anterior lumbar interbody fusion using rhBMP-2: a cohort controlled study. Spine J
11. Kaji H, Naito J, Inoue Y, et al. Statin
suppresses apoptosis in osteoblastic cells: role of transforming growth factor-beta-Smad3 pathway. Horm Metab Res
12. Tsartsalis AN, Dokos C, Kaiafa GD, et al. Statins, bone formation and osteoporosis: hope or hype? Hormones (Athens)
13. Ruan F, Zheng Q, Wang J. Mechanisms of bone anabolism regulated by statins. Biosci Rep
14. Pauly S, Luttosch F, Morawski M, et al. Simvastatin
locally applied from a biodegradable coating of osteosynthetic implants improves fracture healing comparable to BMP-2 application. Bone
15. Ho MH, Chiang CP, Liu YF, et al. Highly efficient release of lovastatin from poly(lactic-co-glycolic acid) nanoparticles enhances bone repair in rats. J Orthop Res
16. Fukui T, Ii M, Shoji T, et al. Therapeutic effect of local administration of low-dose simvastatin
-conjugated gelatin hydrogel for fracture healing. J Bone Miner Res
17. Bostan B, Gunes T, Asci M, et al. Simvastatin
improves spinal fusion
in rats. Acta Orthop Traumatol Turc
18. Yee AJ, Bae HW, Friess D, et al. The use of simvastatin
in rabbit posterolateral lumbar intertransverse process spine fusion. Spine J
19. Yan Q, Xiao LQ, Tan L, et al. Controlled release of simvastatin
-loaded thermo-sensitive PLGA-PEG-PLGA hydrogel for bone tissue regeneration: in vitro and in vivo characteristics. J Biomed Mater Res A
20. Tai IC, Fu YC, Wang CK, et al. Local delivery
of controlled-release simvastatin
/PLGA/HAp microspheres enhances bone repair. Int J Nanomedicine
21. Rosselli JE, Martins DM, Martins JL, et al. The effect of simvastatin
on the regeneration of surgical cavities in the femurs of rabbits. Acta Cir Bras
22. Wang JW, Xu SW, Yang DS, et al. Locally applied simvastatin
promotes fracture healing in ovariectomized rat. Osteoporos Int
23. Papadimitriou K, Karkavelas G, Vouros I, et al. Effects of local application of simvastatin
on bone regeneration in femoral bone defects in rabbit. J Craniomaxillofac Surg
24. Tan J, Yang N, Fu X, et al. Single-dose local simvastatin
injection improves implant fixation via increased angiogenesis and bone formation in an ovariectomized rat model. Med Sci Monit
25. Kaito T, Johnson J, Ellerman J, et al. Synergistic effect of bone morphogenetic proteins 2 and 7 by ex vivo gene therapy in a rat spinal fusion
model. J Bone Joint Surg Am
26. Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics
27. Sirtori CR. The pharmacology of statins. Pharmacol Res
28. Kaji H, Kanatani M, Sugimoto T, et al. Statins modulate the levels of osteoprotegerin/receptor activator of NFkappaB ligand mRNA in mouse bone-cell cultures. Horm Metab Res
29. Skoglund B, Aspenberg P. Locally applied Simvastatin
improves fracture healing in mice. BMC Musculoskelet Disord
30. Skoglund B, Forslund C, Aspenberg P. Simvastatin
improves fracture healing in mice. J Bone Miner Res
31. Oka S, Matsumoto T, Kubo S, et al. Local administration of low-dose simvastatin
-conjugated gelatin hydrogel for tendon-bone healing in anterior cruciate ligament reconstruction. Tissue Eng A
32. Simmonds MC, Brown JV, Heirs MK, et al. Safety and effectiveness of recombinant human bone morphogenetic protein-2 for spinal fusion
: a meta-analysis of individual-participant data. Ann Intern Med
33. Fu R, Selph S, McDonagh M, et al. Effectiveness and harms of recombinant human bone morphogenetic protein-2 in spine fusion: a systematic review and meta-analysis. Ann Intern Med
34. Perri B, Cooper M, Lauryssen C, et al. Adverse swelling associated with use of rh-BMP-2 in anterior cervical discectomy and fusion: a case study. Spine J