Gene Gun-mediated Gene Transfer of Osteogenic Protein-1 (OP-1) to Promote Regeneration of the Intervertebral Disc in Rabbits
H.S. An, MD; S.C. Chang, MD; K. Masuda, MD; G.B.J. Andersson, MD, PhD; E.J. Thonar, PhD
Specific Aims. The specific aim of this study is to correlate the biomechanical characteristics of lumbar motion segments with the morphological appearance of the intervertebral disc, vertebral body and facet joint by: (1) determining the stability of the motion segment by measuring the segmental stiffness in flexion, extension, lateral bending and axial rotation;
(2) determining the pathoanatomical changes before and after loading (ie, foraminal area, nerve root impingement, disc bulging, facet subluxation, etc.) using MRI and the cryomicrotomic techniques; (3) investigating MRI characteristics of the lumbar intervertebral disc and adjacent marrow to quantify the grades of disc degeneration and Modic types of disc degeneration; and (4) investigate possible correlations between above findings.
Transfection of β -galactosidase Reporter Gene. As a first step, we conducted a study to test the feasibility of using the gene gun-mediated gene transfer technique to transfect the β-galactosidase reporter gene in vitro to rabbit NP and AF cells maintained as a high density monolayer and then cultured in alginate beads. Optimization of the particle-mediated gene transfer was initially performed on a monolayer culture of AF cells because these cells are more readily available. The transfection carried out at helium gas pressures above 150 psi resulted in significant cell dislodgement from the tissue culture wells. The range of pressure from 100 psi to 150 psi was found optimal for maximizing transfection efficiency and cell viability. Lower pressures induced no apparent transfection of pCMV--gal. Thus, the 125 psi was selected to test other variable conditions for the transfection. The level of gene expression was highest when using a particle size of 1.6 µm at a DNA loading ratio of 2 µg of DNA per mg gold with the gene gun pressured at 125 psi. Under these conditions the transfection efficiency with the -gal gene was approximately 10%, as measured by manual cell counting after -gal staining. We also have assessed the duration of transgene expression of -galactosidase as well as the effects of particle-mediated gene transfer on the matrix metabolism of rabbit AF cells in three-dimensional alginate cultures. Transgene expression of β-galactosidase by both NP and AF cells in three-dimensional alginate cultures was maintained over the three weeks of culture after transfection. Our study on the biological activity showed that the content of DNA increased in a similar fashion after gene gun transfection, suggesting that the procedure did not cause significant cell damage. There was no significant difference in the rate of proteoglycan (PG) synthesis by AF cells at any time point. PG content in the beads increased similarly in all groups over the 21 days of culture. The results of live/dead cell assays and LDH measurements demonstrated that cell death caused by bombardment of the cells was less than 10%. These findings suggest that this approach does not cause significant deleterious effects on the metabolism of IVD cells.
Effect of Transfection of the OP-1 Gene to NP and AF Cells In Vitro. 
Furthermore, using bovine IVD cells we have recently shown that gene gun-mediated transfection of the OP-1 gene can induce specific metabolic changes in IVD cells (NP and AF cells). Transfer of the β-gal gene was performed to probe the efficiency of transfection in the two different cell sources. Staining of X-gal demonstrated an efficiency of 14.1% for the NP cells and 8.2% for the AF cells. To determine whether gene transfection can be used to alter the metabolism of IVD cells, the human OP-1 gene was transfected using a pW24 vector. On day three after transfection, there were no significant differences in the DNA and PG contents of the cell layer in any group. In contrast, cells transfected with the OP-1 gene synthesized PGs at a faster rate in both AF (124% of control) and NP (144% of control) groups (Figure 2). It was noteworthy that NP cells were more responsive than AF cells to the transfection of the OP-1 gene. To study whether transfection of a growth factor gene can induce metabolic changes in IVD cells, the expression of OP-1 mRNA in OP-1-transfected IVD cells was first evaluated by the RT-PCR method. Both NP and AF cells were transfected with the empty vector (control) or the OP-1 expression vector (pW24), further cultured for two days and total RNA was extracted. While the control groups showed constitutive expression of the OP-1 gene, OP-1-transfected IVD cells showed an increase in expression of the OP-1 gene greater than that found in the control cells.
We also evaluated, by Western Blot analysis, whether the IVD cells transfected with the OP-1 gene showed an increase in the amount of OP-1 protein secreted. Annulus cells showed more spontaneous secretion of OP-1 without gene transfer, which was also observed in RT-PCR. Cells that were transfected with pW24 produced a larger amount of immunoreactive protein than did the control cells. Analysis of the cell lysate showed similar results with multiple bands of 12G3 positive bands that included the proform of OP-1. The results of this study revealed, for the first time, that transfection of the OP-1 gene by the gene gun system to IVD in vitro can alter the metabolism of these cells. Both the efficiency study and the metabolic study provide evidence that the NP cells might be the best targets for transfection. IVD regeneration with cells transfected with the OP-1 gene: an ex vivo gene transfer using the rabbit model.
The potential of a non-viral gene gun-mediated gene transfer method for efficient transfection to the IVD cells has been confirmed. The transfection of the OP-1 gene stimulates IVD matrix synthesis in vitro. The introduction of gene-transfected cells, as an ex vivo transfection with a growth factor or other inhibitors to proteinases or cytokines, might be a practical choice for disc degeneration.
We have conducted a study to monitor the effect of ex 
vivo gene transfer of the OP-1 gene on the in vivo repair of the IVD of rabbits. Twelve New Zealand white rabbits were used and were divided equally into two groups: cells transfected with vacant vector (control group); cells transfected with the OP-1 gene (OP-1 group). After three days, transfected cells with vacant or OP-1 vector were injected, under general anesthesia, by a 25 gauge-needle into three consecutive levels of IVDs (L2-3 to L4-5) for each rabbit in each group. The discs in the L1-2 level were used as noninjection controls. In the intact discs (nontreatment group), there were no significant differences in disc height index (DHI) and %DHI among different time points. In the control group, a significant decrease of %DHI, compared to the nontreatment group, was observed at four weeks and sustained for up to eight weeks (4W: 89 ± 8%, 8W: 81 ± 2%) ( Figure 3) (There are asterisks on Figure 3 that are not defined). In the OP-1 group, the %DHI showed a slight increase (4W: 101 ± 3%, 8W: 107 ± 7%) ( Figure 3). The comparison between the control and OP-1 groups showed significant differences of %DHI as early as four weeks (p < 0.05) that were sustained for up to eight weeks (p < 0.01) after injection (Figure 3). Biochemical analysis also supported the results of X-ray analysis. The PG content of the NP was greater than that of the AF both in the control and OP-1 groups at eight weeks. The PG content of the OP-1 group was greater than that of the control group not only in the NP (132%, p < 0.05), but also in the AF (121%, p <0.05). Measurement of DHI showed that ex vivo gene transfer of OP-1 was very effective in the preservation of disc height when compared to the injection of nontransfected cells. The biochemical analysis of those tissues supports the contention that the delivery of gene-transfected cells with growth factor to IVD can improve the maintenance of IVD matrix.
Gene Gun-mediated Transfer of the OP-1 Gene to IVD and Articular Cartilage (AC) Tissues.
We have conducted a study to compare the efficiency of direct transfection of cells from the IVD [AF and NP] and (AC) tissues using the gene gun transfection system ex vivo. Furthermore, we studied whether or not gene gun-mediated transfer of the OP-1 gene can induce metabolic changes in IVD and AC tissues.
Transfer of the ß-galactosidase gene was performed to probe the efficiency of transfection in the three different tissue sources. After LacZ staining, LacZ positive cells were observed only on the surface of both the IVD and AC tissues (Figure 4). The transfected areas were deeper in the NP than in the AC and AF tissues.
To study whether gene transfection can alter the metabolism of IVD and AC tissues, the human OP-1 gene was transfected to tissues using a pW24 vector while the control groups received a vacant vector. Seven days after transfection, there was no difference between the DNA contents of the OP-1 and control groups, indicating that gene gun-mediated transfer is associated with no deleterious effects on tissues subjected to direct gene transfer [data not shown]. In contrast, the PG content (normalized to wet weight) was significantly higher in all tissues transfected with the OP-1 gene [AC (130%), AF (136%) and NP (170%) tissues]. NP tissues were more responsive than AC and AF tissues to the transfection of the OP-1 gene.
We have shown that the gene gun-mediated gene transfer method can be used to transfect IVD and AC tissues ex vivo without any deleterious effects. Although transfection was limited to the surface layers of the tissues, the transfection of the OP-1 gene successfully unregulated anabolic aspects of the metabolism of sulfated PGs.
Biomechanical Effects of Laminoplasty Versus Laminectomy: Stenosis and Stability
N.R. Crawford, PhD; N. Theodore, MS, MD; V.K.H. Sonntag, MD; C.A. Dickman, MD; R.H. Chamberlain; S. Baek; S. Safavi-Abbasi; M. Senoglu; V. Subramaniam
Introduction: Two neurosurgical techniques for treating cervical stenosis are laminoplasty, where the bony spinal lamina is removed and reattached in a more open orientation, or laminectomy, where the lamina is removed and discarded. Laminectomy eliminates one wall of the spinal canal and theoretically should increase the cross-sectional area of the spinal canal better than laminoplasty. However, because of the loss of the stabilizing contribution of the lamina, it would be expected that laminectomy would leave the spine in a less stable condition than laminoplasty. The purpose of this study was to compare simultaneously the biomechanical stability and the change in cross-section area during flexion and extension after laminectomy versus laminoplasty.
Methods: Seven cadaveric specimens from C2 to T2 were studied. Specimens were loaded in physiological-range flexion and extension using nonconstraining pure moments while recording segmental angles using an optical tracking system. Gore-Tex tubing was placed within the spinal canal and water was continuously pumped through the tubing while measuring back-pressure upstream from the specimen. Spinal canal cross-sectional area correlated to water pressure, allowing continuous monitoring of the smallest cross-sectional area of the canal. Specimens were tested (1) normal, (2) after modeling stenosis by inserting hemispherical wooden beads in the spinal canal at 3 levels, (3) after laminoplasty at 5 levels, and (4) after expanding laminoplasty to laminectomy.
Results: Range of motion (ROM) in the normal, stenotic, and laminoplasty conditions did not differ significantly (p>0.007, ANOVA/Holm-Sidak). However, laminectomy resulted in a significantly greater ROM than other conditions (p<0.02). The ROM after laminectomy was 13% greater than after laminoplasty. After modeling stenosis, the cross-sectional area decreased to 52±12% of normal. Laminoplasty restored the cross-sectional area to 70±12% of normal whereas laminectomy restored cross-sectional area to 101±4% of normal.
Discussion/Conclusion: Laminoplasty has a significant biomechanical advantage over laminectomy, leaving the spine in a more stable condition. However, laminoplasty was unable to completely relieve stenosis. In this study, stenosis was modeled as 52% occlusion of the spinal canal area (71% linear sagittal projection). The degree of stenosis should be considered in the clinical decision of whether laminectomy or laminoplasty is more appropriate.
Acknowledgement: This research is dedicated to Mr. Robert Chamberlain, our dear friend, who created the water flow system and performed the pilot research in this study, but passed away suddenly and unexpectedly in January 2005.
Tissue Engineering for the Intervertebral Disc: Research with the Sand Rat Model
H.E. Gruber, PhD; E.N. Hanley, Jr., MD
The objective of this study was to use a small animal model with spontaneous disc degeneration to illustrate the feasibility of autologous disc cell implantation. The use of autologous disc cells in the potential treatment of disc degeneration offers attractive possibilities for novel therapies. Results are presented on use of the sand rat (Psammomys obesus), a small rodent which spontaneously develops disc degeneration during aging, in experimental studies in which cells were harvested from a lumbar intervertebral disc, expanded in monolayer tissue culture, labeled with agents which allow subsequent immunolocalization of these cells and re-implanted in a second disc site of the donor animal. Tissue culture, disc surgery, histology and immunocytochemistry techniques were employed. Cells were either engrafted in a bioresorbable carrier tested for cell compatibility or injected into the recipient disc. Results were assessed with radiographic examination of the implantation site and with histology and immunocytochemistry. Data from 15 animals were obtained with engraftment resident in the animal for up to 33 weeks. Immunocytologic identification of engrafted cells showed that they integrated into the disc and were surrounded by normal matrix at time points up to eight-seven month’s post-engraftment. Engrafted cells exhibited either a spindle-shaped morphology in the annulus or a rounded chondrocyte-like morphology in the nucleus. Although technically challenging, our experience showed that autologous disc cell implantation can be successful and that the sand rat is a valuable model for autologous disc cell studies.
**The Effects of Doxorubicin on Spinal Fusion: An Animal Model
J.M. Kowalski, DC, MD; J.G. Heller, MD; S.T. Yoon, MD, PhD; S.C. Ludwig, MD; H. Murakami, MD
Introduction. Malignant spinal lesions may require surgical excision and segmental stabilization. The decision to perform a concomitant fusion procedure in part by the patient’s anticipated survival and the need for adjunctive chemotherapy. Based on our review of the literature, no information exists regarding the effect of chemotherapy on spinal fusion healing. The purpose of this study, therefore, was to determine the effect of a standard chemotherapeutic agent, doxorubicin, on posterolateral spinal fusion.
Methods. Thirty-two New Zealand white rabbits underwent posterior intertransverse process arthrodesis at L5-6 with the use of iliac autograft. Rabbits randomly received either intravenous doxorubicin (2.5 mg/kg) via the central vein of the ear at the time of surgery (16 animals) or no treatment (16 animals) (the control group). The animals were killed at five weeks and the spines were denuded of soft tissues. Spine fusion was assessed by manually palpating (by observers blinded to the treatment group) at the level of arthrodesis, and at the adjacent levels proximal and distal. This provided similar information to surgical fusion assessment by palpation in humans. Fusion was defined as complete absence of motion. Postero-anterior radiographs of the denuded spines were graded in a blinded fashion using a five-point scoring system (0-4) devised to describe the amount of bone observed between the L5-6 transverse processes. Grade zero indicated no bone, Grade one indicated only scattered islands of bone, Grade two indicated bridging bone with two or more discontinuous regions or one large gap, Grade three indicated bridging bone with only one narrow line discontinuity, Grade four indicated complete bridging bone at the motion segment. Power analysis conducted prior to initiation of the study indicated that an allocation of 16 animals to each group would permit detection of at least a 20% difference in fusion rates with statistical significance at p=0.05.
Results. Sixty-nine percent (11) of the 16 spines in the control group and 38% (six) of the 16 spines in the doxorubicin group fused. This difference was statistically significant (p=0.03). There was no significant correlation (p>0.05) between the radiographic grade of bone formation (0-4) and fusion as determined by palpation. There were four wound infections in the control group and four in the doxorubicin group, however solid fusions were palpated in three of four spines in both the control and treatment groups.
Discussion/Conclusion. Doxorubicin administered intravenously at the time of surgery appears to play a significant inhibitory role in the process of spinal fusion. Our data suggests that doxorubicin should not be administered perioperatively to patients in whom a spinal arthrodesis has been performed. Further investigation should elucidate whether delaying doxorubicin administration results in improved fusion potential and whether bone morphogenetic proteins could overcome these inhibitory effects.
*Abstracts/permission forms not received at the time of publication
**Current and ongoing research