The effects of recombinant human bone morphogenetic protein-2, recombinant human bone morphogenetic protein-12, and adenoviral bone morphogenetic protein-12 on matrix synthesis in human annulus fibrosis and nucleus pulposus cells

The Spine Journal 8 (2008) 449–456

Review Article

The effects of recombinant human bone morphogenetic protein-2, recombinant human bone morphogenetic protein-12, and adenoviral bone morphogenetic protein-12 on matrix synthesis in human annulus fibrosis and nucleus pulposus cells Lars Gilbertson, PhDa, Sang-Ho Ahn, MD, PhDb, Pang-Ning Teng, BSa, Rebecca K. Studer, PhDc, Christopher Niyibizi, PhDd, James D. Kang, MDa,* a

Department of Orthopedic Surgery, University of Pittsburgh, E1641 BST Pittsburgh, PA 15261, USA b Department of Rehabilitation Medicine, Yeungnam University, Daegu, Korea c VA Pittsburgh Healthcare System, Pittsburgh, PA 15240, USA d Department of Orthopaedics and Rehabilitation, Pennsylvania State University College of Medicine, Hershey, PA 17033, USA

Abstract

BACKGROUND CONTEXT: Bone morphogenetic proteins (BMPs) are potential therapeutic factors for degenerative discs, and BMP-12 does not have the osteogenic potential of BMP-2, making it better suited for intradiscal injection. However, no reports have compared the actions of BMP-2 and -12 on human annulus fibrosus (AF) and nucleus pulposus (NP) cells nor evaluated adenoviral-mediated gene therapy in human AF cells. PURPOSE: To evaluate and compare the effects of recombinant human (rh) BMP-2, rhBMP-12, and adenoviral BMP-12 (Ad-BMP-12) on nucleus pulposus and annulus fibrosis cell matrix protein synthesis. STUDY DESIGN: In vitro study using rhBMP-2 and -12 and adenoviral BMP-12 with human intervertebral disc (IVD) cells. METHODS: Human NP and AF IVD cells were isolated, maintained in monolayer, and incubated with BMP-2 or -12 for 2 days. AF and NP cells were transduced with Ad-BMP-12, pellets formed, and incubated for 6 days. Growth factor–treated cells were labelled with either 35-S or 3H-proline to assay matrix protein synthesis. RESULTS: rhBMP-2 increased NP proteoglycan, collagen, and noncollagen protein synthesis to 355%, 388%, and 234% of control. RhBMP-12 increased the same NP matrix proteins’ synthesis to 140%, 143%, and 160% of control. Effects on AF matrix protein synthesis were minimal. AdBMP-12 significantly increased matrix protein synthesis and DNA content of AF and NP cells in pellet culture. NP synthesis of all matrix proteins and AF synthesis of proteoglycans was increased when the data were normalized to pellet DNA. AF synthesis of noncollagen protein and collagen was not modulated by Ad-BMP-12 if the data are normalized to pellet DNA content. CONCLUSIONS: Both rhBMP-2 and -12 increase human NP cell matrix protein synthesis while having minimal effects on AF cells. However, Ad-BMP-12 did increase matrix protein synthesis in both NP and AF cells, making it a potential therapy for enhancing matrix production in the IVD. These responses plus the proliferative action of Ad-BMP-12 seen in the current studies, and the lack of an osteogenic action noted in other studies justifies future studies to determine if gene therapy with BMP-12 could provide protective and/or reparative actions in degenerating discs. Ó 2008 Elsevier Inc. All rights reserved.

Keywords:

Human intervertebral disc; Nucleus pulposus cells; Annulus fibrosis cells; BMP-2; BMP-12; Matrix protein synthesis; Gene therapy

FDA device/drug status: investigational/not approved (osteogenic protein-1). Nothing of value received from a commercial entity related to this manuscript. 1529-9430/08/$ – see front matter Ó 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.spinee.2006.11.006

* Corresponding author. Ferguson Lab, E1641 BST 200 Lothrop St, Pittsburgh, PA 15261. Tel.: (412) 648-1092; fax: (412) 383-5307. E-mail address: [email protected] (J.D. Kang)

450

L. Gilbertson et al. / The Spine Journal 8 (2008) 449–456

Introduction Age- and injury-related disc degeneration is associated with chronic back pain and disorders of the spine costing billions for medical expenses and disability payments. Current therapies are expensive, and although they provide relief to the accompanying pain and disability, in many cases they do not address the underlying problem of functional disc loss [1]. Thus, maintenance and restoration of normal disc composition, morphology, and function are key objectives of potential molecular therapies using the actions of growth factors to stimulate matrix protein synthesis and deposition by the cells of the disc. Disc proteoglycans confer stiffness and resilience to resist compressive forces [2], and proteoglycan loss is a notable characteristic of the degenerating disc. Collagens give the disc its tensile strength and account for up to 70% of the dry weight of the outer annulus [3]. Collagen and noncollagen protein changes in disc degeneration have received less attention, although a disc with decreased collagen cannot provide the normal stretch function necessary to resist tearing under compression [4], thus contributing to ongoing deterioration. Noncollagen matrix proteins, such as link proteins, elastin, fibronectin, osteonectin, lamin, and tenascin, can modulate cell phenotype and matrix organization; thus, their maintenance is also critical to normal disc function and composition [3,5,6]. Intervertebral discs are avascular and without lymphatic drainage and thus isolated from the systemic circulation. This provides a contained environment suiting them for intradiscal therapeutic injection of factors targeting the cells of the nucleus pulposus and/or the anulus, depending on the nature of the degeneration or injury. The potential for long-term delivery of factors by modification of the disc cells via gene therapy has been shown [7], and the identification and characterization of substances that could arrest or reverse degeneration of the disc remains an area of intense investigation. Nucleus pulposus cells increase proteoglycan synthesis in response to several growth factors. These include transforming growth factor-b (TGF-b) [8], insulin-like growth factor-1 [9], platelet-derived growth factor, epidermal growth factor (EGF) [10], osteogenic protein-1 (OP-1, bone morphogenetic protein [BMP]-7) [11,12], fibroblast growth factor (bFGF) [13], and bone morphogenetic protein-2 (BMP-2) [14,15]. Collagen synthesis is stimulated by OP-1 in rat nucleus pulposus cells [15], and collagen I and II messenger RNA (mRNA) are upregulated by recombinant human BMP-2 (rhBMP-2) in cultures of inner anulus and nucleus pulposus cells from human lumbar and cervical discs [14]. Thus, BMPs could be useful therapeutic agents to increase disc synthesis of critical matrix proteins. The BMPs are members of the TGF-b superfamily with pleitrophic actions on the formation of cartilage, bone, and the connective tissues associated with the skeleton [16–18]

and are thus potential candidates for treatment of disc degeneration. As noted earlier, exogenous BMP-2 can increase matrix protein synthesis by nucleus and annular cells, an action that could potentially slow the course of disc degeneration. Furthermore, we [19] and others [20] have shown the potential of endogenous production of BMP-2, effected by adenoviral vector transduction of disc cells, to increase matrix protein synthesis. BMP-12 is another member of the growth/differentiation factor subgroup of the TGF-b superfamily (also known as growth/differentiation factor-7). It has a crucial role in formation or induction of tissues and organs during development [16,18,21] and is present in the ovoid cells of healthy human patellar tendons [22] where it is associated with proliferation. BMP-12 stimulation of the formation of tendon and cartilage-like tissue [23–27] was recently reported, suggesting its potential for targeting the annular cells for reparative function in intervertebral disc disease. Unlike BMP-2 in certain settings [18,24,25], BMP-12 does not stimulate bone matrix synthesis. The potential of BMP12 to modulate matrix protein synthesis in tissues of the intervertebral disc has not been evaluated and is thus one objective of the current study. Intradiscal administration of OP-1 has shown a modest effect in the discs of normal rabbits [28] and induced restoration of disc height for up to 24 weeks in a rabbit annular needle puncture model of disc degeneration [29]. However, several factors may limit the efficacy of growth factors per se in treatment of chronic degenerative disc diseases [1]. These include (1) the cost associated with the application of purified factors and (2) their brief half-life in vivo. To circumvent these limitations, introduction of therapeutic factors to the disc via ‘‘gene therapy’’ is being explored [7,30–32]. Critical to the development of this approach to treatment is screening the effects of growth factors in the in vitro setting to assess their potential for gene therapy before testing in animal models of disc degeneration [28,33– 35]. Although there are reports showing the efficacy of gene delivery to bovine [20] human NP cells [36], no reports of viral transduction to increase matrix synthesis in degenerated human AF cells have been published. Furthermore, previous studies suggest that growth factors may show different effects in annulus fibrosis (AF) and nucleus pulposus (NP) cells [10,11,33]. Thus, the objectives of this study are to evaluate and compare the potential of recombinant human BMP-2 (rhBMP-2) and rhBMP-12 to upregulate proteoglycan, collagen, and noncollagen protein synthesis in nucleus pulposus and annular cells from human degenerated discs. Cells in both monolayer and pellets were used to compare the behavior of cells in 2- and 3-dimensional culture systems. A second objective was to determine the response of AF and NP cells to adenoviral mediated delivery of BMP12 cDNA (Ad-BMP-12) as an initial step in evaluating the potential of this factor for gene therapy of degenerating discs.

L. Gilbertson et al. / The Spine Journal 8 (2008) 449–456

Methods Disc cell isolation AF and NP tissues were harvested from 18 patients during surgical disc procedures performed for herniated disc, degenerative disc disease, and spinal stenosis. The specimens studied were from anterior sites of cervical and lumbar discs whose overall grading of disc degeneration using magnetic resonance imaging–based Thompson score was III [37]. Samples from nucleus and annular sites were dissected from the specimens, washed with Hank’s balanced salt solution (HBSS), and transported in sterile HBSS to the laboratory within 30 minutes. The experimental protocol was approved by the human subjects Institutional Review Board at the University of Pittsburgh. Specimens were rinsed 3 times with HBSS, minced into small fragments of approximately 2 mm3, and cells from the AF and NP isolated as described by Chelberg et al. [38]. Briefly, tissues were digested in Ham’s F-12 medium with 1% penicillin/streptomycin, 5% fetal bovine serum, and 0.2% pronase (Calbiochem, La Jolla, CA) for 90 minutes. Preparations were washed three times with HBSS and incubated in 0.02% collagenase type II (Sigma, St. Louis, MO, USA) overnight at 37 C with gentle agitation. Cells were passed through a sterile nylon mesh (75-mm pore size) and centrifuged at 2,000 rpm for 5 minutes, resuspended, and cultured with Ham’s F-12 medium (containing 10% FCS, 1% penicillin/streptomycin, 50 mg/mL L-ascorbic acid) in a 5% CO2 incubator, typically for 2 to 3 weeks, until confluent.

451

loci with expression driven by a human cytomegalovirus promoter [39] were provided by Paul D. Robbins, PhD, Department of Molecular Genetics and Biochemistry, University of Pittsburgh. Control preparations were exposed to the same regimen of media changes but without addition of the Ad-BMP-12 vector. After Ad-BMP-12 transduction, AF and NP cells were trypsinized and pelleted by centrifuging at 2,000 rpm for 5 minutes. A hemocytometer was used to count the cells and each pellet contained approximately 150,000 cells. AF and NP pellets were incubated in NT serumless media with 1% insulin transferrin selenium premix for 6 days and matrix protein synthesis measured. Experiments were performed in triplicate and data normalized to pellet DNA content (ds-DNA PicoGreen assay kit; Molecular Probes, Eugene, OR). Proteoglycan synthesis Cells or pellets were incubated in NT serumless medium containing 35S-sulfate (10 mCi/mL) for 7 hours. Conditioned medium was removed and the monolayers and pellets extracted with 4 mol/L guanidine hydrochloride (GuHCl) at 4 C for 48 hours with shaking. Aliquots of conditioned medium, and tissue extracts were eluted on Sephadex G-25M in PD-10 columns with 4 mol/L GuHCl containing 0.05 mol/L Tris and 0.05 mol/L Na2SO4 and the radioactivity in the newly synthesized proteoglycans determined by scintillation counting of appropriate fractions. Collagen and noncollagenase digested protein synthesis

RhBMP-2 and rhBMP-12 stimulation in monolayer cell culture NP and AF cells from 10 patients (cervical and lumbar discs) were processed as described earlier, grown to confluence in 24-well plates, and incubated in Neuman-Tytle (NT) serumless medium, with 1% insulin transferrin selenium premix for 2 days with rhBMP-2 (25, 50, 100, 200, and 300 ng/mL) or rhBMP-12 (25, 50, and 100 ng). Cells were then incubated with 35S or 3H-proline to measure matrix synthesis as described later. Experiments were performed in triplicate and the data were normalized to cell number, determined by trypsinizing and counting cells from wells treated identically to those incubated with the radioactive markers for matrix protein synthesis. Ad-BMP-12 Transduction of NP and AF cells: NP and AF cells from 8 patients (cervical and lumbar discs) were grown to 90% confluence and transduced with Ad-BMP-12 (0, 50, 100, and 150 MOI) for 6 hours, a regimen previously shown to effect a range of transduction efficiencies with 150 multiplicity of infection (MOI) providing 100% transduction [7]. Viral vectors originated from replication-deficient type 5 adenovirus lacking E1 and E3

Cells or pellets were incubated with NT serumless medium containing 50 mg/mL ascorbic acid, 50 mg/mL b-amino-propionitrile, and 10 mCi of 3H-proline/mL for 24 hours. Conditioned medium was collected and the cell samples suspended in homogenizing buffer (0.5 mL/well) and subjected to three freeze/thaw cycles. The samples were stirred at 4 C for overnight, the medium and cell layer combined, and collagen synthesis was determined by 3Hproline incorporation using a modified collagenase digestion method [40]. Briefly, a carrier protein (2 mg/mL of pepsin-solubilized bovine type I collagen) was added, proteins precipitated by addition of 10% trichloroacetic acid (TCA), and the TCA-precipitated proteins recovered by centrifugation at 13,000 rpm for 10 minutes at 4 C. The recovered pellets were suspended in 1 mL of 1 mmol/L proline solution and washed 3 times with 5% TCA to remove the unbound isotope. The pellets were suspended in collagenase buffer (5 mmol/L CaCl2, 3 mmol/L N-ethylmaleimide, and 20 U/mL purified bacterial collagenase ABC form III [Advanced Biofactures, Lynbrook, NY]). The pH was adjusted to 7.0 and pellets incubated at 37 C for 2 hours. The collagenase-digested proteins were separated from nondigested proteins by the addition of 5% TCA followed by centrifugation at 13,000 rpm for 10 minutes. One

452

L. Gilbertson et al. / The Spine Journal 8 (2008) 449–456

milliliter of the supernatant and aliquots of the pellets dissolved in 0.2 mol/L NaOH were subjected to scintillation counting to determine the collagen and noncollagen protein synthesis. Data were normalized to double strand DNA content of the samples. Statistical analysis The data were normalized to cell count or DNA content. Data are given as mean6standard error of the n noted in the Figure legends; p value of Student t test were calculated, and a p value less than 0.05 defined statistically significant differences.

Fig. 2. rhBMP-2 and rhBMP-12 stimulate NP cell collagen synthesis. NP cells were treated as described for Fig. 1 and 24-hour 3H-proline incorporation into collagenase digestible material measured after 24 hours of growth factor activation. Values are mean6SE of n524–28. *p!.05 when compared with control; #p!.05 when compared with rhBMP-2.

Results rhBMP-2 stimulated a dose-dependent increase in NP proteoglycan synthesis with a maximum increase to 388% of control seen at 300 ng/mL; the response to 200 versus 300 ng/mL was not significantly different (Fig. 1). The response to rh-BMP-12, although significantly above control at all concentrations (25–100 ng/mL), was not dose dependent and the maximum stimulation was a modest 158% at 100 ng/mL. No further increase in proteoglycan synthesis was seen in preliminary studies when rhBMP-12 was increased to 200 or 300 ng/mL. Collagen synthesis by NP cells exposed to rhBMP-2 or rhBMP-12 is presented in Figure 2. All rhBMP-2 stimulated values for collagen synthesis are significantly greater than for the same concentration of rhBMP-12. Maximal stimulation (338% of control) was seen at 200 ng/ml rhBMP-2, and the greatest response to rhBMP-12 was an increase to of 158% of control. rhBMP-2 stimulated NP noncollagen protein synthesis to a maximum of 245% at 300 ng/mL, which was significantly less (p!.05 by Student t test) than the increases in proteoglycan and collagen synthesis at the same concentration of growth factor (Fig. 3). The response to rhBMP-12 at

Fig. 1. rhBMP-2 and rhBMP-12 increase NP proteoglycan synthesis. NP cells were grown to confluence, medium changed to NT serum-free medium, and recombinant human growth factors or saline (control) added. Forty-eight hours later, the cells were pulse labelled for 7 hours with 10 uCi/mL 35S-sulfate and proteoglycan synthesis measured. Values are mean6standard error (SE) of n522–30 samples. *p!.05 when compared with control; #p!.05 when compared with rhBMP-2.

25 ng/mL was equivalent to that with BMP-2 (ie,160% of control); however, no further increase was seen with higher concentrations, again showing a modest, non-dose dependent response. Annular cells in monolayer did not increase proteoglycan synthesis in response to rhBMP-2 (100%66% of control at 25, 50, and 100 ng/mL) or rhBMP-12 (98%, 96%, and 98% of control at the same concentrations of growth factor). AF cell collagen synthesis was not consistently changed by rhBMP-2 and -12; the only significant increase was that with 100 ng/mL BMP-2 at 121%610% of control. Noncollagen protein synthesis was increased to 122%613% with rhBMP-2 or rhBMP-12 at 25 ng/mL and to 151%616% with rhBMP-2 at 100 ng/mL, all insignificant effects. The monolayer studies were done with cells at 90% confluence, and there was no significant difference in cells/well after 48 hours of incubation with rhBMP-2 or rhBMP-12. In contrast, NP and AF cells transduced with 50 to 150 MOI of Ad-BMP-12 and maintained in pellet culture for 6 days showed DNA pellet content that was doubled (Fig. 4). This increase was not dose dependent, with DNA content of NP pellets of 216%, 197%, and 203% of control at the noted dose of Ad-BMP-12. AF cells showed a similar response with pellet DNA 208%, 223%, and 198% of control at 50, 100, and 150 MOI.

Fig. 3. rhBMP-2 and rhBMP-12 increase NP noncollagen protein synthesis. Confluent NP cells were treated as described for Figure 2 and noncollagen protein synthesis measured. Values are mean6SE of n523–30. *p!.05 when compared with control; #p!.05 when compared with rhBMP-2.

L. Gilbertson et al. / The Spine Journal 8 (2008) 449–456

Fig. 4. AD-BMP-12 transduction increases DNA content of NP cells in pellet culture. NP cells were grown to 90% confluence, transduced with Ad-BMP-12 at the MOI noted, and cell pellets cultured for 6 days before harvesting for analysis of DNA content of papain digests with Pico green per the molecular probes protocol. Values are mean6SE of n56. *p!.05 when compared with control.

Fifty MOI Ad-BMP-12 increased NP cell pellet matrix protein synthesis (Fig. 5); proteoglycan synthesis was increased 390%, collagen synthesis was increased 370%, and noncollagen protein was increased 280% when compared with sham-transduced cells. Much of this increase in matrix protein synthesis was secondary to the increase in cell number. When the data are normalized to pellet DNA content, the increases were significantly greater than control; however, the apparent stimulation was reduced to 180%, 170%, and 130% of control values. Similar results were seen at the higher doses of Ad-BMP-12 (normalized values for proteoglycan synthesis expressed as percent control were 206%68% at 100 MOI and 215%612% at 150 MOI; normalized values for collagen synthesis were 194%610% and 166%68% at 100 and 150 MOI, respectively; and noncollagen protein was increased to 148%68% and 146%68% of control at the same concentrations of Ad-BMP-12). None of these values are significantly different from the response at 50 MOI. Fifty MOI Ad-BMP-12 increased AF cell pellet matrix protein synthesis 6 days after transduction (Fig. 6). Synthesis of proteoglycan, collagen, and noncollagen protein was increased to 336%, 269%, and 222%, respectively, of control (all significant increased from control values at p!.05).

Fig. 5. Ad-BMP-12 increases matrix protein synthesis by NP cells in pellet culture. NP cells were cultured, transduced at 50 MOI, pelleted, and maintained in serum-free NT culture for 6 days before determination of matrix protein synthesis as described in the Methods section. Values are mean6SE of n512–20. *p!.05 when compared with sham-transduced control.

453

Fig. 6. Ad-BMP-12 increases matrix protein synthesis by AF cells in pellet culture. AF cells were treated as described for the NP cells in Figure 5. Values are mean6SE of n56. All synthesis/pellet are significantly increased. Only proteoglycan synthesis remains significantly elevated above control when the data are normalized to pellet DNA content. *p!.05 when compared with sham-transduced control.

However, when the data are normalized to DNA content (index of cell number), only proteoglycan synthesis/DNA remains significantly increased at 161% of control. Similar to the results in NP cells, no further increase in synthesis was seen at higher MOI of Ad-BMP-12.

Discussion These in vitro studies reaffirm the ability of BMP-2 to enhance matrix protein synthesis in the NP [14,41] and suggest that cells of the AF in monolayer culture are minimally responsive to both BMP-2 and BMP-12. The dosedependent increase in proteoglycan synthesis was equivalent at 200 and 300 ng/mL BMP-2 with synthesis enhanced to over 300% of control. Kim et al. [14] observed a significant increase in proteoglycan synthesis by human NP cells in alginate beads exposed to 300 ng/L BMP-2 (175% of control), a weaker response than seen in our studies. This could be related to differences in activity of the rhBMP-2 used, differences in responsiveness in 2- versus 3 dimensional culture, loss of responsiveness with passage (P2 vs. P-1 used in the current experiments), or the growth factor concentration profile in the pellets that would be different than that on the 2-dimensional surface of the monolayers. Lee et al. [42] showed that NP cells in pellet and alginate bead culture respond to TGF-b with greater increases in proteoglycan synthesis (350% and 200%, respectively) than the same cells in monolayer (150%), consistent with responsiveness being modulated by culture conditions. Wallach et al. [19] saw a dose-dependent increase in proteoglycan synthesis by human NP cells transduced with 50 to 150 MOI of Ad-BMP-2, maintained in pellet culture, and assayed 48 hours later. A 419% increase in synthesis was seen at the highest dose of Ad-BMP-2, a result similar to the current experiments with NP cells in monolayer activated by rhBMP-2. BMP-2 activation of collagen synthesis was similar to that of proteoglycan with similar stimulation at 200 and 300 ng/mL BMP-2. The maximum response for non-collagen

454

L. Gilbertson et al. / The Spine Journal 8 (2008) 449–456

protein synthesis was seen at 100 ng/mL, albeit to a lower value (220% vs. 350%). Although human NP cells responded to BMP-12 with significant increases in matrix protein synthesis, the response was never dose dependent and always less than that seen with a comparable concentration of BMP-2 (with the exception of noncollagen protein synthesis at 25 ng/mL BMP-2 or -12). One can only speculate as to the reasons for the lesser, non–dose-dependent response to BMP-12. The data could suggest that a limited number of high affinity receptors are accessed and activated by BMP-12 in NP cells, resulting in modest increases in synthetic processes even at high concentrations of agonist. Although we have extensive knowledge of BMP type I and II receptors and the binding characteristics of several BMPs to these receptors [16,17], BMP receptors have not been well characterized in intervertebral disc (IVD) tissue. We could find no studies comparing BMP-2 and BMP-12 receptor interactions in any tissues. Although the intracellular signaling pathway(s) involving the SMAD (Mothers Against DPP Homology) proteins and the relationship between their activities and cellular response to BMP-2 and TGF-b have been studied extensively in other cells [16,18], they have not been characterized in disc tissues. Furthermore, Yeh and Lee [43] recently showed that mouse aggrecan gene promoter constructs responded differently to activation by OP-1 and BMP-2 and -4, suggesting specificity of response at the level of nuclear transcriptional regulation as well. Whether the difference in NP response to BMP-2 and BMP-12 lies at the level of receptor or postreceptor actions remains an intriguing question for future studies. The lack of response to both BMP-2 and BMP-12 by human AF cells in monolayer was disappointing. Earlier studies [10] that compared the actions of several growth factors on NP and AF tissue (25-mg pieces) from canine discs showed differences in the response as a function of both cell type and growth factor. Both bFGF and EGF could stimulate proteoglycan synthesis up to threefold; however, bFGF preferentially increased matrix protein synthesis in cells from the annulus and transition zones, whereas responsiveness to EGF was restricted to the NP and transition zone. Insulin-like growth factor modestly increased NP proteoglycan synthesis but had no effect in transition or AF cells. Perhaps more relevant to the current studies, TGF-b increased NP proteoglycan synthesis fivefold but had no effect on AF cells. More recently, Li et al. [41] showed that BMP-2 could increase both aggrecan and collagen II message in annular cells in monolayer culture from rat IVDs. Thus, the differences between our results and previous observations are probably related to species differences; however, the fact that our annular cells were obtained from degenerating human discs could also have affected the outcomes. NP cells transduced with Ad/BMP-12 and maintained in pellet culture for 6 days exhibited a significant proliferative response, with pellet DNA content doubling (Fig. 4).

Furuya et al. [21] reported that osteosarcoma ROS-17/2.8 cells responded to BMP-12 with a twofold increase in cell number by day 6 of culture, similar to our results with both NP and AF cells. Masuda et al. [11] reported a proliferative action of BMP-7 (OP-1) on rabbit NP and AF cells in alginate bead culture. Interestingly, in both of these studies, little change in cell number could be documented before 6 to 8 days of exposure to BMP-7 or BMP-12, consistent with our results with human NP and AF cells in monolayer after 48 hours of growth factor exposure. However, the increase in cell number with the cells maintained in pellet culture was surprising, as Lee et al. [42] had shown no change in human IVD cell pellet DNA from day 5 through 15 when cultured with 10% FCS. The doubling of pellet DNA 6 days after transduction with Ad-BMP-12 suggests a specific action of this growth factor to activate proliferation, even in IVD cells maintained in culture conditions that normally minimize cell division (ie, high density, three dimensional). This effect might be beneficial in the mid- to later stages of disc degeneration when cell numbers in the disc are dwindling [35], and therapies based on growth factor activation of existing cell’s matrix protein synthesis could be less than effective [30,31,35]. However, any actual benefit to the degenerating disc remains to be established by future in vivo studies. Human NP and AF cells in pellet culture did respond to Ad-BMP-12 with significant increases in matrix protein synthesis and proliferation, which contrasts with the results recently published for similar treatment of bovine NP cells [20]. In these experiments, no significant effects on proteoglycan or collagen accumulation or proliferation were detected after transduction with Ad-BMP-12. There are several possible explanations for the differences in results including the following: (1) the bovine cells were in monolayer culture in 20% serum, whereas the human cells were in pellet culture in 0% serum; (2) the bovine cells were maintained with daily medium changes, perhaps precluding accumulation of BMP-12 to concentrations that could affect matrix protein accumulation or proliferation; and (3) there may be real species differences in the response of NP cells to BMP-12. The increased cell number in pellets of NP transduced with Ad-BMP-12 accounts for the apparent greater response to BMP-12 over that seen in monolayer cells treated with rhBMP-12; when the data were normalized to DNA content, the percent increase in synthetic rate was similar to that seen in the monolayer preparations (ie, 130%– 150%) (Fig. 5). The increase in AF protein synthesis after Ad-BMP-12 transduction is also predominantly from the increase in cell number because only proteoglycan synthesis remains significantly elevated when the data are normalized to cell number. Regardless, the 161% increase in normalized proteoglycan synthesis coupled with the proliferative effect shows AF responsiveness to BMP-12 not seen in AF cells in monolayer. Whether this is a function of twoversus three-dimensional cultures, the time of exposure to

L. Gilbertson et al. / The Spine Journal 8 (2008) 449–456

BMP-12, or the effects of a continuous supply of BMP-12 as produced by the transduced cells remains to be determined. It is unlikely to be a function of the concentration of BMP-12 because both the monolayer and pellet culture studies did not show dose-dependent responses to BMP12, as discussed earlier. It is possible that the effects seen by day 6 are secondary to BMP-12 induction of other growth factors and/or their receptors. Li et al. [41] found that BMP-2 (3 days, 200 ng/mL) doubled TGF-b 1 and MMP-7 mRNA and also increased BMP receptor type IB and II mRNA in rat IVD cells. The possibility that similar effects could underlie the actions of Ad-BMP-12 on human NP and AF cells remains to be explored. Extrapolating results from cultured cells to potential in vivo actions or effects has its limitations. NP and AF cells can rapidly dedifferentiate in culture, and their behavior in vitro may or may not be replicated in vivo. However, it is critical to test materials in vitro and use the data to more intelligently design the more difficult in vivo studies that could establish therapeutic potential. Regardless of these limitations and the specific mechanisms involved in NP and AF response to Ad-BMP-12, these studies support further exploration of BMP-12 as a potential therapeutic agent for degenerative disc disease. BMP-2 stimulates higher rates of matrix protein synthesis [19,20] and does not cause osteogenic changes in NP or AF cells per se [14]. However, the possibility of extradiscal tissue exposure to ‘‘escaped’’ BMP-2 during disc gene therapy and potential ossification of these tissues [1,16,18] brings a note of caution to its therapeutic use. BMP-12 has not shown ‘‘osteogenic properties’’ in other studies [23–25]; can effect increases in proteoglycan, collagen, and noncollagen matrix protein synthesis by NP cells; and has a significant proliferative effect on both NP and AF cells. Further studies to understand and exploit these properties of BMP-12 are warranted.

Acknowledgment The authors acknowledge the generous support of the Albert B. Ferguson Jr., MD, Orthopaedic Fund of the Pittsburgh Foundation, and Medtronic Sofamor Danek. References [1] Diwan AD, Parvataneni HK, Khan SN, Sandhu SH, Girardi FP, Cammisa FP Jr. Current concepts in intervertebral disc restoration. Orthop Clin North Am 2000;31:453–64. [2] Guiot BH, Fessler RG. Molecular biology of degenerative disc disease. Neurosurgery 2000;47:1034–40. [3] Buckwalter JA. Aging and degeneration of the human intervertebral disc. Spine 1995;20:1307–14. [4] Gruber HE, Hanley EN Jr. Observations on morphologic changes in the aging and degenerating human disc: secondary collagen alterations. BMC Musculoskelet Disord 2002;3:9. Epub 2002 Mar 21. [5] Roughley PJ. Biology of Intervertebral disc aging and degeneration: Involvement of the extracellular matrix. Spine 2004;29:2691–9.

455

[6] Omlor GW, Lorenz H, Engelleiter K, et al. Changes in gene expression and protein distribution at different stages of mechanically induced disc degeneration-an in vivo study on the New Zealand white rabbit. J Orthop Res 2006;24:385–92. [7] Moon SH, Gilbertson LG, Nishida K, et al. Human intervertebral disc cells are genetically modifiable by adenovirus-mediated gene transfer: implications for the clinical management of intervertebral disc disorders. Spine 2000;25:2573–9. [8] Nishida K, Kang JD, Gilbertson LG, et al. Modulation of the biologic activity of the rabbit intervertebral disc by gene therapy: an in vivo study of adenovirus-mediated transfer of the human transforming growth factor beta 1 encoding gene. Spine 1999;24:2419–25. [9] Osada R, Ohshima H, Ishihara H, et al. Autocrine/paracrine mechanism of insulin-like growth factor-1 secretion, and the effect of insulin-like growth factor-1 on proteoglycan synthesis in bovine intervertebral discs. J Orthop Res 1996;14:690–9. [10] Thompson JP, Oegema TR Jr, Bradford DS. Stimulation of mature canine intervertebral disc by growth factors. Spine 1991;16:253–60. [11] Masuda K, Takegama K, An H, et al. Recombinant osteogenic protein-1 upregulates extracellular matrix metabolism by rabbit annulus fibrosus and nucleus pulposus cells cultured in alginate beads. J Orthop Res 2003;21:922–30. [12] Takegami K, Thonar EJ, An HS, et al. Osteogenic protein-1 enhances matrix replenishment by intervertebral disc cells previously exposed to interleukin-1. Spine 2002;27:1318–25. [13] Nagano T, Yonenobu K, Miyamoto S, Miyamoto S, Tohyama M, Ono K. Distribution of the basic fibroblast growth factor and its receptor gene expression in normal and degenerated rat intervertebral discs. Spine 1995;20:1972–8. [14] Kim D-J, Moon S-H, Kim H, et al. Bone morphogenetic potein-2 facilitates expression of chondrogenic, not osteogenic, phenotype of human intervertebral disc cells. Spine 2003;28:2679–84. [15] Yoon TS, Su Kim K, Li J, et al. The effect of bone morphogenetic protein-2 on rat intervertebral disc cells in vitro. Spine 2003;28: 1773–80. [16] Chen D, Zhao M, Mundy GR. Bone morphogenetic proteins. Growth Factors 2004;22:233–41. [17] Sakou T. Bone morphogenetic proteins: from basic studies to clinical approaches. Bone 1998;22:591–603. [18] Wozney JM, Rosen V. Bone morphogenetic protein and bone morphogenetic protein gene family in bone formation and repair. Clin Orthop 1998;346:26–37. [19] Wallach CJ, Sobajima S, Watanabe Y, et al. Gene transfer of the catabolic inhibitor TIMP-1 increases measured proteoglycans in cells from degenerated human intervertebral discs. Spine 2003;28: 2331–7. [20] Zhang Y, An HS, Thonar EJ-MA, Chubinskaya S, He JC, Phillips FM. Comparative effects of bone morphogenetic proteins and SOX9 overexpression on extracellular matrix metabolism of bovine nucleus pulposus cells. [21] Furuya K, Nifuji A, Rosen V, Noda M. Effects of GDF7/BMP12 on proliferation and alkaline phosphatase expression in rat osteoblastic osteosarcoma ROS 17/2.8 cells. J Cell Biochem 1999;72:177–80. [22] Chuen FS, Chuk CY, Ping WY, Nar WW, Kim HL, Ming CK. Immunohistochemical characterization of cells in adult human patellar tendons. J Histochem Cytochem 2004;52:1151–7. [23] Fu SC, Wong YP, Chan BP, et al. The roles of bone morphogenetic protein (BMP) 12 in stimulating the proliferation and matrix production of human patellar tendon fibroblasts. Life Sci 2003;72:2965–74. [24] Lou J, Tu Y, Ludwig FJ, Zhang J, Manske PR. Effect of bone morphogenetic protein-12 gene transfer on mesenchymal progenitor cells. Clin Orthop Relat Res 1999;369:333–9. [25] Lou J, Tu Y, Burns M, Matthew JS, Manske P. BMP-12 gene transfer augmentation of lacerated tendon repair. J Orthop Res 2001;19: 1199–202. [26] Wikesjo UM, Sorensen RG, Kinoshita A, Jian LX, Wozney JM. Periodontal repair in dogs: effect of recombinant human bone

456

L. Gilbertson / The Spine Journal 8 (2008) 449–456

[27]

[28]

[29]

[30]

[31]

[32] [33]

[34]

morphogenetic protein-12 (rhBMP-12) on regeneration of alveolar bone and periodontal attachment. J Clin Periodontol 2004;31: 662–70. Wolfman NM, Hattersley G, Cox K, et al. Ectopic induction of tendon and ligament in rats by growth and differentiation factors 5, 6, and 7, members of the TGF-beta gene family. J Clin Invest 1997;100:321–30. An HS, Takegami K, Kamada H, et al. Intradiscal administration of osteogenic protein-1 increases intervertebral disc height and proteoglycan content in the nucleus pulposus in normal adolescent rabbits. Spine 2004;30:25–32. Masuda MK, Imai Y, Okuma M, et al. Osteogenic protein-1 injection into a degenerated disc induces the restoration of disc height and structural changes in the rabbit anular puncture model. Spine 2006;31:742–54. Gelse K, von der Mark K, Aigner T, Park J, Schneider H. Articular cartilage repair by gene therapy using growth factor-producing mesenchymal cells. Arthritis Rheum 2003;48:430–41. Shimer AL, Chadderdon RC, Gilbertson LG, Kang JD. Gene therapy approaches for intervertebral disc degeneration. Spine 2004;29: 2770–8. Sobajima S, Kim JS, Gilbertson LG, Kang JD. Gene therapy for degenerative disc disease. Gene Therapy 2004;11:390–401. Horner HA, Roberts S, Bielby RC, et al. Cells from different regions of the intervertebral disc: effect of culture system on matrix expression and cell phenotype. Spine 2002;27:1018–28. Melrose J, Ghosh P, Taylor TK, et al. A longitudinal study of the matrix changes induced in the intervertebral disc by surgical damage to the annulus fibrosus. J Orthop Res 1992;10:665–76.

55 Years Ago in Spine

Verbiest’s syndromedintermittent claudication of the spinedis one of the classic symptoms of spinal stenosis. In 1954, Henk Verbiest of Utrecht, Holland, wrote the first detailed description of the symptoms, signs, and surgical findings of spinal stenosis [1]. Included in his description of seven cases were analyses of the clinical picture, cerebrospinal fluid, radiographic findings, myelography, findings at operation, and the nature of the lesions. Verbiest noted that the symptoms of compression of the lumbar nerve roots were present only on standing or walking but not at rest. The information obtained from radiographs was limited, because neither the

[35] Sobajima S, Kompel JF, Kim JS, et al. A slowly progressive and reproducible animal model of intervertebral disc degeneration characterized by MRI, X-ray, and histology. Spine 2005;30:15–24. [36] Nishida K, Kang JD, Suh JK, Robbins PD, Evans CH, Giobertson LG. Adenovirus-mediated gene transfer to nucleus pulposus cells. Implications for the treatment of intervertebral disc degeneration. Spine 1998;23:2437–43. [37] Thompson JP, Pearce RH, Schechter MT, Adams ME, Tsang IKY, Bishop PB. Preliminary evaluation of a scheme for grading the gross morphology of the human intervertebral disc. Spine 1990;15:411–5. [38] Chelberg MK, Banks GM, Geiger DF, Ogema TR Jr. Identification of heterogeneous cell populations in normal human intervertebral disc. J Anat 1995;186:43–53. [39] Robbins PD, Ghivizzani SC. Viral vectors for gene therapy. Pharmacol Ther 1998;80:35–47. [40] Suzuki K, Oyama M, Faulcon L, Robbins PD, Niyibizi C. In vivo expression of human growth hormone by genetically modified murine bone marrow stromal cells and its effect on the cells in vitro. Cell Transplant 2000;9:319–27. [41] Li J, Yoon ST, Hutton WC. Effect of bone morphogenetic protein2 (BMP-2) on matrix production, other BMPs, and BMP receptors in rat intervertebral disc cells. J Spinal Disord Tech 2004;17: 423–8. [42] Lee JU, Hall R, Pelinkovic D, et al. New use of a three-dimensional pellet culture system for human intervertebral disc cells. Spine 2001;26:2316–22. [43] Yea LC, Lee JC. Identification of an osteogenic protein-1 responsive element in the aggrecan promoter. Biochem Biophys Res Commun 2004;323:223–8.

measurement of the distance between the articular processes and that of the interlaminer angles could be determined accurately, nor could exact measurements be obtained of the diameters of the spinal canal. Myelography revealed a complete block in all seven patients in the lumbar region; this block had the characteristics of extradural compression and was ‘‘toothed’’ or fringed. At operation, Verbiest found narrowing of the spinal canal in part of its lumbar course. He attributed the cause of this narrowing to encroachment of the articular processes on the vertebral canal and noted that the effects were most pronounced at L3 and L4. In this classic paper, Verbiest described a form of narrowing of the lumbar vertebral canal with a typical clinical picture, in which a decompression of the dural sheath was often followed by complete relief. Reference [1] Verbiest H. A radicular syndrome from developmental narrowing of the lumbar vertebral canal. J Bone Joint Surg Br 1954;36-B: 230–7.