Regeneration of Hyaline Cartilage by Cell-Mediated ...

HUMAN GENE THERAPY 12:1805–1813 (September 20, 2001) Mary Ann Liebert, Inc.

Regeneration of Hyaline Cartilage by Cell-Mediated Gene Therapy Using Transforming Growth Factor b1-Producing Fibroblasts KWAN HEE LEE,1,2 SUN U. SONG, 2 TAE SOOK HWANG, 3 YOUNGSUK YI,4 IN SUK OH, 1 JOUNG YOON LEE,1 KYOUNG BAEK CHOI, 2 MI SOOK CHOI, 2 and SEONG-JIN KIM 5

ABSTRACT Transforming growth factor b (TGF-b) has been considered as a candidate for gene therapy of orthopedic diseases. The possible application of cell-mediated TGF-b gene therapy as a new treatment regimen for degenerative arthritis was investigated. In this study, fibroblasts expressing active TGF-b1 were injected into the knee joints of rabbits with artificially made cartilage defects to evaluate the feasibility of this therapy for orthopedic diseases. Two to 3 weeks after the injection there was evidence of cartilage regeneration, and at 4 to 6 weeks the cartilage defect was completely filled with newly grown hyaline cartilage. Histological analyses of the regenerated cartilage suggested that it was well integrated with the adjacent normal cartilage at the sides of the defect and that the newly formed tissue was indeed hyaline cartilage. Our findings suggest that cell-mediated TGF-b1 gene therapy may be a novel treatment for orthopedic diseases in which hyaline cartilage damage has occurred.

OVERVIEW SUMMARY TGF-b is being considered as a candidate for the treatment of orthopedic diseases. However, the clinical application of this protein has been limited because of its short-term effect, resulting from its short half-life, and high cost. In this article, we describe a cell-mediated TGF-b1 gene therapy for generating hyaline cartilage in the knee joints of rabbits with artificially made cartilage defects.

INTRODUCTION

I

N THE ORTHOPEDIC FIELD ,

some cytokines are being considered as suitable candidates for the treatment of orthopedic diseases. For example, bone morphogenic proteins (BMPs) have been identified as effective stimulators of bone formation (Zegzula et al., 1997; Louwerse et al., 2000; Sellers et al., 2000). Analogously, transforming growth factor b proteins (TGF-bs) have been reported to induce osteogenesis and chon-

drogenesis (Sporn and Roberts, 1988; Joyce et al., 1990). TGFb is considered to be a multifunctional cytokine, playing a regulatory role in cell growth, differentiation, and extracellular matrix protein synthesis (Sporn and Roberts, 1988). TGF-b inhibits the growth of epithelial cells and osteoclast-like cells in vitro (Chenu et al., 1988), but stimulates endochondral ossification and eventually bone formation in vivo (Matsumoto et al., 1994). TGF-b-induced bone formation is mediated by stimulating the growth of subperiosteal pluripotential cells, which eventually differentiate into cartilage-forming cells (Miettinen et al., 1994). The results of studies have suggested that TGF-b stimulated chondrocyte proteoglycan synthesis (van Beuningen et al., 1994; van Osch et al., 1998) and the growth of articular chondrocyte cells (Yonekura et al., 1999). The therapeutic value of the TGF-b protein in orthopedics has been reported (Lind et al., 1993; Critchlow et al., 1995; Ripamonti et al., 1997; Glansbeek et al., 1998). However, widespread clinical applications of this protein have been limited because of its short-term effects, resulting from its short halflife, and high cost. There have been reports of articular carti-

1Department

of Orthopedic Surgery, College of Medicine, Inha University, Inchon, South Korea 400-711. Research Center, College of Medicine, Inha University, Inchon, South Korea 400-711. of Pathology, College of Medicine, Inha University, Inchon, South Korea 400-711. 4TissueGene Inc., Gaithersburg, MD 20877. 5Laboratory of Cell Regulation and Carcinogenesis, National Cancer Institute, National Institutes of Health, Bethesda, MD 20892. 2Clinical

3Department

1805

1806 lage regeneration after the transplantation of autologous cartilage cells that were expanded in culture (Brittberg et al., 1994). However, this procedure entails two operations with an excision of the soft tissues and requires a lengthy recovery time, at least 9 to 12 months. Other growth factors, such as epidermal growth factor (EGF), basic fibroblast growth factor (bFGF), and insulin-like growth factor I (IGF-I), have also been tested for their stimulating effects on chondrogenesis in vivo, but none have showed any effect on healing standard cartilage defects (Neidel, 1992). A collagen sponge impregnated with a recombinant human BMP implanted in the knees of rabbits with a full-thickness defect showed a 70% repair of the defect at 24 weeks posttreatment (Sellers et al., 1997). Despite the promising results, this treatment still requires an operation and a long regeneration time. Therefore, a new method for the long-term and cost-effective delivery of such cytokines is essential. Degenerative arthritis is the most frequently encountered orthopedic disease associated with cartilage damage. Almost every joint in the body, such as the knee, the hip, the shoulder, and the wrist, is susceptible to cartilage damage. The pathogenesis of this disease is the degeneration of the hyaline articular cartilage, which becomes deformed, fibrillated, and eventually excavated during the course of degenerative arthritis (Mankin, 1982). If the degenerated cartilage could somehow be regenerated, most patients would be able to enjoy their lives without debilitating pain. A cell-mediated TGF-b1 gene therapeutic approach for regenerating the hyaline cartilage in knee joints of rabbits with artificially made cartilage defects was tested. Here we report that NIH 3T3 fibroblasts containing a transgene, TGF-b1, showed sustained TGF-b1 expression and induced the proliferation and differentiation of the existing chondrocytes and/or chondrocyte precursors residing in the defect base, which resulted in the complete regeneration of hyaline cartilage in 6 weeks.

LEE ET AL.

MATERIALS AND METHODS DNA and virus construction pMTMLVb1 was constructed by cloning a 1.2-kb porcine TGF-b1 cDNA into the polylinker site of the replication-defective retroviral vector pMTMLV. pMTMLV vector was derived from the retroviral vector MFG by deleting entire gag and env sequences as well as some of the c packaging sequence. All of the U3 of 59 long terminal repeat (LTR) except for 36 bp at the 59 end was replaced with the metallothionein (MT) promoter (Kim et al., 1998). pMTMLVb1 and pVSVG were cotransfected into GP293 cells by calcium phosphate method. After 48 hr of culture, the supernatant was filtered through a 0.45-mm pore size filter and the half of it was saved at 270°C for later use. NIH 3T3 cells seeded in 60-mm culture dishes 18 hr before infection were infected with the filtrate plus Polybrene (8 mg/ml; Sigma, St. Louis, MO). After 4 hr of incubation, medium was replaced with fresh medium. Infection was repeated 24 hr later with the saved viral supernatant. Transduced cells were cultured in Dulbecco’s modified Eagle’s medium with 10% fetal bovine serum and selection with neomycin (100 mM/ml) started 48 hr after infection.

Northern blot analysis Total RNA was isolated from cells by a guanidinium–isothiocyanate extraction protocol. Ten micrograms of RNA was electrophoresed on a 1.0% agarose gel containing 0.66 M formaldehyde, transferred to a Duralon-UV membrane, and then cross-linked with a UV Stratalinker (Stratagene, La Jolla, CA). Blots were prehybridized and hybridized in a solution (1% bovine serum albumin, 7% [w/v] sodium dodecyl sulfate [SDS], 0.5 M sodium phosphate, 1 mM EDTA) at 65°C. Hybridized blots were washed in 13 SSC (0.1% SDS) hybridized with 32P-

FIG. 1. Expression of TGF-b1 mRNA and the secretion rate of TGF-b1 protein in NIH 3T3-TGF-b1 fibroblasts. TGF-b1 cDNA sequence was cloned into the polylinker site of the replication-defective retroviral vector pMTMLV, under the control of the metallothionein promoter. NIH 3T3 fibroblasts were infected with recombinant TGF-b1 retroviruses. Stable cell lines were generated by neomycin selection and then subjected to Northern analysis (Miettinen et al., 1994). (A) Northern analysis of mRNA extracted from TGF-b1-producing fibroblasts (NIH 3T3-TGF-b1) and control cells (NIH 3T3-neo). TGF-b1 transcript of NIH 3T3-TGF-b1 cells was elevated in the presence of 100 mM ZnSO4. (B) TGF-b1 protein secretion rate of NIH 3T3-TGF-b1 or NIH 3T3-neo cells was measured by ELISA in the presence or absence of ZnSO4. The average secretion rate of TGF-b1 of several stable cell lines was about 30 ng/106 cells per 24 hr.

TGF- b 1 GENE THERAPY OF CARTILAGE DAMAGE

1807

labeled cDNA probes for porcine TGF-b1. b-Actin was used as a control for the amount of sample loaded.

defect (3 mm 3 6 mm, 1–2 mm deep) was made on the hyaline cartilage layer of the femoral condyle with a surgical knife. Caution was exercised to ensure that the subchondral bone was not damaged in the process. Either control NIH 3T3-neo cells or NIH 3T3-TGF-b1 cells were injected into the rabbit knee joint with a defect. These cells (0.5 ml, containing 2 3 106 cells/ml) were either injected intraarticularly after suturing, or 15–20 ml of cells at the same concentration was loaded to the

Intraarticular injection of TGF-b1-producing cells New Zealand White rabbits weighing 2.0–2.5 kg were selected for the animal study. These rabbits were mature and had a tidemark. The knee joint was exposed and a partial cartilage

1 week

3 weeks

6 weeks

Control

FIG. 2. Regeneration of hyaline cartilage with TGF-b1-producing fibroblasts. Either NIH 3T3-TGF-b1 or control NIH 3T3neo cells were injected into rabbit knee joint with a partial cartilage defect on the femoral condyle, and then specimens were harvested 1, 3, or 6 weeks postinjection. (A) Schematic diagram showing the artificially made partial cartilage defect in the hyaline articular cartilage layer. (B) a, c, e, and g: Photographs of the femoral condyle of rabbit knee injected with either NIH 3T3-TGFb1 cells at 1 week (a), 3 weeks (c), or 6 weeks (e), or with NIH 3T3-neo control cells at 6 weeks (g) postinjection. b, d, f, and h: Overview of the defects stained with hematoxylin–eosin from the knee joint injected with NIH 3T3-TGF-b1 cells at 1 week (b), 3 weeks (d), or 6 weeks (f), or with control NIH 3T3-neo cells at 6 weeks (h) postinjection. Original magnification: (B) b, d, f, and h: 312.5.

1808 top of the defect. In the latter case, the cells were left in the defect for 15–20 min to allow the cells to permeate the wound before suturing. In both cases, no significant difference was observed in the regeneration of cartilage. This study was approved by the Institutional Animal Care and Use Subcommittee of the Inha University College of Medicine (Inchon, South Korea).

Histological staining After harvesting the knee joints, the specimens were fixed in formalin and decalcified in nitric acid. Blocks of the specimens were embedded with paraffin and they were cut into 0.8-nmthick slices. The sections were stained with hematoxylin–eosin, Safranin O, toluidine blue, and Masson’s trichrome for microscopic analyses of the regenerated cartilage.

MRI analysis Magnetic resonance imaging (MRI) of the knee was performed with a 1.5-T magnet (Signa Horizon; General Electric Medical Systems, Milwaukee, WI) and a conventional extremity coil. One sequence was used to assess the articular cartilage in the oblique–axial plane. Each sequence consisted of fast spinecho images that were performed with a long repetition time (4000 msec) a short effective-echo time (108 msec), an echo train length of 12, a bandwidth of 16 kHz, and a slice thickness of 1.5 mm with no interslice gap. Oblique–axial images were obtained with a 256 by 224 matrix and a field of view of 8 cm. The scan time was 3.5 min.

RT-PCR analysis Regenerated hyaline cartilage was removed from the femoral condyle 1, 2, 3, 4, 5, or 6 week(s) after injection of NIH 3T3TGF-b1 or control NIH 3T3-neo cells and mRNA was extracted from the cartilage tissue each time. First-strand cDNA was synthesized by using reverse transcriptase (RT) with oligo(dT) primers and then the polymerase chain reaction (PCR) was performed with an LTR-specific primer (forward, 59-TCT TTC ATT TGG GGG CTC GTC-39) and TGF-b1-specific primer

LEE ET AL. (reverse, 59-AGA GCA ATA CAG GTT CCG GCA C-39) for 35 cycles. Each cycle consists of a denaturation step at 94°C for 1 min, an annealing step at 56°C for 2 min, and an extension step at 72°C for 3 min. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as an internal control. The set of primers for GAPDH included the following: forward primer (59-AAC TCC CTC AAG ATT GTC AGC A-39) and reverse primer (59-TCC ACC ACC CTG TTG CTG TA-39).

Immunohistochemical staining Sections, prepared as described above, were deparaffinized and hydrated by sequential incubations in xylene and ethanol. After washing in 13 phosphate-bufferedsaline (PBS) for 2 min, the sections were blocked with 3% H2O2 for 10 min. The primary antibody against TGF-b1, type I collagen (Novocastra, Newcastle upon Tyne, UK), type II collagen (Southern Biotechnology, Birmingham, AL), mouse proliferating cell nuclear antigen (PCNA; Serotec, Raleigh, NC), or mouse MHC class I protein (Accurate Chemical, Westbury, NY) was applied to the sections and incubated for 1 hr. The control sections were incubated in 13 PBS without the primary antibody at this step. The sections were washed and blocked with 5% milk in 13 PBS for 20 min before incubating with the horseradish peroxidase (HRP)-conjugated secondary antibody. The chromogen reaction was performed with 0.05% diaminobenzidine (DAB) in 13 PBS for 5 min. The sections were subsequently stained with hematoxylin and mounted.

RESULTS Expression of TGF-b1 in NIH 3T3 fibroblasts To explore the possibility of cell-mediated gene therapy using fibroblasts expressing TGF-b1 as a new treatment method for degenerative arthritis, NIH 3T3 murine fibroblast cells were infected with recombinant TGF-b1 retroviruses in which the expression of active TGF-b1 was driven by the metallothionein

FIG. 3. Relative percentage of hyaline cartilage regeneration at 1, 3, or 6 week(s) after injection of TGF-b1 expressing cells. The relative percentage of hyaline cartilage regeneration 1, 3, or 6 week(s) after injection of NIH 3T3-TGF-b1 cells, and 6 weeks after injection of NIH 3T3-neo cells, was measured as the ratio of regenerated cartilage height to that of normal cartilage. The standard deviations are represented by error bars. The statistical data suggest that the regeneration of hyaline cartilage can be accomplished within 6 weeks after injection of the TGF-b1 cells.


1809

FIG. 4. Integration of regenerated cartilage into adjacent normal cartilage and PCNA staining. Regenerated hyaline cartilage was examined for integration into adjacent normal cartilage, integrity, and overgrowth. Shown are the border areas between regenerated cartilage (RC) and remaining normal cartilage (NC) 4 weeks (a, b, d, e, and f) and 6 weeks (c) after injection of NIH 3T3-TGF-b1 cells. (a–c) Hematoxylin staining. (d and e) PCNA immunohistochemistry depicting dense brown staining of the cells in regenerated cartilage, using PCNA antibody. (f) The secondary antibody was used as a control. Original magnification: (d) 320; (a, c, e, and f) 3100; (b) 3200.

FIG. 5. Regenerated hyaline cartilage. Regenerated hyaline cartilage was examined for integration into adjacent normal cartilage, integrity, and overgrowth. Shown are the border areas between regenerated cartilage (RC) and remaining normal cartilage (NC) 6 weeks after injection of NIH 3T3-TGF-b1 cells. Staining: Safranin O (a), toluidine blue (b), Masson’s trichrome (c), type I collagen antibody (d), type II collage antibody (e), and secondary antibody control (f). Original magnification: (a–f) 3100.

1810 (MT) promoter. Several stable cell lines (NIH 3T3-TGF-b1) were generated by G418 selection and subsequent Northern analysis. An inducible promoter was chosen because a high level of constitutive expression of TGF-b1 might inhibit host cell growth. Most of the selected cell lines expressed the TGFb1 mRNA without zinc induction, but the level of transgene expression was markedly elevated when they were cultured in 100 mM zinc sulfate solution (Fig. 1). These TGF-b1-producing cells maintained a high level of TGF-b1 expression for at least a few months in vitro. The average rate of secretion of TGF-b1, measured by enzyme-linked immunosorbent assay (ELISA), from these stable cell lines was about 30 ng/106 cells per 24 hr.

Regeneration of hyaline cartilage with TGF-b1producing fibroblasts To simulate degenerative arthritis of the knee joint, a partial cartilage defect (3 mm 3 6 mm, 1–2 mm deep) was made with a surgical knife in the hyaline cartilage layer of the femoral condyle of New Zealand White rabbits (Fig. 2A). NIH 3T3TGF-b1 cells (5 3 105 to 2 3 106) were subsequently injected into the site of the defect. The same amount of NIH 3T3-neo cells containing the expression vector without the TGF-b1 gene was injected into the wounded contralateral knee joint to serve as a control. Cells were either injected into the knee joint immediately after suturing or loaded into the defective region 15–20 min before suturing. The level of cartilage regeneration did not differ significantly with these two approaches in several of the rabbits tested (data not shown). The regeneration of hyaline cartilage was not obvious 1 week after injection of NIH 3T3-TGF-b1 cells (Fig. 2B, a and b). At 3 weeks postinjection, most of the defect sites were filled with regenerated tissue (Fig. 2B, c and d) and at 6 weeks, the regeneration of hyaline cartilage in the defect was complete (Fig. 2B, e and f). In contrast, no new hyaline cartilage was formed at the site of the cartilage defect of the contralateral knee injected with control NIH 3T3-neo fibroblasts at 6 weeks postinjection (Fig. 2B, g and h). Regeneration of hyaline cartilage was not affected by the addition of zinc in the drinking water even though addition of zinc resulted in a marked increase in TGFb1 expression in vitro (data not shown). The relative degrees

LEE ET AL. of hyaline cartilage regeneration 1, 3, and 6 week(s) after injection of NIH 3T3-TGF-b1 cells were determined by measuring the percentage of regenerated cartilage height relative to that of normal cartilage. The average percentages of cartilage regeneration 1, 3, and 6 week(s) after injection were 24, 69, and 98%, respectively (Fig. 3). These statistical data also suggest that the regeneration of hyaline cartilage in the defect can be accomplished within 6 weeks after injection of the TGF-b1producing cells. Overall, histological examinations demonstrated the regeneration of hyaline cartilage in 80% of NIH 3T3TGF-b1 cell-treated animals at 3 to 6 weeks postinjection (n 5 40).

Histological examination of regenerated cartilage To examine integration of newly generated cartilage with the adjacent normal cartilage, and its integrity and possibility of overgrowth, histological analyses with high-magnification, cartilage-specific stainings, and magnetic resonance imaging (MRI) observation, were performed. Four weeks after injection of NIH 3T3-TGF-b1 cells, microscopic examination of regenerated hyaline cartilage revealed that the cartilage was well integrated with the adjacent normal cartilage, although the architecture of the regenerated tissue was not fully developed into that of normal cartilage at this stage (Fig. 4a and b). PCNApositive cells were markedly increased in regenerated cartilage (Fig. 4d and e). At 6 weeks, staining of the regenerated cartilage with hematoxylin (Fig. 4c), Safranin O, toluidine blue, Masson’s trichrome, type I collagen antibody, and type II collagen antibody revealed that chondrocytes in the regenerated tissue were fully differentiated (Fig. 5a–e). The morphology and staining pattern of the regenerated tissue were the same as those of the adjacent normal hyaline cartilage, indicating that the newly generated tissue was indeed hyaline cartilage. These results suggest that the resulting repair of the defect is almost identical to the natural processes of hyaline cartilage formation. Histological examination of adjacent synovium, vascular channels, subchondral blood vessels, and bone showed no detectable alterations (data not shown). To evaluate the possibility of overgrowth of the regenerated hyaline cartilage in vivo, MRI examination of the knee joint was performed over 12 weeks after injection. MRI examination of the knee 6, 8, 10, and 12

FIG. 6. MRI observation of knee joint. (a and c) MRI pictures of the knee joints taken 12 weeks after injection of either NIH 3T3-TGF-b1 cells (a) or NIH 3T3-neo cells (c). (b and d) Photographs of the femoral condyle of rabbit knee injected with either NIH 3T3-TGF-b1 cells (b) or NIH 3T3-neo cells (d) 12 weeks postinjection. FIG. 7. RT-PCR analysis and immunohistochemical staining of regenerated hyaline cartilage. (A) A schematic diagram showing a part of the pMTMLVb1 vector. The expected RT-PCR products and their estimated sizes from unspliced and spliced transcript are indicated. (B) Regenerated hyaline cartilage was removed from the femoral condyle 1, 2, 3, 4, 5, or 6 week(s) after injection of NIH 3T3-TGF-b1 cells and mRNA was extracted from each cartilage tissue. RT-PCR was then performed with a set of transgene-specific primers. Lanes: (1) normal hyaline cartilage; (2–7) regenerated hyaline cartilage 1, 2, 3, 4, 5, or 6 week(s) after injection. GAPDH was used as an internal control for RT-PCR. (C) Overview of the regenerated hyaline cartilage stained with TGF-b1 antibody 3 weeks after injection of the cells. (D) High magnification of the border area between regenerated cartilage (RC) and adjacent normal cartilage (NC). The result showed a high level of TGF-b1 protein expression only in cells of the regenerated hyaline cartilage. (E) No TGF-b1 staining was observed in a section of the same tissue probed with the secondary antibody alone. (F) Regenerated cartilage tissue stained with mouse MHC class I antibody 3 weeks after injection of the cells. The expression of mouse MHC class I protein is strongly detected only in cells of the regenerated cartilage (RC). (G) No MHC class I staining was observed with the secondary antibody alone. (H) Mouse MHC class I protein was not detected in cells of the regenerated cartilage 6 weeks postinjection. Original magnification: (C) 320; (D–G) 3100; (H) 340.

1811

TGF- b 1 GENE THERAPY OF CARTILAGE DAMAGE weeks after injection of the NIH 3T3-TGF-b1 fibroblasts showed no signs of overgrowth. MRI pictures taken 12 weeks after injection of NIH 3T3-TGF-b1 or NIH 3T3-neo cells are shown in Fig. 6a and c, respectively. The same knee joints were also examined after sacrifice of the animals and confirmed the

results of MRI observation (Fig. 6b and d). The results of histological studies of the regenerated cartilage 12 weeks after injection indicated that the morphology and staining pattern were similar to those of the repaired cartilage at 6 weeks (data not shown).

FIG. 6

FIG. 7

1812

LEE ET AL.

Expression of exogenous TGF-b1 in vivo after injection To investigate the expression of exogenous TGF-b1 transcript after injection of the TGF-b1-producing fibroblasts, RT-PCR was performed with mRNA extracted from the regenerated cartilage tissues up to 6 weeks postinjection. To detect exogenous TGF-b1 transcript only, a set of primers was derived from the retroviral LTR sequence and 59 end of the TGF-b1 gene of pMTMLVb1. Since a splicing event of the full-length retroviral transcript takes place at the splicing donor (SD) and splicing acceptor (SA) sites, located between the LTR and TGF-b1, the primer set generates two RT-PCR products (Fig. 7A). One is from the unspliced transcript and the other is from the spliced transcript. The result showed that the expression of transgenic TGF-b1 mRNA peaked at 2 weeks and continued to 4 weeks after injection of the cells (Fig. 7B). Immunohistochemical staining of regenerated hyaline cartilage performed with TGF-b1 antibody 3 weeks postinjection showed that a high level of TGF-b1 protein expression occurred only in the cells of the regenerated cartilage (Fig. 7C and D). No antibody staining was observed in a section of the same tissue probed with the secondary antibody alone (Fig. 7E). This immunohistochemical staining result correlates well with the RTPCR data. These RT-PCR and immunohistochemical staining results suggest that the long-term expression of transgenic TGFb1 mRNA can be obtained with the current cell-mediated TGFb1 gene therapy approach. This would prove useful in developing a gene therapy protocol that requires long-term production of a protein.

Fate of injected TGF-b1-producing fibroblasts To investigate the fate of the injected NIH 3T3-TGF-b1 cells after injection, immunohistochemical staining of the regenerated cartilage with mouse-specific MHC class I antibody was performed. Only the cells in the regenerated tissue were stained with the mouse-specific MHC class I antibody. The staining result at 3 weeks postinjection is shown in Fig. 7F. The regenerated tissue was stained with the MHC class I antibody up to 4 weeks after injection and the positive MHC class I staining was not detected in the regenerated cartilage at 6 weeks postinjection (Fig. 7H).

DISCUSSION The results of this study suggest that cell-mediated TGF-b1 gene therapy may be a novel treatment for orthopedic diseases in which hyaline cartilage damage has occurred. Here it was demonstrated that fibroblasts containing an exogenous TGF-b1 gene are able to express the transgene for at least 4 weeks after injection in vivo. Furthermore, in a partial cartilage defect model, the TGF-b1 protein produced from the cells can induce a full regeneration of the hyaline cartilage tissue in 6 weeks. The result of RT-PCR analysis of the regenerated tissues showed that the expression of exogenous TGF-b1 transcripts peaked at 2 weeks postinjection and lasted up to 4 weeks. This finding suggests that active TGF-b1 proteins secreted from the injected cells can be available for more than 4 weeks to help

repair the cartilage defect in vivo. This sustained expression of active TGF-b1 protein is a key reason for such rapid regeneration of the cartilage shown in the present study. Sellers et al. (2000) demonstrated that a collagen sponge impregnated with 5 mg of recombinant human BMP-2 lasted for 2 weeks after implantation in vivo. Therefore, it would be difficult for these cytokines to be present for a sustained period of time in vivo without a repeated injection or implantation of purified proteins. The results of immunohistochemical staining of the tissue repaired with the TGF-b1 antibody correlated well with the RT-PCR data. The level of TGF-b1 protein expression was highest 3 weeks after injection (Fig. 5c and d). It became weaker at 4 weeks, and was absent 6 weeks after injection (data not shown). The results of different cartilage-specific histological stainings demonstrated integration of the regenerated cartilage tissue with the normal adjacent cartilage even 3 to 4 weeks after injection of the TGF-b1-producing cells. Previous studies have suggested that integration of the repaired tissue was difficult since chondrocytes in the normal cartilage were not thought to be involved in repair (DePalma et al., 1966; Shapiro et al., 1993). Our results indicated that TGF-b1 produced from the injected cells stimulated the chondrocytes and/or chondrocyte precursors in the remaining normal cartilage to proliferate and/or differentiate into normal chondrocytes. These results also showed that the overall architecture and morphology of the chondrocytes in the newly generated cartilage were similar to those of normal chondrocytes (Figs. 4 and 5). Despite the histological staining data, the ability of the regenerated cartilage to withstand the mechanical forces within the joint still remains to be investigated. The details of the molecular mechanism that controls the regeneration of hyaline cartilage with the present cell-mediated TGF-b1 gene therapy method are still not clear. However, the most likely mechanism is that the TGF-b1 secreted from the injected fibroblasts induces the proliferation and differentiation of the existing chondrocytes and/or chondrocyte precursors residing in the wound base. Immunohistochemical staining of the regenerated cartilage tissue with mouse MHC class I antibodies, shown in Fig. 7F, suggests that the injected fibroblast cells are gradually replaced by growth of the underlying rabbit cartilage. Degenerative arthritis is the most frequently encountered disease with cartilage damage among the orthopedic diseases. The efficacy of the cytokines, including TGF-b, in inducing regeneration of the hyaline articular cartilage has been demonstrated previously. However, the clinical application of these cytokines has been limited because of its short half-life. The results of this study strongly support the feasibility of a cell-mediated gene therapy approach for achieving the long-term production of TGF-b1, using fibroblasts secreting TGF-b1 for cartilage repair. A cell-mediated cytokine gene therapy approach for cartilage regeneration does not require an operation or a lengthy rehabilitation time, whereas the autologous cell-based therapy (Brittberg et al., 1994) and cytokine protein-based therapy involve at least one operation and a substantial rehabilitation time (Sellers et al., 1997, 2000). Therefore, cell-mediated TGF-b1 gene therapy may be a clinically useful and efficient therapy for treating orthopedic diseases with hyaline cartilage damage.

1813


ACKNOWLEDGMENTS We thank Anita B. Roberts, W. Tony Parks, and Isaac Kim for critical reading of the manuscript and helpful suggestions. These studies were partially supported by grant from the South Korean government (HMP-98-M-5-0055) and TissueGene Inc. (Gaithersburg, MD).

REFERENCES BRITTBERG, M., LINDAHL, A., NILSSON, A., OHLSSON, C., ISAKSSON, O., and PETERSON, L. (1994). Treatment of deep cartilage defects in the knee with autologous chondrocyte transplantation. N. Engl. J. Med. 331, 889–895. CHENU, C., PFEILSCHIFTER, J., MUNDY, G.R., and ROODMAN, G.D. (1988). Transforming growth factor-b inhibits formation of osteoclast-like cells in long-term human marrow cultures. Proc. Natl. Acad. Sci. U.S.A. 85, 5683–5687. CRITCHLOW, M.A., BLAND, Y.S., and ASHHURST, D.E. (1995). The effect of exogenous transforming growth factor-b2 on healing fractures in the rabbit. Bone 16, 521–527. DEPALMA, A.F., MCKEEVER, C.D., and SUBIN, D.K. (1966). Process of repair of articular cartilage demonstrated by histology and autoradiography with tritiated thymidine. Clin. Orthop. 48, 229–242. GLANSBEEK, H.L., VAN BEUNINGEN, H.M., VITTERS, E.L., VAN DER KRAAN, P.M., and VAN DEN BERG, W.B. (1998). Stimulation of articular cartilage repair in established arthritis by administration of transforming growth factor-b into murine joints. Lab Invest. 78, 133–142. JOYCE, M.E., ROBERTS, A.B., SPORN, M.B., and BOLANDER, M.E. (1990). Transforming growth factor-b and the initiation of chondrogenesis and osteogenesis in the rat femur. J. Cell Biol. 110, 2195–2207. KIM, S.H., YU, S.S., PARK, J.S., ROBBINS, P.D., AN, C.S., and KIM, S. (1998). Construction of retroviral vectors with improved safety, gene expression, and versatility. J. Virol. 72, 994–1004. LIND, M., SCHUMACKER, B., SOVALLE, K., KELLER, J., MELSEN, F. and BUNGER, C. (1993). Transforming growth factor-b enhances fracture healing in rabbit tibiae. Acta Orthop. Scand. 64, 553–556. LOUWERSE, R.T., HEYLEGER, I.C., KLEIN-NULEND, J., SUGIHARA, S., KAMPEN, G.P., SEMEINS, C.M., GOEI, S.W., DE KONING, M.H., WUISMAN, P.I., and BURGER, E.H. (2000). Use of recombinant human osteogenic protein-1 for the repair of subchondral defects in articular cartilage in goats. J. Biomed. Mater. Res. 49, 506–516. MANKIN, H.J. (1982). The response of articular cartilage to mechanical injury. J. Bone Joint Surg. 64, 460–466. MATSUMOTO, K., MATSUNAGA, S., IMAMURA, T., ISHIDOU, Y., and DERYNCK, R. (1994). Expression and distribution of transforming growth factor-b and decorin during fracture healing. In Vivo 8, 215–220. MIETTINEN, P.J., EBNER, R., LOPEZ, A.R., and DERYNCK, R. (1994). TGF-b induced transdifferentiation of mammary epithelial cells to mesenchymal cells: Involvement of type I receptors. J. Cell Biol. 127, 2021–2036.

NEIDEL, J.J. (1992). Keine Verbesserung der Gelenkknorpelheilung nach Trauma durch intraarticulare Gabe von insulinartigem Wachstumsfaktor I, epidermalem Wachstumsfaktor und FibroblastenWachstumsfaktor beim Kaninchen. Z. Orthop. Ihre Grenzgeb. 130, 73–78. RIPAMONTI, U., DUNEAS, N., VAN DEN HEEVER, B., BOSHC, C., and CROOKS, J. (1997). Recombinant transforming growth factorb1 induces endochondral bone in the baboon and synergizes with recombinant osteogenic protein-1 (bone morphogenetic protein-7) to initiate rapid bone formation. J. Bone Miner. Res. 12, 1584–1595. SELLERS, R.S., PELUSO, D., and MORRIS, E.A. (1997). The effect of recombinant human bone morphogenetic protein-2 (rhBMP-2) on the healing of full-thickness defects of articular cartilage. J. Bone Joint Surg. 79, 1452–1463. SELLERS, R.S., ZHANG, R., GLASSON, S.S., KIM, H.D., PELUSO, D., D’AUGUSTA D.A., BECKWITH, K., and MORRIS, E.A. (2000). Repair of articular cartilage defects one year after treatment with recombinant human bone morphogenetic protein-2 (rhBMP-2). J. Bone Joint Surg. 82, 151–160. SHAPIRO, F., KOIDE, S., and GLIMCHER, M.J. (1993). Cell origin and differentiation in the repair of full-thickness defects of articular cartilage. J. Bone Joint Surg. 75, 532–553. SPORN, M.B., and ROBERTS, A.B. (1988). Peptide growth factors are multifunctional. Nature 332, 217–219. VAN BEUNINGEN, H.M., VAN DER KRAAN, P.M., ARNTZ, O.J., and VAN DER BERG, W.B. (1994). Transforming growth factor-b1 stimulates articular chondrocyte proteoglycan synthesis and induces osteophyte formation in the murine knee joint. Lab. Invest. 71, 279–290. VAN OSCH, G.J., VAN DER VEEN, S.W., BUMA, P., and VERWOERD-VERHOEF, H.L. (1998). Effect of transforming growth factor-b on proteoglycan synthesis by chondrocytes in relation to differentiation stage and the presence of pericellular matrix. Matrix Biol. 17, 413–424. YONEKURA, A., OSAKI, M., HIROTA, Y., TSUKAZAKI, T., MIYAZAKI, Y., MATSUMOTO, T., OHTSURU, A., NAMBA, H., SSHINDO, H., and YANASHITA, S. (1999). Transforming growth factor-b stimulates articular chondrocyte cell growth through p44/42 MAP kinase (ERK) activation. Endocr. J. 46, 545–553. ZEGZULA, H.D., BUCK, D.C., BREKKE, J., WOZNEY, J.M., and HOLLINGER, J.O. (1997). Bone formation with use of rhBMP-2 (recombinant human bone morphogenetic protein-2). J. Bone Joint Surg. 79, 1778–1790.

Address reprint requests to: Dr. Kwan Hee Lee Clinical Research Center College of Medicine, Inha University 7-206, 3-Ga, Shinheung-Dong, Chung-Gu Inchon, South Korea 400-711 E-mail: [email protected]. Received for publication April 19, 2001; accepted after revision August 14, 2001. Published online: August 29, 2001.

This article has been cited by: 1. Onur Bilge, Mahmut Nedim Doral, Kivanc Atesok, Ozgur A. Atay, Gurhan Donmez, Egemen Turhan, Akin Uzumcugil, Gursel Leblebicioglu, Defne Kaya, Hasan Bilgili, Mustafa Sargon. 2011. The effects of the synovium on chondrocyte growth: an experimental study. Knee Surgery, Sports Traumatology, Arthroscopy . [CrossRef] 2. Ericka M. Bueno, Julie Glowacki. 2011. Biologic Foundations for Skeletal Tissue Engineering. Synthesis Lectures on Tissue Engineering 3:1, 1-220. [CrossRef] 3. A Ivkovic, A Pascher, D Hudetz, D Maticic, M Jelic, S Dickinson, M Loparic, M Haspl, R Windhager, M Pecina. 2010. Articular cartilage repair by genetically modified bone marrow aspirate in sheep. Gene Therapy 17:6, 779-789. [CrossRef] 4. Hideyuki Koga, Lars Engebretsen, Jan E. Brinchmann, Takeshi Muneta, Ichiro Sekiya. 2009. Mesenchymal stem cell-based therapy for cartilage repair: a review. Knee Surgery, Sports Traumatology, Arthroscopy 17:11, 1289-1297. [CrossRef] 5. Youngsuk Yi , Kyoung Baek Choi , Chae-Lyul Lim , Jong-Pil Hyun , Hyeon-Youl Lee , Kun Bok Lee , Lillian Yun , Asli Ayverdi , Sally Hwang , Vivian Yip , Moon Jong Noh , Kwan Hee Lee . 2009. Irradiated Human Chondrocytes Expressing Bone Morphogenetic Protein 2 Promote Healing of Osteoporotic Bone Fracture in RatsIrradiated Human Chondrocytes Expressing Bone Morphogenetic Protein 2 Promote Healing of Osteoporotic Bone Fracture in Rats. Tissue Engineering Part A 15:10, 2853-2863. [Abstract] [Full Text] [PDF] [PDF Plus] 6. Christopher H Evans, Steven C Ghivizzani, Paul D Robbins. 2009. Orthopedic Gene Therapy in 2008. Molecular Therapy 17:2, 231-244. [CrossRef] 7. S Trippel, M Cucchiarini, H Madry, S Shi, C Wang. 2007. Gene therapy for articular cartilage repair. Proceedings of the Institution of Mechanical Engineers, Part H: Journal of Engineering in Medicine 221:5, 451-459. [CrossRef] 8. Dean D. Ad-El, Jonah Selah, Gal Goshen, Izhac Dano. 2006. Induction of cartilage growth in a rabbit ear model: a pilot study. European Journal of Plastic Surgery 28:8, 513-516. [CrossRef] 9. Jung Park, Kolja Gelse, Susi Frank, Klaus von der Mark, Thomas Aigner, Holm Schneider. 2006. Transgene-activated mesenchymal cells for articular cartilage repair: a comparison of primary bone marrow-, perichondrium/periosteum- and fat-derived cells. The Journal of Gene Medicine 8:1, 112-125. [CrossRef] 10. Magali Cucchiarini, Henning Madry. 2005. Gene therapy for cartilage defects. The Journal of Gene Medicine 7:12, 1495-1509. [CrossRef] 11. Sun U. Song , Young-Deog Cha , Jeoung-Uk Han , In-Suk Oh , Kyoung Baek Choi , Youngsuk Yi , Jong-Pil Hyun , Hyeon-Youl Lee , Guang Fan Chi , Chae-Lyul Lim , J. Kelly Ganjei , Moon-Jong Noh , Seong-Jin Kim , Dug Keun Lee , Kwan Hee Lee . 2005. Hyaline Cartilage Regeneration Using Mixed Human Chondrocytes and Transforming Growth Factor-β1- Producing ChondrocytesHyaline Cartilage Regeneration Using Mixed Human Chondrocytes and Transforming Growth Factor-β1- Producing Chondrocytes. Tissue Engineering 11:9-10, 1516-1526. [Abstract] [PDF] [PDF Plus] 12. S. Vogt, P. Ueblacker, V. Martinek, B. Gänsbacher, A. B. Imhoff. 2005. Gentherapeutische Möglichkeiten in der Knorpeltherapie. Arthroskopie 18:3, 239-244. [CrossRef] 13. Hyun J. Paek, Anelisa B. Campaner, Jennifer L. Kim, Roy K. Aaron, Deborah M. Ciombor, Jeffrey R. Morgan, Michael J. Lysaght. 2005. In vitro Characterization of TGF-??1 Release from Genetically Modified Fibroblasts in Ca2+-Alginate Microcapsules. ASAIO Journal 51:4, 379-384. [CrossRef] 14. Dug Keun Lee , Kyoung Baek Choi , In Suk Oh , Sun U. Song , Sally Hwang , Chae-Lyul Lim , Jong-Pil Hyun , Hyeon-Youl Lee , Guang Fan Chi , Youngsuk Yi , Vivian Yip , Jeannie Kim , Eun Byul Lee , Moon Jong Noh , Kwan Hee Lee . 2005. Continuous Transforming Growth Factor β1 Secretion by Cell-Mediated Gene Therapy Maintains Chondrocyte RedifferentiationContinuous Transforming Growth

Factor β1 Secretion by Cell-Mediated Gene Therapy Maintains Chondrocyte Redifferentiation. Tissue Engineering 11:1-2, 310-318. [Abstract] [PDF] [PDF Plus] 15. Fei-Zhou Lu, Yusuke Kitazawa, Yuko Hara, Jian-Yuan Jiang, Xiao-Kang Li. 2005. Long-term Gene Expression Using the Lentiviral Vector in Rat Chondrocytes. Clinical Orthopaedics and Related Research 439:&na;, 243. [CrossRef] 16. Sun U. Song , Young-Jin Hong , In-Suk Oh , Youngsuk Yi , Kyoung Baek Choi , Jung Woo Lee , Kwang-Won Park , Jeoung-Uk Han , Jun-Kyu Suh , Kwan Hee Lee . 2004. Regeneration of Hyaline Articular Cartilage with Irradiated Transforming Growth Factor β1-Producing FibroblastsRegeneration of Hyaline Articular Cartilage with Irradiated Transforming Growth Factor β1-Producing Fibroblasts. Tissue Engineering 10:5-6, 665-672. [Abstract] [PDF] [PDF Plus] 17. Kolja Gelse, Klaus von der Mark, Thomas Aigner, Jung Park, Holm Schneider. 2003. Articular cartilage repair by gene therapy using growth factor-producing mesenchymal cells. Arthritis & Rheumatism 48:2, 430-441. [CrossRef] 18. MIKEL S??NCHEZ, JUAN AZOFRA, EDUARDO ANITUA, ISABEL AND??A, SABINO PADILLA, JUANMA SANTISTEBAN, I??IGO MUJIKA. 2003. Plasma Rich in Growth Factors to Treat an Articular Cartilage Avulsion: A Case Report. Medicine & Science in Sports & Exercise 35:10, 1648. [CrossRef]