T

JP

Rotsccttattvaprpir

RtppetmtstpMh

icd

a

R

C

1d

2

issue engineering and rotator cuff tendon healing

oshua S. Dines, MD,a Daniel A. Grande, PhD,b and David M. Dines, MD,b Los Angeles, CA, and New Hyde

ark, NY

ftvaprpir

svdf(chfibPdsdicscpt

bpstehb

otator cuff tears are common soft-tissue injuries thatften require surgical treatment. Initial efforts to bet-

er tendon healing centered on improving thetrength of the repair. More recent studies have fo-used on biologic enhancement of the healing pro-ess. Tissue engineering is a multidisciplinary fieldhat involves the application of scientific principlesoward creating living tissue to replace, repair, orugment diseased tissue. Gene therapy involves the

ransfer of a certain gene into a cell so that the cellranslates the gene into a specific protein. The ad-antage of using a gene-therapy, tissue-engineeredpproach to effect healing rests in the ability of thehysician to select growth factors with documentedoles in the tendon-healing cascade. Ideally, an im-rovement to the current repair technique would yield

mproved tendon healing leading to improved clinicalesults. (J Shoulder Elbow Surg 2007;16:204S-207S.)

otator cuff tears are common soft-tissue injurieshat often require surgical treatment. Operative re-air of these tendon tears significantly improvesain and function.9,12,18 Previous studies, how-ver, have demonstrated that up to 50% of theseears fail to heal when evaluated by ultrasound oragnetic resonance imaging.8,10,16 Initial efforts

o better tendon healing centered on improving thetrength of the repair by using stronger suture ma-erials and knots and restoring the anatomic foot-rint of the rotator cuff through double-row repairs.ore recent studies have focused on biologic en-

ancement of the healing process.4,6,7,17,20

Tissue engineering is a multidisciplinary field thatnvolves the application of scientific principles towardreating living tissue to replace, repair, or augmentiseased tissue.7,13 Gene therapy involves the trans-

From the Kerlan Jobe Orthopaedic Foundation, and the bDe-partment of Orthopaedic Surgery, Long Island Jewish MedicalCenter.

eprint requests: Joshua S. Dines, MD, Kerlan Jobe OrthopaedicFoundation, 6801 Park Terrace, Los Angeles, CA 90045.(E-mail: jdinesmd@gmail.com).opyright © 2007 by Journal of Shoulder and Elbow SurgeryBoard of Trustees.

058-2746/2007/$32.00

oi:10.1016/j.jse.2007.03.004

04S

er of a certain gene into a cell so that the cellranslates the gene into a specific protein.8 The ad-antage of using a gene-therapy, tissue-engineeredpproach to effect healing rests in the ability of thehysician to select growth factors with documentedoles in the tendon-healing cascade. Ideally, an im-rovement to the current repair technique would yield

mproved tendon healing leading to improved clinicalesults.

A number of different growth factors have beentudied for their effects on tendon cells, includingascular endothelial growth factor (VEGF), growthifferentiation factor-5, platelet derived growthactor-� (PDGF-�), and insulin-like growth factor-1IFG-1).1–3,5,11,14,15,21 PDGF-� mediates many pro-esses responsible for accelerating and enhancingealing tissues, including chemotaxis, proliferation ofbroblasts, induction of the extracellular matrix (fi-ronectin), and revascularization. In several studies,DGF-� stimulated DNA and matrix synthesis in ten-on cells,11,14 and it increased the expression of cellurface integrins, which play critical roles in the ten-on repair process.22 IGF-1 also enhances the heal-

ng process by increasing DNA, collagen, and gly-osaminoglycan production. In vitro and in vivotudies have elucidated the ability of IGF-1 to de-rease swelling, while at the same time increase cellroliferation, collagen synthesis, and DNA con-

ent.2,3

If these growth factors could be delivered locallyy gene therapy, it is likely that the tendon repairrocess would be improved. On the basis of thistatement, we have reported our preliminary workhat will serve as our template for producing a tissue-ngineered scaffold that could augment that naturalealing process.6,20 These previous studies wereased on the following hypotheses:

1. Tendon cells could be transduced with specificgenes.

2. These gene-modified cells would demonstratethe capacity to modulate a local environmentin an in vitro setting.

3. The gene-modified cells could be seeded ontoa scaffold that would serve as a bioactivepatch to be incorporated into tendon repairs.

4. This patch would accelerate tendon healingand improve the quality of repair tissue in a

small-animal model.

TPsdv

ododr(iRvffc

mvtcflwswfduwt

dafiat

ompvsw

musfcc(cmi

tc

hsRlnRhsi(ss3biI3ti

fitpatvSaspi

Fg

J Shoulder Elbow Surg Dines, Grande, and Dines 205SVolume 16, Number 5S

he growth factor genes used in this application,DGF- � and IGF-1, were chosen based on previoustudies that suggested a role for their use in geneelivery strategies. To test hypotheses 1, 2, and 3, initro experiments were designed as described below.

Adult Sprague-Dawley rats were used for isolationf tendon fibroblasts (RTFs) from the rotator cuff ten-ons. The RTFs were initiated in culture by explantutgrowth. They were then serially cultured and trans-uced with the genes for PDGF- � or IGF-1 by aetroviral vector. Real-time polymerase chain reactionRT-PCR) reverse transcription systems were used tosolate the PDGF-� complementary DNA (cDNA) fromNA that had been isolated from human umbilicalein endothelial cells and to isolate the IGF-1 cDNArom a human embryonic lung cell line. The isolatedragments were sequenced to verify that they wereorrect.

LNB-PDGF and LNB-IGF-1 retroviral vector plas-ids were constructed and used to generate retroviralector particles. To transduce the cells with the genes,he primary RTFs were plated into well dishes andultured. When the cells reached 25% to 50% con-uence, transductions were performed. Transductionas confirmed using messenger RNA (mRNA) analy-

is by RT-PCR. To detect the genes, specific primersere used in the PCR to generate fragments specific

or PDGF-� or IGF-1. Fragments specific for glyceral-ehyde phosphate dehydrogenase (GAPDH) weresed as controls. The cells containing the active genesere selected by incorporation of a neomycin resis-

ance gene added to the construct.RTF cells maintained their spindle shape in culture

espite the gene transfection. Northern blot analysisnd enzyme-linked immunosorbent assay (ELISA) con-rmed positive gene expression. We then tested thebility of IGF-1 and PDGF-� gene-transfected RTF cells

o modulate the metabolism of surrounding tissue.RTF cells and gene-modified RTFs were seeded

nto a bioabsorbable polymer scaffold that wasade of nonwoven, polyglycolic acid (PGA). It hadore size 200 �m, which corresponded to a 99%oid volume. RTF cells rapidly attached to the polymercaffolds and formed highly cellular tissue constructsithin the PGA scaffolds by 5 days after seeding.To test the ability of the cell–scaffold construct to

odulate a local environment, a transwell system wassed. Two constructs were placed in the systemeparated by a membrane that allowed solubleactors to diffuse across—but not directly touch—ells in the adjoining compartment (Figure 1). Threeonfigurations were tested: (1) single constructRTF/0; control 1), (2) nontransduced RTF (RTF/RTF;ontrol 2), and (3) RTF � IGF or PDGF/RTF (experi-ental 1, 2). The constructs were allowed to incubate

n culture and then were pulse-labeled with tritiated

hymidine and proline to assess cell replication andollagen synthesis, respectively.

Results of the in vitro testing phase confirmed ourypothesis that the gene-modified cells would demon-trate the capacity to modulate a local environment.TF constructs incubated alone exhibited a baseline

evel of collagen synthesis (control 1), and this wasot significantly stimulated by placement of a similarTF construct adjacently (control 2). As early as 24ours after incubation, the PDGF-� transduced cellstimulated RTF cells in their adjacent compartment toncrease collagen synthesis significantly by 3-foldP � .05). At 24 hours, DNA synthesis was notignificantly increased. IGF-1 stimulated collagenynthesis in adjacent RTF cells by approximately0% (P � .05) and DNA synthesis by almost 100%y 24 hours of exposure (P � .05). At 48 hours,

ncreased collagen synthesis in both the PDGF-� andGF-1 groups continued, the PDGF- � group showed a-fold increase in DNA synthesis compared with con-

rols (P � .05), and IGF-1 maintained a 28% increasen DNA synthesis.

On the basis of our ability to engineer tendonbroblasts to successfully deliver therapeutic peptideso a local environment and on the ability of the generoducts to stimulate tendon cell metabolism, we usedpreviously validated rat model of rotator cuff repair

o test the tissue-engineered, gene therapy construct inivo.19 Rotator cuff tears were created in 48 adultprague-Dawley rats. The animals were allowed tombulate for 2 weeks, thus mimicking the clinicaletting of a chronic tear. A second surgery was thenerformed to repair the tears. Animals were random-

zed to 1 of 4 groups of 12 rats each:

igure 1 Diagram of the transwell system. PDGF, Platelet-derivedrowth factor.

1. Suture repair alone (control 1),

Awcihwt

e.gaaTcso

ptasTtsccw

hSet

bhApsgd

R

Fr�

Fbo

Fgld

206S Dines, Grande, and Dines J Shoulder Elbow SurgSeptember/October 2007

2. Suture repair � acellular PGA scaffold (con-trol 2),

3. Suture repair � PDGF-� transduced scaffold(experimental 1), and

4. Suture repair � IGF-1 transduced scaffold (ex-perimental 2).

t 6-weeks postoperatively, the repaired tendonsere harvested and tested for histology and biome-hanics. From each group, 9 animals were random-zed to biomechanical testing, and 3 were evaluatedistologically. Power studies showed that an n � 8ould provide a confidence level of 0.05 between

he different groups.Histology scores were significantly better in both

xperimental groups compared with controls (P �05). A representative photomicrograph of the controlroup 1 tendon is shown in Figure 2. As was seen inll control group slides, the repair site was relativelycellular, and there was a lack of significant repair.his contrasts with Figure 3, which shows a photomi-rograph of experimental group 1 tendon. The repairite is very cellular and shows good repair with well-rganized collagen bundles.

The ability of the tissue-engineered scaffold to im-rove the biomechanical characteristics of repaired

endons was also encouraging (Figure 4). Toughnessnd maximum load to failure were both statisticallyignificantly improved in the IGF-1 repairs (P � .05).he PDGF-� group did have slightly better results thanhe control groups; however, the differences were nottatistically significant. The differences between the 2ontrol groups were not statistically significant, indi-ating that PGA scaffold alone did not improve ororsen tendon healing.The results of this study indicated that IGF-1 en-

anced rotator cuff tendon healing in a rat model.tudies are presently underway in larger-animal mod-ls. The ultimate goal is to use the principles of

igure 2 Histologic specimen representative of control groupepairs. (Hematoxylin and eosin stain; original magnification100.)

issue engineering and gene therapy to develop a

ioactive patch capable of accelerating tendonealing and improving the quality of repair tissue.s our understanding of the healing cascade im-roves, this patch will be a customizable, off-the-helf product with different gene products inte-rated into the patch to work at different time pointsuring the healing process.

EFERENCES

1. Aspenburg P, Forslund C. Enhanced tendon healing with GDF5and 6. Acta Orthop Scand 1999;70:51-54.

2. Dahlgren LA, van der Meulen MC, Bertram JE, Starrak GS, NixonAJ. Insulin-like growth factor-1 improves cellular and molecularaspects of healing in a collagenase-induced model of flexortendonitis. J Orthop Res 2002;20:910-9.

3. Dahlgren LA, Nixon AJ, Brower-Toland BD. Effects of �-aminopropionitrile on equine metabolism in vitro and on ef-fects of insulin-like growth factor-1 on matrix production byequine tenocytes. Am J Veterinary Res 2001;62:1557-62.

4. DeFranco MJ, Derwin K, Iannotti JP. New therapies in tendonreconstruction. J Am Acad Orthop Surg 2004;12:298-304.

5. Desmouliere A, Geinoz A, Gabbiani F, Gabbiani G. Transform-

igure 3 Histologic specimen representative of repairs enhancedy platelet-derived growth factor. (Hematoxylin and eosin stain;riginal magnification �100.)

Max Load (N)

0

0.51

1.52

2.5

33.5

4

PGA + IGF-1

PGA +PDGF

Vicryl Vicryl +PGA

Newtons

Max Load (N)

igure 4 Maximum load to failure of rotator cuff tendons in allroups. Results are presented with the standard error. PGA, Polyg-

ycolic acid; IGF-1, insulin-like growth factor-1; PDGF, platelet-erived growth factor. (Vicryl, Ethicon, Somerville, NJ.)

ing growth factor �1 induces �1 smooth muscle actin expression

1

1

1

1

1

1

1

1

1

1

2

2

2

J Shoulder Elbow Surg Dines, Grande, and Dines 207SVolume 16, Number 5S

in granulation tissue myofibroblasts and in quiescent and growingcultured fibroblasts. J Cell Biol 1993;192:103-11.

6. Dines JS, Uggen C, Dines DM, Razzano P, Mason J, Grande DA.Tendon gene therapy modulates the local repair environmentin vivo: a pilot study. Poster presentation at the American Acad-emy of Orthopaedic Surgeons Annual Meeting; 2004.

7. Evans CH. Gene therapy for ligament healing. In: Douglas W,Jackson DW, editors. The anterior cruciate ligament: current andfuture concepts. Philadelphia, PA: Lippincott Williams & Wilkins;1993, p. 419-22.

8. Galatz LM, Ball CM, Teefey SA, Yamaguchi K. Outcome andrepair integrity of completely arthroscopically repaired largeand massive rotator cuff tears. J Bone Joint Surg Am 2004;86:219-24.

9. Gartsman GM, Khan M, Hammerman SM. Arthroscopic repairof full-thickness tears of the rotator cuff. J Bone Joint Surg Am1998;80:832-40.

0. Gerber C, Fuchs B, Hodler J. The results of repair of massive tearsof the rotator cuff. J Bone Joint Surg Am 2000;82:505-15.

1. Harwood FL, Goomer RS, Gelberman RH, Silva MJ, Amiel D.Regulation of �(v)�-3 and �(v)�-integrin receptors by basic fibro-blast growth factor and platelet-derived growth factor-BB in intra-synovial flexor tendon cells. Wound Repair Regen 1999;7:381-8.

2. Harryman D, Mack LA, Wang KY, Jackins SE, Richardson ML,Matsen FA 3rd. Repairs of the rotator cuff: correlation of func-tional results with integrity of the cuff. J Bone Joint Surg Am1991;73:982-9.

3. Huard J, Fu F, editors. Gene therapy and tissue engineering in

orthopaedic and sports medicine. Boston, MA: Birkhauser;2000.

4. Nakamura N, Shino K, Natsuume T, Horibe S, Kaneda Y, OchiT. Early biological effect of in vivo gene transfer of platelet-derived growth factor (PDGF-�) into healing patellar ligament.Gene Ther 1998;5:1165-70.

5. Nakase T, Sugamoto K, Miyamoto T, Tsumaki N, Inui H, Yo-shikawa H. Activation of cartilage derived morphogenic pro-tein-1 in torn rotator cuff. Clinical Orthop Rel Res 2002:140-5.

6. Postel JM, Goutallier D, Lavau L, Bernageau J. Anatomical resultsof rotator cuff repairs: study of 57 cases controlled by arthrogra-phy. J Shoulder Elbow Surg 1994;3:20.

7. Rodeo S, Potter HG, Turner S, Campbell D, Kim HJ, Atkinson B.Augmentation of rotator cuff tendon healing using an osteoinduc-tive growth factor. Presented at Orthopaedic Research SocietyAnnual Meeting, Dallas, TX; 2002.

8. Rokito AS, Cuomo F, Gallagher MA, Zuckerman JD. Long-termfunctional outcome of repair of large and massive chronic tears ofthe rotator cuff. J Bone Joint Surg Am 1999;81:991-7.

9. Soslowsky LJ, Carpenter CM, DeBano CM, Banerji I, Moalli MR.Development and use of an animal model for investigations onrotator cuff disease. J Shoulder Elbow Surg 1996;5:383-92.

0. Uggen JC, Dines JS, Uggen CW, et al. Tendon gene therapymodulates the local repair environment of the shoulder. J AmOsteopath Assoc 2005;105:20-21.

1. Woo S, An K, Frank C, et al. Anatomy, biology, and biome-chanics of tendon and ligament, in orthopaedic basic science[Chapter 24]. In: Buckwalter JA, Einhorn TA, Simon, SR, editors.Orthopaedic Basic Science. 2nd ed. Rosemont, IL: AmericanAcademy of Orthopaedic Surgeons; 1999. p. 582-614.

2. Yoshikawa Y, Abrahamson SO. Dose-related cellular effects of

platelet–derived growth factor-BB differ in various types of rabbittendons in vitro. Acta Orthop Scand 2001;72:287-92.