Repair and Regeneration of Ligaments, Tendons, and Joint - part 10 doc

Gene Therapy and Ligament Healing 299mesenchyme-derived pleiotropic factor that regulates cell growth, cell motility, and phogenesis of various cells and is thus considered a humoral med

90 Eastwood M, et al Effect of precise mechanical loading on fibroblast populated collagen

lattices: morphological changes Cell Motil Cytoskeleton 1998;40:13–21.

91 Hsieh AH, Sah RL, Paul Sung KL Biomechanical regulation of type I collagen gene

expression in ACLs in organ culture J Orthop Res 2002;20:325–331.

92 Sakai H, et al Biological fixation of the graft within bone after anterior cruciate ligament

reconstruction in rabbits: effects of the duration of postoperative immobilization J Orthop Sci 2000;5:43–51.

93 Kvist J Rehabilitation following anterior cruciate ligament injury: current

recommenda-tions for sports participation Sports Med 2004;34:269–280.

94 Longley D, Lawford P Engineering human tissue and regulation: confronting biology and

law to bridge the gaps Med Law Int 2001;5:101–115.

95 Naughton GK From lab bench to market: critical issues in tissue engineering Ann N Y Acad Sci 2002;961:372–385.

96 Galbraith DN Regulatory and microbiological safety issues surrounding cell and tissue

engineering products Biotechnol Appl Biochem 2004.

97 Badylak SF Xenogeneic extracellular matrix as a scaffold for tissue reconstruction Transpl Immunol 2004;12:367–377.

98 Dornish M, Kaplan D, Skaugrud O Standards and guidelines for biopolymers in engineered medical products: ASTM alginate and chitosan standard guides American

tissue-Society for Testing and Materials Ann N Y Acad Sci 2001;944:388–397.

99 Schutte E Regulatory barriers to tissue-engineered products Med Device Technol 2002;

Gene Therapy and Ligament Healing 297

297

From: Orthopedic Biology and Medicine: Repair and Regeneration of Ligaments, Tendons, and Joint Capsule

Edited by: W R Walsh © Humana Press Inc., Totowa, NJ

15 Gene Therapy and Ligament Healing

Norimasa Nakamura

INTRODUCTION

Although there has been substantial progress in operative techniques, surgical mentation, and rehabilitation programs, based on the wealth of knowledge about thebiology and biomechanics of articular joints, the clinical outcomes following liga-ment injury are often still far from ideal Ligaments take longer to heal than otherconnective soft tissues, and the repaired ligament tissue is scarlike and inferior to

instru-normal ligament tissue both biologically and biomechanically (1) Furthermore, some

ligament-deficient joints subsequently become unstable and can lead to lifelong

dis-ability with osteoarthritis (2) Therefore, a novel therapeutic approach to accelerate

and improve ligament repair is needed One option could be the biological tion of ligament healing by the controlled delivery of biological reagents

manipula-PROBLEMS IN LIGAMENT HEALING

Animal studies on ligament healing have revealed that the same sequence of eventsappears to occur in the ligament as observed in skin wound healing The healing pro-cesses consist of inflammation (days to weeks), repair/proliferation (weeks), and remod-

eling (months to years; see Fig 1) Through these biological processes, ligaments heal

with scarring that is inferior to normal tissue biologically and biomechanically In tion, owing to their relative hypocellularity and hypovascularity, ligaments generallyhave a lower healing potential than other soft tissues, e.g., skin In fact, the tensile

addi-strength of injured skin recovers by 10 wk following injury (3), whereas gap-healing

rabbit medial collateral ligament (MCL) of the knee reaches only about 30% of thenormal ligament strength on a material basis (i.e., per square cross-section of material)

at even 1-yr postinjury (4) Even a completely remodeled ligament at over 2 yr injury remains scar-like (5) Such ligament scar remains different from normal tissue in

post-many aspects: elevated glycosaminoglycan content, decreased collagen content, mal collagen crosslinking different collagen types, and specifically, different ultra-

abnor-structure (1,4,5) Ligament scar has predominantly small-diameter collagen fibrils when

compared with the bimodal distribution (large and small diameter) found in normal

ligament (5) Such differences collectively seem to contribute to the inferior

biome-chanical properties of the ligament Considering that the major role of ligaments is tomechanically stabilize joints, ignoring inferior quality of scar material can lead to seri-

ous clinical problems, such as functional deficits and/or osteoarthritis Furthermore,recent investigation has revealed that healing ligaments show inferior creep behavior

(increase in strain under constant or repetitive stress) under low stress (6) As recent evidence suggests that ligaments are subject to repetitive low loads in vivo (7) and that

irrecoverable creep may result in a permanent stretching out of the ligament over time

(6,8), such inferior creep behavior of the healing ligament might have significant

clini-cal implications

STRATEGIES TO IMPROVE LIGAMENT HEALING

Exogenous Addition of Biological Factors Involved in Tissue Repair

Researchers have tried to develop strategies to improve and speed up the healingprocess of injured ligaments To this end, biological manipulation of scar-tissue forma-tion has predominantly focused on the overexpression of growth factors, which havebeen revealed as an important influence to cutaneous wound healing As describedpreviously, this is because the healing process of the ligament is basically analogous tothat of skin tissue

Normal wound healing begins with the accumulation of fibrin and platelet lation, the latter event involving the release of transforming growth factor-β (TGF-β),platelet-derived growth factor (PDGF), epidermal growth factor (EGF), and insulin-like growth factor 1 (IGF-1), which are chemotactic and mitogenic for inflammatorycells Accordingly, neutrophils and macrophages accumulate during the inflammatoryphase of wound healing with the latter cell type secreting more TGF-β, basic fibroblastgrowth factor (bFGF), and vascular endothelial growth factor (VEGF) These growthfactors stimulate fibroblasts and endothelial cells to proliferate, then fibroblasts andother reparative cells accumulate at the injured site and continue to synthesize andsecrete extracellular matrix (ECM) components Hepatocyte growth factor (HGF) is a

degranu-Fig 1 The healing processes of inflammation, repair/proliferation, and remodeling

Gene Therapy and Ligament Healing 299

mesenchyme-derived pleiotropic factor that regulates cell growth, cell motility, and phogenesis of various cells and is thus considered a humoral mediator of epithelial–

mor-mesenchymal interactions, including wound healing (9) Recent research has revealed

that HGF is expressed in wound fibroblasts, and its expression peaks at 7-d

postwound-ing, suggesting the importance of this growth factor in early wound repair (10) The

early phase of tissue repair is then followed by tissue maturation, remodeling, andreorganization Collectively, the early phases of wound healing depend on the transientand coordinated expression of various growth factors within wounds Therefore, appli-cation of these factors may potentially accelerate and improve wound repair Based onthese findings, various experimental studies have investigated the effect of these growthfactors on the improvement of wound healing Positive results with the administration

of TGF-β (11–13), EGF (14), PDGF-B (3,15), bFGF (16), VEGF (17,18), and HGF (9)

in wound repair have been demonstrated (see Table 1).

Regarding to ligaments, some studies have begun to characterize growth factors andtheir receptors during healing Transcripts for TGF-β1, EGF, bFGF, IGF-1, IGF-2, aswell as insulin and IGF-2 receptors, have been detected in normal and injured liga-

ments (19) Immunohistochemical studies demonstrated the expression of TGF-β (20,

21), EGF (22), bFGF (21,22), PDGF 2, and VEGF (23,24) during the early healing

phases of the ligaments All these findings have led to the experimental use of

exogen-Table 1

Application of Biological Factors to Promote Wound Healing

TGF-β Influx of mononuclear cells and fibroblasts 11

Enhanced collagen depositionIncrease in wound tensile strengthPDGF-B Influx of mononuclear cells and fibroblasts 15

Increase in wound tensile strength

Enhanced collagen depositionIncrease in wound tensile strength

studies showed increased mechanical strength of healing ligaments Administration ofbFGF to the healing ligament has also been conducted, and some positive effects on

matrix formation with enhanced angiogenesis have been demonstrated (29,30) But, both

studies have shown that the response is very dose-dependent and that excess growthfactor could interfere with the healing process Alternatively, growth and differentiationfactors (GDFs) 5, 6, and 7 (identical to bone morphogenic protein [BMP]-12, -13, and -14), members of the TGF-β gene superfamily, were found to induce neotendon/liga-ment-like tissue formation when implanted at ectopic sites in vivo In addition,

comparative in situ localizations of the GDF-5, -6, and -7 mRNAs suggest that these

molecules might be important regulatory components of synovial joint morphogenesis

(31) Their chondrogenic action to mesenchymal cells has also been reported (32)

Fur-ther characterization of these molecules for proper differentiation of mesenchymal stemcells into neotendon/ligament tissue will be needed

Along with these growth factors, the potential feasibility of other biological ecules to improve tissue repair has been suggested Effective tissue repair results from

mol-a rmol-apid, tempormol-ally orchestrmol-ated series of events At the site of locmol-al tissue injury, theproduction of many growth factors and cytokines is partly stimulated by the earlygrowth response transcription factors, which are expressed minutes after acute injury

A recent study has revealed that early growth response factor 1 (EGR-1), a tion factor, stimulates the production of many growth factors involved in early tissuerepair, such as PDGF-AB, HGF, TGF-β1, and VEGF (33) It also promotes angiogen-

transcrip-esis in vitro and in vivo, increases collagen production, and accelerates wound closure

(34; Table 1) These results indicate the potential use for this therapeutic transcription

factor, EGR-1, to improve tissue repair Thus far, this strategy has not been adopted in

a ligament-healing study

Alteration of Scar Tissue Composition Involved in Tissue Repair

This subsection introduces a strategy to improve scar-tissue quality by altering thematrix composition and organization Not many experimental studies have been con-ducted to date using this approach Because collagen (especially type I) is the maintensile element in the ligament, this matrix molecule is focused on as a target of thisstrategy Collagen is a major constituent of all ECM, and it is defined as having alengthy triple-helical domain and as aggregating in an extracellular space to function

as a “supporting element” of the tissue At present, more than 20 genetically distincttypes of collagens have been identified According to the supramolecular forms withinECM, well-characterized collagens are subgrouped into classes: fibrillar (types I, II,

Gene Therapy and Ligament Healing 301

III, V, and XI), fibril-associated (IX and XII), network-forming (IV), filamentous (VI),

short-chain (VIII and X), and long-chain (VII; 35) Among these classes, the fibrillar

collagens are thought to be chiefly responsible for the mechanical properties of thetissues Collagen type I appears to be the major collagen in both normal and injuredligaments, providing the leading component of the millions of collagen fibers (seen onlight microscopy) and their component, microfibrils (seen by transmission electron

microscopy [TEM]; 1) Of the various attributes of collagen (amount, concentration,

alignment, type, and so on.), many investigators have suggested that collagen fibrilthickness may best correlate with the mechanical properties of connective tissues

(36,37) Specifically, collagen fibril diameters seen on TEM may have a relationship to

tissue strength; apparently, larger fibrils are required for greater strength and stiffness.Ligament wounds, even over 2 yr after injury, contain mainly a homogenous popula-tion of small-collagen fibrils as commonly observed in scar tissues, with a few patches

of normal larger fibrils being observed (5) Also, as noted above, these ligaments never achieve their original biomechanical properties (4) Accordingly, production of larger-

diameter collagen fibrils could potentially improve the material strength of the ment scar Relating to growth factor/cytokine therapy, however, there have been nogrowth factor cytokines identified that directly promote collagen fibrillogenesis in vitro

liga-or in vivo Therefliga-ore, another approach is required to achieve this purpose The action of collagen microfibrils with other matrix molecules is one of the mechanismsimplicated in the regulation of collagen fibril diameters Regarding collagens, collagenIII, procollagen III with aminopropeptides (pN collagen III), collagen V, and collagen

inter-VI have been revealed to regulate collagen fibril diameters In addition, members ofthe small leucine-rich proteoglycans (SLRPs), decorin, fibromodulin, and lumican mayinhibit the lateral growth of collagen fibrils Furthermore, involvement of adhesionmolecules, thrombospondin-2 and osteopontin in collagen fibrillogenesis has been in-

dicated by knockout mice studies All these studies are listed in Table 2 (38–51)

Col-lectively, these observations suggest that the alteration of the molar ratio of thesemolecules to collagen microfibrils in ligament scar might result in changes in the lat-eral growth of collagen fibrils and thus potentially improve the mechanical properties

of the ligament scar

METHODS FOR GENE THERAPY

Gene therapy involves the transfer of a gene or genes to tissues within an individualfor a therapeutic purpose This technology offers the ability to manipulate the expres-sion of key molecules in tissue repair by the introduction of genes or gene antagonistsdirectly into the affected tissues Gene therapy not only allows an unprecedented abil-ity to quantify the contributions of various crucial molecules during the healing of jointinjuries, but it also offers ways to control local tissue repair processes in a uniquemanner

Generally, there are two strategies for gene transfer The first strategy involves lation of cells from an organism, establishment of the cells in tissue culture, transfer ofspecific genes into the cells, and subsequent reengraftment of the cells back into thepatient This strategy is termed “ex vivo gene transfer” and has been successful withcells that adapt well to culture and reengraftment The second strategy is to perform thegene transfer directly into somatic cells in the patient, termed “in vivo gene transfer”(Fig 2)

iso-Considerable effort in developing gene therapies has historically focused on genedelivery systems With few exceptions, naked DNA is not well taken up and expressed

by most cells Agents that enable the cellular uptake and expression of genetic materialare known as “vectors.” Vector characteristics have been reviewed in detail in other

comprehensive sources (52–55) Viral vectors (retrovirus, adeno-associated virus

[AAV], and adenovirus [AV]) have been most extensively investigated Their goal is

to infect target cells and to deliver the virally contained genetic material to the nuclei ofcells without permitting viral replication or viral pathology Each vector has certainstrengths and weaknesses in achieving this goal (Table 3) Because retroviruses insert

Fig 2 Gene transfer into somatic cells

Gene Therapy and Ligament Healing

Table 3

Advantages and Disadvantages of Common Vectors for Gene Transfer

• Low immunogenicity • Infection only of dividing cells

• High persistence of gene expression • Oncogenesis

• Infection of nondividing cells • Immunogenicity

• Low immunogenicity

• High persistence of gene expression

• Infection of nondividing cells

• High efficiency of transfection

• Infection of nondividing cells

• High efficiency of transfection

• Infection of nondividing cells

into the cellular genome at random locations, there are safety concerns regarding thepossibility of insertional mutagenesis that leads to cell transformation AAVs insertDNA at a site-specific location at the tip of chromosome 19 and, unlike retroviruses,can infect nondividing cells Furthermore, they generally enable high persistence oftransgene expression; they only accommodate 4 kb of carrier DNA In addition, recom-binant AAV is difficult to produce in high titer and may not retain the site specificity ofthe wild-type virus On the contrary, AV can be produced in very high titers They caninfect both dividing and nondividing cells, and they have high infectivity rates How-ever, their transgene expression is transient and the persistent expression of AV proteincan lead to a high incidence of antibody production, which can then influence infectedcells or subsequent reexposure Thus, this vector does not appear to be suitable forrepetitive delivery Furthermore, infection with high titers of virus can sometimes giverise to cytopathic effects.

As a result of these potential problems, other nonviral vectors have been developed,such as liposomes, DNA–ligand complexes, and colloidal gold (gene gun) Althoughthe alteration of gene expression is more transient, and their efficiency is generallyrecognized to be lower than that of viral vectors, these nonviral options are potentiallysafer than viral vectors The general concept of these techniques is that the carrieragent and plasmid DNA forms complexes that are transported into cells by endocyto-sis, or in the case of the gene gun, by mechanical pressure Among these vectors, lipo-somes are most widely used In fact, certain cationic liposome preparations are currently

in clinical use Recently, a unique fusigenic viral liposome has been developed fordirect introduction of macromolecules into the cytoplasm through cell fusion mediated

by Sendai virus (hemagglutinating virus of Japan [HVJ]; Fig 3; 56,57) With this cell

fusion mechanism, HVJ liposomes have proved to be 100–10,000 times more efficient

Fig 3 Fusigenic viral liposome developed for direct introduction of macromolecules intothe cytoplasm through cell fusion mediated

Gene Therapy and Ligament Healing 305

in gene transfer compared to liposomes without HVJ, and the method has been cessfully used for the introduction of foreign genes and antisense oligonucleotides(ODN) into several organs and tissues

suc-Gene Therapy in Ligament and Tendon Repair

Introduction of Marker Gene

Initial gene therapy studies focused on the introduction of a marker gene (βtosidase; LacZ) into normal and healing ligament and tendon tissues to evaluate theeffectiveness of ex vivo or in vivo gene transfer Nakamura et al injected the lacZ plas-

-galac-mid DNA directly into rat patellar ligament scar using HVJ liposomes (58)

LacZ-bear-ing cells were present in the injured area up to 56 d after transfection, with the peak ofexpression at d 7 (7% of cells at the wound site; Fig 4) With double labelingfor markerantigens for monocyte/macrophage (ED-1) and for collagenI aminopropeptide (pN col-lagen I), it was revealed that fibroblastic(pN collagen I-positive) cells accounted for63% and monocyte/macrophagelineage cells for 32% of the LacZ-labeled cells in thed-7wound On d 28, they formed 58% and 35% of the LacZ-labeledcells in the wound,respectively Moreover, specific labeling of the transfectedcells revealed a biologicalevent, i.e.,that the cells in and around the injured site infiltrate into theuninjured liga-ment substance and come to populate the whole lengthof the ligament substance asrepair progresses Although a potentially less invasive intraarterial delivery of lacZ gene

in HVJ liposomes into the healing rat patellar ligament has been explored (59), a

compa-rable expression of the lacZ gene product was observed This alternative delivery methodcould be advantageous for gene delivery to deeper tissues With regard to viral vectors,

Fig 4 LacZ-bearing cells present in the injured area up to 56 d after transfection, with thepeak of expression at d 7

using the ex vivo and in vivo strategy, the lacZ gene has been introduced into normalrabbit patellar tendon retrovirally and adenovirally, respectively, and the duration of

gene expression was at least 6 wk by both delivery methods (60) The efficiency of transfection in situ by this delivery technique has not been clarified The same research

group also introduced the LacZ gene into normal and injured rabbit MCL and the rior cruciate ligament (ACL) of the knee They confirmed transgene expression for over

ante-6 wk within normal and injured ligaments (ante-61) Adenoviral vector-mediated gene

trans-fer of the lacZ gene to chicken tendon and tendon sheath has also been investigated.Descriptive analysis of lacZ gene expression has been performed, and expression per-

sisted for over 75 d in both tissues (62) Recently, to obtain longer-term gene expression

in the joint, unique myoblast mediated ex vivo gene transfer has also been investigated.The transduced myoblasts were found in the ACL and in the synovial tissue surroundingthe ACL at 4-, 7-, 14-, and 21-d postinjection The myoblasts fused and formed myotubes

in the ligament (63) This unique gene transfer method may be applicable to the

biologi-cal manipulation of ligament healing; further study is expected

Gene Therapy to Accelerate Ligament/Tendon Repair

As described, recent studies have shown the positive effects of growth factors onwound healing and ligament/tendon repair Therefore, initial gene transfer studies havefocused on the overexpression of growth factors to accelerate healing Using the HVJ-liposome method, the PDGF-B gene was introduced into healing rat patellar ligaments

(64) This PDGF-B gene transfer resulted in the enhanced expression of PDGF in the

healing ligament up to 4 wk after transfection, leading to an initial promotion of genesis (Fig 5) and subsequent enhanced collagen I deposition in the wound (Fig 6).There has been no in vitro study showing the direct effect of PDGF on the synthesis ofcollagen I; yet, it is known that PDGF stimulates macrophages to produce TGF-β1 which

angio-stimulates collagen formation (65) Therefore, some stimulating factors of matrix

syn-thesis (e.g., TGF-β1) could be simultaneously overexpressed in the PDGF ferred healing ligament No significant collagen I deposition was detected in the

gene–trans-Fig 5 (A) Using the HVJ-liposome method, the PDGF-B gene was introduced into healing rat patellar ligaments (B) PDGF-B gene transfer resulted in the enhanced expression of PDGF

in the healing ligament up to 4 wk after transfection, leading to an initial promotion of genesis

angio-Gene Therapy and Ligament Healing 307

gene-transferred wound at 8 wk by semiquantitative morphological analysis However,

it should be noted that the amount of collagen deposition in the gene-transferred woundwas a comparable amount with that of the control wound 4 wk earlier This could beinterpreted as the acceleration of ligament healing by gene transfer With focus on astrong in vivo angiogenic action, gene transfer of HGF into healing rat patellar ligament

has been also investigated using the HVJ-liposome method (66) Although the results

were preliminary, in vivo introduction of HGF resulted in enhanced angiogenesis andcollagen synthesis for the first 4 wk following gene transfer Further analysis is pending

As noted previously, the induction of neotendon/ligamentlike tissue by BMP12,

-13, and -14 has been demonstrated (31) Accordingly, to enhance neotendon tissue

formation following injury, the effect of BMP-12 gene transfer on tendon cells andchicken tendon healing has been investigated Adenoviral BMP-12 gene transfer intochicken tendon cells increased type I collagen synthesis and gene transfer into injuredtendon resulted in a twofold increase of tensile strength and stiffness of repaired ten-

dons, indicating improved tendon healing in vivo (67) Adenoviral BMP-13 transfer

into rat thigh muscle also resulted in the formation of collagenous matrix with theultrastructural appearance of neotendon/neoligament At the same time, small foci of

bone and fibrocartilage were also seen within the treated tissue (68) Thus, based on

these results, gene transfer of BMP-12 and BMP-13 might be a promising procedurefor improving the ligament/tendon repair However, it should be emphasized that localadministration of these BMPs into healing tissues, where a number of immature mes-enchymal cells are recruited, has the potential risk of producing chondrogenic and bonytissue, because these BMPs are also known for their strong chondrogenic and bone

morphogenic action (69) More optimization studies would be required for appropriate

tissue induction

Gene Therapy to Prevent Tissue Adhesion

Adhesion is a critical complication in tendon healing Lou et al researched the effect

of the local administration of focal adhesion kinase (pp125FAK) in tendon adhesion (70).

Fig 6 (A) PDGF-B gene transfer resulted in the enhanced expression of PDGF in the

heal-ing ligament up to 4 wk after transfection, leadheal-ing to an initial promotion of angiogenesis and

(B) subsequent enhanced collagen I deposition in the wound.

The intracellular focal adhesion kinase (FAK)-related signaling pathway may be related

to cell–cell and cell–ECM interactions via the cell surface adhesion receptors,

includ-ing the integrins and cadherins (71), and may be one of the mechanisms involved in the

induction of tendon adhesions Gene transfer of pp125FAK to the tendon sheath resulted

in abnormal tendon adhesion This finding suggested a possible relationship betweencell–ECM interaction via the integrins and overproduction of the ECM leading to tis-sue adhesion

Gene Therapy to Alter Collagen Ultrastructure

As mentioned, collagen fibril diameter may correlate with the mechanical properties

of connective tissues (36,37) Healing ligament contains mainly a homogenous lation of small-diameter collagen fibrils as commonly observed in scar tissues (5) and

popu-is inferior to normal tpopu-issue biomechanically (4,6,8) Therefore, the production of

larger-diameter collagen fibrils could be used as a strategy to improve the mechanical ties of healing ligament As shown in Table 3, several matrix molecules have beenidentified to regulate collagen fibril diameter The results of in vitro binding experi-ments, molecular and biochemical analyses of tendon development, and knockoutmouse studies, collectively indicate the involvement of decorin (a member of the

proper-SLRPs) in downregulating collagen fibril diameters (43–45,72) Moreover, the ence of decorin mRNA (73) and protein (74) in the ligament scar has been observed.

pres-Based on these results, it could be hypothesized that decorin inhibition during earlyligament healing would possibly enhance the lateral growth of newly synthesized col-lagen fibrils in the ligament scar To suppress a specific molecule, antisense approacheshave been investigated in a variety of studies Antisense ODN inhibit gene expression

in a sequence-specific manner Complementary ODN specifically bind to mRNA orpre-mRNA by base pairing, thus enhancing the degradation of target mRNA and alsoblocking the translation of the target gene (Fig 7) Therefore, in vivo antisense therapy

could potentially suppress targeted gene expression in a specific tissue (75) Studies

Fig 7 Complementary ODN specifically bind to mRNA or pre-mRNA by base pairing, thusenhancing the degradation of target mRNA and also blocking the translation of the target gene

Gene Therapy and Ligament Healing 309

have shown that the in vivo introduction of fluorescence-labeled ODN into a healingrabbit MCL can be achieved using the HVJ-liposome-mediated gene transfer method

(74) With systematic direct injection into the ligament scars using a dispenser and a

microgrid mesh system to distribute the liposomes, scar cells were effectively fected when assessed 1 d after transfection Furthermore, introduction of antisenseODN for the small proteoglycan decorin has resulted in decorin suppression at boththe mRNA and protein levels over 4 wk following exposure to the HVJ liposomes.Such changes in decorin expression have led to the development of larger-diameter

trans-collagen fibrils within scars (Fig 8; 76) but results are somewhat variable among the

animals with the same treatment The degree of increase in collagen fibril size variedaccording to the location, implying that collagen fibril assembly was not improved inthe whole area of each ligament scar by this single antisense treatment, but it hadclearly altered the morphology of most scars in several locations The average col-lagen fibril diameter in antisense treated scars was 104.7 ± 51.1 nm, whereas the con-trol scars and normal MCL were 74.8 ± 11.0 nm and 189.1 ± 104.0 nm, respectively

In mechanical assessments, this antisense therapy caused enhanced resistance of the

healing ligament scar tissue to elongation by creep (Fig 9; 6,8) Antisense-treated

scars were significantly less suspectible to creep during low stress creep testing (at 2.2Mpa) than control scars (by 18–22%), and they also revealed 33–48% less irrecover-able creep after the same recovery period than control scars (Table 4) Antisense-treated scars also failed at a higher stress than control scars (by 82–84%) (Fig 10;

25,76,77) This study is the first demonstration of in vivo manipulation of collagen

fibrillogenesis during soft connective tissue repair processes, as well as the first reportthat shows such a treatment can improve the functional properties of ligament scar-ring However, it is not likely that there is a simple, direct cause-and-effect relation-ship between downregulation of decorin expression and the increase in collagen fibrildiameters, as well as the improvement in the biomechanical properties of the scartissue Although the ODNs were designed to be specific for decorin based on availableevidence, it is possible that secondary regulation of other molecules could be involved

Fig 8 Development of larger-diameter collagen fibrils within scars