ELSEVIER Journal of Orthopaedic Research 23 (2005) 671-679

Journal of Orthopaedic

Research www.ekevier.codlocdte/orthres

Thrombin peptide (TP508) promotes fracture repair by up-regulating inflammatory mediators, early growth

factors, and increasing angiogenesis

Hali Wang Xinmin Li b, Emre Tomin ', Stephen B. Doty ', Joseph M. Lane ', Darrell H. Carney d, James T. Ryaby a

a Research and Development, OrthoLogic Corp. 1275 W. Washington Street. Tempe, A 2 85281, USA Functional Genomics Facility, University of Chicago, Chicago, IL 60637, USA

' Hospital for Special Surgery, New York, NY 10021, USA The University of Texas Medical Branch, Galveston. TX 77555-0645, USA

Accepted 14 October 2004

Abstract

Previous studies have shown that a single injection of thrombin peptide (TP508) accelerates fracture repair in a closed rat femoral fracture model. The present study was conducted to elucidate the molecular mechanisms of TP508 action using Affymetrix genome- scale profiling and to link early gene expression changes to fracture histology and bone strength changes. Treatment of femoral frac- tures with TP508 accelerated fracture repair as determined by destructive torsion testing. Blinded histological analysis demonstrated that TP508-treated fracture callus had a significant increase in blood vessels relative to the controls. Gene array analysis showed that TP508 significantly induced expression of early growth factors, inflammatory response modifiers, and angiogenesis-related genes. This study therefore suggests that TP508 promotes fracture repair through a mechanism that involves an increased induction of a number of growth factors, enhanced expression of inflammatory mediators, and angiogenesis-related genes. 0 2004 Orthopaedic Research Society. Published by Elsevier Ltd. All rights reserved.

Keywords: Fracture repair; TP508; Gene expression; Early growth factors; Angiogenesis; Microarray analysis

Introduction

Fracture repair is a complex physiological process, which can be enhanced through therapeutic approaches. The fracture repair process is divided into five distin- guishable stages as described by Einhorn: the initial hematoma and inflammation, intramembranous bone formation, angiogenesis and chondrogenesis, endochon- dral bone formation, and bone remodeling [12,24,38]. At the initial phase, inflammatory cells, macrophages, and degranulating platelets infiltrate into the fracture site

Corresponding author. Tel.: + I 602 286 5325; fax: + I 602 286

E-mail address: hwang@olgc.com (H. Wang). 2925.

from peripheral blood. Various growth factors and cyto- kines released from these cells play key roles as initiators of the fracture repair process resulting in osteoprogeni- tor cell differentiation and proliferation, and angiogene- sis [1,4,11,13,18]. A therapeutic molecule may alter any of these stages, potentially enhancing the overall frac- ture repair process.

Angiogenesis is an essential part of fracture repair that may be initiated as early as the hematoma and inflammatory stage [ 17,421. Inadequate or inappropriate vascularity is associated with decreased bone formation, and development of delayed and mal-unions [7,34]. In addition, administration of an angiogenesis inhibitor prevents femoral fracture repair in rats [19]. Angiogene- sis is regulated by many growth factors induced in

0736-0266/$ - see front matter 0 2004 Orthopaedic Research Society. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.orthres.2004.10.002

612 H. Wutig et ul. I Journul of Orthopuedie Researel1 23 (2005) 671479

response to injury, such as vascular endothelial growth factor (VEGF), fibroblast growth factor (FGF) [4,21,32]. Therefore, treatment with angiogenic mole- cules may be expected to promote early angiogenesis and accelerate later stages of fracture repair.

A number of recombinant growth factors, regulatory peptides, and small molecules have been tested in recent years for potential efficacy in stimulating fracture repair [2,20,25,29,36,39,41]. Among the above molecules, re- combinant human bone morphogenetic proteins (BMP), including BMP-2, BMP-4, and BMP-7, have shown sig- nificant enhancement of the repair process in differ- ent animal models of fracture repair [10,35,44]. Although recombinant growth factors have shown effectiveness, the use of peptides or small molecules is a promising alternative due to ease of synthesis, stability, and lower cost.

The thrombin related peptide, TP508 (Chrysalin@), is a synthetic 23 amino acid peptide representing the natu- ral sequence of the receptor-binding domain of human thrombin (prothrombin amino acids 508-530). This peptide was initially identified by its ability to compete with thrombin for binding to a high-affinity thrombin receptor on fibroblasts and to generate receptor occu- pancy-dependent mitogenic signals [ 161. Subsequently, a number of preclinical studies demonstrated a potential therapeutic role of TP508 in tissue repair. A single top- ical application of TP508 enhanced repair of full dermal wounds [9], and accelerated both closure of full-thick- ness excisional wounds in normal rat skin [40] and der- mal excisions in rat skin with surgically induced ischemia [28]. Interestingly, in each of these dermal wound studies, TP508 acceleration of the repair process was associated with early recruitment of inflammatory cells to the site of injury and enhanced early angiogene- sis [27].

More recently, the application of TP508 has been evaluated in orthopedic tissue repair models. A single fracture site injection in a rat closed femoral fracture model [36,39] showed an increase in bone strength of approximately 40% in 2-month-old rats, and 25% in 8- month-old rats. TP508, delivered in controlled release poly-lactoglycolide (PLGA) microspheres, stimulated bone formation in rabbit segmental defects [37]. Inter- estingly, TP508 delivery to rabbit cartilage defects using the same microspheres enhanced repair of full thickness articular cartilage defects in rabbits [22]. These studies show that TP508 enhances musculoskeletal tissue repair, although its underlying mechanisms are unknown.

In this study, we investigated the molecular mecha- nisms of TP508 action in enhancing fracture repair. Since TP508 represents a receptor-binding domain of human thrombin, and thrombin plays a critical role in hematoma and angiogenesis during soft-tissue repair, we hypothesized that TP508 accelerates fracture repair by modulating the expression of early response and

angiogenesis-related genes, such as inflammatory cyto- kines and growth factors. We tested the hypothesis by performing histology and mechanical testing at three and four weeks post-fracture, respectively, and compre- hensive gene expression studies at the early stage of frac- ture repair.

Materials and methods

Aniniul model of fructure

All animal experimental procedures were approved and performed under the guidelines of the Institutional Animal Care and Use Com- mittee (IACUC). Ten-month-old male retired breeder Sprague-Daw- ley rats (Harlan, Indianapolis, IN) weighing 450-550g were randomly divided into three groups in this study. Each rat was sub- jected to a closed transverse fracture on its right femur as described by Bonnarens and Einhorn [5]. Briefly, animals were anesthetized with intramuscular injection of ketamine and xylazine at 120mg and 20mg respectively per kg body weight. A small skin incision was made in the right knee, and the patella was deflected. A 1.1 mm diameter hole was drilled into the intercondylar notch, and a Kirshner wire (1.1 mm diameter) was inserted into the medullary canal. After closing the knee joint, a mid-diaphyseal fracture was induced using a three-point bend- ing device driven by a dropped weight. Fractures were verified via con- tact radiograph using the Hewlett Packard Model #43855-A Faxitron Closed X-ray System. One hour after fracture, TP508 at doses of 0 (sal- ine control), I , and lOpg in loop1 saline was percutaneously injected into the fracture site of each rat in each of the respective groups. The rats were allowed to bear weight as tolerated after recovering from anesthesia. At predetermined time points post surgery, each rat was anaesthetized with isoflurane and killed with 2ml Euthasol (containing Sodium pentobarbital, Delmarva Labs, Midlothian, VA) via intraperi- toneal injection. The right femurs were harvested as required for the different procedures described below.

Biomeclianical testing

This experiment included three groups: control (saline), 1 pg and l0pg TP508 treated groups. Twelve animals were used for each group at each time point. A total of 108 animals were sacrificed at 2, 3 and 4 weeks post-fracture. Femurs were carefully dissected free of soft tis- sues, wrapped in 0.9% saline soaked gauze, frozen and retained at -20°C for destructive torsional testing. Before testing, the exposed ends of the fractured femurs were embedded in Woods metal (Cerro- bend alloy, melting temperature 70°C). Testing was performed with a custom DC servomoter controlled torsion testing apparatus with a deformation rate of approximately 550" or IOradls as described [20]. Biomechanical results are reported as maximum torsional strength [8,30]. Statistical analysis was performed by ANOVA on Ranks (Krus- kal-wallis).

Histology unalysis for blood vessel,formution

This experiment also included three groups: control, 1 pg and 1 0 ~ g TP508 treated groups. Four animals were used in each group and all animals were sacrificed at 3 weeks post-fracture. Fractured femurs were radiographed using the Faxitron, then fixed in 10% buffered for- malin, and subsequently demineralized in 5% nitric acid. Pins were re- moved from the bones before embedding and sectioning. Fixed and decalcified tissues were dehydrated in a series (70%, 90%. 95%. and 100% x2) of ethanol concentrations, transferred to xylene and embed- ded in paraffin. Five-micron thin paraffin sections were cut longitudi- nally and stained using Masson's Trichrome.

Blinded histological analysis of blood vessel formation was per- formed as follows. Digital images of 12 animal fractures, 4 per group and randomly coded, were captured using a Leica DMLB microscope

H . Wang et al. I Journul ojOrtliopuedic Research 23 (2005) 6 7 1 4 7 9 673

equipped with an RT Color Spot Camera (Diagnostic Instruments). Each fracture was captured at 2Sx, and the periosteal soft callus area was identified on both sides of the fractured bone. These callus regions were referred to as the top callus and bottom callus as they were visu- alized in the microscope with the bone going from left to right. Serial images were captured across the callus section at 20x. Each picture was labeled sequentially and the XIY axis recorded. Care was taken to as- sure that the same vessels were not counted two times in cases where the images may have overlapped. Once all images were captured, callus area analyzed, vessel number, and vessel area were calculated from ten images per callus (using five images from the top callus and five from the bottom callus where possible) using Image Pro software. Once the imaged area of the callus had been determined, the perimeter of each blood vessel within this area was traced to determine the vessel area. These data were pooled for each fracture to obtain a value per fracture and were then normalized by taking the total number of vessels or total area occupied by vessels divided by the total callus area analyzed. The data were statistically analyzed using a repeated mixed model ap- proach in the SAS procedure MIXED using the simple covariance structure. Measures from the same animal were considered a cluster of data. Pairwise comparisons were made with least squares means.

considered as candidate genes for further investigations. We also used GeneSpring 5.0 to reveal gene expression patterns and validate the above results.

Quantitative real-time RT-PCR

The same RNA samples at day 1 for microarray hybridizations were used for quantitative RT-PCR. Real-time PCR primers were se- lected for a representative set of selected genes using PRIMER EX- PRESS software (Version 2.0, Applied Biosystems). Reactions were performed in a Sop1 volume that included diluted cDNA sample, pri- mers, and SYBR Green PCR Master mix (Applied Biosystems). Real- time PCR reactions were performed on an Applied Biosystems Prism 7000 sequence detection system. Predicted cycle threshold (Ct) values were exported directly into EXCEL worksheets for analysis. The stan- dard Curve Method was used for the relative quantitation of expres- sion for each gene. Expression of the housekeeping gene GAPDH was used to normalize the expression data.

Results RNA estrcrction

For microarray analysis, three groups were included: control, 1 pg and 10pg TP508 treatment. Three rats were used in each group at each time point. In addition, three intact femurs were used as baseline sub- traction. Femurs were collected at days I , 2, and 4 post-fracture. So a total of 30 rats (3 rats x 3 groups x 3 time points + 3 intact femur base- line) were used in this experiment. The soft tissues around femurs were carefully removed and midshafts were cut off using a sterile Dremel saw blade and frozen in liquid nitrogen until use. Total RNA was iso- lated using Trizol reagent (Life Technologies, Gaithersburg, MD) and RNeasy Mini column purification (Qiagen, Chatsworth, CA) [33]. Integrity of RNA was evaluated with an Agilent 2100 Bioanalyzer (Agilent Technologies, Pa10 Alto, CA, USA) and puritykoncentration was determined using a GeneSpec 111 (Miraibio). All RNA samples used for hybridization had an OD260/280 ratio > 1.8 and total RNA concentration > I pglml.

Microarray liyhridizution and duta unulysis

The Functional Genomics Facility at the University of Chicago performed all Rat U34A Gene Array hybridizations. The target prep- aration protocol followed the Affymetrix GeneChip Expression Anal- ysis Manual (Santa Clara, CA). Briefly, 1Opg of total RNA was used to synthesize double-stranded cDNA using the Superscript Choice Sys- tem (Life Technologies). First strand cDNA synthesis was primed with a T7- (dT24) oligonucleotide. From the phase-log gel-purified cDNA, biotin-labeled antisense cRNA was synthesized using BioArray High Yield RNA Transcript Labeling Kit (Enzo Diagnostics, Farmingdale, NY, USA). After precipitation with 4 M lithium chloride, 20pg of cRNA was fragmented in fragmentation buffer (40mM Tris-acetate, pH8.1, IOOmM KOAc, 30mM MgOAc) for 35min at 94°C and then 12g of fragmented cRNA was hybridized to U34 Arrays for 16h at 45 “C and 60 rpm in an Affymetrix Hybridization Oven 640. The arrays were washed and stained with streptavidin phycoerythrin in Affymetrix Fluidics Station 400 using the Affymetrix GeneChip protocol and scanned using the Affymetrix Agilent GeneArray Scanner.

The acquisition and initial quantification of array images was per- formed using the Affymetrix Microarray Suite Version 5.0 (MAS 5.0) with the default analytic parameters. Subsequent data analysis was used in: ( I ) Evaluation of hybridization quality; (2) Two-step data fil- tration (the first step was to filter genes with signal intensity <I00 intensity units in all samples, and the second step filtration was to re- move the genes that receive an “absent” call in all samples); and (3) Identification of differentially expressed genes. Two separate statistical methods were applied to assess differential gene expression between the compared samples. The first method is Significant Analysis of Micro- arrays (SAM). Significant genes were identified at a delta value with FDR <IS’%, [43]. Following SAM analysis, we independently per- formed t-test with unequal variance. Overlapped genes from the two differentially expressed gene lists identified by SAM and t-test were

Mechanical testing

As shown in Fig. 1, TP508 treatments increased mechanical strength in both low and high dose groups as early as three weeks post-fracture. At 4 weeks, the low dose group (1 pg/lOOpl saline) increased 21% rela- tive to the saline control group, while the high dose group (lOg/lOOpl saline) increased 36% (p c 0.05).

Histology

Histological analysis showed more functional blood vessels containing red blood cells in both 1 pg and 10 pg TP508 treated fracture calluses in comparison with controls (Fig. 2). Quantification of vessel number ( V #) and area (VA) occupied by vessels greater than 50pm2, which included all veins, arteries, venules and arterioles, but not small capillaries, was performed in a blinded fashion using digital morphometric analysis of callus

0.4 - - E

6 0.3 - t C ?? z 2 0.2 - - 0 .- 5 +

0.1 -

4 lug TP508

=Pc.05 0.0

1 2 3 4 5

Time Post-fracture (weeks)

Fig. I. Effect of TP508 on mechanical strength (torsion testing) of rat femurs four weeks after fracture. Statistical analysis was performed by ANOVA on Ranks (Kruskal-Wallis). ‘p < 0.05. n = 12.

674 H. Wang et al. I Journal of Orthopaedic Research 23 (2005) 6 7 1 4 7 9

Fig. 2. Effect of TP508 on blood vessels development in fracture calluses three weeks after treatment. Histological section stained with Masson's Trichrome showed an increased number of large functional blood vessels with red blood cells in TP508 treated fracture callus relative to controls. A and B: controls; C and D: 1 pg; E and F: 1Opg. B, D, and F are from the rectangles of A, C, and E, respectively.

regions within the fractures. As shown in Table 1, 1 pg of TP508 treatment resulted in a 41% increase in the total number of vessels per 1000 pm2 callus area (CA), increasing from the control value of 57.17-80.57 in the TP508 treated callus. In addition, this treatment resulted in an 87% increase in area occupied by vessels, suggest- ing that 1 pg TP508 treatment increased both the num- ber and size of the vessels within the callus region. In contrast to the effects observed with the 1 pg dose and demonstrating an apparent dose response, treatment with 1Opg increased the vessel area by approximately 44%, but did not affect the total number of vessels.

Dc~erential expression of genes related to immune response

The molecular actions of TP508 treatment are be- lieved to start at an early stage of fracture repair. We therefore first focused on the analysis of differential gene expression between TP508 treatment and fractured con- trol one day after fracture. A list of selected genes clearly points to the role of immune response genes in mediation of TP508 effects. The expression of several major genes involved in early immune response was significantly in- creased in the TP508 treated group relative to controls,

Table 1 Effect of TP508 on blood vessel development in fracture calluses"

Group V #/CA St. error V # increase p value VNCA St. error VA increase p value

Control 57.17 5.95 20.74 3.83 1 80.57 6.75 41%) <0.03 38.77 4.34 87% <0.02 I O N 2 57.13 6.03 0% 0.996 29.92 3.88 44% 0.12

a Blinded histological analysis of fracture calluses three weeks post-fracture was performed on 12 rats (n = 4 per group) using ten images per callus to calculate vessel number ( V #), callus area analyzed (CA), and total area occupied by all vessels (VA) relative to CA.

H . Wung et ul. I Journal of Orthopuedic Reveurch 23 (2005) 6 7 1 4 7 9 615

Table 2 Differential expression of genes involved in immune responses

Gene name Accession # Fold change at day 1

1 Pg p value 1oPg p value

MHC class I1 antigen RT1.B beta chain U65217 2.80 0.05 2.50 ns MHC class I1 gene (RT1.DOa) D45240 1.4 0.006 1 ns MHC class I1 A-beta RT1.B-b-beta M36151 1.87 0.03 1.4 ns Interleukin I-beta M98820 1.56 0.05 2.02 0.003 Interleukin I-beta receptor type 2 222812 2.28 0.05 2.03 0.05 Interleukin I-beta converting enzyme U14647 1 .oo ns 1.30 0.003 IL 3 beta U81492 1.10 ns 1.40 0.05 IL 3 receptor beta S79263 1.50 0.05 1 .oo ns Interleukin-6 M26745 1.1 ns 1.73 0.056 Interleukin- 10 X60675 1.60 0.03 1.30 ns Interleukin-I 2 S82489 4.90 0.05 3.70 ns Interleukin-15 U69272 1.35 0.015 1.12 ns IFN-gamma M29317 1.70 0.03 1 .oo ns

Table 3 Differential expression of early growth factor genes and genes involved in angiogenesis

Gene name Accession # Fold change at dav 1

Neutrophil collagenase NO07288 CXCR2 U70988 Adrenomedullin Dl 5069 BI bradykinin receptor A5132230 Smad8 AF012347 Egr- 1 MI8416 c-jun oncogene XI7163 Leptin receptor' U67207 CXCR4 U906 10 Angiopoietin-2 AF030378

iNOS S71597 TNF receptor M63 122 C-FOS X06769 IGF-I1 receptor US9809 FGF2' M22427 Osteopontin M 14656 uPAR X71898 PDGF-A 214120 Neuropilin AFOl6296 Collagen-binding protein (gp46) M69246

* Significantly up-regulated at day 2.

Somatostatin receptor-like protein u77953

including several MHC Class I1 genes, Interferon-y, IL- lp, IL-lp Receptor type 2, ILlO, and IL 12 (Table 2).

Diflerential expression of genes related to early growth and angiogenesis

Twenty-one genes involved in angiogenesis were dif- ferentially expressed, 19 of which were up-regulated and two down-regulated after TP508 treatment (Table 3). Of those up-regulated genes, early growth response factor-1 (Egr-1) is particularly interesting. This gene was expressed at a low level (about 200 intensity units) in the intact bone and significantly increased after frac- ture in all three groups (Fig. 3). Amongst the fractured

~

1 Pg p value p Value

3.6 0.05 3.1 ns 2.3 0.05 1.9 0.08 2.0 0.02 0.8 ns 2.0 0.004 1.3 ns 2.0 0.04 1.5 ns 2.0 0.05 1.20 ns 1.8 0.04 1.1 ns 1.8 ns 1 ns 1.7 0.04 1.7 ns 1.6 0.04 1.3 ns 1.6 0.04 1.1 ns 1.5 O.OOO1 1.6 0.03 1.5 0.05 0.92 ns I .4 0.04 1 .oo ns 1.4 0.02 1.4 ns 1.3 ns 1 ns 1.3 0.58 I .4 0.03 1.2 0.05 1.2 0.07 1.1 0.05 1 ns 0.5 0.02 0.6 0.03 0.5 0.05 0.98 ns

bones, the increase observed with low dose TP508 treat- ment was significantly greater than that of the fractured control (2-fold). The increased Egr-1 expression seen with TP508 treatment was followed by minor increases in expression of a set of Egr-1 target genes including

ICAM-1 and TNF-a. The increase of these Egr-1 target genes varied from 1.2 to 1.8-fold and became statisti- cally significant when analyzing all these genes together. The other two key early response genes regulating angi- ogenesis are c-Fos and c-Jun, and these genes were both significantly up-regulated by TP508. The increase of these genes quickly went back to the same level as the controls at day 2 and thereafter except for FGF2 and

EGF, PDGF-U, PDGF-P, FGF2, IGF-11, CD 44,

676 H . Wung ef ul. I Journul of’0rthi)puedic Research 23 (2005) 6 7 1 6 7 9

Fig. 3. Temporal expression of Egr-l under different conditions using microarray. Expression levels of Egr-l at the fracture site were significantly increased in comparison to the non-fractured control ( * = p < 0.01) at day I . Its expression was further elevated by the TP508 treatment (# = p < 0.05) when compared with fractured saline control. n = 3.

iNOS, which were significantly increased at day 2 (data not shown). In contrast to increased expression of the selected angiogenesis-related genes analyzed, neuropilin and collagen-binding protein (gp46) were down-regu- lated by TP508 at day 1 post fracture.

Validation of microurray data

Verification of microarray-based differential gene expression was examined on nine representative genes by using quantitative real-time RT-PCR. Six of these genes were up-regulated and three down-regulated based on microarray results after TP508 treatment, respec- tively. Fig. 4 compares the relative fold changes of these nine genes with the two methods of measurement. Ex- cept for one gene, type IV collagenase, all other genes showed highly concordant quantitative measurements with the direction of differential expression.

Discussion

Our microarray data support the hypothesis that TP508 promotes fracture repair by initiating an early regulatory cascade response that results in early activa- tion of inflammatory mediators and induction of angio- genic genes. These possible mechanisms derived from transcriptional analysis may provide a key to under- standing the mechanism that leads to the 4 1% increase in blood vessels in fracture callus observed at three weeks by histological evaluation and the 21-36%, in- crease in bone strength as demonstrated by a torsion testing at 4 weeks.

The first phase of fracture repair beginning immedi- ately after fracture is the inflammatory phase, which in- volves neutrophils and macrophage recruitment and

Fig. 4. Comparison of gene expression levels determined by quanti- tative real-time RT-PCR and microarray analysis. RNA samples for both analysis were from day I after treatment with 1 pg TP508. 1. Small espin; 2. IL 1-beta; 3. type IV collagenase (matrix metallopro- teinase-2): 4. IL-12: 5. IL-1 receptor type 2; 6. brush border myosin I: 7. RABl4; 8. calpastatin; 9. ras-related rablB. n = 3.

activation [31]. TP508 seems to play a significant role in mediating immune response through a possible mech- anism depicted in the model proposed in Fig. 5. As pro- posed, one possible scenario is that TP508 induces interferon-gamma expression, which interacts with and consequently activates macrophages as indicated by the increased expression of MHC class I1 marker genes (macrophages only become fully functional when acti- vated) [3,23]. Two pathways are then used by the macro- phage; cell-mediated and factor mediated. In the cell-mediated effect, activated macrophages mediate the immune response directly by protecting the wound site from bacterial infection, and processing antigens and presenting digested peptides to T lymphocytes. The factor mediated, or indirect mediation, involves activated macrophages releasing a group of cytokines including 1L-1 beta that activate other effector cells lead- ing to secretion of essential cytokines required for frac- ture repair. These essential cytokines are key regulators of differentiation during fracture repair, as has been shown for TNF-cl signaling and subsequent cartilage differentiation [ 151. Thus, TP508-mediated activation of macrophages may represent a mechanism of amplifi- cation of fracture repair signals through increasing the magnitude of immune response during the inflammatory phase.

After the initial inflammatory stage, fracture repair enters its growth stage. An essential requirement for tissue repaidgrowth is the formation of new blood ves- sels to sustain the newly formed granulation tissue. Data analysis revealed a clear differential expression pattern of angiogenesis-related genes, which suggests an angiogenesis-mediated mechanism. Egr- 1 , c-Fos and c J u n represent three early response genes impli- cated in angiogenesis and all were up-regulated by TP508. Up-regulation of Egr- 1 is particularly impor- tant and worthy of further discussion. It is known that Egr-1 plays a vital role in immune response, cellular

H. Wung et ul. I Journul of’ Orthopedic Rrsrurcli 23 (2005) 6 7 1 4 7 9

Immunological Effect of TP508

611

1 1 1 Cellular effects Accelerating Increased immune

bone healing response

Fig. 5. Proposed model for TP508 up-regulating immunological response genes in accelerating fracture repair (see description in Discussion).

growth, development, differentiation and angiogenesis [6,14] through regulation of more than 30 target genes. We surveyed eight Egr-1 target genes, all of which are critical for angiogenesis/wound repair, and found that they were consistently up-regulated. As depicted in the cascade model proposed in Fig. 6, this data suggests that the single injection of TP508 used in these studies may act in part through the initiation of a cascade of events mediated by other growth factors, chemokines, and mol- ecules that may effect cellular proliferation and differ- entiation.

A specific example of the downstream effects of TP508 is basic fibroblast growth factor (bFGF/FGF2), one of the Egr-1 target genes known released from mac- rophages and one of the most thoroughly studied key factors for the induction of angiogenesis. It was mod- estly increased by TP508 at day 1 and reached a peak upregulation (2.5-fold, p < 0.05) at day 2 (Table 3). Based on the effects of direct bFGF injection into frac-

TP508 Cascade Effects

TP508

1 PDGFaJP

ICAM-I TNF-a

CXCR2 CXCR4

Early Response Factors Target Genes Cellularrhsue Effects

Fig. 6. Proposed model for TP508 cascade effects promoting early growth and angiogenesis in acceleration of bone healing based on microarray data (see Discussion for details).

tures [25], this up-regulation is expected to enhance the fracture repair process. We did not observe a significant change for another well-known potent angiogenesis- inducer, VEGF, which may be due to the early time points we chose. It is believed that bFGF sets the stage for angiogenesis during the first three days of wound re- pair, whereas VEGF plays a key role during the forma- tion of granulation tissue on days 4 through 7 [26]. We will test this hypothesis in our next series of experiments, which will involve later stages of fracture repair.

Coordinated up-regulation of angiogenic factors by TP508 motivated us to hypothesize that newly formed blood vessels at the TP508 treated fracture site should be increased in comparison to those at the saline-treated control site. An increase of 41% of blood vessels with TP508 treatment using systematic blinded histological analysis of calluses 3 weeks after fracture confirms that the altered expression of angiogenesis-related genes may have translated into increased angiogenesis at the cellular and tissue level. We further questioned to what extent increased angiogenesis has been reflected in the overall fracture repair. The 2 1-36% increase in bone strength with TP508 treatment confirms the therapeutic significance of this synthetic thrombin peptide for the overall fracture repair process.

It appears that enhanced angiogenesis by TP508 may play a major role in the acceleration of fracture repair. However, we cannot exclude other possible mechanisms because many of the angiogenesis-related genes listed in Table 2 have been shown to promote wound repair through angiogenesis-independent path- ways [45]. In addition, we observed several other genes including small espin, brush border myosin and synap- sin Ma, which have no established functions in angio- genesis, but are significantly up-regulated by TP508. Their potential contributions to the acceleration of bone healing deserves future investigation.

678 H. Wung el ul. I Journul of Orthopaedic Research 23 (2005) 6 7 1 4 7 9

In the present study, we have attempted to correlate changes in early gene expression (day 1 and 2) induced by TP508 with later histological and mechanical strength of bone. Although there appears to be a general correlation, the data suggests that there may be certain differences in the magnitude of these responses depend- ing on the dose of TP508 delivered to the fracture site. For example, mechanical testing results show increased bone strength at week 4 with both low dose (1 pg) and high dose (10 pg) treatment although the low dose was not significant. In contrast, the gene array data and his- tological evaluation showed enhanced effects with both low and high doses, but a larger effect with the low dose. It is possible that this is a temporal difference and that at later time points the high dose would have equivalent ef- fects on gene expression and blood vessel development, or that the mechanical testing of bones at a later time would show greater effects with the low concentration. Previous studies looking at TP508 effects on cartilage re- pair, for example, showed that there was a greater effect of a high dose at week 4, but that later the repair initi- ated by low and high dose were equivalent (manuscript submitted, Sheller et al.). It has also been noted previ- ously that at particular times after treatment of dermal wounds, TP508 appears to have a pronounced bi-phasic effect with lower dose having a greater effect than higher doses [9,27,40]. Similarly, TP508 stimulation of cell migration can also be bi-phasic [27]. Therefore, effects on immune cells and angiogenesis which involve cell migration may show more of an early bi-phasic response than later mechanical testing.

In summary, TP508 treatment enhances bone strength and promotes blood vessel formation in frac- ture repair. The increased induction of early growth fac- tors, activation or recruitment of inflammatory cells, and expression of inflammatory mediators and angio- genic genes may represent the molecular mechanisms underlying these changes.

Acknowledgments

We would like to acknowledge Jeremiah Convery and Tammy Bigelow for performing animal surgery, Emma Rousseau for assisting in data analysis, Amber Zwernemann for her technical assistance in blood ves- sel analysis, Mary Campbell, Jin-Ping Yang and David Simmons for their participation in the early phases of this project. This study was supported by OrthoLogic Corp.

References

[I] Barnes G, Kostenuik PJ, Gerstenfeld LC, Einhorn TA. Growth factor regulation of fracture repair. J Bone Miner Res 1999; l4( 1 1):1805-15.

[2] Beck LS, Wong RL, DeGuzman L, et al. Combination of bone marrow and TGF-el augment the healing of critical-sized bone defects. J Pharm Sci 1998;87:1379-85.

[3] Boehm U, Klam T, Groot M, Howard JC. Cellular responses to interferongamma. Annu Rev Immunol 1997;15:749-95.

[4] Bolander ME. Regulation of fracture repair by growth factors. Proc SOC Exp Biol Med 1992;200:165-70.

[5] Bonnarens F, Einhorn TA. Production of a standard closed fracture in laboratory animal bone. J Orthopaed Res 1984;2:

[6] Bryant M. Drew GM, Houston P, et al. Tissue repair with a therapeutic transcription factor. Hum Gene Ther 2000;ll: 2143-58.

[7] Burkhardt R, Kettner G, Bohm W, et al. Changes in trabecular bone, hematopoiesis and bone marrow vessels in aplastic anemia, primary osteoporosis, and old age: a comparative histomorpho- metric study. Bone 1987;8(3):15744.

[8] Burstein A, Frankel V. A standard test for laboratory animal bones. J Biomech 1971;4(2): 15S8.

[9] Carney DH, Mann R, Redin WR, et al. Enhancement of incisional would healing and neovascularization in normal rats by thrombin and synthetic thrombin receptor-activating peptides. J Clin Invest 1992;89:146%77.

[lo] Den Boer FC, Bramer JAM, Blokhuis TJ, et al. Effect of recombinant human osteogenic protein-I on the healing of a freshly closed diaphyseal fracture. Bone 2002;3 1 : 158-64.

[Il l Desai B, Meyer MH, Porter S, et al. The effect of age on gene expression in adult and juvenile rats following femoral fracture. J Orthop Trauma 2003;17(10):689-98.

[12] Einhorn TA. The cell and molecular biology of fracture healing. Clin Orthop 1998;355S:S7-22.

[I31 Einhorn TA, Majeska RJ, Rush EB, et al. The expression of cytokine activity by fracture callus. J Bone Miner Res 1995;

[14] Gashler A, Sukhatme VP. Early growth response protein (Egr-1): Prototype of a zinc-finger family of transcription factors. Nucleic Acid Res Mol Bio 1995;50:191-244.

[I51 Gerstenfeld LC, Cho T-J, Kon T, et al. Impaired fracture healing in the absence of TNF-a signaling: The role of TNF-u in endochondral cartilage resorption. J Bone Miner Res 2003; 18: 158492.

[I61 Glenn KC, Frost GH, Bergmann JS, Carney DH. Synthetic peptides bind to high-affinity thrombin receptors and modulate thrombin mitogenesis. Peptide Res 1988;1:65-73.

[17] Glowacki J. Angiogenesis in fracture repair. Clin Orthop 1998;355:S82-9.

[18] Hadjiargyrou M, Ahrens W, Rubin CT. Temporal expression of the chondrogenic and angiogenic growth factor CYR61 during fracture repair. J Bone Miner Res 2000;15:101423.

[I91 Hausman MR, Schaffler MB, Majeska RJ. Prevention of healing in rats by an inhibitor of angiogenesis. Bone 2001;29:560-1.

[20] Holzer G, Majeska RJ, Lundy MW, et al. Parathyroid hormone enhances fracture healing. A preliminary report. Clin Orthop

97-101.

10(8):1272-8 1.

- . . . 1999;366:258-63. Horner A, Bord S, Kelsall AW, et al. Tie2 ligands angiopoientin- 1 and angiopoietin-2 are coexpressed with vascular endothelial cell growth factor in growing human bone. Bone 2001;25(1):65-71. Karnaugh RD, Ryaby JT, Razzano P, et al. In vitro evaluation of the synthetic thrombin peptide, TP508, in articular cartilage repair. Trans ORS 2002;27:477. Liu L, Wang Y, Fan Y, et al. IFN-y activates cAMP1PKNCREB signaling pathway in murine peritoneal macrophages. J Interferon Res 2004;2433442. McKibbin B. The Biology of Fracture Healing in Long Bones. J Bone Joint Surg l978;60-B:55Ml. Nakamura T, Hara Y, Tagawa M, et al. Recombinant human basic fibroblast growth factor accelerates fracture healing by

H. Wang et ul. I Journal of Orthopardic Reseurcli 23 (2005) 6 7 1 4 7 9 679

enhancing callus remodeling in experimental dog tibial fracture. J Bone Miner Res 1998;13(6):942-9.

[26] Nissen NN, Polverini PJ, Koch AE, et al. Vascular endothelial growth factor mediates angiogenic activity during the proliferative phase of wound healing. AM J Pathol 1998;152(6):144-52.

[27] Norfleet AM, Bergmann JS, Carney DH. Thrombin peptide, TP508, stimulates angiogenic responses in animal models of dermal wound healing, in chick chorioallantoic membranes, and in cultured human aortic and microvascular endothelial cells. Gen Pharmacol 200035:249-54.

[28] Norfleet AM, Huang Y, Sower LE, et al. Thrombin peptide TP508 Accelerate closure of dermal excisions in animal tissue with surgically induced ischemia. Wound Repair Regen 2000;8:517-29.

[29] Paralkar VM, Borovecki F, Ke HZ, et al. An EP2 receptor- selective prostaglandin E2 agonist induces bone healing. PNAS (USA) 2003; 100:6736-40.

[30] Pilla AA, Mont MA, Nasser PR, et al. Non-invasive low-intensity pulsed ultrasound accelerates bone healing in the rabbit. J Orthop Trauma 19904(3):24&53.

[31] Prasad GC, Udupa KN. Studies on ultrastructural pattern of osteogenic cells during bone repair. Acta Orthop Scand 1972;43: 163-75.

[32] Pufe T, Wildemann B, Petersen W, et al. Quantitative measure- ment of the splice variants 120 and 164 of angiogenic peptide vascular endothelial growth factor in the time flow of fracture healing: a study in the rat. Cell Tissue Res 2002;309:387-92.

[33] Reno C, Marchuk L, Sciore P, et al. Rapid isolation of total RNA from small samples of hypocellular, dense connective tissues. BioTechniques 1997;22: 10824.

[34] Rhinelander FW. Tibia1 blood supply in relation to fracture healing. Clin Orthop 1974;105:3481.

[35] Rundle CH, Miyakoshi N. Kasukawa Y, et al. In vivo bone formation in fracture repair induced by direct retroviral-

based gene therapy with bone morphogenetic protein-4. Bone 2003;32:591401.

[36] Ryaby JT, Carney DH, Campbell M, et al. Acceleration of fresh fracture healing with an injectable thrombin peptide in a rat model. Trans ORS 2000;25:877.

[37] Sheller MR, Crowther RS, Kinney JH, et al. Repair of rabbit segmental defects with the thrombin peptide, TP508. J Orthop Res 2004;22: 10949.

[38] Simmons DJ. Fracture healing perspectives. Clin Orthop Relat Res 1 98 5;200 100- 12.

[39] Simmons DJ, Yang J, Yang S, et al. Acceleration of rat femoral fracture healing by a synthetic thrombin peptide. In: Danks J, Dacke C, Flik G, Gay C, editors. Calcium metabolism: compar- ative endocrinology: Proceedings of satellite meeting. Bradley Stoke, Bristol: BioScientifica Ltd.; 1999. p. 1-15.

[40] Stiernberg J, Norfleet AM, Redin WR, et al. Acceleration of full- thickness wound healing in normal rats by the synthetic thrombin peptide, TP508. Wound Repair Regen 2000;8:20415.

[41] Street J, Bao M, DeGuzman L, et al. Vascular endothelial growth factor stimulates bone repair by promoting angiogenesis and bone turnover. PNAS (USA) 2002;99:965&61.

[42] Street J, Winter D, Wang JH, et al. Is human fracture hematoma inherently angiogenic? Clin Orthop 2000;378:224-37.

[43] Tusher VG, Tibshirani R, Chu G. Significance analysis of microarrays applied to the ionizing radiation response. Proc Natl Acad Sci USA 2001;98(9):5116-21.

[44] Welch RD, Jones AL, Bucholz RW, et al. Effect of recombinant human bone morphogenetic protein-2 on fracture healing in a goat tibial fracture model. J Bone Miner Res 1998;13:1483- 90.

[45] Yates S, Rayner TE. Transcription factor activation in response to cutaneous injury: role of AP-I in reepithelialization. Wound Repair Regen 2002(1):5-15.