Expression of matrix metalloprotease and tissue inhibitor of metalloprotease genes in human anterior cruciate ligament

Journal of Orthopaedic Research

Journal of Orthopaedic Research 19 (2001) 642-649

www.elsevier.nl/locate/orthres

Expression of matrix metalloprotease and tissue inhibitor of metalloprotease genes in human anterior cruciate ligament Marcus J. Foos a

John R. Hickox a,b, Paul G. Mansour Daniel M. Hardy a,b**

b,

James R. Slauterbeck ',

Department of Cell Biology and Biochemistry, Texas Tech. Unioersity Health Sciences Center, 3601 Fourth Street, Lubbock, T X 79430. USA Department Orthopaedic Surgery. Tech. University Health Sciences Center, 3601 Fourth Street, Lubbock, T X 79430, USA

Received 23 June 1999; accepted 11 February 2000

Abstract Women are more susceptible to anterior cruciate ligament (ACL) injuries than men performing similar athletic activities. Because tissue remodeling may affect ligament strength, we assessed expression of tissue remodeling effector genes in the human ACL. Specifically, we surveyed ACL for RNAs encoding all known matrix metalloproteases (MMPs) and tissue inhibitors of metalloproteases (TIMPs) by reverse transcription/polymerase chain reaction (RT-PCR). These experiments revealed that mRNAs encoding nine of sixteen MMPs and all four TIMPs are present in the normal ACL. The nine expressed proteases were MMPs 1-3, 7, 9, 11, 14, and 17 (collagenase 1, gelatinase A, stromelysin 1, matrilysin, gelatinase B, stromelysin 3, and membrane types 1 and 4, respectively), and MMP-18. Genes for MMPs 8, 10, 12, 13, 15, and 16 appeared not to be expressed in ACL, as their mRNAs were not detected using RT-PCR conditions that did yield positive signals from other tissues (testis or bone). We conclude that numerous genes encoding tissue remodeling effector proteins are expressed in the human ACL. 0 2001 Orthopaedic Research Society. Published by Elsevier Science Ltd. All rights reserved.

Introduction Females tear their ACLs 3-10 times more frequently than males participating in similar athletic events [2,33]. The reason for this discrepancy is not known. Many possible causative factors such as size, strength, anatomic, social and hormonal differences have been suggested [I 5-17], but the relative contributions of these factors to female ACL injury have not been established. The possible involvement of normal tissue remodeling events in susceptibility to ACL injury has not been evaluated. Tissue remodeling occurs continuously in both normal and injured tissues. In this process, old or damaged structures are degraded and replaced with newly synthesized molecules [7,8,10,1I]. The balance between the degradative and biosynthetic arms of this process is controlled by activities of matrix metalloproteases (MMPs) and their inhibitors (tissue inhibitors of me-

* Corresponding author. Tel.: + 1-806-743-2053; fax: +1-806-7432990. E-mait address: cbbdmh@,ttuhsc.edu (D.M.Hardy).

talloproteases, TIMPs) [7,10,11]. MMPs are Zn2+-dependent endoproteinases that were initially classified into three groups (collagenases, gelatinases, and stromelysins) based on their substrate-specificities[ 1I]. TIMPs are proteins that block the hydrolytic activities of MMPs [9,23]. Together, TIMPs and MMPs function as effector molecules that regulate tissue remodeling. For example, collagenases and TIMPs mediate extracellular matrix remodeling events in the cutaneous reparative process [34,35]. In general, during the degradative component of tissue remodeling, the influence of MMPs is greater than TIMPs [8,18,27,29], and the opposite is true in the reparative process. MMPs are expressed by many cell types, including macrophages, neutrophils, fibroblasts, trophoblasts, endometrial cells, epithelial cells, and various tumor cells. These enzymes are also present in connective tissues (e.g., cartilage), and in synovial fluid [22,24,26]. Estrogen and progesterone regulate the transcription of MMP and TIMP genes in various tissues [24,28,31,371. MMP activity has been demonstrated in the ACL [1,14,22], and results of several studies suggest that steroids affect MMP and/or TIMP gene expression in this tissue [12,19,20,28,32,37].

0736-0266/01/$ - see front matter 0 2001 Orthopaedic Research Society. Published by Elsevier Science Ltd. All rights reserved. PII: S 0 7 3 6 - 0 ~ 6 6 ( 0 0 ) 0 0 0 7 l1-

M.J. Foos et al. I Journal of Orthopaedic Research 19 (2001) 642449

Our working hypothesis is that gender differences in tissue remodeling contribute to the disparate susceptibilities of males and females to ACL injury. A key first step in testing this hypothesis is to determine which effectors of tissue remodeling are present in the human ACL. Accordingly, in this study we used (RT-PCR) to identify all known or putative MMPs and TIMPs in human ACL. The results represent the first systematic assessment of tissue remodeling components in the ACL.

Materials and methods Tissue

With Institutional Review Board approval, whole ACLs were obtained from total knee arthroplasties except for one tissue that was from an amputation (Table 1). Each of these ACLs was undamaged and appeared to be free of aggressive inflammation. The ACLs were rinsed with saline, flash-frozen in N2 [l] within 1 min of removal, and stored at -80°C for up to 1 year. ACLs were discarded if they exhibited gross structural abnormality or a pannus indicative of an inflammatory response. Other human tissues used as controls (foreskin, placenta, and bone) were similarly processed. For histology, ACLs from two total knee arthroplasty (TKA) patients and from a fresh cadaver (60 year old female with healthy knees and an apparently normal ligament) were fixed in formalin and processed for routine H & E staining. Nucleic acid isolations and first-strand cDNA synthesis

Total RNA was isolated from ACL using the guanidinium thiocyanate/acidic phenol extraction method [3] with modifications that proved to be essential for this tissue. Specifically, frozen ACLs were pulverized under N2 [I] prior to suspension in guanidinium thiocyanate solution, and in the precipitation step the equal volume of isopropanol was replaced with 0.025 volumes of 1.2 M sodium citrate, 0.025 volumes of 0.8 M sodium chloride, and 0.25 volumes of isopropanol. Testis total RNA was purchased (Clontech Laboratories, Palo Alto, CA, USA). RNA isolations from other tissues were done as for ACL. RNA quality was assessed by formaldehyde-agarose electrophoresis [13,30], and by RT-PCR using primers for human glyceraldehyde-3phosphate dehydrogenase. RNA yields were estimated either by comparing ethidium bromide staining intensity to known amounts of testis RNA in agarose gels, or fluorimetrically using RiboGreen (Molecular Probes; Eugene, OR). Human genomic DNA was isolated from blood using Proteinase K in a kit format (Gentra Systems, Minneapolis, MN, USA), and quantified by the Saran Wrap method [30]. First-strand cDNA synthesis was by reverse transcription for 60 min at 37°C with an oligo-(dT) primer, murine Moloney leukemia virus reverse transcriptase (RT; Life Technologies, Rockville, MD, USA), and total RNA as template [30,381. Immediately after reverse

643

transcription, 0.5 p1 ( 5 U) RNAse ONE (Promega, Madison, WI, USA) was added to the reaction, and the mixture was incubated 10 min more at 37°C to hydrolyze RNA in DNA-RNA hybrids. Firststrand cDNAs from seven different patients of various age and sex were pooled (Table I ) for use as PCR templates. Polymerase chain reaction ( P C RJ

PCR primer pairs (20-26 bases, Table 2) were designed using the Primerselect program of the Lasergene software suite (DNASTAR, Madison, WI, USA). PCR was performed in an air thermocycler (Rapidcycler, Idaho Technology, Idaho Falls, ID, USA) using glass capillary sample tubes [13]. Reaction composition was: 0.5 pM each primer, 5-10 ng cDNA or 30 ng genomic DNA template, and 0.04 units/pL Taq DNA polymerase in a total volume of 10 pL containing 50 mM TRIS pH 8.3, 250 pg/ml bovine serum albumin, 2%)sucrose, 0.1 mM cresol red, and 2 4 mM MgCI,. Optimal cycling conditions were determined empirically by testing different annealing temperatures in 1" or 2°C increments over a range of up to 26°C in 2, 3 , or 4 mM MgCI,. PCR consisted of an initial 30 s denaturation at 94°C. 30-35 amplification cycles, and a final extension at 72°C for 30 s. Each amplification cycle consisted of template denaturation at 94°C for 0 s. primer annealing for 0 s at the empirically determined optimum, and product extension at 72°C for 15-20 s. Products were resolved by 2%) agarose gel electrophoresis in Tris-borate/EDTA buffer, and visualized by staining with ethidium bromide [30]. cDNA cloning and sequencing

PCR products obtained from human ACL template were cloned directly into pGEM-Teasy by T/A cloning, and both strands of the inserts were sequenced by the dideoxy method [13,30]. The sequences obtained were compared to known MMP and TIMP sequences by pairwise alignment using the Martinez/Needleman-Wunsch algorithm (Lasergene software).

Results Fig. 1 shows the histology of a typical ACL obtained from TKA and of an ACL from a fresh cadaver with no apparent knee pathology. The basic morphology of the TKA specimen (Fig. l(a) and (b)) was indistinguishable from that of the normal ACL (Fig. l(c) and (d)). We have not detected obvious increases in cellular response (lymphocytes, plasma cells, or neutrophils) in either the synovia or the dense connective tissue of ACLs from osteoarthritic knees, although synovial hyperplasia is evident in these specimens (Fig. l(d)). To establish methods for detecting mRNAs encoding known human MMPs and TIMPs, their cDNA sequences (accession numbers listed in Table 2) were used

Table 1 Ages and sexes of patients whose ACL first-strand cDNAs were pooled for analysis of MMP and TIMP expression by RT-PCR Patient

Sex

F A B M C M F D E M F F F G F H a TKA = Total knee arthroplasty.

Age

Procedure

Primary diagnosis

35 40 60 65 69 82 84 87

Amputation TKA" TKA TKA TKA TKA TKA TKA

Trauma Osteoarthritis Osteoarthritis Osteoarthritis Osteoarthritis Osteoarthritis Osteoarthritis Osteoarthritis

M.J. Foos et al. I Journal of Orthopaedic Research 19 (2001) 642449

644

Table 2 Primer sequences and conditions for detecting MMP and TIMP mRNAs by RT-PCR mRNA MMP-1

Accession # X05231

MMP-2

503210

MMP-3

X05232

MMP-7

X07819 YO0728 J05556

MMP-8

Primer sequences (S =sense, A = antisense) S: CAGGTATTGGAGGGGATGCT A: GTGGCCAGAAAACAGAAGTGAA S: GCCTGAGCTCCCGGAAAAGATTGAT A: CAGCAGCCTAGCCAGTCGGATTTGA S: GATGCGCAAGCCCAGGTGTG A: GCCAATTTCATGAGCAGCAACGAG S: TAAACTCCCGCGTCATAGAAATAAT A: TGAGTTGCAGCATACAGGAAGTT S 1: ATATTAAGTCATTGTTTCCCATCACa Al: ACCCACGAAACATATCATCTACATT"

S2:TGACGGGGAAGCCAAATGAGGA MMP-9

505070

MMP- 10

X07820 YO0728 X57766

MMP-11

MMP-12

L23808

MMP-13

X75308

MMP-14

X90925

MMP-15

D86331

MMP- 16

D85511

MMP- 17

X89576

MMP-18

X92521

MMP-20

Y12779

A4

AA4 10822

A8

AA830159

TIMP-1

X03124

TIMP-2

505593

TIMP-3

U02571

A2: AGGAGTGAGCGAGCCCCAAAGAAT S: GGCCAACTACGACACCGACGAC A: CGCCGCCACGAGGAACAAAC S 1: CTTTTGATGGCCCAGGACACAGT A1 : AGGGCCAAAATGCAGAAATCAAA S 1: AGCCTGCCAACACTmCCTCTGACCa A 1: CCCCTCCCCATTTGACTGTGAACTTa S2: CTGGGGTCCCGAGAAGAACAAGAT A2: GTGCCCACAGCCACAAAGATGG S: ACCGGGCAACTGGACACATCTACC A GAGCTCCACGGGCAAAAACCAC S: AGGAGCATGGCGACTTCTACCC A: TTTGTCTGGCGTTTTTGGATGTTTA S: ATGAGGCGCCCCCGATGTGG A: TCCAATGTTGGGGCCTGGGAAGTAG S1: GCCCGGCTACCCCAAGTCCATCCTG" Al: GGCGCCCCCTGAGCACCGTTAGCA* S2: GCTTCGCOOOGAGATGTTCGTGTTCA A2: CTCCATCCGCAGGCGCTCATTGTCG S: GCCCCCACACCGCTCTATTCCT A: TTCCCGACGTCCTCCCACCAA S:TGCCGGAGCCCCCAGACAAC A GAGACGGGGCGCGGGTATCCTT S: GCTCCCCCAAGGCTCCCAGAAATC A: CTCCAGCGGCCCAGCAACAGGT S: GGCCCATTCCACAGACCCATCAG A: TGCCCCTCTCAGCCACTTCGTAA S: CCCTCATGGCCCAGTCTACGA A: GGGCCCGGTCCTGAATCTCTA S: ACGCCTGGGCCTGGGCACTCCT A: GGTTTGGGCGCCTTCCCTTGTGAGG S: CCCCTGGCTTCTGGGCATCCTGTTG A CTCCATGGCGGGGGTGTAGACGAA S: CCCCCTCCTCGGCAGTGT A: CCGGGGGAGGGAGATGTAGCA S: CGGGGCTGTGCAACTTCGTGGAGA

IMa2+1

T,,,,,i"C

Product

3

56

3

62

402 bp (~ro'*~-pro~~~) 444 bP

3

50

3

56

3

46

3

44

3

66

3

62

3

66

3

61

3

62

3

62

2

60

3

62

3

68

3

62

3

66

3

66

3

60

3

66

406 bp ( Met87-Gly22') 421 bp (Le~~~-His~'~) 413 bp (all 3'UTR) 465 bp (~al~~-~er~~*) 424 bp (Ala260-Ala4W) 486 bp ( Ser'72-Phe33') 597 bp (G I u ~ * ~ - ~ ' U T R ) 383 bp (Tr~'~~-Ala'~~) 277 bp (Thr76-Ala'67) 341 bp (Lys'70-Lys2*3) 369 bp (Met88-Gly2'0) 414 bp (Prou' -3'UTR) 499 bp (Le~~~~-Glu~~) 432 bp (~ro~"~-~ys~~~) 331 bp (~eu~~*-~er~~*) 386 bp (5'UTR -ArgIo5) 371 bp ( Ala235-Gly357) 280 bp

2

60

243 bp

3

66

3

56

3

56

255 bp (Pro6-G1u9') 234 bp (A1a92-Pro'69) 335 bp (Thr1"-3'UTR) 556 bp (GlyIS7-3'UTR) 883 bp I11e4*-3'UTRI

( P ~ O ~ ~ * - C) ~ S ~ * ~

A:CAGCGGGAAGGGAGGGAAGTGAG TIMP-4

U76456

S1: GCTGCCAAATCACCACCTGCTACACa 3 Al: GACATTCGCCATTTCTCCCCTACCAGa S2: TTCGGCCCAAAATCTCCAGTG A2: GACATTCGCCATTTCTCCCCTACCA These primer pairs yielded products from control reactions using -RT and/or genomic DNA templates.

to design PCR primer pairs. Major criteria for primer design were sequence uniqueness, absence of stable hairpins, absence of stable self or pair dimers, and matched melting temperatures. In addition, where genomic sequences were available, primer pairs that flanked introns were chosen. The resulting primer pairs were used to establish PCR conditions for detecting the

66 56

various MMP and TIMP mRNAs (Table 2). For all but one gene, the listed PCR primer pairs and conditions successfully amplified the predicted PCR product bands (size range 234-883 bp) from ACL, bone, or testis templates. No MMP-20 RT-PCR product was obtained using templates from ACL, testis, bone, placenta, or foreskin (data not shown). For some targets (TIMP-4

M.J. Foos et al. I Journul of Orthopuedic Research I 9 (2001) 6 4 2 4 4 9

645

Fig. I. Histology of an ACL from a TKA patient (panels a and b) and of the post-mortem ACL (panels c and d). Both tissues were from from females of similar age. Panels a and c: dense, regular connective tissue of the ACL. Panels b and d: synovial tissue associated with the ligments. Note the characteristically sparse, spindle-shaped nuclei of fibroblasts in the dense, regular connective tissue, and the general absence of cellular infiltration in either view of the TKA speciman. Bar = 100 pm

and MMPs 8, 11, & 15) multiple primer pairs were tested because initial experiments yielded products with mock reverse-transcribed RNA (-RT) or genomic templates. When this occurred, new primers were tested until a pair that appeared to flank one or more intron was obtained. However, for the A8 target, available sequence was too limited for designing an intron-flanking primer pair. Using these optimized methods, we detected MMP mRNAs in ACL using ten unique PCR primer pairs. Specifically, RT-PCR products of the expected sizes were obtained using primers specific for MMPs 1,2, 3,7, 9, 11, 14, 17, 18, and A4 (Fig. 2). The DNA sequences of the cloned PCR products were identical to their expected target sequences. Sequence analysis further revealed that the A4 partial sequence is identical to a region of MMP-18, despite its presence in Genbank as a distinct entry. Accordingly, the A4 and MMP-18 primer pairs both yielded products from the ACL (+RT) template. PCR product bands for MMPs 3 and 9 were particularly strong using reverse-transcribed RNA from ACL as template. Four faint products amplified in the control reactions for MMP-3 were judged to be non-

specific because they were not the expected size for this MMP, and we obtained a prominent single product band of the appropriate size and sequence in the +RT reaction (Fig. 2). RT-PCR products for MMPs 8, 10, 12, 13, 15, and 16 were not obtained using reverse-transcribed RNA from ACL as template (Fig. 3). However, products for these MMPs were successfully amplified from either bone or testis. Finally, the predicted PCR products for all four known TIMPs were obtained using reverse-transcribed ACL RNA as template (Fig. 4).

Discussion

Tissue remodeling may contribute to the increased susceptibility of female athletes to ACL injury. In this study we used RT-PCR to assess expression of all known MMPs and TIMPs in the human ACL. The results represent the first comprehensive survey of MMP and TIMP gene expression in human ACL. Sex hormones are known to regulate expression of MMPs and TIMPs in various tissues [24,28,37]. To assure that various hormone profiles were represented in

646

M.J. Foos et al. I Journal of Orthopedic Research 19 (2001) 642449

Fig. 3. Detection of MMP mRNAs in testis or bone, but not ACL total RNA, by RT-PCR. Shown are relevant regions of ethidium bromidestained agarose gels (2%). Tissues from which RNAs were isolated are indicated below each lane. Lane headings specify the templates used for PCR: +RT = reverse-transcribed ACL, bone, or testis RNA; -RT= mock reverse- transcribed bone or testis RNA. Migration and sizes (in bp) of double-stranded DNA markers are indicated to the left of each panel.

Fig. 2. Detection of MMP mRNAs in ACL total RNA by RT-PCR. Shown are relevant regions of ethidium bromide-stained agarose gels (2%). Lane headings specify the templates used for PCR: +RT=reverse-transcribed ACL RNA; -RT = mock reverse-transcribed ACL RNA; G=genomic DNA. Migration and sizes (in bp) of doublestranded DNA markers are indicated to the left of each panel.

our experiments, the template for RT-PCR was a pool of ACL first strand cDNAs from seven different patients of varying age and gender. The results therefore provide a qualitative view of both MMP and/or TIMP gene expression in the human ACL irrespective of the unique sex hormone status of each individual. We detected mRNAs encoding nine of sixteen known MMPs and all four known TIMPs in human ACL. Target mRNAs were considered to be present in ACL if: (1) a PCR product of the correct size and sequence was obtained from a template of reverse-transcribed ACL RNA; (2) no product was obtained using the identical conditions with mock reverse-transcribed (no reverse transcriptase) ACL RNA as template; and (3) no product was obtained using the same PCR conditions

with 30 ng genomic DNA as template (an amount that is larger than any possible genomic DNA contamination of our RNA preparations). The presence of nine MMP and four TIMP mRNAs in the ACL demonstrates that their respective genes are expressed, either constitutively or transiently, in this tissue. Although it is formally possible that the expressed mRNAs are not translated, it seems likely that the protein products of these genes are indeed present in the ACL. Several MMP primer pairs failed to amplify products from ACL template. We considered target mRNAs to be absent from ACL if: (1) a product of the correct size was obtained with a template of reverse-transcribed RNA from a positive control tissue; (2) no product was obtained with reverse-transcribed ACL RNA using the identical PCR conditions; and (3) no product was obtained with a template of mock reverse-transcribed RNA from the positive control tissue. Using these criteria to evaluate the data, we conclude that six known MMPs are either not expressed in human ACL or are present at comparatively low levels. A more sensitive method, such as reamplification with nested PCR primers, might detect expression of additional MMP RNAs. The failure of the MMP-20 primer pair to yield

Af J. Foos el al. I Journal of Orthopaedic Research 19 (2001) 6 4 2 4 4 9

Fig. 4. Detection of TIMP mRNAs in ACL total RNA by RT-PCR. Lane headings and other details are as for Fig. 2.

an RT-PCR product from ACL, testis, bone, placenta, or foreskin is consistent with the very restricted expression of this gene in tooth tissue [21]. Although the results suggest that the MMP-20 gene is not expressed in the ACL, we cannot confirm this without a positive result from another tissue. The structural element of the ACL is dense, regular connective tissue with relatively few cells, nearly all of which are fibroblasts. Because this dense connective tissue is hypovascular, proteins and small molecules move through the extracellular space by diffusion. Synovial tissue associated with the ACL comprises two major cell types, fibroblasts and the macrophage-like synoviocytes. No basal lamina is present between the vascular synovial tissue and the dense, regular connective tissue of the ACL, so secreted products of synoviocytes and fibroblasts are free to diffuse and act on substrates throughout the tissue. MMPs or TIMPs produced by either of these cell types could therefore participate in tissue remodeling events that affect ACL strength. The cloned PCR products we generated in the course of these studies can now be used as probes to identify the cell types that express MMP and TIMP mRNAs. Our results may reflect the complement of MMP and TIMP genes typically expressed in the normal ACL, with some limitations. Truly normal ACLs are rarely

647

available, as they must be obtained post-mortem or from amputations (as was one of our specimens). Consequently, most of our ACLs were from TKA patients, which raises concerns that the mRNAs detected might be present secondary to injury. However, we carefully examined ACLs when they were removed, and discarded all potentially abnormal tissues. Histological evaluation of multiple ACLs selected in this way revealed no immune cell infiltration, so it is unlikely that the MMP and TIMP mRNAs we identified are derived from gross inflammation. We cannot rule out the possibility that a few inflammatory cells are present in occasional specimens; such low-level inflammation may be a normal component of the repairhemodeling process in healthy ACLs. We also cannot rule out the possibility that the synovial hyperplasia we observed reflect an alteration in the physiology of our specimens by cytokines or other external factors. In addition to the implications of this work to ACL physiology, the results provide information about the organization of the TIMP 4 and MMP 8, 1I , and 15 genes. We obtained the same, predicted products from cDNA and genomic templates with our initially designed primer pairs for detecting these mRNAs. Such results are possible only if the primers do not flank introns. Therefore, we can infer that there are no introns in the initially amplified regions of these targets (Table 2). The results also revealed that testis is a rich source of MMPs and TIMPs, as it proved to be a useful control tissue for detection of mRNAs that were not expressed in the ACL. Although the functions of these MMPs and TIMPs in the testis are not known, it seems likely that they mediate the tissue remodeling that must occur for proper spermatid development and spermiation. The biochemical basis of extracellular matrix remodeling has not been fully elucidated. While the literature generally supports the view that the MMP/TIMP axis is a key component of this process, regulation of protease activation and substrate availability could also be important. In addition, ADAM (a disintegrin and metalloprotease) proteins are expressed by chondrocytes in articular cartilage [4,25], and their metalloprotease activities include the ability to hydrolyze the articular proteoglycan aggrecan [36]. ADAMs might also hydrolyze collagen or other important structures in ligament. Further study will be required to determine the relative contributions of MMPs, TIMPs, and other proteins to ACL matrix metabolism. In conclusion, our results reveal that numerous tissue remodeling effector genes are expressed in human ACL. They also extend earlier studies that identified a limited number of MMPs in synovial fluid surrounding injured ACLs [5,6]. The methods established in this study can now be used to determine how sex hormones affect expression of these genes in the ACL and perhaps thereby

M.J. Foos et al. I Journal of Orthopaedic Research 19 (2001) 6 4 2 6 4 9

648

cause gender differences in susceptibility of the ligament to injury.

Acknowledgements

We thank Dr. Greg Stocks and Dr. Eugene Dabezies, TTUHSC Department of Orthopaedics, for providing ACLs, Dr. Daniel McGunegle, TTUHSC Department of Obstetrics & Gynecology, for providing placenta, and Kevin Bi for providing foreskin. We also thank Dr. James Hutson, TTUHSC Department of Cell Biology & Biochemistry, TTUHSC and Dr. Suzanne Graham, TTUHSC Department of Pathology, for help with ACL histology. This work was supported in part by the TTU Center for Applied Biomechanics, by a TTUHSC Basic/ Clinical Seed Grant, (to JRS and DMH), a grant from the Orthopaedic Research and Education Foundation, (to JRS and DMH), and by National Institutes of Health grant HD-35166 (to DMH).

References h i e l D, Ishizue KR, Harwood FL, Kitabayashi L, Akeson WH. Injury of the anterior cruciate ligament: the role of collagenase in ligament degeneration. J Orthop Res 1989;7(4): 486-93. Arendt EA. Orthopaedic issues for active and athletic women. Clin Sports Med 1994;13:483-503. Chomczynski P, Sacchi N. Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem 1987;162:1569. Chubinskaya S, Cs-Szabo G, Kuettner KE. ADAM-I0 message is expressed in human articular cartilage. J Histochem Cytochem 1998;46:723-9. Chubinskaya S, Huch K, Mikecz K, Cs-Szabo G, Hasty KA, Kuettner KE, Cole AA. Chondrocyte matrix metalloproteinase-8: up-regulation of neutrophil collagenase by interleukin-1 beta in human cartilage from knee and ankle joints. Lab Invest 1996;74:23240. Dahlberg L, Friden T, Roos H, Lark MW, Lohmander LS. A longitudinal study of cartilage matrix metabolism in patients with synovial fluid concentrations of cruciate ligament rupture aggrecan fragments, stromelysin-1 and tissue inhibitor of metalloproteinase-1. Br J Rheumatol 1994;33:1107-11. Di Girolamo N, Lloyd A, McCluskey P, Filipic M, Wakefield D. Increased matrix metalloproteinases in the aqueous humor of patients and experimental animals with uveitis. Curr Eye Res 1996;15:1060-8. Edwards DR, Leco KJ, Beaudry PP, Atadja PW, Veillette C. Riabowol KT. Differential effects of transforming growth factorbeta 1 on the expression of matrix metalloproteinases and tissue inhibitors of metalloproteinases in young and old human fibroblasts. Exp Gerontol 1996;31:207-23. Edwards DR, Beaudry PP, Laing TD, Kowal V, Leco KJ, Leco PA, Lim MS. The roles of tissue inhibitors of metalloproteinases in tissue remodelling and cell growth. Int J Obes Relat Metab Disord 1996;2O(Suppl. 3):S9-S15. Everts V, van der Zee E, Creemers L, Beertsen W. Phagocytosis and intracellular digestion of collagen, its role in turnover and remodelling. Histochem J 1996;28:22945. -

[ll] Gaire M, Magbanua Z, McDonnell S, McNeil L, Lovett DH, Matrisian LM. Structure and expression of the human gene for the matrix metalloproteinase matrilysin. J Biol Chem 1994;269: 203240. [12] Hamlet WP, Liu SH, Panossian V, Finerman GA. Primary irnmunolocalization of androgen target cells in the human anterior cruciate ligament. J Orthop Res 1997;15(5):657-63. Hardy DM, Garbers DL. A sperm membrane protein that binds in a species-specific manner to the egg extracellular matrix is homologous to von Willebrand factor. J Biol Chem 1995;270: 26025-8. Harper J, Amid D, Harper E. Collagenase production by rabbit ligaments and tendon. Connective Tissue Res 1988;17:253-9. Haycock CE, Gillette JV. Susceptibility of women athletes to injury. Myths vs reality. JAMA 1976;236:163-5. Ireland ML. Special concerns of the female athlete. Sports Injuries 1994;13:153-88. Knapik JJ, Bauman CL, Jones BH, Harris JM, Vaughan L. Preseason strength and flexibility imbalances associated with athletic injuries in female collegiate athletes. Am J Sports Med 1991;19:76-81. Kozaci LD, Buttle DJ, Hollander AP. Degradation of type I1 collagen, but not proteoglycan, correlates with matrix metalloproteinase activity in cartilage explant cultures. Arth Rheum 1997;40:16&74. Liu SH, AL-Shaikh R, Panossian V, Yang RS, Nelson SD, Soleiman N, Finerman GA, Lane JM. Primary immunolocalization of estrogen and progesterone target cells in the human aneterior cruciate ligament. J Orthop Res 1996;14(4):526-33. Liu SH. AL-Shaikh RA, Panossian V, Finerman GA, Lane JM. Estrogen affects the cellular metabolism of the aneterior cruciate ligament. A potential explanation for female athletic injury. Am J Sports Med 1997;25:704-9. Llano E, Pendas AM, Knauper V, Sorsa T, Salo T, Salido E, et al. Identification and structural and functional characterization of human enamelysin (MMP-20). Biochemistry 1997;36:15101-8. Lohmander LS, Roos H, Dahlberg L, Hoerrner LA, Lark MW. Temporal patterns of stromelysin-1, tissue inhibitor, and proteoglycan fragments in human knee joint fluid after injury to the cruciate ligament or meniscus. J Orthop Res 1994;12(1):21-8. Mankin HJ, Mow VC, Buckwalter JA, Iannotti JP, Ratcliffe A. Form and function of articular cartilage. In: Simon SR, editor. Orthopaedic basic science. American Academy of Orthopaedic Surgeons; 1994. p. 1 4 . Matrisian LM. Matrix metalloproteinase gene expression. Ann N Y Acad Sci 1994;732:42-50. McKie N, Edwards T, Dallas DJ, Houghton A, Stringer B, Graham R, et al. Expression of members of a novel membrane linked metalloproteinase family (ADAM) in human articular chondrocytes. Biochem Biophys Res Commun 1997; 230:335-9. Meikle MC, Hembry RM, Holley J, Horton C, McFarlane CG, Reynolds JJ. Immunolocalization of matrix metalloproteinases and TIMP-1 (tissue inhibitor of metalloproteinases) in human gingival tissues from periodontitis patients. J Periodontal Res 1994;29:1 18-26. Pinals RS. Mechanisms ofjoint destruction, pain and disability in osteoarthritis. Drugs 1996;52(Suppl. 3):14-20. Rajabi MR, Dodge GR, Solomon S, Poole AR. Immunochemical and immunohistochemical evidence of estrogen-mediated collagenolysis as a mechanism of cervical dilatation in the guinea pig at parturition. Endocrinology 1991;128:371-8. Salamonsen LA, Woolley DE. Matrix metalloproteinases in normal menstruation. Hum Reprod 1996;l l(Supp1. 2):124-33. Sambrook J, Fritsch EF, Maniatis T. Molecular cloning: a laboratory manual. Cold Spring Harbor: Cold Spring Harbor Laboratory Press; 1989.

M.J. Foos et al. I Journal of Orthopaedic Research 19 (2001) 64,7449

[31] Schneikert J, Peterziel H. Defossez PA, Klocker H, Launoit Y, Cato AC. Androgen receptor-Ets protein interaction is a novel mechanism for steroid hormone-mediated down-modulation of matrix metalloproteinase expression. J Biol Chem 1996;271: 23907-13. [32] Sciore P, Frank CB, Hart DA. Identification of sex hormone receptors in human and rabbit ligaments of the knee by reverse transcription-polymerase chain reaction: evidence that receptors are present in tissue from both male and female subjects [Citation]. J Orthop Res 1998;16(5):604-10. [33] Slauterbeck J. Epidemiology and comparison of ACL injuries in male and female collegiate soccer athletes. Orthop Trans 1995. [34] Stricklin GP. Localization of mRNAs representing collagenase and TIMP in sections of healing human burn wounds. Am J Path 1993;143:1657-66.

649

[35] Stricklin GP, Li L, Nanney LB. Localization of mRNAs representing interstitial collagenase, 72-kda gelatinase, and TIMP in healing porcine bum wounds. J Invest Dermatol 1994;103: 352-8. [36] Tortorella MD, Pratta MA, Fox JW, Arner EC. The interglobular domain of cartilage aggrecan is cleaved by hemorrhagic metalloproteinase HT-d (atrolysin C) at the matrix metalloproteinase and aggrecanase sites. J Biol Chem 1998;273:5846-50. [37] Wahl LM, Blandau RJ, Page RC. Effect of hormones on collagen metabolism and collagenase activity in the pubic symphysis ligament of the guinea pig. Endocrinology 1977;100:571-9. [38] Williams SC, Du Y, Schwartz RC, Weiler SR, Ortiz M, Keller JR, Johnson PF. C/EBPepsilon is a myeloid-specific activator of cytokine, chemokine, and macrophage-colony-stimulating factor receptor genes. J Biol Chem 1998;273:13493-501.