Response of knee fibrocartilage to joint destabilization.

Osteoarthritis and Cartilage xxx (2015) 1e11

Response of knee fibrocartilage to joint destabilization N.A. Dyment y, Y. Hagiwara y z, X. Jiang y, J. Huang y, D.J. Adams y, D.W. Rowe y * y Center for Regenerative Medicine and Skeletal Development, School of Dental Medicine and Department of Orthopaedic Surgery, New England Musculoskeletal Institute, University of Connecticut Health Center, Farmington, CT 06032, USA z Department of Orthopedic Surgery, Nippon Medical School Hospital, Tokyo 113, Japan

a r t i c l e i n f o

s u m m a r y

Article history: Received 3 July 2014 Accepted 28 January 2015

Objective: A major challenge to understanding osteoarthritis (OA) pathology is identifying the cellular events that precede the onset of cartilage damage. The objective of this study is to determine the effect of joint destabilization on early changes to fibrocartilage in the joint. Design/Methods: The anterior cruciate ligament was transected in collagen reporter mice (Col1CFP and ColXRFP). Mineralization labels were given every 2 weeks to measure new mineralized cartilage apposition. Novel fluorescent histology of mineralized tissue was used to characterize the changes in fibrocartilage at 2 and 4 weeks post-injury. Results: Changes in fibrocartilaginous structures of the joint occur as early as 2 weeks after injury and are well developed by 4 weeks. The alterations are seen in multiple entheses and in the medial surface of the femoral and tibial condyles. In the responding entheses, mineral apposition towards the ligament midsubstance results in thickening of the mineralize fibrocartilage. These changes are associated with increases in ColX-RFP, Col1-CFP reporter activity and alkaline phosphatase enzyme activity. Mineral apposition also occurs in the fibrocartilage of the non-articular regions of the medial condyles by 2 weeks and develops into osteophytes by 4 weeks post-injury. An unexpected observation is punctate expression of tartrate resistant acid phosphatase activity in unmineralized fibrochondrocytes adjacent to active appositional mineralization. Discussion: These observations suggest that fibrocartilage activates prior to degradation of the articular cartilage. Thus clinical and histological imaging of fibrocartilage may be an earlier indicator of disease initiation and may indicate a more appropriate time to start preventative treatment. © 2015 Osteoarthritis Research Society International. Published by Elsevier Ltd. All rights reserved.

Keywords: Knee destabilization GFP reporter Non-decalcified cryohistology TRAP in chondrocytes Mineral apposition of fibrocartilage

Introduction Degenerative joint disease leading to incapacitating osteoarthritis (OA) is a chronic process that is the endpoint of genetic and environmental conditions1, and is a major financial burden of health care2e4. Common to many forms of OA is a past history of trauma leading to joint instability that eventually degrades the articular cartilage. Animal models that are based on induced joint instability require a number of weeks to express the degenerative changes, and these changes are often associated with osteophytes,

* Address correspondence and reprint requests to: D. W. Rowe, Center for Regenerative Medicine and Skeletal Development, Department of Reconstructive Sciences, Biomaterials and Skeletal Development, School of Dental Medicine, University of Connecticut Health Center, Farmington, CT 06030, USA. E-mail addresses: [email protected] (N.A. Dyment), [email protected] (Y. Hagiwara), [email protected] (X. Jiang), [email protected] (J. Huang), [email protected] (D.J. Adams), [email protected] (D.W. Rowe).

which further alter the function of the joint5,6. While there are numerous studies examining the influence of inflammatory cytokines and metalloproteases on the destruction of articular cartilage, less work has focused on other changes to the joint organ that precede and may predict the progression of articular cartilage damage. Some of the impediments for detecting early changes are inherent to paraffin embedded histology, which requires the use of decalcified tissue7 and cannot accommodate fluorescent signals. We have developed several GFP reporters that are expressed in chondrocytes and improve the sensitivity of histological evaluation of articular cartilage and fibrocartilaginous structures. In this study, a double reporter mouse was employed. A type X collagen reporter (ColX-RFPchry) is expressed in cells within the mineralizing regions of articular, fibrocartilage and endochondral cartilage. A type I collagen reporter (Col3.6-CFP) is expressed in fibroblasts, osteoblasts and fibrochondrocytes, but not articular or endochondral chondrocytes. We have also developed a cryohistological approach that maintains GFP signals and enzymatic

http://dx.doi.org/10.1016/j.joca.2015.01.017 1063-4584/© 2015 Osteoarthritis Research Society International. Published by Elsevier Ltd. All rights reserved.

Please cite this article in press as: Dyment NA, et al., Response of knee fibrocartilage to joint destabilization, Osteoarthritis and Cartilage (2015), http://dx.doi.org/10.1016/j.joca.2015.01.017

2

N.A. Dyment et al. / Osteoarthritis and Cartilage xxx (2015) 1e11

activity in mineralized tissues. The method is based on cryotape that adheres to the tissue section. Therefore, the coverslip can be easily removed and multiple rounds of staining and imaging can be performed, leading to colocalization of several response measures on a single section. We employ this approach in the current study to investigate the pathogenesis of OA in the knee joint organ8, with specific focus on fibrocartilage mineralization during osteophyte formation that is often associated with knee destabilization models9. This study employs a histological strategy for consistently sectioning and comparing normal vs destabilized knees taken from

the same GFP reporter mouse. The objectives are to determine the spatiotemporal changes to collagen (Col3.6-CFP and ColX-RFPchry) reporters, mineral apposition, enzymatic indicator of mineralization (alkaline phosphatase, AP), tissue remodeling (tartrate resistance acid phosphatase, TRAP), and cartilage proteoglycan distribution [toluidine blue (TB) staining] during early changes (up to 4 weeks) following joint destabilization in the mouse. We believe understanding the molecular basis for these dramatic changes and how they could directly or indirectly influence the neighboring articular cartilage may contribute to understanding the pathogenesis of OA.

Fig. 1. A,B: Coronal sections of the sham (A) and ACL-transected (B) knees are taken from the same animal at 4 weeks post-injury. Three ROI will be presented in Figs. 2e4: a/b is the femoral enthesis of the MCL with the medial femoral condyle (Fig. 2). c/d is the medial tibial condyle located beneath the MCL (Fig. 3). e/f is the articular cartilage and overlying meniscus (Fig. 4). C,D: Sagittal sections of the sham (C) and ACL-transected (D) knee from the same animal at 4 weeks post-injury. Two ROI will be presented in Fig. 5. g/h is the femoral attachment site of the PCL [Fig. 5(A)e(B)]. i/j is the tibial attachment site of the PCL [Fig. 5(C)e(D)]. Scale bar e 1 mm.



3

Methods

Surgical model

Mice used

All procedures conformed to an approved ACC protocol (100547-1015). The surgical manipulations were performed under a dissecting microscope. Under isoflurane anesthesia, a midline vertical incision of the skin, followed by an incision on the medial edge of patellar tendon was made. The patellar tendon/bone was subluxed laterally, exposing the joint capsule and underlying fat pad. Using micro-forceps, an opening was made through these tissues until the ACL was visible. Care was taken to avoid visible blood vessels within the fat pad. The ACL was cut using the tip of 25G needle, and complete transection was confirmed by anterior tibial translation during an anterior drawer test. An intact PCL was assured by visual inspection and continued posterior stabilization of the tibia. After the desired transection was accomplished, the patellar tendon and patella were returned to its original position and the incision was closed with a 6-0 nylon suture. The contralateral knees were used either as intact or sham-operated (same approach but without ACL transection) controls. The mice were

Two reporters were bred into mice for this study. Col3.6-CFP (Col3.6blue) has been used previously to identify cells within the osteoblast lineage where it generates a very strong signal10, but it is also expressed at a lower level (weak Col3.6blue) in chondrocytes of fibrocartilaginous tissues such as the condylar cartilage of the temporomandibular joint (TMJ)11 and in rapidly growing or injured tendons and ligaments12. It is not expressed in growth plate or articular chondrocytes. ColX-RFPchry (ColXred) is primarily found in the hypertrophic chondrocytes of all cartilaginous tissues13. The knee joints were destabilized in a total of fourteen female mice (age at surgery e mean ± SD: 9.5 ± 1.5 weeks, range: 8.7e12.9 weeks). One limb was destabilized while the other was used as a contralateral sham or unoperated control. Each animal was then assigned to 2 (n ¼ 6) or 4 (n ¼ 8) weeks post-surgery and sectioned in either coronal or sagittal orientation (Table S1).

Fig. 2. The histological changes within the femoral enthesis of the MCL and the medial surface of the femoral condyle that underlies the MCL will be discussed separately. A1eA6 is sham and B1eB6 is transected ACL. Increased appositional mineral formation in the enthesis: Panel A1/B1 shows strong TB stain for proteoglycans in the unmineralized fibrocartilage adjacent to the tidemark. The staining is stronger in the injured side. The thickness of the mineralize fibrocartilage is greater on the injured side as illustrated by the yellow brackets. A2/B2 shows mineral apposition in the enthesis (inset) and subchondral bone via demeclocycline (yellow), calcein (green) and AC (B3) mineralization lines given every 2 weeks. Cells in the unmineralized fibrocartilage adjacent to the tidemark are AP positive (B4) and some of these cells are also punctate TRAP positive (B5, arrow head). Immediately above the enthesis in the sham, the surface shows evidence of a yellow and green mineralization line (A2, #) but not Col3.6blue osteoblasts (A2) indicating earlier appositional bone mineralization but now a quiescent periosteal surface46. However, the surface bone of the injured side has developed a region of increased bone formation as seen by the strong Col3.6blue activity, calcein/demeclocycline labeling (B2), AC labeling (B3), strong AP activity (B4) and TRAP positive osteoclasts (B5). Expansion of the surface of the medial femoral condyle: The surface of the medial condyle extending distally from the enthesis and beneath the MCL of the sham animal shows a thin layer of TB positive cells in the unmineralized cartilage and a deeper layer of TB positive cells within mineralized cartilage (A1). Although there are no ColXred positive cells in this rim of tissue (A2/A3), most of the cells in the unmineralized area are AP positive (A4) and a few of these cells express the punctate TRAP signal (A5). In the injured side the unmineralized TB positive layer is significantly increased in depth and cells in this region express Col3.6blue (B2), AP (B4) and ColXred (B2/B3). Appositional mineralization signals are not prominent in this section but the thickness of the unmineralized fibrocartilage is greater on the injured vs sham side (A1 vs B1). A majority of the cells in this region are also punctate TRAP positive (B5). The subchondral bone adjacent to the periarticular surface appears to be quite active with strong mineralization labeling that occurred between week 0 and 2 (yellow-green) and another region that continues to be active (yellow-green-red-Col3.6blue-osteoclasts). The areas of increased activity show colocalization of TRAP and AP staining (A6/B6). Scale bar: 200 mm.


4


administered demeclocycline on the day prior to surgery, calcein at 2 weeks post injury and alizarin complexone (AC) on the day prior to sacrifice (day 28). Histological analysis The methods used for positioning the knee prior to tissue fixation, cutting the frozen non-decalcified sections and obtaining multiple fluorescent images from the same tissue section are explained in the supplemental section. Statistics Col3.6, ColX, AP, and TRAP area fractions and mineralized cartilage apposition in the MCL and PCL entheses were compared using one-way ANOVA with surgical group as fixed factor (significance level set to P < 0.05) and Tukey's HSD post hoc comparisons. Results Controlled sectioning of the normal, sham, and experimental tissues along with the ability to examine the same tissue section under multiple staining and imaging steps allows us to present the histological information in a consistent manner. Full views of the TB stained tissue (Fig. 1 and Fig. S2) show the regions of interest (ROI) that are detailed in Figs. 2e7 (sham is mirrored 180 to be in

the same orientation as the experimental). Each figure includes six enlarged views in the order of (1) TB only, (2) Col3.6blue, ColXRed, alizarin complexone (AC, red), calcein (green) and demeclocycline (yellow), (3) ColXRed, AC, DAPI, and TB, (4) ColX, AP (magenta) and TB, (5) TRAP (yellow) and TB, and (6) TRAPþ/APþ colocalized area (see Supplemental Methods) and TB. Animals that were terminated 4 weeks after injury were injected with demeclocycline 1 day prior to surgery, calcein 2 weeks post surgery and AC 1 day before sacrifice. Tissues that exhibit continuing appositional mineralization incorporate adjacent yellow, green and red mineralization lines [quantified in Fig. 8(A,C)]. The reader is encouraged to download the primary images from the publisher's weblink for greater resolution and appreciation of the described findings, or via our web site (http://ucsci.uchc.edu/yupaper/) for the primary stacked layers from which the various merged images were derived. This study focused on the MCL femoral enthesis, femoral and tibial medial condyles, and the PCL entheses. The sham procedure showed no significant changes to these specific structures compared to an unoperated knee (Figs. S3e6) at the 2 and 4 week time points.

Coronal section of the knee Alignment of the femur and tibia at 90 flexion and the placement of a fiduciary suture parallel to the MCL enabled cutting a section that includes the load bearing surfaces of the articular cartilage as well as the MCL, its femoral insertion site (medial

Fig. 3. Expansion of fibrocartilage of the medial tibial plateau beneath the MCL. This bone surface is covered with a thin layer of cells that are TB positive (A1) and express a strong Col3.6 signal (A2, arrow head), are AP positive (A4, arrow head) and a weak homogeneous TRAP signal (A5, arrow head) with no active mineralization line. This pattern probably represents quiescent bone lining periosteal cells. In contrast, the injured periarticular region shows a dramatic thickening of fibrocartilaginous tissue extending from the original bone surface (B1, yellow arrow head). The surface and mineralized region of the cartilage outgrowth (bracket) is strongly TB positive but there is no striking layer of APþ cells (B4) or evidence of active mineralization (B2). However it is within the center of this outgrowth that there is a dramatic increase in ColXred cell number (B3) and the presence of strongly expressing Col3.6blue cells (B2), AP activity (B4) and osteoclastic TRAP activity (B5) indicative of a forming osteophyte. Scale bar: 200 mm.



5

fibrochondrocytes in the unmineralized zone [Fig. 2(A1 vs B1)] and a triple mineralization label resulting in increased mineralized fibrocartilage apposition rate [P < 0.05, Fig. 8(A)] on the injured side relative to the control side [Fig. 2(A2 vs B2 insets)]. This mineralizing activity was associated with activation of Col3.6blue fibrochondrocytes (inset) ahead of the mineralizing front and was not seen on the control side [Fig. 2(A2 vs B2)]. Although there was no difference in the number of ColXred cells between the two sides [Fig. 8(B)], there was a greater number of AP positive cells on the injured side [Fig. 2(A4 vs B4), Fig. 8(B)]. Most surprisingly, some of the cells within the AP positive region were also TRAP positive [arrowhead; Fig. 2(A5 vs B5), Fig. 8(B)]. These TRAP positive cells were not osteoclasts because the stain has a punctate intracellular distribution within a cell with chondrocytic morphology in contrast to the homogeneous distribution of TRAP within multinucleated osteoclastic cells (to appreciate, please observe in downloaded full scale images). The net effect of the increase in mineralized matrix apposition was a thickening of the mineralized cartilage which can be appreciated by the TB stain (bracket in a1 and b1) that extends throughout the mineralized region of the fibrocartilage.

Fig. 4. Minimal changes with the articular cartilage and meniscus 4 weeks after injury. A1eA6 is sham and B1eB6 is transected ACL. Panels A1/B1. TB staining shows the proteoglycan within the soft cartilage tissue and outlines an intact articular surface in both samples. A2/B2 shows the demeclocycline, calcein, AC, Col3.6blue and ColXred reporter. These signals indicate strong osteogenic activity in the subchondral bone but no appositional mineralizing activity is seen in the articular cartilage tidemark (arrowheads). A3/B3 includes the ColXred reporter that overlays the DAPI and TB stains and show no difference its number or distribution with the articular cartilage. A4/B4 has the AP stain added to the ColX reporter. The activity appears to be similar in both tissues. Strong AP activity is present in the underlying subcortical bone. No punctate TRAP activity is found in the articular cartilage in A5/B5. In contrast, the solid TRAP activity within osteoclasts is evident in the subchondral bone. A6/B6 illustrates the minimal activity in the articular cartilage, compared to fibrocartilages in the joint, due to a lack of TRAP/AP colocalization. Scale bar: 200 mm.

epicondyle) and the medial condyle adjacent to the articular cartilage and beneath the MCL in both the femur and tibia [Fig. S1(DeG)]. Increased appositional mineral formation in the enthesis The MCL on the injured side is larger in size [Fig. 1(ROI A vs B)] and contains cells on the surface and interior of the body that express the Col3.6blue reporter [Fig. 2(A2 vs B2)]. The femoral insertion of the MCL shows stronger TB staining of the

Activation of the medial condylar surfaces Just proximal to the enthesis is a region of intense osteogenic activity based on the focal accumulation of strong Col3.6blue and AC mineral accumulation [Fig. 2(A2 vs B2)] and AP activity [Fig. 2(A4 vs B4)]. TRAP activity is also present in this region [Fig. 2(A5 vs B5)] but it has the homogeneous distribution pattern characteristic of an osteoclast. On the sham side, the same region contains a yellow and green mineralization label without Col3.6blue osteoblasts or strong AP signal indicative of periosteal bone formation that became quiescent after the 2-week labeling period [Fig. 2(A2#)]. These features, which occurred in 100% of limbs in the destabilized group, may represent the initial steps in osteophyte formation that developed from periosteal lining cells. In the sham animal, a thin layer of fibrocartilage on the medial condyle beneath the MCL extends between the femoral articular cartilage and the MCL enthesis [Fig. 1(ROI A vs B)]. The layer is composed of a single layer of TB positive cells [Fig. 2(A1)] that show weak demeclocycline labeling [Fig. 2(A2)] and AP activity [Fig. 2(A4)] and occasional punctate TRAP positive cells [Fig. 2(A5)] but no consistent reporter activity. The mineralized tissue beneath this surface shows a few ColXred [Fig. 2(A3)] and TB blue cells characteristic of mineralized fibrocartilage [Fig. 2(A1)]. On the injured side, the fibrocartilaginous surface expands in size [Fig. 2(B1)], expresses Col3.6blue [Fig. 2(b)] and ColXred cells [Fig. 2(c)] in the region that precedes the mineralized cartilage and shows remnants of the three mineralization labels [Fig. 2(B2)]. These cells are strongly AP positive [Fig. 2(B4)] and show many cells that are punctate TRAP positive [Fig. 2(B5)]. In fact, colocalization of AP and TRAP depicts the increased activity of this region [Fig. 2(A6 vs B6)]. The corresponding region of the tibia (medial edge of the tibial plateau, see Fig. 1, ROI C and D) that lies beneath the MCL manifests the most dramatic surface alterations. The surface cells of the sham side show dark blue TB staining and strong Col3.6blue signal [Fig. 3(A1,A2, arrowheads)]. However following destabilization, the TB stain illustrates a large increase in mineralized fibrocartilage that has developed by 4 weeks after injury [Fig. 3(A1 vs B1)]. The unmineralized surface chondrocytes contain weakly expressing Col3.6blue cells but no evidence of mineral formation. The most remarkable feature is a disorganized confluence of strong Col3.6blue [Fig. 3(B2)] and ColXred [Fig. 3(B3)] cells that are strongly AP positive [Fig. 3(B4)]. There is intense osteoclastic staining that is distinctly different from TRAP positive chondrocytes


6


Fig. 5. Increased remodeling activity of the entheses of the PCL (see Fig. 1 for placement of the ROI within the original image). The regions defined as the enthesis are bracketed in A/ C (sham) and B/D (injured). A1eD1. TB stain of the each attachment site show comparable proteoglycan accumulation near the tidemark. Similarly there are minimal differences in mineralization or Col3.6blue activity in this region. However the injured side has an increase number of ColXred (B3, D3), APþ cells (B4,D4) and punctate TRAP þ cells (B5,D5) concentrated in the unmineralized fibrocartilage adjacent to the tidemark (solid arrows). Scale bar: 200 mm.

seen in Fig. 2. This central region represents the start of an osteophyte that likely developed from a fibrocartilaginous origin. This osteophyte response was seen in 100% of animals in the destabilized group, compared to zero in the sham and intact groups. Minimal changes in the articular cartilage Despite the dramatic expansion and remodeling activities of the enthesis and surrounding periarticular surfaces, the articular cartilage of the femur and tibia of the sham and injured knee (Fig. 4) show minimal changes. The articular surfaces of both the sham and destabilized animals are intact [Fig. 4(A1 vs B1)]. A faint demeclocycline mineralization line of the tidemark is detected in both [arrowheads; Fig. 4(A2 vs B2)]. ColXred expression is present in unmineralized cartilage adjacent to the tidemark and throughout the mineralized cartilage [Fig. 4(A3 vs B3)], and these cells are AP positive [Fig. 4(A4 vs B4)]. The subchondral bone does not show evidence of increased osteoclastic activity or altered activity of mineralizing bone surfaces [Fig. 4(A2, A4, A5 vs B2, B4, B5)]. Sagittal section of the knee The angular and rotational conformity imposed on the knee structure prior to embedding allows a consistent sagittal sectioning plane that includes the patellar tendon and PCL when the mid portion of the knee is encountered (see Fig. 1 for the ROI of the

femoral (g/h) and tibial (i/j) attachment sites that are discussed in Fig. 5). The sections were obtained from an animal that received a calcein injection 2 weeks post-surgery and AC 1 day prior to sacrifice. Increased remodeling activity of the enthesis of the PCL Focusing on the PCL attachment sites [see brackets in Fig. 5(A1, B1, C1, D1)], evidence of increased cellular and matrix activity can be seen within the femoral [Fig. 5(columns A vs B)] and tibial [Fig. 5(columns C vs D)] insertion sites. Specifically, a strong calcein label is seen at the femoral site but not the tibial site [Fig. 5(B2 vs D2)]. ColXred cells did develop in the unmineralized fibrocartilage adjacent to the mineralization labels [Fig. 5(A3 vs B3 and C3 vs D3)] and there are an increased number of AP positive cells [Fig. 5(A4 vs B4 and C4 vs D4), Fig. 8(d)] within the zone of ColXred cells. However the striking change was the presence of the punctate TRAP positive cells that reside ahead of the mineralizing fibrocartilage [Fig. 5(A5 vs B5 and C5 vs D5)]. The unmineralized fibrocartilage adjacent to the tidemark shows the highest increase in activity, as seen via colocalization of TRAP and AP staining [Fig. 5(A6 vs B6)]. The TRAP and AP activity also corresponds with an increased mineral apposition rate [P < 0.05, Fig. 8(c)]. Histology of knee 2 weeks after injury Based on the analysis of the 4 week post injury histology, the changes surrounding the MCL and PCL appear to be the most



7

Fig. 6. MCL femoral enthesis and medial condyle (A1e6) and the medial tibial plateau (B1e6) at 2 weeks post-injury (sham not illustrated). The mice received demeclocycline the day before surgery and calcein the day prior to sacrifice. See Fig. S2 for selected ROI from the full view image. All of the features that were described in Figs. 2 and 3 have initiated in the 2 week histology. The midsubstance of the MCL is hypercellular and disorganized (A1) with elevated Col3.6blue expression (A2). Increased mineralized cartilage apposition has initiated at the tidemark with strong calcein labeling surrounding ColXred cells (A2, A20 , A3) that are APþ (A4) and TRAPþ (A5) in some cases. The osteophyte that is on the medial tibial plateau (B1) has also initiated at 2 weeks with Col3.6blue cells within the fibrocartilage (B2, B20 ). Some of these cells are also APþ (B4) and TRAPþ (B5) as they prepare to mineralize. A20 /B20 are insets from A2/B2 without the ColXred channel. Scale bar: 200 mm.

advanced and thus could be the most sensitive tissue for detecting early changes in the knee in our ACL transection model. We examined that possibility 2 weeks after injury in coronal and sagittal sections [Fig. S2(a vs b)]. In the coronal view [Fig. S2(ROI a/ b)] there was accelerated appositional mineralization in the enthesis [Fig. 8(a)] with strong calcein labeling and Col3.6blue expression [Fig. 6(A2/A20 )] and an early fibrocartilage expansion on the medial condyle distal to the MCL enthesis [Fig. 6(A2eA4)] compared to sham and normal samples (Fig. S3). Panel set B in Fig. 6 shows a very well developed expansion of the fibrocartilaginous zone of the medial tibial plateau beneath the over-riding MCL [Fig. 6(B1eB6)]. The expression of Col3.6 in the MCL is strong at this time point and evidence for osteophyte formation within the fibrocartilage is not present at this time [Fig. 6(B20 )]. The sagittal view present in Fig. 7 was derived from the ROI of Fig. S2(B). Both insertions sites demonstrate increased mineralizing activity [Fig. 7(a2 and b2)], increased ColXred numbers [Fig. 7(A3 and B3)], increased APþ cell number [Fig. 7(A4 and B4)], and appearance of punctate TRAP positive chondrocytes [Fig. 7A5 and B5)]. Discussion Increasingly, the entire knee joint is thought to act as an integrated organ, such that the pathogenesis of OA involves multiple tissues within the joint8,9,14,15. The evidence provided here shows that the ligaments, their insertion sites, and the periarticular surfaces respond to a destabilizing injury well before there is objective

visual evidence of articular cartilage damage, suggesting that trauma-induced OA pathogenesis involves more tissues than just articular cartilage. This study provides a cellular interpretation to this concept and the role that the fibrocartilaginous structures play in the initial pathogenesis of the degenerative process. We believe that the use of the GFP reporter mice and the cryohistological features of mineralizing fibrocartilage can contribute toward an improved molecular and cellular understanding of the sequence of events that precede articular cartilage destruction. In the model presented here, the changes can be detected as early as 2 weeks after injury and can lead to early osteophyte formation by 4 weeks well before articular cartilage changes, which usually are not evident until 8e12 weeks post injury16e18. To be useful as a cross-laboratory translational model of degenerative OA, there are experimental conditions that have to be well defined and followed6. The nature of the destabilization injury is crucial to the tissue response. Rapid loading protocols designed to rupture a ligament can either tear one or more ligaments or avulse the enthesis and underlying bone17,18. Direct impact of the articular cartilage is also likely to stress the supporting ligaments and cause bone bruising5,7,19e21. When examining the histological consequences of the destabilization model7, there needs to be consistency in the positioning of the knee prior to sectioning the injured and control tissues. Because the ligament transection results in translation of the tibia relative to the femur, it is necessary to return these bones to the normal position so that they can be compared to the control tissue in the same plane of section (see


8


Fig. 7. Onset of remodeling activity of the femoral (A1e6) and tibial (B1e6) entheses of the PCL 2 weeks after transection of the ACL. The mice received demeclocycline the day before surgery and calcein the day prior to sacrifice. See Fig. S2 for selected ROI from the full view image. Both sites show appearance of active cartilage mineralization (A2, B2) with ColXred (A3, B3), APþ (A4, B4) and punctate TRAP positive (A5, B5) cells. Another observation is the activity of the Col3.6blue reporter in the ligament in remodeling regions on the anterior surface of the PCL. Scale bar: 200 mm.

Fig. S1). Placing external fiduciary markers either within the tissues or utilizing internal structures that assist in obtaining similar orientation of the tissue block to the cutting blade is an additional step for consistent sectioning. By comparing sections taken at corresponding positions from the same animal, differences in histological features between injured and sham joint structures can be appreciated with greater confidence. The enthesis plays a central role in modulating the transition of forces from the tensile ligament to the rigid bone15. The collagen fibers of the ligament insertion extend through the

mineralized fibrocartilage, which clearly separates the ligament from the underlying bone (Fig. S7). The unmineralized fibrocartilage zone of the enthesis appears to be the cellular component that responds to joint destabilization. In the intact enthesis, there are minimal numbers of Col3.6blue and ColXred cells in the unmineralized fibrocartilage. However there are ColXred cells at the tidemark and in the mineralized fibrocartilage. Following destabilization, the cells that are located in the unmineralized fibrocartilage zone adjacent to the tidemark activate the Col3.6blue reporter, increase their production of proteoglycans and increase expression of the ColXred reporter. AP positive cells within this region also become prominent. These cellular activities are strongly associated with new appositional mineral formation of the cartilage matrix as indicated by the mineralization labels given prior to the injury and a day prior to sacrifice [see insets in Fig. 2(B2) and 6(A2)]. Thus the functional response of the enthesis may be to embed more of the collagen fibers of the ligament into mineralized fibrocartilage presumably to strengthen its attachment to bone14,22. The unexpected observation in the enthesis is TRAP activity in cells within unmineralized fibrocartilage adjacent to the tidemark. Similar TRAP activity can also be observed in the prehypertrophic and early hypertrophic cells of the growth plate (Fig. S8). These uninucleated chondrocytic cells should not be confused with chondroclasts or osteoclasts23,24 because the punctate cytoplasmic staining pattern is distinct from the homogeneous cytoplasmic distribution within the multinucleated osteoclast. These histological features have been observed previously in chromogenic histology of cartilage and in macrophages25,26. The localization of the activity was confirmed in TRAP-Cre driver mice that mapped a LacZ reporter to macrophages and growth plate chondrocytes27. A murine germline TRAP gene (Acp5) knockout resulted in a mild form of osteopetrosis, but also caused immune deficiency and chondrodystrophy28e30. Similarly, the autosomal recessive human disease due to an inactivating mutation of the Acp5 gene (http:// omim.org/entry/271550) presents as a spondyloenchondrodysplasia that includes combined immunodeficiency with autoimmunity19e21. Because the immune deficiency appears to be secondary to impaired lysosomal degradation of bacteria, it is assumed that TRAP plays a role in lysosomal physiology, which would also explain its punctate cytoplasmic distribution. However its role in forming or remodeling the cartilage matrix prior to mineralization has not been investigated. Another unexpected observation was the expansion of the thin layer of periosteum or fibrocartilage that eventually forms osteophytes over the medial pericondylar surface between the enthesis and articular cartilage that lies directly beneath the MCL. The expanded tissue has features of the condylar fibrocartilage of the TMJ with a thickened fibrocartilaginous zone containing cells that express the Col3.6blue reporter and ColXred cells in the deeper regions near the site of cartilage mineralization11. The condylar cartilage of the TMJ originates from periosteal cells31 and is particularly responsive to shearing forces32 and thus observing a similar response in other periosteal structures might be anticipated. The expansion was exceptionally prominent on the medial condyle of the tibia in which the central region was transitioning to an osteophyte as evidenced by the accumulation of strong Col3.6blue osteoblasts, osteoclasts and bone marrow. There is potential that the osteophyte is periosteal derived, especially since periosteal cells have the potential for fibrocartilaginous fracture callus formation and endochondral ossification33,34. If this is the case, then the formation of the osteophyte should be conceptualized as an internal process within the enlarged fibrocartilaginous tissue rather than an external process initiated within the marrow space.



9

Fig. 8. Mineralized cartilage apposition and area fraction measurements for the femoral MCL and PCL entheses. A,C: Total mineralized cartilage apposition was quantified in the femoral MCL (A) and PCL (C) entheses by measuring the distance between the mineralization labels given the day before surgery, 2 weeks post-surgery, and 4 weeks post-surgery. Both entheses displayed increased mineral apposition in the ACLT transected (ACLT) knees compared to the intact or sham groups. B,D: Area fraction measurements for Col3.6, ColX, AP, and TRAP were also measured in the femoral MCL and PCL entheses within ROIs defined in the supplemental methods. The area fraction of Col3.6þ and ColXþ did not show significant changes amongst the groups. However, AP and TRAP activity showed significant increases in the ACLT knees at both time points compared to the intact and sham groups. * indicates significant difference from intact/sham (P < 0.05). ^ indicates significant difference from ACLT (2 week) group (P < 0.05). P-values are listed in the table below the bar plots. Error bars indicate 95% confidence interval.

How these changes in the tissues adjacent to the articular cartilage relate to subsequent joint damage is not completely understood. The structures of the joint composed of fibrocartilage were far more responsive to joint instability and could serve as an indicator of pending articular cartilage degeneration. The changes in the fibrocartilaginous structures that were evident by 2 weeks in our model were observed as early as 7 days in a collagenaseinduced destabilization model35. Sensitive MRI modalities appear to be capable of detecting the changing mineralization pattern in the enthesis or medial condylar surfaces thus providing a clinical assessment of early disease and progression9,36. If the expanding mineralizing fibrocartilage tissue is a predictor of the degenerative process that will ultimately involve articular cartilage, therapies designed to retard the progression might have clinical benefit. From a mechanical perspective, weight loss and strengthening the surrounding musculature may help to reduce stresses on the ligaments and may slow OA progression37. However once the mineralization initiates, blocking the expansion or mineralizing process would be necessary. The observation that genetic inactivation of the Indian hedgehog (IHH) pathway conveys

resistance to degenerative arthritis might be explained by a reduction in the mineral apposition rate of fibrocartilage and articular cartilage38,39 or a suppression of the development of osteophytes40 within the expanded regions of fibrocartilage. That IHH is required for normal formation and maintenance of the chondral fibrocartilage of the TMJ41,42 supports this concept. In fact, IHH may be upstream of subsequent cartilage degradation as IHH's regulation of RUNX2 can lead to increased expression of ADAMTS5, a major protease found in OA cartilage39. Similarly, the osteoarthritic-sparing effect of genetic inactivation of certain metalloproteinases may be attributed to its role in IHH responsive cells in unmineralized cartilage that is preparing to mineralize43,44. In contrast, blocking the TGFb pathway promotes hypertrophic chondrocyte differentiation and the onset of OA45. Another point of therapeutic attack might be endochondral bone formation that develops in the expanding fibrocartilaginous surface. Agents that inhibit the BMP2 pathway46 or block osteoclastic activity43 might be effective in retarding this damaging development. The majority of drug/genetic intervention studies have focused on treatment after the onset of articular cartilage damage while their impact


10


might be more profound if initiated at the onset of fibrocartilage changes. Perhaps examination of these latter structures would be more informative for identifying tissue responses that impact the later onset of degenerative articular cartilage. Contributions 1. Dyment, NA e Conceived and designed study, collected, assembled, and interpreted data, drafted and approved article. 2. Hagiwara, Y e Conceived and designed study, performed all surgeries, collected, assembled, and interpreted data, drafted and approved article. 3. Jiang, X e Logistical support, collected and assembled data, interpreted data, approved article. 4. Huang, J - Logistical support, collected and assembled data, approved article. 5. Adams, DJ e Conceived and designed study, interpreted data, drafted and approved article. 6. Rowe, DW e Obtained funding, conceived and designed study, interpreted data, drafted and approved article. Role of funding source All funding sources had no influence on the design, collection, or analysis of data in this study. Funding sources are National Institutes of Health R21-AR055750, R01-AR052374, T90-DE021989 (support for NAD).

8.

9.

10.

11.

12.

13.

14.

Competing interests The authors have no competing interests to disclose. Acknowledgments This study was supported by NIH R21-AR055750, R01AR052374, T90-DE021989 (support for NAD).

15.

16.

Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.joca.2015.01.017.

17.

References 1. Madry H, Luyten FP, Facchini A. Biological aspects of early osteoarthritis. Knee Surg Sports Traumatol Arthrosc 2012;20: 407e22. 2. Litwic A, Edwards MH, Dennison EM, Cooper C. Epidemiology and burden of osteoarthritis. Br Med Bull 2013;105:185e99. 3. Ackerman IN, Ademi Z, Osborne RH, Liew D. Comparison of health-related quality of life, work status, and health care utilization and costs according to hip and knee joint disease severity: a national Australian study. Phys Ther 2013;93: 889e99. 4. Nho SJ, Kymes SM, Callaghan JJ, Felson DT. The burden of hip osteoarthritis in the United States: epidemiologic and economic considerations. J Am Acad Orthop Surg 2013;21(Suppl 1):S1e6. 5. Onur TS, Wu R, Chu S, Chang W, Kim HT, Dang AB. Joint instability and cartilage compression in a mouse model of posttraumatic osteoarthritis. J Orthop Res 2014;32:318e23. 6. Fang H, Beier F. Mouse models of osteoarthritis: modelling risk factors and assessing outcomes. Nat Rev Rheumatol 2014;10: 413e21. 7. Glasson SS, Chambers MG, Van Den Berg WB, Little CB. The OARSI histopathology initiative e recommendations for

18.

19.

20.

21.

22.

23.

histological assessments of osteoarthritis in the mouse. Osteoarthritis and Cartilage 2010;18(Suppl 3):S17e23. Man GS, Mologhianu G. Osteoarthritis pathogenesis e a complex process that involves the entire joint. J Med Life 2014;7: 37e41. Mcgonagle D, Tan A, Carey J, Benjamin M. The anatomical basis for a novel classification of osteoarthritis and allied disorders. J Anat 2010;216:279e91. Kalajzic I, Kalajzic Z, Kaliterna M, Gronowicz G, Clark S, Lichtler A, et al. Use of type I collagen green fluorescent protein transgenes to identify subpopulations of cells at different stages of the osteoblast lineage. J Bone Miner Res 2002;17: 15e25. Chen J, Utreja A, Kalajzic Z, Sobue T, Rowe D, Wadhwa S. Isolation and characterization of murine mandibular condylar cartilage cell populations. Cells Tissues Organs 2012;195: 232e43. Dyment NA, Kazemi N, Aschbacher-Smith LE, Barthelery NJ, Kenter K, Gooch C, et al. The relationships among spatiotemporal collagen gene expression, histology, and biomechanics following full-length injury in the murine patellar tendon. J Orthop Res 2012;30:28e36. Maye P, Fu Y, Butler DL, Chokalingam K, Liu Y, Floret J, et al. Generation and characterization of Col10a1-mcherry reporter mice. Genesis 2011;49:410e8. Tan AL, Toumi H, Benjamin M, Grainger AJ, Tanner SF, Emery P, et al. Combined high-resolution magnetic resonance imaging and histological examination to explore the role of ligaments and tendons in the phenotypic expression of early hand osteoarthritis. Ann Rheum Dis 2006;65:1267e72. Benjamin M, McGonagle D. Histopathologic changes at “synovio-entheseal complexes” suggesting a novel mechanism for synovitis in osteoarthritis and spondylarthritis. Arthritis Rheum 2007;56:3601e9. Kamekura S, Hoshi K, Shimoaka T, Chung U, Chikuda H, Yamada T, et al. Osteoarthritis development in novel experimental mouse models induced by knee joint instability. Osteoarthritis and Cartilage 2005;13:632e41. Christiansen BA, Anderson MJ, Lee CA, Williams JC, Yik JH, Haudenschild DR. Musculoskeletal changes following noninvasive knee injury using a novel mouse model of posttraumatic osteoarthritis. Osteoarthritis and Cartilage 2012;20:773e82. Lockwood KA, Chu BT, Anderson MJ, Haudenschild DR, Christiansen BA. Comparison of loading rate-dependent injury modes in a murine model of post-traumatic osteoarthritis. J Orthop Res 2014;32:79e88. Lausch E, Janecke A, Bros M, Trojandt S, Alanay Y, De Laet C, et al. Genetic deficiency of tartrate-resistant acid phosphatase associated with skeletal dysplasia, cerebral calcifications and autoimmunity. Nat Genet 2011;43:132e7. Briggs TA, Rice GI, Daly S, Urquhart J, Gornall H, BaderMeunier B, et al. Tartrate-resistant acid phosphatase deficiency causes a bone dysplasia with autoimmunity and a type I interferon expression signature. Nat Genet 2011;43:127e31. Renella R, Schaefer E, LeMerrer M, Alanay Y, Kandemir N, Eich G, et al. Spondyloenchondrodysplasia with spasticity, cerebral calcifications, and immune dysregulation: clinical and radiographic delineation of a pleiotropic disorder. Am J Med Genet A 2006;140:541e50. Blankenbaker DG, De Smet AA, Fine JP. Is intra-articular pathology associated with MCL edema on MR imaging of the non-traumatic knee? Skeletal Radiol 2005;34:462e7. Turner RT, Evans GL, Wakley GK. Reduced chondroclast differentiation results in increased cancellous bone volume in



24.

25.

26.

27.

28.

29. 30.

31.

32.

33.

34.

35.

36.

estrogen-treated growing rats. Endocrinology 1994;134: 461e6. K Jr S, Vilim V. Biochemical properties of cartilage and multinucleate chondroclast formation. Funct Dev Morphol 1992;2: 163e5. Hayman AR, Bune AJ, Bradley JR, Rashbass J, Cox TM. Osteoclastic tartrate-resistant acid phosphatase (Acp 5): its localization to dendritic cells and diverse murine tissues. J Histochem Cytochem 2000;48:219e28. Hayman AR, Bune AJ, Cox TM. Widespread expression of tartrate-resistant acid phosphatase (Acp 5) in the mouse embryo. J Anat 2000;196(3):433e41. Chiu WS, McManus JF, Notini AJ, Cassady AI, Zajac JD, Davey RA. Transgenic mice that express Cre recombinase in osteoclasts. Genesis 2004;39:178e85. Hayman AR. Tartrate-resistant acid phosphatase (TRAP) and the osteoclast/immune cell dichotomy. Autoimmunity 2008;41: 218e23. Hayman AR, Cox TM. Tartrate-resistant acid phosphatase knockout mice. J Bone Miner Res 2003;18:1905e7. Hayman AR, Jones SJ, Boyde A, Foster D, Colledge WH, Carlton MB, et al. Mice lacking tartrate-resistant acid phosphatase (Acp 5) have disrupted endochondral ossification and mild osteopetrosis. Development 1996;122:3151e62. Fukada K, Shibata S, Suzuki S, Ohya K, Kuroda T. In situ hybridisation study of type I, II, X collagens and aggrecan mRNas in the developing condylar cartilage of fetal mouse mandible. J Anat 1999;195(3):321e9. Habib H, Hatta T, Rahman OI, Yoshimura Y, Otani H. Fetal jaw movement affects development of articular disk in the temporomandibular joint. Congenit Anom 2007;47:53e7. Ushiku C, Adams D, Jiang X, Wang L, Rowe D. Long bone fracture repair in mice Harboring GFP reporters for cells within the osteoblastic lineage. J Orthop Res 2010;28:1338e47. Grcevic D, Pejda S, Matthews BG, Repic D, Wang L, Li H, et al. In vivo fate mapping identifies mesenchymal progenitor cells. Stem Cells 2012;30:187e96. Blaney Davidson E, Vitters EL, van dK, van dB. Expression of transforming growth factor-beta (TGFbeta) and the TGFbeta signalling molecule SMAD-2P in spontaneous and instabilityinduced osteoarthritis: role in cartilage degradation, chondrogenesis and osteophyte formation. Ann Rheum Dis 2006;65:1414e21. Hernandez-Molina G, Guermazi A, Niu J, Gale D, Goggins J, Amin S, et al. Central bone marrow lesions in symptomatic

37.

38.

39.

40.

41.

42.

43.

44.

45.

46.

11

knee osteoarthritis and their relationship to anterior cruciate ligament tears and cartilage loss. Arthritis Rheum 2008;58: 130e6. Jones RK, Chapman GJ, Findlow AH, Forsythe L, Parkes MJ, Sultan J, et al. A new approach to prevention of knee osteoarthritis: reducing medial load in the contralateral knee. J Rheumatol 2013;40:309e15. Zhou J, Chen Q, Lanske B, Fleming BC, Terek R, Wei X, et al. Disrupting the Indian hedgehog signaling pathway in vivo attenuates surgically induced osteoarthritis progression in Col2a1-CreERT2; Ihhfl/fl mice. Arthritis Res Ther 2014;16: R11. Lin AC, Seeto BL, Bartoszko JM, Khoury MA, Whetstone H, Ho L, et al. Modulating hedgehog signaling can attenuate the severity of osteoarthritis. Nat Med 2009;15:1421e5. Baht GS, Silkstone D, Nadesan P, Whetstone H, Alman BA. Activation of hedgehog signaling during fracture repair enhances osteoblastic-dependent matrix formation. J Orthop Res 2014;32:581e6. Shibukawa Y, Young B, Wu C, Yamada S, Long F, Pacifici M, et al. Temporomandibular joint formation and condyle growth require Indian hedgehog signaling. Dev Dyn 2007;236: 426e34. Ochiai T, Shibukawa Y, Nagayama M, Mundy C, Yasuda T, Okabe T, et al. Indian hedgehog roles in post-natal TMJ development and organization. J Dent Res 2010;89:349e54. Kozawa E, Nishida Y, Cheng XW, Urakawa H, Arai E, Futamura N, et al. Osteoarthritic change is delayed in a Ctskknockout mouse model of osteoarthritis. Arthritis Rheum 2012;64:454e64. Wei F, Zhou J, Wei X, Zhang J, Fleming BC, Terek R, et al. Activation of Indian hedgehog promotes chondrocyte hypertrophy and upregulation of MMP-13 in human osteoarthritic cartilage. Osteoarthritis and Cartilage 2012;20:755e63. Shen J, Li J, Wang B, Jin H, Wang M, Zhang Y, et al. Deletion of the transforming growth factor beta receptor type II gene in articular chondrocytes leads to a progressive osteoarthritislike phenotype in mice. Arthritis Rheum 2013;65:3107e19. Blaney Davidson EN, Vitters EL, Bennink MB, van Lent PL, van Caam AP, Blom AB, et al. Inducible chondrocyte-specific overexpression of BMP2 in young mice results in severe aggravation of osteophyte formation in experimental OA without altering cartilage damage. Ann Rheum Dis 2014 [In press].


Age changes in the triangular fibrocartilage of the wrist joint.

Temporomandibular joint fibrocartilage degeneration from unilateral dental splints.

Sensory response following knee joint damage in rabbits.

Pure Varus Injury to the Knee Joint.

Exploiting endogenous fibrocartilage stem cells to regenerate cartilage and repair joint injury.

Resection of the Knee-Joint.

Replacement of the knee joint.

Does Level of Response to SI Joint Block Predict Response to SI Joint Fusion?

[Kinematics of various joint mechanisms for the prosthetic substitution of the knee joint following knee exarticulation].

Delayed gadolinium-enhanced MRI of the fibrocartilage disc of the temporomandibular joint--a feasibility study.

Migration of trochanteric cerclage cable debris to the knee joint.

Hemiresection-interposition arthroplasty of the distal radioulnar joint associated with repair of triangular fibrocartilage complex lesions.

The Dorsal Triangular Fibrocartilage of the Metacarpophalangeal Joint: A Cadaveric Study.

Infection in total knee joint replacement, secondary to tooth abscess.

Case of Excision of Knee-Joint; Recovery.

Case of Excision of the Knee-Joint.

Cases of Injury of the Knee-Joint.

The volume of the human knee joint.

Assessment of inflammatory activity in knee joint.

[Three dimensional kinematics of the knee joint].

[Advantages of arthroscopy of the knee joint].

Response of joint afferent neurons in cat medial articular nerve to active and passive movements of the knee.

[NMR tomography of knee joint injuries].

Helical axes of passive knee joint motions.