Site-Dependent Changes in Structure and Function of Lapine Articular Cartilage 4 Weeks After Anterior Cruciate Ligament Transection

Accepted Manuscript

Site-Dependent Changes in Structure and Function of Lapine Articular Cartilage 4Weeks After Anterior Cruciate Ligament Transection

J.T.A. Mäkelä, Z.S. Rezaeian, S. Mikkonen, R. Madden, S.-K. Han, J.S. Jurvelin, W.Herzog, R.K. Korhonen

PII: S1063-4584(14)01045-0

DOI: 10.1016/j.joca.2014.04.010

Reference: YJOCA 3122

To appear in: Osteoarthritis and Cartilage

Received Date: 11 September 2013

Revised Date: 4 April 2014

Accepted Date: 12 April 2014

Please cite this article as: Mäkelä JTA, Rezaeian ZS, Mikkonen S, Madden R, Han S-K, Jurvelin JS,Herzog W, Korhonen RK, Site-Dependent Changes in Structure and Function of Lapine ArticularCartilage 4 Weeks After Anterior Cruciate Ligament Transection, Osteoarthritis and Cartilage (2014),doi: 10.1016/j.joca.2014.04.010.

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service toour customers we are providing this early version of the manuscript. The manuscript will undergocopyediting, typesetting, and review of the resulting proof before it is published in its final form. Pleasenote that during the production process errors may be discovered which could affect the content, and alllegal disclaimers that apply to the journal pertain.

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

1

Site-Dependent Changes in Structure and Function of Lapine Articular 1

Cartilage 4 Weeks After Anterior Cruciate Ligament Transection 2

3

+1Mäkelä, J T A; 1,2Rezaeian, Z S; 1Mikkonen S; 3Madden, R; 3,4Han, S-K; 1Jurvelin, J S; 4

3Herzog W; 1Korhonen, R K 5

6

+1Department of applied physics, University of Eastern Finland, Kuopio, Finland; 7

2Department of Physical Therapy, Faculty of Rehabilitation Sciences, Isfahan university 8

of Medical Sciences, Isfahan, Iran; Department of Physical Therapy, School of Medical 9

Sciences, Tarbiat modares University, Tehran, Iran; 10

3Human Performance Laboratory, Faculty of Kinesiology, University of Calgary, 11

Calgary, Alberta, Canada; 12

4Advanced Biomedical and Welfare Technology R&BD Group, Korea Institute of 13

Industrial Technology, Cheonan-si, Chungcheongnam-do, Korea. 14

15

Correspondence: 16

Janne Mäkelä 17

Department of Applied Physics 18

University of Eastern Finland 19

Yliopistonranta 1, 20

POB 1627, 70211 Kuopio. 21

Tel: +358 44 020 4034 22

Email: [email protected] 23

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

2

ABSTRACT 1

OBJECTIVE

The aim of this study was to investigate the site-dependent changes in the structure and 2

function of articular cartilage in the lapine knee joint at a very early stage of 3

osteoarthritis, created experimentally by anterior cruciate ligament transection (ACLT). 4

METHODS

Unilateral ACLT was performed in eight mature New Zealand white rabbits. ACL 5

transected and contralateral (C-L) joints were prepared for analysis at four weeks after 6

ACLT. Three rabbits with intact joints were used as a control group (CNTRL). Femoral 7

groove, medial and lateral femoral condyles, and tibial plateaus were harvested and used 8

in the analysis. Biomechanical tests, microscopy and spectroscopy were used to 9

determine the biomechanical properties, composition and structure of the samples. A 10

linear mixed model was chosen for statistical comparisons between the groups. 11

RESULTS

As a result of ACLT, the equilibrium and dynamic moduli were decreased primarily in 12

the femoral condyle cartilage. Up to three times lower moduli (p < 0.05) were observed in 13

the ACLT group compared to the control group. Significant (p < 0.05) proteoglycan loss 14

in the ACLT joint cartilage was observed up to a depth of 20-30% from the cartilage 15

surface in femoral condyles, while significant proteoglycan loss was confined to more 16

superficial regions in tibial plateaus and femoral groove. The collagen orientation angle 17

was increased (p < 0.05) up to a cartilage depth of 60% by ACLT in the lateral femoral 18

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

3

condyle, while smaller effects, but still significant, were observed at other locations. The 1

collagen content was increased (p < 0.05) in the middle and deep zones of the ACLT 2

group compared to the control group samples, especially in the lateral femoral condyle. 3

CONCLUSION

Femoral condyle cartilage experienced the greatest structural and mechanical alterations 4

in very early osteoarthritis, as produced by ACLT. Degenerative alterations were 5

observed especially in the superficial collagen fiber organization and proteoglycan 6

content, while the collagen content was increased in the deep tissue of femoral condyle 7

cartilage. The current findings provide novel information of the early stages of 8

osteoarthritis in different locations of the knee joint. 9

10

KEYWORDS

Articular cartilage; Osteoarthritis; Anterior cruciate ligament transection; Lapine; Knee. 11

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

1

INTRODUCTION 1

Anterior cruciate ligament transection (ACLT) (the Pond-Nuki model1) is a recognized 2

method to induce osteoarthritis (OA) in animal models2-7. Without the ACL resisting 3

primary anterior tibial displacement and internal tibial rotation, loading patterns of the 4

knee joint change8, leading to degenerative changes in metabolism and structure of 5

articular cartilage9. These changes are characteristic markers of early OA10. ACL rupture 6

can lead to OA even with reconstructive surgery11, but if the very first mechanisms 7

responsible for the tissue changes are known, strategies may be developed to stop or limit 8

the progression of OA. 9

Cartilage consists mainly of interstitial water, collagen (type II) fibers, and 10

proteoglycans (PGs)12. The fluid phase (water) primarily resists instantaneous loads and 11

transient, poroelastic responses of the tissue, while collagen fibers resist mainly tensile 12

stresses. Under prolonged load, when fluid flow in the tissue has ceased, the PG mesh 13

resists cartilage compression12. The first degenerative changes of cartilage structure have 14

been observed to start from the superficial layer where PG content decreases and collagen 15

fibrillation occurs6, 7, 13-18. When cartilage also swells, increased fluid content and reduced 16

solid content in the degenerated cartilage allows for easier flow of the interstitial water 17

out of the loaded cartilage. These structural changes lead to alterations in the 18

biomechanical properties of articular cartilage; permeability increases and dynamic and 19

equilibrium stiffness decrease7, 19. 20

In in vivo experimental studies aimed at identifying the first signs of OA, different 21

animals (e.g. dogs14, 20, rabbits3, 13, 21-23, rodents6, sheep24) and mechanical interventions 22

(e.g. ligament transection13, 23, meniscectomy22, 25, traumatic impact26, combination of 23

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

2

these20, 27, 28) have been used. Simultaneous transection of the anterior and posterior 1

cruciate ligaments combined with meniscectomy have been shown to lead to noticeable 2

degradation of articular cartilage as early as two weeks after the intervention29. However, 3

in ACL transected rabbits, the first signs of cartilage deterioration have been reported to 4

occur at four weeks3, 13, 21. Superficial PG loss and degradation of the collagen matrix 5

organization have been detected, while no significant changes in the collagen content 6

have been observed. 7

Earlier studies investigating the effect of ACLT on degenerative changes of 8

cartilage have usually concentrated on either the femoral, the tibial or the patellar 9

cartilage3, 13, 20, 21, 23, 27. Results of structural and mechanical differences between sites in 10

the knee, especially with a depth-wise structural analysis and a very short time period 11

after transection, are lacking. Also in ACLT models, unilateral transection is used as the 12

golden standard, with the contralateral side often used as the control5, 6, 13, 14, 30, 31. 13

However, evidence of systematic changes in contralateral limbs exist23, thus the question 14

arises whether the contralateral joint should be used as an unaffected control. 15

The aim of this study was to investigate the structural and functional changes of 16

articular cartilage in the lapine knee, and the site-dependence of these changes at a very 17

early stage of OA, created experimentally by ACLT. Biomechanical measurements were 18

used to analyze the changes in the elastic properties of cartilage. Microscopic and 19

spectroscopic methods were applied to analyze depth-dependent composition, i.e., 20

collagen and PG content and collagen orientation of the samples. The femoral groove, 21

lateral and medial femoral condyles, and tibial plateaus of transected, contralateral and 22

non-transected control knees were used for site-specific analysis. To our knowledge, this 23

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

3

is the first time these early changes in lapine articular cartilage structure and function 1

have been investigated in a site-specific manner, only at 4 weeks after ACLT. 2

3

METHODS 4

A brief presentation of the materials and methods is given here. More details are 5

presented in the supplementary material. 6

ANIMAL MODEL

Skeletally mature, female New Zealand white rabbits (Oryctolagus cuniculus, age 14 7

months, n = 11) were used for all tests. Unilateral ACLT was performed in eight rabbits 8

and the contralateral joints were used for analysis as a contralateral (C-L) group13. Six 9

knee joints from three non-operated rabbits were used as a separate control group 10

(CNTRL). Animals were sacrificed at 4 weeks following ACLT and knee joints were 11

dissected and immersed into PBS. ACLT and C-L group knees were frozen in dry ice and 12

shipped from Calgary, Alberta, Canada to Kuopio, Finland. The CNTRL group samples 13

were tested biomechanically first in Calgary before shipping them to Kuopio in formalin 14

fixative. ACLT procedures for rabbits were approved by the Animal Ethics committee at 15

the University of Calgary and the guidelines of the Canadian Council on Animal Care 16

were followed. 17

BIOMECHANICAL TESTS

For biomechanical measurements, cartilage-on-bone samples were harvested (Fig. 18

1). Biomechanical measurement locations were defined as the top point (apex of the 19

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

4

posterior curvature) of femoral condyles and groove and the central point of tibial 1

plateaus for femur and tibia, respectively32, 33. Samples were glued to the bottom of a 2

measuring chamber for each measurement and the chamber was filled with PBS. A 3

stress-relaxation protocol was implemented using a ramp rate of 100 %/s (350 ± 150 4

µm/s), 3 x 5 % steps and 15 min relaxation time after each step, after which a sinusoidal 5

dynamic test was performed with a frequency of 1 Hz (amplitude 4 % of remaining tissue 6

thickness, 4 cycles). Both equilibrium and dynamic linearly elastic moduli were then 7

determined similarly as earlier34-36. 8

MICROSCOPIC AND SPECTROSCOPIC ANALYSIS

After biomechanical measurements, the samples were fixed in formalin, 9

decalcified, processed in graded alcohol solutions and embedded in paraffin. Histological 10

sections were then cut perpendicular to the cartilage surface. Digital densitometry (DD) 11

sections (thickness ~3 µm) were stained with Safranin O37 while Fourier transform 12

infrared spectroscopy (FTIR) and polarized light microscopy (PLM) sections (thickness ~ 13

5 µm) underwent enzymatic removal of PGs prior to the measurements using 14

hyaluronidase (1000 U/ml hyaluronidase, SigmaeAldrich, St. Louis, MO, USA)13. 15

Using DD, grayscale images of safranin O stained sections were captured with a 16

light microscope (Olympus CH-2, Olympus, Tokyo, Japan) and a CCD camera (Perkin 17

Elmer, MA, USA). The optical density was used to estimate the distribution of tissue 18

PGs37, 38. PLM was used to analyze collagen orientation angles (0° = parallel to the 19

cartilage surface) based on the Stokes parameters39, 40. The used instrument was Leitz 20

Ortholux II POL-polarized microscope (Leitz, Wetzlar, Germany) equipped with a CCD 21

camera (Photometrics CH 250/A, Photometrics Inc., Tucson, AZ, USA). Collagen 22

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

5

content was estimated with FTIR (PerkinElmer Spotlight 300, PerkinElmer, Shelton, CO, 1

USA) by integration of the Amide I region (1585 – 1720 cm-1)16, 41-43. Transversely 2

averaged depth-wise profiles were created from the sections of each sample at 1% 3

increments of normalized depth. Finally, every sample was given a Mankin score to 4

represent the severity of OA (average value of three blinded scorers)44, 45. 5

STATISTICAL ANALYSIS

Statistical point-by-point comparisons between the groups were made using a linear 6

mixed model ANOVA. Using this approach, potential interrelations between the samples, 7

i.e., dependence between the samples from the same animal, could be taken into account. 8

9

RESULTS 10

MANKIN SCORING

Average Mankin scores ranged from 1.0 to 4.8 in the ACLT group samples, highest 11

values observed for femoral condyles. Group averages in the C-L and CNTRL group 12

samples ranged from 0.4 to 1.6 and from 0 to 0.7, respectively. Except for lateral tibial 13

plateu cartilage, Mankin scores of the samples were significantly greater (p < 0.05) at all 14

sites in the ACLT compared to the CNTRL group (Table 1). In femoral condyles, the 15

ACLT group values were also significantly greater than those in the C-L group (p < 16

0.05). The average Mankin score in the C-L group samples, compared to the CNTRL 17

group samples, was significantly higher only at the medial tibial plateu (p = 0.010). 18

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

6

ELASTIC MODULI

There were no significant differences in the thicknesses of the cartilages between the 1

groups. ACLT caused alterations in the equilibrium and dynamic moduli, especially in 2

the femoral condyle cartilage (Tables 2&3). In the femoral condyles, the equilibrium 3

modulus of cartilage was significantly lower in the ACLT than that observed in the C-L 4

group, for both, the lateral (p = 0.034) and the medial side (p = 0.019). The equilibrium 5

modulus of medial femoral condyle cartilage in the ACLT group was also significantly 6

lower than that obtained in the CNTRL group (p = 0.049). The dynamic elastic modulus 7

of femoral condyle cartilage was significantly lower in the ACLT and C-L groups 8

compared to the CNTRL group animals, both in the lateral (p < 0.001 and p = 0.001, 9

respectively) and medial sides (p < 0.001 and p = 0.006, respectively). Tibial plateau 10

cartilage properties were altered only on the lateral side, where the equilibrium modulus 11

of the ACLT group cartilage was significantly lower than that of the CNTRL group 12

cartilage (p = 0.002). No significant differences in the moduli were observed in the 13

femoral groove cartilage. 14

PROTEOGLYCAN CONTENT

ACLT modulated the amount of PGs in the ACLT group cartilage for all sites (Fig. 3). 15

The amount of PGs in the ACLT group, compared to the other groups, was reduced 16

significantly in the top 20-30 % of tissue thickness in the femoral condyle and femoral 17

groove cartilages, and in the top 10 % in the tibial plateau cartilages. In the medial tibial 18

plateau, the C-L group differed also significantly from the CNTRL group in the top 5 % 19

of cartilage thickness. In the lateral femoral condyle, the ACLT and C-L groups, 20

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

7

compared to the CNTRL group, had significantly greater PG concentration between ~20-1

78 % of normalized tissue depth. The same finding was obtained in the medial tibial 2

plateau (~25-78 % of normalized depth), yet the difference between C-L and CNTRL 3

was not significant. 4

COLLAGEN ORIENTATION

The superficial collagen fibrils were more disorganized, i.e., less parallel to the surface, 5

in the ACLT and C-L groups compared to the CNTRL group at all measurement sites 6

(Fig. 4). Significantly increased collagen orientation angles in the ACLT joint cartilage, 7

compared to the CNTRL group samples, reached up to 57 %, 21 % and 33 % of tissue 8

thickness in lateral femoral condyles, medial tibial plateaus and femoral groove, 9

respectively. In the lateral femoral condyle and tibial plateau, the ACLT group also 10

differed significantly from the C-L group with the difference reaching up to 51 % of the 11

normalized tissue depth. 12

COLLAGEN CONTENT

In the femoral condyles, the collagen content was higher in the middle and deep zones of 13

the ACLT joint samples compared to the other groups (Fig. 5). In the lateral femoral 14

condyle, this increased collagen content reached almost the cartilage surface (~13-100 % 15

of normalized depth), while in the medial side, the increase was only observed between 16

~50 % of cartilage thickness to the cartilage-bone interface. Medial femoral condyle 17

cartilage samples of the ACLT group had greater amount of collagen than those of the 18

CNTRL group also in the most superficial cartilage layer (~3-6 % depth from the 19

surface). 20

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

8

In the tibial plateaus and femoral groove, the collagen content was significantly 1

smaller in the superficial and middle layers of ACLT group cartilages compared to the 2

CNTRL group samples; the difference was observed up to ~50 % of normalized tissue 3

depth. Tibial plateau cartilages of the ACLT group differed significantly also from the C-4

L group cartilages between ~8-36 % and ~10-23 % of tissue thickness in the lateral and 5

medial tibial plateaus, respectively. In the deep layers of C-L group cartilage, the medial 6

tibial plateau had significantly lower collagen content (from ~61 % to ~100 % of tissue 7

thickness) than ACLT and CNTRL group tissues. This was also found to be true for 8

femoral groove cartilages (~40-100 % normalized depth). 9

10

DISCUSSION 11

This is the first study aimed at investigating site-dependent changes in the structure, 12

composition and function of articular cartilage at a very early stage of experimentally 13

induced OA; four weeks after ACL transection of the rabbit knee. Structural and 14

functional changes were observed both in the transected and non-transected contralateral 15

knee joints, and alterations in cartilage properties were highly site-dependent. 16

Degenerative changes of cartilage caused by ACLT were primarily observed in femoral 17

condyles, as demonstrated by alterations in the Mankin score, elastic moduli, collagen 18

fibril orientation and PG content. Significant ACLT-induced changes in the collagen 19

orientation angle and PG content of lateral femoral condyle cartilage were detected as 20

deep as ~60 % and ~20 % of cartilage thickness, respectively, while increases in the 21

collagen content were observed in the middle and deep zones of cartilage. 22

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

9

The causes for the dramatic deterioration of femoral condyle cartilage, and the less 1

severe degradations in other locations, may be associated with the site-specific joint 2

loading. The primary function of the ACL is to restrain anterior tibial displacement and 3

internal tibial rotation. ACL deficiency changes the knee joint kinematics causing the 4

contact locations in the tibio-femoral compartment to shift during loading compared to 5

the ACL intact joint8, 46. This may have caused more dramatic contact and shear stress 6

alterations in femoral condyles compared to other locations. Human follow up study has 7

also indicated the femoral condyle cartilage being the most vulnerable (compared to the 8

other locations used here) to abnormal joint loading, from one to eleven years after ACL 9

injury11. 10

The PG content and collagen orientation angle were most affected by ACLT and 11

changes in these parameters were most pronounced in the superficial and middle layers of 12

cartilage. These alterations have commonly been observed as the first structural signs of 13

early OA6, 7, 13-18. It appears that cartilage may not be able to adapt to the rapid changes in 14

loading conditions following ACLT, and the increased tissue stresses and strains, and the 15

altered fluid pressures may affect the integrity of the superficial zone cartilage. While the 16

collagen content may not be affected in early OA, decreases in the PG content and 17

alterations in the collagen matrix organization may decrease the stiffness of the 18

superficial zone cartilage, thereby increasing the vulnerability of the tissue to additional 19

damage and OA progression. 20

Alterations in the collagen content following ACLT were highly site-specific and 21

surprising. An unexpected change was the increase in the collagen content in the deep 22

zone of cartilage, especially in the lateral femoral condyle cartilage of the ACLT group 23

animals. Increases in the collagen content suggest a fibrotic response to the altered 24

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

10

environment. Results suggesting a synthesis of type II collagen have also been reported in 1

the moderately degenerated areas of very early OA articular cartilage47. However, with 2

progressive cartilage degeneration, this increase in the collagen content could not be 3

identified unequivocally29, 47. As the integrity of the superficial zone cartilage was 4

disturbed here, the deep cartilage zone, and associated cells, may have been subjected to 5

abnormally high strains, stresses and fluid pressures48. This increased mechanical loading 6

may have served as a pathway to an intensified biological response of articular 7

cartilage49-51. However, the mechanical loading may not have been too excessive to cause 8

cell death in these early stages of OA. This idea is supported by the increased PG content 9

of cartilage in the deep zone of femoral condyles in the ACLT group. On the other hand, 10

the superficial zone collagen content of the ACLT group joints was smaller than that 11

observed in the control group cartilage for the tibial plateaus and the femoral groove. This 12

finding may be explained by the excessive increase in stresses and strains at the cartilage 13

surface 46, 52. 14

The modulus analysis indicated that ACLT alters the static properties of cartilage 15

mostly in femoral condyles of the ACLT joints, while the dynamic properties are 16

deteriorated also in the non-operated, contralateral joints. The implication from this could 17

be that static loading of the joint, such as standing, allows for greater deformations of 18

cartilage primarily in femoral condyles, but only in the ACLT joint. On the other hand, 19

dynamic loading of the joint, such as running, may cause greater deformations of 20

cartilage not only in the ACLT but also in the contralateral joint. These different loading 21

conditions can then cause differently amplified deterioration of cartilage in a site- and 22

group-specific manner. 23

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

11

PGs are primarily associated with the equilibrium stiffness of articular cartilage12, 1

53. Consequently, a decrease in the PG content in the superficial zone of femoral condyle 2

cartilage in the ACLT group was associated with a decrease in the equilibrium elastic 3

modulus. In the lateral tibial plateau, the change in the PG content was not as severe as in 4

the femoral condyles, but it was still enough to cause a significant change in the 5

equilibrium modulus. In the medial tibial plateau, the change in the PG content was even 6

smaller than in the lateral tibial plateau, and no change in the equilibrium modulus could 7

be detected. Unaltered modulus might also be partly associated with the slightly increased 8

PG content in the deeper zones. In the femoral groove cartilage, the equilibrium modulus 9

was not altered due to ACLT, even though the PG content had significantly changed. 10

This may be due to the lack of sensitivity of indentation testing to the altered PG content 11

and/or the relatively small sample size. 12

The superficial collagen fiber network is a major contributor to the dynamic 13

stiffness of cartilage54-56, and we observed a strong relationship between these two 14

characteristics. For example, we found a simultaneous increase in the superficial collagen 15

orientation angle with a decrease in the dynamic elastic modulus for the femoral condyle 16

cartilages in the ACLT group compared to those in the control group. However, 17

differences in the dynamic modulus between ACLT and contralateral joint cartilages 18

were not significant, even though the collagen orientation angle was significantly 19

changed in the lateral compartment. We speculate that the increase in the collagen 20

orientation angle was offset by the increase in the collagen content, resulting in no net 21

change in the dynamic stiffness. The dynamic moduli of tibial plateau and femoral 22

groove cartilages were also unaffected by ACLT, even though their collagen orientation 23

angles were increased. In these locations, the collagen content was not altered due to 24

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

12

ACLT in the surface of cartilage. Thus, similarly as in condyles, the collagen content may 1

have controlled the tissue response and the increased orientation angle in the ACLT 2

group may not have been enough to reduce the dynamic stiffness. On the other hand, the 3

dynamic elastic modulus of the lateral condyle cartilage in the contralateral joint was only 4

~30% of that in the control joint, even though changes in the composition and structure 5

were minimal. It could be that the collagen network integrity was changed first. On the 6

other hand, indentation is very sensitive to changes in the most superficial tissue layers, 7

and thus, even small changes in the collagen orientation angles (which were statistically 8

significant here) could explain this difference. 9

The collagen orientation angle was similar for most locations in the ACLT and 10

contralateral joint cartilages, but differed significantly from the control group samples. 11

However, PG and collagen contents were primarily altered in the ACLT group, while the 12

values for the contralateral and control group cartilages were similar for most locations. 13

Mankin scores also indicated that the contralateral joint samples were mainly in good 14

condition with only minor or no signs of osteoarthritis. These findings implicate that the 15

superficial collagen fibers are exquisitely sensitive to changes in loading patterns of the 16

rabbit knee. The injured, ACLT limb may change the gait cycle for both limbs, thereby 17

affecting also the loading of the contralateral knee. 18

In the patellar cartilage from these same animals tested here13, a significant 19

difference was observed in the superficial collagen orientation angle and PG content 20

between the ACLT and contralateral groups. Here, the femoral groove, which is in 21

contact with patella, experienced only changes in the PG content of cartilage, while the 22

collagen orientation angle of the tissue was the same in the ACLT and contralateral 23

joints. As patella slides against femoral groove during the joint motion, while our chosen, 24

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

13

local measurement point in the femoral groove is in contact with patella only 1

occasionally, this may have caused greater alterations to the patellar cartilage than 2

femoral groove cartilage. On the other hand, we did not have a separate control group for 3

patella. Thus, it could also be that the contralateral joint cartilage in the femoral groove 4

had experienced earlier alterations than that in the patella. The comparison between other 5

contacting joint surfaces indicated that the collagen orientation and PG content in the 6

ACLT group cartilage differed more from the contralateral group samples in femoral 7

condyles. On the other hand, the collagen orientation angle between these groups was 8

different only in the lateral joint compartment both in femoral condyle and tibial plateau 9

cartilages. It may be that the contralateral group could be considered as a control group 10

only for some locations. 11

Mankin scores44, 45 of the femoral condyle samples were much greater than those 12

for any of the other location, but they still ranged from the mild to moderate category57. 13

The scores of the ACLT group samples were significantly greater than those of the 14

control group cartilages, except for lateral tibial plateau. These results were consistent 15

with the structural and biomechanical analyses. As Safranin O staining is one of the 16

criteria graded in Mankin scoring, consistently with the reduced PG content, an increased 17

Mankin score from the control to the contralateral group was detected only in the medial 18

tibial plateau cartilage. However, Mankin scoring could not distinguish changes observed 19

in the superficial collagen orientation angle between the control and contralateral groups. 20

There are limitations associated with this study. The ACLT and contralateral joint 21

samples were frozen prior to the biomechanical testing, while the control group samples 22

were tested fresh before embedding in formalin. This should not produce any differences, 23

because a single freeze-thaw cycle should not produce changes in the biomechanical 24

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

14

properties of articular cartilage58, 59. Some superficial PG loss has been observed earlier 1

due to freezing60, 61. However, our results showed similar PG contents for the control (not 2

frozen) and contralateral (frozen) group samples for most of the locations. Although the 3

control group samples were tested with different equipment, the authors could not detect 4

any differences in data between the different testing devices. Estimated cartilage 5

thicknesses prior to the biomechanical tests (see supplementary material) were equally 6

close to those analyzed using microscopy both in Kuopio and Calgary. True 7

displacements and forces were always checked from the raw data. Both instruments had 8

similar resolutions for displacement and force, and they handled the measurements well 9

within the required strain rates. Potentially different initial contact between the indenter 10

and sample, caused possibly by different operator, was checked by confirming the 11

linearity of the equilibrium stress-strain curves. 12

We acknowledge the small sample size in the control group (3 rabbits, 6 knees). 13

However, as the mixed model ANOVA takes into account animal specific variations, we 14

could use both knees of these animals. Yet, the small sample size in the control group 15

might have been the cause for the lack of some significant differences. For instance, the 16

equilibrium modulus of the lateral femoral condyle cartilage was over 50% smaller in the 17

ACLT than control group animals, but this difference was not statistically proven (p = 18

0.117). 19

There are different mechanical testing geometries, all with their advantages and 20

disadvantages. We chose indentation testing, as it has good sensitivity to reveal changes 21

in the mechanical properties of the superficial zone of cartilage, the zone which has been 22

shown to be the first to experience osteoarthritic changes6, 13-16, 55. Specifically, the 23

importance of the collagen fibril network in the superficial zone for the indentation 24

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

15

stiffness has been demonstrated earlier6, 13-16, 18, 21, 54, 55, 62. Also, indentation loading 1

allows for testing intact articular cartilage attached to its native bone. Other mechanical 2

testing geometries, such as unconfined or confined compression testing, might be more 3

sensitive in identifying the predominantly PG defined equilibrium properties, especially 4

those in the deeper cartilage layers35, 55. However, these geometries would have required 5

removing the cartilage samples from their native environment, which might cause 6

loosening of the collagen fibril tension and altering the mechanical response of the 7

tissue35, 63. 8

Test animals received antibiotics and pain medication to reduce inflammation and 9

pain caused by the surgery. When opened, the joints showed no macroscopic signs of 10

cartilage degeneration. However, inflammation in traumatized cartilage has been shown 11

earlier64. We also do not know if the surgery itself caused inflammatory cytokine 12

production. These might have contributed to the results. A sham-operated group of 13

animals might have allowed us to address this question, but that was not done here. 14

The rabbit knee differs in shape from the human knee and exceeds the human 15

knee’s range of motion65. These differences have to be considered when attempting to use 16

the present results in the context of the human knee and its development of early OA. 17

Nevertheless, the basic bony structure, ligamentous stabilization and muscular control are 18

similar between the rabbit and human knee, and OA development, albeit much faster in 19

the rabbit than in humans following ACLT, follows a similar pathological sequence of 20

events3. 21

In conclusion, our findings provide novel information on the site-dependent 22

changes in the mechanical and structural properties of articular cartilage in very early 23

experimental OA. They suggest that the femoral condylar cartilage is affected the most in 24

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

16

early OA induced by ACL injury (Table 4), while tibial plateau and femoral groove 1

cartilages seem more resistant to the altered loading conditions following ACLT. Still, 2

even the non-weight-bearing cartilage in the femoral groove cartilage showed early OA-3

like changes. The change in loading following ACLT led to degenerative alterations in 4

the superficial collagen fiber organization and proteoglycan content. The results of this 5

study also suggests a possible increase in biological response, causing increases in the 6

collagen content in the middle and deep zones of femoral condyle cartilage. ACLT causes 7

specific and site-dependent alterations in cartilage structure and function which must be 8

considered when drawing conclusions from the results of early OA studies. The current 9

findings may help to optimize strategies for the prevention of OA, provide novel ways for 10

predicting the time course of OA onset and progression, and maybe used to develop 11

effective treatment strategies aimed at stopping or slowing OA progression. 12

13

ACKNOWLEDGEMENTS 14

Finnish Cultural Foundation; European Research Council under the European Union’s 15

Seventh Framework Programme (FP/2007–2013) ERC Grant Agreement no. 281180; 16

Academy of Finland (grants no. 140730 and 218038); Kuopio University Hospital (EVO 17

15283); Sigrid Juselius Foundation; AI-HS Team Grant on Osteoarthritis; Dr. Rezaeian 18

was funded by the 6 months research opportunity award from the Ministry of Hygiene, 19

Treatment and Medical Education of Iran. Authors want to thank Tim Leonard, Ph.D, 20

University of Calgary, for performing the ACL transections, Siru Turunen, Ph.D, 21

University of Eastern Finland, for assistance in tissue preparation, and Mr Jari 22

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

17

Rautiainen, M.Sc, and Mr Ari Petteri Ronkainen, M.Sc., University of Eastern Finland 1

for assistance in the measurements. 2

3

CONTRIBUTIONS 4

All authors contributed to the conception and design of the study, the data acquisition, 5

analysis, and interpretation of the results. The manuscript was drafted, revised and finally 6

approved by all authors. Mäkelä JTA ([email protected]) takes responsibility for the 7

integrity of the work. 8

9

ETHICS APPROVAL OF RESEARCH ON ANIMALS 10

All animal procedures were approved by the Animal Ethics committee at the University 11

of Calgary and the guidelines of the Canadian Council on Animal Care were followed. 12

13

ROLE OF THE FUNDING SOURCE 14

Funding sources did not have any involvement in the study design, collection, analysis 15

and interpretation of data, in the writing of the manuscript or in the decision to submit the 16

manuscript for publication. 17

18

COMPETING INTERESTS 19

Authors have no competing interests. 20

21

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

18

REFERENCES 1

1. Pond MJ and Nuki G. Experimentally-induced osteoarthritis in the dog. 2

Ann.Rheum.Dis. 1973;32:387-8. 3

2. Vignon E, Bejui J, Mathieu P, Hartmann JD, Ville G, Evreux JC, et al. Histological 4

cartilage changes in a rabbit model of osteoarthritis. J.Rheumatol. 1987;14 Spec No:104-5

6. 6

3. Altman RD and Dean DD. Osteoarthritis research: animal models. Semin.Arthritis 7

Rheum. 1990;19:21-5. 8

4. Laverty S, Girard CA, Williams JM, Hunziker EB, Pritzker KP. The OARSI 9

histopathology initiative - recommendations for histological assessments of osteoarthritis 10

in the rabbit. Osteoarthritis Cartilage 2010;18 Suppl 3:S53-65. 11

5. Vignon E, Hartmann JD, Vignon G, Moyen B, Arlot M, Ville G. Cartilage destruction 12

in experimentally induced osteoarthritis. J.Rheumatol. 1984;11:202-7. 13

6. Stoop R, Buma P, van der Kraan PM, Hollander AP, Billinghurst RC, Meijers TH, et 14

al. Type II collagen degradation in articular cartilage fibrillation after anterior cruciate 15

ligament transection in rats. Osteoarthritis Cartilage 2001;9:308-15. 16

7. Sah RL, Yang AS, Chen AC, Hant JJ, Halili RB, Yoshioka M, et al. Physical 17

properties of rabbit articular cartilage after transection of the anterior cruciate ligament. 18

J.Orthop.Res. 1997;15:197-203. 19

8. Li G, Moses JM, Papannagari R, Pathare NP, DeFrate LE, Gill TJ. Anterior cruciate 20

ligament deficiency alters the in vivo motion of the tibiofemoral cartilage contact points 21

in both the anteroposterior and mediolateral directions. J.Bone Joint Surg.Am. 22

2006;88:1826-34. 23

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

19

9. Lohmander LS, Englund PM, Dahl LL, Roos EM. The long-term consequence of 1

anterior cruciate ligament and meniscus injuries: osteoarthritis. Am.J.Sports Med. 2

2007;35:1756-69. 3

10. Chaudhari AM, Briant PL, Bevill SL, Koo S, Andriacchi TP. Knee kinematics, 4

cartilage morphology, and osteoarthritis after ACL injury. Med.Sci.Sports Exerc. 5

2008;40:215-22. 6

11. Potter HG, Jain SK, Ma Y, Black BR, Fung S, Lyman S. Cartilage injury after acute, 7

isolated anterior cruciate ligament tear: immediate and longitudinal effect with 8

clinical/MRI follow-up. Am.J.Sports Med. 2012;40:276-85. 9

12. Mow VC and Hayes WC. Basic Orthopaedic Biomechanics. New York: Raven Press, 10

Ltd 1991. 11

13. Turunen SM, Han SK, Herzog W, Korhonen RK. Cell deformation behavior in 12

mechanically loaded rabbit articular cartilage 4 weeks after anterior cruciate ligament 13

transection. Osteoarthritis Cartilage 2013;21:505-13. 14

14. Guilak F, Ratcliffe A, Lane N, Rosenwasser MP, Mow VC. Mechanical and 15

biochemical changes in the superficial zone of articular cartilage in canine experimental 16

osteoarthritis. J.Orthop.Res. 1994;12:474-84. 17

15. Panula HE, Hyttinen MM, Arokoski JP, Langsjo TK, Pelttari A, Kiviranta I, et al. 18

Articular cartilage superficial zone collagen birefringence reduced and cartilage thickness 19

increased before surface fibrillation in experimental osteoarthritis. Ann.Rheum.Dis. 20

1998;57:237-45. 21

16. Saarakkala S, Julkunen P, Kiviranta P, Makitalo J, Jurvelin JS, Korhonen RK. Depth-22

wise progression of osteoarthritis in human articular cartilage: investigation of 23

composition, structure and biomechanics. Osteoarthritis Cartilage 2010;18:73-81. 24

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

20

17. Setton LA, Mow VC, Howell DS. Mechanical behavior of articular cartilage in shear 1

is altered by transection of the anterior cruciate ligament. J.Orthop.Res. 1995;13:473-82. 2

18. Buckwalter JA and Mankin HJ. Articular Cartilage, Part II: Degeneration and 3

osteoarthrosis, repair, regeneration, and transplantation. J.Bone Joint Surg.Am. 4

1997;79:612-632. 5

19. Setton LA, Mow VC, Muller FJ, Pita JC, Howell DS. Mechanical properties of canine 6

articular cartilage are significantly altered following transection of the anterior cruciate 7

ligament. J.Orthop.Res. 1994;12:451-63. 8

20. Intema F, Hazewinkel HA, Gouwens D, Bijlsma JW, Weinans H, Lafeber FP, et al. In 9

early OA, thinning of the subchondral plate is directly related to cartilage damage: results 10

from a canine ACLT-meniscectomy model. Osteoarthritis Cartilage 2010;18:691-8. 11

21. Yoshioka M, Coutts RD, Amiel D, Hacker SA. Characterization of a model of 12

osteoarthritis in the rabbit knee. Osteoarthritis Cartilage 1996;4:87-98. 13

22. Rijk PC, Tigchelaar-Gutter W, Bernoski FP, Van Noorden CJ. Histologic changes in 14

articular cartilage after medial meniscus replacement in rabbits. Arthroscopy 15

2004;20:911-7. 16

23. Han SK, Seerattan R, Herzog W. Mechanical loading of in situ chondrocytes in lapine 17

retropatellar cartilage after anterior cruciate ligament transection. J.R.Soc.Interface 18

2010;7:895-903. 19

24. Barton KI, Ludwig TE, Achari Y, Shrive NG, Frank CB, Schmidt TA. 20

Characterization of proteoglycan 4 and hyaluronan composition and lubrication function 21

of ovine synovial fluid following knee surgery. J.Orthop.Res. 2013; 22

25. Wyland DJ, Guilak F, Elliott DM, Setton LA, Vail TP. Chondropathy after meniscal 23

tear or partial meniscectomy in a canine model. J.Orthop.Res. 2002;20:996-1002. 24

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

21

26. Borrelli J,Jr, Silva MJ, Zaegel MA, Franz C, Sandell LJ. Single high-energy impact 1

load causes posttraumatic OA in young rabbits via a decrease in cellular metabolism. 2

J.Orthop.Res. 2009;27:347-52. 3

27. Rogart JN, Barrach HJ, Chichester CO. Articular collagen degradation in the Hulth-4

Telhag model of osteoarthritis. Osteoarthritis Cartilage 1999;7:539-47. 5

28. Brophy RH, Martinez M, Borrelli J,Jr, Silva MJ. Effect of combined traumatic impact 6

and radial transection of medial meniscus on knee articular cartilage in a rabbit in vivo 7

model. Arthroscopy 2012;28:1490-6. 8

29. Bi X, Yang X, Bostrom MP, Bartusik D, Ramaswamy S, Fishbein KW, et al. Fourier 9

transform infrared imaging and MR microscopy studies detect compositional and 10

structural changes in cartilage in a rabbit model of osteoarthritis. Anal.Bioanal Chem. 11

2007;387:1601-12. 12

30. Altman RD, Tenenbaum J, Latta L, Riskin W, Blanco LN, Howell DS. 13

Biomechanical and biochemical properties of dog cartilage in experimentally induced 14

osteoarthritis. Ann.Rheum.Dis. 1984;43:83-90. 15

31. Carney SL, Billingham ME, Muir H, Sandy JD. Demonstration of increased 16

proteoglycan turnover in cartilage explants from dogs with experimental osteoarthritis. 17

J.Orthop.Res. 1984;2:201-6. 18

32. Wei X, Rasanen T, Messner K. Maturation-related compressive properties of rabbit 19

knee articular cartilage and volume fraction of subchondral tissue. Osteoarthritis 20

Cartilage 1998;6:400-9. 21

33. Julkunen P, Harjula T, Iivarinen J, Marjanen J, Seppanen K, Narhi T, et al. 22

Biomechanical, biochemical and structural correlations in immature and mature rabbit 23

articular cartilage. Osteoarthritis Cartilage 2009;17:1628-38. 24

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

22

34. Hayes WC, Keer LM, Herrmann G, Mockros LF. A mathematical analysis for 1

indentation tests of articular cartilage. J.Biomech. 1972;5:541-51. 2

35. Korhonen RK, Laasanen MS, Toyras J, Rieppo J, Hirvonen J, Helminen HJ, et al. 3

Comparison of the equilibrium response of articular cartilage in unconfined compression, 4

confined compression and indentation. J.Biomech. 2002;35:903-9. 5

36. Brommer H, Brama PA, Laasanen MS, Helminen HJ, van Weeren PR, Jurvelin JS. 6

Functional adaptation of articular cartilage from birth to maturity under the influence of 7

loading: a biomechanical analysis. Equine Vet.J. 2005;37:148-54. 8

37. Kiviranta I, Jurvelin J, Tammi M, Saamanen AM, Helminen HJ. 9

Microspectrophotometric quantitation of glycosaminoglycans in articular cartilage 10

sections stained with Safranin O. Histochemistry 1985;82:249-55. 11

38. Kiraly K, Lapvetelainen T, Arokoski J, Torronen K, Modis L, Kiviranta I, et al. 12

Application of selected cationic dyes for the semiquantitative estimation of 13

glycosaminoglycans in histological sections of articular cartilage by 14

microspectrophotometry. Histochem.J. 1996;28:577-90. 15

39. Rieppo J, Hallikainen J, Jurvelin JS, Kiviranta I, Helminen HJ, Hyttinen MM. 16

Practical considerations in the use of polarized light microscopy in the analysis of the 17

collagen network in articular cartilage. Microsc.Res.Tech. 2008;71:279-87. 18

40. Rieppo J, Hyttinen MM, Halmesmaki E, Ruotsalainen H, Vasara A, Kiviranta I, et al. 19

Changes in spatial collagen content and collagen network architecture in porcine articular 20

cartilage during growth and maturation. Osteoarthritis Cartilage 2009;17:448-55. 21

41. Bi X, Yang X, Bostrom MP, Camacho NP. Fourier transform infrared imaging 22

spectroscopy investigations in the pathogenesis and repair of cartilage. 23

Biochim.Biophys.Acta 2006;1758:934-41. 24

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

23

42. Boskey A and Pleshko Camacho N. FT-IR imaging of native and tissue-engineered 1

bone and cartilage. Biomaterials 2007;28:2465-78. 2

43. Camacho NP, West P, Torzilli PA, Mendelsohn R. FTIR microscopic imaging of 3

collagen and proteoglycan in bovine cartilage. Biopolymers 2001;62:1-8. 4

44. Mankin HJ, Dorfman H, Lippiello L, Zarins A. Biochemical and metabolic 5

abnormalities in articular cartilage from osteo-arthritic human hips. II. Correlation of 6

morphology with biochemical and metabolic data. J.Bone Joint Surg.Am. 1971;53:523-7

37. 8

45. Mankin HJ and Lippiello L. Biochemical and metabolic abnormalities in articular 9

cartilage from osteo-arthritic human hips. J.Bone Joint Surg.Am. 1970;52:424-34. 10

46. Van de Velde SK, Bingham JT, Hosseini A, Kozanek M, DeFrate LE, Gill TJ, et al. 11

Increased tibiofemoral cartilage contact deformation in patients with anterior cruciate 12

ligament deficiency. Arthritis Rheum. 2009;60:3693-702. 13

47. Hotta H, Yamada H, Takaishi H, Abe T, Morioka H, Kikuchi T, et al. Type II 14

collagen synthesis in the articular cartilage of a rabbit model of osteoarthritis: expression 15

of type II collagen C-propeptide and mRNA especially during early-stage osteoarthritis. 16

J.Orthop.Sci. 2005;10:595-607. 17

48. Tanska P, Turunen SM, Han SK, Julkunen P, Herzog W, Korhonen RK. Superficial 18

collagen fibril modulus and pericellular fixed charge density modulate chondrocyte 19

volumetric behaviour in early osteoarthritis. Comput.Math.Methods Med. 20

2013;2013:164146. 21

49. Natoli RM and Athanasiou KA. Traumatic loading of articular cartilage: Mechanical 22

and biological responses and post-injury treatment. Biorheology 2009;46:451-85. 23

50. Urban JP. The chondrocyte: a cell under pressure. Br.J.Rheumatol. 1994;33:901-8. 24

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

24

51. Le Graverand MP, Eggerer J, Vignon E, Otterness IG, Barclay L, Hart DA. 1

Assessment of specific mRNA levels in cartilage regions in a lapine model of 2

osteoarthritis. J.Orthop.Res. 2002;20:535-44. 3

52. Bendjaballah MZ, Shirazi-Adl A, Zukor DJ. Biomechanical response of the passive 4

human knee joint under anterior-posterior forces. Clin.Biomech.(Bristol, Avon) 5

1998;13:625-33. 6

53. Korhonen RK, Laasanen MS, Toyras J, Lappalainen R, Helminen HJ, Jurvelin JS. 7

Fibril reinforced poroelastic model predicts specifically mechanical behavior of normal, 8

proteoglycan depleted and collagen degraded articular cartilage. J.Biomech. 9

2003;36:1373-9. 10

54. Makela JT, Huttu MR, Korhonen RK. Structure-function relationships in 11

osteoarthritic human hip joint articular cartilage. Osteoarthritis Cartilage 2012;20:1268-12

77. 13

55. Korhonen RK, Wong M, Arokoski J, Lindgren R, Helminen HJ, Hunziker EB, et al. 14

Importance of the superficial tissue layer for the indentation stiffness of articular 15

cartilage. Med.Eng.Phys. 2002;24:99-108. 16

56. Gannon AR, Nagel T, Kelly DJ. The role of the superficial region in determining the 17

dynamic properties of articular cartilage. Osteoarthritis Cartilage 2012;20:1417-25. 18

57. Ostergaard K, Andersen CB, Petersen J, Bendtzen K, Salter DM. Validity of 19

histopathological grading of articular cartilage from osteoarthritic knee joints. 20

Ann.Rheum.Dis. 1999;58:208-13. 21

58. Changoor A, Fereydoonzad L, Yaroshinsky A, Buschmann MD. Effects of 22

refrigeration and freezing on the electromechanical and biomechanical properties of 23

articular cartilage. J.Biomech.Eng. 2010;132:064502. 24

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

25

59. Kiefer GN, Sundby K, McAllister D, Shrive NG, Frank CB, Lam T, et al. The effect 1

of cryopreservation on the biomechanical behavior of bovine articular cartilage. 2

J.Orthop.Res. 1989;7:494-501. 3

60. Laouar L, Fishbein K, McGann LE, Horton WE, Spencer RG, Jomha NM. 4

Cryopreservation of porcine articular cartilage: MRI and biochemical results after 5

different freezing protocols. Cryobiology 2007;54:36-43. 6

61. Zheng S, Xia Y, Bidthanapally A, Badar F, Ilsar I, Duvoisin N. Damages to the 7

extracellular matrix in articular cartilage due to cryopreservation by microscopic 8

magnetic resonance imaging and biochemistry. Magn.Reson.Imaging 2009;27:648-55. 9

62. Julkunen P, Jurvelin JS, Isaksson H. Contribution of tissue composition and structure 10

to mechanical response of articular cartilage under different loading geometries and strain 11

rates. Biomech.Model.Mechanobiol 2010;9:237-45. 12

63. Turunen SM, Lammi MJ, Saarakkala S, Koistinen A, Korhonen RK. Hypotonic 13

challenge modulates cell volumes differently in the superficial zone of intact articular 14

cartilage and cartilage explant. Biomech.Model.Mechanobiol 2012;11:665-75. 15

64. Goldring MB and Otero M. Inflammation in osteoarthritis. Curr.Opin.Rheumatol. 16

2011;23:471-8. 17

65. Proffen BL, McElfresh M, Fleming BC, Murray MM. A comparative anatomical 18

study of the human knee and six animal species. Knee 2012;19:493-9. 19

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPTFIGURE LEGENDS

Fig 1: Analyzed locations in the lapine knee joints. Articular cartilage samples in the lateral and medial femoral condyles (A), tibial plateaus (B) and femoral groove (C) were biomechanically tested and used for determinations of tissue structure and composition. Indentation locations are marked with black dots, while the directions of slices prepared for microscopy and spectroscopy are marked with solid lines. Spaces between lines are not to scale. Fig 2: Microscopic images of safranin O stained sections (left), collagen orientation maps (center) and amide I absorption maps (right) used to analyze proteoglycan content, collagen orientation angle and collagen content, respectively, using digital densitometry (DD), polarized light microscopy (PLM) and Fourier Transform Infrared (FTIR) spectroscopy. Representative examples from ACLT, C-L and CNTRL groups have been presented. Fig 3: Mean values (and 95% CI) of PG content profiles in different groups and sites measured with digital densitometry. Dashed line in the top shows spatially significant differences between two groups of those colours (p < 0.05). Fig 4: Mean values (and 95% CI) of collagen orientation profiles in different groups and sites measured with polarized microscopy. Dashed line in the top shows spatially significant differences between two groups of those colours (p < 0.05). Fig 5: Mean values (and 95% CI) of collagen content profiles in different groups and sites measured with FTIR. Dashed line in the top shows spatially significant difference between two groups of those colours (p < 0.05). Table 1. Mean values (95% CI) of Mankin scores. Table 2. Mean values (95% CI) of cartilage equilibrium elastic modulus (MPa) Table 3. Mean values (95% CI) of cartilage dynamic elastic modulus (MPa) Table 4. Summary of the changes in the structure and function of early osteoarthritic cartilage, as caused by ACLT, in the superficial (0-5 %, S), middle (5-15 %, M) and deep (15-100 %, D) zones of the tissue. Tissues were characterized 4 weeks after ACLT. In this comparison, the ACLT group samples have been compared to the control group samples.

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

Table 1

ACLT C-L CTRL

Femoral condyle Lat 4.5 (1.7-7.3)*,** 0.9 (0.2-1.6) 0.7 (0-2)

Med 4.8 (3.0-6.5)*,** 1.4 (0.6-2.1) 0 (0-0)Tibial plateau Lat 1.0 (0.3-1.7) 0.4 (0-0.7) 0.2 (0-0.5)

Med 1.9 (1.4-2.3)* 1.6 (1.3-2.0)* 0.2 (0-0.5)Femoral groove 1.3 (0.8-1.7)* 0.8 (0.3-1.2) 0.2 (0-0.5)* p < 0.05, compared to CTRL** p < 0.05, comparison between ACLT and C-L

Sivu 1

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPTTable 2

ACLT C-L CTRLFemoral condyle Lat 0.58 (0.22-0.95)** 1.12 (0.76-1.48) 1.27 (0.73-1.80)

Med 0.61 (0.11-1.11)*,** 1.26 (0.76-1.76) 1.99 (1.24-2.75)Tibial plateau Lat 0.24 (0-0.6)* 0.77 (0.43-1.11) 1.26 (0.87-1.66)

Med 0.63 (0.45-0.82) 0.81 (0.61-1.01) 0.56 (0.29-0.83)Femoral groove 0.76 (0.35-1.17) 0.79 (0.40-1.18) 0.72 (0.13-1.31)* p < 0.05, compared to CTRL** p < 0.05, comparison between ACLT and C-L

Sivu 1

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPTTable 3

ACLT C-L CTRLFemoral condyle Lat 2.86 (0.87-4.86)* 3.44 (1.44-5.44)* 9.97 (7.66-12.28)

Med 1.98 (0.98-2.98)* 3.51 (2.52-4.51)* 6.49 (5.08-7.90)Tibial plateau Lat 3.26 (1.88-4.65) 2.92 (1.53-4.30) 5.97 (3.85-8.09)

Med 2.25 (1.85-2.64) 1.93 (1.54-2.33) 1.69 (1.10-2.28)Femoral groove 4.80 (2.73-6.87) 4.54 (2.46-6.61) 4.94 (1.73-8.14)* p < 0.05, compared to CTRL** p < 0.05, comparison between ACLT and C-L

Sivu 1

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

Table 4

Femoral condyle Tibial plateau Femoral grooveLat Med Lat Med

Mankin score ↑ ↑ ↔ ↑ ↑

Equilibrium modulus ↔ ↓ ↓ ↔ ↔

Dynamic modulus ↓ ↓ ↔ ↔ ↔

PG content S ↓ ↓ ↓ ↓ ↓

M ↓ ↓ ↔ ↔ ↓

D ↑ ↔ ↔ ↑ ↔

Collagen S ↑ ↑ ↑ ↑ ↑

orientation angle M ↑ ↔ ↔ ↑ ↑

D ↑ ↔ ↔ ↔ ↔

Collagen content S ↑ ↑ ↔ ↔ ↔

M ↑ ↔ ↓ ↓ ↓

D ↑ ↑ ↔ ↔ ↔

Sivu 1

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

DD / PG content

ACLT

Thi

ckne

ss (

µm)

0

100

200

300

400

500C−L CNTRL

Collagen orientation angle (°)0 20 40 60 80

PLM / Collagen orientation

ACLT

Thi

ckne

ss (

µm)

0

100

200

300

400

500C−L

CNTRL

Collagen content (AU)20 40 60

FTIR / Collagen content

ACLT

Thi

ckne

ss (

µm)

0

100

200

300

400

500C−L

CNTRL

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

0 20 40 60 80 1000

0.5

1

1.5

DD Femur latA

Normalized depth (%)

PG

co

nte

nt

(AU

)

ACLTC−LCNTRL

0 20 40 60 80 1000

0.5

1

1.5

DD Femur medB

Normalized depth (%)

PG

co

nte

nt

(AU

)

ACLTC−LCNTRL

0 20 40 60 80 1000

0.5

1

1.5

DD Tibia latC

Normalized depth (%)

PG

co

nte

nt

(AU

)

ACLTC−LCNTRL

0 20 40 60 80 1000

0.5

1

1.5

DD Tibia medD

Normalized depth (%)

PG

co

nte

nt

(AU

)

ACLTC−LCNTRL

0 20 40 60 80 1000

0.5

1

1.5

DD GrooveE

Normalized depth (%)

PG

co

nte

nt

(AU

)

ACLTC−LCNTRL

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

0 20 40 60 80 1000

20

40

FTIR Femur latA

Normalized depth (%)

Co

llag

en c

on

ten

t (A

U)

ACLTC−LCNTRL

0 20 40 60 80 1000

20

40

FTIR Femur medB

Normalized depth (%)

Co

llag

en c

on

ten

t (A

U)

ACLTC−LCNTRL

0 20 40 60 80 1000

20

40

FTIR Tibia latC

Normalized depth (%)

Co

llag

en c

on

ten

t (A

U)

ACLTC−LCNTRL

0 20 40 60 80 1000

20

40

FTIR Tibia medD

Normalized depth (%)

Co

llag

en c

on

ten

t (A

U)

ACLTC−LCNTRL

0 20 40 60 80 1000

20

40

60

FTIR GrooveE

Normalized depth (%)

Co

llag

en c

on

ten

t (A

U)

ACLTC−LCNTRL

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

0 20 40 60 80 1000

20

40

60

80

PLM Femur latA

Normalized depth (%)

Co

llag

en o

rien

tati

on

an

gle

(°)

ACLTC−LCNTRL

0 20 40 60 80 1000

20

40

60

80

PLM Femur medB

Normalized depth (%)

Co

llag

en o

rien

tati

on

an

gle

(°)

ACLTC−LCNTRL

0 20 40 60 80 1000

20

40

60

80

PLM Tibia latC

Normalized depth (%)

Co

llag

en o

rien

tati

on

an

gle

(°)

ACLTC−LCNTRL

0 20 40 60 80 1000

20

40

60

80

PLM Tibia medD

Normalized depth (%)

Co

llag

en o

rien

tati

on

an

gle

(°)

ACLTC−LCNTRL

0 20 40 60 80 1000

20

40

60

80

PLM GrooveE

Normalized depth (%)

Co

llag

en o

rien

tati

on

an

gle

(°)

ACLTC−LCNTRL

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

1

Site-Dependent Changes in Structure and Function of Lapine Articular 1

Cartilage 4 Weeks After Anterior Cruciate Ligament Transection 2

Supplementary material 3

SAMPLES AND PROCESSING

Knee joints were dissected, immersed into PBS and ACLT and C-L groups were frozen 4

in dry ice and shipped from Calgary, Alberta, Canada to Kuopio, Finland. In Kuopio the 5

knee joints were then thawed slowly at room temperature and cartilage-on-bone samples 6

were harvested and used for biomechanical testing (Fig. 1). Throughout the 7

measurements, cartilage samples were kept in phosphate buffered saline (PBS) containing 8

enzyme inhibitors. Biomechanical measurements were conducted on the femoral 9

condyles and tibial plateaus in the center of the weight-bearing areas and on the femoral 10

groove in the center of the contact area with patella with the knee in its neutrally flexed 11

position1, 2. These locations were defined as the top point (apex of the posterior curvature) 12

of femoral condyles and groove and the central point of tibial plateaus for femur and 13

tibia, respectively. Measurement locations in tibial plateaus were also close to the inner 14

edge of meniscus. The measuring spots were marked with fine point permanent marker 15

and the samples were fixed in formalin. The CNTRL group samples were tested 16

biomechanically first in Calgary similarly as the ACLT and C-L joint samples before 17

shipping them to Kuopio in formalin fixative. Finally all of the samples were decalcified, 18

processed in graded alcohol solutions, and embedded in paraffin. Histological sections 19

were then cut using a microtome (LKB 2218 HistoRange mocrotome, LKB produkter 20

Ab, Bromma, Sweden) perpendicular to the cartilage surface. Digital densitometry (DD) 21

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

2

sections (thickness ~3 µm) were stained with Safranin O3 while Fourier transform 1

infrared spectroscopy (FTIR) and polarized light microscopy (PLM) sections (thickness ~ 2

5 µm) underwent enzymatic removal of PGs prior to the measurements using 3

hyaluronidase (1000 U/ml hyaluronidase, SigmaeAldrich, St. Louis, MO, USA)4. ACLT 4

procedures for rabbits were approved by the Animal Ethics committee at the University 5

of Calgary and the guidelines of the Canadian Council on Animal Care were followed. 6

7

BIOMECHANICAL TESTING

Cartilage thickness was first measured using a high resolution ultrasound system5, 6 (Clear 8

View Ultra, Boston Scientific Corporation, San Jose, CA, USA). Thicknesses estimated 9

for CNTRL group cartilages were based on preliminary measurements using a needle-10

probe method. The instruments used for the biomechanical measurements were a custom 11

made high-precision material testing device7 (resolution: 0.1 µm, 0.005 N) in Kuopio and 12

an MTS 858 Mini Bionix II (MTS Systems, Eden Prairie, MN, USA; resolution: 0.1 µm, 13

0.001 N) in Calgary. The bone of the samples was glued to the bottom of a measuring 14

chamber. The chamber was filled with PBS, and an indenter (diameter 1 mm) was 15

brought into contact with the sample surface. Compared to the tissue samples (diameter > 16

5 mm), the indenter was small enough to eliminate any effects from the sample edges8. 17

Perpendicularity between cartilage and the indenter surface was confirmed using a 18

goniometer. Preconditioning of the samples was done using a cyclic 2 % strain (4 full 19

cycles), after which the samples were allowed to relax for ~15 min. A stress-relaxation 20

protocol was implemented using a ramp rate of 100 %/s (350 ± 150 µm/s), 3 x 5 % steps 21

and 15 min relaxation time after each step. Preliminary tests showed that 15 min was 22

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

3

sufficient time to reach equilibrium. This led to the relaxation rate of less than 10 Pa/min. 1

Finally, a sinusoidal dynamic test was performed with a frequency of 1 Hz (amplitude 4 2

% of remaining thickness, 4 cycles). The slope from the equilibrium stress-strain curves 3

and the peak-to-peak stress-strain ratio from the sinusoidal tests (average of 4 cycles) 4

were first determined. Both equilibrium and dynamic linearly elastic moduli were then 5

determined using the equation derived from the solution of Hayes et. al.9: 6

7

8 , (1)

9

where E is the Young´s modulus, ν is the Poisson's ratio, a is the indenter radius, κ is the 10

scaling factor (function of aspect-ratio (a/h) and ν9), h is the sample thickness and P is the 11

measured slope (equilibrium) or stress-strain ratio (dynamic). This equation takes into 12

account the finite thickness of cartilage and has been applied earlier for the calculation of 13

both the equilibrium and dynamic modulus of cartilage7, 10. The Poisson's ratio was 14

assumed to be 0.1 for the equilibrium and 0.5 (incompressible) for the dynamic modulus 15

calculations11, 12. The initially approximated thickness of each sample was confirmed by 16

histology and corrected for the modulus calculations if needed. 17

DIGITAL DENSITOMETRY

The spatial distribution of PGs was determined using DD for Safranin O stained 18

histological sections3, 13. Four sections from each sample were measured using a light 19

microscope (4x magnification) (Olympus CH-2, Olympus, Tokyo, Japan) with a 20

computer-controlled Perkin Elmer UltraPix FSI CCD camera (Perkin Elmer, MA, USA). 21

Grayscale images were captured and the optical density was used to estimate the 22

E=(1− ν2)πa

2κhP

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

4

distribution of tissue fixed charge density. The pixel values of 1.4 µm were averaged in 1

the transverse direction and the average depth-wise profiles were created from the 2

sections of each sample at 1% increments of normalized depth. 3

POLARIZED MICROSCOPY

Collagen orientation angles were analyzed with PLM14, 15 on a Leitz Ortholux II POL-4

polarized microscope (Leitz, Wetzlar, Germany) equipped with a CCD camera 5

(Photometrics CH 250/A, Photometrics Inc., Tucson, AZ, USA). Four sections were 6

measured from each sample. Images were obtained using a magnification of 6.3x and the 7

measured pixel values of 3.65 µm were averaged in the transverse direction. Mean 8

profiles for every sample were calculated from the section profiles at 1% increments of 9

normalized depth. In the profiles, collagen fibers parallel to the cartilage surface were 10

assigned an angle of 0° (superficial zone) and fibers perpendicular to the cartilage surface 11

were given an angle of 90° (deep zone). 12

FOURIER TRANSFORM INFRARED SPECTROSCOPIC IMAGING

Spatial distribution of the collagen content was determined using FTIR16-19 using a 13

PerkinElmer Spotlight 300 (PerkinElmer, Shelton, CO, USA) with CO2-free dry air purge 14

system (FT-IR purge gas generator, Parker Hannifin corporation, Haverhill, MA, USA). 15

Microscopic sections were installed on BaF2 windows and two sections were measured 16

from each sample. 4 cm-1 spectral resolution and 6.25 µm pixel resolution with 2 repeated 17

scans was used. The collagen content was estimated by integration of the amide I region 18

(1585 – 1720 cm-1). Amide I maps were averaged in the transverse direction, and a mean 19

profile of two sections was calculated to create depth-wise collagen profiles for each 20

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

5

sample at 1% increments of normalized depth. A custom made Matlab script was used for 1

IR spectra analysis. 2

STATISTICAL ANALYSIS

Statistical point-by-point comparisons between the groups were made using a linear 3

mixed model ANOVA. Using this approach, potential interrelations between the samples, 4

i.e., dependence between the samples from the same animal, could be taken into account. 5

Knee joint locations (femoral condyle, femoral groove, tibial plateau) were analyzed 6

separately. The sample group was set as a fixed-type variable, in terms of mixed models 7

notation, and the measurement of the standardized depth (1-100) was used both as a 8

fixed-type variable, to detect the overall effect of the depth, and as an animal specific 9

random-type variable, to take into account the variation caused by the individual animals. 10

Restricted maximum likelihood (REML) estimation was used in the model because it 11

produces unbiased estimates for the variance and covariance parameters. Estimated 12

means for the different groups (ACLT, C-L, CNTRL) were then obtained from the fitted 13

model. Main effects between the groups were compared, and 95% confidence intervals 14

for profiles with Bonferroni corrected adjustment are presented. Statistical analyses were 15

carried out using SPSS (ver. 19, SPSS Inc., Chicago, IL). 16

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

6

REFERENCES 1

1. Wei X, Rasanen T, Messner K. Maturation-related compressive properties of rabbit 2

knee articular cartilage and volume fraction of subchondral tissue. Osteoarthritis 3

Cartilage 1998;6:400-9. 4

2. Julkunen P, Harjula T, Iivarinen J, Marjanen J, Seppanen K, Narhi T, et al. 5

Biomechanical, biochemical and structural correlations in immature and mature rabbit 6

articular cartilage. Osteoarthritis Cartilage 2009;17:1628-38. 7

3. Kiviranta I, Jurvelin J, Tammi M, Saamanen AM, Helminen HJ. 8

Microspectrophotometric quantitation of glycosaminoglycans in articular cartilage 9

sections stained with Safranin O. Histochemistry 1985;82:249-55. 10

4. Turunen SM, Han SK, Herzog W, Korhonen RK. Cell deformation behavior in 11

mechanically loaded rabbit articular cartilage 4 weeks after anterior cruciate ligament 12

transection. Osteoarthritis Cartilage 2013;21:505-13. 13

5. Toyras J, Laasanen MS, Saarakkala S, Lammi MJ, Rieppo J, Kurkijarvi J, et al. Speed 14

of sound in normal and degenerated bovine articular cartilage. Ultrasound Med.Biol. 15

2003;29:447-54. 16

6. Huang YP and Zheng YP. Intravascular Ultrasound (IVUS): A Potential Arthroscopic 17

Tool for Quantitative Assessment of Articular Cartilage. Open Biomed.Eng.J. 2009;3:13-18

20. 19

7. Korhonen RK, Laasanen MS, Toyras J, Rieppo J, Hirvonen J, Helminen HJ, et al. 20

Comparison of the equilibrium response of articular cartilage in unconfined compression, 21

confined compression and indentation. J.Biomech. 2002;35:903-9. 22

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

7

8. Spilker RL, Suh JK, Mow VC. A finite element analysis of the indentation stress-1

relaxation response of linear biphasic articular cartilage. J.Biomech.Eng. 1992;114:191-2

201. 3

9. Hayes WC, Keer LM, Herrmann G, Mockros LF. A mathematical analysis for 4

indentation tests of articular cartilage. J.Biomech. 1972;5:541-51. 5

10. Brommer H, Brama PA, Laasanen MS, Helminen HJ, van Weeren PR, Jurvelin JS. 6

Functional adaptation of articular cartilage from birth to maturity under the influence of 7

loading: a biomechanical analysis. Equine Vet.J. 2005;37:148-54. 8

11. Ateshian GA, Ellis BJ, Weiss JA. Equivalence between short-time biphasic and 9

incompressible elastic material responses. J.Biomech.Eng. 2007;129:405-12. 10

12. Mow VC, Kuei SC, Lai WM, Armstrong CG. Biphasic creep and stress relaxation of 11

articular cartilage in compression? Theory and experiments. J.Biomech.Eng. 12

1980;102:73-84. 13

13. Kiraly K, Lapvetelainen T, Arokoski J, Torronen K, Modis L, Kiviranta I, et al. 14

Application of selected cationic dyes for the semiquantitative estimation of 15

glycosaminoglycans in histological sections of articular cartilage by 16

microspectrophotometry. Histochem.J. 1996;28:577-90. 17

14. Rieppo J, Hallikainen J, Jurvelin JS, Kiviranta I, Helminen HJ, Hyttinen MM. 18

Practical considerations in the use of polarized light microscopy in the analysis of the 19

collagen network in articular cartilage. Microsc.Res.Tech. 2008;71:279-87. 20

15. Rieppo J, Hyttinen MM, Halmesmaki E, Ruotsalainen H, Vasara A, Kiviranta I, et al. 21

Changes in spatial collagen content and collagen network architecture in porcine articular 22

cartilage during growth and maturation. Osteoarthritis Cartilage 2009;17:448-55. 23

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

8

16. Bi X, Yang X, Bostrom MP, Camacho NP. Fourier transform infrared imaging 1

spectroscopy investigations in the pathogenesis and repair of cartilage. 2

Biochim.Biophys.Acta 2006;1758:934-41. 3

17. Boskey A and Pleshko Camacho N. FT-IR imaging of native and tissue-engineered 4

bone and cartilage. Biomaterials 2007;28:2465-78. 5

18. Saarakkala S, Julkunen P, Kiviranta P, Makitalo J, Jurvelin JS, Korhonen RK. Depth-6

wise progression of osteoarthritis in human articular cartilage: investigation of 7

composition, structure and biomechanics. Osteoarthritis Cartilage 2010;18:73-81. 8

19. Camacho NP, West P, Torzilli PA, Mendelsohn R. FTIR microscopic imaging of 9

collagen and proteoglycan in bovine cartilage. Biopolymers 2001;62:1-8. 10