FINAL Report to the Southern Consortium for Injury Biomechanics

Effects of Impact Loading on Articular Cartilage

Quarter 4

Year 2

 

 

Submitted to:

Dr. Russ Fine

School of Medicine, University of Alabama, Birmingham

UAB Injury Control Research Center

 

and

 

Dr. Rolf Eppinger

National Highway Traffic Safety Administration

 

 

Submitted by Dr. Kyriacos A. Athanasiou

Department of Bioengineering, Rice University

Houston, TX 77005

 

February 7, 2006

 

 

Address:

Kyriacos A. Athanasiou

Department of Bioengineering

Rice University

6100 Main Street, Keck Hall Suite 116

Houston, TX 77005

athanasiou@rice.edu

 

Telephone:      713-348-6385

Fax:                 713-348-5877

                  I.     Summary

 

The global objective of this research proposal was to establish methodologies for protecting diarthrodial joints from injury following traumatic impact, such as often experienced in motor vehicle accidents or sports injuries. The proposed studies determined tolerance criteria of articular cartilage in response to impact forces and also looked at the use of bioactive agents to ameliorate the destructive effects of trauma. This is the eight, and final, quarterly report of the proposal. The period covered is October 1, 2005 through January 31, 2006.

           

The progress for this report includes the following:

 

1)    Summary of completed work

2)    Statement of ongoing work partly supported by this grant

 

                     II.     Summary of completed work

a)    Impact review

 

This manuscript reviews the effects of mechanical impact on articular cartilage ranging from mathematical models to in vivo and in vitro experiments. The manuscript is currently under review at the journal Critical Reviews in Biomedical Engineering.

b)    Explant study

 

This manuscript will soon be submitted to Annals of Biomedical Engineering. One in vitro model for studying the effect of impact on cartilage is the use of explants. To our knowledge, this model had never been temporally characterized in terms of tissue biomechanical properties, tissue matrix biochemistry, and gene expression. Bovine cartilage explants were cultured for 0, 1, and 4 weeks. The glycosaminoglycan content, relative expression of superficial zone protein, and aggregate modulus all decreased significantly from week 0 to week 4 (p < 0.05), while the relative expression of type I collagen increased significantly. These results parallel disuse atrophy studies and indicate that articular cartilage explants under tissue culture conditions exhibit slow changes in extracellular matrix and mechanical integrity, which become significant by 4 weeks. These changes are consistent with an alteration in phenotype from hyaline articular cartilage to fibrocartilage. Figures 1-5 below show the results of this study.

 

 

 

 

 

 

 

Figure 1. Relative gene expression (mean and standard deviation) of: a.) collagen type I

and b.) collagen type II. The symbol * denotes week 4 is significantly different from

weeks 0 and 1 (p < 0.05).

Figure 2. Relative gene expression (mean and standard deviation) of: a.) aggrecan and

b.) superficial zone protein. The symbol * denotes week 4 is significantly different from

week 0 (p < 0.05).

 

 

 

 

Figure 3. Relative gene expression (mean and standard deviation) of: a.) matrix

metalloproteinase 1 and b.) tissue inhibitor of matrix metalloproteinase 1.

 

 

 

 

Figure 4. Mean and standard deviation of the wet weight percentage of the GAG and

collagen content of the explants at 0, 1, and 4 weeks. a.) The GAG content of the explants

significantly decreased from week 0 to week 4 and week 1 to week 4 (p < 0.05). b.) The

collagen content trended down from 0 to 4 weeks, but was not significant. Symbols * and

† denote statistically significance (p < 0.05).

 

Figure 5. Biomechanical properties (mean and standard deviation) of the explants at 0, 1,

and 4 weeks. a.) The aggregate modulus of the explants at 0, 1, and 4 weeks. The

symbol * denotes statistical significance between week 0 and week 4 (p < 0.05). b.)

Poisson’s ratio of the explants at 0, 1, and 4 weeks.

c)     Cartilage digest study

 

This manuscript was recently accepted for publication in the journal Tissue Engineering. The study looked at the changes in gene expression caused by several standard enzymatic digestion methods for acquiring cells from cartilage.

d)    Impact instrument design and validation and the acute effects of impact on articular cartilage

 

This manuscript will soon be submitted to the journal Osteoarthritis and Cartilage. This study involved developing an instrument to mechanically impact articular cartilage, recording the impact event with a piezoelectric accelerometer, and investigating the changes occurring in articular cartilage in response to impact. After instrument design and fabrication, viscoelastic rubber was used to validate the machine’s ability to deliver impacts of variable peak stresses, time to peak stress, and total duration. We then employed two levels of impact, a low level (1.1 J) that caused no gross surface damage and a high level (2.8 J), causing gross surface damage, to study their effects on cartilage immediately and at 24 hrs. Gross morphology of high impacted specimens was scored as significantly worse than no impact controls or low impact specimens both immediately and at 24 hrs. It was also found that the stiffness of the high impact specimens was significantly less than low and control specimens at these time points. After 24 hrs of culture, specimens were further assayed for cell death, glycosaminoglycan (GAG) release into the media, and changes in gene expression. We found significantly more dead cells in low impact specimens compared to control and significantly more dead cells in high impact specimens compared to low. GAG release into the media showed that the low and high specimens released significantly more than control, but there was not a difference between low and high. Finally, gene expression analysis showed collagen type 1 upregulation in low specimens compared to control and high, and matrix metalloproteinase 1 upregulation in high compared to control and low. The following 8 figures show some results from this study.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Figure 1. The impact instrument with an adjustable specimen clamp and a proximal bovine ulna. The linear bearing (A) has an attached sliding plate (B) that can have variable amounts of mass attached. The base (E) consists of two pieces of plate 304 stainless steel. The specimen clamp (D) holds a bovine ulna, while the impact tip and tip interface (C) are perpendicular to the articular surface.

 

 

Figure 2. The specimen clamp with a harvested, prepared bovine ulna. The specimen clamp has four adjustable leveling feet (A), one at each corner. A strip of stainless steel (B) slides onto two large bolts, one at each end. Three set screws (C) in the middle of the strip of stainless steel allow for variable geometry.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Figure 3. One of the impact tips a) assembled and b) disassembled to be sterilized. The impact tip (C) is stainless steel rod that has been lathed to 5 mm (right) or 3 mm (left) in diameter. The impact interface consists of a small plate (A) that has a partial thickness, tapped-hole that a piece of threaded stainless steel pipe (B) screws into. The impact tip fits snugly into pipe.

 

Figure 4. Two views of a proximal bovine ulnar articular surface, a) superior aspect, and b) anterior aspect. The articular cartilage surface is flat, similar to that of the proximal human tibia.

 

 

Figure 5. Impact measurements for each of six levels of impact on viscoelastic rubber and two levels of impact on articular cartilage. In each column, every value not connected by the same superscripted letter is significantly different (p < 0.05).

 

Tup Weight	Drop height	Energy	Peak stress	Time to peak stress	Duration
			(Mean ± S.D.)	(Mean ± S.D.)	(Mean ± S.D.)
N	cm	J	MPa	msec	msec
18.4	3	0.55	1.46 ± 0.03A	1.53 ± 0.35A	5.11 ± 0.28AB
	6	1.10	1.67 ± 0.04B	0.83 ± 0.13B	4.99 ± 0.17B
	11	2.02	2.03 ± 0.06C	0.65 ± 0.07BC	5.41 ± 0.12A
27.8	4	1.03	4.28 ± 0.28D	0.60 ± 0.08BC	3.10 ± 0.40C
	8	2.06	5.56 ± 0.19E	0.50 ± 0.07C	2.00 ± 0.20D

 

Tup Weight	Drop height	Energy	Peak stress	Time to peak stress	Duration
			(Mean ± S.D.)	(Mean ± S.D.)	(Mean ± S.D.)
N	cm	J	MPa	msec	msec
18.4	6	1.10	3.82 ± 0.30B	0.69 ± 0.31B	1.52 ± 0.60B
27.8	10	2.78	7.23 ± 1.54A	0.53 ± 0.15A	1.01 ± 0.26A

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Figure 6. Histology and morphology of articular cartilage explants for a) control, b) low impact level, and c) high impact level. The morphology pictures were taken after application with a pipette of 2 ml of India ink that was diluted 3:1 in PBS. The white circle indicates the 5 mm diameter area of impact.

 

 

Figure 7. Morphological rating of the articular cartilage. Scoring was based on 4 categories: 1) Tissue morphology, 2) India ink staining, 3) Surface regularity, and 4) Surface indentation. Surface indentation was assessed at the time of impact only. Other scales were graded by two independent blinded researchers. * - denotes statistical significance (p < 0.05).

Figure 8. Aggregate modulus of the mechanically impacted articular cartilage at high (27.8 N tup, 8 cm height, 2.06 J) and low (27.8 N tup, 4 cm height, 1.03 J) levels. * - denotes statistical significance (p < 0.05).

 

e)     Temporal effects of impact on articular cartilage

 

The temporal effects study has been submitted as a manuscript for publication in the journal Arthritis and Rheumatism. The paper describes how, at 1 week, articular cartilage impacted at a low level is not significantly different from non-impacted specimens in terms of cell death and GAG release into the media, but some gene expression changes loss of mechanical integrity has begun to occur. In contrast, at 1 week, high impacted specimens are significantly different from specimens at the no and low impact levels in terms of cell death, GAG release into the media, and mechanical integrity. However, by 4 weeks, the low impact specimens have departed from their similarities to non-impacted specimens to resemble specimens that experienced the high level of impact. Both the low and high impact specimens have increased cell death, with the increase being greater for low compared to high. They have also released similar amounts of GAG to the media and have statistically equivalent mechanical properties. Gene expression changes consistent with cartilage degradation were also observed. Figures 1-9 below describe the results of this study. These findings are significant when considering that post-traumatic osteoarthritis tends to develop several years after an injury, but significant changes in the tissue occur as soon as 1 week after injury.

 

 

 

 

Fig. 1. Viability staining for each experimental group. a) Baseline, b) week 1 culture control, c) week 1 Low impacted specimen, d) week 1 High impacted specimen, e) week 4 culture control, f) week 4 Low impacted specimen, and g) week 4 High impacted specimen. Red indicates dead cells and green indicates living cells. Magnification is 100X for each image. Baseline explants were neither cultured nor impacted. Culture controls were not impacted.

 

Fig. 2. Results from the viability staining expressed as mean ± S.D. percent dead cells. Groups not connected by the same letter are significantly different from one another (p < 0.05).

 

Fig. 3. Gene expression levels for a) collagen I (Col1) and b) aggrecan (AGC). Values are normalized abundance expressed as mean ± S.D. Within a gene, groups not connected by the same letter are significantly different from one another (p < 0.05). (CC = Culture control, LI = Low impact inside the impact zone, LO = Low impact outside the impact zone, HI = High impact inside the impact zone, HO = High impact outside the impact zone)

 

Fig. 4. Gene expression levels for week 1 a) superficial zone protein (SZP) and b) tissue inhibitor of matrix metalloproteinase-1 (TIMP-1). Values are normalized abundance expressed as mean ± S.D. Within a gene, groups not connected by the same letter are significantly different from one another (p < 0.05). (CC = Culture control, LI = Low impact inside the impact zone, LO = Low impact outside the impact zone, HI = High impact inside the impact zone, HO = High impact outside the impact zone)

 

Fig. 5. The ratio of MMP-1 to TIMP-1 gene expression expressed as mean ± S.D. Groups not connected by the same letter are significantly different from one another (p < 0.05). (CC = Culture control, LI = Low impact inside the impact zone, LO = Low impact outside the impact zone, HI = High impact inside the impact zone, HO = High impact outside the impact zone)

 

 

 

 

 

 

 

 

Fig. 6. Data for GAG released into the media displayed as mean ± S.D. a) Comparison of culture control, Low, and High impact levels at week 1, b) Comparison of GAG release per week for the Low impact level, c) Comparison of GAG release per week for the High impact level. Groups not connected by the same letter are significantly different from one another (p < 0.05).

Fig. 7. Wet weight % of GAG in the explant for the a) Low and b) High impact groups. Numbers are mean ± S.D. For both levels, the week 4 inside and outside areas are significantly different from baseline (p < 0.05). (CC = Culture control, LI = Low impact inside the impact zone, LO = Low impact outside the impact zone, HI = High impact inside the impact zone, HO = High impact outside the impact zone)

Fig. 9. DNA content presented as mean ± S.D. number of cells in millions per gram of tissue. Groups not connected by the same letter are significantly different from one another (p < 0.05). (CC = Culture control, LI = Low impact inside the impact zone, LO = Low impact outside the impact zone, HI = High impact inside the impact zone, HO = High impact outside the impact zone)

 

 

 

 

 

 

 

 

Fig. 10. Comparison of mean ± S.D. for a) aggregate modulus and b) permeability at week 1 and week 4 (CC = Culture control, L = Low impact level, H = High impact level). Groups not connected by the same letter are significantly different from each other (p < 0.05).

f)   Summary of completed work

           

The global hypothesis of this work was that low levels of impact to articular cartilage can result in degenerative changes that manifest temporally such that the ensuing damage to the tissue is the same as would occur if the original mechanical insult had been a high level of impact. To study this hypothesis, an appropriate model was chosen after careful review of the literature, an instrument capable of delivering the required impacts was designed and validated, and the effects of impact at two different energy levels was studied at immediate, 24 hour, 1 week, and 4 week time points. Results obtained, i.e., gross morphology and histology, cell viability, tissue matrix and culture media biochemistry, gene expression analyses, and biomechanical properties, generally support the hypothesis allowing us to conclude that clinically “silent” injuries, or those that would not result in identifiable changes in tissue gross morphology or mechanical properties, are equivalent to injuries that manifest clinically from the outset.

g)    Submitted manuscripts

 

The following manuscripts supported by this grant have been, or will be submitted:

 

•       D.M. Hayman, T. B. Blumberg, C.C. Scott, and K.A. Athanasiou, “Chondrocyte Isolation Methods Alter Gene Expression.” Tissue Engineering (Accepted for publication)

•       C.C. Scott and K.A. Athanasiou, “Mechanical Impact and Cartilage.” Critical Reviews in Biomedical Engineering (Under review)

•       C.C. Scott, R.M. Natoli, and K.A. Athanasiou, “Temporal Effects of Mechanical Impact on Articular Cartilage.” Arthritis and Rheumatism (Submitted)

•       C.C. Scott and K.A. Athanasiou, “Acute Effects of Impact on Articular Cartilage.” (To be submitted)

•       C.C. Scott, D.M. Hayman, and K.A. Athanasiou, “Gene Expression, Biochemical and Biomechanical Changes in Articular Cartilage Explant Culture.” Annals of Biomedical Engineering (To be submitted)

 

                       III.     Statement of ongoing work partly supported by this grant

a)    Ameliorating the degenerative effects of impact on articular cartilage

 

Now that we are beginning to understand the effects of impact on articular cartilage, we can move towards the ultimate goal of this work: identifying suitable bioactive agents capable of ameliorating, i.e., preventing or reversing, the degenerative changes identified in the ‘completed work’ section of this report. Two studies looking at the effects of bioactive agents applied to articular cartilage post-impact are underway. In the first study, insulin-like growth factor-1 (IGF-1) and the non-ionic surfactant poloxamer 188 (P188) are being studied. The experimental design is a full-factorial study with 3 factors: impact level, treatment, and delivery method (see Fig. 1).

 

Figure 1: 1^st Study Design

 

The levels of impact are 0, low, and high; treatment levels are none, IGF-1, P188, and their combination; and delivery method is either none, or aided by a compression regiment. Each experimental group will be assayed in terms of gross morphology, cell viability, GAG release into the culture media, and biomechanical properties at 24 hours and 1 week post impact. We expect that the degenerative changes seen will be partially ameliorated. Further, the two bioactive agents are potentially targeting different degenerative pathways. The use of these two agents will allow their direct comparison, and will provide insight into which of the degenerative pathways may play a more significant role.

In another study (see Fig. 2), we are investigating the role of doxycycline in preventing matrix degradation. The same impact levels are being employed. Treatment groups consist of none, 50μM, or 100μM doxycycline. Doxycycline has been shown to inhibit matrixmetalloproteinases, and, thus, we hypothesize that treatment of cartilage after impact will inhibit matrix breakdown and lead to tissue with improved mechanical properties when compared to untreated samples.

Both of these studies are over half complete, and preliminary analysis of the data shows that at least these agents can reverse some of the negative findings in cartilage post impact.

We would like to thank you for the support provided to do the work entailed in this final report.

Figure 2: 2^nd Study Design