Cytokines and apoptosis in supraspinatus tendinopathy

Cytokines and apoptosis in supraspinatus
tendinopathy
N. L. Millar,
A. Q. Wei,
T. J. Molloy,
F. Bonar,
G. A. C. Murrell
From the University
of New South Wales,
Sydney
„ N. L. Millar, MRCS, Clinical
Research Fellow, Specialist
Registrar in Orthopaedics
Division of Immunology,
Infection and Inflammation
Glasgow Biomedical Research
Centre, University of Glasgow,
120 University Avenue,
Glasgow G12 8TA, UK.
„ A. Q. Wei, MB BS, Research
Fellow
„ T. J. Molloy, PhD, Research
Fellow
„ G. A. C. Murrell, DPhil(Oxon),
MD, Professor of Orthopaedic
Surgery
Orthopaedic Research Institute,
Department of Orthopaedic
Surgery
St George Hospital Campus,
University of New South Wales,
4-10 South Road, Kogarah, New
South Wales 2217, Australia.
„ F. Bonar, MRCPI,
FRCPath(London), Consultant
Pathologist
Douglas Hanly Moir, 95 Epping
Road, North Ryde 2113, Sydney,
Australia.
Correspondence should be sent
to Professor G. A. C. Murrell;
e-mail: murrell.g@ori.org.au
©2009 British Editorial Society
of Bone and Joint Surgery
doi:10.1302/0301-620X.91B3.
21652 $2.00
J Bone Joint Surg [Br]
2009;91-B:417-24.
Received 8 August 2008;
Accepted after revision 11
December 2008
VOL. 91-B, No. 3, MARCH 2009
The role of inflammatory cells and their products in tendinopathy is not completely
understood. Pro-inflammatory cytokines are upregulated after oxidative and other forms of
stress. Based on observations that increased cytokine expression has been demonstrated in
cyclically-loaded tendon cells we hypothesised that because of their role in oxidative stress
and apoptosis, pro-inflammatory cytokines may be present in rodent and human models of
tendinopathy. A rat supraspinatus tendinopathy model produced by running overuse was
investigated at the genetic level by custom micro-arrays. Additionally, samples of torn
supraspinatus tendon and matched intact subscapularis tendon were collected from
patients undergoing arthroscopic shoulder surgery for rotator-cuff tears and control samples
of subscapularis tendon from ten patients with normal rotator cuffs undergoing
arthroscopic stabilisation of the shoulder were also obtained. These were all evaluated
using semiquantitative reverse transcription polymerase chain-reaction and
immunohistochemistry.
We identified significant upregulation of pro-inflammatory cytokines and apoptotic genes
in the rodent model (p = 0.005). We further confirmed significantly increased levels of
cytokine and apoptotic genes in human supraspinatus and subscapularis tendon harvested
from patients with rotator cuff tears (p = 0.0008).
These findings suggest that pro-inflammatory cytokines may play a role in tendinopathy
and may provide a target for preventing tendinopathies.
Tears of the rotator cuff are a cause of considerable pain and dysfunction in the shoulder in adults.1 The results of surgical repair
of full-thickness tears have been favourable,2,3 but post-operative imaging has
shown high rates of re-tearing indicating
failure of the biological healing of the
tendon-bone interface.4,5 The subacromial
bursa and the torn margin of the rotator cuff
have been suggested as potential sources of
healing with studies implicating the role of
cytokines,6 matrix metalloproteineases,7
oxygen-free radicals6 and growth factors8 in
the pathophysiology of rotator-cuff tears.
Cytokines are important molecular messengers in the response of soft tissue to injury and
wound healing. They have been shown to be
involved in the regulation of matrix turnover
in tendinopathy9 and may have a direct effect
on the activity of tenocytes10 and the subsequent expression of tendon matrix genes.11
Cytokines also play a key role in oxidative
stress-induced cellular apoptosis12 which is
mediated by the activation of caspases
(cysteine-containing aspirate proteases, a
group of proteolytic enzymes) and is involved in
the stress-induced cascade of tendinopathy.13
Based on our previous observations of
increased apoptosis in tendinopathy we hypothesised that because of their central role in caspase-dependent apoptosis, pro-inflammatory
cytokines may be present in human and rodent
models of tendinopathy.13
Materials and Methods
We randomised equally 24 male, 30-week-old
Sprague-Dawley rats weighing a mean of 362 g
(342 to 388) into an exercise and control
group. The exercise group underwent a daily
treadmill running regimen to model tendon
overuse resulting in degeneration as previously
described.14 The rats initially underwent daily
training which increased in duration over two
weeks to accustom them to the exercise and
surroundings. After this, they were subjected
to exercise which consisted of running on a 10°
decline at 17 m/min for one hour per day, five
days per week. This regimen equates approximately to 7500 strides/day. After running for
four weeks the rats were killed by CO2
417
418
N. L. MILLAR, A. Q. WEI, T. J. MOLLOY, F. BONAR, G. A. C. MURRELL
Table I. Details of the 17 patients
Case
Age
(yrs)
Duration of symptoms History of traumatic
(mths)
rotator cuff injury
Pain intensity
score*
Pain frequency
score†
Supraspinatus
strength (N)
Impingement
sign‡
Size of rotator cuff
tear (cm2)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
58
46
50
64
55
52
70
75
62
51
72
60
39
40
51
64
62
5
3
5
8
7
2
6
3
4
3
6
4
3
5
4
7
4
8
9
7
11
7
7
11
8
6
12
6
8
11
6
6
8
5
9
8
6
10
9
8
10
7
6
12
8
7
10
9
10
10
8
15
17
21
48
38
20
26
29
38
13
30
10
4
17
16
27
0
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
2
2
2
4.5
3
3
6
1
1
2
3
5
1
3
4
6
6
Y
Y
N
Y
Y
Y
Y
Y
Y
Y
N
Y
Y
Y
Y
N
Y
* the pain intensity occurring at night, at rest, and with activities was scored as follows: 0, no pain; 1, mild pain; 2, moderate pain; 3, severe pain and
4, very severe pain, and the three scores combined (maximum 12)
† the frequency of severe pain, pain at night, and pain with activities was scored as 0, never; 1, monthly; 2, weekly; 3, daily; and 4, constantly and
the three scores combined (maximum 12)
‡ Hawkins impingement sign
inhalation and both supraspinatus tendons were collected.
The control group consisted of 12 rats which did not have
exercise. All the procedures and protocols were approved
by the Animal Care and Ethics Committee of the University
of New South Wales under ACEC No. 99/101.
In addition we obtained 17 specimens of supraspinatus
tendon from patients with tears of the rotator cuff undergoing surgery (Table I). Their mean age was 57 years (39 to 75)
and the mean size of tear was 3.2 cm2 (1 to 45). The operation was performed using a standard arthroscopic technique. The size of the tear was recorded at the time of the
operation. Clinical examination included Hawkins
impingement sign5 and strength testing of the supraspinatus using a handheld HFG-45 force gauge (Transducer
Techniques, Temecula, California) as previously described
by Hayes et al.16 Samples of the subscapularis tendon were
also collected from the same patients. They were only
included if there was no evidence of subscapularis tendinopathy on a pre-operative MR scan or of macroscopic
damage to the tendon at the time of arthroscopy.
As an independent control group ten samples of subscapularis tendon were collected from patients undergoing
arthroscopic surgery for shoulder stabilisation who did not
have tears of the rotator-cuff confirmed by arthroscopic
examination. The mean age of the control group was
35 years (20 to 41).
The subscapularis tendons were harvested arthroscopically from the superior border of the tendon 1 cm lateral to
the glenoid labrum and the supraspinatus tendons from
within 1.5 cm of the edge of the tear before surgical repair.
All the specimens were immediately placed in RNAlater
solution (Ambion Inc, Austin, Texas) and stored at -20°C
until RNA extraction. For immunohistochemical staining
the tissue samples were immediately fixed in 10% (v/v) formalin for four to six hours and then embedded in paraffin.
Custom micro-arrays consisting of 5760 rat oligonucleotide features in duplicate were provided by the Ramaciotti
Centre of the University of New South Wales, Sydney, Australia. A total of 15 μg to 20 μg of total RNA was used to
synthesise complimentary DNA (cDNA) in which Cy3 or
Cy5 fluorescent dyes (Amersham Pharmacia Biotech, North
Ryde, Australia) were incorporated using an indirect labelling method. Micro-arrays were scanned using an Axon
GenePix micro-array scanner (Molecular Devices, Union
City, California) and the resulting image analysed using
GenePix prosoftware version 3.0 (Molecular Devices). All
the genes were represented by at least two independent targets on each micro-array so that at least six independent targets were identified across three micro-arrays. The signal
from each target was used to calculate a mean signal for each
gene. The latter was considered to be up- or down-regulated
if it had at least a 1.5-fold difference from the control group.
Total RNA was isolated from tendon tissue using Trizol
reagent (Life Technologies, Grand Island, New York)
according to the manufacturer’s instructions. The RNA yield
and integrity were evaluated by spectrophotometry and agarose gel electrophoresis, respectively. The optical density
260/280 ratio was maintained between 1.7 and 2.1, confirming that the quality of the RNA was satisfactory.
RNA extracted from tendon tissue was reverse transcripted to generate the first strand of cDNA and subsequently amplified using a polymerase chain-reaction
(PCR) to produce amplified targeted segments of cDNA
using the ImProm-II Reverse Transcription System
THE JOURNAL OF BONE AND JOINT SURGERY
CYTOKINES AND APOPTOSIS IN SUPRASPINATUS TENDINOPATHY
419
Table II. Details of the human specific primer sequences and annealing temperatures for polymerase chain reactions
Product size (bp)*
Primer sequence
Annealing temperature (°C)
Caspase 3
398
Forward TCG GTC TGG TAC AGA TGT CG
Reverse CAT ACA AGA AGT CGG CCT CC
58
Caspase 8
375
Forward ACA GTG AAG TCT GGC CTC C
Reverse GCA GGT TCA TGT CAT CAT CC
55
IL-18
348
Forward CCC AAC GAT AAA GAA GAA CGC
Reverse TGT CTG TGC CTC CCG TGC TGG C
50
IL-15
275
Forward GTC TTC ATT TTG GGC TGT TTC AGT
Reverse CCT CAC ATT CTT TGC ATC CAG BATT CT
57
IL-6
341
Forward TAG CCG CCC CAC ACA GAC AG
Reverse CGC TGG CAT TTG TGG TTG GG
58
MIF†
276
Forward CTC TCC GAG CTC ACC CAG CAG
Reverse CGC GTT CAT GTC GTA ATA GTT
54
TNF-α
376
Forward AAG AGT TCC CCA GGG ACC TCT
Reverse CCT GGG AGT AGA TGA GGT ACA
52
* bp, base pairs
† MIF, macrophage inhibitory factor
(Promega, Madison, Wisconsin). All the reverse transcription
steps were performed in a GeneAmp 2400 thermal cycler
(Applied Biosystems, Foster City, California). We mixed 1 μg
to 5 μg of total RNA with 0.5 μg of oligo (dT) nucleotides
primer to make a 5 μl solution and denatured it at 70°C for
five minutes. The reaction mixture was then cooled to 4°C for
at least five minutes. Subsequently, a 15 μl reverse transcription reaction mixture consisting of 1 μl of Improm-II Reverse
Transcriptase, 0.5 mM d nucleoside triphosphates dNTPs,
6 mM MgC12, 4 μl of Improm-II 5× reaction buffer, 20 U of
Recombinant RNasin ribonuclease inhibitor and nucleasefree water, was added to the target RNA and primer combination. The reaction mix was heated to 70°C for five minutes
to allow annealing of the oligo dT primers to the RNA. It was
then incubated at 37°C for one hour in order to create the
cDNA. For each tested RNA sample, a control reaction was
run with Moloney murine leukemia virus (MMLV) reverse
transcriptase omitted from the mixture to exclude falsepositive results from genomic DNA contamination.
We amplified 1 μl (100 ng) of each cDNA sample from
the above extractions by PCR (GeneAmp System 2400) in a
total volume of 20 μl. Platinum Super Mix (22 U/ml of
complexed recombinant Taq Polymerase with platinum
Taq antibody, 22 mH Tris-HC1 (pH 8.4), 55 mM KCl, 1.65
mM MgC12, 220 μM dGTP, 220 μM dATP, 220 μM dTTP,
200 μM cCTP, stabilisers; Invitrogen Life Technologies,
Carlsbad, California) was used together with 0.2 μM each
of the specific oligonucleotide primers using published
sequences (Table II). The thermal cycling programme consisted of an initial denaturation step of 95°C for 1 min 15 s,
followed by 35 cycles of 95°C for 30 s, a 30-second annealing step at varying temperatures and a 30-second extension
VOL. 91-B, No. 3, MARCH 2009
step at 73°C. Reaction products (12 μl) were analysed by
electrophoresis through 3% agarose gels stained with
ethidium bromide and visualised by transillumination
under ultraviolet light at a wavelength of 302 nM. Amplification of the mRNA of β-actin was used as an internal
control in all cases.
Paraffin sections were de-waxed with xylene and graded
ethanol. Sections were stained with haematoxylin and eosin
and Toluidine Blue. The presence or absence of oedema and
degeneration was recorded, together with the degree of
fibroblast cellularity, amyloid deposition and chondroid
metaplasia. Tendinopathy was assessed on a basic histological scale derived from Khan et al17 (grade 4, marked
degeneration; 3, advanced degeneration; 2, mild/moderate
degeneration; 1, normal tendon) by two independent assessors (AQW, FB) who were blinded to the results.
For immunohistochemical techniques antigen retrieval
was achieved using Dako target retrieval solution (Dako,
Carpinteria, California) in accordance with the manufacturer’s instructions. Endogenous peroxidase activity was
scavenged with 3% (v/v) H2O2 and non-specific antibody
binding blocked with 10% (w/v) milk in Tris-buffered
saline (TBS) buffer for 15 minutes. Tissue slides were incubated with primary antibody (Interleukin (IL)-12, IL-15,
IL-18, macrophage inhibitory factor (MIF) and tumour
necrosis factor alpha (TNF-α)) diluted 1:100 in 1% (w/v)
bovine serum albrium (BSA)/TBS at room temperature for
60 minutes. After two washes, the slides were incubated
with Dako LINK consisting of biotinylated anti-mouse
and anti-rabbit immunoglobin (Ig). They were washed
and incubated with streptavidin-peroxidase, followed by
extensive washing and exposure to Dako liquid
N. L. MILLAR, A. Q. WEI, T. J. MOLLOY, F. BONAR, G. A. C. MURRELL
diaminsbenzidine (DAB) for five minutes. Finally, the
sections were counterstained with haematoxylin. Positive
and negative control specimens were included, in addition
to the surgical specimens for each individual antibody
staining technique.
Statistical analysis. The results were reported as the mean
and SEM. Comparisons between groups were made using a
two-way paired Student’s t-test, the Mann-Whitney U test
and Kruskal-Wallis one-way analysis of variance on ranks
using Sigma Stat version 3.1 (Systat Software Inc., Richmond, California). Corrections between groups used the
Pearson’s correlation coefficient. A power analysis was performed with the beta error set at 0.2 (power = 0.8). Based
on the results of this, it was determined that each group
required ten tissue samples to detect a difference of 20%
between each of the groups with regard to the gene expression. Statistical significance was set at a p-value ≤ 0.05.
Torn
supraspinatus
Control
Relative expression to β-actin
420
Matched
subscapularis
1.6
1.4
1.2
1
0.8
0.6
0.4
0.2
0
MIF
IL-18
IL-15
IL-6
TNF Caspase 3 Caspase 8
Fig. 1
Bar chart showing the relative expression of cytokine and apoptosis
genes in the samples of human tendon. The data are displayed as the
mean with the error bars representing the SEM (*** p < 0.001).
Results
Rat micro-array. Of the 5760 independent targets on each of
the micro-arrays, most had less than a twofold change in
expression in the experimental samples compared with the
control group and thus were considered not to be differentially expressed. Overall, 91 genes were found to be significantly upregulated and 37 significantly downregulated as
previously reported by Molloy et al18 from our laboratory.
Inflammation-related genes were also over-represented in
degenerated tendon, with 14 genes (15.4%) significantly
upregulated and five (13.5%) significantly downregulated.
Upregulation (p < 0.01, Student’s t-test) of IL-11 (× 3.6), IL15 (× 2.5) and IL-18 ( 1.8), anti-NGF30 ( 2.6), MIF (× 1.8),
globin transcription factor (GATA) binding protein 3 (× 3.3),
platelet-activating factor acetylhydrolase (× 3.1), CD3
gamma chain (× 3.1), IgG-2b gene (× 3.0), T-cell receptor
variable β chain (× 4.6), IL-6 (× 2.3), attractin (× 2.4), TNF-α
receptor (× 2.5) and receptor activator of NF-kb ligand
(× 3.0) and downregulation (p < 0.05, Student’s t-test) of Ig
heavy chain (× 3.0), CD94 (× 2.8), T cell receptor α chain
(× 6.5), IL-2 (× 3.3) and T-cell receptor (× 2.2) occurred in
degenerative rat supraspinatus tendon subjected to daily
treadmill running for four weeks. The differential expression
of apoptotic-related genes was represented by five upregulated genes (5.5%) and three downregulated genes (8.1%).
Upregulation of heat shock protein 27 (× 3.4 and 70 (× 2.5),
cellular FLICE inability protein (cFLIP) receptor (× 2.2) and
caspase 8 (× 3.1) while downregulation of poly(ADP-robose)
polymerase (× 4.8), type-2 angiotensin II receptor (× 2.6) and
hypoxia-inducible factor 1 (× 2.2) were noted.
Also differentially expressed were cellular communication
genes, including members of the glutamate signalling
machinery, represented by 12 significantly upregulated genes
(13.2%) and three significantly downregulated genes
(8.1%). Several growth, differentiation, and developmental
genes were also differentially regulated in the degenerated
tendons, represented by 13 upregulated genes (14.3%) and
ten downregulated genes (11%).
Human tendon samples. The cytokine genes IL-18, IL-15,
IL-6, MIF and TNF-α were detected in all samples of torn
supraspinatus tendon. The expression levels of IL-18, IL-15,
IL-6 and MIF was higher (p < 0.001, Kruskal-Wallis
ANOVA) in the torn edges of supraspinatus when compared
with matched subscapularis tendon (Fig. 1). The expression
of the apoptotic genes caspases 3 and 8 were also significantly increased in the same tisssue samples (p = 0.0005).
TNF-α mRNA expression was elevated (p < 0.05, Student’s ttest) in subscapularis tendon compared with torn supraspinatus samples. The levels of expression of IL-18, IL-15, IL6, MIF, TNF-α and caspases 3 and 8 were elevated (p < 0.001,
Kruskal-Wallis ANOVA) in torn supraspinatus and matched
subscapularis when compared with normal control subscapularis. We observed no correlation between tear size (r = 0.1,
p = 0.6), patient age (r = 0.2, p = 0.8), pain (r = 0.1, p = 0.9)
and cytokine (r = 0.4, p = 0.8) and apoptotic (r = 0.3, p = 0.5)
gene expression. These apoptotic and cytokine genes were
confirmed at the protein level at which IL-18, IL-15, IL-6,
MIF, TNF-α, and caspases 3 and 8 were confirmed in all
samples of torn supraspinatus tendon (Fig. 2). Increased
(p < 0.001, Kruskal-Wallis ANOVA) immunoactivity of
IL-18, IL-15, IL-6, MIF and caspases 3 and 8 were found in
torn supraspinatus compared with matched and normal subscapularis. The proteins were localised to tendon cells. Negative controls using human epidermis showed no expression
of any apoptotic genes. Positive controls of human tonsil tissue confirmed the presence of IL-18, IL-6, MIF, TNF-α and
caspases 3 and 8. IL-15 was identified in human liver tissue
as a positive control.
Of additional interest haematoxylin and eosin staining of
matched subscapularis samples showed overall appearances
of advanced degenerative change consistent with tendinopathy with seven samples showing grade-3 and nine grade-2
changes despite no evidence of tendinopathy on pre-operative MRI and macroscopically at the time of arthroscopy
(Fig. 3). All torn supraspinatus tendons showed features in
THE JOURNAL OF BONE AND JOINT SURGERY
CYTOKINES AND APOPTOSIS IN SUPRASPINATUS TENDINOPATHY
Fig. 2a
Fig. 2b
Fig. 2e
Fig. 2d
Fig. 2g
Fig. 2h
421
Fig. 2c
Fig. 2f
Fig. 2i
Photomicrographs showing macrophage inhibitory factor (MIF) in a) normal unmatched supraspinatus tendon, b) matched subscapularis tendon and c) torn human supraspinatus tendon, IL-18 in d) normal unmatched supraspinatus tendon, (e) matched subscapularis tendon and f)
torn human supraspinatus tendon and caspase 3 in g) normal unmatched supraspinatus tendon, h) matched subscapularis tendon and i) torn
human supraspinatus tendon (× 200).
keeping with those of advanced mucoid change and chondroid metaplasia as expected in association with marked
degenerative change consistent with tendinopathy (grade 4).
In a previous investigation13 matched subscapularis tendons
from patients with full-thickness tears of the rotator cuff had
approximately 35% of apoptotic cells compared with 20%
in the subscapularis tendon from patients without a tear confirming our histological finding of a degenerative process in
supposedly normal samples of subscapularis tendon. We
observed no correlation between tear size (r = 0.1, p = 0.6),
patient age (r = 0.2, p = 0.5), pain (r = 0.1, p = 0.6) and the
histological grade of tendinopathy (r = 0.8, p = 0.2) Pearson’s
correlation coefficient.
VOL. 91-B, No. 3, MARCH 2009
Discussion
We wished to investigate further the role played by cytokines
in the stress-induced apoptosis pathway of tendinopathy. Our
data were the first to confirm the presence of the cytokines
IL-18, IL-15 and IL-6 and MIF in both rat and human models
of tendinopathy. We also identified significantly increased
levels of key mediators, caspases 3 and 8 and of Fas-Ligand
apoptosis in the margins of the rotator cuff of patients with
full-thickness tears compared with matched subscapularis tendon and normal independent control tendon.
Cytokines have been shown to be important contributors
to the pathology of tendinopathies.19,20 In animal studies
local administration of cytokines induced a histological
Number of samples
422
N. L. MILLAR, A. Q. WEI, T. J. MOLLOY, F. BONAR, G. A. C. MURRELL
18
16
14
12
10
8
6
4
2
0
Grade 4
Grade 3
Grade 2
Grade 1
Control
Torn
Matched
supraspinatus subscapularis
Fig. 3
Bar chart showing histological grades of tendinopathy in torn supraspinatus (n = 17), matched subscapularis (n = 17) and control subscapularis (n = 10).
picture of tendinopathy21 while application of cyclical strain
increased the presence of IL-6 and 1β gene expression in
human flexor tenocytes.22 Lin et al23 have recently shown
that IL-6 knockout mice show inferior mechanical and organisational tendon properties compared with a control group,
while Nakama et al24 highlighted the role of IL-6 in stimulating transcription factors in human rotator-cuff tendon.
Others investigating the therapeutic benefit of blocking
cytokines have shown that in doing so collagenase
concentrations25 and angiogenesis26 can be reduced in rheumatoid tenosynovium which ultimately reduces tendon damage.
Cytokines also play a key role in oxidative-stressinduced cellular apoptosis.27-29 Several groups have highlighted the role of apoptosis in degeneration of tendon
cells with excessive apoptosis noted in the torn edges of
rotator-cuff tendons,13 patellar tendinosis30 and in a
mechanically stained tibialis anterior rodent model.31
Apoptotic cell death is orchestrated by the activation of
caspases, a family of cysteine proteases for aspartic-acid
residues, which cleave specific intracellular substrates to
produce the characteristic features of this form of cell
death.32 Previous work has shown that oxidative-stressinduced apoptosis in human tendon fibroblasts is mediated by the release of the cytochrome C pathway and activation of caspase 3.33 Recently, we discovered increased
levels of heat shock proteins in human rotator cuff. These
are known to protect cells from the cytotoxic effects of
cytokines and apoptotic mediators. 34 Another group has
discovered a genetic component to rotator-cuff tears35
which may also be linked to the apoptotic process.
In both our models of rat and human tendinopathy
increased levels of cytokines involved in oxidative-stressinduced apoptosis were identified as was the expression
of key apoptotic regulatory genes. MIF has been shown to
have pro-inflammatory effects in a wide range of diseases
including atherosclerosis and rheumatoid arthritis.36
MIF induces the gene expression of other inflammatory
cytokines (TNF-α, IL-6), while overexpression suppresses
pro-oxidative stress-induced apoptosis.12 IL-18 increases
the synthesis of nitric oxide37 and induces apoptosis in
human articular chondrocytes38 and bone-marrow
cells.39 IL-15 is a potent inhibitor of several apoptotic
pathways40,41 including ligand-induced association of
TNR receptor 1.42 TNF-α is a classic initiator of Fas
ligand and caspase-8-derived apoptosis and plays a role
in chondrocyte apoptosis43 while IL-6 has been shown to
have several anti-apoptotic effects.28,44
Based on our observations the overexpression of these
cytokines in these models of tendinopathy alongside
increased apoptotic genes suggests that these molecules
play key effector roles in the Fas-Ligand pathways in an
attempt to regulate the degree of apoptosis. They may act
to up- or down-regulate apoptosis and as such may be
either protective or harmful to tendon cells.
Oxygen-free radicals and stress-activated protein
kinases have also been implicated in the stress-induced
apoptotic cell pathway.45,46 IL-18, IL-15 and IL-6 are
known to promote the intensive production of reactive
O2 metabolites and are potent agonists of protein
kinases.47-49 Our finding of significantly increased
cytokine levels may suggest that these molecules when
expressed during the degenerate and healing phases of
tendon injury, result in the subsequent production of
reaction O2 species and protein kinases causing tendon
damage or failure of the normal reparative process.
We found no correlation between size of tear, age, patient
pain scores, or the duration of symptoms with the histological grade and gene expression in all samples. Other groups
have shown a correlation between tear size and the degree
of cellular metabolism50 and cellular and vascular
changes51 in rotator-cuff tears, with smaller tears showing
the greatest potential to heal accompanied by increased
fibroblast cellularity and inflammatory cells, while large to
massive tears may have a reduced ability to repair and
regenerate because of highly degenerate tissue.51 This difference may be due to the fact that 11 of our 17 samples
(64.7%) were from small to medium tears (< 3 cm2) while
only six (35.3%) were from large to massive tears (> 3 cm2),
which may skew our gene and histological grading toward
the more active type of tissue and therefore greater gene
expression. Despite no evidence of MRI-determined tendinopathy or gross macroscopic abnormalities at the time of
arthroscopy, all matched samples of subscapularis tendons
showed evidence of mild to moderate tendinopathy. We
have previously suggested that the subscapularis from
patients with full-thickness rotator-cuff tears could be a
useful model of early human tendinopathy.34,52 The overexpression of cytokines in the subscapularis samples would
seem to confirm that at a molecular level there is an ongoing degenerative process, again suggesting that this may be
an ideal early model to investigate the molecular
pathophysiology of tendinopathy before clinical tendon
rupture occurs.
THE JOURNAL OF BONE AND JOINT SURGERY
CYTOKINES AND APOPTOSIS IN SUPRASPINATUS TENDINOPATHY
423
Ultimately there remains a complex regulatory network
between inflammatory cytokines and apoptosis in tendon
cells. It may be that apoptotic cell death induced by cytokines
in tissues exposed to normal stresses is a routine process used
to remove partially damaged cells while excessive stresses
may cause increased production of cytokine leading to
increased apoptosis and ultimate cell necrosis. We therefore
hypothesise, based on our findings and those of others,17,33,34 that an increase in the amount and duration of
load in a tendon cell may result in the activation of cytokines,
stress-activated protein kinases, oxygen-free radicals and
apoptotic mediators which when persistently activated cause
the tendon cells to undergo apoptosis resulting in a weakened collagenous matrix leading to subsequent tendon
degeneration and clinical tendon rupture.
There are limitations to our study. First, age-related
changes within the tendon samples could contribute to the
overall degenerative picture and gene expression seen in the
matched subscapularis tendons. However, the lack of degenerative change on MRI and arthroscopic examinations suggests that the differences are truly at the cellular level as
suggested by our work. Secondly, the rodent model of tendinopathy is unlikely to represent human age-related degenerative changes and thus would not reflect this in the microarray data. Thirdly, subscapularis tendon is functionally and
organisationally distinct from supraspinatus and thus
responds to mechanical loading in a different manner which
may alter its cytokine gene profile. Its genetic profile may
also be altered from cytokines released from supraspinatus
into the glenohumeral joint. It is reassuring, however, that we
found the same genes expressed in matched subscapularis tissue indicating that gene expression may be uniform within
joints subjected to tendon degeneration. Fourthly, these
genes were evaluated at a single interval and thus the pattern
of expression which we detected may not be uniform for the
whole duration of the tendon overuse exercise. Finally, the
tendon samples were not stained for fibroblast markers
which could have identified precisely which cells were
expressing the cytokine proteins.
Cytokines are favoured targets for intervention in inflammatory diseases since they are key regulatory proteins governing cells which drive inflammation and apoptosis. A
prominent approach to cytokine targeting has been the
development of protein-based biological therapeutics, such
as anticytokine antibodies and soluble cytokine receptors,
which are being tested clinically in other inflammatory conditions with promising results. Our finding of an increase in
inflammatory cytokines in early and late tendinopathy highlights their role in the degenerative process and may provide
novel targets in reducing the degree of tendon damage.
No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article.
Supplementary material
A further opinion by Dr P. Lapner is available with
the electronic version of this article on our website at
www.jbjs.org.uk
23. Lin TW, Cardenas L, Glaser DL, Soslowsky LJ. Tendon healing in interleukin-4
and interleukin-6 knockout mice. J Biomech 2006;39:61-9.
VOL. 91-B, No. 3, MARCH 2009
References
1. Neviaser RJ. Ruptures of the rotator cuff. Orthop Clin North Am 1987;18:387-94.
2. Boileau P, Brassart N, Watkinson DJ, et al. Arthroscopic repair of full-thickness
tears of the supraspinatus: does the tendon really heal? J Bone Joint Surg [Am]
2005;87-A:1229-40.
3. Lafosse L, Brozska R, Toussaint B, Gobezie R. The outcome and structural integrity of arthroscopic rotator cuff repair with use of the double-row suture anchor technique. J Bone Joint Surg [Am] 2007;89-A:1533-41.
4. Calvert PT, Packer NP, Stoker DJ, Bayley JL, Kessel L. Arthrography of the
shoulder after operative repair of the torn rotator cuff. J Bone Joint Surg [Br] 1986;68B:147-50.
5. Galatz LM, Ball CM, Teefey SA, Middleton WD, Yamaguchi K. The outcome and
repair integrity of completely arthroscopically repaired large and massive rotator cuff
tears. J Bone Joint Surg [Am] 2004;86-A:219-24.
6. Szomor ZL, Wang MX, Kruller A, et al. Differential expression of cytokines and
nitric oxide synthase isoforms in human rotator cuff bursae. Ann Rheum Dis
2001;60:431-2.
7. Lo IK, Marchuk LL, Hollinshead R, Hart DA, Frank CB. Matrix metalloproteinase
and tissue inhibitor of matrix metalloproteinase mRNA levels are specifically altered
in torn rotator cuff tendons. Am J Sports Med 2004;32:1223-9.
8. Fenwick SA, Curry V, Harrall RL, et al. Expression of transforming growth factorbeta isoforms and their receptors in chronic tendinosis. J Anat 2001;199:231-40.
9. Riley GP. Gene expression and matrix turnover in overused and damaged tendons.
Scand J Med Sci Sports 2005;15:241-51.
10. Qi J, Chi L, Maloney M, et al. Interleukin-1beta increases elasticity of human bioartificial tendons. Tissue Eng 2006;12:2913-25.
11. Sun HB, Li Y, Fung DT, et al. Coordinate regulation of Il-1beta and MMP-13 in rat
tendons following subrupture fatigue damage. Clin Orthop 2008;466:1555-61.
12. Nguyen MT, Lue H, Kleemann R, et al. The cytokine macrophage migration inhibitory factor reduces pro-oxidative stress-induced apoptosis. J Immunol
2003;170:3337-47.
13. Yuan J, Murrell GA, Wei AQ, Wang MX. Apoptosis in rotator cuff tendonopathy.
J Orthop Res 2002;20:1372-9.
14. Soslowsky LJ, Thomopoulos S, Tun S, et al. Overuse activity injures the supraspinatus tendon in an animal model: a histologic and biomechanical study. J Shoulder
Elbow Surg 2000;9:79-84.
15. Park HB, Yokota A, Gill HS, El Rassi G, McFarland EG. Diagnostic accuracy of
clinical tests for the different degrees of subacromial impingement syndrome. J Bone
Joint Surg [Am] 2005;87-A:1446-55.
16. Hayes K, Walton JR, Szomar ZL, Murrell GA. Reliability of 3 methods for assessing shoulder strength? J Shoulder Elbow Surg 2002;11:33-9.
17. Khan KM, Cook JL, Bonar F, Harcourt P, Astrom M. Histopathology of common
tendinopathies: update and implications for clinical management. Sports Med
1999;27:393-408.
18. Molloy TJ, Wang Y, Horner A, Skerry TM, Murrell GA. Microarray analysis of
healing rat Achilles tendon: evidence for glutamate signaling mechanism and embryonic gene expression in healing tendon tissue. J Orthop Res 2006;24:842-55.
19. Tsuzaki M, Bynum D, Almekinders L, et al. ATP modulates load-inducible IL1beta, COX 2, and MMP-3 gene expression in human tendon cells. J Cell Biochem
2003;89:556-62.
20. Voloshin I, Gelinas J, Maloney MD, et al. Proinflammatory cytokines and metalloproteases are expressed in the subacromial bursa in patients with rotator cuff disease. Arthroscopy 2005;21:1076.
21. Stone D, Green C, Rao U, et al. Cytokine-induced tendinitis: a preliminary study in
rabbits. J Orthop Res 1999;17:168-77.
22. Tsuzaki M, Guyton G, Garrett W, et al. IL-1 beta induces COX2, MMP-1, -3 and -13,
ADAMTS-4, IL-1 beta and IL-6 in human tendon cells. J Orthop Res 2003;21:256-64.
24. Nakama K, Gotoh M, Yamada T, et al. Interleukin-6-induced activation of signal
transducer and activator of transcription-3 in ruptured rotator cuff tendon. J Int Med
Res 2006;34:624-31.
424
N. L. MILLAR, A. Q. WEI, T. J. MOLLOY, F. BONAR, G. A. C. MURRELL
25. Jain A, Brennan F, Nanchahal J. Treatment of rheumatoid tenosynovitis with
cytokine inhibitors. Lancet 2002;360:1565-6.
26. Jain A, Kiriakidis S, Brennan F, et al. Targeting rheumatoid tenosynovial angiogenesis with cytokine inhibitors. Clin Orthop 2006;446:268-77.
27. Lee MW, Park SC, Kim JH, et al. The involvement of oxidative stress in tumor
necrosis factor (TNF)-related apoptosis-inducing ligand (TRAIL)-induced apoptosis in
HeLa cells. Cancer Lett 2002;182:75-82.
28. Lin MT, Juan CY, Chang KJ, Chen WJ, Kuo ML. IL-6 inhibits apoptosis and retains
oxidative DNA lesions in human gastric cancer AGS cells through up-regulation of
anti-apoptotic gene mcl-1. Carcinogenesis 2001;22:1947-53.
29. Machner A, Baier A, Wille A, et al. Higher susceptibility to Fas ligand induced
apoptosis and altered modulation of cell death by tumor necrosis factor-alpha in periarticular tenocytes from patients with knee joint osteoarthritis. Arthritis Res Ther
2003;5:253-61.
30. Lian O, Scott A, Engebretsen L, et al. Excessive apoptosis in patellar tendinopathy
in athletes. Am J Sports Med 2007;35:605-11.
31. Scott A, Khan KM, Heer J, et al. High strain mechanical loading rapidly induces
tendon apoptosis: an ex vivo rat tibialis anterior model. Br J Sports Med 2005;39:25.
32. Wolf BB, Green DR. Suicidal tendencies: apoptotic cell death by caspase family proteinases. J Biol Chem 1999;274:20049-52.
33. Yuan J, Murrell GA, Trickett A, Wang MX. Involvement of cytochrome c release
and caspase-3 activation in the oxidative stress-induced apoptosis in human tendon
fibroblasts. Biochim Biophys Acta 2003;1641:35-41.
34. Millar NL, Wei AQ, Molloy TJ, Bonar F, Murrell GAC. Heat shock protein and
apoptosis in supraspinatus tendinopathy. Clin Orthop 2008;466:1569-76.
35. Harvie P, Ostlere SJ, Teh J, et al. Genetic influences in the aetiology of tears of the
rotator cuff: sibling risk of a full-thickness tear. J Bone Joint Surg [Br] 2004;86-B:696-700.
36. Morand EF, Leech M, Bernhagen J. MIF: a new cytokine link between rheumatoid
arthritis and atherosclerosis. Nat Rev Drug Discov 2006;5:399-410.
37. Leung BP, Culshaw S, Gracie JA, et al. A role for IL-18 in neutrophil activation. J
Immunol 2001;167:2879-86.
38. John T, Kohl B, Mobasheri A, Ertel W, Shakibaei M. Interleukin-18 induces
apoptosis in human articular chondrocytes. Histol Histopathol 2007;22:469-82.
39. Kitaura H, Tatamiya M, Nagata N, et al. IL-18 induces apoptosis of adherent bone
marrow cells in TNF-alpha mediated osteoclast formation in synergy with IL-12.
Immunol Lett 2006;107:22-31.
40. Sotiriadou NN, Perez SA, Gritzapis AD, et al. Beneficial effect of short-term
exposure of human NK cells to IL15/IL12 and IL15/IL18 on cell apoptosis and function.
Cell Immunol 2005;234:67-75.
41. Sparmann G, Glass A, Brock P, et al. Inhibition of lymphocyte apoptosis by pancreatic stellate cells: impact of interleukin-15. Am J Physiol Gastrointest Liver Physiol
2005;289:842-51.
42. Demirci G, Li XC. IL-2 and IL-15 exhibit opposing effects on Fas mediated apoptosis.
Cell Mol Immunol 2004;1:123-8.
43. Lopez-Armada MJ, Carames B, Lires-Dean M, et al. Cytokines, tumor necrosis
factor-alpha and interleukin-1beta, differentially regulate apoptosis in osteoarthritis
cultured human chondrocytes. Osteoarthritis Cartilage 2006;14:660-9.
44. Park H, Ahn Y, Park CK, Chung HY, Park Y. Interleukin-6 protects MIN6 beta cells
from cytokine-induced apoptosis. Ann N Y Acad Sci 2003;1005:242-9.
45. Arnoczky SP, Tian T, Lavagnino M, et al. Activation of stress-activated protein
kinases (SAPK) in tendon cells following cyclic strain: the effects of strain frequency,
strain magnitude, and cytosolic calcium. J Orthop Res 2002;20:947-52.
46. Murrell GA, Szabo C, Hannafin JA, et al. Modulation of tendon healing by nitric
oxide. Inflamm Res 1997;46:19-27.
47. Boileau C, Martel-Pelletier J, Moldovan F, et al. The in situ up-regulation of
chondrocyte interleukin-1-converting enzyme and interleukin-18 levels in experimental osteoarthritis is mediated by nitric oxide. Arthritis Rheum 2002;46:2637-47.
48. Saura M, Zaragoza C,, Bao C, et al. Stat3 mediates interleukin-6 [correction of
interleukin-6] inhibitor of human endothelial nitric-oxide synthase expression. J Biol
Chem 2006;281:30057-62.
49. Yanagita M, Shimabukuro Y, Nozaki T, et al. UL-15 up-regulated iNOS expression
and NO production by gingival epithelial cells. Biochem Biophys Res Commun
2002;297:329-34.
50. Matthews TJ, Smith SR, Peach CA, et al. In vivo measurement of tissue metabolism in tendons of the rotator cuff: implications for surgical management. J Bone Joint
Surg [Br] 2007;89-B:633-8.
51. Matthews TJ, Hand GC, Rees JL, Athanasou NA, Carr AJ. Pathology of the torn
rotator cuff tendon: reduction in potential for repair as tear size increases. J Bone
Joint Surg [Br] 2006;88-B:489-95.
52. Yuan J, Wang MX, Murrell GA. Cell death and tendinopathy. Clin Sports Med
2003;22:693-701.
THE JOURNAL OF BONE AND JOINT SURGERY