close

Вход

Забыли?

вход по аккаунту

?

The response of human anulus fibrosus cells to cyclic tensile strain is frequency-dependent and altered with disc degeneration.

код для вставкиСкачать
ARTHRITIS & RHEUMATISM
Vol. 62, No. 11, November 2010, pp 3385–3394
DOI 10.1002/art.27643
© 2010, American College of Rheumatology
The Response of Human Anulus Fibrosus Cells to
Cyclic Tensile Strain Is Frequency-Dependent and
Altered with Disc Degeneration
Hamish T. J. Gilbert, Judith A. Hoyland, and Sarah J. Millward-Sadler
Results. The expression of catabolic genes
(MMP-3 and ADAMTS-4) in AF cells derived from
nondegenerated tissue decreased in response to 1.0 Hz
of CTS, whereas changing the frequency to 0.33 Hz
resulted in a shift toward matrix catabolism. Application of 1.0 Hz of CTS reduced anabolic gene expression
(aggrecan and type I collagen) in AF cells derived from
degenerated tissue, with 0.33 Hz of CTS resulting in
increased catabolic gene expression.
Conclusion. The response of human AF cells to
CTS is frequency-dependent and is altered by degeneration.
Objective. Mechanical loads are important for
homeostasis of the intervertebral disc (IVD) cell matrix,
with physiologic and nonphysiologic loads leading to
matrix anabolism and catabolism, respectively. Previous investigations into the effects of load on disc cells
have predominantly used animal models, with the limited number of human studies focusing primarily on
nucleus pulposus cells. The aim of this study was to
examine the effect of cyclic tensile strain (CTS) on
human anulus fibrosus (AF) cells to ascertain whether
the response was frequency-dependent and to compare
AF cells derived from nondegenerated and degenerated
tissue samples.
Methods. AF cells were isolated from nondegenerated and degenerated human IVDs, expanded in
monolayer, and cyclically strained for 20 minutes, applying 10% strain at a frequency of 1.0 Hz or 0.33 Hz
with the use of a Flexcell strain device. Total RNA was
extracted from the cells at baseline (control) and at 1, 3,
and 24 hours following application of CTS. Real-time
quantitative polymerase chain reaction was used to
analyze gene expression of matrix proteins (aggrecan,
type I collagen, and type II collagen) and enzymes
(matrix metalloproteinases [MMPs] 3, 9, 13, and
ADAMTS-4).
The intervertebral disc (IVD) can be divided into
2 main regions, the central gelatinous nucleus pulposus
(NP) and the outer anulus fibrosis (AF). The forces that
act on these regions and their component cells are
multiple and complex, with cells in the NP exposed to
hydrostatic pressures and cells within the AF exposed
primarily to tensile strain (1,2). Internal disc pressures
are predicted to range from 0.1 MPa (while lying
down) to 2.5 MPa (while lifting a heavy weight) (3), with
tissue strains reaching 25% during physiologic compression (4).
Mechanical load is known to be important for the
maintenance and integrity of the IVD cell matrix, with
physiologic mechanical stimuli leading to matrix anabolism (5–11). Hutton et al (12) found that in canine NP
cells, 1 MPa of hydrostatic pressure led to increased
collagen and proteoglycan synthesis as compared with
atmospheric pressure, while Matsumoto et al (11) reported increased collagen synthesis in rabbit NP cells
exposed to 20% cyclic tensile strain (CTS). Furthermore, exposure of the IVD to nonphysiologic mechanical load (e.g., overloading or immobilization) results in
altered cellular metabolism, which leads to degenerative
changes (9,13–16). Wang et al (9) showed that the
anabolic response of rabbit IVD explants to dynamic
Supported by Arthritis Research UK (grant 17850). The
Intervertebral Disc Research Group at the University of Manchester
School of Biomedicine is supported by the Manchester Academic
Health Sciences Centre and the National Institute of Health Research,
Manchester Biomedical Research Centre.
Hamish T. J. Gilbert, BSc (Hons), Judith A. Hoyland, PhD,
Sarah J. Millward-Sadler, PhD: University of Manchester, Manchester, UK.
Address correspondence and reprint requests to Sarah J.
Millward-Sadler, PhD, School of Biomedicine, Faculty of Medical and
Human Sciences, The University of Manchester, Stopford Building,
Oxford Road, Manchester M13 9PT, UK. E-mail: Jane.Sadler@
manchester.ac.uk.
Submitted for publication January 29, 2010; accepted in
revised form June 29, 2010.
3385
3386
compression was inhibited when the compression was
changed to a static force and that static in vivo bending
of murine IVDs resulted in increased cell death and
decreased type II collagen gene expression (14).
The type, magnitude, frequency, and duration of
force have all been identified as important factors in
determining IVD cell metabolism in response to load
(9,10,13,15,17–21).
Maclean et al (7) found that NP cells in an in vivo
rat tail model responded in a frequency-dependent
manner, with low-frequency (0.01 Hz) and highfrequency (1 Hz) loading having anabolic and catabolic
effects, respectively, whereas the response of AF cells
was mainly magnitude-dependent. A frequencydependent response was also shown in hydrostatically
loaded porcine NP and AF cells, where a frequency of 5
Hz resulted in poor collagen synthesis and increased
degradation, whereas frequencies above and below this
value had minimal effects (18).
Animal models have been predominantly used in
mechanical loading studies, but differences in solute
transport and matrix biochemistry due to variations in
disc size and age, respectively, prevent their use as an
accurate model for investigating human IVD mechanobiology (22). Thus, to ascertain the effects of load on
human IVD, in vitro mechanical loading systems capable
of directing specific and quantifiable mechanical stimuli
to cultured cells have been used to study magnitudedependent responses. Handa et al (23) found that
hydrostatic pressure influenced proteoglycan synthesis
and matrix metalloproteinase (MMP) gene expression in
human NP explants, with low hydrostatic pressure promoting matrix anabolism and high hydrostatic pressure
leading to matrix catabolism, a response also confirmed
by studies conducted by Neidlinger-Wilke et al (15).
Although these human studies show the importance of mechanical load in IVD cell matrix regulation,
the majority of studies have focused on the response of
hydrostatic pressure–stimulated NP cells, with only a few
studies investigating the response of AF cells to load.
Using human and bovine tissue explants, Ishihara et al
(21) showed that hydrostatic pressure influenced proteoglycan synthesis by NP cells and inner AF cells, while
outer AF cells remained unaffected by this type of load.
Using canine cells, Hutton et al (12) found that proteoglycan and collagen synthesis were stimulated in NP cells
but inhibited in AF cells following hydrostatic pressure.
In studies of human cell cultures, Le Maitre et al (24)
observed that the response of NP cells exposed to
hydrostatic pressure was absent in AF cells exposed to
an identical stimulus. These findings support speculation
GILBERT ET AL
that hydrostatic pressure is not the predominant load
experienced by AF cells in vivo and suggest that investigating a more physiologically relevant load would be
beneficial in progressing our understanding of AF cell
biomechanics.
To date, there are only limited studies of AF cells
treated with CTS, the more physiologically relevant
load. Wuertz et al (25) found that 6% CTS had only a
modest effect on matrix gene expression of human AF
cells in 3-dimensional culture as compared with the
influence of different osmotic potentials. Sowa and
Agarwal (26) found that CTS exhibited a protective
effect against inflammatory cytokine–stimulated catabolism in monolayer cultures of rat AF cells, whereas
Rannou et al (27) found that CTS initiated a catabolic
response in rabbit AF cells.
Although the effect of mechanical load on nondegenerated IVDs has been documented, investigations
into the implications of degeneration during loading
remain limited. However, a recent study showed that NP
cells derived from degenerated IVDs failed to respond
to hydrostatic pressure (24). This inability to respond to
physiologic loads in the normal anabolic way could result
in altered matrix regulation and lead to the progression
of degenerative disc disease. As such, if the response to
load is altered in cells derived from degenerated IVDs
then elucidation of the altered mechanotransduction
pathway could lead to future therapeutic targets for the
prevention or treatment of degenerative disc disease.
To our knowledge, there are currently no studies
investigating the effect of frequency on cyclically
strained human AF cells, nor any studies comparing the
response of AF cells derived from nondegenerated or
degenerated IVDs to load. For the present study, an in
vitro loading system (FX-4000 Flexcell Tension system)
capable of directing CTS to cells in monolayer was used
to assess the effect of a physiologically relevant mechanical stimulus on human AF cells. The aims were to
investigate the effect of CTS on AF cells isolated from
human IVDs to ascertain whether the response was
frequency-dependent and to investigate whether the
response of AF cells derived from nondegenerated and
degenerated IVDs differed.
MATERIALS AND METHODS
IVD tissue samples. Human IVD tissue was collected
at the time of surgery or at postmortem examination (within 18
hours of death), with the consent of the patient or the relatives
of the decedent and with the approval of Central Manchester,
Bury, Rochdale, Salford, and Trafford Research Ethics Committees. A section of formalin-fixed, paraffin-embedded tissue
RESPONSE OF HUMAN AF CELLS TO CYCLIC TENSILE STRAIN
Table 1. Characteristics of the intervertebral disc tissue specimens
examined
Sample
Source
Grade*
Age,
years
Disc
level
1
2
3
4
5
6
Postmortem
Postmortem
Postmortem
Postmortem
Surgical
Surgical
1 (nondegenerated)
1 (nondegenerated)
1 (nondegenerated)
7 (mildly degenerated)
9 (degenerated)
9 (degenerated)
57
46
37
57
29
66
L4/5
L5/S1
L5/S1
L2/3
L4/5
L4/5
* Histologic changes in the tissues were graded on a scale of 0–12,
where 0–3 ⫽ nondegenerated, 4–7 ⫽ mildly degenerated, and 8–12 ⫽
severely degenerated.
derived from the entire IVD containing intact NP and AF was
graded for histologic changes as previously reported (28).
Tissue changes were scored on a scale of 0–12, where 0–3 ⫽
nondegenerated tissue, 4–7 ⫽ mildly degenerated, and 8–12 ⫽
severely degenerated. Characteristics of the tissue samples
used in these studies are detailed in Table 1.
Isolation and culture of AF cells. Within 24 hours of
death or surgical removal, AF tissue (including the inner and
outer anuli) was separated from the IVD, finely minced,
digested with 300–350 PUK/ml of Pronase (Calbiochem) for 1
hour, and then with 0.25% type II collagenase (Invitrogen) and
0.01% hyaluronidase (Sigma) for 4 hours at 37°C with constant
agitation in Dulbecco’s modified Eagle’s medium.
AF cells were cultured in standard medium, which
consisted of Dulbecco’s modified Eagle’s medium, with 4.5
gm/liter glucose, Glutamax and pyruvate (both from Gibco)
containing 50 ␮g/ml of ascorbic acid, 250 ng/ml of amphotericin, 100 units/ml of penicillin, 100 ␮g/ml of streptomycin
(Invitrogen), and 10% fetal calf serum (Invitrogen). Cells were
expanded in monolayer, with medium changes every 2–3 days.
Cells with passage numbers of ⱕ6 were trypsinized and seeded
onto untreated silicone membrane BioFlex culture plates
3387
(Flexcell International) at a density of 1 ⫻ 105 cells/ml in 2 ml
of standard medium and were allowed to adhere for 48 hours
(passage number ⬎6 has been found to influence cell behavior
[Hoyland JA: unpublished observations]). Medium was
changed to serum-free medium at 15–17 hours prior to application of CTS.
Application of CTS using a FX-4000 Flexcell tension
system. BioFlex culture plates containing AF cells in serumfree medium were placed into a vacuum manifold inside an
incubator set to a temperature of 37°C in an atmosphere of 5%
CO2. A precise amount of negative pressure was directed to
the base of the culture plates, causing deformation of the
silicone membranes over the loading posts. The use of 25-mm
loading posts with BioFlex culture plates allows computercontrolled equibiaxial strain of the silicone membranes located
directly above the posts. This strain is directly translated to the
adherent cells, resulting in equibiaxial cellular deformation.
The membrane regions and adherent cells located away from
the loading posts experience biaxial strain with greater elongation in the radial direction as compared with the circumferential direction (29).
For this study, a FX-4000 Flexcell tension system
(Flexcell International) was used to apply CTS of 10% (directed by the computer), resulting in a 10% membrane strain
at the central region and a strain of 14% radial and 5%
circumferential at the membrane periphery (29,30). The AF is
predicted to experience compressive tissue strains of 1–25%
and radial tensile strains of up to 19% during multiaxial
bending motions when compressed with a load that is physiologically similar to that of walking (3,4). Two different loading
regimens were used: either 10% strain at a frequency of 0.33
Hz (1.5 seconds of extension, with 1.5 seconds of relaxation, at
a rate of 20 cycles/minute) for 20 minutes or 10% strain at a
frequency of 1.0 Hz (0.5 seconds of extension, with 0.5 seconds
of relaxation, at a rate of 60 cycles/minute) for 20 minutes.
Frequencies of 1.0 Hz and 0.33 Hz were chosen as being
representative of physiologic locomotion and less than physiologic locomotion, respectively (31). Cells adhering to un-
Table 2. Primer and probe sequences used in the real-time polymerase chain reaction analysis
Target gene
Forward primer
Reverse primer
GAPDH
GGT-GAA-CGG-GAA-GCT-CAC-T
AGG-TCA-GGT-CCA-CCA-CTG-A
Aggrecan
CCG-TGT-GTC-CAA-GGA-GAA-GG
GGG-TAG-TTG-GGC-AGT-GAG-AC
Type I
collagen
Type II
collagen
MMP-3*
AGA-ACA-GCG-TGG-CCT-ACA-TG
GCG-CGG-ATC-TCG-ATC-TCG
ATG-GAG-ACT-GGC-GAG-ACT-TG
GCT-GCT-CCA-CCA-GTT-CTT-CTT
TGA-AGA-GTC-TTC-CAA-TCCTAC-TGT-TG
CCC-GGA-GTG-AGT-TGA-ACC-A
CTA-GAT-ATT-TCT-GAA-CAA-GGTTCA-TGC-A
CAG-GAC-GGG-AGC-CCT-AGT-C
MMP-9
MMP-13
CCC-CAG-GCA-TCA-CCA-TTCAAG
ADAMTS-4 ACT-GGT-GGT-GGC-AGA-TGA-CA
* MMP-3 ⫽ matrix metalloproteinase 3.
Probe
CCC-CAG-TGC-CAACGT-G
CTG-ATA-GGC-ACTGTT-GAC
CAG-CAG-ACT-GGCAAC
CCC-AAT-CCA-GCAAAC-G
TTT-GCT-CAG-CCT-ATCCAT
TAC-GTG-ACC-TATGAC-ATC
CTG-CCT-TCC-TCT-TC
GAC-AAA-TCA-TCT-TCA-TCA-CCACCA-C
TCA-CTG-TTA-GCA-GGT-AGC-GCT-TT ATG-GCC-GCA-TTC-C
GenBank
accession
no.
NM_002046
NM_001135
NM_000088
NM_001844
NM_002422
NM_004994
NM_002427
NM_005099
3388
GILBERT ET AL
stimulated BioFlex culture plates served as controls. Stimulated and unstimulated AF cells were incubated at 37°C for 0
(baseline control), 1, 3, or 24 hours postload in serum-free
medium, and total RNA was extracted.
Cell viability. Cell viability was assessed before and
after cyclic stretching using a trypan blue exclusion assay.
Briefly, 100 ␮l of trypan blue (0.4%; Sigma) was added to each
well of a BioFlex culture plate containing unstimulated or
stimulated AF cells. Stained AF cells were visualized using a
light microscope (Leitz), and the percentage of viable cells
(cells with unstained nuclei) was calculated by counting 100
cells in 5 random fields of view.
Real-time quantitative polymerase chain reaction
(PCR) analysis. Total RNA was extracted from each BioFlex
culture plate well using TRIzol reagent (Invitrogen) according
to the manufacturer’s instructions, and samples were treated
with DNase I (Ambion). RNA quality and quantity were
determined using a NanoDrop ND-1000 spectrophotometer
and 500 ng of RNA that had been reverse transcribed using a
high-capacity reverse transcription kit (Applied Biosystems).
Real-time quantitative PCR was performed in triplicate using
TaqMan Universal PCR Master Mix (Applied Biosystems)
with primers and probes at 900 nM and 250 nM, respectively,
for GAPDH, aggrecan, types I and II collagen, MMPs 3, 9, and
13, and ADAMTS-4 (Table 2). Data were analyzed using the
2–⌬⌬Ct method (32,33), and the results were normalized to the
endogenous control gene GAPDH and unloaded baseline
controls. Primers were designed using Primer Express Software (Applied Biosystems) (Table 2).
Statistical analysis. Nonparametric data, as determined by the Shapiro-Wilk test, were analyzed by MannWhitney U test.
RESULTS
Cell viability and baseline expression. AF cells
isolated from nondegenerated and degenerated IVDs
remained viable (⬎90%) throughout the culture period,
with no change in viability following the application of
CTS. Unloaded controls showed no significant change in
gene expression for any of the genes investigated at any
of the time points analyzed (data not shown). There was
no significant difference in the baseline levels of relative
gene expression for type II collagen, MMP-3, MMP-9,
MMP-13, or ADAMTS-4 between AF cells from nondegenerated or degenerated IVDs. Baseline relative
gene expression of aggrecan and type I collagen was
significantly increased (P ⬍ 0.01) and decreased (P ⬍
0.05), respectively, in AF cells derived from degenerated
IVDs as compared with those from nondegenerated
IVDs.
Effects of CTS at a frequency of 1.0 Hz. Application of CTS at 10% strain and a frequency of 1.0 Hz for
20 minutes to AF cells from nondegenerated IVDs
resulted in no significant change in the relative expression of any of the matrix genes investigated. Exposure of
Figure 1. Effect of cyclic tensile strain on matrix gene expression in
anulus fibrosus cells from nondegenerated and degenerated intervertebral discs (IVDs). Cells derived from nondegenerated or degenerated IVDs were mechanically stimulated (M/S) at 10% strain and a
frequency of 1.0 Hz for 20 minutes and then incubated for up to 24
hours prior to analysis. Real-time quantitative polymerase chain
reaction analysis was used to determine gene expression of matrix
proteins. Levels of aggrecan (A) and type I collagen (B) were
determined in degenerated and nondegenerated tissue relative to the
housekeeping gene GAPDH and were normalized to the unstimulated
baseline control. Values are the mean ⫾ SEM of samples from 3
subjects per group. ⴱ ⫽ P ⱕ 0.05 versus unstimulated baseline control.
AF cells from degenerated IVDs to an identical stimulus
resulted in a significant down-regulation of aggrecan and
type I collagen genes at 24 hours following mechanical
stimulation (5-fold [P ⬍ 0.01] and 6-fold [P ⬍ 0.05],
respectively) (Figures 1A and B), but no change in the
relative expression of the type II collagen gene (data not
shown).
Analysis of matrix-degrading enzymes in AF cells
from nondegenerated IVDs subjected to a 1.0-Hz stimulus showed a significant decrease in the expression of
the MMP-3 gene at 1 hour and at 24 hours poststimulation (6-fold [P ⬍ 0.05] and 7-fold [P ⬍ 0.01], respec-
RESPONSE OF HUMAN AF CELLS TO CYCLIC TENSILE STRAIN
3389
Figure 2. Effect of cyclic tensile strain on matrix-degrading enzyme
gene expression in anulus fibrosus cells from nondegenerated and
degenerated intervertebral discs (IVDs). Cells derived from nondegenerated or degenerated IVDs were mechanically stimulated at 10%
strain and a frequency of 1.0 Hz for 20 minutes and then incubated for
up to 24 hours prior to analysis. Real-time quantitative polymerase
chain reaction analysis was used to determine gene expression of
matrix-degrading enzymes. Levels of matrix metalloproteinase 3
(MMP-3) (A) and ADAMTS-4 (B) were determined in degenerated
and nondegenerated tissue relative to the housekeeping gene GAPDH
and were normalized to the unstimulated baseline control. Values are
the mean ⫾ SEM of samples from 3 subjects per group. ⴱ ⫽ P ⱕ 0.05
versus unstimulated baseline control.
tively), but the decrease at 3 hours did not achieve
significance (P ⫽ 0.07). Relative expression of the
ADAMTS-4 gene was also significantly decreased at 1
hour poststimulation (7-fold; P ⬍ 0.05) (Figure 2B).
Relative expression of the MMP-9 and MMP-13 genes
remained unchanged (data not shown). AF cells from
degenerated IVDs showed no change in the relative
expression of the matrix-degrading enzyme genes at any
time point following 1.0 Hz of stimulation.
Effects of CTS at a frequency of 0.33 Hz. Mechanical stimulation of AF cells from nondegenerated
Figure 3. Effect of cyclic tensile strain on matrix gene expression in
anulus fibrosus cells from nondegenerated and degenerated intervertebral discs (IVDs). Cells derived from nondegenerated or
degenerated IVDs were mechanically stimulated at 10% strain and
a frequency of 0.33 Hz for 20 minutes and then incubated for up to
24 hours prior to analysis. Real-time quantitative polymerase chain
reaction analysis was used to determine gene expression of matrix
proteins. Levels of aggrecan (A), type I collagen (B), and type II
collagen (C) were determined in degenerated and nondegenerated
tissue relative to the housekeeping gene GAPDH and were normalized to the unstimulated baseline control. Values are the mean ⫾
SEM of samples from 3 subjects per group. ⴱ ⫽ P ⱕ 0.05 versus
unstimulated baseline control.
3390
IVDs with 10% strain at a frequency of 0.33 Hz for 20
minutes resulted in decreased relative expression of type
I collagen (9-fold decrease at all 3 time points; P ⬍ 0.05)
and type II collagen (10-fold decrease at 1 hour [P ⬍
0.01], 9-fold decrease at 3 hours [P ⬍ 0.05], and 8-fold
decrease at 24 hours [P ⬍ 0.01]) (Figures 3B and C). AF
cells from nondegenerated IVDs exposed to 0.33 Hz of
CTS showed no difference in the relative gene expression of aggrecan between stimulated tissues and unstimulated controls (Figure 3A).
The expression of types I and II collagen genes in
AF cells from degenerated IVDs following exposure to
0.33 Hz of CTS was similar to that observed in AF cells
from nondegenerated IVDs exposed to the same stimulus. The relative gene expression of type I collagen was
decreased at all 3 time points (⬃8-fold; P ⬍ 0.01),
whereas the relative gene expression of type II collagen
was significantly decreased at 1 hour and at 24 hours
post-CTS (14-fold [P ⬍ 0.01] and 9-fold [P ⬍ 0.05],
respectively), although the decreased expression at the
3-hour time point was not significant as compared with
control (Figures 3B and C). In contrast, AF cells from
degenerated IVDs showed decreased expression of aggrecan 24 hours poststimulus (6-fold; P ⬍ 0.01) (Figure
3A) but showed no significant change at 1 hour or at 3
hours poststimulation.
Exposure of AF cells from nondegenerated and
degenerated IVDs to CTS at a frequency of 0.33 Hz
resulted in an increase in the relative gene expression of
MMP-9 at 3 hours poststimulation (4-fold; P ⬍ 0.05),
with a return toward baseline values by 24 hours (Figure
4A). MMP-13 relative gene expression was decreased at
all 3 time points in AF cells from nondegenerated IVDs,
but the difference did not reach significance. AF cells
from degenerated IVDs subjected to 0.33 Hz of CTS
showed increased relative gene expression of MMP-13
at 3 hours poststimulation (2-fold; P ⬍ 0.05), with
expression returning to baseline values by 24 hours
(Figure 4B). ADAMTS-4 relative gene expression remained unchanged in AF cells from nondegenerated
IVDs following 0.33 Hz of CTS but was decreased at 1
hour poststimulation in AF cells from degenerated IVDs
exposed to the same stimulus (5-fold; P ⬍ 0.05) (Figure
4C). There was no change in gene expression of MMP-3
following cyclic stretch of AF cells from nondegenerated
or degenerated IVDs at a frequency of 0.33 Hz (data not
shown).
DISCUSSION
AF cells are exposed to CTS during normal
physiologic movement of the spine, and this mechanical
GILBERT ET AL
Figure 4. Effect of cyclic tensile strain on matrix-degrading enzyme
gene expression in anulus fibrosus cells from nondegenerated and
degenerated intervertebral discs (IVDs). Cells derived from nondegenerated or degenerated IVDs were mechanically stimulated at 10%
strain and a frequency of 0.33 Hz for 20 minutes and then incubated
for up to 24 hours prior to analysis. Real-time quantitative polymerase
chain reaction analysis was used to determine gene expression of
matrix-degrading enzymes. Levels of matrix metalloproteinase 9
(MMP-9) (A), MMP-13 (B), and ADAMTS-4 (C) were determined in
degenerated and nondegenerated tissue relative to the housekeeping
gene GAPDH and were normalized to the unstimulated baseline
control. Values are the mean ⫾ SEM of samples from 3 subjects per
group. ⴱ ⫽ P ⱕ 0.05 versus unstimulated baseline control.
RESPONSE OF HUMAN AF CELLS TO CYCLIC TENSILE STRAIN
stimulation is important in regulating the AF cell matrix.
This study is the first to compare the effects of CTS on
human AF cells derived from both nondegenerated and
degenerated IVDs, where the aims were to characterize
the response in terms of anabolic and catabolic gene
expression and to compare the effects of loading frequency.
The AF cells derived from both nondegenerated
and degenerated IVDs showed a frequency-dependent
response to CTS, although the response was most dramatic in the nondegenerated AF cells. At a frequency of
1.0 Hz, which is similar to the frequency of normal
physiologic locomotion in humans (31), AF cells derived
from nondegenerated tissue exhibited decreased gene
expression of MMP-3 and ADAMTS-4, suggesting a
shift in matrix regulation toward reduced catabolism. In
contrast, altering the frequency to 0.33 Hz resulted in a
catabolic response of AF cells derived from nondegenerated IVDs, with decreased expression of types I and II
collagen genes and increased expression of the MMP-9
gene.
Frequency has previously been shown to be an
important factor in determining cellular metabolic activity in response to load. MacLean et al (7) found a
distinct frequency-dependent response in NP cells using
a rat tail dynamic compression model, with anabolic
gene expression elevated at a frequency of 0.01 Hz and
catabolic gene expression elevated at a 1.0-Hz frequency. Although the AF cells in that study responded
less dramatically than the NP cells, there was still an
obvious frequency-dependent response to 0.2 MPa of
compression, with anabolic gene expression elevated at
the 1.0-Hz frequency only, suggesting an overall anabolic
response at the tissue level (7). Similarly, the response of
AF cells to CTS at a frequency of 1.0 Hz in our study
suggests an overall anabolic effect as a consequence of
reduced catabolic gene expression. The differences in
the genes regulated between these studies could be due
to species variation and differences in the type and
magnitude of force experienced at the cellular level. The
in vitro model system we used examines the effect of cell
deformation due to CTS only, and although we are
unable to exclude cell stimulation due to fluid flow, our
system allowed the removal of all other types of mechanical stimulation present during in vivo mechanical loading.
The biochemical processes that enable cells to
respond to mechanical stimuli in a frequency-dependent
manner remain largely unknown; however, a study investigating the effects of CTS on bone cells found that
osteoblasts used different integrin receptors at different
3391
frequencies of strain, suggesting that there are different
mechanotransduction pathways operating between frequencies (34). Other factors that could be involved in
the frequency-dependent response observed in the
present study are the differences in the strain rate and
the number of cycles (400 cycles at the 0.33-Hz frequency and 1,200 cycles at the 1.0-Hz frequency). Both
the strain rate and the cycle number could affect which
mechanotransduction pathways are activated and, consequently, could alter the kinetics of such pathways.
The response of AF cells derived from degenerated tissue was also found to be frequency-dependent,
with 1.0 Hz of cyclic strain leading to reduced matrix
anabolism (decreased aggrecan and type I collagen gene
expression) and 0.33 Hz cyclic strain leading to matrix
catabolism (decreased anabolic gene expression and
increased catabolic gene expression). Altering the frequency of the strain led to a switch in gene regulation,
from matrix-degrading enzyme genes to matrix genes for
AF cells derived from nondegenerated tissue and from
matrix genes to matrix-degrading enzyme genes for AF
cells derived from degenerated tissue. Although an
alteration in gene regulation occurred in both cell types,
the frequency-dependent response was most prominent
in the AF cells derived from nondegenerated tissue,
where there was a shift in matrix regulation, from reduced
catabolism at 1.0 Hz (suggesting matrix anabolism at the
tissue level) to increased catabolism at 0.33 Hz.
This change in matrix regulation, which was due
to altered frequency, from an overall anabolic effect to a
catabolic effect was not replicated in the AF cells
derived from degenerated tissue, where both frequencies ultimately led to matrix catabolism. This appears to
be primarily due to the altered response observed between AF cells derived from nondegenerated and degenerated tissue at a 1.0-Hz frequency of strain, suggesting that the response of AF cells to CTS is not only
dependent on the frequency, but also on whether the
cells are derived from nondegenerated or degenerated
tissue. Our data suggest that AF cells derived from
degenerated tissue have not only lost their ability to
down-regulate matrix enzyme gene expression in response to a 1.0-Hz frequency of CTS, but are also
responding by reducing their matrix protein gene expression (decreased aggrecan and type I collagen).
It is also interesting that the differential response
to CTS between AF cells derived from nondegenerated
IVDs and those derived from degenerated IVDs occurred through the regulation of a subset of different
genes, with nondegenerated AF cells altering their
matrix-degrading enzyme gene expression and degener-
3392
ated AF cells altering their matrix protein gene expression. This lack of structural matrix gene regulation with
CTS in AF cells derived from nondegenerated IVDs has
been reported in 2 previous studies, where strains of 1%
or 5% and strains of 6%, respectively, resulted in no
change in the relative gene expression of aggrecan and
biglycan and in the relative gene expression of aggrecan
and types I and II collagen, respectively (26,27). Although the study by Rannou et al (27) found no changes
in metalloproteinase (MMP-1 and MMP-3) expression,
that study and the study by Sowa et al (26) suggested that
CTS modulates matrix regulation in AF cells derived
from nondegenerated tissue by altering the expression of
matrix-degrading enzyme genes as opposed to structural
matrix genes.
Another difference in the cellular response of AF
cells derived from nondegenerated and degenerated
tissue to CTS was the time taken for a change in gene
expression to occur following exposure to 1.0 Hz of
mechanical stimulation. Matrix-degrading enzyme gene
expression was reduced at 1 hour, and matrix protein
gene expression was reduced at 24 hours, following
application of CTS in AF cells derived from nondegenerated and degenerated tissues, respectively. This apparent delay in the CTS-induced change in matrix gene
expression in AF cells derived from degenerated IVDs
could be indicative of an altered or alternative mechanotransduction pathway being in operation.
Both nondegenerated and degenerated AF cells
responded catabolically to 0.33 Hz of CTS; however, the
response of degenerated AF cells included the regulation of additional genes (up-regulation of MMP-13 and
down-regulation of ADAMTS-4), again suggesting potential differences in mechanotransduction.
Although there are no other studies comparing
the effects of load on AF cells derived from nondegenerated and degenerated tissues, the effect of age on
loaded AF cells has been investigated. Korecki et al (20)
found that AF cells from mature (18–24 months) bovine
caudal discs responded anabolically, with increased expression of types I and II collagen genes, while AF cells
from young (4–6 months) bovine caudal discs responded
catabolically, with increased expression of MMP-3 gene.
Thus, the effect of age appears to be dissimilar to the
effect of degeneration on the response of mechanically
loaded AF cells. This suggests that age and degeneration
are separate processes that have differential effects on
the response of AF cells to load. Equally, the differences
observed between the studies could be due to species
variation or limitations in using relatively young cows as
a model for age.
GILBERT ET AL
The difference observed in the cellular response
to CTS at a frequency of 1.0 Hz between nondegenerated and degenerated AF cells suggests that these 2 cell
types may use different mechanotransduction pathways.
A previous study by our group (35) showed an altered
integrin mechanotransduction pathway operating in hydrostatically loaded NP cells derived from degenerated
IVDs. Together, these results suggest that the mechanotransduction pathways differ in disc cells derived from
both the AF and the NP regions of nondegenerated and
degenerated tissues.
The glycosaminoglycan (GAG) content of the AF
has been shown to be decreased in the IVDs of patients
with low back pain (36), as well as with increasing age of
the tissue (37), with loss of the negatively charged GAGs
from the AF matrix potentially leading to a hypoosmotic cellular environment. Furthermore, changes in
the osmotic pressure have been shown to influence
mechanical load–regulated gene expression in AF and
NP cells (25). The mechanotransduction pathway that is
active in AF cells derived from degenerated IVDs may
therefore be altered in order to maintain mechanical
load–regulated gene expression during hypo-osmotic
conditions, leading to the altered response observed
between nondegenerated and degenerated AF cells
when exposed to 1.0 Hz of CTS under physiologically
nondegenerated osmotic conditions.
The amount of strain acting on disc cells during
physiologic loading is likely to differ with degeneration
due to loss of matrix proteins, leading to diminished
elasticity and increased stiffness of the disc (38,39). This
difference between the native mechanical environment
of cells derived from nondegenerated and degenerated
IVDs could lead to differences in the threshold of
mechanical force needed to elicit a cellular response.
Interestingly, this process of altered mechanical threshold has been shown in tenocytes, where the amount of
strain needed to cause a change in gene expression was
increased when tenocytes were initially treated with
tension deprivation (40). If the AF cells derived from
degenerated IVDs had altered their mechanical force
threshold due to loss of tension in their native environment, then 10% strain, a physiologic magnitude capable
of reducing matrix catabolism in nondegenerated AF
cells, could potentially be interpreted as a nonphysiologic understrain, leading to decreased matrix anabolism.
In conclusion, our data are the first to show that
the response of human AF cells to CTS, in terms of
regulation of matrix protein and matrix-degrading enzyme gene expression, is frequency-dependent. This
RESPONSE OF HUMAN AF CELLS TO CYCLIC TENSILE STRAIN
finding supports the concept that physiologic mechanical
loads are important for matrix homeostasis and promotion of matrix anabolism in healthy disc cells. The
frequency of load has been shown to be an important
factor in determining cell metabolic activity, and our
data suggest that frequencies below physiologically “normal” levels could lead to degradation of the IVD matrix.
Interestingly, AF cells derived from degenerated tissue
respond differently to CTS as compared with those
derived from nondegenerated tissue. Specifically, our
data suggest that physiologic mechanical loads could be
detrimental to disc cell matrix homeostasis in degenerated IVDs, potentially leading to the progression of
degenerative disc disease. Although this study is limited
by the in vitro nature of the loading culture system,
removing the AF cells from their native environment
excludes many uncharacterized extrinsic factors, allowing the effect of cell deformation to be considered in
isolation. Such a system will allow further investigation
of the response of IVD cells to CTS and will help
elucidate the mechanotransduction pathways operating
in cells derived from nondegenerated and degenerated
IVDs.
AUTHOR CONTRIBUTIONS
All authors were involved in drafting the article or revising it
critically for important intellectual content, and all authors approved
the final version to be published. Dr. Millward-Sadler had full access to
all of the data in the study and takes responsibility for the integrity of
the data and the accuracy of the data analysis.
Study conception and design. Gilbert, Hoyland, Millward-Sadler.
Acquisition of data. Gilbert.
Analysis and interpretation of data. Gilbert, Hoyland, MillwardSadler.
REFERENCES
1. Adams MA, McNally DS, Dolan P. ‘Stress’ distributions inside
intervertebral discs: the effects of age and degeneration. J Bone
Joint Surg Br 1996;78:965–72.
2. Shirazi-Adl SA, Shrivastava SC, Ahmed AM. Stress analysis of the
lumbar disc-body unit in compression: a three-dimensional nonlinear finite element study. Spine 1984;9:120–34.
3. Wilke HJ, Neef P, Caimi M, Hoogland T, Claes LE. New in vivo
measurements of pressures in the intervertebral disc in daily life.
Spine 1999;24:755–62.
4. O’Connell GD, Johannessen W, Vresilovic EJ, Elliott DM. Human internal disc strains in axial compression measured noninvasively using magnetic resonance imaging. Spine 2007;32:2860–8.
5. Wenger KH, Woods JA, Holecek A, Eckstein EC, Robertson JT,
Hasty KA. Matrix remodeling expression in anulus cells subjected
to increased compressive load. Spine 2005;30:1122–6.
6. Kasra M, Goel V, Martin J, Wang ST, Choi W, Buckwalter J.
Effect of dynamic hydrostatic pressure on rabbit intervertebral disc
cells. J Orthop Res 2003;21:597–603.
7. MacLean JJ, Lee CR, Alini M, Iatridis JC. Anabolic and catabolic
mRNA levels of the intervertebral disc vary with the magnitude
3393
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
and frequency of in vivo dynamic compression. J Orthop Res
2004;22:1193–200.
Hutton WC, Elmer WA, Boden SD, Hyon S, Toribatake Y,
Tomita K, et al. The effect of hydrostatic pressure on intervertebral disc metabolism. Spine 1999;24:1507–15.
Wang DL, Jiang SD, Dai LY. Biologic response of the intervertebral disc to static and dynamic compression in vitro. Spine
2007;32:2521–8.
Wuertz K, Godburn K, MacLean JJ, Barbir A, Stinnett Donnelly
J, Roughley PJ, et al. In vivo remodeling of intervertebral discs in
response to short- and long-term dynamic compression. J Orthop
Res 2009;27:1235–42.
Matsumoto T, Kawakami M, Kuribayashi K, Takenaka T, Tamaki
T. Cyclic mechanical stretch stress increases the growth rate and
collagen synthesis of nucleus pulposus cells in vitro. Spine 1999;
24:315–9.
Hutton WC, Elmer WA, Boden SD, Hyon S, Toribatake Y,
Tomita K, et al. The effect of hydrostatic pressure on intervertebral disc metabolism. Spine 1999;24:1507–15.
MacLean JJ, Lee CR, Grad S, Ito K, Alini M, Iatridis JC. Effects
of immobilization and dynamic compression on intervertebral disc
cell gene expression in vivo. Spine 2003;28:973–81.
Court C, Colliou OK, Chin JR, Liebenberg E, Bradford DS,
Lotz JC. The effect of static in vivo bending on the murine
intervertebral disc. Spine J 2001;1:239–45.
Neidlinger-Wilke C, Wurtz K, Urban JP, Borm W, Arand M,
Ignatius A, et al. Regulation of gene expression in intervertebral
disc cells by low and high hydrostatic pressure. Eur Spine J 2006;15
Suppl 3:S372–8.
Lotz JC, Chin JR. Intervertebral disc cell death is dependent on
the magnitude and duration of spinal loading. Spine 2000;25:
1477–83.
MacLean JJ, Lee CR, Alini M, Iatridis JC. The effects of
short-term load duration on anabolic and catabolic gene expression in the rat tail intervertebral disc. J Orthop Res 2005;23:
1120–7.
Kasra M, Merryman WD, Loveless KN, Goel VK, Martin JD,
Buckwalter JA. Frequency response of pig intervertebral disc cells
subjected to dynamic hydrostatic pressure. J Orthop Res 2006;24:
1967–73.
Hutton WC, Elmer WA, Bryce LM, Kozlowska EE, Boden SD,
Kozlowski M. Do the intervertebral disc cells respond to different
levels of hydrostatic pressure? Clin Biomech (Bristol, Avon)
2001;16:728–34.
Korecki CL, Kuo CK, Tuan RS, Iatridis JC. Intervertebral disc cell
response to dynamic compression is age and frequency dependent.
J Orthop Res 2009;27:800–6.
Ishihara H, McNally DS, Urban JP, Hall AC. Effects of hydrostatic
pressure on matrix synthesis in different regions of the intervertebral disk. J Appl Physiol 1996;80:839–46.
Alini M, Eisenstein SM, Ito K, Little C, Kettler AA, Masuda K, et
al. Are animal models useful for studying human disc disorders/
degeneration? Eur Spine J 2008;17:2–19.
Handa T, Ishihara H, Ohshima H, Osada R, Tsuji H, Obata K.
Effects of hydrostatic pressure on matrix synthesis and matrix
metalloproteinase production in the human lumbar intervertebral
disc. Spine 1997;22:1085–91.
Le Maitre CL, Frain J, Fotheringham AP, Freemont AJ, Hoyland
JA. Human cells derived from degenerate intervertebral discs
respond differently to those derived from non-degenerate intervertebral discs following application of dynamic hydrostatic pressure. Biorheology 2008;45:563–75.
Wuertz K, Urban JP, Klasen J, Ignatius A, Wilke HJ, Claes L, et
al. Influence of extracellular osmolarity and mechanical stimulation on gene expression of intervertebral disc cells. J Orthop Res
2007;25:1513–22.
3394
26. Sowa G, Agarwal S. Cyclic tensile stress exerts a protective effect on
intervertebral disc cells. Am J Phys Med Rehabil 2008;87:537–44.
27. Rannou F, Richette P, Benallaoua M, Francois M, Genries V,
Korwin-Zmijowska C, et al. Cyclic tensile stretch modulates
proteoglycan production by intervertebral disc annulus fibrosus
cells through production of nitrite oxide. J Cell Biochem 2003;90:
148–57.
28. Sive JI, Baird P, Jeziorsk M, Watkins A, Hoyland JA, Freemont
AJ. Expression of chondrocyte markers by cells of normal and
degenerate intervertebral discs. Mol Pathol 2002;55:91–7.
29. Vande Geest JP, Di Martino ES, Vorp DA. An analysis of the
complete strain field within Flexercell membranes. J Biomech
2004;37:1923–8.
30. Liu B, Qu MJ, Qin KR, Li H, Li ZK, Shen BR, et al. Role of cyclic
strain frequency in regulating the alignment of vascular smooth
muscle cells in vitro. Biophys J 2008;94:1497–507.
31. Winter DA. Biomechanical motor patterns in normal walking. J
Mot Behav 1983;15:302–30.
32. Le Maitre CL, Freemont AJ, Hoyland JA. The role of interleukin-1 in the pathogenesis of human intervertebral disc degeneration. Arthritis Res Ther 2005;7:R732–45.
33. Livak KJ, Schmittgen TD. Analysis of relative gene expression
data using real-time quantitative PCR and the 2⫺⌬⌬CT method.
Methods 2001;25:402–8.
34. Salter DM, Robb JE, Wright MO. Electrophysiological responses
of human bone cells to mechanical stimulation: evidence for
GILBERT ET AL
35.
36.
37.
38.
39.
40.
specific integrin function in mechanotransduction. J Bone Miner
Res 1997;12:1133–41.
Le Maitre CL, Frain J, Millward-Sadler J, Fotheringham AP,
Freemont AJ, Hoyland JA. Altered integrin mechanotransduction
in human nucleus pulposus cells derived from degenerated discs.
Arthritis Rheum 2009;60:460–9.
Stevens RL, Ryvar R, Robertson WR, O’Brien JP, Beard HK.
Biological changes in the annulus fibrosus in patients with lowback pain. Spine 1982;7:223–33.
Scott JE, Bosworth TR, Cribb AM, Taylor JR. The chemical
morphology of age-related changes in human intervertebral disc
glycosaminoglycans from cervical, thoracic and lumbar nucleus
pulposus and annulus fibrosus. J Anat 1994;184:73–82.
Iatridis JC, Setton LA, Weidenbaum M, Mow VC. Alterations
in the mechanical behavior of the human lumbar nucleus
pulposus with degeneration and aging. J Orthop Res 1997;15:
318–22.
Iatridis JC, Setton LA, Foster RJ, Rawlins BA, Weidenbaum M,
Mow VC. Degeneration affects the anisotropic and nonlinear
behaviors of human anulus fibrosus in compression. J Biomech
1998;31:535–44.
Arnoczky SP, Lavagnino M, Egerbacher M, Caballero O, Gardner
K, Shender MA. Loss of homeostatic strain alters mechanostat
“set point” of tendon cells in vitro. Clin Orthop Relat Res
2008;466:1583–91.
Документ
Категория
Без категории
Просмотров
0
Размер файла
289 Кб
Теги
cyclic, strait, degeneration, anulus, frequency, disco, tensile, alteren, fibrosus, human, cells, response, dependence
1/--страниц
Пожаловаться на содержимое документа