INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

ARTICULAR CARTILAGE OF THE major load-bearing joints is subjected to repetitive mechanical loading during normal activity, with a significant component applied perpendicularly to the articular surface (1). The mechanical environment to which the chondrocytes are exposed is therefore an important factor affecting the health and function of the diarthrodial joint and influences cell metabolism through a process termed mechanotransduction (14, 23, 34-37, 39). Compressive loading of cartilage results in cell deformation, hydrostatic pressure gradients, fluid flow, and deformation of the charged extracellular matrix with associated changes in osmolarity and pH (19, 42). These extracellular events may influence cell metabolism through the activation of specific intracellular events including nucleus deformation, modulation of the cytoskeleton, and alterations in intracellular concentrations of nitric oxide and Ca²⁺ (17, 18, 22, 26, 29).

Intracellular Ca²⁺ represents an ubiquitous signaling mechanism and is involved in processes as diverse as cell division and apoptosis (7). Cells utilize two sources of Ca²⁺ for the generation of signals: Ca²⁺ released from the intracellular stores and activation of membrane ion channels resulting in Ca²⁺ entry. Ca²⁺ signals can be modulated in space and time for the activation of specific genes (6). Intracellular Ca²⁺ signaling in chondrocytes has previously been investigated in response to both chemical agents and physical stimuli (18, 27).

Particular attention has focused on the role of intracellular Ca²⁺ in early mechanotransduction events. Ca²⁺ signaling in isolated chondrocytes has been shown to be activated by various potential mechanotransduction events, which include fluid flow (46, 47), membrane deformation resulting from "cell poking" (17), hydrostatic pressure (18, 43), and osmotic challenge (20). However, the specific mechanisms of Ca²⁺ mobilization in response to mechanical loading remain unclear. Results vary with different loading regimes and cell type, illustrating varying degrees of involvement from both the intracellular stores and Ca²⁺ influx through voltage-operated Ca²⁺ channels and/or stretch-activated cation (SA-cat) channels (17, 40, 47). Furthermore, previous studies investigating Ca²⁺-mediated mechanotransduction events involved monolayer culture systems, which induce modulation of chondrocytes to a fibroblast-like phenotype with associated cytoskeletal changes (22).

Previous studies by the authors and other groups have utilized a well-characterized and reproducible model system, consisting of chondrocytes isolated from articular cartilage and embedded in agarose constructs (2, 10, 28). The isolated chondrocytes adopt a rounded morphology that is associated with the retention of cytoskeletal organization and the maintenance of chondrocyte phenotype (5). The application of compressive strains to the chondrocyte-agarose constructs produces cell deformation to an oblate ellipsoid morphology (24, 26). The chondrocyte-agarose system thereby provides a more physiological model than monolayer systems for the study of Ca²⁺-mediated compression-induced mechanotransduction events associated with cell deformation. The present study tests the hypothesis that compression-induced chondrocyte deformation influences intracellular Ca²⁺ concentration, thereby suggesting Ca²⁺ signaling as a mediator of mechanotransduction events.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Preparation of chondrocyte-agarose constructs. Full-depth slices of cartilage were removed from the metacarpophalangeal joints of 18-mo-old steers. The cartilage slices were diced finely and incubated at 37°C on rollers for 1 h in DMEM supplemented with 20% (vol/vol) FCS (DMEM + 20% FCS; GIBCO, Paisley, UK) and 700 U/ml pronase (BDH, Poole, UK). Other medium additives were penicillin/streptomycin (5 mg/ml), L-glutamine (2 µM), HEPES (20 mM), and L-ascorbic acid (0.85 mM), all supplied by GIBCO. Subsequently, the minced tissue was incubated at 37°C in DMEM + 20% FCS + 100 U/ml collagenase type XI (Sigma Chemical, Poole, UK) for 16 h on rollers. The supernatant, containing released chondrocytes, was passed through a 70-µm-pore-size sieve (Falcon, Oxford, UK), washed twice in DMEM + 20% FCS, and resuspended at 2 × 10⁷ cells/ml. The chondrocyte suspension was added to an equal volume of 8% (wt/vol) agarose (type IX-A; Sigma Chemical) in Earle's balanced salt solution (EBSS; GIBCO) to give a final concentration of 1 × 10⁷ cells/ml in 4% agarose. The chondrocyte-agarose suspension was plated in specially designed Perspex molds and allowed to gel at 4°C for 30 min to produce rectangular constructs with dimensions of 4 × 3 × 3 mm. The constructs were maintained in DMEM + 20% FCS at 37°C-5% CO₂ for up to 72 h, and the culture medium was changed after 48 h.

Calcium imaging and quantitative analysis. After 1 and 3 days in culture, constructs were incubated for 30 min at 37°C in a 10 µM solution of indo 1-AM (Cambridge Bioscience, Cambridge, UK) prepared in EBSS containing 1.8 mM Ca²⁺ and supplemented with 20 mM HEPES at pH 7.4. The constructs were washed two times in EBSS and subsequently incubated in EBSS for a further 15 min at 37°C. Constructs were placed within a specially designed compression rig (Fig. 1), as detailed in a previous study (26). The unstrained constructs were bathed in EBSS maintained at 37°C and were mounted on the stage of an inverted microscope (Nikon, Kingston-upon-Thames, UK) associated with an Oz real time confocal laser scanning microscope (Noran Instruments, Bicester, UK). Cells were visualized by use of a ×40 oil immersion objective, numerical aperture 1.30 (Nikon, Kingston-upon-Thames, UK). Laser excitation light was provided at a wavelength of 364 nm. Dual emissions were collected simultaneously at 385-425 nm representing indo 1 bound to intracellular Ca²⁺ and 475-515 nm representing indo 1 free of Ca²⁺. Images (512 × 480 pixels) were recorded with a dwell time of 200 ns/pixel and ×4 jump averaging. A zoom factor of 1 was adopted, yielding a pixel size of 0.2 µm. Pixel intensities were measured within the range 0-255, the latter value representing saturation. Photodetector brightness settings were adjusted accordingly for each individual cell and maintained for that cell throughout the experiment.

View larger version (29K): [in this window] [in a new window]   Fig. 1.   Schematic cross section of the test rig mounted on the stage of an inverted microscope, enabling visualization of viable chondrocytes within unstrained and compressed agarose constructs.

Ca²⁺ measurement in unstrained and strained constructs. Individual chondrocytes were located by use of transmitted light microscopy within a plane ~50 µm from the bottom surface of the construct and approximately equidistant between the two loading plungers. Basal Ca²⁺ ratio measurements were calculated from single-ratio images obtained for a total of 170 cells in unstrained agarose constructs at days 1 and 3 of culture.

View larger version (72K): [in this window] [in a new window]   Fig. 2.   a: Confocal time series of an isolated chondrocyte at day 1 labeled with indo 1-AM and visualized within an unstrained agarose construct (top row) and again after the application of a 20% static compression (other seven rows). Every third image is displayed from the full series of 140 images acquired, resulting in an interval of 9 s between images. Scale bar = 5 µm. The concentration of intracellular Ca2+, indicated by a ratio value between 0 and 4, is displayed on a pseudocolor scale as shown on the right hand color bar. Cell X and Y diameters have been indicated. b: Resulting temporal variation in Ca2+ ratio averaged over the whole cell. Compressive strain applied at time = 0 s. Value A is taken as the mean ratio value over a 2-min baseline. Value B is the peak ratio value, with the time taken to reach value B indicated as tB.

(1)

Mechanisms of Ca²⁺ mobilization. To investigate the mechanisms of Ca²⁺ mobilization, two reagents that are known to block Ca²⁺ mobilization pathways were utilized. To examine whether Ca²⁺ pathways involved transport from extracellular Ca²⁺, experiments were performed using constructs maintained in standard EBSS supplemented with 10 mM EGTA (Sigma Chemical). To determine the potential role of mechanosensitive ion channels in deformation-induced Ca²⁺ mobilization, constructs were incubated in EBSS containing 20 µM Gd³⁺ (Sigma Chemical). Constructs were incubated with either EGTA or Gd³⁺ for 1 h before examination and were maintained in solution throughout experiments. Populations of cells (n > 30) in treated and untreated groups were examined in unstrained and strained constructs.

Data analysis. For each cell, a mean Ca²⁺ ratio value (A) was calculated from the initial 2-min unstrained baseline period. A maximum Ca²⁺ ratio value (B) was taken from the subsequent 5-min test period and compared with the mean baseline value (A) to produce a percent increase, termed amplitude of response, as follows

(2)

Throughout the study, a value of amplitude of response above a threshold of 55% was defined as a positive Ca²⁺ response. Cells demonstrating spontaneous transients during baseline measurements were rejected (approximately <1% of total cell population). The time taken to reach a maximum Ca²⁺ ratio calculated from the application of compression was also recorded (see Fig. 2).

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

The mean unstrained diameter ratios at day 1 and day 3 were similar in magnitude with mean ± SD values of 0.97 ± 0.04 and 0.98 ± 0.07, respectively. In constructs compressed at days 1 and 3, there was a statistically significant (P < 0.05) decrease in cell diameter ratios to 0.71 ± 0.07 and 0.75 ± 0.06, respectively.

Mean basal Ca²⁺ ratios were 1.05 ± 0.29 and 1.07 ± 0.27 on days 1 and 3, respectively; values that were not significantly different. A representative time series of images revealing a change in intracellular Ca²⁺ in a single chondrocyte, initially unstrained and then subjected to 20% strain, is presented in Fig. 2a. The corresponding value of Ca²⁺ ratio with time is presented in Fig. 2b. The frequency distribution of amplitude of response in unstrained and strained cells on days 1 and 3 is displayed in Fig. 3. It can be seen that, on day 1, 34% of deformed cells showed a positive Ca²⁺ response, indicated above the 55% threshold value, compared with 7% of unstrained cells (Fig. 3B). The corresponding values for day 3 were 24 and 9%, respectively (Fig. 3D). There was a statistically significantly greater number of positive Ca²⁺ responses in cells subjected to compression compared with those in unstrained cells on both culture days (P < 0.01 for day 1; P < 0.05 for day 3). However, the difference in the number of positive Ca²⁺ responses from day 1 to day 3 in strained chondrocytes was not statistically significant.

View larger version (30K): [in this window] [in a new window]   Fig. 3.   Frequency distributions of amplitude of response values measured in unstrained (A and C) and strained cells (B and D) on day 1 (A and B) and day 3 (C and D). The percentage of cells above and below the 55% threshold value (vertical line), used to discriminate between positive responders and nonresponders, is shown. Median amplitude values for nonresponders and positive responders are indicated for all groups.

The mean times to reach a maximum ratio during a positive Ca²⁺ response in compressed cells were 225 ± 95 and 201 ± 88 s at day 1 and day 3, respectively. Median amplitudes of response values for nonresponders and positive responders in unstrained and strained groups are also indicated in Fig. 3. Examination of the data revealed no significant correlation between the amplitude of response of cells exhibiting a positive response and degree of cell deformation (r = 0.22).

In a separate series of experiments, the individual effects of the two blockers on the proportion of cells exhibiting a positive Ca²⁺ response were investigated (Fig. 4). In the presence of EGTA, 12% of strained cells yielded a positive Ca²⁺ response compared with 0% of unstrained cells, a difference that was statistically significant (P < 0.05). Comparison with strained cells in the absence of EGTA revealed a statistically significant difference (P < 0.05). It should be noted, however, that there were also statistical differences between unstrained cells in the presence or absence of EGTA, making intergroup comparisons difficult (P < 0.05). The presence of Gd³⁺ significantly reduced the number of positive Ca²⁺ responses in strained cells (3%) compared with untreated strained cells (28%) (P < 0.01). There was no significant difference between unstrained and strained cells treated with Gd³⁺, suggesting that Gd³⁺ completely abolished the Ca²⁺ response observed after compression. In addition, unstrained cells in untreated and Gd³⁺ groups showed no significant difference in the number of positive Ca²⁺ responses.

View larger version (13K): [in this window] [in a new window]   Fig. 4.   The percentage of cells on day 1 exhibiting a positive Ca2+ response in unstrained and strained agarose constructs maintained in Earle's balanced salt solution (EBSS) or EBSS supplemented with either EGTA (10 mM) or Gd3+ (20 µM). 2 analysis examined levels of difference between unstrained and strained data for each of the 3 test conditions (not significant at P > 0.05, *P < 0.05, **P < 0.01). Similar analysis was performed between strained data in the supplemented groups compared with the untreated group (+P < 0.05, ++P < 0.01). Number of cells (n) examined at each condition is stated.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

The present study has examined the effects of compressive strain on intracellular Ca²⁺ concentration in chondrocytes seeded in agarose and cultured in the unstrained state for up to 3 days. In addition, the mechanisms of the Ca²⁺ response were investigated on day 1 of culture.

Intracellular Ca²⁺ was measured using a single-excitation, dual-emission ratiometric fluorescent indicator, indo 1, allowing its accurate measurement independent of factors affecting image brightness, such as photobleaching, dye concentration, and specimen path length (15). In this study, intracellular Ca²⁺ concentrations were expressed by using ratio values. It was necessary to establish a well-defined quantitative parameter that would reflect a positive Ca²⁺ response. A percent increase in Ca²⁺ ratio >55% of the mean baseline was selected to define a positive Ca²⁺ response (Fig. 2). Separate studies determined that this threshold value reduced the probability of combined false positive or negative classification to <3% (data not shown).

In the present study, a 20% compressive strain was applied at a rate of 5%/s, which permitted tracking of individual cells during both unstrained and strained periods. Gross 20% compression of cell/agarose constructs at days 1 and 3 resulted in significant changes in cell diameter ratio (X/Y) to mean values of 0.73 and 0.75, respectively. These values are equivalent to reductions in X diameters of ~20%, which lie in the 0-30% range for cell strain in intact cartilage subjected to physiological loads (9, 16, 24, 25). Previous studies within the group have shown that chondrocyte deformation in compressed agarose constructs is also associated with distortion of intracellular organelles and unraveling of the cell membrane over the entire cell surface (30). At both days 1 and 3, compression significantly increased the percentage of cells exhibiting positive Ca²⁺ response compared with the unstrained population, although the exact percentage of responders may be influenced by factors related to the cell isolation procedures. It should be noted that the percentage of positive Ca²⁺ responders in compressed agarose constructs is less than in previous studies in which cells cultured in monolayer were deformed by use of a micropipette (11, 12, 17). This cell-poking technique leads to substantial localized membrane distortion, which may explain the difference between the results obtained using the two methods. Moreover, the response may be attributed to differences in actin organization, known to play a role in the activation of SA-cat channels (38), between cells cultured in monolayer and agarose (4, 22).

After a delay of ~200 s, the general behavior of the positive Ca²⁺ response was similar in the majority of cells, whereby intracellular Ca²⁺ rose rapidly to a single transient peak with subsequent slower decay to approximately prestrain baseline levels (Fig. 2b). Before a positive response, localized areas of elevated Ca²⁺ concentration were observed around the cell periphery, similar to that previously described as "hot spots" (8). This may indicate the initiation of localized Ca²⁺ events that are upstream of the global Ca²⁺ response and may indeed contribute to the observed delay. It is interesting to note that a small percentage of chondrocytes in unstrained constructs displayed spontaneous and transient increases in intracellular Ca²⁺ as previously reported for unstimulated chondrocytes and other cell types (21, 31, 47). Thus compression-induced cell deformation may increase the likelihood of initiation of preexisting spontaneous transients rather than initiation of a unique transduction pathway.

The delay before reaching a maximum ratio value was highly repeatable, with a mean delay time >200 s. This suggests that the Ca²⁺ transient does not represent the initial intracellular mechanotransduction signal but is preceded by mechanically triggered upstream events that may include the activation of nonselective SA-cat channels, stretch-activated K⁺ channels, epithelial sodium channels, or changes in membrane potential, which all may play roles in cartilage physiology (13, 32, 33, 41, 44). Further studies are necessary to elucidate these upstream signaling mechanisms.

Initial studies were performed using EGTA and Gd³⁺ to investigate the mechanisms of the Ca²⁺ response. The eradication of the response with EGTA and Gd³⁺ was similar to that reported in other studies (3, 12, 17, 20, 44, 46, 47). EGTA acts to chelate extracellular Ca²⁺, but prolonged exposure may also lead to depletion of the intracellular Ca²⁺ stores. Treatment with EGTA significantly reduced the percentage of cells exhibiting a positive Ca²⁺ response in both unstrained and strained constructs. This suggests that the positive Ca²⁺ response is associated with Ca²⁺ influx and/or the release of intracellular stores. Gd³⁺ is known to block SA-cat channels as well as voltage-operated Ca²⁺ channels and the Na⁺/Ca²⁺ exchanger (38, 45, 48). Treatment with Gd³⁺ significantly reduced the number of positive responses in both unstrained and strained constructs, suggesting that the Ca²⁺ response to compression is mediated through voltage-gated or SA-cat channels. A previous study has reported that Gd³⁺ also significantly reduced pressure-induced upregulation of proteoglycan synthesis in isolated chondrocytes (3). This further supports the importance of the Gd³⁺-sensitive pathway in the transduction of mechanical signals into downstream alterations in cell metabolism.

In summary, the present study demonstrates for the first time that the application of physiological levels of compressive strain to chondrocytes seeded within three-dimensional agarose constructs results in intracellular Ca²⁺ signaling, which could potentially act within a series of mechanotransduction pathways. The Ca²⁺ response investigated in the present study appears to depend on the activation of a Gd³⁺-sensitive pathway. The delay of the response after cell deformation suggests the involvement of upstream signaling events. These results constitute an important stage in the understanding of the physiological role of intracellular Ca²⁺ in chondrocytes subjected to a mechanical environment.