Histological assessments on the abnormalities of mouse epiphyseal ...

Biomedical Research 28 (4) 191-203, 2007

Histological assessments on the abnormalities of mouse epiphyseal chondrocytes with short term centrifugal loading Paulo Henrique Luiz de FREITAS1, 5, Taku KOJIMA2, 5, Sobhan UBAIDUS1, 5, Minqi LI5, Guangwei SHANG5, 6, Ritsuo TAKAGI1, Takeyasu MAEDA3, 5, Kimimitsu ODA4, 5, Hidehiro OZAWA7 and Norio AMIZUKA5

Divisions of 1 Oral and Maxillofacial Surgery, 2 Reconstructive Surgery for Oral and Maxillofacial Region, 3 Oral Anatomy and 4 Biochemistry, Niigata University Graduate School of Medical and Dental Sciences, Niigata, Japan; 5 Center for Transdisciplinary Research, Niigata University, Niigata, Japan; 6 Center of Health Care in Stomatology, Shanghai 10th Hospital, Tongji University, Shanghai, China; and 7 Institute of Oral Science Matsumoto Dental University, Shiojiri, Japan (Received 2 May 2007; and accepted 31 May 2007)

ABSTRACT We have examined the morphological changes in chondrocytes after exposure to experimental hypergravity. Tibial epiphyseal cartilages of 17-days-old mouse fetuses were exposed to centrifugation at 3G for 16 h mimicking hypergravitational environment (experimental group), or subjected to stationary cultures (control group). Centrifugation did not affect the sizes of epiphyseal cartilage, chondrocyte proliferation, type X collagen-positive hypertrophic zone, and the mRNA expressions of parathyroid hormone-related peptide and fibroblast growth factor receptor III. However, centrifuged chondrocytes showed abnormal morphology and aberrant spatial arrangements, resulting in disrupted chondrocytic columns. Through histochemical assessments, actin filaments were shown to distribute evenly along cell membranes of control proliferative chondrocytes, while chondrocytes subjected to centrifugal force developed a thicker layer of actin filaments. Transmission electron microscopic observations revealed spotty electron-dense materials underlying control chondrocytes’ cell membranes, while experimental chondrocytes showed their thick layer. In the intracolumnar regions of the control cartilage, longitudinal electron-dense fibrils were associated with short cytoplasmic processes of normal chondrocytes, indicating assumed cell-tomatrix interactions. These extracellular fibrils were disrupted in the centrifuged samples. Summarizing, altered actin filaments associated with cell membranes, irregular cell shape and disappearance of intracolumnar extracellular fibrils suggest that hypergravity disturbs cell-to-matrix interactions in our cartilage model.

The unique mechanical properties of the cartilage are conferred by its avascular composition and the complexity of its extracellular matrices including collagen fibrils, aggrecan, link proteins and so forth. Mesenchymal condensation, chondrocytic differentiAddress correspondence to: Norio Amizuka, DDS, PhD Center for Transdisciplinary Research, Niigata University, 2-5274 Gakkocho-dori, Chuo-ku, Niigata 951-8514, Japan Tel: +81-25-227-0900, Fax: +81-25-227-0901 E-mail: [email protected]

ation and subsequent matrix synthesis are genetically programmed, and regulated by microenvironmental influences such as soluble mediators, e.g., parathyroid hormone-related peptide (PTHrP) and fibroblast growth factor receptor III (FGFR3) (1, 2, 14). Currently, however, many investigators are paying attention to the fundamental role played by mechanical loading in chondrocyte regulation and cartilage dysfunction. Gravity, the force of attraction between all masses in the universe, can be understood as a source of loading upon chondrocytes and their surroundings. Altered gravity has been shown to affect

Freitas et al.

192

chondrocyte differentiation and endochondral bone formation (8, 9, 23, 24, 27). More recently, novel strategies of tissue engineering have utilized microgravity (12, 19) and hypergravity (17) as aids for optimizing cartilaginous tissue growth. Nowadays, mankind’s enterprises in space are no longer science fiction, but significant biological, psychological and environmental problems are yet to be addressed (25). The deleterious effects of spaceflights and microgravity upon the musculoskeletal system have been well-documented (13, 20), and hypergravity can be considered a valuable means of counteracting those (36). Cartilage formation is achieved by complex, multi-phased processes that start with mesenchymal condensation (7). Cells in such condensations differentiate into chondrocytes that undergo cell division and synthesize cartilage-specific extracellular matrices. This is followed by further unidirectional proliferation resulting in longitudinal columns of chondrocytes that will eventually exit the cell cycle by hypertrophying (1, 2, 5, 6). As a consequence, the epiphyseal cartilage of long bones shows distinct zones of resting (also referred to as reserving), proliferative and hypertrophic chondrocytes. Structurally, chondrocytic distribution, e.g., the well-arranged longitudinal columns of flattened chondrocytes in the proliferative zone, appears to be regulated and maintained, at least in part, by cell-to-matrix interactions. Physiological tensile or compressive forces may affect the spatial conformation network of the extracellular fibrils and other extracellular components. These strains may be sensed by chondrocytes, resulting in appropriate cellular responses. Depletion of cartilage collagen fibrils in mice carrying a dominant negative α1 (II) procollagen transgene led to an irregular organization of proliferative chondrocytes, which were unable to form the typical longitudinal columns (3). Furthermore, rapid, reversible responses of the chondrocyte cytoskeleton were demonstrated by Durrant et al. (10), who have reported a linkage between actin organization and matrix interactions via focal contacts. Therefore, it seems significant to examine the alterations in chondrocyte and extracellular matrix organization for evaluating the influence of mechanical loading in cartilage. Contemporary research on chondrocyte responses to mechanical loading has been obtained mainly from tissue and/or organ culture experiments, as these approaches provide a more controlled biomechanical environment than the in vivo experiments (32, 35). Rotation and centrifugation experiments using clinostats have often been used to mimic mi-

crogravity and hypergravity, respectively. Furthermore, the mouse embryonic limb has been employed as an excellent model to study morphogenesis and cytodifferentiation (11). In this study, we have modified the abovementioned mouse embryonic cartilage models to make them suitable for centrifugation experiments, mimicking hypergravity. We have observed the ultrastructural alterations of chondrocytes, extracellular matrices and cell-to-matrix association in order to explore the effects of mechanical loading in the growing cartilage. MATERIALS AND METHODS Tissue preparation for the epiphyseal cartilage experiment. Seventeen days-old fetuses were extracted from pregnant female ICR mice (CLEA Japan, Inc., Tokyo) through Caesarian dissection, under anesthesia with diethyl ether, being sacrificed by dislocation shortly thereafter. The tibiae were removed and immediately soaked in α minimal essential medium (αMEM; Flow Laboratories, Irvine, Scotland) containing 10% fetal calf serum (FCS; Cosmo Bio. Ltd., Tokyo, Japan). In order to adjust the direction of the centrifugal force so that it would parallel the longitudinal axis of the specimens, several tibiae were carefully arranged in 0.5 mL micro tubes (Modern Biology, Inc., Lafayette, IN). Tibial distal ends pointed to the bottom of the tubes, which contained αMEM with 10% FCS. These 0.5 mL tubes including the specimens were centrifuged at 3G (200 rpm, with a 0.5 mL tube-adaptor) by a micro hematocrit centrifuge 3220 (Kubota Co., Fujioka, Japan), which was kept in a CO2 incubator (NAPCO 5420-2; Thermo Fisher Scientific, Inc, MA) for 16 h. This procedure provided adequate centrifugal force to the proximal ends of the tibial epiphyseal cartilage while avoiding the reflecting forces from the bottom of the tubes. For the control group, extracted tibiae were cultured in the same fashion, but were not submitted to centrifugation. After loading, experimental and control tibiae were fixed with either 4% paraformaldehyde diluted in 0.1 M phosphate buffer (pH 7.4), or a mixture of 2% paraformaldehyde and 2.5% glutaraldehyde in 0.067 M phosphate buffer (pH 7.4). Stationary cultured specimens were fixed with the same solution. After decalcification with 5% EDTA-2Na solution for 5 days at 4°C, some specimens were soaked in 20% sucrose-containing PBS prior to embedding into OCT compound for cryostat sectioning (Micron, Heidelberg, Germany), and others were dehydrated

Changes on centrifuged cartilage

for paraffin embedding. Other samples were postfixed in a mixture of 1% osmium tetraoxide and 1.5% potassium ferrocyanide for 4 h, dehydrated with ascending concentrations of acetone, and embedded in epoxy resin (Taab, Berkshire, UK) prior to transmission electron microscope (TEM) analysis. Semithin sections of epoxy resin-embedded specimens were stained with toluidine blue. Ultrathin sections were prepared with an ultramicrotome (Sorvall MT-5000; Ivan Sorvall, Inc., Norwalk, CT) and stained with tannic acid, uranyl acetate and lead citrate prior to observation under TEM (Hitachi H-7100; Hitachi Co. Ltd., Tokyo, Japan) at 75 kV. Detection for actin filaments and microtubules on cryostat sections and type X collagen immunohistochemistry on paraffin sections. For histochemistry of actin filaments and microtubules, 10 μm-thick cryostat sections were prepared and pre-incubated with 1% bovine serum albumin in phosphate buffered saline (pH 7.4, 1% BSA-PBS) at room temperature, for 30 min. For actin detection, rhodamine-conjugated phalloidin (Molecular Probes, Inc., Eugene, OR) diluted 1 : 1000 with 1% BSA-PBS was applied to the sections overnight at 4°C prior to fluorescence observation (Eclipse E800; Nikon Ltd, Co, Tokyo, Japan). For microtubules immunodetection, mouse anti-β-tubulin (Abcam Inc., Cambridge, MA) at a dilution of 1 : 100 was applied to the sections, followed by incubation with FITC-conjugated antimouse IgG (Chemicon International, Inc., Temecula, CA) for immunofluorescence observation. Control sections incubated only with the secondary antibody did not show nonspecific binding (data not show). For type X collagen immunohistochemistry, dewaxed paraffin sections were pretreated with 0.5% hyaluronidase (Wako Pure Chemical Industry, Ltd, Osaka, Japan) in PBS for 15 min at room temperature, and then treated with 1% BSA-PBS. They were subsequently incubated with a rabbit antiserum to type X collagen (LSL Ltd, Tokyo, Japan) diluted at 1 : 100, overnight, and then to horseradish peroxidase (HRP)-conjugated anti-rabbit IgG (Calbiochem, San Diego, CA). The reaction was visualized using diaminobenzidine, (DAB; Dojindo Laboratories, Kumamoto, Japan) as a substrate. Reverse transcription-polymerase chain reaction (RTPCR) for PTHrP, FGFR3 and GAPDH transcripts. Total RNA from the age-matched mouse epiphyseal cartilage of stationary culture and centrifuged specimens were obtained by Trizol reagent (Invitrogen BV, Groningen, The Netherlands), after dissecting

193

the tibial chondro-osseous junction. In addition, freshly-obtained epiphyseal cartilage was used for control of our organ culture system. For each specimen, 5 μg of total RNA were subjected to reverse transcription using Superscript II (Invitrogen). PCR reactions (50 μL) contained 5 μL of 10 × reaction buffer, 1 μL of 25 pmol of each primer, 1 μL of 10 mM of dNTPs mixture, 1 U of Taq polymerase (Invitrogen), 3 μL of 25 mM MgCl2 and 2 μL of RTDNA and sterile distilled water. Amplification was performed by denaturation at 94°C for 1 min, annealing at 55°C for 1 min, and extension at 72°C for 1 min using GeneAmp PCR System 9700 (Applied Biosystem Japan Ltd, Tokyo, Japan). The expression of PTHrP and FGFR3 was examined at amplification cycles that do not reach the plateau for PCR reaction (30). Primer construction: rat/mouse PTHrP: sense 5’-TGGTGTTCCTGCTCAGCTA-3’, antisense 5’-CCTCGTCGTCTGACCCAAA-3 (38), rat/mouse FGFR3: sense 5’-AGAAGGCTGCTTTG GACA-3’, antisense 5’-TGAGCTGTTCCTCTGG CA-3’ (16), rat/mouse GAPDH; sense 5’-CATGG AGAAGGCTGGGGCTC-3’, antisense 5’-AACGG ATACATTGGGGGTAG-3’ (33). Quantification of proliferating cell nuclear antigen (PCNA)-positive chondrocytes. Control and experimental dewaxed paraffin sections (n = 10 each) were incubated with a mouse antibody to PCNA (a 36 kD auxiliary protein of DNA polymerase δ; Oncogene Research Products, San Diego, CA), as reported in previous work (2). No additional treatment for blocking non-specific binding was performed. They were then incubated with HRP-conjugated antimouse IgG (Chemicon) diluted at 1 : 100 with 1% BSA-PBS. Visualization of immunostaining was performed with DAB solution. Control sections incubated with only HRP-conjugated secondary antibody did not reveal nonspecific binding (data not show). The number of PCNA-positive chondrocytes were calculated, and then divided by the total number of chondrocytes in the epiphyseal cartilages, with the statistical results represented by the mean ± SD. Statistically significant differences were determined by Student’s t-test. RESULTS Overall appearance of the centrifuged epiphyseal cartilage When analyzed under light microscope, the epiphyseal cartilages were similar in size for both centrifuged and control specimens. Immunohistochemistry

Freitas et al.

194

for type X collagen, a specific marker for the hypertrophic zone, revealed that the distribution pattern of this protein was unaltered in the experimental group (Compare Figs. 1A and 1B). As shown in Fig. 1C, when determined by RT-PCR at cycles before reaching plateau (30), PTHrP and FGFR3, two important molecules regulating chondrocyte proliferation and

differentiation, were similarly expressed in the stationary culture, the centrifugation and the freshlyprepared samples. In addition, statistical analysis for PCNA-positive chondrocytes showed no significant difference in chondrocytic proliferation between the control and experimental groups (Table 1).

Table 1 Statistical analysis for PCNA-positive chondrocytes between the stationary cultured and centrifuged epiphyseal cartilage The percentage of PCNA-positive cells

Stationary culture (control) 10.03 ± 1.79

Centrifuged culture (experiment) 8.74 ± 2.05*

The percentage of PCNA-positive chondrocytes was statistically analyzed (See M&M). All values are expressed as mean ± SD. Significant differences were determined using the Student’s t-test. *NS.

Fig. 1 Immunohistochemistry for type X collagen and RT-PCR for PTHrP and FGFR3 expressions in epiphyseal cartilage. Control (A) and experiment (B) specimens show epiphyseal cartilages of the similar size and with a similar pattern of immunopositivity for type X collagen (brown). In panel C, the PCR products representing GAPDH, PTHrP and FGFR3 are similar among control (Co: freshly-prepared specimen), stationary cultured (S) and centrifuged (Ce) samples. Bars, 0.5 mm


Histological and ultrastructural findings in control and centrifuged chondrocytes When observing each zone of the epiphyseal cartilage, the resting zone of the control specimens featured round, smooth-surfaced cells that were dispersed in the extracellular matrix, while keeping certain distance among them (Fig. 2A). In contrast, the same zone in the experiment group showed increased cell density, with rough-surfaced chondrocytes that sometimes contacted neighboring cells (Fig. 2D). In the control proliferative zone, flattened chondrocytes were evenly distributed in the longitudinal columns like stacked coins (Fig. 2B). Chondrocytes in the experimental proliferative zone were more clustered and the chondrocytic columns were disorganized (Fig. 2E). Interestingly, resting and proliferative zones of the control specimens featured intense toluidine blue-positive fibrils among chondrocytes (Figs. 2A, 2B), whereas in the experimental group such fibrils were not discernible (Figs. 2D, 2E). In addition, centrifuged hypertrophic chondrocytes were irregular in shape (Fig. 2F), while control chondrocytes were all translucent and enlarged (Fig. 2C). TEM imaging unveiled fine, electron-dense extracellular fibrillar structures between the chondrocytes in both the resting and proliferative zones in control samples (Figs. 3A, 3B). At a higher magnification, these fibrillar structures comprised a twisted fine meshwork of cartilaginous fibrils (Figs. 3C, 3D), associating with small cytoplasmic processes of neighboring chondrocytes (Fig. 3D). In contrast, the intercolumnar region displayed a uniform meshwork of extracellular collagen fibrils, most of which ran straight in radial directions (Fig. 3E). Close to the hypertrophic zone, stout fibrils running parallel to the longitudinal axis of long bones appeared in the intercolumnar region (Fig. 3F). Compared with the control samples, the intracolumnar electron-dense fibrillar structures were not clearly visible in the experimental group (Compare Figs. 3A, 3B and 4A, 4B). Under higher magnification, consistent with disorganized chondrocyte distribution, sinuous extracellular fibrils were unevenly distributed among chondrocytes; some extracellular fibrils were loosely packed with remarkably reduced density (Figs. 4C, 4E), while others had a higher density (Fig. 4D). Despite the disorganized assembly of extracellular matrices, intracolumnar electron-dense fibrils were, in part, faintly discernible (Figs. 4C, 4D). In the former intercolumnar region, wavy collagen fibrils were unevenly distributed (Fig. 4F). When observing chondrocytes’ contact,

195

cytoplasmic electron-dense materials had underlain the cell membranes (Fig. 4G). Control hypertrophic chondrocytes featured translucent cytoplasm and nuclei with dispersed cell organelles throughout the cytoplasm (Fig. 5A). In centrifuged samples, however, hypertrophic cells were irregularly shaped, but maintained translucent appearance and organelle dispersion (Fig. 5C). Under higher magnification of control specimen, the extracellular matrix featured the stout fibrils running parallel to the cell membranes (Fig. 5B). These stout fibrils were densely interconnected by transverse ladders. In contrast, under the centrifugal force, these fibrils were bent and partially compressed with no obvious transverse ladders (Fig. 5D). Distribution of actin filaments and microtubules in control and centrifuged chondrocytes Distribution of actin filaments and microtubules on cryostat sections were evident in the proliferative and hypertrophic zones, respectively, in both control and experimental groups (Fig. 6). In control proliferative chondrocytes, spotty actin fluorescence was evenly distributed along the cell membranes (Fig. 6A). TEM observations consistently displayed small cytoplasmic processes with electron-dense materials, which were distributed evenly along the cell membranes (Fig. 6B). In addition, these processes were associated with extracellular fibrils. On the other hand, centrifuged proliferative chondrocytes revealed a more intense and uneven actin fluorescence along with the chondrocyte’s cell membranes (Fig. 6C). Under TEM observation, the centrifuged chondrocytes eventually displayed, beneath their cell membranes, thick electron-dense materials that associated with the extracellular matrix (Fig. 6D). In contrast to actin filaments, tubulin-immunolabeling revealed mesh-like network of microtubules in hypertrophic chondrocytes in both the control and experimental samples (Figs. 6E, 6F, respectively). There was no obvious morphological difference in the distribution and fluorescence intensity of microtubules between the control and experimental groups. DISCUSSION We have examined the histological abnormalities of chondrocytes and their surrounding matrix after centrifugation in our model mimicking a hypergravitational environment. To our knowledge, there are only a few reports on the ultrastructural abnormalities seen in chondrocytes and their extracellular ma-

196

Freitas et al.

Fig. 2 Semithin sections of the epiphyseal cartilages and their distinct zones in control (A, B and C) and experimental (D, E and F) specimens (resting zone; A, D, proliferative zone; B, E, hypertrophic zone; C, F). In the resting zone, control chondrocytes (A) are round, smooth-surfaced cells relatively distant one of the other. Radially spreading, toluidine blue-stained fibrillar strucutures can be seen among chondrocytes (arrows). In a matching area, the experiment sample (D) shows higher cell density, with rough-surfaced chondrocytes contacting neighboring cells. In the proliferative zone, control chondrocytes are flattened and distributed in columns (B), a pattern that cannot be seen in a matching region of the experiment specimen (E). Again, toluidine blue-stained fibrillar structures are present in the control samples, apparently connecting neighboring cells and running parallel to the epiphyseal longitudinal axis (arrows). In the hypertrophic zone, non-hypertrophic, irregularly shaped cells can be seen in the centrifuged specimen (F), contrasted by the control specimen (C), which shows the typical hypertrophic chondrocyte morphology. Bars, 10 µm


197

Fig. 3 TEM imaging of resting (A, C) and proliferative zones (B, D–F) in control samples. Electron-dense extracellular fibrillar structures (arrows in A, B) can be seen among chondrocytes (Ch) in both zones (A, B). C and D are magnified views of extracellular matrices in A and B, respectively, featuring a twisted, fine meshwork of cartilaginous fibrils (arrows in C and D), and its association with small chondrocytic cytoplasmic processes (arrowheads in D). Spreading radially, collagen fibrils can be seen in the intercolumnar regions of proliferative zones (E), while closer to the hypertrophic zone robust fibrils paralleling the cartilage’s longitudinal axis (arrows) populate the intercolumnar spaces (F). Bars, A, B; 10 µm, C, E; 1 µm, D, F; 0.5 µm

198

Freitas et al.

Fig. 4 TEM imaging of resting (A, C) and proliferative (B, D–G) zones in centrifuged specimens. The electron-dense fibrillar structures (See Figs. 3A and 3B) seen in control specimens are virtually absent (A, B). Sinuous extracellular fibrils distribute unevenly among chondrocytes (Ch), some loosely packed and with remarkably reduced density (C in resting zone, and E in the proliferative zone) and others displaying higher electron density (D). Notice electron-dense fibrillar structure in the tightly-packed matrix (an arrow in D), while faint trace of this structure in loose matrix (an arrow in C). Serpentine collagen fibrils populate the intercolumnar regions (F). When observed contacting cells, cytoplasmic electron-dense materials are identified beneath the cell membranes (small arrows in G). Bars, A, B; 10 µm, C-F; 1 µm, G; 0.5 µm


199

Fig. 5 TEM imaging of the hypertrophic zones in control (A, B) and experimental (C, D) specimens. Control hypertrophic chondrocytes (Ch) display translucent cytoplasm and nuclei (A), with the surrounding extracellular matrix featuring robust fibrils running parallel to the cell membranes (B) in a densely interconnected and compartmentalized manner (arrows in B). In contrast, centrifuged hypertrophic chondrocytes, despite being irregularly shaped, maintain the translucent appearance and organelle dispersion (C). Extracellular fibrils, however, are kinky and partially compressed in experimental samples (D). Bars, A, C; 10 µm, B, D; 0.2 µm, inset; 0.1 µm

200

Freitas et al.

Fig. 6 TEM imaging and fluorescence for actin (A–D) and tubulin (E, F) in control (A, B, E) and experimental (C, D, F) epiphyseal cartilages. Actin fluorescence is even along the cellular membranes in control proliferative chondrocytes (A). TEM figure of a control chondrocyte (Ch) shows small cytoplasmic processes that contain electron-dense materials (arrows in B), which are evenly distributed along the cell membranes, associating with extracellular fibrils (B). Centrifuged chondrocytes display a more intense, yet uneven actin labeling accompanying the outline of their membranes (C). These cells display, beneath their cell membranes, a thick layer of electron-dense materials (arrows in D) that associates with the extracellular matrix (arrowheads in D). Tubulin immunofluorescence unveils a mesh-like network of microtubules that is evident at the hypertrophic zones of both control (E) and experiment (F) samples. Bars, A, C, E, F; 10 µm, B, D; 0.5 µm


trix after exposure to mechanical loading. The main findings arising from this work are: 1) although expression levels of important markers for chondrocytic maturation and replication were unaltered, the distribution of centrifuged chondrocytes in their cartilaginous extracellular matrix was not typical, with increased cell density, irregular cell shapes and frequent cell-to-cell contacts, and 2) mechanical loading mimicking hypergravity alters the distribution pattern of actin filaments in chondrocytes, but not that of their microtubules. However, our findings are in contrast to reports of increased levels of mRNAs encoding type X collagen and PTHrP by chondrocytes loaded in vitro (31, 37): rat primary cultured chondrocytes, which were seeded on Bioflex silicone-elastomer with intermittent tensile strains at 30 cycles/min for 12 and 24 h, showed a significant increase in PTHrP expression (31), and hypertrophic chondrocytes of chick sterna, which had been cultured in Gelfoam sponge for 48 h, expressed more mRNA encoding type X collagen after being submitted to uniaxial tensile strains (37). Possible reasons for the discrepancy between these reports and our results may be the different way of applying load, for others groups employed intermittent tensile stress, while we used compressive forces. Likewise, interspecies differences (mice, rat and chick) may exist and should not be overlooked. It is likely that our organ culture model attenuates the amount of stress upon an individual cell, since it conserves the extracellular matrix (15). Also, shortterm application of centrifugal force (16 h) in our study may not suffice to alter type X collagen gene expression. Meanwhile, the developing cartilage was shown to respond to gravitational changes according to Hert’s curve (8, 22, 23), in which an increased baseline loading reduces cartilage differentiation, while a reduced baseline loading leads to increased differentiation, but only within a range, beyond which lack of differentiation results. This may account for the altered cell shapes seen in our centrifuged cartilages, similar to findings that were previously reported by Vico et al. (36) in a study involving excess G loading. At least in our study, the main effects of mechanical stress upon cartilage seem to be changes in extracellular matrix and in cell-to-matrix adhesion. We postulated that the absence of intracolumnar electron-dense fibrillar structures influenced the development of the abnormalities described for the experimental group (Compare Figs. 3B and 4B). These electron-dense fibrillar structures strongly resembled the toluidine blue-stained fibrils seen at light micro-

201

scopic level, and may be a compound of various types of collagens rich in proteoglycans. According to previous reports, these fibrillar structures corelated with the columnar alignment of cells in the epiphyseal cartilage (29). The electron-dense fibrillar structures ran in the columns, and rarely extended transversely. By keeping this network of fibrillar structures that connect to their cell membranes, chondrocytes may be “spatially conscious”, and maintain not only a certain distance among themselves, but also the parallelism among the longitudinal columns. Centrifugation, however, would disassemble this fibrillar network. The direction of centrifugal force was parallel to the longitudinal axis of the bones, so that several chondrocytes had been compressed and contacted each other. Strikingly, centrifuged chondrocytes in the proliferative zone were not so flattened despite compression and, rather, many cells remained round shaped (Fig. 4B). Therefore, centrifugal force appears to target not only chondrocytes, but also cell-to-matrix interactions. Ultrastructural alterations of extracellular fibrils were also found in the intercolumnar region, i.e., wavy and unevenly packed fibrils (Compare Figs. 3E and 4F). Broom and Marra (4), and Speer (29) have observed cartilage fibrils radially oriented, as well as fibril-to-fibril interconnections tying the radial arrays of fibril segments into a coherent structure. We did observe straight cartilaginous fibrils radially oriented, forming a meshwork in the control intercolumnar region. However, the centrifugal force produced wavy and unevenly packed fibrils, probably by disrupting the conformation of the extracellular matrix. In the hypertrophic zone, artificial force had also disordered the stout cartilaginous fibrils that, in control specimens, ran parallel and were interconnected (28). These ultrastructural alterations of the extracellular matrix by centrifugal force may influence the physicochemical composition of the complex cartilaginous matrices, subsequently affecting cell polarity of chondrocyte and cell-to-matrix interaction as well (8, 23, 34). Previous studies have demonstrated that the cytoskeletons can undergo remodeling in response to mechanical stimuli such as tensile strain or fluid flow (18). Yet, microtubules, a mesh-like intracellular network, were similar in hypertrophic chondrocytes of both control and experimental groups (Figs. 6E, 6F). Consistent with our findings, the microtubules of human SH-SY5Y neuroblastoma cells were not considerably affected after exposure to hypergravity (26). Microtubules may participate main-

Freitas et al.

202

ly in vesicular transport and cell polarity, being also involved in cell division and cilia formation, so that a normal distribution of microtubules indicates that hypertrophic chondrocytes are fundamentally intact, despite the compressive force. With respect to actin filaments, a more intense, yet irregular pattern was found in the centrifuged samples. Actin labeling seemed to accompany the outline of the cell membrane in control and centrifuged samples, but TEM examinations revealed an accumulation of electron-dense materials, indicative of actin filaments, right beneath the cell membrane (Fig. 6D). Taken together, the fluorescent and electron microscopic findings seem to indicate an increased accumulation of actin filaments close to the cell membrane, after application of centrifugal forces. Generally, cell-to-matrix adhesion involves integrins that bind to extracellular matrix molecules via their extracellular domains and interact with the cytoplasmic actin filaments (21). Based on this, two putative pathways for mechanotransduction in chondrocytes seem to be possible: unknown mechanosensors would respond to mechanical loads, or mechanical stress would affect the conformation of cell-to-matrix adhesions which includes actin filaments, integrins and accessory proteins. We postulated that the presence of thick actin filaments in centrifuged chondrocytes would be a result of mechanotransduction, a compensatory response to the centrifugal force. Our conjecture is supported by Durrant’s idea (10) that chondrocytic actin filaments are linked to the extracellular matrix by focal contacts, and that the disruption of these bonds may trigger compensatory deposition of actin in the vicinity of the cell membrane as the cell attempts to recreate the adhesion once lost. In summary, our experiments mimicking hypergravity have shown that the distribution of centrifuged chondrocytes in their extracellular matrix was disturbed with markedly increased cell density and frequent cell-to-cell contacts, and that the actin filament of chondrocytes, but not their microtubules, was altered after centrifugation. Further experiments are necessary to analyze the signals involved in chondrocytic load sensing and cytoskeletal responses after exposure to altered gravities. Acknowledgments This work was partially supported by grants from Naito Foundation and the Japan Society for the Promotion of Science (N. Amizuka).

REFERENCES 1. Amizuka N, Davidson D, Liu H, Valverde-Franco G, Chai S, Maeda T, Ozawa H, Hammond V, Ornitz DM, Goltzman D and Henderson JE (2004) Signalling by fibroblast growth factor receptor 3 and parathyroid hormone-related peptide coordinate cartilage and bone development. Bone 34, 13–25. 2. Amizuka N, Henderson JE, Hoshi K, Warshawsky H, Ozawa H, Goltzman D and Karaplis AC (1996) Programmed cell death of chondrocytes and aberrant chondrogenesis in mice homozygous for parathyroid hormone-related peptide gene deletion. Endocrinology 137, 5055–5067. 3. Barbieri O, Astigiano S, Morini M, Tavella S, Schito A, Corsi A, Di Martino D, Bianco P, Cancedda R and Garofalo S (2003) Depletion of cartilage collagen fibrils in mice carrying a dominant negative Col2a1 transgene affects chondrocyte differentiation. Am J Physiol Cell Physiol 285, C1504–C1512. 4. Broom ND and Marra DL (1986) Ultrastructural evidence for fibril-to-fibril associations in articular cartilage and their functional implication. J Anat 146, 185–200. 5. Castagnola P, Dozin B, Moro G and Canceda R (1988) Changes in the expression of collagen genes show two stages in chondrocyte differentiation in vitro. J Cell Biol 106, 461– 467. 6. Chen Q, Johnson DM, Haudenschild DM and Goetinck PF (1995) Progression and recapitulation of the chondrocyte differentiation program: cartilage matrix protein is a marker for cartilage maturation. Dev Biol 172, 293–306. 7. de Crombugghe B, Lefebvre V and Nakashima K (2001) Regulatory mechanisms in the pathways of cartilage and bone formation. Curr Opin Cell Biol 13, 721–727. 8. Duke PJ and Montufar-Solis D (1999) Exposure to altered gravity affects all stages of endochondral cartilage differentiation. Adv Space Res 24, 821–827. 9. Duke J, Sato A, Hamazaki T and Montufar-Solis D (1995) Clinorotation inhibits chondrogenesis in micromass cultures of embryonic mouse limb cells. Environ Med 39, 1–12. 10. Durrant LA, Archer CW, Benjamin M and Ralphs JR (1999) Organisation of the chondrocyte cytoskeleton and its response to changing mechanical conditions in organ culture. J Anat 194, 343–353. 11. Fallon JF and Caplan AI (1983) Limb Development and Regeneration. Part A, Alan R. Liss, Inc., New York. 12. Falsafi S and Koch RJ (2000) Growth of tissue-engineered human nasoseptal cartilage in simulated microgravity. Arch Otolaryngol Head Neck Surg 126, 759–765. 13. Fitts RH, Riley DR and Widrick JJ (2001) Functional and structural adaptations of skeletal muscle to microgravity. J Exp Biol 204, 3201–3208. 14. Grodzinsky AJ, Levenston ME, Jin M and Frank EH (2000) Cartilage tissue remodeling in response to mechanical forces. Ann Rev Biomed Eng 2, 691–713. 15. Ingber DE (2006) Cellular mechanotransduction: putting all the pieces together again. FASEB J 20, 811–827. 16. Henderson JE, Naski MC, Aarts MM, Wang D, Cheng L and Ornitz DM (2000) Expression of FGFR3 with the G380R achondroplasia mutation inhibits proliferation and maturation of CFK2 chondrocytic cells. J Bone Miner Res 15, 155–165. 17. Kelm JM and Fussenegger K (2004) Microscale tissue engineering using gravity-enforced cell assembly. Trends Biotech 22, 195–202. 18. Knight MM, Toyoda T, Lee DA and Bader DL (2006) Mechanical compression and hydrostatic pressure induce re-


19. 20. 21. 22. 23. 24.

25.

26.

27. 28.

29.

versible changes in actin cytoskeletal organization in chondrocytes in agarose. J Biomech 39, 1547–1551. Koch RJ and Gorti GK (2002) Tissue engineering with chondrocytes. Facial Plast Surg 18, 59–68. LeBlanc AD, Spector ER, Evans HJ and Sibonga JD (2007) Skeletal responses to space flight and the bed rest analog: A review. J Musculoskelet Neuronal Interact 7, 33–47. Martin KH, Slack JK, Boerner SA, Martin CC and Parsons JT (2002) Integrin connections map: to infinity and beyond. Science 296, 1652–1653. Montufar-Solis D and Duke PJ (1999) Gravitational changes affect tibial growth plates according to Hert’s curve. Aviat Space Environ Med 70, 245–249. Montufar-Solis D, Duke PJ and D’Aunno D (1996) In vivo and in vitro studies of cartilage differentiation in altered gravities. Adv Space Res 17, 193–199. Negulesco JA (1978) Effects of fracture trauma, estrone treatment, and 2-G environment on the epiphyseal cartilage zones of developing avian radii. Aviat Space Environ Med 49, 1219–1224. Nicogossian AE, Pool SL and Uri JJ (1994) Historical perspectives. In: Space Physiology and Medicine. (Nicogossian AE, Huntoon CL and Pool SE, eds.), pp3–49, Lea & Febiger, Malvern, PA. Rösner H, Wassermann T, Möller W and Hanke W (2006) Effects of altered gravity on the actin and microtubule cytoskeleton of human SH-SY5Y neuroblastoma cells. Protoplasma 229, 225–234. Serova LV (1993) Changing gravity level and the development of animals. Physiologist 36, S31–S33. Shepard N and Mitchell N (1976) Simultaneous localization of proteoglycan by light and electron microscopy using toluidine blue O. A study of epiphyseal cartilage. J Histochem Cytochem 24, 621–629. Speer DP (1982) Collagenous architecture of the growth plate and perichondrial ossification groove. J Bone Joint Surg Am 64, 399–407.

203 30. Suda N, Gillespie MT, Traianedes K, Zhou H, Ho PW, Hards DK, Allan EH, Martin TJ and Moseley JM (1996) Expression of parathyroid hormone-related protein in cells of osteoblast lineage. J Cell Physiol 166, 94–104. 31. Tanaka N, Ohno S, Honda K, Tanimoto K, Doi T, Ohno-Nakahara M, Tafolla E, Kapila S and Tanne K (2005) Cyclic mechanical strain regulates the PTHrP expression in cultured chondrocytes via activation of the Ca2+ channel. J Dent Res 84, 64–68. 32. Torzilli PA, Grigiene R, Huang C, Friedman SM, Doty SB, Boskey AL and Lust G (1997) Characterization of cartilage metabolic response to static and dynamic stress using a mechanical explant test system. J Biomech 30, 1–9. 33. Tso JY, Sun XH, Kao T, Reece KS and Wu R (1985) Isolation and characterization of rat and human glyceraldehydes3-phosphate dehydrogenase cDNA: Genomic complexity and molecular evolusion of the gene. Nucleic Acid Res 13, 2485– 2502. 34. Urban JPG (1994) The chondrocyte: a cell under pressure. Brit J Rheumat 33, 901–908. 35. Valhmu WB, Stazzone EJ, Bachrach NM, Saed-Nejad F, Fischer SG, Mow VC and Ratcliffe A (1998) Load-controlled compression of articular cartilage induces a transient stimulation of aggrecan gene expression. Arch Biochem Biophys 353, 29–36. 36. Vico L, Barou O, Laroche N, Alexandre C and Lafage-Proust MH (1999) Effects of centrifuging at 2 g on rat long bone metaphyses. Eur J Appl Physiol 80, 360–366. 37. Yang X, Vezeridis PS, Nicholas B, Crisco JJ, Moore DC and Chen Q (2006) Differential expression of type X collagen in a mechanically active 3-D chondrocyte culture system: a quantitative study. J Orthop Surg 1, 15–24. 38. Yasuda T, Banville D, Rabbani SA, Hendy GN and Goltzman D (1989) Rat parathyroid hormone-like peptide: comparison with the human homologue and expression in malignant and normal tissue. Mol Endocrinol 3, 518–525.