Structural and mechanical properties of the ... - CyberLeninka

Article MATBIO-1191; No. of pages: 15; 4C:

Structural and mechanical properties of the proliferative zone of the developing murine growth plate cartilage assessed by atomic force microscopy Carina Prein a, b, c , Niklas Warmbold a , Zsuzsanna Farkas b , Matthias Schieker a, b , Attila Aszodi a, b and Hauke Clausen-Schaumann a, c a - Center for Applied Tissue Engineering and Regenerative Medicine (CANTER), Munich University of Applied Sciences, Munich, Germany b - Laboratory of Experimental Surgery and Regenerative Medicine (ExperiMed), Department of Surgery, Clinical Center University of Munich, Munich, Germany c - Center for NanoScience (CeNS), University of Munich, Munich, Germany

Correspondence to Hauke Clausen-Schaumann: Center for Applied Tissue Engineering and Regenerative Medicine (CANTER), Munich University of Applied Sciences, Munich, Germany. [email protected] http://dx.doi.org/10.1016/j.matbio.2015.10.001 Edited by R. Iozzo

Abstract The growth plate (GP) is a dynamic tissue driving bone elongation through chondrocyte proliferation, hypertrophy and matrix production. The extracellular matrix (ECM) is the major determinant of GP biomechanical properties and assumed to play a pivotal role for chondrocyte geometry and arrangement, thereby guiding proper growth plate morphogenesis and bone elongation. To elucidate the relationship between morphology and biomechanics during cartilage morphogenesis, we have investigated age-dependent structural and elastic properties of the proliferative zone of the murine GP by atomic force microscopy (AFM) from the embryonic stage to adulthood. We observed a progressive cell flattening and arrangement into columns from embryonic day 13.5 until postnatal week 2, correlating with an increasing collagen density and ECM stiffness, followed by a nearly constant cell shape, collagen density and ECM stiffness from week 2 to 4 months. At all ages, we found marked differences in the density and organization of the collagen network between the intracolumnar matrix, and the intercolumnar matrix, associated with a roughly two-fold higher stiffness of the intracolumnar matrix compared to the intercolumnar matrix. This difference in local ECM stiffness may force the cells to arrange in a columnar structure upon cell division and drive bone elongation during embryonic and juvenile development. © 2015 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Introduction The growth plate (GP), situated between the epiphysis and diaphysis at both ends of long bones [1,2] is a unique and essential cartilaginous structure which is responsible for the elongation of bones formed by endochondral ossification. In this process, first skeletogenic mesenchymal stem cells condense and differentiate into chondrocytes prefiguring the shape of the future bone. In subsequent steps, the chondrocytes of the cartilaginous template undergo a differentiation cascade [3] within the GP which drives the longitudinal growth of the skeletal elements

through chondrocyte proliferation, cellular enlargements via hypertrophy, extracellular matrix (ECM) synthesis and controlled matrix degradation [4–7]. The fully matured GP is organized into horizontal zones of resting, proliferative and hypertrophic chondrocytes (Fig. 1), which vary in cellular arrangement, function and matrix composition [8–10]. In the resting or germ layer of the GP the chondrocytes are roundish and rarely divide. In the proliferative zone, the cells undergo rapid division, are flattened along the mediolateral axis, and form columns along the proximodistal axis of the long bones. Column elongation occurs through spatially coordinated cell

0022-2836/© 2015 The Author. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Matrix Biol. (2015) xx, xxx–xxx Please cite this article as: C. Prein, et al., Structural and mechanical properties of the proliferative zone of the developing murine growth plate cartilage assessed by atomic ..., Matrix Biol (2015), http://dx.doi.org/10.1016/j.matbio.2015.10.001

2

Properties of the developing murine GP cartilage assessed by AFM

Fig. 1. Zonal arrangement of growth plate cartilage and its ECM. Cartilage growth plate is divided into horizontal zones called resting, proliferative, hypertrophic and calcified. Within the proliferative zone, chondrocytes are flattened and arranged into characteristic columns caused by clonal expansion. Chondrocytes are surrounded by a thin pericellular matrix (PCM, red), which is enveloped by the territorial matrix (TM, blue). The PCM and the TM between the cells of a column define the so called transverse septum (dashed arrow). The ECM around the columns is called the interterritorial matrix (ITM green), which fills up the longitudinal septum between the chondrocyte clusters (solid arrow). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

division and rotational movements [11]. The division axis of the flattened chondrocytes is parallel to the proximodistal direction resulting in semi-circular daughter cells which initially lie along the mediolateral axis. Subsequently, the cells gradually flatten and glide around each other to arrange into the longitudinal column. Chondrocytes at the proximal end of the column then exit from the cell cycle, and increase their volume to become hypertrophic before the zone is calcified and replaced by trabecular bone. One column in the proliferative zone usually consists of four to eight flattened chondrocytes surrounded by different matrix compartments. The pericellular matrix (PCM), which is rich in proteoglycans (PGs), collagen VI and very fine fibrils [12–14], immediately surrounds the chondrocytes. The adjacent territorial matrix (TM) compartment has a fine network of cross-banded, heterotypic fibrils composed of collagen types II, IX, and XI. The PCM and TM together with the clustered chondrocytes define the columnar chondron, the functional unit of the cartilage [15], which is believed to play an important role in regulating the interactions between chondrocytes and their surrounding matrix [16]. The interterritorial matrix (ITM), which is located between the columns contains thick collagen fibrils that largely run parallel to each other in the fully matured growth plate [17]. At a higher structural level the TM–PCM compartments account for the transverse septum (TS) matrix, separating the stacked chondrocytes within a column, while the ITM primarily accumulates in the intercolumnar longitudinal septum (LS) (Fig. 1). In each matrix compartment, the interfibrillar space is filled with various types of proteoglycans like aggrecan, decorin, and fibromodulin as well as non-collagenous glycoproteins such as cartilage oligomeric matrix protein and matrilins [18]. The

tension of the collagen fibrils and the osmotic pressure of the highly charged proteoglycan aggrecan are responsible for the specific mechanical properties of cartilage that withstands tensile, compressive and shear stress. Although the columnar organization of the chondrocytes is essential for bone elongation, and despite the fact that the structure and molecular composition of the different matrix compartments of the GP has been intensively studied, the mechanism guiding the chondrocytes to rotate around each other and arrange in a columnar structure upon cell division still remains unclear. There is an increasing body of evidence that the fate of cells, and therefore the morphology and function of tissue is tightly controlled by the mechanical properties of their micro-environment [19,20]. In the case of cartilage, this mechanical micro-environment is mainly determined by the composition and structure of the cartilage ECM. Numerous studies have revealed that the mechanical properties of articular cartilage correlate with and depend on the collagen/proteoglycan composition, not only during development [21,22] but also in osteoarthritis [23,24]. However, the literature is sparse regarding the mechanobiology of the growth plate, even though there is an emerging perception that limb and growth plate morphogenesis is strongly influenced by both extrinsic and intrinsic mechanical cues, which may play an instructive role for chondrocyte differentiation and organization. In fact, local ECM biomechanics might even guide chondrocytes to arrange into a columnar stack and thus drive the linear elongation of endochondral bones [25–29]. Some recent studies investigating the intrinsic mechanical properties of the GP, using unconfined compression tests, have demonstrated zone- and developmental stage-dependent variations in elasticity [30–32]. Nevertheless, to elucidate how GP

Please cite this article as: C. Prein, et al., Structural and mechanical properties of the proliferative zone of the developing murine growth plate cartilage assessed by atomic ..., Matrix Biol (2015), http://dx.doi.org/10.1016/j.matbio.2015.10.001


chondrocytes and GP morphogenesis are affected by the local physical micro-environment and gain a better understanding of the tight correlation between structure, mechanical properties and biological function of developing cartilage, it is essential to investigate both structural and mechanical parameters of cartilage samples simultaneously, at different developmental stages, and with sub-micrometer resolution. In recent decades, several techniques have emerged which enable the investigation of either structure or mechanical properties of biological samples. Atomic force microscopy allows for the simultaneous analysis of structure and mechanical properties of biological samples on the molecular [33–35], cellular [36–41] and tissue level under native conditions [42]. Consequently, AFM has been extensively used as a dual, high-resolution imaging and stiffness analysis tool for the investigation of normal and diseased articular cartilage in various species [43–45]. The nanometer spatial resolution of AFM not only reveals detailed structural information like individual collagen fibrils with their characteristic D-band structure, but upon indentation also differentiates in elastic moduli of various matrix compartments and ECM constituents such as the soft proteoglycan moiety and the hard collagen fibrils [46,47]. However, despite the large number of studies using AFM to assess structure and mechanical properties of articular cartilage, applications of the AFM technology to rapidly developing and highly dynamic cartilaginous structures like the growth plate in long bones are scarce and limited to static studies, investigating GP cartilage at one single time point Radhakrishnan et al. [9], for example, used indentation-type AFM to study the zone-specific mechanical properties of the GP ECM of 6-weeks-old rabbits. The authors observed a steady increase of matrix stiffness from the reserve zone to the calcified zone, indicating that the GP is an inhomogeneous tissue with biomechanical properties varying according to its molecular and cellular arrangement. Another study revealed an almost three times higher stiffness in the ITM than in the PCM in the proliferative zone of one month old rat GP [14]. Taken together, these results suggest that the different mechanical properties in the different GP zones correlate with different stages of cartilage maturation in these zones. Nevertheless, a systematic study of the changes in the structure and mechanical properties occurring during growth plate cartilage development, which may shed light on the correlation between ECM biomechanics and proper GP development, is still lacking at present. Here, we report the nanoscale AFM investigation of structure and biomechanics of developing cartilage in the tibial GP proliferative zone, from embryonic development to adulthood, in mouse as model animal. The proliferative zone was chosen for the analysis due to its crucial role in the establishment of the GP columnar structure, which is

3

essential for the proper linear growth of endochondral bones.

Materials and methods Sample preparation Hind limbs were collected from euthanized wild-type laboratory mice (strain C57/Bl6) at different embryonic (E) and postnatal developmental stages (E13.5, E15.5, newborn, 2 weeks, 4 weeks and 4 months). After peeling off the skin, hind limbs from the left side were embedded in optimal cutting temperature (OCT, Thermo Scientific, Braunschweig, Germany) tissue freezing medium and were snap-frozen in isopentane chilled in liquid nitrogen bath. Samples were cut in 30 μm-thick sagittal sections using a Microm HM500 cryostat (Thermo Scientific, Braunschweig, Germany). To get a flat sample surface and to preserve the morphology of undecalcified bone specimens, normal adhesive tape was stuck onto the sample block surface before every cutting step. Afterwards the adhesive tape carrying the section upwards was positioned on a positively-charged Superfrost Plus glass slide (Thermo Scientific, Braunschweig, Germany) covered with double-sided adhesive tape. Samples were stored at − 20 °C, until investigated with the AFM. For histological examination of the growth plate structure, samples from the right side were fixed in 4% paraformaldehyde/phosphate buffered saline pH = 7.4 (PFA/PBS), embedded into paraffin, cut in 6 μm and stained with 0.1% Safranin-O (SigmaAldrich, Deisenhofen, Germany) and 0.02% aqueous fast green (Sigma-Aldrich, Deisenhofen, Germany). AFM measurements AFM measurements were carried out using a NanoWizard® I AFM (JPK Instruments, Berlin, Germany) in combination with an inverse optical microscope (Axiovert 200, Carl Zeiss MicroImaging GmbH, Göttingen, Germany). The AFM had a maximum lateral scan range of 100 × 100 μm 2, and a vertical range of 15 μm. A CCD camera (The Imaging Source Europe GmbH, Bremen, Germany) was used to precisely locate the cantilever at the position of interest. The whole set-up was positioned on an active vibration isolation table (Micro 60, Halcyonics, Göttingen, Germany) inside of a 1 m 3 soundproof box for the reduction of external noise [37,38]. For contact mode imaging and indentation-type AFM (IT-AFM) measurements, silicon nitride cantilevers (MLCT Microcantilever, Bruker, Mannheim, Germany) with a nominal spring constant of 100 mN/m and integrated pyramidal tips with nominal radius of 20 nm were used. The actual cantilever


4


spring constant of each cantilever was determined individually after PBS buffer was added by using the thermal noise method [34,48,49]. Every spring constant measurement was repeated three times and the arithmetic average was then used for data analysis. All images were recorded in air with a resolution of 512 × 512 pixels 2 at a line rate of 1 Hz. After the proper orientation of the tibiae cartilage and identification of the growth plate mid-proliferative zone, an initial overview image was recorded. Shape indexes of the chondrocytes were calculated from these overview images by taking the ratio of the long and short axes of the cells. Afterwards the scan area was reduced to obtain structural details of ITM in the intercolumnar longitudinal septa or the TM–PCM in the intracolumnar transverse septa. Because of the limited space between single cells of one column in the developing cartilage, a clear distinction between TM and PCM was not always possible. To quantify the collagen density in the TS and LS, the average number of fibrils contained in a 1 μm 2 area was determined using four independent AFM images for each developmental stage. After AFM imaging, phosphate buffered saline at pH 7.4 was added to the sample for IT-AFM measurements. On an area of 3 × 3 μm 2, 25 × 25 force– indentation curves, consisting of 256 data points each, were recorded. In every age group, three animals and two sections per animal were used to analyze the elastic properties of ECM compartments resulting in 6 × 625 curves for both ITM and TM–PCM. The Young's modulus was extracted from the approach force–indentation curves using a modified Hertz model for a pyramidal indenter. In this model, the force (F), which is required to push the tip into the sample, is a quadratic function of the indentation depth (δ): F¼

2 x tanα E δ2: π x ½1 ν 2

Here E is the Young's elastic modulus; ν is the Poisson's ratio, which was set to 0.5 as is commonly done for cartilage [45,47,50] and other incompress-

ible materials [38,51] and α is the tip half-opening angle (17.5° for our cantilever). Note, that in order to precisely locate matrix compartments (ITM and TM–PCM, respectively) and to compare structural and biomechanical properties at identical regions of the samples, high-resolution AFM images had to be acquired prior to the IT-AFM measurements. Consequently, the samples were first dried in air for high-resolution imaging and then rehydrated in PBS buffer for IT-AFM measurement. To ensure that drying and rehydrating of the cartilage samples does not affect elasticity, we compared the Young's moduli in ITM and TM–PCM of freshly cut tibial sections in PBS at newborn and 4 weeks of age to those of consecutive sections of the same animals after they were dried in air and rehydrated in PBS. Consistent with other publications [52–54], we did not observe an effect of drying/rehydration on the samples Young's moduli (see Fig. S1 in the Supporting Material). To avoid contributions from the underlying substrate to the extracted Young's modulus of the sample, the maximum indentation depth, which was used for data analysis was limited to a few percent (bb 10%) of the sample thickness [55]. As a consequence, the force–distance curves were analyzed only up to a maximum indentation depth of 500 nm, using the JPK Data Processing software (Version 4.2.20, JPK Instruments AG, Berlin, Germany). The contact point of each indentation curve was determined by the fitting procedure and checked and corrected manually, whenever necessary [56]. Statistical analyses Based on the results obtained from the three different animals, stiffness distributions (histograms) were calculated for each developmental stage. To locate the maxima of these histograms, a linear combination of two Gaussian functions was fitted to each histogram using the Igor Pro software (Version 6.3.4.0). The peak positions (locations of the two maxima) and the standard deviations are given in Table 1.

Table 1. Elastic moduli and collagen density (CD) of ITM and TM–PCM at different developmental stages of growth plate proliferative zone. Statistical significance between two developmental stages was established at p ≤ 0.05 (*), p ≤ 0.01 (**) and p ≤ 0.001 (***) ITM (kPa) CD TM–PCM (kPa) CD

1st peak 2nd peak Range 1 μm2 1st peak 2nd peak Range 1 μm2

E13.5

E15.5

NB

2 weeks

4 weeks

4 months

3.6 ± 0.1*** – 1.3–21.6 10.5 ± 1.3 1.6 ± 0.01*** – 0.8–3.6 8.8 ± 1.3

20.5 ± 0.2*** 31.5 ± 1.1*** 7.6–54.3 12.3 ± 1.5* 7.7 ± 0.1*** – 4.2–16.7 9.5 ± 1.3

26.1 ± 0.4*** 59.9 ± 2.1*** 8.9–137.9 16.8 ± 2.9 17.5 ± 0.1*** 23.5 ± 0.9*** 7.5–48.3 10.8 ± 1.3*

47.8 ± 1.1*** 73.9 ± 6.5*** 16.6–164.5 18.3 ± 1.9 23.0 ± 0.1*** 39.6 ± 1.6*** 5.8–110.4 13.5 ± 1.7

56.5 ± 0.8*** 89.9 ± 2.8 20.2–129.2 20.5 ± 1.3 19.3 ± 0.3*** 31.9 ± 1.4*** 7.2–96.9 15.5 ± 2.1

53.0 ± 0.5 89.4 ± 6.7 2.9–169.1 23 ± 1.8 28.0 ± 0.1 49.3 ± 0.8 8.1–61.8 17.8 ± 1.9



Statistical evaluation of Young's moduli and collagen density between two developmental stages was performed using the unpaired t test with the help of the software GraphPad Prism.

Results In mice, the cartilaginous elements of the limbs develop along a proximodistal sequence and appear at the age of E11–E13. At E13.5, the earliest time point investigated here, proliferative and hypertrophic growth plate zones of the tibia were already clearly separated in the optical microscopy images (Fig. S2). Overview AFM images revealed that chondrocytes had a long axis of approximately 5– 7 μm and were moderately flattened, characterized by a shape index of 1.57. Most of the cells already oriented along the mediolateral axis of the bone, however they were not yet arranged into columns (Fig. 2A, and Fig. S2A, A’). At this time-point, the cartilage within the proliferative zone revealed a rather homogenous matrix (Fig. 2B, C), without distinctly visible longitudinal septa (LS) and transverse septa (TS). Usually, the ITM in the GP is defined as the matrix compartment between two individual columns (the longitudinal septum). Since at E13.5 no columnar arrangement could be detected, the matrix area between two cells was chosen for high-resolution AFM imaging and indentation measurements of ITM, while the area with a distance of less than 1 μm away from the cell surface was chosen for close up imaging and indentation measurements of TM–PCM. On high resolution AFM images (3 × 3 μm 2), the collagen fibrils did not show any periodic banding, and appeared as a randomly oriented meshwork without any obvious difference in orientation between TM–PCM and ITM (Fig. 2B, C, and Fig. 4A, B). Nevertheless, quantitative analysis showed a slightly increased collagen fibril density in the ITM compared with the TM–PCM (see Table 1 for collagen density values). These differences in collagen density were reflected by the elasticity measurements, which revealed unimodal frequency distributions of elastic moduli with a maximum at 3.6 ± 0.1 kPa (Fig. 5A) in the ITM and 1.6 ± 0.01 kPa in the TM–PCM areas (Fig. 5B, and Table 1). At E15.5, the proliferative, hypertrophic and mineralized zones were clearly visible as distinct regions in the optical microscopy images (Fig. S2B, B’). AFM overview images within the proliferative zone revealed flattened chondrocytes (shape index: 2.18) which formed short columns (2-to-4 cells) and defined longitudinal and transverse septa between and inside the columns, respectively (Fig. 2D). Higher resolution images of the ITM in the LS and of the TM–PCM in the TS showed a randomly oriented collagen network in both compartments, but the collagen fibrils were less dense in TS compared

5

to LS (Fig. 2E, F, Fig. 4C, D and Table 1). Indentation measurements within the ITM gave rise to a bimodal stiffness distribution characterized by a first peak at 20.5 ± 0.2 kPa and a second peak at 31.5 ± 1.1 kPa. A similar bimodal nanostiffness distribution has been observed in mature articular cartilage, where the first peak is generally attributed to the proteoglycan phase and the second peak to the collagen phase [47]. The indentation measurements in TM–PCM still resulted in a unimodal stiffness distribution (peak: 7.7 ± 0.1 kPa) (Fig. 5C, D). As at E13.5, the TM–PCM was significantly softer then the ITM. Compared to E13.5, the range and peak elastic moduli of both ITM and TM–PCM were greatly increased (Table 1). In newborn (NB) samples, the epiphyseal and growth plate regions of the bone were clearly separated and the AFM cantilever was easily positioned over the central part of the proliferative zone. The overview optical and AFM images (Fig. 2G, S2C, Cˋ) revealed a columnar arrangement of 6 to 8 chondrocytes per column, with a typical cell geometry of 20 μm along the mediolateral axis and 5 to 7 μm along the proximodistal axis (shape index: 3.82). Between individual chondrocytes within one column, the width of the transverse septa was approximately 2 to 3 μm. The longitudinal septa between vertical columns were much broader, ranging from 10 to 30 μm. At a resolution of 3 × 3 μm 2, AFM images revealed a regular axial striation of collagen fibrils, showing the periodic gap and overlap organization of the collagen molecules in the fibril (Fig. 2H). Height profiles along single collagen fibrils, revealed the typical D-band periodicity of 67 nm (Fig. S3). Inside the column, AFM imaging between two chondrocytes showed a less abundant fibrillar material in the TM–PCM/TS (Fig. 2I). In the ITM/LS region, the banded collagen network appeared more distinct and was significantly denser (Table 1). The orientation of the fibrils appeared random in both compartments (Fig. 4 E, F). The ITM of the LS displayed a broad distribution of the elastic moduli with two clearly distinguishable peaks (Fig. 5E). The elastic moduli ranged from 8.9 to 137.9 kPa, and fitting a linear combination of two Gaussian functions rendered two maxima at 26.1 ± 0.4 kPa and 59.9 ± 2.1 kPa, respectively. Indentation measurements in the TM–PCM of the TS resulted in a reduced stiffness and a much narrower distribution ranging from 7.5 kPa to 48.3 kPa with two peaks at 17.5 ± 0.1 kPa and 23.5 ± 0.9 kPa (Table 1). The 2-week-old samples exhibited well-established columns with mediolaterally oriented flattened chondrocytes (shape index: 4.1) (Fig. 3A, S2D, D'). Collagen fibrils within the ITM of the LS were denser than in the TM–PCM of TS (Fig. 3B, C and Table 1). Furthermore, a quantitative analysis of the angular distribution of collagen fibrils revealed that in the ITM


6


Fig. 2. AFM overview and ECM detail images at embryonic and newborn developmental stages. The AFM overview height image (A) of E13.5 tibial growth plate depicted moderately flattened chondrocytes and a lack of columnar arrangement within the proliferative zone. The surrounding ITM (B) revealed the disorganized collagen II network. Within TM–PCM, AFM vertical deflection image (C) indicated a decrease in collagen II density. At E15.5, proliferative chondrocytes were clearly elongated, oriented and organized into short columns (D). At newborn stage, the expanded columnar arrangement of greatly flattened chondrocytes within the proliferative zone was evident (G). The vertical deflection images indicated less dense collagen network in the TM–PCM of the TS (I) compared to the ITM of the LS (H). Arrows indicate the proximodistal axis of the long bones. Dotted lines depict the border between chondrocytes and the adjacent matrix. White squares indicate the positions of the high-resolution images of the ITM, while the blue square indicate the positions of high-resolution images of the TM–PCM. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

fibrils now started to orient preferentially along the proximodistal axis of the bone, while collagen fibrils in the TM–PCM were still randomly oriented

(Fig. 4G, H). The stiffness measurements in the ITM could be fitted with a bimodal distribution in the range of 16.6 kPa to 164.5 kPa. However, the two



7

Fig. 3. AFM overview and ECM detail images at several time points during mouse adulthood. At 2 weeks (A), 4 weeks (D) and 4 months (G) the GP possessed regularly formed columns with flattened chondrocytes oriented at right angles to the proximodistal axis (arrows) of the bone. At all stages, collagen fibrils within the ITM of the longitudinal septa (B, E, H) appeared to be more compact compared to the network within the TM–PCM of the transverse septa (C, F, I). Also note the tendency of bundling and parallel orientation of collagen fibrils in the ITM/LS. Dotted lines depict the border between chondrocytes and the adjacent matrix. White squares indicate the analyzed ITM, while the blue square shows the analyzed TM–PCM region. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

peaks (47.8 ± 1.1 kPa and 73.9 ± 6.5 kPa) were not as clearly separated as in the case of the newborn ITM. Furthermore, compared to the newborn samples, both peaks were shifted to higher values (Fig. 5G). With a range from 5.8 kPa to 110.4 kPa and with peaks at 23.0 ± 0.1 kPa and 39.6 ± 1.6 kPa, the

elastic moduli in TM–PCM of the TS were higher than in the newborn samples and lower than in the ITM (Table 1). At 4 weeks (shape index 4.82), the collagen fibrils in the ITM formed bundles and were oriented parallel to the proximodistal axis of the bone (Fig. 3E, and Fig. 4


8


Fig. 4. Angular distribution of individual collagen fibrils within ITM and TM–PCM at different developmental stages. During growth plate development fibrils within the ITM/LS tend to orient parallel to the proximodistal (0°) axis and parallel to the mediolateral axis in TM–PCM/TS (90°). Newborn = NB; E = embryonic day. Dashed line represents the trend of fibril orientation during growth plate development.



9

Fig. 5. Stiffness distribution of ITM and TM–PCM within growth plate proliferative zone at different developmental stages measured by IT-AFM. Stiffness distributions in the ITM of the longitudinal septa (LS) and in the TM–PCM of the transverse septa (TS) at different developmental stages determined by IT-AFM. Newborn = NB; E = embryonic day. Solid line represents the sum of two Gaussian functions while the dashed lines show independent fits.


10


I). At 4 months (shape index 5.28) most collagen fibrils were aligned along the proximodistal axis in the ITM, while the fibrils in the TM–PCM were preferentially aligned along the mediolateral axis (Fig. 3H, I and Fig. 4 K, L). Like at all other stages, the collagen network was less dense in TM–PCM compared to the ITM (Table 1). The stiffness values in the ITM (4 weeks: 56.5 ± 0.8 kPa and 89.9 ± 2.8 kPa; 4 months: 53.0 ± 0.5 kPa and 89.4 ± 6.7 kPa) and the TM–PCM (4 weeks: 19.3 ± 0.3 kPa and 31.9 ± 1.4 kPa; 4 months: 28.0 ± 0.1 kPa and 49.3 ± 0.8 kPa) were comparable to the 2-weeks-old samples (Fig. 5I − L).

Discussion Temporal evolution of stiffness and morphology of the proliferative zone The aim of this study was to determine the structural and mechanical properties of murine cartilage in the growth plate proliferative zone during development. The experimental data revealed a steady increase in collagen density during cartilage development and maturation, and a preferential orientation of the collagen fibrils along the proximodistal axis in the ITM and the mediolateral axis in the TM–PCM, at the post-natal stages. The increase in collagen density is accompanied by a gradual stiffening of ITM and TM– PCM from E13.5 up to an age of 2 weeks and nearly constant matrix stiffness between 2 weeks and 4 months of age. These observations are consistent with an AFM study on the rabbit GP [9] which demonstrates a steady increase of matrix stiffness from the reserve zone to the calcified zone, suggesting that the mechanical properties correlate with the different stages of cartilage maturation in these zones. Another recent study by Wosu et al., using unconfined compression tests and a transversely isotropic biphasic elastic model, showed an increasing stiffness of porcine GP from newborn to 4 weeks, followed by steady decrease in stiffness to 18 weeks [32]. In this experimental setup, the samples were cylinders of 4 mm diameter comprising several zones of the porcine GP, which means that the fraction of the different zones comprised by the 4 mm cylinder may change during GP development. The authors attributed the stiffness reduction after 4 weeks of age to an increased cell to matrix ratio. In order to accurately determine the mechanical changes in the different zones of the GP over time thus requires a spatial resolution, which is high enough to resolve the different GP zones, as it can be achieved by IT-AFM measurements. Note that our investigation starts at the embryonic stage, where the growth plate acts as a highly dynamic tissue and ends at mouse adulthood, where the process of endochondral ossification is complet-

ed and the growth plate structure and composition is hardly changing any more. In mice, the cartilaginous elements of the limbs develop sequentially along a proximodistal sequence and appear at E11 to E13 [57]. Afterwards, the chondrocytes rapidly proliferate and synthesize the cartilaginous matrix components including collagen fibrils, proteoglycans, hyaluronan and non-collagenous glycoproteins [58]. The cartilage stiffening observed between E13.5 and 2 weeks coincides with a complete GP reorganization and matrix remodeling, as can be seen in the light microscopy and overview AFM images. At E13.5, the chondrocytes in the presumptive proliferative zone are roundish, characterized by a shape index close to 1, and do not form distinct columns but are rather organized into horizontal lines and deposit their matrix predominantly in between these chondrocyte rows. At E15.5, the proliferative chondrocytes flatten and align with their long axis along the mediolateral plane of the developing bone. The mechanisms of mediolateral elongation of the chondrocytes involve morphogenetic signals mediated via the Wnt–Planar Cell Polarity (PCP) pathway [59,60], cell adhesion molecules [12] and matrix mechanics [27]. The formation of longitudinally oriented columns with flattened chondrocytes from E15.5 on is accompanied by extensive synthesis and subsequent maturation of the cartilage matrix resulting in, as we have shown here, increased stiffness of both the longitudinal and the transverse septa. The gradually increasing stiffness in the TS during embryonic and postnatal development may, at least partially, account for the steady flattening of the cells in the columns of the proliferative zone. The transition from roundish to increasingly flattened chondrocyte shape with age correlates well with higher stiffness values of the TS strongly suggesting that the pressure exerted by the matrix will eventually deform the columnar chondrocytes in a mediolateral direction. Our data is in accordance with and mechanistically supports an earlier model proposing that matrix secretion modulates chondrocyte orientation and flattening during early cartilage formation [27]. Spatial variations of stiffness and morphology within the proliferative zone In addition to the temporal changes in the mechanical properties during GP development, we also observed systematic spatial variations in the GP mechanical properties, namely, that the TM–PCM in the transverse septa was always significantly softer than the ITM in the longitudinal septa (roughly by a factor of 1/2 throughout all our measurements). This correlates with a lower collagen density observed in the TM–PCM compared to the ITM, and agrees well with results by Darling et al., [13] who observed by AFM a significantly lower stiffness in the PCM of



murine, porcine and human articular cartilage compared to the ECM. Comparable results were obtained by McLeod et al., who investigated the mechanical properties of the PCM and ECM of porcine articular cartilage by microscale AFM [61]. The different elastic moduli in the different ECM compartments can be attributed to the different biological functions of the different matrix compartments. Characterized by reduced permeability and stiffness, the PCM generates the microenvironment of the chondrocytes and functions as a transducer of mechanical signals between matrix and cells [13,62]. It contains a high amount of hyaluronic acid with fixed negative charges and thus a high amount of water molecules [63–65]. The ITM, on the other hand, has a tissue supporting function [66], including a higher amount of solid material, as was demonstrated by Poole et al. and other groups [65,67,68], who identified collagen type II, IX and XI as well as aggrecan and keratan sulfate within the ITM. In general, with increasing distance from the cell surface, the amount of solid material is increasing, resulting in an increasing Young's modulus in the ITM compared to the PCM [14]. In our study, the higher stiffness of the LS between the chondrocyte stacks compared to the TS inside the column demonstrates a spatially heterogenic mechanical environment within the proliferative zone, which might have a regulatory role for growth plate polarity. The flattened and mediolaterally oriented proliferative chondrocytes in the column divide parallel to the proximodistal axis of the bone and then intercalate to elongate the longitudinal columns [11,12,59]. While primarily the Wnt–PCP pathway regulates chondrocyte polarity and determines the division plane [59,60], the mechanism which forces the daughter cells to rotate around each other, and hence stay in the column, is currently unknown. We suggest that the different mechanical resistance of the ECM between and inside the columns could provide a simple explanation for chondrocyte movement and columnar organization. After division, the stiff matrix of the LS resists the elongation of the semi-circular daughter cells along the mediolateral axis in between the columns and constrains the chondrocytes to move into the softer space of the TS within the column (Fig. 6). Our model suggests that any disturbance of the mechanical environment and differential stiffness of LS/ TS in the proliferative zone should have severe consequence on the columnar organization of the chondrocytes. Indeed, mice or humans with very soft cartilage due to the lack of type II collagen fibrils in the ECM have no discernible columnar structure [69–71]. Bimodal distribution of cartilage stiffness It has been recently demonstrated by Loparic et al. [47] that nanoscale IT-AFM not only measures the

11

Fig. 6. Model for chondrocyte columnar arrangement due to different mechanical resistance of the surrounding ECM. Chondrocyte division and column elongation within the proliferative zone of the growth plate. Ellipsoid cells in the column divide parallel to the longitudinal septum (LS) axis (dashed line), where the stiff matrix resists chondrocyte elongation. As a result, semicircular daughter cells rotate around each other and move into the softer matrix of the transverse septum (TS) where they flatten again.

average local matrix stiffness of porcine articular cartilage but it also probes the elasticity of distinct matrix constituents such as the proteoglycan moiety and the collagen fibrils. As a result, a bimodal Gaussian distribution can be fitted to the histogram of stiffness values, with a lower stiffness peak corresponding to the proteoglycan network, and higher stiffness peak representing the collagen fibrils. In this study, we could confirm such bimodal stiffness distributions also in the proliferative zone of the developing murine growth plate. Interestingly, at an early embryonic stage (E13.5), the entire tissue is very soft and only one peak is detectable in the histograms of both the ITM and the TM–PCM region. In the ITM of the longitudinal septa, two distinct peaks are clearly visible as early as E15.5, while the bimodal stiffness distribution in the TM–PCM of transverse septa appears only from the newborn stage on. The typical diameter of the collagen fibrils in the developing growth plate cartilage is between 10 and 20 nm [72–75], whereas the estimated pore size of the collagen meshwork is around 60–200 nm in mature articular cartilage [76] suggesting that the AFM tip probes the proteoglycans more frequently than the collagen fibrils [47]. Shortly after the formation of the mouse tibia, at E13.5, the cartilage matrix is still immature, and characterized by a fine and less abundant collagen network throughout the growth plate. This may result in less collagen hits by the AFM tip preventing the detection of the collagen


12


peak. At later stages, the higher collagen density due to increased collagen deposition leads to more interaction between the fibrils and the AFM tip, generating, in addition to the first peak representing the proteoglycan component, a second peak representing the collagen fibrils. We have to note that Loparic et al. reported a more clear separation and difference between the peak stiffness of proteoglycans (about 22.3 kPa) and collagen (about 384 kPa) at the surface of adult porcine articular cartilage [47]. Nevertheless, this discrepancy can be explained by species-, location- and age-specific differences in the density, diameter and stiffness of the collagen fibrils. Compared to the transient growth plate, the permanent, mature articular cartilage normally contains thicker collagen fibrils ranging from 30 to 200 nm in diameter [75], and the average fibril diameter in skeletally immature pig articular cartilage is between 40 and 50 nm [77]. In addition, the level of collagen cross-linking in the murine growth plate and the porcine articular cartilage may differ, which can significantly affect the stiffness of individual collagen fibrils. Nevertheless, fibril diameter [73], density and cross-linking increase from embryonic to young adolescence stages [78] of the growth plate cartilage, resulting in an age-dependent stiffening of the collagen network. The stiffer collagen mesh resists the swelling of the proteoglycans which in turn may lead to higher internal pressure in the interfibrillar space. Thus, both the proteoglycan and the collagen fibrillar components of the cartilage ECM exhibit an increasing stiffness during the development of the growth plate.

Outlook In this study, we could demonstrate a close correlation between ECM structure and stiffness and cellular organization of the proliferative zone of the developing growth plate. The results suggest that the higher ECM stiffness in the ITM, compared to the TM–PCM, may force the cells into their columnar structure and thereby drive proper bone elongation and morphogenesis. The data indicate that nanoscale AFM is a powerful tool 1) to monitor structure-biomechanical function relationship both spatially and temporary; and 2) to resolve the mechanical properties of distinct ECM components (proteoglycans and collagen fibrils) in dynamic connective tissues such as the growth plate. We propose that nanoscale AFM imaging and indentation measurements provide a unique biomarker to assess cartilage development in normal murine samples and in genetically-modified mice, lacking particular ECM components. Nanometer-scale AFM imaging and indentation using such genetically modified model systems, could provide essential clues for identifying the molecular determinants of ECM stiffness and understanding the

mechanical influence on longitudinal bone growth in both normal and pathological situations. Furthermore, mapping of nanometer-scale structural and biomechanical properties of the growth plate helps to gain a better understanding of cartilage diseases such as chondrodysplasias, and may help to develop better clinical interventions for corrections of human conditions affecting skeletal growth. Supplementary data to this article can be found online at http://dx.doi.org/10.1016/j.matbio.2015.10. 001.

Author contributions Carina Prein performed the AFM measurements and analysis for all samples except for E15.5. Niklas Warmbold performed AFM measurements for E15.5. Zsuzsanna Farkas performed histology. Matthias Schieker, Attila Aszodi and Hauke ClausenSchaumann designed the project and supervised the experiments. Hauke Clausen-Schaumann and Attila Aszodi wrote the manuscript together with Carina Prein.

Acknowledgments The authors thank Paolo Alberton, Maximilian Saller and Stefanie Sudhop for helpful discussions. CP and HCS acknowledge the financial support from the CANTER Forschungsschwerpunkt of the Bavarian State Ministry for Science and Education, CP, AS and HCS acknowledge financial support from the DFG (grants AS 150/7-1, AS 150/10-1 and CL 409/2-1). Received 10 July 2015; Received in revised form 25 September 2015; Accepted 6 October 2015 Available online xxxx Keywords: Growth plate; Proliferative zone; Atomic force microscopy; AFM; Biomechanics; Elasticity

References [1] B.C.J. Van Der Eerden, M. Karperien, J.M. Wit, Systemic and local regulation of the growth plate, endocr. Rev. 24 (2003) 782–801. [2] F. Rauch, Bone growth in length and width: The Yin and Yang of bone stability, J. Musculoskelet. Neuron. Interact. 5 (2005) 194–201.



[3] H. Akiyama, J.P. Lyons, Y. Mori-Akiyama, X. Yang, R. Zhang, Z. Zhang, et al., Interactions between Sox9 and βcatenin control chondrocyte differentiation, Genes Dev. 18 (2004) 1072–1087. [4] E.B. Hunziker, R.K. Schenk, Physiological mechanisms adopted by chondrocytes in regulating longitudinal bone growth in rats, J. Physiol. 414 (1989) 55–71. [5] N.J. Wilsman, C.E. Farnum, E.M. Leiferman, M. Fry, C. Barreto, Differential growth by growth plates as a function of multiple parameters of chondrocytic kinetics, J. Orthop. Res. 14 (1996) 927–936. [6] G.J. Breur, M.D. Lapierre, K. Kazmierczak, K.M. Stechuchak, G.P. McCabe, The domain of hypertrophic chondrocytes in growth plates growing at different rates Calcif, Tissue Int. 61 (1997) 418–425. [7] C.E. Farnum, A. Nixon, A.O. Lee, D.T. Kwan, L. Belanger, N.J. Wilsman, Quantitative three-dimensional analysis of chondrocytic kinetic responses to short-term stapling of the rat proximal tibial growth plate, Cells Tissues Organs 167 (2000) 247–258. [8] R.T. Ballock, R.J. O'Keefe, Current concepts review: The biology of the growth plate, J. Bone Jt. Surg.-Series A 85 (2003) 715–726. [9] P. Radhakrishnan, N.T. Lewis, J.J. Mao, Zone-specific micromechanical properties of the extracellular matrices of growth plate cartilage, Ann. Biomed. Eng. 32 (2004) 284–291. [10] M. Hauge Bünger, M. Foss, K. Erlacher, M. Bruun Hovgaard, J. Chevallier, B. Langdahl, et al., Nanostructure of the neurocentral growth plate: Insight from scanning small angle X-ray scattering, atomic force microscopy and scanning electron microscopy, Bone 39 (2006) 530–541. [11] G.S. Dodds, Row formation and other types of arrangement of cartilage cells in endochondral ossification, Anat. Rec. 46 (1930) 385–399. [12] A. Aszodi, E.B. Hunziker, C. Brakebusch, R. Fässler, β1 integrins regulate chondrocyte rotation, G1 progression, and cytokinesis, Genes Dev. 17 (2003) 2465–2479. [13] E.M. Darling, R.E. Wilusz, M.P. Bolognesi, S. Zauscher, F. Guilak, Spatial mapping of the biomechanical properties of the pericellular matrix of articular cartilage measured in situ via atomic force microscopy, Biophys. J. 98 (2010) 2848–2856. [14] D.M. Allen, J.J. Mao, Heterogeneous nanostructural and nanoelastic properties of pericellular and interterritorial matrices of chondrocytes by atomic force microscopy, J. Struct. Biol. 145 (2004) 196–204. [15] E.B. Hunziker, Mechanism of longitudinal bone growth and its regulation by growth plate chondrocytes, Microsc. Res. Tech. 28 (1994) 505–519. [16] C.A. Poole, S. Ayad, R.T. Gilbert, Chondrons from articular cartilage. V. Immunohistochemical evaluation of type VI collagen organisation in isolated chondrons by light, confocal and electron microscopy, J. Cell Sci. 103 (1992) 1101–1110. [17] K.J. Noonan, E.B. Hunziker, J. Nessler, J.A. Buckwalter, Changes in cell, matrix compartment, and fibrillar collagen volumes between growth-plate zones, J. Orthop. Res. 16 (1998) 500–508. [18] M. Shakibaei, C. Csaki, A. Mobasheri, Diverse roles of integrin receptors in articular cartilage, Adv. Anat. Embryol. Cell Biol. 197 (2008) 1–60. [19] C. Lee-Thedieck, N. Rauch, R. Fiammengo, G. Klein, J.P. Spatz, Impact of substrate elasticity on human hematopoietic stem and progenitor cell adhesion and motility, J. Cell Sci. 125 (2012) 3765–3775.

13

[20] D.E. Discher, D.J. Mooney, P.W. Zandstra, Growth factors, matrices, and forces combine and control stem cells, Science 324 (2009) 1673–1677. [21] A.K. Williamson, A.C. Chen, R.L. Sah, Compressive properties and function-composition relationships of developing bovine articular cartilage, J. Orth. Res. 19 (2001) 1113–1121. [22] S. Akizuki, V.C. Mow, F. Muller, J.C. Pita, D.S. Howell, D.H. Manicourt, Tensile properties of human knee joint cartilage: I. Influence of ionic conditions, weight bearing, and fibrillation on the tensile modulus, J. Orthop. Res. 4 (1986) 379–392. [23] L.A. Setton, C.H. Perry, M.A. LeRoux, J.Y. Wang, D.S. Howell, H.S. Cheung, Altered mechanics of tibial cartilage following joint immobilization in a canine model American society of mechanical engineers, Bioeng. Div. (Publ.) BED 42 (1999) 71–72. [24] R.U. Kleeman, D. Krocker, A. Cedrano, J. Tuischer, G.N. Duda, Altered cartilage mechanics and histology in knee osteoarthritis: Relation to clinical assessment (ICRS Grade) Osteoarthr, Cartil 13 (2005) 958–963. [25] I. Villemure, I.A.F. Stokes, Growth plate mechanics and mechanobiology. A survey of present understanding, J. Biomech. 42 (2009) 1793–1803. [26] J.H. Henderson, D.R. Carter, Mechanical induction in limb morphogenesis: the role of growth-generated strains and pressures, Bone 31 (2002) 645–653. [27] R.P. Gould, L. Selwood, A. Day, L. Wolpert, The mechanism of cellular orientation during early cartilage formation in the chick limb and regenerating amphibian limb, Exp. Cell Res. 83 (1974) 287–296. [28] J.L. Allen, M.E. Cooke, T. Alliston, ECM stiffness primes the TGFβ pathway to promote chondrocyte differentiation, Mol. Biol. Cell 23 (2012) 3731–3742. [29] J. O'Sullivan, S. D'Arcy, F.P. Barry, J.M. Murphy, C.M. Coleman, Mesenchymal chondroprogenitor cell origin and therapeutic potential, Stem Cell Res. Ther. 2 (2011), http://dx. doi.org/10.1186/scrt49. [30] I. Villemure, L. Cloutier, J.R. Matyas, N.A. Duncan, Nonuniform strain distribution within rat cartilaginous growth plate under uniaxial compression, J. Biomech. 40 (2007) 149–156. [31] K. Sergerie, M.-O. Lacoursière, M. Lévesque, I. Villemure, Mechanical properties of the porcine growth plate and its three zones from unconfined compression tests, J. Biomech. 42 (2009) 510–516. [32] R. Wosu, K. Sergerie, M. Lévesque, I. Villemure, Mechanical properties of the porcine growth plate vary with developmental stage, Biomech. Model. Mechanobiol. 11 (2012) 303–312. [33] I.M. Weiss, F. Lüke, N. Eichner, C. Guth, H. ClausenSchaumann, On the function of chitin synthase extracellular domains in biomineralization, J. Struct. Biol. 183 (2013) 216–225. [34] H. Clausen-Schaumann, M. Rief, C. Tolksdorf, H.E. Gaub, Mechanical Stability of Single DNA Molecules, Biophys. J. 78 (2000) 1997–2007. [35] M. Grandbois, M. Beyer, M. Rief, H. Clausen-Schaumann, H.E. Gaub, How strong is a covalent bond, Science 283 (1999) 1727–1730. [36] E. Sariisik, D. Docheva, D. Padula, C. Popov, J. Opfer, M. Schieker, et al., Probing the interaction forces of prostate cancer cells with collagen I and bone marrow derived stem cells on the single cell level, PLoS One 8 (2013), e57706. [37] D. Docheva, D. Padula, M. Schieker, H. ClausenSchaumann, Effect of collagen I and fibronectin on the adhesion, elasticity and cytoskeletal organization of prostate


14

[38]

[39]

[40]

[41]

[42]

[43]

[44]

[45]

[46]

[47]

[48] [49] [50]

[51]

[52]

[53]


cancer cells, Biochem. Biophys. Res. Commun. 402 (2010) 361–366. D. Docheva, D. Padula, C. Popov, W. Mutschler, H. ClausenSchaumann, M. Schieker, Researching into the cellular shape, volume and elasticity of mesenchymal stem cells, osteoblasts and osteosarcoma cells by atomic force microscopy, J. Cell. Mol. Med. 12 (2008) 537–552. J. Friedrichs, K.R. Legate, R. Schubert, M. Bharadwaj, C. Werner, D.J. Müller, et al., A practical guide to quantify cell adhesion using single-cell force spectroscopy, Methods 60 (2013) 169–178. L.A. Chtcheglova, P. Hinterdorfer, Atomic force microscopy functional imaging on vascular endothelial cells, Methods Mol. Biol. 931 (2013) 331–344. J. Domke, S. Dannöhl, W.J. Parak, O. Müller, W.K. Aicher, M. Radmacher, Substrate dependent differences in morphology and elasticity of living osteoblasts investigated by atomic force microscopy, Colloids Surf. B: Biointerfaces 19 (2000) 367–379. H.P. Riessen, R.D. Linley, I. Altshuler, M. Rabus, T. Söllradl, H. Clausen-Schaumann, et al., Changes in water chemistry can disable plankton prey defenses proc, Proc. Natl. Acad. Sci. U. S. A. 109 (2012) 15377–15382. M. Stolz, R. Gottardi, R. Raiteri, S. Miot, I. Martin, R. Imer, et al., Early detection of aging cartilage and osteoarthritis in mice and patient samples using atomic force microscopy (nature nanotechnology (2009) 4 (186–192) Nat, Nanotechnology 5 (2010) 821. L. Han, A.J. Grodzinsky, C. Ortiz, Nanomechanics of the cartilage extracellular matrix, Annu. Rev. Mater. Res. 41 (2011) 133–168. M. Stolz, R. Raiteri, A.U. Daniels, M.R. VanLandingham, W. Baschong, U. Aebi, dynamic elastic modulus of porcine articular cartilage determined at two different levels of tissue organization by indentation-type atomic force microscopy, Biophys. J. 86 (2004) 3269–3283. M. Stolz, R. Gottardi, R. Raiteri, S. Miot, I. Martin, R. Imer, et al., Early detection of aging cartilage and osteoarthritis in mice and patient samples using atomic force microscopy (nature nanotechnology (2009) 4 (186–192) Nat, Nanotechnology 5 (2009) 821. M. Loparic, D. Wirz, A.U. Daniels, R. Raiteri, M.R. Vanlandingham, G. Guex, et al., Micro- and nanomechanical analysis of articular cartilage by indentation-type atomic force microscopy: validation with a gel-microfiber composite, Biophys. J. 98 (2010) 2731–2740. H.J. Butt, M. Jaschke, Calculation of thermal noise in atomic force microscopy, Nanotechnology 6 (1995) 1–7. J.L. Hutter, J. Bechhoefer, Calibration of atomic-force microscope tips, Rev. Sci. Instrum. 64 (1993) 1868–1873. S.A. Chizhik, K. Wierzcholski, A.V. Trushko, M.A. Zhytkova, A. Miszczak, Properties of Cartilage on Micro- and Nanolevel, Adv. Tribology 2010 (2010), http://dx.doi.org/10.1155/ 2010/243150. M. Radmacher, Measuring the elastic properties of living cells by the atomic force microscope, Methods Cell Biol. 2002 (2002) 67–90. C.H. Liu, M. Singh, J. Li, Z. Han, C. Wu, S. Wang, et al., Quantitative assessment of hyaline cartilage elasticity during optical clearing using optical coherence elastography, Sovremennye Tehnol. Med. 7 (2015) 44–51. P. Zhu, M. Fang, Nano-Morphology of cartilage in hydrated and dehydrated conditions revealed by atomic force microscopy, J. Phys. Chem. Biophys. 2 (2012), http://dx.doi.org/10. 4172/2161–0398.1000106.

[54] J. Xu, P. Zhu, M.D. Morris, A. Ramamoorthy, Solid-State NMR Spectroscopy provides atomic-level insights into the dehydration of cartilage, J. Phys. Chem. B 115 (2011) 9948–9954. [55] J. Domke, M. Radmacher, Measuring the elastic properties of thin polymer films with the atomic force microscope, Langmuir 14 (1998) 3320–3325. [56] H.O.B. Gautier, A.J. Thompson, S. Achouri, D.E. Koser, K. Holtzmann, E. Moeendarbary, et al., Atomic force microscopy-based force measurements on animal cells and tissues, Methods Cell Biol. 125 (2015) 211–235. [57] P. Martin, Tissue patterning in the developing mouse limb, Int. J. Dev. Biol. 34 (1990) 323–336. [58] F. Mwale, E. Tchetina, C.W. Wu, A.R. Poole, The assembly and remodeling of the extracellular matrix in the growth plate in relationship to mineral deposition and cellular hypertrophy: An in situ study of collagens II and IX and proteoglycan, J. Bone Miner. Res. 17 (2002) 275–283. [59] Y. Li, A.T. Dudley, Noncanonical frizzled signaling regulates cell polarity of growth plate chondrocytes, Development 136 (2009) 1083–1092. [60] B. Gao, H. Song, K. Bishop, G. Elliot, L. Garrett, M.A. English, et al., Wnt signaling gradients establish planar cell polarity by inducing Vangl2 phosphorylation through Ror2 Dev, Cell 20 (2011) 163–176. [61] M.A. McLeod, R.E. Wilusz, F. Guilak, Depth-dependent anisotropy of the micromechanical properties of the extracellular and pericellular matrices of articular cartilage evaluated via atomic force microscopy, J. Biomech. 46 (2013) 586–592. [62] L.G. Alexopoulos, L.A. Setton, F. Guilak, The biomechanical role of the chondrocyte pericellular matrix in articular cartilage, Acta Biomater 1 (2005) 317–325. [63] G.M. Lee, B. Johnstone, K. Jacobson, B. Caterson, The dynamic structure of the pericellular matrix on living cells, J. Cell Biol. 123 (1993) 1899–1907. [64] W. Knudson, C.B. Knudson, Assembly of a chondrocyte-like pericellular matrix on non-chondrogenic cells: Role of the cell surface hyaluronan receptors in the assembly of a pericellular matrix, J. Cell Sci. 99 (1991) 227–235. [65] C.B. Knudson, Hyaluronan receptor-directed assembly of chondrocyte pericellular matrix, J. Cell Biol. 120 (1993) 825–834. [66] L.G. Alexopoulos, M.A. Haider, T.P. Vail, F. Guilak, Alterations in the mechanical properties of the human chondrocyte pericellular matrix with osteoarthritis, J. Biomech. Eng. 125 (2003) 323–333. [67] J.M. Clark, A. Norman, H. Nötzli, Postnatal development of the collagen matrix in rabbit tibial plateau articular cartilage, J. Anat. 191 (1997) 215–227. [68] C.A. Poole, M.H. Flint, B.W. Beaumont, Chondrons in cartilage: Ultrastructural analysis of the pericellular microenvironment in adult human articular cartilages, J. Orthop. Res. 5 (1987) 509–522. [69] S.W. Li, D.J. Prockop, H. Helminen, R. Fassler, T. Lapvetelainen, K. Kiraly, et al., Transgenic mice with targeted inactivation of the Col2a1 gene for collagen II develop a skeleton with membranous and periosteal bone but no endochondral bone, Genes Dev. 9 (1995) 2821–2830. [70] A. Aszodi, D. Chan, E. Hunziker, J.F. Bateman, R. Fässler, Collagen II is essential for the removal of the notochord and the formation of intervertebral discs, J. Cell Biol. 143 (1998) 1399–1412. [71] D. Chan, W.G. Cole, C.W. Chow, S. Mundlos, J.F. Bateman, A COL2A1 mutation in achondrogenesis type II results in the



replacement of type II collagen by type I and III collagens in cartilage, J. Biol. Chem. 270 (1995) 1747–1753. [72] D.R. Keene, J.T. Oxford, N.P. Morris, Ultrastructural localization of collagen types II, IX, and XI in the growth plate of human rib and fetal bovine epiphyseal cartilage: Type XI collagen is restricted to thin fibrils, J. Histochem. Cytochem. 43 (1995) 967–979. [73] X. Huang, D.E. Birk, P.F. Goetinck, Mice lacking matrilin-1 (cartilage matrix protein) have alterations in type II collagen fibrillogenesis and fibril organization, Dev. Dyn. 216 (1999) 434–441. [74] C. Nicolae, Y.P. Ko, N. Miosge, A. Niehoff, D. Studer, L. Enggist, et al., Abnormal collagen fibrils in cartilage of matrilin-1/matrilin-3-deficient mice J, Biol. Chem. 282 (2007) 22163–22175.

15

[75] J.M. Lane, C. Weiss, Review of articular cartilage collagen research, Arthritis Rheum. 18 (1975) 553–562. [76] S. Byers, C.J. Handley, D.A. Lowther, Transport of neutral amino acids by adult articular cartilage, Connect. Tissue Res. 11 (1983) 321–330. [77] T.K. Lngsjö, A.I. Vasara, M.M. Hyttinen, M.J. Lammi, A. Kaukinen, H.J. Helminen, et al., Quantitative analysis of collagen network structure and fibril dimensions in cartilage repair with autologous chondrocyte transplantation, Cells Tissues Organs 192 (2010) 351–360. [78] D.R. Eyre, I.R. Dickson, K. Van Ness, Collagen cross-linking in human bone and articular cartilage. Age-related changes in the content of mature hydroxypyridinium residues, Biochem. J. 252 (1988) 495–500.
