Skip to main page content

• HOME
• Current Issue
• Archive
• Alerts
• About ASM
• Contact us
• Tech Support
• Journals.ASM.org

This Article Abstract Full Text (PDF) Supplemental material Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Richardson, S. H. Articles by Kadler, K. E. Search for Related Content PubMed PubMed Citation Articles by Richardson, S. H. Articles by Kadler, K. E.

 Previous Article  |  Next Article 

Molecular and Cellular Biology, September 2007, p. 6218-6228, Vol. 27, No. 17
0270-7306/07/$08.00+0     doi:10.1128/MCB.00261-07
Copyright © 2007, American Society for Microbiology. All Rights Reserved.

Tendon Development Requires Regulation of Cell Condensation and Cell Shape via Cadherin-11-Mediated Cell-Cell Junctions ^,

Susan H. Richardson, Tobias Starborg, Yinhui Lu, Sally M. Humphries, Roger S. Meadows, and Karl E. Kadler^*

Wellcome Trust Centre for Cell-Matrix Research, Faculty of Life Sciences, University of Manchester, Michael Smith Building, Oxford Road, Manchester M13 9PT, United Kingdom

Received 13 February 2007/ Returned for modification 21 March 2007/ Accepted 24 May 2007

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
The ability of tendon to transmit forces from muscle to bone is directly attributable to an extracellular matrix (ECM) containing parallel bundles of collagen fibrils. Although the biosynthesis of collagen is well characterized, how cells deposit the fibrils in regular parallel arrays is not understood. Here we show that cells in the tendon mesenchyme are nearly cylindrical and are aligned side by side and end to end along the proximal-distal axis of the limb. Using three-dimensional reconstruction electron microscopy, we show that the cells have deep channels in their plasma membranes and contain bundles of parallel fibrils that are contiguous from one cell to another along the tendon axis. A combination of electron microscopy, microarray analysis, and immunofluorescence suggested that the cells are held together by cadherin-11-containing cell-cell junctions. Using a combination of RNA interference and electron microscopy, we showed that knockdown of cadherin-11 resulted in cell separation, loss of plasma membrane channels, and misalignment of the collagen fibrils in the ECM. Our results show that tendon formation in the developing limb requires precise regulation of cell shape via cadherin-11-mediated cell-cell junctions and coaxial alignment of plasma membrane channels in longitudinally stacked cells.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Tendon formation begins at embryonic day 14.5 in the mouse and at day 10 in the chick, when the extracellular matrix (ECM) becomes populated with narrow-diameter collagen fibrils (for reviews, see references 6 and 7 and references therein). Toward the end of embryonic development, tendons are readily identified by a dense ECM containing bundles of thick collagen fibrils that are parallel to the tendon long axis. Although gene expression and collagen fibril assembly during limb development have been well documented, little is known about the cellular and molecular mechanisms of tendon morphogenesis and, in particular, how the parallelism of the collagen fibrils is established.

Electron microscope (EM) studies of embryonic chick tendon show the presence of parallel collagen fibrils in extracellular spaces between cells (3) and sometimes within fibripositors (7). Inspection of the EM images in these publications shows junctions between plasma membranes of adjacent cells. Similar junctions have been described for cultured fibroblasts (9, 13, 14, 23, 31) and in situ (16) as regions of intense staining. The occurrence of these junctions in embryonic tendon prompted us to determine if the junctions have a role in tendon organization.

Cell-cell junctions have thus far been characterized only by their roles in cell adhesion, communication, development, differentiation, and tissue homeostasis (for a review, see reference 5). Adhesive cell interactions organize the cells into multicellular tissues (5) and provide regions for dynamic signaling resulting in the activation of pathways controlling cytoskeletal and nuclear dynamics. Cell-cell junctions provide spatial cues to generate cell surface polarity and enable cells to sense and respond to their local environment (4). However, a role for cell-cell junctions in ECM organization has not been shown.

In this study, we show that knockdown of cadherin-11 in embryonic tendon results in a loss of cell-cell contact and disruption of the ECM, thus showing a new role for cadherin-11 in organizing the ECM. Cadherin-11 is a member of a family of structurally related Ca²⁺-dependent transmembrane proteins which interact via extracellular domains on opposing membranes (for reviews of cadherins, see references 26, 28, and 30). ß-Catenin links the cadherin molecule to -catenin, which is attached to the actin cytoskeleton via -actinin and vinculin (25; for a review, see reference 27). Cell-cell adhesion is a critical step in tissue morphogenesis in which cadherin molecules have roles in influencing cell differentiation, growth, and behavior. In this regard, cadherin-11 was recently shown to be a discriminating factor between articular and growth plate chondrocytes (21) and is essential for the development of the synovium (19).

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Source of materials. Phalloidin-fluorescein isothiocyanate (phalloidin-FITC) was obtained from Sigma-Aldrich. Normal goat serum was obtained from Jackson Immuno Research Laboratories Inc. Immunolabeled sections were examined with a Leica light microscope.

Primer and siRNA sequences. The following primers and small interfering RNA (siRNA) sequences were used: chicken ß-actin forward primer, GCC ACA GCT GCC TCT AGC TCT; reverse primer, CAG CAC TGT GTT GGC ATA CAG; chicken cadherin-11 forward primer, GTC ACA CTG ACC TTG AAA GAT; reverse primer, GAT TGC TTC GAG CAT TCT CAC; chicken beta-actin siRNA 411, GCCCAGCAACAGAAAGGAATT; chicken beta-actin siRNA 648, GCUCUCUUUGCUUCUUUAATT; chicken cadherin-11 siRNA 4164, GGAGGACAAUUGUUUACAUTT; chicken cadherin-11 siRNA 6427, GCGACAACAGAUUCUGAUUTT; chick cadherin-11 scrambled siRNA 4164S, GGAAACAUGUUAUUGGCAUTT; and chick cadherin-11 scrambled siRNA 6427S, GCGAAACAGUUUCGACAUUTT.

Preparation for EM. Embryonic mouse tails (15.5 days postcoitus [dpc]) and chick (White Leghorn) metatarsal tendons (13 days) were prepared for transmission EM by being embedded in Spurr's resin as described previously (7). Several series of consecutive sections (typically 120 nm thick) were collected and examined in a Tecnai 12 BioTwin transmission EM operated at a 100-keV accelerating voltage. Images were collected on film at a magnification of x2,900.

3D reconstruction. Transmission EM negatives were scanned using an Imacon Flextight 848 scanner and saved in TIFF format. The digitized images were imported into IMOD software for three-dimensional (3D) reconstruction as described previously (18). The images were aligned based on the arrangement of the individual cells. Features of interest were traced to produce a contour map of specific regions. To create models of the areas of interest, the contours were connected in a 3D volume, with the spacing between adjacent contours dictated by the thickness of the sections.

Isolation of embryonic chick tendon cells. Metatarsal tendons were dissected from day 13 embryonic chicks and rinsed briefly in phosphate-buffered saline (PBS). Fibroblasts were isolated from the tendons by digestion with trypsin (37,000 units) and bacterial collagenase (522 units) in Dulbecco's modified Eagle's medium (DMEM) at 37°C for 2 h. Cells were filtered through a cell strainer, collected by centrifugation (1,500 rpm for 4 min), and washed three times in PBS. Cells were resuspended in DMEM containing 1% L-glutamine, 1% penicillin-streptomycin, and 10% fetal calf serum and seeded into flasks. Cells were grown to 80% confluence at 37°C in 5% carbon dioxide.

Immunofluorescence. Cryosections of chick tendon (9 µm) were fixed in 100% acetone or methanol at –20°C for 10 min and blocked with 10% normal goat serum in PBST (PBS supplemented with 0.1% Triton X-100) for 1 h. The sections were incubated in primary antibody diluted in 5% bovine serum albumin in PBST overnight at 4°C or for 1 h at 37°C, washed three times for 5 min each with PBST, and incubated with goat anti-mouse-Cy3 (1:1,000) for 1 h at room temperature. Tissue was washed three times for 5 min each with PBST and mounted with Vectashield mounting medium containing DAPI (4',6'-diamidino-2-phenylindole). Control experiments involved omission of the relevant primary antibody and incubation with the appropriate secondary antibody in addition to incubation with an inappropriate secondary antibody. Little or no fluorescence was detected in the control sections.

Cells on coverslips were briefly rinsed three times with PBS with calcium ions and fixed with 1% paraformaldehyde in 0.1 M HEPES (pH 7.4) for 15 min at room temperature. After being permeabilized and blocked with 5% skim milk in PBS, cells were incubated for 1 h at room temperature with anti-cadherin-11 antibody (1:100) in 1% skim milk in PBS. They were then rinsed with PBS, incubated with secondary goat anti-mouse Cy3 antibody in 1% milk in PBS containing phalloidin-FITC for 1 h at room temperature, rinsed, and mounted with Vectashield with DAPI.

RNA isolation from whole tendons. Metatarsal tendons were dissected from day 13 embryonic chicks and rinsed briefly in PBS. TRIzol reagent (Invitrogen, Carlsbad, CA) was added and the tissue rapidly frozen in liquid nitrogen prior to disruption using a Mikro-Dismembrator (Sartorius) (twice at 2,000 rpm for 90 seconds). RNAs were extracted from the tissue following the manufacturer's instructions (Invitrogen, Carlsbad, CA).

RNA isolation from cultured chick tenocytes and RT-PCR. Total RNA was extracted from subconfluent (80%) cultures of fibroblasts by using TRIzol reagent followed by DNase treatment. cDNA was transcribed from 2 µg of RNA with TaqMan reverse transcriptase (RT) polymerase (Applied Biosystems), using an oligo(dT)₁₆ primer. RT-PCR analysis was performed on mRNAs which had been isolated from both cultured cells and fresh tendons by using 21-mer primers complementary to ß-actin and cadherin-11 from the chick. Amplification of the correctly sized products was verified by electrophoresis on a 2% Tris-borate-EDTA gel. The identities of the product were confirmed by DNA sequencing.

Microarray analysis of mRNAs from 13-day embryonic chick tendon. Total RNA was isolated from day 13 embryonic chick metatarsal tendons as described above. RNAs (15 µg) were used to produce biotinylated cRNA samples, which were hybridized to chicken oligonucleotide arrays (Affymetrix, Inc., Santa Clara, CA) containing 32,773 gene transcripts, corresponding to over 28,000 chicken genes.

Sequencing of RT-PCR fragments. PCR products (5 ng) were sequenced using a BigDye Terminator v3.1 cycle sequencing kit (Applied Biosystems). Samples were placed in a thermal cycler under the following conditions: initial denaturation was performed at 96°C for 1 min, followed by 25 cycles of 96°C for 10 seconds, 50°C for 5 seconds, and 60°C for 4 min. Samples were precipitated with ethanol-sodium acetate prior to analysis.

RNA interference (RNAi) on cultured chick tendon cells with ß-actin and cadherin-11. Tendon cells were cultured and used at passage number 3 or under. Cells were rinsed once with PBS, trypsinized, and pelleted (1,200 rpm, 4 min). Each pellet of 1.5 x 10⁶ cells was resuspended in 100 µl Nucleofector solution plus supplements containing 1.5 µg of siRNA at room temperature (Amaxa GmbH, Cologne, Germany). The suspension (100 µl) was placed into an Amaxa cuvette and electroporated using the proprietary program U23 (according to the manufacturer's instructions). Immediately after, 0.5 ml of growth medium at 37°C was added to the cells in the cuvette and the cells were transferred to one well of a six-well plate containing a coverslip and 2 ml of growth medium at 37°C. Cells were incubated for up to 48 h under standard cell culture conditions.

RNAi of cells in chick tendon. Metatarsal tendons were dissected from day 13 embryonic chicks and placed into growth medium at 37°C. Two tendons were placed in 100 µl Nucleofector reagent (from basic fibroblast Nucleofector kit; Amaxa) with supplements containing cadherin-11 siRNAs (1.5 µg) or ß-actin siRNAs (1.5 µg). The tissue was electroporated using the proprietary program U23 and then incubated at 37°C in growth medium for 5 h.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Shape and organization of cells in tendon mesenchyme. To determine the spatial organization of cells in the early tendon, we used serial section reconstruction EM to map the shape and relative positions of the cells. We chose to study chick metatarsal tendons at day 13 postfertilization and mouse tail tendons at 15.5 dpc because these represent stages in tendon when the ECM is being deposited. Figure 1A shows 1 of 254 consecutive sections through mouse 15.5 dpc tail tendon in which cells are contoured in different colors. We used the IMOD suite of software (18) to generate 3D reconstructions of the cells. Figure 1B shows a 3D reconstruction of a volume of tendon of 13 µm in diameter by 18 µm in length. The cells are cylindrical and packed closely in side-by-side register. Furthermore, they have plasma membrane protrusions that generate intercellular spaces containing ECM. The plasma membrane protrusions form a physical boundary between collagen fibril bundles in the ECM. Only a few ends of cells were contained within the 3D reconstruction, showing that most cells exceeded 18 µm in length, and collagen fibril bundles were contiguous from one cell to another along the tendon axis (Fig. 2). Higher-magnification images showed the presence of darkly stained regions at the point of contact between plasma membranes (Fig. 3), both involving two cells and the plasma membrane of the same cell. These regions of dense staining were identical in appearance to the cell-cell junctions previously identified (9, 13, 14, 23, 31). The junctions appeared to hold the cells in close proximity and to stabilize the plasma membrane channels.

View larger version (68K): [in this window] [in a new window]

FIG. 1. EM of mouse embryonic tendon. (A) Transmission electron micrograph of 15.5 dpc mouse tail tendon. The image shows 1 of a series of 254 consecutive transverse sections through a single tendon fascicle. The membranes of some cells are shown in color. (B) 3D reconstruction of the fascicle. A single EM image (semitransparent) is superimposed to aid visualization.

View larger version (73K): [in this window] [in a new window]

FIG. 2. Cells contain plasma membrane channels that are contiguous from one cell to another. The 3D reconstruction was generated from EM images of 112- by 120-nm-thick serial (consecutive) sections of 13-day embryonic chick tendon. (Left) View of one cell (dark green) showing bundles of collagen fibrils (gray) located in plasma membrane channels that are aligned with the tendon axis. (Right) Side view of the ends of two cells (green and pink). The ends of cells overlap and the plasma membrane channels in one cell are aligned with the plasma membrane channels in the other cell.

View larger version (120K): [in this window] [in a new window]

FIG. 3. EM images of embryonic tendon showing plasma membrane channels. (A to E) Thirteen-day embryonic chick metatarsal tendon. (F and G) Mouse 15.5 dpc embryonic tail tendon. (A) Six plasma membrane channels (numbered), each containing a bundle of collagen fibrils, positioned between two cells. The boundaries of the channels are provided by one cell. (B and C) Plasma membranes from the same cell contribute to defining the boundaries of plasma membrane channels. (D and E) Plasma membranes from different cells contribute to defining the boundaries of plasma membrane channels. Cell-cell junctions are shown within broken circles. (F) Two plasma membrane leaflets can be seen wrapped around a bundle of collagen fibrils. A darkly stained cell-cell junction (upper broken circle) holds the leaflets together. Another cell-cell junction (lower broken circle) is also visible. A fibripositor (arrow) is located within the plasma membrane channel. (G) A plasma membrane channel is formed between adjacent cells. A cluster of three darkly stained cell-cell junctions is apparent next to a fibripositor (arrow).

Cell-cell junctions are discontinuous strips. From examination of single images, it was not possible to determine the prevalence of the cell-cell junctions in embryonic tendon. Therefore, we obtained several serial sections of chick and mouse tendons and generated 3D reconstructions of the cell-ECM interface. Using this approach, it was possible to visualize the distribution of the cell-cell junctions. A region of one reconstruction from a mouse tail tendon is shown in Fig. 4. The electron micrographs in Fig. 4A and B show a darkly stained cell-cell junction (circled) between plasma membrane channels, with each channel containing a bundle of collagen fibrils (BL1 and BL2).

View larger version (112K): [in this window] [in a new window]

FIG. 4. EM 3D reconstruction of two plasma membrane channels in mouse tendon. (A) Transmission electron micrograph showing the interface between two cells in mouse 15.5 dpc tail tendon. Two collagen fibril bundles are labeled BL1 and BL2. A cell-cell junction is shown within a broken circle. (B) Four cells are color coded red, blue, green, and purple. The cell-cell junction between the blue cell and the purple cell is marked with red and circled. The position of the cell-cell junction is shown in 3D in panel C. (C) Serial section reconstruction of collagen fibril bundles BL1 and BL2. The upper part of the plasma membrane channel containing BL1 is provided by the blue cell only. The cell-cell junctions (shown in red) are strip-like in shape and are parallel to the tendon long axis (top to bottom in the reconstruction). (D) The plasma membrane from the red cell contributes to forming BL1. The plasma membrane from the purple cell contributes to forming BL2. (E) Expanded view of the interfaces between the blue, red, green, and purple cells. Dotted lines indicate the connectivity of cell-cell junctions. The red, green, and purple cells all make connections with the blue cell in forming BL1 and BL2. Interactions between the green, red, and purple cells are not shown. Bars, 1 µm.

For descriptive purposes, the plasma membranes of the four cells adjacent to BL1 and BL2 were color coded and the cells described as blue, red, green, and purple. The 3D reconstruction of BL1 and BL2 shows that the plasma membrane of the blue cell generates two parallel channels that contain BL1 and BL2 (Fig. 4C and D). In the case of BL1, the blue cell pushes out two distinctive plasma membrane leaflets that wrap around the upper part of the bundle. In the lower part of the reconstruction, the plasma membrane of the blue cell is seen forming the sides of the channel. In contrast, the plasma membrane of the blue cell completely envelops BL2 in the lower part of the reconstruction. Examination of the red and purple cells (Fig. 4D) shows that the purple cell contributes to the channel that encompasses BL1 and BL2. Furthermore, the channel that contains BL1 also has contributions of plasma membrane from the red cell. Figure 4E shows an expanded view of the four cells in which the cell-cell junctions are shown as connecting lines. This shows that a cell makes numerous connections to its neighbors and uses cell-cell junctions to define the shape and number of plasma membrane channels.

Figure 4C shows the 3D distribution of cell-cell junctions on the blue cell, in contact with the red, green, and purple cells. The junctions are depicted in red. The reconstruction shows that some junctions extend over 2 to 4 consecutive sections (140 to 280 nm) while others are continuous through 50 or more sections. Therefore, the junctions can exceed 4.25 µm in length. Of particular note, the strip-like junctions are parallel to the tendon long axis.

Cell-cell junctions contain cadherin-11. In initial studies, we examined the calcium ion dependency of the junctions in chick and mouse tendons by incubating the tendons with EGTA. As shown in the supplemental material, the junctions were absent in calcium-ion-depleted tendons. Although EGTA was a crude treatment, the observed loss of junctions suggested that the junctions were adherens junctions. In initial experiments to characterize the junctions, we performed a gene array analysis of 13-day chick tendon. Table 1 contains a subset of the data for genes that encode proteins in cell-cell junctions, cytoskeletal proteins, and ECM proteins. The high level of expression of the 1(I) chain of collagen and decorin was indicative of the rapid rate of synthesis of collagen fibrils during this period of embryonic tendon development. The most conspicuous cell-cell junction gene was that for cadherin-11, as well as that for ß-catenin, which links the cadherin dimer to the cytoskeleton. These observations led us to consider cadherin-11 as a major component of the cell-cell junctions in embryonic tendon.

View this table: [in this window] [in a new window]

TABLE 1. Gene microarray analysis of day 13 embryonic chick metatarsal tendon

To confirm the presence of cadherin-11 mRNA in tendon, we performed an RT-PCR using mRNAs isolated from freshly dissected 13-day embryonic chick tendons. A PCR product of the expected size was detected (Fig. 5A, lane 3). DNA sequencing confirmed the identity of the product as cadherin-11. Similar studies using 13-day chick tendon cells showed that cadherin-11 expression is maintained in culture (Fig. 5A, lane 4). Primers for ß-actin were used in control reactions (Fig. 5A, lanes 1 and 2).

View larger version (47K): [in this window] [in a new window]

FIG. 5. RNAi knockdown of cadherin-11. (A) RT-PCR detected the presence of cadherin-11 mRNA in tendon cells. mRNAs were isolated from cultured 13-day chick tendon cells (lanes 1 and 3) or from 13-day chick embryonic tendon (lanes 2 and 4) and used as a template to detect mRNAs for ß-actin (lanes 1 and 2) and cadherin-11 (lanes 3 and 4). The identities of the PCR products were confirmed by DNA sequencing. M, size markers. (B) Immunofluorescence of 13-day embryonic chick tendon, using an anti-cadherin-11 antibody (red), shows punctate distribution of cadherin-11 within the fascicles (white arrowheads) and staining at the interfascicular boundaries. The positions of the nuclei are shown by DAPI staining (blue). (C to F) Immunofluorescence of cultured 13-day embryonic chick tendon cells before and after treatment with siRNA against cadherin-11. (C and D) Twenty-four and 48 h, respectively, after electroporation without siRNA. (E and F) Twenty-four and 48 h, respectively, after electroporation with siRNA. Nuclei were stained blue with DAPI, actin filaments were stained green with phalloidin-FITC, and cadherin-11 stained red. (G) RNA agarose gel showing two ribosomal subunits in RNAs isolated from cultured 13-day chick tendon fibroblasts. Lane 1, RNAs from cells electroporated in the absence of siRNA; lane 2, RNAs from cells electroporated with siRNA for cadherin-11; lane M, size markers. (H) RT-PCR detected a reduction in the expression of cadherin-11 mRNA in tendon cells after siRNA treatment. Lanes 1 and 2, PCR for beta-actin; lanes 3 and 4, PCR for cadherin-11. Lanes 1 and 3, RNAs from cells electroporated in the absence of siRNA; lanes 2 and 4, RNAs from cells electroporated in the presence of siRNA targeted to cadherin-11. The identities of the PCR products were confirmed by DNA sequencing. M, size markers.

Using a monoclonal antibody specific for chick cadherin-11, we carried out immunofluorescence studies on 13-day embryonic chick tendon. As shown in Fig. 5B, the plasma membranes of the cells reacted strongly. In particular, we noted a punctate distribution, which was consistent with the EM observations of the junctions. In further experiments, we used immunofluorescence to show the presence of cadherin-11-containing junctions in cultured cells. As shown in Fig. 5C and D, 13-day chick tendon cells establish cadherin-11 cell-cell junctions and maintain them for at least 48 h.

Knockdown of cadherin-11 in cultured tendon cells by using siRNA. The presence of cadherin-11 junctions in cultured cells provided an opportunity to test the efficacy of several siRNA sequences targeted against chick cadherin-11 mRNA. We isolated RNAs from control and siRNA-treated cells (Fig. 5G) and confirmed at least 50% knockdown of cadherin-11 mRNA (Fig. 5H) (see Materials and Methods for sequences). Immunofluorescence analysis of cultured cells electroporated with siRNA for cadherin-11 showed efficient knockdown of the protein (Fig. 5E). The absence of cadherin-11 was seen for up to 48 h after electroporation (Fig. 5F).

Knockdown of cadherin-11 in tendon by using siRNA. Having demonstrated the effectiveness of the RNAi approach in cultured cells, we wanted to know the effect of cadherin-11 knockdown on plasma membrane channels and ECM organization in vivo. In preliminary experiments, we electroporated 13-day embryonic chick tendon with a vector encoding green fluorescent protein (GFP). As shown in the supplemental material, electroporation was effective in introducing the vector DNA into cells in situ and the cells remained viable and expressed GFP. It is noteworthy that cells deep within the tendon tissue efficiently expressed GFP. Therefore, the electroporation protocol was effective at delivering nucleic acids to cells in tendon.

To assess the effects of electroporation on cell viability in tendons, we performed a mock electroporation in which siRNA was omitted. EM showed that the plasma membrane, nuclear membrane, and endoplasmic reticulum (indicative of active protein synthesis) membranes were intact, the nucleus had a normal appearance, the cells were closely packed with evidence of cell-cell junctions, and the ECM contained parallel bundles of collagen fibrils (Fig. 6A). Taken together, these observations suggested that electroporation alone had no observable effect on cell viability or on the organization of the ECM.

View larger version (160K): [in this window] [in a new window]

FIG. 6. Electron micrographs of chick tendon after electroporation with siRNAs against ß-actin and cadherin-11. (A) Control sample. The image shows a transverse section through day 13 embryonic chick tendon after incubation in DMEM for 5 h. (B) Transverse section through day 13 embryonic chick tendon taken after treatment with siRNA against ß-actin and incubation in DMEM for 5 h. (C and D) Transverse sections through day 13 embryonic chick tendon taken after treatment with siRNA against cadherin-11. (E and F) Transverse sections through day 13 embryonic chick tendon taken after treatment with scrambled siRNA against cadherin-11 (see Materials and Methods for sequences of siRNAs 4164 and 6427). Closed black arrows, plasma membrane channels; open white arrows, fibripositors; arrowheads, cell-cell junctions.

In a previous study, we had shown that depolymerization of the actin cytoskeleton by cytochalasin resulted in a loss of collagen fibril-containing fibripositors and a loss of parallelism of the tendon ECM (8). Therefore, to validate the RNAi approach, we electroporated tendons with siRNA targeted to ß-actin and examined the tendon tissue by EM (Fig. 6B). Electroporation with siRNA targeted to ß-actin resulted in a loss of parallelism of the collagen fibrils in the ECM, which agreed well with the previously observed loss of parallelism in tendon treated with cytochalasin (8).

To assess the contribution of cadherin-11 to cell-cell junction formation in embryonic tendon, we designed and used two siRNA sequences against cadherin-11 to knock down expression of the protein. Whole tendons from 13-day chick embryos were electroporated separately with siRNA 4164 and siRNA 6427 (see Materials and Methods for sequence information), and we examined the expression of cadherin-11 by immunofluorescence. The expression of cadherin-11 was restricted to the endotenon (the main body of the tendon) (Fig. 7A); no expression was detected in the outermost layer of cells (epitenon) (data not shown). The cadherin-11 staining was punctate and focused at the interface between groups of cells. This observation agrees well with the EM data in which several neighboring cells were shown to act in concert to make plasma membrane channels (Fig. 4E). It is noteworthy that the junctions run in discontinuous strips along the tendon axis, the cells exceed 18 µm in length, and therefore in a single, 7-µm transverse section only parts of individual junctions will be visualized. Expression of cadherin-11 was virtually undetectable in tendons electroporated with siRNAs against cadherin-11 (Fig. 7B).

View larger version (85K): [in this window] [in a new window]

FIG. 7. N-cadherin and cadherin-11 are expressed by different tendon cells. The images show the immunofluorescence of transverse sections of day 13 embryonic chick tendon, using antibodies against cadherin-11 (A and B) and N-cadherin (C and D), in the absence (A and C) and presence (B and D) of siRNA against cadherin-11 (4164).

To determine the effects of siRNA knockdown of cadherin-11 on junction formation, we prepared treated samples for EM. As shown in Fig. 6C and D, cadherin-11 knockdown resulted in disruption of cell-cell junctions, as the cells moved apart and there was a loss of the plasma membrane channels (Fig. 6C and D). A major effect was to disrupt the collagen fibril organization in the ECM. There were no noticeable effects on nuclear membrane integrity or the appearance of the endoplasmic reticulum. Furthermore, fibripositors containing collagen fibrils were present in treated samples, which was indicative of active collagen secretion (7). In control experiments, we electroporated tendon with scrambled siRNAs for cadherin-11 and prepared samples for EM. As shown in Fig. 6E and F, the cells were intact, plasma membrane channels were visible, cell-cell junctions were present, and collagen fibril-containing fibripositors were present in the plasma membrane channels.

N-cadherin and cadherin-11 are expressed by different cells of tendon. Although EM showed that knockdown of cadherin-11 resulted in a loss of junctions and in cells moving apart, we wanted to know if the absence of cadherin-11 resulted in up-regulation of N-cadherin. Immunofluorescence using an anti-N-cadherin antibody showed that N-cadherin was expressed by cells at the surface of the tendon (i.e., the epitenon) and not by cells in the main body of the tendon (i.e., the endotenon) (Fig. 7A). In samples electroporated with siRNAs against cadherin-11, there was no detectable up-regulation of N-cadherin in the endotenon and no apparent change in expression of N-cadherin in the epitenon (Fig. 7B).

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
In this study, we show a new function for cadherin-11 in forming the parallelism of the ECM of tendon. The mechanism of cadherin-11 function in tendon involves cell-cell junction-mediated cellular condensation and plasma membrane-mediated ECM assembly. The two phenomena (cell condensation and plasma membrane-ECM assembly) go hand in hand, which might be expected in the formation of a connective tissue with a highly organized ECM.

In a previous study, we showed that the parallelism of the collagen fibrils in tendon is associated with the occurrence of protrusions of the plasma membrane called fibripositors (7) and that removal of the fibripositors by cytochalasin resulted in some bundles of collagen fibrils being misaligned (8). Importantly, some bundles remained aligned and the cells remained in contact after treatment with cytochalasin. This implied to us that the close contact between cells contributes to the formation of collagen bundles and fibril alignment. Intrigued by these observations, we sought to understand the mechanism of cell-cell interaction in the embryonic tendon.

The first objective of this study was to determine the relationship of the collagen fibrils with the cell surface. We showed that at the onset of collagen fibril formation, the tendon cells had plasma membrane channels that were parallel to the tendon long axis, which agreed with earlier observations of extracellular compartments containing collagen fibrils (1-3). Building on these observations, we showed that plasma membranes from different cells or the same cell could form the channels. Furthermore, 3D reconstructions showed that the cells were nearly cylindrical and coaligned along the tendon long axis. Importantly, the cylindrical cells were stacked side by side with their plasma membrane channels in contact, thereby forming extracellular spaces that are a template for the formation of collagen fibril bundles. Therefore, the cells act as a community to coordinate the patterning of the ECM. Gurdon and colleagues described the community effect as a phenomenon in which cells must contact a sufficient number of neighbors to coordinate differentiation within a developing tissue (10-12). The collagen fibril-producing cells are in contact across the newly forming tendon. Therefore, the community effect appears to also apply to ECM patterning.

Our next series of experiments were focused on understanding the molecular mechanism for cell-cell condensation and stabilization of the plasma membrane channels. Microarray analysis was used to obtain a global profile of genes expressed in the early chick tendon. This analysis suggested that cadherin-11 was a highly expressed cell-cell junction protein. The expression of other cell-cell junction components, e.g., tight and gap junctions, was low. Also, the expression of N-cadherin was relatively low and approximately 15-fold less than that of cadherin-11 (see below). These data gave us the first indication that cadherin-11 could be important in generating cell condensation and plasma membrane channels. Immunofluorescence of embryonic tendon using an anti-cadherin-11 antibody confirmed the presence of cadherin-11 in tendon; the punctate staining was in agreement with EM observations of the junctions at focused points of contact between cells.

A valuable approach to understanding the mechanism of cadherin-11-mediated tendon ECM patterning is to remove the protein and examine the consequences on tendon cell adhesion and ECM organization. The cadherin-11 null mouse is healthy at birth and has reduced bone density in some parts of the skeleton, in particular calvaria and long-bone metaphyses (15). More recent studies have also shown that cadherin-11 is essential for formation of the synovial membrane (19), which is a tough connective tissue with a dense ECM of collagen fibrils. The cadherin-11 null mouse was not available to us for examination, and therefore we chose to use an RNAi approach to knock down expression of cadherin-11 in ex vivo tendon. This had the advantage of specifically targeting cadherin-11 during patterning of the ECM and overcame potential problems of compensation by other cadherins (as discussed in reference 17). N-cadherin and cadherin-11 are reported to be expressed by cells of mesenchymal origin (29). In fact, immunofluorescence studies using an anti-N-cadherin antibody showed that knockdown of cadherin-11 did not result in up-regulation of N-cadherin. During the course of these experiments, we showed that N-cadherin expression is restricted to the epitenon. Therefore, N-cadherin and cadherin-11 are expressed by different cells in the tendon, indicative of these two cadherins having distinct functions during embryonic tendon development. Of particular interest, cadherin-11 has been suggested to compensate for the absence of N-cadherin during limb mesenchymal chondrogenesis (20). However, N-cadherin compensation for cadherin-11 does not appear to occur in embryonic tendon.

We showed that cells isolated from tendons and placed in culture established numerous cell-cell contacts that contain cadherin-11. We were able to knock down cadherin-11 expression sufficiently for the protein not to be detected by immunofluorescence. We went on to show that cells could be electroporated in situ (i.e., within the tendon ECM) with a vector encoding GFP and that GFP could be visualized by fluorescence microscopy. These experiments successfully validated the introduction of siRNA into cells in intact tendon by electroporation. In a further experiment, we showed that siRNA knockdown of cadherin-11 in tendon resulted in the cells losing their cell-cell junctions, the cells moving apart, and the organization of the collagen fibrils in the ECM being disrupted.

The results presented here highlight the part played by cell-cell proteins in defining the architecture of the ECM and the organization of the tissue. In particular, cadherin-11 has a pivotal role in determining the shape and organization of the ECM in developing tendon. Further studies will determine the role of cadherin molecules in other ECM-rich tissues.

Further studies are also needed to explain the predominance of gap junctions in adult tendon (22, 24). In vivo studies of mature rat tendon showed that tendon fibroblasts are arranged in a 3D array linked via gap junctions. These arrays are seen as layers of flattened cells that form a cellular sheet around the tendon proper (endotenon) (22). The gap junctions in adult tendon have been implicated in the maintenance and coordination of tendon cell behavior in response to load. Moreover, the cells in culture have been shown to up-regulate gap junction expression in response to cyclic tensile load (24). Further studies will determine if the cadherin-11 junctions in embryonic tendon change composition during ageing, concomitant with a change in function from tissue patterning to tissue maintenance and repair.

 
We thank Leanne Wardleworth (Microarray Facility) for microarray studies and Cara Verdi for help with image analysis.

This work was supported by the Wellcome Trust and the Biotechnology and Biological Sciences Research Council (United Kingdom).

 
* Corresponding author. Mailing address: Wellcome Trust Centre for Cell-Matrix Research, Faculty of Life Sciences, University of Manchester, Michael Smith Building, Oxford Road, Manchester M13 9PT, United Kingdom. Phone: 44-161-275-5086. Fax: 44-161-275-1505. E-mail: karl.kadler{at}manchester.ac.uk

Published ahead of print on 11 June 2007.

Supplemental material for this article may be found at http://mcb.asm.org/.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 

1
1. Birk, D. E., and R. L. Trelstad. 1984. Extracellular compartments in matrix morphogenesis: collagen fibril, bundle, and lamellar formation by corneal fibroblasts. J. Cell Biol. 99:2024-2033.[Abstract/Free Full Text]
2
2. Birk, D. E., and R. L. Trelstad. 1986. Extracellular compartments in tendon morphogenesis: collagen fibril, bundle, and macroaggregate formation. J. Cell Biol. 103:231-240.[Abstract/Free Full Text]
3
3. Birk, D. E., and R. L. Trelstad. 1985. Fibroblasts create compartments in the extracellular space where collagen polymerizes into fibrils and fibrils associate into bundles. Ann. N. Y. Acad. Sci. 460:258-266.[CrossRef][Medline]
4
4. Boitano, S., Z. Safdar, D. G. Welsh, J. Bhattacharya, and M. Koval. 2004. Cell-cell interactions in regulating lung function. Am. J. Physiol. Lung Cell Mol. Physiol. 287:L455-L459.[Abstract/Free Full Text]
5
5. Braga, V. M., and M. S. Balda. 2004. Regulation of cell-cell adhesion. Semin. Cell Dev. Biol. 15:631-632.[CrossRef][Medline]
6
6. Canty, E. G., and K. E. Kadler. 2005. Procollagen trafficking, processing and fibrillogenesis. J. Cell Sci. 118:1341-1353.[Abstract/Free Full Text]
7
7. Canty, E. G., Y. Lu, R. S. Meadows, M. K. Shaw, D. F. Holmes, and K. E. Kadler. 2004. Coalignment of plasma membrane channels and protrusions (fibripositors) specifies the parallelism of tendon. J. Cell Biol. 165:553-563.[Abstract/Free Full Text]
8
8. Canty, E. G., T. Starborg, Y. Lu, S. M. Humphries, D. F. Holmes, R. S. Meadows, A. Huffman, E. T. O'Toole, and K. E. Kadler. 2006. Actin filaments are required for fibripositor-mediated collagen fibril alignment in tendon. J. Biol. Chem. 281:38592-38598.[Abstract/Free Full Text]
9
9. Chen, X., S. Kojima, G. G. Borisy, and K. J. Green. 2003. p120 catenin associates with kinesin and facilitates the transport of cadherin-catenin complexes to intercellular junctions. J. Cell Biol. 163:547-557.[Abstract/Free Full Text]
10
10. Gurdon, J. B. 1988. A community effect in animal development. Nature 336:772-774.[CrossRef][Medline]
11
11. Gurdon, J. B., P. Lemaire, and K. Kato. 1993. Community effects and related phenomena in development. Cell 75:831-834.[CrossRef][Medline]
12
12. Gurdon, J. B., E. Tiller, J. Roberts, and K. Kato. 1993. A community effect in muscle development. Curr. Biol. 3:1-11.[CrossRef][Medline]
13
13. Heaysman, J. E., and S. M. Pegrum. 1973. Early contacts between fibroblasts. An ultrastructural study. Exp. Cell Res. 78:71-78.[CrossRef][Medline]
14
14. Heaysman, J. E., and S. M. Pegrum. 1973. Early contacts between normal fibroblasts and mouse sarcoma cells. An ultrastructural study. Exp. Cell Res. 78:479-481.[CrossRef][Medline]
15
15. Kawaguchi, J., Y. Azuma, K. Hoshi, I. Kii, S. Takeshita, T. Ohta, H. Ozawa, M. Takeichi, O. Chisaka, and A. Kudo. 2001. Targeted disruption of cadherin-11 leads to a reduction in bone density in calvaria and long bone metaphyses. J. Bone Miner. Res. 16:1265-1271.[CrossRef][Medline]
16
16. Ko, K., P. Arora, W. Lee, and C. McCulloch. 2000. Biochemical and functional characterization of intercellular adhesion and gap junctions in fibroblasts. Am. J. Physiol. Cell Physiol. 279:C147-C157.[Abstract/Free Full Text]
17
17. Krauss, R. S., F. Cole, U. Gaio, G. Takaesu, W. Zhang, and J. S. Kang. 2005. Close encounters: regulation of vertebrate skeletal myogenesis by cell-cell contact. J. Cell Sci. 118:2355-2362.[Abstract/Free Full Text]
18
18. Kremer, J. R., D. N. Mastronarde, and J. R. McIntosh. 1996. Computer visualization of three-dimensional image data using IMOD. J. Struct. Biol. 116:71-76.[CrossRef][Medline]
19
19. Lee, D. M., H. P. Kiener, S. K. Agarwal, E. H. Noss, G. F. Watts, O. Chisaka, M. Takeichi, and M. B. Brenner. 2007. Cadherin-11 in synovial lining formation and pathology in arthritis. Science 315:1006-1010.[Abstract/Free Full Text]
20
20. Luo, Y., I. Kostetskii, and G. L. Radice. 2005. N-cadherin is not essential for limb mesenchymal chondrogenesis. Dev. Dyn. 232:336-344.[CrossRef][Medline]
21
21. Matsusaki, T., T. Aoyama, K. Nishijo, T. Okamoto, T. Nakayama, T. Nakamura, and J. Toguchida. 2006. Expression of the cadherin-11 gene is a discriminative factor between articular and growth plate chondrocytes. Osteoarthritis Cartilage 14:353-366.[CrossRef][Medline]
22
22. McNeilly, C. M., A. J. Banes, M. Benjamin, and J. R. Ralphs. 1996. Tendon cells in vivo form a three dimensional network of cell processes linked by gap junctions. J. Anat. 189:593-600.[Medline]
23
23. Pinto da Silva, P., and N. B. Gilula. 1972. Gap junctions in normal and transformed fibroblasts in culture. Exp. Cell Res. 71:393-401.[CrossRef][Medline]
24
24. Ralphs, J. R., A. D. Waggett, and M. Benjamin. 2002. Actin stress fibres and cell-cell adhesion molecules in tendons: organisation in vivo and response to mechanical loading of tendon cells in vitro. Matrix Biol. 21:67-74.[CrossRef][Medline]
25
25. Straub, B. K., J. Boda, C. Kuhn, M. Schnoelzer, U. Korf, T. Kempf, H. Spring, M. Hatzfeld, and W. W. Franke. 2003. A novel cell-cell junction system: the cortex adhaerens mosaic of lens fiber cells. J. Cell Sci. 116:4985-4995.[Abstract/Free Full Text]
26
26. Takeichi, M. 1995. Morphogenetic roles of classic cadherins. Curr. Opin. Cell Biol. 7:619-627.[CrossRef][Medline]
27
27. Tsukita, S., S. Tsukita, A. Nagafuchi, and S. Yonemura. 1992. Molecular linkage between cadherins and actin filaments in cell-cell adherens junctions. Curr. Opin. Cell Biol. 4:834-839.[CrossRef][Medline]
28
28. Vleminckx, K., and R. Kemler. 1999. Cadherins and tissue formation: integrating adhesion and signaling. Bioessays 21:211-220.[CrossRef][Medline]
29
29. Wuchter, P., J. Boda-Heggemann, B. K. Straub, C. Grund, C. Kuhn, U. Krause, A. Seckinger, W. K. Peitsch, H. Spring, A. D. Ho, and W. W. Franke. 2007. Processus and recessus adhaerentes: giant adherens cell junction systems connect and attract human mesenchymal stem cells. Cell Tissue Res. 328:499-514.[CrossRef][Medline]
30
30. Yagi, T., and M. Takeichi. 2000. Cadherin superfamily genes: functions, genomic organization, and neurologic diversity. Genes Dev. 14:1169-1180.[Free Full Text]
31
31. Yonemura, S., M. Itoh, A. Nagafuchi, and S. Tsukita. 1995. Cell-to-cell adherens junction formation and actin filament organization: similarities and differences between non-polarized fibroblasts and polarized epithelial cells. J. Cell Sci. 108:127-142.[Abstract]

---

Molecular and Cellular Biology, September 2007, p. 6218-6228, Vol. 27, No. 17
0270-7306/07/$08.00+0     doi:10.1128/MCB.00261-07
Copyright © 2007, American Society for Microbiology. All Rights Reserved.

This article has been cited by other articles:

• Gealy, C., Hayes, A. J., Buckwell, R., Young, R. D., Caterson, B., Quantock, A. J., Ralphs, J. R. (2009). Actin and Type I Collagen Propeptide Distribution in the Developing Chick Cornea. IOVS 50: 1653-1658 [Abstract] [Full Text]  
• Humphries, S. M., Lu, Y., Canty, E. G., Kadler, K. E. (2008). Active Negative Control of Collagen Fibrillogenesis in Vivo: INTRACELLULAR CLEAVAGE OF THE TYPE I PROCOLLAGEN PROPEPTIDES IN TENDON FIBROBLASTS WITHOUT INTRACELLULAR FIBRILS. J. Biol. Chem. 283: 12129-12135 [Abstract] [Full Text]  

This Article Abstract Full Text (PDF) Supplemental material Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Richardson, S. H. Articles by Kadler, K. E. Search for Related Content PubMed PubMed Citation Articles by Richardson, S. H. Articles by Kadler, K. E.