STRUCTURE OF THE MOST SUPERFICIAL LAYER OF ARTICULAR CARTILAGE

R. TESHIMA, T. OTSUKA, N. TAKASU, N. YAMAGATA, K. YAMAMOTO

Tottori University, Yonago, Japan

We studied the most superficial layer of macroscopically normal articular cartilage obtained from human femoral heads, using polarising microscopy and SEM. The most superficial layer, 4 to 8 my thick, was acellular consisting of collagen fibrils. This layer could be peeled away as a thin film, with no broken collagen fibrils on its inferior surface or on the surface of subjacent cartilage layers. The orientation and diameter of collagen fibrils were different on these two surfaces. Our findings suggest that the most superficial layer is an independent one which is only loosely connected to the fibrous structure in the layer deep to it.

J Bone Joint Surg [Br] 1995;77-B:460-4.

Receved 24 June 1994.. Accepted after revision 20 October 1994

The superficial structure of articular cartilage has been frequently studied since the 18th century, when John Hunter noted a membrane on the cartilage with a magnifying lens (Ghadially 1983), but there is no clear consensus about this structure. We examined the most superficial layers of human femoral-head articular cartilage, using polarising microscopy and SEM.

MATERIALS AND METHODS
We collected 20 human femoral heads removed at arthro plasty for fracture of the femoral neck or from hips disarticulated after severe trauma. Nine were discarded because of visible pathological changes on the articular surface, such as fine fibrillation or lack of a glossy appearance. Eleven heads with macroscopically smooth and Glossy surfaces and no apparent osteoarthritic changes were selected for study from patients whose ages ranged from 40 to 72 years. Several different investigations were performed:

1) Four femoral heads were fixed in 10% formalin in phosphate-buffered solution at pH 7.0, decalcified in phos­phate-buffered 10% ethylene diamine tetra-acetic acid at pH 7.1, then dehydrated and embedded. Frontal plane sections, 10 gm in thickness, were studied, unstained, by polarising microscopy.

2) From four other femoral heads, strips of cartilage 1.5 x 0.5 cm in size were excised with the subchondral bone. These were fixed in 2.5% glutaraldehyde in phos­phate-buffered solution at pH 7.0. postfixed in buffered 1 % osmium tetraoxide for two hours, and dehydrated. They were then freeze-cracked in a direction perpendicular to the Ion. axis of the strips. The specimens were dried by the critical point method, and the cracked surface was sputter­-coated with gold and examined by SEM (Model HFS-2ST; Hitachi, Tokyo, Japan).

3) Immediately after the three remaining femoral heads had been excised, two slits about 5 mm apart were cut in the head surface in longitudinal and circumferential directions with a scalpel. A third slit was made at the base of these. The superficial edge of the isolated cartilage was held with small rongeurs and separated by peeling it away. The separated layer and the cartilage deep to it were prepared for light microscopy as described above, and 7 my sections vere cut perpendicular to the articular surface in a longitudinal direction. The unstained specimens were examined by polarising microscopy. Parts of the two speci­nens were processed as described above for SEM; the deep surface of the superficial layer and the superficial surface of he deep layer were studied.

4) In a similar fashion, two slits were made from the ;ynovium of the femoral neck into the articular cartilage in i longitudinal direction. The synovium was held by small rongeurs and peeled away towards the articular cartilage. The separated specimens were examined unstained, and also stained with safranin 0 fast green for light microscopy.

5) The strip specimens of cartilage including subchondral bone were fixed in 10% neutral formalin for two days, then macerated with 10% sodium hydroxide at 37gr.C for 24 hours. Some strips were then dehydrated, embedded and cut into 7 my sections perpendicular to the articular surface.

6) Using other macerated strips an attempt was made to prepare specimens similar to those obtained in the surface separation experiment, but the tissues were more friable and only relatively small areas of the superficial layer could be peeled off. The surfaces were then rinsed in phosphate buffer, postfixed in buffered 1% osmium tetroxide for two hours, dehydrated and dried by the critical point method. After sputter-coating, with gold they were examined by SEM.

RESULTS 
Polarising microscopy. Polarising, microscopy revealed a membrane-like structure with strong birefringence over the entire surface of the articular cartilage. This structure was 4 to 6 my thick, acellular, and had separated from a deeper layer consisting of spindle chondrocytes.

Scanning electron microscopy of cracked surfaces. The cracked surface of the articular cartilage, when examined by SEM, showed a layer of dense collagen fibrils as the most superficial layer (Fig. 1). The collagen fibrils ran parallel to the articular surface, and the thickness of the layer was 4 to 8 my.

Figure 1 a - SEM of the cracked surface of human articular cartilage ( x 450). The most superficial structure is distinctly different from the layer deep to it, and its narrow edge is seen below the letter S.

Figure 1 b - High-magnification SEM of the sliperficial layer ( x 9000). It consists of a layer of densely-packed collagen fibrils running parallel to the articular surface.

Surface separation. The superficial layer of articular cartilage could be separated only in a longitudinal direc­tion. A white translucent membrane-like structure 0.5 to 1.0 cm in length was obtained, but it was impossible to obtain more tissue by circumferential separation. The morphological structure of this layer was identical to that seen in non-separated specimens, as were the optical characteristics on polarising microscopy.

Both the deep surface of the separated layer and the opposing surface of the remaining, cartilage were composed of collagen fibrils, but there were distinct differences in the direction and diameter of fibrils and in the amount of granular material adhering to them (Figs 2 and 3). The orientation of the fibrils was clearly seen on the deep surface of the separated layer: most were orientated in a longitudinal direction on the femoral head (Fig. 2a). The fibrils on the deep surface of the superficial layer were 75 to 100 nm in diameter -and showed no adherent granular substance, while those on the surface were 125 to 150 nm in diameter and showed a large amount of an adherent granular substance (Fig. 3). No broken fibrils were observed on either of the originally opposed surfaces.

Specimens obtained in continuity with the synovial tis­sue of the femoral neck had a white translucent membrane­like structure, similar in appearance to that from the surface separation experiment. No specimen obtained by separation from the synovial tissue showed any adherence of cartilage matrix staining red by safranin 0, but a thin layer of chondrocytes was adherent to the part near the synovium.

Figure 2a - SEM of the deep surface of the separated superficial layer ( x 2200). It is composed of collagen fibrils, most of which run in a radial direction (indicated by the long bar) on the femoral head. 

Figure 2b - SEM of the upper surface of the cartilage below the separated surface layer (x 2200). The surface is also composed of collagen fibrils, but no broken fibrils are visible.

Figure 3a - High-magnification SEM of the deep surface of the separated superficial layer (x 9000). The fibrils range in diameter from 75 to 100 nm. 

Figure 3b - High-magnification SEM of the surface of the cartilage portion (x 9000). The fibrils of the surface range from 125 to 150 nm in diameter. A large amount of granular substance is observed on the fibrils

Macerated specimens. The boundary between the layers was seen much more clearly in the macerated cartilage specimens than the fresh specimens on polarising micro­scopy. The superficial layer was more easily separated, although no portions were as larged as those separated from ­the non-macerated cartilage. SEM showed a distinct boundary between the superficial layer and the next layer (fig. 4).

Figure 4 - SEM of the surface of a macerated specimen of articular cartilage with a partially separated surface layer ( X 330). There is a distinct boundary between the intact surface on the left and the surface of the deeper layer on the right. The gently-sloping elevations (arrows) appear to corespond to the spindle cells of the deeper layer.

DISCUSSION
MacConaill (1951) reported the existence of the lamina splendens,a specific superficial structure. A subsequent phase-contrast microscopic study by Aspden and Hukins (1979) led to the conclusion that it was an artifact and questioned its existence. Dunharn et al (1988) confirmed the presence of the lamina splendens by polarising microscopy. Many transmission electron-microscopic studies have indicated that the most superficial layer is composed of collagen fibrils (Weiss, Rosenberg and Helfet 1968; Meachim and Roy 1969; Ghadially 1983; Schenk, Eggli and Hunziker 1986) and that it is morphologically distinct from the next deeper layer (Bullough and Goodfellow 1968; Weiss et al 1968; Wolf 1969).

Meachim and Roy (1969), Ghadially (1983) and Schenk et al (1986) considered, however, that the most superficial layer could not be differentiated from the layer deep to it.

Recent SEM observations have confirmed the presence o layer corresponding to the lamina splendens (Clark 1985 1990; Jeffery et al 1991). Clark (1990) described the lamina splendens as bein. 5 my thick and composed of collagen fibrils, but Jeffery et al (1991) reported it to be an amorphous layer devoid of collagen fibrils.

In our study, the most superficial layer of the articular cartilage was seen to be acellular, and composed of collagen fibrils with a course parallel to the articular surface. The layer was distinctly different from the next deep layer of spindle chondrocytes in terms of its optical characteristics on polarising microscopy and the density and orientation of collagen fibrils on SEM. The thickness of the superficial layer was consistently 4 to 8 my regardless the method of preparation or study. These observations are in close agreement with those obtained by other authors, using a variety of methods of observation (Weiss et al 1968; Dunham et al 1988; Clark 1990). All these findings suggest the existence of an independent acellular super­ficial layer composed of collagen fibrils.

If this layer is an independent one with a different collagen fibril architecture, its boundary surface should be dynamically fragile and physically separable, as has been previously reported (Bullough and Goodfellow 1968; Wolf 1969; Clarke 1971). There have, however, been no previous morphological studies of the boundary surface at the level of separation.

In our study, the separated specimens of the most super­ficial layer showed morphological features identical to those of untreated specimens. No broken fibrils were seen on SEM of the inferior surface of any separated portion or on the surface of any cartilage portion with separated superficial layer. Specimens macerated with sodium hydroxide showed the boundary between the superficial and inferior layers much more clearly.

Our findings confirm that there is a specific lamina splendens on the articular catilage. It is an acellular, independent layer of collacen fibrils which is only loosely connected to the fibrous structure in the deeper layer.

Separation in continuity with the synovium of the neck showed that the most superficial layer is firmly connected to the deeper cartilage matrix only in the vicinity of the synovium and that it undergoes transition to synovial tis­sue. Our findinas indicate that the most superficial layer is anchored firmly in the peripheral region of the articular cartilage and to the synovial tissue and that its orientation tends to resist tensile stress durin. joint movement. It seems that it functions as an outer coating which maintains the orphological integrity of the articular cartilage by withstanding extrinsic compression and intrinsic swelling pressure.

No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article.

REFERENCES

Aspden RM, Hukins DW. The lamina splendens of articular cartilage is an artefact of phase contrast microscopy. Proc R Soc Lotid Biol 1979;206:109-13.

Bullough P, Goodfellow J. The significance of the fine structure of articular cartilage. J Bone Joint Surg [Br] 1968;50-B:852-7.

Clark JM. The organisation of collagen in cryofractured rabbit articular cartilage: a scanning electron microscopic study. J Orthop Res 1985;3:17-29.

Clark JM. The organisation of collagen fibrils in the superficial zones of articular cartilage. J Anat 1990; 171:117-30.

Clarke IC. Articular cartilage: a review and scanning electron microscope study. 1. The interterritorial fibrillar architecture. J Bone Joint Surg [Br1 1971;53-B:732-50.

Dunham J, Shackleton DR, Billingham ME, et al. A reappraisal of the structure of normal canine articular cartilage. J Anat 1988:157:89-99.

Ghadially FN. Fine structure of synovial joints. London, etc: Butterworths 1983:80-102.

Jeffery AK, Blunn GW, Archer CW, Bentley G. Three-dimensional collagen architecture in bovine articular cartilage. J Bone Joint Surg [Br] 1991:73-B:795-801.

MacConaill MA. The movements of bones and joints: 4. The mechanical structure of articulating cartilage. J Bone Joint Surg [Br] 1951;33-B:251-7.

Meachim G, Roy S. Surface ultrastructure of mature adult human articu­lar cartilage. J Bone Joint Surg [Br] 1969;51-B:529-39.

Schenk RK, Eggli PS, Hunziker EB. Articular cartilage morphology. In: Kuettner KE, Schleyerbach R, Hascall VC, eds. Articular cartilage biochemistry. New York, etc: Raven Press, 1986;3-22.

Weiss C, Rosenberg L, Helfet AJ. An ultrastructural study of normal young adult human articular cartilage. J Bone Joint Surg [Am] 1968;50-A:663-74.

Wolf J. Chondrosynovial membrane serving as joint cavity lining with a sliding and barrier function. Folia Morphol Praha 1969;17:291-308.