PRODUCTION OF SPLIT PATTERNS O N T H E ARTICULAR CARTILAGE SURFACES OF RATS

PATRICIA O'CONNOR, CHRISTINE BLAND, A, BJELLE* AND D. L. GARDNER

Department of Histopathology, University of Manchester; and *Department of Rheumatology, University Hospital, UmerO, Sweden

PLATES 1X AND X

W H E N the surfaces of human or animal hyaline articular cartilage are pricked with a sharp, rounded instrument, splitting occurs (Hultkrantz, 1898). If many splits are caused, their orientation is not random : reproducible patterns can be recognised (Benninghoff, 1925; Bullough and Goodfellow, 1968; Clarke, 1972; Meachim et al., 1974). The form of these patterns is constant for individual joints within a species. It is reasonable to deduce that these split (pin-prick) patterns reflect inherent structural characteristics of the cartilage. Since a full understanding of the material properties of cartilage is necessary to appreciate the origins of cartilage failure, the condition termed osteoarthrosis, the pin-prick technique can be adopted as one approach to the investigation of this syndrome.

Rats and other small rodents are frequently selected for the experimental study of joint disease (Gardner, 1980~). Cartilage is invariably affected in inflammatory and non-inflammatory synovial disorders (Gardner, 1980b) and a full knowledge of its physical properties is necessary. No study of the split patterns of rat articular cartilage has been made and it appeared desirable that this material should be examined by the method of pin-pricking that had been applied to the load-bearing cartilage of larger species. The present paper reports the results of this investigation.

MATERIAL AND METHODS

Thirty-seven adult male 250-350 g inbred Wistar rats of a Manchester strain were main- tained on a standard pellet ration and provided with water ad lib. After sacrifice, the soft tissue and muscle were reflected from each femoral shaft and the knee joints opened. Care was taken to avoid touching the condylar articular surfaces. The collateral and cruciate ligaments were cut and the femoral shafts divided 5 mm above the condyles. The articular cartilage, which was up to 0.5 mm thick, was kept moist at all times with isotonic saline. Specimens were collected simultaneously for scanning electron microscopy: the results of these investigations are to be reported subsequently (O'Connor, Bland and Gardner, 1980).

In a first study, 23 femoral condyles from 14 animals were pricked manually using a fine- pointed sewing needle of 0.5 mm shaft diameter. Fifteen of the condyles were then inverted

Accepted for publication 23 April 1979. 1. PATH.-VOL. 130 (1980) 15

16

and the shafts inserted into pieces of plasticine. The articular cartilages were painted with Indian ink and washed with isotonic saline to remove excess ink. The surfaces were then quickly viewed with incident light, using a Zeiss operating microscope at x 16 and x25. The remaining eight condyles were washed in saline and immediately processed for scanning electron microscopy (SEM) according to an " optimum " preparative technique (Cameron, Gardner and Longmore, 1976).

Seven femoral condyles were used to test and modify a device for standardising the mode of pin-pricking. This apparatus (fig. 1) included a ball and socket joint to which the inverted condyle was secured by a fine rod firmly inserted into the residual femoral shaft. Rotation of the specimen was possible about vertical and horizontal axes. An adjustable platform allowed the ball and socket joint to be moved in two mutually perpendicular horizontal directions. The 0.5 mm diameter pin used to penetrate articular cartilage surfaces was secured on the underside of a pivoted arm. The position of weights, situated on the same side of the fulcrum as the pin, was adjusted until the pin penetrated the articular cartilage incompletely when released 5 mm above the condylar surface. The pivoted arm could be raised, lowered and/or rotated in a horizontal plane according to the height and position of the area to be tested. These degrees of movement enabled a series of pricks to be made over an entire articular surface with the pin maintaining a vertical position with respect to the cartilage. A standardised load was determined by the position of the weights and by the height of the pin above the condyle before release.

The articular cartilages of 24 left femoral condyles were subjected to pin-pricking with this device; 14 of these condyles were coated with Indian ink, washed in saline, and examined with the operating microscope. The pattern of pricks observed was mapped manually on two-dimensional diagrams in the manner adopted by Meachim et al. (1974) in their survey of human cartilage surfaces. The surfaces were photographed microscopically with a Nikon F camera, a green filter and coaxial illumination. Three further specimens were treated with sodium hydroxide for 5 hr at room temperature before being pin-pricked and examined. The remaining seven condyles were washed and fixed following pricking; they were processed for SEM. Four of these had been exposed to sodium hydroxide.

A final series of 14 condyles from seven animals was examined in the incident light mode to test the effects of drying, of isotonic saline, of hyaluronic acid (5 mg/ml) and of Indian ink, on the behaviour of individual pin-pricks. Magnifications of x80 and x 140 were selected on a Leitz Orthoplan microscope. The ink was not soluble in hyaluronic acid and could not be used in conjunction with it: in its presence excess ink could not be washed from the surface.

Foranalysis, the articular surfaces of the lower ends of the femurs were divided empirically into four regions. The regions (fig. 2) were delineated by rotating the condyles through 270" anteroposteriorly, selecting the maximum areas in focus at any one time when observed at x 16 with the operating microscope.

P. O'CONNOR, C . BLAND, A. BJELLE AND D. L. GARDNER

RESULTS Pin-prick pattern

A characteristic pin-prick pattern was identified on the rat femoral condylar surfaces. Figure 3 is a composite of two-dimensional drawings of this pattern for 14 rat left femoral condyles.

In region 1, the proximal, anterior articular surface, formed principally of the patellar surface of the femur, the cartilage was covered by uniform, uni- directional elongated splits. The splits were approximately parallel to the longitudinal axis of the cartilage surface in the centre of this region (fig. 4) but gradually deviated from this orientation as the lateral and medial edges were approached. At the extreme edges the splits were almost perpendicular to the longitudinal axis. In region 2, which included the most anterior part of

SPLIT CARTILAGE 17

the condylar notch, the splits were unidirectional and elongated, tending to form lines fanning out around the notch. In region 3 the articular surface was divided into two by the condylar notch. The radius of curvature of this surface was small. The splits were less orderly in arrangement than in regions 1 and 2 and generally lateromedial in direction. Towards the outer and inner edges the splits were directed parallel to the longitudinal axis of the joint; some were multidirectional, forming T-shaped or star-shaped openings. Other pricks formed elliptical or rounded holes with little or no splitting. In region 4, the posterior aspect of the condyle, the surface was also acutely curved.

1 Region 4

Longitudinal

FIG. 2.-Two-dimensional, scaled representation of the rat femoral condylar articulating surfaces showing the four regions of analysis, and the longitudinal axis of the specimen.

Although splits developed after pricking in this part of the condyle, their orientation was much less orderly than in the other regions and they generally displayed a latero-medial orientation. However, some followed the longi- tudinal, antero-posterior axis of the articulating surface, particularly in the extreme posterior part of this region. Multidirectional and rounded apertures were identified frequently.

Eflects of sodium hydroxide Sodium hydroxide treatment before pin-pricking resulted in a similar split

pattern to that observed on the untreated surface but many of the splits appeared longer and wider in regions 1 and 2 than in untreated specimens, with more rounded holes in regions 3 and 4.

Erects of drying, of isotonic saline, of hyaluronic acid and of Indian ink Narrow, undirectional, elongated splits were characteristic of pricked moist

cartilage. The splits were surrounded by a whitened, distorted zone (fig. 5). J . PATH.-VOL. 130 (1980) B

IS P. O'CONNOR, C. BLAND, A. BJELLE AND D. L. GARDNER

Indian ink enhanced the definition of the splits and increased their apparent length. If the articular cartilage was allowed to dry at room temperature for

1-14 hr before testing, elliptical holes resulted from pin-pricks (fig. 6). The major axes of the holes were in the direction of the split pattern. On applying isotonic saline, the holes slowly closed in the direction of the minor axes

SPLIT CARTILAGE 19

leaving splits similar to those produced in moist specimens with associated, localised, distorted areas. Similar results were found if the specimen had first been soaked in hyaluronic acid for 1 hr. If kept moist, these femoral condyles developed a normal split pattern. If allowed to dry before pricking, elliptical holes resulted; these could be made to close on the re-application of hyaluronic acid. In both saline- and hyaluronic acid-treated cartilage, drying after pricking caused the splits gradually to open into elliptical holes.

When the pin tip was coated with ink before pricking, poor delineation of splits often resulted. Ink frequently remained on the cartilage surface around a split or defined only part of the split opening. Painting the cartilage surface with ink after pricking ensured that the carbon entered the splits and accentuated the deeper parts.

DISCUSSION The results presented in this paper demonstrate that reproducible patterns

of splits develop when numerous pin-pricks are made in adult rat articular cartilage surfaces. The evidence suggests that where opposing articular surfaces are congruent, loads relatively small and cartilage relatively thin, the pattern of splits is in the main axis of movement; but that where opposing surfaces, such as those of the femoral condylar/tibial plateaux articulation are non-congruent, loads per unit area relatively high, and cartilage relatively thick, the pattern of splits is transverse, fantail or non-uniform.

As a pin makes contact with a cartilage surface during the production of a split, the compressive force produces an indentation before penetration occurs. As the pin enters the cartilage the tissue in its path is fractured: there is com- pressive deformation at the pin tip and, because of the symmetrically tapered point, in a perpendicular direction outward from the pin. An inverted conical hole results with displacement of tissue and transverse fluid flow. The fact that the eventual aperture is not a rounded tapered hole but an elongated split indicates that crack propagation has taken place in a preferred direction of weakness. The increase in the fracture surface requires the input of energy. The shape of the narrow elongated split also indicates that when the pin is removed, much of the compressive deformation is recovered, closing the hole to a narrow slit.

When splits are caused, cartilage is fractured at the site of penetration and deformed locally. It is assumed that the deformation of rat hyaline cartilage is analogous to that of other articular cartilages and that three types of deforma- tion are possible : instantaneous (elastic) ; time-dependent (anelastic) ; and perwanent (plastic). With the methods used in the present study it was not possible to distinguish between the first two categories. It is assumed that the deformation recoverable after removal of the pin consists of elastic and anelastic components. The evidence, first that pin pricks cause narrow splits in moist specimens whereas open elliptical holes are formed in dry cartilage, and second, that the elliptical holes in dry cartilage close after the addition of saline or hyaluronic acid, suggests that the major part of the time-dependent deforma- tion is due to fluid movement. Thus the degree of hydration of the cartilage, a

20 P. O’CONNOR, C. BLAND, A. BJELLE A N D D . L. CARDNER

function of the proteoglycan population (Maroudas, 1980), is a main deter- minant of the extent of cartilage deformation by pin-prick testing.

Occasionally, splits in regions 3 and 4 remain open even in the presence of excess liquid; this is also the case if a much larger hole is made in any area, due to severe fracture damage. All splits examined in moist cartilage whether fully closed or not, show the persistent white peripheral colouration of per- manent distortion (fig. 5).

Sodium hydroxide acts on articular cartilage to disrupt the proteoglycan matrix; glycosaminoglycan side chains are removed from the proteoglycan molecules, causing protein denaturation.

The results of this treatment suggest that the production and direction of splits are not primarily dependent on the integrity of the proteoglycan matrix but that the disruption of matrix proteoglycan may reduce resistance to crack propagation and reduce or eliminate the capacity to recover from deformation by restricting fluid flow.

It is widely assumed that the shape of individual spIits is determined by the molecular composition of the cartilage and that, in this context, the arrange- ment of collagen fibres and fibre bundles is of pre-eminent importance (Bullough and Goodfellow, 1968; Meachim et al., 1974). However, the dimensions of a typical split in rat condylar articular cartilage, 340 pm deep and 90 pm wide at the surface, greatly exceed those of the mats or arcades of collagen fibre bundles demonstrable by polarised light microscopy (Bullough and Good- fellow, 1968). Preliminary evidence from the transmitted light microscopy of paraffin-embedded sections suggests that the gentle, serpiginous split margin follows the outlines and arrangement of chondrocyte clusters. Thus, the path of the split may be influenced by differences in material properties between the territorial matrix and the interterritorial matrix, and at the cell-matrix interface.

SUMMARY A device has been designed to permit reproducible loads to be applied per-

pendicularly to the acutely curved cartilage surfaces of the diarthrodial joints of small animals. With this apparatus, a sharp, round pin has been used to cause splits in the hyaline articular cartilage of the lower end of the femurs of 37 Wistar rats. In the patellar groove, the splits form a longitudinal pattern. On the condylar surface, the pattern is fantail, radial or non-uniform. It is suggested that cartilage thickness, joint congruence and load per unit area, are variables related to the directional pattern of the splits. These variables, in turn, reflect the gross, microscopic and molecular anatomy of the cartilage.

The effects of pin-pricks are complex: they disrupt surface collagen bundles and other cartilage components. Experiments made to test the effects of dry- ing, isotonic saline, hyaluronic acid and sodium hydroxide support the view that the stateof hydration and fluid flow are important determinants of the shape of cartilage splits. However, sodium hydroxide-induced cartilage disruption, involving glycosaminoglycan loss, did not influence split configuration. Some preliminary evidence suggests that the process of crack propagation, by which splits form, may be related to the position and shape of chondrocyte lacunae.

@CONNOR, BLAND, BJELLE AND GAKDNER

CARTILAGE SPLITS

Fici. I .-The apparatus for producing standardised pin-pricks.

PLATE I X

O'CONNOR, BLAND, BJELLE A N D GARDNER

CARTILAGE SPLITS

PLATE X

FIG. 4.---Region I of moist, Indian ink-covered, articular FIG. 5.-Splits in region I of moist, Indian ink-painted cart i laye. articular cartilage. Note the whitening around the

splits. Incident light microscope. 100.

FIG. 6.-Splits produced i n air-dried carfilage. Incident light microscope. Y 12).

SPLIT CARTILAGE 21

It is concluded that pin-prick testing offers a means by which the structural properties of the cartilage of small diarthrodial joints may be conveniently tested in vitro.

acknowledged. The support of the Arthritis and Rheumatism Council for Research is gratefully

REFERENCES BENNINGHOFF, A. 1925. Form und Bau der Gelekknorpel in ihren Beziehungen zur Funk-

tion. 11. Der Aufbau des Gelenkknorpels in seinen Beziehungen zur Funktion. 2. Zef. Mikrosk. Anat., 2, 783.

BULLOUGH, P., AND GOODFELLOW, J. 1968. The significance of the fine structure of articular cartilage. J. BoneJt. Surg., 50B, 852.

CAMERON, C. H. S., GARDNER, D. L., AND LONGMORE, R. B. 1976. The preparation of human articular cartilage for scanning electron microscopy. J. Micro., 108, 1.

CLARKE, 1. C. 1972. A study of the structure and wear response of hip joint cartilage. Ph.D. Thesis, University of Strathclyde.

GARDNER, D. L. 1980~. Historical aspects. In Models of osteoarthrosis, edited by G. Nuki. Pitman Medical, London (in press).

GARDNER, D. L. 1980b. General pathology of the peripheral joints. In The joints and synovial fluid, vol. 11, edited by L. Sokoloff. Academic Press New York, Sun Francisco, London (in press).

HULTKRANTZ, W. 1898. Uber die Spaltichtungen der Gelenkknorpel. Verh. Anat. Ges., 12, 248.

MAROUDAS, A. 1980. Physical chemistry of articular cartilage. In The joints and synovial fluid, vol. 11, edited by L. Sokoloff. Academic Press, New York, Sun Francisco, London (in press).

MEACHIM, G., DENHAM, D., EMERY, I. H., AND WILKINSON, P. H. 1974. Collagen alignments and artificial splits at the surface of human articular cartilage. J. Anat., 118, 101.

O'CONNOR, P., BLAND, C., AND GARDNER, D. L. 1980. Scanning electron microscopy of artificial splits in the surface of rat articular cartilage (in preparation).