DANIEL CHAPPARD,
CHRISTIAN
ALEXANDRE
UNCALCIFIED CARTILAGE HUMAN FETAL CARTILAGE
Introduction
It is currently believed that uncalcified cxtilape is devoid of blood vessels except in some areas where they are passing through it to reach other tissues (Sledge. 1968). Cartilage vascularization is usually reported as an important process in bone differentiation duringendochondral ossification (Trueta and Buhr. 1963). IIowcver. in these cases. mineralization of the cartilage matrix always occur\ before vascular in&ion (Juster and ,Moscofian. 1971). In the fetus, the presence ot blood vessels wiis first reported in epiphyal uncalcified cartilage in 1743 by Hunter. The term ‘cartilage canal’ has been proposed to describe these vascular intrachondral pathuacs by Howship in 1815 (cited by Waterman. Ic)hl). This system is essentially devoted to nutrition of chondrocytes but also provides additional \tem cclls for interstitial chondrogenesis (Wilcman and Viln Sickle. 1970: Lufti. 1970). A review of
and GEORGES RIFFAT
RESORPTION CANALS
IN
the anatomy and histophysiology of cartilage canals was presented elsewhere (Chappard et al., 1983). In the human fetus, epiphyses appear as solid avascular masses until the eleventh week of development. Around the third month of fetal development, cartilage canals coming from the perichondrium are recognized in the uncalcified cartilage epiphyses. Whether cartilage canals develop by passive inclusion of vessels within the eplphysis or by active chondrolysis is still a matter of controversy: uncalcified cartilage is known to present great resistance to vascular or tumoral invasion which has been related to low molecularweight protease inhibitors, the ‘anti-invasive factor’ (Kuettner and Pauli, 1978). The purpose of the present study was to investigate the developmental mechanisms of cartilage canals in the human fetus at the microscopic level. Material
and Methods
The material consisted of o\er 50 hum;111 fetuses obtained after mi,carriagcs in local obstetrical departments. Specimens ~\\t’re fixed in -I.S(: formaldehyde and transfcrrcd to the Iaboratorv. .Ab
Ct1APPARD
At that time intrachondral canals coming from the perichondrium arc observed within the epiphysis (Fig. 2). At the blind end of cartilage canals numerous areas showing ail the characteristics of an active chondrolysis (loss of matrix metachromasia, lacunae containing cells intimately associated with matrix. presence of granular debris ,) are commonly observed (Figs 3. 4). These areas contain cellular foci composed of ctellate ceils (resembling embryonic fibroblasts) and highly vacuolated macrophages in close contact with the uncalcified cartilage matrix. Behind these resorptive foci. the canal boundaries adopt a perichondrium-like appearante: mitotic figures. are commonly observed. Thus the definitive aspect of a cartilagc canal resembles a match with a blind round end (the resorptive area) followed by a narrowed tail (with pericanalar chondrogenesis) which contains a vascular loop. After the seventeenth week of development and until the appearance of the epiphyseal ossification center. the cartilage canal system develops to a great extent (Fig. 5). The mixed cellular foci are very occasionally observed in a few cartilage canals. Cartilage canals represent a complex vascular network with branching anastnmoses. At that time. the main characteristic is the development of vascular system complexity: each canal contains an arteriole. a venule and numerous capillary glomeruli (Fig. 6). This
internal malformations WLIS congenital carefully checked. The distal femoral epiphysis was selected for the study. Specimens were then dehydrated in graded ethanol and finally embedded in the water-miscible plastic, glycol methacrylate according to the procedures developed in this laboratory (Chappard PI al.. 19X2; Chappard, 1985). Sections 3 !drn in thickness were obtained on a Minot-type microtome equipped with a Ralph knife holder (Chappard et al., 1983).
Results At the end of the eighth week of embryonic development, the definite aspect of the femur is already observable: the hip, diaphysis and condyles are all cartilaginous and avascular (Fig. 1). The extraskeletal mesenchyme and perichondrium are vascularized. Around the ninth week of development, chondrocytes in the centre of the &&IJw~/ shaft hypertrophy and mineralize their currounding matrix, and vascular invasion occurs rapidly in this area. At that time, multinucleated giant cells (osteoclasts/ chondroblasts) are obvious and resorb the calcified cartilage. At the same time, apposition of bone proceeds. However, until the twelfth week of development, the epiphysis remains as a solid highly cellular hyaline cartilage with a faintly metachromatic matrix.
Fig. 3. The
end of a longitudinally
cells: chondrogcncai\ cplthelial
arrangement
ihc canal.
with
Fg
xctwrlcd
canal
in ths cartllagc
of chondrohlasts
(arrows)
The
hhnd bulky end I\ filled
canal wall hchmd
thee
and undlffcrcntlntcd
cells
with rcwrptwc
Note
the pxudo-
mc\cnchymal
ccII\ withln
x 100.
Fig. 1. Dctall Stcllatc
15 ohvioua
L.7AL.
ot the mixed cellular
fihrohla\t\
the uncalclflcd
(F)
cartilage
6. A typical mature
and well-dcvelopcd
populauon
and vacuolatcd matrix.
cartilage
vascular
system.
mwlwd
macrophagc Note
cxtraccllular
canal in trdnsvcnc x 100.
in chondrolyw
(M)
in the rcsorptlon gr.mul;u
xctwn
dchn
ot uncalcdui
cartllagc.
arc’a. m clox (xrow).
contact
~30,
Note the pcrichondnum-hke
wall
vascular
chondrium
system is connected vessels.
with the peri-
Discussion A great deal of literature deals with the histog,enesis of cartilage canals and several theories have been suggested (reviewed extensi\ely in C’happard c[ rrl.. 19X3). Briefly, the following mechanisms have been proposed: ( I ) The ‘evolution theory proposed by Heitzman (1X7.3) and Retterer (1900) (cited by Hurrell. 3933) was based on the angiogenie differentiation of ‘isolated blood islands’ trapped in the cartilage (a mechanism supposed to be very similar to the vascular development in the yolk sac of the early embryo). i 2) The ‘l~rop~tlsion thory (Hurrell. 1934) supposed that vessels penetrate into the cartilage by pushing through the chondroblasts and theu relatively plastic matrix. (3) The ‘inclusiotz tkory was initially proposed by Kajawa in IYIY (cited by Hurrell. 1934) who thought that primary perichondral capillaries were early included in the chondrifying Anlagen. Later, other authors supposed that some particular vessels equipped with a ‘perichondral shell’ could be included in the epiphysis by active subperichondral chondrogenesis (Haines. 1933; Wang. 197.5). However, Levene (1064) demonstrated that this hypothesis was not compatible with a careful examination ot histological data. (4) The ‘in\,asion thcor,v’ was proposed in lY25 by Stump (lY25) who thought that specific intracanalar mesenchymal cells were able to develop a ‘lytic power’ for the uncalcified cartilage matrix. These observations were confirmed later on by tlintzche (lY27), Hurrell (lY34), Levene (1964). Lufti (lY70) and Reddi (1981) who found that a ‘vascularized mesenchyme’ was responsible for the uncalcified cartilage matrix breakdown. A direct chondroblastic chondrolysis has also been proposed (Waterman, 1961). Kugler et al. (1979) described multinucleated cells (chondroclasts) removing the cartilage matrix at the blind end of the cartilage canals. In fact, careful examination of their histological data reveals that they studied mineralized cartilaginous matrix resorption near the epiphyseal ossification center at a later developmental period. In fact, their data are in
agreement with the suggcstlon that multinucleated giant cells (chondroblasts or osteoclasts) can only resorb mineralized malri\ (Savostin-Asling and Asling. 1975). In our serial semithin section scrics. wc have never found such multinucleated giant cells, but mixed cellular foci have in\ari;lhl> been observed. Similar resorptive foci llil\C also been observed in four other approachc\: (a) Mononucleated connective cells ha\c been shown to resorb uncalcificd cartilagc in in Iitro experiments (Fell and Barrett. 1073). (b) In tracheal ring of the rat . ThC endoluminal side is resorbed bv fihrobla\tic cells. while chondrogenesis is obscr\ cd on the external side (Yajima. lY7h). (c) In rheumatoid arthritis. the artlcular cartilage of the joints is destroyed by ‘pmnu~ composed of macrophage-like cells and svnoviocytes (Kobayashi and Ziff, 1975). In tissue culture synociocytes differentiate into ‘dendritic‘ cells prohablv of fibroblastic origin (Gross. 19X1). Immunofluorescencc studies have shown that the degradatlve enzyme collagenasc ~3s secreted by fibroblasts which have been stimulated b! ;I ‘mononuclear ccl1 factor’ ( tiuk hrcchtsGodin et NI., 1Y7Y; Dayert (11 (I/.. I WI): Wooley et cd., 19X0). (d) In the embryonic chick growth plate. the ossification mechanisms are markerll~ different from those in mammals: the cartilage does not mineralize before its replacement by bone. but is rcsorbcd bv ;I mlscd cellular population composed of iibroblasts activated byvicinal macrophages (Sorrel1 and Weiss. IYXO: Sorrel1 and Weiss, lYX3). The degradation of uncalcified cartilage and other extracellular matrices hv macrophage-activated connective tissue cells is pr-esently well documented in patholo~tcal conditions and morphodifferentiation (see review by Vaes. IYXS). Monocytesi’m~lcrophages are able to release soluble factors under a variety of immunological. inflamm;~tory or perhaps other undetermined condtions. Connective tissue cells act as eftector cells under these stimulations. By analog! with the lymphokines released by stimulated 1110110lymphocytes. these soluble cyte/macrophage peptidic factors have bc~n called ‘matrix regulatory monokincs’ ( V.LC~. 19X5). In the human fetus. monocvtcs/macrophages have been recognized In the peri-
CHAPPARD
pheral blood as early as the twelfth week of gestation (Djaldetti et al., 1978), at the same time that cartilage canals appear (Langer, 1876; Bidder, 1906; Hintzche, 1927; Hurrell. 1934; Stanescu et al., 1973). Chondrogenesis is known to occur in the cartilage canal wails (Lufti, 1970; Wilsman and Van Sickle, 1979). Mitosis is often observed immediately behind chondrolytic foci, suggesting a coupling mechanism between chondrolysis and chondrogenesis, a condition very similar to the sequence observed in bone remodeling (Frost, 1973).
ET AL.
Whether this pericanalar chondrogenesis is dependent of other macrophagic mitogenic monokines remains under investigation. Cartilage canal development in mammals appears to be a unique model for the study of cellular interactions during the breakdown and repair of uncalcified cartilage. Acknowledgements The authors are greatly indebted to Mr Bellavia for photographic assistance and to Mr Murigneux for his interest in this study.
References Bidder, A. 1906. Osteobiologie. Arch. Mikr. Ancrr.. 68, 137-213. Chappard. D.. Laurent. J. L., Camps. M. and Monthcard. J. P. 1982 A simpleand reliable method forpurifyingglycol methacrylate for histopathological studies. Acfa Hisfochem.. 71, 05-102. Chappard. D. and Laurent. J. L. 1983. A holder for Ralph kmvea for cutting sections for high performance optxal mxroscopy with a Minot-type microtome. EX~C~ICIIIIU.39, 121-123. Chappard. D.. Alexandre. C., Chol. R. and Riffat, G. 1983. Les canaux intra-chondraux: histog&&e. anatomic et histophysiologle de la vaacularisation cartilagineuse du foetus humain. Lyon Med.. 249, 417-428. Chappard, D. 198.5. Uniform polymernation of large blocks m glycol methacrylate at low temperature wth special reference to enzyme hlstochcmiatry. Mikroscopie. 42, 348-150. Daycr, J. M.. Goldring, S. R.. Robmson. D. R. and Krane, S. M. 1980 Cell-cell interactions and collagenax productlon. In Col1ujienu.w 111Norrnul ccnd Putholog~ul Comective Ticsues (eds D. E. Wollcy and J. M. Evanson), pp. 38-104. Wiley. New York. Djaldctti. M., Goldman. J. A. and Fishman, P. 1978. Ultrastructure of the cells in the peripheral blood of human cmhryos at early gestational stages. Rio/. Neonure. 33, 177-183. Fell, H. B. and Barrett, M E. J. 1973. The role of soft connective tissue in the breakdown of pig articular cartilage in the presence of complement sufficient antiserum to pig crythrocytcs. Int. Arch. Allergy. 44, 441-46X. Frost, H. 1973. Bone Rernodelmg and its Relationship fo Metabolic Bone Dbeaxs. Orthopaedic Lecture Series. Vol. 3. Charles V. Thomas. Springfield. Gross, .I. 1981. An essay on biological degradation of collagen. In Cell Biology ofEx:xtracellulur M&ix (ed. D. Hay), pp. 217-258. Plenum Press. New York, London. Haines. W. lY3s-34. Cartilage canals. J. Anar.. 68, 45-64. Hintzche. E. 1927. Untersuchungen an Sttitgeweben. 1. Ueber die Bedeutung der Geftibkanile m Knorpel nach Befunden am distalen Ende des menschlichen Schenkelbeines. Z. Mikrosk. Amt. Forsch., 12, 61-126. Hunter. W. 1743. On the structure and dwasc of articular cartilage. Phila. Trans., 42, 514521 Hurrell. D. J. l93G-35. The vasculariaatmn of cartilage. .I. Amt., 69, 47-61. Huyhrctchs-Godin. G.. Hauscr. P. and Vats, G. 1979 Macrophage-fibrohlasta intcractlons in collagenase production and cartilage degradation. Biochem. J.. 184, 643-650 Juster. M. and Mowofian. A 1071. Sur la formation du squclctte: croissance d‘un OS long; appantlon premi&c d’un point d‘os\ificatmn. Pathol. Biol.. 19, 21-31. Kobayashi. I. and Ziff, M. lY75. Electron microscopic studies of the cartilage-pannua junction in rheumatoid arthritis. Arrhritis Rhrum.. 18, 475-483. Kuettncr, K. E. and Pauli. B. U. lY7X. Resistance of cartilage to normal and neoplastx invasion. In Mechunr.ms of Localized Bone 1~0s.~ (eds J. E Horton. T. M. Tarplay and W. F. Davis), Calcif. Tissue Abst. Suppl.. pp. 251-278. Kugler. J. H., Tomhnson. A., Wagstaff. A. and Ward. S. M. lY7Y. The role of cartilage canals in the formation of secondary centrcs of owfication. J. Anal.. 129, 4Y4-506. Langcr. K. 1876. Ucbcr das GcfBbsystem dcr Rdhrenknochen usw. Dmkschr. Aknd. W&s. Wit%.. 36, l-40 Levene. C. 1964. The patterns of cartilage canals. J. Amt., 98, 515-538. Luftl. A. M. 1970. Mode of growth. fate and functions of cartllagc canals. .I. Arm.. 106, 135-145.