42 EQUINE VETERINARY JOURNALEquine vet. J. (2003) 35 (1) 42-47

Summary

Reasons for performing study: This study was designed toexamine a new role for cysteine proteinases in the pro c e s sof endochondral ossification.

Objectives: The aim of the present study was to investigate thep resence and distribution of cathepsin B and cathepsin L i nequine art i c u l a rc a rtilage during development.

Methods: Full-depth cartilage samples from a total of 40 horses(age range: 4 month fetuses to 2 years) were examined andenzymes detected by immunocytochemical localisation.

R e s u l t s : Observations on the presence of cathepsins B and Lrevealed significant age-related diff e rences, resulting inc l e a r division of the animals into 2 age groups: i) fetuses andneonates; ii) young growing horses (age 4 weeks to 2 years).Cathepsin B was not detected in cartilage from the majorityof fetuses and neonates but was located characteristically inc h o n d rocytes at the art i c u l a r surface and hypert rophic zonein all growing horses. In contrast, cathepsin L w a sp redominantly present in fetal and neonatal cart i l a g e ,located primarily in proliferating chondro c y t e s.

Conclusions: This study is the first to demonstrate diff e re n t i a land site-specific roles for cathepsin B and cathepsin L i nskeletal development in the horse.

Potential relevance: The demonstrated involvement ofcathepsins B and L in endochondral ossification is of re l e v a n c eto developmental orthopaedic diseases such as osteochondro s i sin which there is a focal failure of bone formation.

Introduction

Extracellular matrix (ECM) turnover is a key physiologicalevent in the metabolism of cartilage and bone and in themorphogenesis and growth of the skeleton. During growth, boneformation and skeletal development require replacement ofcartilage by bone matrix, a complex process known asendochondral ossification. In these processes collagens andproteoglycans, major components of the cartilage ECM, aredegraded through the actions of a variety of proteinases (Murphyand Reynolds 1993). One class of proteinases implicated incartilage and bone matrix turnover is the family of lysosomalcysteine proteinases, namely cathepsins B, H, L, S, C and K.This class of enzymes belongs to the larger group of papain-likeproteinases which also includes a number of plant proteinases

such as papain itself, bromelain, ficin, chymopapain and aleurain( Turk et al. 1997, 2000). Cathepsins B, H, L, S and K haverecently been detected at mRNA and protein levels in the growthplate of rats, mice and man (Ohsawa et al. 1993; Söderström e ta l . 1999; Nakase et al. 2000) and cathepsins B, L and K, inp a r t i c u l a r, have now been proposed to play a role in proteolysisduring endochondral ossification (Söderström et al. 1 9 9 9 ;Nakase et al. 2000).

Generally ubiquitously expressed, cathepsins B and L h a v ebeen shown, by immunolocalisation, in situ hybridisation andNorthern blot analysis, to be abundant in chondrocytes( H e r n a n d e z - Vidal et al. 1998; Söderström et al. 1999; Nakase e ta l . 2000), whereas cathepsin K expression is more restricted,being present predominantly in osteoclasts and, to a lesser extent,in hypertrophic chondrocytes (Rantakokko et al. 1996; Kafienahet al. 1998; Söderström et al. 1999; Gray et al. 2002). All 3enzymes are optimally active at slightly acidic pH (Kirschke e ta l . 1995). Since resorption of bone and cartilage generates a lowpH environment, these cysteine proteinases are thought to befunctionally involved in this process (Baron 1999). T h e s efindings, taken together with the detection of relatively highlevels of these enzymes in chondroclasts, hypertrophicchondrocytes and osteoclastic resorption pits at the osteochondraljunction (Baron 1999; Olsen 1999), their ability to degrade boneand cartilage ECM and the capacity of specific inhibitors toprevent degradation (Esser et al. 1994; Everts et al. 1998), hasprompted the proposal of a new role for these enzymes inendochondral ossification and skeletal development (Söderströmet al. 1999; Nakase et al. 2000). More recently, a specific role forcathepsin L in the degradation of cartilage ECM duringsuccessive developmental stages of the mandibular condyle ofthe rat has further been suggested from mRNA expression studiesusing in situ hybridisation, RT-PCR and Southern blottinganalysis (Ohba et al. 2 0 0 0 ) .

Although all these data provide an indication of involvementof the cysteine proteinases in endochondral ossification, directevidence of their role is not yet available. To date, only twostudies (Söderström et al. 1999; Ohba et al. 2000) have addressedthe involvement of the cysteine proteinases in skeletaldevelopment, by examination of their synthesis and spatialdistribution in growth plate cartilage in long bones of mouse, manand rat. No such investigation of these enzymes has been reportedin the horse. In the present study, we have usedimmunocytochemistry to establish the spatial and temporal

Differential distribution of cathepsins B and L in articularcartilage during skeletal development in the horseK. E. GLÄSER, M. E. DAVIES* and L. B. JEFFCOTT

Department of Clinical Veterinary Medicine, University of Cambridge, Madingley Road, Cambridge CB3 0ES, UK.

Keywords: horse; skeletal development; endochondral ossification; cathepsin B; cathepsin L

*Author to whom correspondence should be addressed.

K. E. Gläser et al. 43

distribution of cathepsins B and L in growth cartilage from thedistal femur of fetal, neonatal and growing horses up to age 2years, with the eventual aim of identifying a role for theseenzymes in equine skeletal development.

Materials and methods

Cartilage samples for immunocytochemistry

Articular cartilage was collected from the distal femurs of fetal, neonatal and young growing horses up to age 2 years. Thestudy was restricted to Thoroughbreds and Thoroughbred-crosses,cartilage was taken only from joints which showed no grossevidence of pathological changes and a total of 40 horses wasexamined. Full-depth (i.e. articular surface down to subchondralbone) cartilage samples were removed within 12 h postmortemand fixed immediately in 10% buffered formalin for at least 24 h,followed by decalcification in 8% formic acid for 1 week. After routine embedding in paraffin, the samples werestored until sectioned for use in immunolocalisation.

Antisera

The affinity-purified polyclonal cathepsin L antibody (sc-6499)was obtained from Santa Cruz1 and had been raised in goatsagainst a peptide mapping at the amino terminus of the heavychain of human cathepsin L. Its cross-reactivity against the equinehomologue was confirmed in this study by SDS-PAGE2 andWestern blotting. The polyclonal anti-cathepsin B serum waskindly donated by Dr D.J. Buttle (University of Sheffield MedicalSchool). It had been raised in sheep against cathepsin B purifiedfrom human liver and characterised as described previously(Buttle et al. 1988). Cross-reactivity with equine cathepsin B wasconfirmed in a previous study (Hernandez-Vidal et al. 1998).

SDS-PAGE and Western blotting

Extracts of horse liver and kidney were prepared following theprotocols of Hernandez-Vidal et al. (1997) and Roth (1996),respectively, and used to confirm the cross-reactivity of thecathepsin L antibody with the equine homologue. Proteins in thesamples and prestained Kaleidoscope molecular weight markers3

were separated by SDS-gel electrophoresis (SDS-PAGE) in 12%reducing gels using the method of Laemmli (1970) as describedpreviously (Hernandez-Vidal et al. 1997, 1998). Proteins wereeither visualised by staining in Coomassie blue or were Westernblotted by electro-transfer onto Hybond ECL n i t r o c e l l u l o s emembrane4 according to Towbin et al. (1979). Immunodetectionof the protein bands was performed using anti-cathepsin L at adilution of 1:50 as primary antibody, followed by horseradishperoxidase-conjugated rabbit anti-goat immunoglobulins5 diluted1:200 as secondary antibody. Bands were visualised bychemiluminescence using a standard kit according to them a n u f a c t u r e r’s instructions6 and recorded on Biomax MR-1imaging film7.

Immunolocalisation of cathepsins B and L

P a r a ffin sections of the cartilage samples were cut at a thickness of6 µm, mounted on poly-L-lysine coated slides and dewaxed.Cathepsins B and L were then immunolocalised on the sections

using a standard biotin/streptavidin-horseradish peroxidase method,essentially as described previously (Davies et al. 1994). Briefly,after removal of endogenous peroxidase activity by treatment with1% hydrogen peroxide in methanol for 15 min, the sections wereincubated with either anti-cathepsin B (1:50) or anti-cathepsin L(1:50) as primary antibody, followed by biotinylated rabbit anti-s h e e p5 or rabbit anti-goat8 immunoglobulins (1:200), respectively,as secondary antibody. Controls were performed using the samedilutions of normal sheep and goat sera to replace the respectiveprimary antibodies. Nuclei were counterstained usinghaematoxylin. Binding of antibodies was detected by addition ofstreptavidin-horseradish peroxidase complex4 and the colourreaction was developed with diaminobenzidine. The distribution ofthe resulting brown staining was determined blindly using a NikonOptiphot microscope and photographed with Fujifilm 100.

Statistics

Data were analysed using Fisher ’s Exact Test (2-tailed), taking assignificant Pvalues <0.05.

Results

C ro s s - reactivity of cathepsin L antibody with the equinehomologue

The commercial anti-human cathepsin L antibody cross-reactedwith a band of approximately 29 kDa in both the equine liver andequine kidney extracts. Since this corresponded to the expectedmolecular weight reported for rat kidney1, we concluded that theantiserum recognised the equine enzyme and was suitable for usein immunolocalisation.

Presence of cathepsins B and Lin articular cartilage samples

A total of 40 horses was examined for positive immunostainingfor cathepsin B and 31 of these were examined for cathepsin Limmunostaining. During examination of immunostaining forcathepsins B and L in the full-depth articular cartilage samples itwas strikingly apparent from the staining pattern observed that thehorses could be divided into 2 groups: (i) fetuses and neonates upto 7 days post partum and (ii) young horses aged between 4 weekspost partum and 2 years (Fig 1). The absence of staining forcathepsin B in the majority of fetal and neonatal articular cartilagesamples was noted. Of the 31 fetuses and neonates examined,cathepsin B staining could be detected in only 8 animals (25.8%)and, in these samples, the staining was of very low intensity. Bycontrast, all young horses (100%) showed strong staining forcathepsin B (Fig 1a). This difference in positive immunostainingbetween the 2 groups was significant (P = 0.002). A similar, butreversed, grouping was observed for cathepsin Limmunostaining.The majority (71.4%) of cartilage samples from fetuses andneonates showed positive staining for cathepsin L and wassignificantly different (P = 0.002) from the same age groupstaining positive for cathepsin B (25.8%) (graph not shown),whereas the comparison of positive cathepsin Lstaining in fetusesvs. young horses (30%) did not reach a significant level (Fig 1b).In addition, among the young horses, cathepsin L showed asignificantly (P = 0.005) lower positive immunostaining whencompared with positive immunostaining for cathepsin B (graphnot shown). Comparison of Figures 1a and b demonstrates clearly

44 Distribution of cathepsins B and Lin articular cartilage

1009080706050403020100

the different expression of these 2 related enzymes in the 2 agegroups. Controls in which normal sheep and goat sera replaced theprimary antibody were consistently negative.

Zonal distribution of cathepsins B and L in articular cartilagefrom fetuses and neonates compared with young horses

The spatial distribution of cathepsins B and L throughout thedifferent zones of the full-depth articular cartilage samples wasexamined carefully and observations summarised in Table 1. Inthe 9 young horses (age 4 weeks post partum to 2 years) thestaining distribution pattern for cathepsin B was the same in allcases, being present (100%) at high levels in chondrocytes at thearticular surface and in the hypertrophic zone (Figs 2d,e). Highmagnification showed the enzyme to be intracellular, located inthe lysosomes. Cathepsin B was never detected in the proliferativezone. By comparison, this typical zonal staining pattern forcathepsin B was totally absent in the fetal and neonatal articularcartilage samples (Figs 2a,b). In the 8 horses in which positivestaining for cathepsin B was observed, it was either (in 4 cases)restricted to a few chondrocytes immediately surroundingcartilage canals (Fig 2c) or (in 4 cases) was present in small fociof chondrocytes randomly distributed throughout the cartilage inareas otherwise negative for cathepsin B (Table 1).

The distribution of cathepsin L in the 2 age groups wasnoticeably different from that of cathepsin B. In contrast tocathepsin B, cathepsin Lwas not generally detected at the articularsurface or in the hypertrophic zone, but was locatedpredominantly in chondrocytes throughout the proliferating zone(Fig 2f). The typical staining pattern for cathepsin L i nproliferating chondrocytes was seen in the majority (71.4%) offetuses and neonates and, in only 3 cases (30%), in the younghorse group (Table 1). As expected, cathepsin L, like cathepsin B,was present in the lysosomes and no staining for the enzyme wasobserved in the cartilage extracellular matrix.

Discussion

The recently suggested involvement of the cysteine proteinases inthe process of endochondral ossification (Söderström et al. 1999;Nakase et al. 2000; Ohba et al. 2000) opens up a novel role forthese enzymes in skeletal development. At present, there is littleinformation available concerning the differential distribution ofthese enzymes during development and, to date, no data have beenreported for the horse. In this study, developmental changes in thezonal distribution of cathepsins B and L in equine articularcartilage have been investigated by immunohistochemistry.

Prior to use for immunolocalisation, the cross-reactivity of theanti-human cathepsin L antibody with the equine enzyme was

confirmed. Western blot analysis showed immunoreactivity of theantibody with a 29 kDa protein present in equine kidney and liverextracts, corresponding presumably to the single chain matureform of cathepsin L (Ishidoh et al. 1998; Xing and Mason 1998)and, therefore, the antibody was deemed suitable for use.

Results from these immunolocalisation studies havedemonstrated a marked age-related presence, in epiphyseal growthcartilage, of cathepsins B and Lat different stages of endochondralossification in growing horses. Our findings confirm and extendthe observations of 2 other studies (Söderström et al. 1999;Nakase et al. 2000) in which these enzymes were localised at theosteochondral junction and secondary ossification centre duringendochondral ossification in mice and man.

The presence of cathepsins B and L, differentially distributedthroughout all zones of the cartilage, is consistent with aninvolvement in the extensive proteolytic degradation of thecartilagenous ECM which is known to occur during bone formation.The precise function of these enzymes in this complex processremains, however, unknown. There is growing evidence that in the

TABLE1: Spatial distribution of cathepsins B and Lthroughout the different zones of equine articular cartilage

Positive immunostaining of articular surface of proliferative around cartilage canals

and hypertrophic zone zone or focal loci No staining

Cathepsin BFetuses and neonates (n = 31) 0 0 8 (25.8%) 23 (74.2%)Young horses (n = 9) 9 (100%) 0 0 0

Cathepsin LFetuses and neonates (n = 21) 0 15 (71.4%) 8 (38.1%) 6 (28.6%)Young horses (n = 10) 0 3 (30%) 0 7 (70%)

1009080706050403020100

100%

25.8%

Fetuses, neonates Young horses

a)

30%

71.4%

Fetuses, neonates Young horses

b)

P = 0.002

P>0.05

Fig 1: Percentage of fetuses and neonates compared with young horses(up to age 2 years) showing positive immunostaining for cathepsins B andL in articular cartilage. (a) Cathepsin B in fetuses and neonates versusyoung horses (P= 0.002). (b) Cathepsin L in fetuses and neonates versusyoung horses (not significant, P> 0.05).

K. E. Gläser et al. 45

early stages of endochondral ossification, in preparation formineralisation, the matrix macromolecules, collagen type II,collagen type X and large aggregating proteoglycan (aggrecan) needto be removed from the cartilage matrix in order that endochondral

ossification can proceed (Buckwalter et al. 1987; Alini et al. 1 9 9 2 ;Sires et al. 1995). Although these matrix components are probablyreduced by inhibition of synthesis, there is now evidence that theycould also be removed from the ECM by the combined action of

a) d)

b)e)

c) f)

Fig 2: Immunolocalisation of cathepsins B and Lin articular cartilage samples from the distal femur of fetal and young horses. Articular surface of a) a9-month-old fetus and d) a 2-year-old young horse immunostained for cathepsin B. Hypertrophic zone/osteochondral junction of b) a 9-month-old fetusand e) a 2-year-old horse immunostained for cathepsin B. Cartilage canals in proliferating zone immunostained for c) cathepsin B and f) cathepsin L ina 10-month-old fetus. Nuclei were counterstained with haematoxylin in a–e. Scale bars 100 µm.

46 Distribution of cathepsins B and Lin articular cartilage

metalloproteinases and cathepsin B (Sires et al. 1995). Certainly,there is no doubt that both cathepsins B and L are able to degradethese ECM macromolecules (Murphy and Reynolds 1993;McIlwraith 1996; Söderström et al. 1999; Ohba et al. 2 0 0 0 ) .

The age-related fluctuation in levels and significantly differentpatterns in the distribution of both enzymes, particularly forcathepsin B, imply distinct developmental regulation and possiblyzonal differences in the function of these enzymes. A cleardifference between the distribution of these 2 cysteine proteinaseswas also revealed by the striking absence of cathepsin B comparedto the widespread presence of cathepsin Lin all zones, particularlythe proliferative, of the growth cartilage in the majority of fetaland neonates. This unexpected finding suggests that cathepsin Bdoes not play an important role in the early, prepartum stages ofskeletal development but might be involved in ECM turnoverinduced by the application of mechanical pressures such asincreased loading and movement following birth and as the animalmatures. Interestingly, we do have some evidence from otherstudies currently in progress that mechanical impact of articularcartilage correlates with increased expression of cathepsin B (E.A.Bowe and M.E. Davies, unpublished data). Cathepsin L, however,being preferentially available at early stages of development, mayindeed have an age-related functional role. To our knowledge,there have been no other studies on the distribution of the cysteineproteinases in fetal and neonatal material and our findings are,therefore, novel and our interpretation speculative at present. Theconsistent localisation of cathepsin B in the articular andhypertrophic zones of the cartilage in all young growing horses inthe age range from 4 weeks post partum to 2 years is in agreementwith our previous reports in which this typical staining pattern wasrecorded for a total of 14 h (Hernandez-Vidal et al. 1996, 1998)and, presumably, already reflects the distribution and zonalfunction of this enzyme in mature articular cartilage.

The abundance of cathepsin B in hypertrophic chondrocyteshas been reported by others (Ohsawa et al. 1993; Söderström et al.1999) and it might be noteworthy that, in the few cases whencathepsin B was detected in fetal and neonatal cartilage, it waslocalised in chondrocytes immediately surrounding cartilagecanals. Vascular canals are more numerous in fetal and newborncartilage and it has been suggested that they could be essential fornutrition of the cartilage and for development of the secondaryossification centre (Wilsman and van Sickle 1970). Interestingly,and in line with our unusual cathepsin B localisation in fetuses andneonates, it has been reported that the chondrocytes in the vicinityof cartilage canals resemble hypertrophic chondrocytes (Horton1993). The precise role of cathepsin B in these particularchondrocytes is, however, open to speculation.

To conclude, this study is the first to present evidence fordifferential and site-specific roles for 2 cysteine proteinases inskeletal development in the horse. Based on temporal and spatialdistribution patterns, it is proposed that cathepsin L might beinvolved in the early stages of endochondral ossification, whereascathepsin B could be more important in post natal, mechanically-induced turnover of cartilage ECM.

Acknowledgements

We wish to thank Dr David Buttle for the generous donation of theanti-cathepsin B serum used in this study. The authors are muchindebted to Rossdale and Partners Veterinary Surgeons, Newmarket,UK, for their continued provision of equine joint tissue.

Manufacturers’addresses

1Santa Cruz, Insight Biotechnology Ltd, Wembley, London, UK. 2Life Technologies, GIBCO-BRL, Paisley, UK.3Bio-Rad, Hemel Hempstead, Hertfordshire, UK.4Amersham Pharmacia Biotech UK Ltd, Little Chalfont, Buckinghamshire, UK.5DAKO Ltd, Ely, Cambridgeshire, UK.6Roche Molecular Biochemicals, Lewes, East Sussex, UK.7Kodak, Rochester, New York, USA8Vector Laboratories, Peterborough, Cambridgeshire, UK

References

Alini, M., Matsui, Y., Dodge, G.R. and Poole, A.R. (1992) The extracellular matrixof cartilage in the growth plate before and during calcification: changes incomposition and degradation of Type II collagen. Calcif. Tiss. Int. 50, 327-335.

Baron, R. (1999) Anatomy and ultrastructure of bone. In: Primer on the MetabolicBone Diseases and Disorders of Mineral Metabolism, Ed: M.J. Favus, LippincottWilliams and Wilkins, Philadelphia. pp 3-10.

Buckwalter, J.A., Rosenberg, L.C. and Ungar, R. (1987) Changes in proteoglycanaggregates during cartilage mineralisation. Calcif. Tiss. Int. 41, 228-234.

Buttle, D.J., Bonner, B.C., Burnett, D. and Barrett, A.J. (1988) A catalytically activehigh Mr form of human cathepsin B from sputum. Biochem. J. 254, 693-699.

Davies, M.E., Horner, A. and Franz, B. (1994) Intercellular adhesion molecule-1(ICAM-1) and MHC Class II on chondrocytes in arthritic joints from pigsexperimentally infected with Erysipelothrix rhusiopathiae. FEMS Immunol.Med. Microbiol. 9, 265-272.

Esser, R.E., Angelo, R.A., Murphey, M.D., Watts, L.M., Thornburn, L.P., Palmer,J.T., Talhouk, J.W. and Smith, R.E. (1994) Cysteine proteinase inhibitorsdecrease articular cartilage and bone destruction in chronic inflammatoryarthritis. Arth. Rheum. 37, 236-247.

Everts, V., Delaisse, J.M., Korper, W. and Beertsen, W. (1998) Cysteine proteinasesand matrix metalloproteinases play distinct roles in the subosteoclastic resorptionzone. J. Bone Min. Res. 13, 1420-1430.

Gray, A.W., Davies, M.E. and Jeffcott, L.B. (2002) Localisation and activity ofcathepsins K and B in equine osteoclasts. Res. vet. Sci. 72, 95-103.

H e r n a n d e z - Vidal, G., Davies, M.E. and Jeffcott, L.B. (1996) Localisation ofcathepsins B and D in equine articular cartilage. Pferdeheilkunde 12, 371-373.

H e r n a n d e z - Vidal, G., Jeffcott, M.E. and Davies, M.E. (1997) Cellular heterogeneity incathepsin D distribution in equine articular cartilage. Equine vet. J. 2 9, 267-273.

Hernandez-Vidal, G., Jeffcott, L.B. and Davies, M.E. (1998) Immunolocalisation ofcathepsin B in equine dyschondroplasic articular cartilage. Vet. J. 156, 193-201.

Horton, W.L. (1993) Morphology of connective tissue. In: Connective Tissue and itsHeritable Disorders, Eds: P. Royce and P. Steinmann, Wiley-Liss, New York. pp73-84.

Ishidoh, K., Saido, T.C., Kawashima, S., Hirose, M., Watanabe, S., Sato, N. andKominami, E. (1998) Multiple processing of procathepsin L to cathepsin L invivo.Biochem. Biophys. Res. Comm. 252, 202-207.

Kafienah, W., Bromme, D., Buttle, D.J., Croucher, L.J. and Hollander, A.P. (1998)Human cathepsin K cleaves native type I and II collagens at the N-terminal endof the triple helix. Biochem. J. 331, 727-732.

Kirschke, H., Barrett, A.J. and Rawlings, N.D. (1995) Proteinases 1: Lysosomalcysteine proteinases. In: Protein Profile , Vol. 2, Ed: P. Sheterline, AcademicPress, London. pp 1587-1643.

Laemmli, U.K. (1970) Cleavage of structural proteins during the assembly of thehead of bacteriophage T4. Nature 227, 680-685.

Murphy, G.M. and Reynolds, J.J. (1993) Extracellular matrix degradation. In:Connective Tissue and its Heritable Disorders, Eds: P. Royce and P. Steinmann,Wiley-Liss, New York. pp 287-316.

McIlwraith, C.W. (1996) General pathobiology of the joint and response to injury. In:Joint Disease in the Horse, Eds: C.W. McIlwraith and G.W. Trotter, W.B.Saunders Co., Philadelphia. pp 40-70.

Nakase, T., Kaneko, M., Tomita, T., Myoui, A., Ariga, K., Sugamoto, K., Uchiyama,Y., Ochi, T. and Yoshikawa, H. (2000) Immunohistochemical detection ofcathepsin D, K and Lin the process of endochondral ossification in the human.Histochem. Cell Biol. 114, 21-27.

Ohba, T., Ohba, Y. and Moriyama, K (2000) Synthesis of mRNAs for cathepsins Land H during development of the rat mandibular condyle cartilage. Cell Tiss. Res.302, 343-352.

Ohsawa, Y., Nitatori, T., Higuchi, S., Kominami, E. and Uchiyama, Y. (1993)Lysosomal cysteine and aspartic proteinases, acid phosphatase and an

K. E. Gläser et al. 47

endogenous cysteine proteinase inhibitor, cystatin-beta, in rat osteoclasts. J.Histochem. Cytochem. 41, 1075-1083.

Olsen, B.R. (1999) Bone morphogenesis and embryonic development. In: Primer onthe Metabolic Bone Diseases and Disorders of Mineral Metabolism, Ed: M.J.Favus, Lippincott Williams and Wilkins, Philadelphia. pp 11-14.

Rantakokko, J., Aro, H.T., Savontaus, M. and Vuorio, E. (1996) Mouse cathepsin K:c D N A cloning and predominant expression of the gene in osteoclasts and in somehypertrophing chondrocytes during mouse development. FEBS Lett. 3 9 3, 307-313.

Roth, W. (1996) Furless, eine Mausmutante mit Haut- und Haaranomalien,verursacht durch eine Missense-Mutation in der lysosomalen Cystein-ProteinaseCathepsin L. PhD Thesis, Georg-August-Universitaet Goettingen, Germany.

Sires, U.I., Schmid, T.M., Fliszar, C.J., Wang, Z.Q., Gluck S.L. and Welgus, H.G.(1995) Complete degradation of type X collagen requires the combined actionof interstitial collagenase and osteoclast-derived cathepsin-B. J. clin. Invest.9 5, 2089-2095.

Söderström, M., Salminen, H., Glumoff, V., Kirschke, H., Hannu, A. and Vuorio, E.(1999) Cathepsin expression during skeletal development. Biochim. Biophys.Acta 1446, 35-46.

Towbin, H., Staehelin, T. and Gordon, J. (1979) Electrophoretic transfer of proteinsfrom polyacrylamide gels to nitrocellulose sheets: procedure and someapplications. Proc. natl. Acad. Sci. USA 76, 4350-4354.

Turk, B., Turk, V. and Turk, D. (1997) Structural and functional aspects of papain-like cysteine proteinases and their protein inhibitors. Biol. Chem. 378, 141-150.

Turk, B., Turk, D. and Turk, V. (2000) Lysosomal cysteine proteases: more thanscavengers. Biochim. Biophys. Acta 1477, 98-111.

Wilsman, N.J. and van Sickle, D.C. (1970) The relationship of cartilage canals to theinitial osteogenesis of secondary centres of ossification. Anat. Rec. 168, 381-392.

Xing, R. and Mason, R.W. (1998) Design of a transferrin-proteinase inhibitorconjugate to probe for active cysteine proteinases in endosomes. Biochem. J.336, 667-673.