The collagens are a family of extracellular matrix proteins that contain the repeating triplet sequence -Gly-X-Y-, in which the Y-position amino acid is often 4-hydroxyproline. Over 20 proteins have been identified as collagens and among other functions, they play an important role in determining size, shape and strength of bone and cartilage. Type I collagen is the major protein in bone, constituting over 90% of its total protein, whereas type II collagen is the principal protein in cartilage.

Prolyl 4-hydroxylase (EC 1.14.11.2) plays a central role in the synthesis of all types of collagens, as 4-hydroxyproline residues are essential for the formation of stable triple-helical collagen molecules. The vertebrate enzymes are ?₂?₂ tetramers. The enzyme requires Fe²⁺, 2-oxoglutarate, O₂ and ascorbate and it acts on proline residues in X-Pro-Gly sequences. The role of PDI in the enzyme is not related with its disulphide isomerase activity and the major function of this polypeptide in prolyl 4-hydroxylase is to keep the highly insoluble ? subunits in a catalytically-active non-aggregated conformation.

Prolyl 4-hydroxylase is expressed as two well-characterized isoenzymes in man. Previously known as ? subunit and enzyme, it is now called the ?(I) subunit and the type I enzyme. The ?(II) subunit forms a [?(II)]₂?₂ tetramer and is called type II enzyme. Data on coexpression in insect cells strongly argue against the existence of a mixed [?(I)?(II)?₂] tetramer in vertebrates. Very recently, the ?(III) subunit was cloned and coexpressed with the ? subunit. This novel type of isoenzyme is especially expressed in smooth muscle cells.

Recent studies have demonstrated that the type I isoenzyme is expressed especially by cells of mesenchymal origin and in developing and malignant tissues such as hepatocellular cancer cells and hepatoblastoma cells, whereas the type II isoenzyme is expressed especially by chondrocytes, endothelial cells and more differentiated cells. The type I isoenzyme represents about 90% or more of the total prolyl 4-hydroxylase activity in mouse kidney, heart, liver, skeletal muscle and skin. Osteoblasts strongly express both types of isoenzyme, but type I isoenzyme slightly more than type I. We report here that type I isoenzyme is the major type expressed in primary bone tumours, and demonstrate that immunostaining for type I and II collagens in primary chondrosarcomas and osteosarcomas is found mainly within cancer cells, whereas the positive signal was rarely found extracellularly.

Immunofluorescence staining

The material comprised of sections of primary bone tumours: chondrosarcomas (7), osteosarcomas (8), chordoma (1), chondroblastoma (1), osteoid osteoma (1), osteochondromas (5), chondromas (4), giant cell tumours of bone (5) and fibrous dysplasia (1). Two chondrosarcomas were grade I tumours and 5 were grade II on a scale from I to III, and the osteosarcomas consisted of one of grade II, 3 of grade III and 4 of grade IV on a scale from I to IV. Grading was based on the WHO classification of bone tumours.

Tissue samples for indirect immunofluorescence were immediately frozen in liquid nitrogen and stored at ?70°C. Cryostat sections (5 ?m thick) were cut, attached to SuperFrost plus glass slides (Menzel Gläser, Braunschweig, Germany) and fixed in precooled methanol for 10 min at ?20°C. After rinsing with PBS, pH 7.2, nonspecific antibody binding was blocked by incubating the sections with 1% bovine serum albumin (BSA) in PBS, pH 7.2, for 1 h at room temp, followed by incubation at 4°C overnight or for 1 h at room temperature with an affinity-purified polyclonal antibody R17 against the ?(I) subunit of type I prolyl 4-hydroxylase, diluted 1:100 (10 ?g/ml), or a monoclonal antibody M14 against the ?(II) subunit of type II prolyl hydroxylase using a nondiluted monoclonal antibody pool. Specificity of staining was further demonstrated by incubating the polyclonal ?(I) antibody overnight with 2 mg of type I prolyl 4-hydroxylase and the monoclonal ?(II) antiserum overnight with 2 mg of the type II enzyme and using these treated antibodies for immunostaining, whereupon no signal was detected. Control sections were also stained with 1:80 diluted mouse and rabbit non-immune isotype immunoglobulins (Dako, Glostrup, Denmark) as primary antibodies. Additionally, sections were stained with a monoclonal antibody against type IV collagen diluted 1:100 (Dako), a monoclonal antibody against CD34, diluted 1:25 (Novocastra Laboratories, Newcastle, UK), a polyclonal antibody against type I collagen, diluted 1:80 (15 ?g/ml; BioDesign International, Bingham, ME, USA), and a polyclonal antibody against type II collagen, diluted 1:40 (10 ?g/ml; BioDesign International). Some sections were also stained with a monoclonal antibody against the ? subunit of prolyl 4-hydroxylase, diluted 1:100 (Dako) and against human osteocalcin, diluted 1:40 (10 ?g/ml, BioDesign International). After thorough washing with PBS, a 1:100 diluted tetramethyl rhodamine isothiocyanate (TRITC)-conjugated polyclonal swine anti-rabbit antibody (Dako) or rabbit anti-mouse antibody (Dako) was applied and the samples were incubated for 60 min at room temp in the dark. After being washed with PBS, sections were mounted with Glycergel (Dako) and examined under an epifluorescence microscope (Aristoplan; Leica, Wetzlar, Germany) equipped with filters for TRITC fluorescence with automatic exposure time. For histological analysis, frozen sections from all the tissues studied were stained with haematoxylin and eosin using routine methods.

Fluorescence intensity was assessed semi-quantitatively: ?, negative; +, faint but detectable signal; ++, signal of moderate intensity; and +++, distinct signal. Staining of collagen was scored separately for extracellular and intracellular compartments.