The preparation of a collagenolytic enzyme from the medium of tissue cultures of tadpole tail fin and back skin is described. The yield of enzyme from the tissues of thyroxine-stimulated tadpoles was the same as that from nontreated tadpoles and very little activity was produced by the skin of mature frogs. The enzyme was concentrated as much as 300-fold by sequential ammonium sulfate precipitation, gel filtration, starch block electrophoresis, and DEAE-cellulose chromatography. Caseinolytic activity was considerably reduced but was still present at low levels in the final product. The purified enzyme attacks native calf skin collagen in solution at pH 7.6 reducing the viscosity 40-50% at 20°and 75% at 27°, and prevents reconstitution p JL^xtracts of tissues in which rapid physiologic resorption of collagen occurs have consistently failed to yield an enzyme which, at neutral pH and physiologic temperature, will attack native collagen (Mandl, 1961;Woessner, 1965; Lapiere and Gross, 1963). However collagenolytic activity has been detected in the medium of cultured tissues such as tadpole tail, gills, and gut, mammalian uterus, bone, and skin wounds (Gross and Lapiere, 1962; Gross et al., 1963;Walker et al., 1964;Grillo and Gross, 1964). This enzyme apparently is synthesized de nooo from living cells and accumulates in the culture chamber to detectable levels during the period of incubation. No enzyme activity was produced by frozen thawed tissues or by those exposed to puromycin (Gross and Lapiere, 1962; Eisen and .The tadpole collagenase has been harvested from the culture medium and purified (Nagai et al., 1963) and its mode of attack on collagen partially characterized. The native molecule is attacked at a single locus and across the three polypeptide chains cleaving off one-quarter length of the molecule at the "B" end.

Type I and type III collagen extracted from skin was purified by differential salt precipitation and chromatography. By heating to 37 degrees, type I formed after a lag phase a floppy and opalescent gel of high optical density and type III formed more rapidly a translucent and rigid gel of low optical density. Addition of type III to type I resulted in formation of gels of reduced optical density and lag phase related to the proportion of type III added. Phase contrast and scanning electronmicroscopy demonstrated the formation of thick bundles of type I, thin fibers of type III and bundles of intermediate size related to the proportion of type III. The relationship between collagen type and bundle architecture might prove most significant in conditioning the mechanical properties of the connective tissues in normal and pathological conditions.

An evaluation of the rheological properties of skin is provided by measuring "in situ" the ability of skin to withstand vertical forces of extension. In physiological conditions the biomechanical parameters are reproducible and related to age, sex, skin thickness and the microarchitecture of the connective tissue studied by scanning electron microscopy. These observations shed some light on the structure-function relationship of the fibrous framework of the connective tissue. Variation in the biomechanical parameters can be used to substantiate clinical diagnosis and to monitor the evolution and the therapy of various diseases of the connective tissue.

Osteoporosis developing during the first weeks after the onset of traumatic paraplegia was studied with cortical and cancellous samples of iliac crest and tibia of 14 patients, and compared to normals. We used a procedure of bone particle fractionation (according to degree of mineralization) that allowed us to establish a profile reflecting the metabolic remodeling of bone and to analyze the organic matrix of the newly synthesized tissue. In paraplegics, we observed a large increase in the proportion of little calcified bone in the cortical as well as in the cancellous bone. Based on amino acid analyses, we found a decreased number of hydroxyproline residues in the newly synthesized organic matrix from paraplegia bone resulting either from an alteration of the prolyl hydroxylation or from the presence of an excess of noncollagen polypeptides. These results, together with previously published data reporting increased urinary hydroxyproline and calcium kinetic parameters, suggest an enhanced rate of skeletal remodeling in acute paraplegia. When investigated 2 years after injury, the patterns of distribution approach that of normal subjects.

The effect of activated factor XIII (FXIIIa), the transglutaminase of blood coagulation, on some cellular functions was studied in skin and lung fibroblasts in vitro. FXIIIa repressed the overall protein synthesis and mainly collagen synthesis in a concentration-dependent manner and induced modifications in the proportion of the different types of newly synthesized collagen. The repression of collagen synthesis occurred in cells cultured on plastic (-40%), on coated fibronectin (-53%), on coated collagens (-38%) and within a collagen lattice (-16%). Preincubation of the cells with FXIIIa and labelling in its absence also resulted in such an inhibition. However, when embedded into a fibrin lattice cross-linked by FXIIIa, fibroblasts displayed a higher biosynthetic activity than in untreated fibrin gel. These results suggest that FXIIIa acts through a modulation of the cell-matrix interactions.