Characterization of procollagen synthesized by matrix-free cells isolated from chick embryo tendons.

The genetic type and molecular structure of the precursor forms of collagen synthesized by matrix-free tendon cells isolated from 17-day old chick embryos were examined by chromatographic and electrophoretic techniques. The [14C]proline-labeled collagenous proteins secreted by the cells resolved on diethylaminoethylcellulose into two peaks, A and B. Both peaks contained type I collagenous proteins since on chromatography on carboxymethylcellulose, after limited pepsin proteolysis, both peaks contained alpha1 and alpha2 chains of collagen in a 2:1 ratio, and cyanogen bromide peptide maps of the 14C-labeled protein in both peaks were similar to cyanogen bromide peptide maps derived from authentic type I collagen. Enzymatic digestion with purified mammalian collagenase demonstrated that the collagen precursor in peak B contained noncollagenous peptide extensions at both the amino- and carboxy-terminal ends of the molecule, while peak A had only carboxy-terminal extension peptides. Although both the amino- and carboxy-terminal extensions incorporated radioactive cystine, only the carboxy-terminal extensions contained interchain disulfide bonds. The carboxy-terminal extensions were also shown to incorporate radioactive tryptophan. Since most of the precursor forms of collagen recovered in the incubation medium chromatographed in peak B, it is concluded that matrix-free tendon cells secrete only type I procollagen with extension peptides at both the amino- and carboxy-terminal ends of the molecule.     

Purification and partial characterization of the type II procollagen synthesized by embryonic cartilage cells

Matrix-free cells were prepared from sternal cartilages of 17-day-old chick embryos, and procollagen synthesized and secreted by the cells was isolated by ion exchange chromatography on carboxymethyl cellulose and by gel filtration. The isolated protein was homogeneous by polyacrylamide gel electrophoresis in sodium dodecyl sulfate and it appeared to consist of identical pro-α chains linked by interchain disulfide bonds. Amino acid analysis and cyanogen bromide peptide mapping of the purified procollagen demonstrated that it had structures similar to Type II collagen. The amino acid composition was also consistent with the conclusion that the peptide extensions on the pro-α chains of procollagen contained amino acid sequences not found in the collagen portion of the molecule. Segment-long-spacing aggregates were prepared from the procollagen, and aggregates demonstrated the same banding pattern as is found in segment-long-spacing aggregates prepared from Type II collagen. The segment-long-spacing aggregates from procollagen revealed, however, the presence of NH₂-terminal extensions of about 150 Å in length. In addition, the procollagen molecules contained irregularly shaped, large extension peptides at the COOH-terminal end of the molecule.     

Synthesis of procollagen by matrix-free cells from embryonic-chick arteries

Cells were isolated from the major arteries of 17-day chick embryos by digestion of the tissue with collagenase and trypsin. The cells, when examined immediately after isolation, exhibited a high degree of viability and they were shown to synthesize and secrete procollagen at a high and constant rate for several hours when incubated in suspension in modified Krebs medium. Continuous labelling of the cells with [(14)C]proline demonstrated a lag of about 30min between the time at which the synthesis of non-diffusible peptide-bound hydroxy[(14)C]proline became linear and the time at which its secretion into the medium became linear. This lag time compares with that of 18min observed for freshly isolated matrix-free cells from embryonic-chick tendon, which synthesize and secrete the same type of collagen. Gel-filtration chromatography and polyacrylamide-gel electrophoresis indicated that the collagenous polypeptides secreted into the medium were in the precursor form, known as procollagen, and that the constituent pro-alpha-chains were linked by interchain disulphide bonds and were also in a triple-helical conformation. Characterization of the secreted procollagen by gel-filtration chromatography, polyacrylamide-gel electrophoresis, DEAE-agarose chromatography, and polyacrylamide-gel electrophoresis of peptides obtained by CNBr cleavage, indicated that the predominant form was type-I procollagen. This work extends the range of freshly isolated matrix-free cell systems, which have been characterized for use in studies on the biosynthesis and secretion of procollagen, and it indicates differences in the rates of secretion of procollagen in different cell types secreting the same type of procollagen.     

The subcellular fractionation of embryonic chick tendon and cartilage cells: a re-examination.

A re-examination of the subcellular fractions obtained from matrix-free chick tendon and cartilage cells has been made since the discovery that three out of four of the micrographs of chick tendon microsomal fractions published in an earlier paper from this laboratory were not authentic. The present studies demonstrate that by using the procedures previously reported it is possible to isolate microsomal and submicrosomal fractions from tendon and cartilage cells which exhibit typical morphology when examined by electron microscopy. These observations are consistent with our original biochemical characterization of subcellular fractions, which we know to be valid. Other publications from this laboratory in which these fractionation procedures have been applied to studies of collagen biosynthesis are in no way compromised, and indeed, most of our data have been confirmed by several other laboratories.     

Mechanism for the regulation of post-translational modifications of procollagens synthesized by matrix-free cells from chick embryos

Three possible mechanisms are considered to account for the variations of post-translational modifications in different collagen types. 1) The cells have different amounts of post-translational modifying enzymes, 2) the rate of prolylhydroxylation of different procollagen types is varied, and 3) the rate of chain association of pro-alpha chains of different collagen types is modulated. In an attempt to examine the three possibilities, we have determined the activities of prolyl hydroxylase and lysyl hydroxylase, and we have examined the kinetics of the secretion of procollagens and the kinetics of pro-gamma chain formation of different procollagen types in matrix-free cells isolated from tissues of 17-day-old chick embryos. Type II collagen synthesized by cartilage cells contains more hydroxylysine than type I collagen synthesized by tendon and cornea cells. It was found, however, that cartilage cells contain significantly less lysyl hydroxylase than tendon and cornea cells. In contrast, we found only a small difference in the amount of prolyl hydroxylase in tendon, cornea, and cartilage cells. The secretion of type I procollagen by tendon and cornea cells can be described by two first order processes. In contrast, the secretion of type II procollagen by cartilage cells, type IV procollagen by lens cells, and type V procollagen by cornea cells can be described by single first order processes. Examination of the formation of pro-gamma components of procollagen types I and II revealed that it occurs via intermediate dimers of two pro-alpha chains. The formation or pro-gamma(I) chains in tendon and cornea cells is about three times faster than the formation of pro-gamma(II) chains in cartilage cells. These results are consistent with the hypothesis that the rate of association of pro-alpha chains regulates the synthesis of procollagens with different degrees of post-translational modifications.     

Isolation and characterization of proteoglycans synthesized in ovo by embryonic chick cartilage and new bone

Cartilage proteoglycans have been well characterized in a number of developing systems, both in vitro and in vivo, but the newly synthesized molecules have been analyzed only from culture material. Because of potential culture artifacts, an attempt was made to characterize the proteoglycans newly synthesized in ovo in chick embryo sternum, tibial epiphysis, and tibial shaft. These in ovo synthesized proteoglycans share many structural features with chick proteoglycans synthesized by chondrocytes in culture including average monomer size, chondroitin sulfate chain size, keratan sulfate chain size, and the ability to aggregate with hyaluronic acid. Moreover, the newly synthesized in ovo proteoglycans, notably those of the tibial epiphysis, display reproducible changes in their structure as a function of embryonic age. These changes correlate with similar changes documented for chick cartilage proteoglycans synthesized in culture. Finally, the proteoglycans synthesized in ovo in the day 17 tibial shaft include, in addition to cartilage proteoglycans, one proteoglycan which seems to be characteristic of bone.     

Studies on the nature of procollagen synthesized by chick embryo polysomes

Analysis of pro-α chains released from chick embryo membrane polysomes indicates that they are not disulfide bonded and have a molecular weight of approximately 120,000. Therefore, disulfide bonding which has been observed between pro-α chains is not essential for their completion and release.     

Properties of highly purified lysyl oxidase from embryonic chick cartilage.

Lysyl oxidase the enzyme which oxidately deaminates lysine residues in collagen and elastin, was purified from embryonic chick cartialge by employing an affinity column of lathyritic rat skin collagen coupled to Sepharose, followed by separation on DEAE-cellulose. An enzyme preparation was obtained which was pure as shown by polyacrylamide gel electrophoresis. The specific activity was 1800-fold higher than that of the original extract. The pure enzyme utilized both collagen and elastin substrate. Furthermore, the ratios of enzyme activity with elastin substrate versus that with collagen substrate were the same at all stages of purity. Only one protein band was found after polyacrylamide gel electrophoresis of the pure lysyl oxidase in sodium dodecyl sulfate and mercaptoethanol. The molecular weight was estimated to be 28000. It was found that the enzyme contained a large number of cysteine and tyrosine residues. Evidence was obtained for molecular heterogeneity of lysyl oxidase. The enzyme eluted from DEAE-cellulsoe in at least four distinct regions. When the peaks were rechromatographed separately, they eluted at salt concentrations similar to those of the original chromatogram. However, the substrate specificity and the electrophoretic mobility on polyacrylamide gel were the same for all enzyme fractions.     

Rate of helix formation by intracellular procollagen and protocollagen. Evidence for a role for disulfide bonds

Matrix-free cells from embryonic tendons were incubated under conditions in which they synthesized and accumulated protocollagen, the unhydroxylated form of procollagen, which is non-helical at 37°. Limited digestion with pepsin demonstrated that when the accumulated protocollagen was hydroxylated intracellularly to procollagen, or when the cells were cooled below the T_m of protocollagen, the protein became triple-helical in about 5 min, or in a fraction of the time required for isolated α chains to become helical. When disulfide bonds in the NH₂-terminal extensions of protocollagen were reduced by treating the cells with dithiothreitol, the rate of helix formation was markedly decreased. The results demonstrated that the NH₂-terminal extensions found in protocollagen and procollagen play an important role in formation of the triple-helix during biosynthesis.     

Regulation of the glycosylations of collagen hydroxylysine in chick embryo tendon and cartilage cells

The regulation of the glycosylations of hydroxylysine was studied in isolated chick-embryo cells by labelling with a [¹⁴C]lysine pulse. The course of the procollagen lysyl modifications was compared in tendon and cartilage cells, and the effect on the glycosylations of the degree of lysyl hydroxylation and the concentration of Mn²⁺ and Fe²⁺ were also studied, in tendon cells. Procollagen triple helix formation was inhibited in most experiments in order to eliminate the effect of this process on the continuation of the reactions.Both in the tendon and cartilage cells the intracellular lysyl modifications proceeded in a biphasic fashion. After an initial sharp linear increase, the reactions did not cease but were protracted at a slower but constant rate. Lysyl hydroxylation was followed by rapid galactosylation in both cell types and this was followed almost immediately by rapid glucosylation, suggesting a close association of the corresponding enzymes. The data further suggest that other factors must also exist, in addition to the differences in the timing of triple helix formation and the actual hydroxylysine content, which are responsible for the different amounts of galactose in the collagens synthesized by these cell types. The amount of glycosylgalactosylhydroxylysine nevertheless seemed to be determined by the available acceptor sites, i.e., the amount of galactosylhydroxylysine.In further experiments wiht tendon cells the oxygen participating in lysyl hydroxylation was displaced by nitrogen at various points in time. When the degree of lysyl hydroxylation was reduced to less than one-third of the original, the total amounts of glycosylated residues decreased correspondingly, but their proportion relative to total hydroxylysine remained unchanged.Extra Mn²⁺ increased the proportion of galactosylated hydroxylysine, suggesting that the activity of hydroxylysyl galactosyltransferase is not saturating in respect of the catalyzed reaction. Experiments on the addition of Fe²⁺ or its chelation by α, α′-dipyridyl gave indications that the presence of this co-factor is not required for either glycosylation reaction in isolated tendon cells.     

Metabolic kinetics of proteoglycans by embryonic chick sternal cartilage in culture.

Explant cultures of embryonic chick sternum have been widely studied, but the kinetics of biosynthesis of proteoglycans by this tissue in culture has not been characterized. Caudal cartilaginous portions of 16-day-old embryonic chick sterna were cultured for 8 days. Histological examination showed that the fresh cartilage contained morphologically homogenous chondrocytes, which were embedded in a uniform extracellular matrix. After culture for 8 days, the histological appearance of the explant remained unchanged but the tissue increased in size with time as indicated by a progressive increase in DNA content and in the content of glycosaminoglycan and collagen. Rates of degradation and release from the tissue of proteoglycans labeled in ovo with 35S were first order during culture, as were the unlabeled proteoglycans. Proteoglycan synthesis was high during the first 2 days of culture, and this then gradually decreased from this high level during the following 2 days. Synthesis was then maintained at a constant level for the remainder of the culture period. After culture for 2 and 7 days, the proteoglycans synthesized by the explants were identical to the preexisting proteoglycans in hydrodynamic size, glycosaminoglycan chain size, and ability to form aggregates. These findings suggest that the embryonic chick sterna maintained a stable cartilage phenotype during the extended culture periods. The initial rapid rate of matrix turnover was probably attributable to an adaptation of the tissue to ex ovo culture conditions and the subsequent maintenance of cellular activities at a lower level indicated the establishment of a steady-state rate of metabolism.     

Sulfate incorporation from ascorbate 2-sulfate into chondroitin sulfate by embryonic chick cartilage epiphyses.

Radioactivity was significantly incorporated from ascorbate 2-[35S]sulfate into chondroitin sulfate by embryonic chick cartilage epiphyses. The extent of incorporation was comparable with that from inorganic [35S]sulfate. The radioactive chondroitin sulfate formed from ascorbate 2-[35S]sulfate gave two radioactive disaccharides on chondroitinase-ABC [EC 4.2.2.4] digestion. The incorporation was markedly decreased by inorganic sulfate. The time course of incorporation from ascorbate 2-[35S]sulfate and inorganic [35S]sulfate into chondroitin sulfate and the constituent disaccharides suggest that the incorporation rates from the two radioactive substances are different.     

Glycoproteins from experimental granulation tissue and their effects on collagen synthesis in embryonic chick tendon cells

Buffer-soluble and pronase-liberated glycoproteins from experimental granulation tissue were fractionated by gel filtration and DEAE-cellulose chromatography. The age of the granuloma was reflected in the gel filtration pattern. Two glycoproteins were isolated, purified to homogeneity and analyzed for their carbohydrate and amino acid compositions.The collagen synthesis in embryonic chick tendon cells was measured in the presence of these fractions, which were found to be inhibiting even at 10^?6 M. These glycoproteins may be significant in the feedback regulation of the development of granulation tissue and fibrosis.     

In vitro stimulation of cartilage in embryonic chick neural crest cells by products of rentinal pigmented epithelium.

The retinal pigmented epithelium of the chick embryo influences head neural crest mesenchymal cells to form the scleral cartilage of the eye. The possible role of extracellular matrix in this interaction was studied. Extracellular matrix was deposited on Millipore filters in vitro by pigmented epithelial cells which were then killed by distilled water lysis. When grown on the Millipore filters which had carried pigmented epithelium, clonal neural crest and periocular mesenchyme “target” cells formed cartilage in 61 of 155 experiments. Cartilage was not formed when the cells were grown on naked filters nor did gels of purified Type I and Type II collagen promote chondrogenesis. It is concluded that extracellular matrix deposited by the pigmented epithelium in vitro is a potent stimulus for the induction of chondrogenesis in competent mesenchyme, and that living pigmented epithelial cells need not be present for such induction.