Burton Goldberg

Radioimmunoassay for human procollagen

Rabbit antiserums were produced against the procollagen molecule secreted into the medium of cultured human skin fibroblasts. The isolated antigenic, amino terminal portion of the procollagen molecule was purified, labeled with iodine-125, and used in a radioimmunoassay which detected nanogram quantities of the same antigen. With the assay, immunologically identical molecules were detected in the culture mediumn of different strains of human fibroblasts and in normal human serums. Serumns from human cord blood contained a 12-fold higher concentration of the antigen than serums from adults, while serums other vertebrates gave reactions to incomplete cross-reactivity or non-reactivity.

The role of the superoxide anion as a toxic species in the erythrocyte.

Abstract The toxic action of the superoxide anion (O2−) toward the erythrocyte was investigated with O2− generated through the autooxidation of dihydroxyfumaric acid (DHF). A suspension of human red cells exposed to DHF undergoes a rapid breakdown of the cellular hemoglobin to methemoglobin and other green pigments. This hemoglobin breakdown is inhibited by superoxide dismutase (SOD) or catalase (CAT) and is accelerated by lactoperoxidase (LP) added externally to the red cell medium. Associated with the hemoglobin breakdown is a hypotonic hemolysis also inhibited by SOD or CAT and initially accelerated but later inhibited by LP. Conversion of the red cell hemoglobin to carbonmonoxyhemoglobin in an aerated medium results in no hemoglobin breakdown or hypotonic lysis in the presence of DHF, even though O2− can be demonstrated in the medium. Although no evidence for membrane sulfhydryl oxidation or lipid peroxidation can be demonstrated in red cells exposed to DHF, the membranes of these cells were found to retain a green pigment. The presence of this green pigment in red cell membranes was inhibited by SOD, CAT, or conversion of the cellular hemoglobin to carbonmonoxyhemoglobin, but was not inhibited by LP. These results have been interpreted as a peroxide-dependent formation of O2− by DHF, followed by attack of O2− on hemoglobin. The reaction of O2− with hemoglobin leads to the formation of a hemoglobin-breakdown product that binds to the red cell membrane, resulting in an increased osmotic fragility of the cell.

Binding of soluble type I collagen molecules to the fibroblast plasma membrane

Soluble 125I-labeled type I collagen binds to cultured fibroblasts but not to cultured epithelia. The binding of the ligand to fibroblasts is reversible, saturable and highly specific for sequences contained within the helical portions of the alpha1 and alpha2 chains. The amount of ligand bound is dependent upon cell number and ligand concentration. Binding is decreased but measurable at 4 degrees C. The steady state binding is greater at 26 degrees than at 37 degrees C due to a more rapid dissociation of the ligand-acceptor complex at 37 degrees C. The half-life of the complex is 46 min at 37 degrees C and approximately 2.5 hr at 26 degrees C. Scatchard plots of binding data indicate a single class of high affinity binding sites (KD = 1.2 X 10(-11) M) with each fibroblast binding approximately 500,000 molecules at saturation. Pretreatment of fibroblasts with bacterial collagenase, chondroitinase ABC or testicular hyaluronidase does not affect the binding reaction, whereas pretreatment of the cells with phospholipase C increases the amount of ligand bound. Ligand binding is decreased but not abolished after fibroblasts are treated with trypsin concentrations which remove surface fibronectin. Fibroblast monolayers treated with antiserum against fibronectin bind the radiolabeled ligand normally. In contrast to collagen, addition of excess fibronectin does not accelerate the dissociation of bound ligand from fibroblasts. Possible functions for surface-bound collagen are discussed.

Collagen synthesis on polyribosomes of cultured mammalian fibroblasts

Abstract Ribosomes attached to the endoplasmic reticulum have been separated from those unattached to cytoplasmic membranes of cultured, stationary mammalian fibroblasts. In linear sucrose gradients, approximately one-half of the unattached ribosomes were recovered as 80 s particles, the rest as polyribosomes. The ribosomes freed by deoxycholate from the endoplasmic reticulum were recovered predominantly as polyribosomes. Free and attached polyribosomes had about the same size distributions in the gradients, were equally active in protein synthesis, and nascent collagen and non-collagen proteins were synthesized over all size distributions of polyribosomes. Polyribosomes sedimenting at 180 s were most abundant, and fractions sedimenting at 210 to 220 s were most active in collagen synthesis. There was no evidence in this system for appreciable synthesis of collagen or non-collagen protein on polyribosomes sedimenting faster than 330 s. Hydroxylation of prolyl residues in nascent collagen occurred while the peptide chains were attached to polyribosomes of the endoplasmic reticulum. Pulse-chase experiments indicated that as these nascent collagen chains neared completion, as judged by their hydroxyproline content, they moved with the fastest sedimenting polyribosomes. Release of completed collagen molecules presumably occurred from such units. Nascent collagen appeared to help stabilize polyribosomes of the endoplasmic reticulum.

Differentiated Cell Types and the Regulation of Collagen Synthesis

COLLAGEN synthesis, the principal differentiated function of the fibroblast, is carried out by homogeneous cell populations cultured under defined conditions1–5. Because hydroxy-proline is formed from proline in collagen, the rate of collagen synthesis can be determined with great sensitivity by measuring the incorporation of radioactive amino-acids3,6. This communication shows that different types of cell vary over a range of at least 104 in the rate at which they synthesize this protein. In cells of fibroblastic origin the property is, of course, most highly developed. In some non-fibroblastic cell types synthesis occurs at about 2–3 per cent of the rate characteristic of the fibroblast, while in others it is completely suppressed. The data indicate three possible levels at which the synthesis may be maintained in different cell types.

THE ACTION OF ANTIBODY AND COMPLEMENT ON MAMMALIAN CELLS

The studies discussed in this paper are concerned principally with the action of heterologous antibodies on mammalian cells. Cell-antibody interactions are most conveniently studied in such systems, for it is a simple matter to obtain powerful heterologous antisera. However, the action of homologous antibodies now has been studied 1-4 (see also, C . A. Stetson and E. Jensen, this monograph) sufficiently to make it clear that they produce the same kind of changes in cells as do heterologous antibodies. There is therefore no reason to doubt that the mechanism of action on cells is substantially the same for the two kinds of antibodies. The effects of antibodies will be described as they are observed in vitro under conditions in which it is possible to distinguish between the effects of antibody alone and those requiring the presence of complement. How these effects may be modified in vivo when the cells are in the form of a solid tissue transplant will not be considered in this paper. Most studies of the in vitro effects of antibodies on animdl cells have utilized erythrocytes because they represent a homogeneous population of cells uniformly exposed to the action of substances in the medium. Methods of obtaining tissue cells in free suspension or in monolayers have provided similar advantages in the use of structurally more complex cells than the anucleate erythrocyte. Although the events occurring in such cells when they are exposed to antibody and complement differ in detail from those in the erythrocyte, the mechanism of immune lysis in the two cases appears to be the same.

Procollagen peptidase: Its mode of action on the native substrate

Procollagen peptidase was recovered from the medium of human and mouse fibroblast cultures by precipitation with ammonium sulfate. The test substrate for the in vitro enzymatic reaction was radioactively-labeled, disulfide-linked procollagen prepared from the medium of human fibroblast cultures. The enzymatic digests were analyzed by electrophoresis in polyacrylamide gets containing sodium dodecyl sulfate and urea. The human and mouse enzymes reacted with the substrate to generate the same intermediates and final products. Procollagen peptidase acts as an endopeptidase which cleaves each of the three procollagen chains in turn. The final products of the reaction are collagen and a three-chain, disulfide-linked fragment derived from the nonhelical aminoterminal residues of procollagen.

Radioimmunoassay of procollagen in serum of patients with Paget's disease of bone.

Summary A new radioimmunoassay for human procollagen showed that the sera of 46 of 50 untreated patients with Pagets disease of bone contained increased concentrations of procollagen protein as compared to normal adults. After therapy with disodium etidronate, all the elevated serum procollagen concentrations decreased significantly, falling to normal levels in 33 of 40 patients. The procollagen levels before and after treatment were coordinate with the values for urinary total hydroxyproline and serum alkaline phosphatase activity. The data show that the radioimmunoassay for procollagen is a dependable and useful adjunct to the study of Pagets disease of bone. The authors thank Dr. Robert E. Canfield of Columbia Presbyterian Medical Center and Dr. Walther Bohne of The Hospital for Special Surgery, New York, N.Y., for making available sera and clinical chemistries from 36 of their patients. Their studies were supported in part by NIH Grant No. AMO 9579. This work was supported by NIH Grant No. 1 R01 HL 17551-01, and by funds from Hoffmann-La Roche Inc. M. B. Taubman is a recipient of NIH Medical Scientist Fellowship No. 5-T05-GM 01668-11. The authors acknowledge the expert technical assistance of Sheila Heitner and Maryann Kunigonis.

Relationship of type I procollagen to corticosteroid therapy in children with inflammatory bowel disease

We determined the serum concentration of the C-terminal propeptide of type I procollagen (pColl-I-C) in 60 children and adolescents (ages 4 to 17 years) with inflammatory bowel disease (24 ulcerative colitis, 36 Crohn disease) and in seven children (ages 2 to 15 years) with nongastrointestinal disease (asthma) during varying regimens of corticosteroid therapy. Patients with inflammatory bowel disease were grouped according to disease severity (mild, and moderate to severe). Significantly lower pColl-I-C concentrations and growth velocities were found in each severity group among those subjects receiving daily corticosteroid therapy compared with those receiving alternate-day or no corticosteroid therapy (P less than 0.01). When daily corticosteroid therapy was initiated and then maintained for 7 to 14 days in 11 patients with exacerbation of inflammatory bowel disease clinical improvement resulted, but mean procollagen concentrations decreased significantly (P less than 0.001). In seven children with asthma receiving methylprednisolone intravenously, significant decreases in pColl-I-C concentrations were noted within 24 to 48 hours of therapy (P less than 0.001). These data indicate that serum procollagen values decrease during both short- and long-term daily administration of corticosteroid therapy. Longitudinal assessment of procollagen concentrations may provide rapid assessment of the effects of different corticosteroid regimens on growth.

Evidence for a Protein that Enhances the Activity of Type I Procollagen C-Proteinase

Gel-filtration separated type I procollagen C-proteinase from a glycoprotein that enhanced the enzyme activity by approximately 4-fold. The enhancer was purified by affinity chromatography on a column of Sepharose coupled to the carboxyl propeptide of type I procollagen. Sodium-dodecyl-sulfate- polyacrylamide gel electrophoresis of the affinity-purified enhancer revealed two active major protein bands with molecular weights of 36 and 34 kdal. Both proteins were glycosylated, as shown by binding to concanavalin-A. The enhancer is extremely heat stable (100 degrees C, 15 min) but its activity is totally abolished by treatment with trypsin or bacterial elastase. The enhancer does not alter the digestion intermediates or final products of the enzymatic reaction but it changes the kinetic properties of the reaction, increasing the apparent Km and Vmax values 16- and 20-fold, respectively. It is suggested that the enhancer might play a regulatory role in procollagen processing.