Variation in Prolyl Hydroxylase Activity of Keloid-Derived and Normal Human Fibroblasts in Response to Hydrocortisone and Ascorbic Acid

Collagen Rel. Res. Vol. 3/1983, pp. 1-12

Assembly of Procollagen mRNA Translation Products into Pepsin-Resistant Structures JANET M. MONSON Department of Biochemistry and Biophysics, University of California, San Francisco, CA 94143, USA.

Abstract Proteolytic digestion was used to probe the conformation of the preproa chains synthesized by a mRNA-dependent reticulocyte lysate from chicken calvarial RNA. Pepsin-resistant a1- and a2-like chains were recovered even from translation reactions that were not preincubated below the reported Tm of the unhydroxylated tripie helix. The pepsin-resistant structures were stable to thermal denaturation at 45 oe and a fraction remained resistant to peptic digestion at 30 oe. Interchain disulfide bonds did not appear to be required for the formation or thermal stability of these structures. Pepsin resistance is normally interpreted as evidence for a triple-helical conformation. Therefore, these results suggest that the in vitra synthesized preproa chains contain the requisite information to associate in register for correct helix folding. The unusual thermal stability of these structures is not understood, but this may indicate assembly into higher orders of structure. Key words: pepsin-resistance, triple-helix folding, registration peptides, preproa chains.

Introduction The mechanism for the assembly of collagen from its constituent polypeptide chains has intrigued and eluded biochemists. Although a chains can be renatured in vitra to form the collagen tri pIe helix, the rate is too slow to be physiologically meaningful (Kuhn et al., 1964; Beier and Engel, 1966; Harrington and Rao, 1970). Ir is generally assumed that the rate-limiting step is the association of the three polypeptides in proper register for correct helix formation. Speakman (1971) postulated that precursor-specific sequences located at either end of each a chain might specifically associate thereby properly registering the chains for helix folding. While the presence of precursor polypeptides at both ends of individual a chains has been weil documented (see Reviews: Fessler and Fessler, 1978; Prockop et al., 1979; Bornstein and Traub, 1979), data supporting their function as "registration peptides" has been lacking. When the pro collagen tripie helix is 1

Collagen 3/1

2

Janet M. Monsor.

denatured without chain dissociation, correct renaturation proceeds in vitro with apparent first order kinetics as would be expected for a unimolecular re action (Fessler et al., 1974; Bornstein, 1974; Veis et al., 1973; Byers et al., 1974; Bachinger et al., 1980; Gerard et al., 1981; Bruckner and Prockop, 1981). However, if the proa chains are completely dissociated by reduction and alkylation, renaturation kinetics are similar to those observed with a chains (Gerard et al., 1981). Two possible explanations are that either the proa chains do not possess the postulated registration peptides or that their function has been destroyed by reduction and alkylation. To further address the question of procollagen assembly and its relationship to structure, attention has turned to an analysis of the structure (Palmiter et al., 1979; Graves et al., 1981) and conformation (Ouellete et al., 1981; Monson, 1982a) of the in vitra synthesized preproa chains. The translation products of type land type 11 pro collagen mRNAs form pepsin-resistant structures, suggesting that they contain the requisite information for proper assembly (Ouellette et al., 1981; Monson, 1982 a). Here are presented some unexpected results regarding the formation and thermal stability of these structures.

Materials and Methods Materials

L-[2,3- 3H]-Proline (43.1 Ci/mmol) was purchased from New England Nuclear. L-[35S]-Methionine (> 1000 Ci/mmol) was purchased from Amersham/Searle. Antiserum against the carboxy-terminal pro peptides of pro collagen (Olsen er al., 1977) was a gift from Dr. B. R. Olsen. Heat-killed, formalin-fixed Staphylococcal aureus was generously supplied by Dr. R. B. Meagher. Micrococcal nUcleasc (17,000 U/mg) and pepsin (2X crystallized) were purchased from Worthington. Urea (Mann, Ultra pure) solutions were deionized immediately before use with a mixed bed ion-exchange resin (Bio-Rad AG 501-X8). Other standard reagents were sodium dodecyl sulfate (Mann, Ultra pure), EGTA 1 (Sigma, analytical grade), EDTA (Mallinckrodt, analytical grade) and acetic acid (Mallinckrodt, analytical grade). Creatine phosphokinase and creatine phosphate were purchased from Sigma and both ATP and GTP were purchased from P-L Biochemicals. Preparation of Total RNA from Chicken Embryo Calvaria

Calvaria (frontal and parietal bones) were surgically removed from day 16.5 Hubbard chicken embryos and placed directly into liquid nitrogen. Total RNA was extracted from the powdered bones by either a SDS-phenol method (Monson and Goodman, 1978) or by a modification (Monson, 1982 b) of the guanidine thiocyanate method (Chirgwin et al., 1979). 1 Abbreviations: EDT A, disodium ethylenediamine tetraacetate; EGT A, ethylene glycol bis (2-aminoethyl ether)-N, N'-tetraacetic acid; mRNA, messenger RNA; Hepes, N-2hydroxyethylpiperazine-N'-2-ethane sulfonic acid; Tris, tris(hydroxymethyl) aminoethane; SDS PAGE, sodium dodecyl sulfate polyacrylamide gel electrophoresis; LRSC, lathyritic rat skin collagen; cDNA, complementary deoxyribonucleic acid.

Assembly of Pro collagen mRNA Translation Products into Pepsin-Resistant Structures 3 Translation Reactions

Preparation of rabbit reticulocyte lysates followed the procedures previously described (Mons on and Goodman, 1978). A mRNA-dependent translation system was genera ted by treating the lysate with micrococcal nuclease according to a modification of the method of Pelham and Jackson (1976). Freshly-thawed lysate (300 ,ul) was digested for 8 min at 18°C with 10 ,ug/ml micrococcal nuclease in the presence of 1 mM CaCI 2, 0.2 mM dithiothreitol. The digestion was terminated by adding EGTA to a final concentration of 2 mM. Aliquots of this treated lysate were used immediately in translations. Translations conditions were similar to those previously reported (Monson and Goodman, 1978). In a standard translation reaction the final concentrations of reagents contributed by the reaction mixture were 50 !tM of each unlabeled amino acid, 20 mM Hepes-KOH, pH 7.6, 80 mM KC1, 2 mM Mg (C 2H 30 2)2, 1 mM Na2ATP, 0.2 mM Na2GTP, 15 mM creatine phosphate, 10 ,ug/ml creatine phosphokinase (83 U/mg) and 0.06 ,uM PhCH 2S0 2F. A typical 625 ,ul translation reaction labeled with 130 ,uCi of [3H]-proline was constituted horn 244 ,ul of reagent mixture, 250 ,ul of treated lysate and 131 ,ul of H 20 containing 120 ,ug of total calvarial RNA. The RNA solution was heat-denatured for 3 min at 90°C and quenched in ice water immediately prior to translation. In a standard, 150 ,ul re action labeled with 111 ,uCi of [35S]-methionine, the isotope solution contributed a final concentration of 2.4 mM K(C 2H 30 2) and 0.012 % 2-mercaptoethanol to the translation reaction. Since the treated lysate constituted 40 % of each translation reaction, all translations were performed in the presence of 0.08 mM dithiothreitol. Translations were performed at 26°C for 90 min and terminated by digesting the reactions with 100 ,u.g/ml pancreatic ribonuclease A and 50 U/ml Tl ribonuclease at 26°C for 15 min in the presence of 10 mM EDT A. Hydroxyproline Analysis

The hydroxyproline content of antibody-precipitated, in vitro-synthesized preproa chains was determined by amino acid analysis. Antiserum against the car-

boxy-terminal propeptides of procollagen was used to precipitate 100 ,ul of a terminated translation reaction labeled with [3H]-proline according to the precipitation procedure detailed previously (Mons on and Goodman, 1978). The preproa chains were eluted from the formalin-fixed Staphylococcus aureus with 200 ,ul 10 M urea, diluted five-fold with H 20 and precipitated with carrier collagen by acetone. The dried precipitate was hydrolyzed in constant-boiling 6N HCl for 22 h at 108°C and radioactive amino acid analysis was performed as described (Siegel, 1977). Slab SDS Polyacrylamide Gel Electrophoresis

The buffer system employed was that of Laemmli (1970) and electrophoresis was performed as described by O'Farrell (1975) using a 3 % stacking and a 6 0/0 separating gel. The slab gels were poured and run in place in a gel apparatus similar to the Hoefer vertical slab apparatus. Sampies were prepared by dissolving pellets or diluting aliquots of translation reactions directly into 5 M urea-SDS sampie buffer (3 Ufo SDS, 0.0625 M Tris-HCI, pH 6.8, 5 M urea, 10 Ufo glycerol and 5 Ufo 2-mercaptoethanol) followed immediately by heating for 5 min in boiling

4

Janet M. Monson

H20. After electrophoresis gels were stained with 0.1 '% eoomassie Blue in 50 Ufo el3eeOOH and destained in 7.5% acetic acid. The dried gels were autoradiographed using No-screen Kodak medical X-ray film for [35S] or for [3H]-prepared for fluorography with preflashed, RP-Royal X-ornat, X-ray film (Bonner and Laskey, 1974; Laskey and Mills, 1975).

Experiment 1: Two sets of three pairs of 25 .ul aliquots of a terminated translation reaction labeled with [3H]-proline were prepared. Two pairs of tub es were preincubated at 4 oe for each of the specified times (0, 30 or 60 min). After preincubation one member of each pair received an equal volume of 1 M acetic containing 3 .ug of lathyritic rat skin collagen (LRSC) and the other member of each pair received an equal volume of the same solution containing in addition 10 .ug pepsin (final concentration 200 .ug/ml). One set of three pairs was digested at 10 oe for 6 h while the other set was digested at 30 oe for 6 h. Digestions were terminated by precipitating each sampIe with 5 volumes of acetone and centrifuging for 2 min in an Eppendorf centrifuge. Each pellet was dissolved in 75 .ul of 5 M urea-SDS sampIe buffer and after heat denaturation at 100 oe; 20 .ul of each sampIe were analyzed by SDS-PAGE. SampIes prepared for electrophoresis in this manner completely penetrated the 3 % stacking gel and entered the 6 % separating gel. All the steps in this and the following experiments were performed in 1.5 ml Eppendorf tubes. Experiment 2: A set of three pairs of 25 .ul aliquots of a terminated translation reaction labeled with [3H]-proline was heat denatured at 45 oe for 30 min while another identical set of sampIes was simply maintained at 26 oe for 30 min. Then a pair from each set was preincubated at 4 oe for each of the specified times (0, 30 or 60 min). After preincubation, one member of each pair received an equal volume of cold 1 M acetic containing 3 fJg LRSe while the other member received the same solution containing in addition 40 fJg pepsin (final concentration 800 fJg/ ml). All 12 sampIes were incubated at 4 oe for 16 h, precipitated with acetone and analyzed by SDS-PAGE as described for Experiment 1. To check for the presence of interchain disulfide bonds, 10 fJl aliquots of the terminated translation reaction were dissolved in 40 fJl of 5 M urea-SDS sampie buffer either with or without 5 % 2-mercaptoethanol. After he at denaturation at 100 oe, 25 ,al of each sampIe was analyzed by SDS-PAGE. Experiment 3: A terminated translation reaction labeled with [35S]-methionine was divided into two pairs of 25 fJl aliquots. One pair of aliquots each received an equal volume of 1 M acetic acid at 10 oe while the other pair received an equal volume of the same solution containing in addition 3 fJg LRSe carrier. One member of each pair also received 40 fJg pepsin (final concentration 800 fJg/ml). All aliquots were incubated at 10 oe for 6 h. Prior to precipitation with acetone, 3 fJg LRSe were added to the two tubes lacking this carrier. The precipitated sampIes were dissolved in 5 M urea-SDS sampIe buffer and analyzed by SDSPAGE as described for Experiments 1 and 2. Results The tripie helix of procollagen is resistant to digestion by pepsin under appropriate conditions (Rubin et al., 1963; RoseDbloom et al., 1973; Bruckner and

Assembly of Procollagen mRNA Translation Products into Pepsin-Resistant Structures 5

Prockop, 1981). This property has been used as a very sensitive assay for triplehelical structures since one can observe by SDS-PAGE the disappearance of proa chains after the globular ends of the molecule are digested by pepsin and the appearance of resistant a1 and a2 chains (Jimenez et al., 1973). Since the stability of the collagen tri pie helix is directly proportional to the extent of hydroxylation of the prolyl residues in the constituent polypeptides (Rosenbloom et al., 1973; Berg and Prockop, 1973), the in vitra synthesized proa chains were tested for their degree of hydroxylation. A procollagen-antibody precipitate of a [3H]-proline-labeled translation reaction was hydrolyzed and subjected to amino acid analysis. Of the 28,000 cpm analyzed, 21,000 cpm eluted with authentic proline while the remaining counts eluted with the base wash. No counts were detected in the hydroxyproline peak. The preproa chains synthesized by the mRNA-dependent reticulocyte lysate were tested for their ability to assume pepsin-resistant conformations. Since there chains were unhydroxylated, tripie helices were not expected to form unless the temperature of the translation reaction was decreased below the Tm (24°C) for the unhydroxylated tri pie helix. Accordingly, sampies were preincubated at 4 oe fot 0, 30, or 60 min prior to digestion by pepsin at 10 oe as described for Experiment 1. Pepsin-resistant, [3H]-proline bands migrating slightly ahead of stained, internal-standard a1 and a2 chains were obtained not only for the sampies preincubated at 4 oe (Fig. 1, E, F) but also for the sampie that was digested im-

Fig. 1: A flllorogram of [3HJ-proline-Iabeled translation prodllcts analyzed by sDs-PAGE is depicted. Sam pies were treated as described for Experiment 1 (Materials and Methods). Each set of triplets, i. e. A, B, C, were preincubated at 4 °C for 0, 30 or 60 min, respectively. Controls were incllbated withollt pepsin for 6 h at either 10 °C (A,B,C) or 30 °C (G,H,I). Sampies were digested with pepsin for 6 h at 10 °C (D,E,F) or 30°C (J,K,L). Arrows labeled 1 and 2 indicate the migration of stained internal-standard a1 and a2 chains. The prominent high molecular weight bands in the controls have been previollsly identified as preproa1 and preproa2 chains (Monson and Goodman, 1978; Graves et al., 1981).

6

Janet M. Monson

mediately (Fig. 1 D). The migration of these pepsin-resistant bands is consistent with their identification as a1- and a2-1ike chains since unhydroxylated a chains migrate slightly faster than their respective hydroxylated farms in the Laemmli buffer system. Evidence for the activity of the pepsin was provided by the fact that the stained bands contributed by the lysate proteins disappeared and the major remaining stained bands corresponded to the carrier a1 and a2 chains and pepsin itself. By comparison, incubation in 0.5 M acetic acid alone did not degrade the in vitra synthesized proa chains (Fig. 1, A-C) and the stained gel patterns of these sam pies were not significantly different from those of untreated sampIes.

,_

2_

I{

B' C' D' E' F' ~-- -...-

Fig. 2: A fluorogram of [3Hl-proline-labeled translation products analyzed by SDS-PAGE is depicted. Sampies were treated as described for Experiment 2 (Materials and Methods). Each set of triplets, i. e. A,B,e, was preincubated at 4 oe for 0, 30 or 60 min, respectively. Sam pies A-F were maintained at 26 oe for 30 min prior to preincubation. Sam pies were treated without (A,B,C) or with (D,E,F) pepsin at 4 oe for 16 h. Sampies A'-F' represent analogous sampies that were heat denatured at 45 oe for 30 min prior to preincubation. Aliquots of the terminated translation reaction that were mixed directly with the sam pie buffer either with (G') or without (H') 2-mercaptoethanol are also shown. Arrows 1 and 2 indicate the migration of stained internal standard al and a 2 chains.

Assembly of Procollagen mRNA Translation Products into Pepsin-Resistant Structures 7 Carrying out an identical experiment except performing the pepsin digestion above the Tm was expected to result in the compiete loss of resistant a chains. However, a fraction of the material was resistant to pepsin even at 30 °C (Fig. 1, J-L). Contral sampIes incubated at 30 °C without pepsin (Fig. 1, G-I) were indistinguishable from controls incubated at 10 °C (Fig. 1, A-C). The thermal stability of the pepsin-resistant structures was tested by he at denaturing sampies at 45 °C for 30 min while controls were maintained at 26 oe. Preincubations at 4 °C and pepsin digestions were performed as described for Experiment 2. The sampies that were heat-treated yielded pepsin-resistant chains regardless of the preincubation time (Fig. 2, D'-F') and the results were indistinguishable from sampies that were not denatured (Fig. 2, D-F). Aliquots of the same translation reaction were directly mixed with sampie buffer either with or without 2mercaptoethanol, denatured and analyzed by SDS-PAGE. Although the migration of the reduced preproa chains (Fig. 2 G ' ) differs slightly from the unreduced chains (Fig. 2, H'), there is no evidence for higher molecular weight species indicative of interchain disulfide bond formation. Furthermore, since the translation reaction was performed under reducing conditions (0.08 mM dithiothreitol) and the pepsin digestions were also conducted under reducing conditions (0.04 mM dithiothreitol) at acidic pH which should minimize disulfide bond formation, it appears unlikely that interchain disulfide

Fig. 3: An autoradiogram of [35S] -methionine-labeled translation products analyzed by SDS-PAGE is depicted. Sampies were treated as described for Experiment 3 (Materials and Methods). Sampies treated without (A and B) oe with (C and D) pepsin at 10 °C for 6 h. Sampies A and C contatined 3 ftg collagen carrier during this treatment, while sampies Band D did not. Arrows indicate the migration of stained internal standard al and a2 chains.

8

Janet M. Monson

bonds are required for the formation or apparent thermal stability of the pepsinresistant structures. In Experiment 3 the translation reaction was carried out under even stronger reducing conditions (0.010f0 2-mercaptoethanol and 0.08 mM dithiothreitol) and the effect of carrier collagen during the digestion was checked. As depicted in Figure 3, [35SJ-methionine-Iabeled al- and a2-like chains were obtained under these conditions whether carrier collagen was present (Fig. 3, C) or absent (Fig. 3, D) during the digestion at 10 oe. This result also suggests that disulfide bonds are not required for the formation of the pepsin-resistant structures. Furthermore, the resistance to proteolysis is not conferred by the presence of carrier collagen during the digestion. Discussion In the present study, calvarial RNA was translated in a mRNA-dependent reticulocyte lysate and the conformation of the in vitra synthesized preproa chains was probed direct1y by peptic digestion. Under these conditions the preproa chains appeared to associate within the milieu of the translation re action to form pepsinresistant structures that were stabilized against complete thermal denaturation. Interchain disulfide bond formation did not appear to be required for this association or the observed thermal stability. Further support for this interpretations is provided by a previous observation by others that preproa chains synthesized in vitra under reducing conditions form pepsin-resistant structures (Ouellette et al., 1981). Pepsin-resistance is normally interpreted as evidence for triple-helical structures provided that prolonged digestions are performed at high enzyme to substrate ratios (Rubin et al., 1963; Rosenbloom et al., 1973; Jimenez et al., 1973; Bruckner and Prockop, 1981). The disappearance of the preproa chains and the appearance of al- and a2-like chains after pepsin digestions is consistent with the interpretation that the a-chain domains were protected from proteolysis whereas the nontriple-helical domains located at the amino and carboxy termini of the preproa chains were not. Although digestion with a mixture of trypsin and chymotrypsin appears to be an even more stringent criterion for a perfect1y folded tripie helix (Bruckner and Prockop, 1981), it seems unlikely that the unhydroxylated helix examined he re could withstand such adernanding test. Therefore, pepsin was selected as an enzymatic probe of the overall conformation with the knowledge that minor imperfections might not be detected. The preproa chains observed in these translations have been previously identified by specific antibody precipitability (Monson and Goodman, 1978) and chemical analysis (Graves et al., 1981). Because procollagen is a secretory protein, it was expected that the proa chains would be initiated with a signal sequence (Blobel and Dobberstein, 1975). The presence of such hydrophobic residues has been confirmed by sequencing in vitra the synthesized preproa chains (Palmiter et al., 1979; Graves et al., 1981) and by DNA sequencing of a proa2(I) gene (Vogeli et al., 1981). A comparison of the collagenase-resistant peptides from the preproal and the proa1 chains has indicated an additional 5,500- to 1O,000-daltons located at the amino terminus of the preproa chain (Palmiter et al., 1979; Graves et al., 1981). Since this is considerably greater than what might be expected for a typical

Assembly of Procollagen mRNA Translation Products into Pepsin-Resistant Structures 9 signal peptide of 15-30 residues (Davis and Tai, 1980), it has been proposed that only a portion functions as a signal sequence while the remaining portion serves some other function (Palmiter et al., 1979; SandeIl and Veis, 1980; Graves et al., 1981). Based on these observations, it is tempting to speculate that there is a region lying between the signal peptide proper and the beginning of the proa chain that might function as a registration peptide. However, the existence of such a region has not yet been confirmed by examining cDNA clones for the corresponding region of the mRNA. Although it is attractive to think of chain association initiating at the amino-termini of preproa chains and there is some data consistent with this view (Veis and Brownwell, 1977 Kirk et al., 1982), the data presented here and elsewhere (Ouellette et al., 1981) simply suggest that the preproa chains in toto contain the requisite information for association into pepsinresistant structures. Several laboratories have reported that in vivo there is a correlation between the appearance of the tripie helix and disulfide bonds linking the carboxy-terminal propeptides of procollagen (Harwood et al. , 1973; Uitto and Prockop, 1974; Schofield et al., 1974; Lukens, 1976; Harwood et al., 1977). In the case of underhydroxylated type III procollagen only the carboxy-terminal propeptides become disulfide-bonded in vivo, but the amino-terminal propeptides do associate sufficicently to genera te antigenic determinants (Fessler et al., 1981). In vivo studies of truncated proa chains largely lacking the carboxy-terminal propeptides have shown that these molecules are not tri pie helical and are not secreted indicating a significant role for the carboxy-terminus in the assembly and secretion process (Rosenbloom et al., 1976). However, as many of the authors of these studies are careful to point out, these observations do not preclude an initial association and chain registration at the amino-termini. The experiments presented here do not distinguish between several possible tri pie helical structures. However, the effective concentration of the nascent chains on the polysomes would favor chain association rather than a concentrationindependent tripie folding of individual chains back upon themselves as proposed for a chains at dilute concentrations (Harrington and Rao, 1970). Perhaps more likely is the formation of tripIe helices from either three preproa1 or three preproa2 chains rather than the conventional composition of two preproa1 chains and one preproa2 chain. Antibodies specific for either chain could distinguish these latter possibilities. The observation that the pepsin-resistant structures apparently form and are stabilized above the Tm (24 °C) reported (Rosenbloom et al., 1973; Fessler and Fessler, 1974) for the unhydroxylated triple-helix of chicken type I procollagen was unexpected and is poody understood. It is possible that the specific chemical composition of the translation reaction alters the Tm of the helix. Another possibility is that the pepsin-resistant structures assemble amongst themselves or with other macromolecules in such a way as to confer thermal stability to the unhydroxylated helix. Consistent with this idea is the fact that pro collagen segmentlong-spacing (SLS) crystallites have been observed within intracellular vacuoles, in medium harvested from cultured cells, in homogenates of several collagen producing tissues, and in partially purified preparations of procollagen (Bruns et al., 1979). Furthermore, it has been suggested that these SLS crystallites are stabilized by some cementing substance (Bruns et al., 1979). The participation of other calvarial translation products in the stabilization of the pepsin-resistant structures

10

Janet M. Monson

is open to investigation by comparing results obtained by translating total calvarial RNA and purifled procollagen mRNAs.

Acknowledgements ].M.M. expresses her gratitude to R. C. Siegel for numerous discussions and performance of the amino acid analysis and to B. R. Olsen for supplying the antiserum against the carboxy-terminal propeptides of type I procollagen. She also wishes to thank A. Veis and G. R. Martin for communicating unpublished results, and ]. Gampell for assistance in preparation of the manuscript.

References Bachinger, H. P., Bruckner, P., TimpI, R., Prockop, D. J., and Engel, J.: Folding mechanism of the tripIe helix in type-III collagen and type-III pN-collagen. Eur. J. Biochem. 106: 619-632, 1980. Beier, G. and Engel, J.: The renaturation of soluble collagen: Products formed at different temperatures. Biochemistry 5: 2744-2755,1966. Berg, R. A. and Prockop, D. J.: The thermal transition of a non-hydroxylated form of collagen: Evidence for a role for hydroxyproline in stabilizing the tripie helix of collagen. Biochem. Biophys. Res. Comm. 52: 115-120, 1973. Blobel, G. and Dobberstein, B.: Transfer of proteins across membranes. I. Presence of proteolytically processed and unprocessed nascent immunoglobulin light chains on membrane-bound ribosomes of murine myeloma. J. Cell Biol. 67: 835-851, 1975. Bonner, W. M. and Laskey, R. A.: A film detection method for tritium-Iabeled proteins and nucleic acids in polyacrylamide gels. Eur. J. Biochem. 46: 83-88, 1974. Bornstein, P.: The structure and assembly of procollagen: A review. }. Supramol. Str. 2: 108-120,1974. Bornstein, P. and Traub, W.: The chemistry and biology of collagen in The Pro teins, 3rd ed. (Neurath, H. and Hill, R. L. eds.) (Academic Press, New York, 1979), Vol. 4, pp. 411-632. Bruckner, P. and Prockop, D. J.: Proteolytic enzymes as probes for the triple-helical conforrnation of procollagen. Anal. Biochem. 110: 360-368, 1981. Bruns, R. R., Hulmes, D. J. S., Therrien , S. F., and Gross, J.: Pro collagen segment-longspacing crystallites: Their role in collagen fibrillo genesis. Proc. Natl. Acad. Sei. USA 76: 313-317, 1979. Byers, P. H ., McKenny, K. H ., Lichtenstein, J. R., and Martin, G. R.: Preparation of type III procoll agen and collagen fro rn rat skin. Biochemistry 13: 5243-5248, 1974. Chirgwin, J. M., Przybyla, A. E., MacDonald, R. ]., and Rutter, W. J.: Isolation of biologically active ribonucleic acid from sources enriched in ribonuclease. Bioehemistry 18: 5294-5299,1979. Davis, B. D. and Tai, P. c.: The mechanism of protein secretion across membranes. Nature 283: 433- 438, 1980. Fessler, L. I. and Fessler, J. H.: Protein assembly of procollagen and effects of hydroxylation. J. Biol. Cem . 249: 7637-7647, 1974. Fessler, J. H. and Fessler, L. 1.: Biosynthesis of procollagen. Ann. Rev. Biochem. 47: 129- 162, 1978. Fessler, L. 1., TimpI, R., and Fessler, J. H.: Assembly and processing of procollagen type III in chick embryo blood vessels. J. Biol. Chem. 256: 2531-2537, 1981. Fessler, L. I., Rudd, c., and Fessler, J. H. : Renaturation of disulfide-linked procollagen. }. Su pramol. Str. 2: 103-107, 1974.

Assembly of Pro collagen mRNA Translation Products into Pepsin-Resistant Structures 11 Gerard, S., Puett, D., and MitchelI, W. M.: Kinetics of collagen fold formation in human type I procollagen and the effect of disulfide bonds. Biochemistry 20: 1857-1865, 1981. Graves, P. N., Olsen, B. R., Fietzek, P. D., Prockop, D. J., and Monson, J. M.: Comparison of the NH 2-terminal sequences of chick type I pro collagen chains synthesized in an mRNA-dependent reticulocyte Iysate. Eur. J. Biochem. 118: 363-369, 1981. Harrington, W. F. and Rao, N. V.: Collagen structure in solution: I. Kinetics of helix regeneration in single-chain gelatins. Biochemistry 9: 3714-3724, 1970. Harwood, R., Grant, M. E., and Jackson, D. S.: The sub-cellular location of interchain disulfide bond formation during pro collagen biosynthesis by embryonic chick tendon cells. Biochem. Biophys. Res. Comm. 55: 1188-1196, 1973. Harwood, R., Merry, A. H., Woolley, D. E., Grant, M. E., and Jackson, D. S.: The disulfide-bonded nature of pro collagen and the role of the extension peptides in the assembly of the molecule. Biochem.}. 161: 405-418,1977. Jimenez, S., Harsch, M., and Rosenbloom, J.: Hydroxyproline stabilizes the tripIe helix of chick tendon collagen. Biochem. Biophys. Res. Comm. 52: 106-114, 1973. Kirk, T. Z., Veis, A., and Krzesichi, R.: Collagen chain association and tripIe helix formation is initiated at the polysome level. J. Biol. Chem., in press. Kuhn, K., Engel, J., Zimmermann, B., and Grassman, W.: Renaturation of soluble collagen. Arch. Biochem. Biophys. 105: 387-403, 1964. Laemmli, U. K.: Cleavage of structural pro teins during the assembly of the head of bacteriophage T4. Nature 227: 680-685, 1970. Laskey, R. A. and Mills, D.: Enhanced autoradiographic detection of 32p and 1251 using intensifying screens and hypersensitized film. FEBS Lett. 82: 314-316, 1977. Lukens, L. N.: Time of occurrence of disulfide linking between pro collagen chains. J. Biol. Chem. 251: 3530-3538,1976. Monson, J. M.: Translation products of procollagen mRNAs. Meth. Enzymol. 82: 222225,1982 a. Monson, J. M.: Preparation and translation of interstitial (calvarial) pro collagen mRNA. Meth. Enzymol. 82: 218-222, 1982 b. Monson, J. M. and Goodman, H. M.: Translation of chick calvarial procollagen messenger RNAs by a messenger RNA dependent reticulocyte lysate. Biochemistry 17: 5122-5128, 1978. O'Farrell, P. H.: High resolution two-dimensional electrophoresis of proteins. J. Bio!. Chem. 250: 4007-4021, 1975. Olsen, B. R., Guzman, N. A., Engel, J., Condit, c., and Aase, S.: Purification and characterization of a peptide from the carboxy-terminal region of chick tendon procollagen type I. Biochemistry 16: 3030-3036, 1977. Ouellette, L. A., Paglia, L. M., and Martin, G. R.: Characterization of the cell-free translation products from types land II procollagen mRNAs. Collagen Rel. Res. 1: 327-335, 1981. Palmiter, R. D., Davidson, J. M., Gagnon, J., Rowe, D. W., and Bornstein, P.: NH 2terminal sequence of the chick proa(l) chain synthesized in the reticulocyte lysate system. J. Biol. Chem. 254: 1433-1436,1979. Pelham, H. R. B. and Jackson, R. J.: An efficient mRNA-dependent translation system from reticulocyte lysates. Eur. }. Biochem. 67: 247-256, 1976. Prockop, D. J., Kiviriko, K. I., Tuderman, L., and Guzman, N. A.: The biosynthesis of collagen and its disorders. New Eng. }. Med. 301: 13-23, 1979. Rosenbloom, J., Harsch, M., and Jimenez, S.: Hydroxyproline content determines the denaturation temperature of chick tendon collagen. Arch. Biochem. Biophys. 158: 478-484, 1973. Rosenbloom, J., Endo, R., and Harsch, M.: Termination of procollagen chain synthesis by puromycin. J. Biol. Chem. 251: 2070-2076, 1976. Rubin, A. L., Pfahl, D., Speakman, P. T., Davison, P. J., and Schmitt, F. 0.: Tropocollagen: Significance of protease-induced alterations. Science 139: 37-39, 1963.

12

Janet M. Monson

SandelI, L. and Veis, A.: The molecular weight of the cell-free translation product of a1 (I) procollagen mRNA. Biochern. Biophys. Res. Cornrn. 92: 554-562, 1980. Schofield, J. D., Uitto, J., and Prockop, D. J.: Formation of interchain disulfide bonds and helical structure during biosynthesis of procollagen by embryonic tendon cells. Biochernistry 13: 1801-1806, 1974. Siegel, R. c.: Collagen cross-linking: Effect of D-penicillamine on cross-linking in vitra. J. Bio!. Chern. 252: 254-259, 1977. Speakman, P. G.: Proposed mechanism for the biological assembly of collagen tripie helix. Nature 229: 241-243, 1971. Uitto, J. and Prockop, D. J.: Biosynthesis of cartilage procollagen: Influence of chain association and hydroxylation of prolyl residues on the folding of the polypeptides into the triple-helical conformation. Biochernistry 13: 4586-4591, 1974. Veis, A., Anesey, J., Yuan, L., and Levy, S. J.: Evidence for an amino-terminal extension in high-molecular-weight collagens from mature bovine skin. Prac. Nat!. Acad. Sei. USA 70: 1464-1467, 1973. Veis, A. and BrownweIl, A. G.: Triple-helix formation on ribosome-bound nascent chains of procollagen: Deuterium-hydrogen exchange studies. Proc. Nat!. Acad. Sei. USA 74: 902-905, 1977. Vogeli, G., Ohkubo, H., Sobel, M. E., Yamada, Y., Pastan, 1., and DeCrombrugghe, B.: Structure of the promoter for chicken a2 type I collagen gene. Proc. Nat!. Acad. Sei. USA 78: 5334-5338, 1981. Dr. Janet M. Monson, Zymos Corporation, 2121 N. 35th Street, Seattle, WA 98103 USA.