Studies on the structure and metabolism of elastin and collagen

The structure and metabolism of elastin and collagen were studied in three sets of experiments. In the first experiment, porcine tropoelastin was digested with the endopeptidase, thermolysin. The digest contained at least 35 groups of peptides. The polydispersed peptides included large nonpolar peptides, 50-60 residues in length, and a series of gradations to polar peptides of 10 residues and less. The largest peptide sequenced was a portion of the tryptic peptide T1, which contained the pentapeptide repeat P G V G V. This region was enzyme resistant, implying that the peptide was in some sort of conformation unavailable to the thermolysin. Very likely, it is similar to the B-spiral which is composed of the pentapeptide repeat proposed by Urry. Two small polar peptides were also sequenced. One was shown to extend 2 residues amino terminal to peptide T7a. The other was previously unknown, its sequence being: I G G K P P K P G. Because of the two lysines and the rigid proline structure, it may be involved in some sort of crosslinking function. Overall, thermolysin appeared to cleave the hydrophilic crosslink regions of tropoelastin more extensively than the large hydrophobic regions. In the second experiment, elastin and collagen metabolism were analyzed from the upper thorax to the abdomen in the young pig aorta, in vitro. Tissue was incubated with either ('14)C leucine, ('14)C valine, ('3)H valine, or ('3)H proline for 1, 2, or 24 hours. The tissue was extracted with detergent, urea, and autoclaving. Total incorporated radioactivity was higher in the upper thorax than in the abdomen during the 1 or 2 hour incubations. Most of the radioactivity was seen in valine-rich soluble proteins. However, after a 24 hour incubation the abdomen was the most radioactive region. The label was primarily in the autoclaved supernate, and it reflected collagen formation as shown by ('3)H hydroxyproline production. From this, it was concluded that collagen formation did not appear to be altered significantly in vitro, while elastin formation was markedly altered. Thus, quantitative questions about elastin metabolism must be answered through experiments conducted in vivo. The third experiment involved analyzing elastin and collagen metabolism from the upper thorax to the abdomen of 4-6 week and 2 day old rat aortas, in vivo. Rats were injected intraperitoneally with ('3)H glycine and then sacrificed at specified times after injection. Soluble and insoluble elastin and collagen were isolated. In the 2 day old rat little radioactivity in tropoelastin was isolated. The elastin showed an increase in the accumulation of radioactivity in the upper thorax as compared to the abdomen. Synthesis of radioactive soluble collagen was higher in the abdomen, but this was not reflected by increased accumulation of radioactivity into insoluble collagen. Finally, radioactivity in both elastin and collagen appeared to accumulate at the same relative rates. In the 6 week old rat, again little radioactivity in tropoelastin was isolated. This indicated that tropoelastin was rapidly crosslinked or otherwise utilized. Accumulation of radioactivity in elastin of the upper thorax was greater than in the abdomen. Synthesis of radioactive collagen was increased in the abdomen along with increased accumulation of radioactivity into insoluble collagen. Finally, accumulation of radioactivity in collagen appeared to proceed slower than in elastin

STUDIES ON THE STRUCTURE AND METABOLISM OF ELAS!IN AND COLLAGEN by John Gr~gory Leslie A dissertation submitted to the faculty of The University of Utah in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Experimental Pathology Department of Pathology The University of Utah December 1980 Copyright © John G. Leslie 1981 All Rights Reserved THE UNIVERSITY OF UTAH GRADUATE SCHOOL SUPERVISORY COMMITTEE APPROVAL of a dissertation submitted by John I Gregory have read / is dissertation and have doctoral d gpee, ! � it to be V / Date ( Leslie Chairman, Supervisory Committee I have read this dissertation and have found it to be of satisfactory quality for a doctoral degree. f/11 ltV I Date I Elvera Ehr eld Member, Superviso y Committee I have read this dissertation and have found it to be of satisfactory quality for a doctoral degree. "1/ I ii Date 0:)1) William R. G Member. Supervisory Committee I have read this dissertation and have found it to be of satisfactory quality for a z doctoral deg (0/1-( Date I t e. 'fa L;:Y�� Member. Supervisory Committee I have read this dissertation and have found it to be of satisfactory quality for a doctoral degree. Date 9/1( /80 Alfred Linker Member, Supervisory Committee THE UNIVERSITY OF UTAH GRADUATE SCHOOL FINAL READING APPROVAL To the Graduate Council of The University of Utah: I have read the dissertation of John Gregory Leslie in its final form and have found that (1) its format, citations, and bibliographic style are consistent and acceptable; (2) its illustrative materials including figures, tables, and charts are in place: and (3) the final manuscript is satisfactory to the Supervisory Committee and is ready for submission to the Graduate School. Date Sandberg Member. Supervisory Committee Approved for the Major Department Reginald G. ason Chairman! Dean Approved for the Graduate Council �;z: a��Z;:;; 7 James L. Clayton Dean of The Graduate School ABSTRACT The structure and metabolism of elastin and collagen were studied in three sets of experiments. The first set of experiments involved digesting porcine tropoelastin with an endopeptidase, thermolysin, which cleaves peptide bonds on the amino terminal end of hydrophobic amino acids. The resultant digest contained at least 35 distinct groups of peptides. The polydis- persed peptides included large nonpolar peptides, 50-60 residues in length, and a series of gradations to polar peptides of 10 residues and less. The largest peptide was sequenced and turned out to be a portion of the tryptic peptide Tl, which contained the pentapeptide repeat P G V G V. It was significant that this region was en- zyme resistant, implying that the peptide was in some sort of conformation unavailable to the thermolysin. Very likely, it is similar to the a-spiral which is composed of the pentapeptide repeat proposed by Urry. small polar peptides were also sequenced. Two One was shown to extend 2 residues amino terminal to peptide T7a. other was previously unknown, K P P K P G. its sequence being: The I G G Because of the two lysines and the rigid proline structure, it may be involved in some sort of crosslinking function. Overall, thermo1ysin appeared to cleave the hydrophilic crosslink regions of tropoe1astin more extensively than the large hydrophobic regions. In the second set of experiments both elastin and collagen metabolism were analyzed from the upper thorax to the abdomen in the young pig, in vitro. Tissue was incubated with either 14C leucine, 14C valine, 3 H valine, or 3H proline for 1, 2, or 24 hours. The tissue was then extracted with detergent followed by urea extraction and autoclaving. The insoluble pellet was defined as elastin. Radioactivity was measured in each fraction. Total in- corporated radioactivity was higher in the upper thorax than in the abdomen during 1 or 2 hour incubations. ~ost of the radioactivity was seen in valine-rich soluble proteins. However, after a 24 hour incubation tae abdomen was the most radioactive region. The label was primarily in the autoclaved supernate, and it reflected collagen formation as shown by 3 H hydroxyproline production. Because the collagen content in the abdomen is higher than in the upper thorax, this result was expected. What was not expected was that little or no increased incorporation was seen in the elastin fraction of the upper thorax, as compared to the abdomen, which would correspond to the higher elastin content of the thorax. From this, it was concluded that collagen formation did not appear to be altered significantly in vitro, while v elastin formation was markedly altered. Thus, quantitative questions about elastin metabolism must be answered through experiments conducted in vivo. The third set of experiments involved analyzing elastin and collagen metabolism from the upper thorax to the abdomen of 4-6 week and 2 day old rat aortas, vivo. in Rats were injected intraperitoneally with 3H gly- cine and then sacrificed at specified times after injection. Soluble and insoluble elastin and collagen were isolated. In the 2 day old rat little radioactivity in the tropoelastin fraction could be isolated. The elas- tin fraction clearly showed an increase in the accumulation of radioactivity in the upper thorax as compared to the abdomen. Synthesis of radioactive collagen was higher in the abdomen, but this was not reflected by increased accumulation of radioactivity in that region. Finally, radioactivity in both elastin and collagen appeared to accumulate at the same relative rates. In the 6 week old rat, radioactivity in the tropoelastin fraction could not be isolated in quantities large enough to account for accumulation in the elastin pellet. Again, accumulation of radioactivity in elastin of the upper thorax was greater than in the abdomen. Synthesis of radioactive collagen was increased in the abdomen along with increased accumulation of radioactivity into the insoluble collagen fraction in the abdomen. vi Finally, accumulation of radioactivity in collagen appeared to proceed slower than in elastin. vii "I can do all things through Christ who strengthens me." Phillipians 4:13 CONTENTS ABSTRACT PREFACE PART I. iv · .. . . .. . . . . .. . . . MOLECULAR STUDIES: THERMOLYSIN DIGESTION OF TROPOELASTIN Introduction Methods . . • Results • . Discussion References PART II. · 1 ·...... ....... ·...... METABOLIC STUDIES: ELASTIN AND COLLAGEN METABOLISM IN THE YOUNG PIG AORTA IN VITRO VITA ....·· 74 79 92 125 127 --- · · · · 16 71 76 METABOLIC STUDIES: ELASTIN AND COLLAGEN METABOLISM IN THE YOUNG RAT AORTA IN VIVO Introduction Methods Results Discussion References 1 3 76 Introduction Methods . • . Results . . . Discussion References PART III. x · · ·· · · · · · ·· · ·· ·· · · ·· · · · ······· ·· · · · · · ·· 129 · · · · ···· · · · · ···· · · · 129 131 148 238 241 ··· 243 PREFACE Elastin and collagen are two structural proteins that work synergistically, at least in the mammalian aorta, to produce a flexible and yet durib1e, tensile organ. They both appear to be produced by cells of similar origin. Each contains hydroxyproline and approximately 300 residues/1000 residues of glycine. They are similar in many aspects, but do they follow similar metabolic pathways during synthesis, processing, and formation of the mature protein? Collagen metabolism has been clearly outlined in recent years, but little is known about elastin metabolism. Does it relate to collagen metabolism in any way? These studies were designed to investigate primarily the metabolism of elastin, and secondly, its relationship to collagen metabolism. I am indebted to Dr. L. B. Sandberg for his patience, care, and guidance of my work. To each of my committee members, Dr. Alfred Linker, Dr. Elvera Ehrenfeld, Dr. Leroy Kuehl, and Dr. Bill Gray, I am grateful for guidance and for their desire for quality research. Barbara Hansen Leslie, Chuck Leach, Dr. Vern Alvarez, Dr. Ron Torres, Terril Wolt, Dr. Norman Soskel, Dr. Jeff Davidson, Dr. E. G. Cleary, Dr. Bob Mecham, and Dr. Barry Oakes have all given me technical assistance. family: To my Mr. and Mrs. John C. Leslie, Doris Leslie, Nancy Conlee, Greg Leslie, Jennifer Leslie, and Jay Leslie, I am indebted for love and encouragement. To Howard Larson, Pastor Jim Powers, and other Christian friends, I am thankful for prayers. I To my Lord Jesus Christ, give all glory. This work has been supported by the National Heart, Lung, and Blood Institute grants (HL 22446, HL 20546, and HL 11963). xi PART I MOLECULAR STUDIES: THERMOLYSIN DIGESTION OF TROPOELASTIN Introduction Elastin has been a difficult protein to study structurally. It is not easily dissociated by enzyme or chem- ical means; however, some efforts have met with success. The first breakthrough occurred in the early 1950's when Lansing isolated elastin by mild alkaline hydrolysis and established the first accurate amino acid analysis of elastin. l Partridge, a few years later, developed a technique for solubilizing elastin by refluxing it repeatedly in oxalic acid. 2 This resulted in a heterogeneous population of peptides, and revealed the characteristic of coacervation common in soluble elastin. Quite soon after, Thomas isolated the major crosslinks of elastin, desmosine and isodesmosine. 3 Extensive work followed on determining how the crosslinks were formed. Anwar showed that four lysines were required for one desmosine. 4 others isolated Miller and ~-aminoadipic-o-semialdehyde (allysine) and showed that it was a precursor of desmosine. 5 Pinnell then isolated lysyl oxidase, an enzyme necessary for 2 oxidatively deaminating the lysine, a necessary step leading to the formation of a1lysine. 6 Therefore, an under- standing of the crosslinking nature of elastin has been well outlined. Also needed for structure determination was amino acid sequence data. Sequence data has come from insoluble elastin peptides isolated in a few fortuitous experiments. Keller et ale sequenced 17 steps of a nonpolar peptide from bovine ligamentum nuchae. 7 It was isolated by hot alcoholic extrac- tion and then separated on several different chromatographGerber 8 and Foster 9 have both obtained some ic columns. sequence data, by isolating small crosslinked peptides, but the sequences were very short. The first really big breakthrough in sequencing came with the discovery of tropoelastin. Since then Sandberg lO , Torres ll , Grayl2, and Foster 13 have done extensive sequencing. At present, about 75% of the amino acid sequence of the porcine molecu1e (~580 residues) is known. done using tryptic ~r Most of the work has been thrombic digests of tropoelastin. Trypsin is lysine and arginine specific, cleaving on the carboxyl side. Thrombin is arginine specific, cleaving on the carboxyl side. Since most of these peptides have now been sequenced, a new enzyme approach was needed to obtain overlapping sequence data. It was determined that thermo- lysin, an enzyme isolated from a thermophilic bacterium, Bacillus thermoproteolyticus, hydrolyzed the peptide bonds 3 of proteins and peptides at the amino sites of hydrophobic amino acids with bulky side chains. 14 This enzyme promised to be a good choice since there are quite a few bulky hydrophobic amino acids in tropoelastin. Thermolysin is a metallo endopeptidase (EC 3.4.4 group) with a zinc ion in the active site along with a histidine residue. 15 It preferentially cleaves on the amino terminal side of isoleucine, leucine, valine, and phenylalanine in insulin and the tobacco mosaic virus protein. 14 Peptidyl proline inhibits the cleavage reaction when it is in the position -X-proline, but not -X-Y-proline or proline-Y-X-, where X is any bulky hydrophobic amino acid and Y is any amino acid except proline or hydroxyproline. 14 ,16,17 Th~rmolysin react? best at a neutral pH, and is stable at room temperature up to ~80oC in the presence of Ca 2 +. l8 Because of these characteristics, it was decided that thermolysin digestion could produce peptides with overlap information. This was the purpose of the experiments: to isolate new overlapping peptides to tryptic and thrombic fragments of porcine tropoelastin. Methods Digestion £f protein ~ thermo lysin A control experiment was carried out using oxidized insulin. 30 mg of protein was dissolved in 2 ml H2 0. While stirring, 0.25 ml of thermolysin (0.01%) was added 4 to the solvent, while controlling the temperature at 36 0 C and pH at 8.9 by a pH stat. The reaction was stopped after 30 minutes, and the digest was used for two dimensional peptide mapping. A pilot experiment of the digestion of tropoelastin by thermolysin was performed by adding 17 mg tropoelastin in 1.5 ml H20. While stirring, 0.13 ml of thermolysin (0.01%) was added to the solvent, and the temperature was controlled at 36 0 C for 30 mintues. pH was not controlled. The digestion products were then lyophilized. After drying, they were dissolved in 10% acetic acid and separated from salt by chromatography on Sephadex G-lS. The salt free digestion products were lyophilized and later characterized by two dimensional peptide mapping. A preparative digest of tropoelastin was performed to provide enough peptide material for sequencing. 209 mg tropoelastin in 14 ml H2 0 was adjusted to pH 8.9 with 0.01 N NaOH. The pH was kept constant by a pH stat throughout the experiment. While the solution was stirred, 1.681 ml of thermolysin (0.01%) was added. tinued for 30 minutes at 36 o C. glacial acetic acid. The reaction was con- It was terminated with The products were lyophilized, de- salted on Sephadex G-lS, lyophilized again, and used for two dimensional peptide mapping and peptide isolation. Two dimensional peptide mapping. The two dimensional peptide analysis involved electrophoresis in the first 5 dimension. This was carried out using either a large 10,000 volt Gilson two tank electrophoretic apparatus 19 , or a smaller flat bet electrophoretic apparatus. Samples of the whole digest of tropoelastin were separated electrophoretically on the large apparatus. ma, 3000 V, for 3 hours, on Whatman 3 MM paper 5 ft by 18 in. The electro- Electrophoresis was carried out at ~110 phoresis buffer was 1.64 M formic acid, pH 1.58. After running samples in duplicate in the first dimension, one strip was stained with ninhydrin to locate peptides, and the other was sewn on a rectangular piece of Whatman 3 MM paper. The rectangular piece of paper was 45 by 58 inches. Descending chromatography was then performed with a two phase system of n-butanol:glacial acetic acid:water (3.4: 1.0:5.0). A large chromatographic chamber was first equil- ibrated with the lower phase, after which the paper was attached to a trough overhead and upper phase added to the trough. The paper did not touch the lower phase. Chroma- tography usually took about eighteen hours, after which the paper was dried and stained with ninhydrin. Flatbed (microelectrophoresis) was also carried out on some peptides. Control experiments were performed using standard amino acids. The electrophoresis buffer was 0.2 M pyridine acetate, pH 3.5. at a constant 2500 V and ~200 Electrophoresis was performed ma for 30 minutes. Dimen- sions of the 3 .HM Whatman paper were 40 cm by 20 cm. After 6 electrophoresis, the paper was dried and stained with ninhydrin, or cut into strips and chromatographed in a second dimension and then stained with ninhydrin. Fraction 3 of the experiment was separated by electrophoresis in the first dimension using the above conditions, and then separated in the second dimension by paper chromatrography. Peptide purification £y molecular sieving. The crude digest of tropoelastin as described on page 4 was first separated at 4°C on a Sephadex G-75 column, 2.S cm by 1.0 m. The elution buffer was 0.2 M pyridine acetate, pH 5.0 with 0.1 mM EDTA. was collected. The flow rate was 24 ml/hour, and 3 ml/tube Before purification of the digested tropo- elastin, the Vo (void volume) and Vt (total volume) were determined respectively, by sieving blue dextran and DNPalanine through the column. Once column conditions were determined, the digested tropoelastin was sieved and collected. Samples of each tube were used for ninhydrin anal- ysis. Peptides 2, 3, and 4 were separated by molecular sieving on Sephadex G-SO, similarly to the G-75 separations. Using the same column size, the flow rate was 32 ml/hour, and 3.2 ml/tube was collected. changed to 50% acetic acid. The elution buffer was Tubes were again assayed for protein by the ninhydrin reaction. Sephadex G-2S and G-IS were used many times to desalt 7 peptides. Usually protein elution was monitored by the ninhydrin assay or OPT, and salt by DNP-alanine. In later purifications, conductivity of the solvent was used to measure salt concentration. Peptide purification £z ion exchange chromatography. Ion exchange chromatography was extensively utilized with different resins and solvents for various peptide mixtures. Peptide 4.1,.2 (pooled) was separated on a TP resin (Technicon) at 20 0 C, using a linear salt gradient of 0.01 M pyridine acetate, pH 5.0, to 0.5 M pyridine acetate, pH 5.0, plus 1.0 M KCl. Protei~ elution was monitored by OPT, and the eluant was collected at 6 ml/tube/IO minutes. Peptide (4.1,.2)3.1 was purified by ion exchange on ECG resin (epich10rhydrin glycine cellulose) developed with a linear pH gradient of 0.02 M pyridine acetate, pH 6.1, to 0.02 M pyridine acetate, pH 4.2. Sixteen hours was the length of the experiment, with a flow rate of 4.5 ml/hour. The column size was 20 cm by 1 cm. Peptide 3.2 was separated on Whatman CM Cellulose developed with a slightly concave pH and salt gradient of 0.2 M pyridine acetate, pH 6.1, to 0.2 M pyridine acetate, pH 4.2, with 0.5 M KC1. Protein elution was monitored by OPT, and the eluant was collected at 5 ml/tube/20 minutes. Peptide 3.1.1 was separated on Whatman CM Cellulose developed with a linear pH gradient of 0.2 M pyridine acetate, pH 9.1, to 0.2 M pyridine acetate, pH 3.0. Sixteen 8 hours was the length of the experiment. Protein elution was monitored by OPT, and the eluant was collected at 3 mIl tube/20 minutes. Peptide 2 was also purified under similar conditions except that the column was developed with a concave gradient. Peptide 3.1.1.3 was separated on DEAE Sephadex developed with a linear pH gradient of 0.2 M pyridine acetate, pH 4.5, to 0.2 M pyridine acetate, pH 8.3. Peptide 2.2 was purified on CM Cellulose developed with a gradient of 0.05 M pyridine acetate, pH 8.0, to 0.05 M pyridine acetate, pH 5.0. The gradient was linear during most of the separation with a rapid increase at the end. Protein elution was monitored by OPT and the eluant was collected at 4 ml/tube/10 minutes for 24 hours. Peptide 3.1 was separated on a desalting ion exchange column. The column was 92 cm by 1.75 cm (O.D.). The lower 79.5 cm was filled with Sephadex G-15, and the 1.9 cm on top was filled with DEAE Sephadex A-50. The column was equilibrated with 0.2 M pyridine acetate, pH 5.0. Protein elution was monitored by OPT and the eluant collected at 3 ml/tube/6 minutes. Peptide 3.3 was purified on a column containing Bio Rex 70. The column was developed with a slightly concave pH gradient from 0.2 M pyridine acetate, pH 8.3, to 0.2 M pyridine acetate, pH 3.0. by OPT. Protein elution was monitored 9 Ninhydrin color reaction for tration. 50 ~l measuri~g protein concen- from each tube containing peptide sample was added to a polypropylene tube along with 50 NaOH (100 mixed. ~l of 10 N if the peptide was in an acid solvent) and Base hydrolysis was performed by autoclaving the mixture 30 minutes at 2l0 o C, and 20 psi. 100 ~l ~l After cooling, glacial acetic acid was added to each tube and mixed. 0.5 ml of ninhydrin reagent was added, and the tube was heated in a boiling water bath for 20 minutes. After cooling, 2.0 ml of 50% propanol was added and absorbance was measured at 570 nm. 2 ~l, 5 ~l, and 10 ~l Leucine standards (1 mg/ml) of were also evaluated in duplicate. The ninhydrin reagent was comprised of the following: to 50 ml of citrate buffer (0.2 M sodium citrate, pH 5.0) was added 80 mg SnC12' After dissolving the Sne1 2 in the buffer, 2 g of ninhydrin in 50 ml Piersolve (Pierce) was added and thoroughly mixed. To stain the peptides separated on paper by either electrophoresis or chromatography, a different ninhydrin solution was used. It consisted of 200 ml acetone plus 24 ml cadmium solution (20 g cadmium acetate, 400 ml glacial acetic acid, and 2000 ml H20) along with 2 g ninhydrin. The dry strip of paper was soaked in this solution similarly to wetting shipping tape. an upright oven. It was then dried in Gloves were worn throughout the procedure to avoid artifacts caused by fingerprints. 10 OPT analysis. To assay peptides eluting from columns, OPT (o-phthalaldehyde) was used employing a Chromatronix automatic sampler which would sample 10 eluant every few minutes. 20 ~l or 50 ~l of the The sample was mixed with OPT reagent (500 ml 0.5 M lithium borate, pH 9.3; 1 ml S-mercaptoethanol; 400 mg OPT in 5 ml 90% ethanol; and 1.5 ml 30% Brig detergent). After reacting in continuous tubing, fluorescence was measured on an Aminco fluorimeter with a Corning #7-15 primary filter, and Wratten #3 secondary filter. Optimal excitation of the OPT compounds was 340 nm and fluorescence at 450 nm. Amino acid analysis. Amino acid analyses of protein were performed on a Beckman Model 121 Automatic Amino Acid Analyzer using a two column system. The acidic resin was type A-5; the basic resin was type PA-35. A standard was routinely evaluated in which 12.5 nmol/column of each amino acid was injected per 700 calculated by C = net ~l. A constant, C, was height of standard amino acid x 80 (except hydroxyproline, which was 40). Micromoles of each amino acid in the experimental analysis was determined by: micromoles = net height experimental. K, a constant was C determined by: K = 1000 total micromoles of all amino acids Therefore residues/lOOO was determined by: K x micromoles (for each amino acid). The constant C for desmosine and isodesmosine was 11 calculated as 3.65 leucine equivalents 3 : cine x net height leucine x 80 x 3.65. ~ area under ogram.) C = ~ height leu(Where ~ height = the "dots" of the leucine peak on the chromat- The constant for lysinonorleucine (LNL) was deter- mined as 2 leucine equivalents: net height leucine x 80 x 2.0. C =~ height leucine x Micromoles of each (des- mosine, isodesmosine, and lysinonorleucine) net height x ~ = height . C - All acid hydrolyses for amino acid analyses were carried out either at 1100C for 16-24 hours, or 1350C for 3 hours, in 6 N HCl. Peptide purification EY countercurrent chromatography. The countercurrent column consisted of 100 coils, each coil which consisted of alternating differently sized tubing. 2l ,22 The large tubing had an inside diameter of 2.0 mm and was 20 cm long. This was alternated with small tubing with an inside diameter of 0.5 mm, and was 34 mm long. The column was first filled with "light" buffer. After completely filling all the tubing with "light" buffer, heavy buffer was added at a constant rate (28/27 on Harvard pump 975). The column was then equilibrated so that all the large tubing contained light buffer with droplets of heavy buffer dropping through in the direction of downflow. Once equilibrated, a peptide mixture in 1.0 ml of both phases was added. the heavy buffer. Fractions were collected of When it appeared that all the peptide 12 had either flowed through the column, or migrated to the light phase and stayed in the column, the column was "blown out" by a gentle stream of nitrogen and the fractions were collected. Many two phase (light/heavy) systems were used. Among them were: 1) n-butanol:lO% acetic ~cid:pyridine (20:4.4:1.5); 2) 2-butanol:H ZO:TFA (trifluoroacetic acid) (120:160:1); 3) 2-butanol:DCA (dichloroacetic acid) :3% acetic acid (Z:l:l); 4) n-propanol:3 M sodium acetate (1.5:1); 5) pyridine:IO% acetic acid:ethyl acetate (0.7: 1:1); and 6) n-hexanol:HZO (1:1). The solvents were mixed well so that they became saturated, and they were then separated into light and heavy phases. Dansyl endgroup analysis ~ peptides. Peptides were dansylated according to the method of Gray.23 This in- volved using 0.5-5 nmol of peptide and drying it down in vacuo in a small glass test tube (1 ml in size). 15 ~l of 0.2 M sodium bicarbonate solution was added, and the sample was redried. This was dissolved in 15 ized water and the pH determined. approximately 8.5-9.0. (5 mg dansyl 15 chloride/~l ~l ~l of deion- The pH needed to be of dansyl chloride solution acetone) was then added. was allowed to react for I hour at 37°C. then dried and 100 ~l 6.1 N HCl was added. This The sample was The tube was sealed and the sample was hydrolyzed for 16-18 hours at 105 0 C. After hydrolysis, the HCl was removed and the DNS 13 amino acids were extracted two times using 100 acetate saturated with water. redissolved in 20 ~l ~l ethyl The extract was dried and of 50% pyridine. It was then spotted on Whatman 3 MM paper (46 cm by 57 cm) along with standards. The paper was wetted with electrophoresis buffer (pyridine:acetic acid:water (10:20:2500», pH 4.40. The DNS amino acids were separated on a flat bed electrophoresis plate for 2.5 hours at 80 V/cm. The paper was dried and observed under ultraviolet light. Peptide purification ~ recycling. Recycling was performed by a circular flow system consisting of a molecular sieving column, OPT monitor, and pump.20 A peptide was recycled until separation was achieved, and then it was collected. Peptides 1.1 and 1.2 (pooled) were recycled on Sephadex G-75 fine in a solvent of 0.2 M pyridine acetate, pH 5.0. The peptides were recycled twice and then collected at 2 ml/tube/20 minutes. Amino acid sequencing. Peptides (4.1,.2)5.1, (4.1,.2)8, and (1.1,.2)1 were sequenced on a Sequemat Mini-IS solid state sequencer. (4.1,.2)5.1, (1.1,.2)1, and (4.1,.2)8 were coupled by water soluble carbodiimide to APG (aminopropyl glass). (4.1,.2)8 was also coupled to DITe (diisothiocyanate) glass beads. Sequencing was accomplished by the sequential Edman degradation chemistry. The resultant thiazolinone residues were converted to PTH 14 (pheny1thiohydantoin) derivatives of the amino acids and separated on high pressure liquid chromatography (HPLC) using a Waters Liquid Chromatograph. Carbodiimide coupling of (4.1,.2)5.1 was accomplished as follows: to 100 nmol of ammonia free peptide was added two drops of triethylamine; the peptide was then dried under vacuum. To the dry peptide was added 500 (dimethy1formamide), 50 ~l ~l DMF N-methy1morpho1ine, and 75 T-Boc (t-butoxycarbonylazide). ~l After heating in a covered tube at 50 0 C for 5 hours, the sample was dried under vacuum. 500 ~l dried again. 1-2 ~l of DMF was then added and the peptide was It was diluted in 200 of N-ethylmorpholine. ~l DMF along with 5 mg EDC-HCl (n-ethyl-n-di- methylaminopropylcarbodiimide hydrochloride) was added, and the reaction was allowed to proceed at 40 0 C for 90 minutes. APG (45 mg) was then added and the mixture was stirred at 40 0 C for 3 hours. The excess amino groups of the APG were blocked with T-Boc as above, and the APG was dried. sample was then re~dy The for sequencing. Carbodiimide coupling of (1.1,.2)1 and (4.1, .2)8 was performed according to the method of Appe1la 24 with several modifications. The N-termina1 blocking reagent employed was BOC-ON (Aldrich). done with pentane. Trituration of excess BOC-ON was The EDC-HCl (1-ethyl-3(3-dimethylamino- propyl)-carbodiimide hydrochloride) was obtained from Pierce. 50 mg of APG was used as the support. In blocking 15 the coupled glass, N-acetylimidazole (Pierce) was used instead of MITC (methylisothiocyanate). DITC coupling was carried out as follows: to 100 nmol of ammonia free peptide was added 0.3 m1 of N-methylmorpholine:water (1:1, adjusted to pH 9.5 with TFA), of DITC glass. and 200 mg The mixture was evacuated in a vacuum des- sicator for 30 minutes using a water aspirator. The pep- tide-glass mixture was then removed and stirred for 2 hours. Ethanolamine (0.1 ml) was added to block excess isothiocyanate groups. After stirring for 1 hour, the sample was washed with methanol and was dried. Coupled peptides were subjected to degradation as described by Laursen. 25 an~~! 10% phenyl isothiocyana~e, meth- dichlorothane, and 99+% TFA (Pierce), were the es- sential reagents, along with Sequemat's sequence buffer, which was a 3:2 mixture of pyridine and N-methylmorpholine trifluoroacetate buffer (pH 8.1). The sequencer program used was the basic one-column program designed by Sequemat. The manual conversion of the thiazolinones to PTH's involved the following procedure: the sample was dried down with nitrogen at room temperature. 0.2 ml of 20% TFA was added, and the sample was vortexed. The sample was heated at aooc for 10 minutes and then dried on an N-Evap. After cooling, 0.5-0.7 ml methanol was added, and the sample was sieved through a Dowex 50W-X2 column (200-400 mesh, H+ form, in disposable pipettes). The sample tube was 16 washed with methanol two times, and the washes were also passed over the column. The resulting effluent was col- lected in a 3 ml conical test tube. The sample was dried down under nitrogen and diluted appropriately in acetonitrile for running on HPLC. The thiazolinones from the sequencing experiment of (1.1,.2)1 were converted to their respective PTH's using Sequemat's P-6 Autoconverter. They were also sieved over Dowex columns before HPLC analysis. HPLC separation of PTH amino acids was performed on a Waters Liquid Chromatograph using a at 55 0 C. ~C18 Bondapak column The program was 0.02 M sodium acetate, pH 5.0 (A), and CH3GN (B) with a fifteen minute gradient period (curve 6 on the Waters Model 660 Solvent Programmer). The equilibration time between samples was seven minutes, and the flow rate was 2.5 ml/min. BTH-norleucine (BTH = The internal standard was benzylthiohydantoin). Results As a control, insulin was the first protein digested by thermolysin. The two dimensional map of ~he digest seen in Figure 1 compared favorably to that of Matsubara. 14 A pilot study was then done on tropoelastin (Figure 2). The results were promising with several discrete pep- tide areas resolved. Therefore, a large scale digestion of tropoelastin was done. The extent of digestion of tro- poelastin was indicated by the two dimensional peptide map 17 Figure 1. A two dimensional peptide map of thermolysin digested insulin. The first dimension was paper electro- phoresis, and the second descending chromatography. Peptides were detected by ninhydrin stain. Figure 2. A two dimensional peptide map of a pilot study on thermolysin digested tropoelastin. The first dimension was electrophoresis, and the second descending chromatography. Peptides were detected by ninhydrin stain. < ~"~d \, 't\ I , i \j (" -, 1/~ \ ! '.......... .,/ 19 as seen in Figure 3. Much of the chemical nature of these peptides was indicated in the two dimensional map. There were some large hydrophobic peptides (A) which migrated slowly on electrophoresis and fast on descending paper chromatography. There were also small hydrophilic pep- tides (B) that migrated fast on electrophoresis and slowly on descending paper chromatography. In between these two populations was a heterogeneous population of peptides. The peptides in the region A probably correspond to the large peptides that eluted in the first two fractions of the Sephadex G-7S filtration of the digest (Figure 4). Much of the material in the B region may have corresponded to protein that eluted in fraction 4 of the G-7S separation. As can be seen in Figure 4, there were three to four discrete regions of peptides according primarily to size. region. Amino acid analyses were done on samples from each As can be seen in Table I, the peptides that eluted early (large) from the G-75 contained higher percentages of valine, and those that eluted later (smaller) contained higher amounts of alanine. This was consistent with the idea of there being large valine rich hydrophobic regions interspersed with smaller alanine and lysine rich regions, where crosslinking often occurs. were then pooled into 4 fractions. 9 The samples Since extensive work followed on each of the four peptide fractions they will be considered individually. Figure 5a,b is a flow diagram 20 Figure 3. A two dimensional peptide map of the prepara- tive thermolysin digest oftropoelastin. The first dimen- sion was electrophoresis, and the second descending chromatography. Region (A) was composed of large hydrophobic peptides, and region (B) was composed of small hydrophilic peptides. The peptides were detected by ninhydrin stain. 22 Figure 4. Thermolysin digested tropoelastin sieved over a Sephadex G-7S column. Amino acid analyses were done on sequential tubes between tube #106 to tube #166, after which all the tubes were pooled in four main fractions, as shown on the graph. drin stain assay. Protein was assayed for by the ninhy- 23 o o N o +-I 00> ~ o o ~ '. o 00 o > 00 TABLE I AMINO ACID COMPOSITION OF SEQUENTIAL FRACTIONS OF THERMOLYSIN DIGESTED TROPOELASTIN SEPARATED ON SEPHADEX G-75 EXPRESSED AS RESIDUES PER 1000 RESIDUES The thermo1ysin digest of tropoe1astin was dissolved in 0.2 M pyridine acetate (pH 5.U) and run over a Sephadex G-75 column. A1iquots of sample from sequential fraction collector tubes were taken for amino acid analysis (see Figure 4). Amino Acid Lys Arg Asp Thr Ser G1u Pro Gly Ala Val lIe Leu Tyr Phe llmo1es Total Amino Acids *112 118 20 11 20 10 51 152 343 121 283 Fraction Collector Tube Number 124 130 136 142 148 6 13 20 7 10 10 17 13 124 352 144 232 13 34 14 27 17 12 14 14 136 334 203 164 19 49 12 21 7.79 8.05 6.38 13.74 3 6 11 6 140 365 88 349 11 5 8 19 11 114 375 88 330 3 8 2.12 154 160 48 50 11 24 24 21 120 336 170 105 29 60 23 26 39.61 16 25 39 7 7 5 6 9 10 6 14 27 22 15 93 351 186 127 22 61 12 22 37.88 159 312 183 145 19 101 5 15 25 17 17 138 169 259 189 25 73 12 23 21.64 20.74 9 * low values in analysis. N .p.. 25 Figure 5a. A flow diagram of the major separation pro- cedures and the resultant peptide pools. 26 TROPOELASTIN DIGEST I Sephadex G-7~ 1 1 I ) '4 I Sephadex Sephadex G-50 I Sephadex G-S 0 G-50 Counter 'current (n-Butanol/lO%HAC/Pyridine) I r I I 3.1 3.2 3.3 Counter Current Ion Exchange (2-Butanol/H 2 o)TFA) Ion CM Exchange Counter Current Cellulose Bio Rad 70 (Ethyl Acetate/lO%HAC/ ~ I 2.1 2.2 PYI id ine) ·1 3.3: 2 3: 3. 3 i , ( t 1.1 1.2 1.3 1.4 3.3.1 Ion ~ Exchange Recycle CM Sephadex Cellulose G-75 Counter Current (n-Butanol/lO%HAC/Pyridine) I I I I (1:1,.2)1 (1.'1,.2)2 Ion Exchange DEAE I 3.i.l 3.i.2 I I rl--------~I--------~'~ I 3.1.1. 2 Ion Exchange DEAE I t t 3.1.1.3.1 3.1.1.3.2 4.2 4.3 '---y-/ Ion Exchange CM Cellulose 3.1.1.3 4.1 . 3.1.1.1 Ion Exchange TP Resin I I 1 2 3 4 5 6 7 8 9 10 11 12 13 I 3.1.1.3.3 27 Figure 5b. A flow diagram of separation procedures and resultant peptide pools of fraction (4.1,.2). 4·.1 4. 2 I Ion Exchange TP Resin I j I I I I I , 1 (4.1,.2)1 (4.1,.2)2 (4.1, .2)3 (4.1,.2)4 (4.1,.2)5 (4.1,.2)6 (4.1,.2)7 (4.1,.2)8 I I Counter Current (n-Butano1/10%HAC/Pyridine) I I (4.1,.2)3.1 Counter Current (n-Butano1/10%HAC/Pyridine) I I (4.1,.2)3.2 (4.1,.2)6.1 Counter Current (n-Butano1/10%HAC/Pyridine) I Ion Exchange ECG Resin r(4.1,.2)3.1.1 I (4.1,.2)5.1 I (4.1,.2)3.1.2 rI I (4.1, .2)9 (4.1, .2)10 (4.1, .2)11 (4.1, .2)12 (4.1, .2)13 Counter Current (n-Butano1/10%HAC/Pyridine) I (.1,.2)11.1 IV 00 29 of the entire separation procedure. Purification of peak! Fraction 4 eluted near the Vt of G-75, and contained peptides of 30,000 MW or less. Since the average molecular weight of an amino acid residue in tropoelastin was 100 this would imply peptides of around 30 residues or less. Therefore, it was decided to separate fraction 4 on Sephadex G-50 with a fractionation range of 1,500-30,000 MW. The results of this are seen in Figure 6. It appeared very polydisperse, and much of the ninhydrin positive material migrated just before the Vt . three fractions: It was divided into 4.1, 4.2, and 4.3. Nothing further was done with 4.3, since it probably contained free amino acids and very small peptides, e~en di- and tri-peptides. In the analyses of 4.1 and 4.2 (Table II) proline was consistently high which was interesting since thermolysin does not cleave within one residue next to a proline, and yet some of these were small peptides. also higher than expected. Phenylalanine was This may indicate that peptides containing phenylalanine were retarded in elution from the Sephadex. The aromatic ring may have interacted with some of the hydroxyl groups of the dextran, so that some peptides could have been larger than their apparent size as determined by molecular sieving. 4.1 and 4.2 were pooled after looking at their respective amino acid analyses, in which little difference was seen. This pool was then 30 Figure 6. column. Fraction (4.1,.2) sieved over a Sephadex G-50 Amino acid analyses were done on sequential tubes from tube #105 to tube #125, after which all the tubes were pooled into three fractions. for by the ninhydrin assay. Protein was assayed 31 o .-l .-l o o .-l 1 t 00 I o I I \0 mu I N CO OL~ aJNVg~OSHV 0 "'> 32 TABLE II AMINO ACID COMPOSITION OF SEQUENTIAL FRACTIONS OF FRACTION 4 OF THERMOLYSIN DIGESTED TROPOELASTIN SEPARATED ON SEPHADEX G-50 EXPRESSED AS RESIDUES PER 1000 RESIDUES Fraction 4 of pooled samples from the previous G-75 fractionation was run over a Sephadex G-50 column. Aliquots of sample from sequential fraction collector tubes were taken for amino acid analysis (see Figure 6). Amino Acid Fraction Collector Tube Number *105 110 115 120 125 Lys 18 24 32 31 Hyp 14 14 16 14 Asp 20 10 9 11 10 Thr 39 18 19 18 17 Ser 31 10 9 23 19 G1u 28 39 25 23 19 Pro 177 166 169 135 123 G1y 318 329 330 325 327 Ala 122 112 137 172 192 Val 67 112 120 99 109 Ile 31 28 19 29 31 Leu 82 51 49 56 55 Tyr 59 29 24 23 12 Phe 28 67 54 38 43 0.92 3.69 5.69 8.53 j.Jmo1es Total Amino Acids *No basics were done. 15.24 33 separated on a TP (Technicon P) resin and gave a good separation of 13 peaks (Figure 7). Each sample was lyo- philized, desalted, and an amino acid analysis performed (Table III). Endgroup analyses of the peaks indicated that most still had a mixture of peptides, i.e. more than one dansyl amino acid was present (Table IV). However, (4.1,.2)5 and (4.1,.2)8 were pure, with glycine and isoleucine as respective endgroups. These peptides also migrated as single spots on flatbed electrophoresis. Pep- tide (4.1, .2)5 was further purified by countercurrent chromatography (Table V, Figure 8). Then both peptides were sequenced, and the results can be seen in Table VI and Figure 9. As can be seen in Table VI, peptide (4.1, .2)5 extended three residues amino terminal to peptide T7a. 10 peptide, (4.1,.2)8, was unique. The other Both peptides along with (1.1,.2)1 were present in low amounts, on a molar basis. The only reasonable explanations for this, since all three peptides were present in similar amounts, are that either the initial amount of tropoe1astin used contained salt, that a quantitative loss of all peptides occurred in one of the early separation steps, or occurred. t~at incomplete cleavage However, since (4.1, .2)8 was present in similar quantity to 4.1,.2)5, a known tropoelastin peptide, it was taken to be a real peptide of tropoe1astin. Fractions (4.1,.2)7,9,10,12,13 were not pure as seen 34 Figure 7. TP resin ion exchange of (4.1,.2). sulted in thirteen discrete peptide fractions. was assayed for by the OPT assay. This reProtein Also indicated is the salt gradient, from the resin, used to elute peptides (---). 35 CONCENTRATION(MOLAR) NACL 11'1 0 .-I .-I I 11'1 I 0 .-I "'-.. N "'-.. ~ ==:::j 1 0 .-I ~ .-I '" co == -= ----; \ ..j '".-I co ".-I \ N \0 " .-I \ "l::: ~ j:Q \ ::J e-; 11'1 \0 \ ..:t .-I -:t ~ -==1 \ N \ N .-1- o co \ o N ~ 0 ~ .-I - o TABLE III AMINO ACID COMPOSITION OF PEPTIDES ISOLATED BY ION EXCHANGE CHROMATOGRAPHY ON TP RESIN OF FRACTION (4.1,.2) EXPRESSED AS RESIDUES PER 1000 RESIDUES Fraction (4.1,.2) was run on TP Resin ion exchange chromatography with a linear salt gradient of 0.01 M pyridine acetate (pH 5.0) to 0.01 M pyridine acetate (pH 5.0) plus 1.0 M KCl. Thirteen fractions were collected and pooled (see Figure 7). After desalting and lyophilizing, an aliquot of each was taken for amino acid analysis. Peptide Number of Amino Acid 1 4 5 6 3 *7 2 16 67 Lys 55 75 Arg 65 20 Hyp 7 12 Asp 2 11 4 3 37 42 Thr 29 62 37 5 49 8 8 4 262 Ser 35 8 7 12 G1u 24 35 43 9 Pro 82 108 146 61 98 409 338 G1y 342 253 342 337 277 287 130 209 85 Ala 232 Cys 87 63 Val 154 134 97 37 12 61 20 24 52 lIe 11 42 56 52 31 Leu 36 8 31 18 23 Tyr 17 18 20 3 14 35 Phe 5 l1mo1es Total Amino Acids 2.61 11.77 8.40 6.06 .836 18.63 apossib1y an aldol condensation product. *Basics not injected. (4.1,.2) 8 9 191 79 47 9 9 10 75 30 11 51 25 13 63 39 4 5 2 5 29 47 84 23 14 14 54 30 85 10 50 49 63 35 23 41 48 71 333 80 a 71 50 5 110 62 27 29 9 9 302 251 37 16 6 5 172 312 131 2.41 1. 90 1.58 9 12 92 288 247 57 9 10.72 7 8 95 325 231 50 7 70 30 40 1. 69 w '" 37 TABLE IV DANSYL ENDGROUP ANALYSES OF PEP TIDES Dansy1 end group analyses of various peptides were done according to Gray's method. 23 Peptide fraction Endgroups Present (4.1,.2}1 Several spots (4.1,.2}3 Val (4.1,.2}5 G1y predominate, with some Ala (4.1,.2}6.1 (4.1,.2}8 Ala predominate, with some I1e or Leu I1e predominate, with some Leu (4.1,.2}9 3 spots (4.1,.2}10 2 spots (4.1,.2}11 2 spots, maybe Val and I1e (4.1,.2}13 One broad spot (1.1, .2) 1 I1e or Leu (1.1,.2}2 lle or Leu 3.1.1 3 spots 3.1. 2 G1y 3.2.3 2 spots, one large and one small 38 TABLE V AMINO ACID COMPOSITION OF PEP TIDES PURIFIED FURTHER FROM THE PARENT PEPTIDE ISOLATED DURING ION EXCHANGE OF (4.1,.2) Four peptides from the TP Resin ion exchange of (4.1,.2) were further purified and the resulting fraction analyzed (see Figure 5b). Amino Acid Lys 3.1 62 Peptide Number of (4.1,.2) 5.1 6.1 3.1.1 3.2 65 49 11.1 10 51 81 67 24 5 1 8 7 20 His Arg Asp 4 Thr 100 104 16 65 40 4 Ser 21 10 49 5 11 81 G1u 46 46 22 4 16 88 Pro 54 65 49 82 91 11 G1y 311 316 375 429 368 398 Ala 274 274 261 111 195 95 25 a Cys Val 66 lle 12 Leu 21 Tyr 25 Phe 4 llmo1es Total Amino Acids 1. 23 65 20 16 .25 92 146 101 5 65 24 14 76 8 35 144 22 7 29 53 5 2 8 1. 38 6.88 apossibly an aldol condensation product. 3.26 2.13 39 Figure 8. Purification of (4.1,.2)5 by countercurrent chromatography (n-butano1:10% acetic acid:pyridine). Protein was assayed by the ninhydrin reaction. 40 .8 - .6 - s ~ 0 r-lI"'I /:;r.:l U :z; .4 - •2 - <: !Xl p:: 0 til !Xl <: o 20 40 TUBE If 60 80 BLOW OUT TABLE VI -. '1'Y .,. c.--: 1rT' rr. = AMINO ACID SEQUENCE DATA ON PEPTIDES (4.1,.2)5, (4.1,.2)5, AND (1.1, .2)1 Purified peptides (4.1,.2)5, (1.1,.2)1, and (4.1, .2)8 were coupled by carbodiimide Solid state sequencing was performed by sequential Edman to aminopropy1 glass. degradations and the resultant PTH amino acids analyzed by HPLC (see Methods and Figure 9). Molar Ratio Tropoelastin:Peptide i·T'. .):.1t \""""' Sequence J>~~tide - 10 5 1 . ""'I"'" 15 20 00 C'"'j (4.1,.2)5 G A R G G V G V G f'T"1 (4.1, .2)8 I G G K P P K P G (1.1,.2)1 f G P G V G v P G V G V P G V G v P G = 21 x V p G v G x p G 30 V G V P G 35 V G v P G 40 V :I> 41 g V P G 45 V G V P G 50 V x V x g 55 v g x P G / 2 -,., C,",· r..n 25 G I P t 1.0 0.17 1.0 0.07 V 1. 0 0.11 ::0 ox:; <: ~ I--' 42 Figure 9. Representative tracings of the recorder print- out of the results of sequencing peptides (4.1,.2)5 and (4 • 1, . 2 ) 8 . (a ) A s tan dar d run Step 10 of peptide (4.1, .2)5. (4.1,.2)8. 0 f P THam i no a c ids . (c) Step 6 of peptide (b ) 43 A 50 B 5 c 5 10 TIME (MINUTES) 16.5 44 by endgroup analyses and amino acid analyses. was contaminated and (4.1, .2)2 lost. (4.1,.2)4 (4.1, .2)1 was a group of small nonpolar peptides as determined by endgroup analysis and amino acid analysis. (4.1,.2)3 appeared to be predominately one peptide with an amino terminal valine, as determined by dansyl endgroup analysis. It was further purified by countercur- rent chromatography, Figure 10, and by ion exchange chromatrography, Figure 11. The primary peak was used for amino acid analysis and solid state sequencing by carbodiimide coupling. The analysis showed one glutamic acid, threo- nine, and lysine and therefore implied that this was a new peptide of about 15 residues. Unfortunately, the first attempt to sequence this peptide was unsuccessful. (4.1,.2)6 was further purified by countercurrent chromatography (Table V, Figure 12). It may be one or two predominate peptides that could possibly be sequenced together. r Since arginine and tyrosine are often found together and are present in the amino acid analysis in equal amounts, it is possible that the arginine peptide is present in one half the quantity of the lysine peptide. (4.1,.2)11 was further purified by countercurrent chromatrography (Figure 13). The major peak was subjected to amino acid analysis and several amino acids such as valine, arginine, and proline, were no longer present. However, lysine, serine, and glutamic acid were present 45 Figure 10. Purification of peptide (4.1,.2)3 by counter- current chromatography (n-butano1:10% acetic acid:pyridine). This resulted in two pooled fractions. Protein was assayed by the ninhydrin stain assay. Figure 11. Purification of peptide (4.1, .2)3.1 by ECG resin ion exchange. This resulted in two pooled fractions. Protein was assayed for by OPT. Figure 12. Purification of peptide (4.1, .2)6 by counter- current chromatography (n-butano1:10% acetic acid:pyridine). Protein was assayed by the ninhydrin stain assay. 46 1 .1 .) "- s~ 0 ..... u; I:Ll t) .05 z <: I=Q ~ 0 tf.) I=Q <: I a 20 TURE IF 20- I 1 I 2 I 10- o 26 10 TURE II s ~ 0 ..... u; ~ t) .1- z <: I=Q ~ 0 tf.) cq <: 0 I 30 TUBE 11 47 Figure 13. Purification of peptide (4.1, .2)11 by counter- current chromatography (n-butanol:10% acetic acid:pyridine). Protein was assayed by the ninhydrin stain assay. 48 o .-! .-! o ("") o .-! I N I .-! o 49 which made this possibly a new peptide. It was concluded that this fraction was very profitable in terms of new information, both in sequence and the amino acid composition of distinctive peptides. Purification of peak 1 Fraction 3 was separated on a Sephadex G-50 column with little success. It appeared homogeneous (Figure 14), but on a two dimensional map, it was heterogeneous (Figure 15). Therefore it was purified by countercurrent chroma- tography (Figure 16). Amino acid analyses were performed (Table VII), which showed that the leading peak was more hydrophilic than the peaks eluting later. Once the analy- ses performed, appropriate fractions were pooled: 3.1, 3.2, and 3.3. Peptide fraction 3.1 was separated over a DEAE column (Figure 17). 3.1.1 Dansy1 endgroup analyses indicated that contain~d three spots not clearly defined, and 3.1.2 contained one dansy1ated terminus co-migrating with glycine (Table IV). The amino acid analysis of 3.1.1 (Table VIII) and its endgroup analysis indicated that it was still heterogeneous, while 3.1.2 was probably ready for sequencing. column. 3.1.1 was then separated on a CM Cellulose This gave two distinct peaks: (Figure 18). 3.1.1.1 and 3.1.1.3 Amino acid analyses still indicated some heterogeneity in both. 3.1.1.3 was further purified on DEAE ion exchange (Figure 19). Amino acid analyses were 50 Figure 14. umn. Fraction 3 separated on a Sephadex G-50 co1- Protein was assayed by the ninhydrin stain assay. Figure 15. fraction 3. A tracing of a two dimensional peptide map of The first dimension was paper electrophoresis, and the second was descending paper chromatography. 51 1.5- a I=l 0 ,..... ll"l W u z < ~ 1.0- p:: 0 CJ) ~ < .5- a +E h If I I I 60 V 0 90 100 TUBE If 150 Vt >< c:.!l p:: ....:l c:.!l - - - - -0 - - -- -- C 120 110 tI) >< ..... <t! -- - - ( 0<:0 - ------ I 1" SCALE 52 Figure 16. Fraction 3 separated on countercurrent chromat- ography (n-butanol:lO% acetic acid:pyridine). This re- sulted in peptide material in both the heavy and light phases. tions. Therefore it was divided into three pooled fracProtein was assayed by the ninhydrin stain assay. 53 o N M o o M .0 co E-l :;J O~ ::;:I'<:! O~ ,..:l:;J ~E-l 0 \0 o N o co o TABLE VII AMINO ACID COMPOSITION OF SEQUENTIAL FRACTIONS OF FRACTION 3 OF THERMOLYSIN DIGESTED TROPOELASTIN SEPARATED ON SEPHADEX G-50 EXPRESSED AS RESIDUES PER 1000 RESIDUES Fraction 3 was dissolved in 0.2 M pyridine acetate and separated on a G-50 column. quots of sequential fraction collector tubes were taken for amino acid analysis. Then tubes were pooled into three fractions and aliquots of these pools were taken for analysis (see Figure 16). Amino Acid Lys Arg Asp Thr Ser G1u Pro Cly Ala Val lIe Leu Tyr Phe l1moles Total Amino Acids Fraction Collector Tube Number 29 40 23 26 33 Pooled Samples 3.1 3.3 96 85 67 47 35 103 7 61 17 24 126 304 201 96 7 48 10 3 5 56 9 21 120 333 193 99 5 54 12 9 6 57 12 27 97 354 199 88 9 48 18 18 8 23 27 23 90 358 218 101 27 62 17 13 17 17 112 323 228 125 22 60 13 17 3 27 3 8 66 444 265 45 3 24 16.93 24.62 19.13 14.85 8 8 13.41 Ali- 5 5 9 2 8 3 9 20 134 349 146 155 15 64 18 67 193.93 202.13 lJ1 .j::'- 55 Figure 17. tion 3.1. DEAE ion exchange separation of peptide fracThis resulted in two pooled fractions. Protein was assayed for by OPT. Figure 18. CM Cellulose ion exchange separation of pep- tide fraction 3.1.1. tions. This resulted in three pooled frac- Protein was assayed for by OPT. Figure 19. DEAE ion exchange separation of peptide frac- tion 3.1.1.3. The peak tubes were pooled. assayed for by OPT. Protein was 56 1 2 I I I I J_~ I II 50J- I'l I o I I 10 20 I I 30 40 TUBE II I 1 I 2 I 3 I o LI'\ ..;r - II o I I I I 10 20 30 40 TUBE II 50- o I I 10 20 I 30 TUBE II 40 57 TABLE VIII AMINO ACID COMPOSITION OF PEPTIDES PURIFIED FURTHER FROM THE PARENT PEPTIDE ISOLATED DURING COUNTER CURRENT FRACTIONATION OF FRACTION 3 EXPRESSED AS RESIDUES PER 1000 RESIDUES Fraction 3 was pooled into three samples after counter current chromatography. These samples were further purified, and from selected samples an aliquot was taken for amino acid analysis (see Figure Sa), Amino Acid Peptide Number of Fraction 3 3.3.2 3.1.1.3 3.1. 1 Lys 94 86 7 Thr 59 57 2 Ser 9 6 11 Glu 20 16 25 Pro 116 115 135 G1y 316 344 366 Ala 207 194 167 Val 102 104 147 lIe 7 4 16 Leu 48 53 32 Tyr 15 12 23 Phe 9 8 70 ]..Imoles Total Amino Acids 64.31 31. 39 23.70 58 attempted on these fractions and the results were unsatisfactory. Many of these peptides were present in small quantities «200 nmol) and theref9re only a limited number of assays could be performed. Peptide fraction 3.2 was separated by ion exchange chromatography to give three discrete peaks (Figure 20). The last peak was contaminated by bound copper (from tubing) and was not examined further. were pooled into four fractions. The other two peaks Endgroup analysis on 3.2.3 indicated it was still a mixture. Therefore, frac- tion 3.2.1 was the only fraction suitable for further study. Peptide 3.3 was purified on Bio Rex 70 (a weak carboxylic cation exchanger) ion exchange to give one discrete peak, 3.3.1 (Figure 21). Purification £i peak Nothing further was done with it. ~ Peptide fraction 2 was first separated on a Sephadex G-50 column (Figure 22). As was expected it was hetero- geneous and contained peptides slightly larger than those found in fraction 3 (compare Figure 22 with Figure 14). Amino acid analysis of fraction 2 indicated that at least some of the peptides contained small amounts of lysine. For this reason the fraction was chromatographed through a CM Cellulose column. 23. The results can be seen in Figure The majority of the material eluted directly through the column and was low in lYSine content. However, a small 59 Figure 20. CM Cellulose ion exchange separation of pep- tide fraction 3.2. This resulted in four pooled fractions. Protein was assayed by the ninhydrin reaction. Figure 21. ~io fraction 3.3. Rad 70 ion exchange separation of peptide This resulted in three pooled fractions. Protein was assayed for by OPT. 60 1 3 2 4 e ~ 2.0 o ...... Ll"\ o 20 10 TUBE 1 3 30 40 1 I il 2 e ~ 0 Ll"\ 20- ~ f4 U Z f4 U CI) f4 p::: 0 lO- :;:J ~ ~ III I I 0 10 I 30 20 TUBE II 40 61 Figure 22. umn. Fraction 2 separation on a Sephadex G-SO co1- Protein was assayed by the ninhydrin reaction. Figure 23. tion 2. CM Cellulose ion exchange separation of frac- This resulted in two discrete peaks. assayed for by the ninhydrin reaction. Protein was 62 ·4 .3 .2 ~/~/--~I--------~I----------~I----------~I~// o 60 Vo 80 90 TUBE II 100 110 2 1 2.0- 1.0- o 10 TUBE II 20 I 150 Vt 63 amount of material did have a significant amount of lysine causing retention and was eluted later in the gradient. The retarded fraction was labeled as 2.2 (see Table IX also). Both peaks appeared fairly heterogeneous and con- sisted primarily of moderately large hydrophobic peptides. Purification ~ peak 1 Probably one of the most interesting fractions was fraction 1. Fraction 1 contained the largest pep t id esand also the most nonpolar; therefore countercurrent chromatography was used to separate it. A two phase system con- taining n-butanol, 10% acetic acid, and pyridine was tried first. Some material came out in the heavy phase with the predominate amount left behind in the light (Figure 24). The material left in the light phase was run on a different two phase system of 2-butanol, water, and TFA. Little peptide came out this time (Figure 25); therefore, a different system involving a more polar light phase (ethyl acetate, 10% acetic acid, and pyridine) was used. The result was the extraction of the peptide material from the light phase into the heavy phase with a good separation of material. Four pools were made, and after analysis of 1.1 and 1.2 (Table X) these two were repooled. 1.2 had some copper bound to it from a faulty injection needle and that may have caused it to migrate slightly differently. 1.1 and 1.2 were pooled (1.1,.2) and recycled on a 64 TARLE IX AMINO ACID COMPOSITION OF FRACTION 2 AND PEPTIDES ISOLATED AFTER CM CELLULOSE ION EXCHANGE EXPRESSED AS RESIDUES PER 1000 RESIDUES Fraction 2 was separated on a G-50 column after which an aliquot for amino acid analysis was taken from the pooled sample. Fraction 2 was further purified on a eM Cellulose column into two fractions. An aliquot from each fraction was taken for amino acid analysis. Amino Acid Fraction 2 Lys 26 His 2 Arg 5 Fraction 2 Peptides 2.1 2.2 13 56 25 Hyp 13 Asp 4 Thr 17 5 38 Ser 11 7 2 G1u 12 13 3 Pro 151 123 147 G1y 390 341 350 Ala 200 221 118 Val 97 183 140 I1e 18 23 2 Leu 53 42 71 Tyr 9 4 24 Phe 15 12 26 llmo1es Total Amino Acids 33.20 65 Figure 24. ogaphy Fraction 1 separated by countercurrent chromat- (n-butanol:lO% acetic acid:pyridine). This resulted in peptide material in both the heavy and light phases. Protein was assayed by the ninhydrin stain. Figure 25. Fraction 1 purified by countercurrent chromat- ography (2-butanol:water:TFA). This resulted in peptide material primarily in the light phase. Protein was as- sayed by the ninhydrin stain assay. Figure 26. Fraction 1 separated by countercurrent chromat- ography (ethyl acetate:lO% acetic acid:pyridine). This resulted in a good separation of peptide material primarily in the heavy phase. Four pooled fractions were made. Protein was assayed for by the ninhydrin stain assay. 66 ·4 •3 ·2 .1 I 0 20 40 I 20 I 40 60 TUBE II 100 BLOW OUT 120 •4 sc:: •3 c r-. lJ"\ ~ .2 u z <: i:Q p::: ·1 0 tr.I i:Q <: 0 I 60 I 80 TUBE II 3 ·8 I I 100 BLOW OUT 120 4 ·6 ·4 •2 a 20 40 60 TUBE f) 80 100 BLOW OUT 120 67 TARLE X AMINO ACID COMPOSITION OF FRACTION 1 PEPTIDES ISOLATED AFTER COUNTER CURRENT AND RECYCLING EXPRESSED AS RESIDUES PER 1000 RESIDUES Fraction 1 was purified by counter current and four fractions isolated. A1iquots of the first two were taken for amino acid analysis. Then the first two fractions were pooled and separated by recycling G-75 chromatography. Two fractions were obtained and amino acid analysis was performed on one of them. Amino Acid 1.1 Lys 20 His 8 Asp Fraction 1 Peptides 1.2 (1.1, .2)2 13 3 12 13 2 Thr 4 3 2 Ser 31 25 15 G1u 16 16 7 Pro 152 166 149 G1y 354 366 383 Ala 82 59 65 Val 304 319 366 lIe 4 6 2 Leu 8 6 5 9 Tyr j.lmo1es Total Amino Acids 7.80 13.09 68 G-75 recycling system with ho~es of separating and then cutting out any contaminants between recycling. this did not appear necessary (Figure 27). However, The two peaks were reisolated and amino acid analyses were performed. (1.1,.2)2 was very similar to that of 1.1 and 1.2 (Table X). Most of the copper was removed from (1.1, .2)2 by gel filtration on G-25 in 10% acetic acid. Endgroup analyses were done on (1.1,.2)1 and (1.1,.2)2 and one spot co-migrating with isoleucine or leucine was seen in each case (however there was probably no isoleucine or leucine in the peptide). It was therefore concluded that (1.1,.2)1 and (1.1,.2)2 were the same peptide, or close to it. What was most interesting was that the residue composition taken from (1.1,.2)2 appeared to be (1 serine, 10 proline, 25 glycine, 4 alanine, and 24 valine) or approximately 64 residues. Because of its unique composition of high val- ine, low alanine, and one serine, few known sequences of tropoelastin could match it. However from residue 15 to residue 80 of tryptic peptide W3,4 G-50 (i.e. Tl) the composition is (1 serine, 12-13 proline, 23 glycine, 4 alanine, and 23 valine). This region contained the eleven pentapeptide repeating units of P G V G V. 10 (1.1, .2)1, because it was purer than (1.1,.2)2, was then sequenced for 59 residues. As was expected, it did turn out to be the pentapeptide repeat portion of Tl (Table VI). mole to mole ratio of (1.1,.2)1 plus (1.1,.2)2 to The 69 Figure 27. Fraction (1.1,.2) recycled on Sephadex G-7S. The sample was recycled over the column two times and then isolated into two pooled fractions. assayed for by OPT. Protein was 70 o I' I I o o M N 71 tropoelastin turned out to be very similar to that for (4.1, .2)5.1 and (4.1,.2)8, as seen in Table VI. Discussion Thermolysin, a metallo-endopeptidase, was used to digest porcine tropoelastin in hope of obtaining overlap sequence information to thrombic and tryptic peptide fragmenta of tropoelastin. Under the conditions employed, extensive digestion of the tropoelastin occurred and resuIted in a polydispersed group of peptides. The largest peptides, around 60 residues, were nonpolar, while most of the smaller peptides, 2 to 15 residues in length, were polar in amino acid composition. These results can be best interpreted in light of Urry's beta spiral configuration for tropoelastin. The large nonpolar peptides were virtually identical in amino acid composition and sequence to a portion of Tl which contains the pentapeptide repeat P G V G V. This repeat has been shown to fit into the beta 9piral. 26 It was therefore possible that the spiral configuration confirred on the tropoelastin a secondary structure "unsuitable" for thermolysin digestion. It is known that proline within one amino acid, carboxyl terminal, to the site of cleavage inhibits the ability of thermolysin to cleave a peptide bond that it would normally cleave. This may be due to the "kink" that proline places in a polypeptide 72 chain. However, thermolysin will cleave a peptide chain with a proline one amino acid amino terminal or two to three amino acids carboxyl terminal to the cleavage site. 14 ,17 Therefore, the pentapeptide should have been digested were it not for some other conformation than just the proline "kink". The beta spiral may be this conforma- tion. The fact that most of the smaller peptides were rich in lysine and alanine indicates that these may come from crosslink regions. A possible explanation for thermolysin activity in these regions may have been due to the polar lysine residues "opening up" the peptide chain to a polar solvent and allowing exposure to the enzyme. This may occur in vivo too, so that a(ter tropoelastin is synthe- ---- .. sized, the large hydrophobic regions aggregate together as an enzyme resistant mass or coacervate, while the lysine rich regions "open up" to the polar environment. It has been shown that tropoelastin coacervates (aggregates) under physiologic conditions and that electron dense material, similar to elastin, has been seen in intracellular vacuoles. 27 Therefore, it is quite possible that this aggregation occurs intrac~11u1ar1y. While this oc- curs, the lysines which have been brought into juxtaposition by the aggregation, can be oxidatively deaminated by lysy1 oxidase. This would be feasible since this part of the molecule would be open to enzymatic attack. Once 73 deamination has occurred, modified and unmodified lysines could rapidly condense to form desmosine as perceived by Thomas et al. 3 This could occur intra- or extra-cellular- ly, or both. Two other peptides were sequenced during this study. One, a peptide containing arginine, was shown to extend 3 residues amino terminally to T7a. was completely new and unique. The other, however, It contained two lysine residues separated by 2 proline residues. In collagen the presence of at least 2 pralines together is common. Because of the pralines, the peptide must be very similar to a rigid "knot" tied in a rope. The. lysines may be in- valved in some sort of binding or crosslinking function. However, the exact function of such a peptide in tropoelastin is not readily understood. Thermolysin has proven to be a very useful enzyme for studying the structure of tropoelastin. It has pro- vided not only sequence information, but also configuration information of the tropoelastin molecule. It therefore could be used further as a tool in investigating the structure of elastin. 74 References 1 A. I. Lansing, T. B. Rosenthal, M. Alex, and E. W. Dempsey, Anat. Rec., 114 (1952) 555. 2 S. M. Partridge, H. F. Davis, and G. S. Adair, Biochem. J., 61 (1955) 11. 3 J. Thomas, D. F. Elsden, and S. M. Partridge, Nature, 200 (1963) 651. 4 R. A. Anwar, and G. Oda, Biochim. Biophys. Acta, 133 (1967) 151. 5 E. J. Miller, S. R. Pinnell, G. R. Martin, and E. Schiffmann, Biochem. Biophys. Res. Commun., 26 (1967) 132. 6 S. R. Pinnell, and G. R. Martin, Proc. Natl. Acad. Sci. USA, 61 (1968) 708. 7 S. Keller, I. Mandl, S. Birken, and R. Canfield, Biochem. Biophys. Res. Commun., 70 (1976) 174. 8 G. E. Gerber, and R. A. Anwar, J. Bio1. Chem., 249 (1974) 5200. 9 J. A. Foster, L. Rubin, H. M. Kagan, C. Franzb1au, E. Bruenger, and L. B. Sandberg, J. Biol. Chem., 249 (1974) 6191. 10 L. B. Sandberg, C. T. Leach, J. G. Leslie, R. A. Torres, and V. L. Alvarez, Front. Matrix Biol., in the press. 11 A. R. Torres, V. L. Alvarez, L. B. Sandberg, and W. R. Gray, in L. B. Sandberg, W. R. Gray, and C. Franzb1au, Elastin and Elastic Tissue, Plenum Press, New York, 1977, p. 267. 12 W. R. Gray, L. B. Sandberg, and J. A. Foster, Nature, 246 (1973) 461. 13 J. A. Foster, R. Shapiro, P. Voynow, G. Crombie, B. Faris, and C. Franzb1au, Biochemistry, 14 (1975) 5343. 75 14 H. Matsubara, R. Sasaki, A. Singer, and T. H. Jukes, Arch. Biochem. Biophys., 115 (1966) 324. 15 T. Abe, K. Takahashi, and T. Ando, J. Biochem., 69 (1971) 363. 16 H. Matsubara, A. Singer, and R. M. Sasaki, Biochem. Biophys. Res. Commun., 34 (1969) 719. 17 H. Matsubara, Biochem. Biophys. Res. Commun., 24 (1966) 427. 18 H. Matsubara, A. Singer, R. Sasaki, and T. H. Jukes, Biochem. Biophys. Res. Commun., 21 (1965) 242. 19 W. J. Dreyer, and E. Bynum, in C. H. W. Hirs, Methods Enzymo1., Vol. 11, Academic Press, New York, 1967, p. 32. 20 A. R. Torres, V. L. Alvarez, and L. B. Sandberg, Biochim. Biophys. Acta, 434 (1976) 209. 21 Y. Ito, and R. L. Bowman, J. Chromatogr. Sci., 11 (1973) 284. 22 L. C. Craig, and T. P. King, Methods Biochem. Anal., 10 (1962) 201. 23 W. R. Gray, in C. H. W. Hirs, Methods Enzymo1., Vol. 11, Academic Press, New York, 1967, p. 121. 24 E. Appe11a, J. K. Inman, and G. C. Dubois, INSERM No.1, Montpe11ier, 1977, Elsevier North-Holland, Amsterdam, 1977, p. 121. ~. 25 R. A. Laursen, Eur. J. Biochem., 20 (1971) 89. 26 D. W. Urry, Perspect. BioI. Med. , Winter (1978) 265. 27 W. H. Fahrenbach, L. B. Sandberg, and E. G. Cleary, Anat. Rec. , 155 (1966) 563. PART II METABOLIC STUDIES: ELASTIN AND COLLAGEN METABOLISM IN THE YOUNG PIG AORTA IN VITRO Introduction Studies on the metabolism of elastin and collagen have been fairly recent in terms of looking at soluble precursors. Soluble precursors of collagen were identified as recently as the 1950's, whereas with elastin this did not occur until 1967. Since then much information about the formation of elastin and collagen has been obtained through cell and tissue culture work. Cell and tissue culture work on collagen has established several steps in the process leading to the formation of insoluble collagen. It has been shown that colla- gen is translated as a large monomer which immediately proceeds into the rough endoplasmic reticulum (RER) of fibroblasts. l The signal peptide is rapidly cleaved as this monomer entered the RER. 2 ,3 Within the RER, hydrox- ylation of prolyl residues occurs 4 , glycosylation 5 , and helix formation among three collagen monomers to make a collagen molecule 6 . This molecule proceeds to the Golgi apparatus where it is further glycosylated and packaged 77 more tight1y.7 From there it is transported to the exterior of the cell. Microtubu1es appear to be involved in this process. S The collagen molecule is further trimmed at the amino and carboxyl terminal ends during transport to an extracellular site. 9 Once the soluble collagen has been posi- tioned at an extracellular site, cross1inking occurs and eventually soluble collagen becomes insoluble collagen. The term, soluble, in collagen chemistry means any collagen soluble in a neutral salt extract, a low pH citrate buffer (acid soluble), or guanidine chloride at neutral pH. In- soluble collagen is any collagen left after extraction with these solvents. lO ,ll It has been shown clearly that solubility is directly related to intermolecular crosslinking; therefore, insoluble collagen has the highest degree of crosslinking.l 1 Elastin formation is not as clearly defined as collagen formation. However, in the last two or three years much light has been shed on this subject. The first solu- b1e precursor of elastin to be isolated was tropoelastin. l2 It was determined to be a precursor of elastin by several criteria: 1) it had an amino acid composition identical to elastin except for lysiae and lysine derived cross1inks (desmosine and isodesmosine)l2; 2) in pulse/chase experiments using l4 C lysine labeled tropoelastin, the radioactivity was chased into desmosinel 3 ; 3) antibody directed 78 against insoluble elastin crossreacted with tropoelastin 14 ; and 4) in disease states like copper deficiency, there is a buildup of tropoelastin in the aorta along with a concomitant decrease in elastin 15 . Since that time much effort has been spent investigating the possibility that there is a larger primary precursor than tropoelastin. One lab identified in vitro, a molecule of 130,000 daltons 16 which was: 1) collagenase resistant, 2) incorporated significant amounts of 3 H valine, and 3) was pulse/chased into a 70,000 dalton protein. It also appeared to contain histidine which was not found in the tropoelastin "proper". This study, however, did not provide the supporting evidence needed to prove their isolation product was indeed a larger precursor. Several other labs also investigated approaches to finding a larger precursor. Burnett et al. 17 isolated mRNA from chick aortas and found the primary cell free trans lation product to be around 70,000 daltons. It incorporated large amounts of 3H glycine or 3 H valine, but not methionine. It was not collagenase sensitive. At the same time another lab isolated polysomes from chick aorta and found the primary translation product to be approximately 70,000 daltons. Several valuable observations, however, have resulted from these precursor studies. For example, tropoelastin was synthesized and isolated in cell l3 and tissue 18 79 culture. BAPN will inhibit its crosslinking into insoluble elastin. 19 The peak time for synthesis of tropoelastin appears to be within the first week after birth in the chick. 20 Therefore, cell and tissue culture appear to be good tools for analyzing the quantity of tropoelastin present in different regions of the seven day old pig aorta, and its subsequent incorporation into the insoluble elastin pellet. Most mammals have an elastin to collagen ratio of 2:1 in the upper thoracic region of the aorta, and the inverse is true in the abdomen. 2l Therefore, is there an increase of tropoelastin synthesis in the upper thoracic aorta as compared to the abdominal region? soluble collagen? Is the inverse true for If tropoelastin is efficiently utilized in insoluble elastin formation, there should be a corres- ponding gradient of synthesis that reflects the accumulation of elastin seen in the insoluble pellet. should be true for collagen. The same Thus it was decided to study the precursor/product reiationships of elastin and collagen in vitro. Methods Experiment 1 One young pig (seven days old) was killed by exsanguination. The aorta was marked at the diaphragm with surgi- cal thread in situ, and was removed from the branch of the 80 innominate artery to the iliac bifurcation. 100 mg (wet tissue) samples were taken from the upper thoracic, lower thoracic, and abdominal regions of the aorta. Each was finely minced and added to 1.0 ml aMEM (alpha modified Earl's medium) without leucine. 12 ~Ci of l4C leucine were added, and the tissue was incubated for 1.0 hour at 37°C in air. After the incubation, protein synthesis was stopped by adding SDS (sodium dodecyl sulfate) to 1%, BME (B-mercaptoethanol) at 10 ~l/ml, and EDTA (ethylenedinitrilo- tetraacetic acid) at 9.3 mg/ml. The sample was placed on a boiling water bath for 5 minutes. Next it was placed in a Virtis 5 ml container (a Virtis "45" homogenizer was used) and 1.0 ml aMEM was added. ized until it appeared creamy. The sample was homogenThis was transferred to a 6 ml centrifuge tube along with two 1.0 ml washes of the homogenizer. hour. The sample was centrifuged at 100 KG for one The supernate was removed and the pellet was washed two times with 1.0 ml PBS (phosphate buffered saline). Supernate. The supernate, which totaled 6 mI, was dialyzed against 0.01 M sodium phosphate, pH 7.4 with 1% SDS. After exhaustive dialysis, aliquots were taken for measurement of radioactivity and PAGE (polyacrylamide gel electrophoresis). Autoradiography was performed on some gels. Pellet. The pellet was dried and then autoclaved for 5 hours at 220°C, 20 pSi, in 10 ml H20. The sample was 81 then filtered on a Millipore filter (3 MM Whatman filter paper). Samples of the filtrate were used for the measure- ment of radioactivity, and the rest was hydrolyzed in 6 N HCl at 1100C for 24 hours. The autoclave insoluble pellet was also hydrolyzed in 6 N HCl and aliquots were used for the measurement of radioactivity. Amino acid analyses were done on all the hydrolysates. Scintillation counting. 0.2 ml of each sample was added to 10 ml of Aquasol (New England Nuclear). The sam- ple was counted for 1 minute. ~ (Polyacrylamide ~ electrophoresis). nique used was described by Clark. 22 reagents were: The tech- Stock solutions of A) acrylamide solution - 31.08 g acryla- mide, 0.61 g DATD (N, N'-diallyltartardiamide), and H2 0 to a final volume of 100 ml; B) ammonium per sulfate solution (prepared fresh daily) - 30 mg (NH4)2S208 and H2 0 to a final volume of 2 ml; C) gel buffer solution - 0.2 M phosphate , pH 7.2, 0.2% SDS, 1 M urea; and D) sample buffer solution - 0.01 M phosphate, pH 7.2, 1% SDS, 0.5 M urea. The polymerizing solution consisted of 4.5 ml A, 10 ml C, 15 ~l TEMED (N,N,N',N'-tetramethylethylenediamine), 4.5 ml H20, and 1 ml B. this. Both tube and plate gels were made with The reservoir buffer was 0.1 M phosphate, pH 7.2, 0.1% SDS, and 0.5 M urea. Once the gel was polymerized and placed in the gel machine along with the reservoir buffer, the sample was added in sample buffer D. Prior to 82 adding it onto the gel, for 1 hour. the sample was heated at 55-60 o C Tube gels were electrophoresed at 7 ma/tube until the tracking dye was near the bottom of the gel. Dansylation of proteins. A sample of protein (0.1 mg) was dissolved in 50 Ul of 1% SDS; 50 Ul of N-ethylmorpholine was then added, and the sample was mixed well. 75 ul of DNS-Cl (dansyl chloride, 25 mg/ml in dimethylformamide) was added and the sample was incubated at either 37 0 C for 1 hour or 50°C for 15 minutes. 0.5 ml acetone (cold) was then added and the proteins were precipitated at -20 o C for 1 hour or more. Centrifugation was carried out on a clin- ical centrifuge and the supernate was poured off. The pellet was washed with 0.5 ml of 90% acetone (cold), centrifuged, and the supernate was poured off. The dansylated pellet was then dissolved in sample buffer. Autoradiography. man 3 MM paper. Gels were dried on a piece of What- The dried gel was placed for varying amounts of time, on x-ray film at -SOoC. Amino acid analyses. Experiment See methods section of Part I. l A seven day old pig was killed by exsanguination. The aorta was marked with surgical thread through the aorta in situ at the branch of the renal arteries, the diaphragm, and the level of every other thoracic vertebral disc from the region of the 2nd through the 12th. The aorta was removed and placed in 50 ml of aMEM with 10% fetal calf 83 serum (FCS). times. It was then sectioned in three areas three The areas were: 1) between the branch of the in- nominate artery and the seventh thoracic vertebral disc, 2) between the seventh thoracic vertebral disc and the diaphragm, and 3) from just below the diaphragm to just below the branch of the renal arteries. approximately 200 mg wet tissue. The sample sizes were These were minced to less than one cubic mm and were incubated 45 minutes at 37 0 C in a CO 2 (5%) incubator with aMEM plus 10% FCS. After incu- bation the medium was removed and the tissue mince was washed three times with PBS. aMEM without leucine was then added to the minced tissue along with 10.8 ~Ci l4C leucine and 65 ~g/ml BAPN. This was incubated at 37 0 C for 2 hours in a CO 2 incubator. The incubation was stopped by addi- tion of 9.3 mg/ml EDTA, 20 ~l BME, and 10 ~l DFP. The sample was then boiled for 5 minutes, after which it was chilled on ice until homogenization. Homogenization was accomplished using a Virtis homogenizer with a 5 ml container. The chilled sample was placed in the container along with 1.0 ml of 0.1 M pyridine acetate buffer, pH 7.5. Homogenization was then performed at a setting of 4 for 10 minutes. 1.0 ml of 1.5% NP-40 (v/v in pyridine acetate) was added, and the sample was vortexed momentarily. The homogenate was removed and the container was rinsed one time with 1.0 ml of pyridine acetate. It was added to the homogenate. The sample was 84 then centrifuged at 25,000 rpm for 1.5 hours at 4 o C. The supernate was removed and the pellet was washed twice with 5 ml each of pyridine acetate. Both washings and the su- pernate were added together (total volume, 12 ml). Both the pellet and the supernate were kept separately. Supernate. The supernates from each aortic section were dialyzed exhaustively against 0.1 M pyridine acetate with 0.3% BME at 4°C. Total radioactivity in each of the supernate samples was determined by taking an aliquot of each sample. PAGE was carried out on each of the samples similarly to that described in Experiment 1 except the gels were tris/glycine gradient gels of 6% to 12.5%. also had 6 M urea in them. Some gels After PAGE, each gel was sliced and radioactivity was measured. Pellet. The pellet from each aortic sample was dia- lyzed against 0.1 M pyridine acetate with 0.3% BME at 4 o C. For each sample, after dialysis the pellet was centrifuged for 1 hour at 10,000 rpm and the supernate was removed. The pellet was then extracted in 5 ml of 5 M guanidine chloride (GuCl), pH 7.5 for 72 hours, after which it was centrifuged at 10,000 rpm for 1 hour. Five ml more of GuCl were added. The sample was vortexed and allowed to stand overnight. It was again centrifuged and the super- nates were combined, and dialyzed. samples were evaluated on PAGE. (Some GuCl soluble The gels were sliced and radioactivity was measured in each slice.) The pellet was 85 then autoc1aved as in Experiment 1. The pellet and auto- clave supernate were separated after centrifugation. All samples (combined supernates, pellets, and autoc1aved supernates) were hydrolyzed for amino acid analyses and measurement of radioactivity. Gel slicing and measurement of radioactivity. Slab gels (or tube) were sliced in 1-2 mm thick sections and dissolved in 0.5 m1 2% periodic acid overnight. Five m1 of Formula 950A (New England Nuclear) was added, after which the samples were cooled down in the scintillation counter. Five minute counts were taken, and adjusted to counts per minute. Tris/g1ycine~. Tris/g1ycine gels were made up from the following reagents: 1) Solution B - 1.5 M tris, pH 8.8; 2) Solution D - 0.5 M tris, pH 6.8; 3) acry1amide 100 g acry1amide, 3.9 g DATD, and H2 0 to 330 m1; 4) sample buffer - 10% glycerol, 5% BME, 3% SDS, 0.0625 M tris, pH 6.8, and bromophenol blue (1:50 in H2 0); and 5) running buffer - 0.024 M tris, 0.37 M glycine, and 1% SDS. For a 6% gel 0.66 m1 of 10% SDS, 16.0 m1 solution B, 12.8 m1 acry1amide, and 34.2 m1 of H2 0 were added together and mixed. Then 0.32 m1 of 10% ammonium persulfate and 30 ~1 of TEMED were added to cause polymerization of the gels. A stacking gel was added just before electrophoresis. had the following composition: It 5.0 m1 solution D, 0.2 m1 10% SDS, 2.5 m1 acry1amide, 21.3 m1 H2 0, 20 ~1 TEMED, and 86 200 ~l 10% ammonium persulfate. allowed to polymerize. The stacking gel was then Gradient gels were made by mixing two concentrations of acrylamide per gel. Proteins were subjected to electrophoresis at 15 ma/gel through the stacking gel, and 25 ma/gel through the running gel. Experiment 1 This experiment was performed essentially the same as Experiment 2 except for the following alterations: 1) a l4C valine label was added at 8.3 ~Ci/section; 2) BAPN was not added to the incubation; 3) the GuCl and NP-40 extracts were filtered on an Amicon Diaflo PM-30 filter (molecules <500 MW are separated out) rather than dialyzed, and 4) the GuCl extracts were evaluated on slab PAGE and autoradiographed using the Bonner and Laskey technique 23 • This involved soaking the gel in DMSO (dimethyl sulfoxide) for 30 minutes, then in PPO/DMSO (2,5 diphenyloxazole at 20% w/v) for 3 hours, and then precipitating in water for I hour. The gel was dried and autoradiography was accom- plished at -SOoC with medical x-ray film. Experiment i Experiment 4 was significantly different from 2 and 3, so it will be described in detail. rapidly killed by exsanguination. Two 7 day old pigs were The entire operation was performed sterilely because the samples were to be incubated for 24 hours. Their aortas were removed after 87 marking them anatomically as described in Experiment 1. After removing the aortas, one sample of 300 mg wet tissue was taken from each region. divided in half (~150 or 12 samples total. mg). Each sample was minced and This resulted in 6 samples/pig Each sample was added to 5 ml of aMEM with 20% FCS and incubated for 45 minutes in a CO 2 incubator. Afterwards, the medium was removed and the tissue was washed with PBS. of aMEM, 75 ~g/ml It was then incubated in 9 ml glutamine, 75 ~l of Fe(N03)3·9H20 (0.72 mg/l), and 8.0 ~Ci/ml 3H valine for 24 hours. tion was stopped with 2.7 The incuba- aa-dipyridyl, 1.7 mM PMSF, ~M 10 mM N-ethylmaleimide and 270 ~g cycloheximide. The samples were frozen and thawed three times and then one sample at a time was homogenized on a Virtis homogenizer at a setting of 4 for 10 minutes. The homogenate was placed in a 36 ml polyallomer centrifuge tube, and the homogenizer was washed with 8.5 ml of H2 0. volume approximately 23 mI. with 30 ~g/ml This made the The sample was then treated RNase (Sigma R-5500) for 10 minutes, at 37 0 C in an H2 0 bath; after which, it was cooled down on ice. 7.5 ml of cold 20% TCA was added to make a final concentration of 5%, and the sample was allowed to stand at 4 0 C for 10-20 minutes. To finish filling each centrifuge tube to capacity, 5% TCA was added and the tube was then centrifuged at 10 KG for 5 minutes. 36 ml of 5% TCA two times. The pellet was washed with 5 M GuCl, pH 7.5 with 0.1 M 88 BME was then added and the tissue was extracted for 72 hours. This was centrifuged at 10,000 rpm for 5 minutes. The supernate was removed and the pellet was washed with 8 to 10 ml H2 0. combined. After centrifugation, the supernates were The pellet was autoclaved in 5 ml H2 0, 5 hours, 220 o C, at 20 psi. The autoclave supernate was removed and the pellet was washed with 5 ml H2 0 two times. The pellet and autoclave supernate were hydrolyzed and measured for radioactivity. The GuCl extract was dialyzed and prepared for scintillation counting by the addition of 200 plus 1.5 ml of NCS. dissolved. ~l of extract This was heated at SOoC until it was Then 14 ml of scintillator fluid was added (6 g/l PPO, 75 mg/l POPOP, and 1 liter toluene). The GuCl extract was counted in duplicate. Experiment 1 Experiment 5 was performed essentially the same as Experiment 4 until the TCA pellet was isolated; also a different label, 8.3 ~Ci/ml 3H proline per sample, was used instead of valine. The isolated TeA precipitate was auto- claved for 5 hours at 220°C at 20 psi in 5 ml H2 0. The supernate was removed and the pellet was washed twice with 10 ml hot H2 0. ized. The supernates were combined and lyophil- The specific activity of hydroxyproline and proline was determined as described below on both the supernate and pellet. 89 Separation £i 3H proline and 3 H hydroxyproline. Separation of proline and hydroxyproline was performed by the method described by Woessner. 24 ~tock reagents were: 1) hydroxyproline and proline mixture - 20 mg hydroxyproline and 10 mg proline in 20 m1 H2 0; 2) pyrophosphate buffer - 0.2 M sodium pyrophosphate adjusted to pH 8.0 with HC1; 3) chloramine T - 1.4 g chloramine T in 25 m1 H2 0; 4) sodium thiosulfate - 3.6 M in H2 0; 5) phenolphthalein 0.1% in ethanol; 6) tris buffer - 1 M tris (hydroxymethy1) aminomethane adjusted to pH 8.0 with HC1; 7) scintillation fluid - 15 g PPO, 59 mg POPOP, 1 liter toluene; and 8) controls: A50 - 5 mg hydrolyzed albumin, 0.5 mg hydroxypro- line, 5 x 10 6 dpm hydroxyproline or 3.6 x 10 6 dpm proline; A75 - (same as A50 but with 7.5 x 10 6 dpm hydroxyproline or 5.4 x 10 6 dpm proline); C50 - 5 mg hydrolyzed collagen, 0.5 mg hydroxyproline, 5 x 10 6 dpm hydroxyproline or 3.6 x 10 6 dpm hydroxyproline; C75 - (same as C50 except with 7.5 x 10 6 dpm hydroxyproline or 5.4 x 10 6 dpm proline). The final volume was approximately 1.1 m1 for each control. These were made up to 5 m1 with H2 0. 0.5 m1 of the appro- priate sample, control or unknown, was then added to a .large 60 ml screw cap test tube. To both the controls and unknown samples the following procedure was carried out in duplicate. Two m1 of the cold hydroxyproline/proline mix- ture was added. NaOH. The pH was adjusted to pH 8.0 with 1 N This volume was brought up to approximately 8.0 m1 90 Six ml pyrophosphate and 1.0 ml chloramine T were added and the mixture was incubated at room temperature for 20 minutes. Six ml of sodium thiosulfate (freshly made) was then added. Next, 1-4 drops of phenolphthalein were added and the color was adjusted to a faint pink with 1 N NaOH. Four ml of tris buffer was added and the mixture became murky. Approximately 15 g of NaCl was added to each tube along with 10 ml of toluene, after which it was shaken for 5 minutes. This was allowed to stand until it was separated into two phases. The light phase (toluene) con- taining modified proline was removed and 5 ml of it was used for measurement of radioactivity. Each tube contain- ing the heavy phase (H 2 0) was placed on a waterbath at 8S o C for 25 minutes after which each was cooled. Twelve ml of toluene was added and the tube was shaken for 5 minutes. The two phases were again allowed to separate. Ten ml of the light phase (toluene), containing modified hydroxyproline, was removed with a volumetric pipette and placed in a 20 ml screw cap test tube with 800 mg of silicic acid (Riorad). It was shaken gently and filtered into a 25 ml test tube. Five ml of the sample was removed with a volumetric pipette and placed in a scintillation vial along with 10 ml of scintillator fluid. The assay of the stock solutions of 3H hydroxyproline and 3 H proline was done by addition of 100 ~l of each stock to 6 ml methyl cellusolve, and 10 ml toluene scintillator 91 fluid. DNA assay (diphenylamine) of aorta. DNA was assayed using a modification of Schneider's diphenylamine assay.25 For the seven day old pig aorta, the assay was as follows: one aorta was marked anatomically in situ and then removed. The adventitia was scraped off the media. Sections were taken of 100-150 mg wet tissue and their anatomical locations were noted. This involved 18 samples. Each section was gently blotted dry and frozen in liquid N2 • One at a time, the frozen sections were placed in a cold (-50 0 C) tissue press and pulverized. The tissue, yet frozen, was scraped into a cold test tube (pre-weighed) and placed on dry ice. When all the sections were pulverized, each was capped and allowed to warm to roo~.~emperature. Each sam- ple was then weighed, and 3.5 ml of cold 10% TCA was added. The sample was vortexed and centrifuged. The supernate was discarded and the pellet was washed once with TCA. Five ml of 95% ethanol was added after which the sample ~ was centrifuged. The supernate was discarded. extraction was repeated. The ethanol 2.6 ml of 5% TCA was added to the pellet and the sample was heated at 90 0 C for 15 minutes with stirring. kept. This was centrifuged and the supernate was The pellet was washed with 1.0 ml of warm 5% TCA, and the supernates were combined. A colormetric assay was then performed in triplicate using 1.0 ml of the sample in 2.0 ml of diphenylamine reagent (0.5 g diphenylamine, 50 ml 92 content. The pellets were then Lansing hydro1yzed 26 by adding 10 m1 0.1 N NaOH and heating the samples to 98 0 C for 50 minutes. Afterwards the pellets were washed two times with 10 m1 0.1 N NaOH, two times with 10 m1 H2 0, two times with 10 m1 ethanol/ether (3:1), and once with 10 m1 ether. There were then dried over P20S' The pellets were weighed to estimate elastin content. Hydroxyproline assay. The hydroxyproline content was determined by an automated method developed by Grant 27 using a Technicon system. The procedure used was identical to his, except that the PDAB (p-dimethy1aminobenza1dehyde) was purified by H2 0 precipitation before us~. This was done by adding 100 g of PDAB to 200-400 ml of n-propano1. Once it was dissolved, H2 0 was added until a white creamy precipitate formed. The precipitate was collected on a Buchner funnel and dried in an evacuated vacuum dessicator with P 2 0 5 , This purified PDAB was used in the assay, Duplicate standards of 1, 2, 6. and 10 ~g were evaluated with each set of samples. hydroxypro1ine/m1 Quench curves were developed by adding a known amount of standard hydroxyproline to various concentrations of each chemical and then evaluating it on the analyzer, Results Experiment 1 was a pilot study to evaluate overall protein synthesis in different regions of the aorta. As 93 can be seen from Figure 1, most of the 14C leucine radioactivity appeared in the dialyzed SDS extract. There was a significant increase in radioactivity in the upper thoracic region. When this sample was subjected to PAGE, most of the label migrated similarly as DNS-tropoe1astin (Figure 2). However, amino acid analyses indicated this band was a mixture of proteins (Table I). Incorporation of radio- activity in the autoclave supernate was less than in the SDS supernate. The elastin pellet (as determined by amino acid analyses in Table I) was lowest in radioactivity and contained no gradient of incorporation throughout the aorta. To understand and possibly confirm the results seen in Experiment 1, Experiment 2 was done using the same label, 14C leucine, but the incubation was increased to 2 hours. Also NP-40 was used instead of SDS to extract proteins, and a GuC1 extraction step was added. rather than three samples were used. Nine The supernate (NP-40) did not contain the majority of radioactivity this time (Figure 3) and no gradient was present. Autoradiography of a PAGE gel revealed a faint band where tropoelastin would migrate. measured. Gels were also sliced and radioactivity was Although the peaks were not predominate (Figure 4), the presence of radioactivity where collagen and tropoelastin migrate confirmed the above observation. Amino acid analyses revealed that the supernate was a mixture of 94 Figure 1. A diagram of the various protein fractions isolated during Experiment 1 and the respective radioactivity of l4C leucine incorporated into each. media (~lOO Aortic mg) from upper thorax (UT), lower thorax (LT), and abdomen (AD) were incubated with 12 ~Ci/sample of l4C leucine for 1 hour. Each section was then separated into SDS soluble ( . ) , autoclave soluble ( . ) , autoclave insoluble proteins (It;.) , and total radioactivity ( X ). 95 22 ~ \ \ \ 18 \ \ \ ~ 0 U) 14 \ U) H \ E-< \ E-< ~ :::: - \ 10 \ ~ ::.:: \ N 0 .-i :xl ~..,..; ........ ..,..; ::.:: p... ~~ 0 ~ ..,..;....-X 6 u .-l ..,..; 2 u / • -:j" .-i 0 DT SECTION V AD LT ~ ( 96 Figu~e 2. PAGE/autoradiography of the SDS soluble proteins present in the UT sample. SDS soluble protein was ex- tracted and run on PAGE as described in Methods. A dansyl tropoelastin standard (DNS-TE) was also run with the sample. The dye front is also indicated. UT TOP OF ~ GEL h \ .~~ BOTTOM OF ~ ...-. -- GEL j DNS TE TABLE I AMINO ACID ANALYSIS OF VARIOUS PROTEIN FRACTIONS OF AORTA FROM A 7 DAY OLD PIG EXPRESSED AS RESIDUES PER 1000 RESIDUES A 7 day old pig aorta was first homogenized and extracted with 1% SDS (Supernate) and The insoluble residue was used for elastin content (Pellet). autoc1aved (Autoclave). UT = upper thorax, LT = lower thorax, and AD = abdomen. Amino Acid UT Lys His Arg Hyp Asp Thr Ser G1u Pro G1y Ala Val Het lIe Leu Tyr Phe Des Ides 75 20 59 26 92 47 58 104 74 136 98 58 4 29 71 20 29 Supernate AD LT 64 19 62 43 76 34 46 74 91 195 107 56 2 25 62 15 28 54 10 55 48 93 32 38 78 94 221 110 51 23 56 13 24 Autoclave AD LT 32 6 63 84 53 21 34 70 116 319 115 24 3 13 27 5 14 31 6 58 93 49 16 35 65 122 321 121 26 2 12 26 3 14 UT Pellet LT AD 9 8 3 8 21 5 15 13 19 125 295 216 139 9 18 10 15 13 21 118 310 225 126 7 23 9 15 12 19 119 314 230 129 19 58 19 36 1 1 19 56 17 32 1 1 17 55 16 32 1 1 \0 co 99 Figure 3. A diagram of the various protein fractions isolated during Experiment 2 and the respective radioactivity of 14C leucine incorporated into each. media (~100 Aortic mg) from 8 samples throughout the aorta were incubated with 10.8 ~Ci 14C leucine/sample for 2 hours. Each sample was then separated into NP-40 soluble ( . ) , GuC1 soluble (0), autoclave soluble ( . ) , autoclave insoluble proteins (A), and total radioactivity ( X ). 100 18 16 14 12 10 8 6 4 2 a 987 6 5 SECTION 4 3 2 1 101 Figure 4. Graphs of each NP-40 soluble sample (1-9) from Experiment 2 run on PAGE which were sliced and counted. Dansylated salt soluble collagen (C) and tropoelastin (E) were run as standard markers. front (D). Also included is the dye (---) indicates offscale. E I 1) , , 1---1 3) I ~ J I I I I I Ii< 'U I H ....:I en ....:I Ii< ~ N 0 .-I 1\ 7) ~ 4) ~\ !\ >'l ~ t:4 8) U :::> Ii< ....:I U ..;t .-I 9) 5 10 15 20 25 GEL SLICE NUMBER \ 103 proteins, but primarily collagen (Table II). As can be seen in Figure 3, most of the radioactivity appeared in the GuCl extract in which little or no gradient was present. When these samples were evaluated on PAGE followed by autoradiography, a band migrated with DNS-tropoelastin. Also when gels were sliced and measured for radioactivity, the labeled proteins migrated similarly as DNS-tropoelastin and soluble DNS-collagen (Figure 5). not predominate. Again, the peaks were From amino acid analyses the GuCl frac- tion was primarily collagenous in content (Table II) with other protein contaminants. The autoclave supernate was lower in radioactivity than other fractions except for the elastin pellet, which was lowest. Also no gradient of radioactivity was seen in either of these fractions. Experiment 3 was done similarly as Experiment 2, except that l4C valine was used as a more specific label to examine the elastin. As seen in Figure 6, little radioac- tivity was seen in the supernate fraction. tract again contained the majority of The GuCl ex- radioact~vity, but now there was a distinct increasing gradient from the abdomen to the upper thorax. When these samples were evalu- ated on PAGE, and autoradiography performed, a band was present where tropoelastin would run (Figure 7). The autoclave supernate and elastin pellet were similar as before, lower in radioactivity with no gradient of incorporated ra~ioactivity present. TABLE II AMINO ACID ANALYSIS OF VARIOUS PROTEIN FRACTIONS OF AORTA FROM A 7 DAY OLD PIG EXPRESSED AS RESIDUES PER 1000 RESIDUES A 7 day old pig aorta was sectioned into 9 samples, homogenized and extracted with 1% NP-40 (Supernate) and then with 6 M GuCl pH 7.5 (GuCl Extract). Amino Acid 1 2 3 Lys His Arg Hyp Asp Thr Ser G1u Pro G1y Ala Val Met lIe Leu Tyr Phe 39 9 60 66 93 33 48 78 95 240 98 37 7 22 48 7 20 34 7 70 91 71 27 42 77 116 236 110 33 7 17 39 6 18 36 7 68 87 75 29 45 80 113 228 109 33 6 18 41 6 18 SUEernate 5 6 49 12 53 56 89 46 52 86 92 210 93 43 8 28 57 7 24 47 9 52 69 79 34 49 87 97 244 98 37 6 22 48 4 19 7 8 4 1 50 11 69 53 90 41 . '·53 95 85 196 94 44 8 28 52 9 22 59 13 64 54 71 28 49 80 88 223 104 41 7 27 60 11 23 43 10 17 63 96 37 50 73 92 225 90 40 7 24 52 10 21 27 3 79 102 69 20 42 63 105 317 107 23 10 24 10 GuCl Extract 6 7 9 37 6 64 76 86 27 43 92 103 259 89 33 7 17 39 6 17 32 9 52 68 109 30 55 115 66 244 93 36 52 18 55 45 21 66 28 71 95 100 291 104 36 9 21 45 16 14 4 30 2 60 81 83 27 51 78 106 274 103 29 5 14 36 5 16 ~ o .f;- 105 Figure 5. Graphs of each GuCl soluble sample (1-9) from Rxperiment 2 PAGE which were sliced and the radioactivity was measured. Dansylated salt soluble collagen (C) and tropoelastin eE) were run as standard markers. eluded is the dye front. Also in- (---) indicates offscale. 106 1) 6) 2) H ~ U H ..... U) 3) H 8) , "- ..... ~ 0 N 0 r-i :>< ::0:: ~ u ;:J ~ ..... ,..\ 4) \ I \ '.J \ 9) -\rll~ I I r\ v \ u ..;t r-i 5) 1 5 10 15 20 GEL SLICE NUMBER H 107 Figure 6. A diagram of various protein fractions isolated in Experiment 3 and the respective radioactivity of 14C valine incorporated into each. Aortic media (~l50 mg) from 9 samples throughout the aorta were incubated with 8.31 ~Ci l4C valine/sample for 2 hours. Each sample was then sepa ra ted in to NP-40 so 1uble ( . ) , GuC1 so 1 ub Ie ( 0 ) , autoclave soluble ( . ) , autoclave insoluble pro teins (A), and total radioactivity ( X ). 108 14 12 10 j;;cl !:l u:l u:l H 8 H H ~ ~ c.!l ::a: 6 -- N 0 r-I >:: ~ ~ U 4 ....:I < :> u ..-:t r-I 2 o 4 9 8 7 6 5 3 2 1 SECTION ~----~--~ 109 Figure 7. PAGE/autoradiography of the GuCl soluble pro- teins present in sample 6 (A), 3. 7 (B), and 8 (C) of Experiment Samples were run on PAGE as described in Methods. dansylated tropoelastin standard (DNS-TE) was also run with the sample. The dye front is also indicated. A rr A i 01' B c OF----? CEL ~ DNS TE 111 Experiment 4 was similar to Experiment 3 in that 3 H valine was the label, but now 2 pig aortas were used, and duplicate samples were used from each of the regions as described in Experiment 1. to 24 hours. The labeling time was increased This was done to see if the gradient of valine proteins seen in the GuCI extract of Experiment 3 could be chased into either the insoluble collagen fraction or the elastin fraction. in vitro similarly to ~ If elastin synthesis was occuring vivo, there should be a gradient of radioactivity increasing from the abdomen to the upper thorax in the elastin pellet in a prolonged incubation since elastin has been observed not to turnover. As seen in Figure 8, the elastin pellet had increased radioactivity but reflected little or no gradient. The GuCl extract still contained a significant amount of radioactivity but no gradient was present. was entirely different. The autoclave fraction however, There was significantly more ra- dioactivity present, and also an increasing gradient from the upper thorax to the abdomen. This was expected, and it appeared that while elastin production was altered, collagen production was not altered. Because of the results of Experiment 4, Experiment 5 was designed to determine whether collagen or elastin was responsible for the results in 4. For this reason 3 H pro- line was used, and radioactivity in proline and hydroxyproline was determined in the autoclave fraction after a 24 112 hour incubation with the label. As seen in the autoclave supernate, both 3 H proline and 3 H hydroxyproline were increased in the abdominal region (Figure 9a,b) which indicated that collagen was responsible for the gradient. In- corporation of both proline and hydroxyproline was significant in the elastin pellet; but again, little or no gradient was seen, which indicated altered elastin production. The elastin/collagen (E/C) ratio, and water content throughout the 7 day old pig aorta were also determined. The E/C content of the young pig aorta was similar to what has been observed for other mammals. It was 2:1 in the upper thorax and 1:2 in the abdomen as shown in Figure lOa. The water content was fairly constant throughout the aorta (Figure lOb). DNA content was also determined for the 7 day old pig aorta. As can be seen from Figure 11, the DNA content/wet tissue is relatively constant throughout the aorta. A DNA standard which showed the reliability of the assay is pictured in Figure l2. The reliability of the hy- droxyproline assay was shown by the results in Figure 13. The quenching effect of various solvents in which hydroxyproline was often present was evaluated as seen in Figure 14. Recause the water content and DNA content were fairly constant throughout the 7 day old pig aorta (Figures lOb, 11), radioactivity could be normalized to the weight of 113 Figure 8. A diagram of various protein fractions isolated in Experiment 4 and the respective radioactivity of 3H valine (~150 whic~ was incorporated into each. Aortic media mg) was isolated in duplicate from three different regions of the aorta in two different pigs (pig 1 pig 2 = ... ) • = • , The samples were incubated with 8.0 ~Ci 3 H valine/sample for 24 hours. They were then separated into the GuCl soluble (a), autoclave soluble (b), and autoclave insoluble proteins (c). 114 24 16 8 24 ~ ;:J CI.l CI.l H H 16 H ~ :::: C-' ::E: N 8 ., 0 ...-! ::< ::E: /l.o ~ ...:I < :> 24 ::z:: C"1 16 8 o UT LT AD ~======~Z:==========~ V SECTION 115 Figure 9. A diagram of various protein fractions isolated in Experiment 5 and the respective radioactivity of 3 H proline and hydroxyproline which was incorporated into each. Aortic media (~150 mg) was isolated in duplicate from three different regions of the aorta in two different pigs (pig 1 = ., pig 2 = A). The samples were incubated with 8.3 ~Ci 3 H proline/sample for 24 hours. then separated into the 3 They were H hydroxyproline autoclave solu- ble (a), 3 H proline autoclave soluble (b), and 3 H proline and hydroxyproline autoclave insoluble proteins (c). 116 A 6 8 lO,---------.----------r-----------------r---------- ~ C 8 ~ P-I >< ::r: ::r: - C"1 0 p::: P-I ::r: C"1 UT 0 V AD LT SECTION \ ( 117 Figure 10. Elastin/collagen and water content throughout the 7 day old pig aorta. a) Both elastin and collagen were related to mg/IOO mg dry fat free tissue (DFFT). Elastin ce) was determined gravimetrically, collagen ( . ) was determined by hydroxyproline analysis, and total protein C.) by adding the two together. b) Water content was determined by drying the tissue to constant weight under vacuum and P20S' Figure 11. DNA content throughout the 7 day old pig aorta. DNA was determined by the diphenylamine assay as described in Methods with triplicate measurement of the extract. 118 !:xl A 60 ;:::J en en H ~ ~ ~ ~ t.:l ::E:: - 40 20 0 0 .-l t.:l ::E:: I=>:J ;:::J en 8 tI) H ~ 160 ~ !:xl ::: t.:l ::E:: 0 0 .-l 120 80 • • . • • • • • p:: I=>:J ~ < ::: 40 t.:l ::E:: 3.0 !:xl ;:::J tI) tI) H ~ 2.0 ~ !:xl ~ t.:l ::E:: < z 1.0 ~ t.:l ';:::J V ----r-z--:~ SECTION 119 Figure 12. A standard DNA assay with diphenylamine. ples were assayed in duplicate. at 600 nm. Sam- Absorbance was measured 120 o o ('f") o N N o 00 .-l o o .-l o \0 o N o WN 009 a~NVH~OSHV 121 Figure 13. A standard hydroxyproline assay with PDAB. Absorbance was measured at 550 nm. 122 o N rl rl ,-., ....:l ~ '" 0 0 '-' ~ ~ :xl ,.... o rl N o 123 Figure 14. Quench curves of NaCl, HCl, and TCA on the hydroxyproline assay. Various concentrations of each compound had a standard amount of hydroxyproline mixed with it which was then reacted in the hydroxyproline assay. The samples were measured in duplicate. was measured at 550 nm. Absorbance 124 ·6 ·4 ·2 o •2 .4 •6 .8 NACL (M) .6 ~ z o .4 UI UI .2 . o .12 .48 1.0 2.0 ReL 3.0 3.6 (N) .6 .4 .2 o 8 16 TCA (%) 24 32 ' 125 the wet tissue or DNA content, with the same pattern resulting. Discussion In vitro experiments have indicated that the young mammal synthesizes large amounts of soluble e1astin 20 and co1lagen 28 . This is reflected in rapid accumulation of elastin and collagen in the developing aorta. 29 Even with- in the young animal inverse gradients of accumulation of elastin and collagen are seen (Figure lOa) throughout the aorta. tion, In this experiment, after 1 to 2 hours of incubather~ did appear to be a gradient of synthesis of soluble elastin-like protein that would correspond to the gradient seen upon dry weight analysis of the insoluble elastin. However, by 24 hours the early gradient of syn- thesis seen could not be related to incorporation into insoluble elastin. While there was increased radioactivity present in the elastin pellet, little or no gradient was seen. This definitely indicated that some sort of altered elastin metabolism had occurred in vitro. This is sup- ported by Keeley who showed that within 2-3 hours soluble elastin synthesis in chick aorta organ culture decreased. 30 This was not seen in the collagen fraction. By 24 hours the abdominal region had accumulated more collagen than the upper thoracic. This result correlated with the hy- droxyproline analysis of insoluble collagen. Therefore 126 collagen metabolism in vitro appeared to mimic in vivo results. It was therefore concluded from this study that elastin metabolism appeared altered in vitro, while collagen metabolism was not. Therefore, the precursor/product relationships in elastin metabolism need to be studied in vivo. 127 References 1 R. Harwood, M. E. Grant, and D. S. Jackson, Biochem. Biophys. Res. Commun., 59 (1974) 947. 2 G. R1obe1, and B. Dobberstein, J. Cell BioI., 67 (1975) 835. 3 R. D. Palmiter, J. M. Davidson, J. Gagnon, D. W. Rowe, and P. Bornstein, J. BioI. Chem., 254 (1979) 1433. 4 J. Uitto, and D. J. Prockop, Arch. Biochem. Biophys., 164 (1974) 210. 5 H. Anttinen, A. Oikarinen, L. Ryhanen, and K. I. Kivirikko, ~ Lett., 87 (1978) 222. 6 J. Uitto, and J. R. Lichtenstein, J. Invest. Dermato1., 66 (1976) 59. 7 M. Neutra, and C. P. Leblond, Sci. Am., 220 (1969) 100. 8 J. Uitto, H. Hoffmann, and D. J. Prockop, Arch. Biochem. Biophys.,173 (1976) 187. 9 J. H. Fessler, and L. I. Fessler, Annu. Rev. Biochem., 47 (1978) 129. 10 D. S. Jackson, and J. P. Bentley, J. Biophys. Biochem. Cyto1., 7 (1960) 37. 11 D. A. Hall, The Methodology of Connective Tissue Research, Joynson-Bruvvers Ltd,. Oxford, First Edition, 1976. 12 L. B. Sandberg, N. Weissman, and D. W. Smith, Biochemistry, 8 (1969) 2940. 13 A. S. Narayanan, L. B. Sandberg, R. Ross, and D. L. Layman, ~. Cell BioI., 68 (1976) 411. 14 R. A. Daynes, M. Thomas, V. L. Alvarez, and L. B. Sandberg, Connect. Tissue Res., 5 (1977) 75. 128 15 N. Weissman, G. S. Shields, and W. H. Carnes, J. BioI. Chem., 238 (1963) 3115. 16 J. A. Foster, R. P. Mecham, C. B. Rich, M. F. Cronin, A. Levine, M. Imberman, and L. L. Salcedo, J. BioI. Chem., 253 (1978) 2797. 17 W. Burnett, and J. Rosenbloom, Biochem. Biophys. Res. Commun., 86 (1979) 478. 18 D. W. Smith, and W. H. Carnes, J. BioI. Chem., 248 (19 73 ) 8157. 19 G. M. Bressan, and D. J. Prockop, Biochemistry, 16 (1977) 1406. 20 F. W. Keeley, Can. J. Biochem., 57 (1979) 1273. 21 E. G. Cleary, Doctoral Dissertation, University of Sydney, 1963. 22 C. C. Clark, in D. A. Hall, The Methodology of Connective Tissue Research, Joynson-Bruvvers Ltd., Oxford, First Edition, 1976, p. 205. 23 W. M. Bonner, and R. A. Laskey, Eur. J. Biochem., 46 (1974) 83. 24 J. F. Woessner Jr., in D. A. Hall, The Methodology of Connective Tissue Research, Joynson-Bruvvers Ltd., Oxford, First Edition, 1976, p. 247. 25 w. 26 A. I. Lansing, T. B. Rosenthal, M. Alex, and E. W. Dempsey, Anat. Rec., 114 (1952) 555. 27 R. A. Grant, 28 J. Vuust, and K. A. Piez, J. Biol. Chem., 245 (1970) 6201. 29 T. Looker, and C. L. Berry, J. Anat., 113 (1972) 17. 30 F. W. Keeley, Connect. Tissue Res., 4 (1976) 193. C. Schneider, Methods Enzymol., 3 (1957) 680. ~. C1in. Pathol., 17 (1964) 685. PART III METABOLIC STUDIES: ELASTIN AND COLLAGEN METABOLISM IN THE YOUNG RAT AORTA IN VIVO Introduction In vivo studies have produced much valuable information on the metabolism of elastin and collagen. Walford l injected young rats intraperitoneally (IP) with 3 H glycine and then isolated elastin at various ages after injection. During the growth of the animal, the specific activity of elastin went down proportionately to the increase in body weight; however, in older animals the specific activity leveled off indicating little turnover of elastin. also found little turnover of elastin in rats. 2 appear that collagen gradually turned over. Kao It did Sodek has shown that in adult rats collagen isolated from different tissues, i.e. skin and peridontal tissues, turned over at different rates. 3 Other investigators, Looker 4 and Gerrity5, have both shown that there was a rapid phase of elastin and co1lagen accumulation in the rat aorta from just after birth to about twelve weeks of age. A much slower phase of accum- ulation of both proteins was seen at ages older than twelve 130 weeks. Therefore to study elastin formation, the young rat was most suitable. Ross6 and Gerrity5 have both done experiments in which they injected young rats IP with 3H proline and then isolated and sectioned the aorta from animals at various times after injection for examination by electron microscopy and autoradiography. Both studies gave similar re- sults with the label appearing in the cell at approximately 15 minutes, the extracellular space by 30 minutes, and the extracellular elastin or collagen fiber by 3 hours. Finally some authors have analyzed, in vivo, various disease states which involve the aorta. Increased choles- tero1 7 , hypertension 8 , and estrogen 9 all directly stimulate the accumulation of collagen, but also there was an increase in elastin. Fisher found that there were differ- ences in the effect of cholesterol on the accumulation of elastin and collagen at various levels of the aorta. 10 Much has been done looking at the accumulation of elastin and collagen. Often it was incorrectly labeled as synthesis rather than accumulation. Increased accumu- lation does not always imply increased synthesis, but may reflect increased utilization of the precursor. Few ex- periments in vitro, and only one in viv0 2 , have been reported which analyze synthesis and accumulation of elastin and collagen in the aorta. The experiments below were designed to measure known precursors of elastin and 131 collagen, and their subsequent conversion into the insoluble forms in vivo in the young growing rat. Methods Four week old ~ experiment 1 Four week old rats were injected with 9 weight with 3 H glycine IP. ~Ci/g body Four rats were injected per group, and the groups were sacrificed at 30 minutes, 1 hour, 2 hours, 4 hours, 10 hours, and 24 hours after injection of isotope. Rats were killed by guillotine. Once the animals were decapitated they were rapidly dissected and the aortas were sectioned into three regions: 1) abdomen - from the bifurcation to the diaphragm, 2) lower thorax - from the diaphragm to the seventh verte- bral disc, and 3) upper thorax - from the seventh vertebral disc to the branch of the innominate artery. The sections were removed and freed of adventitia by blunt dissection. They were frozen on dry ice until further processing. Homogenization. To 0.81 m1 of a 0.14 M NaCl cock- tail, pH 7.2, with 0.1 mg/m1 BAPN (S-aminopropionitri1e), 1.0 ~M PMSF (pheny1methy1su1phony1f1uoride), and 0.1 mM BME CS-mercaptoethano1), was added 1 mg gammag1obu1in II (yGB-II, Sigma BG-II) and 0.5 mg porcine tropoe1astin, along with the appropriate sample. The mixture was homo- genized in a glass conical test tube with a ground glass 132 pestel driven by a slip clutch motor at 4 0 C until a fine particulate matter was obtained. The sample was trans- ferred to a 1.5 ml microfuge tube. The homogenizer was washed with 0.4 ml of the NaCl cocktail, and this wash too was added to the microfuge tube. The homogenate was then placed on a boiling water bath for 3 minutes, after which it was sampled for DNA (50 ~l or 75 ~l) while stirring. The DNA sample was placed in 1 ml cold 10% TCA (trichloroacetic acid) until it was assayed. then stirred for 3 hours at 4 o C. The homogenate was It was centrifuged on a Beckman Microfuge B for 4 minutes, and the supernate and the pellet were separated. The pellet was washed with 0.2 ml of NaCl cocktail, and the supernate, after centrifugation, was combined with the previous supernate. Supernate. To the supernate was added cold 100% TCA until a 7.7% final concentration of TCA was attained. After thorough mixing, the supernate was allowed to stand overnight at 4 o C. It was then centrifuged, and the TCA soluble fraction was removed. The TCA insoluble pellet was then washed with 1.0 ml of cold 5% TCA. The washes were combined, and an aliquot was taken for measurement of radioactivity in Biofluor (New England Nuclear). The TCA insoluble pellet was then solubilized in 1.2 ml of collagenase buffer (0.05 M tris, 0.01 M calcium acetate, pH 7.2, 10 mM NEM (n-ethylma1eimide), and 0.1 mM PMSF), and divided into three portions. Two portions 133 .received collagenase and one served as a blank. portions received 5 ~l The two or 15 units of collagenase (Ad- vanced Biofactors Type III). All samples were incubated at 37 0 C for 90 minutes with stirring. stopped by the addition of 50 ~l The reaction was of 100% TCA, and the samples were cooled for 2 hours at 4 o C. The samples were centrifuged and the supernate was removed. The pellets were washed with 1.0 ml cold 5% TCA and the supernates were combined. Aliquots of the supernates were used for measurement of radioactivity and hydroxyproline analysis. The pellet was then dissolved in 200 ~l of 0.5 M ammonium formate, pH 5.2 at 4 0 C (the blank was not continued). 350 ~l of n-propanol was added dropwise followed by addition of 500 ~l of n-butanol with rapid stirring. The sample was then centrifuged and the supernate was removed. Measurement of radioactivity and PAGE were carried out on the supernate. The pellet was stirred to a homogenous mixture in 0.5 ml of ammonium formate. PAGE and measure- ment of radioactivity were carried out on the resolubilized pellet. Pellet. The NaCl insoluble pellets were treated with collagenase similarly as the supernates were, except that 60 units of enzyme were added and the incubation was 24 hours. After collagenase treatment the pellets were ex- tracted with 1.0 ml 8 M urea/O.l M BME for 24 hours. The sample was then centrifuged, the supernate was removed, 134 and the pellet was washed with 1.0 ml H2 0. The extracts were combined and aliquots were taken for measurement of radioactivity and hydroxyproline analysiS (see Methods of Part II). The insoluble pellet was then hydrolyzed in 6 N HCl and aliquots used for measurement of radioactivity, amino acid analysis, and hydroxyproline analysis. Qli! assay ~ DABA (diaminobenzoic acid). The samples of homogenized tissue that were set apart for DNA analysis in 10% TCA were collected on a Whatman GF/C filter (2.4 cm) by suction using a M~llipore filter holder (XX 10 025 14). The filters were washed with 10% TCA, and then with 95% ethanol five times (2-4 ml!time), suctioning slowly each time. Finally, ether was passed through 2 or 3 times. The filters were placed in clean scintillation vials. The samples were kept dry until assayed. of To assay, 250 ~l DAB A reagent (0.4 g purified DABA/ml H2 0) was added to a 11 sample. The DABA was purified according to Foster. All samples were heated at 58-60°C for 40 minutes. After heating, the samples were cooled and 4 ml of 1 N HCl was added. Next, the samples were vortexed well and allowed to mix for 15 minutes. They were filtered over a coarse filter to collect fibers of the GF/C filter. The resul- tant solute was measured for fluorescence at 510 nm emission (410 nm excitation). analyzed in duplicate. Standards of 1-10 ~g DNA were Usually one to two time points were assayed at a given time. All samples were done in 135 either duplicate or triplicate. PAGE. Tris gels were prepared and electrophoresed as described in Methods of Part II. Samples and standards were dansylated as described in Methods of Part II. Inhibitors of the collagenase assay. Reaction sam- ples were made up of 5 mg of bovine tendon collagen in 1 ml of collagenase buffer (with no inhibitors). To 3 sam- ples was added BAPN at 0.01, 0.1, and 1.0 mg/ml. To another 3 samples PM SF was added at 0.1, 1.0, and 10.0 ~M/ml. DTT (dithiothreitol) was added at 1.0, 10.0, and 100.0 mg/ml to a final three samples. Collagenase, 12.5 units, was added and the samples were incubated at 37 0 C for 22 hours, after which 100 added. ~l of cold 100% TCA was The samples were cooled to 4 0 C for 2 hours. The samples were centrifuged, and ninhydrin analyses (see Methods of Part I) were performed on the supernates. A zero time control with no inhibitors was performed, as well as blanks with and without inhibitors. A sample with enzyme and no inhibitors was the final control. Control assay of the collagenase digest. 19.4 mg wet tissue of the abdominal parts of the aortas of 4 week old Wistar rats was homogenized in the NaCl cocktail, and placed in a boiling water bath 3 minutes. supernate was removed from the pellet. The resulting To the pellet was added 1.9 ml of collagenase buffer (0.025 M tris and 0.33 M calcium acetate, pH 7.2). While stirring, it was 136 sampled 8 times at 125 ~l per time. To consecutive samples was added no enzyme, enzyme for a zero time point, 4 vI (10 units) enzyme, 10 VI, 20 vI, 40 vI, 100 VI, and 200 VI. The final volume with the addition of enzyme was 500 vI, the difference being made up by addition of collagenase buffer. The incubation was for 22 hours at 37 0 C, with stirring in a water bath (except for the zero time sample which was stopped immediately). with 10% TCA. The reaction was stopped The supernate was removed and both super- nate and pellet were hydrolyzed. Hydroxyproline analyses were performed on each. Two ~ old rat experiment 1 10 baby rats/time period were injected IP with 9 vCi/ g body weight with 3H glycine. Fractionation of proteins was similar to the four week old rat (experiment 1) except that the NaCl cocktail was modified to 0.45 M NaCl, 0.1 mg/ml BAPN, 0.1 mM PMSF, 10 mM NEM, 10 mM EDTA «ethylenedinitrilo)-tetraacetic acid), and 0.1 M tris, pH 7.2. salt extraction time was also reduced to 30 minutes. The No heating of the sample after homogenization was performed this time. 95% ethanol was used to wash the TCA precipi- tated pellets instead of 5% TCA. Control experiment solubilization E! ~ collagen. the effect of homogenization on One adult rat aorta was homo- genized in the standard NaCl cocktail and the homogenate was allowed to stir overnight. The conditions of the 137 homogenization were: 1) mincing with scissors, 2) rough homogenization with a ground glass pestel homogenizer, and 3) fine homogenization with a ground glass pestle homogenizer. The homogenate was centrifuged, and the supernate and pellet were separated. Each supernate was precipitated with TCA at 10% TCA concentration. After centrifugation, the supernate and pellet were separated and hydroxyproline analyses were performed on all samples. Control experiment on the effect of 3 minute heating on boiling water bath ££ solubilizing collagen in homogenized aortic samples. in the 0~14 One adult rat aorta was homogenized M NaCl cocktail. tions, I and 2. It was divided into two frac- Fraction I was centrifuged immediately and hydroxyproline determination was performed on both the supernate and pellet. Fraction 2 was heated for 3 minutes and then the supernate and pellet were separated after centrifugation. Hydroxyproline determinations were per- formed on both. l4C collagen/collagenase assay. soluble collagen (1 mg/ml, 1 mg in 4 tubes, blank. with 5 20,000 cpm) were placed two for collagenase digestion and two for a Then 100 ~l = 20 ~l samples of salt ~l of collagenase buffer was added along of collagenase to the appropriate tubes. The incubation period was 30 minutes at 37°C with stirring. To stop the reaction, 100 ~l of RSA (bovine serum albumin at 5 mg/ml of collagenase buffer) along with 20 ~l of cold 138 100% TCA were added to the samples. They were placed on ice 15 minutes, then centrifuged, and the supernates were used for measurement of radioactivity. Zero time, blank, and total radioactivity samples were also evaluated. EDTA inhibition of collagenase. The same l4C colla- gen/collagenase assay as above was used with the inclusion of two samples which contained collagenase as well as 10 ~l of 10 mM EDTA. Effect £i collagenase ~ BSA. One mg of BSA in 0.5 ml collagenase buffer was reacted with 12.5 units (5 ~l) of collagenase for 0 minutes, 15 minutes, 35 minutes, 1 hour, 2 hours, 4 hours, 18 hours, and 23.5 hours at 37 o C. Also assayed were blanks without collagenase for the same time periods. Then 50 ~l of cold 100% TCA was added, and each sample was placed on ice 15 minutes. fuged, They were centri- and the ninhydrin assay was performed on the super- nate. The effect of collagenase on porcine tropoelastin. The standard collagenase assay was performed except that 3 H tropoelastin was added instead of14C collagen. As a control, an elastase digest of the 3H tropoelastin was performed concurrently. To 100 ~l of elastase buffer (0.2 M tris, pH 8.8) was added 20 ~l 3 H tropoelastin and 5 ~l elastase (0.5 units). also performed. Blanks and a zero time control were The incubation time was 2 hours at 37°C. After incubation, 20 ~l of 100% TCA was added and the 139 samples were cooled to 4°C for 15 minutes. The samples were centrifuged and the supernates were used for measurement of radioactivity in 10 ml Biofluor. Rate £i digestion of soluble collagen under standard conditions. ~l was added 20 To 100 14 salt soluble soluble collagen, and 10 ~l ~l collagenase ~ collagenase buffer C collagen, 100 collagenase. ~g citrate Samples were incubated at 37 0 C for 0 minutes, 15 minutes, 30 minutes, 1 hour, 2 hours, and 3.5 hours. 25 ~l of 100% TCA was added and samples were precipitated at 4 o C. The super- nates were measured for radioactivity. Rate £i digestion £i insoluble collagen ase under standard conditions. ~ collagen- 120 mg wet aortic tissue from an adult rat was homogenized in 2 ml collagenase buffer. It was centrifuged in a microfuge for 4 minutes at 8700 XG. The supernate was removed and the pellet was redissolved in 8 ml collagenase buffer. ~l pled for 500 It was then sam- 16 times while vigorously stirring. The 16 samples were 8 time points, one set for collagenase and one set for a blank. The time points were: 0 minutes, 30 minutes, 1 hour, 2 hours, 4 hours, 8 hours, 18 hours, and 24 hours. at 37°C, 50 ~l After incubating for the appropriate time of 100% TCA was added and the samples were cooled to 4°C for 15 minutes. They were then centrifuged, hydrolyzed as before, and hydroxyproline analysis was done on both the pellet and the supernate. 140 £! Determination the rat aorta. elastin/collagen content throughout Rat aortic media (10 mg) from different anatomical regions was homogenized in 0.6 ml of the 0.45 M NaCl cocktail. It was transferred to a microfuge tube and the homogenizer was washed with 0.6 ml more NaCl cocktail. This was added to the previous homogenate. Aliquots of 50 pI (3 times) were used for DNA analysis performed by the DABA assay. 100 ~l of 100% TCA was added to the sam- ples which were next placed in the cold for 15 minutes. They were centrifuged and the supernates were removed. The pellet was washed with ethanol, after which it was Lansing hydrolyzed in 1.0 ml 0.1 N NaOH in a screw cap test tube for 45 minutes in boiling water. The mixture was centrifuged, the supernate was removed, and the pellet was washed with 1.0 ml hot water which was then added to the first extract. The supernate (collagen) and pellet (elastin) were acid hydrolyzed and hydroxyproline analyses were performed on both. Histological observations ££ the week old ~ aorta. ~ ~ and four Rat aortas were sectioned, removed by sharp dissection and cleaned of adventitia as in previous experiments. They were fixed in 10% neutral formalin, sliced, and stained with EVG (Eosin Van Gieson). Two ~ old ~ experiment 1 In most ways this experiment was similar to the first two day old rat experiment. However the following changes 141 were implemented: 1) a given section was divided immedi- ately into 3 samples after homogenization and sampling for DNA; this way the pipetting errors of the insoluble pellet were avoided, 2) 5% TCA was used instead of 95% ethanol for washing TCA precipitated proteins, 3) DFP, 10 ~l/sam­ pIe (diisopropylfluorophosphate, 1 g/3.62 ml isopropanol), was included with the collagenase digest of the salt soluble proteins, and 4) the tris concentration of the collagenase buffer was increased to 0.5 M. Assay for DNA. ~ quantitative sampling ~ aortic homogenate 32 mg of wet rat aortic tissue was homogenized in 1.0 ml of 0.45 M NaCl cocktail. sampled for 25 ~l, 50 ~1, and 100 The homogenate was ~l in triplicate. The DABA assay was performed on each. Evaluation analysis. ~ various methods to extract DNA for DABA Potassium acetate/ethanol 12 and hot 5% TCA 13 have been used to improve DNA assays. of triplicate samples were set up: as previously described, Therefore, 3 sets 1) standard isolation 2) extraction of DNA by hot 5% TCA, and 3) purification of DNA by potassium acetate/ethanol. Sample 3 was processed as follows: aorta, ~70-120 mg wet weight, was homogenized in 1;0 ml of the 0.45 M NaCl cocktail. times at 50 ~l/time. microfuge tube. a whole adult rat The homogenate was sampled 3 Each sample was placed in a 1.5 ml Precipitation was accomplished by adding 142 1.0 ml 10% TCA. After centrifuging at 8700 XG for 4 min- utes, the supernate was removed and 1.0 ml 0.1 N potassium acetate in absolute ethanol was added to the pellet at 4 o C. This was stirred for 5-10 minutes, and then centrifuged. The supernate was removed, and the pellet was extracted at 60 0 C with 0.5 ml absolute ethanol for 15 minutes. The sample was again centrifuged, and the pellet was extracted with 0.5 ml absolute ethanol two times at room temperature. One ml of cold 10% TCA was added and the sample was collected on a GF/C filter under suction as before. The DABA assay was then performed on it. The procedure for hot TCA extraction on sample 2 was exactly the same as 3 until the 10% TCA precipitation step at the end. Instead, 200 ~l of 5% TCA was added and the sample was heated at 90 0 C for 15 minutes. It was then centrifuged, and the supernate was pipetted onto a GF/C filter with no suction. 1) 200 ~l of 5% TCA was lay- and 2) 200 ~l of 5% TCA plus 5 Two controls were run: ered onto a GF/C filter, DNA was layered onto a GF/C filter. ~g The DAB A assay was performed on all samples. Efficiency of GF/C filter to retain DNA. triplicate samples were set up: 1) 100 ~g Two sets of RNA plus 10 ~g DNA in 1.0 ml 10% TCA, filtered on a GF/C filter, and 2) 10 ~g DNA layered directly on a GF/C filter. for 1 consisted of filtering 100 ~g A control RNA on a GF/C filter, 143 and for 2 a blank filter, done in duplicate. The DABA assay was performed on all samples. Solubilities of various proteins in different solvents. 1 mg of protein was placed in a test tube along with 0.5 ml solvent. They were mixed, and let stand 15 minutes at 65 0 e (H), 22 0 C (R), or 4 0 C (e). They were then checked for solubility (S) or insolubility (I) by visual inspection. The proteins tested were: 1) yGB-II, 2) por- cine tropoelastin, 3) methylated casein, and 4) insulin. If the sample was TCA or ethanol treated, this was done first. After removing the TCA or ethanol, the solvent was added. Recovery of 125 1 tropoelastin. Approximately 20,000 cpm of 125 1 tropoelastin was added to aortic tissue. The tissue was homogenized and fractionated in duplicate as in the two day old rat (experiment 1). The experiment was repeated using 5% TeA instead of ethanol. DFP (10 ~l) added to the 0.5 M tris collagenase buffer digest. propanol/butanol extr~ction was The was increased from 15 minutes to 45 minutes. Assay of solvents for ~ ~ ~ extractant of tropo- elastin from the propanol/butanol insoluble protein pellet. Four solvents were used as extractants. There were: the 0.45 M NaCl cocktail, 2) ammonium formate, 1) pH 5.2, 3) 0.5 M acetic acid, pH 2.0, and 4) n-methylmorpholine: water (1:1), pH 10. 144 Each set of duplicate samples consisted of 200 ~l of yGB-II (5 mg/ml in H2 0), 25 ~l' l25 I tropoelastin, 50 III yGB-II/tropoelastin (10 mg yGB-II/5 mg tropoelastin in 1.0 ml HZO), and finally 300 III cold 10% TCA. was placed on ice for 15 minutes. This mixture Each sample was centri- fuged on a microfuge at 8700 XG for 4 minutes and the supernate was removed. cold 5% TCA. fugation. The pellet was washed with 1.0 ml The supernate was disposed of after centri- To the appropriate sample was added ZOO III ex- tract ant plus 10 III PMSF. The sample was stirred for 15 minutes at 4 o C; then, 350 jll of propanol was added, and the sample was stirred 15 minutes. 500 III of n-butanol was added, and this was stirred for 15 minutes. After centrifugation, the pellet and supernate were separated, and measured for radioactivity. The effect ~ time ~ the extraction of tropoelastin from TCA precipitated proteins Ez ammonium formate. Four sets of duplicate samples were assayed as in the previous experiment. pH 5.Z. The extractant was 0.5 M ammonium formate, Each of the four sets of samples was extracted for different amounts of time. B = If A = ammonium formate, prop·anol, and C = butanol, then the time was as fol- lows: 1) 15 minutes A, 15 mintues B, 15 minutes C; 2) 45 minutes A, 45 minutes B, 45 minutes C; 3) 2.Z5 hours A, 2.25 hours B, 2.25 hours C; and 4) 5 hours A, 5 hours B, 5 hours C. 145 Amino acid analyses of 3H glycine in various samples. Amino acid analysis was performed as previously described in Methods of Part I except that a program for the acidic separation only was implemented. The effluent line, after the spectrophotometer, was connected to a fraction collector, and aliquots of 4.6 ml/2.7 minutes were collected. One ml of the effluent in each tube under the glycine peak was then mixed with 10 ml of Biofluor, and measured for radioactivity. This was compared to the total radioactiv- ity put on the column. A control run with 3H glycine and 3H valine was performed to evaluate separation and recovery. The valine peak was isolated similarly as the glycine peak. The TCA supernates of the salt extracts of some experiments were performed (unhydrolyzed). Six week old ~ experiment Six week old rats, l 3 per group, were injected with 9 ~Ci/g body weight 3 H glycine and sacrificed at 30 minutes, 1 hour, 2 hours, 5 hours, and 10 hours after injection. The samples were processed as in the two day old rat (experiment 2). Quick extract of the ~ ~ old rat. Five 2 day old rats were injected IP with 9 ~Ci/g body weight 3H glycine. After one hour they were killed by decapitation. The chest cavity was opened and an inhibitor cocktail (0.45 M NaCl, 0.1 mM PMSF, 10.0 mM NEM, and 10.0 mM EDTA) was 146 poured over the aorta in ~. The aorta was sectioned into three pieces, removed, and frozen on dry ice. Imme- diately after, sections of all five aortas were pooled, and one sample at a time was homogenized. thawing the sample and adding 10 ately to the sample. This involved DFP and PMSF immedi- ~l Then 1.3 ml of 0.45 M NaCl cocktail was added, along with I mg yGB-II and 0.5 mg tropoelastin. The sample was homogenized as before at 4 o C, and was divided into three groups. Extraction was continued for 15 minutes with stirring, and then the samples were centrifuged. The supernate was removed, and the pellet was washed with 0.8 m1 NaCI cocktail. The supernates were combined and precipitated with 100 ~l of 100% TCA. They were allowed to coolon ice 20 minutes before centrifugation. Next, a collagenase digestion and propanol/butanol extraction were performed on the supernate samples as before except that 10 ~l DFP and PMSF were added to the ammonium formate. The salt insoluble pellet was hydrolyzed, and both the insoluble and soluble fractions were measured for radioactivity. Quick extract £i the six week old rat. old rats were killed and rapidly dissected. Four 6 week As soon as the chest cavities were opened, enzyme inhibitors were placed on the whole aortas in situ. removed they were frozen on dry ice. Once the aortas were After all 4 aortas 147 were collected, they were rapidly thawed in the presence of DFP, PMSF, and added to the NaC1 cocktail. Homogeniza- tion was performed and the supernate and pellet were separated. The supernate was dialyzed for 2 hours in the presence of NEM and PMSF, and then was lyophilized. was diluted in distilled water and used for: It 1) antibody assay labeled with 125 I Rolton-Hunter reagent, 2) separation on PAGE, and 3) extraction with propanol/butanol first, and then separation on PAGE. performed on some fractions. Autoradiography was Amino acid analyses were performed on various samples also. Antibody analysis for the presence of tropoe1astin in the 0.45 M NaC1 extract of the young rat. The antibody used was rabbit anti bovine ligamentum nuchae alpha e1astin. It was reconstituted from a lyophilized form by dil- ution with water. The antigen, bovine ligamentum nuchae alpha elastin, was produced by oxalic acid digestion as described by Partridge. 14 The fourth extract was taken for coupling to l25 I Rolton-Hunter reagent. 15 l25 I BH rat alpha elastin was also prepared, and l25 I tropoelastin was labeled by the peroxidase method 16 . The 0.45 M NaCl aortic extract of the six week old rat was also assayed for antigenic activity. Both antigens and antibody were diluted to the proper radioactivity and/or concentration. After antibody/antigen incubation, which was overnight at 4 0 C, most samples were precipitated by IgGsorb (Enzyme 148 Center; killed, whole Staphylococcus aureus). In one ex- periment goat anti IgG was the second precipitating antibody. All samples were done in duplicate. Results Figure 1 is a flow diagram of the techniques used in isolating soluble and insoluble elastin and collagen. However, before the actual fractionation of the radioactively labeled aortic medium, several control experiments were performed. DNA was assayed by the DABA assay as seen in Figure 2. It was both quantitative and reproducible. The DABA assay was sensitive by the methods used to less than 1 ~g DNA; whereas DNA analysis by diphenylamine, the more commonly used DNA reagent, was sensitive to about 10 ~g DNA. The use of the most sensitive and yet reliable assay was advantageous when working with 50-100 ~g DNA total/time point. The DABA assay was therefore chosen. Collagenase, by definition, is a specific enzyme class which will digest undenatured collagen. Table I shows the results of an experiment done to measure how many units of collagenase were needed to solubilize approximately the amount of aortic collagen that would be present under experimental conditons. As shown, even 10 units completely solubilized the collagen present within 24 hours. The hydroxyproline in the pellet was due 149 Figure 1. A flow diagram of the procedures used to iso- late soluble and insoluble elastin and collagen. 150 AORTIC MEDIA I HOMOGENIZED IN 0.14 M NaC1, pH 7.2 WITH INHIBITORS SAMPLE FOR DNA I SUPERNATE I PRECIPITATE WITH TCA UP TO 5% SUPERNATE ~(---RE S IDUE FREE AMINO I ACIDS, SMALL COLLAGENASE PEPTIDES DIGEST 90 MINUTES, 37°C, TCA PRECIPITATE SUPERNATE SALT SOLUBLE COLLAGEN ~(---RESIDUE I DISSOLVE IN AMMONIUM FORMATE, EXTRACT WITH PROPANOL/BUTANOL TROPOELASTIN 1 RESIDUE SOLUBLE INTRA- AND EXTRACELLULAR PROTEINS SUPERNATE~(--- I RESIDUE COLLAbENASE DIGEST 8 HOURS 37°C TCA PRECIPITATE RESIDUE----~)~~S~U~P~E~R~N~A~T~E I SALT EXTRACT WITH INSOLUBLE 8.0 M UREA COLLAGEN PLUS 2-MERCAPTOETHANOL 1 RESIDUE--~)~3S~U~P~E~R~N~A~T~E INSOLUBLE ELASTIN MICROFIBRILLAR PROTEINS 151 Figure 2. acid). A standard DNA assay using DABA (diaminobenzoic The assay is sensitive to 1.0 the average of 2 samples ± SD. ~g DNA. Values are 152 co o N N o 153 TABLE I CONTROL ASSAY OF THE COLLAGENASE DIGEST Samples of 2.4 mg of wet abdominal aortic tissue, which had already been salt extracted, were placed in 500 ~l of collagenase buffer (0.05 M Tris, pH 7.2, 10 mM calcium acetate) along with varying amounts of collagenase. The reaction was carried out at 37 0 C for 22 hours and stopped with 10% TCA. The pellet and supernate were separated, hydrolyzed, and hydroxyproline analyses were performed. S = supernate; P = pellet; 0, 4, 10, 40, 100, 200 = volume (~l) of enzyme added (10 ~l = 25 units). Sample ~g Hyp S4 ·'V10.6 P4 6.4 S40 'VlO.6 P40 6.4 SlOO 9.8 P100 6.1 SO 'V 0 llg Collagen 79.1 79.1 73.1 'V 0 10.8 80.6 S200 8.9 66.4 S blank 2.2 16.4 SlO P blank offscale 154 primarily to elastin, and was not affected by enzyme concentration. In all subsequent experiments, an excess of collagenase was used to assure complete solubilization of collagen. Of concern also was degradation of soluble elastin during the collagenase assay by nonspecific proteases. Tropoelastin has been shown to be very sensitive to proteolytic degradation. Therefore, it was desirable to add proteolytic inhibitors that would not inhibit the collagenase, but would inhibit other proteases. As seen in Table II, PMSF and BAPN did not inhibit the collagenase but DTT did. Therefore PMSF was included with collagenase diges- tions. After having concluded the preliminary control experiments, experiment I using 4 week old rats was performed. Figure 3 shows the results of the TCA soluble radioactivity present in free amino acids and small peptides. was surprising at first, This until the results of the soluble and insoluble collagen fractions were analyzed (Figure 4). It was clear that a precursor/product relationship could not be established and that the radioactivity was much too high in the soluble fraction at the wrong times. This was confirmed by hydroxyproline analyses of the fractions. Up to 50% of the total collagen hydroxyproline could be found in the soluble fraction. Most experimenters have shown there is approximately 2% of the total collagen 155 TABLE II EFFECT OF INHIBITORS ON THE COLLAGENASE ASSAY 5 mg of bovine tendon collagen was added to 1 rol of collagenase buffer (10 roM calcium acetate, 0.05 M Tris, pH 7.2). Three different enzyme inhibitors were added and the reaction allowed to proceed at 37°C for 24 hours. A sample of the supernate was then taken for ninhydrin analysis. All samples were performed in triplicate. + = offscale ninhydrin reaction. Values are the average of 3 samples ± SD. Absorbance 570 nm First Assay Second Assay Sample + + .19 ± .12 .14 ± .01 Blank + inhibitors .13 ± .03 .13 ± .02 Blank .11 ± .04 .10 ± .02 BAPN .01 mg/ml + + .10 + + 1. 00 + + + .40 ± .07 1.0 + + 10.0 + + .41 ± .04 .48 ± .07 10.0 .22 ± .08 .27 ± .02 100.0 .28 ± .03 .49 ± .04 Control o Time PMSF 0.1 llM/ml DTT '. 1.0 mg/ml 156 Figure 3. TCA soluble fraction of the salt extract of the 4 week old rat experiment 1. This fraction conta~ned marily free amino acids and small peptides. ax CA), lower thorax ce), and abdomen C.). the average of 2 samples ± SD. pri- Upper thorValues are 157 ~ ~ ~ ~ 0 ~ '-' 0 ~ ~ ~ H H o 158 Figure 4. Radioactivity present in soluble (a) and insol- uble (b) collagen in the 4 week old rat experiment 1. Both the salt soluble extract and insoluble pellet were treated with collagenase and then precipitated with cold TCA. ty. (.). The TCA supernate was then measured for radioactiviUpper thorax ( ... ), lower thorax ( . ) , and abdomen Values are the average of 2 samples ± SD. 159 32 24 r 16 < z -I 8 0 0 ::;J 0 0 ..... ""0 M ..... ~ ~ p.. 0 >-< 24 ....:l C!l ::r:: M 18 012 4 10 TIME (HOURS) 24 160 hydroxyproline in the soluble fraction. 3 It was surmised that the 3 minute heating carried out to denature proteolytic enzymes was also solubilizing the salt insoluble collagen. While precursor/product relationships could not be established with the collagen, it was clear that the abdom- inal region had far more radioactive collagen present than the thoracic regions. The propanol/butanol (P/B) pellet had an initial pulse of radioactivity which was removed rapidly (Figure Sa). This fraction probably contained soluble cellular and extracellular proteins. The urea fraction (Figure Sb), which contained little collagen as seen by hydroxyproline analysis, had a pulse of radioactivity at 1-2 hours and thereafter remained stable. This was thought to contain primarily microfibril- lar material. The elastin fractions should not have been degraded by heating as was the collagen. As seen in Figure 6b, there was rapid accumulation with the upper thoracic region value being the highest. There also appeared to be some turnover as indicated by the decrease at 24 hours. This was unexpected when compared to other data which indicated little turnover of mature elastin at these times. l ,2 Ii was also unlikely that it was due to contaminant proteins, since the abdomen was the most highly contaminated fraction but did not have the highest radioactivity (Table 161 Figure 5. Radioactivity present in the propanol/butanol pellet (a) and urea extract (b) of the 4 week old rat experiment 1. Upper thorax ( .. ), lower thorax ( . ) , and C.). Values are the average of 2 samples ± SD. abdomen 162 20 16 12 8 < Z 4 Q C!I ;:J 0 0 ,...; M 0 ,...; >< ~ Po. Q >< ...:I C!I M 20 ::t: 12 4 o 1 2 4 10 TIME (HOURS) 24 163 Figure 6. Radioactivity present in the propanol/butanol supernate (a) and elastin pellet (b) of the 4 week old rat experiment 1. Upper thorax ( . ) , lower thorax ( . ) , and abdomen ( . ) . Values are the average of 2 samples ± SD. 164 A <: z 6 ~ C,!) :::J 0 0 .-I ....... 4 ("") 0 .-I ~ ~ p.. ~ ;:.:. 2 ....:I C,!) =: ("") 14 <: z ~ C,!) :::J 10 0 0 .-I ....... --r 0 .-I 6 ~ ::E: p.. ~ ~ 2 ....:I C,!) ::I:l ("") 0 1 2 4 10 TIME (HOURS) 24 165 III). It was therefore not easily explained. The PIB supernate, which should contain primarily tropoelastin, did have an early pulse of radioactivity at 1 hour (Figure 6a), however the total radioactivity was far below that required to be a precursor to the elastin fraction. PAGE (Figure 7) showed that co~d carrier tropo- elastin was isolated, so that it was not clear whether little tropoelastin occurred freely in vivo or whether degradation had occurred. Because of several questions raised about the results of the 4 week old rat experiment 1, a number of control experiments were carried out. the solubilization of collagen. The first set dealt with By definition, salt sol- uble collagen is that fraction of collagen soluble in a neutral salt buffer. In Table IV the effect of homogeni- zation on solubilization was evaluated. Under the condi- tions used, it was apparent homogenization did not affect the amount of collagen solubilized. Heating the homogen- ate was next evaluated as shown in Table V. As can be seen, heating definitely affected the amount of collagen solubilized. from further Therefore, the 3 minute heating was omitted experimen~s. It became apparent that a quick evaluation of the success of digestion of collagen was necessary before the fractionation procedure proceeded further. For this, a l4C collagen control assay was performed as seen in Table 166 TABLE III AMINO ACID ANALYSIS OF FOUR WEEK RAT AORTIC ELASTIN EXPRESSED AS RESIDUES PER 1000 RESIDUES Analysis was performed on each section of aorta with a Beckman 121 Amino Acid Analyzer using a 2 column physiologic system with Beckman A-IS and PA-34 resins. The elastin was purified by sequential extractions with NaCl, collagenase, and urea as described in Methods. Amino Acid Lys His Arg Hyp Asp Thr Ser Glu Pro Gly Ala Val Met lIe Leu Tyr Phe Des b Ides Aortic Sesmenta Upper Lower Thoracic Thoracic Abdominal 19 0 20 14 24 27 25 37 97 294 194 95 3 32 73 24 21 0.6 0.3 22 0 22 15 27 29 28 42 95 285 185 94 4' 31 73 27 21 0.3 0.3 aSee Methods ~~r sectioning. Thomas et ale 29 a 26 15 37 34 35 54 91 255 166 93 4 34 77 25 24 0.6 0.4 Two Day Rat Elastin 9 1 10 13 8 16 15 21 98 331 238 96 25 68 24 19 1.4 0.3 bCalculated according to 167 Figure 7. PAGE of proteins isolated in the propanol/bu- tanol fractionation step. seen in Figure 1. The protein was isolated as Protein soluble in propanol/butanol (a), and protein precipitated in propanol/butanol (b). Fractions are 1) 5 2) 10 ~g ~g dansylated (DNS) tropoelastin, DNS tropoelastin, 3) DNS upper thoracic region proteins, 4) DNS lower thoracic region proteins, 5) DNS abdominal region proteins, and 6) DNS lower thoracic region proteins. ..c o 169 TABLE IV EFFECT OF HOMOGENIZATION ON THE SOLUBILIZATION OF COLLAGEN Adult rat aorta was homogenized at 4°C in 0.45 M NaCl cocktail 3 different ways: 1) mincing with scissors, 2) coarse grinding, and.3) fine grinding with a ground glass pestel homogenizer. The homogenate was centrifuged and the pellet (c) was separated from the supernate, which was then divided into TCA non-precipitable (b) and precipitable (a) fractions. Sample 1 2 3 ~g Hyp < 1 a b c 0 123 a < 1 b c 0 107 a b c < 1 0 123 170 TABLE V EFFECT OF HEATING ON THE SOLUBILITY OF AORTIC COLLAGEN An adult rat aorta was homogenized in 0.14 M NaCl cocktail. It was divided into 2 portions, 1 and 2. 1 was centrifuged and hydroxyproline determinations were done on the supernate (a) and pellet (b). 2 was heated on a boiling water bath 3 minutes and then treated as 1. Sample ~g Hyp 1 Total llg Hyp 142.2 a b 2.2 ± .02 140.0 ± 0 a b 47.3 ± 1.1 90.0 ± 0 2 137.3 171 VI. Evaluation of this control allowed rapid determination of whether collagenase digestion was achieved with each set of samples. Next it was desirable to have EDTA (10 mM) present in the initial homogenization buffer as an enzyme inhibitor of aortic collagenase. It was necessary to keep the collagen intact until the time of collagenase digest, especially the salt soluble collagen. However, since EDTA is a powerful inhibitor of collagenase, could enough of it be removed during the TCA wash to allow the collagenase digest to work? It was estimated that after washing with TCA the maximum EDTA concentration left behind would be a 0.7 mM concentration. In Table VII are shown the results of a collagenase digest in the presence of 0.7 mM EDTA. The assay worked without inhibition; therefore, EDTA was added to the enzyme inhibitors present in the homogenization buffer. The specificity of the collagenase used was evaluated against BSA (Table VIII) and porcine tropoelastin (Table IX). As can be seen, neither proteins were digested by the collagenase. It was also clear that soluble collagen may digest more rapidly than insoluble collagen and that the time required for both digestions might be different. Since it was desirable to keep the digestion time to a minimum, experiments were carried out to evaluate the rate of 172 TABLE VI COLLAGENASE ASSAY USING l4C COLLAGEN AS SUBSTRATE l4C collagen was incubated with collagenase at 37 0 C for 30 minutes. Cold TCA was added to give a 10% concentration and the samples were centrifuged. The supernates were then counted. Values are the average of 2 samples ± SD. Sample Collagenase o Time Blank Total CPM ± Background 208 ± 9 43 ± 21 6 ± 4 237 ± 33 TABLE VII EFFECT OF 0.7 mM EDTA ON COLLAGENASE ACTIVITY IN THE PRESENCE OF 10 mM CALCIUM ACETATE A typical collagenase assay as in Table VI was run along with one extra set of samples which contained 0.7 mM EDTA. Values are the average of 2 samples ± SD. Sample CPM ± Background Collagenase 227 ± 11 Collagenase + EDTA 261 ± 42 0 Time 26 ± 17 Blank 7 ± Total 226 ± 17 0 TABLE VIII EFFECT OF COLLAGENASE ON BOVINE SERUM ALBUMIN Collagenase was incubated with 1 mg albumin in 0.5 ml collagenase buffer for varying amounts of time after which 25 ~l aliquots were taken in duplicate for the ninhydrin assay. A st~ndard was also run by taking a1iquots of a stock of 1 mg Leu/ml water. Values are the average of 2 samples ± SD. BSA Sample o Absorbance at 570 nM Collagenase Blank .01 ± .01 minutes ± 0 o ~g Absorbance at 570 nM 0 0 2 .12 ± .01 0 0 5 .21 ± 0 1 hour 0 0 10 .41 ± .02 2 0 .03 ± .04 20 .79 ± .01 4 0 .06 ± .06 30 1.19 ± .01 18 0 .02 ± .03 23.5 0 .02 ± .01 15 0 35 ± 0 0 Leu Standard ..... ....... w 174 TABLE IX EFFECT OF COLLAGENASE AND ELASTASE ON DIGESTION OF 3H TROPOELASTIN (PORCINE) 3H tropoelastin was incubated with collagenase or elastase in the appropriate buffers for 30 minutes at 37°C. The samples were then TCA precipitated and the supernates counted. Digestion of l4C collagen by collagenase was done as a control. Values are the average of 2 samples ± SD. Sample CPM 3H Tropoe1astin 14C Collagen Collagenase 198 ± 13 o 200 ± 16 IHank 208 ± 12 Total 625 ± 101 Elastase 514 ± 139 0 Time 198 ± 45 Blank 209 ± 32 Total 767 ± 28 Time Collagenase Blank 216 ± 7 29 ± 4 175 digestion of soluble and insoluble collagen. Table X con- tains the results of an experiment in which soluble collagen was digested under approximate conditions of the aortic extract. There was rapid digestion by 1-2 hours. It was therefore decided to incubate the soluble fraction for 2 hours in further experiments. In Table XI, the results of an experiment evaluating the rate of digestion of insoluble collagen are shown. By 8 hours 80% of the hydroxy- proline was TeA soluble. Since most of the hydroxyproline present in the pellet was probably associated with elastin, the extent of collagen digestion was higher than 80% and therefore 8 hours was used as the incubation time in further experiments. The next experiment to be carried out was that using 2 day old rats. At this age, the rat is in a rapid growth phase as seen in Figure 8. Looker 4 has shown that body weight increases and aortic weight increases correspond during this growth phase. Therefore one could expect to see rapid accumulations of both elastin and collagen. The elastin and collagen composition in different regions of the aortic media were different in the 2 day old rat as seen in Figure 9a. Both elastin and collagen were in- creased per amount DNA in the upper thoracic as compared to the abdominal region of the 2 day old rat. the difference is more significant for elastin. However This is not seen in the 4 week old rat where reverse gradients of 176 . TABLE X RATE OF DIGESTION OF SOLUBLE COLLAGEN BY COLLAGENASE 15 ~g of l4C collagen (salt soluble) and 100 ~g of citrate soluble collagen were incubated with 10 units of collagenase for increasing amounts of time at 37 o C. After TCA precipitation the supernates were taken for measurement of radioactivity. Values are the average of 2 samples ± SD. Sample 0 *% Digestion CPM minutes 210 ± 10 30 ± 15 406 ± 54 84 ± 14 30 431 ± 10 91 ± 2 457 ± 15 98 ± 3 1 hour 3 2 477 ± 3 103 ± 1 3.5 473 ± 7 102 ± 2 Total 465 ± 37 Blank *% Digestion = 99 ± ((CPM - 8 background) + (Total CPM) ) x 100. TABLE XI RATE OF DIGESTION OF INSOLUBLE COLLAGEN BY COLLAGENASE 7.5 mg wet weight salt extracted rat aortic tissue was incubated with 63 units of collagenase for increasing amounts of time. After TCA precipitation hydroxyproline content was measured in both the supernate and the pellet. A control experiment was performed with no collagenase. Time o Collagenase }lg Hyp TCA Pellet TCA Supernate % Digestion Control 11 g H~:I~ TCA Supernate TCA Pellet 4 61 6 2 63 20 61 25 3 68 1 hour 30 50 38 3 71 2 48 42 53 3 64 4 64 23 74 4 58 8 72 18 80 4 61 18 76 15 84 4 53 24 78 10 89 4 55 minutes 30 I-' '-l '-l 178 Figure 8. age. Graph of body weight increases of rats with There are at least 6 specimens per time point. graph is semi-1ogrithmic. animals ± SD. Values are the av~rage The of 6-15 179 N C""\ co N ...:t N ~~ 0 N ~~ '""' en >-< < CI ........, ~ l::: t--8 -.0 1'004 r-I E4 '--8--1 t--e~ t--8----i '. co (SMa) LHDIlM xao~ 180 Figure 9. Elastin and collagen content in the aorta of the 2 day old Ca} and 4 week old rat (b), as related to DNA. Elastin (.2, and collagen scale is differe~t· for each age. of 3 samples ± SD. (tl). Note that the Values are the average 181 1 A o o r-I N o r-I 1 B o o r-I <"1 o r-I ~~---,.--~--~ ~ SECTION 182 elastin and collagen content are seen (Figure 9b). To be assured that only the media and endothelium were being isolated for analyses, histology was obtained on appropriate samples isolated by the normal procedure. As seen in Figures lOa,b, both the 2 day old rat and the 4 week old rat aortas were relatively free of extraneous matter. The results of the 2 day old rat experiment 1 are shown in Figures 11 through 14. As expected, the radioac- tivity present in the TCA supernate containing free amino acids and small pep tides decreased rapidly (Figure 11). Soluble collagen synthesis (Figure l2a) was increased in the abdominal region compared with the upper thoracic. However, measurement of subsequent incorporation into insoluble collagen (Figure l2b) was not reliable due to losses during pipetting of the insoluble fraction. This was corrected in the next experiment (two day old rat experiment 2). The P/B extract supernate revealed a pulse of radioactivity at 1 hour similar as that seen in experiment 1 (Figure l3a). The radioactivity was-again low; also no gradient of synthesis was seen as might have been expected. Subsequent incorporation into the elastin pellet (Figure l3b) could not be quantitated for the same reason as that given for the collagen fraction. However it was clear that more radioactivity had accumulated in the upper thoracic than abdominal region. The P/R pellet presented an interesting problem 183 Figure 10. rat (b). Histology on the 2 day old (a) and 4 week old The aortas were stained with Eosin-Van Gieson in cross-section. The magnification is 197x for the 2 day old, and 30x for the 4 week old rat. (UT), lower thorax (LT), and abdomen (AD). Upper thorax A B UT , ..... - LT AD 185 Figure 11. TCA soluble fraction of the salt extract of the 2 day old rat experiment 1. This fraction contained primarily free amino acids and small peptides. thorax CA), lower thorax ce), Upper and abdomen C.). are the average of 2 samples ± SD. Values 186 ~ '"' ~ ~ ~ 0 ~ '-' ~ ~ H ~ o 187 Figure 12. Radioactivity present in soluble (a) and in- soluble (b) collagen in the 2 day old rat experiment 1. Both the salt soluble extract and insoluble pellet were treated with collagenase and then precipitated with cold TCA. The TCA supernate was then measured for radioactiv- ity. Upper thorax (. ... ), lower thorax ( . ) , and abdomen (.). Values are the average of 2 samples ± SD. 188 20 A 16 12 8 < z Q ~ 4 c.!l ;:J 0 0 ...-I - <"'l 0 rl ~ ~ Po. Q >< ,..J 40 B c.!l :::t: <"'l 32 24 16 8 o 1 2 TIME 4 (HOURS) 6 189 Figure 13. Radioactivity present in the propanol/butanol supernate (a) and elastin pellet (b) of the 2 day old rat experimen t 1. Upper thorax ( . ) , lower thorax ( . ) , and abdomen ( . ) . Values are the average of 2 samples ± SD. 190 8 5 B 4 < z 0 C,!) :;:J - 3 0 0 M ~ 2 0 M X ::;:: ~ 1 0 >< t-l C,!) ::t:: M 0 1 2 TIME (HOURS) 4 6 191 (Figure l4a). It did not solubilize sufficiently in any solvent that was tried. Therefore, the 4 and 6 hour frac- tions were hydroly~ed in acid, and then measured for radio- activity. The results were different from what was ex- pected (compare with Figure 5a). Therefore it was decided tQ acid hydrolyze the plB pellet before measuring radioactivity in further experiments. The urea fraction (Figure l4b) was unreliable for the same reason given for the insoluble collagen. However it was clear that more radioactivity was present in the upper thoracic than the abdominal region. Amino acid analyses were performed on each fraction as shown in Table XII through XIV. The soluble collagen fraction was clearly collagenous with minor contamination, as was the insoluble collagen fraction (Table XII). The elastin fraction was quite free of collagen, as shown by hydroxyproline measurement, but it did contain a minor acidic protein contaminant (Table XIII). The plB super- nate fraction which contained all the tropoe1astin also contained a fair amount of contaminating proteins (Table XIII). The plB pellet was very similar to the yGB-II, and the urea extract did indeed contain primarily microfibrillar protein (Table XIV). Before the next rat experiment was done, several more control experiments were conducted. Because accurate eval- uation of the DNA content was essential, three more 192 Figure 14. Radioactivity present in the propanol/butanol pellet (a) and urea extract (b) of the 2 day old rat experiment 1. abdomen ( . ) . Upper thorax (i.), lower thorax ( . ) , and Values are the average of 2 samples ± SD. '. 193 20 16 12 8 A 16 8 o 1 2 TIME (HOURS) 4 6 TABLE XII AMINO ACID ANALYSIS OF 2 DAY RAT AORTIC COLLAGEN EXPRESSED AS RESIDUES PER 1000 RESIDUES Analysis was performed on each section of aorta with a Beckman 121 Amino Acid Analyzer using a 2 column physiologic system with Beckman A-15 and PA-35 resins. UT = upper thorax, LT = lower thorax, and AD = abdomen. Amino Acid Lys His Arg Hyp Asp Thr Ser G1u Pro G1y Ala Val Met lIe Leu Tyr Phe Soluble Collagen UT LT AD 26 28 0 0 ND ND 48 b 74 46 72 26 36 91 53 71 60 96 113 344 243 121 108 62 63 ND ND 16 19 70 57 10 11 18 18 27 0 ND 89 61 27 59 67 106 283 115 61 ND 18 56 4 25 ND ~ Not determined. aFrom Chung et a1.21 bLow value due to difficulty of measurement. Insoluble Collagen UT LT AD 66 0 33 ND 69 14 45 66 98 349 161 34 ND 12 24 37 3 29 101 42 14 30 53 113 311 155 45 ND 12 30 8 6 24 18 29 4 25 82 49 22 40 56 131 344 131 40 ND 12 28 0 10 Type III Co11age n a 30 6 45 126 42 13 37 70 117 347 93 13 13 22 2 8 I-' 0.0 .;:-. TABLE XIII AMINO ACID ANALYSIS OF 2 DAY RAT AORTIC INSOLUBLE ELASTIN AND PROPANOL/BUTANOL EXTRACTION SOLUBLE PROTEINS EXPRESSED AS RESIDUES PER 1000 RESIDUES Analysis was performed on each section of aorta with a Beckman 121 Amino Acid Analyzer using a 2 column physiologic system with Beckman A-15 and PA-35 resins. UT ; upper thorax, LT = lower thorax, and AD = abdomen. Amino Acid Lys His Arg Hyp Asp Thr Ser Glu Pro G1y Ala Val Met lIe Leu Tyr Phe Des Ides c Insoluble Elastin AD UT LT 11 17 13 6 5 7 26 32 19 18 17 14 18 23 23 24 25 25 2S 23 25 30 33 37 86 92 91 309 320 312 218 215 213 83 83 79 0 0 a 25 25 27 65 70 67 22 22 21 18 16 17 0.2 0.2 0.2 0.3 0.3 0.2 2 Day Rat Aortic E1astin a 9 1 10 13 8 16 15 21 98 331 238 96 a 25 68 24 19 1.4 0.3 P/B Supernate UT AD LT 51 12 12 ND 33 22 32 37 99 257 219 114 0 20 59 11 25 52 9 9 ND 22 21 28 34 102 263 231 125 0 19 56 8 22 56 11 12 ND 22 19 23 31 96 254 228 126 0 20 66 8 27 ND ; Not determined. aElastin prepared by the Lansing method. 24 et al. 17 CDetermined according to Thomas et a1. 23 Porcine Tropoe1astin b 45 5 8 4 14 10 16 106 333 227 125 0 16 47 16 28 ~ bFrom Sandberg \0 VI TABLE XIV AMINO ACID ANALYSIS OF UREA EXTRACTION AND PROPANOL/BUTANOL EXTRACTION INSOLUBLE PROTEINS EXPRESSED AS RESIDUES PER 1000 RESIDUES Analysis was performed on each section of aorta with a Beckman 121 Amino Acid Analyzer using a 2 column physiologic system with Beckman A-IS and PA-35 resins. UT ~ upper thorax, LT = lower thorax, and AD = abdomen. Amino Acid Lys His Arg Hyp Asp Thr Ser G1u Pro Gly Ala Cys Val Met lIe Leu Tyr Phe Urea Extract UT AD LT 41 56 50 14 11 13 0 0 0 0 0 0 72 92 85 38 45 44 62 63 54 98 135 120 86 96 72 139 166 195 145 123 123 b Ob Ob 0 84 77 74 ND ND ND 37 34 30 83 90 77 26 24 23 36 36 35 Microfibrillar Protein a 45 15 45 114 56 62 114 64 110 65 48 56 16 48 69 36 38 P/B Insoluble Extract LT UT AD 56 16 0 0 86 56 91 96 75 128 121 0 88 0 27 83 37 37 60 14 0 0 79 60 100 89 82 142 98 0 94 0 28 77 35 35 61 19 0 0 95 71 121 113 67 97 74 0 90 0 32 84 39 36 yGB-II 57 15 0 0 97 56 176 107 77 105 68 0 62 0 27 77 52 28 ND = Not determined. aFrom Ross et al. 22 b Not resolved from glycine. t-' "" 0\ 197 experiments were carried out to evaluate the DNA assay. Table XV shows the results of an experiment where both repeated sampling, and sampling of different amounts was performed. tive. Sampling the homogenate appeared fairly quantitaNext, three different methods of isolation of DNA were evaluated as shown in Table XVI. Potassium acetate/ ethanol has been used to remove sugars that can give a f a 1 se rea d ~ng. 12 o However, it was clear that with the ex- traction being used, this was not a problem. The DNA con- tent of the TCA extraction sample was lower than the DNA content of the extract obtained by the present procedure. Therefore, ued. the extraction procedure being used was contin- Also questioned was the ability of the GF/C filter to retain cold TCA precipitated DNA. XVII, it did an adequate job. As seen in Table Therefore it was concluded that the DNA sampling, extraction, and quantitation procedures were reproducible and accurate. The next set of control experiments centered on the isolation and quantitation of soluble (tropo-) elastin. In Table XVIII, it can be seen that both tropoelastin (porcine) and yGB-II became insoluble after contact with TCA. However tropoelastin could be resolubilized with ammonium formate at 4°C after precipitation with TCA. The next three experiments utilized 125 1 tropoelastin to evaluate quantitative recovery of tropoelastin. Table XIX, experiment 1, revealed that tropoelastin was not isolated 198 TABLE XV QUANTITATIVE SAMPLING OF AORTIC HOMOGENATE FOR DNA Aortic tissue homogenized in 0.45 M NaC1 cocktail was sampled in triplicate at 25 ~1, 50 ~1, and 100 ~1. The DABA assay was then run on each sample. Values are the average of 3 samples ± SD. Sample Volume 25 DNA Content Expected (~g) Observed 11 11 ± 0.5 50 22 25 ± 3.0 100 44 54 ± 4.0 ~1 199 TABLE XVI EVALUATION OF VARIOUS METHODS TO EXTRACT DNA FOR DABA ANALYSIS Three methods were used to extract and evaluate DNA content. There were: 1) sampling the 0.45 M NaCl homogenate, 2) sampling the 0.45 M NaCl homogenate, followed by potassium acetate/ethanol extraction, and 3) sampling the 0.45 M NaCl homogenate, followed by potassium acetate/ ethanol extraction, and finally extraction with 5% TCA at 90 0 C for 15 minutes. All samples were evaluated by the DABA assay. Values are the average of 2 samples ± SD. Sample DNA (Fluorescence 510 nm) 1 42.1 ± 1.8 2 40.2 ± 1.3 3 24.7 ± 2.1 1.7 ± 0.2 12.3 ± 3.2 TCA blank TCA blank + 5 Ilg DNA TABLE XVII EFFICIENCY OF A GF/C FILTER TO RETAIN DNA Two sets of triplicate samples were set up: 1) 100 Ilg RNA (carrier) plus 10 Ilg DNA precipitated in TCA and then filtered by vacuum through a GF/C filter, and 2) 10 Ilg DNA layered directly on a GF/C filter. Values are the average of 2 samples ± SD. Sample 1 (DNA + RNA) DNA (Fluorescence 510 nm) 35.7 ± 0.5 1 (RNA) 3.1 ± 0.1 2 (DNA) 39.0 ± 1.7 2 (Blank) 1.2 ± 0.2 TABLE XVIII SOLUBILITY OF yGB-11 AND TROPOELAST1N IN VARIOUS SOLVENTS yGB-II and tropoe1astin were mixed with various solvents with or without prior treatment with 10% TCA and ethanol. Solubility was then visually observed at 65 0 C (H), 22°C (R), and 4°C (C). The protein was considered insoluble (I) if a particulate matter was observed and soluble (S) if clear. Methylated casein and insulin were also included for comparison. Sample 10% TCA 95% Ethanol yGB-I1 + + + + + + + + a b + Methylated Casein Insulin + + + + e f + + + Temperature H R C + + + + + + + + ~ 45 M NaC1 cocktail; + + + + + + + + + + + + + + + + + + + + + + + + Tropoe1astin Solvents c d + + + + + + Solubility I S + + + + + + + + + + + + + + + + + + + +(R) +(R) +(R) +(R) + + + + + + + + + bammonium formate; cO.2 M sad ium- ace ta te,-Plf 4.9; eO.5 M acetic acid; fO.5 M pyridine, pH 8.3. <1 1% SDS; +(C) +(C) +(C) +(C) + N o o TABLE XIX FRACTIONATION OF 125 1 TROPOELASTIN 125 1 tropoelastin was added to aortic ti.ssue which was then homogenized and fractionated as in Figure 1. A second fractionation was performed with certain improvements: 1) 5% TCA instead of 100% ethanol, 2) 10 ~l DFP and 0.5 M Tris in the collagenase assay, and 3) longer propanol/butanol extraction. Values 'are the average of 2 samples ± SD. Experiment 1 Procedure Total Pellet CPM/Fraction 19160 ± o 352 ± 19 TCA Supernate 1659 ± 126 100% Ethanol 2183 ± 45 Experiment 2 % TCA Precipitable Counts CPM/Fraction % TCA Precipitable Counts 13165 ± 203 Not measured 2768 ± 291 13.8 5% TCA 256 ± 10 2.5 DFP + 0.5 M Tris 600 ± 1 5.8 8231 ± 498 80.0 Collagenase Digest Supernate 2213 ± 76 14.0 Propanol/ Butanol Supernate 8785 ± 387 55.5 Propanol/ Butanol Pellet Alterations in the Procedure N 2656 ± 385 16.8 Longer Extraction o 1197 ± 51 11.6 ~ 202 along with the insoluble elastin, but that considerable amounts were lost into the ethanol wash, collagenase digest, and P/B insoluble pellet. As can be seen in experi- ment 2, this procedure was corrected by replacing the ethanol wash with 5% TCA, adding DFP and 0.5 M tris (to buffer residual TCA) to the collagenase digest, and bly by longer P/B extraction. p~ssi­ Significant radioactivity was still present in the P/B pellet so various solvents were tried to increase extraction as shown in Table XX. Ammonium formate was still the best extractant, so longer extraction times were tried as seen in Table XXI. had little effect on the degree of extraction. Time Under the old conditions approximately 55% of the tropoelastin was extracted, and with slight modifications this was increased to 80%. The 2 day old rat experiment was then repeated and the results are seen in Figures 15 through 18. The upper thoracic aorta appeared to contain more free radioactivity initially as seen in Figure 15. This was not similar to that seen in the previous experiment, however the radioactivity rapidly decreased as before. The gradient of syn- thesis of soluble collagen seen in the previous experiment was again observed (Figure l6a), i.e. greater radioactivity in the abdominal region was seen initially. However, this did not account for the accumulation seen in the insoluble collagen fraction (Figure l6b). This may reflect TABLE XX ASSAY OF SOLVENTS FOR USE AS AN EXTRACTANT OF TROPOELASTIN FROM THE PROPANOL/BUTANOL INSOLUBLE PROTEIN PELLET TCA precipitated yGB-II (1 mg) and 125 1 tropoe1astin (0.5 mg) were extracted with one of 4 solvents for 15 minutes before propanol/butanol extraction. The 4 extractants were: A) 0.45 M NaC1 cocktail, B) ammonium formate, pH 5.2, C) 0.5 M acetic acid, and D) n-methy1morpho1ine/H 2 0 (1:1), pH ~10. The resultant extract supernate (S) and pellet (P) were measured for radioactivity. Values are the average of 2 samples ± SD. Sample % TCA Precipitable Counts Extracted CPM Total TCA Precipitable Counts 13032 ± 2948 A 76.7 ± 3.5 P 9944 ± 1804 3088 ± 1144 S P 9328 ± 1774 ± 36 82 84.0 ± 0.6 S 122 97 79.6 ± 0.5 P 8802 ± 2266 ± S P 2152 ± 8731 ± 381 59 19.8 ± 2.9 S B C D 11102 ± 118 11067 ± 219 10883 ± 322 N o w TABLE XXI EFFECT OF TIME ON THE EXTRACTION OF 125 1 TROPOELASTIN FROM TCA PRECIPITATED PROTEINS BY AMMONIUM FORMATE TCA precipitated yGB-II (1 mg) and tropoelastin (0.5 mg) were extracted with 0.5 M ammonium formate, pH 5.2, propanol, and then butanol for varying amounts of time. Sample A was extracted for 15 minutes in each, B for 45 minutes, C for 2.25 hours, and D for 5 hours. Both the supernate (S), and the insoluble pellet (P) were measured for radioactivity. Values are the average of 2 samples ± SD. S a m p l e _ _ ___ CPM in Supernate %TCA Precipitable Coun ts in Superna te Total TCA Precip itable Counts 9224 ± A 8 16 82.8 ± 0.1 P 7635 ± 1590 ± S P 7374 ± 325 1393 ± 78 84.1 ± 0.2 S 83.3 ± 0.4 P 7305 ± 120 1464 ± 68 S P 7270 ± 216 1478 ± 33 83.1 ± 0.1 S 24 8767 ± 402 B 8769 ± 187 C 8748 ± 249 D N o .f:- 205 Figure 15. TeA soluble fraction of the 2 day old rat experiment 2. th~ salt extract of This fraction contained primarily free amino acids and small peptides. ax (A), lower thorax (..), and abdomen ( . ) . the average of 2 samples ± SD. Upper thorValues are 206 o 207 Figure 16. Radioactivity present in soluble Ca) and in- soluble (b) collagen in the 2 day old rat experiment 2. Both the salt soluble extract and insoluble pellet were treated with collagenase and then precipitated with cold The TCA supernate was then measured for radioactivUpper thorax CA.), lower thorax ( . ) , and abdomen Values are the average of 2 samples ± SD. 208 12 ~ 10 :z; Cl 0 = 0 0 r-i 8 6 C ") 0 r-i >< ~ p.. Q 4 2 >- ...:I 0 :::t: . C") B 5 4 ~ :z; Q 0 :;J 3 0 0 r-i - ...::t 0 r-i 2 >< ~ p.. Q 1 >...:I 0 :::t: C") Q 1 2 3 4 6 TIME (HOURS) 15 209 incomplete extraction and may be explained by the "hump" seen in the early time points of the insoluble collagen fraction. It is interesting to note that no gradient of accumulation of collagen occurred, soluble collagen being made. so it did not reflect the It also accumulated quite rapidly, being completed by 6 hours. The plR supernate (Figure l7a) was similar to that seen in the 2 day old rat (experiment 1). was still far below that seen in the (Figure 17b). Radioactivity inso1~ble elastin This could be partially due to incomplete extraction, as a "hump" was observed in the early time points, but this was unlikely since the 125 1 tropoe1astin tracer study previously showed that tropoelastin didn't remain with the salt insoluble pellet. The "humps" seen in the insoluble pellet were probably artifactual. How- ever, it was quite possible that partially crosslinked elastin was isolated either in the plB pellet or urea extract which showed a pulse in radioactivity during the early time periods (Figures 18a,b). Amino acid analyses (Table XIV) revealed the possibility of a minor contamition with elastin. The valine, alanine, -glycine, and pro- line amino acids were fairly rich in these fractions. It may be that tropoelastin was not present in sufficient quantities as a free entity in the normal animal because it was rapidly crosslinked intrace11ularly. It was clear 210 Figure 17. Radioactivity present in the propanol/butanol supernate Ca) and elastin pellet (b) of the 2 day old rat experiment 2. Upper thorax CA), lower thorax C.), and abdomen ( . ) . Values are the average of 2 samples ± SD. 211 A 8 <: z Cl c.!l ~ - 6 0 0 M C""l 4 0 M >: ~ ~ 2 Cl >< ...:l c.!l ::r:: C""l 1 2 3 4 6 TIME (HOURS) 15 212 Figure 18. Radioactivity present in the propanol/butanol pellet Ca) and urea extract (bl of the 2 day old rat experiment 2. abdomen C.). Upper th'O.rax CA), lower thorax ( . ) , and Values are the average of 2 samples ± SD. 213 20 16 12 8 < z 4 0 C,!) ::J 0 0 ..., C"") ..., 0 >:: ;s:: p.. 0 :>< .... C,!) B 36 = C"") 28 20 12 4 o 1 2 3 4 6 TIME (HOURS) 15 214 from Figure 17b that accumulation of elastin was greater in the upper thoracic than the abdominal region, and that it was quite rapid, having been completed by 6 hours. Ac- cumulation of radioactivity into elastin and collagen occurred at similar relative rates; however, quantitatively, as determined by total accumulated radioactivity, the ratio of elastin to collagen was approximately 2.2:1 in the upper thorax, 1.8:1 in the lower thorax, and 1:1 in the abdomen. The next experiment was a repeat of the 4 week old rat (experiment 1), however, were used. in this case 6 week old rats Figures 19 through 22 display the results. Figure 19 shows that approximately the same amount of label was present in tissue from each region, and that it rapidly decreased with time. Table XXII reveals that by the 30 minute time point of the 6 week old rat the amount of label in free glycine was altered appreciably, especially in the abdominal region. However, this was not true of the 2 day old rat where all the label could be freegly~ine at 30 minutes. accoun~ed for in This reflected very different rates of metabolism of glycine not only at different ages but in different regions of the aorta. The results shown in Figure 20a revealed a significant gradient of soluble collagen synthesis which gradually decreased. This was reflected quantitatively by a gradual accumulation of insoluble collagen. A gradient of 215 Figure 19. TCA soluble fraction of the salt soluble ex- tract of the 6 week old rat experiment 1. This fraction contained primarily free amino acids and small peptides. Upper thorax (A), lower thorax ( . ) , and abdomen ( . ) . Values are the average of 2 samples ± SD. 216 ~ ~ ~ 0 0 ~ ~ ~ ~ ~ H H 217 TABLE XXII AMINO ACID ANALYSIS OF 3H GLYCINE IN VARIOUS FRACTIONS A1iquots of the TCA supernate of the salt extract of the 2 day rat experiment 1 and the 6 week rat experiment were separated, unhydro1yzed, on a Beckman 121 Amino Acid Analyzer using only an acidic program. The TCA supernates from each region were evaluated. UT = upper thorax, LT = lower thorax, and AD = abdomen. A = Free 3H j1 y cine, 30 minute time point of the 2 day rat, B = free H glycine, 30 minute time point of the 6 week rat, and C = control 3U glycine, and 3H valine. Values are the average of 2 samples ± SD taken from a single amino acid analysis. Sample CPM Total Added % Total A UT LT AD 1229 ± 71 849 ± 10 1326 ± 136 1208 ± 104 840 ± 49 1355 ± 15 101.7 101.1 97.9 B UT LT AD 4646 ± 195 3015 ± 16 3482 ± 46 5639 ± 25 5460 ± 119 7777 ± 69 82.4 55.2 44.8 C Control 3H glycine 3H valine 448743 285223 459220 314663 97.7 90.6 218 Figure 20. Radioactivity present in soluble (a) and in- soluble (b) collagen of the 6 week old rat experiment 1. Both the salt soluble extract and insoluble pellet were treated with collagenase and then precipitated with cold TCA. The TCA supernate was then measured for radioactiv- ity. Upper thorax C A ), lower thorax ( . ) , and abdomen C.) . Values are the average of 2 samples ± SD. 219 5 4 3 2 1 <11 Z t::l c..? ::;J 0 0 r-l ..;r 0 r-l >:: B ~ Po< T t::l ~ ~ 0 6 =: C"") 4 2 30 t 1 2 5 TIME (HOU!.S) 10 220 accumulation similar to the gradient of synthesis (Figure 20b) also occurred. The P/B supernate again rapidly decreased and was low in radioactivity (Figure 21a) when compared with that seen in the elastin pellet (Figure 2Ib). Partial crosslinked elastin (soluble) may again be reflected by radioactivity in the P/B pellet and urea fractions (Figures 22a,b). The elastin pellet did show a gradient of accumulation. The upper and lower thoracic regions showed greater accumulation than the abdominal region (Figure 2Ib). The accumulation was "faster" than that for the insoluble collagen as the curve was convex (Figure 21b) compared with the concave curve of insoluble collagen formation (Figure 20b). Again accumulation was greater in the elastin frac- tion than the collagen fraction. The total activity ratios of elastin to collagen were in the abdominal region 1.3:1, in the lower thoracic region 3:1, and in the upper thorcic region 8:1. These results are confirmed by data ob- tained for total elastin and collagen content determined by hydroxyproline content (Figure 9b). 3H glycine was chosen so that comparisons between elastin and collagen metabolism could be made by incorporated radioactivity since the residues/IOOO residues of glycine is approximately the same in each. Amino acid analyses of each fraction revealed much the same results as seen in the 2 day old rat results 221 Figure 21. Radioactivity present in the propanol/butanol supernate (a) experiment 1. abdomen C.). and elastin pellet Upper thorax (A), (b) of the 6 week old rat lower thorax ( . ) , and Values are the average of 2 samples ± SD. 222 28 24 -< z 20 C!> l:I 16 Q 0 0 .-! ...... 12 C"1 0 .-! ~ 8 ~ Po. Q >< 4 ..J C!> ::r: C"1 < Z 16 Q C!> l:I 12 0 0 ~ ...... ..;:t 0 ~ 8 ~ ~ Po. Q 4 >< ..J C!> Jrl B ::r: C"1 0 30 t 1 2 5 TIME (HOURS) 10 223 Figure 22. Radioactivity present in the propanol/butanol pellet (a) and urea extract (b) of the 6 week old rat experiment 1. abdomen ( . ) . Upper thorax CA), lower thorax ( . ) , and Values are the average of 2 samples ± SD. 224 6 B -, o 10 1 TIHE (HOURS) 225 (Tables XXIII through XXV). The elastin and collagen frac- tions were quite pure. The inability to isolate free tropoelastin in quantities large enough to account for that incorporated into elastin led to a series of control experiments. Table XXVI contains the results of an experiment where the entire procedure of isolation of tropoelastin was done within one day. Enzyme inhibitors were present from the moment the chest cavity was opened until plB extraction. no tropoelastin was found to be present. Little or However, signi- ficant amounts of radioactivity were found in the plB pellets, but the standard deviations were very large, making the results- questionable. out. Further experiments were carried Figure 23 is an autoradiogram of a PAGE gel contain- ing the 125 1 Bolton-Hunter reagent labeled soluble salt extract and the subsequent plB extract. In most of the fractions there was a band that migrated similarly as tropoelastin. However upon amino acid analyses of both the supernate and the pellet of the PIB extract, there was little indication of tropoe1astin (Table XXVII). Valine, alanine, and proline were low in both fractions, while hydroxyproline and hydroxy1ysine were rich in the pellet indicating the presence of collagen. Several antibody experiments were also carried out. Since antibody to rat elastin or tropoelastin was not available, rabbit antibody to bovine ligamentum nuchae TABLE XXIII AMINO ACID ANALYSIS OF 6 WEEK RAT AORTIC COLLAGEN EXPRESSED AS RESIDUES PER 1000 RESIDUES Analysis was performed on each section of aorta with a Beckman 121 Amino Acid Analyzer using a 2 column physiologic system with Beckman A-15 and PA-35 resins. UT = upper thorax, LT = lower thorax, and AD = abdomen. Amino Acid Lys His Arg Hyp Asp ·Thr Ser G1u Pro G1y Ala Val Met I1e Leu Tyr Phe HDL Soluble Collagen AD UT LT 41 48 14 0 19 33 64 b ND 80 61 26 19 50 43 53 37 109 115 350 300 112 126 67 63 ND ND 21 18 44 33 9 9 15 21 ND ND 38 12 46 ND 73 20 51 56 144 327 117 41 ND 16 31 6 22 ND aFrom Chung et a1.22 ND = Not determined. measurement. HDL = Hydroxy1ysine. Insoluble Collagen UT LT AD 28 8 54 96 48 15 42 60 121 311 112 24 ND 14 25 1 11 31 28 8 53 110 48 15 44 59 112 305 113 25 ND 14 27 1 13 27 30 8 54 101 48 16 41 62 123 309 115 23 ND 13 25 2 12 17 Type III Co11agena 30 6 45 126 42 13 37 70 117 347 93 13 13 22 2 8 bLow value due to difficulty of ~ ~ 0\ TABLE XXIV AMINO ACID ANALYSIS OF 6 WEEK RAT AORTIC INSOLUBLE ELASTIN AND PROPANOL/BUTANOL EXTRACTION SOLUBLE PROTEINS EXPRESSED AS RESIDUES PER 1000 RESIDUES Analysis was performed on each section of aorta with a Beckman 121 Amino Acid Analyzer using a 2 column physiologic system with Beckman A-IS and PA-3S resins. UT = upper thorax, LT = lower thorax, and AD = abdomen. Insoluble Elastin Amino Acid UT LT AD Lys His Arg Hyp Asp Thr Ser G1u Pro G1y Ala Val Het lIe Leu Tyr Phe Des Ides 12 2 Day Rat Aortic E1astin a 14 15 9 0 0 9 ND 23 24 22 26 103 368 200 85 12 15 26 23 24 25 94 356 201 83 16 14 26 23 25 28 118 342 192 81 1 10 13 18 16 15 21 98 331 238 96 0 0 26 66 22 15 ND ND 28 68 16 16 ND ND 26 64 17 14 ND ND o o o 25 68 24 19 0.3 1.4 P/B Supernate UT LT AD 48 o 36 0 6 15 9 14 15 18 120 294 219 134 6 9 12 14 15 22 96 403 171 108 18 51 11 29 o 45 o 6 6d Porcine Tropoe1astin b 45 5 8 4 13 13 23 20 112 360 186 114 0 o 14 10 16 106 333 227 125 22 50 11 26 17 47 13 26 16 47 16 28 ND = Not determined. aElastin prepared by the Lansing method. Z3 -bFrom Sandberg 17 et a1. CDetermined according to Thomas et a1.21 dLow value due to difficulty of measurement. N N ...... TABLE XXV AMINO ACID ANALYSIS OF UREA EXTRACTION AND PROPANOL/BUTANOL EXTRACTION INSOLUBLE PROTEINS EXPRESSED AS RESIDUES PER 1000 RESIDUES Analysis was performed on each section of aorta with a Beckman 121 Amino Acid Analyzer using a 2 column physiologic system with Beckman A-1S and PA-35 resins. UT = upper thorax, LT = lower thorax, and AD = abdomen. Amino Acid Lys His Arg Hyp Asp Thr Ser G1u Pro G1y Ala Cys Val Met lle Leu Tyr Phe Urea Extract AD UT LT 55 55 54 0 0 0 55 54 55 0 0 0 82 86 87 52 51 49 71 71 66 118 119 .114 77 79 77 120 120 142 88 85 86 Ob Ob Ob 67 64 67 12 21 17 43 44 45 84 86 87 32 31 31 36 34 34 ND = Not determined. aFrom Ross et a1.24 bNot resolved from glycine. Microfibrillar Protein a 45 15 45 114 56 62 114 64 110 65 48 56 16 48 69 36 38 P/B Insoluble Extract UT LT AD 66 21 42 0 90 65 104 103 74 108 87 0 75 ' 1 26 75 31 32 62 19 41 0 90 64 110 105 74 107 87 0 71 7 25 75 33 32 64 19 44 0 90 60 94 110 73 106 89 0 73 10 29 79 30 35 yGB-II 57 15 0 0 97 56 176 107 77 105 68 62 0 27 77 52 28 N N 00 229 TABLE XXVI QUICK EXTRACTION OF TROPOELASTIN FROM 2 DAY RAT AORTA Five 2 day rats were injected with 9.0 ~Ci of 3 H glycine/g body weight. The aortas were removed and fractionated as described in the Methods. UT = upper thorax, LT = lower thorax, and AD = abdomen. Values are the average of 2 samples ± SD. Fraction DPM/100 ~g DNA TCA Supernate UT LT AD 131145 ± 116799 ± 127923 ± 4464 5759 4766 Collagenase Digest Supernate (of soluble fraction) UT LT AD 9961 ± 6524 ± 19636 ± 100 1131 4457 Propanol/Butanol Supernate UT LT AD 12747 ± 7314 ± 7911 ± 4794 5835 1655 Propanol/Butanol Pellet UT LT AD 0.45 M NaCl Insoluble Proteins UT LT AD 142579 ± 80349 36405 ± 35559 39705 ± 14400 50866 ± 38804 ± 28803 ± 3287 8641 7093 230 Figure 23. Autoradiogram of 125 1 BH reagent labeled salt soluble extract and subsequent propanol/butanol extract proteins. Whole aorta was homogenized in a NaCl buffer and soluble proteins isolated. After dialysis and 1yo- philization, the dry sample was resolubilized in water, followed by propanol/butanol extraction. protein (a), water soluble protein (b), Water insoluble 125 1 tropoelas- tin (c), propanol/butanol supernate Cd), and propanol/butanol pellet (e). DNS-tropoelastin. The arrow indicates the position of A c D E 232 TAB-LE XXVII AMINO ACID ANALYSIS OF THE SALT EXTRACT AND SUB-SEQUENT PROPANOL/BUTANOL EXTRACT SOLUBLE AND INSOLUBLE PROTEINS OF 6 WEEK OLD WHOLE RAT AORTA EXPRESSED AS RESIDUES PER 1000 RESIDUES Amino acid analysis was performed on a Beckman 121 Amino Acid Analyzer using a 2 column physiologic system with B-eckman A-15 and PA-35 resins. PIB- Soluble Amino Acid Lys His Arg Hyp Asp Thr Ser Glu Pro Gly + Cys Ala Val Met Ile Leu Tyr Phe HDLc ~mo1es Total Amino Acids Protein a Porcine P/B Insoluble Protein Tropoelastin b 5 ± 1 45 11 ± 2 5 8 4 14 10 16 106 333 227 125 54 26 60 178 40 445 63 36 ± ± ± ± ± ± ± ± 32 4 10 7 13 38 13 6 13 38 15 16 ± ± ± ± 3 6 6 6 16 47 16 28 35.9 aDetermined by two analyses on one sample. et a1. 17 cHDL = Hydroxy1ysine. 51 24 42 43 86 40 52 103 74 177 99 53 2 25 69 17 25 11 1075.9 bFrom Sandberg 233 alpha elastin was used. Table XXVIII shows the results of an assay using constant antigen and serially diluted antibody concentration. Maximum precipitation of the alpha elastin was approximately 39%. This was probably due to the fact that alpha elastin is a very heterogeneous population of peptides which give a long smear from top to bottom of a 12% PAGE gel. Therefore, many peptides probably did not contain an antigenic determinant. Table XXIX shows an assay that used a constant high concentration of antibody and serially diluted antigen. increased to 51%. Maximum precipitation was Table XXX shows the results of a com- petitive binding assay. It was nece$sary to show that the antibody would crossreact with rat alpha elastin and possibly porcine tropoe1astin, and at what concentrations. This was carried out and a significant difference was seen at 0.08 ~g elastin. rat alpha elastin, and 0.003 porcine tropo- Also evaluated were samples of the previous salt extract of aorta. served. ~g No elastin-like material could be ob- Since IgGsorb was used as the second precipitating substance, and since it is known not to precipitate all IgG's, a double immune precipitation using goat anti rabbit IgG was also carried out (Table XXXI). material could be detected again. No elastin-like It is important to note that the salt extraction was done in the presence of DFP, and with PMSF and NEM present during the short 2 hour dialysis. No collagenase assay was done. Therefore no 234 TABLE XXVIII BOVINE LIGAMENTUM NUCHAE ALPHA ELASTIN BINDING BY RABBIT ANTI BOVINE LIGAMENTUM NUCHAE ALPHA ELASTIN SERUM - ANTIBODY DILUTION Serial dilutions of antibody were reacted with a constant amount of antigen at 4 0 c overnight. IgG was then precipitated by IgGsorb. Values are the average of 2 samples ± SD. Ag (Dilution) 1:480 Ab (Dilution) CPM± Background % Immune Precipitable Counts 1037 ± 120 38.5 1: 10 1: 20 834 ± 69 31.0 1: 40 688 ± 37 25.6 1: 80 573 ± 3 21. 3 1 : 160 318 ± 16 11.8 1 : 320 146 ± 4 5.4 1 : 640 30 ± 2 1.1 1:1280 0 0 1:2560 0 0 NRS 1:10 (Background) NRS 1:10 (plus 10% TCA) 262 ± 6 2690 ± 96 235 TABLE XXIX BOVINE LIGAMENTUM NUCHAE (BLN) ALPHA ELASTIN BINDING BY RABBIT ANTI BOVINE LIGAMENTUM NUCHAE ALPHA ELASTIN SERUM - ANTIGEN DILUTION 125 Serial dilutions of antigen IBo1ton labeled BLN a-elastin was reacted with of antibody at 4 0 C overnight. IgG was by IgGsorb. Values are the average of Ag (Dilution) Ab (Dilution) Hunter reagent a constant amount then precipitated 2 samples ± SD. % Immune Precipitable Counts CPM± Background 9317 ± 103 14.2 1: 60 6826 ± 21 31. 2 1:120 4117 ± 33 35.5 1:240 3023 ± 39 51.2 1:480 1037 ± 120 33.7 495 ± 40 1: 20 1:120 1:10 NRS 1:10 (Background) 1: 20 65600 ± 1491 1: 60 21867 ± 497 1:120 11589 ± 64 1:240 5902 ± 26 1:480 3075 ± 63 TABLE XXX COMPETATIVE BINDING ANTIBODY ASSAY FOR THE PRESENCE OF TROPOELASTIN IN THE 0.45 M NACL EXTRACT OF THE YOUNG RAT AORTA 5 Whole aortas from 6 week old rats were homogenized in 0.45 M NaCl cocktail. The soluble proteins were incubated with rabbit anti bovine ligamentum nuchae alpha elastin antibody, along with l25IBolton Hunter reagent labeled bovine ligamentum nuchae (BLN) . alpha elastin, cold competing antigens, rat alpha aortic elastin, and porcine aortic tropoelastin. IgG was precipitated by IgGsorb. Values are the average of 2 samples ± SD. Antibody Rabbit aBLN Alpha Elastin Antigen 125IBH BL Elastin Cold Comptetative Antigen 1:10 dilution .05 11g .000 11g .003 .016 .080 .400 2.000 10.000 50.000 250.000 10 111 25 % Immune Precipitable Counts in the Presence of Cold Antigen Rat a Elastin Pig Tropoelastin Aortic Extract 100 96.1 90.1 82.2 68.6 64.7 68.9 55.7 51.4 ± ± ± ± ± ± ± ± ± 16.4 22.0 16.5 1.1 19.6 1.0 9.0 23.0 16.0 100 86.7 83.6 84.3 80.1 69.7 68.7 58.6 55.2 ± ± ± ± ± ± ± ± ± .27 .68 0 1.37 3.28 3.69 2.46 .27 2.60 118.40 ± 2.19 103.10 ± 10.15 tv W 0\ 237 TABLE XXXI DOUBLE IMMUNE PRECIPITATION OF 125 IBH AORTIC EXTRACT BY RABBIT ALPHA LIGUMENTUM NUCHAE ELASTIN ANTIBODY Antibody was reacted with 125IBolton Hunter (BH) reagent labeled aortic extract at 4°C overnight. It was then double immune precipitated by addition of goat aIgG for 2 hours at 4°C. After centrifugation the pellets were measured for radioactivity. Values are the average of 2 samples ± SD. 125IBH Aortic Extract + Ab NRS + + + + + + TCA + % Immune Precipitable Counts CPM in Pellet 591 ± 54 735 ± 285 15582 ± 1237 56409 ± 195 0 238 degradation of soluble elastin should have occurred. Yet little, if any free tropoelastin was found. Discussion The growing young rat rapidly increases in body weight and total aortic weight between one day to twelve weeks of age. 4 Much of the increase in aortic weight is due to increased accumulation of elastin and collagen, the primary proteins by weight found in the aorta. It was clear in the present experiments that there were significant changes in the ratios of elastin to collagen at two days and four weeks of age (see Figure 9). The total amount of elastin per mg DNA was shown to increase, as did the amount of elastin from the upper thorax to the abdomen. Collagen, on the other hand, did increase in total amount, and also relative accumulation in different regions changed between two days and four weeks. The foregoing experiments shed light on the reasons for these observations. It was found in the two day rat that synthesis of collagen did not correlate with subsequent incorporation into insoluble collagen. It was shown that increased syn- thesis in the abdomen was not reflected in increased accumulation in the abdomen. This suggested different rates of crosslinking or degradation in different regions of the aorta. A precursor to elastin, tropoelastin, could not be 239 isolated in quantities large enough to account for that seen in the insoluble elastin. This was repeatedly the case, even in several control experiments. From this it was concluded that tropoelastin was not present as a free monomer in vivo in the normal animal. plained by two different reasons: This could be ex- 1) rapid intracellular crosslinking, and/or 2) a precursor with different chemical properties than tropoelastin, of which tropoelastin is a subunit. That tropoelastin must be part of a precursor has been substantiated by three criteria: 1) its amino acid composition is identical to that of elastin except for lysine and subsequent crosslinking amino acids 17 ,18, 2) antibody crossreactivity19, and 3) 14C lysine labeled tropoelastin can be chased into l4C desmosine labeled elastin 20 . Elastin accumulation in the two day old rat was seen to be greatest in the upper thoracic as compared to the abdominal regions. This was substantiated by results of the elastin content obtained by hydroxyproline analyses. Elastin accumulation was also greater than collagen accumulation except in the abdominal region, while percent rates of formation appeared similar. The above results are different from those found in the six week old rat. Soluble collagen synthesis was again increased in the abdominal region but in the six week old animal the soluble fraction decreased with gradual 240 subsequent incorporation into insoluble collagen. The gradient of synthesis was reflected in a gradient of accumulation. Soluble elastin again could not be isolated in large enough quantities to account for accumulation into elastin. In the elastin fraction of the six week old rat a gradient of accumulation was seen approximating that seen by hydroxyproline analyses. The accumulation of elastin was again greater than that of collagen throughout the aorta. However, in the six week old rat instead of the formation proceeding at similar relative rates as in the two day old rat, collagen appeared to form much more gradually than did elastin. It is concluded therefore, that collagen synthesis is not always reflected by accumulation into insoluble collagen. Soluble elastin, tropoelastin, may be rapidly cross- linked, even intracellularly, or may be present in a larger precursor so that little if any free tropoelastin exists Depending on the age of the rat, elastin and collagen mayor may not form at similar relative rates. Finally, overall accumulation of elastin is greater than that of collagen in the young growing rat aorta. 241 References 1 R. L. Walford, P. K. Carter, and R. B. Schneider, Arch. Pathol., 78 (1964) 43. 2 K. T. Kao, D. M. Hilker, and T. H. McGavack, Proc. Soc. ~. Rio1. Med., 106 (1961) 335. 3 J. Sodek, Arch. Oral BioI., 23 (1978) 977. 4 T. Looker, and C. L. Berry, J. Anat., 113 (1972) 17. 5 R. G. Gerrity, E. P. Adams, and W. J. Cliff, Lab. Invest., 32 (1975) 601. 6 R. Ross, and S. J. Klebanoff, J. Cell BioI., 50 (1971) 159. 7 R. B. Rucker, W. Goetlich-Riemann, K. Tom, M. Chen, J. Poaster, and S. Koerner, J. ~., 105 (1975) 46. 8 G. M. Fischer, Exp. Geronto1., 11 (1976) 209. 9 G. M. Fischer, Endocrinology, 93 (1973) 1216. 10 G. M. Fischer, M. L. Swain, and K. Cherian, Atherosclerosis, 35 (1980) 11. 11 D. N. Foster, and T. Gurney, J. BioI. Chem., 251 (1976) 7893. 12 J. M. Kissane, and E. Robins, J. BioI. Chem., 233 (1958) 184. 13 W. C. Schneider, J. BioI. Chem., 161 (1945) 293. 14 S. M. Partridge, H. F. Davis, and G. S •. Adair, Biochem. J., 43 (1948) 387. 15 A. E. Bolton, and W. M. Hunter, Biochem. J., 133 (1973) 529. 16 J. I. Thorell, and B. G. Johansson, Biochim. Biophys. Acta, 251 (1971) 363. 242 17 L.~. 18 B. L. Rasmussen, E. Bruenger, and L. B. Sandberg, Anal. Biochem., 64 (1975) 255. 19 R. H. Daynes, M. Thomas, V. L. Alvarez, and L. B. Sandberg, Connect. Tissue Res., 5 (1977) 75. 20 A. S. Narayanan, and R. C. Page, {. BioI. Chem., 251 (1976), 1125. 21 J. Thomas, D. F. Elsden, and S. M. Partridge, Nature, 200 (1963) 651. 22 E. Chung, and E. J. Miller, Science, 183 (1974) 1200. 23 A. I. Lansing, T. B. Rosenthal, M. Alex, and E. W. Dempsey, Anat. Rec., 114 (1952) 555. 24 R. Ross, and P. Bornstein, J. Cell BioI., 40 (1969) 366. Sandberg, N. Weissman, and W. R. Gray, Riochemistry, 10 (1971) 52.