ARCHIVES

OF BIOCHEMISTRY

AND

BIOPHYSICS

Vol. 278, No. 2, May 1, pp. 326-3321990

Elastin Metabolism during Recovery from Impaired Crosslink Formation Donald Tinker,*

Nadia Romero-Chapman,*

Karen Reiser,? Dallas Hyde,* and Robert Rucker*,l

*Department of Nutrition, College of Agriculture and Environmental Sciences, TDivision of Pulmonary Medicine, School of Medicine, and *Department of Anatomy, School of Veterinary Medicine, University of California, Davis, California 95616

Received August 23,1989, and in revised form November

20,1989

Accelerated proteolysis of tropoelastin and elastin occurs in the arteries of chicks rendered nutritionally copper-deficient. The process results in part from decreased elastin crosslinking. Repletion of copper-deficient chicks with copper causes a deposition of elastin that is proteinase resistant. Resistance to proteolysis is conferred within 48 h of dietary copper repletion. Deposition of aorta elastin to near normal values occurs after 3-4 days in copper-repleted chicks. Moreover, elastolysis was enhanced when the content of dehydrolysinonorleucine in elastin was abnormally low. The chemical modification of lysyl residue in elastin by citroconylation, however, did not influence the rate of elastolysis. We have shown previously that tropoelastin messenger RNA activity and synthesis are not influenced by dietary copper deprivation (1986, Biochem. J. 236,17-23). Rather, as demonstrated herein, the decrease in elastin content in arteries of copper-deficient birds appears to be more the result of enhanced degradation. Restoration of normal crosslinking restores deposition and imparts resistance to elastolysis. Moreover, serum appears to be a good source of elastolytic proteinases when the elastin substrate is partially or e lsso Academic PESS, IN. abnormally crosslinked.

Elastic fibers contribute to the elasticity of blood vessels, ligaments, skin, and lung (l-3). The major component of elastic fibers, elastin, is an unusual protein composed of apolar polypeptide sequences, which are crosslinked extensively by means of lysyl-derived crosslinks (4). Elastin is also a very long-lived protein, whose extracellular repair appears to occur by reduplication of damaged fibers, rather than complete degradation and i To whom correspondence

should be addressed.

resynthesis (5-8). Further, elastin degradation contributes to important pathological and physiological events. In major arteries, focal destruction of the internal elastic lamina is a key event in the atherosclerotic process (9). Loss of elastin is important in emphysema, since dilation of terminal airways can occur when the loss is extensive (10-12). That elastolytic enzymes are important in these processes is well-established. However, it is less appreciated that in uiuo chemical modification, such as lysine-derived crosslink formation, substantially influences the potential for elastolysis. Previously, we demonstrated that inhibition of elastin crosslinking does not alter tropoelastin messenger RNA’ activity (13), but does enhance degradation (14). Serum was identified as a potential source of elastolytic proteinase activity (14). Further, using nutritional copper deficiency as an in uiuo model, it was possible to demonstrate that when crosslinking is severely impaired, the pool of soluble elastin is enhanced lo- to 20-fold. Some of this pool is derived from the continual proteolysis of insoluble, but partially crosslinked elastin (13,14). Herein, we address in more detail elastin degradation as influenced by inhibition of elastin crosslinking. Dietary copper repletion following severe depletion is examined to determine the extent to which elastin fiber repair reconfers resistance to elastolysis. When dietary copper is deficient, lysyl oxidase activity and, as a consequence, the oxidative deamination and condensation of lysyl residues in elastin to crosslinks are compromised (8, 15-18). In the thoracic aorta, repletion is associated ’ Abbreviations used: BAPN, fl-aminopropionitrile; RNA, ribonucleic acid; EDTA, ethylenediamine tetraacetic acid; TBNS, 2,4,6Tris(hydroxymethyl)-aminotrinitrobenzenesulfonic acid; Tris, methane; ELISA, enzyme linked immunosorbent assay; ACP, aldol condensation products; A-LNL, dehydrolysinonorleucine; LNL, lysinonorleucine.

326 All

OOO3-9861/90 $3.00 Copyright 8 1990 by Academic Press, Inc. rights of reproduction in any form reserved.

ELASTIN

METABOLISM

AND

with increased deposition of crosslinked elastin. Also, the levels of soluble elastin diminish and signs of elastolysis are not obvious. EXPERIMENTAL

PROCEDURES

Materials. Groups of l-day-old White Leghorn cockerel chicks were fed either a basal diet (co.4 pg copper/g diet) based on spraydried skim milk (16), or the basal diet with copper added as cupric sulfate (25 pg/g diet). Copper was assayed by flame atomic absorption spectrophotometry (19). Details regarding husbandry and selected morphologic features of the copper-deprived chicks have been described (13, 15,20). Chemical reagents were purchased from Bio-Rad Laboratories (Richmond, CA), Pierce Chemical Co. (Rockford, IL), Bethesda Research Laboratories (Bethesda, MD), and Sigma Chemical Co. (St. Louis, MO). L-[G-“Hlvaline (15 Ci/mmol) was purchased from ICN Radiochemicals (Irvine, CA). Tissue-culture supplies were obtained from Grand Island Biochemical Co. (Grand Island, NY). Chromatographic supplies were purchased from Beckman, Palo Alto, CA; Altex, Berkeley, CA; and Rainin, Emeryville, CA. Tropoelastin was isolated as previous described (17). Fresh serum from young adult Leghorn cockerels was used as the source of elastolytic enzymes. of soluble and insoluble elastin. One to two aortae Determination (50-100 mg) were homogenized in guanidine thiocyanate reagent: 5 M guanidine thiocyanate containing 0.5% N-lauryl sarcosine, 10 mM EDTA, 0.7% 2-mercaptoethanol, 0.25% antiform A, and 10 mM Tris buffer, pH 7.5. After 5 h of extraction, the homogenate was centrifuged (lS,OOOg, 20 min) and the supernatant fraction collected. For Western blots, the guanidine thiocyanate in aorta extracts was exchanged for urea by passage through PD.10 columns (Pharmacia, Uppsala, Sweden) equilibrated with 4 M urea. The eluates were then electrophoresed (21), and elastin was immunodetected (22,23) using a polyclonal antielastin antibody from rabbit serum and horseradish peroxidase conjugated to goat anti-rabbit immunoglobulins (14) and 4-chloro-l-naphthol as the peroxidase substrate (22). Samples were also assayed by enzyme-linked immunosorbent assays (ELISA) following methods outlined by Mecham and Lange (24), Buckingham et ul. (15), and Voller et al. (25). Chick tropoelastin was used to coat the ELISA plates and establish standard curves. The guanidine thiocyanate reagent was added to the standard solutions in amounts equivalent to those in t,issue extracts to compensate for any effects of the reagent on the binding capacity of the antibodies. For insoluble elastin, the residue remaining after extraction with guanidine thiocyanate was extracted further with 95% (v/v) ethanol overnight. The residue was then freeze-dried, and insoluble elastin was isolated by a modified alkali-extraction procedure and measured as described previously (1.5). Morphology. Aortae from chicks were cut transversely so that six samples of full-wall thickness were taken for each aorta (initially fixed in buffered formalin). Four blocks were cut from each of the samples at random. These samples (postfixed in osmium tetroxide) were dehydrated in increasing concentrations of ethanol and embedded in glycol methacrylate. Sections (1.5 pm) were cut using a Sorvall JR4 microtome and stained with hernotoxylin and eosin. All sections were evaluated without knowledge of their experimental grouping. Organ culture. Two aortae (minced into l-mm3 pieces) were used per culture (13, 26). The medium (2 ml) was Eagle’s basal medium (Lvaline-free) with Earle’s salts. 1,.[G-“Hlvaiine (10 pCi/ml) was the only source of L-valine. The medium was flushed continually with CO,/O, (19:l) and the aortae were incubated at 40°C. After 1.5 h of incubation, the L-[“Hlvaline-labeled medium was decanted and a sample (about 20 mg from each of the cultures) was transferred to the same medium with no label, but L-valine was added at 5 mM. The incubations were continued for an additional 4 h. Radioactivity incorporated

CROSSLINK

327

FORMATION

into soluble and insoluble protein was measured by scintillation counting and the soluble proteins were also partially characterized following polyacrylamide gel electrophoresis (21) and detection by fluorograPhY (27). Ten milligrams of Chemical modification and amino acid analyses. insoluble elastin was suspended in 20 ml of 0.5 M disodium phosphate butfer, pH 9.0, containing 200 ~1 of citraconic anhydride. The pH was maintained at 9.0 by addition of 2 M NaOH. Samples were incubated for 2 h to ensure full modification. Elastin was then recovered by centrifugation and washed with 95% ethanol and then acetone. Success of derivatization was verified by testing for free amino groups by reacting the modified protein with TNBS (28). Citraconyl groups were also removed from citraconylated protein by incubating the modified protein in 10 mM HCl at 25°C for 10 h. Aliquots of unmodified insoluble elastin from copper-sufficient and -deficient chicks, treated identically, served as controls. All samples were neutralized and fully equilibrated in 100 mM Tris buffer, pH 7.8, prior to elastolytic susceptibility assays. For routine amino acid analysis, elastin samples were hydrolyzed under nitrogen in 6 N HCl for 48-72 h at 104°C. For analysis of reducible crosslinks, aldol condensation products (ACP) and dehydrolysinonorleucine (A-LDL) samples were hydrolyzed in 4 M KOH for 36 h at 104°C (cf. next section). Amino acids and radiochemiQuantitation of crosslinks by HPLC. tally labeled aldol condensation products (NaBH, reduced) and dehydrolysinonorleucine (as lysinonorleucine) were analyzed by reversephase HPLC (29,30) using a Cl8 column, 0.46 X 25 cm. Hydrolysates were injected onto the column with a Hamilton microsyringe via an Altex Model 210 injector and were eluted isocratically with 24% npropanol in 0.01 M phosphoric acid, 0.3% sodium dodecyl sulfate, pH 2.84, at a flow rate of 0.8 ml/min. Altex Model 100 pumps controlled by an Altex microprocessor (Model 420) were used for elution. Amino acids were detected by postcolumn derivatization with o-phthalaldehyde, which was added to the efhuent stream via a 3-way valve and an Altex Model 1OOA pump operated manually. Fluorescent adducts were quantified with a fluorometer with filters designed to detect o-phthalaldehyde derivatives (excitation at 360 nm, emission at 455 nm). Elution times and areas of peaks were recorded. Fractions were collected at 0.5. or 1.0.min intervals with a fraction collector. Ready-Safe cocktail was added as scintillant and the entire sample was counted by liquid scintillation counting in a Beckman 8000 counter using preprogrammed settings recommended by the manufacturer. For each assay, elastin samples were reduced with a mixture of sodium borohydride, 10 mg/ml in dimethylformamide, and NaB3H, (5 mCi/mg NaBH,) for 1 h. Borohydride was added at 1/30th of the dry weight of the sample in each assay. Reduction was stopped by the addition of glacial acetic acid to a pH of 3. Samples were left to stand for 15 min and then washed exhaustively with distilled water prior to hydrolysis, Data were evaluated statistically using Statistical and sampling. the Student t-test or Dunnett’s modification of the Student t-test.

RESULTS Changes

in Soluble

and Insoluble

Elastin

After 14 days of copper deprivation, a 25-35% decrease in insoluble elastin and a lo- to 20-fold increase in soluble elastin were observed in aortae from copperdeficient chicks compared to controls. Values for soluble elastin (tropoelastin and elastin-derived peptides) and insoluble elastin are given in Fig. 1 and 2. The redeposition of insoluble elastin began to occur within 24-48 h following dietary copper repletion. The rate of normal elastin accumulation during the experimental period, 400-500 pg per aorta per day, was linear and identical to

328

TINKER

i**

.*;;/------

i ; -

z50

&_.-'

t

t -~.-.-.-.-.-,-,-,~-.-,-.-.-.-.-.-~

0 0

5

10

Days Following

15

Copper

20

25

Repletion

of tropoelastin and FIG. 1. Dietary copper and the concentrations soluble peptides in the developing chick aorta. The continuous line connects values obtained from copper-supplemented chicks fed ad libiturn (14 days plus 21 days). The hroken lines connect values from copper-deficient chicks (0) or copper-deficient chicks that were repleted with copper (a). The latter was accomplished by transferring one-half of the chicks originally assigned to the copper-deficient diet to the copper-supplemented diet. Each value is the mean f SEM for a minimum of six determinations for any given time point.

the previous estimate of 480 yg per aorta per day by Keeley et al. (31). Accumulation of insoluble elastin in arteries of copper-deficient chicks was also relatively linear, though about 2530% the rate of normal accumulation, i.e., 100-200 pg per aorta per day. In the recovering artery from repleted chicks, the earliest response observed was the decrease in the level of soluble elastin upon copper repletion (Fig. 2). Within 48 h following repletion, the levels of soluble elastin decreased to normal values. This change preceded the increase in elastin accumulation. In this regard, the rate of elastin accumulation markedly exceeded the normal rate of accumulation following 4 to 5 days of repletion. The period of 4 to 7 days of repletion was particularly impressive, with a total accumulation of about 4000 pugof new elastin. Figure 3A shows Western blots for aorta extracts of birds rendered copper-sufficient and -deficient, or copper-repleted for 2, 4, 6, or 14 days following the initial depletion. Note the absence of degraded products (both high- and low-molecular-weight fragments) after 48 h of repletion. Copper repletion and reinitiation of elastin fibrogenesis reconferred elastolytic resistance. Alkaliinsoluble elastin from copper-deficient chick aorta also served as an excellent substrate for serum proteinases, whereas alkali-insoluble elastin from normal chick aorta or from copper-repleted chicks was resistant to elastolysis (Fig. 3B). Moreover, the changes in aortic elastin occurred substantially before changes in growth, measured as weight gain, and were stimulated in response to copper repletion. Copper-deficient chicks weighed 102 + 12 g at Day 14 compared to 125 + 6 g for copper-supplemented chicks. After 7 days of dietary copper repletion or continual copper deprivation, copper-deficient

ET AL.

and repleted chicks weighed 102 to 135 g (115 * 6 g or 126 + 9 g, for the groups of copper-deficient or repleted chicks, respectively). The group of copper-supplemented chicks increased their weight to 165 k 5 g (P < 0.01). However, after 14 days of dietary copper repletion, weights for the copper-repleted and copper-supplemented birds were no longer significant (245 * 10 g or 266 + 15 g for the groups of repleted or supplemented birds, respectively). The group of copper-deficient birds weighed 164 + 16 g (P < 0.05). Further, when aortic elastin was expressed as a percentage, the values for copperdeficient birds remain relatively constant from Day 14 to 21, i.e., 6-7 percent. Upon copper repletion, aortic elastin increased from this value to a normal value of lo-12 percent during the first 7 days of dietary copper repletion. Synthesis of Tropoelastin by Aorta Explants

and Insoluble Protein

Data in Table I indicate rates at which soluble aorta protein is rendered insoluble in explant cultures. When aortae from 14-day-old chicks were incubated in the presence of [3H]valine, about one-third to one-half of the labeled soluble protein becomes insoluble after 34 h of incubation. This process of insolubilization was clearly more rapid than the equilibration of lysyl groups to A-LNL in viuo (Table II), or the equilibration of lysine to desmosine, which requires 48 to 72 h in explant culture as has been previously published (32). Crosslink-Amino Acid Analysis and Chemical Modification Data for changes in the lysyl residue content, the relative ACP and A-LDL content, and elastolysis are given in Table II. Elastolysis as influenced by the chemical modification of lysyl groups, i.e., the citraconylation of insoluble elastin, was also examined (Table III).

0 0

5

10

Days Following

15

Copper

20

25

Repletion

of insoluble elastin in FIG. 2. Dietary copper and the concentration the developing chick aorta. The procedure and symbols are the same as in the legend to Fig. 1.

ELASTIN

w

inDays

l4

16

18

METABOLISM

21

AND

28

CROSSLINK

0

329

FORMATION

Copper Copper Copper SupplementalRepleted Deficient

Experimental

Groups

FIG. 3. Dietary copper and elastin-derived peptides in aorta extracts from chicks following copper repletion. Part A depicts immunoblots of protein from aorta extracts from chicks either deficient (Cup), supplemented (Cu+), or deficient and then repleted with copper (Repleted). These immunoblots corroborate the ELISA values given in Fig. 1. Part I3 are the data for the release ofpeptides when aorta insoluble elastin from deficient, supplemented, or repleted chicks was used in assays in vitro. The insoluble elastin was isolated as described by Buckingham et al. (15), and no peptides were released without a proteinase source. For the assays, chicken serum provided the source of elastolytic proteinase activity. Each assay consisted of 500 pg insoluble elastin, 200 ~1 chicken serum, and 100 ~1 Tris buffer, pH 7.8. The samples were incubated for 60 min at 41°C. An aliquot of the supernatant fraction (50 ~1) was assayed for soluble elastin by an ELISA.

There was a significant increase in the lysine content of elastin from copper-deficient chicks compared to elastin from copper-supplemented chicks at Day 14; corresponding A-LNL estimates were reduced in elastin samples from copper-deficient chicks (Table II). Values for lysine and A-LNL from elastin samples for repleted chicks (Day 14 to 18) were intermediate between corresponding values for normal chicks versus the chicks maintained on copper-deficient diets. Surprisingly, these changes were not accompanied by significant changes in ACP. However, the lack of statistical significance was due in part to the high variability of control estimates. With respect to crosslinking, elastin degradation effected by chick serum was most pronounced for the samples with an elevated lysine content and abnormally depressed A-LNL content. This process, however, involved more than accessibility of lysyl function. The chemical modification of lysyl groups in elastin by citraconylation did not decrease the potential for elastolysis. Unmodified elastin and citraconylated insoluble elastin from copper-deficient chicks (before and after deblocking) were not particularly different with respect to their ability to serve as substrates in assays in vitro (Table III). If anything, citraconylation enhanced elastolytic susceptibility. All of the insoluble elastin samples from copper-deficient chicks remained 4 to 5 times more susceptible to proteolysis than insoluble elastin for normal

chick aorta. It could also be inferred that cessation of proteolysis during recovery may reflect changes in exposure to proteinases. Isolated elastin from repleted chicks (48- or 96-h estimates) was still vulnerable to elastolysis in vitro (Table II), whereas no signs of elastolysis in uiuo were observed in corresponding Western blots (Fig. 3A). DISCUSSION

Previous investigations have shown that elastin synthesis in arterial tissue is proportional to elastin-specific mRNA levels (33-35). Further, in aorta from copper-deficient or copper-sufficient chicks, elastin-specific messenger RNA levels and elastin synthesis in cell-free translation and in tissue culture systems were the same despite copper depletion or tissue tropoelastin concentrations that were varied over a 5- to 20-fold range (13). Consequently, it was assumed for the work herein that the decrease in elastin accumulation associated with copper deficiency was due primarily to extracellular proteolysis, and not changes in synthesis. At the level of light microscopic examination, decreased cellularity and signs of elastic fiber fragmentation were observed following copper depletion. In contrast, an apparent increase in tissue cellularity occurred upon copper repletion. This change was associated with the increase in elastin accumulation that occurred during the recovery phase following copper repletion. Al-

TINKER TABLE Incorporation

of [“H]Valine

ET AL.

I

into Chick

Aorta

permeability increases in chronic lathyrism. Changes in permeability could represent one possible mechanism for serum proteinase infiltration into vessels. Jensen et al. (37) have also shown that D-penicillamine, another inhibitor of crosslink formation, induces angiopathy in rats with accompanying elastin fragmentation. Decreased crosslinking may alter permeability so as to increase both the amounts and rate of proteinase infiltration into vessels. Further, it was noted that in this study and a previous study (14) phagocytic cells, e.g., monocytes and macrophages, were not observed in any of the tissue sections examined histochemically. Although sources of elastolytic activity obviously include proteinases secreted from arterial cells and phagocytic cells, the observations in this report and those reported previously (14) clearly suggest that an additional source of elastolytic activity is serum. Stone et al. (5) have observed that elastin repair involves more than enzymatic debridement of elastin fibers followed by incorporation of tropoelastin into elastin. The reutilization and reinsertion of elastin fragments in the elastin was observed in cultured rat aorta smooth muscle cells following proteolytic injury. In their studies, over 90% of the elastin (prelabeled with [“Hllysine prior to proteolytic injury) was converted from hot alkali (0.1 M NaOH)-extractable (immediately following injury) to alkali-resistant protein following repair in culture. In addition, frayed elastic fibers, observed at the electron microscope level following injury were replaced immediately during the repair process by continuous bands of elastin fibers that resembled those in control cultures (5, also cf ref. (10)). Our observations are also consistent with some reutilization of elastin, since virtually all of the soluble elastin in aorta, including fragments arising from degradation, appeared to be rendered insoluble upon copper repletion. Further, the daily loss of soluble elastin from aorta was about equal to the tissue pool. Thus, we estimate the turnover or utilization of the soluble elastin pool to be on the order of 12 to 24 h in the chick aorta during periods of rapid elastin accumulation.

Proteins”

Radioactivity (10m5dpm per 100 mg) Aorta Treatment

Incubations

No BAPN BAPN added

1.5 h 1.5 h 1.5 h 1.5 h

Soluble

(pulse) (pulse + 4 h chase) (pulse) (pulse + 4 h chase)

3.3 1.7 3.0 2.0

* 0.8” + 0.5h f 0.6 +- 0.7h

Insoluble 0.8 2.4 0.3 0.9

+ f * +

0.4h 0.2” O.lh 0.5h

“Following a 1.5-h incubation in the presence of [“Hlvaline (see text), one-half of the aortae in each of four incubation flasks (per treatment group) were taken and immediately homogenized in 0.1 M NaOH at 4°C. Radioactivity was estimated associated with soluble protein and an insoluble pellet was recovered by centrifugation. Prior to being counted, the samples containing soluble proteins were neutralized with 0.1 M HCl and 0.05 M sodium phosphate buffer (pH 7.0), and unincorporated [“Hlvaline was removed by gel filtration utilizing columns of G-10 Sephadex. The pellet was washed extensively with 6% trichloroacetic acid containing 50 mM L-valine prior to counting. The remaining aortae were incubated further in nonlabeled medium in order to obtain a relative estimate of labeled protein distribution after a chase period (4 h). From previous experiments, it has been estimated that about 70-80% of the [“Hlvaline-labeled soluble protein is tropoelastin (13). Values represent the means + 1 SEM for three to four determinations. When added, BAPN, an inhibitor of lysyl oxidase, was present throughout the incubations at 50 @g/ml of medium. Within a column, values with a differing superscript are different at P i 0.05.

though the accumulation was correlated with an increase in aorta weight and growth of repleted chicks, the enhanced rate of elastin accumulation was at times 3- to 4-fold greater than corresponding control rates. Without repletion, chicks rendered copper-deficient died within 3 to 4 weeks. Most of the deaths could be attributed to aortic aneurysms, abdominal hemorrhages, or other connective tissue disorders and signs previously associated with the decreased crosslinking of collagen and elastin (l&16, l&20). With respect to possible mechanisms for the elastolysis, Pieraggi et al. (36) have reported that endothelial TABLE Effect

Dietary

treatment

Cu-supplemented Cu-deficient Cu-deficient, repleted 48 h Cu-deficient, repleted 96 h

of Diet

and Elastin

Crosslinking

II

on Elastolytic

Susceptibility

by Serum

Amount released (ng elastin)

ACP content (dpm X 10m5/mg)

&LNL content (dpm X 10-“/mg)

Lysine content (mol/lOOO total residues)

185od 926-1034” 383-495h 116-254”

5.2 f 1.2” 4.1 f 0.3”

1.8 * 0.16” 0.8 k 0.03h

6.9 i 0.7h 11.5 f 0.3”

4.3 f 0.5”

1.4 * 0.2+

8.9 t 0.4”,b

Note. The elastolytic assay is described in the text and the legend to Fig. 3. The values for ACP and A-LNL represent estimates following reduction with NaB3H, and chromatographic separation. Within a column, values with a differing superscript are significant at P < 0.05. Values are the averages of three determinations f SEM per assay.

ELASTIN TABLE

METABOLISM

AND

III

Effect of Chemical Modification on Elastolytic Susceptibility of Insoluble Elastins from Copper-Supplemented and -Deficient Chicks

Dietary

treatment

Cu-supplemented Cu-deficient Cu-deficient Cu-deficient

Chemical

modification

None None Citraconylation Citraconylation, by deblocking

followed

Amount of elastin released per assay (ng) 29 155 226 184

+ 1.0’ -t 10.1” 2 6.3” f 16.9”

Note. The assay conditions for estimating elastin peptide release are described in the legend to Fig. 3. Each value is the average of four separate incubations 2 1 SEM. Values with a differing superscript are significant at P < 0.05.

The availability of elastin samples, each with a different susceptibility to plasma proteases, made it possible to study selected features responsible for elastolytic resistance. The increased availability of lysyl functions in partially crosslinked elastin at elastolytic sites was considered one possibility. Blocking accessibility to peptidyl lysine by citraconylation, however, did not decrease proteolytic susceptibility of partially crosslinked elastin. Consequently, lysyl residue availability did not appear to be a factor. We also focused on the chemically reducible crosslinks, ACP and A-LNL, since they serve as precursors to polyfunctional crosslinks, such as desmosine. Though ACP did not change, &LNL was reduced to one-half of normal values upon copper depletion. A-LNL could be important to conferring resistance to elastolysis in several ways. First, A-LNL and its reduction product, LNL, are uniquely situated in the crosslinking domains of elastin (38). Second, in addition to serving as a desmosine precursor, it is the reduction of A-LNL to LNL that appears to coincide with and facilitate the oxidation of dihydroxydesmosines to desmosines (38,39). Third, A-LNL has been proposed as a possible crosslinkage between elastin and microfibrillar components of elastic fibers (40). Structural alterations of elastin through A-LNL formation, or alternatively the covalent attachment of microfibrillar components at sites critical to proteinase resistance, may represent important possibilities. Regardless of mechanism, our observations are important in that they underscore that the degree of crosslinking or state of elastin maturation is as important to elastolysis as the presence of proteinases with elastolytic activity. This work also corroborates observations using a uterine model to study elastin synthesis and degradation that signs of degradation were closely related to the relatively low levels of crosslinking amino acids in uterine elastin (41). In particular, the relative degree of elas-

CROSSLINK

FORMATION

331

tin maturation may be very important to elastic fiber repair, when inhibition of lysyl oxidase is a component of a biochemical lesion (12,42). The copper depletion and repletion model offers unique possibilities for the study of arterial elastin in ho. For example, elastolytic repair processes that involve reinsertion of elastin peptides into elastic fibers may give rise to fibers with abnormal mechanical properties. The extent to which this may be important to the continued function of such fibers deserves attention and may be studied in viva using the model. The changes that occur are also distributed throughout the vessel and are temporally influenced. As such they are amenable to biochemical or developmental investigations. Additionally, a copper-deficient and repletion model, or similarly, an experimental lathyrism and recovery model may provide the opportunity to study in uivo the function of a recently reported elastin receptor transduction system (43). ACKNOWLEDGMENTS This work was supported by National Institutes of Health Grants HL 15956 and AM 25358 and in part by a grant from the USDA.

REFERENCES 1. Rucker, R. B., and Dubick, M. A. (1984) Enuiron. Health Perspect. 55,179-191. 2. Sandberg, L. B., Soskel, N. T., and Leslie, J. G. (1981) N. Engl. J. Med. 304,566-579. :?I. Sandberg, L. B., and Davidson, J. M. (1984) Pept. and Protein Reu. 3, 1699226. 4. Kagan, H. M. (1986) in Biology of the Extracellular Matrix (Mecham, R. P., Ed.), pp. 3222398, Academic Press, New York. -5. Stone, P. J., Morris, S. M., Martin, B. M., McMahon, M. P., Faris, B., and Franzblau, C. (1989) J. Clin. Znuest. 82,1644&1654. 6. Lefevre, M., and Rucker, R. B. (1980) Biochim. Biophys. Acta 630,519-529. 7. Lefevre, M., and Rucker, R. B. (1983) Biochim. Biophys. Acta 743,338-342. 8. Dubick, M. A., Keen, C., and Rucker, B. R. (1985) Exp. Lung Res. 8,227-241. 9. Ross, R., and Glomset, J. A. (1973) Science 180, 133221339. 10. Kuhn, C., III, Yu, S. Y., Chraplyvy, M., Linder, H. E., and Senior, R. M. (1976) Lab. Invest. 34,372-380. 11. Kuhn, C., III, and Starcher, B. C. (1980) Amer. Reu. Respir. Dis. 122,453%460. 12. Clark, J. G., Kuhn, C., McDonald, d. A., and Mecham, R. P. (1983) Znt. RPU. Connect.

Tim.

Res. 10, 249-320.

13. Tinker, L)., Geller, J., Romero, N., Cross, C. E., and Rucker, R. B. (1986) Biochem J. 236,17-23. 14. Romero, N., Tinker, D., Hyde, D., and Rucker, R. B. 91986) Arch. Biochem.

Biophys.

244,161-168.

15. Buckingham, K., Khoo, C. S., Dubick, M., Lefevre, M., Cross, C., cJulian, L., and Rucker, R. (1981) Proc. Sot. Exp. Bid. Med. 166, 310-319. 16. O’Dell, B. L., Hardwick, B. C., Reynolds, G., and Savage, J. E. (1961) Proc. Sot. Enp. Biol. Med. 108, 402-405. 17. Rucker, R. B. (1982) in Methods in Enzymology (Cunningham, L. W., and Frederiksen, D. W., Eds.), 82,650-657.

332

TINKER

18. Tinker, D., and Rucker, R. B. (1985) Physiol. Reu. 65,607?657. 19. Clegg, M. S., Keen, C. L., Lonnerdal, B., and Hurley, L. S. (1981) Biol. Trace Elem. Res. 3, 107-115.

20. Lefevre,

M., Heng, H., and Rucker, 1344-1352.

R. B. (1982) J. Nutr.

21. Laemmli, U. K. (1970) Nature (London) 227,680&685. 22. Bowen, B., Steinberg, J., Laemmli, U. K., and Weintraub,

112,

H.

(1980) Nucleic Acids Res. 8, l-20.

23. Towbin,

H., Stakelin, T., and Gordon, J. (1979) Proc. Natl. Acad. Sci. USA 76,4350-4354.

24. Mecham, R. P., and Lange, G. (1982) in Methods in Enzymology (Cunningham, L. W., and Frederickson, 744-759, Academic Press, San Diego.

D. W., Eds.), Vol. 82, pp.

25. Voller, A., Bidwell, D., and Barlett, A. (1980) in Manual of Clinical Immunology (Rose, N. R., and Friedman, Academic Press, New York.

H., Eds.), pp. 359-371,

26. Heng-Khoo, C. S., Rucker, R. B., and Buckingham, K. (1979) Biothem. J. 177,559-567. 27. Chamberlain, J. P. (1979) Anal. Biochem. 98,132-135. 28. Gibson, I., and Perham, R. N. (1970) Biochem. J. 116,843-849. 29. Last, J. S., Summers, P., and Reiser, K. M. (1989) Biochim. Biophys. Acta, in press. 30. Reiser, K. M., Tyler, W. S., Hennessy, S. M., Dominguez, and Last, J. A. (1987) Toxicol. Appl. Pharmacol. 89,314-322.

J. J.,

ET AL. 31. Keeley, F. W., and Johnson, D. J. (1983) Canad. J. Biochem. Cell Biol. 61,1079-1084. 32 Snider, R., Faris, B., Verbitzki, V., Moscaritolo, R., Salcedo, L. L., and Franzblau, C. (1981) Biochemistry 20,2614-2618. 33. Shibahara, S., Davidson, J. M., Smith, K., and Crystal, R. G. (1981) Biochemistry 20,6577-6584. 34. Davidson, J. M., Smith, K., Shibahara, S., Tolstoshev, P., and Crystal, R. G. (1982) J. Biol. Chem. 257,747-754. 35. Davidson, J. M., Hill, K. E., Mason, M. L., and Giro, M. G. (1985) J. Biol. Chem. 260,1901~1908. 36. Pieraggi, M. Th., Julian, M., and Bouissou, H. (1979) Virchows Arch. A. Pathol. Anat. Histopathol. 384,337-346. 37. Jensen, B. A., Chemnitz, J., Christensen, B. C., Junker, P., and Lorenzen, I. (1983) Acta Pathol. Microbial. Zmmunol. Stand. Sect. A 91,4033411. 38. Raju, K., and Anwar, R. A. (1987) J. Biol. Chem. 262,5755-5762. 39. Rich, C. B., and Foster, J. A. (1989) Arch. Biochem. Biophys. 248, 551-558. 40. Richmond, V. L. (1981) Biochim. Biophys. Acta 669, 193-205. 41. Sharrow, L., Tinker, D., Davidson, J., and Rucker, R. B. (1989) Proc. Sot. Exp. Biol. Med., in press. 42. Osman, M., Cantor, J. O., Roffman, S., Keller, S., Turino, G. M., and Mandl, I. (1985) Amer. Reu. Respir. Dis. 132,640-643. 43. Mecham, R. P., Hinek, A., Entwistle, R., Wrenn, D. S., Griffin, G. L., and Senior, R. M. (1989) Biochemistry 28,3716-3722.