encoded by exon 6 of the pro-a2(1) collagen gene (COL. 1 A2), and .... following folding of the triple helix and pepsin digestion, ... 11). hne 6 is control fibroblast collagen containing Type I and. Type 111. ... 2:1 ratio. Repeat proteolytic digestion.
THEJOURNAL OF BIOLOGICAL CHEMISTRY 0 1987 by The American Society for Biochemistry and Molecular Biology, Inc.
Ehlers-Danlos Syndrome Type
Vol. 262, No. 34, Issue of December 5, pp. 1637616385,1987 Printed in U.S.A.
VIIB
DELETION OF 18 AMINO ACIDS COMPRISING THE N-TELOPEPTIDEREGION OF A PRO-aS(1) CHAIN* (Received for publication, May 15, 1987)
Mary K. WirtzSg, Robert W. GlanvilleSg, Beat Steinmannll, Velidi H. RaoS, and David W. HollisterSg11 ** From the $Research Department, Portland Unit, ShrinersHospital for Crippled Children, Portland,Oregon 97201, the Departments of §Biochemistry and 11 Medicine, Oregon Health Sciences University, Portland,Oregon 97201, and the llDepartment of Pediatrics, University of Zurich, Zurich, Switzerland
A patient with Ehlers-Danlos syndrome Type VIIB neous hyperextensibility, joint hypermobility, and easy bruiswas found to have an interstitialdeletion of 18.amino ability, and these features are associated with a wide variety acids in approximately half of the pro-a2(I) chains of of other manifestations (herniae, prolapses, easily torn skin, Type I procollagen. Analysis of pepsin-solubilized tis- rupture of blood vessels, and so forth) whose common theme sue and fibroblast collagen revealed an abnormal ad- is the diminished biomechanical integrity of various connecditional chain, a2(I)’, which migratedin sodium dode- tive tissue matrices. Variations in the cardinal features, difcy1 sulfate-5% polyacrylamide gel electrophoresis between the normalal(1)and a2(I)chains. The apparent fering modes of inheritance, and, in particular, the presence ratio of normal al(1):mutant (u2(I)’:normal a2(I) was of distinctive associated features serve to distinguish the nine 4:l:l. Procollagen studies and enzyme digestion stud- types of EDS currently recognized. In EDS Type VII, the ies of native mutant collagen suggested defective re- clinical hallmark is extreme joint laxity, and patientstypically moval of the aminopropeptide. Sieve chromatography present with bilateral hip dislocations at birth andmay suffer of CNBr peptides from purified a2(I)’ chains revealed recurrent joint dislocations thereafter. Skin hyperextensibilthe absence of the normal amino telopeptide fragment ity and easy bruisability are associated with minimal cutaCB 1 and the appearance of a larger new peptide of neous fragility, and bone appears to be clinically normal (1approximately 60 residues (CB X). Compositional and 3). Early studies of three patients with EDS Type VI1 found sequencing studies of this peptide identified normal that connective tissues contained, in addition to the normal amino propeptide sequences. However, the most car- al(1) and a2(I) chains of Type I collagen, larger precursor boxyl-terminal tryptic peptide of CB X differed sub- chains of Type I collagen not found in normal tissues. Initially stantially in composition and sequence from the ex- identified as procollagen chains,thesematerials are now pected and was found to have an interstitial deletion considered to represent (modified) pN-al(1) and/or pN-a2(I) of 18 amino acids corresponding to the N-telopeptide chains, intermediates in the processing of Type I procollagen of the pro-a2(1) chain. This deletion removes the normal sites of cleavage of the N-proteinase andalso re- to collagen which retain their respective amino propeptide moves a critical cross-linking lysine residue. The 18 extensions. Since a specific enzyme, procollagen N-proteinamino acids deleted correspond exactly to the residues ase, is required to cleave the propeptide extensions, these encoded by exon 6 of the pro-a2(1) collagen gene (COL patients were considered to have a deficiency of this enzyme 1A2), and, therefore, the protein defect may be due to (2, 4). The pathogenesis of this disorder was reported to be a genomic deletion, or alternatively, anRNA splicing defective and incomplete removal of the amino propeptide of Type I collagen, and an analogy drawn with an apparently defect. similar disease of cattle and sheep (dermatosparaxis) (5) in which a deficiency of the procollagen N-proteinase had been demonstrated (6). The Ehlers-Danlos syndromes (EDS)’ form a clinically and Subsequent detailed study of one of these patients provided genetically heterogeneous group of human inheriteddisorders evidence for a structural mutation of half of the a2(I)chains of connective tissue. The cardinal features consist of cutaaccumulating in tissue. The abnormal a2(I) chain exhibited * This work was supported by grants from the Shriners Hospitals delayed electrophoretic migration as compared with the norof North America, the Louis Gerlinger, Jr. andBeatrice Lee Gerlinger mal a2(I) chain and resembled a pN-a2(1)chain. Digestion of Request Awards, and by Swiss National Science Foundation Grant fibroblast procollagen with N-proteinase excised only about 3.861.086. The costs of publication of this article were defrayed in half of the pro-a2(1) aminopropeptide; digestion with pepsin part by the payment of page charges. This article must therefore be yielded a similar result. These data suggested a structural hereby marked “advertisement” in accordance with 18U.S.C. Section mutation in or near the normal sites at which these enzymes 1734 solelyto indicate this fact. ** To whom correspondence should be addressed Research Dept., normally cleave the pro-a2(I) chain to remove the amino Shriners Hospital for Crippled Children, 3101 S. W. Sam Jackson Park Rd., Portland, OR 97201. ’ The abbreviations used are: EDS, Ehlers-Danlos syndrome; proal(1) and pro-a2(I), type I procollagen chains bearing amino- and carboxyl-terminal extension propeptides; pN-cul(1) and pN-aB(I), partially processed type I procollagen chains that retain their aminoterminal extension peptides; al(1) and a2(I), fully processed type I collagen chains; a2(1)’, the mutantaZ(1)chain from the patient; proa2(I)’, thecorresponding procollagen chain; N-telopeptide, nontriple helical domain between the amino propeptide and the major helix;
CB 4,CB 3-5 etc., designation of CNBr-derivedpeptides; SDS, sodium dodecyl sulfate; PAGE, polyacrylamide gel electrophoresis; DMEM, Dulbecco’s modified Eagle’s medium; procollagen N-proteinase, the specific proteinase that cleaves the amino propeptide from pN and procollagen; TPCK, ~-l-tosylamido-2-phenylethyl chloromethyl ketone; HPLC, high pressure liquid chromatography; TLCK, Nu-ptosyl-L-lysine chloromethyl ketone; PTH, phenylthiohydantoin; PMSF, phenylmethanesulfonyl fluoride; TFA, trifluoroacetic acid; PITC, phenylisothiocyanate; RP, reverse-phase.
16376
N-telopeptide Deletion in
a Pro-cuB(I) Chain EDS in
VIIB
16377
propeptide. The parents were normal clinically, and their fibroblast collagen was normal, suggesting that the patient represented a new dominant mutation (7). Detailed characterization of this mutation has not yet been reported. Two additional unrelated patients with similar heterozygous defects of a pro-a2(I) chain due to presumed structural alterations have now been reported; the first of these patients is the subject of the present communication (8). Eyre et al. (9) reported the second case. Comparison of CNBr peptides from normal pN-a2(I) chains and the abnormal larger a2 chain provided evidence that the abnormal chain contained an amino-terminal CNBr peptide which was 15-20 residues smaller than thecomparable normal peptide. These data were interpreted as indicating the probable deletion of the NFIG. 1. The Ehlers-Danlos Type VIIB patient. Extreme hytelopeptide domain connecting the amino propeptide to the permobility of major and minor joints is well illustrated; the skin was major helix. soft, velvety, and hyperextensible. While the present studies were in progress, a 3-month-old patient with features of EDS VI1 was found to have a deletion 1 2 3 4 5 6 7 0 9 of 24 amino acids of a pro-al(1) chain (10). In addition to the normal al(1) chain, equivalent amounts of a larger pN-al(1)like chain were observed in collagen extracted from skin or produced by dermal fibroblasts. The interstitial deletion was shown by amino acid sequencing to consist of the entireamino telopeptide domain of an al(1) chain. The deletion removed the normal procollagen N-proteinase cleavage site as well as an important cross-linking lysine residue. The deleted amino acids are normally coded by a single exon of the pro-al(1) gene (11). The parents were normal, suggesting that the patient was a new mutation. This report documents detailed amino acid sequencing studies of a mutant pro-a2(1) chain from a patient with EDS VIIB.2 MATERIALS AND METHODS3 RESULTS AND DISCUSSION
Clinical Phenotype-The patient was a 1-year-old girl, the second child born to nonconsanguineous young parents of Libyan extraction. Both parents and an older and younger sibling were clinically normal, and paternity was ascertained bybloodgroup analysis. Clinical findings included severe generalized joint hypermobility with bilateral hip dislocations at birth requiring several operative procedures and velvety hyperelastic skin whichwas only moderately bruisable or fragile. Fig. 1demonstrates the remarkable joint hypermobility exhibited by thispatient.Further clinical detailsare contained in Ref. 8. Electrophoretic Analysis of Fibroblast and Dermal Collagens-Pepsin-digestedcollagen produced by dermal fibroblasts or liberated from the patient’s skin yielded the expected al(1) and a2(I) chains of Type I collagen and an additional collagenous chain (designated a2(I)’) migrating on electro-
FIG. 2. SDS-5% PAGE of procollagen a n d collagen from controls a n d the patient w i t h E D SVIIB. A, fluorogram of radiolabeled disulfide-reduced intracellular collagenous proteins synthesized by control ( l a n e I ) and EDS VIIB fibroblasts ( l a n e 2) in the presence of 0.5 m M a,a’-dipyridyl. B, fluorogram of radiolabeled pepsin-treated media procollagen synthesized by control ( l a n e 3) and EDS VIIB ( l a n e 4 ) fibroblasts. C, Coomassie Brilliant Blue stained pepsin-solubilized skin collagens from control ( l a n e 5)and EDSVIIB ( l a n e 6). D, directly extracted skin collagens stained with Coomassie Brilliant Blue. Control is lathyritic rat skin collagen ( l a n e 7).Lune 8 shows an SDS extract of patient skin, and lane 9 is pepsin-solubilized collagens from patient skin.
phoresis between these normal chains and close to the expected position for a pN-a2(I) chain. Fig. 2, B and C,depicts pepsin-solubilized fibroblast and skin collagen, respectively, and is compared with similarly prepared normal control collagens. In contrast to the expected 2:l ratio of al(I):a2(1) found in normal collagen, the observed chain ratio of al(I):a2(I)‘:a2(1) was approximately4:1:1,suggestingthat the patient was heterozygous for a larger abnormal a2 chain * Emerging biochemical and structuralinformation has resulted in (compare lanes 3 and 4, and 5 and 6 of Fig. 2). The electroconflicting subclassifications of variants of EDS Type VI1 (12-14). A phoretic mobility of the (u2(1)’ chain is just slightly greater recent international committee has proposed a uniform nosology in than that of the normal pN-a2(I) chain (data not shown). which EDS VIIA is caused by structural defects of pro-al(I), EDS Estimates of the molecular size of the a2(I)’chain based on VIIB by defects of pro-a2(1) and EDS VIIC by deficiency of procolelectrophoretic mobility suggested the addition of approxilagen N-proteinase activity (15). This communication will use this mately 40 amino acid residues as compared with the normal proposed nomenclature. ’Portions of this paper (including “Materials and Methods,” Figs. a2(I) chain. The mutant a2(I)’ chain is not an artifact of 9-13, and Tables I11 and IV) are presented in miniprintat the end of pepsin digestion, since direct extraction of skin with SDS this paper. Miniprint is easily read with the aid of a standard yields a similar abnormally migrating chain (Fig. 20). It is magnifying glass. Full size photocopies are available from the Journal apparent that SDS extraction yields al(1) and mutant a2(I)‘ of Biological Chemistry, 9650 Rockville Pike, Bethesda, MD 20814. chains in an apparent 2:l ratio; little normal a2(I) was soluRequest Document No. 87M-1633, cite the authors, and include a check or money order for $5.60 per set of photocopies. Full size bilized by this technique. Thesedata suggest that native photocopies are also included in the microfilm edition of the Journal collagen molecules containing the a2(I)‘chain accumulate in skin andare more soluble than theircounterpart normal that is available from Waverly Press.
16378
N-telopeptide Deletion in
a Pro-aB(I) Chain EDS in
molecules. Assuming that a2(I)’ represents a modified pNa2(I) chain, the persistence of this chain in tissue indicates that themutant chain cannot be cleaved in vivoby the physiologic enzyme procollagen N-proteinase. This enzyme has recently been shown to be normally active in fibroblast cultures from this patient (16). ProcollagenStudies-Unhydroxylatedcollagen (protocollagen) producedby fibroblasts incubated with a,a‘-dipyridyl, followingfolding of the triple helix and pepsin digestion, demonstrated an electrophoretic pattern similar to that observed for collagen extracted from skin, and approximately equal amounts of the mutant a2(I)’ and a2(1) chains were observed (8). Since a,a’-dipyridyl blocks thepost-translational hydroxylation (and glycosylation) of collagen, this result implies that the electrophoretic behavior of the a2(I)’ chain is due to differences in primary structure (i.e. insertion of peptide material) rather than post-translational modifications. Previous studies exploiting the asymmetric cleavage of collagen chains by mammalian collagenase indicated that the peptide insertion occurred in the amino-terminal threefourths of the a2(I)’chain (8). If such an insertion of peptide material has occurred, the initial biosynthetic product of the pro-a2(1)’ allele should demonstrate a similar, albeit smaller, electrophoretic difference ’from the normal pro-a2(1) chain as that observed for a2(I)’ and a2(I) chains. To explore this possibility, newly synthesized radiolabeled intracellular protocollagen chains from EDS VIIB fibroblasts treated with a,a’-dipyridyl were isolated and subjected to electrophoresis. As depicted in Fig. 2A, the apparent mobility and apparent relative amount of the EDS proto-pro-al(1) and -pro-a2(1) chains (lane 2) were essentially identical to those of normal fibroblast proto-proa chains (lane 1) produced under identical conditions. These data dispute the hypothesis that an insertion of about 40 amino acids has occurred in half of the pro-aP(1) chains and indicate that a defect inthe pepsin-mediated and procollagen N-proteinase-mediatedconversion of pro-a2(1)’ chain to normal a2 chains is responsible forthe observed results. Collagen studies of the parents were normal, suggesting that thepatient represented a new dominant mutation (8). Isolation of Native Mutant Collagen and the Mutant a2(I)’ Chain-As described in the appended Miniprint, total procollagens wereisolated from large-scale roller bottle cultures of EDS fibroblasts, subjected to limited pepsin digestion, and purified by native C-18 RP and DEAE chromatography. Fig. 3 depicts the results obtained at various steps in the purification scheme. Reverse-phase chromatography in the native state yielded a complex elution profile in which collagenous proteins appeared in the early portions of the gradient (Fig. 9, Miniprint). Electrophoretic analysis of the pooled materials from this chromatogram (Fig. 9, bar) is shown in lane 1 of Fig. 3. In addition to the normally migrating chains of al(1) and a2(I), the a2(I)’ chain is apparent, andType I11 collagen is present as disulfide-bonded trimers barely entering the gel. Stepwise DEAEchromatography of total reverse-phase-purified EDS collagens isdemonstrated in Fig. 10 of the Miniprint, and various pools analyzedin Fig. 3 are indicated. Lane 2 of Fig. 3 depicts the collagenous proteins which did not bind to the column under the initial buffer conditions (Pool 1).The major materials are normally migrating chains of Type I and I11 collagen together with a relatively smaller amount of the a2(I)’ chain. Under these chromatographic conditions, normal Types I and I11 collagen do not bind to this column. Collagenousproteins binding to theDEAE columnand eluted with 125 mM Tris-HC1 buffer are analyzed in Fig. 3, lane 3. Essentially, the only collagen chains observed are the al(1)
1 Type
VIIB
2
3
4
5
6
III
FIG. 3. SDS-5% PAGE stained with Coomassie Brilliant Blue demonstrating purification of a2(I)‘ from EDS VIIB fibroblast collagen. The samples were not disulfide-reduced prior to electrophoresis. Lune I is native RP chromatography isolation of whole collagen (Fig. 9, bar). Lunes 2 and 3 showpools 1 and 3, respectively, from DEAE chromatography (see Fig. 10). Lune 3 has been deliberately overloaded. Lune 4 shows pool 1 and lane 5 shows pool 2 of denaturing RP separation of al(1) and a2(I)’ chains (see n e 6 is control fibroblast collagen containing Type I and Fig. 11).h Type 111. The identity of the observed bands is indicated a t left.
and a2(I)’chains in a 2:1 ratio. Repeat proteolytic digestion of this material followed by electrophoretic analysis did not alter the apparent amounts or mobility of these chains, demonstrating that the recovered mutant collagen retained the native conformation. Native mutant collagen may be resolvedinto the constituent a chains by denaturing reverse-phase chromatography as described in the Miniprint and illustrated in Fig. 11 (Miniprint). Electrophoretic analysis of the individual peaks represented by pools 1 and 2of Fig. 11are shown in Fig. 3, lanes 4 and 5, respectively. Lane 4 consists of the al(1) chain, and lane 5 consists exclusively of the a2(I)’chain. Minor amounts of small molecular weightpeptides were found to coelute with these chains and were routinely removed by sievechromatography prior to further analysis. Enzymatic Digestion of Native Mutant Collagen and Amimterminal Sequencing of the Product a2(I)’ Chains-Available evidence suggested that the mutational event producing the a2(I)’ chain was not an insertion of peptide material but rather a small structural change which blocked cleavage of the amino propeptide. This could be due to amino acid substitution(s) at or near the cleavage sites of the procollagen Nproteinase and pepsin in the N-telopeptide. To gain further information, native mutant collagen (prepared by pepsin digestion) was subjected to additional proteolysis with enzymes known to cleave the normal &(I) telopeptide domain (17) and then analyzed by SDS-PAGE. In addition, the constituent a chains were purified by denaturing RP chromatography, and the mutant a2(I)’chain was subjected in toto to amino-terminal amino acid sequencing. Repeat pepsin digestion did not alter the mobility or apparent amount of native mutant collagen. Under conditions yielding reproducible amino-terminal amino acid sequences from normal pepsin-solubilizeda2(I) chains, no reproducible sequences were obtained from the a2(I)’ chain despite repeated attempts. Instead, low levels of (presumably back-
N-telopeptide Deletion
in a Pro-aB(I) C h i n in EDS VIIB
ground) sequences wereobtained which varied with differing preparations. The reasonfor this result is not clear but presumably reflects a blocked amino terminus. Trypsin or chymotrypsin digestion of native mutant collagen did not alter the electrophoretic mobility of the al(1) or a2(I)’chains under conditions sufficient to completely cleave normal native pN or procollagen to collagen (data not shown (8)).However, after purification, the trypsin product aP(1)’ yielded a clean amino-terminal sequence beginning with a lysine residueand continuing for 23 residues. Comparison of this sequence with the normal a2(I) amino propeptide sequence predicted by the nucleic acid sequence revealed complete identity from residues 10-33 (Fig. 7, TI sequence). This sequence was repeated four times on the same or similarly prepared materials. The apparent anomaly of a trypticdigestion product containing an NHp-terminal lysine residue is presumably due to preferential cleavage betweenArg and Lys at positions 9 and 10. Thermolysin digestion of native mutant collagen resulted in a time-, temperature-, and enzyme concentration-dependent conversion of a2(I)’ chains to a limit product which comigrated with normal a2(I) chains or normal a2(I) chains similarly treated with thermolysin. Fig. 4 demonstrates the enzyme concentration dependence of this conversion. Following limit thermolysin digestion, the mutant a2(I)’ product was recovered and yielded an amino acid sequencebeginning with leucine (residue 73) followed by methionine and continuing for18 residues (Fig.7, Th sequence). Comparison of this sequence with the normal nucleic acid-derived sequence of the pro-a2(1) chain revealed an exact match from residue 73 to 92; this sequence defines the beginning of the major triple helical domain of the pro-a2(1) chain (Fig. 7). In addition, thermolysin-prepared normal a2(I) chains yielded an identical electrophoretic mobility and initial sequence (data not shown), indicating that both mutant and normal a2 chains were cleaved at the same site by this enzyme. The studies of enzymatic digestion of native mutant collagen demonstrated that, like the pepsin site(s), the cleavage sites of trypsin and chymotrypsin in or near the N-telopeptide domain were missing or inaccessible to enzyme whereas the thermolysin cleavage site was retained. The amino acid sequencing of the a2(I)‘ products producedby trypsin and thermolysin digestion demonstrates retention of N-propep-
16379
tide sequences and suggests that the structural alteration is NH2-terminal to themajor triple helical domain. Fragmentation and Sequencing of the a2(I)’ Chin-Pepsinsolubilized a2(I)‘ chains were isolated, further purified by sieve chromatography, and digested with CNBr. The resultant peptides were resolved by sieve chromatography on a tandem TSK 125column. Fig. 5 illustrates the elution profiles of CNBr peptides derived fromthe normal a2(I) chain compared with similar peptides from themutanta2(I)’ chain. The identifications of the observed chromatographic peaks are based upon electrophoretic mobility in the case of CB 4, and amino acid compositions and sequencing for CB 4, CB 2, and CB 1. The ratios of the elution volume to total volume for CB 4 and CB 2 derived from a2(I) and a2(I)’ were identical. The amino acid compositions and a partial amino acid sequence for CB4 (residues 75-96, see Fig. 7) and acomplete sequence for CB2 derived fromthe a2(I)’chain are normal (Tables I11 and IV, Miniprint). Two major chromatographic differences are apparent in the a2(I)’ CB peptide profile: the CB 1 peak representing the normal amino telopeptide is missing and a new peak (labeled X)is present. This new peak is estimated to contain 58 f 4 residues from plots of log residue number uersus elution volume calibrated with standard a2(I) CNBr peptides (datanot shown). This peptide was assumed to represent a new amino-terminal CNBr peptide. Electrophoretic resolutions of a2(I)‘-derived CNBr peptides are compared in Fig. 6 with similar peptides from the a2(I) chain. This 7.5-12.5% gradient gel is deliberately overloaded to emphasize minor peptides. The major peptides, CB 3-5 and CB 4, comigrate with their normal counterparts. Two minor peptides derived from the a2(I)’ chain migrate some-
1 2 3 4 5 6 7 8 9 1 0
0.041 0.02
01
’
0.500
RG.4. SDS-5%PAGE of thermolysin-digested native mutant collagen stained with Coomassie Brilliant Blue.The observed bands are identified at left. Lane I shows control fibroblast
collagenincubatedwiththermolysin (1:50) plusinhibitor (50 mM EDTA). Lane 2 is a sample of mutant collagen treated similarly. Lanes 3-10 show mutant collagen incubated with varying ratios 01 enzyme:substrate from1:12,800 in lane 3 and continuing 1:6,400, 1:1,600, 1:800,1:400, 1:200, and 1:100to a final 1:50 concentration in lane 10 for 24 hat 30 “Cin the absence of EDTA.
0.625
0.750
v.
I
0.875
1.000
I vt
FIG.5. Tandem TSK 125 sieve chromatography of CNBr peptides from normal a2(I) and mutant a2(I)’.For the normal a2(I) CNBr peptides,550 pg was appliedin 100 pl, and for the a2(I)’ CNBr peptides, 160 pg of CNBr peptides wasapplied in 100 pl. Individual peaks are identified and labeled as described in the text. The new peak present in the EDS VI1 chromatogram is labeled X andrepresents the NHt-terminal CB peptide of a2(I)’. V, is the elution volume, and V, is the total volume.
N-telopeptide Deletion in
16380
a Pro-aB(I) ChuinEDS in
what slower than CB 4 (Fig. 6, lane 1 ) . The first of these is slightly slower than CB 4, has a normal counterpart, and is identified as uncleaved peptide CB 4-2. The slower migrating minor peptide, however, has no normal counterpart and appears to be unique to the a2(I)’ chain. The estimated number of residues of this minor peptide from plots of log residue number uersus log %T (18) is 370, a value in reasonable agreement with the sum of residues of CB 4 and the estimate derived for the new amino-terminal CNBr peptide (321 and 58, respectively, yielding a sum of 379). This peptide is tentatively identified as the noncleaved peptide of these two CNBr fragments and is labeled X-4. The appearance of this minor peptide suggests that only one of the 2 methionine residues normally occurring between the N-telopeptide and helical domains of a2(I) is present in the mutantchain, since noncleavage at 2 methionine residues would be a relatively rare event. Since the methionine at residue 74 is present (as shownby sequencing), this implies that the mutant chain lacks the methionine at position 71 of the telopeptide (see Fig. 7) and, therefore, lacks the normal CB tripeptide CB 0. Fragmentation and Sequencing of the Amim-terminal CNBr Peptide (CB X) of a2fZ)’”The a2(I)’ CNBr peptide identified as theamino-terminal peptide (labeled X in Fig. 5)
1 F -
2 _ I
VIIB
was further purified by reverse-phase chromatography. The amino acid composition of this peptide is recorded inTable I and compared with the expected composition of the aminoterminal CNBr peptide of normal pN-a2(1) deduced fromthe nucleic acid sequence.Since the post-translational hydroxylation of proline to hydroxyprolinecannot be ascertained from the DNA sequence, the derived proline value should be compared with the sum of proline and hydroxyproline. Several differences are noteworthy, namely the absence of tyrosine and phenylalanine, which normally occur in the telopeptide domain. Also listed in Table I is an expected composition for an amino-terminal CNBr peptide lacking the amino telopeptide (residues 54-71). This composition, although differing slightly, is much more comparable to the observed composition of the CB X peptide. Attempts to sequence CB X yielded no reproducible sequences, probably due to blockage of the amino-terminal residue. When this peptide was digested with trypsin at an enzyme:substrate ratio of 1:10,000, a product was isolated which yielded a sequence identical to sequence T1 (see Fig. 7). These data established this peptide as the NH2-terminal peptide and confirmed the sensitivity of the Arg-Lys bondto trypsin cleavage. When the peptide was limit-digested with trypsin at 150 (enzymembstrate), adifferent peptide profile was obtained. These peptides were fractionated by reversephase chromatography and the major peptide rechromatographed and pooled as depicted in Fig. 12. A single sharp peak with minor asymmetry was obtained. The amino acid comTABLEI
3- 5
x-4 4- 2 4
Amino acid composition of the amino-terminalCNBr peptide (GB X) of a2(1)’ Comparison of the observed amino acid composition of the aminoterminal CNBr peptide(CB X ) of a2(I)’ isolated by sieve chromatography (see Fig. 5) with the calculated compositions of the aminoterminal CNBr peptide from a normal pro-a2(1) chain (“normal”) and of an amino-terminal CNBr peptidefrom a pro-a2(I) missing the telopeptide (listed in the “calculated” column). NormaI‘
Mutant CB X*
Calculated‘
Aspartic acid 5 3.1 (3) 3 7 6 Glutamic acid 3.6 (4) d d 5.8 (6) 4-Hydroxyproline 1 Serine 1 0.3 (0) 21 16.9 (17) 17 Glycine 1 1 1.0 (1) Homoserine‘ 5 Arginine 5 4.8 (5) Threonine 2 2 2.0 (2) 3 1.4 (1) Alanine 1 Proline 15 17 9.0 (9) 1 0.3 ( 0 ) 0 Tyrosine 2 1.0 (1) Valine 1 Leucine 2.0 (2) 3 3 d d 0.1 (0) Hydroxylysine 1 0 (0) 0 Phenvlalanine 2 0.8 il j 1 Lysine 0 Total 71 52 56 1 / 4 2 3 5 ’Composition of the amino-terminal CNBr peptide residues (1d a2111 71) derived from the human DNA sequence of the pro-a2(1) chain (20). * The observed amino acid composition is the average of uncorX 4 2 3 5 a21 11’ rected triplicate analyses of the presumed amino-terminalCNBr peptide (CBX ) of the a2(I)’chain. Complete recovery of homoserine FIG.6. 7.5-12.5% gradient SDS-PAGE of CNBr peptides was assumed. Integer values are shown in parentheses. from a2(I)’ a n d a2(I) stained with Coomassie Brilliant Blue. The predicted composition of an amino-terminal CNBr peptide Lune I contains the a2(I)‘ CNBr peptides, and lune 2 shows the consisting of residues 1-53, 72-74 compiled from the human DNA normal a2(I) CNBr peptides. Individual a2-derived CNBr peptides sequences of the pro-a2(1) (20). or uncleaved double peptides are labeled, and the minor uncleaved Post-translational modifications of proline and lysine could not X - 4 ) are described in the text. An alignment be determined from the DNA sequence. double peptides (4-2; map comparing the CNBr peptides of a normal a2(I) and the mutant e Methionine residue predicted by DNA sequence; put atunity for a2(I)’ is depicted below the SDS-PAGE. the amino acid analysis.
Pro-aB(I) Chain in EDS VIIB
N-telopeptide Deletion a in
position of this material (designated CB Tm) is recorded in Table 11. The peptide contains homoserine, while lysine, arginine, and methionineareabsent,indicating that this peptide is both carboxyl-terminal and fully cleaved. Assuming virtually complete recovery of homoserine, the peptide was found to contain 25-26 residues. This finding was in marked contrast to the expected 9 residues for the fully cleaved carboxyl-terminal tryptic peptide of the amino-terminal CNBr peptide of the normal pN-aZ(1) chain. Because of previous enzyme and CNBr cleavage studies, it seemed likely that a portion or all of the normal telopeptide domain was deleted in the mutant chain. To explore this possibility, the amino acid compositions of a series of carboxyl-terminal tryptic peptides with contiguous deletions of portions or all of the telopeptide sequence were calculated. The calculated composition of a deleted tryptic peptide starting at residue 31 of the normal propeptide and extending to residue 74 but deleting residues 54-71 matched the observed composition and number of residues. This composition is listed in Table I1 for comparison. This result suggested deletion of the entiretelopeptide region. Initial attempts tosequence the CB Tm peptide were frustrated by gradual loss of the peptide from the sequenator filter. Modification of the standard program as described in the accompanying Miniprint resulted in successful identification of 25 residues. The CB Tm tryptic peptide yielded a normal amino propeptide sequence from residues 31 to 53. However, instead of the Asn-Phe sequence expected at positions 54-55, a Gly-Leu sequence was found. This Gly-Leu sequence presumably originated from residues 72 and 73 (Fig.
TABLE I1 Amino acid composition of CB Tm Comparison of the amino acid composition of the carboxyl-terminal tryptic peptide isolated from the amino-terminal CNBr peptide from a2(I)’ with the expected composition of the normal carboxylterminal tryptic peptide from the amino-terminal CNBr of a normal pro-a2(1) chain and with a similar peptide missing residues 54-71. Mutant
Normal”
CB Tmb
16381
Calculated‘
Aspartic acid 1.75 (2) 2 Glutamic acid 0.55 (1) 1 d d 4.01 (4) 4-Hydroxyproline 4 9.98 (10) 10 Glycine 1 1.03 (1) 1 Homoserine‘ Threonine 0.84 (1) 1 Alanine 0.13 (0) Proline 2 5.10 ( 5 ) 9 Valine 1 0 (0) -1 2.08 (2j 2 Leucine 26 9 26 Total Composition of the carboxyl-terminal tryptic peptide of the amino-terminal CNBr peptide of normal pro-a2(I) (residues 64-72) (seeFig. 7) (20). *Observed composition of the carboxyl-terminal tryptic peptide isolated from the amino-terminal CNBr peptide from a2(1)’. This represents one of three independent analyses of the same peptide. Integer values are given in parentheses. No amino acids other than those listed were found. Predicted composition of residues 31-53,72-74 derived from the human DNA sequence of pro-a2(1) chain (see Fig. 7) (20). dPost-translational modifications of proline could not be determined from the DNA sequence. Methionine residue predicted by DNA sequence.
10
I1
TGC CAA TCTTTA CAA G M GAAACTGTAAGAAAG GGC CCA GCC GGAGATAGAGGA Asp ArgGly CysGlnSerLeuGlnGluGlu ThrValArgLysGlyProAlaGly ”
T1
T1 CB Tm
70
”
”
30 20 CCA CGT GGA GAA AGG GGT CCA CCAGGCCCC CCAGGCAGAGATGGT GAA GAT GGT Asp GlyGlu Asp Gly ProArgGlyGluArgGlyProProGlyProProGlyArg ” ” ” Hyp Hyp 7
--
- - -
7
”
* CB Tm CB Tmc
”
-
50
40
CCC ACA GGC CCT CCT GGT CCA CCT GGT CCT CCT GGC CCC CCT GGT CTC GGTGGG ProThrGlyProProGlyProProGlyProProGlyProProGlyLeuGlyGly HYP HYP Hyp -, H y P
-
” . - 9
--
I
6 5 55
-
-
- -
-
--
60
AAC TTT GCTGCT GAGTATGATGGA AAA GGA GTT GGA CTT GGC CCT GGACCAATG A s p GlyLysGly Val G l yL e u G l yP r oG l yP r oM e t Asn PheAlaAlaGlnTyr
75
80
N-TELOPEPTIDE
85
GGC TTAATG GGA CCT AGA GGC CCA CCT GGT GCA GCT GGA GCC CCA
GGC CCT CAA
Gly Leu Met Gly Pro Arg Gly Pro Pro Gly Ala Ala Gly Ala Pro Gly Pro Gln C B Tm CB Tmc Th
- AMINO P R O P E P T I D E
7
”
MAJOR H E L I X
” C
“
- - - ” ” H y p ” ”
-
Hyp
-,
-. -.
FIG. 7. DNA and derived amino acid sequence of normal human pro-a2(1) (20,21) and comparison with the determined amino acid sequences from a2(I)’. Lines I , 2, and 3 show the numbering of the amino acid residues, the nucleotide sequence, and the derived amino acid sequence, respectively. Post-translational modifications of proline or lysine could not be determined from the DNA sequence. Lines 4, 5, and 6 record the observed amino acid sequences of different peptides derived from the a2(I)’chain. The two arrows (1)indicate the cleavage sites of the signal peptidase and the N-proteinase, respectively. The half-arrows indicate identity of the observed a2(I)’ sequence with the DNA-derived amino acid sequence. All Y position proline residues predicted by the DNA sequence were found to be hydroxyproline. TI is the amino-terminal sequence of the a2(1)’ chain following trypsin digestion of native mutant collagen. Th is the comparable sequence following limit thermolysin digestion. CB Tm is the tryptic peptide shown in Fig. 12, and CB Tm, denotes the results of carboxypeptidase Y digestion of CB X (reverse half-arrows). The italicized amino acids correspond to the deleted amino acids in the a2(I)’ chain. * refers to the revised nucleotide and amino acid sequence misprinted in Ref. 20 (F. Ramirez, personal communication).
16382
N-telopeptide Deletion ina Pro-a2(I) Chain inEDS VIIB
7, sequence CB Tm). This resultwas entirely consistent with the expected deletion of residues 54-71, yielding a “fusion” peptide of residues 31-53,72-74. The amino acids recovered in the sequence account for thecomplete composition of the CB Tmpeptide with theexception of homoserine and, therefore, established homoserine as the next residue. Compositional studies notwithstanding, the formal possibility of a smaller deletion, namely residues 54-64, and the production of a “fusion”peptide 31-53,6571is not completely excluded by these data. The last two sequenced residues of CB T m (-Gly-Leu-) could be residues 65 and 66, respectively (see Fig. 7). To confirm the COOH-terminalsequence, CB X was digested withcarboxypeptidase Y (19),andthetime course and identityof the liberatedresidues were determined by amino acid analysis. As depicted in Fig. 13, the liberation of homoserine is followed by the appearance of free leucine and subsequently substantial amounts of glycine. Two additional independent experimentsyielded similar results. These data confirm that the COOH-terminal sequence is Gly-GlyLeu-Hse (where Hse is homoserine) and that the deletion involves 18 amino acids of the telopeptide domain, residues 54-71. The combined sequence data define 66 contiguous amino acid residues beginningat residue 10 of the amino propeptide and extending into the triple helical domain of the a2(I)‘ chain. The criticalsequence overlapping the deletion site has beenconfirmed by carboxypeptidase Y sequencing. Fig. 8 compares the normal aminoacid sequence of pro-a2(1) with the defined sequenceof the mutant pro-a2(1)’ chain. Residues 35-53 and residues 72-86 are present, normal and contiguous in the mutant a2(I)’ chain. The deleted telopeptide domain is depicted above the normal sequence. Consequences of the Deletion of the Telopeptide-Deletion of the telopeptide results in theloss of at least two important physiologic sites: the procollagen N-proteinase cleavage site (residue 57-58, Ala-Gln)andthecross-linking lysine site (residue 62) (see Fig. 8). In addition, the loss of this region explains the enzyme digestion results since pepsin, trypsin, and chymotrypsinalso cleave in the telopeptide domain. The result, either in vivo or in uitro, is retention of the amino propeptide of the pro-a2(1)’ chain. The apparent paradoxof a deletion yielding a larger protein is thus resolved. It is of interest that direct extraction of skin does not yield pN-al(1) chains (Fig. 2, lane 8 ) , suggesting that the N-proteinase or (less likely) nonspecific proteases cleave the pN-al(1) chains of mutant collagen in uiuo. Similarly, proteolytic digestion of mutant procollagen in uitro also converts pN-al(1) chains to d ( 1 ) chains, indicating that al(1) telopeptide cleavage sites are accessible. This report is the first case of EDS VIIB demonstrated by amino acid sequencing to be a deletion of the N-telopeptide in approximately half of the pro-a2(I) chains. The 18 amino acids deleted are specified by and form the entire segment encoded by exon 6 (54 base pairs) of the pro-a2(1) gene (see Fig. 8) (20, 21). This suggests either a genomic deletion of exon 6 from one allele or, alternatively, a missplicing event during mRNA maturation. It is noteworthy that the EDS VIIB patient of Eyre et al. (9) may well have an identical deietion, and the EDS VIIA patient of Cole et ai. (10) was shown to have a 24-residue deletion of the al(1) amino telopeptide correspondingexactly to thesegment encodedby exon 6 of the pro-al(1) gene (11). Furthermore, a patient with overlapping features of osteogenesis imperfecta and EDSVI1 was found tohave ashortened a2(I) chain (22) duedeletion to of all or most of exon 11 (54 base pairs) from mRNA by Rloop analysis (23). A recent report indicates that this mutant
exon 6 ,755
IV
NFAAOYDGKGVGLGPGP 1 \.., . ffi
exon 7
exon 5 35
normal mutant
. . . G P T G P P G P P45G P P G P P G L t t ‘ . ~ L M ~ P R G P P G A A 85t I P . . . GPTGPPGPPGPPGPPGLGG-GLMGPRGPPGAAGAP
FIG. 8. Comparison of the amino acid sequence of the normal a2(I)chain with the mutant a2(I)’chain from residues 35 to 86. The residues are numbered relative to the first residue of the pro-a2(1) chain. The telopeptide sequence, which has been deleted from the mutant chain, is shown above the normal sequence. The exons corresponding to the amino acid sequence are numbered, and the exon boundaries are indicated by V. The dot above P indicates hydroxyproline.
procollagen is relatively resistant to cleavage by procollagen N-proteinase (24),presumably due to misalignment of the Nproteinase cleavage site. Other examples of single or multiexon deletions of type I collagen genes have been reported in osteogenesis imperfecta (23,25,26).Deletion of exon units either in genomic material or via the missplicing of mRNA precursors may prove to be a common defect in many Type I collagen diseases. Acknowtedgments-We wish to thank Dr. Francesco Ramirez for providing unpublished nucleotide sequences of the amino-terminal portion of the pro-a2(1) chain and Dr. David Eyre for helpful advice and discussion. We also thank Drs. Hisae Hori and Maurice Godfrey for their assistance. N. Donna Gaudette and Mathis Zopfi provided invaluable expert technical support throughout this project. We are grateful to B. Kerry Maddox and Kenine Comstock for amino acid sequencing and amino acid analysis,respectively. The administrative assistance of Barbara Nagle is also gratefully acknowledged. REFERENCES 1. McKusick, V. A. (1972) in Heritable Disorders of Connective Tissue, 4th Ed., pp. 292-371, C. V. Mosby CO., St. Louis, MO 2. Lichtenstein, J. R., Kohn, L. D., Martin, G. R., Byers, P., and McKusick, V. A, (1973) Trans. Assoc. Am. Phys. 86, 333-339 3. Hollister, D. W., Byers, P. H., and Holbrook, K. A. (1982) Adu. Hum. Genet. 12,l-85 4. Lichtenstein, J . R., Martin, G. R., Kohn, L. D., Byers, P. H., and McKusick, V. A, (1973) Science 182,298-300 5. Lenaers, A., Ansay, M., Nusgens, B. V., and Lapiere, C. M. (1971) Eur. J. Bwchem. 23,533-543 6. Lapiere, C. M., Lenaers, A., and Kohn, L. D. (1971) Proc. Natl. Acad. Sci. U. S. A. 68,3054-3058 7. Steinmann, B., Tuderman, L., Peltonen, L., Martin, G. R., McKusick. V. A.. and Prockou. D. J. (1980) . . J. Bid. Chem. 2 5 5 , 8887-8893 8. Steinmann, B., Rao, V. H., and Gitzelmann, R. (1985) Ann. N. Y. Acad. Sci. 460,506-509 9. Eyre, D. R., Shapiro, F. D., and Aldridge, J. F. (1985) J. Biol. Chem. 260,11322-11329 10. Cole, W. G., Chan, D., Chambers, G . W., Walker, I. D., and Bateman, J. F. (1986) J. Biol. Chem. 261, 5496-5503 11. Chu, M-L., de Wet, W., Bernard, M., Ding, J-F., Morabito, M., Myers, J., Williams, C., and Ramirez, F. (1984) Nature 3 1 0 , 337-340 12. Prockop, D. J., and Kivirikko, K. I. (19M) N. Engl. J. Med. 311, 376-386 13. Kaplan, J., Maroteaux, P., and Frezal, J. (1986) Ann. Biol. Clin. 44,289-295 14. McKusick, V. A. (1986) in Mendelian Inheritance of Man (McKusick, V. A., ed) 7th Ed., pp. 217,950, The JohnsHopkins Press Ltd., London 15. Beighton, P. H.(1988) J. Med. Genet., in press 16. Halila. R.. Steinmann.. B... and Peltonen. L. (1986) Am. J. Hum. Genet. 39,222-231 17, Bornstein, P., Kang, A. H., and Piez, K. A. (1966) Biochemistry 5,3803-3812 18. Poduslo, J. F., and Rodbard, D. (1980) Anal. Biochem. 101,394406 19. Hayashi, R. (1977) Methods Enzymol. 47, 84-93 _
I
N-telopeptide Deletion in
a Pro-a2(I) Chain in
20. Dickson, L. A., de Wet, W., Di Liberto, M., Weil, D., and Ramirez, F. (1985) Nucleic Acids Res. 13,3427-3438 21. de Wet, W., Bernard, M., Benson-Chanda, V., Chu, M-L., Dickson, L., Weil, D., and Ramirez, F. (1987) J. Bid. Chem. 262, 16032-16036 22. Sippola, M., Kaffe, S., and Prockop, D. J. (1984) J. Bwl. Chem. 259,14094-14100 23. de Wet, W., Sippola, M., Tromp, G., Prockop, D., Chu, M-L., and Ramirez, F. (1986) J. Biol. Chem. 261, 3857-3862 24. Minor, R. R.,Sippola-Thiele, M., McKeon, J., Berger, J., and Prockop, D. J. (1986) J. Bid. Chem. 261, 10006-10014 and Gelinas, 25. Barsh, G. S., Roush, C. L., Bonadio, J., Byers, P. H., R. E. (1985) Proc. Natl. Acad. Sci. U. S. A. 82, 2870-2874
EDS VIIB
16383
26. Chu, M-L., Gargiulo, V., Williams, C. J., and Ramirez, F. (1985) J. Biol. Chem. 260, 691-694 27. Laemmli, U. K. (1970) Nature 227,680-685 28. Laskey, R. A., and Mills, A. D. (1975) Eur. J. Biochem. 56,335341 29. Chen, T. R. (1977) Exp. Cell Res. 104, 255-262 H., Jr., Scott, R. D., Miller, E. J., and Piez, K. A. 30. Epstein, E. (1971)J. Bid. Chem. 246, 1718-1724 31. Morris, N. P., Keene, D. R., Glanville, R. W., Bentz, H., and Burgeson, R. E. (1986) J . Biol. Chem. 261, 5638-5644 32. Click, E. M., and Bornstein, P. (1970) Biochemistry 9,4699-4706 33. Fietzek, P. P., Furthmayr, H., andKuhn, K. (1974) Eur. J. Biochem. 47, 257-261
/
N-telopeptide Deletion ina Pro-a2(I) Chain inEDS VIIB T h e r m o l y s i n P u r l f ~ c d native mutant collagen containing the m2(1Ir chain ras digested with therrolysln under a variety of conditions. A time course of digestion vas perfocmed by incubating for 1. 2. 4, 6 , 8 . and 24 hours at 22°C rich 1 : 5 0 0 (enzyme:collagenl i n 0.2 II a u o n l u m acetate, pH 8.5. The Ceaction with 20 M EDTA and the pcoducts analysed by SDS-58 PAGE after adding l/lOth volume of 2% SOS i n 60% qlycerol with 0.041 bromophenol blue and 160 M dithzathreitol and boiling far three minutes. In addition, mutant and
was stopped
sample6 were digested with dilutions of thecmolyrin (1:50, 1:200, 1:400. 1:800. 1:1600, 1:6400 and 1:12,500; enzyme:collagenl at l o r 24 hours. Control samples contained t h e r m l y b i n plus S O mB EDTA. 3 0 ' C The reaction -as stopped by the additlo" of 50 M EDTA dnd freezing. Samples w e r e analysed by PAGE a 6 described above ( s e e F l g . 4 1 . C a n t l M collagen
1:lOO.
roc amino acid sequencing. mutant collagen was digested with thermalysin enzpe:collagenl and the reaction was stopped with 20 u1 EDTA. The 0 2 f 1 1 , c h a m swere separated f r o m the d ( I I c h a m s by denaturing RP a= described in the legend to n g . 11 except that a 4.6 x 250 u column was used vith a flow rate of 1 lal/mlnute. The fractLons containing the isolated 02111' 11:50;
chain were pooled, lyaphxlized and a m n o acid sequencing performed. Cyanogen Bromide Digestionof Collagenous Pcotelns
0 '
I
I
10
20
f-
L
3,
I
I
30
The mutant a2 andnormal = chaln6 rere Cleaved rich 20 mq/ml iesuhlimated cyanogen bromide (20:l; cyanogen bromide:collagenl ~n 701 formxc acid at 30°c for four hovra ( 3 0 1 . The reaction was stopped by dzluting five fold with water followed by lyophllizatlon and relyophilzzation.
IL
40
50 Sieve
F r a c t i o n Number
Chromatography Of Cyanogen Bromide Peptldes
The CB peptzdes were resolved by sieve chromatograpmy on tandem si0 si1 TSK 125 (600 x 21.5 U l columns u61ng 50 m H trxb-HCl, pH 6.8. 6 tl urea, 6 0 111 sodium sulfate at room re-peratuce with a flow rate of 0.1 d/alnute. Absorbance at 220 nm was followed and 0.5 01 fractions w e r e collected. The Iho1at.d peptides were desalted on a 4.6 ram x 250 ClS RP column Y L I n g slollar CondltlOnS as described i n the legend for Plg. 1 2 . Praction size: 6 m 1 Absorbance: (-); 220 nm
Tc.perai"re:
4% Tryprxn Dlgestxan of the Amino ~ e r n l n a lCNBr Peptide ( C E X I
1.c
-
100
80
a
ru L
0.6
a l l ather conditlona as dencrlbed above.
0.4-
60
a,
m
u c
2
volume of 1 II acetic acid and then frozen . The tryptlc peptides were isolated by RP chromatography a s described in the legend of rig. 12. CB X 10.15 mg/m11 was also digested with t r y p s m at a i:lO,OOO dxlution (c0llaqen:trypsinl with
i
0.E
df11'
The mutant amino terminal peptlde f C B X I *ab dissolved 14 0.2 I a v o n i u a bicarbonate, pB 8.0 containing 1.25 u1 calcluo chloride at a protein concentration Of 0.60 mg/al and incubated at 37O c overnight ritb trypsin (TPCK treated) f1:501. The reactlo" was stopped by addition Of an equal
1'
C
Of
7
E
2?
c 0.3-
0.4
40
L
a cu
0
2
w
Q
0.2
al 0.2-
20
I
01'
'
I";ect;ons
C m
2 20 30 Fraction Number
'0
40
t: 0.1 -
0
m z y m e Digestion of Native Collage" Carhorypeptldase Y Digestlan Of them i n o Terminal CNBr Peptide CB X Native pepsin-solupilired collagens from patient and Control fibroblasts treated rxth trypsin. chymotrypsin oc thelmolysin.
were
Native collagen samples r e r e dxgeetcd with trypsin fTPCK treated, sigma1 (1:500; enryme:collagenl far 24 hours at 22Oc in 0.2 II NB4BC03 pB 8.0. containing 0.25 u1 CaC12. The reaction was stopped by boxling for three .InUtes after addicton ofI/lOth volume o f 2 8 SO5 i n 608 glycerol with 0 . 0 4 % bromophenol blue and 160M dithiothreltol. For sequencing of the .2(I)* chain after trypsin treatlsnt the ceaction was stopped by acidifyxng with acetlc acld. The digested Samplewas applied t o a denaturing RP column uslng the same conditions as described in Fig. 11 except that a 4.6 % 250 Y Column was used with a I ml/min flow rate. Chymotrypsin Digestion with Chymotrypsin ITLCK treated. sigma1 11:50; enzyme collagenl was c a r r ~ e dout using the Came condltionh as trypsin for Sixteen hOYCS.
To deterrlne the CaIboXyl termLnal sequence of the Dvtant CBX peptide. a modification of the method of Bayash1 ( 1 9 ) was used. 300 proles of f b e
peptide was treated with lt piperidine followed by lyophilization. The peptide w a s dissolved in 0.06 m1 of 50 U a ~ o n i u macetate pH 6.5 containing 8 . 3 M L o? amino n-butyric a c i d a s an lnternal standard. An aliquot was removed tor the lero time and then 0.01 n1 of 1 mg/m1 of Carboxypeptidase Y Inoehringer~mannhe~.) in H20 was added to the remaining 0.05 m 1 . Timed aliquotb were re-aved after Incubating at 37OC far 2. 4. 6. 8, 12 and 16 hours. The reaction was stopped by boxling for three minutes. The samples were lyophilized. re-treated rlth 11 plperzdineand the free anins acids w e r e analysed as described below. Each allquat was normalized by multlPlylng the pmoles of recovcced amino acid by the ratio of thearea of L-ol amino n-butyric acid of the 0 time aliquot/ area of the L-. amino n-butyric acid i n each a114"ot.
N-telopeptide Deletion in a Pro-aB(I) Chain inEDS VIIB
*0°
1
-1
/
E loo Q
0
a
Y
I
I
2
4
6
-
012 CB 4
-
"
%heem
%e*=
02CB2
Observed
Expected'
11.2
11
2.0
21.1
23
1.2
29.2
29
2.5
3
10.1
11
1.9
2
105.3
108
-3.9
10
1
2 1
1
1.0
1.6
2
0
0
23.5
17
2.8
3
5.9
0.3
0
34.5
39
2.8
3
24.1
32
2.6
3
0.2
0
1.0
5.1
0 11.4
0 13
0.8
1
4.3
3.9
0.1
0
11.4
12
1.0
1
8.6
4
0
0
3.8
4
0
4 2
6322
312
"
0
0 -
0 -
30
30
"
