Mutations Which Impede Loop/Sheet Polymerization Enhance the Secretion of Human α1-Antitrypsin Deficiency Variants
1995; Elsevier BV; Volume: 270; Issue: 15 Linguagem: Inglês
10.1074/jbc.270.15.8393
ISSN1083-351X
AutoresSanjiv Sidhar, David A. Lomas, Robin W. Carrell, Richard C. Foreman,
Tópico(s)Blood Coagulation and Thrombosis Mechanisms
Resumoα1-Antitrypsin plasma deficiency variants which form hepatic inclusion bodies within the endoplasmic pathway include the common Z variant (Glu342→ Lys) and the rarer α1-antitrypsin Siiyama(Ser53→ Phe). It has been proposed that retention of both abnormal proteins is accompanied by a common mechanism of loop-sheet polymerization with the insertion of the reactive center loop of one molecule into a β-pleated sheet of another. We have compared the biosynthesis, glycosylation, and secretion of normal, Z and Siiyamavariants of α1-antitrypsin using Xenopus oocytes. Siiyamaand Z α1-antitrypsin both duplicated the secretory defect seen in hepatocytes that results in decreased plasma α1-antitrypsin levels. Digestion with endoglycosidase H localized both variants to a pre-Golgi compartment. The mutation Phe51→ Leu abolished completely the intracellular blockage of Siiyamaα1-antitrypsin and reduced significantly the retention of Z α1-antitrypsin. The secretory properties of M and Z α1-antitrypsin variants containing amino acid substitutions designed to decrease loop mobility and sheet insertion were investigated. A reduction in intracellular levels of Z α1-antitrypsin was achieved with the replacement of P11/12alanines by valines. Thus a decrease in Z and Siiyamaα1-antitrypsin retention was observed with mutations which either closed the A sheet or decreased loop mobility at the loop hinge region. α1-Antitrypsin plasma deficiency variants which form hepatic inclusion bodies within the endoplasmic pathway include the common Z variant (Glu342→ Lys) and the rarer α1-antitrypsin Siiyama(Ser53→ Phe). It has been proposed that retention of both abnormal proteins is accompanied by a common mechanism of loop-sheet polymerization with the insertion of the reactive center loop of one molecule into a β-pleated sheet of another. We have compared the biosynthesis, glycosylation, and secretion of normal, Z and Siiyamavariants of α1-antitrypsin using Xenopus oocytes. Siiyamaand Z α1-antitrypsin both duplicated the secretory defect seen in hepatocytes that results in decreased plasma α1-antitrypsin levels. Digestion with endoglycosidase H localized both variants to a pre-Golgi compartment. The mutation Phe51→ Leu abolished completely the intracellular blockage of Siiyamaα1-antitrypsin and reduced significantly the retention of Z α1-antitrypsin. The secretory properties of M and Z α1-antitrypsin variants containing amino acid substitutions designed to decrease loop mobility and sheet insertion were investigated. A reduction in intracellular levels of Z α1-antitrypsin was achieved with the replacement of P11/12alanines by valines. Thus a decrease in Z and Siiyamaα1-antitrypsin retention was observed with mutations which either closed the A sheet or decreased loop mobility at the loop hinge region. α1-Antitrypsin variants Z (1Jeppsson J.-O. FEBS Lett. 1976; 65: 195-197Crossref PubMed Scopus (152) Google Scholar, 2Eriksson S. Larsson C. N. Engl. J. Med. 1975; 292: 176-180Crossref PubMed Scopus (72) Google Scholar) and Siiyama(3Seyama K. Nukiwa T. Takabe K. Takahashi H. Miyake K. Kira S. J. Biol. Chem. 1991; 266: 12627-12632Abstract Full Text PDF PubMed Google Scholar) are associated with a deficiency of the inhibitor in the blood and its partial retention within the endoplasmic reticulum of synthesizing hepatocytes. Recent evidence indicates that this retention is accompanied by polymerization of the abnormal protein by the process of loop-sheet linkage in which the reactive center loop of one molecule inserts into a β-pleated sheet of another (4Lomas D.A. Evans D.L. Finch J.T. Carrell R.W. Nature. 1992; 357: 605-607Crossref PubMed Scopus (897) Google Scholar, 5Lomas D.A. Evans D.L. Stone S.R. Chang W.-S. Carrell R.W. Biochemistry. 1993; 32: 500-508Crossref PubMed Scopus (208) Google Scholar). The mutation in Z α1-antitrypsin, Glu342→ Lys, occurs at the junction of strand 5 of the six-membered sheet A of the molecule (6Loebermann H. Tokuoka R. Deisenhofer J. Huber R. J. Mol. Biol. 1984; 177: 531-556Crossref PubMed Scopus (610) Google Scholar) with the base of the mobile loop. This alteration in the hinge region is thought to interfere with normal refolding of the reactive center into the A sheet, thus favoring intermolecular loop insertion with the sequential formation of loop-sheet polymers (5Lomas D.A. Evans D.L. Stone S.R. Chang W.-S. Carrell R.W. Biochemistry. 1993; 32: 500-508Crossref PubMed Scopus (208) Google Scholar). The A sheet polymerization model is also compatible with structural studies on α1-antitrypsin Siiyamain which a Ser53→ Phe mutation is predicted to cause an opening of the A sheet between strands s3A and s5A (7Lomas D.A. Finch J.T. Seyama K. Nukiwa T. Carrell R.W. J. Biol. Chem. 1993; 268: 15333-15335Abstract Full Text PDF PubMed Google Scholar, 8Stein P. Chothia C. J. Mol. Biol. 1991; 221: 615-621Crossref PubMed Scopus (134) Google Scholar). Recent structural studies (9Carrell R.W. Stein P.E. Fermi G. Wardell M. Structure. 1994; 2: 257-270Abstract Full Text Full Text PDF PubMed Scopus (368) Google Scholar, 10Schreuder H.A. de Boer B. Dijkema R. Mulders J. Theunissen H.J.M. Grootenhuis P.D.J. Hol W.G.J. Nature Struct. Biol. 1994; 1: 48-55Crossref PubMed Scopus (270) Google Scholar) suggest a related but more complex process may be involved in the loop-sheet polymerization of the serpins. The structure of a dimer of antithrombin shows one molecule in the latent form in which the reactive loop is totally incorporated into the A sheet of the molecule. This releases a strand from the C sheet and it is this strand that is replaced by the reactive loop of the second molecule in the dimer. This alternative mechanism of C sheet polymerization is not incompatible with the A sheet model, since the common first step in either polymerization process would be increased rate of refolding of the loop into the A sheet. Here we examine the comparative secretion of M, Z, and Siiyamaα1-antitrypsins from Xenopus oocytes to further explore the link between loop-sheet interactions and defects in secretion. Yu and colleagues (11Kwon K.-S. Kim J. Shin H.S. Yu M.-H. J. Biol. Chem. 1994; 269: 9627-9631Abstract Full Text PDF PubMed Google Scholar) have identified a mutation, Phe51→ Leu, which stabilizes α1-antitrypsin against polymerization, predictably by locking its A sheet in the closed position. In particular we examine the secretion of this mutant α1-antitrypsin and of its chimer with Z α1-antitrypsin (Phe51→ Leu/Glu342→ Lys) and Siiyamaα1-antitrypsin (Phe51→ Leu/Ser53→ Phe) to establish whether, as predicted, stabilization of the A sheet will prevent the polymerization that otherwise occurs in these variants with a consequent failure in protein export. Although the structural and other evidence strongly indicates that the underlying defect in Siiyamais due to an opening of the A sheet allowing the ready formation of polymers, the structural deductions with respect to the Z mutant are more ambivalent. The Z mutation is at residue P17, at the base of the hinge region on which the reactive center loop pivots, and may result in either an opening of the A sheet or an impedance of loop insertion into the sheet. An inhibition of loop folding would be expected to favor polymerization, but if this is so, the mechanism would be that of A sheet rather than C sheet polymerization. We test this possibility here using constructs of M and Z antitrypsin containing replacements in the hinge region designed to constrain loop mobility but which should not effect the mechanism of A sheet opening. The secretory properties of these M and Z chimeric mutants provide evidence as to the mechanism underlying the intracellular polymerization of Z antitrypsin. DNA and RNA modifying enzymes were from Promega Corp. or Boehringer Mannheim. α-35S-dATP (specific activity > 1000 Ci−1mmol−1) and L-[35S]methionine (specific activity > 1000 Ci−1mmol−1) were supplied by Amersham International plc. Oligonucleotides were synthesized by Oswel DNA Service, Edinburgh, United Kingdom. Endoglycosidase H (cloned from Streptomyces plicatus) was from Boehringer Mannheim. All other reagents were analytical grade or better and provided by Sigma. Mutants were generated from full-length α1-antitrypsin cDNA (12Ciliberto G. Dente L. Cortese R. Cell. 1985; 41: 531-540Abstract Full Text PDF PubMed Scopus (99) Google Scholar) using the polymerase chain reaction-based procedure for site-directed mutagenesis described by Landt et al. (13Landt O. Grunert H.-P. Hahn U. Gene (Amst.). 1990; 96: 125-128Crossref PubMed Scopus (638) Google Scholar). PstI-compatible ends were added to the 5′- and 3′-flanking primers to facilitate cloning of the mutated CDNA into the PstI site of SP64T-RCF, a modified SP64T transcription vector (14Foreman R.C. FEBS Lett. 1987; 216: 79-82Crossref PubMed Scopus (21) Google Scholar). Clones were sequenced using Sequenase reagents (Amersham International plc.) to ensure that the desired mutation was in place. cDNAs were transcribed in vitro as described previously (14Foreman R.C. FEBS Lett. 1987; 216: 79-82Crossref PubMed Scopus (21) Google Scholar) using the Promega Ribomax transcription system. The preparation and micro-injection of Xenopus laevis oocytes were as described by Colman (15Colman A. Hames B.D. Higgins S. Transcription and Translation: A Practical Approach. IRL Press, Oxford1984: 271-302Google Scholar). Healthy oocytes were incubated for 7 h in Barth's saline supplemented with 0.2 mCi−1ml−1L-[35S]methionine. At the end of this period radiolabeling medium was replaced with Barth's saline containing 10 m M methionine and the incubation continued overnight. Oocytes were homogenized and the oocyte extracts and incubation media immunoprecipitated using anti-human α1-antitrypsin (Dako) as described previously (15Colman A. Hames B.D. Higgins S. Transcription and Translation: A Practical Approach. IRL Press, Oxford1984: 271-302Google Scholar, 16Foreman R.C. Judah J.D. Colman A. FEBS Lett. 1984; 168: 84-88Crossref PubMed Scopus (34) Google Scholar). The products were analyzed on 12.5% (w/v) SDS-polyacrylamide gels (SDS-PAGE) 1The abbreviations used are:PAGEpolyacrylamide gel electrophoresisERendoplasmic reticulum. followed by fluorography using "Amplify" (Amersham International plc). Quantitation of radio-labeled proteins was performed by scintillation counting of excised gel fragments. polyacrylamide gel electrophoresis endoplasmic reticulum. immunoadsorbed pellets were suspended in 60 μl of 50 m M Tris-HCl, pH 5.5, 1% (w/v) SDS, 20% (v/v) glycerol, 1% (v/v) 2-mercaptoethanol and incubated for 5 min at 95°C. After cooling, 10 μl of 1 milliunit/μl endoglycosidase H containing 1 m M phenylmethylsulfonyl fluoride was added, and samples were digested for 16 h at 37°C (17Trimble R.B. Maley F. Anal. Biochem. 1984; 141: 515-522Crossref PubMed Scopus (200) Google Scholar) and then analyzed by SDS-PAGE as described above. Expression of PiM-, PiZ-, and Siiyama-encoding RNAs in Xenopus oocytes produced a 54-kDa partially glycosylated intracellular form of the inhibitor (Fig. 1) and a 56-kDa secreted protein. However, it was evident from the fluorograph that secretion of normal, M α1-antitrypsin was far greater than that of both Z and Siiyamavariants. The estimated molecular mass of secreted α1-antitrypsin as reported by various authors, ranges over 52-58 kDa, depending on the electrophoretic conditions and molecular mass standards used. Immunoprecipitated protein was treated with endoglycosidase H (18Lodish H.F. Kong N. Snider M. Strous G.J.A.M. Nature. 1983; 304: 80-83Crossref PubMed Scopus (236) Google Scholar) prior to analysis by SDS-PAGE to determine the intracellular location of the 54-kDa α1-antitrypsin seen in oocytes. As shown in Fig. 1, the intracellular 54-kDa species was converted to a single band of molecular mass 46 kDa on endoglycosidase H digestion. Intracellular PiZ and Siiyamaproteins were sensitive to digestion, indicating that both mutant proteins are retained in a pre-Golgi compartment, probably the ER. Digestion with the enzyme had no effect on secreted M, Z, and Siiyamaproteins which were terminally glycosylated. To quantify differences in the extent of secretion of α1-antitrypsin PiM, PiZ, and Siiyamabands were excised and counted and the experiment repeated with different batches of oocytes to eliminate, as far as possible, oocyte variation. Fig. 2(open bars) shows the amount of inhibitor secreted expressed as a percentage of the total immunoprecipitable material. Siiyama(13.7% secreted ± 1.2) and Z (10.0% secreted ± 1.0) variants are similarly retained in oocytes whereas M α1-antitrypsin (63.3% secreted ± 3.9) is more readily secreted from these cells during the incubation. The Phe51→ Leu mutant was constructed using polymerase chain reaction mutagenesis of M and Z cDNAs. The double mutant Phe51→ Leu, Ser53→ Phe was constructing using M cDNA as template. The effect of the Phe51→ Leu mutation on M, Z, and Siiyamaantitrypsin secretion is shown in Fig. 2 (hatched bars). A 3-fold enhancement of Z antitrypsin secretion was recorded in the chimer Phe51→ Leu, Glu342→ Lys (28.9% secreted ± 4.2). Moreover, the SiiyamaLeu51chimer was secreted (68.9%±3.1) as efficiently as normal M antitrypsin. Thus the Phe51→ Leu mutation fully restores secretion of the Siiyamavariant to that of the normal phenotype and reduces retention of Z antitrypsin. The 20-residue reactive center loop of α1-antitrypsin extends 15 residues (P15→ P1) amino-terminal and 5 residues (P1′ → P5′) carboxyl-terminal to the reactive center. The influence of residues P14and P11/12(near the base of the loop) on the secretory properties of M and Z antitrypsins was examined by replacing the normal residues with those present in the noninhibitory serpin ovalbumin; namely P14Thr → Arg and P11/12Ala → Val. Fig. 3 displays the extent of secretion of antitrypsin as a percentage of the total immunoprecipitable protein. M P14Arg shows a significant reduction in secretion (41.7%± 2.3) compared with normal M antitrypsin (62.4%± 3.1). No significant difference was observed between Z antitrypsin (12.4%± 1.3) and the Z P14Arg double mutant (13.7%± 1.3). The secretion defect observed with Z antitrypsin is partially corrected if the P11/12alanines are substituted for the larger valine residues as indicated by the increased secretion of Z P11/12Val (25.2%± 2.7) compared with Z antitrypsin. It is now realized that the abnormalities of serpins associated with deficiency have a common molecular pathology in that they can spontaneously undergo a conformational transition which results in partial retention within the endoplasmic pathway and, particularly with α1-antitrypsin, is accompanied by polymer formation. The liver disease in Z homozygotes is coincident with the intracellular retention of polymerized inhibitor in inclusion bodies within the endoplasmic reticulum of the hepatocytes (2Eriksson S. Larsson C. N. Engl. J. Med. 1975; 292: 176-180Crossref PubMed Scopus (72) Google Scholar, 19Sharp H.L. Hosp. Pract. 1971; 6: 83-96Crossref Google Scholar). Recently another variant, α1-antitrypsin Siiyama, was shown to have the same association with plasma deficiency and the identical histological finding of hepatic inclusions of mutant inhibitor (3Seyama K. Nukiwa T. Takabe K. Takahashi H. Miyake K. Kira S. J. Biol. Chem. 1991; 266: 12627-12632Abstract Full Text PDF PubMed Google Scholar, 7Lomas D.A. Finch J.T. Seyama K. Nukiwa T. Carrell R.W. J. Biol. Chem. 1993; 268: 15333-15335Abstract Full Text PDF PubMed Google Scholar). Both the Z and Siiyamamutants have amino acid substitutions, that although well separated from each other, will theoretically have the same effect, that is to open the A sheet between the third and fifth strands and thus promote the process of loop sheet polymerization (4Lomas D.A. Evans D.L. Finch J.T. Carrell R.W. Nature. 1992; 357: 605-607Crossref PubMed Scopus (897) Google Scholar). Z and Siiyamaα1-antitrypsin will form long chain polymers with equal facility (7Lomas D.A. Finch J.T. Seyama K. Nukiwa T. Carrell R.W. J. Biol. Chem. 1993; 268: 15333-15335Abstract Full Text PDF PubMed Google Scholar), and Fig. 1 demonstrates that they are secreted from oocytes with comparable efficiency, Z 10.0% and Siiyama13.7% of immunoprecipitable material. In mammalian cells only a small proportion of nonsecreted Z protein accumulates in the ER; the majority is rapidly degraded (20Graham K.S. Le A. Sifers R.N. J. Biol. Chem. 1990; 265: 20463-20468Abstract Full Text PDF PubMed Google Scholar). We attempted to measure the degradation of nonsecreted material in oocytes by quantifying α1-antitrypsin before and after the chase incubation. We were unable to detect significant proteolysis, mainly because the slow rate of equilibration of the large intracellular amino acid pool means that labeling continues well into the chase incubation. Other experiments following the fate of retained Z protein in oocytes suggest that little degradation occurs over the time scale of our experiments (21Perlmutter D.H. Kay R.M. Sessions Cole F. Rossing T.H. Van Thiel D. Colten H.R. Proc. Natl. Acad. Sci. U. S. A. 1985; 82: 6918-6921Crossref PubMed Scopus (43) Google Scholar). The synthesis of Z antitrypsin does stimulate the activity of a number of lysosomal enzymes in injected oocytes (22Bathurst I.C. Errington D.M. Foreman R.C. Judah J.D. Carrell R.W. FEBS Lett. 1985; 183: 304-308Crossref PubMed Scopus (17) Google Scholar), but this response may not facilitate the removal of retained protein since, in mammalian cells, degradation has been shown to occur in a post ER nonlysosomal compartment (23Le A. Graham K.S. Sifers R.N. J. Biol. Chem. 1990; 265: 14001-14007Abstract Full Text PDF PubMed Google Scholar). Both Z and Siiyamavariants are retained in a similar, if not identical, pre-Golgi compartment within the oocyte as indicated by the pattern of endoglycosidase H digestion. These findings emphasize the link between loop sheet insertion and the secretory block and suggest that, as with Z, the plasma deficiency of the Siiyamavariant can be explained solely in terms of a failure in protein export. We have investigated the phenomenon of loop sheet polymerization by two approaches involving the prevention of entry of the reactive center loop into the A sheet. First, mutations which close the gap between strands s3A and s5A and then mutations in the loop-hinge region which restrict entry into the sheet due to steric hindrance. The Siiyamamutation (Ser53→ Phe) is located in the B helix which underlies the A sheet and provides a surface on which strand s3A slides in order for the sheet to open (8Stein P. Chothia C. J. Mol. Biol. 1991; 221: 615-621Crossref PubMed Scopus (134) Google Scholar). Substitution with a large, aromatic side chain at this position is thought to lock the sheet in the open conformation and thereby promote loop sheet polymerization (7Lomas D.A. Finch J.T. Seyama K. Nukiwa T. Carrell R.W. J. Biol. Chem. 1993; 268: 15333-15335Abstract Full Text PDF PubMed Google Scholar, 8Stein P. Chothia C. J. Mol. Biol. 1991; 221: 615-621Crossref PubMed Scopus (134) Google Scholar). Recently, Yu and colleagues (11Kwon K.-S. Kim J. Shin H.S. Yu M.-H. J. Biol. Chem. 1994; 269: 9627-9631Abstract Full Text PDF PubMed Google Scholar) have reported that substitution of the Phe residue at position 51 by a small, nonpolar residue such as leucine enhances the thermal stability and decreases heat-induced polymerization of wild type M α1-antitrypsin. Here we show that the stabilizing properties of this change at position 51 have an ameliorating influence on the Z and Siiyamasecretion-defective mutations (Fig. 2). This effect, however, is not equivalent for both dysfunctional proteins; secretion of the Z/Phe51→ Leu double mutant is increased nearly 3-fold, whereas secretion of Siiyama/Phe51→ Leu is equal to M α1-antitrypsin (a greater than 5-fold increase). The increased effect on Siiyamamay simply be a matter of proximity in that removal of Phe51may correct an aberrant conformation of the B helix, induced by the introduction of a third contiguous phenylalanine at position 53, and thus allow closure of the A sheet. The location of the Z mutation, at the hinge of the reactive center loop, will influence both A sheet opening and loop mobility. Changes at position 51 are liable to reverse the former but not the latter and would be predicted to allow only a partial correction of the Z secretory defect. A profound structural transformation from a stressed S conformation to a more ordered, heat-stable, and relaxed R state is observed upon reactive center cleavage of inhibitory serpins. This S to R transition is dependent upon the insertion of the mobile reactive center loop into β sheet A after cleavage of the P1-P1′ peptide bond. Inappropriate insertion of the intact loop into β sheet A is a feature of the polymerization of aberrant serpins, but partial insertion of the uncleaved loop into the β sheet is thought to facilitate formation of the canonical form of the active inhibitor (24Stein P.G. Leslie A.G.W. Finch J.T. Carrell R.W. J. Mol. Biol. 1991; 221: 941-959Crossref PubMed Scopus (406) Google Scholar, 25Engh R.A. Wright H.T. Huber R. Protein Eng. 1990; 3: 469-477Crossref PubMed Scopus (54) Google Scholar, 26Skriver K. Wikoff W.R. Patston P.A. Tausk F. Schapira M. Kaplan A.P. Bock P.E. J. Biol. Chem. 1991; 266: 9216-9221Abstract Full Text PDF PubMed Google Scholar). The reactive center loops of all inhibitory serpins are characterized by the conservation of small hydrophobic amino acids, particularly at positions P10, P11/12, and P14at the base of the loop (27Stein P.E. Tewkesbury D.A. Carrell R.W. Biochem. J. 1989; 262: 103-107Crossref PubMed Scopus (125) Google Scholar). These residues are orientated with their side chains facing the hydrophobic interior of the molecule (6Loebermann H. Tokuoka R. Deisenhofer J. Huber R. J. Mol. Biol. 1984; 177: 531-556Crossref PubMed Scopus (610) Google Scholar), and as a consequence, there is a constraint on their size and polarity if loop insertion is to occur. The absence of inhibitory activity and the S to R transition in ovalbumin and angiotensinogen can be explained by the appearance of larger and/or more polar residues in these critical positions (27Stein P.E. Tewkesbury D.A. Carrell R.W. Biochem. J. 1989; 262: 103-107Crossref PubMed Scopus (125) Google Scholar). Similarly, several natural mutants of antithrombin III, C1-inhibitor, and other serpins have been identified with point mutations at positions P12and P10, and in most cases these are proteinase substrates, not inhibitors (26Skriver K. Wikoff W.R. Patston P.A. Tausk F. Schapira M. Kaplan A.P. Bock P.E. J. Biol. Chem. 1991; 266: 9216-9221Abstract Full Text PDF PubMed Google Scholar, 28Stein P.E. Carrell R.W. Nature Struct. Biol. 1995; 2: 96-113Crossref PubMed Scopus (394) Google Scholar). Schulze et al. (29Schulze A.J. Huber R. Degryse E. Speck D. Bischoff R. Eur. J. Biochem. 1991; 202: 1147-1155Crossref PubMed Scopus (50) Google Scholar) have shown that substitution of P14Thr by Arg converts a reactive center mutant of α1-antitrypsin from an inhibitor to a substrate that fails to undergo a detectable conformational change. However, a contradictory case has been made by Hood et al. (30Hood D.B. Huntington J.A. Gettins P.G.W. Biochemistry. 1994; 33: 8538-8547Crossref PubMed Scopus (102) Google Scholar), who constructed a P14Thr → Arg α1-antitrypsin which retained the ability to complex with several cognate proteinases and underwent the S to R transition. Although the effect of such mutations on the thermal stability and inhibitory activity of normal serpins are well established, their influence on polymerization and secretion have yet to be explored. The secretion of Z α1-antitrypsin from oocytes was not improved by the replacement of P14Thr by arginine. This may mean that aggregation is unhampered by partial exclusion of strand 4A from the sheet, alternatively the gap between strands 3A and 5A may be wider as a result of the lysine residue at position 342 and thus able to accommodate the larger and more polar arginine side chain. The assertion that an arginine residue at P14does not prevent polymerization of the Z variant, and presumably does not materially affect loop insertion, is in general agreement with the observation that such a change in the normal inhibitor is compatible with the S → R transition and maintenance of inhibitory function (30Hood D.B. Huntington J.A. Gettins P.G.W. Biochemistry. 1994; 33: 8538-8547Crossref PubMed Scopus (102) Google Scholar). Nevertheless, the P14Arg mutation may have other structural consequences in addition to its effect on loop mobility, since the secretion of M type α1-antitrypsin P14Thr → Arg was significantly reduced, but not to the level observed for Z α1-antitrypsin. Alanine to valine substitutions at P11and P12were more effective in overcoming the block in secretion imposed by the Z mutation, causing a 2-fold increase in export of this mutant α1-antitrypsin (Fig. 3). Recent results (28Stein P.E. Carrell R.W. Nature Struct. Biol. 1995; 2: 96-113Crossref PubMed Scopus (394) Google Scholar) indicate that insertion up to P10is required for the release of strand 1 from the C sheet. The presence of valine at P12, and perhaps P11, will inhibit this insertion and hence the availability of the S1C position for C sheet polymerization. Similarly, a naturally occurring mutation of the human C1-inhibitor with P12Ala → Glu was not an effective inhibitor and did not undergo the S to R conformational change and also showed no tendency to polymerize (26Skriver K. Wikoff W.R. Patston P.A. Tausk F. Schapira M. Kaplan A.P. Bock P.E. J. Biol. Chem. 1991; 266: 9216-9221Abstract Full Text PDF PubMed Google Scholar). Alanine to valine substitutions at both P11and P12will consequently serve to hinder this more extensive incorporation of the loop, although single replacement with threonine at P12does not prevent M type antitrypsin from undergoing the S to R transition (31Hopkins P.C.R. Carrell R.W. Stone S.R. Biochemistry. 1993; 32: 7650-7657Crossref PubMed Scopus (169) Google Scholar). Thus an increase in secretion of Z antitrypsin with valine residues in both the P11and P12positions is most readily explained by a C sheet mechanism of polymerization.
Referência(s)