Portal de Periódicos da CAPES

Latent S49P Neuroserpin Forms Polymers in the Dementia Familial Encephalopathy with Neuroserpin Inclusion Bodies

2005; Elsevier BV; Volume: 280; Issue: 14 Linguagem: Inglês

10.1074/jbc.m413282200

1083-351X

Maki Onda, Didier Belorgey, Lynda K. Sharp, David A. Lomas,

Signaling Pathways in Disease

The serpinopathies result from conformational transitions in members of the serine proteinase inhibitor superfamily with aberrant tissue deposition or loss of function. They are typified by mutants of neuroserpin that are retained within the endoplasmic reticulum of neurons as ordered polymers in association with dementia. We show here that the S49P mutant of neuroserpin that causes the dementia familial encephalopathy with neuroserpin inclusion bodies (FENIB) forms a latent species in vitro and in vivo in addition to the formation of polymers. Latent neuroserpin is thermostable and inactive as a proteinase inhibitor, but activity can be restored by refolding. Strikingly, latent S49P neuroserpin is unlike any other latent serine proteinase inhibitor (serpin) in that it spontaneously forms polymers under physiological conditions. These data provide an alternative method for the inactivation of mutant neuroserpin as a proteinase inhibitor in FENIB and demonstrate a second pathway for the formation of intracellular polymers in association with disease. The serpinopathies result from conformational transitions in members of the serine proteinase inhibitor superfamily with aberrant tissue deposition or loss of function. They are typified by mutants of neuroserpin that are retained within the endoplasmic reticulum of neurons as ordered polymers in association with dementia. We show here that the S49P mutant of neuroserpin that causes the dementia familial encephalopathy with neuroserpin inclusion bodies (FENIB) forms a latent species in vitro and in vivo in addition to the formation of polymers. Latent neuroserpin is thermostable and inactive as a proteinase inhibitor, but activity can be restored by refolding. Strikingly, latent S49P neuroserpin is unlike any other latent serine proteinase inhibitor (serpin) in that it spontaneously forms polymers under physiological conditions. These data provide an alternative method for the inactivation of mutant neuroserpin as a proteinase inhibitor in FENIB and demonstrate a second pathway for the formation of intracellular polymers in association with disease. Neuroserpin is a member of the serine proteinase inhibitor (serpin) 1The abbreviations and trivial names used are: serpin, serine proteinase inhibitor; S49P, the S49P mutant of neuroserpin; FENIB, familial encephalopathy with neuroserpin inclusion bodies; NS, wild-type neuroserpin; tPA, tissue plasminogen activator; PBS, phosphate-buffered saline. 1The abbreviations and trivial names used are: serpin, serine proteinase inhibitor; S49P, the S49P mutant of neuroserpin; FENIB, familial encephalopathy with neuroserpin inclusion bodies; NS, wild-type neuroserpin; tPA, tissue plasminogen activator; PBS, phosphate-buffered saline. superfamily that is predominantly expressed by neurons in the developing and adult brain. It is secreted from the axonal growth cones of the central and peripheral nervous system, where it inhibits the enzyme tissue plasminogen activator (tPA) (1.Osterwalder T. Contartese J. Stoeckli E.T. Kuhn T.B. Sonderegger P. EMBO J. 1996; 15: 2944-2953Crossref PubMed Scopus (166) Google Scholar, 2.Osterwalder T. Cinelli P. Baici A. Pennella A. Krueger S.R. Schrimpf S.P. Meins M. Sonderegger P. J. Biol. Chem. 1998; 237: 2312-2321Abstract Full Text Full Text PDF Scopus (122) Google Scholar, 3.Hastings G.A. Coleman T.A. Haudenschild C.C. Stefansson S. Smith E.P. Barthlow R. Cherry S. Sandkvist M. Lawrence D.A. J. Biol. Chem. 1997; 272: 33062-33067Abstract Full Text Full Text PDF PubMed Scopus (201) Google Scholar, 4.Belorgey D. Crowther D.C. Mahadeva R. Lomas D.A. J. Biol. Chem. 2002; 277: 17367-17373Abstract Full Text Full Text PDF PubMed Scopus (88) Google Scholar, 5.Barker-Carlson K. Lawrence D.A. Schwartz B.S. J. Biol. Chem. 2002; 277: 46852-46857Abstract Full Text Full Text PDF PubMed Scopus (54) Google Scholar). The expression pattern of neuroserpin and its in vitro inhibitory activity against tPA suggest that neuroserpin has a role in controlling axonal growth, regulating emotional behavior and memory, reducing epileptic seizure activity, and limiting damage in cerebral infarction (6.Parmar P.K. Coates L.C. Pearson J.F. Hill R.M. Birch N.P. J. Neurochem. 2002; 82: 1406-1415Crossref PubMed Scopus (57) Google Scholar, 7.Hill R.M. Coates L.C. Parmar P.K. Mezey E. Pearson J.F. Birch N.P. Ann. N. Y. Acad. Sci. 2002; 971: 406-415Crossref PubMed Scopus (17) Google Scholar, 8.Yepes M. Sandkvist M. Wong M.K. Coleman T.A. Smith E. Cohan S.L. Lawrence D.A. Blood. 2000; 96: 569-576Crossref PubMed Google Scholar, 9.Yepes M. Sandkvist M. Coleman T.A. Moore E. Wu J.Y. Mitola D. Bugge T.H. Lawrence D.A. J. Clin. Investig. 2002; 109: 1571-1578Crossref PubMed Scopus (124) Google Scholar, 10.Madani R. Kozlov S. Akhmedov A. Cinelli P. Kinter J. Lipp H.P. Sonderegger P. Wolfer D.P. Mol. Cell. Neurosci. 2003; 23: 473-494Crossref PubMed Scopus (123) Google Scholar). We have recently described an autosomal dominant dementia, FENIB, that is characterized by inclusions of mutant neuroserpin as Collins' bodies within cortical and subcortical neurons (11.Davis R.L. Shrimpton A.E. Holohan P.D. Bradshaw C. Feiglin D. Collins G.H. Sonderegger P. Kinter J. Becker L.M. Lacbawan F. Krasnewich D. Muenke M. Lawrence D.A. Yerby M.S. Shaw C.M. Gooptu B. Elliott P.R. Finch J.T. Carrell R.W. Lomas D.A. Nature. 1999; 401: 376-379Crossref PubMed Google Scholar). This dementia is unusual in that the inclusions result from the retention of ordered polymers of neuroserpin within the endoplasmic reticulum of neurons (4.Belorgey D. Crowther D.C. Mahadeva R. Lomas D.A. J. Biol. Chem. 2002; 277: 17367-17373Abstract Full Text Full Text PDF PubMed Scopus (88) Google Scholar, 12.Belorgey D. Sharp L.K. Crowther D.C. Onda M. Johansson J. Lomas D.A. Eur. J. Biochem. 2004; 271: 3360-3367Crossref PubMed Scopus (48) Google Scholar, 13.Miranda E. Römisch K. Lomas D.A. J. Biol. Chem. 2004; 279: 28283-28291Abstract Full Text Full Text PDF PubMed Scopus (99) Google Scholar). Moreover, the number of inclusions is directly related to the rate of polymer formation and inversely proportional to the age of onset of dementia (4.Belorgey D. Crowther D.C. Mahadeva R. Lomas D.A. J. Biol. Chem. 2002; 277: 17367-17373Abstract Full Text Full Text PDF PubMed Scopus (88) Google Scholar, 12.Belorgey D. Sharp L.K. Crowther D.C. Onda M. Johansson J. Lomas D.A. Eur. J. Biochem. 2004; 271: 3360-3367Crossref PubMed Scopus (48) Google Scholar, 13.Miranda E. Römisch K. Lomas D.A. J. Biol. Chem. 2004; 279: 28283-28291Abstract Full Text Full Text PDF PubMed Scopus (99) Google Scholar, 14.Davis R.L. Shrimpton A.E. Carrell R.W. Lomas D.A. Gerhard L. Baumann B. Lawrence D.A. Yepes M. Kim T.S. Ghetti B. Piccardo P. Takao M. Lacbawan F. Muenke M. Sifers R.N. Bradshaw C.B. Kent P.F. Collins G.H. Larocca D. Holohan P.D. Lancet. 2002; 359: 2242-2247Abstract Full Text Full Text PDF PubMed Scopus (133) Google Scholar). For example, the Syracuse mutation (S49P) causes dementia in middle age, whereas the more rapidly polymerizing Portland mutant (S52R) causes more inclusions and an onset of dementia in the early twenties. Polymers of the serpins result from the sequential linkage between the reactive center loop of one molecule and the β-sheet A of another (4.Belorgey D. Crowther D.C. Mahadeva R. Lomas D.A. J. Biol. Chem. 2002; 277: 17367-17373Abstract Full Text Full Text PDF PubMed Scopus (88) Google Scholar, 11.Davis R.L. Shrimpton A.E. Holohan P.D. Bradshaw C. Feiglin D. Collins G.H. Sonderegger P. Kinter J. Becker L.M. Lacbawan F. Krasnewich D. Muenke M. Lawrence D.A. Yerby M.S. Shaw C.M. Gooptu B. Elliott P.R. Finch J.T. Carrell R.W. Lomas D.A. Nature. 1999; 401: 376-379Crossref PubMed Google Scholar, 12.Belorgey D. Sharp L.K. Crowther D.C. Onda M. Johansson J. Lomas D.A. Eur. J. Biochem. 2004; 271: 3360-3367Crossref PubMed Scopus (48) Google Scholar, 13.Miranda E. Römisch K. Lomas D.A. J. Biol. Chem. 2004; 279: 28283-28291Abstract Full Text Full Text PDF PubMed Scopus (99) Google Scholar, 15.Lomas D.A. Evans D.L. Finch J.T. Carrell R.W. Nature. 1992; 357: 605-607Crossref PubMed Scopus (893) Google Scholar, 16.Elliott P.R. Lomas D.A. Carrell R.W. Abrahams J.-P. Nat. Struct. Biol. 1996; 3: 676-681Crossref PubMed Scopus (244) Google Scholar, 17.Sivasothy P. Dafforn T.R. Gettins P.G.W. Lomas D.A. J. Biol. Chem. 2000; 275: 33663-33668Abstract Full Text Full Text PDF PubMed Scopus (128) Google Scholar). The polymers that form in FENIB are identical to those that are formed by a mutant of another member of the serpin superfamily, the Z variant of α1-antitrypsin (E342K), in the endoplasmic reticulum of hepatocytes in association with liver disease (15.Lomas D.A. Evans D.L. Finch J.T. Carrell R.W. Nature. 1992; 357: 605-607Crossref PubMed Scopus (893) Google Scholar). This common mechanism of disease has allowed us to group these conditions and others that result from polymerization of serpin mutants, angio-edema (C1-inhibitor), thrombosis (antithrombin), and emphysema (α1-antichymotrypsin), as the serpinopathies (18.Lomas D.A. Mahadeva R. J. Clin. Investig. 2002; 110: 1585-1590Crossref PubMed Scopus (236) Google Scholar, 19.Carrell R.W. Lomas D.A. New. Engl. J. Med. 2002; 346: 45-53Crossref PubMed Scopus (360) Google Scholar). Serpinopathies differ from many other conformational diseases (20.Carrell R.W. Lomas D.A. Lancet. 1997; 350: 134-138Abstract Full Text Full Text PDF PubMed Scopus (810) Google Scholar) in several fundamental aspects. The mutants are retained as ordered, rather than disordered structures (11.Davis R.L. Shrimpton A.E. Holohan P.D. Bradshaw C. Feiglin D. Collins G.H. Sonderegger P. Kinter J. Becker L.M. Lacbawan F. Krasnewich D. Muenke M. Lawrence D.A. Yerby M.S. Shaw C.M. Gooptu B. Elliott P.R. Finch J.T. Carrell R.W. Lomas D.A. Nature. 1999; 401: 376-379Crossref PubMed Google Scholar, 13.Miranda E. Römisch K. Lomas D.A. J. Biol. Chem. 2004; 279: 28283-28291Abstract Full Text Full Text PDF PubMed Scopus (99) Google Scholar, 15.Lomas D.A. Evans D.L. Finch J.T. Carrell R.W. Nature. 1992; 357: 605-607Crossref PubMed Scopus (893) Google Scholar, 21.Janciauskiene S. Dominaitiene R. Sternby N.H. Piitulainen E. Eriksson S. J. Biol. Chem. 2002; 277: 26540-26546Abstract Full Text Full Text PDF PubMed Scopus (71) Google Scholar); they are retained within the endoplasmic reticulum (13.Miranda E. Römisch K. Lomas D.A. J. Biol. Chem. 2004; 279: 28283-28291Abstract Full Text Full Text PDF PubMed Scopus (99) Google Scholar, 22.Sidhar S.K. Lomas D.A. Carrell R.W. Foreman R.C. J. Biol. Chem. 1995; 270: 8393-8396Abstract Full Text Full Text PDF PubMed Scopus (63) Google Scholar); and they do not induce an unfolded protein response (23.Graham K.S. Le A. Sifers R.N. J. Biol. Chem. 1990; 265: 20463-20468Abstract Full Text PDF PubMed Google Scholar). Polymers of the serpins usually, but not always (24.Mikus P. Urano T. Liljeström P. Ny T. Eur. J. Biochem. 1993; 218: 1071-1082Crossref PubMed Scopus (75) Google Scholar, 25.Wilczynska M. Lobov S. Ohlsson P.-I. Ny T. EMBO J. 2003; 22: 1753-1761Crossref PubMed Scopus (32) Google Scholar), form in vivo in association with point mutations. We have demonstrated that this is via an unstable intermediate M* (26.Dafforn T.R. Mahadeva R. Elliott P.R. Sivasothy P. Lomas D.A. J. Biol. Chem. 1999; 274: 9548-9555Abstract Full Text Full Text PDF PubMed Scopus (214) Google Scholar, 27.Gooptu B. Hazes B. Chang W.S. Dafforn T.R. Carrell R.W. Read R.J. Lomas D.A. Proc. Natl. Acad. Sci (U. S. A.). 2000; 97: 67-72Crossref PubMed Scopus (176) Google Scholar, 28.Mahadeva R. Dafforn T.R. Carrell R.W. Lomas D.A. J. Biol. Chem. 2002; 277: 6771-6774Abstract Full Text Full Text PDF PubMed Scopus (146) Google Scholar). However, in serpins such as plasminogen activator inhibitor-1, antithrombin, and α1-antichymotrypsin population of M* can result in intramolecular loop insertion and an inert latent configuration (29.Lawrence D.A. Strandberg L. Ericson J. Ny T. J. Biol. Chem. 1990; 265: 20293-20301Abstract Full Text PDF PubMed Google Scholar, 30.Bruce D. Perry D.J. Borg J.-Y. Carrell R.W. Wardell M.R. J. Clin. Investig. 1994; 94: 2265-2274Crossref PubMed Scopus (150) Google Scholar, 31.Zhou A. Huntington J.A. Carrell R.W. Blood. 1999; 94: 3388-3396Crossref PubMed Google Scholar, 32.Chang W.-S.W. Lomas D.A. J. Biol. Chem. 1998; 273: 3695-3701Abstract Full Text Full Text PDF PubMed Scopus (45) Google Scholar). Other serpins, such as α1-antitrypsin and antithrombin, can also be induced to form the latent conformer by heating in stabilizing concentrations of sodium citrate (33.Lomas D.A. Elliott P.R. Chang W.-S.W. Wardell M.R. Carrell R.W. J. Biol. Chem. 1995; 270: 5282-5288Abstract Full Text Full Text PDF PubMed Scopus (141) Google Scholar, 34.Wardell M.R. Chang W.-S.W. Bruce D. Skinner R. Lesk A. Carrell R.W. Biochemistry. 1997; 36: 13133-13142Crossref PubMed Scopus (76) Google Scholar). This latent conformation cannot revert back to the stable species without incubation at high temperatures or refolding from denaturants (33.Lomas D.A. Elliott P.R. Chang W.-S.W. Wardell M.R. Carrell R.W. J. Biol. Chem. 1995; 270: 5282-5288Abstract Full Text Full Text PDF PubMed Scopus (141) Google Scholar, 35.Hekman C.M. Loskutoff D.J. J. Biol. Chem. 1985; 260: 11581-11587Abstract Full Text PDF PubMed Google Scholar). We show here that both polymers and latent species occur with the S49P mutant of neuroserpin in vitro and in vivo. Moreover, we demonstrate that the latent conformation can spontaneously form polymers under physiological conditions. These data demonstrate that latency is an alternative pathway for inactivation of mutant neuroserpin as a proteinase inhibitor as well as providing an additional pathway for the formation of polymers in vivo. Expression and Purification of Recombinant Proteins—Recombinant wild-type neuroserpin (NS) and the S49P mutant of neuroserpin were expressed with a His6 tag at the N terminus and purified as described previously (4.Belorgey D. Crowther D.C. Mahadeva R. Lomas D.A. J. Biol. Chem. 2002; 277: 17367-17373Abstract Full Text Full Text PDF PubMed Scopus (88) Google Scholar, 12.Belorgey D. Sharp L.K. Crowther D.C. Onda M. Johansson J. Lomas D.A. Eur. J. Biochem. 2004; 271: 3360-3367Crossref PubMed Scopus (48) Google Scholar) with the following modifications. After purification on HiTrap Q-Sepharose (Amersham Biosciences), the protein was purified on a UNO-Q6 column (Bio-Rad) with an NaCl gradient (0.02–1 m) in 20 mm Tris-HCl buffer (pH 7.4). The monomeric protein was then isolated by gel filtration using a Superpose 12HR column (Amersham Biosciences) equilibrated with PBS (137 mm NaCl, 2.7 mm KCl, 1.5 mm KH2PO4, 8 mm Na2HPO4, pH 7.4). The resulting proteins were assessed by SDS and non-denaturing and transverse urea gradient PAGE, and activity was assessed against tPA (4.Belorgey D. Crowther D.C. Mahadeva R. Lomas D.A. J. Biol. Chem. 2002; 277: 17367-17373Abstract Full Text Full Text PDF PubMed Scopus (88) Google Scholar). The protein concentration was determined by Bio-Rad protein assay. Preparation of Latent Neuroserpin—Native NS or native S49P was incubated at 1 μg/ml and 55 °C for 24 h in PBS and then concentrated to 1–5 mg/ml at 4 °C with a Vivaspin concentrator (Vivasciene). The latent protein was purified with a Superpose 12HR column (Amersham Biosciences) equilibrated with PBS, concentrated with a Vivaspin concentrator, and then stored at –80 °C. The resulting proteins were assessed by SDS and non-denaturing PAGE. Assessment of Latent and Polymeric Neuroserpin by Non-denaturing and SDS-PAGE—1–2 μg of proteins were separated on a 7.5% w/v non-denaturing gel or 10% w/v SDS gel and then visualized by staining with GelCode Blue stain reagent (Pierce) or by silver staining. The density of bands was assessed by densitometry scanning with Quantity One (Bio-Rad) software. Western blot analysis was performed as described previously (13.Miranda E. Römisch K. Lomas D.A. J. Biol. Chem. 2004; 279: 28283-28291Abstract Full Text Full Text PDF PubMed Scopus (99) Google Scholar). Briefly, proteins were separated on a 7.5% w/v non-denaturing gel or 10% w/v SDS gel and then transferred onto Immobilon P membrane (Millipore). The blots were probed with 1/25,000 or 1/5000 rabbit polyclonal anti-neuroserpin antibody and 1/10,000 goat polyclonal anti-rabbit IgG-HRP (horseradish peroxidase) (Sigma) and detected with an ECL SuperSignal West Femto maximum sensitivity substrate or a SuperSignal West Pico chemiluminescent substrate (Pierce). Circular Dichroism—CD experiments were performed using a JASCO J-810 spectropolarimeter. The far-UV CD spectrum was recorded at 0.5 mg/ml protein and 10 °C in 20 mm sodium phosphate buffer (pH 7.4), and the average of 20 traces was determined. The spectra were analyzed in terms of secondary structure content by a variable selection method with 43 reference proteins (36.Manavalan P. Johnson Jr., W.C. Anal. Biochem. 1987; 167: 76-85Crossref PubMed Scopus (661) Google Scholar, 37.Sreerama N. Woody R.W. Anal. Biochem. 2000; 287: 252-260Crossref PubMed Scopus (2487) Google Scholar). Thermal unfolding experiments were performed by monitoring the CD signal at 216 nm between 25 and 100 °C using a heating rate of 1 °C/min and a protein concentration of 1 mg/ml in PBS. Melting temperatures were calculated using an expression for a two-state transition as described previously (26.Dafforn T.R. Mahadeva R. Elliott P.R. Sivasothy P. Lomas D.A. J. Biol. Chem. 1999; 274: 9548-9555Abstract Full Text Full Text PDF PubMed Scopus (214) Google Scholar, 38.Lawrence D.A. Olson S.T. Palaniappan S. Ginsburg D. Biochemistry. 1994; 33: 3643-3648Crossref PubMed Scopus (79) Google Scholar, 39.Dafforn T.R. Della M. Miller A.D. J. Biol. Chem. 2001; 276: 49310-49319Abstract Full Text Full Text PDF PubMed Scopus (99) Google Scholar), and the results are the average of three experiments. Complex Formation Assays—NS, S49P, or Collins' body extraction was incubated with 1.7 μm tPA at 25 °C as described previously (4.Belorgey D. Crowther D.C. Mahadeva R. Lomas D.A. J. Biol. Chem. 2002; 277: 17367-17373Abstract Full Text Full Text PDF PubMed Scopus (88) Google Scholar). Samples were taken at different time intervals, and the reaction was stopped by the addition of 1 mm 1,5-dansyl-Glu-Gly-Arg-chloromethylketone (final concentration) to inhibit any free tPA (40.Renatus M. Engh R.A. Stubbs M.T. Huber R. Fischer S. Kohnert U. Bode W. EMBO J. 1997; 16: 4797-4805Crossref PubMed Scopus (93) Google Scholar). The samples were then mixed with loading buffer, snap-frozen in liquid nitrogen, and stored until the completion of the experiment. The samples for SDS-PAGE analysis were boiled for 3 min before loading onto the gel. Proteins were separated by 10% w/v SDS-PAGE or 7.5% non-denaturing PAGE and visualized by staining with GelCode Blue stain reagent (Pierce) or Western blot analysis. Refolding of Neuroserpin—Native or latent neuroserpin was denatured at 0.1 mg/ml and 37 °C for 30 min in 0.2 m HCl containing 9 m urea, pH 2.2, and then diluted with a 30-fold volume of 0.1 m Tris-HCl, pH 8.0, at 4 °C. The proteins were allowed to refold at 4 °C for 8 h and then concentrated with a Vivaspin concentrator at 4 °C. Preparation of Collins' Bodies Extraction—Collins' bodies were isolated from the brain of a patient with FENIB as detailed previously (11.Davis R.L. Shrimpton A.E. Holohan P.D. Bradshaw C. Feiglin D. Collins G.H. Sonderegger P. Kinter J. Becker L.M. Lacbawan F. Krasnewich D. Muenke M. Lawrence D.A. Yerby M.S. Shaw C.M. Gooptu B. Elliott P.R. Finch J.T. Carrell R.W. Lomas D.A. Nature. 1999; 401: 376-379Crossref PubMed Google Scholar). They were treated with Tris-HCl buffer (0.1 m Tris, 0.15 m NaCl, 50 mm dithiothreitol, 5 mm EDTA, 3 mg/ml protease inhibitor mixture, pH 8.0) containing 1% v/v Triton X-100 and sonicated at 1 °C for 3 min. The tissue suspension was centrifuged at 20,000 × g and 4 °C for 30 min, and the supernatant was dialyzed against PBS at 4 °C for 5 h with a Slide-A-Lyzer dialysis cassette (Pierce). This extraction process had no effect on the conformation of purified recombinant native, latent, and polymeric S49P at a protein concentration of 1 mg/ml (data not shown). Thus the extraction process did not interfere with the analysis of neuroserpin from Collins' bodies. The protein concentration was determined by Bio-Rad protein assay. Neuroserpin Forms Polymers and a Latent Species in Vitro— Incubation of NS at 1.0 mg/ml and 45 °C in PBS resulted in the formation of polymers (Fig. 1a, right) as described previously (4.Belorgey D. Crowther D.C. Mahadeva R. Lomas D.A. J. Biol. Chem. 2002; 277: 17367-17373Abstract Full Text Full Text PDF PubMed Scopus (88) Google Scholar, 12.Belorgey D. Sharp L.K. Crowther D.C. Onda M. Johansson J. Lomas D.A. Eur. J. Biochem. 2004; 271: 3360-3367Crossref PubMed Scopus (48) Google Scholar). These polymers dissociated to monomeric protein following heating in SDS in keeping with a non-covalent reactive loop-β-sheet A linkage (Fig. 1a, left). However, after prolonged incubation, a faster migrating band was apparent on the non-denaturing gel (arrow). The intensity of this band increased if the polymerization experiments were performed at low concentrations of NS (Fig. 1b), whereas at higher concentrations, the polymerization reaction prevailed. Incubation of NS at 10 μg/ml and 55 °C resulted in the formation of almost pure monomeric neuroserpin that migrated as the lower band on non-denaturing PAGE (Fig. 1b). This same monomeric conformer was also formed by incubation of native S49P at 1 μg/ml and 55 °C (data not shown) and incubation of both NS and S49P under physiological conditions at 37 °C (Fig. 1c). S49P formed the new conformer more readily than wild-type protein at 37 °C as assessed by densitometry scanning analysis. S49P and NS were completely transformed to the new conformer after 48 and 120 h, respectively (Fig. 1c). Characterization of Latent Neuroserpin—The faster migrating species was prepared by heating native NS or S49P at 1 μg/ml and 55 °C in PBS for 24 h. It was then purified by gel filtration (Fig. 1b) and analyzed by CD spectroscopy (Fig. 2a). There were differences in the far-UV CD profile consistent with an increase in β-structure upon transformation of the native form into a latent species. On secondary structure prediction based upon deconvolution of these spectra, latent NS had 4% less α-helices and 14% more β-strands than native NS, and latent S49P had 4% less α-helices and 9% more β-strands than native S49P. The latent conformer of NS was inactive as a proteinase inhibitor against the target enzyme tPA (Fig. 2b) under conditions in which native NS formed a characteristic SDS-stable complex. Similarly, the latent species of S49P, prepared under the same conditions as NS, was inactive against tPA, whereas native S49P formed a stable complex (Fig. 2b). The cardinal feature of latent serpins is the restoration of the native species and inhibitory activity following refolding (35.Hekman C.M. Loskutoff D.J. J. Biol. Chem. 1985; 260: 11581-11587Abstract Full Text PDF PubMed Google Scholar). Refolding of urea-denatured latent NS resulted in a species that had the same migration profile as native NS (Fig. 2c). Likewise, refolding of urea-denatured latent S49P also resulted in a species with the same electrophoretic mobility as native S49P on non-denaturing PAGE (data not shown). Refolding of urea-denatured latent NS or latent S49P resulted in a restoration of inhibitory activity as demonstrated by the ability of the refolded species to once again form an SDS-stable complex with tPA (Fig. 2b). Taken together, these data show that both NS and S49P can form an inactive latent species and that this species can be reactivated by refolding. Stability of Latent Neuroserpin—Previous studies have shown that latent serpins are more stable than their native equivalents. Latent NS and latent S49P were therefore assessed by transverse urea gradient PAGE (Fig. 3a). The upper gels clearly showed that latent NS was more stable than native NS. In contrast, there was no significant difference between latent S49P and native S49P (lower gels). Stability was also assessed by measuring the melting temperature (Tm) of each of the conformers by monitoring the change in CD signal at 216 nm. The Tm values of native NS and native S49P were 56.6 ± 0.3 and 49.9 ± 1.2 °C, respectively (4.Belorgey D. Crowther D.C. Mahadeva R. Lomas D.A. J. Biol. Chem. 2002; 277: 17367-17373Abstract Full Text Full Text PDF PubMed Scopus (88) Google Scholar), whereas those of latent NS and latent S49P were 89.4 ± 0.4 and 61.6 ± 0.5 °C, respectively (Fig. 3b). It was striking that the melting temperature of latent S49P was much lower than that of latent NS. Polymers formed from either native NS or native S49P did not melt at temperatures up to 100 °C. Polymerization of Latent Neuroserpin—The reactive loop is fully incorporated into β-sheet A in a latent serpin, and so it is unable to accept the loop of another molecule to form polymers. In keeping with this, latent NS did not form polymers when incubated at 1 mg/ml and 45 °C for 24 h (Fig. 4a). Indeed, latent NS was resistant to polymer formation when incubated at a higher protein concentration (3 mg/ml) and a higher temperature (55 °C) for 48 h. In marked contrast, latent S49P formed polymers when incubated at 0.25 mg/ml and 37 °C (Fig. 4b). Latent S49P formed polymers less readily than native S49P but more readily than native NS when incubated under these physiological conditions (Fig. 4b). We have shown previously that heating α1-antitrypsin in 0.7 m sodium citrate at 67 °C forces it to adopt the latent conformer and to form short chain polymers (33.Lomas D.A. Elliott P.R. Chang W.-S.W. Wardell M.R. Carrell R.W. J. Biol. Chem. 1995; 270: 5282-5288Abstract Full Text Full Text PDF PubMed Scopus (141) Google Scholar). Heating native NS or native S49P between 37 and 55 °C in 0.5 m citrate resulted in suppression of polymerization and slower formation of the latent conformer (Fig. 5, a and b). In contrast, latent S49P was able to form oligomers in the presence of citrate at temperatures ranging from 37 to 55 °C (Fig. 5c). These oligomers were not formed by heating latent NS in the presence of citrate (Fig. 5d). It is of interest that latent S49P also formed oligomers in the absence of citrate (Figs. 4b and 5c). They were formed even when this conformer was incubated at 55 °C. In contrast, incubation of native NS and native S49P in the same conditions resulted in the formation of long chain polymers and higher ordered aggregates. These results demonstrate that latent S49P can form polymers through a different pathway from either NS or S49P. Conformers of Neuroserpin in Collins' Bodies from Patients with FENIB—The observation that S49P forms the latent species under physiological conditions (Fig. 1c), and remarkably, that it can then polymerize (Fig. 4b), raised the question as to whether the latent conformer is present in inclusion (Collins‘) bodies in patients with FENIB. Collins’ bodies were isolated from the brain of an individual with FENIB as a consequence of the S49P neuroserpin mutation as described previously (11.Davis R.L. Shrimpton A.E. Holohan P.D. Bradshaw C. Feiglin D. Collins G.H. Sonderegger P. Kinter J. Becker L.M. Lacbawan F. Krasnewich D. Muenke M. Lawrence D.A. Yerby M.S. Shaw C.M. Gooptu B. Elliott P.R. Finch J.T. Carrell R.W. Lomas D.A. Nature. 1999; 401: 376-379Crossref PubMed Google Scholar) and then assessed by non-denaturing PAGE and Western blot analysis. In addition to polymers, a monomeric band was also detected in Collins' bodies that had the same migration profile as latent S49P (Fig. 6a, upper left and right). There was no evidence of reactive loop cleavage when neuroserpin from Collins' bodies was assessed by SDS-PAGE (Fig. 6a, lower left). It is difficult to compare the migration of neuroserpin from Collins' bodies with that of recombinant S49P as the later lacks carbohydrate side chains. The electrophoretic mobility of neuroserpin from Collins' bodies was therefore assessed in comparison with neuroserpin secreted by mammalian COS-7 cells transfected with S49P (13.Miranda E. Römisch K. Lomas D.A. J. Biol. Chem. 2004; 279: 28283-28291Abstract Full Text Full Text PDF PubMed Scopus (99) Google Scholar). As shown in Fig. 6a, in the lower left gel, neuroserpin from Collins' bodies had a molecular mass between that of recombinant protein and protein secreted by transfected COS-7 cells in keeping with the glycosylation expected as a result of retention within the endoplasmic reticulum (13.Miranda E. Römisch K. Lomas D.A. J. Biol. Chem. 2004; 279: 28283-28291Abstract Full Text Full Text PDF PubMed Scopus (99) Google Scholar). However, monomeric neuroserpin from Collins' bodies migrated more anodally on non-denaturing gels than both the native recombinant protein and the monomeric protein secreted from the cells (Fig. 6a, upper left). This is in keeping with neuroserpin within Collins' bodies adopting a latent conformation as well as forming polymers in vivo. The monomeric neuroserpin from Collins' bodies was inactive as a proteinase inhibitor when incubated with tPA (Fig. 6b). There was no difference in the electrophoretic migration of monomeric neuroserpin from Collins' bodies before and after treatment with tPA (Fig. 6b, lanes 11 and 12), and furthermore, there was no evidence of cleavage of the reactive center loop (Fig. 6b, lanes 4–6). These results demonstrate that the reactive loop of monomeric neuroserpin from Collins' bodies is inaccessible through formation of a latent species. The polymers identified from the Collins' bodies were a mix of both short and long chain polymers (Fig. 6a, upper right). This compares with only long chain polymers found in the medium from COS-7 cells transfected with S49P (Fig. 6a, upper left gel) (13.Miranda E. Römisch K. Lomas D.A. J. Biol. Chem. 2004; 279: 28283-28291Abstract Full Text Full Text PDF PubMed Scopus (99) Google Scholar) and is in keeping with polymerization of latent S49P. Taken together, these observations provide strong support for the presence of latent S49P within the Collins' bodies of FENIB and are in keeping with a proportion of the polymers in Collins' bodies being formed by oligomerization of latent S49P. The serpinopathies result from conformational transitions of members of the serpin superfamily with aberrant tissue deposition or loss of function. They are best exemplified by point mutations of neuroserpin to cause the inclusion body dementia FENIB and the Z allele of α1-antitrypsin in association with hepatic inclusions and liver disease. Our previous work has defined the pathway of polymerization (Fig. 7, inside broken line). Point mutations in the shutter region (blue circle) allow the formation of an unstable intermediate (M*) that is characterized by inactivity, a raised melting temperature, and the rapid formation of polymers (27.Gooptu B. Hazes B. Chang W.S. Dafforn T.R. Carrell R.W. Read R.J. Lomas D.A. Proc. Natl. Acad. Sci (U. S. A.). 2000; 97: 67-72Crossref PubMed Scopus (176) Google Scholar). In some serpins, the population of M* results in the formation of a latent species in which the reactive loop is fully inserted into β-sheet A (41.Mottonen J. Strand A. Symersky J. Sweet R.M. Danley D.E. Geoghegan K.F. Gerard R.D. Goldsmith E.J. Nature. 1992; 355: 270-273Crossref PubMed Scopus (526) Google Scholar, 42.Carrell R.W. Stein P.E. Fermi G. Wardell M.R. Structure. 1994; 2: 257-270Abstract Full Text Full Text PDF PubMed Scopus (368) Google Scholar, 43.Im H. Woo M.-S. Hwang K.Y. Yu M.-H. J. Biol. Chem. 2002; 277: 46347-46354Abstract Full Text Full Text PDF PubMed Scopus (65) Google Scholar). Such a conformer is inactive as a proteinase inhibitor, thermostable, and cannot polymerize as the annealing site of β-sheet A is occupied. This species can only be induced to form polymers by denaturation and refolding to the native conformer (33.Lomas D.A. Elliott P.R. Chang W.-S.W. Wardell M.R. Carrell R.W. J. Biol. Chem. 1995; 270: 5282-5288Abstract Full Text Full Text PDF PubMed Scopus (141) Google Scholar, 35.Hekman C.M. Loskutoff D.J. J. Biol. Chem. 1985; 260: 11581-11587Abstract Full Text PDF PubMed Google Scholar). We show here that wild-type neuroserpin can form both polymers and a latent species in vitro. Latent NS has 14% more β-strands than native NS (Fig. 2a), is thermostable (Fig. 3b), cannot polymerize (Fig. 4a), and is inactive as a proteinase inhibitor (Fig. 2b). The activity of latent NS can be restored by refolding (Fig. 2b). The formation of latent and polymeric NS is in keeping with the conventional pathway illustrated in Fig. 7, inside the broken line. The predominant species that forms depends upon protein concentration, with concentrated solutions favoring step 2 and the formation of polymers, whereas dilute solutions favor step 3 and latency. It is surprising that latent NS can be produced under physiological conditions (Fig. 1c). The only other serpin that is inactivated by spontaneous transition to a latent species is plasminogen activator inhibitor-1 (35.Hekman C.M. Loskutoff D.J. J. Biol. Chem. 1985; 260: 11581-11587Abstract Full Text PDF PubMed Google Scholar). Both neuroserpin (2.Osterwalder T. Cinelli P. Baici A. Pennella A. Krueger S.R. Schrimpf S.P. Meins M. Sonderegger P. J. Biol. Chem. 1998; 237: 2312-2321Abstract Full Text Full Text PDF Scopus (122) Google Scholar, 3.Hastings G.A. Coleman T.A. Haudenschild C.C. Stefansson S. Smith E.P. Barthlow R. Cherry S. Sandkvist M. Lawrence D.A. J. Biol. Chem. 1997; 272: 33062-33067Abstract Full Text Full Text PDF PubMed Scopus (201) Google Scholar, 4.Belorgey D. Crowther D.C. Mahadeva R. Lomas D.A. J. Biol. Chem. 2002; 277: 17367-17373Abstract Full Text Full Text PDF PubMed Scopus (88) Google Scholar, 5.Barker-Carlson K. Lawrence D.A. Schwartz B.S. J. Biol. Chem. 2002; 277: 46852-46857Abstract Full Text Full Text PDF PubMed Scopus (54) Google Scholar) and plasminogen activator inhibitor-1 (29.Lawrence D.A. Strandberg L. Ericson J. Ny T. J. Biol. Chem. 1990; 265: 20293-20301Abstract Full Text PDF PubMed Google Scholar, 44.Strickland S. Thromb. Haemostasis. 2001; 86: 138-143Crossref PubMed Scopus (46) Google Scholar, 45.Makarova A. Mikhailenko I. Bugge T.H. List K. Lawrence D.A. Strickland D.K. J. Biol. Chem. 2003; 278: 50250-50258Abstract Full Text Full Text PDF PubMed Scopus (57) Google Scholar) inhibit tPA at the neuronal synapse, and therefore, transition to latency may be a common mechanism to regulate the activity of these neuronal serpins in vivo. However, the length of time required to form the latent conformer in vitro means that wild-type neuroserpin may be turned over at the synapse before the latent species forms in vivo. The Syracuse (S49P) mutation of neuroserpin is in the shutter domain (Fig. 7, blue circle). This mutant perturbs β-sheet A to allow the spontaneous formation of polymers that are retained as inclusion bodies in the cerebral cortex in association with dementia (4.Belorgey D. Crowther D.C. Mahadeva R. Lomas D.A. J. Biol. Chem. 2002; 277: 17367-17373Abstract Full Text Full Text PDF PubMed Scopus (88) Google Scholar, 11.Davis R.L. Shrimpton A.E. Holohan P.D. Bradshaw C. Feiglin D. Collins G.H. Sonderegger P. Kinter J. Becker L.M. Lacbawan F. Krasnewich D. Muenke M. Lawrence D.A. Yerby M.S. Shaw C.M. Gooptu B. Elliott P.R. Finch J.T. Carrell R.W. Lomas D.A. Nature. 1999; 401: 376-379Crossref PubMed Google Scholar, 12.Belorgey D. Sharp L.K. Crowther D.C. Onda M. Johansson J. Lomas D.A. Eur. J. Biochem. 2004; 271: 3360-3367Crossref PubMed Scopus (48) Google Scholar, 13.Miranda E. Römisch K. Lomas D.A. J. Biol. Chem. 2004; 279: 28283-28291Abstract Full Text Full Text PDF PubMed Scopus (99) Google Scholar, 14.Davis R.L. Shrimpton A.E. Carrell R.W. Lomas D.A. Gerhard L. Baumann B. Lawrence D.A. Yepes M. Kim T.S. Ghetti B. Piccardo P. Takao M. Lacbawan F. Muenke M. Sifers R.N. Bradshaw C.B. Kent P.F. Collins G.H. Larocca D. Holohan P.D. Lancet. 2002; 359: 2242-2247Abstract Full Text Full Text PDF PubMed Scopus (133) Google Scholar). Our data demonstrate that this mutation also favors the formation of the latent conformer at low concentrations under physiological conditions (Fig. 1c). This transition, like that of polymerization, also serves to inactivate neuroserpin as an inhibitor of its target proteinase tPA (Fig. 2b). The importance of this finding is underscored by the finding of latent S49P in Collins' bodies isolated from patients with disease (Fig. 6). This inactivation of S49P will exacerbate the deficiency of neuroserpin at the synapse that arises from polymerization and the 100-fold decrease in inhibitory activity caused by the S49P mutation (4.Belorgey D. Crowther D.C. Mahadeva R. Lomas D.A. J. Biol. Chem. 2002; 277: 17367-17373Abstract Full Text Full Text PDF PubMed Scopus (88) Google Scholar). These factors combine to increase the concentration of synaptic tPA, which may exacerbate the neuropsychiatric features of FENIB (10.Madani R. Kozlov S. Akhmedov A. Cinelli P. Kinter J. Lipp H.P. Sonderegger P. Wolfer D.P. Mol. Cell. Neurosci. 2003; 23: 473-494Crossref PubMed Scopus (123) Google Scholar, 46.Bradshaw C.B. Davis R.L. Shrimpton A.E. Holohan P.D. Rea C.B. Fieglin D. Kent P. Collins G.H. Arch. Neurol. 2001; 58: 1429-1434Crossref PubMed Scopus (47) Google Scholar). The most striking finding is that far from being inert (Fig. 2b), latent S49P spontaneously forms polymers under physiological conditions (Fig. 4b). Latent S49P has a Tm 27.8 °C lower, and less β-structure, than latent NS. Thus the reactive loop must be less stably incorporated into β-sheet A in latent S49P than it is in latent NS (Fig. 7, compare Lf and Lp). Thus we must now modify the pathways of latency and polymerization. Polymers may form via a back reaction as shown in steps 5 and 2 or directly as shown in step 6. Only latent S49P has the ability to form oligomers in the presence of 0.5 m citrate (Fig. 5), indicating that the second polymerization pathway from step 6 is distinct from that of step 2. The finding of oligomers of latent S49P in vitro and from Collins bodies in vivo highlights the importance of this second pathway in disease. It is currently not possible to define which pathway (either steps 5 and 2 or step 6) is most important in the polymerization of latent S49P in vivo. Nevertheless, the ability of latent S49P to polymerize will increase the burden of polymers within neurones and so exacerbate the neuronal dysfunction and death that underlie FENIB. The structure of the polymers that form from oligomerization of latent S49P (step 6) is unknown. It is recognized that citrate prevents the β-sheet A-linked polymerization of α1-antitrypsin (Fig. 7, step 2) (33.Lomas D.A. Elliott P.R. Chang W.-S.W. Wardell M.R. Carrell R.W. J. Biol. Chem. 1995; 270: 5282-5288Abstract Full Text Full Text PDF PubMed Scopus (141) Google Scholar) and induces linkage via β-sheet C (47.Bottomley S.P. Hopkins P.C. Whisstock J.C. Biochem. Biophys. Res. Commun. 1998; 251: 1-5Crossref PubMed Scopus (24) Google Scholar). Indeed, polymerization of native NS and native S49P, which proceeds via β-sheet A, was suppressed by citrate (Fig. 5, a and b). The formation of the partially loop inserted latent species (Fig. 7, Lp) by S49P causes the liberation of strand 1C, which can then be replaced by the reactive loop of a donor molecule to form C-sheet polymers (42.Carrell R.W. Stein P.E. Fermi G. Wardell M.R. Structure. 1994; 2: 257-270Abstract Full Text Full Text PDF PubMed Scopus (368) Google Scholar). Thus polymerization of latent S49P in the presence of citrate is consistent with the formation of polymers via a C β-sheet pathway. In summary, we have shown that S49P can be inactivated by both polymerization and the formation of a latent species in vivo. This latent conformer is unlike any other latent serpin in that it can spontaneously form polymers under physiological conditions. This allows us to extend the pathways of polymer formation within the serpinopathies and to further define the pathogenic conformers of neuroserpin that contribute to the dementia FENIB. We are grateful to Dr Richard L. Davis, State University of New York Health Science Center, Syracuse, New York for providing the Collins' bodies, Dr. Elena Miranda, Department of Medicine, University of Cambridge, Cambridge Institute for Medical Research, for the culture medium from cells expressing S49P neuroserpin, and all members of the Lomas laboratory for helpful discussions.