Erythropoietic protoporphyria (EPP) is caused by a defect in ferrochelatase, leading to the accumulation of protoporphyrin predominantly in erythrocytes and hepatocytes, and resulting in skin photosensitivity upon leaching of blood protoporphyrin into the skin. Some patients also develop severe liver damage. Because the respective contributions of hepatic and erythrocytic protoporphyrin to the pathophysiology of EPP remain unclear, we investigated this question using the murine model of EPP. Transplantation of bone marrow from EPP mice to normal recipients resulted in elevated erythrocyte and plasma protoporphyrin levels. However, quantification of serum liver enzymes and bilirubin together with histopathologic examination of liver sections of mice up to 16 months post-transplantation showed no evidence of liver damage. Moreover, despite massive elevation of serum protoporphyrin, transplanted mice showed minimal evidence of skin photosensitivity. Photosensitivity could also be prevented locally by implanting skin grafts from normal mice onto the backs of EPP recipients. These data validate the hypothesis that the main source of toxic protoporphyrin originates from the erythrocytes. However, we unexpectedly observed that normal ferrochelatase activity in hepatic and dermal cells of wild-type mice is sufficient to prevent liver disease and significant skin photosensitivity. These findings may provide new strategies for the treatment of EPP.

Definitive cure of an animal model of a human disease by gene transfer into hematopoietic stem cells has not yet been accomplished in the absence of spontaneous in vivo selection for transduced cells. Erythropoietic protoporphyria is a genetic disease in which ferrochelatase is defective. Protoporphyrin accumulates in erythrocytes, leaks into the plasma and results in severe skin photosensitivity. Using a mouse model of erythropoietic protoporphyria, we demonstrate here that ex vivo preselection of hematopoietic stem cells transduced with a polycistronic retrovirus expressing both human ferrochelatase and green fluorescent protein results in complete and long-term correction of skin photosensitivity in all transplanted mice.

The von Willebrand factor (vWF) mediates platelet adhesion to the vascular subendothelium by binding to collagen, other matrix constituents, and the platelet receptor glycoproteins Ib/IX and IIb/IIIa. Although substantial progress has been made in defining vWF structure-function relationships, there are conflicting data regarding the location of its collagen-binding site(s). Possible collagen-binding sites have been localized in the A1 and A3 domains of vWF. To study the proposed binding sites, we have expressed cDNA sequences encoding the A1 and A3 domains of vWF in Escherichia coli and purified the resulting proteins from bacterial inclusion bodies. In addition, a chimeric molecule containing residues 465-598 of the vWF A1 domain polypeptide (vWF-A1) fused in frame to residues 1018-1114 of the vWF A3 domain polypeptide (vWF-A3) was also expressed. Each of the three recombinant proteins purified as a monomer and contained a single disulfide bond. As previously reported (Cruz, M. A., Handin, R. I., and Wise, R. J. (1993) J. Biol. Chem. 268, 21238-21245), recombinant vWF-A1 inhibited ristocetin-induced platelet agglutination, but did not compete with vWF multimers for collagen binding. In contrast, vWF-A3 inhibited the binding of multimeric vWF to immobilized collagen, but did not inhibit ristocetin-induced platelet agglutination. Metabolically labeled vWF-A3 bound to immobilized collagen in a saturable and reversible manner with a Kdof 1.8 × 10-6M. The vWF-A1/A3 chimera was bifunctional. It inhibited vWF binding to platelet glycoprotein Ib/IX with an IC50 of 0.6 × 10-6M and inhibited vWF binding to collagen with an IC50 of 0.5-1.0 × 10-6M. These results, taken together, provide firm evidence that the major collagen-binding site in vWF resides in the A3 domain. The von Willebrand factor (vWF) mediates platelet adhesion to the vascular subendothelium by binding to collagen, other matrix constituents, and the platelet receptor glycoproteins Ib/IX and IIb/IIIa. Although substantial progress has been made in defining vWF structure-function relationships, there are conflicting data regarding the location of its collagen-binding site(s). Possible collagen-binding sites have been localized in the A1 and A3 domains of vWF. To study the proposed binding sites, we have expressed cDNA sequences encoding the A1 and A3 domains of vWF in Escherichia coli and purified the resulting proteins from bacterial inclusion bodies. In addition, a chimeric molecule containing residues 465-598 of the vWF A1 domain polypeptide (vWF-A1) fused in frame to residues 1018-1114 of the vWF A3 domain polypeptide (vWF-A3) was also expressed. Each of the three recombinant proteins purified as a monomer and contained a single disulfide bond. As previously reported (Cruz, M. A., Handin, R. I., and Wise, R. J. (1993) J. Biol. Chem. 268, 21238-21245), recombinant vWF-A1 inhibited ristocetin-induced platelet agglutination, but did not compete with vWF multimers for collagen binding. In contrast, vWF-A3 inhibited the binding of multimeric vWF to immobilized collagen, but did not inhibit ristocetin-induced platelet agglutination. Metabolically labeled vWF-A3 bound to immobilized collagen in a saturable and reversible manner with a Kdof 1.8 × 10-6M. The vWF-A1/A3 chimera was bifunctional. It inhibited vWF binding to platelet glycoprotein Ib/IX with an IC50 of 0.6 × 10-6M and inhibited vWF binding to collagen with an IC50 of 0.5-1.0 × 10-6M. These results, taken together, provide firm evidence that the major collagen-binding site in vWF resides in the A3 domain. INTRODUCTIONThe von Willebrand factor (vWF) 1The abbreviations used are: vWFvon Willebrand factorGPglycoproteinvWF-A1vWF A1 domain polypeptidevWF-A3vWF A3 domain polypeptidevWF-A1/A3recombinant vWF polypeptide containing sequences from the A1 and A3 domainsIPTGisopropyl-β-D-thiogalactopyranosideTBSTris-buffered salinePAGEpolyacrylamide gel electrophoresis. is a multimeric plasma glycoprotein that plays an important role in primary hemostasis (1Turitto V.T. Weiss H.J. Baumgartner H.R. J. Clin. Invest. 1984; 74: 1730-1741Crossref PubMed Scopus (93) Google Scholar, 2Girma J.P. Meyer D. Verweij C.L. Pannekoek H. Sixma J.J. Blood. 1987; 70: 605-611Crossref PubMed Google Scholar). vWF mediates the adhesion of platelets to exposed subendothelium by forming a bridge between collagen, heparin-like glycosaminoglycans and other components of the subendothelium, and platelet receptor sites on glycoproteins Ib/IX and IIb/IIIa. vWF stabilizes adherent platelets under conditions of high flow and shear stress (3De Marco L. Girolami A. Zimmerman T.S. Ruggeri Z.M. J. Clin. Invest. 1985; 75: 1198-1203Crossref PubMed Scopus (113) Google Scholar, 4De Marco L. Girolami A. Zimmerman T.S. Ruggeri Z.M. Proc. Natl. Acad. Sci. U. S. A. 1985; 82: 7424-7428Crossref PubMed Scopus (96) Google Scholar, 5Gralnick H.R. Williams S.B. Coller B.S. J. Clin. Invest. 1985; 75: 19-25Crossref PubMed Google Scholar, 6Weiss H.J. Turitto V.T. Baumgartner H.R. J. Lab. Clin. Med. 1978; 92: 750-764PubMed Google Scholar). Although the sites on vWF that bind to platelet GPIb/IX and GPIIb/IIIa have been well characterized, the collagen-binding site(s) within vWF is not well defined. Results vary with the source of vWF, the type of collagen, and the nature of the binding assay. Previous investigators have reported that vWF binds to collagen types I, III, and VI (7Santoro S.A. Cowan J.F. Collagen Relat. Res. 1982; 2: 31-43Crossref PubMed Scopus (29) Google Scholar, 8Nyman D. Thromb. Res. 1980; 17: 209-214Abstract Full Text PDF PubMed Scopus (19) Google Scholar, 9Santoro S.A. Thromb. Res. 1981; 21: 689-693Abstract Full Text PDF PubMed Scopus (39) Google Scholar, 10Morton L.F. Griffin B. Pepper D.S. Barnes M.J. Thromb. Res. 1983; 32: 545-556Abstract Full Text PDF PubMed Scopus (29) Google Scholar, 11Sixma J.J. Sakariassen K.S. Stel H.V. Houdijk W.P.M. In Der Maur D.W. Hamer R.J. De Groot P.G. Van Mourik J.A. J. Clin. Invest. 1984; 74: 736-744Crossref PubMed Scopus (64) Google Scholar, 12Bockenstedt P. McDonagh J. Handin R.I. J. Clin. Invest. 1986; 78: 551-556Crossref PubMed Scopus (43) Google Scholar, 13Pareti F.I. Fujimura Y. Dent J.A. Holland L.Z. Zimmerman T.S. Ruggeri Z.M. J. Biol. Chem. 1986; 261: 15310-15315Abstract Full Text PDF PubMed Google Scholar, 14Rand J.H. Patel N.D. Schwartz E. Zhou S.L. Potter B.J. J. Clin. Invest. 1991; 88: 253-259Crossref PubMed Scopus (88) Google Scholar, 15Denis C. Baruch D. Kielty C.M. Ajzenberg N. Christophe O. Meyer D. Arterioscler. Thromb. 1993; 13: 398-406Crossref PubMed Google Scholar). Studies with proteolytic fragments of vWF have defined three potential collagen-binding sites in vWF. One is localized in the propeptide, which is cleaved during the assembly of vWF multimers and is unlikely to play a major role in platelet adhesion (16Takagi J. Sekiya F. Kasahara K. Inada Y. Saito Y. J. Biol. Chem. 1989; 264: 6017-6020Abstract Full Text PDF PubMed Google Scholar). The other two sites have been localized to amino acids 542-622 and 948-998 of the mature vWF subunit polypeptide (17Roth G.J. Titani K. Hoyer L.W. Hickey M.J. Biochemistry. 1986; 25: 8357-8361Crossref PubMed Scopus (97) Google Scholar).Analysis of vWF cDNA and its predicted amino acid sequence shows an interesting pattern of homologous repeats (20Shelton-Inloes B.B. Titani K. Sadler J.E. Biochemistry. 1986; 25: 3164-3171Crossref PubMed Scopus (135) Google Scholar, 21Bonthron D.T. Handin R.I. Kaufman R.J. Wasley L.C. Orr E.C. Mitsock L.M. Ewenstein B. Loscalzo J. Ginsburg D. Orkin S.H. Nature. 1986; 324: 270-273Crossref PubMed Scopus (127) Google Scholar, 22Verweij C.L. Diergaarde P.J. Hart M. Pannekoek H. EMBO J. 1986; 5: 1839-1847Crossref PubMed Scopus (186) Google Scholar). There is strong evidence that the vWF-A1 repeat, which encodes amino acids 479-717, contains binding sites for GPIb/IX, glycosaminoglycans, sulfatides, and collagen (23Mohri H. Yoshioka A. Zimmerman T.S. Ruggeri Z.M. J. Biol. Chem. 1989; 264: 17361-17367Abstract Full Text PDF PubMed Google Scholar, 24Fujimura Y. Titani K. Holland L.Z. Roberts J.R. Kostel P. Ruggeri Z.M. Zimmerman T.S. J. Biol. Chem. 1987; 262: 1734-1739Abstract Full Text PDF PubMed Google Scholar, 25Mohri H. Fujimura Y. Shima M. Yoshioka A. Houghten R.A. Ruggeri Z.M. Zimmerman T.S. J. Biol. Chem. 1988; 263: 17901-17904Abstract Full Text PDF PubMed Google Scholar, 26Fujimura Y. Usami Y. Titani K. Niinomi K. Nishio K. Takase T. Yoshioka A. Fukui H. Blood. 1991; 77: 113-120Crossref PubMed Google Scholar, 27Christophe O. Obert B. Meyer D. Girma J.P. Blood. 1991; 78: 2310-2317Crossref PubMed Google Scholar). The vWF-A3 repeat, which encodes amino acids 910-1111 (28Bonthron D. Orr E.C. Mitsock L.M. Ginsburg D. Handin R.I. Orkin S.H. Nucleic Acids Res. 1986; 14: 7125-7127Crossref PubMed Scopus (96) Google Scholar), is also reported to have a collagen-binding site (15Denis C. Baruch D. Kielty C.M. Ajzenberg N. Christophe O. Meyer D. Arterioscler. Thromb. 1993; 13: 398-406Crossref PubMed Google Scholar, 17Roth G.J. Titani K. Hoyer L.W. Hickey M.J. Biochemistry. 1986; 25: 8357-8361Crossref PubMed Scopus (97) Google Scholar, 18Pareti F.I. Niiya K. McPherson J.M. Ruggeri Z.M. J. Biol. Chem. 1987; 262: 13835-13841Abstract Full Text PDF PubMed Google Scholar, 29Kalafatis M. Takahashi Y. Girma J.P. Meyer D. Blood. 1987; 5: 1577-1583Crossref Google Scholar, 30Jorieux S. Pietu G. Gaucher C. Mischler F. Meulien P. Meyer D. Mazurier C. Blood Coagul. & Fibrinolysis. 1990; 1: 375-384Crossref PubMed Scopus (2) Google Scholar). Thus, two of the triplicated A repeats contain sequences that have been implicated in collagen binding. In addition to their sequence similarity, the A1 and A3 repeats each contain a single intrachain disulfide bond. The A1 disulfide bond links Cys-509 and Cys-695, and the A3 disulfide links Cys-923 and Cys-1109, forming a 185-amino acid “loop” structure in each domain (19Andrews R.K. Gorman J.J. Booth W.J. Corino G.L. Castaldi P.A. Berndt M.C. Biochemistry. 1989; 28: 8326-8336Crossref PubMed Scopus (103) Google Scholar, 31Marti T. Rosselet S.J. Titani K. Walsh K.A. Biochemistry. 1987; 26: 8099-8109Crossref PubMed Scopus (202) Google Scholar). Given the similarities in primary and secondary structure, the A1 and A3 domains could easily have overlapping functions.Despite the data obtained with tryptic fragments of vWF (13Pareti F.I. Fujimura Y. Dent J.A. Holland L.Z. Zimmerman T.S. Ruggeri Z.M. J. Biol. Chem. 1986; 261: 15310-15315Abstract Full Text PDF PubMed Google Scholar, 15Denis C. Baruch D. Kielty C.M. Ajzenberg N. Christophe O. Meyer D. Arterioscler. Thromb. 1993; 13: 398-406Crossref PubMed Google Scholar, 17Roth G.J. Titani K. Hoyer L.W. Hickey M.J. Biochemistry. 1986; 25: 8357-8361Crossref PubMed Scopus (97) Google Scholar, 18Pareti F.I. Niiya K. McPherson J.M. Ruggeri Z.M. J. Biol. Chem. 1987; 262: 13835-13841Abstract Full Text PDF PubMed Google Scholar, 19Andrews R.K. Gorman J.J. Booth W.J. Corino G.L. Castaldi P.A. Berndt M.C. Biochemistry. 1989; 28: 8326-8336Crossref PubMed Scopus (103) Google Scholar), there is still some uncertainty regarding the existence of a collagen-binding site in the A1 domain. One early study, utilizing unpurified bacterial lysate, reported an interaction between recombinant vWF-A1 protein in the bacterial lysate and collagen (32Pietu G. Meulien P. Cherel G. Diaz J. Baruch D. Courtney M. Meyer D. Biochem. Biophys. Res. Commun. 1989; 164: 1339-1347Crossref PubMed Scopus (28) Google Scholar). Other investigators who have subsequently expressed and studied the function of purified vWF-A1 protein have reported conflicting results (33Azuma H. Dent J.A. Sugimoto M. Ruggeri Z.M. Ware J. J. Biol. Chem. 1991; 266: 12342-12347Abstract Full Text PDF PubMed Google Scholar, 34Gralnick H.R. Williams S. McKeown L. Kramer W. Krutzsh H. Gorecki M. Pinet A. Garfinkel L.I. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 7880-7884Crossref PubMed Scopus (58) Google Scholar, 35Sugimoto M. Ricca G. Hrinda M.E. Schreiber A.B. Searfoss G.H. Bottini E. Ruggeri Z.M. Biochemistry. 1991; 30: 5202-5209Crossref PubMed Scopus (79) Google Scholar). For example, our laboratory recently reported that a highly purified monomeric vWF-A1 protein bound to platelet GPIb/IX and heparin, but did not bind to collagen (36Cruz M.A. Handin R.I. Wise R.J. J. Biol. Chem. 1993; 268: 21238-21245Abstract Full Text PDF PubMed Google Scholar). In addition, it has been reported that recombinant vWF from which the entire A1 domain has been deleted, which forms the normal spectrum of multimers, no longer binds to platelet GPIb/IX, but still binds to collagen (37Sixma J.J. Schiphorst M.E. Verweij C.L. Pannekoek H. Eur. J. Biochem. 1991; 688: 1-7Google Scholar).To resolve these conflicting results and to learn more about the collagen-vWF interaction, we have cloned and expressed vWF A1 and A3 domain polypeptides and compared their biochemical properties with a chimeric vWF A1 domain polypeptide containing sequences derived from the vWF A1 and A3 domains (vWF-A1/A3). The studies reported here clearly demonstrate that vWF-A3 protein and the vWF-A1/A3 chimera both bind to type I collagen. Results obtained with vWF-A1/A3 help to localize the collagen-binding sequence in the A3 domain to a sequence between amino acids 1018 and 1114. We have concluded that the vWF-A3 binding site can account for all of the interactions between multimeric vWF and type I collagen and that this site probably represents the physiologically relevant collagen-binding site in vWF.EXPERIMENTAL PROCEDURESConstruction of vWF Expression VectorsA map for each of the three vWF cDNAs expressed is shown in Fig. 1. For expression of the vWF A3 domain in Escherichia coli, a cDNA fragment encoding amino acids 908-1111 of mature vWF was constructed by mutagenesis of vWF cDNA in M13 with two oligonucleotides. Oligonucleotide 1 spanned codons 1666-1674 (28Bonthron D. Orr E.C. Mitsock L.M. Ginsburg D. Handin R.I. Orkin S.H. Nucleic Acids Res. 1986; 14: 7125-7127Crossref PubMed Scopus (96) Google Scholar) and introduced a BamHI restriction site (encoding Gly and Ser) at codons 1670 and 1671. Oligonucleotide 2 spanned codons 1870-1882 and introduced a termination codon and a HindIII restriction site. Following sequence confirmation, the vWF-A3 cDNA fragment was isolated by digestion with BamHI and HindIII and inserted into the expression vector pQE9 (QIAGEN Inc.). Insertion of the vWF-A3 fragment in pQE9 produces an amino-terminal fusion protein containing the vWF sequence fused in frame to 10 amino acids (6 histidines) contributed by the vector. vWF-A1 cDNA was prepared as described previously and used to transform pQE9 as well (36Cruz M.A. Handin R.I. Wise R.J. J. Biol. Chem. 1993; 268: 21238-21245Abstract Full Text PDF PubMed Google Scholar).Chimeric vWF-A1/A3 cDNA was constructed with a polymerase chain reaction-based mutagenesis strategy. In the first round of amplification, the amino-terminal half of the A1 domain (residues 475-598) and the carboxyl-terminal half of the A3 domain (residues 1018-1114) were amplified. The 3’-primer for amplifying the A1 domain was designed to contain 12 extra bases of A3 domain sequence (5’-TCGCACAGCAAACAAGACCTCGCTGGTGGA-3’ (vWF-A1 sequence is underlined)). Thus, the amplified A1 cDNA contained a short A3 sequence at its 3’-end. Similarly, A3 cDNA contained a short A1 sequence at its 5’-end (5’-AGCGAGGTCTTGGCTGTGCGATACTTG-3’ (vWF-A1 sequence is underlined)). The two chimeric cDNA fragments were then incubated together so that they were annealed via their overlapping sequences. The annealed material was used as the template for the second round of polymerase chain reaction. The chimeric DNA fragment was amplified using 5’-A1 (5’-CCTCACCTGTGAAGGATCCCAGGAGCCGGGAG-3’) and 3’-A3 (5’-CATTCCAAGCTTGAATTCATCAAGATCTAACAAATCCAGAGC-3’) primers designed to introduce BamHI and HindIII restriction sites, respectively, for cloning as described previously above. The chimeric vWF-A1/A3 cDNA fragment was isolated by digestion with BamHI and HindIII and used to transform pQE9.Purification of Recombinant ProteinsE. coli M15(pREP4) cells (QIAGEN Inc.) containing pQE9-vWF-A3, pQE9-vWF-A1, or pQE9-vWF-A1/A3 were cultured overnight at 37°C in 8 liters of 25 g/liter Tryptone, 15 g/liter yeast extract, 5 g/liter NaCl, pH 7.3, containing 100 μg/ml ampicillin and 25 μg/ml kanamycin. The overnight culture was diluted 1:20 and grown to A595 = 0.7. The culture was adjusted to 1.5 mM IPTG and incubated for 5 h at 37°C. The cells were then harvested, resuspended in 125 ml of lysis buffer (50 mM Tris-HCl, 0.1 M NaCl, 1 mM EDTA, pH 8.0) containing lysozyme at a final concentration of 250 μg/ml, and allowed to stand for 1 h at 4°C. The bacterial cells were lysed in the presence of 1.25 mg/ml deoxycholic acid and 7 μg/ml DNase I. The lysate was centrifuged at 12,000 × g for 15 min, and the resulting pellet was washed with lysis buffer containing 0.5% Triton X-100 and 10 mM EDTA, followed by recentrifugation.For purification of vWF-A3 protein, the washed pellet was solubilized by the addition of 7.5 M urea in 25 mM Tris-HCl, pH 8.8, and the solubilized proteins were dialyzed against 25 mM Tris-HCl, pH 8.2. The solubilized proteins were passed over a Q-Sepharose column (Pharmacia Biotech Inc.) equilibrated with 25 mM Tris-HCl, pH 8.2. vWF-A3 eluted from the column with sodium chloride. The fractions containing vWF-A3 were pooled, concentrated by ultrafiltration, and dialyzed against Tris-buffered saline (TBS; 25 mM Tris-HCl, 150 mM NaCl, pH 7.4)Recombinant vWF-A1 or chimeric vWF-A1/A3 protein was expressed in E. coli as described above. They were both purified as described previously (36Cruz M.A. Handin R.I. Wise R.J. J. Biol. Chem. 1993; 268: 21238-21245Abstract Full Text PDF PubMed Google Scholar). Briefly, after induction with IPTG and collection of inclusion bodies from lysed E. coli cells, the washed inclusion bodies were solubilized in 6 M guanidine hydrochloride, dialyzed into 6 M urea, and purified by fast protein liquid chromatography on an S-Sepharose column (Pharmacia Biotech Inc.). Urea was removed by slow dialysis, and the recombinant proteins were dialyzed against TBS (36Cruz M.A. Handin R.I. Wise R.J. J. Biol. Chem. 1993; 268: 21238-21245Abstract Full Text PDF PubMed Google Scholar). Each of the purified recombinant proteins was then concentrated by adsorption to and elution from a heparin column (Bio-Rad) and/or ultrafiltration and dialyzed against TBS.Radiolabeling of vWF-A3 ProteinE. coli cells were grown overnight and diluted in medium described above. When grown to A595 = 0.7, the cells (500 ml) were pelleted by centrifugation, washed with M9 minimal medium (6 g of Na2HPO4/liter, 3 g of KH2PO4/liter, 0.5 g of NaCl/liter, 1 g of NH4Cl/liter, 1 mM MgCl, 0.1 mM CaCl2, and 0.2% glucose), and pelleted again. The pellet was resuspended in 500 ml of M9 minimal medium supplemented with a 0.02% concentration of 18 amino acids except Met and Cys. Bacteria were grown for 45 min; IPTG was added; and the medium was supplemented with 1.5 μCi/ml [35S]H2SO4 (DuPont NEN) overnight at 37°C. Labeled vWF-A3 was purified as described above. Specific activity of 35S-labeled vWF-A3 was 3.6 × 104 cpm/μg.Production and Purification of Multimeric Recombinant vWFRecombinant vWF was purified from the conditioned medium of Chinese hamster ovary cells that had been stably transformed with full-length vWF cDNA as described previously (21Bonthron D.T. Handin R.I. Kaufman R.J. Wasley L.C. Orr E.C. Mitsock L.M. Ewenstein B. Loscalzo J. Ginsburg D. Orkin S.H. Nature. 1986; 324: 270-273Crossref PubMed Scopus (127) Google Scholar, 39Wise R.J. Pittman D.D. Handin R.I. Kaufman R.J. Orkin S.H. Cell. 1988; 52: 229-236Abstract Full Text PDF PubMed Scopus (120) Google Scholar, 40Wise R.J. Dorner A.J. Krane M. Pittman D.D. Kaufman R.J. J. Biol. Chem. 1991; 266: 21948-21955Abstract Full Text PDF PubMed Google Scholar). Confluent cells were rinsed and incubated for 24-48 h in α-minimal essential medium containing 0.05% bovine serum albumin, insulin/transferrin/sodium selenite supplement (Sigma), and 1% (v/v) aprotinin (Sigma, A-6279). The conditioned medium was collected and centrifuged to remove cellular debris, and Na2EDTA and phenylmethanesulfonyl fluoride were added to final concentrations of 5 and 2 mM, respectively. To isolate vWF wild-type protein, Chinese hamster ovary cell-conditioned medium was concentrated in dialysis tubing by using Aquacide II (Calbiochem) and passed over a Sepharose CL-4B gel filtration chromatography column (Pharmacia Biotech Inc.; 5 × 100 cm). Fractions containing vWF protein were pooled, concentrated by Aquacide II, and dialyzed against TBS. For radiolabeling of vWF produced in Chinese hamster ovary cells, Cys- and Met-deficient serum-free medium was supplemented with 10 μCi/ml [35S]Met and [35S]Cys (39Wise R.J. Pittman D.D. Handin R.I. Kaufman R.J. Orkin S.H. Cell. 1988; 52: 229-236Abstract Full Text PDF PubMed Scopus (120) Google Scholar, 40Wise R.J. Dorner A.J. Krane M. Pittman D.D. Kaufman R.J. J. Biol. Chem. 1991; 266: 21948-21955Abstract Full Text PDF PubMed Google Scholar). Radiolabeled vWF was purified as described above. Specific activity of 35S-labeled vWF was 2.0 × 104 cpm/μg.Collagen Binding AssayA final concentration of 1.8 mg/ml acid-soluble bovine type I collagen (Collaborative Biomedical, Boston) was added to microtiter wells in 20 mM sodium citrate buffer, pH 6.0, for 90 min. An acid-soluble calfskin type I collagen (Sigma) was also used in some experiments. After washing three times with TBS to remove nonadsorbed collagen, wells were blocked with 1% bovine serum albumin in TBS for 30 min. Increasing concentrations of 35S-labeled vWF proteins were added to the wells and incubated for 60 min at room temperature. For competition assays, a constant concentration (1-4 μg/ml) of 35S-labeled vWF was added to the wells with increasing concentrations of the unlabeled ligands. Wells were washed with TBS, and bound radioactivity was removed for scintillation counting by overnight incubation in 1% SDS/TBS. Nonspecific binding was determined in the presence of a 40-fold excess of nonradioactive vWF-A3 or a 50-fold excess of nonradioactive vWF. In each case, nonspecific binding was always <5% of total binding. The amount of collagen bound to wells was measured in duplicate as described previously (12Bockenstedt P. McDonagh J. Handin R.I. J. Clin. Invest. 1986; 78: 551-556Crossref PubMed Scopus (43) Google Scholar). Bound collagen was removed with 1% SDS and submitted to protein quantitation. At the concentration used in the assays, 6 ± 1.0 μg of collagen bound to each well.Platelet Agglutination AssayRistocetin-induced platelet agglutination was carried out in siliconized glass cuvettes at 37°C with constant stirring at 1200 rpm in a four-channel aggregometer (Bio/Data Corp.). A suspension of 2 × 108/ml formaldehyde-fixed platelets containing 8 μg/ml purified vWF (41Cruz M.A. Petersen E. Turci S.M. Handin R.I. J. Biol. Chem. 1992; 267: 1303-1309Abstract Full Text PDF PubMed Google Scholar) and increasing concentrations of the recombinant proteins was prepared. After 5 min of incubation at 37°C, agglutination was initiated by the addition of ristocetin (Sigma) to a final concentration of 1 mg/ml.Protein QuantitationProtein concentrations were determined by the bicinchoninic acid method (Pierce). Purity was assessed by Coomassie Blue staining of SDS-polyacrylamide gels (42Laemmli U.K. Nature. 1970; 227: 680-685Crossref PubMed Scopus (206048) Google Scholar). Radiolabeled protein was visualized by autoradiography utilizing ENHANCE (DuPont NEN). Gel filtration analysis was carried out in a Sephacryl 300-HR column (Sigma; 0.8 × 30 cm) using a Waters 650E-APPS apparatus (36Cruz M.A. Handin R.I. Wise R.J. J. Biol. Chem. 1993; 268: 21238-21245Abstract Full Text PDF PubMed Google Scholar).RESULTSProduction, Purification, and Collagen-binding Activity of vWF-A3After induction with IPTG, transformed bacteria expressing vWF-A3 cDNA, pQE9-vWF-A3, were lysed, and their inclusion bodies were collected. As shown in Fig. 2 A, when analyzed by SDS-PAGE under reducing conditions, the washed inclusion bodies contained a prominent protein band of 27,000 Da. The inclusion bodies could be solubilized in 7.5 M urea, and all of the protein remained in solution after dialysis against 20 mM Tris, pH 8.2. The soluble proteins were then fractionated by fast protein liquid chromatography using a Q-Sepharose ion-exchange column and a linear NaCl gradient. As shown in Fig. 3, vWF-A3 eluted in a sharp peak at 160 mM NaCl. This peak represented ~50% of the total vWF-A3 protein expressed. The remaining 50%, which formed high molecular mass aggregates, eluted in the flow-through volume of the column. The final yield of purified monomeric protein was 8 mg/liter of bacterial culture. To assess the purity of vWF-A3, radiolabeled protein was produced by incubating E. coli containing pQE9-vWF-A3 in medium containing 35SO4 and IPTG. Radiolabeled vWF-A3 was then purified by the procedure described above. A single radiolabeled band was seen after purification by SDS-PAGE and autoradiography (Fig. 2 B, lane4).Figure 2:SDS-PAGE analysis of recombinant vWF-A3 protein. Bacterial inclusion bodies and purified bacterial vWF-A3 protein were analyzed by SDS-PAGE (12.5%). A, the gel, analyzed under reducing conditions, shows molecular mass markers (lane1) and washed bacterial inclusion bodies (lane2) from E. coli transformed with pQE-vWF-A3 and induced with IPTG. B, lanes 1, 2, and 4 were analyzed under reducing conditions. Lane1 shows molecular mass markers. Lane 2 shows vWF-A3 eluted from a Q-Sepharose column. Lane 3 shows vWF-A3 analyzed under nonreducing conditions. Lane 4 shows an autoradiograph of vWF-A3 metabolically labeled with 35SO4.View Large Image Figure ViewerDownload Hi-res image Download (PPT)Figure 3:Ion-exchange chromatography of vWF-A3. Following solubilization of inclusion bodies, vWF-A3 was purified using a Q-Sepharose column equilibrated with 20 mM Tris-HCl, pH 8.2. The bound protein was eluted with a gradient of NaCl. The eluted protein was concentrated by ultrafiltration and dialyzed against TBS. Abs, absorbance.View Large Image Figure ViewerDownload Hi-res image Download (PPT)The calculated molecular mass for the sequence between Ser-908 and Gly-1111 is 21,538 Da. The 10 additional amino acids from the vector sequence add another 1254 Da, bringing the estimated molecular mass to 22,792 Da. This is in good agreement with the estimated molecular mass of the purified material of 27,000 Da (Fig. 2 B, lane2). In addition, purified vWF-A3 eluted from a Sephacryl 300 column with the Kav of a globular monomeric 24-kDa protein (data not shown). When purified vWF-A3 was analyzed by SDS-PAGE under nonreducing conditions (Fig. 2 B, lane3), it migrated slightly faster than the reduced form, suggesting a compact globular structure that is extended following reduction of the single disulfide bond between Cys-923 and Cys-1109.As shown in Fig. 4, 35S-labeled vWF-A3 bound to immobilized soluble type I collagen derived from bovine Achilles tendon or calfskin collagen in a saturable and reversible manner. At the highest concentration of added 35S-labeled vWF-A3, nonspecific binding accounted for <5% of total bound radioactivity. The relevant binding parameters for vWF-A3 were derived by Scatchard analysis of binding isotherms (Fig. 4, inset). The Scatchard plot demonstrated a single class of binding sites with a Kdof 1.8 ± 0.3 μM. At saturation, there were 300 fmol of vWF-A3 bound per μg of immobilized collagen.Figure 4:Binding of 35S-labeled recombinant vWF-A3 protein to collagen-coated microtiter wells. A final concentration of 1.8 mg/ml acid-soluble bovine type I collagen was added to microtiter wells in 20 mM sodium citrate buffer, pH 6.0, for 90 min at 37°C. After washing with TBS to remove nonadsorbed collagen, wells were blocked by the addition of 1% bovine serum albumin for 60 min at room temperature. Increasing concentrations of 35S-labeled recombinant vWF-A3 were added to the wells and incubated for 30 min at 37°C. Wells were washed with TBS, and bound radioactivity was removed with 1% SDS/TBS and counted. Nonspecific binding, measured in parallel wells with the addition of a 40-fold excess of unlabeled vWF-A3 protein, was subtracted from each point. The values shown represent specific binding from five separate experiments. B, bound; F, free.View Large Image Figure ViewerDownload Hi-res image Download (PPT)Characterization of Chimeric vWF-A1/A3 ProteinThe purified vWF-A1/A3 chimera, when analyzed by SDS-PAGE under reducing conditions, had a band of 32,000 Da. This is in good agreement with the predicted molecular mass of 27,000 Da. The anticipated shift in mobility between reduced and unreduced protein was also demonstrated, suggesting formation of a disulfide bond between Cys-509 and Cys-1109 (data not shown). As shown in Fig. 5, both vWF-A1 and vWF-A1/A3 proteins inhibited ristocetin-induced platelet agglutination in a dose-dependent manner. The IC50 for both proteins was between 200 and 600 nM. This finding provides additional evidence for proper folding of the chimeric protein and helps to localize the GPIb/IX-binding domain to the first 98 residues of the vWF A1 domain. In contrast, at concentrations up to 2 μM, vW

The cDNA for human ferrochelatase, the enzyme that is defective in the rare genetic disease erythropoietic protoporphyria (EPP), was tested for its ability to allow the expression of ferrochelatase in mammalian cells. The cDNA was ligated to the plasmid expression vectors pCD and pED6 and transfected into COS-1 and CHO-DUKX cells, respectively. In each case, ferrochelatase activity increased. The cDNA was also ligated into the retroviral vector pLXSN, and virus-packaging cells were produced. Supernatants from these cells were used to infect fibroblasts in vitro from a patient with EPP. We found that the infected cells containing the ferrochelatase cDNA had enzyme levels in the range of normal fibroblasts and that they did not accumulate protoporphyrin when grown in the presence of delta-aminolevulinic acid. We conclude that introducing the cDNA for normal ferrochelatase into fibroblasts from an EPP patient restores ferrochelatase enzyme activity to the normal range. These experiments suggest potential for genetic therapy in EPP.

The differential diagnosis of the genetic bleeding disorders, hemophilia A and von Willebrand disease, is occasionally confounded by the close molecular relationship of coagulation factor VIII and von Willebrand factor (vWF). This report describes the autosomal inheritance of a hemophilia A phenotype due to a mutation of vWF that results in defective factor VIII binding. The proband was a female patient with low levels of factor VIII activity. Polymerase chain reaction (PCR) amplification and DNA sequencing were employed to examine exons encoding the putative factor VIII binding domain of vWF. The patient was found to be homozygous for a single point mutation causing a Thr --> Met substitution at amino acid position 28 in the mature vWF subunit. The phenotypic expression of the mutation was determined to be recessive because heterozygous family members were clinically unaffected. Recombinant vWF containing the observed amino acid substitution was expressed in COS-1 cells. The mutant vWF was processed and secreted normally, and was functionally equivalent to wild-type vWF in its ability to bind to platelets. However, the mutant failed to bind factor VIII, demonstrating that the mutation was functionally related to the observed hemophilia phenotype. The family we describe demonstrates the recessive inheritance of a recently recognized class of genetic bleeding disorders, we call ''autosomal hemophilia.'' We conclude that vWF mutation may be an under recognized cause of hemophilia, especially in cases where the inheritance pattern is not consistent with X-linked transmission.

Paleoecological analysis of the sediment record of 12 Adirondack lakes reveals that the 8 clearwater lakes with current pH < 5.5 and alkalinity < 10 μeq l-1 have acidified recently. The onset of this acidification occurred between 1920 and 1970. Loss of alkalinity, based on quanitative analysis of diatom assemblages, ranged from 2 to 35 μeq l-1. The acidification trends are substantiated by several lines of evidence including stratigraphies of diatom, chrysophyte, chironomid, and cladoceran remains, Ca:Ti and Mn:Ti ratios, sequentially extracted forms of Al, and historical fish data. Acidification trends appear to be continuing in some lakes, despite reductions in atmospheric sulfur loading that began in the early 1970s. The primary cause of the acidification trend is clearly increased atmospheric deposition of strong acids derived from the combustion of fossil fuels. Natural processes and watershed disturbances cannot account for the changes in water chemistry that have occurred, but they may play a role. Sediment core profiles of Pb, Cu, V, Zn, S, polycyclic aromatic hydrocarbons, magnetic particles, and coal and oil soot provide a clear record of increased atmospheric input of materials associated with the combustion of fossil fuels beginning in the late 1800s and early 1900s. The primary evidence for acidification occurs after that period, and the pattern of water chemistry response to increased acid inputs is consistent with current understanding of lake-watershed acidification processes.

Platelet-derived growth factor (PDGF) is a cationic dimer composed of two chains, designated A and B. All three dimeric isotypes of PDGF, PDGF-AA, -AB and -BB, are biologically active but may have distinct functional activities. Two A-chain precursors which differ by the presence of a highly basic 15 amino acid C-terminal extension are derived from the A-chain by alternative RNA splicing. To compare the functional properties of these two different forms of the A-chain, expression vectors were generated in which the cDNAs were placed under the transcriptional control of a viral promoter (pSV2). Surprisingly, cryptic RNA splice donor sites were identified in both forms of the PDGF A-chain which modify the A-chain open reading frame and alter the structure of the expressed protein. Recognition of this phenomenon appears to explain the discrepancies between previous results regarding the secretory properties of the PDGF A-chain and may explain difficulties in expression vectors containing splice acceptor sites between the inserted sequence and the polyadenylation site.

The biosynthesis of von Willebrand Factor (vWF) by vascular endothelial cells involves a complex series of processing steps that includes proteolytic cleavage of a 741-residue propeptide and the assembly of disulfide-linked multimers. Using a model system in which experimentally altered vWF cDNAs are expressed in COS-1 cells, we have shown that the vWF propeptide contains determinants that govern the assembly of vWF multimers. Furthermore, the role of the propeptide (in the assembly process) does not require it to be a contiguous part of the pro-vWF primary structure, since independently expressed propeptide was shown to promote the assembly of mature vWF subunits into multimers. Pulse-chase experiments indicated that the independently expressed propeptide formed a transient association with the mature vWF subunit inside the cell. Thus, it appears that the vWF propeptide segment can act in "trans" to direct the assembly of disulfide-linked vWF multimers.

The PIRLA project is an interdisciplinary paleoecological study designed to provide reconstructions of the recent acidification histories of a representative set of lakes in four acid-sensitive regions in North America. We are trying to determine if lakes in the study regions have acidified, and if so, to what extent, over what time period and why. Sediment cores from 5 to 15 lakes in each region are being analyzed for several characteristics. Diatoms and chrysophytes are being used to reconstruct lakewater pH. Results for three Adirondack lakes with current pH of 4.8 to 5.0 indicate a decrease in pH beginning in the 1930's–1950's. Increased atmospheric deposition of strong acids appears to be the primary factor responsible for the pH decline. Two lakes (pH 4.4 and 4.7) in New England show clear evidence of acidification probably due to acidic deposition. Preliminary reconstructions for two lakes in Michigan (pH 4.4 and 5.6), one in Wisconsin (pH 5.3), and one in Minnesota (pH 6.8) suggest no recent pH decrease. For, the one Florida lake (pH 4.4) analyzed, inferred pH decreases by about 0.5 unit, beginning in the 1950s; the cause has not been determined.

ORIGINAL ARTICLEINTERNATIONAL ABSTRACTS OF PLASTIC AND RECONSTRUCTIVE SURGERY: GENITOURINARY: PDF Only

Houston, Texas * Department of Plastic Surgery, Baylor University School of Medicine, Houston, Texas.