Bovine cardiac troponin T: amino acid sequences of the two isoforms

Troponin T (TnT) is the tropomyosin-binding subunit of troponin, the thin filament regulatory complex that confers calcium sensitivity to striated muscle contraction and actomyosin ATPase activity. Bovine cardiac muscle contains two isoforms (TnT-1 and TnT-2) of TnT that differ in sequence near their amino termini. Thin filaments containing TnT-2 require less calcium to activate the MgATPase rate of myosin than do thin filaments containing TnT-1. Using whole troponin T purified from adult bovine cardiac muscle, we have determined the complete amino acid sequence of the larger, more abundant isoform TnT-1. We confirmed that sequence differences between TnT-1 and TnT-2 are confined to the amino-terminal regions and found that TnT-1 makes up approximately 75% of the total troponin T isolated. Partial sequencing of the separated isoforms showed that the difference between them is due solely to residues 15-19 (Glu-Ala-Ala-Glu-Glu) of TnT-1 being absent from TnT-2. The deleted segment may correspond to the product of exon 4 of the chicken cardiac TnT gene [Cooper, T.A., & Ordahl, C.P. (1985) J. Biol. Chem. 260, 11140-11148]. Exon 5, which is developmentally regulated in the chicken, is not expressed in either TnT-1 or TnT-2. TnT-1 contains 284 amino acid residues and has a Mr of 33,808, while TnT-2 contains 279 amino acid residues and has a Mr of 33,279. Bovine cardiac TnT contains the only known thiol group in any isolated TnT (Cys-39 of TnT-1, Cys-34 of TnT-2). Comparison of bovine, rabbit, and chicken cardiac TnT sequences shows near identity of the amino-terminal 13 amino acid residues (exons 2 and 3 of the chicken cardiac gene), many differences in the following 60 residues (exons 4-8), and great similarity in the C-terminal 230 residues (exons 9-18).     

Primary structure of human alpha 2-macroglobulin. I. Isolation of the 26 CNBr fragments, amino acid sequence of 13 small CNBr fragments, amino acid sequence of methionine-containing peptides, and alignment of all CNBr fragments

The isolation of the 26 CNBr fragments from the identical Mr = 180,000 subunits of human alpha 2-macroglobulin is described. The fragments have been purified by combinations of gel chromatography, ion-exchange chromatography, high voltage paper electrophoresis, paper chromatography, and high performance liquid chromatography. The complete amino acid sequences of 13 small CNBr fragments have been determined. These fragments include CB1 (residues 1-9), CB3 (residues 79-98), CB4 (residues 99-128), CB9 (residues 442-477), CB10 (residues 478-497), CB13 (residues 644-650), CB14 (residues 651-665), CB15 (residues 666-674), CB16 (residues 675-690), CB19 (residues 937-945), CB20 (residues 946-954), CB24 (residues 1356-1362), and CB25 (residues 1363-1375). The fragments determined account for 200 of the 1451 residues of the subunits of alpha 2-macroglobulin. Most likely, Cys-6 of CB9 is bound to the corresponding residue in CB9 from another subunit, thus forming an interchain disulfide bridge in alpha 2-macroglobulin. Cys-1 of CB15 is bound to Cys-35 of CB12. CB15 contains a pair of Gln residues that can react covalently with amines in a factor XIIIa-catalyzed process (Gln-5 and Gln-6). CB16 contains the primary cleavage sites for proteinases in the bait region of alpha 2-macroglobulin (-Arg7-Val-Gly-Phe-Tyr-Glu-). CB20 contains the residues which in native alpha 2-macroglobulin presumably form an internal reactive beta-cysteinyl-gamma-glutamyl thiol ester (Cys-4 and Glx-7). Partial NH2- and COOH-terminal sequence data are given for the 13 large CNBr fragments. Complete or partial sequence determination of 19 methionine-containing peptides or variants thereof allow the alignment of all the CNBr fragments.     

Multiforms of mammalian adenylate kinase and its monoclonal antibody against AK1

An attempt has been made to determine the intracellular distribution of the multiforms of the adenylate kinase (AK) isoenzymes in mammalian tissues, to shed some light on their physiological roles, especially in energy metabolism. The adenylate kinase zymograms obtained from isoelectric focusing yielded two typical isoform patterns: (1) with a pI greater than or equal to 9 and 8.6, specific for bovine skeletal muscle, heart, aorta and brain, and (2) with a pI = 7.9 and 7.1, specific for liver and kidney. Pattern (1) was attributed to the cytosolic isoenzyme (AK1) as demonstrated by immunostaining with anti-AK1. Pattern (2) was attributed to the mitochondrial isoenzyme (AK2). These results were largely confirmed by chromatofocusing experiments. The AK1 isoenzyme was partially purified from the cytosol fraction of bovine aortic smooth muscle and had an apparent Mr of 23.5 kilodaltons. Its kinetic features are discussed from a comparative standpoint. Finally, the human serum AK1 isoform was also detected by Western blotting with a monoclonal antibody directed against crystalline porcine muscle AK1. These results are to form the basis of further studies on the 'aberrant' adenylate kinase isoenzyme from the serum of Duchenne muscular dystrophics.     

Primary structure of hemorrhagic protein, HR2a, isolated from the venom of Trimeresurus flavoviridis

The complete amino acid sequence and disulfide bridge location of HR2a, one of the hemorrhagic proteins isolated from the snake venom of Trimeresurus flavoviridis, have been determined by analysis of peptides derived from digests with cyanogen bromide, lysyl endopeptidase, trypsin, and Staphylococcus aureus V8 protease. Peptides were purified by gel filtration followed by reversed-phase HPLC. HR2a has the amino-terminal sequence of less than Glu-Gln-Arg- and consists of a total of 202 residues with a calculated molecular weight of 23,015. Sequence analysis indicates the presence of another isoform which lacks the amino-terminal residue, making 201 amino acid residues with a molecular weight of 22,887. Three disulfide bridges of HR2a link Cys-118 to Cys-197, Cys-159 to Cys-181, and Cys-161 to Cys-164. HR2a contains a segment which is similar to the zinc-chelating sequences found in thermolysin and several mammalian metalloproteinases, suggesting that HR2a is a metalloproteinase with limited substrate specificity. However, there is no other significant sequence homology with thermolysin except for the zinc-ligand region.     

Mutational analysis of disulfide bridges in the Mr 46,000 mannose 6-phosphate receptor. Localization and role for ligand binding

Formation of intramolecular disulfide bonds is a key step in the early maturation of newly synthesized Mr 46,000 mannose 6-phosphate receptors to acquire ligand-binding activity (Hille, A., Waheed, A., and von Figura, K. (1990) J. Cell Biol. 110, 963-972). The luminal domain of the receptor, which carries the ligand-binding site, contains 6 cysteine residues. We have analyzed the function of individual cysteine residues for the ligand-binding conformation by exchanging cysteine for glycine. In each case, the replacement of cysteine resulted in a complete loss of binding activity, indicating that all 6 luminal cysteine residues are required for the ligand-binding conformation. The cysteine mutants displayed a greatly reduced immunoreactivity, decreased stability, and a blocked or delayed transport to the trans Golgi. The glycosylation pattern allowed the distinguishing of three phenotypes, each of which was represented by one pair of cysteine mutants. Based on the assumption that replacement of either of the 2 cysteine residues forming a disulfide bond results in an identical phenotype, we postulate that disulfide bonds are formed between Cys-32 and Cys-78 and between Cys-132 and Cys-167, as well as between Cys-145 and Cys-179. This assumption was supported by the observation that the simultaneous exchange of the 2 cysteine residues of a putative pair resulted in the same phenotypes as the single exchange of either of the 2 cysteine residues.     

Thiol peptides from the aspartate transaminase of chicken heart cytosol

Aspartate transaminase from chicken heart cytosol was immobilized covalently on activated thiol-Sepharose and digested with trypsin. After washing, the thiol-containing peptides were eluted with 2-mercaptoethanol and further purified by gel-filtration and paper chromatography. Three pure cysteinyl peptides were isolated. One of them may be represented as Ile-(Asp, Met, Cys, Gly, Leu, Thr2)-Lys; this peptide is identical to the fragment comprizing residues 387--395 in the peptide chain of aspartate transaminase from pig heart cytosol. It thus contains a cysteine residue homologous to Cys-390 of the pig heart enzyme. The second cysteinyl peptide had the following composition and partial sequence: Tyr-Phe-Val-Ser-Glu-Gly-Phe-Glu-Leu-Phe (Cys, Ala, Glu, Ser2, Phe)Lys, which corresponds to the sequence 242--258 of the pig enzyme and thus contains a cysteine residue homologous to Cys-252. The third cysteinyl peptide was similar to the tryptic peptide of the pig enzyme containing Cys-191.     

Methylmalonyl-CoA mutase from Propionibacterium shermanii. Evidence for the presence of two masked cysteine residues.

Adenosylcobalamin-dependent methylmalonyl-CoA mutase from Propionibacterium shermanii contains no intramolecular disulphide bridges, but two of the six thiol groups in the heterodimer are only revealed after reduction of the denatured enzyme with dithiothreitol. The available evidence suggests that they are present in disulphide linkages to unknown thiols of low Mr. The two specifically masked cysteine residues are Cys-535 in the alpha-subunit and Cys-517 in the beta-subunit, which occupy exactly homologous positions in each chain.     

Effects of interchain disulfide cross-links on the trypsin cleavage pattern and conformation of myosin subfragment 2

The ability of 5,5'-dithiobis(2-nitrobenzoate) (Nbs2) to produce interchain disulfide cross-links in both the long and short forms of myosin subfragment 2 (S2) and the conformational effects of these cross-links have been investigated. Short S2 (residues 3-287) contains two pairs of Cys residues at positions 66 and 108, and long S2 (residues 1-440) contains an additional pair at position 410. The reaction kinetics of each form of S2 with Nbs2 was biphasic. During the fast kinetic phase the reaction resulted in un-cross-linked species having Nbs-blocked Cys. During the slow phase disulfide-cross-linked species were formed via interchain S-Nbs/SH exchange. For short S2, Cys-66 appeared to react without forming disulfide cross-links, and the Cys- 108 pair reacted with partial cross-linking. For long S2, the Cys-66 pair appeared to react with partial cross-linking, and the Cys pairs at 108 and 410 reacted with complete cross-linking. Mild tryptic digestion of disulfide-cross-linked long S2, under conditions that resulted in partial production of short S2 from un-cross-linked LS2, produced peptides T1a and T1b (residues 1 to approximately 360), with one and two disulfide cross-links, respectively. Further digestion of cross-linked long S2 or cross-linked short S2 resulted in the same shorter fragment, T2, with an NH2-terminus beginning at 103 consistent with a sequence of residues 103-287. Circular dichroism studies on long S2 indicated that the presence of disulfide cross-links changed the thermal unfolding profile of the helix.(ABSTRACT TRUNCATED AT 250 WORDS)     

Analysis of amino acid residues involved in catalysis of polyethylene glycol dehydrogenase from Sphingopyxis terrae, using three-dimensional molecular modeling-based kinetic characterization of mutants

Polyethylene glycol dehydrogenase (PEGDH) from Sphingopyxis terrae (formerly Sphingomonas terrae) is composed of 535 amino acid residues and one flavin adenine dinucleotide per monomer protein in a homodimeric structure. Its amino acid sequence shows 28.5 to 30.5% identity with glucose oxidases from Aspergillus niger and Penicillium amagasakiense. The ADP-binding site and the signature 1 and 2 consensus sequences of glucose-methanol-choline oxidoreductases are present in PEGDH. Based on three-dimensional molecular modeling and kinetic characterization of wild-type PEGDH and mutant PEGDHs constructed by site-directed mutagenesis, residues potentially involved in catalysis and substrate binding were found in the vicinity of the flavin ring. The catalytically important active sites were assigned to His-467 and Asn-511. One disulfide bridge between Cys-379 and Cys-382 existed in PEGDH and seemed to play roles in both substrate binding and electron mediation. The Cys-297 mutant showed decreased activity, suggesting the residue's importance in both substrate binding and electron mediation, as well as Cys-379 and Cys-382. PEGDH also contains a motif of a ubiquinone-binding site, and coenzyme Q10 was utilized as an electron acceptor. Thus, we propose several important amino acid residues involved in the electron transfer pathway from the substrate to ubiquinone.     

Sequence motif-specific assignment of two [2Fe-2S] clusters in rat xanthine oxidoreductase studied by site-directed mutagenesis

The sequence motif-specific assignment of the two distinct [2Fe-2S] clusters in rat xanthine oxidoreductase (XOR) was unequivocally established by site-directed mutagenesis of recombinant enzymes expressed in a baculovirus-insect cell system and electron paramagnetic resonance (EPR) spectroscopy. The conserved cysteine residues, including Cys-115, in the unusual C-terminal -Cys-Xaa(2)-Cys-//-Cys-Xaa(1)-Cys- motif serve as ligands to the Fe/S I center, which is probably located in close proximity to the Mo-pterin center. Other conserved cysteine residues, including Cys-43 and Cys-51, in the N-terminal plant ferredoxin-like motif serve as ligands to the Fe/S II center, which is distantly located from the Mo-pterin center. The present sequence motif-specific assignment of the Fe/S I and II centers is discussed in the light of the structural features of XOR.     

Primary structure of a hemorrhagic metalloproteinase, HT-2, isolated from the venom of Crotalus ruber ruber

Crotalidae and Viperidae snake venoms contains several kinds of metalloproteinases which cause localized hemorrhage by direct action on blood vessel walls. We report here the entire amino acid sequence and the disulfide bridge locations of HT-2, one of the hemorrhagic toxins isolated from the venom of Crotalus ruber ruber (red rattlesnake). The non-reduced protein was first cleaved at methionine residues to provide a set of 8 fragments, which covered the entire sequence of HT-2. The disulfide bridge locations of HT-2 were also determined by using these primary fragments. The unambiguous sequence for the whole protein was then established by conventional methods using lysyl endopeptidase and thermolysin digests. HT-2 consisted of 202 amino acid residues with two disulfide bridges, which were assigned to Cys-117-Cys-197 and Cys-157-Cys-164. HT-2 had a typical zinc-chelating sequence His-Glu-X-X-His (residues 142-146) found in thermolysin, and its overall sequence showed, respectively, 50, 52, and 53% identities to those of HR2a, H2-proteinase, and the metalloproteinase domain of HR1B. However, the disulfide bridge locations of HT-2 were different from those in the other metalloproteinases. The primary structure of HT-2 was more closely related to that of Ht-d from Crotalus atrox recently determined (81% identity). From the structural comparison with five metalloproteinases so far elucidated, six conservative amino acid residues, which may possibly be related to the induction of the hemorrhagic activity, were suggested to be present in these toxins.     

Expression of human glutathione S-transferase 2 in Escherichia coli. Immunological comparison with the basic glutathione S-transferases isoenzymes from human liver.

A plasmid, termed pTacGST2, which contains the complete coding sequence of a GST2 (glutathione S-transferase 2) subunit and permits the expression of the protein in Escherichia coli was constructed. The expressed protein had the same subunit Mr as the enzyme from normal human liver and retained its catalytic function with both GST and glutathione peroxidase activity. Antiserum raised against the bacterially synthesized protein cross-reacted with all the basic GST isoenzymes in human liver. The electrophoretic mobility in agarose of the bacterially expressed isoenzyme suggested that its pI is identical with that of the cationic isoenzyme from human liver previously termed GST2 type 1. The available evidence suggests that the three common cationic isoenzymes found in human liver are the products of two very similar gene loci.     

Assignment of disulfide bridges in the fusion glycoprotein of Sendai virus.

The mature fusion (F) glycoprotein of the paramyxovirus family consists of two disulfide-linked subunits, the N-terminal F2 and the C-terminal F1 subunits, and contains 10 cysteine residues which are highly conserved at specific positions. The high level of conservation strongly suggests that they are indeed disulfide linked and play important roles in the folding and functioning of the molecule. However, it has not even been clarified which cysteine residues link the F2 and F1 subunits. This report describes our assignment of the disulfide bridges in purified Sendai virus F glycoprotein by fragmentation of the polypeptide and isolation of cystine-containing peptides and determination of their N-terminal sequences. The data demonstrate that all of the 10 cysteine residues participate in disulfide bridges and that Cys-70, the only cysteine in F2, and Cys-199, the most upstream cysteine in F1, form the interchain bond. Of the remaining eight cysteine residues clustered near the transmembrane domain of F1, the specific bridges identified are Cys-338 to Cys-347 and Cys-362 to Cys-370. Although no exact pairings between the subsequent four residues were defined, it seems likely that the most downstream, Cys-424, is linked to Cys-394, Cys-399, or Cys-401. Thus, we conclude that the cysteine-rich domain indeed contributes to the formation of a bunched structure containing at least two tandem cystine loops.     

Characterization of human adenylate kinase 3 (AK3) cDNA and mapping of the AK3 pseudogene to an intron of the NF1 gene.

We have isolated cDNA clones for human adenylate kinase isozyme 3 (AK3) with a genomic probe from the neurofibromatosis type 1 (NF1) region. Three overlapping clones isolated from a human frontal-cortex cDNA library gave rise to a consensus sequence of 1.7 kb. The open reading frame identified in this sequence predicted a peptide of 223 residues. A database search revealed striking homology, about 58% amino acid sequence identity, between this predicted protein and bovine AK3. Human AK3 protein also showed significant homology to other members of the adenylate kinase family isolated from various species. Genomic Southern analysis suggested that multiple AK3 loci exist in the human genome, including one located in an intron of NF1 on chromosome 17. The chromosome-17 locus appears to be a processed pseudogene, since it is intronless and contains a polyadenylate tract; it nevertheless retains coding potential because the open reading frame is not impaired by any observed base substitutions.     

Covalent structure, disulfide bonding, and identification of reactive surface and active site residues of human prostatic acid phosphatase

The pairing of the half-cysteine residues of human prostatic acid phosphatase was established by proteolytic digestion and analysis of the resulting peptide mixtures by fast atom bombardment mass spectrometry (FAB-MS). An independently derived, full length cDNA clone was used as the basis for the interpretation of the FAB-MS data. The sequence of the native protein is that predicted from the present cDNA sequence, except for the carboxyl-terminal end and some possible post-translational deamidations. Isolated human prostatic acid phosphatase was found to have multiple carboxyl-terminal ends, terminating in Thr, Glu, and Asp, corresponding to residues 349-351 of the 354-residue protein that is predicted from the cDNA sequence after removal of a leader peptide. The protein contains no free sulfhydryl groups. The identical monomer chains of the dimeric native enzyme are found to contain three disulfide bonds, specifically Cys-129 to Cys-340, Cys-183 to Cys-281, and Cys-315 to Cys-319. In view of the conserved positions of cysteines in the homologous human and rat liver lysosomal acid phosphatases, an identical disulfide bonding pattern may be predicted for those proteins. The location of a potential antigenic site was established by selective labeling of proximate tyrosine residues predicted to be on the surface. A conserved RHGXRXP sequence is present in the prostatic, lysosomal, Escherichia coli, and yeast acid phosphatases and is predicted to be of mechanistic significance. In addition, residue Arg-54 is shown to be an active site residue by reaction of the enzyme with phenylglyoxal. Interestingly, this residue is present in a sequence RXRY (R,H) that is also present in lysosomal phosphatase and in recently described protein tyrosine phosphatases.     

Complete nucleotide sequence of the hexokinase PI gene (HXK1) of Saccharomyces cerevisiae

The nucleotide sequence of the yeast glycolytic hexokinase isoenzyme PI-gene, HXK1, has been determined by sequencing the yeast DNA insert of the previously isolated plasmid HXK1 clone [Entian et al., Mol. Gen. Genet. 198 (1984) 50-54]. The structural gene sequence included 1452 bp coding for 484 amino acid (aa) residues corresponding to the Mr of 153 605 for the HXK1 monomer. Several initiation regions and termination points were located using nuclease S1 mapping. The HXK1 sequence was 76% homologous with that of HXK2, which is responsible for triggering glucose repression in yeasts. Since HXK1 is not involved in this regulatory system, the regulatory function of HXK2 must correspond to one or more of the differences between both isoenzymes. Most changes in the amino acid sequence were statistically distributed; however, four clustered regions with more than five altered aa residues were identified.     

The covalent structure of mitochondrial aspartate aminotransferase from chicken. Identification of segments of the polypeptide chain invariant specifically in the mitochondrial isoenzyme

The primary structure of mitochondrial aspartate aminotransferase from chicken is reported. The enzyme is a dimer of identical subunits. Each subunit contains 401 amino acid residues; the calculated subunit molecular weight of the apoform is 44,866. The degree of sequence identity with the homologous cytosolic isoenzyme from chicken is 46%. A comparison of the primary structures of the mitochondrial and the cytosolic isoenzyme from pig and chicken shows that 40% of all residues are invariant. The degree of interspecies sequence identity both of the mitochondrial and the cytosolic isoenzyme from chicken and pig (86% and 83%, respectively) markedly exceeds that of the intraspecies identity between mitochondrial and cytosolic aspartate aminotransferase in chicken (46%) or in pig (48%). Based on these values, the duplication of the aspartate aminotransferase ancestral gene is estimated to have occurred approximately 1000 million years ago, i.e. at the time of the emergence of eukaryotic cells. By sequence comparison it is possible to identify amino acid residues and segments of the polypeptide chain that have been conserved specifically in the mitochondrial isoenzyme during phylogenetic evolution. These segments comprise about a third of the total polypeptide chain and appear to cluster in a certain surface region. The cluster carries an excess of positively charged residues which exceeds the overall charge difference between the cytosolic (pI approximately 6) and the mitochondrial isoenzyme (pI approximately 9).     

Importance of cysteines in the LDLR-related domain of the subgroup A avian leukosis and sarcoma virus receptor for viral entry.

The extracellular domain of the subgroup A avian sarcoma and leukemia virus (ALSV-A) receptor contains a region that is related in sequence to the ligand-binding motifs of the low-density lipoprotein receptor (LDLR). This domain contains six cysteines that are highly conserved between different members of the LDLR protein superfamily, and these residues are presumed to participate in intrachain disulfide bonds. To assess the importance of each cysteine in the ALSV-A receptor, individual or multiple cysteines were mutated to alanines and the altered receptors were tested for the ability to confer susceptibility to viral infection. Receptors bearing single mutations allowed subgroup A viral entry, albeit at less than wild-type levels. Receptors containing two or three substitutions were completely inactive if one of the changed residues was Cys-35 or Cys-50. Of the altered receptors tested, the only exception to this rule was a functional receptor which lacked both Cys-35 and Cys-50, an activity that was dependent on the presence of other cysteines in this protein. Most interestingly, a receptor containing both Cys-35 and Cys-50 but lacking the other four cysteines was completely functional. These results demonstrate the importance of Cys-35 and Cys-50 for viral entry mediated by the ALSV-A receptor and show that in the presence of these two residues, all of the other cysteines in this protein can be removed without loss of this function.     

The amino acid sequence of GTP:AMP phosphotransferase from beef-heart mitochondria. Extensive homology with cytosolic adenylate kinase

The amino acid sequence of GTP:AMP phosphotransferase (AK3) from beef-heart mitochondria has been determined, except for one segment of about 33 residues in the middle of the polypeptide chain. The established sequence has been unambiguously aligned to the sequence of cytosolic ATP:AMP phosphotransferase (AK1) from pig muscle, allowing for six insertions and deletions. With 30% of all aligned residues being identical, the homology between AK3 and AK1 is well established. As derived from the known three-dimensional structure of AK1, the missing segment is localized at a small surface area of the molecule, far apart from the active center. The pattern of conserved residues demonstrates that earlier views on substrate binding have to be modified. The observation of three different consecutive N-termini indicates enzyme processing.     

Purification, characterization, and the complete amino acid sequence of porcine pancreatic deoxyribonuclease

Porcine pancreatic DNase has been purified to homogeneity. The polypeptide exhibits a single band of Mr = 34,000 on sodium dodecyl sulfate-polyacrylamide gel electrophoresis. The enzyme is a glycoprotein containing glucosamine. The results of end group analyses show leucine at the NH2 terminus and alanine at the COOH terminus. The enzymatic properties of the purified porcine DNase are very similar to those of bovine and ovine DNases. The sequence data on the tryptic and chymotryptic peptides derived from CNBr fragments of porcine DNase, along with the results of automated Edman degradation of the intact polypeptide and of the two largest CNBr fragments, indicate the complete amino acid sequence of porcine DNase to be as follows:L-R- I-A-F-N-I-R-T-F-G-E-T-K-M-S-N-A-T-S-N-Y-I-V-R-I-L-S-R-Y-D-I-A-L-I-Q- E-V-R-D-S-H-L-T-A-V-G-K-L-L-N-E-L-N-Q-D-D-P-N-N-Y-H-H-V-V-S-E-P-L-G-R- S-T-Y-K-E-R-Y-L-F-V-F-R-P-N-Q-V-S-V-L-D-S-Y-L-Y-D-D-G-C-E-P-C-G-N-D-T- F-N-R-E-P-S-V-V-K-F-S-S-P-F-T-Q-V-K-E-F-A-I-V-P-L-H-A-A-P-S-D-A-A-A-E- I-N-S-L-Y-D-V-Y-L-N-V-R-Q-K-W-D-L-Q-D-I-M-L-M-G-D-F-N-A-G-C-S-Y-V-T- T-S-H-W-S-S-I-R-L-R-E-S-P-P-F-Q-W-L-I-P-D-T-A-D-T-T-V-S-S-H-T-C-A-Y- D-R-I-V-V-A-G-P-L-L-Q-R-A-V-V-P-D-S-A-A-P-F-D-F-Q-A-A-F-G-L-S-Q-E-T- A-L-A-I-S-D-H-Y-P-V-E-V-T-L-K-R-A. The polypeptide consists of 262 amino acid residues. One of the two disulfide loops links Cys-101 and Cys-104 and the other Cys-173 and Cys-209. Two carbohydrate side chains are attached at Asn-18 and Asn-106.