The acidic C-terminal domain of protein disulfide isomerase is not critical for the enzyme subunit function or for the chaperone or disulfide isomerase activities of the polypeptide
1999; Springer Nature; Volume: 18; Issue: 1 Linguagem: Inglês
10.1093/emboj/18.1.65
ISSN1460-2075
Autores Tópico(s)Microbial Inactivation Methods
ResumoArticle4 January 1999free access The acidic C-terminal domain of protein disulfide isomerase is not critical for the enzyme subunit function or for the chaperone or disulfide isomerase activities of the polypeptide Peppi Koivunen Peppi Koivunen Collagen Research Unit, Biocenter Oulu and Department of Medical Biochemistry, University of Oulu, Kajaanintie 52A, FIN-90220 Oulu, Finland Search for more papers by this author Annamari Pirneskoski Annamari Pirneskoski Collagen Research Unit, Biocenter Oulu and Department of Medical Biochemistry, University of Oulu, Kajaanintie 52A, FIN-90220 Oulu, Finland Search for more papers by this author Päivi Karvonen Päivi Karvonen Collagen Research Unit, Biocenter Oulu and Department of Medical Biochemistry, University of Oulu, Kajaanintie 52A, FIN-90220 Oulu, Finland Search for more papers by this author Johanna Ljung Johanna Ljung Department of Medical Biochemistry and Biophysics, Karolinska Institutet, S-17177 Stockholm, Sweden Search for more papers by this author Tarja Helaakoski Tarja Helaakoski Collagen Research Unit, Biocenter Oulu and Department of Medical Biochemistry, University of Oulu, Kajaanintie 52A, FIN-90220 Oulu, Finland Search for more papers by this author Holger Notbohm Holger Notbohm Institute for Medical Molecular Biology, Medical University of Lübeck, D-23538 Lübeck, Germany Search for more papers by this author Kari I. Kivirikko Corresponding Author Kari I. Kivirikko Collagen Research Unit, Biocenter Oulu and Department of Medical Biochemistry, University of Oulu, Kajaanintie 52A, FIN-90220 Oulu, Finland Search for more papers by this author Peppi Koivunen Peppi Koivunen Collagen Research Unit, Biocenter Oulu and Department of Medical Biochemistry, University of Oulu, Kajaanintie 52A, FIN-90220 Oulu, Finland Search for more papers by this author Annamari Pirneskoski Annamari Pirneskoski Collagen Research Unit, Biocenter Oulu and Department of Medical Biochemistry, University of Oulu, Kajaanintie 52A, FIN-90220 Oulu, Finland Search for more papers by this author Päivi Karvonen Päivi Karvonen Collagen Research Unit, Biocenter Oulu and Department of Medical Biochemistry, University of Oulu, Kajaanintie 52A, FIN-90220 Oulu, Finland Search for more papers by this author Johanna Ljung Johanna Ljung Department of Medical Biochemistry and Biophysics, Karolinska Institutet, S-17177 Stockholm, Sweden Search for more papers by this author Tarja Helaakoski Tarja Helaakoski Collagen Research Unit, Biocenter Oulu and Department of Medical Biochemistry, University of Oulu, Kajaanintie 52A, FIN-90220 Oulu, Finland Search for more papers by this author Holger Notbohm Holger Notbohm Institute for Medical Molecular Biology, Medical University of Lübeck, D-23538 Lübeck, Germany Search for more papers by this author Kari I. Kivirikko Corresponding Author Kari I. Kivirikko Collagen Research Unit, Biocenter Oulu and Department of Medical Biochemistry, University of Oulu, Kajaanintie 52A, FIN-90220 Oulu, Finland Search for more papers by this author Author Information Peppi Koivunen1, Annamari Pirneskoski1, Päivi Karvonen1, Johanna Ljung2, Tarja Helaakoski1, Holger Notbohm3 and Kari I. Kivirikko 1 1Collagen Research Unit, Biocenter Oulu and Department of Medical Biochemistry, University of Oulu, Kajaanintie 52A, FIN-90220 Oulu, Finland 2Department of Medical Biochemistry and Biophysics, Karolinska Institutet, S-17177 Stockholm, Sweden 3Institute for Medical Molecular Biology, Medical University of Lübeck, D-23538 Lübeck, Germany *[email protected] The EMBO Journal (1999)18:65-74https://doi.org/10.1093/emboj/18.1.65 PDFDownload PDF of article text and main figures. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info Protein disulfide isomerase (PDI) is a multifunctional polypeptide that acts as a subunit in the animal prolyl 4-hydroxylases and the microsomal triglyceride transfer protein, and as a chaperone that binds various peptides and assists their folding. We report here that deletion of PDI sequences corresponding to the entire C-terminal domain c, previously thought to be critical for chaperone activity, had no inhibitory effect on the assembly of recombinant prolyl 4-hydroxylase in insect cells or on the in vitro chaperone activity or disulfide isomerase activity of purified PDI. However, partially overlapping critical regions for all these functions were identified at the C-terminal end of the preceding thioredoxin-like domain a′. Point mutations introduced into this region identified several residues as critical for prolyl 4-hydroxylase assembly. Circular dichroism spectra of three mutants suggested that two of these mutations may have caused only local alterations, whereas one of them may have led to more extensive structural changes. The critical region identified here corresponds to the C-terminal α helix of domain a′, but this is not the only critical region for any of these functions. Introduction Protein disulfide isomerase (PDI, EC 5.3.4.1), a major protein within the lumen of the endoplasmic reticulum (ER), catalyzes disulfide bond formation in the biosynthesis of proteins translocated into this cell compartment (for reviews, see Noiva and Lennarz, 1992; Freedman et al., 1994). It is a modular protein that consists of domains a, b, b′, a′ and c (Edman et al., 1985; Pihlajaniemi et al., 1987; Kemmink et al., 1995). The existence of an additional domain e, thought to be located between domains a and b, is now very unlikely (Kemmink et al., 1995; Darby et al., 1996). The a and a′ domains are similar to thioredoxin (Edman et al., 1985), and both contain the sequence -CGHC-, representing two independently acting catalytic sites for PDI activity (Hawkins and Freedman, 1991; Vuori et al., 1992a, c; LaMantia and Lennarz, 1993; Lyles and Gilbert, 1994; Darby and Creighton, 1995). Nuclear magnetic resonance (NMR) characterization of human domain a produced in Escherichia coli indicates that it also has the thioredoxin fold (Kemmink et al., 1995, 1996). Recent NMR studies on domain b and preliminary data on domain b′ suggest that these likewise may have the thioredoxin fold, even though they have no catalytic site sequences and show no significant amino acid sequence similarity to thioredoxin (Kemmink et al., 1997). The main part of the PDI polypeptide may thus consist of two catalytically active thioredoxin modules and two inactive ones (Kemmink et al., 1997). The C-terminal domain c, which represents a putative Ca2+-binding region (Freedman et al., 1994), is rich in acidic amino acids; 18 out of the 29 extreme C-terminal residues in the human polypeptide being either aspartate or glutamate (Pihlajaniemi et al., 1987). Its C-terminus contains the -KDEL motif, which is both necessary and sufficient for the retention of a polypeptide within the lumen of the ER (Pelham, 1990). The PDI polypeptide has a number of additional functions. It serves as the β subunit of the animal prolyl 4-hydroxylases (Pihlajaniemi et al., 1987; Veijola et al., 1994; Helaakoski et al., 1995) and of the microsomal triglyceride transfer protein (Wetterau et al., 1990), and as a chaperone-like polypeptide that binds various peptides within the lumen of the ER and probably assists in folding of many newly synthesized proteins (LaMantia and Lennarz, 1993; Cai et al., 1994; Otsu et al., 1994; Puig and Gilbert, 1994; Puig et al, 1994; Rupp et al., 1994; Hayano et al., 1995; Yao et al., 1997). PDI may also have other, less well characterized functions (Freedman et al., 1994; Prockop and Kivirikko, 1995; Kivirikko and Myllyharju, 1998; Kivirikko and Pihlajaniemi, 1998). One major function of the PDI polypeptide in the prolyl 4-hydroxylases (Vuori et al., 1992b; John et al., 1993; Veijola et al., 1994; Helaakoski et al., 1995) and the microsomal triglyceride transfer protein (Wetterau et al., 1991; Ricci et al., 1995; Lamberg et al., 1996) is to keep their highly insoluble α subunits in a catalytically active, non-aggregated conformation. This function is similar to that of some molecular chaperones such as Hsp90 in other proteins (Freedman et al., 1994) and may thus be related to the chaperone function of the polypeptide. Photoaffinity labelling studies with a radioactive tripeptide probe have localized a peptide-binding site in rat PDI to a single 26 amino acid tryptic peptide that begins at the end of domain a′ and includes most of domain c (Noiva et al., 1993). Binding of this photoaffinity-labelled peptide to PDI inhibits the chaperone activity of the latter (Puig et al., 1994). The present work sets out to study whether domain c and the C-terminal end of domain a′ are critical for the prolyl 4-hydroxylase subunit function and the chaperone and disulfide isomerase activities of the PDI polypeptide. Our data indicate that deletion of the entire domain c has no inhibitory effect on any of these functions, whereas the extreme C-terminal α helix of domain a′ appears to constitute one critical region for all these functions. Results Effect of deletions in the C-terminal region of PDI on assembly of the human type I prolyl 4-hydroxylase tetramer Prolyl 4-hydroxylase from vertebrates is an α2β2 tetramer (Kivirikko et al., 1989, 1992) in which PDI functions as the β subunit (Pihlajaniemi et al., 1987). The vertebrate α subunit has two isoforms, α(I) and α(II), which contribute to the [α(I)]2β2, type I, and [α(II)]2β2, type II, tetramers (Helaakoski et al., 1995). The photoaffinity-labelled 26 amino acid tryptic peptide of rat PDI (Noiva et al., 1993) corresponds to a 27 amino acid sequence, residues 452–478, in the 491 amino acid human PDI polypeptide (Figure 1). Secondary structure predictions and modelling based on the structures of thioredoxin (see Martin, 1995) and domain a of PDI (Kemmink et al., 1995, 1996) suggest that the first five residues in this sequence correspond to the end of the last α helix of domain a′, while most of the residues correspond to those of domain c (Figure 1). Figure 1.Schematic representation of the C-terminal region of human PDI. The amino acid sequence is shown in one-letter codes and the corresponding numbering is indicated below. The black bars show the various deletions introduced in the polypeptide, while the point mutations are shown above the amino acid sequence. The predicted protein secondary structure elements present are indicated by a white bar (α helix) and a jagged line (β sheet), whereas the unmarked region between them is likely to form a loop. Download figure Download PowerPoint The 27 amino acid sequence was deleted from human PDI (Δ452–478) and the mutant polypeptide was expressed together with the wild-type human α(I) subunit in insect cells. The cells were harvested 72 h after infection, homogenized in a buffer containing Triton X-100 and centrifuged. The Triton X-100-soluble proteins were then studied by non-denaturing PAGE followed by Coomassie staining (Figure 2A) or by Western blotting with a polyclonal antibody to human PDI (Figure 2B), and assayed for prolyl 4-hydroxylase activity by a procedure based on the hydroxylation-coupled decarboxylation of 2-oxo[1-14C]glutarate (Table I). The Triton X-100-soluble proteins from all the experiments described below were also studied by SDS–PAGE under reducing conditions to verify the expression levels and sizes of the various mutant polypeptides. None of the mutations reported in the present study was found to influence the expression level (details not shown). Figure 2.Non-denaturing PAGE analysis of prolyl 4-hydroxylase tetramer formation from the wild-type human prolyl 4-hydroxylase α(I) subunit and the wild-type or various deletion mutants of human PDI. The samples were extracted with a buffer containing 0.1% Triton X-100 and analysed by non-denaturing 8% PAGE followed by Coomassie staining in (A) or Western blotting with a polyclonal antibody to human PDI in (B). The sample in lane 18 was run separately. T indicates the migration of tetramers and PDIs those of the deletion mutant PDI polypeptides. Lanes 1–18 show Triton X-100-soluble proteins from cells co-infected with the human α(I) virus together with viruses PDI-wt (1), PDIΔ452–478 (2), PDIΔ452–465 (3), PDIΔ466–478 (4), PDIΔ452–454 (5), PDIΔ455–457 (6), PDIΔ458–461 (7), PDIΔ462–465 (8), PDIΔ449–451 (9), PDIΔ446–448 (10), PDIΔ446–449 (11), PDIΔ444–449 (12), PDIΔ440–443 (13), PDIΔ436–439 (14), PDIΔ432–435 (15), PDIΔ431–433 (16), PDIΔ479–487 (17) and PDIΔ462–491 (18). Download figure Download PowerPoint Table 1. Prolyl 4-hydroxylase activity of Triton X-100-soluble proteins from insect cells expressing the α(I) subunit of human prolyl 4-hydroxylase together with either the wild-type PDI or its deleted forms Polypeptides expressed Prolyl 4-hydroxylase activity % (d.p.m./10 μg) α(I) + PDI-wt 4510 ± 910 100 α(I) + PDIΔ452–478 <90 <2** α(I) + PDIΔ452–465 210 ± 290 5** α(I) + PDIΔ466–478 4480 ± 750 99 α(I) + PDIΔ462–465 3950 ± 1130 88 α(I) + PDIΔ458–461 5370 ± 800 119 α(I) + PDIΔ455–457 5990 ± 1590 133 α(I) + PDIΔ452–454 <90 <2** α(I) + PDIΔ449–451 <90 <2** α(I) + PDIΔ446–449 <90 <2** α(I) + PDIΔ446–448 <90 <2** α(I) + PDIΔ444–449 <90 <2** α(I) + PDIΔ440–443 <90 <2** α(I) + PDIΔ436–439 <90 <2** α(I) + PDIΔ432–435 2710 ± 730 60* α(I) + PDIΔ431–433 4130 ± 260 92 α(I) + PDIΔ479–487 4420 ± 860 98 α(I) + PDIΔ462–491 5180 ± 650 125 Values are given in d.p.m./10 μg of extractable cell protein (mean ± SD for three experiments) and as percentages of the value obtained with the wild-type PDI (PDI-wt). The amount of prolyl 4-hydroxylase activity generated by association of the human α(I) subunit with the endogenous insect cell PDI, as determined in each experiment by expressing the α(I) subunit alone, was subtracted from all the values. This enzyme activity level was typically ∼3–5% of that obtained by co-expression of the α(I) subunit with the wild-type PDI. Significances of the differences relative to PDI-wt are indicated by asterisks (*p <0.05; **p <0.001). Deletion of the 27 amino acid sequence from PDI was found to prevent prolyl 4-hydroxylase assembly, as no band corresponding to the enzyme tetramer was seen in non-denaturing PAGE (Figure 2A and B, lanes 2), and no prolyl 4-hydroxylase activity was generated in insect cells (Table I). Deletion of the N-terminal half of this sequence, residues 452–465 (Figure 1), likewise prevented tetramer assembly virtually entirely, as no band corresponding to the tetramer was seen in non-denaturing PAGE (Figure 2A and B, lanes 3), and only a very minor amount of prolyl 4-hydroxylase activity was generated in insect cells (Table I). In contrast, deletion of the C-terminal half, residues 466–478, had no inhibitory effect (Figure 2A and B, lanes 4; Table I). Region 452–465 subsequently was studied by means of four further deletions. A three residue deletion, Δ452–454, involving an FLE triplet (Figure 1), totally prevented tetramer assembly (Figure 2A and B, lanes 5), whereas deletion of residues 455–457 (SGG), 458–461 (QDGA) or 462–465 (GDDD) had no inhibitory effect (Figure 2A and B, lanes 6–8; Table I). Thus the only critical region within the 27 residue sequence was its N-terminal FLE triplet. Additional deletion experiments indicated that the critical region when studied by means of small deletions extended in the N-terminal direction from the 27 amino acid sequence, as the deletions 449–451 (FKK), 446–448 (LDG), 446–449 (LDGF), 444–449 (RTLDGF), 440–443 (YNGE) and 436–439 (TVID) also prevented tetramer assembly (Figure 2A and B, lanes 9–14; Table I). In contrast, additional deletions in the N-terminal direction, 432–435 (SADR) and 431–433 (ASA), did not prevent tetramer assembly (Figure 2A and B, lanes 15 and 16, and Table I), except that the enzyme tetramer formed with the Δ432–435 (SADR) deletion mutant had an abnormal mobility (Figure 2B, lane 15) and could not be detected in the Coomassie-stained gel (Figure 2A, lane 15), and the amount of enzyme activity generated in the insect cells was only 60% of that obtained with the wild-type PDI (Table I). The role of domain c was studied further by deleting residues 479–487 (Figure 1) or by creating a translation termination codon after residue 461, thus deleting the 30 extreme C-terminal residues of the polypeptide. Neither of these deletions inhibited assembly of an active prolyl 4-hydroxylase tetramer (Figure 2A and B, lanes 17 and 18; Table I). Effect of deletions in the C-terminal region of PDI on assembly of the hybrid C.elegans α subunit–human PDI prolyl 4-hydroxylase dimer and the human type II enzyme tetramer The α subunit of prolyl 4-hydroxylase from Caenorhabditis elegans forms an active enzyme with both the C.elegans and human PDI polypeptides in insect cells, but the prolyl 4-hydroxylases containing the C.elegans α subunit differ from vertebrate enzymes in being αβ dimers (Veijola et al., 1994, 1996). The role of the C-terminal region of the human PDI polypeptide in this dimer assembly was studied by expressing all the mutant polypeptides described above in insect cells together with the C.elegans α subunit. Dimer assembly was assessed by non-denaturing PAGE of the Triton X-100-soluble proteins followed by Western blotting with a polyclonal antibody to human PDI. The data obtained were identical to those with the human α(I) subunit in that the only critical region among the amino acids studied covered residues 436–454 (Figure 3). Additional experiments demonstrated that the critical region among the residues studied for assembly of the human type II prolyl 4-hydroxylase [α(II)]2β2 tetramer was identical (details not shown). Figure 3.Non-denaturing PAGE analysis of prolyl 4-hydroxylase dimer formation from the wild-type C.elegans prolyl 4-hydroxylase α subunit and the wild-type or mutant human PDI polypeptides. The samples were extracted as in Figure 2 and analysed by 8% PAGE followed by Western blotting with a polyclonal antibody to human PDI. D indicates the migration of dimers and PDIs those of the deletion mutant PDI polypeptides. Lanes 1–17 show the 0.1% Triton X-100-soluble proteins from cells co-infected with the C.elegans α virus together with viruses human PDI-wt (1), PDIΔ452–478 (2), PDIΔ452–465 (3), PDIΔ466–478 (4), PDIΔ452–454 (5), PDIΔ455–457 (6), PDIΔ458–461 (7), PDIΔ462–465 (8), PDIΔ449–451 (9), PDIΔ446–448 (10), PDIΔ446–449 (11), PDIΔ444–449 (12), PDIΔ440–443 (13), PDIΔ436–439 (14), PDIΔ432–435 (15), PDIΔ431–433 (16) and PDIΔ479–487 (17). Download figure Download PowerPoint Effect of point mutations in the C-terminal α helix of domain a′ Thirteen residues in the C-terminal region of domain a′ were converted individually to other amino acids (Figure 1). Nine of them were within the predicted extreme C-terminal α helix of domain a′, while four residues were in the preceding region predicted to form a loop structure. One of these residues, F449, was converted to five different amino acids (Figure 1). The mutant PDI polypeptides were then expressed together with the wild-type human α(I) subunit in insect cells, and tetramer assembly was studied by non-denaturing PAGE analysis followed by Western blotting and prolyl 4-hydroxylase activity assays as above. In the initial experiments, three mutations were found to either markedly reduce (R444A and L453E) or totally prevent (F449R) prolyl 4-hydroxylase assembly (Figure 4, lanes 5, 16 and 9, respectively). These mutations also caused a concomitant decrease in or loss of the amount of prolyl 4-hydroxylase activity generated in the cells to 43% (R444A), 13% (L453E) or 0% (F449R) of that obtained with the wild-type PDI (Table II). Mutations G448R and K451A also significantly decreased the amount of prolyl 4-hydroxylase activity (Table II), whereas mutation of any of the eight other residues gave no significant decrease in tetramer assembly (Figure 4) or in the amount of prolyl 4-hydroxylase activity, D439A giving a statistically significant minor increase in the amount of activity (Table II). Figure 4.Analysis of prolyl 4-hydroxylase tetramer formation from human α(I) and mutant PDI/β subunits expressed in insect cells. The Triton X-100-soluble samples were electrophoresed on non-denaturing 8% PAGE and analysed by Western blotting with a polyclonal antibody to the human PDI polypeptide. The samples were run on two separate gels. T indicates the migration of tetramers and PDI that of the mutant PDI/β subunits. Lanes 1–17 show samples from co-expression of human α(I) together with PDI-wt (1), PDI (V437D) (2), PDI (I438E) (3), PDI (D439A) (4), PDI (R444A) (5), PDI (L446E) (6), PDI (D447A) (7), PDI (G448R) (8), PDI (F449R) (9), PDI (F449Y) (10), PDI (F449E) (11), PDI (F449W) (12), PDI (K450A) (13), PDI (K451A) (14), PDI (F452R) (15), PDI (L453E) (16) and PDI (E454A) (17). Download figure Download PowerPoint Table 2. Prolyl 4-hydroxylase activity of Triton X-100-soluble proteins from insect cells expressing the α(I) subunit of human prolyl 4-hydroxylase together with either the wild-type PDI or its point mutant forms Polypeptides expressed Prolyl 4-hydroxylase activity % (d.p.m./10 μg) α(I) + PDI-wt 3490 ± 560 100 α(I) + PDI(V437D) 3560 ± 200 102 α(I) + PDI(I438E) 2770 ± 1170 79 α(I) + PDI(D439A) 4150 ± 110 119** α(I) + PDI(R444A) 1500 ± 1140 43* α(I) + PDI(L446E) 3750 ± 1760 107 α(I) + PDI(D447A) 3200 ± 860 92 α(I) + PDI(G448R) 2420 ± 300 69** α(I) + PDI(F449R) 0 ± 0 0*** α(I) + PDI(F449E) 320 ± 170 9*** α(I) + PDI(F449Y) 2750 ± 380 79* α(I) + PDI(F449W) 2050 ± 1340 59* α(I) + PDI(F449A) 2200 ± 160 63** α(I) + PDI(K450A) 3930 ± 280 113 α(I) + PDI(K451A) 2600 ± 530 74* α(I) + PDI(F452R) 4030 ± 1440 115 α(I) + PDI(L453E) 470 ± 310 13*** α(I) + PDI(E454A) 2640 ± 1170 76 Values are given in d.p.m./10 μg of extractable cell protein (mean ± SD for at least three experiments) and as percentages of the value obtained with the wild-type PDI (PDI-wt). The enzyme activity generated by expressing the α(I) subunit alone was subtracted from all the values. Significances of the differences relative to PDI-wt are indicated by asterisks (*p <0.05; **p <0.01; ***p <0.001). The role of F449 was studied further by individual mutations to glutamate, tyrosine, tryptophan and alanine. F449E inhibited tetramer assembly essentially just as completely as F449R, in that no tetramer could be detected in the non-denaturing PAGE (Figure 4, lane 11) and the amount of prolyl 4-hydroxylase activity generated in insect cells was only 9% of that obtained with the wild-type PDI (Table II). F449W and F449A also reduced the amount of prolyl 4-hydroxylase activity generated in the insect cells to 59 and 63%, respectively, lower values than those obtained with any of the mutants involving amino acids other than R444, F449 or L453, whereas F449Y reduced the activity only to 79% (Table II). Effect of deletions in the C-terminal region of PDI on the chaperone activity of the polypeptide in vitro To study the effect of C-terminal deletions on the chaperone activity of PDI, selected mutants were expressed in insect cells and the recombinant polypeptides purified by a procedure consisting of gel filtration and anion exchange chromatography. The purified polypeptides were then assayed for chaperone activity by measuring the reactivation of denatured rhodanese, a protein containing no disulfide bonds (Song and Wang, 1995). Measurements were made at three concentrations of the wild-type or mutant PDI polypeptides, the data obtained at 10 μM being shown in Table III. Table 3. Chaperone activity of purified wild-type PDI and its mutants Polypeptide studied Reactivation level of rhodanase % PDI-wt % rhodanase reactivated PDI-wt 35 ± 9 100 PDIΔ462–491 32 ± 9 91 PDIΔ458–461 30 ± 4 86 PDIΔ455–457 13 ± 5 37** PDIΔ452–454 14 ± 4 40** PDIΔ449–451 11 ± 4 31** PDIΔ446–448 28 ± 4 80* PDIΔ440–443 41 ± 5 117 PDIΔ431–433 36 ± 6 103 F449R 43 ± 12 123 Lysozyme 0 0 No polypeptide 2 ± 3 6** Values are given as percentages of the activity obtained with native rhodanese (mean ± SD) and as percentages of the value obtained with wild-type PDI (PDI-wt) for at least three experiments. Significances of the differences relative to PDI-wt are indicated by asterisks (*p <0.05; **p <0.001). All measurements were performed in the presence of 10 μM PDI-wt or its mutant forms, or 10 μM lysozyme. Domain c appeared to have no role in the chaperone activity, as the rate of rhodanese reactivation obtained with PDIΔ462–491 was almost identical to that obtained with the wild-type PDI (as shown for 10 μM concentrations in Table III). The three amino acid deletion 458–461 on the N-terminal side of the 30 amino acid deletion likewise had no effect on the chaperone activity, whereas four deletions, 455–457 (SGG), 452–454 (FLE), 449–451 (FKK) and 446–448 (LDG), had a distinct inhibitory effect (Table III). As described above, three of these deletions also prevented prolyl 4-hydroxylase assembly, whereas the Δ455–457 deletion had no inhibitory effect. The critical region for the chaperone activity appeared to be shorter than for the subunit function, as the deletions 440–443 (YNGE) and 431–433 (ASA) gave no inhibition (Table III). The F449R mutation that totally prevented prolyl 4-hydroxylase assembly had no inhibitory effect on the chaperone activity (Table III). Effect of deletions and point mutations in the C-terminal region on the disulfide isomerase and reductase activities of PDI Nine PDI deletion mutants were expressed in E.coli in a form containing an N-terminal histidine affinity tag and purified by an affinity chromatography based on this tag. The purified polypeptides were then used to study their disulfide isomerase and reductase activities. The former activity was measured by studying oxidative refolding of a fully reduced RNase (Lyles and Gilbert, 1991) and the latter by an insulin disulfide reduction assay (Holmgren, 1979). Domain c had no role in the disulfide isomerase or reductase activities of PDI, as the former activity of the PDIΔ462–491 mutant was identical and the latter very similar to those of the wild-type PDI (Table IV). Deletions 458–461 (QDGA) and 455–457 (SGG), which caused no inhibition of prolyl 4-hydroxylase assembly, reduced the disulfide isomerase activity by ∼20 and 30%, respectively, whereas the insulin reductase activity was not reduced (Table IV). The Δ455–457 mutant was also purified from insect cells and found to have an isomerase activity that was 75% of that of the wild-type PDI (details not shown). Table 4. Disulfide isomerase and reductase activities of purified PDI mutants expressed in E.coli Polypeptide Isomerase activity Reductase activity % PDI-wt % PDI-wt PDI-wt 100 100 PDIΔ462–491 98 ± 22 86 ± 7* PDIΔ458–461 77 ± 7* 125 ± 2** PDIΔ455–457 67 ± 3** 114 ± 6 PDIΔ452–454 17 ± 2*** 49 ± 4** PDIΔ449–451 20 ± 8*** 50 ± 1* PDIΔ446–448 19 ± 6*** 47 ± 0** PDIΔ440–443 22 ± 1*** 44 ± 3*** PDIΔ436–439 24 ± 0*** 43 ± 1** PDIΔ431–433 12 ± 1*** 60 ± 3** PDI (R444A) 83 ± 35 79 ± 1* PDI (G448R) 85 ± 10** 93 ± 1 PDI (F449R) 21 ± 1*** 43 ± 3** PDI (F449Y) 42 ± 17*** 87 ± 6 PDI (F449W) 133 ± 3* 94 ± 2 PDI (L453E) 23 ± 8*** 48 ± 6* Values are given as percentages of the activity obtained with PDI-wt, mean ± SD for at least two experiments. Significances of the differences relative to PDI-wt are indicated by asterisks (*p <0.05; **p <0.01; ***p <0.001). Isomerase activity and reductase activity were measured in the presence of 1 μM PDI-wt or its mutant forms. Deletions in the region 431–454 caused a major decrease in the disulfide isomerase activity of the polypeptide to ∼20% (Table IV). These deletions include Δ431–433 (ASA), which caused no inhibition of prolyl 4-hydroxylase tetramer assembly. All these deletions also lowered the insulin disulfide reductase activity, although to a lesser extent, by ∼50% (Table IV). The Δ452–454 mutant was also purified from insect cells and found to have an isomerase activity equal to 41% of that of the wild-type PDI (details not shown), while the polypeptide purified from E.coli had an activity level of only ∼17% (Table IV). It is possible, therefore, that the deletions have a stronger effect on the properties of the polypeptides expressed in E.coli than in insect cells. The effects of six of the 17 point mutations were also studied as above. Two of these, F449R and L453E, which inhibited prolyl 4-hydroxylase assembly completely or almost completely (Table II), also had major effects on the disulfide isomerase and reductase activities (Table IV). The R444A mutation, which inhibited prolyl 4-hydroxylase assembly by ∼60% (Table II), had only a minor effect on the disulfide isomerase and the reductase activities (Table IV), as did G448R, which inhibited prolyl 4-hydroxylase assembly by ∼30% (Table IV). Mutation F449Y, causing ∼20% reduction in prolyl 4-hydroxylase activity, reduced disulfide isomerase activity by ∼60% but had little if any effect on the reductase activity, whereas F449W had no inhibitory effect on either (Table IV), although it did reduce prolyl 4-hydroxylase assembly by ∼40% (Table II). Circular dichroism spectrum analysis of PDI and its mutant forms To study whether the loss of subunit function, isomerase and/or chaperone activity in the case of some of the PDI mutants was due to an overall change in the secondary structure rather than a local effect, circular dichroism (CD) spectra in the far
Referência(s)