Protein disulfide–isomerase,a folding catalyst and a redox-regulated chaperone

Protein disulfide–isomerase (PDI) was the first protein-folding catalyst to be characterized, half a century ago. It plays critical roles in a variety of physiological events by displaying oxidoreductase and redox-regulated chaperone activities. This review provides a brief history of the identification of PDI as both an enzyme and a molecular chaperone and of the recent advances in studies on the structure and dynamics of PDI, the substrate binding and release, and the cooperation with its partners to catalyze oxidative protein folding and maintain ER redox homeostasis. In this review, we highlight the structural features of PDI, including the high interdomain flexibility, the multiple binding sites, the two synergic active sites, and the redox-dependent conformational changes.     

Cdc37: a protein kinase chaperone?

The activity of most protein kinases is highly regulated, typically via phosphorylation and/or subunit association. However, the folding of protein kinases into an active state or a form capable of activation is now emerging as another important step through which they can be regulated. The 50-kDa protein Cdc37 and the associated heat-shock protein Hsp90 have been found to bind to, and be required for the activity of, diverse protein kinases, including Cdk4, v-Src, Raf and SEVENLESS. Together, Cdc37 and Hsp90 may act as a general chaperone for protein kinases, in particular those involved in signal-transduction pathways and cell-cycle control.     

α-Crystallin: molecular chaperone and protein surfactant

Bovine lens α-crystallin has recently been shown to function as a molecular chaperone by stabilizing proteins against heat denaturation (Horwitz, J. (1992) Proc. Natl. Acad. Sci. USA, 89, 10449–10453). An investigation, using a variety of physico-chemical methods, is presented into the mechanism of stabilization. α-Crystallin exhibits properties of a surfactant. Firstly, a plot of conductivity of α-crystallin versus concentration shows a distinct inflection in its profile, i.e., a critical micelle concentration (cmc), over a concentration range from 0.15 to 0.17 mM. Gel chromatographic and ¹H-NMR spectroscopic studies spanning the cmc indicate no change in the aggregated state of α-crystallin implying that a change in conformation of the aggregate occurs at the cmc. Secondly, spectrophotometric studies of the rate of heat-induced aggregation and precipitation of alcohol dehydrogenase (ADH), β_L- and γ-crystallin in the presence of α-crystallin and a variety of synthetic surfactants show that stabilization against precipitation results from hydrophobic interactions with α-crystallin and monomeric anionic surfactants. Per mole of subunit or monomer, α-crystallin is the most efficient at stabilization. α-Crystallin, however, does not preserve the activity of ADH after heating. After heat inactivation, gel permeation HPLC indicates that ADH and α-crystallin form a high molecular weight aggregate. Similar results are obtained following incubation of β_L- and γ-crystallin with α-crystallin. ¹H-NMR spectroscopy of mixtures of α- and β_L-crystallin, in their native states, reveals that the C-terminus of βB2-crystallin is involved in interaction with α-crystallin. In the case of γ- and α-crystallin mixtures, a specific interaction occurs between α-crystallin and the C-terminal region of γ_B-crystallin, an area which is known from the crystal structure to be relatively hydrophobic and to be involved in intermolecular interactions. The short, flexible C-terminal extensions of α-crystallin are not involved in specific interactions with these proteins. It is concluded that α-crystallin interacts with native proteins in a weak manner. Once a protein has become denatured, however, the soluble complex with α-crystallin cannot be readily dissociated. In the aging lens this finding may have relevance to the formation of high molecular weight crystallin aggregates.     

Is NSF a fusion protein?

N-ethylmaleimide-sensitive fusion protein (NSF) is an ATPase required for vesicular transport throughout the constitutive secretory and endocytic pathways. Recently, NSF has also been implicated in regulated exocytosis in synapses--based on SNAP-mediated binding in vitro to a complex of neurotoxin substrates (termed 'SNAREs'). This work has generated an hypothesis in which the interaction of SNAREs (SNAP receptors) on the vesicle membrane with those on the target membrane forms a docking complex to which SNAPs bind, thus allowing NSF to bind and elicit membrane fusion. However, current evidence supports an earlier, pre-fusion role for NSF. We speculate that this role may be as a molecular chaperone for the membrane docking/fusion machinery.     

Is AKR2A an essential molecular chaperone for a class of membrane-bound proteins in plants?

The Arabidopsis ankyrin-repeat containing protein 2A (AKR2A) was shown to be an essential molecular chaperone for the peroxisomal membrane-bound ascorbate peroxidase 3 (APX3), because the biogenesis of APX3 depends on the function of AKR2A in plant cells. AKR2A binds specifically to a sequence in APX3 that is made up of a transmembrane domain followed by a few positively charged amino acid residues; this sequence is named as AKR2A-binding sequence or ABS. Interestingly, a sequence in the chloroplast outer envelope protein 7 (OEP7) shares similar features to ABS and is able to bind specifically to AKR2A, suggesting a possibility that proteins with a sequence similar to ABS could bind to AKR2A and they are all likely ligand proteins of AKR2A. This hypothesis was supported by analyzing five additional proteins that contain sequences similar to ABS using the yeast two-hybrid technique. A preliminary survey in the Arabidopsis genome indicates that there are at least 500 genes encoding proteins that contain sequences similar to ABS, which raises interesting questions: are these proteins AKR2A''s ligand proteins and does AKR2A play a critical role in the biogenesis of these proteins in plants?Key words: Arabidopsis, membrane protein, molecular chaperone, protein targeting, transmembrane domainThe Arabidopsis ankyrin-repeat containing protein 2A (AKR2A) is an essential molecular chaperone for the peroxisomal membrane-bound ascorbate peroxidase 3 (APX3).¹ Both AKR2A and APX3 were identified as GF14λ-interacting proteins^2,3 when the mode of action of a 14-3-3 protein, GF14λ⁴ was studied. In characterizing the enzymatic property of APX3, there was some initial difficulty in purifying the expressed APX3 from a bacterial expression system. Although APX3 could be expressed in E. coli cells in large quantities, as evidenced by directly boiling the bacterial cells and analyzing the bacterial cells by SDS-PAGE and Western blot analysis (Fig. 1), APX3 enzymatic activity in the supernatant fraction was not detectable after cells were broken by sonication (Fig. 1). The reason that APX3 activity was not detectable in the supernatant fraction was likely caused by the transmembrane domain that occurs at the C-terminal end of APX3; because these hydrophobic domains could interact with one another, forming insoluble aggregates in bacterial cells. When a truncated APX3 was expressed, i.e., APX3 without the transmembrane domain (APX3Δ in Fig. 1), APX3 activity was then detectable in the supernatant fraction of bacterial cellular extracts. If a protein is able to bind to APX3''s transmembrane domain immediately after or during translation of APX3, this protein could prevent APX3 from forming insoluble aggregates among themselves. APX3 activity would then be detectable in the supernatant fraction. Because some 14-3-3-interacting proteins were shown to interact with one another,⁵ the best candidate that could interact with APX3 should be AKR2A (because they are both GF14λ-interacting proteins). This possibility was tested by simultaneously expressing both APX3 and AKR2A in the same bacterial cell; APX3 activity was indeed detectable in the supernatant fraction of bacterial cellular extracts (Fig. 1).Open in a separate windowFigure 1Protein-protein interaction between AKR2A and APX3 in bacterial cells. (A) Analysis of APX3 activity in supernatant fractions of various bacterial cells. In lanes, APX3, supernatant from cells that express full-length APX3; APX3 + OMT 1, supernatant from cells that express both full-length APX3 and OMT 1 (O-methyltransferase1,⁷); APX3 + AKR2A, supernatant from cells that express both full-length APX3 and AKR2A; APX3Δ, supernatant from cells that express a partial APX3 (i.e., lacking the transmembrane domain and the last seven amino acid residues); APX3Δ + OMT 1, supernatant from cells that express both APX3Δ and OMT 1; APX3Δ + AKR2A, supernatant from cells that express both APX3Δ and AKR2A; OMT 1, supernatant from cells that express OMT1; AKR2A, supernatant from cells that express AKR2A. The white bands in the gel represent APX3 activities as assayed by using the method of Mittler and Zilinskas.⁸ (B) Bacterial cells expressing various target proteins were analyzed directly by using SDS-PAGE method and the positions of the expressed target proteins are marked on the right. (C) Bacterial cells expressing various target proteins were analyzed by western blot. The antibodies used are listed on the right.This was the first evidence that AKR2A interacts with APX3 and the interaction site involves the C-terminal transmembrane domain of APX3. To further define the amino acid residues involved in the AKR2A-APX3 interaction, yeast two-hybrid experiments were conducted with various deletion fragments of AKR2A and APX3.1 It was found that in addition to the transmembrane domain, the positively charged amino acid residues following the transmembrane domain also play a role in the AKR2A-APX3 interaction.¹ This sequence in APX3 was designated as AKR2A-binding sequence (ABS). In order to understand the biological function of the AKR2A-APX3 interaction, several akr2a mutants that displayed reduced or altered interaction with APX3 were created and analyzed. Results indicated that reduced AKR2A activity leads to severe developmental, phenotypic, and physiological abnormalities including reduced steady-state level of APX3 and reduced targeting of APX3 to peroxisomal membranes in Arabidopsis.¹ The pleiotropic nature of akr2a mutants indicated that AKR2A plays more roles in addition to chaperoning APX3. Indeed this work was corroborated by a finding that AKR2A is also required for the biogenesis of the chloroplast outer envelope protein 7 (OEP7).⁶ More importantly, the interaction between AKR2A and OEP7 also involves a sequence in OEP7 that is similar to the ABS found in APX3.There is no apparent similarity, at the amino acid level, between the sequences of the AKR2A-binding site found in APX3 and OEP7; it appears that what AKR2A recognizes in its ligand proteins is the structural feature: single transmembrane domain followed by one or a few positively charged amino acid residues. Therefore, these AKR2A-binding sequences should all be designated as ABS, and it was predicted that any protein with an ABS could be AKR2A''s interacting protein. Five such proteins, APX5, TOC34, TOM20, cytochrome b5 (CB5) and cytochrome b₅ reductase (CB5R) were tested, and indeed all five proteins interacted with AKR2A in the yeast two-hybrid system.¹ More importantly, the interaction sites of these proteins are their ABS in every case tested.¹ Based on these discoveries, it is proposed that AKR2A is a molecular chaperone for this group of ABS-containing proteins.Among the seven AKR2A-interacting proteins that were characterized, the ABS is found at C-terminal end of four proteins (APX3, APX5, CB5 and TOM20), near N-terminal end of two proteins (OEP7 and CB5R), and near C-terminal end of one protein (TOC34), suggesting that the position of ABS in these membrane proteins does not affect its interaction with AKR2A. Furthermore, in all cases, AKR2A binds to its ligand proteins that contain only one ABS. AKR2A does not appear to bind to proteins that contain multiple transmembrane domains such as PMP22,^1,6 even though these transmembrane domains are followed by a few positively charged amino acid residues.APX3 and APX5 are peroxisomal membrane-bound, OEP7 and TOC34 are chloroplast outer envelope proteins, TOM20 is a mitochondrion outer membrane protein and CB5 and CB5R are microsomal membrane (ER-membrane) proteins. Therefore, AKR2A is clearly not responsible for targeting these proteins to their specific membranes; instead AKR2A serves as a molecular chaperone to prevent these proteins from forming aggregates through their hydrophobic domain in ABS after translation (Fig. 2). Perhaps, AKR2A''s binding to the ABS of these membrane proteins also keeps these proteins in insertion competent state before they are sent to their specific destinations. It is clear that other factors, such as organellar membrane-specific receptors, must be required for sending these proteins to their specific membranes (Fig. 2).Open in a separate windowFigure 2Model on how AKR2A chaperones its ligand proteins. (1) AKR2A binds to ABS of a nascent protein that is being synthesized from a free ribosome. (2) AKR2A keeps its ligand protein (L) in the cytoplasm. (3) With the help of membrane-specific receptors, AKR2A''s ligand proteins are sent to their specific membranes.The Arabidopsis proteome was analyzed and it was found that there are at least 500 proteins that contain sequences similar to ABS (http://bio.scu.edu.cn/list.xls). Would these proteins be AKR2A''s ligand proteins? Some of them, if not all, might be, but it will be a challenging task to experimentally test these proteins one by one. A better bioinformatics tool that can provide clues on the mode of action of the protein-protein interactions between AKR2A and its known ligand proteins should help us designing next set of experiments in order to answer the above question in an efficient way.     

Enhancing functional expression of β-glucosidase in Pichia pastoris by co-expressing protein disulfide isomerase

The expression of heterologous proteins may exert severe stress on the host cells at different levels. Protein folding and disulfide bond formation were identified as rate-limited steps in recombinant protein secretion in yeast cells. For the production of β-glucosidase in Pichia pastoris, final β-glucosidase activity reached 1,749 U/mL after fermentation optimization in a 3 L bioreactor, while the specific activity decreased from 620 to 467 U/mg, indicating a potential protein misfolding. To solve this problem, protein disulfide isomerase, a chaperone protein which may effectively regulate disulfide bond formation and protein folding, was co-expressed with β-glucosidase. In the co-expression system, a β-glucosidase production level of 2,553 U/mL was achieved and the specific activity of the enzyme reached 721 U/mg, which is 1.54 fold that of the control.     

Targeting HSP70: The second potentially druggable heat shock protein and molecular chaperone?

The HSF1-mediated stress response pathway is steadily gaining momentum as a critical source of targets for cancer therapy. Key mediators of this pathway include molecular chaperones such as heat shock protein (HSP) 90. There has been considerable progress in targeting HSP90 and the preclinical efficacy and signs of early clinical activity of HSP90 inhibitors have provided proof-of-concept for targeting this group of proteins. The HSP70 family of molecular chaperones are also key mediators of the HSF-1-stress response pathway and have multiple additional roles in protein folding, trafficking and degradation, as well as regulating apoptosis. Genetic and biochemical studies have supported the discovery of HSP70 inhibitors which have the potential for use as single agents or in combination to enhance the effects of classical chemotherapeutic or molecularly targeted agents including HSP90 inhibitors. Here we provide a perspective on the progress made so far in designing agents which target the HSP70 family.     

Action of protein disulfide isomerase on proinsulin exit from endoplasmic reticulum of pancreatic β-cells

For insulin synthesis, the proinsulin precursor is translated at the endoplasmic reticulum (ER), folds to include its three native disulfide bonds, and is exported to secretory granules for processing and secretion. Protein disulfide isomerase (PDI) has long been assumed to assist proinsulin in this process. Herein we have examined the effect of PDI knockdown (PDI-KD) in β-cells. The data establish that upon PDI-KD, oxidation of proinsulin to form native disulfide bonds is unimpaired and in fact enhanced. This is accompanied by improved proinsulin exit from the ER and increased total insulin secretion, with no evidence of ER stress. We provide evidence for direct physical interaction between PDI and proinsulin in the ER of pancreatic β-cells, in a manner requiring the catalytic activity of PDI. In β-cells after PDI-KD, enhanced export is selective for proinsulin over other secretory proteins, but the same effect is observed for recombinant proinsulin trafficking upon PDI-KD in heterologous cells. We hypothesize that PDI exhibits unfoldase activity for proinsulin, increasing retention of proinsulin within the ER of pancreatic β-cells.     

Is prostate-specific membrane antigen a multifunctional protein?

Prostate-specific membrane antigen (PSMA) is a metallopeptidase expressed predominantly in prostate cancer (PCa) cells. PSMA is considered a biomarker for PCa and is under intense investigation for use as an imaging and therapeutic target. Although the clinical utility of PSMA in the detection and treatment of PCa is evident and is being pursued, very little is known about its basic biological function in PCa cells. The purpose of this review is to highlight the possibility that PSMA might be a multifunctional protein. We suggest that PSMA may function as a receptor internalizing a putative ligand, an enzyme playing a role in nutrient uptake, and a peptidase involved in signal transduction in prostate epithelial cells. Insights into the possible functions of PSMA should improve the diagnostic and therapeutic values of this clinically important molecule. prostate cancer; receptor; peptidase; endocytosis     

Is apolipoprotein D a mammalian bilin-binding protein?

Human apolipoprotein D (APO-D) is a serum glycoprotein that has no sequence similarity with other apolipoproteins but rather belongs to the alpha 2-microglobulin superfamily whose other members transport small hydrophobic ligands in a wide variety of biological contexts. To investigate the ligand specificity of APO-D, we analyzed its relationship with the other members of this superfamily and constructed a detailed molecular model using the atomic coordinates of its most closely related homolog--insecticyanin from the tobacco hornworm, Manduca sexta. We studied the geometry of the binding pocket of APO-D and the topology of characteristic patches of both hydrophobic and polar side chains that also occur in crystal structures of insecticyanin and bilin-binding protein from the butterfly Pieris brassicae. From the data obtained we hypothesize that heme-related compounds may be more favorable ligands for APO-D than either cholesterol or cholesteryl ester. Preliminary experiments showed that purified human APO-D binds bilirubin in an approximately one-to-one molar ratio. These results suggest a new biological role for APO-D that is more congruent with its tissue distribution and evolutionary history.     

The collagen-specific molecular chaperone HSP47: is there a role in fibrosis?

Heat shock protein 47 (HSP47) is a collagen-specific molecular chaperone that is required for molecular maturation of various types of collagens. Recent studies have shown a close association between increased expression of HSP47 and excessive accumulation of collagens in scar tissues of various human and experimental fibrotic diseases. It is presumed that the increased levels of HSP47 in fibrotic diseases assist in excessive assembly and intracellular processing of procollagen molecules and, thereby, contribute to the formation of fibrotic lesions. Studies have also shown that suppression of HSP47 expression can reduce accumulation of collagens to delay the progression of fibrotic diseases in experimental animal models. Because HSP47 is a specific chaperone for collagen synthesis, it provides a selective target to manipulate collagen production, a phenomenon that might have enormous clinical impact in controlling a wide range of fibrotic diseases. Here, we outline the fibrogenic role of HSP47 and discuss the potential usefulness of HSP47 as an anti-fibrotic therapeutic target.     

Is there a unifying mechanism for protein folding?

Proteins appear to fold by diverse pathways, but variations of a simple mechanism - nucleation-condensation - describe the overall features of folding of most domains. In general, secondary structure is inherently unstable and its stability is enhanced by tertiary interactions. Consequently, an extensive interplay of secondary and tertiary interactions determines the transition-state for folding, which is structurally similar to the native state, being formed in a general collapse (condensation) around a diffuse nucleus. As the propensity for stable secondary structure increases, folding becomes more hierarchical and eventually follows a framework mechanism where the transition state is assembled from pre-formed secondary structural elements.     

Is tensegrity a unifying concept of protein folds?

We suggest that the three-dimensional architecture of globular proteins can be described in terms of tensegrets, i.e. structural elements that are held together through attractive and repulsive forces. Hard elements of tensegrets are represented by secondary structure elements, i.e. alpha-helices and beta-strands, while the role of elastic elements is played by attractive and repulsive atomic forces. Characteristics of tensegrets is that they can auto-assemble and that they respond to changes of tension in some part of the entire object through a deformation in another part, thus partially preserving their structure, despite their deformation. This latter property well explains both the folding process and the behavior of globular proteins under mild denaturing conditions, as revealed by the molten globule state.     

Escaping the endoplasmic reticulum: why does a molecular chaperone leave home for greener pastures?

Molecular chaperones reside in nearly every organelle within a eukaryotic cell, and in each of these compartments, they ensure that protein homeostasis (or proteostasis) is maintained. In this issue, Wiseman and colleagues find that an ER lumenal chaperone escapes this compartment when a specific stress pathway is activated. The chaperone, an Hsp40 homolog known as ERdj3, transits through the secretory pathway to the extracellular space. During this journey, ERdj3 can escort an aggregation‐prone protein or it can identify aggregation‐prone proteins extracellularly, thereby functioning outside of its normal environment.