Enhanced stability of heterologous proteins by supramolecular self-assembly

Recently, we reported on the dual function of human ferritin heavy chain (hFTN-H) used for the fusion expression and solubility enhancement of various heterologous proteins: (1) high-affinity interaction with HSP70 chaperone DnaK and (2) formation of self-assembled supramolecules with limited and constant sizes. Especially the latter, the self-assembly function of hFTN-H is highly useful in avoiding the undesirable formation of insoluble macroaggregates of heterologous proteins in bacterial cytoplasm. In this study, using enhanced green fluorescent protein (eGFP) and several deletion mutants of Mycoplasma arginine deiminase (ADI_132–410) as reporter proteins, we confirmed through TEM image analysis that the recombinant fusion proteins (hFTN-H::eGFP and hFTN-H::ADI_132–410) formed intracellular spherical particles with nanoscale diameter (≈10 nm), i.e., noncovalently cross-linked supramolecules. Surprisingly, the supramolecular eGFP and ADI showed much enhanced stability in bioactivity. That is, the activity level was much more stably maintained for the prolonged period of time even at high temperature, at high concentration of Gdn–HCl, and in wide range of pH. The stability enhancement by supramolecular self-assembly may make it possible to utilize the protein supramolecules as novel means for drug delivery, enzymatic material conversion (biotransformation), protein chip/sensor, etc. where the maintenance of protein/enzyme stability is strictly required. Jin-Seung Park and Ji-Young Ahn contributed equally to this work.     

Glutamate decarboxylase-derived IDDM autoantigens displayed on self-assembled protein nanoparticles

The recombinant ferritin heavy chain (FTN-H) formed self-assembled spherical nanoparticles with the size comparable to native one. We tried to express the GAD65 COOH-terminal fragments, i.e., 448-585 (GAD65(448-585)), 487-585 (GAD65(487-585)), and 512-585 (GAD65(512-585)) amino acid fragments, using FTN-H as N-terminus fusion expression partner in Escherichia coli. All of recombinant fusion proteins (FTN-H::GAD65(448-585), FTN-H::GAD65(487-585), and FTN-H::GAD65(512-585)) also formed spherical nanoparticles due probably to the self-assembly function of the fused ferritin heavy chain. The antigenic epitopes within GAD65(448-585), GAD65(487-585), and GAD65(512-585) against insulin-dependent diabetes mellitus (IDDM) marker (autoantibodies against GAD65) were localized at the surface of the spherical protein nanoparticles so that anti-GAD65 Ab could recognize them. Protein nanoparticles like FTN-H seem to provide distinct advantages over other inorganic nanoparticles (e.g., Au, Ag, CdSe, etc.) in that through the bacterial synthesis, the active capture probes can be located at the nanoparticle surface with constant orientation/conformation via covalent cross-linking without complex chemistry. Also it is possible for the protein nanoparticles to have uniform particle size, which is rarely achieved in the chemical synthesis of inorganic nanoparticles. Thus, the recombinant ferritin particles can be used as a three-dimensional (spherical) and nanometer-scale probe structure that is a key component in ultra-sensitive protein chip for detecting protein-small molecule interactions and protein-protein interactions.     

cis-Effect of DnaJ on DnaK in ternary complexes with chimeric DnaK/DnaJ-binding peptides

Chimeric peptides, comprising a DnaK-binding sequence of L-amino acid residues (motif k) and an exclusive DnaJ-binding sequence of D-amino acid residues (motif j) connected through a 22-residue linker, were examined as minisubstrates for the DnaK chaperone system. The DnaJ-stimulated ATPase activity of DnaK was three times higher in the presence of the chimeric peptides pjk or pkj than in the simultaneous presence of the corresponding single-motif peptides ala-p5 (k motif) plus D-p5 (j motif). Apparently, pjk and pkj mimic unfolded proteins by forming ternary (ATP x DnaK) x peptide x DnaJ complexes which favor cis-interaction of DnaJ with DnaK. Consistent with this interpretation, the specific stimulatory effect of the chimeric peptides was abolished by either single-motif peptide in excess.     

Chaperone-fusion expression plasmid vectors for improved solubility of recombinant proteins in Escherichia coli

The enteric bacterium Escherichia coli is the most extensively used prokaryotic organism for production of proteins of therapeutic or commercial interest. However, it is common that heterologous over-expressed recombinant proteins fail to properly fold resulting in formation of insoluble aggregates known as inclusion bodies. Complex systems have been developed that employ simultaneous over-expression of chaperone proteins to aid proper folding and solubility during bacterial expression. Here we describe a simple method whereby a protein of interest, when fused in frame to the E. coli chaperones DnaK or GroEL, is readily expressed in large amounts in a soluble form. This system was tested using expression of the mouse prion protein PrP, which is normally insoluble in bacteria. We show that while in trans over-expression of the chaperone DnaK failed to alter partitioning of PrP from the insoluble inclusion body fraction to the soluble cytosol, expression of a DnaK–PrP fusion protein yielded large amounts of soluble protein. Similar results were achieved with a fragment of insoluble Varicella Zoster virus protein ORF21p. In theory this approach could be applied to any protein that partitions with inclusion bodies to render it soluble for production in E. coli.     

An essential role for the DnaK molecular chaperone in stabilizing over-expressed substrate proteins of the bacterial twin-arginine translocation pathway

All secreted proteins in Escherichia coli must be maintained in an export-competent state before translocation across the inner membrane. In the case of the Sec pathway, this function is carried out by the dedicated SecB chaperone and the general chaperones DnaK-DnaJ-GrpE and GroEL-GroES, whose job collectively is to render substrate proteins partially or entirely unfolded before engagement of the translocon. To determine whether these or other general molecular chaperones are similarly involved in the translocation of folded proteins through the twin-arginine translocation (Tat) system, we screened a collection of E. coli mutant strains for their ability to transport a green fluorescent protein (GFP) reporter through the Tat pathway. We found that the molecular chaperone DnaK was essential for cytoplasmic stability of GFP bearing an N-terminal Tat signal peptide, as well as for numerous other recombinantly expressed endogenous and heterologous Tat substrates. Interestingly, the stability conferred by DnaK did not require a fully functional Tat signal as substrates bearing translocation defective twin lysine substitutions in the consensus Tat motif were equally unstable in the absence of DnaK. These findings were corroborated by crosslinking experiments that revealed an in vivo association between DnaK and a truncated version of the Tat substrate trimethylamine N-oxide reductase (TorA502) bearing an RR or a KK signal peptide. Since TorA502 lacks nine molybdo-cofactor ligands essential for cofactor attachment, the involvement of DnaK is apparently independent of cofactor acquisition. Finally, we show that the stabilizing effects of DnaK can be exploited to increase the expression and translocation of Tat substrates under conditions where the substrate production level exceeds the capacity of the Tat translocase. This latter observation is expected to have important consequences for the use of the Tat system in biotechnology applications where high levels of periplasmic expression are desirable.     

Enhanced Fusion Pore Expansion Mediated by the Trans-Acting Endodomain of the Reovirus FAST Proteins

The reovirus fusion-associated small transmembrane (FAST) proteins are virus-encoded membrane fusion proteins that function as dedicated cell–cell fusogens. The topology of these small, single-pass membrane proteins orients the majority of the protein on the distal side of the membrane (i.e., inside the cell). We now show that ectopic expression of the endodomains of the p10, p14, and p15 FAST proteins enhances syncytiogenesis induced by the full-length FAST proteins, both homotypically and heterotypically. Results further indicate that the 68-residue cytoplasmic endodomain of the p14 FAST protein (1) is endogenously generated from full-length p14 protein expressed in virus-infected or transfected cells; (2) enhances syncytiogenesis subsequent to stable pore formation; (3) increases the syncytiogenic activity of heterologous fusion proteins, including the differentiation-dependent fusion of murine myoblasts; (4) exerts its enhancing activity from the cytosol, independent of direct interactions with either the fusogen or the membranes being fused; and (5) contains several regions with protein–protein interaction motifs that influence enhancing activity. We propose that the unique evolution of the FAST proteins as virus-encoded cellular fusogens has allowed them to generate a trans-acting, soluble endodomain peptide to harness a cellular pathway or process involved in the poorly understood process that facilitates the transition from microfusion pores to macrofusion and syncytiogenesis.     

Removal of DnaK contamination during fusion protein purifications

The use of fusion proteins for recombinant protein expression in Escherichia coli has become popular because the carrier increases protein solubility, standardizes expression levels, and facilitates purification of the fusion products. However, we have observed that the peptide regions that fuse the carrier to the protein of interest bind E. coli Hsp70 molecular chaperones (DnaK) depending on their amino acid composition, resulting in an unwanted contamination during protein purification. Here we describe an approach that helps to circumvent this unwanted contamination. First, the appropriate amino acids surrounding and comprising the cloning site are chosen by using a software based on an algorithm already developed to decrease to a minimum the propensity of the fusion protein to bind DnaK. Second, DnaK contamination is significantly reduced by washing the fusion protein bound to the purification resin with MgATP plus soluble denatured E. coli proteins before elution. The approach can also be applied to eliminate other molecular chaperones.     

Interaction of the targeting sequence of chloroplast precursors with Hsp70 molecular chaperones.

We have analyzed the interaction of DnaK and plant Hsp70 proteins with the wild-type ferredoxin-NADP+ reductase precursor (preFNR) and mutants containing amino-acid replacements in the targeting sequence. Using an algorithm already developed [Rüdiger, S., Germeroth, L., Schneider-Mergener, J. & Bukau, B. (1997) EMBO J. 16, 1501-1507] we observed that 75% of the 727 plastid precursor proteins analyzed contained at least one site with high likelihood of DnaK binding in their transit peptides. Statistical analysis showed a decrease of DnaK binding site frequency within the first 15 amino-acid residues of the transit peptides. Using fusion proteins we detected the interaction of DnaK with the transit peptide of the folded preFNR but not with the mature region of the protein. Discharge of DnaK from the presequence was favored by addition of MgATP. When a putative DnaK binding site was artificially added at the N-terminus of the mature protein, we observed formation of complexes with bacterial and plant Hsp70 molecular chaperones. Reducing the likelihood of DnaK binding by directed mutagenesis of the presequence increased the release of bound DnaK. The Hsp70 proteins from plastids and plant cell cytosol also interacted with the preFNR transit peptide. Overall results are discussed in the context of the proposed models to explain the organelle protein import.     

Substrate specificity of the DnaK chaperone determined by screening cellulose-bound peptide libraries.

Hsp70 chaperones assist protein folding by ATP-dependent association with linear peptide segments of a large variety of folding intermediates. The molecular basis for this ability to differentiate between native and non-native conformers was investigated for the DnaK homolog of Escherichia coli. We identified binding sites and the recognition motif in substrates by screening 4360 cellulose-bound peptides scanning the sequences of 37 biologically relevant proteins. DnaK binding sites in protein sequences occurred statistically every 36 residues. In the folded proteins these sites are mostly buried and in the majority found in beta-sheet elements. The binding motif consists of a hydrophobic core of four to five residues enriched particularly in Leu, but also in Ile, Val, Phe and Tyr, and two flanking regions enriched in basic residues. Acidic residues are excluded from the core and disfavored in flanking regions. The energetic contribution of all 20 amino acids for DnaK binding was determined. On the basis of these data an algorithm was established that predicts DnaK binding sites in protein sequences with high accuracy.     

Formation in vitro of complexes between an abnormal fusion protein and the heat shock proteins from Escherichia coli and yeast mitochondria.

Heat shock proteins (HSPs) of the Hsp70 and GroEL families associate with a variety of cell proteins in vivo. However, the formation of such complexes has not been systematically studied. A 31-kDa fusion protein (CRAG), which contains 12 residues of cro repressor, truncated protein A, and 14 residues of beta-galactosidase, when expressed in Escherichia coli, was found in complexes with DnaK, GrpE, protease La, and GroEL. When an E. coli extract not containing CRAG was applied to an affinity column containing CRAG, DnaK, GroEL, and GrpE were selectively bound. These HSPs did not bind to a normal protein A column. DnaK, GrpE, and the fraction of GroEL could be eluted from the CRAG column with ATP but not with a nonhydrolyzable ATP analog. The ATP-dependent release of DnaK and GroEL also required Mg2+, but GrpE dissociated with ATP alone. The binding and release of DnaK and GroEL were independent events, but the binding of GrpE required DnaK. Inactivation of DnaJ, GrpE, and GroES did not affect the association or dissociation of DnaK or GroEL from CRAG. The DnaK and GrpE proteins could be eluted with 10(-6) M ATP, but 10(-4) M was required for GroEL release. This approach allows a one-step purification of these proteins from E. coli and also the isolation of the DnaK and GroEL homologs from yeast mitochondria. Competition experiments with oligopeptide fragments of CRAG showed that DnaK and GroEL interact with different sites on CRAG and that the cro-derived domain of CRAG contains the DnaK-binding site.     

Heterologous protein expression using a novel stress-responsive protein of E. coli RpoA as fusion expression partner

E. coli proteome response to the stressor 2-HEDS was analyzed through two-dimensional gel electrophoresis (2-DE), and we identified DNA-directed RNA polymerase -subunit (RpoA) as stress-responsive protein. Even under stress situation where the total number of soluble proteins decreased by 9.8%, the synthesis level of RpoA was increased 1.5-fold. As a fusion expression partner as well as solubility enhancer, RpoA facilitated the folding and increased significantly the solubility of many aggregation-prone heterologous proteins (human minipro-insulin, human epidermal growth factor, human prepro-ghrelin, human interleukin-2, human activation induced cytidine deaminase, human glutamate decarboxylase, Pseudomonas putida cutinase, human ferritin light chain, human granulocyte colony-stimulating factor, and cold inflammatory syndrome1 protein Nacht domain) in E. coli cytoplasm. Due probably to intrinsic high folding efficiency and/or chaperone-like activity, RpoA was very effective in shielding interactive surfaces of heterologous proteins that are associated with non-specific protein–protein interaction leading to the formation of inclusion bodies. RpoA was also well suited for the production of biologically active fusion mutant of Pseudomonas putida cutinase that is of much biotechnological and commercial interest.     

Heterologous expression and purification of SpaB involved in subtilin biosynthesis

Lantibiotic peptides contain thioether bridges termed lanthionines that are putatively generated by dehydration of Ser and Thr residues followed by Michael addition of cysteine residues within the peptide. The LanB and LanC proteins have been proposed to catalyze the dehydration and formation of the thioether rings, respectively. We report here the first heterologous overexpression in Escherichia coli of SpaB, the putative dehydratase for subtilin. Sequence analysis of spaB revealed several nucleotide differences with current gene database entries. The solubility of SpaB was increased dramatically when co-expressed with GroEL/ES, and soluble His(6)-tagged SpaB was purified. The protein is at least a dimer, and interaction between SpaB and SpaC was observed. SpaS the putative substrate for SpaB was overexpressed in E. coli as an intein fusion protein, and after cleavage, the peptide was obtained in good yield.     

A novel fusion partner for enhanced secretion of recombinant proteins in <Emphasis Type="Italic">Saccharomyces cerevisiae</Emphasis>

Expressing proteins with fusion partners improves yield and simplifies the purification process. We developed a novel fusion partner to improve the secretion of heterologous proteins that are otherwise poorly excreted in yeast. The VOA1 (YGR106C) gene of Saccharomyces cerevisiae encodes a subunit of vacuolar ATPase. We found that C-terminally truncated Voa1p was highly secreted into the culture medium, even when fused with rarely secreted heterologous proteins such as human interleukin-2 (hIL-2). Deletion mapping of C-terminally truncated Voa1p, identified a hydrophilic 28-amino acid peptide (HL peptide) that was responsible for the enhanced secretion of target protein. A purification tag and a protease cleavage site were added to use HL peptide as a multi-purpose fusion partner. The utility of this system was tested via the expression and purification of various heterologous proteins. In many cases, the yield of target proteins fused with the peptide was significantly increased, and fusion proteins could be directly purified with affinity chromatography. The fusion partner was removed by in vitro processing, and intact proteins were purified by re-application of samples to affinity chromatography.     

叶绿体分裂蛋白PDV1胞质侧结构域可溶性表达条件的研究及纯化

目的:探索叶绿体分裂蛋白PLASTID DIVISION1(PDV1)胞质侧结构域的高效可溶性表达条件,并得到高纯度目的蛋白。方法:通过改变表达载体种类、基因片段大小、诱导剂浓度、诱导温度的方法,以及运用分子伴侣的协助,实现目的蛋白高效可溶性表达。通过镍柱亲和层析和分子筛层析纯化目的蛋白。结果:(1)带His标签的目的蛋白大部分以包涵体形式存在于沉淀中;(2)截掉疏水区域并与增溶标签GST或NusA融合表达,再通过改变诱导表达条件,可以实现PDV1胞质侧结构域的可溶性表达;(3)比较目的蛋白可溶性表达量,选择高效可溶性表达体系,并在该条件下纯化得到高纯度目的蛋白。结论:PDV1胞质侧结构域的高效可溶性表达及纯化,为进一步研究该蛋白的结构及其在叶绿体分裂过程中的作用奠定了一定基础。     

Refolding of substrates bound to small Hsps relies on a disaggregation reaction mediated most efficiently by ClpB/DnaK

Small heat shock proteins (sHsps) are ubiquitous molecular chaperones that bind denatured proteins in vitro, thereby facilitating their subsequent refolding by ATP-dependent chaperones. The mechanistic basis of this refolding process is poorly defined. We demonstrate that substrates complexed to sHsps from various sources are not released spontaneously. Dissociation and refolding of sHsp bound substrates relies on a disaggregation reaction mediated by the DnaK system, or, more efficiently, by ClpB/DnaK. While the DnaK system alone works for small, soluble sHsp/substrate complexes, ClpB/DnaK-mediated protein refolding is fastest for large, insoluble protein aggregates with incorporated sHsps. Such conditions reflect the situation in vivo, where sHsps are usually associated with insoluble proteins during heat stress. We therefore propose that sHsp function in cellular protein quality control is to promote rapid resolubilization of aggregated proteins, formed upon severe heat stress, by DnaK or ClpB/DnaK.     

Growth-associated synthesis of recombinant human glucagon and human growth hormone in high-cell-density cultures of Escherichia coli

Synthesis of two recombinant proteins (human glucagon and human growth hormone) was investigated in fed-batch cultures at high cell concentrations of recombinant Escherichia coli. The glucose-limited growth was achieved without accumulation of metabolic by-products and hence the cellular environment is presumed invariable during growth and recombinant protein synthesis. Via exponential feeding in the two-phase fed-batch operation, the specific cell growth rate was successfully controlled at the desired rates and the fed-batch mode employed is considered appropriate for examining the correlation between the specific growth rate and the efficiency of recombinant product formation in the recombinant E. coli strains. The two recombinant proteins were expressed as fusion proteins and the concentration in the culture broth was increased to 15 g fusion growth hormone l⁻¹ and 7 g fusion glucagon l⁻¹. The fusion growth hormone was initially expressed as soluble protein but seemed to be gradually aggregated into inclusion bodies as the expression level increased, whereas the synthesized fusion glucagon existed as a cytoplasmic soluble protein during the whole induction period. The stressful conditions of cultivation employed (i.e. high-cell-density cultivation at low growth rate) may induce the increased production of various host-derived chaperones and thereby enhance the folding efficiency of synthesized heterologous proteins. The synthesis of the recombinant fusion proteins was strongly growth-dependent and more efficient at a higher specific growth rate. The mechanism linking specific growth rate with recombinant protein productivity is likely to be related to the change in cellular ribosomal content. Received: 27 May 1997 / Received last revision: 31 October 1997 / Accepted: 21 November 1997     

Aromatic di-alanine repeats (AdAR) are structural motifs characteristic of the soluble N-ethylmaleimide-sensitive factor attachment protein (SNAP) family

The aromatic di-alanine repeat is a novel 12-amino acid-long motif constituting alternate small and large hydrophobic residues that mediate the close packing of alpha-helices. A hidden Markov model profile was constructed from the motifs initially described in Soluble N-ethyl maleimide-sensitive factor attachment proteins (SNAP), a family of soluble proteins involved in intracellular membrane fusion. Scanning different sets of protein sequences showed unambiguously that this profile defines a structural motif independent of the tetratrico peptide repeat, another widespread alpha-helical motif. In addition to SNAP, aromatic di-alanine repeats are found in selective LIM homeodomain binding proteins (SLB) and in proteins from the Pyrococcus and Archaeoglobus prokaryotes.     

A disulfide-bonded DnaK dimer is maintained in an ATP-bound state

DnaK, a major Hsp70 molecular chaperones in Escherichia coli, is a widely used model for studying Hsp70s. We recently solved a crystal structure of DnaK in complex with ATP and showed that DnaK was packed as a dimer in the crystal structure. Our previous biochemical studies supported the formation of a specific DnaK dimer as observed in the crystal structure in solution in the presence of ATP and suggested an important role of this dimer in efficient interaction with Hsp40 co-chaperones. In this study, we dissected the biochemical properties of this DnaK dimer. To restrict DnaK in this dimer form, we mutated two residues on the dimer interface to cysteine, A303C, and H541C. Upon oxidation, this DnaK-A303C-H541C protein formed a specific dimer linked by disulfide bonds formed between A303C and H541C only in the presence of ATP, consistent with the crystal structure. Intriguingly, this disulfide-bond-linked dimer of DnaK-A303C-H541C has reduced ATPase activity and decreased affinity for peptide substrate. More interestingly, unlike wild-type DnaK, the peptide substrate-binding kinetics of this dimer is drastically accelerated even in the absence of ATP, suggesting this dimer is restricted in an ATP-bound conformation regardless of nucleotide bound, which was further supported by our analysis using tryptophan fluorescence and ATP-induced peptide release. Thus, formation of the dimer restricted DnaK in an ATP-bound state and blocked the progression through the chaperone cycle. Productive progression through the chaperone cycle requires the dissociation of this transient dimer. Surprisingly, a significantly compromised interaction with Hsp40 co-chaperone was observed for this disulfide-bond-linked dimer. Thus, dissociation of this DnaK dimer is equally crucial for efficient Hsp40 interaction. An initial interaction between Hsp70 and Hsp40 requires the formation of DnaK dimer; but a stable Hsp70-Hsp40 interaction may follow the dissociation of the dimer.     

Recombinant Fusion Proteins for the Industrial Production of Disulfide Bridge Containing Peptides: Purification, Oxidation without Concatamer Formation, and Selective Cleavage

We report the biotechnical production of peptides of approximately 35–50 amino acids in length containing one intramolecular disulfide bridge, using a recombinant fusion tail approach. This method fills the technological gap when either (a) chemical synthesis fails due to known problematic peptide sequences or (b) if simple recombinant expression is unsuccessful due to degradation. The fusion tail described here serves several purposes: (i) it enables high expression levels inEscherichia colito be achieved; (ii) it renders the fusion protein fairly soluble; (iii) it contains a histidine affinity tag for easy purification on Ni-chelate resins, which also serves as a catalyst for the oxygen-dependent formation of the disulfide bridge; and (iv) it suppresses the formation of concatamers during the oxidation process through steric hindrance. The purified fusion protein is then immobilized on a reversed phase column for two purposes: (i) chemical cleavage of the fusion tail by cyanogen bromide and (ii) subsequent purification of the peptide. A very hydrophilic fusion partner is required so that immobilization on the reversed phase column always occurs due to the peptide. Sensitive hydrophobic residues are thereby protected from the cleavage reagent while the cleaved hydrophilic fusion tail is easily separated from the hydrophobic peptide. The method is exemplified by eight peptides representing an immunodominant epitope of the human immunodeficiency virus, but may be useful for a significant variety of similar peptides.     

The N-domain of Escherichia coli phosphoglycerate kinase is a novel fusion partner to express aggregation-prone heterologous proteins

As a fusion partner to express aggregation-prone heterologous proteins, we investigated the efficacy of Escherichia coli phosphoglycerate kinase (ePGK) that consists of two functional domains (N- and C-domain) and reportedly has a high structural stability. When the full-length ePGK (F-ePGK) was used as a fusion partner, the solubility of the heterologous proteins increased, but some of them still had a large fraction of insoluble aggregates. Surprisingly, the fusion expression using the N-domain of ePGK (N-ePGK) made the insoluble fraction significantly reduce to less than 10% for all the heterologous fusion proteins tested. Also, we evaluated the efficacy of N-ePGK in making the target proteins be expressed with their own native function or structure. It was found that of human ferritin light chain, bacterial arginine deiminase, human granulocyte colony stimulating factor were synthesized evidently with the self-assembly function, L-arginine-degrading activity, and the correct secondary structure, respectively, through the fusion expression using N-ePGK. These results indicate that N-ePGK is a highly potent fusion partner that can be widely used for the synthesis of a variety of heterologous proteins in E. coli.