Overproduced Brucella abortus PdhS-mCherry forms soluble aggregates in Escherichia coli, partially associating with mobile foci of IbpA-YFP
© Van der Henst et al; licensee BioMed Central Ltd. 2010
Received: 14 July 2010
Accepted: 28 September 2010
Published: 28 September 2010
When heterologous recombinant proteins are produced in Escherichia coli, they often precipitate to form insoluble aggregates of unfolded polypeptides called inclusion bodies. These structures are associated with chaperones like IbpA. However, there are reported cases of "non-classical" inclusion bodies in which proteins are soluble, folded and active.
We report that the Brucella abortus PdhS histidine kinase fused to the mCherry fluorescent protein forms intermediate aggregates resembling "non-classical" inclusion bodies when overproduced in E. coli, before forming "classical" inclusion bodies. The intermediate aggregates of PdhS-mCherry are characterized by the solubility of PdhS-mCherry, its ability to specifically recruit known partners fused to YFP, suggesting that PdhS is folded in these conditions, and the quick elimination (in less than 10 min) of these structures when bacterial cells are placed on fresh rich medium. Moreover, soluble PdhS-mCherry foci do not systematically colocalize with IpbA-YFP, a marker of inclusion bodies. Instead, time-lapse experiments show that IbpA-YFP exhibits rapid pole-to-pole shuttling, until it partially colocalizes with PdhS-mCherry aggregates.
The data reported here suggest that, in E. coli, recombinant proteins like PdhS-mCherry may transit through a soluble and folded state, resembling previously reported "non-classical" inclusion bodies, before forming "classical" inclusion bodies. The dynamic localization of IbpA-YFP foci suggests that the IbpA chaperone could scan the E. coli cell to find its substrates.
Escherichia coli is widely used to produce recombinant proteins of interest. One of the major concerns in the overproduction process is the formation of insoluble structures called inclusions bodies (IB) [1, 2]. IB formation results from the aggregation of misfolded polypeptides that have escaped quality control by chaperones and proteases to interact through their exposed hydrophobic regions before precipitating . Aggregate formation and features are influenced by various growth conditions such as temperature and pH , culture phase  and glucose/oxygen availability .
In vivo protein aggregation is a dynamic reversible process . Chaperones involved in aggregate dissociation, e.g. DnaK/DnaJ/ClpB and IbpA/IbpB, colocalize with IB in E. coli[8–11]. Recently, it has been reported that aggregate cellular localization is not random . Small protein aggregates are delivered to a cell pole to form larger structures that are further dissolved by an energy dependent process . All proteins in IB were initially considered as unfolded, but it has been shown that some polypeptides inside aggregates are present in an active form [2, 13, 14]. Several groups reported the formation of "non-classical" IB mainly characterized by the presence of folded and soluble recombinant proteins [15, 16].
Here, we report a novel example of "non-classical" IB that contain folded and soluble recombinant proteins and only transiently interact with the IpbA chaperone. Indeed, overproduction of Brucella abortus PdhS cytoplasmic histidine kinase  in E. coli revealed that PdhS-mCherry fusions were first folded and soluble in aggregates formed during the stationary phase of culture before forming insoluble structures having all the characteristics of "classical" IB. These "classical" IB recruited IpbA-YFP, as previously reported for other IB in E. coli, unlike the intermediate "non classical" IB. We observed that IbpA-YFP was able to form foci with very dynamic properties inside E. coli and to reach and colocalize with soluble PdhS-mCherry aggregates.
PdhS-mCherry forms growth phase-dependent aggregates in E. coli
Given that bacteria growth conditions strongly influence aggregate formation, we checked whether the fluorescent foci were dependent on the growth phase, as previously reported for IB . Using the pdhS-mCherry overexpressing strain, we observed bacteria grown until the early, mid and late stationary phase, corresponding to bacteria having just reached the maximal turbidity of the culture (t0), the bacteria 12 h later (t12), and the bacteria 36 h later (t36), respectively. At t0 of the stationary culture phase, very few bacteria (4%, n = 100) showed polar fluorescent foci as many were associated with a bright diffuse cytoplasmic fluorescent signal (Fig. 1A). Twelve hours later in the same medium (t12), polar fluorescent foci were observed (in 98% of the observed bacteria, n = 100), together with a decrease of the diffuse cytoplasmic fluorescent signal (Fig. 1B). No detectable refractile bodies were observed in these conditions. After 24 additional hours (t36), larger and brighter fluorescent polar foci were formed, colocalizing with dense refractile bodies typical of "classical" IB, and accompanied by a strong decrease of the diffuse fluorescent signal (Fig. 1C).
Colocalization assays between PdhS-mCherry fluorescent aggregates and IbpA-YFP fusions
IbpA (for Inclusion body protein A) is a small heat shock chaperone discovered in E. coli. The IbpA-YFP fusion was already successfully used to label inclusion bodies in vivo, in single cells of E. coli. As PdhS-mCherry fluorescent polar foci generated during the mid and late stationary culture phases could differ from each other, we tested their possible colocalization with the IbpA-YFP fusion.
PdhS-mCherry fusions in fluorescent foci of mid stationary phase cells display properties of folded proteins
Since the PdhS-mCherry foci observed during the mid stationary phase did not colocalize with IbpA-YFP, it was tempting to speculate that PdhS-mCherry fusions were correctly folded in these aggregates. In keeping with this idea, Western blot analysis using anti-mCherry antibodies showed that PdhS-mCherry was mainly found in the soluble fraction of bacteria grown until the late stationary phase (Additional file 2, Figure S1). When soluble extracts were examined by gel permeation combined with fluorescence and Western blot analysis, soluble PdhS-mCherry proteins were identified as a single peak, with a predicted molecular weight between 669 kDa and 20,000 kDa, the upper limit of the fractionation range (Additional file 2, Figure S2). This suggests that the fusion is able to form multimers with a defined number of monomers, further implying that PdhS-mCherry is folded.
We report that, when overproduced in E. coli, B. abortus PdhS fused to mCherry is able to form intermediate aggregates of soluble proteins resembling previously reported "non-classical" IB [3, 15], before forming "classical" IB. These intermediate aggregates are very different from "classical" IB because they are soluble, are quickly removed when bacteria are placed in rich medium (Fig. 2A), do not systematically colocalize with IbpA-YFP (Fig. 3B) and are still able to recruit known PdhS partners (Fig. 6). The observation of "intermediate" aggregates of soluble proteins does not fit with a simple model of IB formation in which unfolded proteins precipitate to form IB immediately after translation. Our observations thus suggest that some proteins could form aggregates of folded and soluble polypeptides before their precipitation into "classical" IB. The initial solubility of heterologous PdhS-mCherry could be due to a slow expression (as suggested by the slow accumulation of PdhS-mCherry, Additional file 2, Figure S1), since there is no predicted promoter identified upstream the CDS in the pCVDH07 plasmid, and the codon bias of pdhS CDS is probably not optimal for sustained translation. Strong accumulation could lead to the saturation of chaperones and proteolysis activities, explaining the slow transition between soluble and "classical" IB.
The data we report suggests that PdhS-mCherry is folded in aggregates resembling "non-classical" IB. The data supporting the folded state of PdhS in E. coli are that PdhS-mCherry (i) is soluble and forms multimers of homogeneous size, and (ii) is still able to interact with partners like the fumarase FumC and the response regulator DivK. The recent resolution of a complex between a histidine kinase and its cognate response regulator  strongly suggests that the dimerization and histidine-containing phosphotransfer (DHp) domain of the kinase needs to be folded to allow interaction with the response regulator. It is therefore predictable that at least the DHp domain of PdhS-mCherry is folded to allow interaction with DivK-YFP. Interestingly, we previously reported that B. abortus PdhS was able to colocalize with B. abortus fumarase FumC, but not with C. crescentus FumC , and here the recruitment of FumC proteins by PdhS-mCherry is consistent with this specificity (Fig. 6A and 6B). Moreover, it means that fusions to YFP are not all aspecifically associated to soluble aggregates of PdhS-mCherry resembling "non-classical" IB.
Additional file 1:Movement of IbpA-YFP in E. coli cells producing PdhS-mCherry. Time lapse movie of E. coli cells at stationary (t12) phase, producing PdhS-mCherry (red) and IbpA-YFP (yellow). The time interval between two pictures is 2 min. (AVI 7 MB)
PdhS-mCherry is a new example of a protein able to form soluble "non-classical" inclusion bodies in E. coli. Here we report a detailed characterization of these particular IB using several approaches. These IB are able to recruit partners of PdhS, suggesting that PdhS remains folded in these IB, at least during a first step of IB maturation. The "non-classical" IB are probably highly sensitive to proteolysis, since they are quickly cleared from the cells when the environmental conditions change. Time lapse analysis of E. coli cells containing PdhS-mCherry "non-classical" IB indicates that IbpA-YFP foci move rapidly inside the bacteria until they reach fluorescent aggregates. The characterization of IbpA-YFP movement inside E. coli should be investigated further as it could indicate how the IbpA chaperone is able to scan the cytoplasm to recognize intracellular protein aggregates.
Strains, plasmids and media
E. coli strains MG1655 expressing the ibpA coding sequence (CDS) fused to the enhanced version of YFP CDS (13) and S17-1, TOP10 and DH10B were grown in liquid Luria-Bertani (LB) broth medium at 37°C. Antibiotics were used at the following concentrations when appropriate: kanamycin, 50 μg/ml and chloramphenicol, 20 μg/ml. The pdhS CDS was inserted in fusion with the mCherry CDS on a high-copy number plasmid, in the opposite orientation of the lac promoter, derived from the pBluescriptKS vector (Stratagene); this plasmid was named pCVDH07. The E. coli strains transformed with pCVDH07 were grown in liquid LB with kanamycin for times indicated in the text, without induction of gene expression for the PdhS-mCherry fusion. The growth was followed by measuring the optical density at 600 nm.
For fluorescence imaging, E. coli S17-1 and MG1655 strains were placed on a microscope slide that was layered with 1% agarose containing either PBS or 1% agarose containing LB medium (40 g/l). Time-lapse microscopy was performed by placing strains on a microscope slide that was layered with a 1% agarose pad containing LB medium. Fluorescence corresponding to the mCherry reporter was observed at 583 nm using a TxRed filter. Fluorescence corresponding to the YFP signal was observed using an emission filter centered on 535 nanometers and an excitation from 490 to 510 nanometers. Samples were observed every 2 min using a Nikon i80 fluorescence microscope and the NIS software from Nikon with a Hamamatsu camera.
Protein extracts and Western blotting
Cultures at the mid stationary phase (optical density at 600 nm of 1.5) were centrifuged and then washed twice in 20 mM Tris-HCl 100 mM NaCl buffer at pH 7.9, lysed by sonication carried out over periods of 30 s with 1 min intervals in cooled tube on ice using a Branson sonifier 150. The cells were disrupted as observed microscopically to obtain total bacterial lysates that were centrifuged for 15 minutes at 13,000 rpm at 4°C. After centrifugation, the supernatant was harvested and considered as the soluble fraction of the bacterial cell lysate. The pellet was resuspended in PBS to reach the same volume as the supernatant, and was considered as the insoluble fraction. The soluble and insoluble fractions were then analysed by Western blot using polyclonal anti-DsRed antibodies (Clontech Laboratories, Inc) recognizing the mCherry protein, as previously reported (16).
The soluble fraction of bacterial lysate (500 μl) was injected into a HiPrep 16/60 Sephacryl S-500 HR column (GE Healthcare). The calibration curve was obtained using thyroglobulin (669 kDa), apoferritin (443 kDa) and amylase (200 kDa). One milliliter fractions were collected and tested for the presence of the mCherry fluorochrome using a fluorimeter equipped with a TxRed filter. Positive fluorescent fractions were then tested by Western blot analysis using anti-DsRed antibodies.
We thank Ariel B. Lindner for kindly providing the E. coli strain expressing the chromosomal ibpA-yfp fusion and Etienne Maisonneuve for fruitful discussions. This work was supported by the FRFC (Collective Fundamental Research Fund, agreements 2.4521.04 and 2.4541.08) and by the University of Namur. C. Van der Henst and M. Deghelt held PhD fellowships from the FRIA (Industrial and Agricultural Research Training Fund). C. Charlier held a fellowship from the FRS-FNRS.
- Speed MA, Wang DI, King J: Specific aggregation of partially folded polypeptide chains: the molecular basis of inclusion body composition. Nat Biotechnol. 1996, 14 (10): 1283-1287. 10.1038/nbt1096-1283.View ArticlePubMedGoogle Scholar
- Villaverde A, Carrio MM: Protein aggregation in recombinant bacteria: biological role of inclusion bodies. Biotechnol Lett. 2003, 25 (17): 1385-1395. 10.1023/A:1025024104862.View ArticlePubMedGoogle Scholar
- Ventura S, Villaverde A: Protein quality in bacterial inclusion bodies. Trends Biotechnol. 2006, 24 (4): 179-185. 10.1016/j.tibtech.2006.02.007.View ArticlePubMedGoogle Scholar
- Strandberg L, Enfors SO: Factors influencing inclusion body formation in the production of a fused protein in Escherichia coli. Appl Environ Microbiol. 1991, 57 (6): 1669-1674.PubMed CentralPubMedGoogle Scholar
- Maisonneuve E, Ezraty B, Dukan S: Protein aggregates: an aging factor involved in cell death. J Bacteriol. 2008, 190 (18): 6070-6075. 10.1128/JB.00736-08.PubMed CentralView ArticlePubMedGoogle Scholar
- Kwiatkowska J, Matuszewska E, Kuczynska-Wisnik D, Laskowska E: Aggregation of Escherichia coli proteins during stationary phase depends on glucose and oxygen availability. Res Microbiol. 2008, 159 (9-10): 651-657. 10.1016/j.resmic.2008.09.008.View ArticlePubMedGoogle Scholar
- Carrio MM, Villaverde A: Construction and deconstruction of bacterial inclusion bodies. J Biotechnol. 2002, 96 (1): 3-12. 10.1016/S0168-1656(02)00032-9.View ArticlePubMedGoogle Scholar
- Allen SP, Polazzi JO, Gierse JK, Easton AM: Two novel heat shock genes encoding proteins produced in response to heterologous protein expression in Escherichia coli. J Bacteriol. 1992, 174 (21): 6938-6947.PubMed CentralPubMedGoogle Scholar
- Winkler J, Seybert A, Konig L, Pruggnaller S, Haselmann U, Sourjik V, Weiss M, Frangakis AS, Mogk A, Bukau B: Quantitative and spatio-temporal features of protein aggregation in Escherichia coli and consequences on protein quality control and cellular ageing. Embo J. 29 (5): 910-923. 10.1038/emboj.2009.412.
- Kuczynska-Wisnik D, Kedzierska S, Matuszewska E, Lund P, Taylor A, Lipinska B, Laskowska E: The Escherichia coli small heat-shock proteins IbpA and IbpB prevent the aggregation of endogenous proteins denatured in vivo during extreme heat shock. Microbiology. 2002, 148 (Pt 6): 1757-1765.View ArticlePubMedGoogle Scholar
- Lindner AB, Madden R, Demarez A, Stewart EJ, Taddei F: Asymmetric segregation of protein aggregates is associated with cellular aging and rejuvenation. Proc Natl Acad Sci USA. 2008, 105 (8): 3076-3081. 10.1073/pnas.0708931105.PubMed CentralView ArticlePubMedGoogle Scholar
- Rokney A, Shagan M, Kessel M, Smith Y, Rosenshine I, Oppenheim AB: E. coli transports aggregated proteins to the poles by a specific and energy-dependent process. J Mol Biol. 2009, 392 (3): 589-601. 10.1016/j.jmb.2009.07.009.View ArticlePubMedGoogle Scholar
- Oberg K, Chrunyk BA, Wetzel R, Fink AL: Nativelike secondary structure in interleukin-1 beta inclusion bodies by attenuated total reflectance FTIR. Biochemistry. 1994, 33 (9): 2628-2634. 10.1021/bi00175a035.View ArticlePubMedGoogle Scholar
- Gonzalez-Montalban N, Garcia-Fruitos E, Ventura S, Aris A, Villaverde A: The chaperone DnaK controls the fractioning of functional protein between soluble and insoluble cell fractions in inclusion body-forming cells. Microb Cell Fact. 2006, 5: 26-10.1186/1475-2859-5-26.PubMed CentralView ArticlePubMedGoogle Scholar
- Stampolidis P, Kaderbhai NN, Kaderbhai MA: Periplasmically-exported lupanine hydroxylase undergoes transition from soluble to functional inclusion bodies in Escherichia coli. Arch Biochem Biophys. 2009, 484 (1): 8-15. 10.1016/j.abb.2009.01.017.View ArticlePubMedGoogle Scholar
- Jevsevar S, Gaberc-Porekar V, Fonda I, Podobnik B, Grdadolnik J, Menart V: Production of nonclassical inclusion bodies from which correctly folded protein can be extracted. Biotechnol Prog. 2005, 21 (2): 632-639. 10.1021/bp0497839.View ArticlePubMedGoogle Scholar
- Hallez R, Mignolet J, Van Mullem V, Wery M, Vandenhaute J, Letesson JJ, Jacobs-Wagner C, De Bolle X: The asymmetric distribution of the essential histidine kinase PdhS indicates a differentiation event in Brucella abortus. Embo J. 2007, 26 (5): 1444-1455. 10.1038/sj.emboj.7601577.PubMed CentralView ArticlePubMedGoogle Scholar
- Mignolet J, Van der Henst C, Nicolas C, Deghelt M, Dotreppe D, Letesson JJ, De Bolle X: PdhS, an old-pole-localized histidine kinase, recruits the fumarase FumC in Brucella abortus. J Bacteriol. 2010, 192 (12): 3235-3239. 10.1128/JB.00066-10.PubMed CentralView ArticlePubMedGoogle Scholar
- Casino P, Rubio V, Marina A: Structural insight into partner specificity and phosphoryl transfer in two-component signal transduction. Cell. 2009, 139 (2): 325-336. 10.1016/j.cell.2009.08.032.View ArticlePubMedGoogle Scholar
- Ratajczak E, Strozecka J, Matuszewska M, Zietkiewicz S, Kuczynska-Wisnik D, Laskowska E, Liberek K: IbpA the small heat shock protein from Escherichia coli forms fibrils in the absence of its cochaperone IbpB. FEBS Lett. 2010, 584 (11): 2253-2257. 10.1016/j.febslet.2010.04.060.View ArticlePubMedGoogle Scholar