Overexpression, refolding, and purification of the

Protein Expression and Purification 40 (2005) 126–133 www.elsevier.com/locate/yprep

Overexpression, refolding, and purification of the major immunodominant outer membrane porin OmpC from Salmonella typhi: characterization of refolded OmpC P.D. Kumar, S. Krishnaswamy* Department of Genetic Engineering, School of Biotechnology, Madurai Kamaraj University, Madurai 625 021, India Received 12 September 2004, and in revised form 11 December 2004 Available online 8 January 2005

Abstract The major immunodominant integral outer membrane protein C (OmpC) from Salmonella typhi Ty21a was overexpressed, without the signal peptide, in Escherichia coli. The protein aggregates as inclusion bodies (IBs) in the cytoplasm. OmpC from IBs was solubilized with 4 M urea and refolded. This involved rapid dilution of unfolded OmpC into a refolding buffer containing polyoxyethylene-9-lauryl ether (C12E9) and glycerol. The refolded OmpC (rfOmpC) was shown to be structurally similar to the native OmpC by SDS–PAGE, Western blotting, tryptic digestion, ultrafiltration, circular dichroism, and fluorescence spectroscopic techniques. Crystals of rfOmpC were obtained in preliminary crystallization trials. The rfOmpC also sets a stage for rational design by recombinant DNA technology for vaccine design and high resolution structure determination. Ó 2005 Elsevier Inc. All rights reserved. Keywords: Membrane protein; OmpC; Salmonella typhi; Refolding

Porins of Gram-negative bacteria are outer membrane proteins that act as receptors for bacteriophages [1,2] and are involved in a variety of functions like solute transport [3], pathogenesis and immunity [4]. The trimeric porins show stability to temperature and denaturants. They are also resistant to proteolysis. These properties make them good candidates for industrial applications towards the development of oral vaccines, biosensors, and nanoreactors [5,6]. Porins can be used for the development of engineered multivalent vaccines by heterologous epitope presentation through their loops [7–9]. Structural characterization and protein engineering of porins is enabled by the development of an overexpression system. Heterologous expression in Escherichia coli along with signal peptide is toxic to cells, while the removal of signal peptide leads to the formation of cyto*

Corresponding author. Fax: +91 452 2459105. E-mail addresses: [email protected] (S. Krishnaswamy), mkukrishna@rediffmail.com (S. Krishnaswamy). 1046-5928/$ - see front matter Ó 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.pep.2004.12.023

plasmic inclusion bodies (IBs) [10]. These IBs can be denatured with protein denaturants and subsequently refolded into buffers containing detergents that mimic the hydrophobic environment. Refolding studies on various porins such as E. coli outer membrane protein A (OmpA), Hemophilus influenza porin [11], Rhodobacter capsulatus porin, and PorA of Neisseria meningitides [12] suggest that refolding can be achieved in the presence of detergent alone and lipopolysaccharides (LPS) or other membrane components are not required. On the contrary, assembly as trimers of in vitro synthesized outer membrane protein F (OmpF) of E. coli was achieved only in the presence of LPS and membrane fractions. LPS and outer membrane fractions helped in native like trimerization even in the absence of detergents [13]. Biochemical evidence was provided for the role of phospholipids and LPS at distinct stages of assembly process of E. coli phosphate porin (PhoE) [14]. In the case of Salmonella typhi OmpC, we have earlier shown the role of LPS on the stability of the trimer [15]. Functional refolding

P.D. Kumar, S. Krishnaswamy / Protein Expression and Purification 40 (2005) 126–133

studies of Campylobacter jejuni [16], major outer membrane protein (MOMP), porin showed 95% refolding in the presence of homologous GroEL, suggesting the application of chaperons in the efficient in vitro refolding of porins. Refolding was used as a successful strategy in the case of Rhodopseudomonas blastica porin [10] and OpCA from N. meningitidis [17] where they were heterologously expressed as IBs in E. coli, refolded, and shown to be structurally similar to their native structures. Despite the fact that a variety of membrane proteins were refolded successfully and structurally characterized, refolding of membrane proteins still remains a bottleneck. Successful refolding into native conformation with higher yields becomes crucial for the structural characterization of membrane proteins that are underrepresented in the protein data bank (PDB). OmpC from S. typhi is a major surface antigen, expressed throughout the infection period [18] in typhoid patients. It is a good candidate to display heterologous epitopes on the cell surface [8,18]. The functional and mature OmpC is a homotrimer. The monomer without the signal peptide has 357 amino acids and a molecular weight of 39 kDa. The purification and crystallization of native Ty21a OmpC has been described earlier [19]. Crystals from native S. typhi OmpC and recombinant OmpC, expressed in E. coli HB101, earlier diffracted ˚ only [20]. Heterogeneity due to the bound up to 7 A LPS and improper packing of molecules in the crystal due to the longer loops were suspected to be possible reasons for low resolution. S. typhi Ty21a ompC gene was cloned and overexpressed to achieve higher yields without having bound LPS and to provide a system for protein engineering of the loops. The refolded OmpC (rfOmpC) was shown to be structurally similar to native Ty21a OmpC by Western blot using MAbs, circular dichroism (CD), and fluorescence spectroscopy. Stability of rfOmpC was analyzed with tryptic digestion, SDS–PAGE, and temperature scan with CD.

127

was based on the sequence information available for S. typhi OmpC [22]. NdeI and BamHI sites were introduced through primers at 5 0 and 3 0 end of the amplicon, respectively. Reverse primer incorporates a stop codon upstream to BamHI site. OmpC amplicon was subsequently cloned into NdeI, BamHI sites of pET20 b (+) vector (Novagen) for overexpression. The overexpression construct for S. typhi Ty21a OmpC was named as pLF2 and confirmed by sequencing Forward primer: 5 0 GCAGCGAATCATATGGAA ATTTATAATAAAGAC3 0 Reverse primer: 5 0 AACATCTTTGGATCCTTAGA ACTGGTAAAC3 0 Overexpression and extraction of inclusion bodies Escherichia coli BL21 (DE3) was transformed with pLF2. BL21/pLF2 was grown overnight in Luria–Bertani (LB) medium. One percent of this inoculum was added to 1 L of LB medium and the culture was grown in a shaker at 37 °C. At 0.6 OD, culture was induced with 0.2 mM isopropyl-b-D -thiogalactopyranoside (IPTG) and kept for expression for 5 h. Cell pellet was obtained by centrifuging at 7000 rpm for 13 min at 20 °C (rotor RPR-12-2, Hitachi high speed centrifuge). Cell pellet was washed twice with 0.8% (w/v) saline. Cells were disrupted by sonication (SONICS, Vibra cell 300 W, total run time of 3 min with pulsar of 9 s on and 5 s off). Crude IBs were obtained by centrifugation of cell lysate at 7000 rpm for 10 min at 20 °C (Fig. 1).

Materials and methods Bacterial strains and culture conditions The bacterial strains used in this study are the vaccine strain of S. typhi Ty21a (galE mutant [21], which produces aberrant LPS) and E. coli BL21 (DE3). BL21/ pLF2 (OmpC overexpression construct) was grown in LB medium supplemented with 50 lg/ml ampicillin. Overexpression construct Primer design ompC gene was PCR amplified from S. typhi Ty21a genomic DNA. Primers were designed to omit the region corresponding to signal peptide. Primer design

Fig. 1. Purification of rfOmpC. Lane 1, Uninduced cells; 2, cells induced with 0.2 mM IPTG; 3, supernatant of cell lysate; 4, crude inclusion bodies; 5 and 6, supernatant of IBs washes with TTN and TN buffers, respectively; 7, unfolded OmpC; 8 and 9, rfOmpC; 10, protein marker. Samples in all the lanes were boiled except for lane 9. The 12% SDS–PAGE was Coomassie stained.

128


Table 1 Purification of rfOmpC from inclusion bodies Steps

Amount of protein (mg from 1 L culture)

Step yield (%)

Overall yield (%)

Crude IB (wet weight) Purified IB (wet weight) UF OmpC (4 M urea) rfOmpC rfOmpC after concentration and ultracentrifugation rfOmpC bound to Q-Sepharose

1000 900 120 100 80 15

— — 100 83.3 80 18.75

— — 100 83.3 66.6 12.5

IBs were further washed two times with TTN buffer (50 mM Tris, pH 8.5, 0.1 M NaCl, and 2% Triton X100) and with TN buffer (50 mM Tris, pH 8.5, and 0.1 M NaCl). After each wash, the resuspended solution was centrifuged at 7000 rpm for 7 min at 20 °C. Typical yield of purified IBs was 0.9 g/L culture (Table 1).

Tryptic digestion Trypsin was added in 1:400 (trypsin to protein weight ratio) and incubated at 37 °C for 1 h. Samples were loaded onto 12% SDS–PAGE. Circular dichroism spectroscopy

Solubilization of IBs IBs were solubilized in Tris–urea buffer (50 mM Tris, pH 8.5, 0.1 M NaCl, and 4 M urea). Solubilization was done for 5 h on moderate shaking at 37 °C. This suspension was centrifuged at 13,000 rpm for 20 min at 20 °C to obtain unfolded OmpC (UFOmpC) in the supernatant. UFOmpC was passed through 0.2 lm filter to remove aggregates and particulate matter. Refolding Refolding buffer [50 mM Tris, pH 8.5, 0.1 M NaCl, 10% (v/v) glycerol, and 0.2% (v/v) polyoxyethylene-9laurylether (C12E9, Sigma)] was kept at 4 °C with stirring, into which the UFOmpC was rapidly added in 1:5 protein to buffer ratio. The solution was incubated at 4 °C overnight and concentrated using ultrafiltration stirred cell (Amicon, Millipore) with 50 kDa MWCO membrane. The protein solution was subjected to ultracentrifugation at 30,000 rpm for 90 min at 20 °C using RPR 50-T rotor (Hitachi ultracentrifuge) to remove small aggregates and particulate matter. rfOmpC was further purified by ion-exchange chromatography.

The far UV spectra were recorded using JASCO spectropolarimeter J-715 attached with peltier type temperature control system (PTC 348 WI) at Molecular Biophysics Unit, Indian Institute of Science, Bangalore, India. Cuvette with 1 cm path length was used. Native Ty21a OmpC was in 50 mM sodium phosphate, pH 7.6, and 0.1% C12E9. This was diluted in a buffer containing 50 mM Tris, pH 8.5, 0.1 M NaCl, 10% glycerol, and 0.2% C12E9. rfOmpC (column bound and unbound) samples were used at 50 lg/ml concentration in a buffer containing 50 mM Tris, pH 8.5, 0.1 M NaCl, 10% glycerol, and 0.2% C12E9 for wavelength and temperature scan experiments. Wavelength scan for unfolded proteins was performed at 100 mg/ml protein concentration in Tris–urea buffer. The parameters used were: resolution 0.5 nm, bandwidth 1.0 nm, sensitivity 50 m deg, response 4 s and accumulation 3. Temperature scan was performed from 25 °C to 90 °C at 218 nm. The parameters used were: temperature slope 3 °C/min, bandwidth 2 nm, sensitivity 100 m deg, and response 4 s. The CD values were expressed in [h] MRW deg cm2/dmol. Fluorescence spectroscopy

Ion-exchange chromatography Q-Sepharose fast flow (Sigma) ion exchange column was equilibrated with binding buffer (50 mM Tris, pH 8.5, 0.1 M NaCl, 10% glycerol, and 0.2% C12E9). rfOmpC was loaded onto the column at 10 ml/h flow rate. Column was washed with the binding buffer (five bed volumes) to remove unbound protein and eluted with binding buffer containing 1 M NaCl. To reduce the salt concentration in the eluent to 0.1 M, the eluted protein was diluted with binding buffer without NaCl and concentrated, in a stepwise manner, in ultrafiltration stirred cell with 50 kDa MWCO membrane.

The fluorescence spectra (emission scan) were taken using the Fluoro Max-3 (JOBIN YVON HORIBA) spectrophotometer at the Molecular Biophysics Unit, Indian Institute of Science, Bangalore. Samples were excited at 280 nm in a 1 cm path length cuvette. Western blot Protein from SDS–PAGE (12%) was transferred to PVDF membrane (Immobilon-P 0.45 lm pore size, Millipore) using a homemade wet gel transfer apparatus. Transfer was carried out at 100 V for 1 h. Salmonella


porin specific MAb (MPN5) was used as primary antibody and 3,3 0 -diaminobenzidine (DAB, Sigma) in citrate phosphate buffer (pH 5.5) at 1 mg/ml concentration was used as substrate for horse radish peroxidase (HRP) linked secondary antibody (Genei, Bangalore, India).

Results and discussion OmpC was expressed without the signal peptide, as overexpression with signal peptide was shown to be toxic to cells leading to premature cell death in the case of other outer membrane proteins such as R. blastica porin [10]. From the available homology model of S. typhi OmpC [18] (PDB Id: 1IIV) it is known that N and C termini are linked with a salt bridge making the molecule essentially closed. Addition of any residue at either of the termini could disrupt the salt bridge and possibly perturb the structure. So, to keep the number of amino acids constant, the first amino acid, alanine, was deleted and methionine, which is required for translational start, was introduced. Overexpression without signal peptide does not lead the protein into the outer membrane as a result of which the misfolded protein aggregates resulting in the formation of cytoplasmic IBs (Fig. 1). IBs were solubilized using denaturants (urea or guanidine HCl) in different concentrations. Urea and guanidine HCl show concentration dependent binding to protein [23]. So, modulation of denaturant concentration determines the degree of unfolding. Denaturant concentration has been critical in obtaining higher yields of natively folded purified protein. High denaturant concentration solubilizes other contaminating proteins also along with OmpC. These impurities affected the efficiency and yields of refolding. Moderate concentration of urea (4 M) possibly disrupts the intermolecular inter-

129

actions with partial unfolding of porin and was found to be most suitable. The process of refolding OmpC into the native b-barrel structure and oligomer formation as a trimer can be affected by misfolding and aggregation. Different approaches were tried for refolding such as dialysis, reverse dialysis, rapid dilution, and slow dilution. Rapid dilution was found to be efficient in terms of yield and time required for purification on scaling up. This method is advantageous over slow dilution, which results in insufficient concentration of refolded monomers for a prolonged period in the beginning of refolding experimental process [24]. Refolding additives like sucrose (folding enhancer), PEG 3350 (aggregation suppressor), glycerol (protein stability enhancer) [25], and different ionic and nonionic detergents were also screened. These various combinations of solubilizing agents, refolding buffers, and methods of refolding were combined to optimize the conditions for refolding. The combination that worked well for OmpC was glycerol and C12E9. Refolding in the presence of SDS, an anionic detergent, led to aggregation. C12E9, a nonionic detergent, provided the hydrophobic environment required for refolding, suggesting that proper refolding of membrane proteins is possible only in the presence of nonionic detergents. Glycerol probably helped in preventing the competing process of aggregation during refolding, possibly by enhancing protein stability [26]. Characterization of rfOmpC Salmonella typhi OmpC belongs to the class of 16 stranded b-barrel outer membrane proteins (E. coli OmpF, E. coli PhoE), which are trimeric in nature [3]. General properties of these porins, arising from the strong trimeric association, include a high thermal stability, and resistance to proteolysis and detergents such as SDS. These properties are essential for the survival

Fig. 2. Characterization of rfOmpC. (A) 12% SDS–PAGE Coomassie stained Lane 1, protein marker; 2 and 3 rfOmpC; 4, tryptic digested rfOmpC; 5 and 6, column bound rfOmpC; 7 and 8, column unbound rfOmpC; 9 column unbound rfOmpC; 10, column bound rfOmpC; 11and 12 native OmpC, samples in lanes 1, 3, 5, 7, and 11 were boiled for 5 min at 100 °C. The other samples were not boiled before loading. Lanes 9 and 10 are a repeat of lanes 6 and 8 given for ease of comparison. (B) Western blot with Salmonella specific MAb Lanes 1 and 2, native OmpC; 3 and 4 rfOmpC. Alternate wells contain unboiled and boiled samples, respectively.

130


of Gram-negative bacteria in harsh environments of the intestinal tract [27]. Sensitivity to trypsin, reaction with conformation specific ligands such as monoclonal antibodies, and electrophoretic mobility on SDS–PAGE are the simple methods generally used to assess the oligomeric state of porins apart from ultracentrifugation and gel filtration. These methods were extensively used for the characterization of folding intermediates of E. coli OmpF [13] and PhoE [28,29], and to dissect their assembly process in vivo and in vitro. Refolding and reassembly of OmpF from E. coli was studied by lowangle laser light scattering photometry coupled with high performance gel chromatography [30]. Native S.

typhi OmpC, like other Gram-negative bacterial porins, is a trimer which has been earlier characterized [19] by gel chromatography and ultrafiltration using appropriate molecular weight cutoff membranes. It is stable in 2% SDS, resistant to proteolysis, and thermally stable up to 80 °C and reacts with conformation specific MAb, MPN5, raised against the Salmonella native porins [31,32]. Using these characteristic properties of native OmpC, the rfOmpC was characterized to confirm that it is equivalent to native OmpC in terms of structure and conformation. There were two populations identified in the refolded OmpC (rfOmpC): one that is in relatively minor

Fig. 3. Secondary structure and thermal stability analysis. (A) Wavelength scan was taken from 250 to 210 nm for rfOmpC B (column bound) (solid line), rfOmpC UB (column unbound) (dotted line), Native Ty21a OmpC (thin line), unfolded native OmpC (h), and unfolded rfOmpC (n). Column bound rfOmpC exhibits similar secondary structure as that of native OmpC as is seen from CD spectrum, giving a peak at 218 nm whereas the column unbound rfOmpC gives peak at 213 nm though it also exhibits overall b-structure. Unfolded samples of native OmpC and rfOmpC show altered conformation with a-helical content. B Comparison of temperature induced transition for S. typhi Ty21a OmpC and rfOmpC in the presence of non-ionic detergent C12E9. Native (smooth line) and rfOmpC (Solid line). The ratio of ½hT218 =½h25 218 plotted here, indicates the change (loss) in bstructure with the increase in temperature. ½hT218 refers to the value of [h]218 at increasing temperatures. ½h25 218 refers to the value of [h]218 at 25 °C.


proportion (20%), which is resistant to tryptic digestion and the other in major proportion (80%), which is digested by trypsin treatment (Fig. 2A, lane 4) These two populations could be separated on Q-Sepharose ion exchange column. The trypsin sensitive fraction does not bind to the column and comes in the flowthrough while the trypsin resistant fraction binds to the column and can be eluted with NaCl. The column bound rfOmpC runs equivalent to the native OmpC trimer on both native PAGE (data not shown) and SDS–PAGE (Fig. 2A, lane 6). This fraction as earlier noted was not digested by trypsin, which is a characteristic feature for the native porin trimers. The column bound rfOmpC, like the native OmpC, does not pass through 100 kDa MWCO membrane during ultrafiltration. Secondary structure analysis with CD spectroscopy shows similar spectrum as that of native OmpC (Fig. 3A). The column bound rfOmpC also shows similar fluorescence emission spectra as the native OmpC with a emission maximum at 334 nm for an excitation of 280 nm. Reaction of rfOmpC with Salmonella porins specific MAb, MPN5 (Fig. 2B, lane 3) also suggest proper refolding of OmpC. These experiments together suggest that the column bound rfOmpC is structurally equivalent to native OmpC. The column unbound population of rfOmpC runs the equivalent of trimer on native PAGE (data not shown), does not pass through 100 MWCO membrane, and reacts with the Salmonella specific MAb (data not shown), suggesting the formation of a trimeric structure (120 kDa). However, the column unbound rfOmpC is not stable in 2% SDS as seen in the SDS–PAGE sample that was not boiled (Fig. 2A, lane 8). Moreover, the fraction is also susceptible to tryptic digestion (Fig. 2A, lane 4).The column unbound rfOmpC also shows a CD spectrum different from that of the folded and the unfolded populations (Fig. 3A), and aggregates at concen-

131

trations greater than 1 mg/ml, thereby suggesting a misfolded monomer. Taken together, these observations suggest the formation of a possible metastable trimer formed by weak interaction of not completely folded monomers. The property of resistance to proteolysis can also be taken as advantage for the purification of properly folded stable trimer. When refolded OmpC was subjected to tryptic digestion, the metastable trimer was digested leaving behind the properly folded stable OmpC. Digested and aggregated products were removed by centrifugation, and further passage through 0.2 lm filter. Concentration of rfOmpC was achieved by ultrafiltration in stirred cell through 50 kDa MWCO membrane. Buffer exchange was done to remove the residual urea and trypsin from the rfOmpC. The yield by this procedure is comparable to the procedure detailed above using ion-exchange chromatography. This forms an alternate simple purification procedure consuming less time to obtain enough protein for further structural studies. Thermal stability of rfOmpC Porins exhibit high stability towards temperature and denaturants because of the barrel architecture and possible association of lipopolysaccharides. The oligomeric structure of crude preparations of MspA from M. smegmatis [33] is stable up to 120 °C. On further purification, the stability was shown to be reduced and the reason was attributed to removal of lipids during the course of purification. Thermal stability of rfOmpC was relatively less when compared to native OmpC as is seen from the temperature scan CD spectrum (Fig. 3B). Native OmpC is stable up to 80 °C whereas rfOmpC undergoes structural collapse after 60 °C itself indicating lower thermal stability due to the lack of LPS [15].

Fig. 4. Crystals of rfOmpC. Crystals grew in sitting drop method under similar conditions, where crystals were obtained earlier for native and recombinant OmpC. Buffer condition: 75 mM Tris, pH 7.5, 13% PEG 3350, 0.1 M NaCl, 0.1% C12E9, 0.8% b-OG, 5% glycerol, and 0.02% NaN3.

132


Anomalous mobility Like other porins of Gram-negative bacteria, native OmpC exhibits anomalous mobility showing multiple isoforms on SDS–PAGE. This anomalous mobility is due to the heterogeneity in the bound LPS. These isoforms could be separated by ion-exchange and gelfiltration chromatography [18]. rfOmpC, that is completely devoid of LPS, runs as a single band both in boiled and unboiled states (Fig. 2, lanes 9 and 10 in comparison with lanes 11 and 12). Heterogeneity due to the bound LPS was believed to be one of the possible reasons for poor diffraction of native S. typhi OmpC crystals [19]. rfOmpC, in the total absence of LPS, eliminates the problem of heterogeneity and possibly helps in improving the diffraction of crystals. Preliminary crystallization trials were done with rfOmpC and small crystals were obtained (Fig. 4). This further confirms the conformational homogeneity of the rfOmpC. Large-scale crystallization trials are being carried out for obtaining bigger crystals for X-ray diffraction studies. Thus, rfOmpC sets the stage for further modification of protein by loop engineering (deletions, insertions, and point mutations) that can help in understanding the structural biology of S. typhi OmpC.

Acknowledgments We acknowledge the Genetic Engineering Research Unit, Centre for Plant Molecular Biology, Bioinformatics Centre at School of Biotechnology, MKU for the use of facilities. Molecular Biophysics Unit, IISc., Bangalore, India for spectroscopic facilities. We thank UGC for support under the UGC-SAP to School of Biotechnology and Centre of Excellence in Genomics, and Proteomics program to Madurai Kamaraj University. P.D. Kumar thanks CSIR for research fellowship.

References [1] T.D. Ho, J.M. Slauch, OmpC is the receptor for Gifsy-1 and Gifsy-2 bacteriophages of Salmonella, J. Bacteriol. 183 (2001) 1495–1498. [2] M. Traurig, R. Misra, Identification of bacteriophage K20 binding regions of OmpF and lipopolysaccharide in Escherichia coli K-12, FEMS Microbiol. Lett. 181 (1999) 101–108. [3] R. Koebnik, K.P. Locher, P. Van Gelder, Structure and function of bacterial outer membrane proteins: barrels in a nutshell, Mol. Microbiol. 37 (2000) 239–257. [4] P. Massari, S. Ram, H. Macleod, Wetzler, M. Lee, The role of porins in Neisserial pathogenesis and immunity, Trends Microbiol. 11 (2001) 87–93. [5] H. Bayley, P.S. Cremer, Stochastic sensing based on nature, Nature 413 (2001) 226–230. [6] M. Niederweis, C. Heinz, K. Janik, S.H. Bossmann, Nanostructuring by deposition of protein channels formed on carbon surfaces, Nano Lett. 11 (2002) 1263–1268.

[7] H. Lang, Outer membrane proteins as surface display systems, Int. J. Med. Microbiol. 290 (2000) 579–585. [8] J.L. Puente, D. Juarez, M. Bobadilla, C.F. Arias, E. Calva, The Salmonella typhi OmpC gene: structure and uses as a carrier for heterologous sequences, Gene 156 (1995) 1–9. [9] S. Sujatha, A. Arokiasamy, S. Krishnaswamy, R. Usha, Molecular modeling of epitope presentation using membrane protein OmpC, Indian J. Biochem. Biophys. 38 (2001) 294–297. [10] B. Schmid, M. Kromer, G.E. Schulz, Expression of porin from Rhodopseudomonas blastica in Escherichia coli inclusion bodies and folding into exact native structure, FEBS Lett. 381 (1996) 111–114. [11] D. Dahan, R. Srikumar, R. Laprade, J.W. Coulton, Purification and refolding of recombinant Haemophilus influenza type b porin produced in Bacillus subtilis, FEBS Lett. 392 (1996) 304–308. [12] C. Jansen, A. Wiese, L. Reubsaet, N. Dekker, H. de Cock, U. Seydel, J. Tommassen, Biochemical and biophysical characterization of in vitro folded outer membrane protein PorA of Neisseria meningitides, Biochem. Biophys. Acta 1464 (2000) 284–298. [13] K. Sen, H. Nikaido, Trimerization of an in vitro synthesized OmpF porin of Escherichia coli outer membrane, J. Biol. Chem. 266 (1991) 11295–11300. [14] H. de Cock, M. Pasveer, J. Tommassen, E. Bouveret, Identification of phospholipids as new components that assist in the in vitro trimerization of a bacterial pore protein, Eur. J. Biochem. 268 (2001) 865–875. [15] A. Arokiasamy, P.D. Kumar, N. Sundara Baalaji, M.R. Rukmini, S. Krishnaswamy, Folding and structural stability of OmpC from Salmonella typhi: role of LPS and environment, Curr. Sci. 87 (2004) 197–202. [16] F. Goulhen, E. De, J.M. Pages, J.M. Bolla, Functional refolding of the Campylobacter jejuni MOMP (major outer membrane protein) porin by homlogous GroEL from the same species, Biochem. J. 378 (2004) 851–856. [17] S.M. Prince, C. Feron, D. Janssens, Y. Lobet, M. Achtman, B. Kusecek, P.A. Bullough, J.P. Derricka, Expression, refolding and crystallization of the OpcA invasin from Neisseria meningitides, Acta Cryst. D 57 (2001) 1164–1166. [18] A. Arockiasamy, S. Krishnaswamy, Homology model of surface antigen OmpC from Salmonella typhi and its functional implications, J. Biomol. Struct. Dyn. 18 (2000) 261–271. [19] A. Arockiasamy, S. Krishnaswamy, Purification of integral outermembrane protein OmpC, a surface antigen from Salmonella typhi for structure-function studies: a method applicable to enterobacterial major outer-membrane protein, Anal. Biochem. 283 (2000) 64–70. [20] A. Arockiasamy, S. Krishnaswamy, Crystallization of the immunodominant outer membrane protein OmpC; the first protein crystals from Salmonella typhi, a human pathogen, FEBS Lett. 453 (1999) 380–382. [21] R. Germanier, E. Furer, Isolation and characterization of Gal E mutant Ty21a of Salmonella typhi: a candidate strain for a live, oral typhoid vaccine, J. Infect. Dis. 131 (1975) 553–558. [22] J. Puente, V.A. Scherer, G. Gosset, E. Calva, Comparative analysis of Salmonella typhi and Escherichia coli ompC genes, Gene 83 (1989) 197–206. [23] K. Tsumoto, D. Ejima, I. Kumagai, T. Arakawa, Practical considerations in refolding proteins from inclusion bodies, Protein. Expr. Purif. 28 (2003) 1–8. [24] E.B. Clark, Protein refolding for industrial process, Curr. Opin. Biotechnol. 12 (2001) 202–207. [25] K. Gekko, S.N. Timasheff, Mechanism of protein stabilization by glycerol: preferential hydration in glycerol-water mixtures, Biochemistry 20 (1981) 4667–4676. [26] R. Sousa, Use of Glycerol, polyols and other protein structure stabilizing agents in protein crystallization, Acta Cryst. D 51 (1995) 271–277.

P.D. Kumar, S. Krishnaswamy / Protein Expression and Purification 40 (2005) 126–133 [27] T. Schirmer, J.P. Rosenbusch, Prokaryotic and eukaryotic porins, Curr. Opin. Struct. Biol. 1 (1991) 539–545. [28] H. de Cock, K. Brandenburg, A. Wiese, O. Holst, U. Seydel, Non-lamellar structure and negative charges of lipopolysaccharides required for efficient folding of outer membrane protein PhoE of Escherichia coli, J. Biol. Chem. 274 (1999) 5114–5119. [29] C. Jansen, M. Heutink, J. Tommassen, H. de Cock, The assembly pathway of outer membrane protein PhoE of Escherichia coli, Eur. J. Biochem. 267 (2000) 3792–3800. [30] Y. Watanabe, Characterization of the refolding and reassembly of an integral membrane protein OmpF porin by low-angle laser

133

light scattering photometry coupled with high-performance gel chromatography, J. Chromatogr. A 961 (2002) 137–146. [31] V.R. Muthukkaruppan, K.S. Nandakumar, V. Palanivel, Monoclonal antibodies against Salmonella porins: generation and characterization, Immunol. Lett. 33 (1992) 201–206. [32] A. Arokiasamy, G.S. Murthy, M.R. Rukmini, N. Sundara Baalaji, Umesh Chandra Katpally, S. Krishnaswamy, Conformational epitope mapping of OmpC, a major cell surface antigen from Salmonella typhi, J. Struct. Biol. 148 (2004) 22–33. [33] C. Heinz, H. Engelhardt, M. Niederweis, The core of tetrameric Mycobacterial porin MspA is an extremely stable b-sheet domain, J. Biol. Chem. 278 (2003) 8678–8685.