Open Collections

UBC Faculty Research and Publications

Overexpression and purification of U24 from human herpesvirus type-6 in E. coli: unconventional use of… Tait, Andrew R; Straus, Suzana K Jun 29, 2011

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


52383-12934_2011_Article_544.pdf [ 1.97MB ]
JSON: 52383-1.0223605.json
JSON-LD: 52383-1.0223605-ld.json
RDF/XML (Pretty): 52383-1.0223605-rdf.xml
RDF/JSON: 52383-1.0223605-rdf.json
Turtle: 52383-1.0223605-turtle.txt
N-Triples: 52383-1.0223605-rdf-ntriples.txt
Original Record: 52383-1.0223605-source.json
Full Text

Full Text

RESEARCH Open AccessOverexpression and purification of U24 fromhuman herpesvirus type-6 in E. coli:unconventional use of oxidizing environmentswith a maltose binding protein-hexahistine dualtag to enhance membrane protein yieldAndrew R Tait and Suzana K Straus*AbstractBackground: Obtaining membrane proteins in sufficient quantity for biophysical study and biotechnologicalapplications has been a difficult task. Use of the maltose binding protein/hexahistidine dual tag system with E.colias an expression host is emerging as a high throughput method to enhance membrane protein yield, solubility,and purity, but fails to be effective for certain proteins. Optimizing the variables in this system to fine-tune forefficiency can ultimately be a daunting task. To identify factors critical to success in this expression system, wehave selected to study U24, a novel membrane protein from Human Herpesvirus type-6 with potentimmunosuppressive ability and a possible role in the pathogenesis of the disease multiple sclerosis.Results: We expressed full-length U24 as a C-terminal fusion to a maltose binding protein/hexahistidine tag andexamined the effects of temperature, growth medium type, cell strain type, oxidizing vs. reducing conditions andperiplasmic vs. cytoplasmic expression location. Temperature appeared to have the greatest effect on yield; at 37°Cfull-length protein was either poorly expressed (periplasm) or degraded (cytoplasm) whereas at 18°C, expressionwas improved especially in the periplasm of C41(DE3) cells and in the cytoplasm of oxidizing Δtrx/Δgor mutantstrains, Origami 2 and SHuffle. Expression of the fusion protein in these strains were estimated to be 3.2, 5.3 and4.3 times greater, respectively, compared to commonly-used BL21(DE3) cells. We found that U24 is isolated with anintramolecular disulfide bond under these conditions, and we probed whether this disulfide bond was critical tohigh yield expression of full-length protein. Expression analysis of a C21SC37S cysteine-free mutant U24demonstrated that this disulfide was not critical for full-length protein expression, but it is more likely that strainedmetabolic conditions favour factors which promote protein expression. This hypothesis is supported by the factthat use of minimal media could enhance protein production compared to nutrient-rich LB media.Conclusions: We have found optimal conditions for heterologous expression of U24 from Human Herpesvirustype-6 in E.coli and have demonstrated that milligram quantities of pure protein can be obtained. Strainedmetabolic conditions such as low temperature, minimal media and an oxidizing environment appeared essentialfor high-level, full-length protein production and this information may be useful for expressing other membraneproteins of interest.* Correspondence: sstraus@chem.ubc.caDepartment of Chemistry, University of British Columbia, 2036 Main Mall,Vancouver, BC, V6T 1Z1, CanadaTait and Straus Microbial Cell Factories 2011, 10:51© 2011 Tait and Straus; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the CreativeCommons Attribution License (, which permits unrestricted use, distribution, andreproduction in any medium, provided the original work is properly cited.BackgroundU24, a membrane glycoprotein from Human Herpes-virus Type-6A (HHV-6A), has garnered recent interestbecause a N-terminal fragment of the protein wasshown by Tejada-Simon et al. to activate T-cells [1],and cause them to cross-react with myelin basic protein,an autoantigen targeted in the pathogenesis of multiplesclerosis (MS). Sullivan et al. showed that in vivoexpression of U24 alone could downregulate CD3 T-cellreceptor and transferrin receptor cell-surface expression,and impair T-cell activation [2,3]. We have previouslydemonstrated U24 as an in vitro kinase target for ERK2MAP kinase, further implicating a potential role for U24in immune-modulating activity [1,4,5]. Using theTMMH Server Version 2 [6], open reading frame analy-sis for U24 from HHV-6A (Strain U1102) predictsexpression of an 87 amino acid, 10 kDa glycoprotein,identified to have a single transmembrane pass (residues57-79). U24 also has two cysteines (Cys21 and Cys37),whose oxidation state is unknown.Structure/function studies of membrane proteins ofbiological interest such as U24 has historically been adifficult task, primarily because of the limited amountof material available [7]. Proteins extracted fromnatural sources can also have a backdrop of post-trans-lational modifications such as glycosylation, phosphor-ylation, etc., thereby precluding any meaningfulstructural study of the heterogeneous pool of protein.It was demonstrated that U24 appeared to have exten-sive post-translational modifications if expressed inhuman cells, consistently giving molecular weights of20 and 23 kDa by SDS-PAGE [3], which is more thandouble the mass predicted by the primary sequence.Heterologous expression of U24 in a prokaryoticsystem such as E. coli can therefore represent a cost-effective and relatively easy way to obtain large yieldsof homogeneous membrane proteins where post-trans-lational modifications can be added subsequently in acontrolled manner.Protein fusion tags have a direct bearing on expressionlevels, solubility and/or efficiency of purification. Typicalfusion tags used include glutathione-S-transferase (GST)[8], maltose binding protein (MBP) [9] and hexahistidine(6 × His) tags [10], amongst many others. Since no sin-gle bacterial strain or fusion tag has been shown towork in all cases for membrane proteins, the trial-and-error task of discovering which selection will work toproduce enough pure protein for study can be quitedaunting [11]. In a high-throughput screening sense, anapproximate yield of 0.5 mg of purified membrane pro-tein per L of E.coli culture has been considered theacceptable lower expression limit for cost-effective scaleup with the intent on performing structural studies [12].A yield of 3-5 mg of purified membrane protein per Lof culture is deemed to be a high level [13].Combining the solubility-enhancing chaperone abilityof MBP [9,14] with the high affinity of a polyhistidinetag (≥ 6 consecutive histidines) for efficient proteinexpression and purification is not a recent development[15], but has seen a recent revival in the context ofhigh-throughput protein production [16]. Even so, thecombinatorial use of MBP and 6 × His tags is onlybeginning to be extensively explored with membraneproteins. For instance, Korepanova et al. demonstratedthat this system could enable a 70% success rate inexpressing 16 of 22 integral membrane proteins fromMycobacterium tuberculosis in the cytoplasm of E. coli,proteins which would otherwise be unobtainable due topoor or undetectable expression levels [17].An added benefit of using MBP as a fusion tag is thatone can generally choose the location within the E. coliwhere the expressed protein resides: either in the cyto-plasm or periplasm [16]. MBP contains a signal sequenceand thus naturally resides in the periplasm of E. coli, yet ifthe signal sequence is deleted, MBP remains in the cyto-plasm. While the most common approach is to expressfusion proteins in the cytoplasm [16], because the yield isgenerally higher than those expressed in the periplasm,there may still be valid reasons why the expression of aprotein of interest should be directed to the periplasm. Forinstance, the guinea pig sigma-1 receptor, when engi-neered as a fusion to the MBP-6 × His dual tag, couldonly be expressed as a periplasmic-directed fusion protein[18]. In this case, it was hypothesized that the receptorportion becomes stably embedded in the E. coli membraneas a result of the periplasmic transport process. Thishypothesis is consistent with what was observed for theexpression of the aquaporin Z membrane protein as aperiplasmic MBP fusion, where 200 mg/L of protein wereobtained, but was not found to be secreted out into theperiplasmic space [19].If there are cysteines in the primary sequence of aprotein and their oxidation state is unknown, as is thecase with U24, it can be beneficial to express the proteinin the oxidizing environment of the periplasm [20] oruse an oxidizing strain of E.coli [21]. It is generallyaccepted that disulfide bonds cannot be formed in thecytoplasm of E. coli unless strains such as Origami(Novagen), which are defective in reductases ((Δtrx/Δgor) are used. More recently however, it was shownthat disruption of reducing pathways in E.coli was notabsolutely essential to obtain disulfide-bonded proteinsin the cytoplasm. Introduction of the sulfhydryl oxidaseErv1p into a native reducing cytoplasm was used toachieve greater yields of a disulfide-bonded protein thanwith reductase-deficient strains alone [22,23].Tait and Straus Microbial Cell Factories 2011, 10:51 2 of 12Interestingly, MBP may help promote correctdisulfide bond formation in fused proteins. Planson etal. [24] showed that the C-terminal fragment of Plas-modium falciparum merozoite surface protein 1 wasmisfolded and formed disulfide linked polymers whenexpressed untagged in the periplasm, but was a cor-rectly folded monomer with six intramolecular disul-fides when expressed fused to MBP. This result clearlydemonstrates the potential that MBP has for introdu-cing correct disulfide bonding into globular proteins, aproperty which would be important in membrane pro-teins as well.In this study, we demonstrate how using a fusion to amaltose binding protein/hexahistidine tag can be benefi-cial for the expression of good quantities of pure U24.This work also examines the effects of temperature,growth medium type, cell strain type, oxidizing vs. redu-cing conditions and periplasmic vs. cytoplasmic expres-sion location on expression yields.ResultsCloning and ExpressionOligonucleotide fragments representing the U24 genewere synthesized and the full-length gene was obtainedby overlap PCR, before being cloned into two expressionvectors: pMAL-c2x, pMAL-p2x (New England Biolabs)(Figure 1). Vectors pMAL-c2x and pMAL-p2x direct theMBP-6 × His-U24 fusion protein to the E. coli cyto-plasm and periplasm, respectively. The commerciallyavailable E. coli cell strains that were chosen for expres-sion were XL1-Blue (Stratagene), BL21(DE3) (Strata-gene), C41(DE3) (Lucigen), Origami 2 (Novagen), andSHuffle (New England Biolabs). The C41(DE3) strain[25] is a derivative of the commonly used BL21(DE3)strain which allows high expression yields of membraneproteins by reducing their toxicity to the host [11].Origami 2 is deficient in the glutathione and thioredoxinreductases (Δtrx/Δgor) and therefore provides an oxidiz-ing environment conducive to disulfide formation. SHuf-fle is deficient in the same reductases, but expresses asignal-truncated disulfide bond isomerase (DsbC) in thecytoplasm. Normally found in the periplasm, DsbC rear-ranges incorrect disulfide bonds [26] and can also act asa chaperone to assist in the proper folding of proteinsthat do not require disulfide bonds [20].Our attempts to express MBP-6 × His-U24 in thecytoplasm of various cell lines at 37°C yielded a proteinthat was truncated (Figure 2C&Figure 2D, indicated bylower arrow). This truncated product could be easilypurified by Ni2+ affinity and gave a mass of ~45 kDa byMALDI-TOF MS (data not shown), suggesting that thisisolated form is the MBP-6 × His tag with most of theU24 portion absent. Of the numerous parameters wetested, lowering the induction temperature to 18°Cappeared to have the greatest effect on alleviating theproblem of protein truncation, yet the expression offull-length protein was still somewhat poor in the cyto-plasm of XL1-Blue, BL21(DE3) and C41(DE3) strains(Figure 2A&Figure 2B, indicated by arrow). Whenexpression of MBP-6 × His-U24 was directed to theperiplasm, low levels of full-length protein could beobserved at either 18°C or 37°C, but a substantialincrease in yield of the full-length protein was observedfor C41(DE3), especially over either the XL1 Blue orBL21(DE3) strains. Densitometric analysis suggest thatperiplasmic MBP-6 × His-U24 expression in C41(DE3)at 18°C is 3.2 × greater than in BL21(DE3) whenexpressed under identical conditions (Figure 2E). Aninteresting yet unexplained phenomenon is that cyto-plasmic expression of MBP-6 × His-U24 in C41(DE3) isactually 3 × less than BL21(DE3) at 18°C. It is unclearwhy C41(DE3) should fare any better (or worse) than itsparent strain BL21(DE3), since C41(DE3) is presumedFigure 1 U24 codon-optimized gene, amino acid sequence andgraphical representation of expressed protein construct. A)BamHI/HindIII cut sites are indicated and used to clone the PCR-amplified duplex DNA into the corresponding sites of pMAL-p2xand pMAL-c2x vectors, from which the MBP-6 × His-U24 fusionprotein is expressed. The U24 gene was designed to be precededby a hexahistidine tag (6 × His) and LVPRGS thrombin cleavage site(indicated by an arrow). Final thrombin-cleaved and purified U24protein will include an additional two amino acids (Gly-Ser) at theN-terminus. B) Cartoon representation of expressed protein. Thedifference in constructs is a signal sequence at the N-terminus ofthe protein expressed by pMAL-p2x-U24, directing expression to theperiplasm. The Factor Xa cleavage site is vector-encoded.Tait and Straus Microbial Cell Factories 2011, 10:51 3 of 12to only enhance protein production from a T7-basedpromoter [11]; here a Ptac promoter was used. We wereprompted to question whether it was specifically an oxi-dizing environment like the periplasm that had a posi-tive effect on expression levels. When MBP-6 × His-U24was expressed in the oxidizing cytoplasm of Origami 2and SHuffle, higher expression was observed in thesestrains over the reducing cytoplasm of the other strainstested. Densitometry revealed that MBP-6 × His-U24expression is an estimated 5.3 × higher in Origami 2and 4.3 × higher in SHuffle compared to BL21(DE3)(Figure 2E). Low temperature was still critical for thegeneration of full-length protein. The presence of DsbCin the SHuffle strain did not enhance the apparent pro-tein yield over Origami 2 cells. Since MBP has nocysteines, these findings suggested that the oxidationstate of the cysteines in U24 may have an effect onexpression levels. The chaperone-like qualities of theMBP moiety may help protect the U24 passenger pro-tein against proteolysis under oxidative conditions andlow temperature, enabling optimal conditions for foldingand possible disulfide-formation. An alternative factorFigure 2 Comparison of MBP-6 × His-U24 expression in various E. coli strains, at low (18°C) and high temperature (37°C*). Culturesinduced at 18°C (A&B) and 37°C* (C&D) and visualized with His-tag-specific Invision stain (A&C) and Coomassie Blue stain (B&D). Cytoplasmic andperiplasmic expressions are denoted by c and p, respectively. While at high temperature (C&D) MBP-6 × His-U24 appears degraded (c lanes) orpoorly expressed (p lanes), optimal expression conditions are observed at low temperature especially in the periplasm of C41 (DE3), andcytoplasm of Origami 2 and SHuffle. (*)SHuffle was grown at 30°C, as per vendor’s recommendations, perhaps explaining why at this lowertemperature some trace full length MBP-6 × His-U24 is observed. Position of full-length MBP-6 × His-U24 (right of the marker lane) is denotedby arrows in A&B, upper arrow in C&D (lower arrows point to truncated/degraded protein). Summary of fold-enhancement for full-length MBP-6× His-U24 expression at 18°C in the various strains/vectors is listed in E), based on densitometric analysis of bands represented A&B, andnormalized against corresponding BL21(DE3) cytoplasmic/periplasmic expression.Tait and Straus Microbial Cell Factories 2011, 10:51 4 of 12for increased yields is that degrading proteases are eitherinactive or absent under the oxidizing, low-temperatureexpression conditions that were tested.Purification and Isolation of U24MBP-6 × His-U24 expressed in the C41(DE3) periplasm,Origami 2 and SHuffle cytoplasm at low temperature wereextracted in soluble form and purified by Ni2+-affinity(Histrap) chromatography. After digestion by thrombin,U24 was liberated from MBP-6 × His and further purifiedto homogeneity by applying the digest mixture to tandemHistrap and Q-sepharose columns, and collecting U24 inthe flowthrough. U24 protein that had been purified fromthe three E. coli strains ran as a monomer under bothreducing and non-reducing conditions, indicating thatU24 does not form significant amounts of disulfide-linkedpolymers (Figure 3), although some trace amounts of whatis presumed to be a U24 dimer was observed under someconditions. The dimer disappears upon addition of redu-cing agent. Since our recombinant form of U24 runs at~10 kDa by SDS-PAGE, we conclude that we have avoidedthe unknown post-translational modifications which causeU24 to run as a doublet of 20 and 23 kDa when expressedfrom human cells. Highest yields of U24 (Table 1) werefrom C41(DE3) grown in M9 minimal media. Pryor et al.[15] suggested that use of minimal media reduces thelevels of endogenous E.coli proteases which cleave the sen-sitive region between the MBP-6 × His tag and protein ofinterest.Secondary Structure Analysis of Purified U24 by CDSpectroscopyIn order to determine whether the strains used have animpact on secondary structure of the purified protein,we determined the structure of the purified U24obtained above in the presence of membrane-mimeticSDS detergent, observed by circular dichroism (CD)(Figure 4). U24 was found to be highly a-helical underall conditions tested (Table 2). The relative proportionsof secondary structure components (a-helix. b-sheet,etc.), as determined by fitting the spectra in Figure 5using the programs CONTILL, SELCON3, and CDSSTR[27], remain highly similar in the presence and absenceof TCEP reducing agent.Disulfide AnalysisMALDI-TOF MS was used to analyze purified U24obtained from the three E.coli strains representing thehighest overexpressed proteins levels, confirming thatU24 existed in a monomeric form of expected mole-cular mass (experimental = 10,235 ± 0.2% [4]). Whenthe cysteine modifying reageant N-ethyl maleimide(NEM) was added alone, no increase in mass wasobserved. NEM cannot directly react with disulfide-bonded cysteines. If a reducing agent is added such astris(2-carboxyethyl)phosphine (TCEP), the cysteinedisulfide in U24 is broken, and then the mass shiftattributed to both cysteine thiols being modified byNEM could be observed (+250 m/z) (Figure 5A&Fig-ure 5C). These results are the same for U24 samplesobtained from all strains tested. To further confirmthe presence of the disulfide between Cys21-Cys37, asample of U24 from SHuffle was subjected to GluCprotease digestion in the presence and absence ofreducing agents, and the resulting peptide fragmentswere analyzed by MALDI-TOF MS and SDS-PAGE(Figure 5B & Figure 5C).Analysis of Cys-free U24 to Determine the Effects of theDisulfide on Protein ExpressionAfter confirming that our recombinant U24 forms anintramolecular disulfide bond, another objective was tosee if formation of this bond directly contributed to thestability of the expressed full-length MBP-6 × His-U24.When Cys-free C21SC37S mutant U24 plasmids wereexpressed under ideal periplasmic and cytoplasmic con-ditions (pMAL-p2x-U24 in C41(DE3) and pMAL-c2x-U24 in Origami 2) in LB media, we observed a similartrend in expression as compared to wild-type by SDS-PAGE (Figure 6 and 2A). Removal of the disulfide bondhad no apparent effect on the ability to produce full-length MBP-6 × His-U24, neither promoting degrada-tion at low-temperature nor enhancing yield of full-length at high temperature.Figure 3 SDS-PAGE analysis of U24 purified from C41(DE3),Origami 2 and SHuffle E.coli, in the presence and absence of b-mercaptoethanol reducing agent. Isolated U24 was fromindicated strains, cultured in LB media. Protein solution was mixed1:1 (vol:vol) with 2X SDS-PAGE buffer ± 2.5% b-mercaptoethanol(bME) final concentration, and analysis carried out by Tris-TricineSDS-PAGE. Protein amounts were approximately as follows: U24C41(DE3) , 0.1 μg; U24Origami2 and U24SHuffle, 0.4 μg. M: molecularmarkers. Arrow points to location of monomeric U24 (10.2 kDa); *indicates U24 dimer.Tait and Straus Microbial Cell Factories 2011, 10:51 5 of 12Comparison of Cellular Mass YieldsWe observed a trend in the amount of protein-expres-sing cells from large scale preparations of U24, wherepMAL-p2x-U24 is expressed in C41(DE3) and pMAL-c2x-U24 in Origami 2 (Figure 7). Despite C41(DE3)being grown in nutrient-limiting M9 media, these cellsregularly accumulated to a much higher final mass thanOrigami 2, which were grown in nutrient-rich LBmedia. In our hands, Origami 2 cells expressing MBP-6× His-U24 grew so slowly in M9 media that furtherattempts with this combination were not attempted(data not shown). It is clear that Origami 2 cells experi-ence extreme metabolic stress compared to other strainslike C41(DE3). However, with high yields of U24 proteinisolated Origami 2 (Table 1) with low background ofother endogenous E.coli proteins (Figure 2A&Figure 2B),we hypothesize that the unique metabolism of oxidizingstrains like Origami 2 may favor production of recombi-nant protein while minimizing expression of proteasesand other factors which lower the expression of full-length protein.DiscussionIn the course of expressing and isolating U24, we haveshown that membrane protein expression levels canincrease when the E.coli periplasm or an oxidizing cyto-plasm is used in conjunction with the MBP-6 × Hisdual tag, and that highly pure, soluble membrane pro-tein can be readily obtained in good yields (2-3 mg U24protein per L of culture, see Table 1). Using this as amodel system, we conclude that use of the MBP-6 × Histag, together with low temperature, and either the peri-plasm of C41(DE3) or oxidizing cytoplasm of strains(Origami 2, SHuffle, etc.) is a worthwhile approach forobtaining other difficult-to-express membrane proteinsin E. coli. Based on these results, one might expect aseveral-fold expression level increase over the com-monly-used BL21(DE3) cell strain when employingthese conditions to certain other membrane protein sys-tems. In the current work, we have also determined thatdisulfide bonding of the target protein need not be aTable 1 Expression summary and purification yields for cultures grown at 18°CVector ExpressedproteinCellularlocationC41(DE3) E.coli isolated U24(mg/L)Origami 2 E.coli isolated U24(mg/L)Shuffle E.coli isolatedU24(mg/L)pMAL-c2xMBP-6 × His-U24 Cytoplasm N/A 1.84 ± 0.31 1.60 ± 0.24pMAL-p2xMBP-6 × His-U24 Periplasm 0.46 ± 0.272.81 ± 0.32 (M9)N/A N/AResults are based on the intensity of expected bands observed by SDS-PAGE and by the BCA assay. U24 was purified as described from the indicated strainsgrown in 4 × 1 L of LB (or M9), induced at 18°C. N/A: Not applicable. Concentration is the mean ± s.d. of n = 4-6 independent assay measurements.Figure 4 Far-UV CD spectra of U24 obtained from the differentcell strains used in this study. U24 protein that had been isolatedfrom various E. coli cell strains was reconstituted in 10 mM Tris·HCl,10 mM NaCl, 10 mM SDS, pH 7.5 and far-UV CD was run in thepresence and absence of tris(2-carboxyethyl)phosphine (TCEP)reducing agent. Concentration of protein in the samples weredetermined by BCA assay to be [U24]C41(DE3) = 0.106 ± 0.004 mg/ml,[U24]Origami = 0.093 ± 0.006 mg/ml, and [U24]SHuffle = 0.118 ± 0.004mg/ml, where the standard deviation is based on four independentassay measurements.Table 2 Secondary Structure Analysis of Purified U24 byCD SpectroscopyU24 samplea Secondary Structure Fractionb NRMSDcaR aD bR bD T UC41(DE3) 0.412 0.208 0.006 0.032 0.113 0.234 0.150C41(DE3) + TCEP 0.367 0.212 0.018 0.035 0.127 0.243 0.143Origami 0.379 0.223 0.023 0.032 0.120 0.222 0.130Origami + TCEP 0.341 0.201 0.021 0.041 0.134 0.256 0.130SHuffle 0.411 0.204 0.003 0.03 0.117 0.231 0.156SHuffle + TCEP 0.401 0.212 0.013 0.032 0.112 0.239 0.147aU24 protein that had been isolated from various E.coli cell strains wasreconstituted in 10 mM Tris·HCl, 10 mM NaCl, 10 mM SDS, pH 7.5 and far-UVCD was run (260-195 nm) in the presence and absence of tris(2-carboxyethyl)phosphine (TCEP) reducing agent.bFractions of secondary structure were determined by combining the softwareanalysis results of the CONTIN/LL, SELCON3 and CDSSTR programs [27] usingthe SMP56 reference data set. Structural classes are given as: aR, regular a-helix; aD, distorted a-helix; bR, regular b-strand; bD, distorted b-strand; T, turns;U, unordered.cNormalized Root-Mean-Square Deviation. NRMSD is defined as Σ[(θexp - θcal)2/(θexp)2]1/2, summed over all wavelengths, and where θexp and θcal are,respectively, the experimental ellipticities and ellipticities of the back-calculatedspectra for the derived structure [27,35]. It has been suggested [36] thatexperimental and calculated spectra are in good agreement if NRMSD < 0.1, aresimilar if 0.1 < NMRSD < 0.2, and are in poor agreement if NRMSD > 0.2.Tait and Straus Microbial Cell Factories 2011, 10:51 6 of 12Figure 5 Characterization of the disulfide in isolated U24 protein. A) Illustration of isolated U24 protein, demonstrating the formation of asingle intramolecular disulfide bond between the only two cysteines in the molecule (i). N-ethylmaleimide (NEM) does not react with theoxidized, disulfide bonded protein (ii) and NEM only modifies cysteines in the reduced form (+125 Da per cysteine) once they are reduced bytris(2-carboxyethyl)phosphine (TCEP) (iii). Use of glutamyl endoproteinase (GluC), which cleaves peptide bonds primarily after glutamic acid underthese conditions, yielded two fragments if U24 contained an intramolecular disulfide (iv). Using dithiothreitol (DTT) to reduce the disulfide in U24(v), GluC cleavage then gave three proteolytic fragments for U24 (vi). B) Tris-Tricine SDS-PAGE results of U24 (isolated from SHuffle) ± DTT andGluC-digested. M: molecular markers; oxidized U24 ((i), 10.2 kDa) shifts to a lower mass once cleaved ((iv), 8.6 kDa) and U24 reduced by DTT ((v),10.2 kDa) shifts to even lower mass when cleaved ((vi), 6.1 kDa). C) MALDI TOF MS analysis of U24 modified with NEM ± TCEP, and cleaved byGluC ± DTT. Tabulated theoretical masses represent U24 species indicated in A) (i-vi). Experimental masses that were detected are ± 0.2% thetheoretical mass.Tait and Straus Microbial Cell Factories 2011, 10:51 7 of 12prerequisite for full-length expression of U24. Ulti-mately, we utilize the C41(DE3) in M9 media regime toproduce isotopically labeled U24 protein for NMRexperiments, while reserving the Origami 2 in LBapproach to routinely produce unlabeled U24 for allother biophysical studies.While we are capable of forming (and reducing) adisulfide in our recombinant version of U24 fromHHV-6A we must consider the fact that the cysteinesare not conserved across the U24 proteins from otherviruses, HHV-6B and HHV-7. All three versions ofU24 have the ability to sequester cellular receptors inearly and intermediate endosomes of human cells [3],yet while U24HHV-6A has two cytoplasmic cysteines,U24HHV-6B has only one and U24HHV-7 has two whichare exclusively located in the transmembrane region(and thus are not likely to form disulfide bonds). Wecannot rule out whether these differences can affectthe activity of U24 on a more precise level, or some-how are involved in regulatory mechanisms that haveyet to be discovered.Using the MBP-6 × His tagging system for a mem-brane protein of interest with a generic expression strain(i.e.: C41(DE3)), one should quickly be able to test bothcytoplasmic and periplasmic expression strategies athigh and low temperature. If there is the potential toform a disulfide bond in the membrane protein of inter-est, yet the oxidation state of the cysteines are unknown,one may benefit from using the reductase deficientSHuffle strain. Should the protein not have any disul-fides, DsbC in SHuffle may still act as a chaperone toinduce proper folding and avoid unwanted disulfidebonds. If co-overexpression of DsbC should fail toinduce native disulfide bonding patterns of an MBP-fusion protein, it may be possible to correct disulfidemispairings by an in vitro incubation of MBP-fusionswith DsbC [28], or by co-expression with the recentlycharacterized sulfhydryl-oxidase, Erv1p [23].Impaired disulfide bonding may be but one of severalkey obstacles in obtaining stable full-length membraneproteins in E. coli, especially those from mammaliansources. Polypeptide synthesis is 4-10 times faster inprokaryotes than eukaryotes, and the apparatus bywhich membrane proteins are inserted into the mem-brane is different [29]. One approach may be to reducethe rate of synthesis by lowering expression tempera-ture, or to use a weaker promoter. Although MBPappears superior to most other fusion tags in the litera-ture, improvements in yield can result from using otherunique tags [30]. Autoexpression media [17] or novelenzyme-based-substrate delivery media (EnBase, [31])also represents an attractive approach for increasingboth cell mass and protein yield. Still, other advancesinvolve the reduction in toxicity associated with mem-brane protein overexpression [11,25]. While these andother trial-and-error optimization techniques may give amodest improvement in membrane protein yield,increasing our mechanistic understanding of prokaryoticsystems of membrane protein translation and folding isclearly warranted. Directed evolution studies [32] andgenetic analysis of upregulation, downregulation andmutations [33] made to house-keeping genes [34] duringmembrane protein expression will undoubtedly continueto have an impact on yields of functional membraneproteins in E. coli.Figure 6 Examination of cysteine-free mutant U24 expression.The C21SC37S mutant U24 constructs were expressed: pMAL-p2x-U24 in C41 (DE3) and pMAL-c2x-U24 in Origami 2 cells, at 18°C and37°C. Removal of disulfide bond potential appeared to have noeffect on in vivo stability of expressed MBP-6 × His-U24, whichexhibited the same expression characteristics as wild-type; a similarloss in mass of the degraded fusion protein is observed at thehigher temperature.Figure 7 Mass of isolated cells expressing MBP-6 × His-U24.Cell masses which correspond to highest yields of protein arenotably different between the cell types; although grown in rich LBmedia, Origami 2 cells (pMAL-c2x-U24) are consistently ~3 × lowerin mass than C41(DE3) (pMAL-p2x-U24), which is grown in M9minimal media. Data is based on mean ± s.d. for n = 4-5 batches ofcells isolated from 4 × 1L of culture, expressed at 18°C overnight.Tait and Straus Microbial Cell Factories 2011, 10:51 8 of 12MethodsGene constructionThe gene sequence for U24 from HHV-6A wasobtained from NCBI [GenBank: Q69559]. Oligonucleo-tide fragments that represent the full-length U24 geneand primers were synthesized (Nucleic Acid ProteinService Unit, University of British Columbia), replacingcodons which are rare in E. coli with ones more fre-quently used. The codon optimized U24 gene wasassembled by overlap PCR for insertion into thepMAL-c2x and pMAL-p2x vectors (New England Bio-labs). For use with the pMAL-c2x and pMAL-p2x vec-tors, the amplified gene segment included a BamHIsite at the 5’ end, followed by a sequence coding ahexahistidine tag followed by a thrombin cleavage site(LVPRGS), then the U24 sequence, followed by anamber stop codon then a HindIII site at the 3’ end.DNA fragments were purified by agarose gel electro-phoresis and subsequently extracted using a QiaquickGel Extraction Kit (Qiagen).CloningThe isolated gene fragments described above were incu-bated with Taq polymerase, pCR®2.1-TOPO® vector anddATP (all from Invitrogen) and the reaction mixtureswere used to transform competent SURE® E. coli (Strata-gene). The E. coli were then plated on Luria Bertani (LB)agar plates containing 50 μg/ml carbenecillin to growovernight at 37°C. Colonies were selected and subjected toPCR using M13 forward and reverse primers to identifywhich colonies contained the vector with the correctinserted gene. PCR reactions for colonies which did harborthe gene insert yielded a single band between 500-600 bpwhen run by agarose gel electrophoresis. These colonieswere then grown in 5 ml of LB media with 50 μg/ml car-benecillin overnight and their plasmids were harvestedusing a QIAprep Spin Miniprep Kit (Qiagen). Isolatedplasmid DNA was sent for sequencing (Nucleic Acid Pro-tein Service Unit, University of British Columbia) to con-firm that the correct gene sequences were obtained.The pCR®2.1-TOPO® vectors containing the expectedU24 gene inserts were digested by the appropriate endo-nucleases, HindIII/BamHI. The digested fragments wereisolated as previously described, then ligated into thecorresponding pre-digested gel-purified pMAL vectorsusing T4 DNA ligase (Invitrogen). SURE® E. coli weretransformed with the ligation products, the plasmidswere isolated and sequenced as before, and designatedpMAL-c2x-U24 and pMAL-p2x-U24.Site-directed MutagenesisSynthetic primers were purchased (Integrated DNATechnologies) and used according to Quikchange SiteDirected Mutagenesis Kit instructions (Stratagene).Vectors pMAL-c2x-U24 and pMAL-p2x-U24 were usedas the starting template. To construct the C21S mutant,the primers were 5’-CTGATGATGGACGTTATGTCCGGTCAGGTTTCC-3’ and 5’-GGAAACCT-GACCGGACATAACGTCCATCATCAG-3’, whichintroduces an Eam1105I-sensitive restriction site. For theC37S mutant, the primers used were 5’-CCTTCGTTGAATCTATTCCGCCACCG-3’and 5’-GGACTGCGGTGGCGGAATAGATTCA-3’ and they introduced aXmnI-sensitive restriction site. The C21S mutant plasmidwas the starting template for constructing the C21SC37Sdouble mutant. All plasmids were isolated and sequencedas described above for wild-type plasmids.Small Scale Protein ExpressionThe expression vector constructs pMAL-c2x-U24 andpMAL-p2x-U24 were used to transform XL1-Blue (Stra-tagene), BL21(DE3) (Stratagene), C41(DE3) (Lugicen),Origami 2 (EMD Biosciences) and SHuffle (New Eng-land Biolabs) E. coli strains. Transformed E. coli werethen plated on LB agar plates containing 50 μg/ml car-benecillin to grow overnight at 37°C (30°C in the case ofSHuffle, as per manufacturer’s instructions). Single colo-nies were selected and grown at these temperaturesovernight with shaking (225 r.p.m.) in 5 ml of LB, con-taining 50 μg/ml carbenecillin. Cells were pelleted bycentrifugation, then resuspended in 100 ml of freshmedium with antibiotics and again grown for severalhours until reaching an OD600 = 0.5-1.0. Isopropyl b-D-1-thiogalactopyranoside (IPTG) was added at a finalconcentration of 0.3 mM and the cells continued togrow for 3 hours at the same temperature (30 or 37°C).For growths carried out at lower temperature, the cul-tures were cooled in a cold-water bath for 30 minutesbefore addition of the IPTG, and these cultures werethen grown at 18°C for 16-20 hours. All cultures wereharvested by centrifugation at 4°C and the cells werestored at -80°C until further use.Large Scale Protein ExpressionSmall starter cultures (5 ml) were set up as described forthe small scale expression. In this case, the cultureswere grown for only three to five hours before one mlof culture was used to inoculate 100 ml of fresh LB orM9 minimal media containing 100 μg/ml ampicillin. M9minimal media was supplemented with 50 μg/ml thia-mine. These cultures were grown overnight under thesame conditions as the 5 ml cultures described in thesection above, and then harvested by centrifugation. Cellpellets were resuspended in 4 × 1 L of fresh mediumwith antibiotics and grown to an OD600 = 0.5-1.0. Cul-tures were cooled in a cold-water bath for 30 minutesbefore addition of IPTG at a final concentration of 0.3mM, and they then continued to grow at 18°C forTait and Straus Microbial Cell Factories 2011, 10:51 9 of 12another 16-20 hours. Cultures were harvested by centri-fugation at 4°C and the cells were stored at -80°C untilfurther use.Protein Extraction and PurificationC41(DE3) cell pellets from 4 L of culture, harboringexpressed MBP-6 × His-U24, were thawed on ice andresuspended in lysis buffer (20 mM Na+/K+ phosphate,0.5 M NaCl, pH 7.4), adding DNase I (Roche) andEDTA-free protease inhibitor cocktail (Sigma). Cellswere lysed by three passes through a French press, andthe lysate was centrifuged for 90 minutes at 25,000 ×g, 4°C. The pellet was resuspended in solubilizationbuffer (20 mM Na+/K+ phosphate, 0.5 M NaCl, 1%Triton X-100, 10 mM imidazole, pH 7.4), gently stirredfor three hours at 4°C, and again centrifuged for 90minutes at 25,000 × g, 4°C. Since substantial amountof target protein was found in both the supernatantand pellet of Origami 2 and SHuffle cells treated withlysis buffer, the lysis buffer step was omitted and thesecells were resuspended directly into solubilization buf-fer with protease inhibitor cocktail prior to being lysedby French press. Supernatants were filtered through a45 μm filter and loaded on a 5 ml Histrap HP column(GE Biosciences) equilibrated with wash buffer (20mM Na+/K+ phosphate, 0.5 M NaCl, 0.5% Triton X-100, 10 mM imidazole, pH 7.4). The column waswashed with 30 column volumes of wash buffer. Theimidazole concentration of the wash buffer was raisedto 45 mM, and the column was further washed withanother 10 column volumes. The MBP-6 × His-U24protein was eluted from the column in 25 ml of elu-tion buffer (20 mM Na+/K+ phosphate, 0.5 M NaCl,0.5% Triton X-100, 500 mM imidazole, pH 7.4) anddialyzed overnight at 4°C against 1 L of dialysis buffer(20 mM Na+/K+ phosphate, 62.5 mM NaCl, 0.5% Tri-ton X-100) using a dialysis membrane with MWCO of1000 (Spectrum Laboratories).Dialyzed solution containing the MBP-6 × His-U24was transferred to a 50 ml polypropylene tube, and 50units of bovine thrombin (GE Healthcare) were added.The digest was carried out for 48 hours at ambienttemperature.The digest solution was filtered through a 45 μm fil-ter and re-loaded on a 5 ml Histrap HP column equili-brated with dialysis buffer. U24 was collected in theflowthrough, while MBP-6 × His and any undigestedMBP-6 × His-U24 were eluted with elution buffer. Toremove the thrombin and any trace contaminants thatremained, U24 solution was loaded on a 5 ml Q-Sepharose column (GE Biosciences) equilibrated withdialysis buffer and pure U24 was collected in theflowthrough. Purity was assessed by MALDI-TOFmass spectrometry and SDS-PAGE, and proteinconcentration was measured with the bicinchoninicacid (BCA) assay (Pierce).GluC DigestsU24 protein isolated from SHuffle E.coli was dialyzedovernight at 4°C against 50 mM Tris·HCl, 100 mMNaCl, pH 7.5 with a 1000 MWCO membrane. Staphylo-coccus aureus Protease V8 (GluC, from New EnglandBiolabs) was dissolved in deionized water to a concen-tration of 0.1 μg/ml then mixed with U24 (1:17 w/w), ±20 mM DTT, and incubated for two hours at 25°C.Reactions were quenched by addition of trichloroaceticacid (TCA) at a final concentration of 20% (w/v). Pelletswere collected by centrifugation and washed twice withice-cold acetone then air-dried.Far-UV Circular DichroismProtein stock solutions of U24 were precipitated byaddition of TCA, 10% (w/v), and collected by centrifuga-tion. The protein pellets were then washed with twoadditions of ice-cold acetone, vortexing and centrifuga-tion. After acetone was removed and the sample air-dried, the protein was reconstituted in 10 mM Tris·HCl,10 mM NaCl, 10 mM SDS, pH 7.5, +/- 0.5 mM tris(2-carboxyethyl)phosphine (TCEP) with an estimated finalprotein concentration of 0.11 mg/ml. Protein solutionswere sonicated in a water bath until the protein wascompletely dissolved, then centrifuged briefly to ensurethe removal of any particulate matter. Final protein con-centrations of the samples were determined by BCAassay.Spectra were recorded with a J-810 spectropolarimeterflushed with nitrogen gas. Using a cell with a pathlength of 0.2 cm and a sample compartment that wasthermostated to 20°C, samples were scanned at a rate of50 nm/min with a step size of 1 nm. Spectra were aver-aged over three scans and corrected for background bysubtracting the scans of buffer without protein.SDS-PAGE Analysis of Protein ExpressionCell pellets from 0.5 ml of culture were mixed with 50μl 2 × Novex dye (Invitrogen) supplemented with 5%b-mercaptoethanol and 50 μl of 50% glycerol, vortexedand heated at 95°C for 5 minutes. BenchMark His-tagged molecular weight markers (Invitrogen) and a 5μl volume of each prepared sample were loaded on a13% acrylamide gel and run by Tris-Glycine SDS-PAGE. The initial voltage was set to 50 V for onehour, and 100 V for the remainder of the run until thedye front reached the bottom of the gel. Gels werestained with a Coomassie Blue-G250 solution for 1hour and destained with 50% methanol/10% acetic acidor deionized water alone. For Invision staining, thegels were treated according to manufacturer ’sTait and Straus Microbial Cell Factories 2011, 10:51 10 of 12instructions (Invitrogen). SDS-PAGE image data wascaptured using a GelDoc XR imager (Biorad) and den-sitometric analysis of protein expression was carriedout using Quantity One software, V4.6.7 (Biorad). Thecontour tool and global background subtraction meth-ods were used to select bands and calculate their rela-tive intensities.Purified U24 samples were mixed with equal volumesof 2 × Novex dye that contained 5% b-mercaptoethanoland then were heated to 95°C for 5 minutes. Samplesused in the GluC digest experiment were not heatedprior to loading on the gel, and only those that werepreviously exposed to DTT contained 2.5% b-mercap-toethanol. Gels were run, stained and destained asbefore, with the exception that a Tris-Tricine gel bufferwas used in order to resolve the lower molecular weightproteins and peptides. In these gels, Mark12 molecularweight markers (Invitrogen) were used to estimate pro-tein masses.NEM modificationU24 protein isolated from C41(DE3), Origami 2 andSHuffle E. coli strains were dialyzed overnight at 4°Cagainst 50 mM Tris·HCl, 100 mM NaCl, pH 7.5 with a1000 MWCO membrane. N-ethylmaleimide (NEM) wasdissolved in the same buffer and diluted to 4 mM.TCEP was dissolved separately in this buffer to a con-centration of 20 mM and the pH was made slightlybasic by addition of NaOH before diluting further to aTCEP concentration of 2 mM. Samples of U24 weremixed 1:1 (vol:vol) with TCEP solution (+ TCEP sam-ples) or with buffer (- TCEP samples) and incubated for10 minutes at 37°C. These samples were then mixed 1:1(vol:vol) with NEM solution (+ NEM samples) or buffer(- NEM samples) and incubated in the dark for 30 min-utes at room temperature. Reactions were quenched byaddition of trichloroacetic acid (TCA) to a final concen-tration of 20% (w/v). Pellets were collected by centrifu-gation and washed twice with ice-cold acetone, and thenair-dried.MALDI-TOF Mass SpectrometryProtein samples were dissolved in 50% acetonitrile with0.1% trifluoroacetic acid. The matrix, sinapinic acid, wasdissolved in the same solvent to a concentration of 10mg/ml. The matrix was spotted on the MALDI targetplate followed by the protein solution and another layerof matrix. Volumes added were 1 μL, and the spot wasair-dried between each addition. Experiments were per-formed on Bruker Biflex IV (Bruker Daltonics) MALDI-TOF mass spectrometer, operating in linear ion modeand externally calibrated with horse heart cytochrome Cand bovine ubiquitin (Sigma), giving a mass accuracy of± 0.2%.AcknowledgementsThe authors would like to acknowledge funding from NSERC (Discoverygrant) and the Michael Smith Foundation for Health Research (CareerInvestigator Scholar).Authors’ contributionsART performed all experiments reported herein, as well as the data analysis.He wrote the first draft of the manuscript. SKS edited the manuscript. Bothauthors read and approved the final manuscript.Competing interestsThe authors declare that they have no competing interests.Received: 21 March 2011 Accepted: 29 June 2011Published: 29 June 2011References1. Tejada-Simon MV, Zang YC, Hong J, Rivera VM, Zhang JZ: Cross-reactivitywith myelin basic protein and human herpesvirus-6 in multiple sclerosis.Ann Neurol 2003, 53:189-197.2. Sullivan BM, Coscoy L: Downregulation of the T-cell receptor complexand impairment of T-cell activation by human herpesvirus 6 u24protein. J Virol 2008, 82:602-608.3. Sullivan BM, Coscoy L: The U24 protein from human herpesvirus 6 and 7affects endocytic recycling. J Virol 2010, 84:1265-1275.4. Tait AR, Straus SK: Phosphorylation of U24 from Human Herpes Virustype 6 (HHV-6) and its potential role in mimicking myelin basic protein(MBP) in multiple sclerosis. FEBS Lett 2008, 582:2685-2688.5. Japelj B, Zorko M, Majerle A, Pristovsek P, Sanchez-Gomez S, Martinez deTejada G, Moriyon I, Blondelle SE, Brandenburg K, Andra J, et al: The acylgroup as the central element of the structural organization ofantimicrobial lipopeptide. J Am Chem Soc 2007, 129:1022-1023.6. Sonnhammer EL, von Heijne G, Krogh A: A hidden Markov model forpredicting transmembrane helices in protein sequences. Proc Int ConfIntell Syst Mol Biol 1998, 6:175-182.7. Midgett CR, Madden DR: Breaking the bottleneck: eukaryotic membraneprotein expression for high-resolution structural studies. J Struct Biol2007, 160:265-274.8. Smith DB, Johnson KS: Single-step purification of polypeptides expressedin Escherichia coli as fusions with glutathione S-transferase. Gene 1988,67:31-40.9. Maina CV, Riggs PD, Grandea AG, Slatko BE, Moran LS, Tagliamonte JA,McReynolds LA, Guan CD: An Escherichia coli vector to express andpurify foreign proteins by fusion to and separation from maltose-binding protein. Gene 1988, 74:365-373.10. Janknecht R, de Martynoff G, Lou J, Hipskind RA, Nordheim A,Stunnenberg HG: Rapid and efficient purification of native histidine-tagged protein expressed by recombinant vaccinia virus. Proc Natl AcadSci USA 1991, 88:8972-8976.11. Pryka RD, Novak RM, Wagner DK, Rodvold KA: Clinical pharmacokinetics ofdaptomycin. Dicp 1990, 24:255-256.12. Mancia F, Love J: High-throughput expression and purification ofmembrane proteins. Journal of Structural Biology 2010, 172:85-93.13. Eshaghi S, Hedren M, Nasser MIA, Hammarberg T, Thornell A, Nordlund P:An efficient strategy for high-throughput expression screening ofrecombinant integral membrane proteins. Protein Science 2005,14:676-683.14. Kapust RB, Waugh DS: Escherichia coli maltose-binding protein isuncommonly effective at promoting the solubility of polypeptides towhich it is fused. Protein Sci 1999, 8:1668-1674.15. Pryor KD, Leiting B: High-level expression of soluble protein inEscherichia coli using a His6-tag and maltose-binding-protein double-affinity fusion system. Protein Expr Purif 1997, 10:309-319.16. Nallamsetty S, Austin BP, Penrose KJ, Waugh DS: Gateway vectors for theproduction of combinatorially-tagged His6-MBP fusion proteins in thecytoplasm and periplasm of Escherichia coli. Protein Sci 2005,14:2964-2971.17. Korepanova A, Moore JD, Nguyen HB, Hua Y, Cross TA, Gao F: Expressionof membrane proteins from Mycobacterium tuberculosis in Escherichiacoli as fusions with maltose binding protein. Protein Expr Purif 2007,53:24-30.Tait and Straus Microbial Cell Factories 2011, 10:51 11 of 1218. Ramachandran S, Lu H, Prabhu U, Ruoho AE: Purification andcharacterization of the guinea pig sigma-1 receptor functionallyexpressed in Escherichia coli. Protein Expr Purif 2007, 51:283-292.19. Lian J, Ding S, Cai J, Zhang D, Xu Z, Wang X: Improving aquaporin Zexpression in Escherichia coli by fusion partners and subsequentcondition optimization. Appl Microbiol Biotechnol 2009, 82:463-470.20. Kadokura H, Katzen F, Beckwith J: Protein disulfide bond formation inprokaryotes. Annu Rev Biochem 2003, 72:111-135.21. Bessette PH, Aslund F, Beckwith J, Georgiou G: Efficient folding of proteinswith multiple disulfide bonds in the Escherichia coli cytoplasm. Proc NatlAcad Sci USA 1999, 96:13703-13708.22. Hatahet F, Nguyen VD, Salo KE, Ruddock LW: Disruption of reducingpathways is not essential for efficient disulfide bond formation in thecytoplasm of E. coli. Microb Cell Fact 2010, 9:67.23. Nguyen VD, Hatahet F, Salo KE, Enlund E, Zhang C, Ruddock LW: Pre-expression of a sulfhydryl oxidase significantly increases the yields ofeukaryotic disulfide bond containing proteins expressed in thecytoplasm of E.coli. Microb Cell Fact 2011, 10:1.24. Planson AG, Guijarro JI, Goldberg ME, Chaffotte AF: Assistance of maltosebinding protein to the in vivo folding of the disulfide-rich C-terminalfragment from Plasmodium falciparum merozoite surface protein 1expressed in Escherichia coli. Biochemistry 2003, 42:13202-13211.25. Miroux B, Walker JE: Over-production of proteins in Escherichia coli:mutant hosts that allow synthesis of some membrane proteins andglobular proteins at high levels. J Mol Biol 1996, 260:289-298.26. Maskos K, Huber-Wunderlich M, Glockshuber R: DsbA and DsbC-catalyzedoxidative folding of proteins with complex disulfide bridge patterns invitro and in vivo. J Mol Biol 2003, 325:495-513.27. Sreerama N, Woody RW: On the analysis of membrane protein circulardichroism spectra. Protein Sci 2004, 13:100-112.28. Pioszak AA, Xu HE: Molecular recognition of parathyroid hormone by itsG protein-coupled receptor. Proc Natl Acad Sci USA 2008, 105:5034-5039.29. Tate CG: Overexpression of mammalian integral membrane proteins forstructural studies. FEBS Lett 2001, 504:94-98.30. Roosild TP, Greenwald J, Vega M, Castronovo S, Riek R, Choe S: NMRstructure of Mistic, a membrane-integrating protein for membraneprotein expression. Science 2005, 307:1317-1321.31. Panula-Perala J, Siurkus J, Vasala A, Wilmanowski R, Casteleijn MG,Neubauer P: Enzyme controlled glucose auto-delivery for high celldensity cultivations in microplates and shake flasks. Microb Cell Fact 2008,7:31.32. Massey-Gendel E, Zhao A, Boulting G, Kim HY, Balamotis MA, Seligman LM,Nakamoto RK, Bowie JU: Genetic selection system for improvingrecombinant membrane protein expression in E. coli. Protein Sci 2009,18:372-383.33. Skretas G, Georgiou G: Genetic analysis of G protein-coupled receptorexpression in Escherichia coli: inhibitory role of DnaJ on the membraneintegration of the human central cannabinoid receptor. Biotechnol Bioeng2009, 102:357-367.34. Link AJ, Skretas G, Strauch EM, Chari NS, Georgiou G: Efficient productionof membrane-integrated and detergent-soluble G protein-coupledreceptors in Escherichia coli. Protein Sci 2008, 17:1857-1863.35. Wallace BA, Lees JG, Orry AJ, Lobley A, Janes RW: Analyses of circulardichroism spectra of membrane proteins. Protein Sci 2003, 12:875-884.36. Khrapunov S, Cheng H, Hegde S, Blanchard J, Brenowitz M: Solutionstructure and refolding of the Mycobacterium tuberculosis pentapeptiderepeat protein MfpA. J Biol Chem 2008, 283:36290-36299.doi:10.1186/1475-2859-10-51Cite this article as: Tait and Straus: Overexpression and purification ofU24 from human herpesvirus type-6 in E. coli: unconventional use ofoxidizing environments with a maltose binding protein-hexahistinedual tag to enhance membrane protein yield. Microbial Cell Factories2011 10:51.Submit your next manuscript to BioMed Centraland take full advantage of: • Convenient online submission• Thorough peer review• No space constraints or color figure charges• Immediate publication on acceptance• Inclusion in PubMed, CAS, Scopus and Google Scholar• Research which is freely available for redistributionSubmit your manuscript at and Straus Microbial Cell Factories 2011, 10:51 12 of 12


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items