You are on page 1of 9

Protein Expression and Purication 124 (2016) 32e40

Contents lists available at ScienceDirect

Protein Expression and Purication

journal homepage:

A simplied protocol for high-yield expression and purication of

bacterial topoisomerase I
Jesse A. Jones, Emily Price, Donovan Miller, Kirk E. Hevener*
Department of Biomedical and Pharmaceutical Sciences, Idaho State University, 1311 E. Central Drive, Meridian, ID 83642-7991, USA

a r t i c l e i n f o a b s t r a c t

Article history: Type IA topoisomerases represent promising antibacterial drug targets. Data exists suggesting that the
Received 6 February 2016 two bacterial type IA topoisomerase enzymesdtopoisomerase I and topoisomerase IIIdshare an over-
Received in revised form lapping biological role. Furthermore, topoisomerase I has been shown to be essential for the survival of
21 April 2016
certain organisms lacking topoisomerase III. With this in mind, it is plausible that topoisomerase I may
Accepted 22 April 2016
represent a potential target for selective antibacterial drug development. As many reported bacterial
Available online 23 April 2016
topoisomerase I purication protocols have either suffered from relatively low yield, numerous steps, or
a simple failure to report target protein yield altogether, a high-yield and high-purity bacterial topo-
Type IA topoisomerase
isomerase I expression and purication protocol is highly desirable. The goal of this study was therefore
Topoisomerase I to optimize the expression and purication of topoisomerase I from Streptococcus mutans, a clinically
Streptococcus mutans relevant organism that plays a signicant role in oral and extra-oral infection, in order to quickly and
Purication easily attain the requisite quantities of pure target enzyme suitable for use in assay development,
Characterization compound library screening, and carrying out further structural and biochemical characterization ana-
Autoinduction lyses. Herein we report the systematic implementation and analysis of various expression and puri-
cation techniques leading to the development and optimization of a rapid and straightforward protocol
for the auto-induced expression and two-step, afnity tag purication of Streptococcus mutans topo-
isomerase I yielding >20 mg/L of enzyme at over 95% purity.
2016 Elsevier Inc. All rights reserved.

1. Introduction gastrointestinal tract and female genitourinary tract [1,2]. With

historically low consideration for pathogenicity, VGS have been
Viridans Group Streptococci (VGS) are represented by a plurality increasingly implicated in a wide array of problematic diseases
of heterogeneous streptococcal organisms that constitute a com- such as endocarditis, pneumonia, cellulitis, bacteremia, and vir-
mon component of the human microbiome found primarily within idans streptococcal shock syndrome (VGSS), among others [1,3].
the oral cavity and upper respiratory tract, as well as in the Moreover, VGS stand as the predominant group of microorganisms
within the oropharynx, with a member of the group, Streptococcus
mutans, representing the main culprit in dental cariogenesis [4].
Abbreviations: SmTopI, Streptococcus mutans topoisomerase I; VGS, Viridans
Considering the vast clinical and economic burden now known to
Group Streptococci; VGSS, Viridans Group Streptococcal Shock Syndrome; SmTo- be caused by VGS, the necessity for effective antibacterials targeting
pA15b, Streptococcus mutans topA in pET15b vector; SmTopA16b, Streptococcus these pathogens has escalated. While the majority of VGS infections
mutans topA in pET16b vector; SmTopA21d, Streptococcus mutans topA in pET21d are still susceptible to penicillin, the proliferation of drug-resistant
vector; SmTopA-SUMO, Streptococcus mutans topA in pET SUMO vector; SmTopI15b,
VGS organisms, including those resistant to penicillins, macrolides,
N-terminus hexa-Histidine-tagged Streptococcus mutans topoisomerase I enzyme;
SmTopI16b, N-terminus deca-Histidine-tagged Streptococcus mutans topoisomerase and lincosamide antibiotics, is becoming more of a concern [1,2].
I enzyme; SmTopI21d, C-terminus hexa-Histidine-tagged Streptococcus mutans Additionally, as the benecial bacteria that cohabitate the usual
topoisomerase I enzyme; SmTopI-SUMO, N-terminus hexa-Histidine-tagged and sites of VGS residence are concomitantly eradicated by the tradi-
SUMO-tagged Streptococcus mutans topoisomerase I enzyme; IMAC, immobilized tional use of these broad-spectrum antibacterials, signicant
metal-ion afnity chromatography; Ni2-NTA, nickel-nitrilotriacetic acid; IPTG,
isopropyl beta-D-1-thiogalactopyranoside; OTG, n-octyl beta-D-thioglucopyrano-
collateral consequences such as candidiasis and Clostridium difcile
side; TAE, Tris-Acetate-EDTA. infections have increased over time [5e7]. Therefore, novel targets
* Corresponding author. for new VGS-focused antibacterials are in dire need.
E-mail address: (K.E. Hevener).
1046-5928/ 2016 Elsevier Inc. All rights reserved.
J.A. Jones et al. / Protein Expression and Purication 124 (2016) 32e40 33

One such potential target lies within the topoisomerase family antibacterial agents capable of selectively targeting pathogens,
of enzymes. Topoisomerases are ubiquitous enzymes that manage drug-resistant or otherwise, while mitigating the impact on bene-
the topological complications DNA undergoes during biological cial and commensal organisms. With evidence supporting the
processes and, accordingly, are indispensable in their duties and bacterial essentiality of at least one type IA topoisomerase (topo-
represent highly attractive and efcacious chemotherapeutic tar- isomerase I or topoisomerase III), an overlapping role between the
gets. Up to four topoisomerases can be present in the prokaryotic two enzymes, and the fact that a select few bacterial species solely
genome, but only two (the type IIA topoisomerases, gyrase and possess topoisomerase I without topoisomerase III, the catalytic
topoisomerase IV) have inhibitors that have been developed and inhibition of bacterial topoisomerase I represents an attractive
marketed as antibacterial drugs. While these type IIA top- prospect for novel, selective antibacterial development [8,11e14].
oisomerases stand as well-established and well-validated anti- The streptococci are one of the few bacterial species that possess
bacterial targets via the quinolone class of topoisomerase poisons topoisomerase I as their sole type IA topoisomerase and a repre-
(which form bactericidal topoisomerase-DNA-quinolone ternary sentative of the VGS, Streptococcus mutans, exists as a particularly
complexes), the type IA topoisomerases (topoisomerase I and III) promising candidate for the exploration and validation of topo-
remain relatively unexplored and without respective approved isomerase I as a novel, selective antibacterial target [15e21].
inhibitors [8]. Current efforts are being undertaken, however, to As expected, inherent difculties are involved with the puri-
discover novel bacterial topoisomerase IA inhibitors [9]. Unlike the cation of DNA-binding proteins such as bacterial topoisomerase I.
ATP-dependent type IIA topoisomerases, which cleave both strands The initial need to separate cellular DNA from the target enzyme
of DNA to introduce strand passage and relieve or generate super- introduces an early obstacle in the protocol. To overcome such an
coiling, the type IA topoisomerases are ATP-independent and obstacle, previous methods including treating extracts to batch
cleave only one strand of DNA to produce strand passage and anion exchange, DNase treatments, and DNA precipitations using
relieve hyper-negative supercoiling or decatenate interlinked DNA different compounds like streptomycin sulfate and poly-
[8]. Also unlike type IIA topoisomerases, which are heterotetramers ethyleneimine have been employed [22]. While each DNA-
encoded by multiple genes (gyrA/B and parC/E), type IA top- separation method has its own advantages and disadvantages,
oisomerases are monomers encoded only by single genesdtopA for they all share a certain degree of costliness with respect to time and
topoisomerase I and topB for topoisomerase III (Fig. 1) [8,10]. resources. Previous studies regarding the expression and purica-
As our understanding of the human microbiome evolves, the tion of bacterial topoisomerase I have yielded varied results. While
deciencies in traditional, broad-spectrum antibacterial therapy truncated N-terminal fragments of E. coli topoisomerase I have
are becoming increasingly apparent. Both the advent of multi-drug been well studied and characterized, the full-length enzyme is
resistant bacterial infections and the devastating disturbance of our desirable for thorough characterization and inhibitor development
normal bacterial ecology testify of the alarming need for novel [23,24]. In one study, the purication and analysis of full-length
topoisomerase I from the extremophile Thermotoga maritima was
carried out (expressed in Escherichia coli with IPTG induction;
heated to precipitate non-thermophile proteins and DNA; puried
using a heparin column followed by size exclusion chromatog-
raphy) and yielded 2e3 mg/L [25,26]. However, information
gleaned from an extremophile would not necessarily be expected
to provide data that could be reliably applied to clinically relevant
bacteria. Another study detailed the purication of full-length
topoisomerase I from Streptococcus pneumonia (expressed in
E. coli with IPTG induction followed by dialysis, IMAC purication,
and a second dialysis step), but did not report yield [27]. Despite
their noteworthy contributions to our understanding of bacterial
topoisomerase I, these studies either failed to report the amount of
protein yielded from their respective protocols, or the yield was
relatively low, making biophysical and biochemical assay studies
logistically difcult. A study developing the purication of topo-
isomerase I from E. coli yielded a signicant amount of the enzyme,
but would not be appropriate for expressing and purifying topo-
isomerase I from non-E. coli bacteria as the protocol lacked a
method for separating non-native enzyme from native [28]. More
recent studies discussed the purication of full-length E. coli and
Mycobacterium tuberculosis topoisomerase I used for the subse-
quent crystallization and characterization of the respective en-
zymes, but did not report the yields [29,30].
Herein we report the expression, rapid two-step purication,
and preliminary characterization of topoisomerase I from Strepto-
coccus mutans with yields signicantly greater than previously re-
ported purications of bacterial topoisomerase I. Notably, the
following expression protocol requires signicantly less effort than
classical induction methods and the purication protocol bypasses
traditionally cumbersome DNA-removal methods and can be car-
ried out in a single day with an estimated purity of 95% or higher.
Fig. 1. S. mutans topoisomerase I structural model with single-stranded DNA bound.
Full structure of the enzyme (blue ribbons), and single strand DNA bound to active site
We also report our work to date regarding the development of
(yellow ribbons). biophysical and biochemical topoisomerase I assays.
34 J.A. Jones et al. / Protein Expression and Purication 124 (2016) 32e40

2. Material and methods 8.0, 50 mM NaCl, and 6 mM MgCl2. The assay stop buffer is
composed of 50 mM EDTA, 50% glycerol, and 0.5% w/v bromo-
2.1. Reagents, chemicals, biologicals, and equipment phenol blue.

Biological materials, equipment, and reagent-grade chemicals 2.3. Cloning and plasmid construction
used were from the following sources: Streptococcus mutans
genomic DNA, ATCC 700610D-5 from ATCC (Manassas, VA); Initially, the topA gene was cloned from Streptococcus mutans
ArcticExpress(DE3) Competent Cells, pUC18 plasmid, XL1-Blue UA159 genomic DNA for insertion into the pET-15b vector based for
Supercompetent cells, and BL21(DE3)-Gold Competent cells from the NdeI and BamHI restriction sites using the following primers:
Agilent Technologies (Santa Clara, CA); Champion pET SUMO
Protein Expression System, including One Shot Mach1-T1R and TopA_Forward: 50 -TGT AGA CAT ATG ACA AGT AAA ACA ACG ACA
BL21(DE3) One Shot Chemically Competent E. coli cells from ACA-30
invitrogen by Life Technologies (Carlsbad, CA); HyperLadder TopA_Reverse: 50 -TCA TCA GGA TCC TTA TTT AAC AGC TTT TTC
1 kb Plus from Bioline USA Inc (Taunton, MA); pET-15b vector, pET- CTT-30
16b vector, and pET-21d vector from Novagen EMD Millipore
(Billerica, MA); Quick Ligation Kit, BamHI, NdeI, and NcoI re- PCR was carried out via Platinum PCR SuperMix High Fidelity
striction enzymes from New England Biolabs, Inc. (Ipswich, MA); protocol and denatured at 94  C for 30 s, annealed at 58  C for 40 s,
Difco LB, Miller and Terric Broth from BD (Franklin Lakes, NJ); and extended at 68  C for two minutes for 30 cycles and then held
Thermo Scientic GelCode Blue Safe Protein Stain and at 4  C. Success was analyzed via 1% agarose gel electrophoresis
Spectra Multicolor High Range Protein Ladder from Thermo based upon the conrmed presence of the ~2 Kbp SmTopA_pET PCR
Fisher Scientic (Waltham, MA); HiMark Pre-stained Protein product. SmTopA_pET was then puried using QIAQuick PCR Pu-
Standard, Novex InVision His-Tag In-Gel Stain, Platinum PCR rication Kit per protocol. The pET-15b vector was then digested
SuperMix High Fidelity, NuPAGE Novex 3e8% Tris-Acetate Pro- with NdeI and BamHI restriction enzymes per New England Biolabs
tein Gels, Novex NuPAGE LDS Sample Buffer (4X), and Novex (NEB) protocol, puried, then mixed and ligated with the SmTo-
NuPAGE Tris-Acetate SDS Running Buffer and Molecular Probes pA_pET PCR product via NEB Quick Ligation Kit protocol. The
SYPRO Orange Protein Gel Stain from Invitrogen by Life Tech- ligation product (SmTopA15b) was then transformed into XL1-Blue
nologies (Carlsbad, CA); QIAGEN Plasmid Maxi Kit, QIAquick PCR Supercompetent cells per protocol. Upon successful trans-
Purication Kit, and QIAprep Spin Miniprep Kit from Qiagen formation, the plasmid was isolated with the QIAprep Spin Mini-
(Valencia, CA); DTT, glycerol, IPTG, lysozyme, NaCl, MgCl2, protease prep Kit via protocol and the sequence was conrmed via Retrogen,
inhibitor tablets (EDTA-free), Tris, Tris base, and Triton X-100 from Inc. (San Diego, CA). The plasmid was also transformed into
Fisher Scientic (Pittsburgh, PA); DNase I from Worthington BL21(DE3)-Gold Competent cells via protocol.
Biochemical Corporation (Lakewood, NJ); ampicillin, kanamycin, The topA gene was sub-cloned from the pET-15b vector in XL-1
and KH2PO4 from Acros Organics, part of Thermo Fisher Scientic Blue into the pET-16b vector using NdeI and BamHI restriction
(Waltham, MA); 30,000 MWCO Amicon Ultra-15 Centrifugal Fil- enzymes and NEB Quick Ligation Kit per manufacturer protocols.
ter Units and Imidazole from EMD Millipore (Billerica, MA); For the pET 21d vector, the SmTopA gene was amplied via PCR
K2HPO4 from Sigma Aldrich (St. Louis, MO); GelRed nucleic acid around NcoI and BamHI restriction sites using the following
gel stain from Biotium, Inc. (Hayward, CA); primers synthesized by primers:
Integrated DNA Technologies (Coralville, IA); Hampton Solubility &
Stability Screen 2 from Hampton Research (Aliso Viejo, CA). NcoI_Site_For: TAT ACG ATG GGC AGC AGC CAT CAT C

AKTA Purier FPLC, HisTrap-HP 5 mL column, and HiLoad 26/ NdeI_Site_For: GGC AGC CCC ATG GCA AGT AAA AC
600 Superdex 200 PG size exclusion column from GE Healthcare BamHI_Site_Rev: TTC GGA TCC TTT TTA ACA GCT TTT TCC TTA
Life Sciences (Pittsburgh, PA). TAG TCA CAC

2.2. Buffers PCR was carried out via Platinum PCR SuperMix High Fidelity
protocol and denatured at 94  C for 30 s, annealed at 58  C for 45 s,
The mini-expression lysis buffer is composed of 75 mL of 1% n- and extended at 68  C for two minutes and thirty seconds for 35
octyl beta-D-thioglucopyranoside (OTG) in 20 mM Tris HCl (pH 7.5), cycles and then held at 4  C.
0.1 mg Lysozyme, 0.1 mg DNase I, and 0.5 mL of 0.5 M MgCl2 for one For the pET-SUMO vector, the topA gene was cloned from PCR
cell pellet. The 2x LDS dye is composed of 5 mL Novex NuPAGE with hanging T ends per Champion pET SUMO Protein
LDS Sample Buffer (4X), 500 mM DTT, and brought to 10 mL with Expression System protocol using the following primers:
ddH2O. The standard lysis buffer is composed of 20 mM sodium
phosphate (pH 7.4), one Pierce EDTA-free protease inhibitor tablet TopA_SUMO_For: AGC ACA AGT AAA ACA ACG ACA ACA GTA
per 50 mL, 1 mM DTT, 0.5% Triton X, 0.5 M NaCl, 5% glycerol, 0.5 mg/ AAA AAG ACG
mL lysozyme, 5 mg/mL DNase, and 5 mM imidazole. Lysis buffer 2 is TopA_SUMO_Rev: TTA TTT AAC AGC TTT TTC CTT ATA GTC ACA
composed of standard lysis buffer without DNase I. Buffer A is CTC CTT ATT GC
composed of 20 mM sodium phosphate (pH 7.4), 0.5 M NaCl, 5 mM
imidazole, and 5% glycerol. Buffer B is composed of 20 mM sodium PCR was carried out via Platinum PCR SuperMix High Fidelity
phosphate (pH 7.4), 0.15 M NaCl, 250 mM imidazole, and 5% glyc- protocol and denatured at 94  C for 30 s, annealed at 60  C for 45 s,
erol. The size exclusion buffer is composed of 50 mM sodium and extended at 68  C for two minutes and thirty seconds for 35
phosphate (pH 7.4) and 150 mM NaCl. Buffer C is composed of cycles and then held at 4  C.
buffer A with 1M NaCl. Buffer D is composed of buffer B with 0.5M All gene and respective plasmids were ligated and transformed
NaCl and 350 mM imidazole. Size exclusion buffer 2 is composed of per respective protocols per respective protocols into XL1-Blue
20 mM Tris (pH 8.0), 1 mM MgCl2, 0.5M NaCl, and 1 mM DTT. The Supercompetent cells and BL21(DE3)-Gold Competent cells
gel assay buffer and assay stop buffer were prepared as previously except the SmTopA_SUMO plasmid used with the Champion pET
reported [31]. The gel assay buffer is composed of 10 mM Tris pH SUMO Protein Expression System, which utilized included One
J.A. Jones et al. / Protein Expression and Purication 124 (2016) 32e40 35

Shot Mach1-T1R and BL21(DE3) One Shot Chemically Compe- via auto-induction (SmTopI16b_AI) was puried similarly, except it
tent E. coli cells. Sequences were conrmed by sequencing via was loaded and washed with Buffer C for 40 CVs and eluted with a
Retrogen, Inc. (San Diego, CA). step gradient of Buffer D using 7CV steps at 15%, 20%, and 100% with
the sample eluting in a sharp peak in the 100% step. The sample was
2.4. Mini-expression of SmTopI15b then loaded onto a pre-equilibrated GE HiLoad 26/600 Superdex
200 PG size exclusion column and eluted with SE Buffer 2. All
A mini-expression was initially conducted in order to determine sample concentrations were attained via Bradford assay with BSA
preliminary expression conditions, starting with a 5-mL LB and as the standard.
100 mg/mL starter culture of SmTopI15b in BL21(DE3)-Gold grown
at 32  C overnight. Upon the next day, the sample was diluted 100- 2.7. Biochemical assays
fold into three 5-mL sterile TB and 100 mg/mL ampicillin samples
for each of three different temperatures: 37  C, 25  C, and 18  C. Gel assays were used to conrm target protein activity using
Cultures were then grown for two hours, then one sample from published methods [31]. The pUC18 plasmid was isolated via QIA-
each temperature was induced with 1 mM IPTG, one with 0.5 mM GEN Plasmid Maxi Kit per protocol. Enzyme assays were conducted
IPTG, and one remained un-induced. Samples were then grown an in standard volumes of 20 mL comprised of 25, 50, 75, 100, 150, 200,
additional twenty-four hours, then centrifuged at 10,000 rpm for or 400 ng SmTopI; 500 ng negatively supercoiled pUC18 plasmid;
ve minutes. The supernatant was then removed and each pellet and a sufcient quantity of gel assay buffer to bring the reaction
was dissolved in Mini_Lysis_Buffer for ve minutes. Cells were re- volume to 20 mL. One unit of commercially obtained E. coli topo-
suspended and vortexed for 30 s to completely dissolve and the isomerase I was substituted for SmTopI as a positive control. Assays
samples were centrifuged at 13,000 rpm for ve minutes. The su- were incubated at 37  C for 30 min and then terminated using 5 mL
pernatants were then separated into clean 1.5 mL microcentrifuge of assay stop buffer per reaction. DNA relaxation activity was
tubes as the soluble fractions. Pellets were washed by re- monitored on 0.7% agarose gels in Tris-Acetate-EDTA TAE buffer at
suspending in 30 mL of 1% OTG in 20 mM Tris (pH 7.5) and centri- 20 V for 10 h and then stained with GelRed nucleic acid gel stain
fuging at 13,000 rpm for ve minutes. The supernatant was then per protocol. Freeze trials were conducted similarly using 1 mg (high
discarded. Pellets were once again re-suspended in 30 mL of 1% OTG concentration) or 400 ng (low concentration) of enzyme in 50%
in 20 mM Tris (pH 7.5) and mixed with 30 mL of LDS_Sample_Buffer, glycerol or 0% glycerol and either snap-frozen in liquid nitrogen,
then centrifuged again at 13,000 rpm for ve minutes with the frozen drop-wise in liquid nitrogen and stored at 80  C overnight,
resulting supernatant containing the insoluble fraction. 10 mL or simply refrigerated. Non-specic DNAse activity assay was con-
samples were then heated at 95  C for ve minutes and loaded on ducted using published methods and incubated for 6 h [28].
SDS-PAGE gels for analysis. Mini-expression was repeated for 48-h
growths as well. (Data not shown.) Mini-expression was again 2.8. Thermal stability assay
repeated with SmTopI16b at 18  C.
Protein stability and solubility conditions were analyzed via the
2.5. Expression of SmTopI constructs and cell lysis use of Hampton Research Solubility and Stability Screen 2 per
Expression and purication experiments were conducted in
duplicate. SmTopI15b initial starter cultures were grown in 2.9. Peptide ngerprinting
BL21 cell lines at 32  C overnight in 5-mL LB with 100 mg/mL
ampicillin (pET constructs) or 50 mg/mL kanamycin (SUMO Target protein identity was conrmed by standard in-gel tryptic
construct) and used to inoculate 500 mL TB with respective anti- digest and LC-MS analysis [33e35]. Peptide spectra matching and
biotics. Cultures were grown to OD600 ~0.6, induced with 0.5 mM protein identication was achieved by database search using
IPTG, and expressed at 18  C for 24 h. Auto-induction was carried Sequest HT algorithm in Proteome Discoverer 1.4 (Thermo
out per Studier protocol with SmTopI16b and grown for 24 h at Scientic).
37  C [32]. All samples were then centrifuged at 15,000 g for 30 min
at 4  C and the supernatants were discarded. Pellets were stored 2.10. Protein homology modeling
at 80  C.
Cell pellets for all constructs were thawed at room temperature A comparative homology model of the SmTopI protein was built
and immediately lysed in the standard lysis buffer by stirring for 1 h using the E. coli (3PX7) and Thermotoga maritima (2GAI) topo-
at 4  C. Samples were then sonicated for 8 min, centrifuged at isomerase I crystal structures as templates, the Streptococcus
18K rpm at 4  C for 15 min, ltered at 0.22 mm, and collected. Auto- mutans (UA159 strain) protein sequence, and the program MOD-
induction samples were lysed in the same fashion, except lysis ELLER [36,37]. The model was rened by MD simulation using the
buffer 2 was used, and then samples were sonicated and ltered as program Amber [38].
3. Results
2.6. Purication of SmTopI constructs
3.1. Cloning and expression of recombinant SmTopI
Each initial construct sample was puried using a HisTrap-HP
5 mL column on an AKTA Purier FPLC. Samples were loaded at Considering the need to rapidly isolate SmTopI from host E. coli
1 mL/min, washed with 20 column volumes (CVs) Buffer A at topoisomerase I, cloning into a poly-histidine afnity tag system
2.5 mL/min, and eluted via linear gradient from 100% Buffer A to was chosen and implemented. The Streptococcus mutans topA gene
100% Buffer B over 20 CVs at 2.5 mL/min. Samples were then was rst cloned and inserted into the pET15b vector (SmTopA15b),
collected and concentrated using a 30,000 MWCO Amicon Ultra- introducing a hexa-Histidine (hexa-His) tag at the N-terminus of
15 Centrifugal Filter Unit. The SmTopI16b sample was collected and the target enzyme [39]. The target gene was also cloned or sub-
loaded onto a GE HiLoad 26/600 Superdex 200 PG size exclusion cloned into several alternate vectors in order to analyze the ef-
column and eluted with 1.5 CV of SE Buffer. The sample expressed fects of different constructs, including afnity tag location and size,
36 J.A. Jones et al. / Protein Expression and Purication 124 (2016) 32e40

as well as the usefulness of a protein fusion tag. BL21(DE3)-Gold the hexa-His tags (Fig. 2). Therefore, the precise point of elution of
Competent cells were used for the over-expression of target pro- SmTopI16b from the linear gradient was observed to be at 25%
tein. The target gene was also transformed into the ArcticExpress elution buffer (about 90 mM imidazole), and this aspect was used
Competent Cell system to evaluate the utility of expressing at lower to develop a step elution protocol. Upon implementation of a 15%,
temperatures (12  C) in order to increase solubility, but was not 20%, 100% step elution protocol, SmTopI16b eluted in a sharp peak
found to signicantly improve results in this instance (data not at the 100% elution step and was successfully separated from
shown) [40,41]. contaminating proteins, including E. coli catalase HPII, as deter-
Preliminary mini-expression results from the SmTopI15b mined by SDS-PAGE analysis and His-staining (Fig. 3). As such, the
construct led to a traditional IPTG induction and expression that pET16b N-term deca-His tag system was deemed the most suc-
was implemented across all constructs in duplicate. Expression cessful and carried forward for further experimentation.
products were then carried into initial immobilized metal-ion af- While analyzing mini-expression data, it was determined that
nity chromatography (IMAC) nickel column purication trials target protein solubility could likely be improved. A simple tran-
using a standard linear imidazole elution protocol. While SmTo- sition from 50 mM Tris-HCl pH 8.0e20 mM phosphate buffer pH 7.4
pI16b from the N-term deca-His construct and SmTopI-SUMO both in the lysis and nickel column buffers showed an apparent increase
signicantly outperformed the others, SmTopI16b was ultimately in yield and increasing the ionic strength of the nickel column
selected and carried forward (Table 1). Of note, as the SmTopI- binding and wash buffers (from 150 mM to 1 M NaCl) and the nickel
SUMO enzyme approached the customary 100-kDa E. coli expres- column elution buffer (from 150 mM to 500 mM NaCl) greatly
sion cutoff, the additional 11 kDa in size added by the SUMO fusion increased the overall purication yield of SmTopI16b (Table 1 and
tag showed no apparent deleterious effect on the expression in this Fig. 3). A long wash step over 40 CVs with the higher ionic strength
respective system, demonstrated by the average calculated binding buffer also resulted in satisfactory removal of DNA from the
amounts of expressed recombinant protein from the construct target protein. Absorbance at 260 nm and 280 nm was monitored,
trials resulting in 88.9 pmol for the ~83 kDa SmTopI16b and ensuring the maximal removal of DNA. As stated above, tran-
100 pmol for the ~93 kDa SmTopI-SUMO, respectively. sitioning to an auto-induction expression protocol also signicantly
As further loss of target protein was anticipated with additional improved yield and convenience.
purication steps, different conditions were also employed in order Following successful IMAC purication, SmTopI16b was identi-
to further increase target protein expression and convenience, ed upon SDS-PAGE analysis as an overexpressed band (Fig. 3a).
including the implementation of auto-induction for protein Target protein identity was conrmed initially using mass spec-
expression [32]. Upon transitioning from a classical IPTG-induction trometry peptide ngerprinting (Fig. S3), secondarily upon His-
method to a standard auto-induction method, the yield of SmTo- epitope tag antibody directed western blot (data not shown), and
pI16b increased dramatically, as did the convenience of the proto- again using His-tag staining (Fig. 3b). Protein concentrations and
col (Table 1) [32]. A subsequent IPTG mini-expression prole of yield were calculated using the standard Bradford protein assay
SmTopI16b was also conducted in order to determine favorable with bovine serum albumin (BSA) as the protein standard, which is
IPTG expression conditions, which was also compared to auto- compatible with the IMAC elution buffer [42].
induced expression and illustrated no apparent loss of solubility Following optimization of the IMAC purication step, analyses
from adopting an auto-induction method (Fig. S1). with several alternative secondary purication methods, including
the use of a heparin column, cation and anion exchange columns, a
Cibacron Blue column, and a size exclusion (SE) column were
3.2. Purication of recombinant SmTopI conducted. Ultimately, transitioning straight from IMAC to a sec-
ond, polishing size exclusion step was chosen due to simplicity and
Early exploratory nickel-nitrilotriacetic acid (Ni2-NTA) column the fact that it showed the overall greatest yield. The target protein
IMAC purication with the pET15b N-term hexa-His tag system size of ~83 kDa was judged to be distinct enough from any of the
showed the afnity of the tagged target protein for the Ni2-NTA remaining ~97 kDa native E. coli topoisomerase I to separate them
column to be markedly weak, leading to both an unacceptable loss using size exclusion chromatography. After increasing the ionic
of target protein during wash steps and very early dissociation strength of the size exclusion buffer (from 150 mM NaCl to 500 mM
upon elution. Furthermore, it was discovered upon peptide NaCl), the target protein solubility was found to be sufcient for
ngerprinting analysis that what was initially presumed to be the purication without the need for an intermediate buffer exchange
~83 kDa SmTopI15b target protein upon SDS-PAGE analysis was or dialysis step of the Ni2-NTA purication product, which again
actually the ~84 kDa native E. coli catalase HPII protein co-eluting afforded an additional gain in speed and convenience. Following
with the target protein (Fig. S2). the size exclusion chromatography polishing step, the target pro-
Comparative analysis of chromatographic linear elution proles tein was collected and again analyzed via SDS-PAGE, indicating
from the SmTopI IMAC purication trials showed a slight increase in purity deemed greater than 95%. Yield was again determined from
the retention time for the deca-His tagged target protein relative to

Table 1
Comparative summary of purication yields from SmTopI constructs.

Vector system System characteristics Growth/induction method Overall yield (mg/500 mL)a

pET-15b N-term 6X His tag 500 mL TB; 0.5 mM IPTG; 18 C  24 h 4.94 (IMAC)
pET-16b N-term 10X His tag 500 mL TB; 0.5 mM IPTG; 18  C  24 h 7.11 (IMAC)
6.24 (SE)
pET-21d C-term 6X His tag 500 mL TB; 0.5 mM IPTG; 18  C  24 h 4.11 (IMAC)
SUMO N-term SUMO fusion tag and N-term 6X His tag 500 mL TB; 0.5 mM IPTG; 18  C  24 h 9.12 (IMAC)
pET-16b N-term 10X His tag 500 mL TB; 0.5 mM IPTG; 18  C  24 h; high salt 8.30 (IMAC)
pET-16b N-term 10X His tag 500 mL auto-induction media; 37  C  24 h; high salt 14.21 (IMAC)
11.66 (SE)
Overall yields are an average of two 500 mL growths in respective media.
J.A. Jones et al. / Protein Expression and Purication 124 (2016) 32e40 37

Fig. 2. Chromatogram of nickel column linear elution from preliminary construct trials. (A) N-term hexa-His-tagged SmTopI_15b construct. (B) N-term deca-His-tagged SmTopI_16b
construct. (C) C-term hexa-His-tagged SmTopI_21d construct. (D) N-term SUMO and hexa-his-tagged SUMO construct.

Fig. 3. SDS-PAGE gels of SmTopI_16b purication results. L, ladder; CE, claried extract from auto-induction growth; FT, nickel column ow-through; P1, nickel column rst peak
from rst step elution using high salt buffers; Ni2, target fraction from nickel column; SE, target fraction from size exclusion; Ec; commercially available E. coli topoisomerase I; TB,
target fraction from nickel column from TB growth with high salt buffers. (A) SDS-PAGE after blue-staining. (B) SDS-PAGE after His-staining in order to verify target protein bands.
(For interpretation of the references to colour in this gure legend, the reader is referred to the web version of this article.)

duplicate trials by Bradford protein assay with BSA as a standard, Buffer 2 refrigerated at 4  C, as well as after overnight freeze trials
which indicated over 23 mg/L of target protein (Tables 1 and 2). at 80  C including enzyme with and without 50% glycerol using
both snap-freezing and drop-wise freezing methods in liquid ni-
trogen (Fig. S4). Protein activity after cleavage of the deca-his tag
3.3. Characterization of SmTopI was also analyzed, showing no apparent increase in activity
(Fig. S5). Furthermore, the absence of DNase activity was deter-
In order to track protein activity, a gel-based DNA-relaxation mined via the implementation of a non-specic DNase activity
assay was employed using published methods (Fig. 4) [31]. The assay using published methods (Fig. S6) [28].
activity of SmTopI16b was maintained over three months in SE
38 J.A. Jones et al. / Protein Expression and Purication 124 (2016) 32e40

Table 2
Summary of S. mutans topoisomerase I purication.

Step Total protein (mg)a Target protein (mg)b Total activity (units)c Specic activity (units/mg) Yield (%) Purity (%)d

Claried extract 1445 107.3 N/A N/A 100 7.4

Nickel column 28.4 25.3 N/A N/A 24 89
Size exclusion column 23.3 22.6 58,250 2575 21 97
Protein concentration determined by Bradford assay using BSA as the standard protein.
Amount of target protein determined based off of SDS-PAGE gel analysis.
Activity measured as described under Materials and Methods section.
Purity determined based off of SDS-PAGE gel analysis.

protein expression levels, the target was cloned into the pET SUMO
vector system (SmTopA-SUMO) in order to analyze the usefulness
of an N-term SUMO fusion protein along with an N-term hexa-His
tag [46,47].
Of the various constructs examined, the product of the deca-his-
tagged pET16b construct (SmTopI16b) resulted in enough of an
apparent increase in afnity for the nickel column to be success-
fully exploited via a step-wise elution to satisfactorily isolate the
tagged target protein from the co-eluting contaminant proteins
that plagued the hexa-his-tagged pET15b construct. Furthermore,
Fig. 4. Gel-based DNA relaxation assay. Lane 1, ladder; lane 2, supercoiled pUC18
enzyme activity was maintained with the deca-his-tagged
plasmid (DNA substrate as negative control); lane 3, commercially available E. coli
topoisomerase I (positive control); lanes 4e10, SmTopI16b enzyme in increasing construct, bypassing the immediate need for cleaving the afnity
concentration (nM). N, nicked; FR, fully relaxed; PR, partially relaxed; SC, supercoiled. tag for downstream applicationsda luxury not necessarily ex-
pected with the larger SUMO fusion tag construct (which also
showed promising results in the purication trials).
To characterize the stability of SmTopI16b, a number of methods Because the separation of DNA from DNA-binding proteins like
were employed. Hampton Research Solubility & Stability Screen topoisomerase I is essential, yet particularly onerous, a protocol
2 was used to assess conditions that mitigate target protein implementing a rapid and efcient means of addressing this
precipitation and increase its overall stability [43]. Optimal buffer obstacle was highly desired. As an alternative to more traditional
conditions as determined via Differential Scanning Fluorimetry DNA removal methods, the ionic strength of the buffers used in the
(DSF) from the screen were found to be 0.05 M sodium citrate IMAC purication step were increased and the Ni2-NTA-bound
tribasic dihydrate (pH 5.0) with 4.0 M sodium chloride as show- target protein was exposed to a high salt wash over an extended
cased by an optimal melting temperature (Tm) of 48  C (Fig. S7). An period of time in order to successfully disrupt the largely electro-
apparent trend was seen showing optimal buffer conditions static nature of DNA-protein binding [48]. Therefore, in order to
included relatively low pH and high salt content. Protein solubility further optimize SmTopI16b yield, higher ionic strength buffers
was also analyzed via overnight dialysis trials into lower-salt- were employed to both increase the solubility of the target protein
containing buffers across a wide range of pH values. Results as well as to aid in the expedited removal of DNA during purica-
showed considerable visual precipitation regardless of buffer and tion steps. Additionally, auto-induction was successfully adopted in
pH when ionic strength was decreased below 200 mM NaCl with order to increase yield and greatly decrease the effort required
protein concentration at 1 mg/mL, indicating the necessity of a during the expression and purication protocol. As mini-expression
higher ionic strength buffer for the solubility of this particular analysis showed no apparent gain or loss in target solubility by
enzyme and construct (data not shown). adopting auto-induction, it was deduced that the overall increase in
target protein yield seen between auto-induction and IPTG-
induction likely resulted from greater overall growth.
4. Discussion

An active bacterial topoisomerase I enzyme has been puried to 5. Conclusion

a relatively high yield and purity from the clinically relevant bac-
teria S. mutans. A systematic analysis of different protein constructs Using auto-induction for expression and IMAC and size exclu-
was necessary as an unacceptable amount of the prototype hexa- sion for purication, a simple and rapid protocol has been devel-
his tagged protein showed weak afnity for the nickel column oped for acquiring a substantial amount of high-purity bacterial
and was therefore lost to ow-through and overly contaminated topoisomerase I in just over two days. Via the systematic analysis
with early co-eluting proteins. In order to analyze the possibility of and implementation of different methods, the yield of SmTopI16b
the obstruction of the N-term His tag in the protein's folded state, was increased dramatically to a nal and consistent yield of over
the target gene was cloned into a pET21d vector (SmTopA21d), 20 mg/L after two steps of purication resulting in an estimated
placing a hexa-His tag at the C-terminus. As increasing the length of 95% purity or greaterda consistent yield that is both higher than
histidine tags has been shown to increase afnity to Ni2-NTA, those previously reported for bacterial topoisomerase I and one
allowing for the use of more rigorous wash steps, and generally that results from a substantially faster and less arduous purication
resulting in improved removal of non-specic binding contaminant protocol than those reported to date.
proteins, this concept was also explored by sub-cloning the target With the ease and speed of this protocol, a substantial amount of
gene from the pET15b vector into a pET16b vector, placing and bacterial topoisomerase I can be easily obtained and used in assay
extended deca-His tag at the N-terminus [44,45]. Finally, as the development and screening large libraries for further bacterial
small ubiquitin-related modier (SUMO) fusion system has been topoisomerase I characterization and drug development. With the
reported to increase both target protein solubility and target addition of an intermediate purication step, i.e., ion exchange or
J.A. Jones et al. / Protein Expression and Purication 124 (2016) 32e40 39

hydrophobic interaction chromatography, it is presumed that even antibacterial compounds, Nucleic Acids Res. 37 (3) (2009) 731e737.
[12] A. Stockum, R.G. Lloyd, C.J. Rudolph, On the viability of Escherichia coli cells
greater purity could be attained. The substantial yield of this pro-
lacking DNA topoisomerase I, BMC Microbiol. 12 (2012) 26.
tocol affords the ability to sacrice target protein yield for overall [13] I.F. Liu, J.H. Sutherland, B. Cheng, Y.C. Tse-Dinh, Topoisomerase I function
purity if necessary, for example, for protein crystallography trials. during Escherichia coli response to antibiotics and stress enhances cell killing
Overall, the optimized expression and purication protocol for from stabilization of its cleavage complex, J. Antimicrob. Chemother. 66 (7)
(2011) 1518e1524.
topoisomerase I from Streptococcus mutans here represents a rapid [14] W. Ahmed, S. Menon, A.A. Godbole, P.V. Karthik, V. Nagaraja, Conditional
and convenient tool in the continued study of bacterial type 1A silencing of topoisomerase I gene of Mycobacterium tuberculosis validates its
topoisomerases. essentiality for cell survival, FEMS Microbiol. Lett. 353 (2) (2014) 116e123.
[15] D. Ajdic, W.M. McShan, R.E. McLaughlin, G. Savic, J. Chang, M.B. Carson,
C. Primeaux, R. Tian, S. Kenton, Jia H and others. Genome sequence of
Authors' contributions Streptococcus mutans UA159, a cariogenic dental pathogen, Proc. Natl. Acad.
Sci. U. S. A. 99 (22) (2002) 14434e14439.
[16] F. Maruyama, M. Kobata, K. Kurokawa, K. Nishida, A. Sakurai, K. Nakano,
K.E.H. and J.A.J. designed all experiments and analyzed the data. R. Nomura, S. Kawabata, T. Ooshima, Nakai K and others. Comparative
K.E.H. provided laboratory space and resources. J.A.J., E.P., and D.M. genomic analyses of Streptococcus mutans provide insights into chromosomal
conducted the experimental studies. K.E.H. performed the protein shufing and species-specic content, BMC Genom. 10 (2009) 358.
[17] C. Aikawa, N. Furukawa, T. Watanabe, K. Minegishi, A. Furukawa, Y. Eishi,
comparative homology modeling. J.A.J. wrote the manuscript. K. Oshima, K. Kurokawa, M. Hattori, K. Nakano, F. Maruyama, I. Nakagawa,
T. Ooshima, Complete genome sequence of the serotype k Streptococcus
Conicts of interests mutans strain LJ23, J. Bacteriol. 194 (10) (2012) 2754e2755.
[18] P. Xu, X. Ge, L. Chen, X. Wang, Y. Dou, J.Z. Xu, J.R. Patel, V. Stone, M. Trinh,
K. Evans, T. Kitten, D. Bonchev, G.A. Buck, Genome-wide essential gene
The authors declare no conict of interest. identication in Streptococcus sanguinis, Sci. Rep. 1 (2011) 125.
[19] T.E. Molzen, P. Burghout, H.J. Bootsma, C.T. Brandt, C.E. van der Gaast-de
Jongh, M.J. Eleveld, M.M. Verbeek, N. Frimodt-Moller, C. Ostergaard,
Acknowledgements P.W. Hermans, Genome-wide identication of Streptococcus pneumoniae
genes essential for bacterial replication during experimental meningitis,
The project described was supported, in part, by a grant from the Infect. Immun. 79 (1) (2011) 288e297.
[20] J.H. Song, K.S. Ko, Detection of essential genes in Streptococcus pneumoniae
National Institute of General Medical Sciences (5 U54 GM104944).
using bioinformatics and allelic replacement mutagenesis, Methods Mol. Biol.
Additional funding was received from the ISU, Ofce of Research 416 (2008) 401e408.
and Economic Development, Faculty Seed Grant Program and the [21] J.H. Song, K.S. Ko, J.Y. Lee, J.Y. Baek, W.S. Oh, H.S. Yoon, J.Y. Jeong, J. Chun,
Identication of essential genes in Streptococcus pneumoniae by allelic
ISU, College of Pharmacy. The authors gratefully acknowledge the
replacement mutagenesis, Mol. Cells 19 (3) (2005) 365e374.
ALSAM Foundation, which provided funds for infrastructure and [22] R.R. Burgess, Use of polyethyleneimine in purication of DNA-binding pro-
laboratory renovations at ISU Meridian Health Science Center. teins, Methods Enzymol. 208 (1991) 3e10.
Portions of this research were made possible by the INBRE Program, [23] C.D. Lima, J.C. Wang, A. Mondragon, Crystallization of a 67 kDa fragment of
Escherichia coli DNA topoisomerase I, J. Mol. Biol. 232 (4) (1993) 1213e1216.
National Institutes of Health Grant NOs: P20 RR016454 (National [24] C.D. Lima, J.C. Wang, A. Mondragon, Three-dimensional structure of the 67K
Center for Research Resources) and P20 GM103408 (National N-terminal fragment of E. coli DNA topoisomerase I, Nature 367 (6459) (1994)
Institute of General Medical Sciences), and the Biomolecular 138e146.
[25] G. Hansen, A. Harrenga, B. Wieland, D. Schomburg, P. Reinemer, Crystal
Research Center at Boise State University. structure of full length topoisomerase I from Thermotoga maritima, J. Mol.
Biol. 358 (5) (2006) 1328e1340.
Appendix A. Supplementary data [26] T. Viard, V. Lamour, M. Duguet, C. Bouthier de la Tour, Hyperthermophilic
topoisomerase I from Thermotoga maritima. A very efcient enzyme that
functions independently of zinc binding, J. Biol. Chem. 276 (49) (2001)
Supplementary data related to this article can be found at http:// 46495e46503. [27] M.T. Garcia, M.A. Blazquez, M.J. Ferrandiz, M.J. Sanz, N. Silva-Martin,
J.A. Hermoso, A.G. de la Campa, New alkaloid antibiotics that target the DNA
topoisomerase I of Streptococcus pneumoniae, J. Biol. Chem. 286 (8) (2011)
References 6402e6413.
[28] X. Xu, F. Leng, A rapid procedure to purify Escherichia coli DNA topoisomerase
[1] A. Brenciani, E. Tiberi, E. Tili, M. Mingoia, C. Palmieri, P.E. Varaldo, I, Protein Expr. Purif. 77 (2) (2011) 214e219.
E. Giovanetti, Genetic determinants and elements associated with antibiotic [29] K. Tan, Q. Zhou, B. Cheng, Z. Zhang, A. Joachimiak, Y.C. Tse-Dinh, Structural
resistance in viridans group streptococci, J. Antimicrob. Chemother. 69 (5) basis for suppression of hypernegative DNA supercoiling by E. coli topo-
(2014) 1197e1204. isomerase I, Nucleic Acids Res. 43 (22) (2015) 11031e11046.
[2] C.D. Doern, C.A. Burnham, It's not easy being green: the viridans group [30] K. Tan, N. Cao, B. Cheng, A. Joachimiak, Y.C. Tse-Dinh, Insights from the
streptococci, with a focus on pediatric clinical manifestations, J. Clin. Micro- Structure of Mycobacterium tuberculosis Topoisomerase I with a Novel Pro-
biol. 48 (11) (2010) 3829e3835. tein Fold, J. Mol. Biol. 428 (1) (2016) 182e193.
[3] S.A. Shelburne, P. Sahasrabhojane, M. Saldana, H. Yao, X. Su, N. Horstmann, [31] D. Strahs, C.X. Zhu, B. Cheng, J. Chen, Y.C. Tse-Dinh, Experimental and
E. Thompson, A.R. Flores, Streptococcus mitis strains causing severe clinical computational investigations of Ser10 and Lys13 in the binding and cleavage
disease in cancer patients, Emerg. Infect. Dis. 20 (5) (2014) 762e771. of DNA substrates by Escherichia coli DNA topoisomerase I, Nucleic Acids Res.
[4] R. Sullivan, P. Santarpia, S. Lavender, E. Gittins, Z. Liu, M.H. Anderson, J. He, 34 (6) (2006) 1785e1797.
W. Shi, R. Eckert, Clinical efcacy of a specically targeted antimicrobial [32] F.W. Studier, Stable expression clones and auto-induction for protein pro-
peptide mouth rinse: targeted elimination of Streptococcus mutans and pre- duction in E. coli, Methods Mol. Biol. 1091 (2014) 17e32.
vention of demineralization, Caries Res. 45 (5) (2011) 415e428. [33] D.J. Pappin, P. Hojrup, A.J. Bleasby, Rapid identication of proteins by peptide-
[5] E.T. Stoopler, T.P. Sollecito, Oral mucosal diseases: evaluation and manage- mass ngerprinting, Curr. Biol. 3 (6) (1993) 327e332.
ment, Med. Clin. North Am. 98 (6) (2014) 1323e1352. [34] W.J. Henzel, T.M. Billeci, J.T. Stults, S.C. Wong, C. Grimley, C. Watanabe,
[6] M.A. Pfaller, D.J. Diekema, Epidemiology of invasive candidiasis: a persistent Identifying proteins from two-dimensional gels by molecular mass searching
public health problem, Clin. Microbiol. Rev. 20 (1) (2007) 133e163. of peptide fragments in protein sequence databases, Proc. Natl. Acad. Sci. U. S.
[7] A. Rineh, M.J. Kelso, F. Vatansever, G.P. Tegos, M.R. Hamblin, Clostridium A. 90 (11) (1993) 5011e5015.
difcile infection: molecular pathogenesis and novel therapeutics, Expert Rev. [35] M. Mann, P. Hojrup, P. Roepstorff, Use of mass spectrometric molecular
Anti Infect. Ther. 12 (1) (2014) 131e150. weight information to identify proteins in sequence databases, Biol. Mass
[8] Y. Pommier, Drugging topoisomerases: lessons and challenges, ACS Chem. Spectrom. 22 (6) (1993) 338e345.
Biol. 8 (1) (2013) 82e95. [36] N. Eswar, B. Webb, M.A. Marti-Renom, M.S. Madhusudhan, D. Eramian,
[9] H. Nimesh, S. Sur, D. Sinha, P. Yadav, P. Anand, P. Bajaj, J.S. Virdi, V. Tandon, M.Y. Shen, U. Pieper, A. Sali, Comparative protein structure modeling using
Synthesis and biological evaluation of novel bisbenzimidazoles as Escherichia MODELLER, Curr. Protoc. Protein Sci. (2007) (Chapter 2):2.9.1e2.9.31.
coli topoisomerase IA inhibitors and potential antibacterial agents, J. Med. [37] N. Eswar, D. Eramian, B. Webb, M.Y. Shen, A. Sali, Protein structure modeling
Chem. 57 (12) (2014) 5238e5257. with MODELLER, Methods Mol. Biol. 426 (2008) 145e159.
[10] C. Mayer, Y.L. Janin, Non-quinolone inhibitors of bacterial type IIA top- [38] D.A. Case, T.A. D, T.E. Cheatham III, C.L. Simmerling, J. Wang, R.E. Duke, R. Luo,
oisomerases: a feat of bioisosterism, Chem. Rev. 114 (4) (2014) 2313e2342. R.C. Walker, W. Zhang, K.M. Merz, B. Roberts, S. Hayik, A. Roitberg, G. Seabra,
[11] Y.C. Tse-Dinh, Bacterial topoisomerase I as a target for discovery of J. Swails, A.W. Goetz, I. Kolossvary, K.F. Wong, F. Paesani, J. Vanicek, R.M. Wolf,
40 J.A. Jones et al. / Protein Expression and Purication 124 (2016) 32e40

J. Liu, X. Wu, S.R. Brozell, T. Steinbrecher, H. Gohlke, Q. Cai, X. Ye, J. Wang, M.- studies, Anal. Biochem. 357 (2) (2006) 289e298.
J. Hsieh, G. Cui, D.R. Roe, D.H. Mathews, M.G. Seetin, R. Salomon-Ferrer, [44] M. Fischer, A.P. Leech, R.E. Hubbard, Comparative assessment of different
C. Sagui, V. Babin, T. Luchko, S. Gusarov, A. Kovalenko, P.A. Kollman, AMBER histidine-tags for immobilization of protein onto surface plasmon resonance
12, University of California, San Francisco, 2012. sensorchips, Anal. Chem. 83 (5) (2011) 1800e1807.
[39] F.W. Studier, B.A. Moffatt, Use of bacteriophage T7 RNA polymerase to direct [45] R. Grisshammer, J. Tucker, Quantitative evaluation of neurotensin receptor
selective high-level expression of cloned genes, J. Mol. Biol. 189 (1) (1986) purication by immobilized metal afnity chromatography, Protein Expr.
113e130. Purif. 11 (1) (1997) 53e60.
[40] M. Ferrer, T.N. Chernikova, M.M. Yakimov, P.N. Golyshin, K.N. Timmis, Chap- [46] J.G. Marblestone, S.C. Edavettal, Y. Lim, P. Lim, X. Zuo, T.R. Butt, Comparison of
eronins govern growth of Escherichia coli at low temperatures, Nat. Bio- SUMO fusion technology with traditional gene fusion systems: enhanced
technol. 21 (11) (2003) 1266e1267. expression and solubility with SUMO, Protein Sci. 15 (1) (2006) 182e189.
[41] M. Ferrer, H. Lunsdorf, T.N. Chernikova, M. Yakimov, K.N. Timmis, [47] L. Truong, K.E. Hevener, A.J. Rice, K. Patel, M.E. Johnson, H. Lee, High-level
P.N. Golyshin, Functional consequences of single:double ring transitions in expression, purication, and characterization of Staphylococcus aureus
chaperonins: life in the cold, Mol. Microbiol. 53 (1) (2004) 167e182. dihydroorotase (PyrC) as a cleavable His-SUMO fusion, Protein Expr. Purif. 88
[42] M.M. Bradford, A rapid and sensitive method for the quantitation of micro- (1) (2013) 98e106.
gram quantities of protein utilizing the principle of protein-dye binding, Anal. [48] O.G. Berg, C. Blomberg, Association kinetics with coupled diffusion III. Ionic-
Biochem. 72 (1976) 248e254. strength dependence of the lac repressor-operator association, Biophys.
[43] U.B. Ericsson, B.M. Hallberg, G.T. Detitta, N. Dekker, P. Nordlund, Thermouor- Chem. 8 (4) (1978) 271e280.
based high-throughput stability optimization of proteins for structural