Use of IPTG inducible promoters for anchoring recombinant proteins on the Bacillus subtilis spore surface

Use of IPTG inducible promoters for anchoring recombinant proteins on the Bacillus subtilis spore surface tài liệu, giáo...

Use of IPTG-inducible promoters for anchoring recombinant proteins

on the Bacillus subtilis spore surface

a

Institute of Genetics, University of Bayreuth, D-95440 Bayreuth, Germany

b

Department of Molecular and Environmental Biotechnology, Faculty of Biology, University of Science, Vietnam National University, HCMC, Viet Nam

a r t i c l e i n f o

Article history:

Received 6 September 2013

and in revised form 24 November 2013

Available online 8 December 2013

Keywords:

IPTG

CotB

CotC

CotG

a b s t r a c t

The method of surface display allows the fusion of passenger proteins to a carrier protein displayed on the outside of bioparticles such as spores Here, we used spores of Bacillus subtilis, the outer surface pro-teins CotB, CotC, and CotG as carrier and the amyQ-encodeda-amylase and GFPuv as passenger proteins The different translational fusions were fused to two different IPTG-inducible promoters, and the regu-lated expression level of both passenger proteins were measured in relation to the inducer concentration added to sporulating cells It turned out that the amount of fusion protein on the outside of spores was dependent on the amount of IPTG added, but the optimal amount of inducer varied depending on the car-rier and the passenger proteins These experiments demonstrate that a regulatable expression of passen-ger proteins on the surface of spores is possible This will help to adjust the amount of any passenpassen-ger protein to that needed for specific purposes

Ó 2013 Elsevier Inc All rights reserved

Introduction

Surface display is a powerful technique allowing the

expres-sion of peptides and proteins on the outside of phages, bacterial

and eukaryotic cells and on endospores collectively designated

as bioparticles These peptides and proteins called passenger

pro-teins are translationally fused to a carrier protein leading to the

synthesis of a hybrid protein, which is located on the outside of

these bioparticles The passenger proteins displayed on the

bio-particle surface are freely accessible to substrates or binding

partners in activity studies There are several excellent review

articles dealing with the application of surface display using the

different types of bioparticles[1–4] Surface display has various

applications such as in biocatalyst and bioadsorbent reactions

[5,6], vaccine production [7], peptide screening[8]and the

pro-duction of antibodies[9]

When cells and filamentous phages are used for surface display,

the hybrid proteins have to cross at least one membrane

Further-more, correct folding of the passenger proteins cannot be aided by

the general folder chaperones such as DnaK and GroEL These

dis-advantages are not present when endospores are used for surface

display Endospores or spores are produced intracellular by the

groups of Bacillus and Clostridia, and anchoring of the passenger

proteins occurs within the cytoplasm, and these proteins do not

have to cross any membrane Furthermore, the molecular folder chaperones present within the cytoplasm can assist correct folding

of the recombinant proteins[10,11] Spores are tough, non-productive and highly resistant struc-tures which allow the organism to survive a wide range of ex-treme stresses, harsh conditions such as starvation or terrestrial environments and are stable for millions of years [12,13] The Bacillus spore consists of three main concentric compartments: the core, the cortex and the coat The single bacterial chromo-some, coated by low-molecular-weight proteins, is condensed within the core compartment The core is surrounded by the spore cortex, which consists of a modified form of peptidoglycan [14] The spore coat is assembled around the cortex, playing the most important role in the spore’s resistance to organic solvents and lysozyme In thin sections, the coat appears as a series of con-centric layers which are divided into two major layers: a thick, highly electron-dense outer coat and a thin, less electron-dense inner coat [15] Recently, a spore crust structure was demon-strated to be the outermost layer of the spore This layer is found

to locate outside of the outer coat and is composed of CgeA and CotZ proteins[16,17]

Bacillus subtilis has several advantages in developing spore sur-face display systems due to the detailed knowledge of its spore structure [15,18], the availability and ease of advanced genetic tools[19]and genomic data[20]that allow the construction of re-combinant spores For the construction of B subtilis spore surface display systems, several outer coat proteins such as CotB, CotG,

1046-5928/$ - see front matter Ó 2013 Elsevier Inc All rights reserved.

⇑ Corresponding author Tel.: +49 921 552708; fax: +49 921 552710.

E-mail address: wschumann@uni-bayreuth.de (W Schumann).

Contents lists available atScienceDirect Protein Expression and Purification

j o u r n a l h o m e p a g e : w w w e l s e v i e r c o m / l o c a t e / y p r e p

CotC and an inner coat protein OxdD were successfully used as

carrier proteins for displaying different heterologous proteins on

the spore surface (recently reviewed in [21–23] In all reported

cases, a common approach was applied This includes fusion of a

coding region of the passenger protein with a gene encoding the

spore coat protein (carrier protein carrying an anchoring motif)

and using the native promoter of the coat gene to drive expression

of the fusion gene The translational gene fusion is then integrated

into the B subtilis chromosome, usually into the coding sequence of

the non-essential gene amyE, to obtain genetic stability [19]

Afterwards, the passenger protein will be produced together with

the coat proteins in the cytoplasm and anchored on the spore

surface As the mother cells lyses, spores will be released with

the target protein displayed on the surface This method

guaran-tees the proper timing of the fusion gene expression during

sporulation but the amount of produced fusion proteins cannot

be controlled

The objective of this work was to construct a regulatable spore

surface expression system with two different b-D-thiogalactoside

(IPTG)-inducible promoters, Pgracand PSgrac The coat proteins CotB,

CotC, and CotG were used as carrier protein, and thea-amylase Q

(AmyQ)1 from B amyloliquefaciens [24,25] and the GFPuv, an

en-hanced version of the GFP protein of the jellyfish Aquoria victoria

[26]were used as passenger proteins Our results demonstrate that

different gene dosage effects depend on the carrier and the passen-ger proteins

Materials and methods Bacterial strains, plasmids, and growth conditions The bacterial strains and plasmids used are listed inTable 1 Escherichia coli strain DH10B (Bethesda Research Laboratories) was used as recipient in all cloning experiments The eight-fold protease-deficient strain B subtilis WB800N[27]was used as host strain for surface display of the recombinant proteins on Bacillus spores The vectors pDG1730[28], pDG364[29], pHT01 [30]and pBG01 (M Helfrich, unpublished) were used as cloning vectors Cells were routinely grown in Luria broth (LB) medium (1% 2tryptone, 0.5% yeast extract, 1% NaCl) at 37 °C under aeration To prepare spores,

B subtilis cells were grown in sporulation medium 2  SG (1.6% Difco Nutrient Broth, 0.2% KCl, and 0.05% MgSO4 7H2O, pH 7, supplemented with 0.1% glucose, 1 mM Ca(NO3)2, 0.1 mM MnCl2

and 0.01 mM FeSO4) Antibiotics were added into media at concen-tration of 10lg/ml for chloramphenicol, 100lg/ml for ampicillin and spectinomycin when culturing plasmid-based strains Construction of the plasmids

For displaying recombinant proteins on the spore surface, the coat proteins CotB, CotC and CotG were chosen as carrier proteins

Table 1

Bacterial strains and plasmids used.

E coli DH10B F- mcrA D(mrr-hsdRMS-mcrBC)U80dlacZDM15 DlacX74 endA1 recA1 deoR D(ara,leu)7697 araD139 galU galK

nupG rpsL

k-(Bethesda Research Laboratories) WB800N nprE aprE epr bpr mpr::ble nprB::bsr Dvpr wprA::hyg cm::neo (Neo R ) (prototrophic) (Nguyen et al., 2011b) WB800N

amyE::spc

nprE aprE epr bpr mpr::ble nprB::bsr Dvpr wprA::hyg cm::neo amyE::spc (Neo R

, Spec R

QAS03 amyE::P cotB -cotBD (Cm R

QAS06 amyE::P cotB -cotBD-amyQ (Cm R

QAS52 amyE::P cotB -cotBD-gfp uv (Cm R

QAS43 amyE::P grac -cotBD-gfp uv (Spec R

QAS49 amyE::P grac -cotG-gfp uv (Spec R

QAS32 pQAS32 (P grac -cotBD-amyQ) (Cm R

QAS34 pQAS34 (P Sgrac -cotBD-amyQ) (Cm R

QASK1 pSK01 (P grac -cotBD-gfp uv ) (Cm R

QAS23 pQAS23 (P grac -cotG-gfp uv ) (Cm R

pSDJH-cotG-GFPuv

1

The DNA coding for each protein was amplified by PCR using

chromosomal DNA of B subtilis 1012 as template and appropriate

primer pairs for each backbone plasmid (Table 2)

First, the coat protein CotB was chosen as carrier protein using a

controllable promoter for display of heterologous proteins on the

spore surface Its N-terminal version devoid of three

27-amino-acids repeats [31] was used in all experiments The shortened

version cotBD gene was amplified using the primer pair ON01/

ON02, digested with BglII and BamHI and inserted into the BamHI

site of vector pHT01 and pBG01, respectively, so that regulated

expression of cotBDgene depended on either Pgrac(in pHT01) or

PSgrac (in pBG01) The two new plasmids were named pQAS17

and pQAS19, respectively The cotBDgene with its own promoter

was also amplified from the genome of B subtilis 1012 and

inserted into the integration vector pDG364[19] A flexible linker

(G-G-G-G-S) was then added downstream of cotB, resulting in

pQAS03 Thea-amylase protein AmyQ from B amyloliquefaciens

was employed as a passenger protein for evaluating the system

The coding region of amyQ was amplified using plasmid pKTH10

[25] as a template and fused in-frame to cotBD in the three

plasmids mentioned above The primer pair ON03/ON04 was used

for amplification of the amyQ coding region which was then

inserted at the ClaI site into the plasmid pQAS03, resulting in

pQAS06 The amyQ coding region was also amplified using the

primer pair ON05/ON06, sequentially cleaved by BamHI and AatII,

and cloned into the pQAS17 and pQAS19 plasmid, yielding

plasmids pQAS32 and pQAS34, respectively

Second, the coding regions of cotC and cotG were amplified

using the primer pair ON07/ON08 (cotC) and ON09/ON10 (cotG),

digested with BglII and BamHI and inserted into the BamHI site of

vector pHT01 The new plasmids were constructed in such a way

that transcription of cotC and cotG were controlled by the

IPTG-inducible Pgracpromoter, resulting in pQAS18 and pQAS20,

respec-tively The pSDJH-cotG-GFPuv plasmid[32]was used as template

together with the primer pair ON11/ON12 for amplification of

the gfpuvgene The PCR product was digested with AatII and fused

in-frame to the 30 ends of cotB, cotC and cotG present in pQAS17,

pQAS18 and pQAS20 resulting in the plasmid-based pSK01,

pQAS26 and pQAS23 which carry cotB, cotC and cotG, respectively

The EcoRI fragments containing the Pgrac-cot(BD/C/G)-gfpuvfusion

from plasmid pSK01, pQAS26 and pQAS23 were then inserted into

the integration vector pDG1730[28]resulting in pQAS43, pQAS48

and pQAS49, which carry cotB, cotC and cotG, respectively The gfpuv

coding region was also amplified using the primer pair ON13/

ON14, cleaved by EcoRI and inserted in-frame at the 3’ end of cotBD

in plasmid pQAS03 yielding pQAS52

Construction of strain WB800N amyE::spc

To construct strain WB800N amyE::spc, plasmid pDG1730 was transformed into strain WB800N where replacement of amyE by amyE::spc can occur by a double crossing-over event The candi-date colonies were resistant to spectinomycin, sensitive to erythro-mycin and did not show halo formation when plated on medium containing 0.5% starch stained with I2/KI Correct integration was confirmed by PCR (result not shown) and one correct transformant was kept for further studies (WB800N amyE::spc)

Preparation of spores Single colonies of appropriate B subtilis strains were grown overnight in LB at 37 °C Sporulation was induced by the exhaus-tion method using 2  SG medium [33] After 6 h of inoculation

at 37 °C, 0.2x of a protease inhibitor cocktail (Roche Diagnostics) solution was added (1x is equal to one tablet of protease inhibitor cocktail in 50 ml volume) The culture was then divided into subcultures where one was further grown in the absence and the others in the presence of the inducer IPTG as indicated in each experiment After 24 h of inoculation at 37 °C, whole cells, sporulating cells and spores were harvested by centrifugation The spores were purified by the Renografin (sodium diatrizoate, Sigma) gradient method[33]

Extraction of spore coat proteins Purified spores corresponding to an OD600of 5 were decoated

by treatment for 60 min at 70 °C with 100ll of ST solution (1% so-dium dodecyl sulphate (SDS), 50 mM dithiothreitol) Then, the spores were sedimented by centrifugation (13,000 rpm, 5 min at

4 °C) The coat protein extract was then used directly for SDS–PAGE

by addition of 20ll of 6x loading buffer and heating at 95 °C for

5 min

Western blot analysis Extracted proteins were first separated in 10% denaturing poly-acrylamide gels, then transferred onto a nitrocellulose membrane using the electro blotting technique [34] To compare the regu-lated expression level of proteins on the spore surface, coat ex-tracts from purified spores with an OD600 of 1 were applied in each lane

Table 2

Oligonucleotides used.

Determination ofa-amylase activity

a-amylase activity was determined as described [35] with

whole spores and with supernatants and presented in units per

OD600 One unit is defined as a decrease in OD620of 0.1 All

exper-iments were repeated at least twice

FACS analysis

The spores displaying GFPuv from different cultures were

puri-fied and washed three times with ddH2O Dilutions of 105spores

per ml in ddH2O were directly examined using a Beckman-Coulter

FC-500 MCL System, and the SXP Software was used for data

analysis

Visualization of surface immobilized proteins by confocal microscopy

For immunofluorescence staining, purified spores were first

washed twice in AP buffer (100 mM Tris–HCl, pH 7.4, 100 mM

NaCl, 0.25 mM MgCl2), and then blocked in AP buffer containing

3% BSA (APB buffer) for 1 h on ice Then, the spores were

resus-pended and incubated subsequently in APB buffer containing the

primary antibodies raised in rabbits against B amyloliquefaciens

a-amylase (1:1000), then with anti-rabbit IgG Alexa conjugate

(Alexa FluorÒ488 donkey anti-rabbit IgG; Molecular Probes) at a

dilution of 1:600 in APB buffer for 2 h on ice followed by washing

three times with cold AP buffer Generally, no staining process was

needed for the samples with GFPuv protein; these spores were

used directly for visualization A 5ll aliquot of the stained spore

suspension was mixed with 10ll of 1% agarose and spread onto

a glass slide The fluorescence images were acquired by a SP5

con-focal microscope (Leica Microsystems, Germany) and processed by

using the Adobe Photoshop (Adobe Systems Inc., San Jose, CA, USA)

Results and discussion

Construction of plasmids and strains for display of recombinant

proteins on the B subtilis spore surface using IPTG-inducible

promoters

To examine the idea of using a controllable promoter for display

of heterologous proteins on the spore surface, the coat protein CotB

was first chosen as a carrier protein Its N-terminal version devoid

of three 27-amino-acids repeats was used in all experiments since

it has been shown that fusions of recombinant proteins to this cotB

version are correctly assembled and exposed on the spore surface

[29] The shortened version of the cotB gene (cotBD) was amplified

and inserted into the vector pHT01 Expression of cotBDcan be

reg-ulated by either Pgrac, an IPTG-inducible promoter, which consists

of the PgroES promoter and the lac operator (present in pHT01),

and the promoter-up mutation PSgrac (in pBG01) The two new

plasmids, named pQAS17 and pQAS19, respectively, were able to

replicate autonomously The cotBD gene with its own promoter

was also amplified and inserted into the integration vector

pDG364 with a flexible linker (G-G-G-G-S) translationally added

downstream of CotBD, resulting in pQAS03 This plasmid allowed

ectopic insertion of fusion genes at the non-essential amyE locus

of the B subtilis chromosome [19] Thea-amylase (AmyQ) from

B amyloliquefaciens was employed here as a reporter protein for

evaluation of the surface display system The coding region of

amyQ was amplified using plasmid pKTH10[25]as template and

fused in-frame with cotBDin the three plasmids pQAS17, pQAS19

and pQAS03, resulting in pQAS32, pQAS34 and pQAS06

In a comparable way, cotC and cotG were amplified and inserted

into pHT01 (pQAS18 and pQAS20) Next, the GFPuv coding

sequence was fused in-frame with cotBDin pQAS17 and pQAS03, cotC in pQAS18 and cotG in pQAS20 The EcoRI fragments contain-ing the translational fusion Pgrac-cot(BD/C/G)-gfpuv from the plasmid-based pSK01, pQAS26 and pQAS23 were then inserted into the integration vector pDG1730[28], resulting in a series of plasmids (including pQAS43, pQAS48 and pQAS49, respectively) that allow integration of the hybrid genes with the Pgracpromoter

at the amyE locus

B subtilis cells produce a variety of intra- and extracellular pro-teases[36], which can interfere with the stability of heterologous proteins present either in the cytoplasm or after secretion into the medium Proteins anchored on the spore surface are first exposed to intracellular proteases and, after lysis of the mother cells, to extracellular proteases To prevent degradation by extra-cellular proteases, the eight-fold protease-deficient strain B subtilis WB800N [27] was used in all experiments All the newly con-structed plasmids were transformed into strain WB800N amyE::spc, and first screened by plate assay Individual clones of each transformation were tested by PCR (data not shown) and used for further experiments

To cope with intracellular protease activities including those produced during the sporulation process[37], we decided to add

an protease inhibitor (Pi) cocktail to inactivate them Since many spore coat proteins are synthesized at stage II (at t2–3) of sporula-tion, t2was chosen for adding the Pi cocktail

The optimal concentration of the Pi cocktail was determined experimentally We used strain QAS06, a derivative of WB800N with PcotB-cotBD-amyQ translational fusion inserted ectopically at the amyE locus The molecular mass of the fusion protein was cal-culated to be approximately 100 kDa (42.9 kDa of CotBD plus

55 kDa of AmyQ) Cells of QAS06 were treated with different Pi cocktail concentrations, the total spore coat protein fraction was extracted and separated by SDS–PAGE As shown inFig 1A, there

is no significant increase in a protein band of about 100 kDa in the absence (0) and presence (0.1 to 0.8) of the Pi cocktail

To verify the fusion protein, an immunoblot was carried out using polyclonal antibodies raised against the AmyQ protein Cross-reacting material is present in all lanes corresponding to a protein

of about 100 kDa, and the highest amount was visible in the pres-ence of 0.2 Pi cocktail (Fig 1B) We also determined the number

of spores in each sample and measured a slight reduction to about 25% in the presence of 0.4X Pi cocktail (Fig 1C) This reduction is concomitant with the appearance of a strong protein band of about

30 kDa We cannot correlate this protein to any gene and also do not know whether there is a connection between this protein band and the reduced number of spores

The IPTG-inducible promoters are able to increase the amount of heterologous proteins anchored on the spore surface

To examine the possibility of using an IPTG-inducible promoter for display of heterologous proteins on the spore surface, sporula-tion of the strains harbouring either the construct with the Pgracor the PSgrac promoter were induced by the exhaustion method in

2  SG medium, then challenged with different IPTG-concentrations

at t2 of sporulation The result was verified by SDS–PAGE and immunoblot analysis of total spore coat extracts The anti-AmyQ and anti-GFP antibodies were used in immunoblot experiments The molecular masses of the fusion proteins were calculated

to be approximately 100 kDa for CotBD-AmyQ, 70 kDa for CotBD-GFPuv, 39 kDa for CotC-GFPuv and 63 kDa for CotG-GFPuv The immunoblot analysis of the strain QAS32 carrying the Pgrac -cotBD-amyQ translational fusion revealed the appearance of an about 100 kDa band when IPTG was added to the sporulating culture This band was not observed in the absence of IPTG (Fig 2, lane 2–5) With strain QAS34, carrying the P -cotBD

-amyQ translational fusion in pQAS34, the immunoblot analysis re-vealed the presence of this band with the strongest intensity in the presence of 0.25 mM inducer and reduced intensity when higher IPTG-concentrations were added to the sporulating cultures (Fig 2, lane 7–9) Comparison of the CotBD-AmyQ expression level

in the total coat extract using three different promoters indicated the highest amount of the fusion protein with the PSgracpromoter and the lowest one with the PcotBpromoter (Fig 2, lane 1) This re-sult agrees with the hypothesis that the PSgracpromoter is stronger than the Pgracpromoter in controlling gene expression However, due to the leakiness of the PSgracpromoter a band of the CotBD -AmyQ is visible even the absence of IPTG (Fig 2, lane 6)

The immunoblot analysis of strain QAS43 (Pgrac-cotBD-gfpuv in-serted at the amyE locus;Fig 3A, lane 2–5) and strain QASK1 (Pgrac -cotBD-gfpuvin plasmid pSK01;Fig 3A, lane 7 to 10), exhibited the presence of an about 70 kDa band (representing the CotBD-GFPuv fusion protein) which reacted with anti-GFP antibodies when IPTG was supplied In the absence of the inducer, this band was not present (Fig 3A, lane 1 and 6) Similar results were also obtained when inspecting the fusion proteins CotC-GFPuv and CotG-GFPuv:

no detectable band was seen in the absence of IPTG in the culture, and an about 39 kDa band (correlated to CotC-GFPuv;Fig 3B) or a

63 kDa band (correlated to CotG-GFPuv;Fig 3C) appeared after addition of IPTG In all three cases, the intensity of the protein band increased when higher concentrations of IPTG were added until an optimal IPTG-concentration was reached At higher concentrations, the cross reacting band either remained constant (Fig 3B, lane 7–10) or decreased when more IPTG was added

Besides, some alterations in total coat proteins were observed via SDS–PAGE analysis when using different IPTG concentrations For example, notable differences in the intensity of bands between

27 and 35 kDa and/or between 55 and 70 kDa were seen in the gels with all three CotBD-GFPuv fusions (Fig 3A, lane 5), CotC-GFPuv (Fig 3B, lane 1,3,4,6 and 7) and CotG-GFPuv (Fig 3C, lane 2,5 and 7) The optimal IPTG concentrations, identified via the analysis of SDS–PAGEs and immunoblots based on the amount of fusion pro-teins, existed in the total coat proteins from different spore sam-ples, were 0.5 mM for strain QAS32 (Pgrac-cotBD-amyQ) and QAS49 (amyE::Pgrac-cotG-gfpuv), 0.25 mM for strain QAS34 (PSgrac -cotBD-amyQ) and QAS43 (amyE::Pgrac-cotBD-gfpuv), 0.1 mM for

Fig 1 Stabilization of heterologous proteins on the spore surface The strain QAS06

(WB800N amyE::P cotB -cotBD-amyQ) was grown in 2  SG medium, well shaken at

37 °C until t 2 of sporulation (2 h after entering the transition phase, about 6 h after

inoculation) The culture was then divided into subcultures where one was further

grown in the absence of any Pi, while the others were supplemented with different

concentrations of Pi (Protease inhibitors cocktail) as indicated at the top of each

lane The spores were purified at t 18 (about 24 h after inoculation) and decoated.

The coat extracts from purified spores at an OD 600 of 1 per sample were applied per

lane, fractionated by 12% SDS–PAGE and subjected to either Coomassie blue

staining (A) or immunoblotting using rabbit antia-amylase antibodies (B) The

number of spores obtained after purification was calculated by measuring the OD 600

of the spore solution and direct counting with a Thoma counting chamber under a

microscope and given as number of spores 10 11

per 1 l of 2  SG medium (C).

Fig 2 Comparison of regulated CotBD-AmyQ expression The B subtilis strain WB800N amyE::spc harboring one of the fusions, P grac -cotBD-amyQ (QAS32) or P Sgrac -cotBD-amyQ (QAS34) was grown in 2  SG medium, divided into subcultures, which were then induced with different IPTG-concentrations (indicated at the top of each lane) Uninduced cultures of QAS06 (P cotB -cotBD-amyQ), QAS32 and QAS34 were used as controls (A) Coat extracts from purified spores were subjected to immunoblotting using the rabbit antia-amylase antibodies (B).

Fig 3 Identification of the optimal IPTG-concentration for CotBD-GFPuv (A), CotC-GFPuv (B) and CotG-GFPuv (C) expression The B subtilis strains QAS43 (amyE:: P grac -cotC-gfp uv ), QASK1 (P grac -cotBD-gfp uv ), QAS48 (amyE:: P grac -cotC-gfp uv ), QAS26 (P grac -cotC-gfp uv ), QAS49 (amyE:: P grac -cotG-gfp uv ) and QAS23 (P grac -cotG-gfp uv ) were grown in 2  SG medium, divided into subcultures, which were then induced with different IPTG-concentrations (indicated at the top of each lane) The spores were purified at t 18 (about 24 h after inoculation) and decoated The coat extracts from spores of an OD 600 of 1 per sample were applied per well, fractionated by 12% SDS–PAGE and subjected to immunoblotting using rabbit anti-GFP antibodies Coomassie stained gels were prepared in parallel.

Fig 4 Flow cytometric analyses of GFPuv expression on the spore surface The purified spores from the following cultures were analyzed: QAS52 (b), QAS43 induced with 0.25 mM IPTG (c), QASK1; 0.3 mM IPTG (d), QAS48; 0.3 mM IPTG (e), QAS26; 0.1 mM IPTG (f), QAS49; 0.5 mM IPTG (g) and QAS23; 0.1 mM IPTG (h) Spores were prepared and directly used in FACS analysis Spores prepared from strain WB800N (a) were used as a negative control FL1log indicates units of GFP fluorescence intensity; a total of 20,000 spores per sample were counted The percentage of positive fluorescence spores, gated by panel B, was used for comparison of the GFPuv expression on the spore surface from

strain QAS26 (Pgrac-cotC-gfpuv) and QAS23 (Pgrac-cotG-gfpuv),

0.3 mM for strain QASK1 (Pgrac-cotBD-gfpuv), and 0.35 mM for

strain QAS48 (amyE::Pgrac-cotC-gfpuv) These concentrations were

used in further experiments

In addition to immunoblot analysis, a flow cytometric analysis

of GFPuv expression on the spore surface was performed Dilutions

of 105spores per ml in ddH2O were examined directly using the

Beckman-Coulter FC-500 MCL System flow cytometer A total of

20,000 spores per sample were counted and panel B was used for

estimating the percentage of spores exhibiting fluorescence The

data were analyzed by using the CXP Software and presented in

Fig 4 Compared to the WB800N spores (Fig 4a), the histogram

of all seven constructs exhibited an increased signal intensity,

indi-cating that the Cot-GFPuv fusions were expressed as active fusion

proteins

In case of the CotBD-GFPuv constructs, a significant higher GFP

intensity was observed when using the Pgracpromoter: 59.06% for

QAS43 (Fig 4c) and 62.37% for QASK1 (Fig 4d) as compared to

18.06% for QAS52 (PcotB promoter) (Fig 4b) This indicated that

more protein molecules could be indeed displayed on the spore

surface when its regulated expression level was increased This

observation was comparable with results obtained with CotBD

-AmyQ constructs The slight differences in the expression level

between the intensity of chromosomally located (Fig 4c) and

plasmid-based (Fig 4d) constructs showed the minor effect of

increasing the copy number of the fusion gene at the level of

dis-played protein A minor gene dosage effect was also seen with

the CotC-GFPuv constructs: the plasmid-based construct QAS26

(Fig 4e) gave less intensity than the integration construct QAS48

(Fig 4f) (49.15% as compared to 57.35%) On the contrary, the

CotG-GFPuv samples exhibited an increase in the percentage of

gated spores from 18.41% of strain QAS49 (chromosome) (Fig 4g)

to 46.55% of QAS23 (plasmid) (Fig 4h) This result demonstrates the gene dosage effect of the translational fusion present on a plas-mid (4–6 copies per chromosome) versus the chromosomal inser-tion (one copy) These results suggest that each strategy has different effects on displaying proteins on the spore surface when different carrier proteins were used

Considering the histograms from different spore samples, there was another observation with the two CotC-GFPuv constructs (Fig 4e and f) and CotG-GFPuv integration construct There are two peaks in the histogram representing two populations of spores which emitted different intensities in each construct This was

Fig 5 Visualization of AmyQ on the spore surface The purified spores from three different cultures (QAS06, QAS32; 0.5 mM IPTG and QAS34; 0.25 mM IPTG) were treated with the primary antibodies againsta-amylase, then with Alexa 488-conjugated secondary antibodies for immunofluorescence microscopy The AmyQ protein displayed on the spore surface was visualized by transmission and fluorescence Leica SP5 microscopy The purified spore from WB800N amyE::spc were treated the same way and used as a

Fig 6 Comparison of the a-amylase activities from different spore samples displayed the fusion CotBD-AmyQ on the surface Purified spores from strains QAS32 (P grac -cotBD-amyQ) and QAS34 (P Sgrac -cotBD-amyQ) were analyzed The spores corresponding to an OD 600 of 1 from each sample were used for determi-nation ofa-amylase activity The result is presented in units per OD 600 of spores.

previously observed when the CotC protein was utilized for

expres-sion of Clonorchis sinensis tegumental protein 22.3 kDa on the spore

surface[38] Still, no convincing explanation is possible for this

observation

IPTG-inducible promoters can be used for displaying heterologous

proteins on the surface of the spores

To analyse the surface exposure of CotBD-AmyQ using

IPTG-inducible promoters, the purified spores from strains QAS06,

QAS32, QAS34 and wild type strain WB800N were first incubated with antibodies against AmyQ, followed by Alexa-conjugated secondary antibodies and then visualized under the Leica SP5 confocal microscope The secondary antibodies bind to the primary antibodies raised against AmyQ and emit fluorescence light after excitation with the appropriate wavelength Fluorescence was observed around the spores of all three strains displaying AmyQ

on the spore coat (Fig 5) This indicated interaction of the antibody with AmyQ, hence proved the display of this protein on the spore surface using the CotBD as an anchoring motif and an IPTG-inducible promoter for controlling the regulated expression of heterologous proteins on the spore surface

In addition to immunofluorescence microscopic analyses,

a-amylase assays were performed using purified spores displaying the fusion CotBD-AmyQ The activities, presented as units per

OD600, were calculated and compared among the different samples Spores from strain WB800N amyE::spc were used as a reference As shown inFig 6, about 2.5 units were measured when using spores from strain QAS06 (PcotB promoter), while the activity of spores using the Pgracpromoter for regulated expression (strain QAS32) was increased to 3 units An unexpected result was observed with strain QAS34 (PSgrac promoter) The spores showed the lowest activity (about 1.5 units) although the amount of the fusion CotBD-AmyQ expressed on the spore coat using the PSgracpromoter was the highest as compared to other samples (Fig 2) This result

Fig 7 Expression of GFPuv with/without a carrier protein for display on the spore surface The spores from WB800N strain harboring either plasmid pHT01-GFPuv (without carrier) or pSK01 (CotBD as carrier) were purified (0.3 mM IPTG was used for induction) These spores were treated with the primary antibodies against GFP (Clontech, 1:500), followed by Cy3-conjugated secondary antibodies (Molecular Probes, 1:600) for immunofluorescence microscopy The GFPuv displayed on the spore surface was visualized

by Leica SP5 microscopy.

Scheme 1 Determination of the number ofa-amylase molecules on the spore

surface The spores from strains QAS06 and QAS32 were prepared using the

exhaustion method The spore coat fraction from purified spores was prepared and

the contents of 1.5  10 8 spores were applied per lane Defined amounts of purified

a-amylase from 22.5 to 60 ng (corresponding to 0.45–1.2 pmol) were run on the

same gel Only the material in the upper band in two lanes of QAS06 and QAS32 was

quantified.

Fig 8 Localization of Cot proteins on the spore surface The spores from different strains QAS52 (a), QASK1 (b), QAS26 (c) and QAS23 (d) were prepared and directly observed

might indicate the presence of an enzymatically inactive form of

AmyQ on the spore surface

A confocal microscopic analysis with the spores from strains

that expressed GFPuv on the spore coat also showed that GFPuv

was observed around the shape of the examined spores (data not

shown) This result demonstrates the success of using the Pgrac

promoter on management of the cot-gfpuvfusion genes for spore

surface display purposes

It was previously described that the Pgrac promoter could be

used in the absence of a carrier protein and was able to display

recombinant proteins on the spore surface[39] Therefore, a

con-struct with GFPuv without a carrier protein was generated and

tested These spores exhibited GFP fluorescence, however a very

weak signal was detected when stained with primary antibodies

against GFP and Cy3-conjugated secondary antibodies A clear

red signal was observed with spores from strain QASK1 (using

CotBDas carrier protein) treated the same way (Fig 7) Taken

to-gether, those phenomena and the observations from the previous

report might be the result from overproduction of the recombinant

protein in the mother-cell compartment leading to the random

assembly of the target protein in the spore coat The red signal

de-tected from the QASK1 spores also indicated the GFPuv displayed

on the spore surface using the CotBDas anchoring motif

Substitution of the native promoter by Pgracresulted in a two-fold

increase in the amount of the fusion proteins displayed on the spore coat

The amount ofa-amylase molecules displayed on the spore

sur-face was determined by densitometrical analysis of the

immuno-blot The soluble coat fractions were extracted from a defined

amount of purified spores (1.5  108) of strains QAS06 and

QAS32, the coat proteins were separated by SDS-PAGE, and

CotBD-AmyQ was identified by immunoblotting To calculate the

amount ofa-amylase, increasing amounts of purified enzyme were

applied on the same gel (Scheme 1) From the densitometric

scan-ning of the different bands, the numbers ofa-amylase molecules

were calculated to be 4.6  103 molecules per QAS06 spore and

9.73  103 molecules per QAS32 spore This means that the

amount of proteins displayed per spore can be increased about

two-fold by substitution of the native by the Pgracpromoter

Overproduced Cot proteins are accumulated at different positions on

the surface of spores

An interesting phenomenon dealing with the localization of the

coat proteins once overproduced was observed when using GFPuv

as a reporter protein When PcotBwas used, GFPuv was visualized

around the shape of the spore indicating the presence of the CotBD

protein all over the surface of spore (Fig 8a) Using the Pgrac

pro-moter for overexpression, the CotBD protein was concentrated

on the two poles of the spore (Fig 8b), while the CotC protein

accu-mulated into a spot on spore surface (Fig 8c) and the CotG protein

tended to arrange at the midpoints of the spores (Fig 8d) These

arrangements might be the consequence of self-interaction for

each species of coat protein once overproduced and their tendency

of localization on the spore coat since the interaction of the

pro-teins during coat assembly has been previously reported[40–44]

Further studies should be carried out to understand more about

this interesting phenomenon

Acknowledgments

Q A Nguyen was funded by the Bavarian Research Foundation

We would like to express our thanks to Prof J Kim from the

Dong-A University in Busan, South-Korea, for valuable advices

References

[1] J Löfblom, Bacterial display in combinatorial protein engineering, Biotechnol J.

6 (2011) 1115–1129 [2] O Schneewind, D.M Missiakas, Protein secretion and surface display in Gram-positive bacteria, Philos Trans R Soc Lond B Biol Sci 367 (2012) 1123–1139 [3] J Pande, M.M Szewczyk, A.K Grover, Phage display: concept, innovations, applications and future, Biotechnol Adv 28 (2010) 849–858

[4] M Paschke, Phage display systems and their applications, Appl Microbiol Biotechnol 70 (2006) 2–11

[5] S.Y Lee, J.H Choi, Z Xu, Microbial cell-surface display, Trends Biotechnol 21 (2003) 45–52

[6] Z Xu, S.Y Lee, Display of polyhistidine peptides on the Escherichia coli cell surface by using outer membrane protein C as an anchoring motif, Appl Environ Microbiol 65 (1999) 5142–5147

[7] M Hansson, P Samuelson, E Gunneriusson, S Stahl, Surface display on gram positive bacteria, Comb Chem High Throughput Screen 4 (2001) 171–184

[8] E.T Boder, K.D Wittrup, Yeast surface display for screening combinatorial polypeptide libraries, Nat Biotechnol 15 (1997) 553–557

[9] P Martineau, A Charbit, C Leclerc, C Werts, D O’Callaghan, M Hofnung, A genetic system to elicit and monitor antipeptide antibodies without peptide synthesis, Biotechnology (NY) 9 (1991) 170–172

[10] A Mogk, D Huber, B Bukau, Integrating protein homeostasis strategies in prokaryotes, Cold Spring Harb Perspect Biol 3 (2011) 1–19

[11] F.U Hartl, A Bracher, M Hayer-Hartl, Molecular chaperones in protein folding and proteostasis, Nature 475 (2011) 324–332

[12] W.L Nicholson, N Munakata, G Horneck, H.J Melosh, P Setlow, Resistance of Bacillus endospores to extreme terrestrial and extraterrestrial environments, Microbiol Mol Biol Rev 64 (2000) 548–572

[13] A.I Aronson, P.C Fitz-James, Structure and morphogenesis of the bacterial spore coat, Bacteriol Rev 40 (1976) 360–402

[14] J Errington, Regulation of endospore formation in Bacillus subtilis, Nat Rev Microbiol 1 (2003) 117–126

[15] A Driks, Bacillus subtilis spore coat, Microbiol Mol Biol Rev 63 (1999) 1–20 [16] D Imamura, R Kuwana, H Takamatsu, K Watabe, Localization of proteins to different layers and regions of Bacillus subtilis spore coats, J Bacteriol 192 (2010) 518–524

[17] D Imamura, R Kuwana, H Takamatsu, K Watabe, Proteins involved in formation of the outermost layer of Bacillus subtilis spores, J Bacteriol 193 (2011) 4075–4080

[18] A.O Henriques, C.P Moran Jr., Structure, assembly, and function of the spore surface layers, Annu Rev Microbiol 61 (2007) 555–588

[19] S.M Cutting, P.B Vander Horn, Genetic analysis, in: C.R Harwood, S.M Cutting (Eds.), Molecular Biological Methods for Bacillus, John Wiley & Sons, Chichester, 1990, pp 27–60

[20] F Kunst, N Ogasawara, I Moszer, et al., The complete genome sequence of the Gram-positive bacterium Bacillus subtilis, Nature 390 (1997) 249–256 [21] L.D Knecht, P Pasini, S Daunert, Bacterial spores as platforms for bioanalytical and biomedical applications, Anal Bioanal Chem 400 (2011) 977–989 [22] J Kim, W Schumann, Display of proteins on Bacillus subtilis endospores, Cell Mol Life Sci 66 (2009) 3127–3136

[23] J.G Pan, E.J Kim, C.H Yun, Bacillus spore display, Trends Biotechnol 30 (2012) 610–612

[24] I Palva, M Sarvas, P Lehtovaara, M Sibakov, L Kaariainen, Secretion of Escherichia coli beta-lactamase from Bacillus subtilis by the aid of alpha-amylase signal sequence, Proc Natl Acad Sci USA 79 (1982) 5582–

5586 [25] I Palva, Molecular cloning of alpha-amylase gene from Bacillus amyloliquefaciens and its expression in B subtilis, Gene 19 (1982) 81–87 [26] A Crameri, E.A Whitehorn, E Tate, W.P.C Stemmer, Improved green fluorescent protein by molecular evolution using DNA shuffling, Nat Biotechnol 14 (1996) 315–319

[27] H.D Nguyen, T.T Phan, W Schumann, Analysis and application of Bacillus subtilis sortases to anchor recombinant proteins on the cell wall, AMB Express

1 (2011) 22 [28] A.M Guérout-Fleury, N Frandsen, P Stragier, Plasmids for ectopic integration

in Bacillus subtilis, Gene 180 (1996) 57–61 [29] C.R Harwood, S Cutting, Molecular Biological Methods for Bacillus, John Wiley

& Sons, Chichester, England, 1990 [30] D.H Nguyen, T.T.P Phan, W Schumann, Expression vectors for the rapid purification of recombinant proteins in Bacillus subtilis, Curr Microbiol 55 (2007) 89–93

[31] R Isticato, G Cangiano, H.T Tran, A Ciabattini, D Medaglini, M.R Oggioni, F.M.

De, G Pozzi, E Ricca, Surface display of recombinant proteins on Bacillus subtilis spores, J Bacteriol 183 (2001) 6294–6301

[32] J.H Kim, C Roh, C.W Lee, D Kyung, S.K Choi, H.C Jung, J.G Pan, B.G Kim, Bacterial surface display of GFP(uv) on Bacillus subtilis spores, J Microbiol Biotechnol 17 (2007) 677–680

[33] W.L Nicholson, P Setlow, Sporulation, germination and outgrowth, in: C.R Harwood, S.M Cutting (Eds.), Molecular Biological Methods for Bacillus, John Wiley & Sons, New York, 1990, pp 391–450

[34] H Towbin, T Staehelin, J Gordon, Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications, Proc Natl Acad Sci USA 76 (1979) 4350–4354

[35] W.L Nicholson, G.H Chambliss, Isolation and characterization of a cis-acting

mutation conferring catabolite repression resistance toa-amylase synthesis in

Bacillus subtilis, J Bacteriol 161 (1985) 875–881

[36] P Schaeffer, J Millet, J.P Aubert, Catabolic repression of bacterial sporulation,

Proc Natl Acad Sci USA 54 (1965) 704–711

[37] J Mandelstam, W.M Waites, Sporulation in Bacillus subtilis The role of

exoprotease, Biochem J 109 (1968) 793–801

[38] Z Zhou, H Xia, X Hu, Y Huang, C Ma, X Chen, F Hu, J Xu, F Lu, Z Wu, X Yu,

Immunogenicity of recombinant Bacillus subtilis spores expressing Clonorchis

sinensis tegumental protein, Parasitol Res 102 (2008) 293–297

[39] S.K Yim, H.C Jung, C.H Yun, J.G Pan, Functional expression in Bacillus subtilis

of mammalian NADPH-cytochrome P450 oxidoreductase and its

spore-display, Protein Expr Purif 63 (2009) 5–11

[40] H Kim, M Hahn, P Grabowski, D.C McPherson, M.M Otte, R Wang, C.C Ferguson, P Eichenberger, A Driks, The Bacillus subtilis spore coat protein interaction network, Mol Microbiol 59 (2006) 487–502

[41] A Driks, The Bacillus spore coat, Phytopathology 94 (2004) 1249–1251 [42] A Driks, S Roels, B Beall, C.P Moran Jr., R Losick, Subcellular localization of proteins involved in the assembly of the spore coat of Bacillus subtilis, Genes Dev 8 (1994) 234–244

[43] R Isticato, A Pelosi, R Zilhao, L Baccigalupi, A.O Henriques, F.M De, E Ricca, CotC-CotU heterodimerization during assembly of the Bacillus subtilis spore coat, J Bacteriol 190 (2008) 1267–1275

[44] R Isticato, A Pelosi, F.M De, E Ricca, CotE binds to CotC and CotU and mediates their interaction during spore coat formation in Bacillus subtilis, J Bacteriol 192 (2010) 949–954