Scholarly article on topic 'IcsA autotransporter passenger promotes increased fusion protein expression on the cell surface'

IcsA autotransporter passenger promotes increased fusion protein expression on the cell surface Academic research paper on "Biological sciences"

0
0
Share paper
Academic journal
Microb Cell Fact
OECD Field of science
Keywords
{""}

Academic research paper on topic "IcsA autotransporter passenger promotes increased fusion protein expression on the cell surface"

GjGjCj MICROBIAL CELL

GjGjCj FACTORIES ®00

RESEARCH Open Access

IcsA autotransporter passenger promotes increased fusion protein expression on the cell surface

Mabel Lum and Renato Morona*

Abstract

Background: Autotransporters are attractive cell surface display vehicles as they lack complex adaptor proteins necessary for protein export. Recent reports have suggested that the native effector domain (a domain) and translocation domain (p domain) interact with each other to drive translocation of the effector domain to the outer membrane. In this report we compared the expression, surface localisation and folding of TEM-1 p-lactamase (Bla) and maltose binding protein (MalE or MBP) fused to either full length Shigella flexneri IcsA (IcsA) autotransporter or to the b domain alone (IcsAp) to determine the contribution of the native IcsA a domain in presenting the fusion proteins on the surface of E. coli K-12 UT5600 (AompT).

Results: Expression of IcsA-Bla was greater than IcsAp-Bla. High levels of IcsA-MalE were detected but IcsAp-MalE was not expressed. All fusion proteins other than IcsAp-MalE were localised to the outer membrane and were detected on the surface of UT5600 via immunofluorescence microscopy. All bacteria expressing IcsA-MalE were labelled with both a-IcsA and a-MBP. UT5600 expressing IcsAp-MalE was not labelled with a-MBP. A third of UT5600 expressing IcsA-Bla were detectable with a-Bla but only 5% of UT5600 (IcsAp-Bla) were labelled with a-Bla. The correct folding of the Bla moiety when fused to IcsA and IcsAp was also retained as UT5600 expressing either fusion protein exhibited a decreased zone of inhibition in the presence of ampicillin. UT5600 expressing IcsA-Bla was more resistant compared to UT5600 expressing IcsAp-Bla.

Conclusions: The export mechanism of autotransporters is not well understood but accumulating evidence suggest a critical role for the native effector or a domain in facilitating its own export via interactions with the translocation or p domain. This is the first report directly comparing expression of heterologous proteins fused to the full length IcsA autotransporter and fusion to the p domain alone. Protein expression and surface presentation of the fusion proteins were dramatically improved when fused to IcsA rather than IcsAp. Future studies involved in designing autotransporters as cell surface display vehicles would benefit from including the native a domain. This work also provides further evidence for a key interaction between the autotransporter a and p domains.

Background

Expression of heterologous proteins and peptides on bacterial surface are important for a wide range of applications such as vaccine production, bioremediation, biocatalysis, peptide library screening and translocation studies [1]. The autotransporter or type Va secretary pathway is an attractive cell surface display vehicle due to the lack of complex adaptor machineries necessary for protein export compared to other Gram negative export pathways [2]. High expression of recombinant molecules

* Correspondence: renato.morona@adelaide.edu.au Australian BacterialPathogenesis Program, Discipline of Microbiology and Immunology, Schoolof Molecular and BiomedicalScience, University of Adelaide, Adelaide, South Australia, Australia

Bio Med Central

(> 105) per cell without adverse effects on cell viability have been reported [3]. Fusion proteins displayed on bacterial surface can be subjected to FACS (fluorescence-activated cell sorting) and ELISA (enzyme-linked immu-nosorbent assay) to facilitate high throughput screenings [4]. The strict genotype to phenotype linkage of the autotransporter system aids phenotypic selections [5]. Furthermore autotransporters can also be expressed in other Gram negative bacteria making it highly versatile [6,7]. Epitopes displayed have been reported to elicit strong immune stimulation, making it a suitable medium for potential live vaccine development [8].

Cell surface display with autotransporters has its drawbacks. Disulphide bond formation within the

© 2012 Lum and Morona; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

exogenous protein has been reported to affect protein translocation but this has been addressed by inactivating dsbA, a dithiol oxidase, which assists in disulphide bond formations in periplasmic proteins [7,9,10]. The successful translocation of the single chain immunoglobulin fragment with a disulfide bridge to the bacterial surface [11] suggest that the final dimension of the heterologous protein is the limiting factor rather than the presence of the disulfide bridge. OmpT, an outer membrane (OM) protease, has been shown to cleave exogenous proteins after translocation, thus it is important to use a host strain with an ompT deletion [7,9].

The autotransporter protein IcsA (or VirG) is a key virulence factor of Shigella flexneri. It is polarly distributed on the OM and is essential for mediating Shigella's intra- and intercellular motility in the host colonic epithelium through activation of the host neural Wiskott-Aldrich syndrome protein (N-WASP) and subsequent actin nucleation through the Arp2/3 complex [12-16]. The 120 kDa IcsA protein is a typical autotransporter with three distinct regions; an N-terminal signal sequence (amino acids (aa) 1-52), a central effector domain or a domain (aa 53-758) and a C-terminal translocation domain or b domain (aa 759-1102) [17]. The translocation of the IcsA polypeptide from the cytoplasm to the peri-plasm is directed by its unusually long signal sequence via the Sec pathway [18]. The signal sequence is also crucial in maintaining the stability of fusion constructs [19]. In the periplasm, an intramolecular disulphide bridge is formed in the IcsA a domain [20]. The 80 kDa a domain is exported to the extracellular milieu when the 37 kDa b domain inserts itself into the OM [17]. The translocation unit of NalP (Neisseria meningitides), EspP (E. coli strain O157:H7) and the trimetric EstA (Pseudomonas aerugi-nosa) autotransporters are made up of a 12 stranded C-terminal b-barrel pore with a N-terminal a-helix inserted into the pore itself [21-23]. The IcsA translocation unit is likely to adopt a similar confirmation with a predicted N-terminal a-helix [15]. IcsAa can be cleaved on the surface by its specific protease IcsP (SopA) between Arg758-Arg759 [24-26].

In most examples of autotransporter-based display systems, the native a domain is replaced with the pep-tide or protein of interest and is expressed as a single polypeptide with the b domain [3,6,15]. Previously the FimH lectin domain was fused to a portion of the E. coli Antigen 43 (Ag43) autotransporter a domain and was exported to the cell surface by the Ag43 b domain. The chimeric protein was expressed, its function retained and was presented on the surface of the Salmonella enterica host strain [6]. This suggest that the presence of the a domain, at least a portion of it is not detrimental to the stability of the chimeric polypeptide.

Recently Saurf et al. suggested that the native a and b domains of ATs form specific interactions which facilitate efficient translocation of the a domain [27]. Replacement of the haemoglobin protease (Hbp) b domain with the closely related EspP b domain, resulted in the secretion of the Hbp a domain, albeit with reduced efficiency [27]. Increasing or decreasing the size of the Hbp b domain by modifying the number of b strands also affected translocation of the a domain [27]. Furthermore the conserved residues of the a-helix embedded inside the EspP b domain is not crucial for efficient translocation but is indispensable for proteolytic release of the Hbp a domain [28].

In this report, the stability and expression in E. coli K-12 UT5600 (AompT) of TEM-1 b-lactamase (Bla) and E. coli maltose binding protein (MalE or MBP) fused to either the full length IcsA protein (IcsA) or to the b domain alone (IcsAb) was compared to determine the contribution of the a domain to the stability, expression and surface localisation of the chimeric proteins. The outcomes of this study will aid in future cell surface display design and also in improving our understanding of the autotransporter export pathway.

Results & discussion

Construction of IcsA and IcsAp fusion proteins

In a previous study, a NotI site was inserted into icsA, as part of a 15 nucleotide in-frame insertion at amino acid 87 (icsAi87, TGCGGCCGCAATGGA: amino acid 88-92, nucleotide 261-270) [29]. icsAi87 will herein be referred as icsA. icsA was cloned via digestion with EcoRI and SalI with likewise digested pBAD33 to generate pMLRM1 (Figure 1). In a previous study, IcsAb (amino acid 7641102) was fused to maltose binding protein (MalE/MBP) and alkaline phosphatase to dissect the IcsA export pathway [15]. The C-terminal b-barrel pore and the embedded N-terminal a-helix, herein referred as icsAp (amino acid 764-1102), was amplified from pMLRM1 with a forward primer incorporating a NotI restriction site and a reverse primer incorporating a SalI restriction site. icsAp was cloned via digestion with NotI and SalI with likewise digested pBAD33 to generate pMLRM2. The expression of the icsA and icsAp coding regions in pMLRM1 and pMLRM2, respectively, is under the control of the tightly regulated arabinose-inducible araBAD promoter [30].

IcsA and IcsAb proteins fused to the 29 kDa Bla and 44 kDa MalE were generated to investigate the contribution of the IcsA passenger domain to the overall stability of the chimeric proteins (Figure 2). Both Bla and MalE are relatively globular and have been used in various autotransporter studies [15,31]. The bla and malE coding regions were subcloned from pKMRM1 and pMAL-c2,

NotI (aa87)

bla or malE

ss a icsAe

ß SalI

pMLRM2 (pBAD33::icsA6) 6.7 kbp

PCR with 2256_KpnI and REPL.BIO-pBAD30_Sal

NotI (aa87)

NotI SalI

PCR with IcsA_764_NotI and REPL.BIO-pBAD30_Sal

pBAD33 5.4 kbp

bla or malE

Figure 1 Schematic representation of the construction of pMLRM1 and pMLRM2. icsA was amplified from pKMRM1 and was cloned into pBAD33, resulting in pMLRM1. icsAp was amplified from pMLRM1 and was cloned into pBAD33, resulting in pMLRM2. The NotI site (TGCGGCCGC: amino acid 88-91, nucleotide 262-270) of icsA and icsAp serves as the insertion site for bla (TEM-1 beta lactamase) and malE (E coli maltose binding protein). pBAD is the arabinose inducible promoter, ss represents the IcsA signalpeptide, a represents the IcsA passenger domain and b represents the IcsA translocator domain. Drawing is not to scale.

Figure 2 Summary of IcsA and IcsAp fusion proteins. Black boxes represent the IcsA signalpeptide (ss), light grey boxes represent the 80 kDa IcsA„ (a) and white boxes represent the 37 kDa IcsAp (p). TEM-1 beta lactamase (Bla) and maltose binding protein (MalE/MBP) are represented by dark grey boxes. Predicted molecular weights are shown in parentheses.

A ВО' SO' SO' ПО' О'

ПО kDa -SS kDa

1 Z В 4 S B О' ВО' SO' SO' 1Ю' C О' ВО' SO' SO' 1Ю'

1SO kDa IcsA-Bla

123 4 5123 4 5

1 Z В 4 S

■ 7В kDa IcsAp-Bla

■ SO kDa IcsA„-Bla *

ISO kDa — 120 kDa — 85 kDa — 60 kDa —

E 0' 30' 60' SO' 120' F 0' 30' 60' SO' 120'

1S4 kDa IcsA-MalE

123 4 5 12345

60 kDa—. 50 kDa —

Non-specific binding ■ 44 kDa MalEt

H О' ВО' SO' SO' 1ZO' _------ — ~ ^— Non-specific binding

Figure 3 Analysis of IcsA/IcsAp fusion protein production.

UT5600 strains were grown to mid-log phase and induced with 0.2% (w/v) arabinose. Samples were taken at times indicated. Approximately 1 x 108 bacteria were solubilised, electrophoresed on a 12% SDS-polyacrylamide geland subjected to Western immunoblotting. Immunoblotting of (A) MLRM4 (pMLRMI) with anti-IcsA; (B) MLRM28 (pMLRMI::bla) with anti-IcsA and (C) anti-Bla; (D) MLRM20 (pMLRM2::bla) with anti-Bla; (E) MLRM39 (pMLRMI:: malE) with anti-IcsA and (F) anti-MBP; (G) MLRM37 (pMLRM2::malE) with anti-MBP; (H) UT5600 with anti-MBP. The apparent molecular sizes of the proteins are indicated. *Protein degradation products fMalE induced due to chromosomalgene in UT5600

1ZOkDa IcsA

1ZO kDa

D О' ВО' SO' SO' HO SS kDa —

SO kDa —

4O kDa —

4O kDa

respectively, with forward and reverse primers incorporating a NotI restriction site resulting in pMLRM28 (pMLRM1::Wa), pMLRM20 (pMLRM2::b/a), pMLRM39 (pMLRM1::ma/£) and pMLRM37 (pMLRM2::ma/£). Figure 1 outlines the construction of these plasmids. The plasmids were transformed into an AompTE. coli K-12 strain, UT5600, to prevent fusion protein cleavage by OmpT, an OM protease [7,9].

Effects of pMLRM1 and pMLRM2 fusion protein expression on the growth of E. coli K-12 UT5600

The effects of inducing the pBAD promoter in pMLRM1::bla and pMLRM2::bla were compared to pMLRM1 and pBAD33 in the E. coli K-12 UT5600 strain. The growth rates of UT5600 carrying pMLRM1, pMLRM1::bla and pMLRM2::bla were comparable to UT5600 carrying pBAD33 (data not shown, Additional file 1: Figure S1).

Expression of chimeric IcsA and IcsAp fusion proteins

Production of the chimeric IcsA-Bla, IcsAp-Bla, IcsA-MalE and IcsAp-MalE in UT5600 were examined by SDS-PAGE and Western immunoblotting after arabinose induction. At every 30 min, a 1 mL sample was taken and kept on ice before they were standardised to 1 x 108 bacteria/mL with 2x sample buffer. All pMLRM1 chimeric constructs were immunoblotted with both a-IcsA and also the respective specific antibodies (a-Bla, a-MBP). All pMLRM2 chimeras were only blotted with the respective specific antibodies since our anti-IcsA serum does not recognise the p domain of IcsA.

IcsA induced from UT5600 (pMLRM4) was immuno-blotted with a-IcsA as shown in Figure 3A. The amount of IcsA detected did not appear to be increase over the 2 h time course; the smaller MW bands were probably IcsA breakdown products.

IcsA-Bla induced from UT5600 (pMLRM28) was immunoblotted with a-IcsA and a-Bla (Figure 3B and 3C). Increasing amounts of proteins were detected over the time course of pMLRM28 induction. Small amounts of degraded proteins were observed after 2 h (Figure 3B

and 3C, lane 5). When IcsAp-Bla induced from UT5600 (pMLRM20) was imunoblotted with a-Bla (Figure 3B), low levels of IcsAp-Bla was detected without any increase in protein levels over time. A smaller 60 kDa IcsAp-Bla was also consistently observed.

IcsA-MalE induced from UT5600 (pMLRM39) was detected by immunoblotting with a-IcsA and a-MBP (Figure 3E and 3F). Increasing amounts of the fusion protein were detected over the time course of pMLRM39 induction. Degraded IcsA-MalE proteins appeared after 1 h induction of pMLRM39 arabinose induction, consistent with the observed increased amount of protein production (Figure 3E and 3F, lanes 3-5). When IcsAp-MalE induced from UT5600 (pMLRM37) was immunoblotted with a-MBP (Figure 3G), the 81 kDa IcsAp-MalE was not detected. Instead two bands of approximately 44 kDa and 60 kDa were observed.

It was expected that the expression of all constructs subcloned into the pBAD33-based vectors would increase when the strains were induced with arabinose. This was observed for IcsA-Bla (Figure 3B and 3C). In the case of IcsAp-Bla, the expression level of the protein remained constant over a period of 2 h (Figure 3D). The consistency of the protein levels induced from the

pBAD promoter was also observed when IcsA was induced from pMLRM1 (Figure 3A).

Due to the rapid induction of the pBAD promoter, it is likely that the maximal protein induction has been reached by 30 min. Consistent with this hypothesis, degraded proteins were observed for both 120 kDa IcsA (Figure 3A, lanes 1-4) and 150 kDa IcsA-Bla (Figure 3B and 3C, lane 5). The sizes of the degraded proteins were greater than 85 kDa in both cases. Compared to IcsA-Bla, very low levels of the 73 kDa IcsAp-Bla was expressed (Figure 3D). The presence of a smaller 60 kDa protein suggests degradation of IcsAb-Bla was rapid and smaller polypeptides were subsequently not detected by Western immunoblot-ting. It can be concluded that Bla fused to IcsA resulted in higher levels of but not necessarily more stable chimeras compared to fusion to IcsAp.

Over the 2 h induction with arabinose, increasing amounts of proteins, both full length and degraded (> 60 kDa) were observed for IcsA-MalE (Figure 3E and 3F, lanes 3-5). IcsA-MalE was cleaved either during or after translocation from the periplasm to the OM. No IcsAb-MalE was observed by whole cell western immunoblotting even though a slower induction at 25°C was attempted (not shown). Increasing amounts of a 44 kDa protein, corresponding to MalE was present at all time points after arabinose induction (Figure 3G). As MalE expression was induced in UT5600 alone by the presence of arabinose (Figure 3H), hence the 44 kDa MalE protein was background MalE expression by UT5600. The lack of IcsAb-MalE expression cannot be explained at this stage. No mutations were identified in DNA sequencing, and previous attempts to detect MalE fusions to the translocation domain of IcsA and other autotransporters have been successful [15,32].

OM localisation

To determine if the fusion chimeras were presented on the surface of UT5600, two different approaches were undertaken; cell fractionation and immunofluorescence (IF) microscopy.

The UT5600 strains (MLRM28 - pMLRM1::bla, MLRM20 - pMLRM2::bla, MLRM39 - pMLRM1::malE, MLRM37 - pMLRM2::malE) were subjected to cell frac-tionation to determine if the IcsA and IcsAb chimeras were located in the OM.

The 150 kDa band corresponding to IcsA-Bla was detected in the whole membrane (WM) and OM when immunoblotted with a-Bla and a-IcsA (Figure 4A and 4B, lanes 2 and 4). IcsAb-Bla was also detected in the WM and OM (Figure 4A, lanes 6 and 8). A band corresponding to the size of the mature 29 kDa Bla was observed in the soluble fraction (S) and the inner membrane (IM) of UT5600 strains expressing IcsA-Bla (MLRM28) and IcsAb-Bla (MLRM20) (Figure 4A, lanes

1, 3, 5 and 7). The association of the Bla protein at the IM has not been reported previously and is most likely due to an artefact from fractionation. Degraded proteins (< 120 kDa) were observed for IcsA-Bla (Figure 4A and 4B, lanes 2 and 4) but not IcsAb-Bla in the WM and OM fractions (Figure 4A, lanes 6 and 8).

IcsA-MalE (pMLRM1::malE) was detected in both the WM and OM fractions of UT5600 when immuno-blotted with a-IcsA and a-MBP (Figure 4C and 4D, lanes 2 and 4). MalE-related bands (> 70 kDa) were also observed, similar to immunoblotting results following induction with arabinose (Figure 3E and 3F).

Surface localisation

IF microscopy was carried out to determine if the chimera IcsA and IcsAb were presented on the surface of UT5600. The chimeric IcsA autotransporters were detected with a-IcsA (Figure 5A). As expected the positive control (pMLRM4 - pBAD33::icsA) was detected on the cell surface on both the lateral and polar regions of UT5600. The negative control for a-IcsA was UT5600 (pBAD33) (pMLRM3) and cell surface labelling was not

190 kDa _ 120 kDa _ 85 kDa —

60 kDa — 1 50 kDa —

40 kDa —

25kDa — -

WM IM OM

WM IM OM

■ 150 kDa IcsA-Bla

■ 73 kDa IcsA„-Bla

190 kDa — 120 kDa — 85 kDa —

190 kDa — 120 kDa _

WM IM OM

WM i ;

150 kDa IcsA-Bla

■ 164 kDa IcsA-MalE

190 kDa — 120 kDa _

S WM IM OM

164 kDa IcsA-MalE

Figure 4 Detection of IcsA and IcsAp fusion proteins in the OM.

UT5600 strains were grown to mid-log phase, induced with 0.2% (w/v) arabinose and were fractionated into soluble fraction (S), whole membrane (WM), inner membrane (IM) and outer membrane (OM). Equivalent amounts of protein extracts from each fractior were loaded and analysed by Western immunoblotting. Immunobloting of fractions from (A) MLRM28 (pMLRM1::b/a) and MLRM20 (pMLRM2::b/a) with anti-Bla; (B) MLRM28 (pMLRM1::b/a) with anti-IcsA; (C) MLRM39 (pMLRM1::ma/£) with anti-MBP and (D) anti-IcsA. The apparent molecular sizes of the proteins are indicated.

csA -Bla

29 kDa Bla

85 kDa -

85 kDa -

■» t < £2 v r ¿VA* .orMI v

100% 100% IcsA IcsA-Bla IcsAB-Bla 100%

f V ft • 1 f

35°% 5°% IcsA IcsA-MalE IcsA.-MalE

H w fm v- ■ " № f? -

Figure 5 Detection of IcsA and IcsAp fusion proteins on the surface of UT5600 strains. Immunofluorescence microscopy showing surface presentation of pBAD33 (empty vector) (MLRM3), IcsA (MLRM4), IcsA-Bla (MLRM28), IcsAp-Bla (MLRM20), IcsA-MalE (MLRM39) and IcsAp-MalE (MLRM37) with (A) anti-IcsA; (B) anti-Bla; (C) anti-MBP. Phase contrast microscopy (upper panels) and fluorescence microscopy (lower panels) were performed. Specific proteins were detected with their respective antibodies and were detected by FITC-labelled anti-rabbit immunoglobulin serum. The relative percentage of FITC-positive cells are indicated (n = 100)

observed. UT5600 strains expressing IcsA-Bla and IcsA-MalE were labelled with a-IcsA on both the lateral and polar regions, consistent with the OM localisation observations from cell fractionation (Figure 4B and 4D). UT5600 strains expressing Bla chimeras were immunos-tained with a-Bla to determine if the Bla moiety was presented on the surface of UT5600 (Figure 5B). UT5600 strain expressing IcsA (pMLRM4) was not labelled with a-Bla, as expected. Both Bla chimeras, IcsA-Bla and IcsAb-Bla were detected on the lateral and

polar regions of UT5600. The staining of UT5600 (pMLRM28 - IcsA-Bla) (35% of bacteria labelled) was greater than UT5600 (pMLRM20 - IcsAp-Bla) (5% of bacteria labelled). However compared to a-IcsA staining (Figure 5A), fewer IcsA-Bla expressing cells were detected with a-Bla, suggesting that the Bla moiety was cleaved, either during or after transport from the cytoplasm. The low level of IcsAb-Bla detected for UT5600 (pMLRM20 - IcsAb-Bla) was not surprising, given that a low level IcsAp-Bla was detected by Western immuno-blotting (Figure 3D).

UT5600 strains expressing MalE chimeras (pMLRM39 - IcsA-MalE and pMLRM37- IcsAp-MalE) were immu-nostained with a-MBP to determine if MalE was presented on the bacterial surface (Figure 5C). As expected the negative control, UT5600 (pMLRM4 - pBAD33:: icsA) was not labelled with a-MBP. IcsA-MalE was detected on both the lateral and polar regions of UT5600 (pMLRM39) in all the cells, similar to labelling with a-IcsA (Figure 5A). None of UT5600 (pMLRM37) expressing IcsAp-MalE had reacted with a-MBP.

Activity of bla fusion proteins

The activity of the Bla fusion proteins was assessed by measuring the ampicillin resistance of MLRM28 (IcsA-Bla) and MLRM20 (IcsAp-Bla). Bacterial cells were mixed with soft agar and were overlaid on solidified media. The diameter of the zone of bacterial inhibition in the presence of 150 ^g ampicillin was measured. Mean values representing three independent experiments and SEM are presented in Figure 6.

The ampicillin resistance of MLRM28 (IcsA-Bla) and MLRM20 (IcsAp-Bla) were compared to MLRM4 (IcsA) and KMRM1, the original strain from which bla was originally amplified from in three different conditions: LB alone, LB with 0.2% (w/v) glucose and LB with 0.02% (w/v) arabinose. Glucose represses the pBAD promoter, hence the resistance of the Bla constructs when grown on LB agar should be similar to LB with 0.2% (w/v) glucose agar. Higher resistance levels towards ampicillin was anticipated for bacterial cultures overlaid on 0.02% (w/v) arabinose plates.

Strain KMRM1 expressing Bla displayed 100% resistance towards ampicillin and was not affected by the growth media. No observable differences in resistance were noted for Bla fusion proteins grown on LB and LB + 0.2% (w/v) glucose, as expected. MLRM28 (IcsA-Bla) was more resistant to ampicillin compared to MLRM20 (IcsAp-Bla). This is not surprising given that more IcsA-Bla was detected via Western immunoblotting compared to IcsAp-Bla (Figure 3B, C and 3D). Furthermore a third of UT5600 expressing IcsA-Bla was presented on the bacterial surface compared to 5% of cells expressing IcsAp-Bla (Figure 5B). Overall the ability of the UT5600

Figure 6 Ampicillin susceptibility of UT5600 strains expressing IcsA and IcsAp fusion proteins are resistant to 150 ^g ampicillin.

Bacterial strains expressing Bla (KMRM1), IcsA (MLRM4), IcsA-Bla (MLRM28) and IcsAp-Bla (MLRM20) were added to 5 mL soft agar and were overlaid onto three different solidified media: LB alone, LB + 0.2% glucose or LB + 0.2% arabinose. 150 |jg ampicillin was spotted onto sterile Whatman filters, and the agar plate was incubated at 37°C for 16 h. The diameter of the zone of inhibition was measured in millimetre (mm). KMRM1 is resistant to ampicillin. Data are represented as mear ± SEM of three independent experiments. *p < 0.05

strains expressing IcsA-Bla and IcsAp-Bla to grow in the presence of ampicillin suggest proper folding of the Bla moiety and retention of enzymatic activity when fused to IcsA and IcsAb.

Conclusions

The export pathway of autotransporters is not well understood but recent evidence suggests that the native effector domain plays a critical role in facilitating its own translocation when it traverses through the translocation domain via interactions with specific residues [27]. A direct comparison between heterologous proteins fused to full length autotransporter and heterologous proteins fused to the translocation domain has previously not been reported. In this study, protein expression and surface localization of Bla and MalE fused to full length IcsA and IcsAp alone was compared.

The overall data suggest that fusion of exogenous proteins close to the amino terminal of the IcsA autotransporter passenger domain significantly improves protein expression and surface presentation when compared to chimeras fused directly to the IcsA translocation domain. Differences in protein expression, OM localisation and surface presentation between chimera proteins of full

length IcsA and IcsAp was to some extent dramatic. For both Bla and MalE, expression of IcsA fusions was greater than IcsAp chimeras. However the overall expression of IcsAp itself could not be assessed, therefore its contribution to the lack of IcsAp chimera expression cannot be ruled out. Surface presentation of IcsA chimeras were also significantly greater compared to IcsAp as determined by IF microscopy. Degraded MalE and Bla fusion proteins suggest that some degree of protein degradation occurred during protein export. Nonetheless the Bla protein was highly active as UT5600 (MLRM28 -pMLRM1::bla) and UT5600 (MLRM20 - pMLRM2::bla) showed resistance towards ampicillin.

Overall, the results here suggest that the autotransporter display system can be improved dramatically by fusing proteins to the native passenger domain, providing further evidence for specific interactions between the a and p domain in protein translocation.

Methods

Bacterial strains, plasmids and growth conditions

The strains and plasmids used in this study are listed in Table 1. Escherichia coli K-12 strain DH5a was used for routine cloning and E. coli K-12 strain UT5600 was used as host cells for the surface display of fusion proteins. All bacterial strains were routinely cultured in Luria Bertani (LB). Bacteria were grown in media with antibiotics for 16 h with aeration and then subcultured 1/20 and grown to mid-exponential growth phase by incubation with aeration for 2 h at 37°C. Bacterial strains harbouring pBAD33 and derivatives were cultivated in LB broth with 0.2% (w/v) glucose. Where appropriate, media were supplemented with ampicillin (100 ^g/mL) and chloramphenicol (25 ^g/mL).

Plasmid construction and DNA methods

In a previous study, a NotI site was inserted into icsA, as part of a 15 nucleotide in-frame insertion at aa 87 [icsA, TGCGGCCGCAATGGA: amino acid 88-92, nucleotide 261-270] [29]. The icsA gene was amplified by PCR from pKMRM1 with primers 2156_KpnI (5'-GCCGGTAC-CAACGGAATCTTTTCAGGGG) and REPL.BIO-pBAD30_Sal (5'-CACGCCCTGTCGACTTATTATCA-GAAGG) containing KpnI and SalI sites, respectively. The resulting PCR product was digested with KpnI and SalI, and ligated into the KpnI and SalI sites of the pBAD33 plasmid to produce pMLRM1 (Figure 1).

The icsAp region of icsA was amplified by PCR from pMLRM1 with primers IcsA_764_NotI (5'-CTAGCGG CCGCCTTGTATCTTCAC) and REPL.BIO-pBAD30_Sal containing NotI and SalI sites, respectively. The PCR product was digested with NotI and SalI and ligated between the NotI and SalI sites of the pMLRM1 plasmid to produce pMLRM2. The bla gene was amplified by PCR from

Table 1 Bacterial strains and plasmids

Strain or plasmid' Relevant characteristics

Reference or source

E. coli K-12

MLRM20

MLRM28

MLRM37

MLRM39

RMA3045

UT5600

Plasmids

pBAD33

pKMRM1

pMLRM1

pMLRM2

pMAL-c2

F- 080/acZAM15 A(/acZYA-argF)U169 recAl endA1 hsdR17(rk-, mk+) phoAsupE44 thi-1 gyrA96 re/A1 l-

DH5a [pKMRMl]; AmpR

UT5600 [pBAD33]; CmR

UT5600 [pMLRMl]; CmR

UT5600 [pMLRM2::b/a (NotI - NotI)], CmR

UT5600 [pMLRM1::b/a (NotI - NotI)], CmR

UT5600 [pMLRM2::ma/E (NotI - NotI)], CmR

UT5600 [pMLRM1::ma/E (NotI - NotI)], CmR

E1315 [pMAL-c2], AmpR

F- ara-14 /euB6 secA6 /acYl proC14 tsx-67 0(ompT-fepC)266 entA403 trpE38 rfbDI rpsL109 xy/-5 mt/-1 thi-1 pBAD expression; pACYC184 ori; CmR

icsA (EcoRI - Sa/I) gene cloned into pBR322; medium copy no.; ColE1 ori; AmpR

pIcsA::5aa insertion at amino acid position 87 (icsAi87 or icsA) (b/a); AmpR

icsA (KpnI - Sa/I) gene cloned into pBAD33 (pBAD33::icsA), CmR

icsAß (amino acid 764-1102) (NotI - Sa/I) cloned into pMLRM1 (pBAD33::icsAb); CmR

pMAL-c2 (ma/E); AmpR

Gibco-BRL

This study This study This study This study This study This study

Laboratory collection Laboratory collection

[30] [33] [29]

This study This study

New England Biolabs

a plasmid preparation of pKMRM1 with primers Bla_No-tI_f (5'-CTGGCGGCCGCCACCCAGAAACGC) and Bla_NotI_rev3 (5'-GTCGCGGCCGCCCCAATGCT-TAATCAGTGAG) containing NotI sites. The resulting PCR product was digested with NotI and ligated into the NotI site of the pMLRM1 and pMLRM2 plasmids to produce pMLRM28 and pMLRM20, respectively. The malE gene was amplified by PCR from a plasmid preparation of pMAL-c2 with primers MalE_NotI_f (5'- GACGCG GCCGCAAAATCGAAGAAGGTA) and MalE_NotI_rev5 (5'-CGAGCGGCCGCCAGTCTGCGCGT CTTTC) containing NotI sites. The resulting PCR product was digested with NotI and ligated into the NotI site of the pMLRM1 and pMLRM2 plasmids to produce pMLRM39 and pMLRM37, respectively. DNA sequencing was used to confirm the integrity of the inserts within the constructed plasmids. Isolation of plasmid DNA was carried out with the QIAprep Miniprep kit (Qiagen). Restriction endonu-cleases were used according to the manufacturer's specifications (NEB). General cloning techniques, PCR and DNA sequencing were performed as described previously [29].

Antibodies and antisera

Affinity-purified rabbit polyclonal anti-IcsA antiserum was as described previously [34]. Rabbit a-Bla polyclonal antibody was from Chemicon, and rabbit a-MBP antiserum was from NEB. Anti-IcsA was used at 1:1000, and anti-Bla and anti-MBP were used at 1:10,000.

Measuring growth rate

Ten milliliters 18 h bacterial culture were subcultured (1/20) into LB + 0.2% (w/v) glucose and were grown for

1 h, washed in 2 x 20 mL pre-warmed fresh LB before induction with 0.2% (w/v) arabinose in a side arm flask in an orbital (200 rpm) water bath at 37°C. A spectro-photometer (Spectronic 20 D+, Milton Roy) was used to take absorbance readings at 600 nm every 30 min until the ABS600 reading reached 1.999. The experiment was repeated twice.

SDS-PAGE and western immunoblotting

Ten milliliters bacterial cultures were subcultured (1/20) for 2 h, 37°C in LB + 0.2% (w/v) glucose, washed once in 20 mL fresh LB before induction with 0.2% (w/v) arabinose for 2 h at 37°C. 1 mL culture was centrifuged (13,000 rpm, 1 min, Eppendorf 5415R) and was standardised to 1 x 108 bacteria/mL with 2x sample buffer (0.125 M Tris-HCl, pH 6.8, 4% [w/v] SDS, 20% [v/v] gly-cerol, 10% [v/v] p-mercaptoethanol, 0.04% [w/v] Bromo-phenol blue). SDS-PAGE and Western immunoblotting were carried out as described previously [29]. Molecular weight markers used were BenchMark™ Pre-Stained Protein Ladder (Invitrogen).

Cell fractionation

One hundred milliliters bacterial cultures were subcul-tured (1/20) for 2 h, 37°C in LB + 0.2% (w/v) glucose and were pelleted by centrifugation (8,000 rpm, 15 min, 4°C, Beckman Model J2-21 M). Cultures were washed in 2 x 20 mL fresh LB and induced with 0.2% (w/v) arabinose for 2 h at 37°C. Bacteria were pelleted by centrifugation (8,000 rpm, 15 min, 4°C, Beckman Model J2-21M), pellet washed with 3 mL 50 mM Tris-HCl (pH 8.0) and resus-pended in 10 mL HEPES/1 mM MgCl2. Cultures were

passaged through a French press and cell debris was pelleted by centrifugation (4,500 rpm, 15 min, RT, Sigma 3K15). The supernatant was pelleted by centrifugation (35,000 rpm, 1 h, 4°C, Beckman Coulter Optima™ L-100 Ultracentrifuge). A sample of the pellet (WM - whole membrane) was kept and resuspended in 100 ^L 2x sample buffer. 10 ^L of the supernatant (S/N) (soluble fraction) was mixed with 100 ^L 1x sample buffer. The WM pellet was resuspended in 2.5 mL Triton/MgCl2 and was incubated at RT, shaking for 30 min. The mixture was pelleted by centrifugation (35,000 rpm, 1 h, 4°C, Beck-man Coulter Optima™ L-100 Ultracentrifuge). A sample of the pellet (outer membrane) was kept and resuspended in 100 ^L 2x sample buffer. 10 ^L of the S/N (inner membrane) was mixed with 100 ^L 1x sample buffer. Proteins were subjected to SDS-PAGE and detected via Western immunoblotting with specific antibodies.

Indirect immunofluorescence (IF) of whole bacteria

Ten milliliters bacterial cultures were subcultured (1/20) for 2 h, 37°C in LB + 0.2% (w/v) glucose and were pelleted by centrifugation (13,000 rpm, 1 min, Eppendorf, 5415 R). Cultures were resuspended in 20 mL fresh LB and were pelleted by centrifugation (13,000 rpm, 1 min). Bacterial cultures were resuspended in 10 mL LB + 0.2% (w/v) arabinose and were induced for 2 h at 37°C. 500 ^L of bacteria were pelleted by centrifugation (13,200 rpm, 1 min) and the supernatant discarded. Bacteria were fixed in 500 ^L formalin (3.7% paraformaldehyde in 0.85% saline) for 30 min at room temperature. Fixed bacteria were pelleted by centrifugation (13,000 rpm, 1 min) and were resuspended in 500 ^L phosphate-buffered saline (PBS). Fixed bacteria were pelleted by centrifugation (13,000 rpm, 1 min) and were resuspended in 100 ^L PBS. Sterile cover-slips were placed in a 24-well tray and were incubated with filter sterilised poly-L-lysine (0.01% (w/v) in PBS) for 1 min and was aspirated. The coverslips were washed with PBS before 4 ^L of formalin fixed bacteria were centrifuged onto the cover-slips (2,500 rpm, 5 min, Heraeus Labofuge 400 R). The bacteria were washed twice in PBS and were incubated with the specific primary antibody diluted 1:100 with 10% foetal calf serum (FCS). Bacteria were washed thrice in PBS and incubated with Alexa 488-conjugated donkey anti-rabbit secondary antibody (Molecular Probes) in PBS with 10% FCS. Cover slips were washed thrice in PBS were mounted on glass slides with Mowiol 4-88 (Calbiochem) containing 1 mg/mL />-phenylenediamine (Sigma) and examined with an Olympus IX-70 microscope with phase-contrast optics using a 100x oil immersion objective and a 1.5x enlarger as required. Fluorescence and phase contrast images were false colour merged using the Metamorph software program (version 6.3r7; Molecular Devices).

Ampicillin susceptibility testing

Ten milliliters bacterial cultures were subcultured (1/20) for 2 h, 37°C in LB + 0.2% (w/v) glucose and were pelleted by centrifugation (13,000 rpm, 1 min, Eppendorf, 5415 R). Cultures were resuspended in 20 mL fresh LB and were pelleted by centrifugation (13,000 rpm, 1 min). Bacterial cultures were resuspended in 10 mL LB + 0.2% (w/v) arabinose and were induced for 2 h at 37°C. 100 ^L of bacteria (5 x 107 cells total) were added to the 5 mL soft agar (1:1 LB agar to LB broth, 56°C), swirled before it was overlayed onto pre-warmed LB agar (42°C) and allowed to set. 6 mm sterile Whatman filters were placed onto the agar and were soaked with a 150 ^g ampicillin solution. The filter was allowed to dry before incubation at 37°C for 16 h. The diameter of the zone of inhibition was measured. The experiment was repeated twice.

Statistical analysis

Statistical analysis was carried out using GraphPad Prism 5. Results are expressed as means ± SEM of data obtained in independent experiments. Statistical differences between groups were determined with two-tailed unpaired t-test. Statistical significance was set at p < 0.05.

Additional material

Additional file 1: Figure S1. Expression of IcsA-Bla and IcsAp-Bla does not affect UT5600 growth. The growth of UT5600 strains carrying pMLRM1::b/a (MLRM28) and pMLRM2::b/a (MLRM20) were comparable to pBAD33 (MLRM3) and pMLRM1 (MLRM4). An absorbance reading at 600 nm for each culture was taken every 30 min until the reading reached 1.999. Data are represented as mean ± SEM of three independent experiments.

Acknowledgements

We gratefully acknowledge Dr Marcin Grabowicz for his advice. This work was supported by a Program grant from the NationalHealth and Medical Research Councilof Australia awarded to RM. ML was the recipient of the Australian Postgraduate Award (APA).

Authors' contributions

ML performed allthe experimentalwork and wrote to the manuscript. RM conceived of the study, and participated in its design and coordination and helped to draft the manuscript. Allauthors read and approved the final manuscript.

Competing interests

The authors declare that they have no competing interests.

Received: 17 November 2011 Accepted: 7 February 2012 Published: 7 February 2012

References

1. van Bloois E, Winter RT, Kolmar H, Fraaije MW: Decorating microbes: surface display of proteins on Escherichia coli. Trends Biotechnol 2011, 29:79-86.

2. Henderson IR, Navarro-Garcia F, Desvaux M, Fernandez RC, Ala'Aldeen D: Type V protein secretion pathway: the Autotransporter story. Microbiol Mol Biol Rev 2004, 68:692-744.

3. Jose J, Bernhardt R, Hannemann F: Functional display of active bovine adrenodoxin on the surface of E. coli by Chemical Incorporation of the [2Fe-2S] Cluster. Chem Bio Chem 2001, 2:695-701.

4. Jose J, Betscheider D, Zangen D: Bacterial surface display library screening by target enzyme labeling: Identification of new human cathepsin G inhibitors. Anal Biochem 2005, 346:258-267.

5. Maurer J, Jose J, Meyer T: Autodisplay: one-component system for efficient surface display and release of soluble recombinant proteins from Escherichia coli. J Bacteriol 1997, 179:794-804.

6. Kj^rgaard K, Hasman H, Schembri MA, Klemm P: Antigen 43-mediated Autotransporter display, a versatile bacterial cell surface presentation system. J Bacteriol 2002, 184:4197-4204.

7. Klauser T, Pohlner J, Meyer TF: Selective extracellular release of cholera toxin B subunit by Escherichia coli: dissection of Neisseria Iga betamediated outer membrane transport. EMBO J 1992, 11:2327-2335.

8. Kramer U, Rozos K, Apfel H, Autenrieth IB, Lattemann CT: Autodisplay: development of an efficacious system for surface display of antigenic determinants in Salmonella vaccine strains. Infect Immun 2003, 71:1944-1952.

9. Jose J, Kramer J, Klauser T, Pohlner J, Meyer TF: Absence of periplasmic DsbA oxidoreductase facilitates export of cysteine-containing passenger proteins to the Escherichia coli cell surface via the Iga beta autotransporter pathway. Gene 1996, 178:107-110.

10. Kadokura H, Beckwith J: Four cysteines of the membrane protein DsbB act in concert to oxidize its substrate DsbA. EMBO J 2002, 21:2354-2363.

11. Veiga E, De Lorenzo V, Fernandez LA: Structural tolerance of bacterial autotransporters for folded passenger protein domains. Mol Microbiol 2004, 52:1069-1080.

12. Lett MC, Sasakawa C, Okada N, Sakai T, Makino S, Yamada M, Komatsu K, Yoshikawa M: virG, a plasmid-coded virulence gene of Shigella flexneri: identification of the virG protein and determination of the complete coding sequence. J Bacteriol 1989, 171:353-359.

13. Makino S, Sasakawa C, Kumata K, Kurata T, Yishikawa M: A genetic determinant required for continuous reinfection of adjacent cells on large plasmid in S. flexneri 2a. Cell 1986, 46:551-555.

14. Goldberg MB: Actin-based Motility of Intracellular Microbial Pathogens. Microbiol Mol Biol Rev 2001, 65:595-626.

15. Suzuki T, Lett MC, Sasakawa C: Extracellular Transport of VirG Protein in Shigella. J Biol Chem 1995, 270:30874-30880.

16. Stevens JM, Galyov EE, Stevens MP: Actin-dependent movement of bacterial pathogens. Nat Rev Microbiol 2006, 4:91-101.

17. Suzuki T, Lett MC, Sasakawa C: Extracellular transport of VirG Protein in Shigella. J Biol Chem 1995, 270:30874-30880.

18. Brandon LD, Goehring N, Janakiraman A, Yan AW, Wu T, Beckwith J, Goldberg MB: IcsA, a polarly localized autotransporter with an atypical signal peptide, uses the Sec apparatus for secretion, although the Sec apparatus is circumferentially distributed. Mol Microbiol 2003, 50:45-60.

19. Charles M, Pérez M, Konil JH, Goldberg MB: Polar targeting of Shigella virulence factor IcsA in Enterobacteriacae and Vibrio. Proc Natl Acad Sci USA 2001, 98:9871-9876.

20. Brandon LD, Goldberg MB: Periplasmic Transit and Disulfide Bond Formation of the Autotransporter Shigella Protein IcsA. J Bacteriol 2001, 183:951-958.

21. Oomen CJ, van Ulsen P, van Gelder P, Feijen M, Tommassen J, Gros P: Structure of the translocator domain of a bacterial autotransporter. EMBO J 2004, 23:1257-1266.

22. Barnard TJ, Dauntin N, Lukacik P, Bernstein HD, Buchanan SK: Autotransporter structure reveals intra-barrel cleavage followed by conformational changes. Nat Struct Mol Biol 2007, 4:1214-1220.

23. van den Berg B: Crystal Structure of a Full-Length Autotransporter. J Mol Biol 2010, 396:627-633.

24. Egile C, d'Hauteville H, Parsot C, Sansonetti PJ: SopA, the outer membrane protease responsible for polar localization of IcsA in Shigella flexneri. Mol Microbiol 1997, 23:1063-1073.

25. Shere KD, Sallustio S, Manessis A, D'Aversa TG, Goldberg MB: Disruption of IcsP, the major Shigella protease that cleaves IcsA, accelerates actin-based motility. Mol Microbiol 1997, 25:451-462.

26. Fukuda I, Suzuki T, Munakata H, Hayashi N, Katayama E, Yoshikawa M, Sasakawa C: Cleavage of Shigella surface protein VirG occurs at a specific site, but the secretion is not essential for intracellular spreading. J Bacteriol 1995, 177:1719-1726.

27. Sauri A, Oreshkova N, Soprova Z, Jong WS, Sani M, Peters PJ, Lulrlnk J, van Ulsen P: Autotransporter ß-Domains have a specific function in protein secretion beyond outer-membrane targeting. J Mol Biol 2011, 412:553-567.

28. Dautin N, Bernstein HD: Residues in a Conserved a-Helical Segment are required for cleavage but not secretion of an Escherichia coli serine protease autotransporter passenger domain. J Bacteriol 2011, 193:3748-3756.

29. May KL, Morona R: Mutagenesis of the Shigella flexneri autotransporter IcsA reveals novel functional regions involved in IcsA biogenesis and recruitment of host neural Wiscott-Aldrich syndrome protein. J Bacterio 2008, 190:4666-4676.

30. Guzman LM, Belin D, Carson MJ, Beckwith J: Tight regulation, modulation, and high-level expression by vectors containing the arabinose PBAD promoter. J Bacteriol 1995, 177:4121-4130.

31. Lattemann CT, Maurer J, Gerland E, Meyer TF: Autodisplay: functional display of active beta-lactamase on the surface of Escherichia coli by the AIDA-I autotransporter. J Bacteriol 2000, 182:3726-3733.

32. Rutherford N, Charbonneau ME, Berthiaume F, Betton JM, Mourez M: The periplasmic folding of a cysteineless autotransporter passenger domain interferes with its outer membrane translocation. J Bacteriol 2006, 188:4111-4116.

33. Morona R, Van Den Bosch L: Lipopolysaccharide O antigen chains mask IcsA (VirG) in Shigella flexneri. FEMS Microbiol Lett 2003, 221:173-180.

34. Van Den Bosch L, Manning PA, Morona R: Regulation of O-antigen chain length is required for Shigella flexneri virulence. Mol Microbiol 1997, 23:765-775.

doi:10.1186/1475-2859-11-20

Cite this article as: Lum and Morona: IcsA autotransporter passenger promotes increased fusion protein expression on the cell surface. Microbial Cell Factories 2012 11:20.

Submit your next manuscript to BioMed Central and take full advantage of:

• Convenient online submission

• Thorough peer review

• No space constraints or color figure charges

• Immediate publication on acceptance

• Inclusion in PubMed, CAS, Scopus and Google Scholar

• Research which is freely available for redistribution

Submit your manuscript at www.biomedcentral.com/submit

Central