Skip to main page content

• HOME
• Current Issue
• Archive
• Alerts
• About ASM
• Contact us
• Tech Support
• Journals.ASM.org

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Hoischen, C. Articles by Fahnert, B. Search for Related Content PubMed PubMed Citation Articles by Hoischen, C. Articles by Fahnert, B. Agricola Articles by Hoischen, C. Articles by Fahnert, B.

 Previous Article  |  Next Article 

Applied and Environmental Microbiology, February 2002, p. 525-531, Vol. 68, No. 2
0099-2240/02/$04.00+0     DOI: 10.1128/AEM.68.2.525-531.2002
Copyright © 2002, American Society for Microbiology. All Rights Reserved.

Novel Bacterial Membrane Surface Display System Using Cell Wall-Less L-Forms of Proteus mirabilis and Escherichia coli

Christian Hoischen,^1* Christine Fritsche,¹ Johannes Gumpert,¹ Martin Westermann,² Katleen Gura,¹ and Beatrix Fahnert³

Department of Molecular Biology, Institute of Molecular Biotechnology,¹ Department of Applied Microbiology, Hans-Knöll-Institute for Natural Products Research, D-07745 Jena,³ Department of Ultrastructure Research, Friedrich-Schiller-University Jena, D-07740 Jena, Germany²

Received 12 June 2001/ Accepted 5 November 2001

Top Abstract Introduction Materials and Methods Results Discussion References

 
We describe a novel membrane surface display system that allows the anchoring of foreign proteins in the cytoplasmic membrane (CM) of stable, cell wall-less L-form cells of Escherichia coli and Proteus mirabilis. The reporter protein, staphylokinase (Sak), was fused to transmembrane domains of integral membrane proteins from E. coli (lactose permease LacY, preprotein translocase SecY) and P. mirabilis (curved cell morphology protein CcmA). Both L-form strains overexpressed fusion proteins in amounts of 1 to 100 µg ml^-1, with higher expression for those with homologous anchor motifs. Various experimental approaches, e.g., cell fractionation, Percoll gradient purification, and solubilization of the CM, demonstrated that the fusion proteins are tightly bound to the CM and do not form aggregates. Trypsin digestion, as well as electron microscopy of immunogold-labeled replicas, confirmed that the protein was localized on the outside surface. The displayed Sak showed functional activity, indicating correct folding. This membrane surface display system features endotoxin-poor organisms and can provide a novel platform for numerous applications.

Top Abstract Introduction Materials and Methods Results Discussion References

 
The overexpression of recombinant proteins, which remain bound to the outer surface of the bacterial cells as accessible and functional active molecules, offers new applications in biotechnology and medicine. Among these are the development of diagnostics and vaccines, adhesin-receptor interaction studies, the generation of peptide libraries, the immobilization of enzymes, and the expression of heavy metal-binding peptides and antibody fragments (6, 10, 25, 33). The surface display systems follow various strategies for anchoring. In gram-negative bacteria, outer membrane proteins (21), pili and flagella (26), modified lipoproteins (6, 7, 11), ice nucleation proteins (17, 23), and autotransporters (19, 27, 35) have been used as anchors. In gram-positive bacteria, surface anchors have been derived from lipoproteins, cell wall proteins, or S-layer proteins (24, 32, 33). However, depending on the displayed protein and the desired application, each system has its own advantages and disadvantages, such as the size limitation of the displayed protein, mislocalization or formation of inclusion bodies, association with lipopolysaccharides (LPS), or destabilization of the outer membrane (10, 17, 22). There is still a need for further developments to increase the repertoire of applications of surface display systems (21).

Stable protoplast type L-form bacteria have up to now not been considered as a system for surface display, although they exhibit several interesting features for such applications. They have lost irreversibly the ability to form cell wall structures and periplasmic compartments, and their cells are surrounded only by a cytoplasmic membrane (CM). Cell biological properties of the strains and the lipid composition of their CM are well characterized (13, 16). Furthermore, the L-form strains of Proteus mirabilis LVIWEI and Escherichia coli LWF+WEI have been used for the efficient overexpression of numerous recombinant proteins (5, 12, 20, 30). In contrast to surface display systems with walled bacteria, both L-form strains lack extracellular proteolytic activities (12). Moreover, E. coli LWF+WEI does not synthesize any endotoxic LPS, and only small quantities of LPS were found in P. mirabilis LVIWEI (O. Holst, unpublished data). Due to the lack of a cell wall, the L-form membrane is easy to isolate and to handle.

We describe here a novel membrane surface display system based on the L-form strains. Staphylokinase (Sak), a plasminogen activator with potential for medical application, was chosen as reporter protein. Sak is functionally and structurally well characterized (31), and it has been shown in previous studies to be overexpressed and secreted by L-form cells as a soluble protein (12). Transmembrane domains from integral membrane proteins of E. coli (lactose permease LacY, preprotein translocase SecY) and P. mirabilis (curved cell morphology protein CcmA) were selected as membrane anchors. The construction and the expression of the fusion proteins (membrane anchor and Sak) in the L-form strains were demonstrated. The fusion proteins were localized in the CM as surface displayed and functionally active products.

Top Abstract Introduction Materials and Methods Results Discussion References

 
Bacterial strains, growth conditions, and protein expression.
E. coli DH5 was used for cloning and construction of vectors. DH5 and P. mirabilis VI were used for genomic PCR. Both were cultivated in Luria-Bertani (LB) medium or on LB agar plates. If necessary, the media were supplemented with the appropriate antibiotics (2).

Surface display of Sak was examined in stable protoplast-type L-form strains of P. mirabilis LVIWEI and E. coli LWF+WEI (13; J. Gumpert, C. Hoischen, M. J. Kujau, C. Fritsche, G. Elske, B. Fahnert, S. Sieben, and H. Müller, German patent FN 100 11 358.3 [pending]). Cultivation in L-form standard (LFS) medium, transformation, selection of transformants, and adaptation to growth in liquid medium were carried out as described previously (14). For the expression of fusion proteins, 100-ml bottles containing 30 ml of LFS medium, supplemented with kanamycin (10 to 20 µg ml^-1), were inoculated with 1 to 2 ml of an overnight culture. After 3 h of cultivation on a rotary shaker (220 rpm, 26°C) at an optical density at 540 nm of 0.4 to 0.8, protein expression was induced by the addition of IPTG (isopropyl-ß-D-thiogalactopyranoside) to a final concentration of 1 mM. Cells were generally analyzed for the expression of fusion proteins 20 h after induction.

Construction of expression plasmids.
The DNA cloning procedures were performed by standard methods (2). For the fusion of Sak with the relevant transmembrane protein domains, a vector, pF003, was constructed. This vector was derived from plasmid pGEX-4T-2 (Amersham Pharmacia Biotech), which was digested with FspI-SapI and from a modified version (M. Hartmann, unpublished data) of pMEX6 (U.S. Biochemicals), which was cleaved with Eco47III-SapI. The isolated fragments FspI-SapI containing the lacI^q gene and fragment Eco47III-SapI containing promoter P tac, the terminator of the rrnB operon (rrnBT), an ampicillin resistance region (bla), and the ColE1ori were ligated, resulting in plasmid pBF003 (4,245 bp). After ScaI digestion of the single ScaI restriction site of pBF003 and treatment of the blunt ends with alkaline phosphatase, a SmaI fragment from vector pACK02sckan (20) was inserted, which contained a kanamycin resistance region (kan). pF003 harbored a multicloning site containing the restriction sites NdeI, PstI, and HindIII for insertion of expression cassettes under the control of the tac promoter. The DNA sequence coding for the mature Sak (amino acids 28 to 163) without signal peptide (3) was amplified by overlap PCR with the upstream primer 5"-CGCCTGCAGTCAAGTTCATTCGACAAAGGA-3" and the downstream primer 5"-CGCCTGCAGTCAAGTTCATTCGACAAAGGA-3" from vector pSIS2 (S. Sieben, unpublished data). After PstI-HindIII restriction of the PCR product, the resulting fragment was cloned into the PstI-HindIII fragment of pF003, resulting in plasmid pFsak. The introduction of the PstI site at the 5" end caused the insertion of two linker amino acids (Leu-Gln) at the N terminus of Sak. The coding DNA of the following transmembrane domains of membrane proteins was amplified by overlap PCR from genomic DNA: helix 1 (amino acids 1 to 41; LacYH1) and helices 1 to 3 (amino acids 1 to 101; LacYH1-3) of E. coli LacY (8, 18), helix 1 (amino acids 1 to 74; SecYH1) and helices 1 to 3 (amino acids 1 to 153; SecYH1-3) of E. coli SecY (1), and helix 1 (amino acids 1 to 34; CcmAH1) of the P. mirabilis CcmA (15). The following upstream primers were used: 5"-GGAATTCCATATGTACTATTTAAAAAACACAAACTTTT-3" for LacYH1 and LacYH1-3, 5"-CGCGCATATGGCTAAACAACCGGGATTAGATTTCAAA-3" for SecYH1 and SecYH1-3, and 5"-CGCCATATGGATAATAAGCGAACACAGCGG-3" for CcmAH1. The following downstream primers were used: 5"-CGCCTGCAGGCTGATATGGTTGATGTCATG-3" for LacYH1, 5"-CGCCTGCAGGTATTGTAACAGTGGCCCGAAGATAAA-3" for LacYH1-3, 5"-CGCCTGCAGACGGCTGAGAGCACCACCAGAGAACATGTT-3" for SecYH1, 5"-CGCCTGCAGGCCCGGGTTAATCACCAGGCCTTGCATACC-3" for SecYH1-3, and 5"-CGCCTGCAGTTCTCTGGTGGTGCTCTCACCAGCTATTTGATATCG-3" for CcmAH1. The primers introduced to all amplified DNA fragments an NdeI site at the 5" end and a PstI site at the 3" end, respectively. In addition, a DNA sequence coding for a linker of five amino acids (Ser-Thr-Thr-Arg-Glu) was introduced to the 5" end of CcmAH1. After the NdeI-PstI restrictions, the amplified fragments were cloned into the NdeI-PstI fragment of pFsak, resulting in the expression plasmids pFLacYH1-Sak, pFLacYH1-3-Sak, pFSecYH1-Sak, pFSecYH1-3-Sak, and pFCcmAH1-Sak.

Fractionation of cells.
Cells were separated from growth medium by centrifugation at 14,000 rpm in an Eppendorf centrifuge (5415C) for 10 min. The supernatants were directly used for the detection of soluble proteins. The pellets containing the cells were washed once in sucrose (0.4 M) and resuspended in an equal volume of buffer A (20 mM Tris-HCl, 15 mM MgCl₂; pH 7.2). For membrane preparation, the washed cells were disrupted by ultrasonication with a Branson 250 Sonifier (50 kHz, two 2-min intervals; Emerson Technologies, Dietbach, Germany). Membranes were isolated by ultracentrifugation (100,000 x g, 30 min). The resulting supernatants, which were almost depleted in membranes and lipids, represented the cytosolic fraction. It should be noted that the 100,000 x g supernatant can still contain certain amounts of membrane enzymes suspended as small particles. Membranes were used for analysis after two washes in buffer A and resuspension in the initial volume of buffer A.

Alternatively, the membranes obtained by ultracentrifugation were further purified by Percoll gradient ultracentrifugation (28) in buffer A containing 30% (vol/vol) Percoll. Membranes from 10 ml of a CcmAH1-Sak-expressing P. mirabilis LVIWEI culture were isolated as described above and resuspended in 28 ml of buffer A supplemented with 30% (vol/vol) Percoll. After centrifugation (30 min, 15,000 rpm) in a Beckman 50.2Ti rotor at 4°C, a self-generating buoyant density gradient was established. Eight fractions with increasing buoyant densities were collected by pipetting 3.5-ml aliquots from the top of the centrifugation tubes. The NADH-oxidase activity of the resulting fractions was monitored at 30°C in buffer A supplemented with 1 mM dithiothreitol and 0.2 mM NADH₂ at 340 nm by using a Spekol 1100 spectrophotometer (Zeiss, Jena, Germany).

For solubilization of the fusion proteins, membranes or cells from 1 ml of culture were washed and resuspended in 1 ml of buffer A supplemented with 0.5 or 1% (vol/vol) Triton X-100 and then gently shaken for 60 min at 30°C. The solution was subsequently centrifuged at 14,000 x g for 30 min to separate insoluble material.

Protein methods.
Proteins of culture supernatants, washed cells, cytosolic fractions, and washed membranes were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) on a 15% polyacrylamide gel and then visualized by Coomassie brilliant blue. For Western blot analysis, the gels were electroblotted to polyvinylidene difluoride membranes (Millipore, Bedford, Mass.). To compare the amounts of fusion proteins in the fractions, all samples were equalized for the same volumes of the initial culture. The amounts of applied proteins per lane were determined by the method of Bradford (4). The blotted membranes were incubated in TTBS (Tris-HCl-buffered saline [TBS] containing 1% [vol/vol] Tween 20) supplemented with 5% (wt/vol) skim milk and then washed four times with TTBS (2). For immunostaining the membranes were incubated in TTBS containing purified diluted (1:2,500) rabbit anti-Sak as primary antibody. The membranes were rinsed four times with TTBS prior to the addition of the secondary alkaline phosphatase-conjugated anti-rabbit antibody (Bio-Rad) diluted 1:5,000 in TTBS. Antigen-antibody conjugates were visualized by the color reaction recommended by the manufacturer. The intensity of the bands was determined by densitometry with Molecular Analyst Software (Bio-Rad).

Functional activity of Sak.
In the standard assay for soluble Sak, 50-µl aliquots of Sak-containing samples and defined standards of purified Sak (1 to 100 µg ml^-1) were applied to holes (9 mm in diameter) in agar layers (2% [wt/vol]), which contained skim milk (1% [wt/vol]) and plasminogen (10 µg ml^-1), and then incubated for 3 to 8 h at 37°C. The Sak molecules activate plasminogen, which causes digestion of milk casein and the formation of clearing zones (3). To determine the amount of functional active Sak, the diameters of the clearing zones were compared with those of the standards. However, it was impossible to quantify the activity of the cell-bound Sak with the standard assay because whole cells could not diffuse freely in agar layers. To facilitate the diffusion, the CM was solubilized with Triton X-100 (0.5% [vol/vol]) prior to the functionality assays. The obtained values demonstrate that the fusion proteins were functionally active. The determined functional activities corresponded to the activity of soluble purified Sak, but they did not reflect an accurate estimation of the activity because even in the presence of a detergent the fusion proteins still have lipids attached to their membrane anchors, which might hinder the diffusion in the agar layer.

Trypsin digestion of displayed fusion proteins.
Culture aliquots of 900 µl were mixed with 100 µl of buffer B (200 mM Tris-HCl, 150 mM MgCl₂; pH 7.5) and supplemented with trypsin to a final concentration of 2 mg ml^-1. After incubation for 60 min at 37°C, trypsin inhibitor (8 mg ml^-1) was added, and the samples were analyzed by SDS-PAGE and Sak-specific immunoblotting. To digest nonaccessible (intracellular) fusion protein, aliquots were treated with ultrasonication or 0.5% (vol/vol) Triton X-100 prior to trypsin digestion.

Freeze-fracturing, replica immunogold-labeling, and electron microscopy.
At a temperature of 30°C, cells were separated from growth medium, washed, and resuspended in 0.4 M sucrose (10% of the initial volume). Aliquots were enclosed between two 0.1-mm copper profiles as used for the sandwich double-replica technique. The sandwiches were rapidly frozen in liquid propane cooled by liquid nitrogen. Freeze fracturing was performed in a BAF400T freeze-fracture unit (BAL-TEC; Liechtenstein) at -150°C by using a double-replica stage. The samples were shadowed without etching and with a 2- to 2.5-nm-thick layer of Pt/C at an angle of 35°.

For immunogold labeling the replicas were transferred to an SDS digestion solution (2.5% [wt/vol] SDS in 10 mM Tris-HCl buffer [pH 8.3] and 30 mM sucrose) and incubated overnight at room temperature (9). After four washes in TBS and 30 min of incubation in TBS plus 1% (wt/vol) bovine serum albumin (TBS-1% BSA), the replicas were incubated for 1 h in TBS-0.5% BSA plus Sak-specific primary antibody (diluted 1:50). Subsequently, the replicas were washed four times in TBS and incubated for 1 h in TBS-0.5% BSA plus secondary antibody (1:50-diluted goat anti-rabbit immunoglobulin G conjugated with 10-nm gold [British Biocell International, Cardiff, United Kingdom]). After four washes in TBS and fixation with 0.5% (vol/vol) glutaraldehyde in TBS, the replicas were washed twice in distilled water and picked onto Formvar-coated grids for observation in an EM902 electron microscope (Zeiss, Oberkochen, Germany).

Top Abstract Introduction Materials and Methods Results Discussion References

 
Construction of fusion proteins and expression in L-forms.
In comparison to currently used surface display systems, which anchor the exposed proteins in the outer membrane of gram-negative bacteria or in the cell wall of gram-positive bacteria, the strategy employed here was the anchoring in the CM of stable protoplast type L-forms (Fig. 1). Transmembrane domains of integral proteins of the CM were used as membrane anchors, i.e., helix 1 (LacYH1) or helices 1 to 3 (LacYH1-3) of E. coli LacY (8, 18), helix 1 (SecYH1) or helices 1 to 3 (SecYH1-3) of E. coli SecY (1), and helix 1 (CcmAH1) of P. mirabilis CcmA (15). The reporter protein Sak (31) was fused to the C termini of external regions of the various anchor motifs.

View larger version (49K): [in this window] [in a new window]

FIG. 1. Schematic illustration of the L-form membrane surface display system and commonly used bacterial systems. In endotoxin-poor L-form cells the protein is anchored in the CM by transmembrane domains of integral proteins of the CM, whereas in gram-negative bacteria the proteins are anchored in the outer membrane or bound to surface components such as pili or flagella. For surface display in gram-positive bacteria, the proteins are bound to protein components of the cell wall.

We intended to express all five constructs in the L-form strains E. coli LWF+WEI and P. mirabilis LVIWEI by using expression plasmids that bear the fusion proteins under the control of the tac promoter (Fig. 2A). In E. coli LWF+WEI, the fusion proteins LacYH1-Sak, LacYH1-3-Sak, and SecYH1-Sak were overexpressed (Fig. 2B), whereas CcmAH1-Sak with a membrane anchor derived from P. mirabilis and SecYH1-3-Sak were not expressed (data not shown). P. mirabilis LVIWEI overexpressed CcmAH1-Sak with a homologous membrane anchor and SecYH1-Sak with an anchor derived from E. coli (Fig. 2B). The molecular masses of the fusion proteins corresponded to the values calculated from the amino acid sequences (LacYH1-Sak, 20.5 kDa; LacYH1-3-Sak, 27.9 kDa; SecYH1-Sak, 23.7 kDa; CcmAH1-Sak, 20.5 kDa). The growth rate of the cells was not affected by the expression of the fusion protein, and the cells remained viable during prolonged cultivation over 48 h (data not shown).

View larger version (33K): [in this window] [in a new window]

FIG. 2. Principle of vector constructs and expression of fusion proteins. (A) Plasmid pFCcmAH1-Sak illustrates the expression vectors used in this study. It carries DNA sequences encoding for ColE1 origin, lacIq repressor, kanamycin resistance (kan), terminator (rrnBT), and the fusion protein (CcmAH1-Sak) under control of the tac promoter (P tac). (B) Synthesis of fusion proteins by L-form cells. Shown are a Sak-specific immunoblot of purified mature Sak (lane 1); washed cells of E. coli LWF+WEI expressing LacYH1-Sak (lane 2), LacYH1-3-Sak (lane 3), and SecYH1-Sak (lane 4); and cells of P. mirabilis LVIWEI expressing CcmAH1-Sak (lane 5) and SecYH1-Sak (lane 6). The following total protein amounts were applied: 2 µg, lane 1; 31 µg, lane 2; 26 µg, lane 3; 32 µg, lane 4; 27 µg, lane 5; and 35 µg, lane 6.

Quantification of fusion proteins.
Synthesized fusion proteins were quantified by densitometry of Sak-specific immunostained Western blots. Different amounts of overexpressing cells and defined amounts of purified Sak were blotted, and the signal intensities were compared. If we assume that the fusion proteins and the soluble reference protein were transferred and detected equally, the following approximate molar yields and corresponding weight yields were calculated: (i) <1 µg ml^-1 for SecYH1-Sak, 1 to 5 µg ml^-1 for LacYH1-3-Sak, and 5 to 15 µg ml^-1 for LacYH1-Sak in E. coli LWF+WEI and (ii) <1 µg ml^-1 for SecYH1-Sak and 70 to 100 µg ml^-1 for CcmAH1-Sak in P. mirabilis LVIWEI.

Detection of fusion proteins in cells and membranes.
The fusion proteins were determined in the culture supernatant, in the washed cells, in the membrane fraction, and in the 100,000 x g supernatant representing the cytosol. Figure 3A compares the band intensities of CcmAH1-Sak in the fractions of P. mirabilis LVIWEI. The total recombinant CcmAH1-Sak was cell bound, and only traces were found in the culture supernatant. The band intensity of the purified membranes was equal to the signal of the total cell-bound fusion protein, and only minor amounts were detected in the 100,000 x g supernatant. This demonstrates the exclusive location of CcmAH1-Sak in the CM. In the case of LacYH1-Sak and SecYH1-Sak expressed in E. coli LWF+WEI, 80 to 90% of the total fusion protein was located in the membrane fraction, and the remaining 10 to 20% was found in the culture supernatant (data not shown). LacYH1-3-Sak was located exclusively in the membrane fractions.

View larger version (55K): [in this window] [in a new window]

FIG. 3. Localization of CcmAH1-Sak synthesized by P. mirabilis LVIWEI. (A) Sak-specific immunoblot of culture supernatant (lane 1), washed cells (lane 2), 100,000 x g supernatant representing the cytosol (lane 3), and washed membranes (lane 4). The following protein amounts were applied: 18 µg, lane 1; 58 µg, lane 2; 31 µg, lane 3; and 24 µg, lane 4. (B) Density gradient centrifugation of washed membranes. Eight fractions with increasing buoyant densities were collected. The densities of the eight fractions (•) and the corresponding relative amounts of totally recovered NADH-oxidase activity (), which serves as the marker enzyme for the CM, are presented. (C) Sak-specific immunoblot analysis of the eight fractions collected after density gradient centrifugation. Equal volumes of each fraction were applied, and, therefore, the intensity of the immunostained CcmAH1-Sak bands represents the amounts of fusion protein in each fraction. The majority of CcmAH1-Sak is located in fraction 2 (density, 1.03 g ml-1). This finding correlates with the occurrence of NADH-oxidase activity, which was also mainly found in fraction 2 (70% of total activity). (D) Coomassie blue-stained SDS-PAGE gels of the eight fractions. The arrow indicates CcmAH1-Sak in lane 2 that corresponds to fraction 2. Equal volumes of each fraction were applied.

Overexpressed membrane proteins or proteins with hydrophobic domains often tend to form insoluble self-aggregates of high density when they are not properly integrated into the CM (34). These aggregates cannot be separated from cells or membranes by conventional centrifugation. Therefore, the membranes, which were prepared by conventional ultracentrifugation, were purified further by Percoll density gradient centrifugation (Fig. 3B, C, and D). Among the eight fractions collected after centrifugation, fraction 2, with a buoyant density of 1.03 g ml^-1, contained ca. 70% of the total recovered NADH-oxidase activity, a marker enzyme for the CM (Fig. 3B). This demonstrates a concentration of the CM in fraction 2, which also contained by far the majority of the total CcmAH1-Sak, as shown by immunoblot analysis of the fractions (Fig. 3C). The corresponding Coomassie blue-stained SDS-PAGE gels showed that CcmAH1-Sak is a major protein of the purified CM fraction (Fig. 3D). The similar band patterns of the different fractions in Fig. 3D can be explained by the fact that the applied membranes consist of vesicles and membrane fragments of different sizes. The membranes are not completely concentrated in one fraction, and certain minor populations with different sedimentation properties, which cannot be concentrated by prolonged centrifugation, are also found in the other fractions. A second reason might be the occurrence of proteins, which are loosely bound to the membrane. They can be released during the membrane preparations and are found evenly distributed in the gradient (28).

Another indication for the localization of fusion proteins in the CM is their solubilization by detergents. It was found that all expressed fusion proteins could be solubilized from washed membranes or cells by using the nonionic detergent Triton X-100 (data not shown). These findings show that the fusion proteins are embedded in the lipid bilayer of the membrane.

Digestion by trypsin.
Trypsin does not penetrate or destroy L-form cells (data not shown). Therefore, the proteolytic digestion of fusion proteins was used as an indication of their surface exposure and accessibility. Western blot analysis of trypsin-treated and untreated whole cells of CcmAH1-Sak-expressing P. mirabilis LVIWEI revealed that the fusion protein disappeared almost completely after trypsin treatment (Fig. 4), indicating that it was located on the outside surface of the cells. Analysis of the signal intensities revealed that >95% of the Sak signal disappeared after trypsin treatment. The fusion protein, which was not accessible to trypsin, could be digested when the cells were sonicated or solubilized with Triton X-100 prior to trypsin addition (Fig. 4). Comparable results were obtained with E. coli LWF+WEI expressing LacYH1-Sak, SecYH1-Sak, and LacYH1-3-Sak (data not shown).

View larger version (48K): [in this window] [in a new window]

FIG. 4. Accessibility of CcmAH1-Sak for trypsin digestion. The Sak-specific immunoblot shows cells of P. mirabilis LVIWEI expressing CcmAH1-Sak: without trypsin treatment (lane 1), after treatment with trypsin (lane 2), after ultrasonication prior to trypsin treatment (lane 3), and after Triton X-100 addition prior to trypsin digestion (lane 4). To lane 1 was added 28 µg of protein.

Localization of CcmAH1-Sak by electron microscopy.
The surface display was verified further by freeze-fracturing, replica immunogold labeling, and electron microscopy. It was demonstrated by using P. mirabilis LVIWEI that CcmAH1-Sak was localized on the exoplasmic fracture (EF) face of the CM. A large number of gold particles indicating Sak molecules were found on this EF face (Fig. 5B), whereas almost no gold particles were located on the protoplasmic fracture (PF) face (Fig. 5D). Cross-fractures (Fig. 5C) showed only a few particles in the cytoplasm, but a high concentration of Sak molecules was observed in defined areas on the outside of the CM. The anchored proteins were clustered in domains of the CM (Fig. 5B), which were not found in nonexpressing cells (Fig. 5A). The background labeling in cells without expression plasmid was rather low (Fig. 5A).

View larger version (144K): [in this window] [in a new window]

FIG. 5. Localization of CcmAH1-Sak in P. mirabilis LVIWEI by freeze-fracture, replica immunogold labeling, and electron microscopy. (A) Cell without plasmid and CcmAH1-Sak. Intramembrane particles on the EF faces of the plasma membrane are randomly distributed, and the background level of 10-nm immunogold particles (arrowhead) is very low. (B) EF face of a CcmAH1-Sak-expressing cell. Numerous Sak-indicating gold particles are seen (arrowhead), which are almost exclusively found in areas of clustered intramembrane particles. The inset shows a more detailed view of labeled Sak and particle clusters. (C) Cross-fracture view of a Sak-expressing cell. The labeling density of Sak in the cytoplasm is very low in contrast to the high density in the region of the CM (arrowhead). (D) PF face of a Sak-expressing cell. Almost no gold particles are located in the PF. The arrowhead indicates structures corresponding to the particle clusters on the EF faces. Bar, 100 nm.

The distribution of gold particles was determined for the various surfaces of expressing cells and also for the nonexpressing cells. For nonexpressing cells, an average of three particles per square micrometer was determined (minimum of two, maximum of four) out of five frames. This value represents the unspecific background. A total of 15 frames from nine expressing cells were analyzed, showing an average density of 670 particles µm^-2 (minimum of 520, maximum of 960) in the domains containing protein clusters of the EF. Outside of the clusters was found an average of 10 particles µm^-2 (minimum of 4, maximum of 20), as calculated for 20 frames from 15 cells. Analysis of 20 frames from 15 cells resulted in an average of 7 particles µm^-2 (minimum of 4, maximum of 16) on the PF.

Functional activity of Sak.
In order to test whether the membrane-bound fusion proteins were correctly folded, the functional activity of Sak was determined. A modified version of the standard assay, which is based on the diffusion of Sak in agar layers, was used. After solubilization of the CM from the various L-form cells the functional activities corresponded to at least 40 µg of soluble Sak per ml of culture for CcmAH1-Sak expressed in P. mirabilis LVIWEI and 1 to 5 µg ml^-1 for LacYH1-Sak and LacYH1-3-Sak expressed in E. coli LWF+WEI. The functional activities determined for SecYH1-Sak and SecYH1-3-Sak expressed in E. coli LWF+WEI and in P. mirabilis LVIWEI were below values sufficient for quantification.

Top Abstract Introduction Materials and Methods Results Discussion References

 
Two novel approaches for bacterial surface display have been introduced: (i) protoplast type, endotoxin-poor L-form bacteria, which lack the cell wall and the periplasmic compartment (13), and (ii) transmembrane domains of integral proteins of the CM as membrane anchors. Integral membrane proteins from the CM or their transmembrane domains have not hitherto been successfully used as anchor motifs. In gram-negative bacteria the proteins are usually anchored in the outer membrane or bound to surface proteins (6, 10, 25, 33) (Fig. 1). In gram-positive bacteria there have been attempts to anchor proteins in the CM by fusion to transmembrane domains; however, the free accessibility of the fusion proteins from the outside could not be demonstrated due to the cell wall structure (24).

Transmembrane domains (helix 1 and helices 1 to 3) derived from membrane proteins of E. coli (LacY and SecY) and of P. mirabilis (CcmA) were used as membrane anchors, and Sak was used as a fusion partner. The fusion proteins LacYH1-Sak, LacYH1-3-Sak, and SecYH1-Sak were thus overexpressed in E. coli LWF+WEI, while SecYH1-Sak and CcmAH1-Sak were overexpressed in P. mirabilis LVIWEI. Both strains expressed those fusion proteins with higher efficiency, proteins which contained homologous and small anchor sequences with one transmembrane domain. From previous studies it is known (12) that Sak is expressed in L-form cells as a soluble extracellular product with yields of 150 to 250 µg ml^-1. The yields of the modified surface displayed Sak varied depending on the membrane anchor and the strain. Maximum values of 100 µg of total membrane-anchored CcmAH1-Sak ml^-1 were detected in P. mirabilis LVIWEI by quantitative Western blot analysis. The functional activity of this fusion protein corresponded to 40 µg of soluble Sak ml^-1. Also, the other fusion proteins were functionally active. The lower values, which were determined by functionality assays, can be caused by the treatment of the membranes prior to the assays or by limited diffusion of the proteins, since they have lipids attached to the integral membrane moiety even in the presence of a detergent.

All overexpressed fusion proteins remained cell bound, except small quantities of LacYH1-Sak and SecYH1-Sak, which were also found in the culture supernatant. The fusion proteins detected in the cell fractions were bound to the CM of the cells. CcmAH1-Sak expressed in P. mirabilis LVIWEI was exclusively detected in membrane fractions, which were purified by ultracentrifugation or by Percoll density gradient centrifugation (Fig. 3). In combination with solubilization experiments, these results gave proof that the fusion proteins were embedded in the CM and did not form considerable amounts of aggregates as described for other surface display systems (22).

The majority of the membrane-anchored Sak molecules were freely accessible to the protease trypsin, indicating that they were located on the outside of the cells. These findings were further verified by freeze-fracture replica-labeling electron microscopy. The Sak molecules were mainly found in the EF face, and only traces could be detected in the cytoplasm or in the PF face. Obviously, the Sak moieties of the fusion proteins were translocated immediately after translation to the outer surface of the CM, where they remained fixed by their membrane anchors. Furthermore, comparison of the EF faces from overproducing cells and those which did not synthesize fusion proteins indicated that overexpression of CcmAH1-Sak caused the formation of protein clusters in the EF face, which consisted of overexpressed fusion proteins and endogenous membrane proteins. The molecules remained bound tightly to the CM, even after repeated washes and ultrasonication. The membrane-anchored Sak was functionally active. This indicated that the Sak molecules, when fused to anchor peptides, were correctly folded after translocation, although the cells lacked a periplasmic compartment and associated folding factors.

The L-form strains are free of or reduced in surface components, such as endotoxic LPS, pili, flagella, adhesins, and proteases, which are responsible for numerous interactions at the cellular level, among them the binding to receptors, as well as the induction of inflammatory processes or immune responses (25, 29). This novel membrane surface display system therefore has potential in applications, such as the development of diagnostics and vaccines, and specific adhesin-receptor interaction studies between bacterial and eukaryotic cells at the cellular and molecular level (Gumpert et al., patent FN 100 11 358.3 [pending]).

 
C.F. and C.H. contributed equally to this study.

We gratefully thank B. Küntzel, G. Elske, and S. Pfeiffer for excellent technical assistance; M. Hartmann and S. Sieben for plasmids; E. J. Allan and J. Rippmann for critical reading of the manuscript; and Stephan Diekmann for support.

This work was supported by a grant from the Deutsche Forschungsgemeinschaft (Sonderforschungsbereich 197, TP A2).

 
* Corresponding author. Mailing address: Department of Molecular Biology, Institute of Molecular Biotechnology, Beutenbergstr. 11, D-07745 Jena, Germany. Phone: 49-3641-656305. Fax: 49-3641-656310. E-mail: hoischen{at}imb-jena.de.

Top Abstract Introduction Materials and Methods Results Discussion References

 

1
1. Akiyama, Y., and K. Ito. 1987. Topology analysis of the SecY protein, an integral membrane protein involved in protein export in Escherichia coli. EMBO J. 6:3465-3470.[Medline]
2
2. Ausubel, F. M., R. Brent, R. E. Kingston, D. D. Moore, J. G. Seidman, J. A. Smith, and K. Struhl (ed.). 1994. Current protocols in molecular biology. John Wiley & Sons, Inc., New York, N.Y.
3
3. Behnke, D., and D. Gerlach. 1987. Cloning and expression in Escherichia coli, Bacillus subtilis, and Streptococcus sanguis of a gene for staphylokinase—a bacterial plasminogen activator. Mol. Gen. Genet. 210:528-534.[CrossRef][Medline]
4
4. Bradford, M. M. 1976. A rapid and sensitive method for the quantification of microgram quantities of proteins utilizing the principle of protein-dye binding. Anal. Biochem. 72:248-254.[CrossRef][Medline]
5
5. Bushueva, A. M., A. B. Shevelev, J. Gumpert, G. G. Chestukhina, A. V. Serkina, C. Hoischen, M. V. Matz, M. V. Kuryatova, and V. M. Stepanov. 1998. Expression of the carboxypeptidase T gene from Thermoactinomyces vulgaris in stable cell wall-less L-forms of Proteus mirabilis. FEMS Microbiol. Lett. 159:145-150.[CrossRef][Medline]
6
6. Cornelis, P. 2000. Expressing genes in different Escherichia coli compartments. Curr. Opin. Biotechnol. 11:450-454.[CrossRef][Medline]
7
7. Cote-Sierra, J., E. Jongert, A. Bredan, D. Gautam, M. Parkhouse, P. Cornelis, P. De Baetselier, and H. Revets. 1998. A new membrane-bound OprI lipoprotein expression vector. High production of heterologous fusion proteins in gram-negative bacteria and the implications for oral vaccination. Gene 221:25-34.[CrossRef][Medline]
8
8. Foster, D. L., M. Boublik, and H. R. Kaback. 1983. Structure of the lac carrier protein of Escherichia coli. J. Biol. Chem. 258:31-34.[Abstract/Free Full Text]
9
9. Fujimoto, K. 1997. SDS-digested freeze-fracture replica labeling electron microscopy to study the two-dimensional distribution of integral membrane proteins and phospholipids in biomembranes: practical procedure, interpretation, and application. Histochem. Cell Biol. 107:87-96.[CrossRef][Medline]
10
10. Georgiou, G., C. Stathopoulos, P. S. Daugherty, A. R. Nayak, B. L. Iverson, and R. Curtiss III. 1997. Display of heterologous proteins on the surface of microorganisms: from the screening of combinatorial libraries to live recombinant vaccines. Nat. Biotechnol. 15:29-34.[CrossRef][Medline]
11
11. Georgiou, G., D. L. Stephens, C. Stathopoulos, H. L. Poetschke, J. Mendenhall, and C. F. Earhart. 1996. Display of ß-lactamase on the Escherichia coli surface: outer membrane phenotypes conferred by Lpp"-OmpA"-ß-lactamase fusions. Protein Eng. 9:239-247.[Abstract/Free Full Text]
12
12. Gumpert, J., and C. Hoischen. 1998. Use of cell wall-less bacteria (L-forms) for efficient expression and secretion of heterologous gene products. Curr. Opin. Biotechnol. 9:506-509.[CrossRef][Medline]
13
13. Gumpert, J., and U. Taubeneck. 1983. Characteristic properties and biological significance of stable protoplast type L-forms. Experientia Suppl. 46:227-241.[Medline]
14
14. Gumpert, J., H. Cron, R. Plapp, H. Niersbach, and C. Hoischen. 1996. Synthesis and secretion of recombinant penicillin G acylase in bacterial L-forms. J. Basic Microbiol. 36:89-98.[Medline]
15
15. Hay, N. A., D. J. Tipper, D. Gygi, and C. Hughes. 1999. A novel membrane protein influencing cell shape and multicellular swarming of Proteus mirabilis. J. Bacteriol. 181:2008-2016.[Abstract/Free Full Text]
16
16. Hoischen, C., K. Gura, C. Luge, and J. Gumpert. 1997. Lipid and fatty acid composition of cytoplasmic membranes from Streptomyces hygroscopicus and its stable protoplast-type L-form. J. Bacteriol. 179:3430-3436.[Abstract/Free Full Text]
17
17. Jung, H.-C., J.-M. Lebeault, and J.-G. Pan. 1998. Surface display of Zymomonas mobilis levansucrase by using the ice-nucleation protein of Pseudomonas syringae. Nat. Biotechnol. 16:576-580.[CrossRef][Medline]
18
18. King, S. C., C. L. Hansen, and T. H. Wilson. 1991. The interaction between aspartic acid 237 and lysine 358 in the lactose carrier of Escherichia coli. Biochim. Biophys. Acta 25:177-186.
19
19. Konieczny, M. P., M. Suhr, A. Noll, I. B. Autenrieth, and A. M. Schmidt. 2000. Cell surface presentation of recombinant (poly)peptides including functional T-cell epitopes by the AIDA autotransporter system. FEMS Immunol. Med. Microbiol. 27:321-332.[CrossRef][Medline]
20
20. Kujau, M. J., C. Hoischen, D. Riesenberg, and J. Gumpert. 1998. Expression and secretion of functional miniantibodies McPC603scFvDhlx in cell wall-less L-form strains of Proteus mirabilis and Escherichia coli. Appl. Microbiol. Biotechnol. 45:51-58.[CrossRef]
21
21. Lång, H. 2000. Outer membrane proteins as surface display systems. Int. J. Med. Microbiol. 290:579-585.[Medline]
22
22. Lång, H., M. Mäki, A. Rantakari, and T. K. Korhonen. 2000. Characterization of adhesive epitopes with the OmpS display system. Eur. J. Biochem. 267:163-170.[Medline]
23
23. Lee, J.-S., K.-S. Shin, J.-G. Pan, and C.-J. Kim. 2000. Surface-displayed viral antigens on Salmonella carrier vaccine. Nat. Biotechnol. 18:645-648.[CrossRef][Medline]
24
24. Leenhouts, K., G. Buist, and J. Kok. 1999. Anchoring of proteins to lactic acid bacteria. Antonie Leeuwenhoek 76:367-376.
25
25. Liljeqvist, S., and S. Ståhl. 1999. Production of recombinant subunit vaccines: protein immunogens, live delivery systems and nucleic acid vaccination. J. Biotechnol. 73:1-33.[CrossRef][Medline]
26
26. Lu, Z., K. S. Murray, V. Van Cleave, E. R. LaVallie, M. L. Stahl, and J. M. McCoy. 1995. Expression of thioredoxin random peptide libraries on the Escherichia coli cell surface as functional fusions to flagellin: a system designed for exploring protein-protein interactions. Bio/Technology 13:366-372.[CrossRef][Medline]
27
27. Maurer, J., J. Jose, and T. F. Meyer. 2000. Autodisplay: one-component system for efficient surface display and release of soluble recombinant proteins from Escherichia coli. J. Bacteriol. 179:794-804.[Abstract/Free Full Text]
28
28. Morein, S., D. Henricson, and L. Riffors. 1994. Separation of inner and outer membrane vesicles from Escherichia coli in self-generating percoll gradients. Anal. Biochem. 216:47-51.[CrossRef][Medline]
29
29. Oelschlaeger, T. A., and J. Hacker. 2000. Bacterial invasion into eukaryotic cells. Plenum Publishing Co., New York, N.Y.
30
30. Rippmann, J. F., M. Klein, C. Hoischen, B. Brocks, W. J. Rettig, J. Gumpert, K. Pfizenmaier, R. Mattes, and D. Moosmayer. 1998. Expression and secretion of recombinant scFv antibodies in L-form cells lead to active product and overcomes the limitations of periplasmic expression in E. coli. Appl. Environ. Microbiol. 64:4862-4869.[Abstract/Free Full Text]
31
31. Schlott, B., K. H. Gührs, M. Hartmann, A. Röcker, and D. Collen. 1998. NH2-terminal structural motifs in staphylokinase required for plasminogen activation. J. Biol. Chem. 273:22346-22350.[Abstract/Free Full Text]
32
32. Sleytr, U. B., H. Bayley, M. Sára, A. Breitwieser, S. Küpcü, C. Mader, S. Weigert, F. M. Unger, P. Messner, B. Jahn-Schmid, B. Schuster, D. Pum, K. Douglas, N. A. Clark, J. T. Moore, T. A. Winningham, S. Levy, I. Frithsen, J. Pankovc, P. Beale, H. P. Gillis, D. A. Choutov, and K. P. Martin. 1997. Applications of S-layers. FEMS Microbiol. Rev. 20:151-175.[Medline]
33
33. Ståhl, S., and M. Uhlén. 1997. Bacterial surface display: trends and progress. Trends Biotechnol. 15:185-192.[CrossRef][Medline]
34
34. Tadayyon, M., J. R. Gittins, J. M. Pratt, and J. K. Broome-Smith. 1994. Expression of membrane proteins in Escherichia coli, p. 29-83. In G. W. Gould (ed.), Membrane protein expression systems: a user's guide. Portland Press, London, United Kingdom.
35
35. Valls, M., S. Atrian, V. de Lorenzo, and L. A. Fernández. 2000. Engineering a mouse metallothionein on the cell surface of Ralstonia eutropha CH34 for immobilization of heavy metals in soil. Nat. Biotechnol. 18:661-665.[CrossRef][Medline]

---

Applied and Environmental Microbiology, February 2002, p. 525-531, Vol. 68, No. 2
0099-2240/02/$04.00+0     DOI: 10.1128/AEM.68.2.525-531.2002
Copyright © 2002, American Society for Microbiology. All Rights Reserved.

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Hoischen, C. Articles by Fahnert, B. Search for Related Content PubMed PubMed Citation Articles by Hoischen, C. Articles by Fahnert, B. Agricola Articles by Hoischen, C. Articles by Fahnert, B.