Specific binding of recombinant Listeria monocytogenes p60 protein to Caco-2 cells

Abstract

The Listeria monocytogenes p60 is a major extracellular protein, which is believed to be involved in the invasion of these bacteria into their host cells. So far the mechanism by which p60 participates in the internalization or penetration of L. monocytogenes is still veiled. To determine the possibility of a direct interaction of p60 with the host cell surface, the iap gene was recombinantly expressed in Escherichia coli and used for binding studies with the enterocyte-like Caco-2 cells. Fluorescence activated flow cytometry and confocal laser scanning microscopy revealed a cell membrane specific staining with p60, which implications in Listeria virulence are discussed.

Maltose binding protein, Invasion associated protein, p60, Recombinant protein, Listeria monocytogenes

1 Introduction

Listeria monocytogenes is a Gram-positive, facultative intracellular bacteria species that is the etiological agent for listeriosis in human [1]. Infection of L. monocytogenes can lead to meningitis, meningoencephalitis, gastroenteritis, and even abortions with an overall mortality rate of over 20%[2], whereby other species of the Listeria genus are known to be mostly non-pathogenic. The transmission of Listeria infection is food-borne [3], and after entering the host through the gastrointestinal tract, bacteria spread to various tissues such as liver, brain and placenta by the lymph and blood stream. The broad range of host cell specificity of L. monocytogenes is mainly due to its unique characteristic having the ability to cross the intestinal barrier as well as the blood brain and placenta barrier [4] using a still poorly characterized cell invasion machinery. Recent studies have identified a few genes, which are believed to play central roles in the invasion of Listeria into their host cell, nevertheless, the precise mechanism has still to be unveiled. Depending on the type of the host cell, it is currently accepted that at least two different ways of invasion exist where different molecules are required for penetration. The cell surface protein internalin, which is encoded by inlA, is essential for the invasion of the enterocyte-like epithelial cell line Caco-2 [5], whereas another homologous surface protein InlB is thought to be necessary for the internalization into hepatocytes [6] and some fibroblast-like cell lines such as Vero, HeLa and HEp-2 [7,8]. Also, ActA, a protein involved in actin polymerization might participate in cell invasion [9], as do maybe other members of the internalin gene family such as inlC and inlF[10]. Furthermore, the identification of spontaneous mutants that have lost the ability to invade 3T6 mouse fibroblast cells had lead to the identification of a major extracellular protein of 60 kDa that is necessary for a successful invasion into host cells [11]. The production of this protein, termed p60 or invasion associated protein (iap), is impaired in those mutated L. monocytogenes strains, and it was subsequently shown that p60 possesses murein hydrolase activity without which cells tend to form long chains, where septum formation between individual cells was intact but cells do not separate [11]. Further studies revealed that not this physical defect but actually the lack of p60 protein was responsible for the loss of cell invasion ability [11,12]. Also, since the addition of purified p60 into culture media restored the cell internalization capacity of these mutant cells as well as conferred increased cellular uptake when expressed in Salmonella typhimurium[13], it was evident that p60 is an essential factor in the invasion into non-phagocytic host cells. However despite such overwhelming evidences, in contrast to other bacterial factors involved in internalization, the operating mechanism for p60 is still largely uninvestigated. To address the mode of p60 functioning in cell internalization, in the present study, the iap gene of L. monocytogenes was cloned and, for the first time, overexpressed as a recombinant protein in Escherichia coli. The expression as a maltose binding protein (MBP) fusion protein in E. coli enabled the detection of p60 protein in several systems using MBP specific antibodies, which was then visualized by fluorescence activated flow cytometry and confocal laser scanning microscopy. Using this strategy, Caco-2 human enterocyte-like cells were analyzed upon their binding activity with this recombinant protein.

2 Materials and methods

2.1 Materials

Oligonucleotide primers were purchased from Bioneer Inc., Chungwon, Korea, and all reagents for performing PCR, except for the Taq DNA polymerase (Stratagene, La Jolla, CA, USA), were ordered from Promega, Madison, WI, USA. Restriction enzymes and T4 DNA ligase were products of Roche Molecular Biochemicals, Mannheim, Germany. The pMAL-c2 vector as well as amylose resins for the purification of MBP fusion proteins were obtained from New England Biolabs, Beverly, MA, USA. All other chemicals were ordered from Sigma, St. Louis, MO, USA, if not else indicated.

2.2 Molecular cloning of the MBP-iap fusion protein expression vector

For the recombinant expression of the L. monocytogenes p60 protein, the whole coding region of the iap was amplified by PCR using the following primers, iap-up (5′-tatgaattcgctacagctgggattgcggt-3′) and iap-down (5′-atatctagattatacgcgaccgaagccaa-3′). The PCR reaction was performed in a total volume of 100 μl containing 10 ng of chromosomal DNA from L. monocytogenes as template with a final concentration of the following reagents: 20 mM Tris/HCl (pH 8.3), 50 mM KCl, 1.5 mM MgCl₂, 0.2 mM of dNTP, 2.5 units of Taq DNA polymerase and 10 pmol of each primer. After a preincubation for 5 min at 95°C, PCR was performed in a Thermal Cycler 9600 (Perkin Elmer, Foster, CA, USA) for 30 cycles (denaturation for 30 s at 94°C; annealing for 60 s at 52°C; elongation for 60 s at 72°C) followed by a final elongation step for 10 min at 72°C. After amplification, the PCR product was purified over a 1.2% TAE-buffered agarose gel and eluted using a QIAquick gel extraction kit (Qiagen, Hilden, Germany). For further subcloning, the DNA fragment was digested with the restriction enzymes EcoRI and XbaI and transferred into the corresponding enzyme sites of the MBP fusion protein vector, pMAL-c2 (New England Biolabs, Beverly, MA, USA). The resulting clone was termed pMAL-iap/mono and was used for the further expression of the recombinant p60 protein in E. coli.

2.3 Expression and purification of the recombinant iap protein

The recombinant expression of MBP-iap fusion products was performed with E. coli (strain JM109) cells, which were transformed with the newly constructed pMAL-iap/mono vector. Cells were induced in their log growth phase with 1 mM IPTG (end concentration) and harvested after further 3 h of culture. Purification of recombinant proteins was performed using amylose resins as described elsewhere [14]. In brief, the pellet was resuspended in 1/20 culture volume of column buffer (20 mM Tris–HCl pH 7.4, 200 mM NaCl, 1 mM EDTA, 1 mM PMSF), and the cells were disrupted by sonication. After centrifugation at 10 000×g for 15 min, the supernatant was loaded on an amylose affinity column, and after binding of the MBP fusion proteins, column was washed extensively with column buffer. Specifically bound proteins were eluted by competition with 20 mM of free maltose in column buffer. The purity of the isolated MBP-iap fusion proteins was examined in SDS–PAGE [15], and the identity of MBP fusion proteins was confirmed in Western blot [16] using anti-MBP antibodies [14].

2.4 FACS analysis and confocal microscopy

To examine a possible interaction of recombinant MBP-iap proteins with human host cells, fluorescence activated flow cytometry (FACS) was performed. The human enterocytic cell line, Caco-2, was obtained from the ATCC (Rockville, MA, USA). Cells were maintained in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal bovine serum at 37°C in a humidified CO₂ incubator. For each analysis, 5×10⁵ Caco-2 cells were harvested at 50% confluence, washed once in Hanks balanced salt solution (HBSS) and resuspended in 200 μl of cell staining buffer (0.1% BSA, 0.02% Na-azide in PBS). MBP-iap or as negative control MBP-preS1(1–56) proteins were added to an end concentration of 50 μg ml⁻¹ and incubated for 1 h at 4°C. Excess ligands were then washed out by centrifugation at 250×g for 7 min, and the cells were incubated with saturating amounts of rabbit anti-MBP antibodies (New England Biolabs) at 4°C. After 30 min, excess antibodies were washed out with cell staining buffer, and cells were stained with FITC-conjugated goat anti-rabbit IgG antibodies. Analysis was performed with a FACScalibur flow cytometer (Becton Dickinson, Mountain view, CA, USA).

For confocal microscopy, the day before analysis, Caco-2 cells were harvested at 50% confluence using a trypsin/EDTA detachment protocol and transferred into 6-well plates (Corning Costar Corp., Cambridge, MA, USA) containing autoclaved coverslips to a concentration of 4×10⁵ cells ml⁻¹ DMEM. The next day, the coverslips were carefully transferred into new 6-well plates, and cells were incubated with the recombinant MBP-iap or as control with MBP-preS1. Detection of cell-bound ligands was performed using the anti-MBP monoclonal antibody (mAb) HAM-19 [14], and mAb HAM-19 was visualized using FITC-conjugated rabbit anti-mouse Ig antibodies (Jackson ImmunoResearch Laboratories Inc., West Grove, PA, USA). Analysis was performed using a Leica TCS 4D confocal microscope (Leica Lasertech., Heidelberg, Germany).

3 Results and discussion

The recombinant expression of the invasion associated protein (iap) p60 protein was achieved using the E. coli expression vector pMAL-c2, which enables the expression of proteins in interest in fusion to the maltose binding protein [17]. This expression system has been repeatedly confirmed to be effective in the overexpression of various proteins [14,18], and the purification protocol of MBP fusion proteins using amylose resins is attractive over other expression systems in both its simplicity and purity. Such characteristics of the pMAL system strongly supported its use in the first expression of the iap gene of L. monocytogenes in a genetically engineered recombinant form as presented in this study.

Molecular cloning and construction of the iap expression vector, pMAL-iap/mono, was performed by conventional molecular biological techniques, which procedures are described in details in Section 2. The vector map of this plasmid is shown in Fig. 1. Expression of the MBP-iap fusion protein is under the control of a strong tac promoter, and this was induced by addition of IPTG. Since the pMAL-c2 vector encodes for a genetically engineered maltose binding protein, where the malE signal sequence had been deleted, recombinant fusion proteins are expressed in the cytoplasm instead of the periplasm as in the case of the original MBP.

1

Schematic presentation of the MBP-iap expression vector construction.

The optimal conditions for the quantitative expression of recombinant p60 were then determined by testing variable IPTG concentrations and induction times (data not shown). The addition of 1 mM IPTG (final concentration) for induction and 3 h of further culture resulted in the highest yield of recombinant MBP-iap, which was determined to be of 2 mg l⁻¹ original culture media. The successful induction and purification is documented in Fig. 2.

2

Expression and purification of the iap in E. coli. For the quantitative preparation of MBP-iap fusion proteins, large scale culture up to 1 l was induced for expression and then subjected to purification using amylose resins. The SDS–polyacrylamide gel shows the result of the purification. Lane M, marker; lane 1, uninduced cells; lane 2, IPTG-induced cells; lane 3, purified MBP-iap protein. Arrow indicates the position of the induced recombinant protein.

The recombinant production of the L. monocytogenes p60 protein as shown in the present study is the first description of a successful overexpression of the p60 protein in any form. Before this, p60 proteins were either purified from culture media of L. monocytogenes[11,12], or, in a single study, recombinantly produced for functional assays in low amounts in S. typhimurium[13], both of which are unsatisfactory for the quantitative isolation of p60 proteins. The demand for p60 proteins in preparative scale is however increasing. The emergence and establishment of the pathogenic role of L. monocytogenes in food contamination together with the property that p60 proteins are secreted into their environment have lead to the development of some Listeria spp. detection systems using p60 specific antibodies [19]. Immunization of rabbits or mice with recombinant p60, and the use of the polyclonal serum or potential monoclonal antibodies for the detection of Listeria spp. is the base for the engineering of such systems, and the overexpressed p60 as described in this study will serve as excellent raw material in the production of such Listeria detection assays.

The availability of large amounts of recombinant p60 proteins also enabled a more detailed study on the mechanism of p60-mediated Listeria cell invasion. In particular, the p60 protein as expressed in this study has an MBP moiety fused to the N-terminal region of the p60 protein, which not only ensures a convenient purification protocol but also represents an effective tag for the detection of p60 when using MBP specific antibodies [14]. The use of MBP-directed antibodies enables the detection of MBP and their fused proteins in a variety of systems including ELISA and immunoblots. Using this system, in the present study, indirect immunostaining was performed, and the results were analyzed by flow cytometry and confocal microscopy. These two techniques are compensatory to each other since flow cytometry can analyze a large number of cells upon their p60 binding activity, however without the knowledge about the specific cellular distribution, while confocal laser scanning microscopy reveals the cell morphology without statistical information.

The main question addressed using the present system was the determination of a possible interaction of free p60 proteins with the cell membrane of Caco-2 cells, a good established model cell line for the in vitro infection of L. monocytogenes[20–22]. Originally, p60 was identified as an extracellular protein in culture media of L. monocytogenes, however recent studies have also documented the presence of this protein on the cell surface [12]. Regarding the invasion mechanism via InlA or ActA, which mediate direct interactions of Listeria with the host cell membrane via cellular receptors like E-cadherin [23] and heparan sulfate receptors [9], respectively, it had been suggested that p60 might also mediate host cell attachment of Listeria via bacterial cell surface displayed p60 proteins. However, further studies had revealed that such a direct interaction between L. monocytogenes and their host cells via p60 is rather unrealistic, since p60 secretion mutants with normal level of cell surface p60 proteins [12] were still unable to penetrate into cells. In conclusion from these previous observations, it is evident that soluble, free p60 protein is required for a successful penetration of the cell membrane. However beside this report, only few studies had been performed on the p60-mediated cell invasion process, so that it has not been even established whether p60 acts on the bacterial side or on the host cell side to facilitate the internalization into target cells. The indispensability of p60 in the L. monocytogenes life cycle had been demonstrated previously [24] as has been its function as a murein hydrolase.

To examine whether there might be any interactions of free p60 with the host cell, in the present study, Caco-2 cells were incubated with recombinant MBP-iap and detected with fluorescence-conjugated antibodies. As shown in Fig. 3, fluorescence activated cell sorting (FACS) revealed that p60 directly binds to Caco-2 cells with high specificity. The filled histogram indicates the distribution of the relative fluorescence intensity of 10 000 individual Caco-2 cells stained with recombinant p60 proteins and detected with fluorescence-conjugated secondary reagents. Cell staining with negative control proteins (MBP-hepatitis B virus preS(1–56) fusion protein [25]) on the other hand showed no binding activities under the same conditions, which is indicated in the distribution of fluorescence intensity of the open histogram. The cellular distribution of MBP-iap binding to Caco-2 cells was then further analyzed by confocal laser scanning microscopy (CLSM). Using attached growing Caco-2 cells, MBP-iap binding was visualized, and here it was shown that p60 specifically interacts with the cell membrane in a uniform distribution (Fig. 4). The transmission light image, which shows the overall morphology of the cells as seen in a conventional light microscope, is displayed in panel A of Fig. 4, and the fluorescence image of the same cells is shown in panel B. The observations as described above contain several implications. First, since p60 obviously binds to the cell surface of Caco-2 cells, there must be some cellular structures that serve as receptor for these proteins. Second, depending on the nature of these receptors, the binding and/or internalization of free p60 protein must release a cellular signal that facilitates the internalization of Listeria. Whether cell surface displayed p60 proteins might also play a role in this step or other internalization involved proteins might take over will need further investigations. Nevertheless, the determination of p60 binding on host cell membranes is of great impact for the understanding of Listeria invasion. Also with the observation that at least for human umbilical vein endothelial cells L. monocytogenes might also invade and grow intracellular independent of InlA, B, C and ActA [26], the role of free p60 proteins is even more emphasized. Further studies regarding the p60-mediated internalization will first involve the identification of the cellular receptor for this protein. Also, a screening of various human tissues for MBP-iap binding activity will give a clue about a possible correlation about p60 protein receptor expression and host cell restriction of L. monocytogenes.

3

Determination of recombinant p60 binding onto Caco-2 cells by FACS. Using fluorescence activated flow cytometry (FACS), the specific binding of recombinant MBP-iap proteins to the human enterocyte-like cell line Caco-2 was determined. After detachment of the cells at 50% confluence from culture flasks, each 5×10⁵ cells were used for one staining reaction. Histogram shows the relative log fluorescence intensity of Caco-2 cells incubated either with the same amount of MBP-iap fusion proteins or with a control protein (MBP-preS1(1–56) fusion protein). Specific binding was detected with anti-MBP antibodies and FITC-conjugated secondary antibodies. Open histogram indicates the fluorescence signal of MBP-preS1(1–56) staining and the filled histogram shows the fluorescence intensity of MBP-iap binding, respectively. Histogram shows the result of analysis from 10 000 cells.

4

Confocal microscopic analysis of recombinant p60 binding in Caco-2 cells. The cellular distribution of recombinant p60 binding was analyzed by confocal laser scanning microscopy, which enables the optical sectioning of the sample. For this Caco-2 cells were grown on glass cover slips, and incubated in this state with recombinant MBP-iap fusion proteins. Specific binding was detected with anti-MBP antibodies and FITC-conjugated secondary antibodies. A: Transmission light image of Caco-2 cells at 400-fold magnification. B: The same cells as in panel A, this time analyzed for fluorescence signals by laser scanning at 488 nm, and detected using a FITC band pass filter.

In summary, the present study reports for the first time the successful overexpression of the L. monocytogenes p60 protein in E. coli, and its use in the determination of free p60 protein binding activity onto Caco-2 human enterocyte-like cells. The specific binding of p60 to host cell membrane indicates a possible role of this protein in the invasion of L. monocytogenes by modulation of the host cell physiology, which nature will be revealed by further identification of the natural receptor.

Acknowledgements

This work was supported by the special grant research program no. 198043-3 from the Ministry of Agriculture and Forestry, Korea.