Arch Microbiol DOI 10.1007/s00203-015-1115-2


Membrane cholesterol plays an important role in enteropathogen adhesion and the activation of innate immunity via flagellin– TLR5 signaling Mingxu Zhou1,2 · Qiangde Duan1,3 · Yinchau Li1,2 · Yang Yang1,2 · Philip R. Hardwidge4 · Guoqiang Zhu1,2 

Received: 4 January 2015 / Revised: 17 March 2015 / Accepted: 23 April 2015 © Springer-Verlag Berlin Heidelberg 2015

Abstract  Lipid rafts are cholesterol- and sphingolipidrich ordered microdomains distributed in the plasma membrane that participates in mammalian signal transduction pathways. To determine the role of lipid rafts in mediating interactions between enteropathogens and intestinal epithelial cells, membrane cholesterol was depleted from Caco-2 and IPEC-J2 cells using methyl-β-cyclodextrin. Cholesterol depletion significantly reduced Escherichia coli and Salmonella enteritidis adhesion and invasion into intestinal epithelial cells. Complementation with exogenous cholesterol restored bacterial adhesion to basal levels. We also evaluated the role of lipid rafts in the activation of Toll-like receptor 5 signaling by bacterial flagellin. Depleting membrane cholesterol reduced the ability of purified recombinant E. coli flagellin to activate TLR5 signaling in intestinal cells. These data suggest that both membrane cholesterol and lipid rafts play important roles in enteropathogen

Communicated by Djamel Drider. Mingxu Zhou and Qiangde Duan have contributed equally to this work.

adhesion and contribute to the activation of innate immunity via flagellin–TLR5 signaling. Keywords  Cholesterol · Flagellin · Innate immunity · Lipid raft

Introduction The bacterial pathogens enterotoxigenic Escherichia coli (ETEC) and Salmonella enteritidis cause diarrheal disease in humans and livestock. Mammalian cells communicate with the outside environment via proteins embedded in the plasma membrane. Lipid rafts also participate in numerous biological processes, including signal transduction, apoptosis, cholesterol transport, and membrane trafficking (Simons and van Meer 1988; Brown and London 1998; Simons and Ikonen 2000; Simons and Toomre 2000; Ikonen 2001). Some bacteria adhere to and invade into host cells via lipid rafts to avoid the perils of endocytosis and lysosomal degradation (Shin and Abraham 2001; Garner

* Guoqiang Zhu [email protected]


College of Veterinary Medicine, Yangzhou University, 12 East Wenhui Road, Yangzhou 225009, China

Mingxu Zhou [email protected]


Qiangde Duan [email protected]

Jiangsu Co-innovation Center for Prevention and Control of Important Animal Infectious Diseases and Zoonoses, Yangzhou 225009, China


Yinchau Li [email protected]

Agriculture College, Weinan Vocational and Technical College, Weinan 714000, China


College of Veterinary Medicine, Kansas State University, Manhattan, KS 66506, USA

Yang Yang [email protected] Philip R. Hardwidge [email protected]


Arch Microbiol

et al. 2002; Lafont et al. 2002b; Duncan et al. 2004; Lafont and van der Goot 2005). For example, CD44 in lipid rafts is implicated in Shigella binding and entry into CHO cells (Lafont et al. 2002a). GPI-anchored proteins, which are located in lipid rafts, are involved in the invasion of Brucella and Mycobacterium into macrophages and neutrophils (Peyron et al. 2000; Watarai et al. 2002). Lipid rafts and GPI-anchored proteins also affect Ehrlichia chaffeensis and Anaplasma phagocytophilum binding to and infection of human leukemia cell lines (Lin and Rikihisa 2003). Lipid rafts also mediate invasion of bladder epithelial cells by UPEC (Zaas et al. 2005). Cholesterol has an important influence on the phase of lipid rafts (Ahmed et al. 1997), and thus, cholesterol depletion can disrupt lipid rafts and affect its biological function (Brown and London 2000). Innate immunity serves as an essential first-line defense against microbial pathogens. Pattern recognition receptors (PRRs) on mammalian cells recognize conserved pathogen-associated molecular patterns (PAMPs), which are shared by large groups of microorganisms (Akira et al. 2001). Toll-like receptor 5 (TLR5) is expressed on the surface of monocytes, immature dendritic cells, and epithelial cells (Vijay-Kumar and Gewirtz 2009). TLR5 recognizes bacterial flagellin and stimulates the production of proinflammatory cytokines (Hayashi et al. 2001). The flagella of ETEC and Shiga-like toxin-producing E. coli (SLTEC) have accessory functions as both adhesins and invasins (Duan et al. 2013a, b) and contribute to biofilm formation (Duan et al. 2012; Zhou et al. 2014). Flagellin, the major subunit of this bacterial surface organelle, plays an important role in mediating bacterial interactions with the host cell membrane. Previous studies have shown that TLR4 signaling is dependent on cholesterol-rich lipid rafts (Lu et al. 2012; Kim et al. 2013; Mutoh and Ueda 2013). Lipid rafts are involved in the TLR5 activation in the human alveolar type II epithelial cells (Im et al. 2009). Whether lipid rafts mediate interaction between intestinal epithelial cells and enteropathogens is still unclear. We therefore examined the impact of depleting membrane

cholesterol with methyl-β-cyclodextrin (MβCD) on the adhesion and invasion of several human and porcine pathogenic bacteria on human and porcine intestinal cell lines (Ilangumaran and Hoessli 1998). We also expressed and purified recombinant flagellin from E. coli to quantify the impact of depleting membrane cholesterol and disrupting lipid rafts with MβCD on the ability of flagellin to activate interleukin-8 and tumor necrosis factor expression.

Materials and methods Bacterial strains, plasmids, and culture conditions The strains and plasmids used are listed in Table 1. Strains were grown in LB broth (tryptone 10.0 g/l, yeast extract 5.0 g/l, NaCl 10 g/l) or on LB agar plates at 37 °C, in the presence of ampicillin (100 μg/ml) or kanamycin (30 μg/ ml) where appropriate. Mammalian cell culture Porcine neonatal jejunal epithelial IPEC-J2 cells (Berschneider 1989) were cultured in antibiotic-free F12-RPMI 1640 mixed media (Gibco, USA), supplemented with 10 % newborn calf serum (NCS) (Gibco). Caco-2 cells derived from human colonic adenocarcinoma were cultured in antibiotic-free DMEM (Gibco), supplemented with 10 % fetal bovine serum (FBS) (Gibco). Cells were maintained at 37 °C in a humidified incubator with 5 % CO2. Methyl‑β‑cyclodextrin cytotoxicity We used MβCD to deplete cholesterol from cell membranes (Simons and Toomre 2000; Vercauteren et al. 2010; Zhu et al. 2010). Methyl-β-cyclodextrin (MβCD) cytotoxicity to Caco-2 and IPEC-J2 cells was evaluated using the Cell Proliferation Reagent WST-1 (Roche, Germany). Cells were seeded at 5 × 104 cells/well in a 96-well plate (Corning,

Table 1  Strains and plasmids used in this study Strain


Source or reference − (r− B mB )

E. coli F107/86

gal dcm (DE3) F− ompT hsdSB Porcine ETEC, F18ab+, O139:K12(B):H1

Bertschinger et al. (1990)

E. coli C83902

Porcine ETEC, F4ac+, O8:K87:H19

China Institute of Veterinary Drugs Control

E. coli BL21(DE3)


E. coli H10407

Human ETEC, O78:H11 CFA/I LT ST (ST-H, ST-P)

Evans et al. (1975)

S. enteritidis 50336

SEF14+, SEF21+

National Institutes for Food and Drug Control of China

Plasmids  pET28a  pET28a-FliC(H1)

Bacterial hexahistidine fusion expression FliC(H1)-His6

Novagen This study


This study



Arch Microbiol

USA) for 12 h and then incubated with 0–10 Mm MβCD for 1 h at 37 °C. WST-1 (10 µl) was added, and after 4 h at 37 °C, the absorbance values of the samples were measured using a spectrophotometer (BioTek, USA) at 440 nm and corrected for nonspecific absorbance at 600 nm. The percent viability of cells was calculated with respect to that of untreated control cells. Efficiency of cholesterol removal The ganglioside GM1 is a commonly used raft marker (Janes et al. 1999). The efficiency of cholesterol removal by MβCD was determined by detecting fluorescent CTBFITC binding to GM1 using flow cytometry. Caco-2 and IPEC-J2 cells were seeded in six-well plates for 12 h and then incubated with MβCD for 1 h at 37 °C. After washing three times with ice-cold PBS, the detached cells were incubated with 10 μg/ml CTB-FITC (Sigma, USA) for 30 min on ice. Flow cytometry was carried out using a Becton–Dickinson FACSCalibur flow cytometer with 10,000 events collected, and data were analyzed using FlowJo software, version 7.1.3. Adhesion and invasion assays Bacterial adhesion to Caco-2 and IPEC-J2 cells was determined as described previously (Jouve et al. 1997), with minor modifications. A monolayer of 1 × 105 cells/well in a 96-well tissue culture plate (Corning) was pre-incubated with 2 mM MβCD for 1 h, and where indicated, supplemented with 400 μg/ml cholesterol (Sigma, USA) for 1 h at 37 °C (Zhu et al. 2010). Cells were washed gently three times with PBS, and then 1 × 107 CFUs of bacterial strains were added to each well. After 1 h, cells were washed three times with PBS and then lysed with 0.5 % Triton X-100 for 20 min. The lysates containing total cell-associated bacteria were diluted 1:10 in PBS and plated onto LB agar plates to enumerate adherent bacteria. Invasion assays were performed similarly to adhesion assays, but the incubation time after adding bacteria to mammalian cells was extended to 2 h. After washing with PBS, 130 μg/ml gentamicin was added for 2 h to kill extracellular bacteria. Generation of recombinant flagellin proteins The fliC genes from E. coli F107/86 (serotype H1) and E. coli C83902 (serotype H19) were PCR-amplified (Table 2) and cloned into the BamHI-SalI sites of pET-28a. Recombinant His6-flagellins were expressed and purified from E. coli BL21 (DE3) cells. Protein expression was induced at an OD600 of 0.5 with 0.1 mM IPTG (Invitrogen, USA) for 4 h. After centrifugation, bacteria were lysed using an ultrasonic processor (Scientz JY92-11DN, China) and

Table 2  Primers used in this study Primer

Sequence (5′–3′)






Restriction sites are underlined

recombinant proteins were purified using Protino® NiTED 2000 Packed Columns (Macherey–Nagel, Germany). Eluted proteins were dialyzed against PBS and analyzed using SDS-PAGE and the BCA Protein Assay Reagent (APPLYGEN, China). Purified His6-tag proteins expressed from E. coli harboring an empty pET-28a vector served as a negative control. Chromogenic endpoint tachypleus amebocyte lysate assays indicated that there was less than 0.05 EU/ml (~0.02 ng/ml) of LPS per 100 ng/ml of recombinant His6-flagellin. Quantitative PCR and ELISA Caco-2 cells were seeded at 1 × 106 cells/ml on 6-well plates (Corning, USA) and cultured overnight. For cholesterol depletion and replenishment studies, Caco-2 cells were pre-treated with 2 mM MβCD for 1 h and then either supplemented with or without 400 μg/ml cholesterol for 1 h at 37 °C. Cell monolayers were washed three times with PBS and stimulated with recombinant His6-flagellin (100 ng/ml) in a serum-free cell medium containing 0.1 % FBS at 37 °C for 3 h. Total RNA from the cells of each well was extracted using TRIzol (Invitrogen, USA) according to the manufacturer’s instructions. cDNA was synthesized using the PrimeScript®RT reagent Kit with gDNA Eraser (Takara Bio, Japan). Real-time PCR amplification was performed using the Faststart Universal SYBR GREEN Master (ROX) (Roche, Germany). Assays were performed in quadruplicate in a 7500 Real-Time System (Applied Biosystems, USA). The 2−CT method was used for data normalization (Pfaffl 2001; Schmittgen and Livak 2008). For ELISAs, cell-free supernatants were collected after 3 h and utilized in Human IL-8 and TNF-alpha Valukine ELISA Kits (R&D Systems, USA).


Arch Microbiol

Statistical analysis Data were analyzed using the Student’s t test for independent samples with SPSS 17 software. p values

Membrane cholesterol plays an important role in enteropathogen adhesion and the activation of innate immunity via flagellin-TLR5 signaling.

Lipid rafts are cholesterol- and sphingolipid-rich ordered microdomains distributed in the plasma membrane that participates in mammalian signal trans...
612KB Sizes 2 Downloads 6 Views