Monolayer Systems Evaluating Bacterial Invasiveness 15

15

Tohoku J. Exp. Med., 2010, 220, 15-19

Monolayer Culture Systems with Respiratory Epithelial Cells for Evaluation of Bacterial Invasiveness

Yoichi Hirakata,1,2 Hisakazu Yano,1,2 Kazuaki Arai,1,2 Shiro Endo,1 Hajime Kanamori,1 Tetsuji Aoyagi,1 Ayako Hirotani,1 Miho Kitagawa,1 Masumitsu Hatta,1 Natsuo Yamamoto,1 Hiroyuki Kunishima,1 Kazuyoshi Kawakami3 and Mitsuo Kaku1,2

1Department of Infection Control and Laboratory Diagnostics, Tohoku University Graduate School of Medicine, Sendai, Japan

2Department of Clinical Microbiology with Epidemiological Research & Management and Analysis of Infectious Diseases (C-MERMAID) , Tohoku University Graduate School of Medicine, Sendai, Japan

3Department of Medical Microbiology, Mycology and Immunology, Tohoku University Graduate School of Medicine, Sendai, Japan

Pseudomonas (P. ) aeruginosa is a major opportunistic pathogen especially in immunocompromised patients. To evaluate the invasiveness of respiratory pathogens, we developed monolayer culture systems and examined the degree of invasion by P. aeruginosa and invasive Salmonella (S. ) typhimurium strains using human respiratory cell lines: A549 (derived from lung cancer), BEAS-2B (normal bronchial epithelium), and Calu-3 (pleural effusion of a patient with adenocarcinoma of the lung). Cells were seeded into filter units containing 0.33 cm2 filter membranes with 3.0 µm pores, and were incubated at 37°C under 5% CO2 for 4-10 days. By monitoring the trans-monolayer electrical resistance (TER), we judged that BEAS-2B cells (TER values: 436.2 ± 16.8 to 628.8 ± 66.3 Ω cm2) and Calu-3 cells (TER values: 490.5 ± 25.2 to 547.8 ± 21.6 Ω cm2) formed monolayers with tight junctions, but not A549 cells. On day 8 of culture, monolayer cultures were infected with bacteria, and the number of microorganisms penetrating into the basolateral medium was counted. Wild-type P. aeruginosa PAO1 (PAO1 WT) and S. typhimurium SL1344 were detected in the basolateral medium of BEAS-2B monolayer system by 3 h after inoculation, while only P. aeruginosa PAO1 WT was detected in the basolateral medium of Calu-3 monolayer, indicating poor invasiveness of S. typhimurium SL1344 in the Calu-3 system. These findings suggest that BEAS-2B or Calu-3 monolayer system could be useful for evaluating the invasiveness of respiratory pathogens. Because of the difference in bacterial invasiveness, we may need to choose a suitable cell system for each target pathogen.

Keywords: Pseudomonas aeruginosa/invasiveness/MDCK/BEAS-2B/Calu-3Tohoku J. Exp. Med., 2010, 220 (1), 15-19.　© 2010 Tohoku University Medical Press

Pseudomonas (P.) aeruginosa is a major opportunistic pathogen, particularly in immunocompromised patients. It is especially invasive in neutropenic patients and causes bacteremia with a high mortality rate. We previously estab-lished two systems for evaluating the invasiveness of clini-cal isolates of P. aeruginosa that employed the monolayer of human intestinal Caco-2 cells (Hirakata et al. 1998) or Madin-Darby canine kidney (MDCK) cells (Hirakata et al. 2000) grown on filter units. The MDCK monolayer system has already been used to evaluate differences in the inva-siveness of wild-type P. aeruginosa PAO1 (PAO1 WT) and its efflux pump mutations, including a mexAB-oprM dele-tion mutant (Hirakata et al. 2002), as well as the invasive-ness of quorum-sensing mutations (Kondo et al. 2006).

PAO1 WT is an invasive microorganism that penetrates MDCK cell monolayers within 3 h of infection, while the mexAB-oprM deletion mutant shows reduced invasiveness of MDCK cells.

In addition to P. aeruginosa, it has recently been reported that other respiratory pathogens, such as Streptococcus pneumoniae (Attali et al. 2008; Rajam et al. 2008; Bergmann et al. 2009), Streptococcus pyogenes (Spinaci et al. 2004; Hyland et al. 2007), and Haemophilus influenzae (Swords et al. 2000; Okabe et al., Epub ahead of print 2008), can invade epithelial cells. Klebsiella pneumo-nia and Acinetobacter species may also be invasive, because these microorganisms cause pneumonia associated with bacteremia. The ability to enter epithelial cells is necessary

Received September 29, 2009; revision accepted for publication November 13, 2009. doi:10.1620/tjem.220.15Correspondence: Yoichi Hirakata, M.D., FJSIM, Ph.D., Department of Clinical Microbiology with Epidemiological Research &

Management and Analysis of Infectious Diseases (C-MERMAID), Seiryo-machi, Aoba-ku, Sendai 980-8574, Japan.e-mail: hiraichi@mail.tains.tohoku.ac.jp

Y. Hirakata et al.16

for bacteria to cause invasive infection, and it can also be a mechanism for escaping from antimicrobial agents that can-not penetrate into cells, such as β -lactams and aminoglyco-sides. Accordingly, an in vitro system is needed that allows us to evaluate the ability of bacterial pathogens to invade respiratory epithelial cells.

For this purpose, we established monolayer systems using BEAS-2B and Calu-3 cells derived from human lungs. In the present study, we compared the invasiveness of P. aeruginosa PAO1 WT and its mexAB-oprM deletion mutant, as well as Salmonella typhimurium SL1344 and a noninvasive rabbit enterotoxigenic Escherichia coli RDEC-1, by using BEAS-2B and Calu-3 cell monolayer systems according to the method that we previously employed for evaluation of P. aeruginosa isolates with our MDCK monolayer system (Hirakata et al. 2000).

Materials and MethodsBacterial strains

The bacterial strains employed were P. aeruginosa PAO1 WT (Li et al. 1998) and its mexAB-oprM deletion mutant (∆ABM) (Hirakata et al. 2002), as well as Salmonella (S.) typhimurium SL1344 (Leung and Finlay 1991) and a noninvasive rabbit strain of enterotoxigenic Escherichia coli (RDEC-1) (Stein et al. 1996). P. aeruginosa PAO1 WT and S. typhimurium SL1344 are invasive strains that penetrate MDCK cell monolayers within 3 h of infection, while P. aeruginosa ∆ABM shows reduced invasiveness and E. coli RDEC-1 is unable to invade MDCK cells (Hirakata et al. 2000, 2002). Therefore, noninva-sive E. coli RDEC-1 was employed as a negative control. The strains were stored at −80°C until use in the study.

Cell linesMDCK cells, A549 cells (ATCC® CCL-185™) derived from

human lung cancer tissue, BEAS-2B cells (ATCC® CRL-9609™) isolated from normal human bronchial epithelium, and Calu-3 cells (ATCC® HTB-55™) derived from the pleural effusion of a patient with adenocarcinoma of the lung were used. All cells were grown in minimal essential medium with 10% fetal bovine serum.

Monolayer penetration assayFig. 1 shows a diagram of the monolayer penetration assay. The

cells were seeded at 1.5 × 105/well in Transwell filter units (Costar, Cambridge, MA, USA) containing 0.33 cm2 filter membranes (3.0 µm pores). Cells were incubated at 37°C in 5% CO2 for 4-10 days to allow the formation of monolayers, while monitoring the trans-mono-layer electrical resistance (TER) with a Millicell-ESR apparatus (Millipore, Bedford, MA, USA).

The monolayer that had been cultured for 8 days were infected by adding 5 µ l (3.5 × 106 colony-forming units [CFUs]) of freshly grown bacteria, which had been cultured in LB broth overnight at 37°C with shaking at 150 rpm. Bacteria that penetrated the monolay-er to reach the basolateral medium were counted by plating appropri-ate dilutions of medium harvested at several different times.

StatisticsAll assays were done in triplicate and experiments were repeat-

ed at least three times. Representative results are expressed as the mean ± S.D. The Mann-Whitney U test was used for comparison between groups and P < 0.05 was accepted as indicating statistical

significance.

ResultsDaily changes of TER

TER was monitored daily between days 4 and 10 after seeding the four types of cells. With MDCK cells as a posi-tive control, the TER reached a range of 1,522.5 ± 939.9 to 1,706.4 ± 878.91 Ω cm2 between day 4 and 8, and then declined to 1,244.7 ± 828.0 Ω cm2 on day 10. In the case of A549 cells, TER did not increase, indicating that these cells do not form monolayer with tight junctions under the pres-ent culture conditions. On the other hand, BEAS-2B cells and Calu-3 cells yielded TER values ranging from 436.2 ± 16.8 to 628.8 ± 66.3 and 490.5 ± 25.2 to 547.8 ± 21.6 Ω cm2, respectively, between days 4 and 10, suggesting that these two cell lines formed monolayers with tight junctions.

Fig. 1. Monolayer invasion assay. The bottom of apical chamber is closed by a filter membrane with 3.0 µm pores, which allows the passage of bacteria. If cells seeded into the apical chamber form a monolayer with tight junctions, the trans-monolayer electrical resistance (TER) between increases apical and basolateral chambers.

Monolayer Systems Evaluating Bacterial Invasiveness 17

Penetration of the cell monolayer by representative isolates

Using the MDCK cell monolayer, we initially con-firmed that P. aeruginosa PAO1 WT and S. typhimurium SL1344 were detected in the basolateral medium by 3 h after inoculation onto MDCK cell monolayers that had been cultured for 8 days, whereas P. aeruginosa ∆ABM was not detected until 6 h and E. coli RDEC-1 was not found in the basolateral medium until at least 12 h after inoculation.

When BEAS-2B cell monolayer was employed, P. aeruginosa PAO1 WT and S. typhimurium SL1344 were

detected in the basolateral medium by 3 h after inoculation (10,275.0 ± 6,242.5 and 2,257.5 ± 1,837.0 CFU/ml, respec-tively), with the number of P. aeruginosa PAO1 WT being greater than that of S. typhimurium SL1344 at each time of assessment (eg., 67,000.0 ± 9,850.2 versus 28,250.0 ± 7,675.7 CFU/ml at 10 h after inoculation, P = 0.00217) (Fig. 2). In contrast, only a small number of P. aeruginosa ∆ABM bacteria were detected in the basolateral medium at 7 h after inoculation (2,450.0 ± 1,424.3 CFU/ml), but E. coli RDEC-1 was not found in the basolateral medium until at least 12 h after inoculation (Fig. 2).

Fig. 2. BEAS-2B cell monolayer system. Penetration of Pseudomonas aeruginosa PAO1 WT and its mexAB-oprM deletion mutant ∆ABM, Salmonella typhimurium SL1344, and Escherichia coli RDEC-1 through BEAS-2B cell monolayer. Bacteria were inoculated at 3.5 × 106 CFU/well onto the monolayer on day 8 of culture. Bacteria in the basolateral medium were counted by plating appropriate dilutions at several times up to 12 h. Assays were done in triplicate, and results are expressed as the mean ± S.D.

Fig. 3. Calu-3 cell monolayer system. Penetration of Pseudomonas aeruginosa PAO1 WT and its mexAB-oprM deletion mutant ∆ABM, Salmonella typhimurium SL1344, and Escherichia coli RDEC-1 through the Calu-3 cell monolayer. Bacteria were inoculated at 3.5 × 106 CFU/well onto the monolayer on day 8. Bacteria in the basolateral medium were counted by plating appropriate dilutions at several times up to 12 h. Assays were done in triplicate, and results are expressed as the mean ± S.D.

Y. Hirakata et al.18

In case of the Calu-3 cell monolayer sytem, P. aerugi-nosa PAO1 WT was detected in the basolateral medium by 3 h after inoculation (14,750.0 ± 9,912.1 CFU/ml) and P. aeruginosa ∆ABM was found by 4 h after inoculation (17,250.0 ± 1,707.8 CFU/ml) (Fig. 3). In contrast to the results obtained with the MDCK and BEAS-2B cells mono-layer systems, S. typhimurium SL1344 only appeared in the basolateral medium from 6 h after inoculation, and the num-ber of bacteria was always smaller than for the two P. aeru-ginosa strains at each time of assessment (eg., 47,000.0 ± 12,850.5 [WT] and 37,500.0 ± 2,886.8 [∆ABM] versus 17,250.0 ± 9,142.4 [SL1344] CFU/ml at 7 h after inocula-tion, P < 0.0001 and 0.00315, respectively). Similarly, only a small number of E. coli RDEC-1 were detected in the basolateral medium at 8 hours after inoculation (9,375.0 ± 2,173.4 CFU/ml) (Fig. 3).

DiscussionThe present study has demonstrated that BEAS-2B and

Calu-3 cells form monolayers with tight junctions, as judged by the TER values of the BEAS-2B and Calu-3 cells. Because the TER for A549 cells did not increase above baseline, these cells did not seem to form a monolayer with tight junctions in the present culture system. These findings suggest that monolayers of BEAS-2B and Calu-3 cells could be used for evaluation of bacterial invasiveness, while A549 cells are not suitable for this purpose. In addition, BEAS-2B and Calu-3 cell monolayers showed smaller vari-ations of their TER values than the MDCK cell monolayer.

Interestingly, the number of bacteria recovered from the basolateral chamber showed some differences between BEAS-2B and Calu-3 cell monolayer systems, depending on the test strains. The timing of detection and the bacterial count in the basolateral chamber were similar for the BEAS-2B and MDCK cell monolayer systems, but the results obtained with the Calu-3 cell monolayer system were quite different. S. typhimurium SL1344 showed poor invasive-ness in the Calu-3 system, while this microorganism dem-onstrated strong invasiveness in the MDCK and BEAS-2B systems. The BEAS-2B system seemed to be more stable since noninvasive E. coli was not found in the basolateral medium within 12 h and the mexAB-oprM deletion mutant (∆ABM) of P. aeruginosa PAO1 was not detected in the basolateral medium at an early time.

We evaluated bacterial invasion in the present study, but these systems may also be useful to examine changes of infected epithelial cells such as apoptosis or necrosis, and to assess virulence factors other than invasiveness. Because our findings showed clear differences of bacterial kinetics, except for P. aeruginosa, between BEAS-2B and Calu-3 cells, it may be important to choose the most suitable sys-tem for each bacterial pathogen.

In conclusion, both the BEAS-2B and Calu-3 cell monolayer systems could be useful for in vitro evaluation of the invasiveness of respiratory pathogens. Future investiga-tions may apply these systems to assess respiratory patho-

gens, such as S. pneumoniae, Haemophilus influenzae, Klebsiella pneumoniae, and Acinetobacter species, in addi-tion to P. aeruginosa, as well as comparing bacterial viru-lence between these in vitro systems and in vivo models such as mice with pneumonia.

ReferencesAttali, C., Durmort, C., Vernet, T. & Di Guilmi, A.M. (2008) The

interaction of Streptococcus pneumoniae with plasmin medi-ates transmigration across endothelial and epithelial monolay-ers by intercellular junction cleavage. Infect. Immun., 76, 5350-5356.

Bergmann, S., Lang, A., Rohde, A.M., Agarwal, V., Rennemeier, C., Grashoff, C., Preissner, K.T. & Hammerschmidt, S. (2009) Integrin-linked kinase is required for vitronectin-mediated internalization of Streptococcus pneumoniae by host cells. J. Cell Sci., 122, 256-267.

Hirakata, Y., Finlay, B.B., Simpson, D.A., Kohno, S., Kamihira, S. & Speert, D.P. (2000) Penetration of clinical isolates of Pseu-domonas aeruginosa through MDCK epithelial cell monolay-ers. J. Infect. Dis., 181, 765-769.

Hirakata, Y., Izumikawa, K., Yamaguchi, T., Igimi, S., Furuya, N., Maesaki, S., Tomono, K., Yamada, Y., Kohno, S., Yamaguchi, K. & Kamihira, S. (1998) Adherence to and penetration of human intestinal Caco-2 epithelial cell monolayers by Pseudo-monas aeruginosa. Infect. Immun., 66, 1748-1751.

Hirakata, Y., Srikumar, R., Poole, K., Gotoh, N., Suematsu, T., Kohno, S., Kamihira, S., Hancock, R.E. & Speert, D.P. (2002) Multidrug efflux systems play an important role in the inva-siveness of Pseudomonas aeruginosa. J. Exp. Med., 196, 109-118.

Hyland, K.A., Wang, B. & Cleary, P.P. (2007) Protein F1 and Streptococcus pyogenes resistance to phagocytosis. Infect. Immun., 75, 3188-3191.

Kondo, A., Hirakata, Y., Gotoh, N., Fukushima, K., Yanagihara, K., Ohno, H., Higashiyama, Y., Miyazaki, Y., Nishide, K., Node, M., Yamada, Y., Kohno, S. & Kamihira, S. (2006) Quorum sensing system lactones do not increase invasiveness of a MexAB-OprM efflux mutant but do play a partial role in Pseu-domonas aeruginosa invasion. Microbiol. Immunol., 50, 395-401.

Leung, K.Y. & Finlay, B.B. (1991) Intracellular replication is essential for the virulence of Salmonella typhimurium. Proc. Natl. Acad. Sci. USA, 88, 11470-11474.

Li, X.Z., Zhang, L., Srikumar, R. & Poole, K. (1998) β -lactamase inhibitors are substrates for the multidrug efflux pumps of Pseudomonas aeruginosa. Antimicrob. Agents Chemother., 42, 399-403.

Okabe, T., Yamazaki, Y., Shiotani, M., Suzuki, T., Shiohara, M., Kasuga, E., Notake, S. & Yanagisawa, H. (2008) An amino acid substitution in PBP-3 in Haemophilus influenzae associate with the invasion to bronchial epithelial cells. Microbiol. Res. [Epub ahead of print]

Rajam, G., Phillips, D.J., White, E., Anderton, J., Hooper, C.W., Sampson, J.S., Carlone, G.M., Ades, E.W. & Romero-Steiner, S. (2008) A functional epitope of the pneumococcal surface adhesin A activates nasopharyngeal cells and increases bacteri-al internalization. Microb. Pathog., 44, 186-196.

Spinaci, C., Magi, G., Zampaloni, C., Vitali, L.A., Paoletti, C., Catania, M.R., Prenna, M., Ferrante, L., Ripa, S., Varaldo, P.E. & Facinelli, B. (2004) Genetic diversity of cell-invasive erythromycin-resistant and -susceptible group A streptococci determined by analysis of the RD2 region of the prtF1 gene. J. Clin. Microbiol., 42, 639-644.

Stein, M., Kenny, B., Stein, M.A. & Finlay, B.B. (1996) Characterization of EspC, a 110-kilodalton protein secreted by enteropathogenic Escherichia coli which is homologous to

Monolayer Systems Evaluating Bacterial Invasiveness 19

members of the immunoglobulin A protease-like family of secreted proteins. J. Bacteriol., 178, 6546-6554.

Swords, W.E., Buscher, B.A., Ver Steeg Ii, K., Preston, A., Nichols, W.A., Weiser, J.N., Gibson, B.W. & Apicella, M.A. (2000)

Non-typeable Haemophilus influenzae adhere to and invade human bronchial epithelial cells via an interaction of lipooli-gosaccharide with the PAF receptor. Mol. Microbiol., 37, 13-27.