Quantitative invasion assay values were calculated as follows: We used an in vitro assay according to Trombert et al. (2010), modified from the method described by McCormick et al. (1993). Briefly, the colon carcinoma HT-29 cell line was grown to confluence (18–21 days) on 3.0-μm Lapatinib molecular weight pore-size filters (or transwells, Millicell®, Millipore) with glucose-free RPMI (Gibco). Each transwell was inoculated individually to the

apical surface with 400 μL of approximately 1 × 107 CFU mL−1 of bacterial cultures and immediately incubated for 60 min at 37 °C. After extensive washing with sterile PBS, the extracellular bacteria were killed by treatment of monolayers with gentamicin (50 μg mL−1). Immediately after gentamicin treatment, the medium from basal compartment

of the epithelial cell monolayer was collected and plated for CFU to assess the number of bacteria that passed through the cell monolayer. The polarization of cells was confirmed by transepithelial electrical Antiinfection Compound Library purchase resistance and transmission electron microscopy (data not shown). All results are expressed as means±SD of an individual experiment performed in triplicate. P-values were calculated according to Student’s t-test, and P<0.05 or P<0.01 values were considered statistically significant. To assess whether the sopD2 locus is a pseudogene in the serovar Typhi, we compared the available sequences of S. Typhi CT18 and S. Typhi Ty2 (McClelland et al., 2001; Parkhill et al., 2001; Deng et al., 2003). We observed that the sequence of sopD2 in S. Typhi contains a frameshift at codon 48 resulting in

a premature stop codon that disrupts the expected ORF. Accordingly, the online databases report sopD2 as a pseudogene (Fig. 1). To confirm the presence of this frameshift in our S. Typhi clinical strain collection, PCR assays were carried out using SopD21 and SopD22 primer pairs. The primers yield an 1800-bp amplicon in both S. Typhi and S. Typhimurium and were used to test the clinical strains obtained from Chilean typhoid patients (STH collection, Hospital Dr Lucio Córdova, Chile). The PCR products were sequenced and in silico comparison was performed with the reference strains (S. Typhi CT18, S. Typhimurium LT2 and Salmonella enterica serovar Paratyphi A ATCC 9150) using Fossariinae the bioinformatic available program (clustalw; http://www.ebi.ac.uk/Tools/msa/clustalw2/). Our results indicated that the deletion described in S. Typhi CT18 is conserved among S. Typhi clinical strains (Fig. 2). Therefore, the sopD2 frameshift mutation seems to be a feature in the genome of serovar Typhi. Recently, we reported that S. Typhi harboring the sseJ gene from S. Typhimurium significantly disrupted HT-29 monolayer compared with the wild-type strain (Trombert et al., 2010). In the same way, we infected polarized HT-29 monolayers with the S. Typhi wild-type and S. Typhi sopD2STM strains.