Helicobacter pylori uptake and efflux: basis for intrinsic susceptibility to antibiotics in vitro

Abstract

We previously demonstrated (M. M. Exner, P. Doig, T. J. Trust, and R. E. W. Hancock, Infect. Immun. 63:1567-1572, 1995) that Helicobacter pylori has at least one nonspecific porin, HopE, which has a low abundance in the outer membrane but forms large channels. H. pylori is relatively susceptible to most antimicrobial agents but less susceptible to the polycationic antibiotic polymyxin B. We demonstrate here that H. pylori is able to take up higher basal levels of the hydrophobic fluorescent probe 1-N-phenylnaphthylamine (NPN) than Pseudomonas aeruginosa or Escherichia coli, consistent with its enhanced susceptibility to hydrophobic agents. Addition of polymyxin B led to a further increase in NPN uptake, indicative of a self-promoted uptake pathway, but it required a much higher amount of polymyxin B to yield a 50% increase in NPN uptake in H. pylori (6 to 8 microg/ml) than in P. aeruginosa or E. coli (0.3 to 0.5 microg/ml), suggesting that H. pylori has a less efficient self-promoted uptake pathway. Since intrinsic resistance involves the collaboration of restricted outer membrane permeability and secondary defense mechanisms, such as periplasmic beta-lactamase (which H. pylori lacks) or efflux, we examined the possible role of efflux in antibiotic susceptibility. We had previously identified in H. pylori 11637 the presence of portions of three genes with homology to potential restriction-nodulation-division (RND) efflux systems. It was confirmed that H. pylori contained only these three putative RND efflux systems, named here hefABC, hefDEF, and hefGHI, and that the hefGHI system was expressed only in vivo while the two other RND systems were expressed both in vivo and in vitro. In uptake studies, there was no observable energy-dependent tetracycline, chloramphenicol, or NPN efflux activity in H. pylori. Independent mutagenesis of the three putative RND efflux operons in the chromosome of H. pylori had no effect on the in vitro susceptibility of H. pylori to 19 antibiotics. These results, in contrast to what is observed in E. coli, P. aeruginosa, and other clinically important gram-negative bacteria, suggest that active efflux does not play a role in the intrinsic resistance of H. pylori to antibiotics.

FIG. 1

Uptake of NPN by E. coli (A) and H. pylori (B) in the presence of NPN alone (dashed lines), after the addition of NPN and polymyxin B (solid lines), and after incubation of E. coli with CCCP prior to the addition of NPN and polymyxin B (dotted lines). a. u., arbitrary unit.

FIG. 2

Phylogenetic tree showing the relationship of the H. pylori RND pump proteins to other RND pump proteins. The three putative H. pylori RND efflux pump proteins (HefC, HefF, and HefI) are divergent from the clusters of RND efflux pump proteins involved in the efflux of divalent cations (CnrA [Ralstonia eutropha], NccA [Alcaligenes denitrificans], HelA [Legionella pneumophila], and CzcA [R. eutropha]) and those pumps involved in multiple-drug efflux (AcrA [E. coli], MexB [P. aeruginosa], MexD [P. aeruginosa], and MtrD [N. gonorrhoeae]).

FIG. 3

In vitro and in vivo expression of the H. pylori hef operons; RT-PCR of H. pylori grown in vitro (A) and in vivo (B). Lanes 1 to 3 were done with hefF-specific PCR primers, while lanes 4 to 6 were done with the hefI-specific PCR primers. The RT-PCR mixtures in lanes 1 and 4 contained reverse transcriptase, Taq DNA polymerase, and RNA; lanes 2 and 5 are the negative controls and contained Taq DNA polymerase and RNA; lanes 3 and 6 are the positive controls and contained all the nucleic acids (prior to DNase treatment) plus Taq DNA polymerase. Lanes S, size standards.

FIG. 4

CCCP inhibition of [³H]chloramphenicol uptake by H. pylori (solid bars) and H. pylori pretreated with 40 μM CCCP (open bars). The cells were incubated separately in the presence of [³H]chloramphenicol. Aliquots were removed at 1, 15, 30, and 45 min following the addition of [³H]chloramphenicol and processed as stated in Materials and Methods. The amount of radioactivity on the filter was determined by scintillation counting. The relative uptake is defined as the percentage of radioactivity (counts per minute) accumulated by the cells divided by the total radioactivity (counts per minute) present. Representative results from one of three experiments are shown.