The majority of Helicobacter pylori reside on gastric epithelial cell surfaces and in the overlying mucus, but a small fraction of H. pylori enter host epithelial and immune cells. To explore the role of the nudA invasin in host cell entry, a ΔnudA deletion derivative of strain J99 was constructed and transformants were verified by PCR and by fluorescence in situ hybridization. AGS cells were inoculated with either wild type (WT) strain J99 or its ΔnudA mutant to determine the fraction of bacteria that were bound to the cells and were present inside these cells using the gentamicin protection assay. We observed no significant difference between either the density of H. pylori bound to AGS cell membranes or the density of intracellular H. pylori. To further explore this finding, separate chambers of each culture were fixed in glutaraldehyde for transmission electron microscopy (TEM) and immunogold TEM. This addition to the "classical" gentamicin assay demonstrated that there were significantly more intracellular, and fewer membrane-bound, H. pylori in WT-infected AGS cells than in ΔnudA allele infected cells. Thus, the sum of intracellular and membrane-bound H. pylori was similar in the two groups. Since no other similar TEM study has been performed, it is at present unknown whether our observations can be reproduced by others Taken together however, our observations suggest that the "classical" gentamicin protection assay is not sufficiently sensitive to analyze H. pylori cell entry and that the addition of TEM to the test demonstrates that nudA plays a role in H. pylori entry into AGS cells in vitro. In addition, deletion of the invasin gene appears to limit H. pylori to the AGS cell surface, where it may be partly protected against gentamicin. In contrast, this specific environment may render H. pylori more vulnerable to host defense and therapeutic intervention, and less prone to trigger normal immune, carcinogenic, and other developmental response pathways.