Inhaltszusammenfassung:
Many antibiotics against Gram-positive bacteria show no activity against Gram-negative species, as uptake is hindered by the additional outer membrane. If the antibiotic’s target is within the cytoplasm, the antibiotic has to overcome both the outer and cytoplasmic membranes. For diffusion across these two membranes, ideal molecule attributes are strongly opposing. For outer membrane diffusion through porins, molecules should be small, hydrophilic and charged, whereas for passive diffusion across the cytoplasmic membrane, molecules should be hydrophobic and uncharged. Negamycin is a natural product antibiotic, which showed better activity against Gram-negative than Gram-positive species, which makes it an interesting molecule for uptake studies, as it has to overcome both outer and inner membranes to reach its target binding site at the ribosome. Negamycin binds both the small ribosomal subunit as well as the nearby tRNA, inhibiting translocation and stabilizing unspecific tRNAs, which leads to a miscoding activity. In derivatization studies, most negamycin analogues did not show improved whole cell activity, although some showed stronger target interactions, indicating that uptake was reduced upon modification. We were therefore interested in studying the uptake pathways of negamycin and used Escherichia coli as a model organism. The first barrier for antibiotics is the outer membrane. As negamycin is small, hydrophilic and positively charged at neutral pH, it has the ideal molecular properties to diffuse through porins, which are water- filled channels spanning the whole outer membrane. In this thesis, it could be shown for the first time that an aminoglycoside antibiotic uses the major E. coli porins OmpF and OmpC, as kanamycin translocation through these porins was demonstrated. For negamycin, only a small decrease in bioactivity was detected when both porin genes ompF and ompC were deleted. Alternative pathways across the outer membrane, like the self-promoted uptake mechanism, could not be observed for negamycin. The deletion of the genes of other porins, which so far have not been associated with antibiotic translocation, had an effect on negamycin susceptibility, namely of OmpN, ChiP, LamB, OmpG, OmpW and OmpA. In vitro, electrophysiological experiments showed translocation of negamycin through the purified proteins OmpN, ChiP, OmpF and OmpC. When measured in parallel, ertapenem, cefotaxime and kanamycin were not able to pass through OmpN. The selective barrier of OmpN was investigated further. To this end, the crystal structure of OmpN was resolved. The constriction region of OmpN, which is the narrowest region of a porin, showed a similar size as OmpF and OmpC. However, amino acid residues ranging inside the barrel structure of the OmpN protein create an electrostatic barrier, which hinders the passage of, e.g., kanamycin, but not of negamycin, as molecular dynamics simulations could show. Investigations of the E. coli BW25113 transcriptome revealed that the expression of the ompN and chiP genes was low, but increased when E. coli was treated with negamycin. These increased transcript levels, however, did not raise the protein amounts to an extent, where the proteins could be detected by Western blot analysis. At the inner membrane, multiple uptake pathways and environmental conditions affecting negamycin activity could be identified. In a peptide-free medium, negamycin preferably uses the ATP-dependent dipeptide transporter Dpp. A decrease in negamycin accumulation was shown by [3H]negamycin measurements in a dppA-deficient mutant. Additionally, all mutants selected on peptide-free agar containing negamycin showed alterations in the dpp operon. Further ATP-dependent peptide transporters and proton-dependent oligopeptide transporters are involved in negamycin uptake as well. The effect of salts on negamycin was investigated and it could be observed that CaCl 2 increases negamycin activity. Other salts, like MgCl2, NaCl, KCl or NH4Cl, did not increase negamycin activity, but NaCl, KCl or NH4Cl actually decreased susceptibility at high salt concentrations. A shift to alkaline pH increased negamycin activity and an additive effect could be observed, when CaCl 2 was added to a peptide-rich medium at pH 8.5. CaCl2 increases binding of negamycin to phospholipids, which was shown by surface acoustic wave measurements. Negamycin forms a complex with Ca2+, which was demonstrated by a reduction of the retardation factor during thin layer chromatography in the presence of CaCl2 and by binding studies using isothermal titration calorimetry. Negamycin activity was also affected by the deletion of genes involved in the respiratory chain and in an anaerobic growth environment, but only in a growth medium with peptides, as here the Dpp-mediated uptake pathway is not available. These results indicate that negamycin activity could be affected by the membrane potential across the cytoplasmic membrane, which may also explain the observations in the pH experiments. This abundance of multiple uptake pathways resulted in very low resistance rates, when E. coli cells were treated with negamycin concentrations equivalent to four times the minimal inhibitory concentration. Finally, a protocol for negamycin fermentation and batch purification could be established. From the supernatant of a Streptomyces purpeofuscus culture, negamycin was purified in two steps using cation exchange chromatography. The purified negamycin was less impure than material generated by total synthesis, as judged by thin layer chromatography and high-performance liquid chromatography. For the biosynthetic negamycin, a purity of >95% could be confirmed by nuclear magnetic resonance spectroscopy. The work summarized in this thesis provides new insights into the uptake of pseudopeptidic natural products into Gram-negative bacteria and on the opportunities of negamycin for entering E. coli. New uptake routes could be identified, both at the outer as well as the inner membranes, which so far have not been associated with negamycin accumulation. These findings improve our understanding of uptake of peptide-like natural products into bacteria and can support the further development of negamycin derivatives.
