Multidrug-resistant enterobacterales in Oman : molecular epidemiology and therapeutic insights

Sammanfattning: The spread of antibiotic resistance is a concerning issue causing limited treatment options for bacterial infections, particularly with Gram-negative bacteria. Surveillance and epidemiological studies help to determine the magnitude of the problem as well as to establish early measures to slow down the spread of resistance and consequently increase antibiotic lifespan. Currently, there is a visible paucity of published data about resistance from the Arabian Peninsula. In this thesis, we studied a collection of carbapenem nonsusceptible E. coli (n=35) and K. pneumoniae (n=237) isolated in 2015 from various hospitals in Oman. We aimed at identifying resistance mechanisms, mapping the bacterial population structure, investigating bacterial fitness, and studying potential treatment options available to tackle infections caused by such multidrug-resistant strains. These aims were addressed in five papers as discussed below. NDM and OXA-48 were the only carbapenemases we found in this collection, both among E. coli (Paper I) and K. pneumoniae (Paper II). The pattern of resistance among the isolates from Arabian Peninsula mimics the pattern reported from the Indian subcontinent, most likely due to the close socioeconomic interactions between them. Both regions lack KPC enzymes, which are commonly seen in China and the US from strains belong to ST11 and ST258, respectively. Despite ST11 being predominant in this collection, we did not detect KPC. Yet, we detected a high-risk clone of E. coli, ST131-H30Rx-CTX-M-15. Additionally, we identified newly emerging clones of K. pneumoniae and E. coli such as ST231 and ST1193-H64RxC, respectively. Nearly 10% of the K. pneumoniae isolates in our collection were colistin resistant which prompted us to study the mechanisms of colistin resistance (Paper III). MgrB-inactivation by insertion elements was seen in 8 isolates while other mutations were seen in other chromosomal genes known to be involved in colistin resistance e.g. pmrB, phoPQ and crrB. However, we did not detect mcr genes. Collectively, the genetic alterations are thought to reduce the net negative charge in bacterial cell wall, hence lowering the binding affinity of colistin. Our data underscores that there is no reduction in the surface charge in colistinresistant K. pneumoniae, due to the MgrB-insertion (Paper IV). The genetic alteration might lead to other structural changes in the cell wall such as altering hydrophobicity, which required further investigation. Also, our data shows no difference in the survival rates of colistin resistant and susceptible strains in blood, serum and zebrafish model. Thus, gaining resistance against colistin does not infer a fitness cost in K. pneumoniae with MgrB-insertion (Paper IV). Additionally, colistin and LL-37 share similar binding mechanism which suggest there might be a cross-resistance between them. Our data supported this hypothesis, but only at high concentrations of LL-37 ( ≥ 50 mg/L) (Paper IV). Finally, we studied available options to treat infections caused by multidrug-resistant strains. Combining colistin and rifampicin showed good in vitro activity against multidrug-resistant strains of E. coli (Paper V) and K. pneumoniae (Paper III). To summarise, we conducted comprehensive genomic analysis of E. coli and K. pneumoniae isolates from Oman to reveal the resistance mechanism, their impact on bacterial cell structural and if there is a fitness cost inferred by the resistance mechanisms. Finally, we studied combination therapy as an available option at hand for tackling infections caused by multidrug-resistant strains.