α-Haemolysin (HlyA) from uropathogenic Escherichia coli has been demonstrated to be a significant virulence factor for ascending urinary tract infections. Once the E. coli reach the well-vascularised kidneys, there is a high risk of bacteraemia and a subsequent septic host response. Despite this, HlyA has the potential to accelerate the host response both directly and via its ability to facilitate adenosine triphosphate release from cells. It has not been settled whether HlyA aggravates bacteraemia into a septic state. To address this, we used an E. coli strain in a model of acute urosepsis that was either transfected with a plasmid containing the full HlyA operon or one with deletion in the HlyA gene. Here, we show that HlyA accelerates the host response to E. coli in the circulation. Mice exposed to HlyA-producing E. coli showed massively increased proinflammatory cytokines, a substantial fall in circulating thrombocytes, extensive haematuria, and intravascular haemolysis. This was not seen in mice exposed to either E. coli that do not secrete HlyA or vehicle controls. Consistent with the massive host response to the bacteria, the mice exposed to HlyA-producing E. coli died exceedingly early, whereas mice exposed to E. coli without HlyA production and vehicle controls survived the entire observation period. These data allow us to conclude that HlyA is a virulence factor that accelerates a state of bacteraemia into fulminant sepsis in a mouse model.

Urosepsis is a potentially life-threatening, systemic reaction to uropathogenic bacteria entering the bloodstream of the host. One of the hallmarks of sepsis is early thrombocyte activation with a following fall in circulating thrombocytes as a result of intravascular aggregation and sequestering of thrombocytes in the major organs. Development of a thrombocytopenic state is associated with a poorer outcome of sepsis. Uropathogenic Escherichia coli frequently produce the pore-forming, virulence factor α-haemolysin (HlyA), of which the biological effects are mediated by ATP release and subsequent activation of P2 receptors. Thus, we speculated that inhibition of thrombocyte P2Y1 and P2Y12 receptors might ameliorate the septic response to HlyA-producing E. coli. The study combined in vitro measurements of toxin-induced thrombocyte activation assessed as increased membrane abundance of P-selectin, fibronectin and CD63 and data from in vivo murine model of sepsis-induced by HlyA-producing E. coli under infusion of P2Y1 and P2Y12 antagonists. Our data show that the P2Y1 receptor antagonist almost abolishes thrombocyte activation by pore-forming bacterial toxins. Inhibition of P2Y1, by constant infusion of MRS2500, markedly increased the survival in mice with induced sepsis. Moreover, MRS2500 partially prevented the sepsis-induced depletion of circulating thrombocytes and dampened the sepsis-associated increase in proinflammatory cytokines. In contrast, P2Y12 receptor inhibition had only a marginal effect in vivo and in vitro. Taken together, inhibition of the P2Y1 receptor gives a subtle dampening of the thrombocyte activation and the cytokine response to bacteraemia, which may explain the improved survival observed by P2Y1 receptor antagonists.

The urinary tract is frequently a primary infection site in patients with severe sepsis or septic shock 1. Urinary tract infections are most commonly caused by E.coli sub‐strains that produce several virulence factors 4– 8, of which α‐haemolysin (HlyA) is the most abundant 9. Our previous data demonstrate that the biological effects of HlyA are intimately associated with extracellular ATP signalling secondarily to ATP release from cells through the HlyA‐pore. Our previous data support that HlyA is largely responsible for the septic symptoms observed in response to bacteraemia with uropathogenic E.coli. ATP and its degradation products are aggressive activators of immune cells 10 and platelets 11– 13, and prime candidates for promoting the intravascular coagulation observed during septic shock. We tested the effect of antagonising platelet P2Y receptors (P2Y12 and P2Y1) during sepsis with uropathogenic E.coli(ARD6, O6:K13:H1) in anaesthetised mice. In aggreement with the literature, infusion of the P2Y12 antagonist cangrelor (8.6 or 86mg/hour) did not change survival or any tested sepsis parameters. However, the P2Y1 inhibitor MRS2500 reduced platelet activation by pore‐forming bacterial toxins (70–80%) in vitro, measured as surface exposure of fibronectin, DC63 and P‐selectin. In vivo, infusion of MRS2500 (624μg/hour) increased the survival in mice exposed to uropathogenic E. coli. Moreover, infusion of MRS2500 partially prevented the 60% fall in platelet count otherwise seen in response to sepsis in control mice. Interestingly, targeting of P2Y2 receptors, which is highly expressed on neutrophils and monocytes 14 but not in platelets, showed that lack of P2Y2 receptors entirely prevented the fall in circulating platelets in response to sepsis. Moreover, P2Y2 receptor‐deficient mice die much earlier in response to intravenous application of uropathogenic E. coli compared to controls. By flow cytometry, we are able to detect increased complex formation between thrombocytes (CD42) and neutrophils (Ly6‐G) /monocyte (Ly6‐G/CD11B). These data could potentially suggest that complex formation between thrombocytes and neutrophils and/or monocytes are important for the observed reduction in free thrombocyte numbers during sepsis and that the complexes could dampen the septic reaction to the circulating bacteria and improve survival. Support or Funding Information This project is funded byThe Independent Research Foundation ‐ DenmarkThe Augustinus Foundation NicolleCritCareClin29699715KennedyChest15625568GottsMatthayBMJ353i1585JohnsonClinMicrobiolRev480128ConnellPNAS93982732SvanborgAdvExpMedBiol76420510WulltMolMicrobiol3845664JohnsonStellJInfectDis18126172CavalieriMicrobiolRev4832643LandAmJTransplant16332237JinJBiolChem273203034DanielJBiolChem273202429JarvisBrJPharmacol12927582CliffordAmJPhysiol273C97387

UTI is an everyday encounter in the primary health sector, and in the US, UTI accounts for 8.1% of all hospital admissions 1. In Denmark, with 5.6 mill citizens, over 500,000 prescriptions of antibiotics for UTIs are issued per year 2and even in the primary sector, one finds a high incidence of antibiotic resistance in bacteria that cause UTI 3. This stresses the need for new approaches to supplement standard antibiotic treatment to control UTIs. UTIs are most frequently caused by E. coli that produce a variety of virulence factors 4– 8, of which α‐haemolysin (HlyA) is the most consistent. Our previous in vitro studies demonstrate that the biological effects of HlyA are to a large extent are secondary to extracellular ATP signalling. Interestingly, insertion of HlyA into a cells membrane inflicts non‐lytic ATP release from cells directly through the HlyA‐pore. In our murine model of ascending UTI, it is easily confirmed that HlyA is a virulence factor that promotes development of pyelonephritis and we have shown that HlyA is responsible for the majority of the septic symptoms in a murine model of urosepsis. Therefore, we hypothesise that release of host ATP may be essential for the virulence promoted by HlyA. We test the hypothesis by a combination of bacterial growth studies (flow cytometry and OD660) and in vivo murine model of pyelonephritis. Our data show that E.coli grow exceedingly well in urine from healthy volunteers, despite its low iron content17. Within the first couple of hours, the growth rate of E.coli is significantly higher in urine than in Lysogeny Broth (LB). In contrast, the E.coli practically does not grow in physiological saline (HEPES buffered salt solution, HBS) containing 5.6 mM glucose. Moreover, we found that E.coli grows faster in hyperosmotic compared to dilute urine and addition of as little as 25μl human urine per ml HBS was sufficient to stimulate E.coli growth maximally. This supports that the initial accelerated growth in urine compared to LB medium must be a substance found explicitly in the urine. By screening, we exclude urea, iron, creatinine, proteins/polypeptides and taurin to have any effect on bacterial growth. We did, however, find that NH4+ was necessary as a nitrogen donor. Moreover, it was observed that ATP degradation products ADP and AMP, which are present in the urine in micromolar concentrations, markedly supported and accelerated E. coli growth in HBS. Our in vitro data signifies that dilution of the urine potentially could protect against ascending UTI and our initial in vivo studies of ascending UTI in mice support that the urine has to be diluted to values below 300 mosmol/l to be able to detect this effect. Therefore, simple dilution would not provide sufficient protection against ascending UTIs. Consequently, we are currently specifically targeting the ADP/AMP and NH4+ pathways in our strive to develop new treatment strategies for prevention of UTI. Support or Funding Information This project was funden byThe Independent Research fund ‐DenmarkThe Augustinus F...