It has been shown previously that binding of vesicles and monolayers containing PE (phosphatidylethanolamine) by either erythroid or non-erythroid spectrin proved sensitive to inhibition by purified erythrocyte ankyrin. We tested the lipid-binding affinities of the purified ankyrin-binding domain of beta-spectrin and of its truncated mutants in four ways, by analysing: (1) penetration of 'loose' PE/PC (phosphatidylcholine) monolayers; (2) binding to liposomes in suspension; (3) competition with spectrin for liposomes; and (4) binding of a PE/PC monolayer in a surface plasmon resonance system. The results obtained indicated that the full-length ankyrin-binding domain bound PE/PC mono- and bi-layers with moderate affinity, penetrated monolayers and competed with spectrin for liposomes. Moreover, its truncated mutants that retained the N-terminal part, in contrast with those lacking eight or 38 N-terminal residues (which bound lipid mono- and bi-layers with lower affinity), bound PE/PC mono- and bi-layers with an affinity and capacity comparable with those of the full-length ankyrin-binding domain, and this activity was inhibited by purified erythrocyte ankyrin. The full-length domain, in contrast with the mutant lacking 38 N-terminal residues, induced a small increase in the fluidity of PE/PC membranes when probed with 5'-doxyl stearate, similar to the effect of purified spectrin. Therefore we conclude that the binding site for PE-rich lipids, which is sensitive to ankyrin inhibition, is located in a 38-residue N-terminal fragment of the beta-spectrin ankyrin-binding domain, and that the first eight residues play a key role in this activity.

It was shown previously that an ankyrin-sensitive, phosphatidylethanolamine/phosphatidylcholine (PE/PC) binding site maps to the N-terminal part of the ankyrin-binding domain of β-spectrin (ankBDn). Here we have identified the amino acid residues within this domain which are responsible for recognizing monolayers and bilayers composed of PE/PC mixtures. In vitro binding studies revealed that a quadruple mutant with substituted hydrophobic residues W1771, L1775, M1778 and W1779 not only failed to effectively bind PE/PC, but its residual PE/PC-binding activity was insensitive to inhibition with ankyrin. Structure prediction and analysis, supported by in vitro experiments, suggests that “opening” of the coiled-coil structure underlies the mechanism of this interaction. Experiments on red blood cells and HeLa cells supported the conclusions derived from the model and in vitro lipid-protein interaction results, and showed the potential physiological role of this binding. We postulate that direct interactions between spectrin ankBDn and PE-rich domains play an important role in stabilizing the structure of the spectrin-based membrane skeleton.

Cattle is a reservoir of potentially pathogenic E. coli, bacteria that can represent a significant threat to public health, hence it is crucial to monitor the prevalence of the genetic determinants of virulence and antimicrobial resistance among the E. coli population. The aim of this study was the analysis of the phylogenetic structure, distribution of virulence factors (VFs) and prevalence of antimicrobial resistance among E. coli isolated from two groups of healthy cattle: 50 cows housed in the conventional barn (147 isolates) and 42 cows living on the ecological pasture (118 isolates). The phylogenetic analysis, identification of VFs and antimicrobial resistance genes were based on either multiplex or simplex PCR. The antimicrobial susceptibilities of E. coli were examined using the broth microdilution method. Two statistical approaches were used to analyse the results obtained for two groups of cattle. The relations between the dependent (VFs profiles, antibiotics) and the independent variables were described using the two models. The mixed logit model was used to characterise the prevalence of the analysed factors in the sets of isolates. The univariate logistic regression model was used to characterise the prevalence of these factors in particular animals. Given each model, the odds ratio (OR) and the 95% confidence interval for the population were estimated. The phylogroup B1 was predominant among isolates from beef cattle, while the phylogroups A, B1 and D occurred with equal frequency among isolates from dairy cattle. The frequency of VFs-positive isolates was significantly higher among isolates from beef cattle. E. coli from dairy cattle revealed significantly higher resistance to antibiotics. Some of the tested resistance genes were present among isolates from dairy cattle. Our study showed that the habitat and diet may affect the genetic diversity of commensal E. coli in the cattle. The results suggest that the ecological pasture habitat is related to the increased spreading rate of the VFs, while the barn habitat is characterised by the higher levels of antimicrobial resistance among E. coli.

Commensal Escherichia coli, naturally occurring in the intestinal tract, can be the origin of extraintestinal pathogenic E. coli (ExPEC) strains. ExPEC causes high mortality and significant economic losses in the swine industry in several countries and poses a serious threat to public health worldwide. The aim of this study was to analyze the extended phylogenetic structure and extraintestinal virulence potential in two groups of commensal E. coli isolates from post-weaning piglets and sows. The phylogenetic assignment to eight groups was determined using the revised Clermont phylogenetic typing method in quadruplex PCR. Identification of extraintestinal virulence genes (VGs) and adhesin operon genes was performed using multiplex or simplex PCR. The revised phylogenetic assignment allowed us to distinguish E. coli with significantly higher (groups C and F) or lower (group E) virulence potential in isolates from piglets. The majority of the tested VGs occurred more frequently in isolates from piglets than from sows, with statistically significant differences for seven genes: fimH, papAH, iutA, iroN, ompT, traT, and iss. Complete operons for type I and P fimbriae significantly prevailed among E. coli from piglets. This study provides insight into the extended phylogenetic structure of porcine commensal E. coli and showed that these strains, particularly from piglets, constitute a considerable reservoir of extraintestinal VGs and may increase the potential risk of extraintestinal infections.

It is known that erythroid and non-erythroid spectrins binding of vesicles and monolayers containing PE proved sensitive to inhibition by red blood cell ankyrin. We now show that the bacterially-expressed recombinant peptides representing betaII(brain)-spectrin's ankyrin-binding domain and its truncated mutants showed lipid-binding activity, although only those containing a full-length amino terminal fragment showed high to moderate affinity towards phospholipid mono- and bilayers and a substantial sensitivity of this binding to inhibition by ankyrin. These results are in accordance with our published data on betaI-spectrin's ankyrin-binding domain [Hryniewicz-Jankowska A, et al. Mapping of ankyrin-sensitive, PE/PC mono- and bilayer binding site in erythroid beta-spectrin. Biochem J 2004;382:677-85]. Moreover, we tested also the effect of transient transfection of living cells of several cell-lines with vectors coding for GFP-conjugates including betaII and also betaI full-length ankyrin-binding domain and their truncated fragments on the membrane skeleton organization. The transfection with constructs encoding full-length ankyrin-binding domain of betaII and betaI spectrin resulted in increased aggregation of membrane skeleton and its punctate appearance in contrast to near normal appearance of membrane skeleton of cells transiently transfected with GFP control or construct encoding ankyrin-binding domain truncated at their N-terminal region. Our results therefore indicate the importance of N-terminal region for lipid-binding activity of the beta-spectrin ankyrin-binding domain and its substantial role in maintaining the spectrin-based skeleton distribution.