Comparative analysis of new peptide conjugates of antitubercular drug candidates-Model membrane and in vitro studies.

Novel peptide conjugates of two antitubercular drug candidates were synthesised and characterised using new tuftsin peptide derivative (OT14) as carrier moiety. As antitubercular drug candidates two pyridopyrimidine derivatives, TB803 (2-allylamino-4-oxopyrido[1,2-a]pyrimidine-3-carbaldehyde) and TB820 (4-oxo-2-(pyrrolidin-1-yl)-pyrido[1,2-a]pyrimidin-3-carbaldehyde) inhibiting vital enzyme of Mycobacterium tuberculosis were applied. Membrane affinity of the compounds TB803 and TB820 and their peptide conjugates was evaluated using experimental lipid mono- and bilayer models. Penetration ability was assessed tensiometrically from Langmuir monolayer study and applying quartz crystal microbalance for the supported lipid bilayer (SLB) system. Minimal inhibitory concentration (MIC) values remained in a similar micromolar range for both of the conjugates while their cellular uptake rate was improved significantly compared to the drug candidates. A correlation was found between membrane affinity properties and results of in vitro biological investigations. Analysis of physical/structural properties of SLB in contact with bioactive components and visualization of the structural change by atomic-force microscopy (AFM) provided information on the type and route of molecular interaction of drug construction with lipid layers. The possible role of electrostatic interactions between lipid layer and drug candidates was tested in Langmuir-balance experiments using negatively charged lipid mixture (DPPC+DPPG). Especially the peptide conjugates presented increased membrane affinity due to cationic character of the peptide sequence selected for the conjugate formation. That is supposed to be one reason for the enhanced cellular uptake observed in vitro on MonoMac6 cell line. The conjugation of antitubercular agents to a peptidic carrier is a promising approach to enhance membrane affinity, cellular uptake rate and in vitro selectivity.

Prohepcidin levels during human perinatal adaptation.

The aim of this study was to test for the presence of prohepcidin in cord blood, to gauge its alteration during the early postnatal period, and to look for a possible association with neonatal iron homeostasis. Cord blood and postnatal venous blood samples were taken from 20 healthy neonates. In both kinds of samples the presence prohepcidin could be detected. No association was found between cord blood and postnatal samples prohepcidin and iron homeostasis. However, an association is demonstrated between cord blood prohepcidin values and mean cell hemoglobin concentration (MCHC). Prohepcidin increased postnatally in half of the neonates, indicating the active synthesis of the molecule. Interestingly, neonates with detectable non-protein-bound iron levels in cord blood were presented with lower prohepcidin concentrations. Association between cord blood prohepcidin and MCHC may suggest a possible link between hepcidin and fetal iron homeostasis.

Validation of in silico prediction by in vitro immunoserological results of fine epitope mapping on citrate synthase specific autoantibodies.

In silico antibody-antigen binding predictions are generally employed in research to rationalize epitope development. These techniques are widely spread despite their technical limitations. To validate the results of these bioinformatic calculations evidence based comparative in vitro studies are necessary. We have used a well-conserved mitochondrial inner membrane antigen-citrate synthase to develop a model for comparative analysis of the predicted and the immunoserologically verified epitopes of circulating autoantibodies. Epitopes were predicted using accepted tools: the GCG Wisconsin package and TEPITOPE 2000. An overlapping multipin ELISA assay--covering 49% of the citrate synthase molecule--was developed to map autoantibody epitopes of individuals (healthy, systemic autoimmune, and heart transplanted) in different immunopathological conditions. From the 40 synthesized decapeptides 34 were predicted in silico and 27 were validated in vitro. Thirty-two percent of epitopes were recognized by majority of sera 47% by at least one sera. False positive predictions were 21%. There was major difference in the recognized epitope pattern under different immunopathological conditions. Our results suggest that special databases are needed for training and weighing prediction methods by clinically well-characterized samples, due to the differences in the immune response under different health status. The development of these special algorithms needs a new approach. A high number of samples under these special immunological conditions are to be mapped and then used for the "fine tuning" of different prediction algorithms.