The transport of the antiparasitic agents, ivermectin, selamectin and moxidectin was studied in human intestinal epithelial cell monolayers (Caco-2) and canine peripheral blood lymphocytes (PBL). Both models expressed the mdr1-coded 170 kDa ATP-binding cassette (ABC) transporter P-glycoprotein (P-gp). Fluxes of the P-gp substrate rhodamine-123 (Rh-123) across Caco-2 monolayers showed that ivermectin and selamectin acted as potent P-gp inhibitors with IC50 values of 0.1 microm. In contrast, moxidectin was a weaker P-gp inhibitor with an IC50 of 10 microm. The transport of radiolabelled ivermectin, selamectin and moxidectin through Caco-2 monolayers showed that ivermectin, selamectin and moxidectin were P-gp substrates with secretory/absorptive ratios of 7.5, 4.7 and 2.6 respectively. Secretory transport of [3H]-ivermectin and [3H]-selamectin was blocked by the P-gp inhibitor, verapamil. Ivermectin and selamectin inhibited the efflux of Rh-123 from PBL and the concentration of inhibition was similar to that of verapamil. In contrast, moxidectin did not have a significant effect on Rh-123 efflux from PBL. The data suggest that ivermectin and selamectin are potent P-gp substrates, while moxidectin is a weak one.

No abstract

Fluxes of the anti-parasitic agents, [(3)H]-ivermectin, [(3)H]-selamectin and [(3)H]-moxidectin were studied across non-transfected and transfected canine kidney epithelial monolayers, MDCK II/wt, MDCK II-MDR1, MDCK II-MRP1 and MDCK II-MRP2. All four lines surprisingly expressed significant levels of P-glycoprotein (P-gp), coded for by MDR1, but MDCK II-MDR1 expressed increased levels compared to the other lines. MDCK II-MRP1 and MDCK II-MRP2 expressed increased levels of MRP1 and MRP2 respectively. Fluxes of [(3)H]-ivermectin, [(3)H]-selamectin, [(3)H]-moxidectin, and the P-gp substrates, rhodamine-123 and DiOC(2), were polarized in the basolateral-to-apical (secretory) direction across the four lines. Selected MRP inhibitors used in relevant pharmacological concentrations did not block the secretory fluxes of either [(3)H]-ivermectin or [(3)H]-selamectin in either the non-transfected or MRP-transfected lines. In contrast, secretory fluxes of ivermectin and selamectin were inhibited in all four lines by the P-gp inhibitor, verapamil. These data confirm that ivermectin and selamectin are substrates for P-gp in four additional cell lines, but suggest that they are not significant substrates for MRP1 or MRP2 where there is background expression of P-gp. Since this pattern of expression also pertains on the blood-brain barrier, it is unlikely that MRP1 and MRP2 play a significant role in ivermectin and selamectin blood: brain distribution in vivo.

Sequence and structure comparisons of various glutamate dehydrogenases (GDH) and other nicotinamide nucleotide-dependent dehydrogenases have potentially implicated certain residues in coenzyme binding and discrimination. We have mutated key residues in Clostridium symbiosum NAD+-specific GDH to investigate their contribution to specificity and to enhance acceptance of NADPH. Comparisons with E. coli NADPH-dependent GDH prompted design of mutants F238S, P262S, and F238S/P262S, which were purified and assessed at pH 6.0, 7.0, and 8.0. They showed markedly increased catalytic efficiency with NADPH, especially at pH 8.0 (∼170-fold for P262S and F238S/P262S with relatively small changes for NADH). A positive charge introduced through the D263K mutation also greatly increased catalytic efficiency with NADPH (over 100-fold at pH 8) and slightly decreased activity with NADH. At position 242, “P6” of the “core fingerprint,” where NAD+- and NADP+-dependent enzymes normally have Gly or Ala, respectively, clostridial GDH already has Ala. Replacement with Gly produced negligible shift in coenzyme specificity.

Clostridial glutamate dehydrogenase mutants, designed to accommodate the 2′‐phosphate of disfavoured NADPH, showed the expected large specificity shifts with NAD(P)H. Puzzlingly, similar assays with oxidized cofactors initially revealed little improvement with NADP+, although rates with NAD+ were markedly diminished. This article reveals that the enzyme’s discrimination in favour of NAD+ and against NADP+ had been greatly underestimated and has indeed been abated by a factor of > 16 000 by the mutagenesis. Initially, stopped‐flow studies of the wild‐type enzyme showed a burst increase of A340 with NADP+ but not NAD+, with amplitude depending on the concentration of the coenzyme, rather than enzyme. Amplitude also varied with the commercial source of the NADP+. FPLC, HPLC and mass spectrometry identified NAD+ contamination ranging from 0.04 to 0.37% in different commercial samples. It is now clear that apparent rates of NADP+ utilization mainly reflected the reduction of contaminating NAD+, creating an entirely false view of the initial coenzyme specificity and also of the effects of mutagenesis. Purification of the NADP+ eliminated the burst. With freshly purified NADP+, the NAD+ : NADP+ activity ratio under standard conditions, previously estimated as 300 : 1, is 11 000. The catalytic efficiency ratio is even higher at 80 000. Retested with pure cofactor, mutants showed marked specificity shifts in the expected direction, for example, 16 200 fold change in catalytic efficiency ratio for the mutant F238S/P262S, confirming that the key structural determinants of specificity have been successfully identified. Of wider significance, these results underline that, without purification, even the best commercial coenzyme preparations are inadequate for such studies.

We have utilised a Chemometrics based method to predict the function of enzymes taken from the ENZYME database from primary sequence alone. 1621 parameters were derived from the primary sequences, which include sequence length, number of observations of each amino acid in the sequence and pairwise amino acid distributions. This data was then fed into a software package (Pirouette, Infometrix) along with the EC number of each enzyme. A Soft Independent Modelling of Class Analogy (SIMCA) model was then built from this data and several jack-knife tests were conducted to test the prediction power of the model. Overall, the model performed significantly better (>40%) than the 16.7% accuracy of a random model.This approach to the prediction of protein function has several advantages over other methods. It does not require that the protein has a known and characterised homologue, and so is superior to homology based methods in this respect. This method is semi-empirical and allows determination of the statistical significance of any result and is therefore extremely robust. We are now utilising this method to predict the functions of protein coding genes in the Yeast genome.

No abstract

No abstract