Use of membrane filters for measurement of mycobactericidal activity of alkaline glutaraldehyde solution

Bactericidal assays of 2% alkaline glutaraldehyde solution were carried out using Millipore membranes, and the rate of kill was compared with that of mycobacteria in suspension and on Penicylinder surfaces when using the methods recommended for the official tuberculocidal test. The rate of inactivation observed on the membrane filter surface was similar to that achieved using Penicylinders. The absence of visible colonies on the treated membranes provided a direct demonstration of sterility. The use of filter membranes in tuberculocidal tests provides a simple quantitative assay.     

A Rapid Method for Determining the Tuberculocidal Activity of Liquid Chemical Germicides

A rapid, quantitative method has been developed for determining the tuberculocidal activity of liquid chemical germicides. In this method, a test strain of Mycobacterium bovis that carries the firefly luciferase gene is exposed to a germicide, and the surviving bacteria are detected by bioluminescence. The tuberculocidal activities of five commercially available glutaraldehyde-based disinfectants were tested, and all reduced the number of surviving mycobacteria by greater than five orders of magnitude. In contrast, a phenol-based disinfectant with tuberculocidal claims gave less than one order of magnitude reduction of the test organism. With this method for determining tuberculocidal activity, results can be obtained in less than one day, compared with weeks or months for the standard tuberculocidal assays. Received: 6 November 2000 / Accepted: 28 November 2000     

Kinetics of the tuberculocidal response by alkaline glutaraldehyde in solution and on an inert surface

Single cell suspensions of BCG and Mycobacterium tuberculosis were exposed to 2% alkaline glutaraldehyde solution (pH 8·0) and the rate of kill measured at intervals up to 30 min. Residual glutaraldehyde was neutralized with freshly prepared 1% sodium bisulphite. The rate of kill was directly proportional to the temperature and independent of the inoculum size whether the organism was tested in suspension or attached to an inert surface. Glutaraldehyde was slightly more bactericidal for the virulent M. tuberculosis than for the attenuated BCG. A substantial proportion of the mycobacterial population on an inert surface floated off during its exposure to the glutaraldehyde solution but the 'floaters'were killed at an equivalent rate to the attached bacilli. Complete sterility of a standardized suspension of M. tuberculosis could not be achieved within the 10 min period specified by the tuberculocidal assay, although it was usually attained within 20 min.     

Kinetics of the tuberculocidal response by alkaline glutaraldehyde in solution and on an inert surface

Single cell suspensions of BCG and Mycobacterium tuberculosis were exposed to 2% alkaline glutaraldehyde solution (pH 8.0) and the rate of kill measured at intervals up to 30 min. Residual glutaraldehyde was neutralized with freshly prepared 1% sodium bisulphite. The rate of kill was directly proportional to the temperature and independent of the inoculum size whether the organism was tested in suspension or attached to an inert surface. Glutaraldehyde was slightly more bactericidal for the virulent M. tuberculosis than for the attenuated BCG. A substantial proportion of the mycobacterial population on an inert surface floated off during its exposure to the glutaraldehyde solution but the 'floaters' were killed at an equivalent rate to the attached bacilli. Complete sterility of a standardized suspension of M. tuberculosis could not be achieved within the 10 min period specified by the tuberculocidal assay, although it was usually attained within 20 min.     

Biocidal Activities of Glutaraldehyde and Related Compounds

S ummary . Glutaraldehyde in 2% aqueous alkaline solution is rapidly bactericidal and sporicidal, killing 99.99% of the spores of Bacillus anthracis and Clostridium tetani in 15 and 30 min, respectively. It exhibits some degree of tuberculocidal activity, but it appears to be inferior to several other disinfectants when challenged with a large inoculum. Fungicidal action has been demonstrated but not assessed quantitatively. Evidence is presented to show that biocidal activity depends on the availability of two free aldehyde groups in the molecule.     

A more accurate method for measurement of tuberculocidal activity of disinfectants

The current Association of Official Analytical Chemists method for testing tuberculocidal activity of disinfectants has been shown to be inaccurate and to have a high degree of variability. An alternate test method is proposed which is more accurate, more precise, and quantitative. A suspension of Mycobacterium bovis BCG was exposed to a variety of disinfectant chemicals and a kill curve was constructed from quantitative data. Data are presented that show the discrepancy between current claims, determined by the Association of Official Analytical Chemists method, of selected commercially available products and claims generated by the proposed method. The effects of different recovery media were examined. The data indicated that Mycobacteria 7H11 and Middlebrook 7H10 agars were equal in recovery of the different chemically treated cells, with Lowenstein-Jensen agar having approximately the same recovery rate but requiring incubation for up to 3 weeks longer for countability. The kill curves generated for several different chemicals were reproducible, as indicated by the standard deviations of the slopes and intercepts of the linear regression curves.     

Virulence of Mycobacterium tuberculosis and susceptibility to peroxidative killing systems.

At sub-bactericidal concentrations of hydrogen peroxide, Mycobacterium tuberculosis was killed by hydrogen peroxide/peroxidase/halide microbicidal systems. The halide cofactor could be either iodide or, with much lower efficiency, chloride. Omission of any one of the reactants eliminated the tuberculocidal effect. Differences in susceptibility between different strains of M. tuberculosis did not correlate with virulence differences. The observations are discussed in the context of host defence mechanisms against tuberculosis.     

A more accurate method for measurement of tuberculocidal activity of disinfectants.

The current Association of Official Analytical Chemists method for testing tuberculocidal activity of disinfectants has been shown to be inaccurate and to have a high degree of variability. An alternate test method is proposed which is more accurate, more precise, and quantitative. A suspension of Mycobacterium bovis BCG was exposed to a variety of disinfectant chemicals and a kill curve was constructed from quantitative data. Data are presented that show the discrepancy between current claims, determined by the Association of Official Analytical Chemists method, of selected commercially available products and claims generated by the proposed method. The effects of different recovery media were examined. The data indicated that Mycobacteria 7H11 and Middlebrook 7H10 agars were equal in recovery of the different chemically treated cells, with Lowenstein-Jensen agar having approximately the same recovery rate but requiring incubation for up to 3 weeks longer for countability. The kill curves generated for several different chemicals were reproducible, as indicated by the standard deviations of the slopes and intercepts of the linear regression curves.     

Increase of the molecular rigidity of the protein conformation in the intestinal brush-border membranes by lipid peroxidation

The effect of lipid peroxidation on the protein conformation of the porcine intestinal brush-border membranes was studied using a fluorogenic thiol reagent, N-[7-dimethylamino-4-methylcoumarinyl]maleimide (DACM). By a kinetic analysis of the reaction of the membranes with DACM, it was shown that the reaction rate of the SH groups (SHf) of the membrane proteins, whose reaction with the dye is very fast, decreases in proportion to the extent of thiobarbituric acid-reactive substance formation. The difference in the rate of the reaction of the SHf groups for DACM between the control and peroxidized membranes completely disappeared after denaturation of the proteins by treatment with guanidine hydrochloride. The reaction of DACM with the SHf groups of the control membranes accelerated when the temperature was increased with an apparent transition temperature between 25 degrees C and 30 degrees C. On the other hand, no transition was observed in the peroxidized membranes over the temperature range 20-43 degrees C. These results suggest that the conformation around the SHf groups of the proteins in the peroxidized membranes is apparently different from that in the control membranes. A modification of the conformation around the SH groups in the membrane proteins associated with lipid peroxidation was further demonstrated by finding that the quenching efficiency of the fluorescence of the DACM-labeled membranes by Tl+ was markedly decreased after lipid peroxidation. Based on these results, changes in the protein conformation of the porcine intestinal brush-border membranes by lipid peroxidation are discussed.     

Transfer of cholestane spin label between lipid bilayer membranes and its molecular motion in membranes.

The relation between the molecular motion of a steroid in lipid membranes and the transfer rate between membranes was examined using radioactive cholestane spin label. Order parameters of the molecule were determined in bilayers composedof dipalmitoylglycerophosphocholine or egg yolk phosphatidylcholine at various temperatures. The line widths of the ESR signal of the cholestane spin label in membranes, which depend upon the rate of molecular axial rotation in the membranes, were also measured. The temperature dependences of these two parameters and of the transfer rate suggest a close correlation between the rate of molecular axial rotation and the transfer rate.     

Structural requirement for the rapid movement of charged molecules across membranes. Experiments with tetraphenylborate analogues.

Charge-pulse experiments were performed in the presence of structural analogues of tetraphenylborate (TPB) on membranes made of dioleoyl phosphatidylethanolamine and dioleoyl phosphatidylcholine. The analysis of the experimental results using a previously proposed model allowed the calculation of the partition coefficient, beta, and of the translocation rate constant, kappa i. The temperature dependence of the partition coefficients was used to calculate the thermodynamics of the adsorption of the lipophilic ions to the membranes. The analysis of the translocation rate constants obtained at different temperatures yielded detailed information on the free energy of the TPB-analogues within artificial lipid bilayer membranes, and on the activation energy of the translocation rate constants. The adsorption of the different TPB-analogues to the membranes was only slightly affected by their structure, whereas a dramatic influence of the structure on the free energy of the lipophilic ions within the membranes was observed. The free energy of the ions in the membranes decreased from triphenylcyanoborate (TPCB) to tetrakis(3-trifluoromethylphenyl)borate (TTFPB) by more than 31 kJ/mol (7.4 kcal/mol). This could be concluded from the observed increase in the translocation rate constant by almost six orders of magnitude. The change of the free energy in the membrane was used for the estimation of an effective radius of the TPB-analogues with respect to TPB.     

Effect of neuraminidase treatment on the lipid fluidity of the intestinal brush-border membranes

The effect of neuraminidase treatment on the lipid fluidity of the porcine intestinal brush-border membranes was studied using two fluorescence dyes, pyrene and 1,6-diphenyl-1,3,5-hexatriene. By treatment of the membranes with neuraminidase, the fluorescence parameters of pyrene-labeled membranes changed; i.e., a shift of thermal transition temperature, an increase in the fluorescence quenching rate for Tl+ and a decrease in the fluorescence lifetime. These results suggest that the environmental properties around the dye molecules in the membranes change sensitively upon neuraminidase treatment. Perturbation of the lipid domain in the membranes associated with neuraminidase treatment is also demonstrated by a stimulated solubilization of diphenylhexatriene molecules in the membrane lipids, an increased quenching efficiency with Tl+ and a decreased rotational correlation time of diphenylhexatriene-labeled membranes. Based on these results, we conclude that the lipid organization of the membranes is susceptible to neuraminidase treatment and that the membrane lipid fluidity increases by desialylation by the enzyme treatment.     

Mechanism of free fatty acid transfer from rat heart fatty acid-binding protein to phospholipid membranes. Evidence for a collisional process.

Fatty acid binding proteins (FABP) are a family of low molecular weight proteins found in many tissues that actively utilize free fatty acids (ffa). FABP would be expected to have a particularly important role in the heart, where over 80% of energy requirements are derived from oxidation of long chain fatty acids. The precise physiological function of heart FABP (H-FABP) has not been definitively identified, although it is thought to play a role in intracellular ffa transport. To examine the possible role of H-FABP in cardiac myocyte transfer of ffa, we examined the transfer of fluorescent anthroyloxy ffa (AOffa) from H-FABP to model phospholipid membranes, using a resonance energy transfer assay. In contrast to previous observations of ffa transfer from liver FABP and from membranes, transfer from H-FABP to membranes appears to occur by a different mechanism. AO-palmitate (16:0) transfer was 1.5-fold slower than AO-stearate (18:0) transfer, and mono-unsaturation did not affect the transfer rate. The AOffa transfer rate from H-FABP increased with increasing ionic strength and decreased slightly between pH 7 and 9. These results suggest that the rate of ffa transfer from H-FABP to membranes is independent of the ffa aqueous solubility. Thermodynamic analysis showed that the free energy of activation for the ffa transfer process arises primarily from an enthalpic component, with only a small entropic contribution, again suggesting the lack of an aqueous phase route of ffa delivery. Finally, the ffa transfer rate was found to be directly dependent on the concentration of acceptor membranes. These data therefore suggest that transfer of AOffa from H-FABP to membranes may occur via collisional interactions between the protein and membranes.     

[3H]Citalopram Binding to Brain and Platelet Membranes of Human and Rat

Citalopram, a selective serotonin (5-HT) uptake inhibitor with antidepressant properties, was found to bind with high affinity to the 5-HT transporter from human neuronal and platelet membranes. At 20 degrees C, KD was about 1.5 nM in both tissues. [3H]Citalopram bound to rat neuronal membranes with higher affinity than to human neuronal and platelet membranes; at 20 degrees C KD was about 0.7 nM. The Bmax value for the binding of [3H]citalopram to platelet membranes was close to that found using the 5-HT uptake inhibitors [3H]imipramine and [3H]paroxetine, suggesting that all three 5-HT uptake inhibitors bind to the 5-HT transporter. The dissociation rate of [3H]citalopram increased twofold with each 4-5 degree C increase in temperature in both human and rat membranes, although at any given temperature, the dissociation rate was about four times faster in the human neuronal and platelet membranes than in rat neuronal membranes.     

Effects of alpha-tocopherol on the lipid peroxidation and fluidity of porcine intestinal brush-border membranes

The effect of alpha-tocopherol on the lipid fluidity of porcine intestinal brush-border membranes was studied using pyrene as a fluorescent probe. Addition of alpha-tocopherol to the medium decreased fluorescence intensity and lifetime, but increased the fluorescence polarization of pyrene-labeled membranes. beta-, gamma-, and delta-Tocopherols gave no appreciable effect on the fluorescence intensity and polarization of the complex. The apparent dissociation constant (3.1 +/- 0.12 microM) of the interaction of alpha-tocopherol with the membranes, estimated from the change in the fluorescence intensity with varying concentrations of alpha-tocopherol, was in good agreement with the concentration required to cause the half-maximal inhibition of lipid peroxidation of the membranes performed by incubation with 100 microM ascorbic acid and 10 microM Fe2+. Decrease of the slope in the thermal Perrin plot of the polarization of pyrene-labeled membranes by alpha-tocopherol suggests that the movement of pyrene molecules in the membranes is restricted by binding of the tocopherol. This interpretation was confirmed by an increased harmonic mean of the rotational relaxation time of the dye molecules in the membranes from 10.9 +/- 0.16 to 18.5 +/- 0.51 microseconds after addition of 25 microM alpha-tocopherol to the medium. The perturbation of lipid phase in the membranes induced by alpha-tocopherol was also suggested from a decreased quenching rate constant of pyrene fluorescence in the membranes for Tl+. Based on these results, the effect of alpha-tocopherol on the lipid fluidity of the membranes is discussed.     

A new concept in polymeric thin-film composite nanofiltration membranes with antibacterial properties

A new, thin film, biofouling resistant, nanofiltration (NF) membrane was fabricated with two key characteristics, viz. a low rate of silver (Ag) release and long-lasting antibacterial properties. In the new approach, nanoparticles were embedded completely in a polymeric thin-film layer. A comparison was made between the new thin-film composite (TFC), NF membrane and thin-film nanocomposite (TFN), and antibacterial NF membranes. Both types of NF membrane were fabricated by interfacial polymerization on a polysulphone sublayer using m-phenylenediamine and trimesoyl chloride as an amine monomer and an acid chloride monomer, respectively. Energy dispersive X-ray (EDX) microanalysis demonstrated the presence of Ag nanoparticles. Scanning electron microscopy (SEM) and atomic force microscopy (AFM) were used to study the cross-sectional and surface morphological properties of the NF membranes. Permeability and salt rejection were tested using a dead-end filtration cell. Ag leaching from the membranes was measured using inductively coupled mass spectrometry (ICP–MS). Morphological studies showed that the TFC NF membranes had better thin-film formation (a more compact structure and a smoother surface) than TFN NF membranes. Performance experiments on TFC NF membranes revealed that permeability was good, without sacrificing salt rejection. The antibacterial properties of the fabricated membranes were tested using the disk diffusion method and viable plate counts. The antibiofouling properties of the membranes were examined by measuring the quantity of bacterial cells released from the biofilm formed (as a function of the amount of biofilm present). A more sensitive surface was observed compared to that of a typical antibacterial NF membrane. The Ag leaching rates were low, which will likely result in long-lasting antibacterial and biofouling resistant properties.     

Culture variability associated with the U.S. Environmental Protection Agency Tuberculocidal Activity Test Method.

Tuberculosis continues to be a major world health threat. The etiologic agent is among the vegetative organisms most resistant to chemical disinfection. Tuberculocidal efficacy testing for regulatory approval of chemical germicides has evolved considerably over the past decade. A method currently in use is the Environmental Protection Agency Tuberculocidal Activity Test Method, a suspension test using a Mycobacterium bovis culture grown under specific conditions and stored frozen until used. Differing tuberculocidal label claims on products with similar formulations have raised questions concerning the equivalence of test suspensions prepared by different laboratories. Five M. bovis suspensions from laboratories currently performing this test were compared against a battery of three disinfectants at a single test site. A significant difference between test cultures was found, with two of the five exhibiting a significant difference from the other three and also from each other. There was a significant culture-by-disinfectant interaction, indicating that the five cultures did not respond in a consistent manner across the different disinfectants used. However, these differences were due to cultures that were not prepared in accordance with the standard procedure or otherwise did not meet the test suspension criteria. In addition, a 0.55% sodium hypochlorite solution was found to be a sensitive indicator of culture variability. These data reinforce the need to adhere to published procedures and guidelines when growing and preparing a tuberculocidal test suspension and shed light on the variables associated with this type of testing.     

Quantitative molecular hybridization on nylon membranes

A study of DNA hybridization to DNA covalently bound to nylon membranes was made in order to develop a quantitative method for molecular hybridization using a nylon-based matrix. Chloroplast DNA was covalently attached to nylon membranes by irradiation at 254 nm. Under hybridization conditions the initial rate of DNA loss from the nylon membranes was 5-10% per 24 h, while under comparable conditions DNA bound to nitrocellulose membranes was lost at a rate of 38 to 61% per 24 h. Several sets of hybridization conditions were examined to select one giving reasonable hybridization rates and minimal loss of bound DNA. Under the conditions selected [Denhardt's solution (D. Denhardt, 1966, Biochem. Biophys. Res. Commun. 23, 641-646), 0.5 M NaCl, 0.1% sodium dodecyl sulfate, and 31.4% formamide at 50 degrees C for 92 h], hybridization was observed to be 29% more efficient on nylon membranes than on nitrocellulose. Several attempts to remove previously hybridized DNA from nylon membranes proved only partially successful. Reuse of the membranes, therefore, was of limited value. Quantitative hybridization of total radiolabeled tobacco cellular DNA to cloned tobacco chloroplast DNA attached to nylon yielded results similar to those previously reported using nitrocellulose membranes. However, use of nylon membranes greatly facilitated the manipulations required in the procedure.     

Inhibition of phospholipid hydrolysis by phospholipase A2 in microsomal and mitochondrial membranes subjected to lipid peroxidation

The rate of phospholipid hydrolysis in rat liver microsomal and mitochondrial membranes catalyzed by phospholipase A2 was shown to decrease after ascorbate + Fe2+-induced lipid peroxidation. The degree of inhibition was linearly dependent on the amount of lipid peroxidation products (malonyl dialdehyde) accumulated in the membrane. The decreased phospholipid hydrolysis rate in membranes after lipid peroxidation was registered using phospholipases A2 from two sources: porcine pancreas and bee venom. It was established that the inhibitory action of phospholipid peroxidation products was not linked with a direct effect on the enzyme and was not caused by depletion of phospholipase reaction substrates (as a result of lipid peroxidation). A possible role of lateral separation of oxidized and non-oxidized lipid phases in the mechanisms of inhibition of phospholipid hydrolysis by phospholipase A2 is discussed.     

Effect of methodology, dilution, and exposure time on the tuberculocidal activity of glutaraldehyde-based disinfectants

The Association of Official Analytical Chemists (AOAC) test for assessing the tuberculocidal activity of disinfectants has been shown to be variable. A modified AOAC test, which substituted Middlebrook 7H9 broth as the primary subculture medium and used neutralization by dilution, was compared with the standard AOAC method to assess the mycobactericidal activity of three glutaraldehyde-based disinfectants at 20 degrees C and various exposure times. These changes had a marked effect on results, with the modified AOAC test providing more positive penicylinders per 10 replicates in 12 of the 13 comparisons that provided positive results. These differences were observed with both Mycobacterium bovis (ATCC 35743) and a clinical isolate of Mycobacterium tuberculosis. The effects of various exposure times to and dilutions of the glutaraldehyde-based disinfectants were also examined. The minimum exposure time needed to inactivate reliably M. bovis or M. tuberculosis with 2% glutaraldehyde was 20 min at 20 degrees C. Diluting 2% glutaraldehyde caused a significant decline in mycobactericidal activity. Modification of the standard AOAC test to improve its sensitivity in detecting the failure of disinfectants to inactivate mycobacteria is indicated.