Search

We have shown previously that the 134-residue endonuclease domain of the bacterial cytotoxin colicin E9 (E9 DNase) forms channels in planar lipid bilayers (Mosbahi, K., Lemaitre, C., Keeble, A. H., Mobasheri, H., Morel, B., James, R., Moore, G. R., Lea, E. J., and Kleanthous, C. (2002) Nat. Struct. Biol. 9, 476-484). It was proposed that the E9 DNase mediates its own translocation across the cytoplasmic membrane and that the formation of ion channels is essential to this process. Here we describe changes to the structure and stability of the E9 DNase that accompany interaction of the protein with phospholipid vesicles. Formation of the protein-lipid complex at pH 7.5 resulted in a red-shift of the intrinsic protein fluorescence emission maximum (lambda(max)) from 333 to 346 nm. At pH 4.0, where the E9 DNase lacks tertiary structure but retains secondary structure, DOPG induced a blue-shift in lambda(max), from 354 to 342 nm. Changes in lambda(max) were specific for anionic phospholipid vesicles at both pHs, suggesting electrostatics play a role in this association. The effects of phospholipid were negated by Im9 binding, the high affinity, acidic, exosite inhibitor protein, but not by zinc, which binds at the active site. Fluorescence-quenching experiments further demonstrated that similar protein-phospholipid complexes are formed regardless of whether the E9 DNase is initially in its native conformation. Consistent with these observations, chemical and thermal denaturation data as well as proteolytic susceptibility experiments showed that association with negatively charged phospholipids destabilize the E9 DNase. We suggest that formation of a destabilizing protein-lipid complex pre-empts channel formation by the E9 DNase and constitutes the initial step in its translocation across the Escherichia coli inner membrane.

The effects of protein kinase A (PKA) on the inositol 1,4,5-trisphosphate (InsP 3 ) receptor isoforms type I and type III were studied. The effects of PKA on the extent and rate constants for InsP 3 -induced Ca 2+ release (IICR) were different for the two isoforms. The effects of PKA on the type I isoform showed a biphasic relationship dependent upon the concentration of PKA used. At low concentrations of PKA ( 50 U / ml ), both the extent and rate constants for IICR increased, while at higher concentrations (> 200 U / ml ) the extent and rate constants decreased. The type III isoform showed only an increase in the extent of IICR and not in the rate constants. The effects of PKA on the type I InsP 3 receptor using single channel electrophysiological studies were also investigated. The stimulatory effect of PKA is due to an increase in conductance levels and not to a change in the mean open time of the channel.

The mechanism by which the membrane potential closes and opens voltage-dependent beta-barrel membrane channels is not fully understood. OmpC porins form trimeric water-filled channels when incorporated into artificial bilayers, each monomer having a conductance of approximately 510 pS in 1 M KCl. These channels are relatively insensitive to membrane potential difference (pd) and close only when the pd exceeds +/-250 mV. Another well-known trimer, OmpF, has a monomer conductance of approximately 780 pS in 1 M NaCl, is more sensitive to pd, and can be closed reversibly when a pd of more than +/-150 mV is applied to the channel-containing membranes. With the aid of the 3D atomic structure of these channels determined by X-ray crystallography, and using site-directed mutagenesis, specific amino acids can be substituted in desired locations in the channel lumen. In this study we have used mutants 37C and 74C and attached fluorescence probes to them to monitor polarity changes in the channel lumen during gating. From the observed changes in polarity, we conclude that conformational changes occur in the channel which interrupt the electrolyte conducting pathway.

Ca(2+) within intracellular stores (luminal Ca(2+)) is believed to play a role in regulating Ca(2+) release into the cytosol via the inositol (1,4,5)-trisphosphate (Ins(1,4,5)P(3))-gated Ca(2+) channel (or Ins(1,4,5)P(3) receptor). To investigate this, we incorporated purified Type 1 Ins(1,4,5)P(3) receptor from rat cerebellum into planar lipid bilayers and monitored effects at altered luminal [Ca(2+)] using K(+) as the current carrier. At a high luminal [Ca(2+)] and in the presence of optimal [Ins(1,4,5)P(3)] and cytosolic [Ca(2+)], a short burst of Ins(1,4,5)P(3) receptor channel activity was followed by complete inactivation. Lowering the luminal [Ca(2+)] caused the channel to reactivate indefinitely. At luminal [Ca(2+)], reflecting a partially empty store, channel activity did not inactivate. The addition of cytosolic ATP to a channel inactivated by high luminal [Ca(2+)] caused reactivation. We provide evidence that luminal Ca(2+) is exerting its effects via a direct interaction with the luminal face of the receptor. Activation of the receptor by ATP may act as a device by which cytosolic Ca(2+) overload is prevented when the energy state of the cell is compromised.

An open reading frame in the genomic database of Mycobacterium tuberculosis H37Rv was identified as having homology with an outer membrane protein. We found that the gene specified a protein belonging to the OmpA family, which includes some porins of gram-negative organisms. The gene was amplified by PCR and cloned into Escherichia coli . Overexpression of the gene was toxic to the host, but limited amounts could be purified from cells before growth ceased. A truncated gene devoid of the code for a presumed signal sequence was well expressed, but the protein had no pore-forming activity in the liposome swelling assay. However, the intact protein, OmpATb, behaved as a porin of low specific activity, with a pore diameter of 1.4 to 1.8 nm, and was also active in planar lipid bilayers, showing a single-channel conductance of 700 pS. The protein had a molecular mass of about 38 kDa in sodium dodecyl sulfate-polyacrylamide gel electrophoresis. A polyclonal rabbit antiserum raised to the truncated protein recognized a protein of similar molecular mass in detergent extracts of broken M. tuberculosis cells. Reverse transcription-PCR confirmed that the gene for OmpATb was expressed in M. tuberculosis cells growing in culture. Comparison of the purified protein with that in the detergent-extracted preparation using liposomes and planar lipid bilayers showed that the two materials had similar pore-forming properties. OmpATb is different from either of the mycobacterial porins described so far. This is the first report of a porin-like molecule from M. tuberculosis ; the porin is likely to be important in controlling the access of hydrophilic molecules to the bacterial cell.

Porins, such as Escherichia coli OmpF, provide the only reported example of a voltage-gated channel where the three-dimensional structure is known to high resolution. Mutations that affect voltage-gating are clustered around the eyelet region, which is a mid-channel constriction caused by a polypeptide loop (L3) folding inside the lumen of this β-barrel pore. These data, combined with molecular dynamics simulations, indicate that voltage-gating may involve L3 displacement. We have constructed six double cysteine OmpF mutants, five of which form disulphide bonds fixing L3 in the conformation determined by X-ray crystallography. These channels have altered single-channel conductances but unimpaired voltage-gating. The data show that L3 movement is not required for voltage-gating.

The outer membrane of Gram-negative bacteria contains aqueous channels, porins, which aid the diffusion of small hydrophilic molecules across it. Escherischia coli, as enteric bacteria, are able to survive a hostile environment of proteases, surfactants, and drastic changes of osmotic pressure. Rhodobacter capsulatus is not an enteric bacterium and as such has not evolved to resist the same challenges. Porins, which have molecular weight of approximately 35 kDa, form trimeric channels with a solute exclusion limit of about 600 Da. Most of them open and close in a controlled manner as a function of p.d. This function is little understood at present. The functional properties of single trimers of the major porin of Rhodobacter capsulatus 37b4 have been investigated in planar artificial bilayers. On application of a suitable p.d. the observed trimer closes in approximately three equal steps. The behaviour is completely symmetrical as regards closure in response to p.d.'s of opposite polarity and is strongly cation selective.

Mycobacteria are characterized by an unusual cell wall that controls nutrient and small hydrophilic compound permeability. Porin-like proteins are necessary to ensure the transport of molecules into the cell. Here, we investigated the pore-forming properties of OmpATb, a porin from Mycobacterium tuberculosis, in lipid bilayers. Multi-channel experiments showed an asymmetric behaviour with channel closures at negative critical voltages (Vc) and a strong decrease in Vc at acidic pH. Single-channel experiments gave conductance values of about 850 +/- 80 pS in 1 M KCl and displayed a weak cationic selectivity in 4-8 pH range. The production and characterization of a series of truncated OmpATb proteins, showed that the central domain (OmpATb73-220) was sufficient to induce the ion channel properties of the native protein in lipid bilayers, i.e. asymmetric insertion, pH-dependent voltage closure, cationic selectivity and similar conductance values in 1 M KCl. Western blot analysis suggests that the presence of OmpATb is only restricted to certain pathogenic species. Therefore, the propensity of channels of native OmpATb to close at low pH may represent an intrinsic property allowing pathogenic mycobacteria to adapt and survive to mildly acidic conditions, such as those encountered within the macrophage phagosome.Mycobacteria are characterized by an unusual cell wall that controls nutrient and small hydrophilic compound permeability. Porin-like proteins are necessary to ensure the transport of molecules into the cell. Here, we investigated the pore-forming properties of OmpATb, a porin from Mycobacterium tuberculosis, in lipid bilayers. Multi-channel experiments showed an asymmetric behaviour with channel closures at negative critical voltages (Vc) and a strong decrease in Vc at acidic pH. Single-channel experiments gave conductance values of about 850 +/- 80 pS in 1 M KCl and displayed a weak cationic selectivity in 4-8 pH range. The production and characterization of a series of truncated OmpATb proteins, showed that the central domain (OmpATb73-220) was sufficient to induce the ion channel properties of the native protein in lipid bilayers, i.e. asymmetric insertion, pH-dependent voltage closure, cationic selectivity and similar conductance values in 1 M KCl. Western blot analysis suggests that the presence of OmpATb is only restricted to certain pathogenic species. Therefore, the propensity of channels of native OmpATb to close at low pH may represent an intrinsic property allowing pathogenic mycobacteria to adapt and survive to mildly acidic conditions, such as those encountered within the macrophage phagosome.

The release of Ca(2+) from intracellular stores via InsP(3) receptors shows anomalous kinetics. Successive additions of low concentrations of InsP(3) cause successive rapid transients of Ca(2+) release. These quantal responses have been ascribed to all-or-none release from stores with differing sensitivities to InsP(3) or, alternatively, to a steady-state mechanism where complex kinetic properties of the InsP(3) receptor allow partial emptying of all the stores. We present here an adaptive model of the InsP(3) receptor that can show either pattern, depending on the imposed experimental conditions. The model proposes two interconvertible conformational states of the receptor: one state binds InsP(3) rapidly, but with low affinity, whereas the other state binds slowly, but with high affinity. The model shows repetitive increments of Ca(2+) release in the absence of a Ca(2+) gradient, but more pronounced incremental behaviour when released Ca(2+) builds up at the mouth of the channel. The sensitivity to Ins P (3) is critically dependent on the density of InsP(3) receptors, so that different stores can respond to different concentration ranges of Ins P (3). Since the model generates very high Hill coefficients (h approximately 7), it allows all-or-none release of Ca(2+) from stores of differing receptor density, but questions the validity of the use of h values as a guide to the number of InsP(3) molecules needed to open the channel. The model presents a mechanism for terminating Ca(2+) release in the presence of positive feedback from released Ca(2+), thereby providing an explanation of why elementary Ca(2+) signals ('blips' and 'puffs') do not inevitably turn into regenerative waves.

Bacterial toxins commonly translocate cytotoxic enzymes into cells using channel-forming subunits or domains as conduits. Here we demonstrate that the small cytotoxic endonuclease domain from the bacterial toxin colicin E9 (E9 DNase) shows nonvoltage-gated, channel-forming activity in planar lipid bilayers that is linked to toxin translocation into cells. A disulfide bond engineered into the DNase abolished channel activity and colicin toxicity but left endonuclease activity unaffected; NMR experiments suggest decreased conformational flexibility as the likely reason for these alterations. Concomitant with the reduction of the disulfide bond is the restoration of conformational flexibility, DNase channel activity and colicin toxicity. Our data suggest that endonuclease domains of colicins may mediate their own translocation across the bacterial inner membrane through an intrinsic channel activity that is dependent on structural plasticity in the protein.

Brucella usually carry two highly homologous genes (omp2a and omp2b) for porin-like proteins. In several B. abortus biovars the omp2a gene has a large deletion compared to other Brucella omp2's. In this study we have measured Omp2 pore activity in planar bilayers. Omp2b exhibits well-defined trimeric channel activity whilst Omp2a forms monomeric pores of variable size which are smaller than Omp2b. No sequence homology exists between Omp2 and porins of known structure, so hydrophobic moment analysis has been used to model their membrane topology. From this it appears likely that the deletion removes the crucial L3 internal loop.

The solubilized partially purified Ins(1,4,5)P3-sensitive Ca2+ channel from rat cerebellum has been reconstituted into planar lipid bilayer membranes by the ‘tip-dip’ method [Ehrlich (1992) Methods Enzymol. 207, 463–471] allowing low noise current records. Single-channel events have been recorded. In the presence of 10 µM Ins(1,4,5)P3, 50 µM ATP, and 0.2 µM Ca2+ the Ins(1,4,5)P3 receptor channel opens to a conductance level of 53 pS. In the presence of 100 µM thimerosal (TMS), a sulphydryl-oxidizing agent, three subconductance levels (60 pS, 80 pS and 120 pS) were observed. More than one population of mean open times was found, both in the absence and presence of TMS, although TMS affected the length of the open time by decreasing the short opening significantly from 4.05 ms to 2.78 ms and increasing the longer open time from 27.8 ms to 94.8 ms. The results indicate that TMS enhances Ins(1,4,5)P3-induced Ca2+ release by both altering the open times of the channel significantly and causing a shift to higher subconductance levels.

Single OmpC porin channels have been reconstituted in planar bilayer membranes. Wild-type OmpC forms trimers which are largely insensitive to voltages below 250 mV. A single-point mutation of the ompC gene has been prepared resulting in replacement of Trp56 by Cys in the eyelet region of the channel wall in a highly conserved segment of the polypeptide. The monomeric channels of which the trimer is composed have smaller conductivity in 1 M NaCl (400 ± 20 pS, mean and S.E.M., n = 30) and increased voltage sensitivity by comparison with the wild-type under similar conditions, whereas other (Dex) mutants form larger channels and display different behaviour. Further, by treatment in SDS solutions at different temperatures, the W56C mutant has been shown to be less stable than either the wild-type or the Dex mutants.

Purified OmpF, OmpC, NmpC, PhoE and Lc (Protein 2) porins from the Escherichia coli outer membrane were incorporated into planar phospholipid bilayer membranes and the permeability properties of the pores studied. Triton X-100 solubilised porin samples showed large and reproducible increases in membrane conductivity composed of discreet single-channel events. The magnitude of the cation selectivity found for the porins was in the order OmpC greater than OmpF greater than NmpC = Lc; PhoE was anion selective. For the cation selective porins the cation/anion permeability ratios in a variety of solutes ranged from 6 to 35. Further information on the internal structure of the porins was obtained by examination of the single-channel conductance and this was used to interpret macroscopic observations and to estimate single-channel diameters. The same porins solubilised in SDS exhibited slight conductance increase with no observable single-channel activity. Use of on-line microcomputer techniques confirmed the ohmic current vs. voltage behaviour for all the single porin channels examined.

Detergents are widely used for extracting and purifying membrane proteins. Four such detergents have been studied to find the extent to which they alone can alter black lipid film conductances. The slope of the plot of conductivity versus concentration for Triton X-100 is 4.54 in the range 0.025--0.15 mM; dodecyl sulphate 0.82 in the range 0.01--1 mM; sodium deoxycholate 1.03 in the range 0.01--1 mM and sodium cholate 1.37 in the range 0.1--10 mM. These ranges are below the respective critical micelle concentrations; above these concentrations the membranes break. Bilayer lipid membrane conductivity measured at constant detergent concentration increases with the conductivity of the bathing salt solution with a slope greater than 1, indicating an effect on the putative pore structures induced by detergents.

1. 1. Bilayers have been made from phosphatidylcholine and phosphatidylserine. 2. 2. By means of an osmotic technique, the water permeability coefficient Pos has been measured for different values of pH. 3. 3. The value of Pos for phosphatidylcholine bilayers (1.8·10−5 m·s−1) does not vary with pH within the range of 3.0–7.4. 4. 4. The value of Pos for phosphatidylserine varies from 0.86·10−5 m·s−1 at pH 3.0 to 1.25·10−5 m·s−1 at pH 9.0. 5. 5. The increase in Pos with pH for phosphatidylserine bilayers has been interpreted in terms of an increase of configurational freedom of hydrocarbon chains resulting from an increase in surface area per molecule. 6. 6. The large difference between the phosphatidylcholine value for the water permeability coefficient and the phosphatidylserine values is interpreted in terms of differences in hydrocarbon chain unsaturation.

Misra and Benson [(1988) J. Bacteriol. 170, 3611-3617] showed that point mutations in theompC gene can allowEscherichia coli to grow on maltotriose in the absence of LamB. This report shows that these mutants produce OmpC porins with increased single channel conductance compared to the wild type. The mutants showed similar voltage dependence to each other and to PhoE by being totally closed at 200 mV. The wild type from various sources was largely insensitive to voltages below 200 mV and thus 6 point mutations at 3 sites appear to increase the voltage dependence of OmpC channels.

An acidic lipopeptide A21978C has previously been shown to have a powerful antibiotic activity against Gram-positive organisms. Due to its ability to increase the K+ permeability of bacterial cells and its specific calcium requirement, which is similar to a previously described ionophore CDA, its effect on planar bilayer membranes has been studied. Although it produces significant increases in the conductivity of lipid bilayers it is shown that this alone cannot account for its in vivo activity. Similarly, unlike the in vivo results, the Ca2+-induced increases in bilayer conductivity can be mimicked by Mg2+ and charged lipids. Results from a series of homologues differing in the length of the acyl moiety show a close similarity between bilayer conductance and LD50 trends from in vivo studies. A complex activity is proposed which depends upon incorporation in, rather than disruption of, the bilayer membrane.

The insoluble peptide, T(is), prepared by trypsin hydrolysis of the MN-glycoprotein (glycophorin) of the human erythrocyte has been incorporated into phospolipid membranes in the form of liposomes and black lipid membranes. The permeability of liposome membranes to 42 K + and of black lipid membranes to water and ions is increased significantly by the presence of the T(is) peptide. Electrophoresis measurements indicate that these effects are not due to the T(is) peptide carrying a net charge. The results suggest that the peptide causes local disordering of the bilayer membrane structures. This is considered in the light of findings published elsewhere: that the MN-glycoprotein penetrates through the cell membrane via a non-polar segment of its polypeptide chain, which is contained intact within the T(is) peptide; that the T(is) peptide is partially helical when associated with phospholipid and forms multimeric 8.0 nm structures within the hydrophobic plane of phospholipid bilayers.

Summary: A recently discovered antibiotic (CDA; calcium-dependent antibiotic) of Streptomyces coelicolor A3(2) was found to be effective against a wide range of Gram-positive bacteria only in the presence of calcium ions. Producer and non-producer strains were identified and several media tested for their ability to support antibiotic production. The action of calcium was not simulated by any of the other cations tested. The antibiotic was found to induce discrete conductance fluctuations in planar lipid bilayer consistent with a channel-forming action. The electrical potential difference caused by a concentration difference of various salts across the CDA-containing bilayer, showed the channel to be cation-selective but of a size that discriminated against tetramethyl ammonium and choline ions. The data indicate that the antibiotic activity of CDA is due to its action as a calcium-dependent ionophore.