ABSTRACT
The activity of the membrane-bound H+-ATPase of the beer spoilage bacterium Lactobacillus brevis ABBC45 increased upon adaptation to bacteriostatic hop compounds. The ATPase activity was optimal around pH 5.6 and increased up to fourfold when L. brevis was exposed to 666 μM hop compounds. The extent of activation depended on the concentration of hop compounds and was maximal at the highest concentration tested. The ATPase activity was strongly inhibited by N,N′-dicyclohexylcarbodiimide, a known inhibitor of FoF1-ATPase. Western blots of membrane proteins of L. brevis with antisera raised against the α- and β-subunits of FoF1-ATPase from Enterococcus hirae showed that there was increased expression of the ATPase after hop adaptation. The expression levels, as well as the ATPase activity, decreased to the initial nonadapted levels when the hop-adapted cells were cultured further without hop compounds. These observations strongly indicate that proton pumping by the membrane-bound ATPase contributes considerably to the resistance of L. brevis to hop compounds.
The hop plant, Humulus lupulus L., is used in beer fermentation because of its contribution to the bitter flavor of beer. Furthermore, the use of hops in the brewing industry is preferred because hops have antibacterial activity and prevent beer from bacterial spoilage. Hop compounds are weak acids, which can cross cytoplasmic membranes in undissociated form in response to the transmembrane pH gradient (16). Due to the higher internal pH, these compounds dissociate internally, thereby dissipating the pH gradient across the membrane. As a result of this protonophoric action of hop compounds, the viability of the exposed bacteria decreases (14-16). Some bacteria, however, are able to grow in beer in spite of the presence of hop compounds. Sami et al. (12) reported that Lactobacillus brevis strain ABBC45 could adapt to hop treatment and develop a high level of resistance to hop compounds. During the development of hop resistance the copy number of plasmid pRH45 harboring the horA gene increased (12). Subsequent studies revealed that horA encodes a bacterial ATP-binding cassette (ABC) multidrug resistance transporter (MDR) which can extrude hop compounds from the cell membranes upon ATP hydrolysis (11). As a result of exogenous expression of HorA in Lactococcus lactis, the resistance of this organism to hop compounds increased up to twofold. Microorganisms have been found to increase the proton motive force (PMF)-generating activities in their cytoplasmic membranes when they are confronted with a high influx of protons (20). The thermophilic bacterium Bacillus stearothermophilus (4) increases proton-pumping respiratory chain activities when the proton permeability of its cytoplasmic membrane increases drastically at higher temperatures. In Enterococcus hirae (formerly Streptococcus faecalis) (6, 7) and Saccharomyces cerevisiae (20) the proton-translocating ATPase levels in the membranes were found to increase upon exposure to protonophores such as carbonyl cyanide-m-chlorophenylhydrazone or weak acids. Obviously, the main reason for this increase in proton-pumping activities is to maintain the PMF and the internal pH at viable levels. In view of the protonophoric activities of hop compounds, it was of interest to investigate whether the hop-resistant organism L. brevis would respond in a similar way to the action of hop compounds and whether functional expression of its proton-translocating ATPase in addition to expression of the MDR HorA would increase. In this study, we found that this is indeed the case and that functional expression of the proton-translocating ATPase of L. brevis increases during growth in the presence of hop compounds.
MATERIALS AND METHODS
Bacterial strains and growth conditions. L. brevis ABBC45 was grown anaerobically at 30°C in MRS broth (Merck, Darmstadt, Germany). The initial pH of the growth medium was adjusted to 5.5 with HCl. Hop resistance and expression of HorA were achieved by growing L. brevis in the presence of hop compounds at concentrations up to 666 μM, as described previously (12). Cells grown in the presence of 666 μM hop compounds were subcultured without hop compounds added in order to monitor the ATPase activity under these growth conditions.
Hop compounds.A concentrated isomerized hop extract (Hopsteiner GmbH, Mainburg, Germany) was the hop compound preparation used. The iso-α-acid contents were determined by high-performance liquid chromatography (10). The concentration of hop compounds in the medium was expressed as the concentration of iso-α-acids.
Preparation of the membrane. L. brevis was grown to the late exponential phase in the absence and in the presence of 100 and 666 μM hop compounds. Cells of L. brevis were harvested by centrifugation at 7,000 × g for 15 min and washed twice at room temperature in 50 mM potassium HEPES (pH 7.4) containing 5 mM MgSO4. The cells, suspended in the same buffer, were lysed at 37°C by treatment for 1.5 h with 1 mg of lysozyme (Sigma Chemical Co., St. Louis, Mo.) per ml and 50 μg of mutanolysin (Sigma) per ml in the presence of a cocktail of proteinase inhibitors (Complete; Boehringer, Mannheim, Germany). After addition of DNase I (50 μg/ml) and RNase (1 μg/ml), the suspension was passed three times through an ice-cold French pressure cell at 70 MPa. Unbroken cells were subsequently removed by centrifugation at 7,000 × g for 15 min at room temperature. The supernatant was centrifuged at 200,000 × g for 45 min at 4°C, and the pellet was suspended in the same buffer. This membrane fraction was used for ATPase assays and Western blot analysis. The concentration of the membrane proteins was determined with a DC protein assay kit (Bio-Rad Laboratories, Richmond, Calif.) by using bovine serum albumin as a quantitative standard.
ATPase assay.ATPase activity was estimated from the release of inorganic phosphate as measured by a modification of the method of Driessen et al. (5). One or two micrograms of membrane protein was incubated at 30°C for 10 min in 50 mM potassium MES (morpholineethanesulfonic acid) buffer (usually at pH 5.5) containing 5 mM MgCl2. ATP (potassium salt) was added at a final concentration of 2 mM to initiate the reaction. The reaction (total volume, 40 μl) was stopped after 5 min by immediately cooling the test tubes on ice. A malachite green solution (200 μl of a 0.034% solution) was added, and after 40 min color development was terminated by adding 30 μl of a citric acid solution (34%, wt/vol). The absorbance at 660 nm was measured immediately with a multiscan photometer (Multiskan MS; Labsystems, Vantaa, Finland). One unit of ATPase activity was defined as the amount of enzyme that released 1 μmol of inorganic phosphate in 1 min. Calibration was done by using a series of Pi standards (Sigma). To determine the pH dependence of the ATPase activity, membranes were incubated for 60 min on ice in 50 mM potassium MES buffer adjusted to various pH values. The ATPase activity was assayed at the different pH values as described above. To measure the effects of inhibitors on the ATPase activity, the membranes were preincubated with N,N′-dicyclohexylcarbodiimide (final concentration, 0.2 mM), ortho-vanadate (final concentration, 0.2 mM), or nitrate (K2NO3) (final concentration, 25 mM) for 10 min at 30°C and subsequently for 60 min on ice. A membrane sample without inhibitor was used as the control.
Western blot analysis.The membrane protein of L. brevis, prepared as described above, was solubilized in Laemmli sample buffer containing 2% sodium dodecyl sulfate (9) and was separated by electrophoresis (20 μg of protein/lane) through a sodium dodecyl sulfate-10% polyacrylamide gel by the method of Laemmli (9). The protein bands were transferred to a polyvinylidene difluoride filter membrane and detected with antisera raised against the F1 complex of E. hirae H+-ATPase (3), which can also bind with FoF1-ATPase from L. lactis (1). Membranes from E. hirae prepared as previously described (3) were used as a control. The antibody-bound proteins were visualized with nitroblue tetrazolium and 5-bromo-4-chloro-3-indolylphosphate (Gibco BRL, Gaithersburg, Md.). The intensities of the bands were measured by densitometric analysis with NIH Image software, version 1.61 (National Institutes of Health).
RESULTS
Effect of hops on ATPase activity.Previously, it has been demonstrated that under the conditions described above L. brevis develops hop resistance by overexpressing the MDR HorA (13). The cytoplasmic membranes of cells were isolated as described in Materials and Methods, and the ATPase activities in these membranes were determined as a function of pH at pH values ranging from 4.4 to 7.0. All membranes of L. brevis grown in the presence of different levels of hop compounds showed maximum ATPase activity at around pH 5.6 (Fig. 1). At pH 5.6 membranes from the cells adapted to 666 μM hop compounds had the highest activity, which was about fourfold greater than the ATPase activity of membranes from nonadapted cells. The ATPase activities of membranes from cells adapted to 100 μM hop compounds were between these extremes and were about 1.7-fold greater than the activity of the membranes from the nonadapted cells. Once the cells adapted to hop compounds (666 μM) were subcultured in medium without hop compounds, the ATPase activities of their membranes decreased rapidly (Fig. 1).
pH profile of the ATPase activity in membranes of L. brevis. The ATPase activities at pH values ranging from 4.4 to 7.0 were measured for membranes prepared from cells grown without hop compounds (W0) (○), from cells adapted to 100 μM hop compounds (W100) (×) or 666 μM hop compounds (R666) (▪), and from cells deadapted by growth in the presence of 666 μM hop compounds and then growth for 2 days in the absence of hop compounds (R0) (▴). The ATPase activity is expressed as the amount of inorganic phosphate (Pi) released per minute per milligram of protein.
Effects of inhibitors on the ATPase activity.To characterize the type of ATPase present in the membrane of L. brevis, the effects of several kinds of inhibitors on the ATPase activity were studied (Fig. 2). The ATPase activities of membranes from nonadapted cells and from cells adapted to different concentrations of hop compounds were all significantly inhibited by the FoF1-type inhibitor N,N′-dicyclohexylcarbodiimide. Moderate inhibition was observed with the P-type inhibitor ortho-vanadate, while the V-type inhibitor K2NO3 showed the least inhibition or even activation with membranes from cells grown in the presence of 100 μM hop compounds (Fig. 2, W100). These results correspond to the observations made for the enterococcal FoF1-type ATPase activity, which is slightly inhibited by ortho-vanadate and slightly enhanced by K2NO3 (Y. Kakinuma, personal communication), indicating that FoF1-type ATPase is the major ATPase in the membranes of L. brevis.
Effects of inhibitors on the ATPase activity of L. brevis. The ATPase activities of the membranes of W0, W100, and R666 (see the legend to Fig. 1) were measured at pH 5.6 in the presence of 0.2 mM N,N′-dicyclohexylcarbodiimide (DCCD) (solid bars), 0.2 mM ortho-vanadate (vertically striped bars), or 25 mM nitrate (horizontally striped bars). The activity without any inhibitor was also measured as a control (open bars).
Western blot analysis.Two strong bands were detected with the membranes of L. brevis with the antisera against the α- and β-subunits of the F1-ATPase complex from E. hirae, which strongly indicates the FoF1-type nature of the ATPase of L. brevis. The apparent molecular weights of these bands were slightly higher than those of the α- and β-subunits of F1 from E. hirae (Fig. 3A). The intensities of both bands were higher in membranes isolated from cells grown in the presence of higher concentrations of hop compounds and were lower in membranes from cells adapted to hop compounds (666 μM) and subcultured in medium without hop compounds (Fig. 3B.). The intensities of both bands correlated well (correlation coefficient, 0.990) with the ATPase activities of the different membranes. The rate and extent of growth in MRS broth of hop-adapted cells were less than the rate and extent of growth of nonadapted cells (12). Also, hop-adapted cells were smaller than cells grown in the absence of hop compounds (data not shown).
Western blot analysis of membranes of L. brevis and E. hirae with antisera against F1 of E. hirae. (A) Membranes of L. brevis were solubilized and separated by electrophoresis through a 10% polyacrylamide gel (lanes 2 to 5). For comparison the results obtained with membranes from E. hirae are shown in lane 1. The proteins were transferred to a polyvinylidene difluoride filter membrane and reacted with the antisera raised against the F1 complex of E. hirae H+-ATPase. Lane 1, E. hirae cultured at pH 6.0; lane 2, L. brevis grown without hop compounds (W0); lane 3, L. brevis adapted to 100 μM hop compounds (W100); lane 4, L. brevis adapted to 666 μM hop compounds (R666); lane 5, L. brevis deadapted from 666 to 0 μM hop compounds (R0). The arrows indicate the positions of the α- and β-subunits of H+-ATPase from E. hirae. (B) Intensities of the lower bands of the ATPase from L. brevis. The intensities of the bands were measured with the NIH Image software and are expressed in arbitrary units (a.u.).
DISCUSSION
The beer spoilage bacterium L. brevis ABBC45 develops hop resistance upon growth in hop-containing media (12). This resistance was found to be mediated by the functionally expressed multidrug resistance ABC transporter HorA (11, 13). Studies of HorA, functionally expressed in L. lactis, revealed that HorA can excrete the lipophilic hop compounds and several other MDR substrates from the membrane into the external medium (11). Recently, a second PMF-driven MDR with affinity for hop compounds has been found in L. brevis ABBC45 lacking HorA (19). The activity of HorA and this PMF-driven MDR results in a reduced influx of the undissociated and membrane-permeable iso-α-acids into the cytoplasm and thereby limits the antibacterial PMF-dissipating effect of hop compounds. Since L. brevis develops resistance against rather high concentrations of hop compounds, the question arose whether functional expression of HorA and the PMF-driven MDR was sufficient to confer this resistance or whether additional activities could contribute to hop resistance. Anaerobic gram-positive lactic acid bacteria such as L. brevis depend strongly on their membrane-bound H+-FoF1-ATPase for generation of their PMF (6, 8). In this study, we demonstrated that functional expression of a membrane-bound H+-FoF1-ATPase increased during development of hop resistance and decreased again when exposure to hop compounds was stopped. Previously, it was demonstrated that expression of the HorA transporter increased during development of hop resistance (13). The H+-FoF1 nature of the ATPase was confirmed by H+-FoF1-ATPase effectors and especially by immunological studies with the antisera against the α- and β-subunits of H+-FoF1-ATPase from E. hirae. In accordance with the observations of Kobayashi et al. (6, 7) made with the anaerobic gram-positive bacterium E. hirae, the increased functional expression of H+-FoF1-ATPase most likely allows L. brevis to maintain a viable PMF and intracellular pH in the presence of the protonophoric hop compounds.
The results of this study, together with those of previous studies (11, 13, 19), indicate that L. brevis becomes resistant to hop compounds due to the combined action of two ATP-driven systems, the H+-ATPase and MDR pump HorA (11, 13) and a PMF-driven MDR (19). HorA and the PMF-driven MDR reduce the influx of the weakly acidic hop compounds by pumping undissociated hop compounds from the membrane environment into the external medium. The H+-ATPase compensates for the PMF-dissipating and internal pH-decreasing effects of hop compounds which have escaped the MDR activities by pumping more protons from the cytoplasm across the membrane. As a result of the higher expression of ATPase and HorA and the energy dissipation by hop compounds, the rate and extent of growth in MRS broth of hop-adapted cells are less than the rate and extent of growth in MRS broth of nonadapted cells (12). The various hop resistance mechanisms (Fig. 4) provide another demonstration of the versatility of bacteria and their capacity to develop a variety of mechanisms to cope with toxic compounds in their environments.
Proposed mechanisms of hop resistance in L. brevis ABBC45 due to the combined action of two ATP-driven systems and one PMF-driven MDR. The undissociated hop compounds (Hop-H) intercalate into the cytoplasmic membrane and are pumped out by the multidrug resistance ABC-type transporter HorA (a) (11, 13) and by a secondary MDR (b) (19). A fraction of Hop-H escapes the pumping activity of the transporters and enters the cytoplasm. In the cytoplasm, Hop-H dissociates into the anion (Hop−) and H+ due to the higher internal pH. H+ also enters the cytoplasm in antiport with Hop-H by means of the secondary transporter. Hop− may bind to cations such as Mn2+ (2, 14, 15, 17, 18), while the increased H+-ATPase activity excretes H+ across the membrane (c) (this study).
ACKNOWLEDGMENTS
We thank Asahi Breweries, Ltd., for its support during this study.
FOOTNOTES
- Received 20 February 2002.
- Accepted 2 August 2002.
- Copyright © 2002 American Society for Microbiology
