Role of Porins for Uptake of Antibiotics by Mycobacterium smegmatis

The outer membrane of mycobacteria presents an effective permeabilitybarrier for many antibiotics. Transport pathways across thismembrane are unknown for most drugs. Here, we examined whichantibiotics utilize the porin pathway across the outer membraneof the model organism Mycobacterium smegmatis. Deletion of theporins MspA and MspC drastically increased the resistance ofM. smegmatis ML10 to β-lactam antibiotics, while its β-lactamaseactivity remained unchanged. These results are consistent withthe ninefold-reduced outer membrane permeability of the M. smegmatisporin mutants for cephaloridine and strongly indicate that β-lactamantibiotics rely on the porin pathway. The porin mutant ML10accumulated less chloramphenicol and norfloxacin and was lesssusceptible to these antibiotics than wild-type M. smegmatis.These results demonstrated that small and hydrophilic antibioticsuse the Msp porins for entering the cell. In contrast to norfloxacin,the hydrophobic moxifloxacin was 32-fold more effective in inhibitingthe growth of M. smegmatis, presumably because it was able todiffuse through the lipid membrane. Structural models indicatedthat erythromycin, kanamycin, and vancomycin are too large tomove through the MspA channel. This study presents the firstexperimental evidence that hydrophilic fluoroquinolones andchloramphenicol diffuse through porins in mycobacteria. Thus,mutations resulting in less efficient porins or lower porinexpression levels are likely to represent a mechanism for theopportunistic pathogens M. avium, M. chelonae, and M. fortuitum,which have Msp-like porins, to acquire resistance to fluoroquinolones.

INTRODUCTION

Top

INTRODUCTION

Mycobacterium tuberculosis has infected about 2 billion peopleand causes the death of about 2 million people every year, morethan any other bacterial pathogen (20). Chemotherapy of tuberculosis(TB) is effective in patients infected with susceptible M. tuberculosisstrains. However, infections with multidrug-resistant and extensivelydrug-resistant strains of M. tuberculosis are increasing throughoutthe world (51). M. tuberculosis quickly becomes resistant intherapeutic regimens based on a single drug (31, 57) due totarget mutations (56). In addition to these acquired resistancemechanisms, mycobacteria, including M. tuberculosis, are intrinsicallyresistant to many antibiotics and drugs. This has been attributedto the low permeability of the cell envelope (3), which actsin combination with drug-inactivating, target-modifying enzymesand multidrug efflux pumps (10, 16, 24, 34). Considering theimportance of the few drugs and antibiotics available for TBchemotherapy, it is surprising that the mechanism by which thesemolecules enter the cell is still unknown.

For a long time, the architecture of the mycobacterial cellwall has been controversial (11). Recently, we demonstratedby cryoelectron microscopy that mycobacteria contain a secondlipid bilayer in addition to the inner membrane. Long-chainmycolic acids are essential components of this unusual outermembrane (21). Thus, antibiotics first need to overcome thepermeability barrier established by the outer membrane. Thisrequirement explains previous findings that mutants or treatmentsaffecting mycolic acid biosynthesis and the production of extractablelipids show increased permeability and a concomitant increasein susceptibility to hydrophobic drugs (12, 43, 58). In addition,M. tuberculosis mutants with defects in the synthesis of theouter membrane were impaired in virulence, thus underliningthe importance of its protective function for intracellularsurvival (1). Hydrophobic compounds can directly diffuse throughlipid membranes, while the permeability of lipid membranes forhydrophilic drugs is extremely low. Therefore, they often utilizechannel proteins, such as porins, to cross the outer membranesof bacteria. Antibiotics, such as β-lactams, tetracyclines,fluoroquinolones, chloramphenicol, and cycloserine, have beenshown to use porins for access to the periplasm in gram-negativebacteria (37).

To date, only two classes of channel-forming outer membraneproteins are known in mycobacteria: OmpA-like proteins and MspA-likeporins in slow- and fast-growing mycobacteria, respectively.OmpA is a channel protein in the outer membrane of M. tuberculosis(50). The ompA mutant of M. tuberculosis does not show increaseddrug resistance (44), and no other evidence for involvementof OmpA in the uptake of antibiotics is available. MspA is themain porin, constituting more than 70% of all pores of M. smegmatis(52). X-ray analysis of MspA revealed a unique structure withan octameric channel that is more than twice as long as thechannel of the porin OmpF and of other porins of Escherichiacoli (13). The genes mspB, -C, and -D encode porins that arevery similar to MspA (52). Only mspA and mspC are expressedin wild-type (wt) M. smegmatis, while transcription of mspBand mspD is activated upon deletion of mspA by an unknown mechanism(53). Although the uptake rate for glucose was sixfold reducedin the mspA mutant MN01 (52), the growth of M. smegmatis wasnot much affected (52). Lack of MspA also increased the MICsof ampicillin and cephaloridine for M. smegmatis, indicatingthat MspA plays an important role in the uptake of hydrophilicβ-lactam antibiotics (54). In addition, this study suggestedthat the MICs of vancomycin, erythromycin, and rifampin wereincreased for the mspA mutant. This was surprising, becausethese antibiotics were thought to be too large to diffuse throughthe MspA pore (54). We hypothesized that M. smegmatis strainswith significantly fewer porins than the ${Delta}$ mspA mutant MN01 shouldexhibit a stronger resistance phenotype if the increased resistancewas indeed due to impaired uptake. To this end, we utilizedthe porin mutants ML10 ( ${Delta}$ mspA ${Delta}$ mspC) and ML16 ( ${Delta}$ mspA ${Delta}$ mspC ${Delta}$ mspD).The numbers of Msp porins in these mutants are at least 15-and 5-fold lower than in wt M. smegmatis and the ${Delta}$ mspA mutantMN01, respectively (53). This coincides with 75-fold lower permeabilityof ML10 and ML16 to nutrient molecules, such as glucose. Thegrowth of both strains was much slower than that of the wt,demonstrating that porin-mediated influx of nutrients is a majordeterminant of the growth rate of M. smegmatis (53).

In this study we show that the loss of Msp porins does not significantlyalter the susceptibility of M. smegmatis to vancomycin, kanamycin,and erythromycin. Further, uptake experiments provide directevidence that chloramphenicol and fluoroquinolones use Msp porinsto cross the outer membrane of M. smegmatis. These findingsin the model organism M. smegmatis have important implicationsfor an understanding of how antibiotics can cross mycobacterialouter membranes.

MATERIALS AND METHODS

Top

MATERIALS AND METHODS

Bacterial strains, media, and growth conditions. M. smegmatis SMR5 is a derivative of M. smegmatis mc²155 andis streptomycin resistant due to a mutation in ribosomal proteinS12 (48). The construction of M. smegmatis porin mutants MN01( ${Delta}$ mspA), ML10 ( ${Delta}$ mspA ${Delta}$ mspC), and ML16 ( ${Delta}$ mspA ${Delta}$ mspC ${Delta}$ mspD) was describedpreviously (52, 53). All mycobacterial strains were grown inMiddlebrook 7H9 liquid medium (Difco) supplemented with 0.2%glycerol and 0.05% Tween 80 or on Middlebrook 7H10 plates (Difco)containing 0.5% glycerol at 37°C.

Drug susceptibility by agar dilution. Initial drug solutions were prepared in ethanol or distilleddeionized water, or in 0.1 M NaOH for fluoroquinolones, andstored at –80°C. MICs for M. smegmatis wt and porinmutants were determined in triplicate by agar dilution experimentson Middlebrook 7H10 agar as described previously (53). A preculturein Middlebrook 7H9 liquid medium (Difco) was filtered through5-µm filters (Sartorius AG) to remove large cell aggregatesand then grown in the same medium to an optical density at 600nm (OD₆₀₀) of 0.6 to 0.8. After dilution in 7H9 medium, approximately500 CFU was streaked on agar plates. For all drugs, the concentrationthat resulted in the complete inhibition of bacterial growthafter 5 days of incubation at 37°C was defined as the MIC.Colonies of the porin mutants ML10 and ML16 were clearly visibleafter 5 days on Middlebrook 7H10 agar plates.

Drug susceptibility by microplate Alamar blue assay. M. smegmatis wt and porin mutant strains were grown in 4-mlcultures at 37°C until an OD₆₀₀ of 0.5 to 0.9 was reached.The cultures were diluted to an OD₆₀₀ of 0.01 to 0.02 in medium,and 100 µl of these dilutions was added per well of black,clear-bottom 96-well microplates (Greiner Bio-One) (27). Themicroplate Alamar blue assay was performed and analyzed as describedpreviously (7). For all experiments, the fluorometric endpointwas the same within a twofold drug dilution as the visible endpoint,which was determined by a color change from blue to pink uponreduction of Alamar blue by viable bacteria. The slower growthof the porin mutants ML10 and ML16 did not influence the resultsbecause the fluorescence intensities were normalized to thoseof the drug-free control. Aggregation of cells during liquidculture is a known problem for mycobacteria, which is likelyto reduce the accessibility and hence the susceptibility ofthe bacteria to antibiotics. Therefore, all bacterial cultureswere examined by phase-contrast microscopy before susceptibilityexperiments were performed. The vast majority of all cells inthese cultures were single cells, and they did not have aggregateslarger than approximately 10 cells. Cultures with more or largerclumps were discarded.

Uptake of fluoroquinolones by M. smegmatis. Uptake of fluoroquinolones was determined by measuring theirintrinsic fluorescence as described previously (8). In a firststep, the fluorescences of four fluoroquinolones were examined.Norfloxacin, moxifloxacin, sparfloxacin, and ofloxacin weredissolved in 0.02 M NaOH and diluted in 0.1 M glycine hydrochloride(pH 3.0) to give a final concentration of 0.32 µg/ml.Absorbance and fluorescence spectra were recorded at 25°Cusing a photometer (Lambda; Perkin Elmer) and a fluorimeter(FP-6500; Jasco). All fluoroquinolones had an absorbance maximumat 283 nm, but only norfloxacin and moxifloxacin showed detectablefluorescence with an emission maximum at 440 nm when excitedat 283 nm. The fluorescence of norfloxacin was approximately100-fold higher than that of moxifloxacin under those conditions.

For the uptake experiments, M. smegmatis cells were harvestedby centrifugation at 4°C, washed, and resuspended in sodiumphosphate buffer (50 mM; pH 7.0; 0.05% [vol/vol] Tween 80) to4 x 10⁹ cells per ml. Samples were placed in flasks and preincubatedat 37°C for 10 min in a shaking bath for uptake measurementsor placed on ice for adsorption assays. Norfloxacin or moxifloxacinwas added to a final concentration of 100 µg/ml. Triplicatesamples of 0.8 ml were removed at different times and immediatelyput on ice. Following centrifugation at 16,000 x g (6°C;1 min), the cells were washed twice with 0.1 M sodium phosphatebuffer containing 0.1% (vol/vol) Tween 80 (pH 9.0). The cellpellets were lysed with 1.5 ml of 0.1 M glycine hydrochloride(pH 3.0) overnight at room temperature. The samples were centrifugedat 16,000 x g (20°C; 40 min). The fluorescence of the supernatantat 440 nm was measured after excitation at 283 nm. The fluorescenceof control cell lysates that were not treated with fluoroquinoloneswas subtracted from the fluorescence measured for cells incubatedwith fluoroquinolones. The cells were dried in vacuum at 60°Cfor 30 min to determine the dry weight. The results were expressedas µg of fluoroquinolone per mg of cells (dry weight).

Uptake of chloramphenicol by M. smegmatis. The cells were harvested at an OD₆₀₀ of 0.6 to 1.0 by centrifugation(3,000 x g at 4°C for 10 min), washed twice in uptake buffer[50 mM Tris, pH 6.9, 15 mM KCl, 10 mM (NH₄)₂SO₄, 1 mM MgSO₄,0.1% Tween 80] and resuspended in the same buffer. The cultureswere preincubated at 20°C for 5 min before the additionof [¹⁴C]chloramphenicol (specific activity, 59 mCi/mmol). Thefinal concentration of the antibiotic in the sample was 20 µM.At the indicated time intervals, 0.2-ml samples were removedand filtered immediately through a 0.45-µm-pore-size filterusing a Spin-X centrifuge tube filter (Costar). The cells werewashed twice with a killing buffer (1:2 [vol/vol] 0.1 M LiCl/10%formalin). The radioactivity associated with the cells was countedin a liquid scintillation counter (Beckman). The cells weredried in vacuum at 60°C for 30 min to determine the dryweight. The uptake rate was expressed as pmol of antibioticper milligram of cells.

Nitrocefin hydrolysis assay. To determine the total β-lactamase activity in M. smegmatis,the hydrolysis of the cephalosporin antibiotic nitrocefin wasmeasured as previously described (14) with some modifications.M. smegmatis wt and porin mutant strains were grown at 37°Cuntil late logarithmic phase (OD₆₀₀ $≥$ 2.0). To induce β-lactamaseexpression, 100 µg/ml ampicillin was added to the stationary-phasecultures. The cells were further incubated at 37°C for 2h to allow protein expression. To obtain cell-free culture filtrates,500 µl of the culture supernatants was filtered twicethrough 0.2-µm filters (Pall Corporation). To obtain lysates,cells were pelleted and washed twice in ice-cold 1x phosphate-bufferedsaline (PBS) buffer (pH 7.0). The cell pellets were resuspendedin 1/100 volume of 1x PBS containing corresponding amounts ofprotease inhibitor cocktail (Sigma) and DNase I (NEB). The cellswere disrupted by agitation with glass beads (FastRNA Tubes-Blue)in a FastPrep FP120 bead beater apparatus (Bio-101) twice for30 seconds each time at level 6.0 with a 5-min rest on ice betweenagitation steps. Cell debris was removed by centrifugation.The lysates obtained were filtered twice through 0.2-µmfilters. Protein concentrations were determined using a bicinchoninicacid protein assay kit (Pierce). Nitrocefin was added to a finalconcentration of 100 µM in 1x PBS (pH 7.0), and hydrolysiswas monitored as a change in absorbance at 490 nm using a microplatereader (Synergy HT; Bio-Tek). The activities of β-lactamasesfor each strain were determined as A₄₉₀ min^–1 mg totalprotein^–1.

Phase-contrast microscopy. Cells were viewed with a Zeiss Axiovert 200 microscope usingdifferential interference contrast. Images were recorded witha Zeiss AxioCam MRc camera. To allow visualization of the cultureswithout any fixing and staining artifacts, 5 µl of cellswas added to Delta T Dishes (Bioptechs Inc.) and pictures weretaken from unprocessed samples obtained directly from the cultures.

Structural models of MspA with antibiotics. The MspA structure was downloaded from the Protein Data Bank(identifier, 1uun). The two-dimensional structures of ampicillin(compound accession identifier [CID], 441087), chloramphenicol(CID, 5959), erythromycin (CID, 12560), kanamycin (CID, 6032),norfloxacin (CID, 4539), and vancomycin (CID, 441141) were downloadedfrom the PubChemCompound database at NCBI. Chem 3D Pro 8.0 softwarewas used to obtain three-dimensional structures of the antibioticsand to minimize their energy using the built-in MOPAC algorithm.Finally, the antibiotic molecules were visualized using Chimera(41) (http://www.cgl.ucsf.edu/chimera). All molecules were drawnto scale and displayed as surface representations that showedboth surface-accessible and solvent-excluded areas. Antibioticswere colored by atom. Molecular models of antibiotics were manuallyplaced into the constriction zone of MspA.

RESULTS AND DISCUSSION

Top

RESULTS AND DISCUSSION

Role of porins in susceptibility of M. smegmatis to β-lactam antibiotics. The aim of this study was to examine which antibiotics utilizethe porin pathway to enter mycobacteria. Since general porinsof M. tuberculosis are still unknown, we used the model organismM. smegmatis, whose principal architecture of the outer membraneas the main permeability barrier is similar to that of slow-growingmycobacteria (21), to examine the physical properties of uptakepathways for several classes of antibiotics and TB drugs. Thesusceptibilities of two porin mutants, ML10 ( ${Delta}$ mspA ${Delta}$ mspC) andML16 ( ${Delta}$ mspA ${Delta}$ mspC ${Delta}$ mspD), to these agents were tested by Alamarblue assays and in agar dilution experiments (Table 1). Sincethe number of Msp porins in these mutants is at least 15- and5-fold lower than in wt M. smegmatis and the ${Delta}$ mspA mutant MN01,respectively (53), we expected that the mutants should showa more pronounced resistance phenotype if an antibiotic usedthe Msp pores for entry into M. smegmatis.

View this table:
[in this window]
[in a new window]

TABLE 1. Efficacies of antibiotics against M. smegmatis

As a positive control, we determined the susceptibilities ofML10 and ML16 to hydrophilic β-lactam antibiotics, suchas ampicillin and cephaloridine, which are known to diffusethrough porins in gram-negative bacteria (39). The MICs of ampicillinfor the porin mutants ML10 and ML16 were increased by 16- and32-fold to 1,000 µg/ml in Alamar blue assays and in agardilution experiments, respectively (Table 1; see Fig. S1 inthe supplemental material). In comparison, deletion of the majorporins OmpF and OmpC only moderately increased resistance toampicillin to MICs of 8 and 16 µg/ml in E. coli (19, 23).The double and triple porin mutants ML10 and ML16 were alsocompletely resistant to all other β-lactam antibioticstested (Table 1).

Resistance to β-lactam antibiotics is determined by thesynergy of the outer membrane permeability barrier and enzymatichydrolysis by β-lactamases. M. smegmatis encodes severalchromosomal β-lactamases. Alterations in their expressioncan drastically change the resistance of the strain, as wasobserved for TEM β-lactamase-producing E. coli and Salmonellaenterica (6, 47). Therefore, we analyzed the activities of β-lactamasesin the constructed porin mutants. M. smegmatis whole-cell extractsand culture filtrates were prepared as described earlier (30),and the activities of β-lactamases were determined usingthe chromogenic β-lactam nitrocefin. The vast majorityof the β-lactamase activity of wt M. smegmatis was cellassociated and was 20-fold higher than the activity of the culturefiltrate (not shown), consistent with previously published results(30). Importantly, the β-lactamase activities of the porinmutants MN01, ML10, and ML16 did not differ significantly fromthat of wt M. smegmatis (not shown). Moreover, addition of 100mg/ml ampicillin did not cause a significant increase in theβ-lactamase activities of the wt and the porin mutant ML10(see Fig. 3). It was concluded that the increased resistanceof the porin mutants to β-lactam antibiotics is not causedby faster hydrolysis of the drugs but is due to slower porin-mediateddiffusion across the outer membrane. We used the Zimmermann-Rosseletassay (25, 60) to quantify the permeability of the outer membraneof M. smegmatis ML10 to cephaloridine. Hydrolysis of 0.8 mMcephaloridine was ninefold slower than that of wt M. smegmatis(not shown). Since the total β-lactamase activity in thesupernatant and inside M. smegmatis cells did not change inthe porin mutants, we concluded that deletion of the mspA andmspC genes significantly reduced the outer membrane permeabilityof ML10 to an extent similar to that shown previously for the ${Delta}$ mspA mutant, MN01 (52). These results demonstrated a directlink between the porin-mediated outer membrane permeabilityand the susceptibility of M. smegmatis to cephaloridine andprovided an immediate explanation for the increased resistanceof M. smegmatis porin mutants to β-lactam antibiotics (Table1). The observation that expression of only 15 to 30 MspA poresper µm² of cell wall substantially increased the susceptibilityof M. tuberculosis and M. bovis BCG to several β-lactamantibiotics (28) supports the conclusion that porins providethe major uptake pathway for β-lactams in mycobacteria.

View larger version (25K):
[in this window]
[in a new window]

FIG. 3. β-Lactamase activities of wt M. smegmatis and porin deletion mutants. Hydrolysis of nitrocefin by uninduced (black bars) and induced (gray bars) cultures was measured for the SMR5 and ML10 strains; 100 µg/ml ampicillin was added to stationary-phase cultures for induction of β-lactamase expression. The cells were further incubated at 37°C for 2 h to allow protein expression. The cells were pelleted, washed twice, and resuspended in PBS (pH 7.0), followed by disruption by agitation with glass beads, centrifugation, and filtration. Protein concentrations in the supernatants were determined using bicinchoninic acid. Nitrocefin was added to a final concentration of 100 µM, and hydrolysis was monitored as a change in absorbance at 490 nm using a microplate reader. The activities of β-lactamases for each strain were determined as the A₄₉₀ min^–1 mg total protein^–1. The background activity was determined using PBS as a negative control. All assays were performed in triplicate. The error bars represent standard deviations.

Susceptibility of M. smegmatis wt and porin mutants to fluoroquinolones. Fluoroquinolones are very important second-line drugs for treatmentof TB (32). However, it is unknown how fluoroquinolones crossthe outer membranes of mycobacteria. In gram-negative bacteria,porins at least partially contribute to the uptake of fluoroquinolones(2, 4, 5). To assess a potential role of porins in this processin mycobacteria, we first determined the MICs of several fluoroquinolonesfor the wt strain and the porin mutants. M. smegmatis was verysusceptible to all fluoroquinolones tested in this study, withMICs ranging from 0.1 to 3.2 µg/ml (Table 1). The MICsof ciprofloxacin, sparfloxacin, and moxifloxacin for M. smegmatisdiffered by not more than twofold from the MICs obtained earlierby agar dilution experiments (29, 45). The lowest MIC for M.smegmatis was measured for moxifloxacin, the most hydrophobicfluoroquinolone in this study (Table 1). The MIC of norfloxacin,the smallest and most hydrophilic fluoroquinolone, was 32-foldincreased compared to that of moxifloxacin (Table 1; see Fig.S3 in the supplemental material). This is consistent with theprevious observation that the activities of fluoroquinolonesagainst mycobacteria increase with the hydrophobicity of thedrugs (17) and may indicate a preference of the fluoroquinolonesfor direct diffusion through the lipid membranes of mycobacteria.On the other hand, increasing lipophilicity of the substituentat position N-1 did not correlate with increased potency againstmycobacteria (46). We observed a fourfold increase in the resistanceof the porin triple mutant ML16 to sparfloxacin and moxifloxacincompared to wt M. smegmatis (Table 1). To examine whether porinsplay a role in the uptake of fluoroquinolones, we employed anuptake assay based on the intrinsic fluorescence of fluoroquinolones.Therefore, we determined the absorption and fluorescence emissionmaxima of the four fluoroquinolones tested in this study. Onlymoxifloxacin and norfloxacin exhibited detectable fluorescenceat 0.32 µg/ml when excited at the absorption maximum (notshown). Therefore, we chose both fluoroquinolones for the uptakeassay. Figure 1 shows that the intracellular accumulation ofnorfloxacin in wt M. smegmatis is approximately four times higherthan that in the porin mutant ML10, in contrast to the low adsorptionof norfloxacin measured at 0°C (Fig. 1). Uptake of moxifloxacincould not be determined due to the high adsorption of the hydrophobicmoxifloxacin to the cell surface of M. smegmatis (not shown).Taken together, these results show that porins play a role inthe uptake of hydrophilic fluoroquinolones, such as norfloxacin,by M. smegmatis.

View larger version (17K):
[in this window]
[in a new window]

FIG. 1. Effects of Msp porins on uptake of norfloxacin by M. smegmatis. Accumulation of norfloxacin by wt M. smegmatis (wt) and the ${Delta}$ mspA ${Delta}$ mspC mutant (ML10) was measured. The assay was performed at 37°C at a final norfloxacin concentration of 100 µg/ml. Intracellular norfloxacin was released by cell lysis, and the fluorescence of the supernatant was measured (excitation at 283 nm; emission at 440 nm) using a fluorimeter. Adsorption of norfloxacin to the cell surfaces of wt M. smegmatis (open circles) and the ${Delta}$ mspA ${Delta}$ mspC mutant (open triangles) was determined at 0°C. The fluorescence values were expressed as µg of norfloxacin per mg of cells (dry weight). The values are averages of two independent experiments, each done in triplicate, and are shown with their standard deviations.

Susceptibility of M. smegmatis porin mutants to small antibiotics and TB drugs. Deletion of Msp porins increased the resistance of M. smegmatisto chloramphenicol fourfold (Table 1), suggesting that the Mspporins contribute to the passage of chloramphenicol across theouter membrane of M. smegmatis. The uptake of [¹⁴C]chloramphenicolby both the wt and the porin mutant ML10 of M. smegmatis wasassessed to examine the role of Msp porins in this process.Chloramphenicol was rapidly accumulated by wt M. smegmatis andreached a saturating concentration 2 minutes after additionof the antibiotic (Fig. 2). Importantly, accumulation of chloramphenicolwas significantly slower in the ML10 strain, demonstrating thatthe Msp porins significantly contribute to the uptake of chloramphenicolby M. smegmatis. It should be noted that the concentration ofchloramphenicol decreased after 2 minutes, suggesting that anefflux mechanism contributes to the resistance of M. smegmatisto the antibiotic (Fig. 2). A role of porins in the uptake ofchloramphenicol and susceptibility was also demonstrated forE. coli (33) and Pseudomonas aeruginosa (22), suggesting thatsimilar physical principles govern the passage of this antibioticacross the outer membranes of gram-negative bacteria and mycobacteria.

View larger version (16K):
[in this window]
[in a new window]

FIG. 2. Effects of Msp porins on uptake of chloramphenicol by M. smegmatis. Accumulation of [¹⁴C]chloramphenicol by wt M. smegmatis (wt) and the ${Delta}$ mspA ${Delta}$ mspC mutant (ML10) was measured. Cells were incubated at 20°C for 5 min before the addition of 20 µM chloramphenicol. At the indicated times, samples were removed, filtered, and washed twice. The temperature was kept at 20°C throughout the experiment. The radioactivity associated with the cells was determined in a liquid scintillation counter. The uptake rate was expressed as pmol of antibiotic per milligram of cells. Representative results from one of three independent experiments are shown.

Deletion of the porin genes did not change the MICs of tetracycline,ethambutol, and isoniazid for M. smegmatis. The permeabilitymediated by the remaining porin, MspB, which differs from MspAby only two surface-exposed amino acids (13, 52), might stillbe sufficient for these small drugs to inactivate their targetenzymes at similar rates in both M. smegmatis wt and the tripleporin mutant, ML16. This might also be caused by the lack ofanother resistance mechanism, such as an efflux system, whichis often required for significant resistance to be observedin porin mutants (55). Thus, the possibility that Msp porinscontribute to the uptake of these drugs cannot be excluded.

Susceptibilities of M. smegmatis porin mutants to large and hydrophilic antibiotics. In a previous study, it was reported that the MIC of vancomycinfor the ${Delta}$ mspA mutant MN01 was 10-fold increased compared to thatof wt M. smegmatis (54). This was a surprising result, becausevancomycin has molecular dimensions of 2.9 by 2.4 nm (49) andrequires a channel at least 3.5 nm in diameter to cross theouter membrane of E. coli (26). However, the crystal structureof MspA revealed a channel constriction with a diameter of only1 nm (13). Here, we examined the vancomycin susceptibilitiesof the porin mutant ML10, which lacks the two porin genes mspAand mspC that are expressed in wt M. smegmatis, and of the porintriple mutant ML16, which additionally lacks the mspD gene (53).Using the microplate Alamar blue assay and the agar dilutionassay, we observed only twofold-increased resistance of theporin mutants ML10 and ML16 to vancomycin (Table 1). The increasedMIC of the ${Delta}$ mspA mutant, MN01, obtained in the previous studymay have resulted from cell aggregates, which were observedmore frequently for porin mutants (unpublished observations).The liquid cultures of M. smegmatis used in this study werefree of aggregates larger than 5 to 10 cells as observed bymicroscopy (not shown). The MIC of kanamycin as a representativeof aminoglycoside antibiotics was also only twofold increasedfor the porin double mutant ML10 (Table 1). Since uptake assaysare not available for vancomycin and kanamycin, one cannot examinedirectly whether the Msp porins play any role in the transportof these antibiotics across the outer membrane of M. smegmatis.However, considering the large sizes of these antibiotics andthe drastic reduction in the number of porins in the ML10 andML16 mutants, a more significant MIC increase would be expectedif these molecules diffused through the Msp porins. Taken together,these results suggest that Msp porins do not play a significantrole in the uptake of vancomycin and aminoglycosides by M. smegmatis.This is similar to findings for E. coli porin mutants, whosesusceptibilities to aminoglycosides and vancomycin were notaltered (18). Since aminoglycosides and vancomycin do not appearto be affected by an impaired porin pathway and are too hydrophilicto diffuse through lipid membranes, other entry pathways, suchas self-promoted uptake, as described for E. coli (18), needto be considered for mycobacteria as well.

Susceptibilities of M. smegmatis porin mutants to large and hydrophobic antibiotics. The susceptibilities of E. coli and other gram-negative bacteriato hydrophobic antibiotics were not affected by deletion ofporins (37). This was explained by the decreasing penetrationrate through porins with increasing hydrophobicity of the solutes.The same physical principle also applies to mycobacterial porins.However, previous agar dilution experiments showed a two- tothreefold-increased resistance to erythromycin and rifampinof the porin mutant MN01 compared to that of wt M. smegmatis(54). Using the Alamar blue assay, we reproducibly determinedtwofold-higher MICs of erythromycin and rifampin for all porinmutants compared to those of wt M. smegmatis (Table 1). Computermodeling indicated that erythromycin is too large to diffusethrough the MspA pore (see Fig. 4D). This appears to excludediffusion through Msp pores as the uptake pathway of these antibioticsin M. smegmatis. An alternative explanation is that the lackof porins in the outer membrane may indirectly decrease itspermeability for large and hydrophobic solutes. Indeed, theM. smegmatis mspA mutant showed decreased permeability for thehydrophobic chenodeoxycholate (54). In conclusion, the efficaciesof erythromycin and rifampin against M. smegmatis were not stronglyaffected by the absence of porins, consistent with the largesize and the hydrophobicity of these antibiotics, which favordirect diffusion through the outer membrane lipid bilayer.

View larger version (66K):
[in this window]
[in a new window]

FIG. 4. Structural models of antibiotics and the MspA pore. Visualization of ampicillin (A), chloramphenicol (B), norfloxacin (C), erythromycin (D), kanamycin (E), and vancomycin (F) within the MspA pore. Note that aspartic residues at positions 90 and 91 form the constriction zone and are negatively charged (magnified and shown in red). All antibiotics and MspA are drawn to scale. The images were made using the Chimera software.

It should be noted that the previously determined MICs of erythromycinand rifampin for M. smegmatis were much higher than those inthis study (54). We think that this was due to the formationof clumps and a concomitant decrease in accessibility of thecells in the previous study. However, this did not affect wtand porin mutants differently (54).

Comparison of the agar dilution method with the Alamar blue assay. For most antibiotics tested in this study, the agar dilutionmethod and the Alamar blue assay yielded the same or similarMICs for the different M. smegmatis strains (not shown). Thisis consistent with previous results obtained for many antibioticsfor M. tuberculosis (7, 15). However, we noticed that the MICsof β-lactam antibiotics determined by the microplate Alamarblue assay were always higher than those determined by the agardilution method (Table 1 shows ampicillin; data not shown forother β-lactam antibiotics). This is probably caused bythe secretion of β-lactamases by M. smegmatis (Fig. 3),which decreases the effective concentrations of other β-lactamsin the liquid medium in the Alamar blue assay more rapidly thanin agar plates, resulting in higher MICs. Interestingly, higherMICs for β-lactam antibiotics in the Alamar blue assaywere not observed for M. tuberculosis (7) and for M. bovis BCG(not shown). This may be due to lower β-lactamase activitiesof the slow-growing mycobacteria (14). We concluded that MICsfor β-lactam antibiotics obtained by the Alamar blue assayshould be interpreted with caution.

Can structural models be used to predict utilization of the porin pathway by antibiotics? The size, hydrophobicity, and net charge are critical parametersdetermining the diffusion rate of a solute through a water-filledchannel (37, 38). However, the primary function of porins isto convert the outer membrane into a molecular sieve. Thus,solutes too large to fit through the pores are excluded frompassage. We therefore, set out to determine which of the antibioticstested in this study would fit through the MspA channel. Tothis end, the antibiotic molecules were drawn to scale and displayedas surface representations that showed both surface-accessibleand solvent-excluded areas (Fig. 4). The radius of the MspAconstriction zone calculated by Chimera is about 1.6 nm. Thesecond-largest dimensions of ampicillin, chloramphenicol, andnorfloxacin when oriented along their longest axes are 0.5,0.44, and 0.8 nm, respectively. These models clearly show thatthese antibiotics are significantly smaller than the constrictionzone of MspA (Fig. 4A, B, and C) and corroborate experimentaldata that show that porins are used for uptake of these compounds(52) (Fig. 1). By contrast, the second-largest dimensions oferythromycin, kanamycin, and vancomycin when oriented alongtheir longest axes are about 1.0 nm. It can be easily seen thatthese antibiotics either have the same dimensions as or aretoo large for the constriction zone of MspA. It has been shownby Nikaido and Rosenberg and Cowan et al. that the diffusionrate through porins is reduced drastically with increasing solutesize and was close to zero for lactose, which has a hydratedradius only half of the size of the constriction zone of themain porin, OmpF, of E. coli (9, 40). Therefore, we concludedthat erythromycin, kanamycin, and vancomycin are too large fordiffusion through MspA. This is consistent with experimentsin which wt and porin mutants of M. smegmatis showed similarsusceptibilities to these antibiotics.

Conclusions. This is the first experimental evidence that hydrophilic fluoroquinolonesand chloramphenicol diffuse through porins in mycobacteria.The finding that the hydrophilic norfloxacin uses mainly theMsp porins, while the hydrophobic moxifloxacin was the moreeffective antibiotic, indicates that direct diffusion throughthe outer membrane of M. smegmatis might be faster than diffusionthrough porins for the more hydrophobic fluoroquinolones. Thus,mutations resulting in less efficient porins or lower porinexpression levels likely represent a mechanism for the opportunisticpathogens M. avium, M. chelonae, and M. fortuitum (59), whichhave Msp-like porins (36), to acquire resistance to fluoroquinolones.M. tuberculosis lacks Msp-like porins (35) but is neverthelesssusceptible to both moxifloxacin and norfloxacin, which haveMICs of 0.5 and 2 µg/ml, respectively (42). This may indicatethat a porin-like protein of M. tuberculosis enables diffusionof norfloxacin across the outer membrane or that the slow growthof M. tuberculosis makes it susceptible to compounds with lowouter membrane permeability. Obviously, more work is neededto distinguish between these possibilities and to understandthe uptake of fluoroquinolones and of other TB drugs acrossthe outer membrane of M. tuberculosis.

ACKNOWLEDGMENTS

We thank Joachim Stephan and Jennifer Bender for initial experimentsin this study and K. C. Walls for critically reading the manuscript.

This work was funded by grant AI63432 from the National Institutesof Health to M.N.

FOOTNOTES

* Corresponding author. Mailing address: Department of Microbiology, University of Alabama at Birmingham, 609 Bevill Biomedical Research Building, 845 19th Street South, Birmingham, AL 35294. Phone: (205) 996-2711. Fax: (205) 934-9256. E-mail: mnieder{at}uab.edu

Published ahead of print on 16 June 2008.

Supplemental material for this article may be found at http://aac.asm.org/.

REFERENCES

Top