Plasmid-determined resistance to antimicrobial drugs and toxic metal ions in bacteria

Cloning of pMOL28-encoded nickel resistance genes and expression of the genes in Alcaligenes eutrophus and Pseudomonas spp

The 163-kilobase-pair (kb) plasmid pMOL28, which determines inducible resistance to nickel, cobalt, chromate, and mercury salts in its native host Alcaligenes eutrophus CH34, was transferred to a derivative of A. eutrophus H16 and subjected to cloning procedures. After Tn5 transposon mutagenesis, restriction endonuclease analysis, and DNA-DNA hybridization, two DNA fragments, a 9.5-kb KpnI fragment and a 13.5-kb HindIII fragment (HKI), were isolated. HKI contained EK1, the KpnI fragment, as a subfragment flanked on both sides by short regions. Both fragments were ligated into the suicide vector pSUP202, the broad-host-range vector pVK101, and pUC19. Both fragments restored a nickel-sensitive Tn5 mutant to full nickel and cobalt resistance. The hybrid plasmid pVK101::HKI expressed full nickel resistance in all nickel-sensitive derivatives, either pMOL28-deficient or -defective, of the native host CH34. The hybrid plasmid pVK101::HKI also conferred nickel and cobalt resistance to A. eutrophus strains H16 and JMP222, Alcaligenes hydrogenophilus, Pseudomonas putida, and Pseudomonas oleovorans, but to a lower level of resistance. In all transconjugants the metal resistances coded by pVK101::HKI were expressed constitutively rather than inducibly. The hybrid plasmid metal resistance was not expressed in Escherichia coli. DNA sequences responsible for nickel resistance in newly isolated strains showed homology to the cloned pMOL28-encoded nickel and cobalt resistance determinant.

Plasmid expression and maintenance during long-term starvation-survival of bacteria in well water

Strains of enteric bacteria and pseudomonads containing plasmid R388::Tnl721 (Tpr, Tcr) or pRO101 (Hgr, Tcr) were starved for over 250 days in sterile well water to evaluate effects of starvation-survival on plasmid expression and maintenance. Viable populations dropped to between approximately 0.1 and 1% of the initial populations. Escherichia coli(pRO101) and Pseudomonas cepacia(pRO101) lost both viability and plasmid expression at a lower rate than strains containing R388::Tnl721. Three patterns of host-plasmid interaction were detected: (i) no apparent loss of plasmid expression, (ii) loss of plasmid expression on initial recovery with subsequent expression upon resuscitation, and (iii) loss of capability to produce functional plasmid resistance.

Cloning and expression of Thiobacillus ferrooxidans mercury ion resistance genes in Escherichia coli

A search of various domestic isolates of Thiobacillus ferrooxidans revealed that some were fairly resistant to mercury ion. A proportion of mercury-resistant clones were able to volatilize mercury, and their corresponding gene was localized not in the plasmid DNA but in chromosomal DNA. This mercury ion resistance gene was cloned in Escherichia coli. E. coli carrying the recombinant plasmid was able to grow in the presence of more than 40 micrograms of HgCl2 per ml. Deletion analysis of the recombinant plasmid showed that the entire coding sequence of the mercury ion resistance gene was located within a 2.3-kilobase fragment of the chromosomal DNA from strain E-15. At least two polypeptides (molecular mass, 56 and 16 kDa, respectively) were coded by this fragment.

Chancroid and Haemophilus ducreyi

Haemophilus ducreyi is the causative agent of chancroid, one of the genital ulcerative diseases. H. ducreyi is the major cause of genital ulcer disease in Africa and Southeast Asia and is of increasing concern in the United States. Definitive diagnosis of chancroid requires the isolation and identification of H. ducreyi, but isolation of this organism is difficult and the available medium is not optimal for all strains. Fluorescent antibody and serologic tests are of limited value. In general, our knowledge of this organism is rather limited, and indeed, recent studies have questioned the placement of H. ducreyi in the genus Haemophilus. H. ducreyi has relatively few biochemical activities, and epidemiologic studies are limited because there are limited phenotypic markers available for strain typing. Specific virulence factors of H. ducreyi have yet to be identified. Antimicrobial resistance in H. ducreyi is of special concern, as this organism has acquired both gram-negative and gram-positive resistance determinants. In addition, some of these determinants can be mobilized and transferred to other Haemophilus species or to Neisseria gonorrhoeae.

Plasmids in the bacterial assemblage of a dystrophic Lake: Evidence for plasmid-encoded nickel resistance

Sixty-two aerobic bacterial strains isolated from the unproductive dystrophic Lake Skärshultsjön (South Sweden) were screened for plasmids. The lake is considered to be an extreme environment because of its high concentration of persistent but nontoxic humic compounds. One-third of the isolates harbored multiple plasmids usually of similar high molecular weights (>25 Mdal). The plasmid-bearing strains were members of the common aquatic taxaPseudomonas spp.,Acinetobacter sp.,Alcaligenes sp.,Aeromonas/Vibrio group, andEnterobacteriaceae (taxonomy is tentative). The majority of isolates displayed multiple resistance to antibiotics and heavy metals. Some of them were capable of degrading aromatic compounds. Three isolates were chosen for curing experiments. Only strain S-68, anAlcaligenes sp., could be cured of one of its two plasmids. It harbored the two cryptic plasmids pQQ32 and pQQ70 of 32 and ca. 70 Mdal, and the latter was segregated during ethidium bromide treatment. Parental strain S-68 was capable of degrading some of nonchlorinated phenolic compounds and displayed resistance to a broad spectrum of antibiotics and the heavy metals Co(2+), Ni(2+), Zn(2+), Cd(2+), and Hg(2+). Derivative strain S-68-41 lost its resistance to nickel, suggesting segregated plasmid PQQ70 coded for nickel resistance. Transformation experiments to restore nickel resistance in the cured derivative strain were not successful.

Cloning and analysis of genes involved in coenzyme B12 biosynthesis in Pseudomonas denitrificans

Cobalamin synthesis probably requires 20 to 30 different enzymatic steps. Pseudomonas putida and Agrobacterium tumefaciens mutants deficient in cobalamin synthesis (Cob have been isolated. In P. putida, Cob mutants were identified as being unable to use ethanolamine as a source of nitrogen in the absence of added cobalamin (deamination of ethanolamine requires coenzyme B12 as a cofactor). In A. tumefaciens, Cob mutants were simply screened for their reduced cobalamin synthesis. A genomic library of Pseudomonas denitrificans was constructed on a mobilizable wide-host-range vector. Eleven plasmids from this library were able to complement most of these mutants. By complementation and restriction mapping analysis, four genomic loci of P. denitrificans were found to be responsible for complementation of the Cob mutants. By subcloning fragments from the four genomic loci, we identified at least 14 different genes involved in cobalamin synthesis.

Novel aerobic tetracycline resistance gene that chemically modifies tetracycline

A tetracycline resistance gene that was found originally on the Bacteroides plasmid pBF4 confers resistance on Escherichia coli but only when cells are growing aerobically. When E. coli EM24 carrying this aerobic tetracycline resistance (*Tcr) gene is grown in medium containing tetracycline, the resulting spent medium is no longer toxic to tetracycline-sensitive (Tcs) E. coli EM24 (B.S. Speer and A.A. Salyers, J. Bacteriol. 170: 1423-1429, 1988). To determine whether the *Tcr gene product modified tetracycline, we characterized the material resulting from incubation of E. coli (*Tcr) with tetracycline. When [7-3H(N)]tetracycline was added to cultures of E. coli (*Tcr), at least 90% of the label was recovered in the extracellular fluid. Therefore, tetracycline was not being sequestered by the cells. The labeled material behaved similarly to tetracycline with respect to solubility in various organic solvents. However, the UV-visible light spectrum had a single peak at 258 nm, whereas the tetracycline spectrum had a peak at 364 nm. The labeled material also had a faster migration rate than did tetracycline on thin-layer plates in a solvent system of butanol-methanol-10% citric acid (4:1:2, vol/vol/vol) and was separable from tetracycline by reverse-phase high-pressure liquid chromatography, using an acetronitrile-0.1% trifluoroacetic acid solvent system. These results demonstrate that the *Tcr gene product chemically modifies tetracycline. The *Tcr gene is the first example of a chemically modifying tetracycline resistance mechanism.

Plasmid Incidence in Bacteria from Deep Subsurface Sediments

Bacteria were isolated from deep terrestrial subsurface sediments underlying the coastal plain of South Carolina. A total of 163 isolates from deep sediments, surface soil, and return drill muds were examined for plasmid DNA content and resistance to the antibiotics penicillin, ampicillin, carbenicillin, streptomycin, kanamycin, and tetracycline. MICs of Cu 2+ , Cr 3+ , and Hg 2+ for each isolate were also determined. The overall frequency of plasmid occurrence in the subsurface bacteria was 33%. Resistance was most frequent to penicillin (70% of all isolates), ampicillin (49%), and carbenicillin (32%) and was concluded to be related to the concentrations of the individual antibiotics in the disks used for assaying resistance and to the production of low levels of β-lactamase. The frequencies of resistance to penicillin and ampicillin were significantly greater for isolates bearing plasmids than for plasmidless isolates; however, resistance was not transferable to penicillin-sensitive Escherichia coli . Hybridization of subsurface bacterial plasmids and chromosomal DNA with a whole-TOL-plasmid (pWWO) probe revealed some homology of subsurface bacterial plasmid and chromosomal DNAs, indicating a potential for those bacteria to harbor catabolic genes on plasmids or chromosomes. The incidences of antibiotic resistance and MICs of metals for subsurface bacteria were significantly different from those for drill mud bacteria, ruling out the possibility that bacteria from sediments were derived from drill muds.

Synergistic effect of dosage and bacterial inoculum in TEM-1 mediated antibiotic resistance

The effect of inoculum size and gene dosage on the level of antibiotic resistance mediated by TEM-1 beta-lactamase was measured. From the results it seemed that gene dosage is a more efficient mechanism than inoculum size for increasing TEM-1 mediated resistance to beta-lactam antibiotics. It also seemed that the two mechanisms for enhancing antibiotic resistance are synergistic. The clinical implications of these results are discussed.

Plasmid Frequency Fluctuations in Bacterial Populations from Chemically Stressed Soil Communities

The frequency of plasmids in chemically stressed bacterial populations was investigated by individually adding various concentration of kanamycin, ampicillin, and mercuric chloride to soil samples. Viable bacterial populations were enumerated, soil respiration was monitored for up to 6 weeks as an indicator of physiological stress, and bacterial isolates from stressed and control soils were screened for the presence of plasmids. Low levels of the chemical stress factors did not for the most part significantly alter population viability, soil respiration, or plasmid frequency. Exposure to high stress levels of mercury and ampicillin, however, resulted in altered numbers of viable organisms, soil respiration, and plasmid frequency. Plasmid frequency increased in response to ampicillin exposure but was not significantly changed after exposure to kanamycin. In mercuric chloride-stressed soils, there was a decrease in plasmid frequency despite an increase in overall mercury resistance of the isolates, suggesting that mercury resistance in these populations is largely, if not completely, chromosome encoded. Chemical stress did not cause an increase in plasmid-mediated multiple resistance. A genetic response (change in plasmid frequency) was not found unless a physiological (phenotypic) response (change in viable cells and respiratory activity) was also observed. The results indicate that a change in plasmid frequency is dependent on both the amount and type of chemical stress.

Isolation of Plasmid-Harboring Serratia plymuthica from Facultative Gut Microflora of the Tobacco Hornworm, Manduca sexta

Aseptic isolation of the facultative gut microflora of the tobacco hornworm, Manduca sexta , yielded four microorganisms. Two were gram-positive Bacillus spp., one was Serratia plymuthica , and another was the yeast Candida guilliermondii . The three bacterial species were screened for extrachromosomal DNA, and S. plymuthica was found to have a 6.4-kilobase plasmid, which was designated pCP-1.

Purification and biochemical characterization of tellurite-reducing activities from Thermus thermophilus HB8

Cell-free extracts of Thermus thermophilus HB8 catalyze the in vitro, NADH-dependent reduction of potassium tellurite (K2TeO3). Three different protein fractions with tellurite-reducing activities were identified. Two exhibited high molecular weight and were composed of at least two different polypeptides. The protein in the third fraction was purified to homogeneity and had a single polypeptide chain of 53 to 54 kilodaltons, with an isoelectric point of 8.1. Each enzyme was thermostable, the temperature optimum was 75 degrees C, and 30 mM NaCl, 1.5 M urea, or 0.004% sodium dodecyl sulfate caused 50% inhibition of the enzymes. However, 2% Triton X-100 did not have an inhibitory effect. The enzymes were also able to catalyze the reduction of sodium selenite and sodium sulfite in vitro. NADH was replaceable by NADPH. Divalent cations, such as Ca2+ and Ba2+, had no effect on the activity, while similar concentrations of Zn2+, Ni2+, and Cu2+ abolished the activity. This reductase activity could enable these bacteria both to reduce K2TeO3 and to increase their tolerance toward this salt.

Characterization of a novel tetracycline resistance that functions only in aerobically grown Escherichia coli

A tetracycline resistance (Tcr) gene that was found originally on two Bacteroides plasmids (pBF4 and pCP1) confers tetracycline resistance on Escherichia coli, but only when it is grown aerobically. Using maxicells, we have identified a 44-kilodalton protein which is encoded by the region that carries the Tcr gene and which may be the Tcr gene product. Localization experiments indicate that this 44-kilodalton protein is cytoplasmic. To determine whether the tetracycline resistance gene is expressed under anaerobic conditions, we have constructed a protein fusion between the Tcr gene and lacZ. In strains of E. coli carrying the fusion, beta-galactosidase activity was the same when the cells were grown under anaerobic conditions as when the cells were grown under aerobic conditions. This indicates that the tetracycline resistance gene product is made under anaerobic conditions but does not work. The failure of the Tcr protein to function under anaerobic conditions was not due to a requirement for function of the anaerobic electron transport system, because neither nitrate nor fumarate added to anaerobic media restored tetracycline resistance. Inhibition of the aerobic electron transport system with potassium cyanide did not prevent growth on tetracycline of cells containing the Tcr gene. A heme-deficient mutant, E. coli SHSP19, which carries the Tcr gene, was still resistant to tetracycline even when grown in heme-free medium. These results indicate that functioning of the Tcr gene product is not dependent on the aerobic electron transport system. Thus the requirement for aerobic conditions appears to reflect a requirement for oxygen. Spent medium from an E. coli strain carrying the Tcr gene, which was grown in medium containing tetracycline (50 micrograms/ml), did not inhibit growth of a tetracycline-susceptible strain of E. coli. Thus, the Tcr gene product may be detoxifying tetracycline.

Bleomycin resistance conferred by a drug-binding protein

The protein coded by a bleomycin-resistance gene (ble) cloned from producing actinomycetes was purified from a culture of a recombinant E. coli strain and its action on bleomycin was determined by in vitro assays. The protein binds reversibly in a one to one ratio to bleomycin which can no longer cleave DNA. The bleomycin resistance of cells harboring a ble gene could be accounted for by a sequestering effect of the bleomycin-binding protein.

Precise insertion of antibiotic resistance determinants into Tn21-like transposons: nucleotide sequence of the OXA-1 beta-lactamase gene

Several plasmid-encoded beta-lactamases are on multiresistance transposable elements. The OXA-1 beta-lactamase gene is part of Tn2603, which is borne on the R plasmid RGN238. We report here the complete nucleotide sequence of the OXA-1 beta-lactamase gene and flanking sequences. The OXA-1 gene shows a greater than 50% sequence divergence from the OXA-2 gene, yet there is significant functional similarity at the peptide level. Analysis of 5' and 3' flanking sequences shows that Tn2603 differs from its probable precursor, Tn21, by a precise 1004-base-pair insertion, containing the OXA-1 structural gene, at the target sequence AAAGTT, which is located between the Tn21 streptomycin/spectinomycin (aadA) promoter and its structural gene. A 5- for 6-base repeat of the target sequence is found at the end of the insertion. The same precise insertion and repeat of the target sequence are found for the OXA-2 gene from R46. The 5' flanking regions of two other genes, the trimethoprim-resistance gene from R388 and the gentamicin resistance (aadB) gene from pDGO100, are greater than 98% homologous to the 5' flanking sequences of the OXA-1, OXA-2, and aadA genes until they diverge at the target sequence. From the available sequence data a recombinational hot spot is defined at the nucleotide level 5' of the aadA gene of Tn21, and a second potential hot spot is proposed 3' of this gene.

Cloning and expression in Escherichia coli of chromosomal mercury resistance genes from a Bacillus sp

A 7.9-kilobase (kb) chromosomal fragment was cloned from a mercury-resistant Bacillus sp. In Escherichia coli, in the presence of a second plasmid carrying functional transport genes, resistance to HgCl2 and to phenylmercury acetate (PMA) was expressed. Shortening the cloned fragment to 3.8 kb abolished resistance to PMA but not to HgCl2. In Bacillus subtilis, the 3.8-kb fragment produced mercuric reductase constitutively but did not produce resistance to HgCl2 or to PMA.

Plasmid-mediated sulfonamide resistance in Neisseria meningitidis

An 8.5-megadalton plasmid coding for sulfonamide resistance was found in a clinical isolate of Neisseria meningitidis, as demonstrated by plasmid elimination and transformation experiments. The plasmid complemented a mutation which determines the production of a thermosensitive dihydropteroate synthetase in Escherichia coli, thus suggesting that the mechanism of resistance involved a plasmid-encoded dihydropteroate synthetase.

Frequency of antibiotic and heavy metal resistance, pigmentation, and plasmids in bacteria of the marine air-water interface

A field investigation of marine coastal waters revealed that the frequency of pigmented bacteria and the occurrence of bacterial antibiotic resistance were higher at the air-water interface than in the bulk water. The differences in the frequency of pigmented colonies at the surface and in the bulk-water samples could not be explained by the degree of cell surface hydrophobicity or by bacterial adhesion to air-water interfaces. Pigmented strains exhibited a higher degree of multiple drug resistance than did nonpigmented strains. However, the frequency of multiple drug resistance in nonpigmented strains was also substantial. An average of 91% of all strains were resistant to more than one antibiotic, and 21% of the bacteria isolated were resistant to five of the eight antibiotics tested. High numbers of plasmid-carrying strains were found among selected surface isolates, but the presence of detectable plasmids could not be correlated with either pigmentation or multiple drug resistance. Furthermore, selected surface isolates were significantly more resistant to mercury than were bulk-water bacteria. The higher frequency of pigmented, antibiotic-resistant, and mercury resistant strains at the air-water interface than in the bulk water are discussed in terms of various forms of selective pressure and genetic exchange at the surface.

Detection of a second mechanism of resistance to gentamicin in animal strains of Escherichia coli

One mechanism of plasmid-mediated resistance to gentamicin in Escherichia coli strains isolated from animals is due to the synthesis of the aminoglycoside 3-N-acetyltransferase type IV. A second mechanism of plasmid-mediated resistance to gentamicin was detected in animal strains of E. coli in France and is due to the production of the aminoglycoside 3-N-acetyltransferase type II. The molecular relationships among plasmids encoding this enzyme were studied.

Nucleotide sequence and expression of the mercurial-resistance operon from Staphylococcus aureus plasmid pI258

The mercurial-resistance determinant from Staphylococcus aureus plasmid pI258 is located on a 6.4-kilobase-pair Bgl II fragment. The determinant was cloned into both Bacillus subtilis and Escherichia coli. Mercury resistance was found only in B. subtilis. The 6404-base-pair DNA sequence of the Bgl II fragment was determined. The mer DNA sequence includes seven open reading frames, two of which have been identified by homology with the merA (mercuric reductase) and merB (organomercurial lyase) genes from the mercurial-resistance determinants of Gram-negative bacteria. Whereas 40% of the amino acid residues overall were identical between the pI258 merA polypeptide product and mercuric reductases from Gram-negative bacteria, the percentage identity in the active-site positions and those thought to be involved in NADPH and FAD contacts was above 90%. The 216 amino acid organomercurial lyase sequence was 39% identical with that from a Serratia plasmid, with higher conservation in the middle of the sequences and lower homologies at the amino and carboxyl termini. The remaining five open reading frames in the pI258 mer sequence have no significant homologies with the genes from previously sequenced Gram-negative mer operons.

Genetic and physical analysis of plasmid genes expressing inducible resistance of tellurite in Escherichia coli

A large (greater than 250 kb) conjugative plasmid, pMER610, specifying resistance to tellurium and mercury was isolated from an Alcaligenes strain and transferred by conjugation to Escherichia coli AB1157. The acquisition of pMER610 by AB1157 increased the resistance to both telurite and tellurate by 100-fold. Expression of tellurite resistance by pMER610 and the cloned Ter determinant was inducible by prior exposure to tellurite at levels sub-toxic to the sensitive AB1157. Physical analysis of the cloned Ter fragment located the resistance determinant to a 3.55 kb region. Insertion of Tn 1000 (gamma delta) into this region produced two classes of sensitive mutations, fully sensitive and intermediate or hyposensitive, which map in adjacent regions and form two complementation groups. Maxicell analysis identified four polypeptides (15.5, 22, 23 and 41 kDa) expressed by the Ter clone. The 23 kDa polypeptide may not be required for resistance since tellurium-sensitive gamma delta insertion mutations were not detected in the 23 kDa coding region.

Location on RP4 of a tellurite resistance determinant not normally expressed in IncP alpha plasmids

The tellurite resistance (Ter) determinant of RP4 is not normally expressed unless variants are selected on medium containing tellurite. The determinant was mapped in the variant plasmid RP4Ter by Tn7 insertion mutagenesis. Based on a 56.4-kilobase (kb) replicon, it covered the region from 56 kb, across the EcoRI site at 0 kb, to 1.5 kb.

Cloning and DNA sequence of the mercuric- and organomercurial-resistance determinants of plasmid pDU1358

The broad-spectrum mercurial-resistance plasmid pDU1358 was analyzed by cloning the resistance determinants and preparing a physical and genetic map of a 45-kilobase (kb) region of the plasmid that contains two separate mercurial-resistance operons that mapped about 20 kb apart. One encoded narrow-spectrum mercurial resistance to Hg2+ and a few organomercurials; the other specified broad-spectrum resistance to phenylmercury and additional organomercurials. Each determinant governed mercurial transport functions. Southern DNA X DNA hybridization experiments using gene-specific probes from the plasmid R100 mer operon indicated close homology with the R100 determinant. The 2153 base pairs of the promoter-distal part of the broad-spectrum Hg2+-resistance operon of pDU1358 were sequenced. This region included the 3'-terminal part of the merA gene, merD, unidentified reading frame URF1, and a part of URF2 homologous to previously sequenced determinants of plasmid R100. Between the merA and merD genes, an open reading frame encoding a 212 amino acid polypeptide was identified as the merB gene that determines the enzyme organomercurial lyase that cleaves the C--Hg bond of phenylmercury.

Amplification of drug resistance genes flanked by inversely repeated IS1 elements: involvement of IS1-promoted DNA rearrangements before amplification

Tn2653 contains one copy of the tet gene and two copies of the cat gene derived from plasmid pBR325 and is flanked by inverted repeats of IS1. Transposed onto the P1-15 prophage, it confers a chloramphenicol resistance phenotype to the Escherichia coli host. Because the prophage is perpetuated as a plasmid at about one copy per host chromosome, the host cell is still tetracycline sensitive even though P1-15 is carrying one copy of the tet gene. We isolated P1-15::Tn2653 mutants conferring a tetracycline resistance phenotype, in which the whole transposon and variable flanking P1-15 DNA segments were amplified. Amplification was most probably preceded by IS1-mediated DNA rearrangements which led to long direct repeats containing Tn2653 sequences and P1-15 DNA. Subsequent recombination events between these direct repeats led to amplification of a segment containing the tetracycline resistance gene in tandem arrays.

Antimicrobial resistance of Staphylococcus aureus: genetic basis

Images

High incidence of selenite-resistant bacteria from a site polluted with selenium

The level of selenium-resistant bacteria in water, algal mats, and sediment from Kesterson reservoir, Calif., a site with known selenium pollution, was compared with that in nearby Volta reservoir, a site with low selenium levels. A high percentage (greater than 50%) of all isolates from the Kesterson samples were resistant to 10 mM selenite. In contrast, only a small percentage of the Volta isolates were resistant to this level of selenite. The identity of some selenite-resistant isolates and MICs of selenite, selenate, arsenate, tellurite, and tellurate were determined.

Metal resistance and accumulation in bacteria

Recent research on the ecology, physiology and genetics of metal resistance and accumulation in bacteria has significantly increased the basic understanding of microbiology in these areas. Research has clearly demonstrated the versatility of bacteria to cope with toxic metal ions. For example, certain strains of bacteria can efficiently efflux toxic ions such as cadmium, that normally exert an inhibitory effect on bacteria. Some bacteria such as Escherichia coli and Staphylococcus sp. can volatilize mercury via enzymatic transformations. It is also noteworthy that many of these resistance mechanisms are encoded on plasmids or transposons. By expanding the knowledge on metal-resistance and accumulation mechanisms in bacteria, it may be possible to utilize certain strains to recover precious metals such as gold and silver, or alternatively remove toxic metal ions from environments or products where their presence is undesirable.

The genetics and biochemistry of mercury resistance

The ability of bacteria to detoxify mercurial compounds by reduction and volatilization is conferred by mer genes, which are usually plasmid located. The narrow spectrum (Hg2+ detoxifying) Tn501 and R100 determinants have been subjected to molecular genetic and DNA sequence analysis. Biochemical studies on the flavoprotein mercuric reductase have elucidated the mechanism of reduction of Hg2+ to Hg0. The mer genes have been mapped and sequenced and their protein products studied in minicells. Based on the deduced amino acid sequences, these proteins have been assigned a role in a mechanistic scheme for mercury flux in resistant bacteria. The mer genes are inducible, with regulatory control being exerted at the transcriptional level both positively and negatively. Attention is now focusing on broad-spectrum resistance involving detoxification of organomercurials by an additional enzyme, organomercurial lyase. Lyase genes have recently been cloned and sequencing studies are in progress.

Mechanisms of resistance to cephalosporin antibiotics

Cephalosporins, like other beta-lactams, bind to the bacterial penicillin-binding proteins (PBPs). These correspond to the D-ala-D-ala trans-, carboxy- and endo-peptidases responsible for catalysing the cross-linking of newly formed peptidoglycan. Resistance arises when the PBPs-and particularly the transpeptidases-are modified, or when they are protected by beta-lactamases or 'permeability barriers'. Target-mediated cephalosporin resistance can involve either reduced affinity of an existing PBP component, or the acquisition of a supplementary beta-lactam-insensitive PBP. beta-lactamases are produced widely by bacteria and may be determined by chromosomal or plasmid DNA. The chromosomal beta-lactamases are species-specific, but can be classified into a few broad groups. The plasmid-mediated enzymes cross interspecific and intergeneric boundaries. The level of beta-lactamase-mediated resistance relates to the amount of enzyme produced with or without induction; to the location of the enzyme (extracellular for Gram-positive organisms and periplasmic in Gram-negative ones); and to the kinetics of the enzyme's activity. In Gram-positive organisms the PBPs are located on the outer aspect of the cytoplasmic membrane and so shielding by permeability barriers is minimal. In Gram-negative cells, however, the PBPs are protected by the outer membrane, which most beta-lactams cross by diffusion through aqueous pores composed of 'porin' proteins. In enterobacteria, a clear correlation exists between porin quantity and cephalosporin resistance, suggesting that the outer membrane is the sole barrier to drug entry. Such relationships are less clear for Pseudomonas aeruginosa, where the cell may contain additional barriers between the outer membrane and the PBPs. Although elevated cephalosporin resistance often is attributed to a single factor (PBP-modification, beta-lactamase action or impermeability) an organism's response to a drug often reflects the interplay of several factors. Mathematical models can be proposed to describe this interplay.

Cadmium- and mercury-resistant Bacillus strains from a salt marsh and from Boston Harbor

Bacteria resistant to cadmium or mercury or both were isolated from the Great Sippewissett Marsh (Cape Cod, Mass.) and from Boston Harbor. Many of these metal-resistant isolates were gram-positive aerobic sporeformers, although not necessarily isolated as spores. Although several of the isolated strains bore plasmids, cadmium and mercury resistances appeared to be, for the most part, chromosomally encoded. DNA sequence homology of the gram-positive cadmium- and mercury-resistant isolates was not demonstrable with metal resistance genes from plasmids of either gram-positive (pI258) or gram-negative (pDB7) origin. Cadmium resistance of all the marsh isolates tested resulted from reduced Cd2+ transport. On the other hand, three cadmium-resistant harbor isolates displayed considerable influx but no efflux of Cd2+. Hg-resistant strains detoxified mercury by transforming Hg2+ to volatile Hg0 via mercuric reductase.

Frequency of four classes of tetracycline resistance determinants in Salmonella and Shigella spp. clinical isolates

The frequencies of tetracycline resistance determinants of the classes A, B, C, and D were determined in 53 non-lactose-fermenting clinical isolates. The most frequent class of determinant in Salmonella typhimurium and Shigella flexneri strains was class B; however, the predominant determinant in Shigella sonnei strains was class C.

Coexistence of two competitors on one resource and one inhibitor: a chemostat model based on bacteria and antibiotics

We present a continuous time model of the dynamics of two species competing for a single limiting resource in the presence of a substance that inhibits the growth of one of the species. Resource and inhibitor are both derived from external sources. These inputs, and all other model parameters, are assumed to be constant in space and time. There exist conditions that permit the stable coexistence of the competitors, provided that the sensitive species is more efficient in exploiting the limiting resource, and the resistant species removes the inhibitor from the environment. There exists a subset of these conditions wherein the sensitive species can become established if and only if the resistant species is already established. If the resistant species does not remove the inhibitor from the environment, then coexistence of sensitive and resistant species is structurally unstable. If the resistant species produces the inhibitor, then coexistence is dynamically unstable. We review several studies of bacterial competition in the presence of antibiotics that support these conclusions.

Analysis by using DNA probes of the OXA-1 beta-lactamase gene and its transposon

From recombinant clone pTY27, encoding an OXA-1 beta-lactamase gene, we performed subcloning experiments to more precisely delimit the gene. We describe the use as probes of six different restriction fragments known from subcloning experiments to be within the structural gene or part of the transposable element, Tn2603, flanking the OXA-1 determinant. We showed that the OXA-1 structural gene is slightly related to the OXA-2 determinant and also that sequences within Tn2603 are common to all the OXA- and PSE-producing strains tested. For epidemiological purposes, we began nucleotide sequencing of the OXA-1 determinant, and from preliminary sequence data we synthesized an oligonucleotide 15 bases in length, corresponding to a sequence within the OXA-1 gene. This oligonucleotide was found to be specific for the OXA-1 determinant, because it hybridized only to bacteria producing that beta-lactamase.

Lincomycin stimulates synthesis of TEM-2 beta-lactamase by Escherichia coli

Lincomycin increased the TEM-2 beta-lactamase activity of Escherichia coli K-12 cells carrying plasmid RP4 at a concentration which slightly inhibited cell growth. In a control culture beta-lactamase activity reached its maximal level in late log phase, whereas when lincomycin was present beta-lactamase activity continued to increase into the stationary phase. Lincomycin (100 micrograms/ml) inhibited both cell growth and protein synthesis by about 35% but stimulated beta-lactamase activity 2.5-fold per ml of culture and about 4-fold per cell after 20 h of growth. The amount of beta-lactamase produced in each culture was also compared by densitophotometry of a stained sodium dodecyl sulfate-polyacrylamide gel. The relative values were in good agreement with the relative enzyme activities, indicating that the stimulatory effect of lincomycin was due to an increase in the amount of beta-lactamase protein. Inactivation of beta-lactamase appeared to be faster when lincomycin was present. This was determined by measuring the decrease in beta-lactamase activity when phenethyl alcohol was present to prevent maturation of the enzyme. There was no significant difference in plasmid copy number between the cells grown in the presence or absence of lincomycin. These results indicate that lincomycin stimulates transcription, translation, or translocation of beta-lactamase.

Tellurite susceptibility and non-plasmid-mediated resistance in Escherichia coli

Tellurite (TeO3(2-)) is highly toxic toward Escherichia coli (MIC, approximately 1 microgram ml-1). Mutants (Tel) that were resistant to low levels of TeO3(2-) (MIC, approximately 10 micrograms ml-1) and collaterally resistant to arsenate were isolated. These Tel mutants were unable to grow on media containing low levels of Pi, which supported growth of the parent strain. When grown at much higher Pi levels they exhibited depressed levels of the outer membrane phoE protein and the periplasmic phoS protein, as well as several other proteins indicative of Pi starvation. Tel mutants were markedly defective in 32Pi transport, and TeO3(2-) was shown to be a potent competitive inhibitor of 32Pi transport in the parent strain. The Tel phenotype could be complemented by an F' plasmid harboring the phoR, phoB, and phoA loci, and curing of the F' plasmid completely restored TeO3(2-) resistance. Of a variety of well-characterized Pi transport mutants, only phoB mutants were equally resistant to TeO3(2-), and susceptibility could also be restored in strains carrying an F' plasmid for the phoB region and lost once more after F' curing. The tel and phoB loci were equally cotransducible with lac. Tel mutants still synthesized alkaline phosphatase, the phoA gene product, after Pi starvation, suggesting that the phoB locus per se was not involved because phoB is a positive regulatory gene for phoA expression. The results indicate that TeO3(2-) is transported into E. coli by a phosphate transport system and that resistance to TeO3(2-) specifically selects for as yet uncharacterized mutants in the phoB-phoA region of the chromosome.

Previously undescribed 6.6-kilobase R plasmid in penicillinase-producing Neisseria gonorrhoeae

A penicillin-resistant Neisseria gonorrhoeae strain was isolated. The resistance was due to the production of TEM-1 beta-lactamase encoded by a plasmid. This 6.6-kilobase plasmid was compared with the previously known 7.4- and 5.3-kilobase penicillin R plasmids of N. gonorrhoeae.

Transfer of amikacin resistance by closely related plasmids in members of the family Enterobacteriaceae isolated in Chile

During a 9-month period when amikacin was the sole aminoglycoside used clinically in a hospital in Santiago, Chile, resistance to amikacin and other antibiotics was encountered in 42 strains of the family Enterobacteriaceae, including Escherichia coli, Klebsiella pneumoniae, Citrobacter freundii, Enterobacter cloacae, Serratia marcescens, and Serratia liquefaciens. Amikacin resistance was transferable by conjugation and carried by IncM plasmids ranging in size from ca. 48.4 to 58.1 kilobase pairs. The plasmids had ca. 70 to 80% of their structure in common, as judged after digestion with restriction endonucleases. The resistance was mediated by a 6' aminoglycoside acetyltransferase. We conclude that selective pressure has favored the dissemination of a wide-host-range amikacin resistance plasmid and its derivatives.

High-level amikacin resistance in Escherichia coli due to phosphorylation and impaired aminoglycoside uptake

Plasmid pMP1-1 in Escherichia coli L-0 encodes aminoglycoside (AG) 3'-phosphotransferase II [APH(3')-II]. This enzyme modifies and confers high-level resistance to kanamycin. Although amikacin is a substrate for APH(3')-II, strain L-0(pMP1-1) is susceptible to amikacin. Plasmid pMP1-2 is a spontaneous mutant of pMP1-1 which determines increased APH(3')-II activity for amikacin, apparently as a result of an increase in the copy number of the plasmid. From amikacin-susceptible, gentamicin-susceptible transformants and transconjugants that bear the APH(3')-II gene on plasmid pMP1-1 or pMP1-2 or cloned into multicopy plasmid pBR322, we selected spontaneous mutants at concentrations of amikacin or gentamicin that were two to four times higher than the MICs of these antibiotics. In each case, whether they were selected by using amikacin or gentamicin, the mutants exhibited modest (two- to eightfold) increases in the MIC of gentamicin and major (64- to 128-fold) increases in the MIC of amikacin. Using these laboratory strains of E. coli, we examined the effects on AG susceptibility of the interaction of AG-modifying enzyme activity and generalized AG uptake. Increasing the level of activity of an AG phosphotransferase in these strains lowered their susceptibility to AGs which were substrates for which the enzyme had low Kms. However, an increase in AG-modifying activity alone did not result in large increases in the MICs for poor substrates of the enzyme. In strains which lacked AG-modifying enzymes, a decrease in the rate of AG uptake increased the MICs modestly for a broad spectrum of AGs. When a strain bore the phosphotransferase, a decrease in generalized AG uptake could raise the MIC further, not only for low-Km substrates, but even for AG substrates for which the enzyme had high Kms. Thus, increased modifying activity, together with a diminished rate of uptake, could produce even higher MICs for poor AG substrates.

Indigenous plasmids in Pseudomonas syringae pv. tomato: conjugative transfer and role in copper resistance

Twenty strains of Pseudomonas syringae pv. tomato were examined for the presence of plasmid DNA. P. syringae pv. tomato plasmids were grouped into five size classes: class A ranged from 95 to 103 kilobases (kb); class B ranged from 71 to 83 kb; class C ranged from 59 to 67 kb; class D ranged from 37 to 39 kb; and class E was 29 kb. All strains contained at least two plasmids in classes A and B. The conjugative ability of P. syringae pv. tomato plasmids in three strains was demonstrated by mobilization of the nonconjugative plasmid RSF1010 into Pseudomonas syringae pv. syringae recipients. Plasmids from the three conjugative strains were labeled with Tn5. Four conjugative plasmids were identified by their repeated transfer to P. syringae pv. syringae recipients. P. syringae pv. tomato strains varied in sensitivity to copper sulfate (CuSO4): MICs were 0.4 to 0.6 mM for sensitive strains, 1.2 mM for moderately resistant strains, and 1.6 to 2.0 mM for very resistant strains. One very resistant strain, PT23, functioned as a donor of copper resistance. Recipient P. syringae pv. syringae strains PS51 and PS61 were inhibited by 0.1 mM CuSO4, whereas the CuSO4 MICs for transconjugant strains PS51(pPT23A) and PS61(pPT23C) were 1.8 and 2.6 mM, respectively. P. syringae pv. tomato strains PT12.2 and PT17.2 were inhibited by 0.6 mM copper sulfate, but their copper sulfate MICs were 2.6 and 1.8 mM, respectively, when they acquired pPT23C. Therefore, copper resistance in PT23 was controlled by two conjugative plasmids, designated pPT23A (101 kb) and pPT23C (67 kb).

Sequence and interspecies transfer of an aminoglycoside phosphotransferase gene (APH) of Bacillus circulans. Self-defence mechanism in antibiotic-producing organisms

The APH gene of a butirosin-producing Bacillus circulans was cloned and shown to be expressed in Escherichia coli and Streptomyces lividans. The gene was sequenced and a possible developmentally regulated promoter identified. When the deduced protein sequence was compared with those from transposon Tn5, transposon Tn903, Streptomyces fradiae, Staphylococcus aureus and Streptococcus faecalis, significant homology was found, indicating that the genes may have a common origin.

Plasmid-determined cadmium resistance in Pseudomonas putida GAM-1 isolated from soil

Cadmium-resistant Pseudomonas putida GAM-1, which was able to grow in concentrations of CdCl2 as high as 7 mM, was isolated from soil in a rice paddy. This bacterium harbored a DNA plasmid of about 52 kilobases. The plasmid (pGU100) transformed Escherichia coli C600 to cadmium resistance. A cadmium-resistant transformant of E. coli C600 contained a plasmid corresponding to that seen in P. putida GAM-1. The transformant did not take up cadmium as well as P. putida GAM-1 did.

The nucleotide sequence of the mercuric resistance operons of plasmid R100 and transposon Tn501: further evidence for mer genes which enhance the activity of the mercuric ion detoxification system

The DNA sequences of the mercuric resistance determinants of plasmid R100 and transposon Tn501 distal to the gene (merA) coding for mercuric reductase have been determined. These 1.4 kilobase (kb) regions show 79% identity in their nucleotide sequence, and in both sequences two common potential coding sequences have been identified. In R100, the end of the homologous sequence is disrupted by an 11.2 kb segment of DNA which encodes the sulfonamide and streptomycin resistance determinants of Tn21. This insert contains terminal inverted repeat sequences and is flanked by a 5 base pair (bp) direct repeat. The first of the common potential coding sequences is likely to be that of the merD gene. Induction experiments and mercury volatilization studies demonstrate an enhancing but non-essential role for these merA-distal coding sequences in mercury resistance and volatilization. The potential coding sequences have predicted codon usages similar to those found in other Tn501 and R100 mer genes.