Comparison of Tetracycline Action on Staphylococcus aureus

Transcription

1 ANTIMICROBIAL AGENTS AND CHEMOTHERAPY, Aug. 1976, p Copyright C 1976 American Society for Microbiology Vol. 10, No. 2 Printed in U.S.A. Comparison of Tetracycline Action on Staphylococcus aureus and Escherichia coli by Microbial Kinetics SAMUEL M. HEMAN-ACKAH Department ofpharmacy, College ofpharmacy and Pharmacal Sciences, Howard University, Washington, D.C Received for publication 7 August 1975 Cultures of tetracycline-treated Staphylococcus aureus exhibited monophasic steady-state growth curves similar to that observed for tetracycline-treated Escherichia coli. Apparent growth rate constants of the respective drug-treated cultures showed the same formal dependence on drug concentration, which was linear at a low concentration but asymptotically approached zero at higher concentration levels and implied the saturation of a limited number of receptor sites engaged in microbial protein synthesis. The relative potency oftetracycline action ofs. aureusle. coli was 6.50:1 at 37.5 C and ph This is attributed to relative differences in drug permeation and/or binding affinity for biophase receptors in the respective organisms. It is concluded from kinetic dependencies ofgrowth inhibition of the cultures that tetracycline has the same mode of action on S. aureus and E. coli. It is bacteriostatic at concentrations below the minimal inhibitory concentration level but bactericidal at the higher concentration levels. Tetracycline is a broad-spectrum antibiotic, bial growth kinetics (16) have revealed relative which is active against gram-positive cocci and differences in the kinetics and dependencies some gram-negative organisms, in addition to related to modes of drug action on gram-positive and gram-negative organisms. This paper rickettsias and clamydia (15). It has been shown to bind to the 30S ribosomal subunit (5, presents the results of similar studies on tetracycline action. It considers the functional de- 18, 22, 23, 29) and to inhibit the formation of peptide linkages in microbial protein synthesis pendency of kinetic parameters, derived from (4, 21, 27, 29). Tetracycline is reported to produce bacteriostatic effects on microorganisms, cline concentrations, the inoculum size, broth growth inhibition of the cultures, on tetracy- even though high concentrations of the drug constituents, and other varying conditions. In have shown bactericidal effects (15). addition, the modes of tetracycline action in the The relative potency of the tetracyclines subinhibitory concentration range and at high against susceptible and resistant strains of organisms (15) has been attributed to differences concentration (MIC) values are considered. concentrations above the minimal inhibitory in the drug permeation through cell membranes, which result in increased ability of sus- MATERIALS AND METHODS ceptible strains to accumulate the drug in the Antibiotics. Assayed samples of tetracycline hydrochloride (U.S.P.), supplied by courtesy of the biophase (6-8, 19, 25, 28). Since tetracycline is more active against Staphylococcus aureus Upjohn Co., were used for the experiments. The concentrations than against Escherichia coli (15), it would be employed in all experiments refer to the hydrochloride salt of the antibiotic. Stock solutions were aseptically prepared by membrane filtra- expected that the drug would have a higher partitioning and/or binding affinity for biophase receptors in the former than in the latter Culture media. Broth was antibiotic medium 3 tion and stored at 40C. organism. (Difco Laboratories) rehydrated to peptone broth Studies on the action of tetracycline against (U.S.P.). The broth was clarified by initial filtration E. coli by the kinetics of microbial growth (2, through 22-,um membrane filters (HA type; Millipore Corp.) and finally through 0.45-,um membrane 10-14, 24) permitted not only the characterization of drug-receptor interactions in the biophase but also an elucidation of the filters (HA type; Millipore Corp.). It was sterilized by autoclaving at 15 lb/in2 for 15 min. The ph of the possible broth was Nutrient agar was broth mode of tetracycline action on the cells. Recently, comparative studies on the action of sterilized by autoclaving at 15 lb/in2 for 15 min. containing 1.5% agar (Difco Laboratories) and was some drugs on S. aureus and E. coli by micro- Bacterial cultures. Replicate cultures of S. au- 223

2 224 HEMAN-ACKAH reus (ATCC 6538) ande. coli (ATCC 12407) on nutrient agar slants were stored at 4 C and used for daily experiments. When required, a 15- to 18-h growth of the respective culture in broth was diluted 50-fold (S. aureus) or 100-fold (E. coli) in 50 ml of fresh broth contained in a 125-ml, loosely capped Erlenmeyer flask. They were grown aerobically at 37.50C in a thermostatically controlled shaker water bath (model R. W. 65, New Brunswick Scientific Co.) to a cell count of 2 x 107/ml. A portion of this culture (1.0 ml of S. aureus cultures per 50 ml of broth or 1.0 ml ofe. coli culture per 100 ml of broth) was added to a bulk amount of broth contained in a 500-ml Pyrex flask fitted with a 50-ml dispenser to form the "seeded" broth. This was incubated at 37.50C for 30 min with intermittent shaking. Quantities (50 ml) of the "seeded" broth were then aseptically transferred with the aid of the dispenser into replicate sterile Erlenmeyer flasks and maintained at 37.5 C in the shaker water bath. A portion (0.5 ml) of the suitably diluted tetracycline solution was added to each culture to yield a desired concentration of the antibiotic. The solution was added when the culture in the flask reached the logarithmic phase of growth at a cell concentration of 2.0 x 106/ml. The culture without antibiotic served as the control. Viable count methods. Samples (1.0 ml) of drugfree and drug-treated cultures were withdrawn at 30-min intervals and diluted serially in sterile 0.9% NaCl and finally in broth to a cell concentration of 50 to 150/ml. Portions (1.0 ml) from the final broth dilution were plated in each of three replicate sandwich agar plates and incubated at 37.50C for 48 h. The bacterial colonies were counted on the colony counter (Lab-line Instruments, Inc.). Total count methods. Samples (0.25 to 1.0 ml) of drug-free and drug-treated cultures were withdrawn at 30-min intervals and diluted in Formol-saline to obtain counts within a range of 2,000 to 70,000 counts/50,ul on a particle-size counter (Coulter counter, model ZB1, Coulter Electronics Co.) that was equipped with a 30-,um orifice. The settings of the instrument were: 1/aperture current, 1.0; 1/amplification, 1/2; gain, 8.0; matching switch, 40 K; lower threshold, 5; and upper threshold, at maximum. The Formol-saline diluent was 0.85% sodium chloride in 1.0% formaldehyde, filtered twice through a 0.2-,um membrane filter (GA-8 type, Metricel), which promptly arrested growth of the culture sample without lytic or clumping effects on the cells. RESULTS Shape of growth curves. The steady-state generation of S. aureus and E. coli cultures affected with graded concentrations of tetracycline demonstrated linear semilogarithmic plots in accordance with the expression: logn = log N0 + kapp t/2.303 (1) where N is the number of organisms per milliliter at anytime (t), No is the number of organisms per milliliter at some initial time (0), ANTIMICROB. AGENTS CHEMOTHER. and kapp is the apparent generation rate constant. In all cases, the tetracycline-treated cultures showed an initial lag phase of approximately 30 min before emerging into a steadystate phase of growth. Figure 1 shows typical monophasic generation curves of tetracyclineaffected S. aureus, determined by the total count method. The curves are similar to those obtained for E. coli and confirm previous observations (2, 10-14, 16, 24). Functional dependency of apparent growth rate constants on drug concentrations. The relationship between the kapp and the tetracycline concentration is plotted in Fig. 2. There was a linear dependency in the low concentration range from 0 to gg/ml for tetracycline action on E. coli (curve A) and from 0 to 0.02,ug/ml for its action on S. aureus (curve B) in accordance with the expression (2, 10-13, 24): kapp = ko-kt T (2) where k, is the specific inhibitory rate constant and T is tetracycline concentration. In the absence of drug, kapp = k),, which is the growth rate constant of drug-free culture. At the a -MNUTES FIG. 1. Typical growth curves of S. aureus in antibiotic medium 3 at ph 7.05 and 37.5 C in the absence and presence of graded concentrations of tetracycline, obtained by the total count method. The curves are labeled according to drug concentrations (micrograms per milliliter).

3 VOL. 10, 1976 TETRACYCLINE ACTION ON S. AUREUS AND E. COLI xc3 4C 29 Q2, 3C3 3 IC3 k \ A u z 4 i ia ik,, bi, m TETRAYCLIE CONCENTRATION IT) XN FIG. 2. Dependence of apparent growth rate constant, kapp, for S. aureus and E. coli on concentrations of tetracycline at ph 7.05 and 37.5 C. The actual concentrations are multiplied by a factor, n, where n = 20 in curve A for E. coli (a) and n = 100 in curve B for S. aureus (U). higher concentration levels, the kapp was not a linear function of increasing tetracycline concentration but asymptotically approached zero. The plots of TI(ko - kapp) versus T (Fig. 3) were reasonably linear for the action of tetracycline concentrations >0.075,ug/ml on E. coli (curve A) and >0.02 ug/ml on S. aureus (curve B), but showed deviations at the lower concentration levels. This implied quantitative adherence of the action of tetracycline on E. coli and S. aureus to an expression previously derived from a saturation kinetic model (13): T/(ko- kapp) = T(ka/k&) + 1/ka (3) where ka and kb are constants of proportionality related to the tetracycline partition through cell membranes and its binding affinity for biophase receptors. At the MIC, when kapp = 0, equation 3 simplifies to: Tm = 1/(ka/ko - kb) (4) where Tm is the MIC for tetracycline. The values of ka and kj, derived from the slopes and intercepts of the plots in Fig. 3, and the Tm, calculated in accordance with equation 4, are given in Table 1. Effect of the counting method on growth curves. Semilogarithmic plots of total and viable counts versus time for the action of selected concentrations of tetracycline, 0, 0.014, 0.06, 0.30, 0.60, and 1.20 gg/ml, respectively, on S. aureus are shown in Fig. 4. Parallel growth curves were obtained from the plots of the total and viable counts on cultures treated with low concentrations (0 to 0.06 ug/ml) of 0 tetracycline (curves A1, A; B, B; and C1, C). The total count was 70 to 80% of the viable count at each concentration level. Cultures treated with high concentrations, 0.30,g/ml or greater, showed coincidence of plots for the total counts that remained constant for the entire period of the drug-bacteria contact time (curves D, E, and F). The corresponding plots for the viable count showed an exponential decrease with time as a function ofthe tetracycline concentration (curves D1, E1, and F,). 7 0.X LC =1 lll TETACYCLIE CWOTRATION to) XN FIG. 3. Applicability of saturation kinetic model to the dependency ofapparent growth rate constants, kapp, for drug-treated cultures at high concentrations of tetracycline. The actual concentrations are multiplied by a factor of n, where n = 20 in curve A for E. coli (O) and n = 100 in curve B for S. aureus (U). TABLE 1. Kinetic parameters deriveda from tetracycline inhibition of S. aureus and E. coli cultures Kinetic parameter S. aureus E. coli 105 kt (ml/,g per sp) ka (mlv,g per s)y 1, kb (ml/,.tgy kalkb (s-')" (kalkb)lkoc MICd Ome a Derived from the data of Fig. 2. b Calculated from the plot of k.pp versus T for the action of ,ug of tetracycline per ml on S. aureus or jig of tetracycline per ml on E. coli in accordance with the expression in equation 2. c Estimated from the slopes and intercepts of the plots of T/(k,, - kapp) versus T for the action of tetracycline on S. aureus at concentrations > 0.02 Ag/ml or for its action on E. coli at concentrations >0.075,ug/ml according to the expression in equation 3. d Calculated from the expression in equation 4. e Calculated from the expression in equation 10.

4 226 HEMAN-ACKAH I W o 2 10MTES FIG. 4. Growth curves ofs. aureus atph 7.05 and 37.50C in the presence and absence of tetracycline, obtained by total count ( ) and viable count (- - -) methods. The curves and respective drug concentrations (micrograms per milliliter) are A, Al, 0.0; B, B,, 0.014; C, C,, 0.06; D, Di, 0.30; E, E,, 0.60; and F, F,, These observations clearly indicated that the action of tetracycline was bactericidal at concentrations above the MIC. Similar results were obtained for the action of tetracycline on E. coli (12). Effect of changes in inoculum size and broth composition on apparent growth rates of tetracycline-treated cultures. The apparent growth rate constants for S. aureus cultures that were treated with 0, 0.01, 0.02, 0.04, 0.06, and 0.08 jig of tetracycline per ml, respectively, did not show significant variations from those obtained at corresponding concentration levels in Fig. 2, when either (i) the inoculum size was changed 0.1- to 10.0-fold at the time of drug addition in a single-strength broth, or (ii) the broth composition was changed 0.5- to 2.0-fold with the organism population fixed at 2.0 x 106/ml at the time of drug addition. DISCUSSION There was no dependence of the apparent growth rate constants of drug-treated S. aureus on variations in the broth composition or the inoculum size at the time of drug addition, suggesting drug interaction with broth constituents or drug depletion/inactivation by microbial metabolic activities as was similarly ANTIMICROB. AGENTS CHEMOTHER. observed (12) for the action of tetracycline on E. coli. The parallelism of growth curves, obtained by the total and viable counts, indicates suitability of either method for determining growth rates of drug-free and drug-treated S. aureus cultures. It also provides evidence that there was no killing or lytic phenomenon superimposed on normal growth inhibition by the action of subinhibitory concentrations oftetracycline on S. aureus. Therefore, tetracycline exerts a "true bacteriostatic" action (3) on S. aureus, causing a decrease in the rate of cell multiplication, presumably by competitive binding on ribosomes to lower cell protein synthesis. At concentrations above the inhibitory level, the total count remains invariant with time, but the viable count decreased logarithmically with time as a function of tetracycline concentration. If the death of the organism was due to effects of prolonged bacteriostasis, then the death rate should show deviation from the monomolecular law (17, 20, 26, 30). The fact that there is an exponential relationship between the v'iable count and drug-bacteria contact time suggests a bactericidal phenomenon. The formal dependence of kapp on drug concentration for tetracycline-treated S. aureus is the same as the tetracycline-treated E. coli (Fig. 2) and suggests the same mechanism of action. The relationship is described by a kinetic expression that is derived from model compartmental analysis (13) on the premise that: (i) the rate of increase in microbial numbers (dn/dt) is proportional to the net rate of protein synthesis (dp'/dt) above a certain minimum rate (Ap) required for survival and the numbers of organisms (N) present, and (ii) the overall rate of protein synthesis (dp/dt) is proportional to the free or unbound fraction (1-0) of total ribosomal sites, RT, where 0 is the fraction of total ribosomal sites bound by tetracycline. Therefore, the following equations are applicable: dn/dt = km(dp'/dt)n = km(dp/dt - Ap)N = kapp N and dp/dt = q (1-0)RT (6) where k, and q are constants of proportionality, and kapp is the growth rate constant of tetracycline-treated cultures. 0 is implicitly defined (9, 13) as follows: (5)

5 VOL. 10, = K1K2T/(1 + K1K2T) (7) where K, is tetracycline partition constant through cell membranes, and K2 is the ribosomal affinity constant for tetracycline (T). A combination of equations 5, 6, and 7 yields: kapp = ko - kat/(1 + kbt) (8) where k0 = q kmrt - km Ap is the growth rate constant of the drug-free culture, kb = KK2, and ka = q kmrtklk2. Therefore TETRACYCLINE ACTION ON S. AUREUS AND E. COLI 227 (ka/kb)/ko > 1 (9) Equation 8 may describe the functional dependency of the kapp on drug concentration for tetracycline-treated S. aureus and E. coli, respectively (Fig. 2). The action of low concentrations (0 to 0.02,ug/ml) of tetracycline on S. aureus or (0 to 0.75,ug/ml) oftetracycline on E. coli shows linear dependency of the kapp on drug concentration, possibly because few receptors interact with the drug and kb T < 1, which therefore simplifies equation 8 to equation 2. At high concentrations (>0.02 jig of tetracycline per ml on S. aureus or >0.075 ug of tetracycline per ml on E. coli), a nonlinear dependency is applicable in accordance with equation 8. Arithmetical transformation of equation 8 is equation 3, for which linear plots of T/(ko - kapp) versus T are obtained in Fig. 3 from the data of Fig. 2 at the high concentration levels. This indicates quantitative adherence to a saturation kinetic model (13). Deviations occur in the low concentration range when the receptor sites are not saturated and the extent of growth inhibition (ko - kapp) is directly proportional to drug concentration. In earlier studies oftetracycline action on E. coli, a linear dependence of the kapp on drug concentration that could be quantified by equation 2 was reported (2, 10, 12, 13). The apparent discrepancy could be attributed to the fact that earlier studies did not employ a wider concentration than was presently investigated. Alternatively, the laboratory strain B/r ofe. coli employed for earlier studies could be a substrain of E. coli (ATCC 12407), which is more susceptible to tetracycline action. Other workers (G. H. Miller, H. L. Smith, W. L. Rock, and S. Hedbert, J. Pharm. Sci., in press), using resistant strains of E. coli, have also reported a nonlinear dependency of the kapp on concentration that can be quantified by equation 8 at high concentration levels for various tetracyclines. The values of ka and kb, calculated from the data of Fig. 3 in accordance with equation 3, are given in Table 1. The experimentally determined values of (ka/kb)/k,, > 1 agree very well with theoretical predictions in equation 9. The calculated ratio of the kb of S. aureusle. coli is 6.35:1. The MIC values of tetracycline against the organisms calculated from ka, kb, and k,, in accordance with equation 4, are also given in Table 1. The relative potency of tetracycline action, as determined by the MIC values, is of the order S. aureusle. coli as 6.50:1. Although kb is a parameter derived from K, and K,, as defined in equation 7, K, and K. cannot be kinetically differentiated. However, the close agreement between the comparative values of kb and the relative potency of tetracycline action on S. aureus and E. coli indicate that the drug partitioning through cell membranes and/ or drug binding affinity for receptor sites are the determining factors for the differences shown in tetracycline activity against susceptible strains of organisms. At the MIC level, the following is applicable: 6,,, = KjK2T,,,/(1 + KIK2T,,,) = (10) kbt,,,/(1 + kbtw,) where 6,,, is a fraction of total ribosomal sites bound by tetracycline. Therefore, 1 - On is the fraction engaged in the minimum rate of protein synthesis for survival. Tm is the MIC, as defined in equation 4. Table 1 shows that O., calculated from kinetic parameters for the action of tetracycline, competitively binds 85% of transfer ribonucleic acid-binding sites to produce true bacteriostasis (12) on the organisms. At concentrations greater than the MIC, there must be a change in the overt mechanism of tetracycline action that results in a change from truly bacteriostatic to a highly bactericidal agent, as observed experimentally (Fig. 4). ACKNOWLEDGMENTS I thank Patricia Ann Wright for excellent technical assistance. LITERATURE CITED 1. Brock, T. D Biology of microorganisms, p , Prentice-Hall, Englewood Cliffs, N.J. 2. Brown, M. R. W., and C. R. Garrett Kinetics and mechanisms of action of antibiotics. T. Reproducibility ofescherichia coli growth curves and dependence upon tetracycline concentrations. J. Pharm. Sci. 53: Cook, A. M Comparative studies of methods of

6 228 HEMAN-ACKAH evaluating antibacterial substance. I. Evaluation of bacteriostatic action. J. Pharm. Pharmacol. 6: Cudliffe, E., and K. McQuillen Bacterial protein synthesis: the effects of antibiotics. J. Mol. Biol. 30: Franklin, T. J Mode of action of tetracycline. Symp. Soc. Gen. Microbiol. 16: Franklin, T. J Resistance of Escherichia coli to tetracyclines-changes in permeability to tetracyclines in Escherichia coli bearing transferable resistance factors. Biochem. J. 105: Franklin, T. J., and A. Godfrey Resistance of Escherichia coli to tetracyclines. Biochem. J. 94: Franklin, T. J., and B. Higginson Active accumulation of tetracycline by Escherichia coli. Biochem. J. 116: Garrett, E. R Drug action and assay by microbial kinetics, p In E. Jucker (ed.), Progress in drug research, vol. 15. Birkhauser Verlag, Basel, Switzerland. 10. Garrett, E. R., and M. R. W. Brown The action of tetracycline and chloramphenicol alone and in admixture on the growth ofescherichia coli. J. Pharm. Pharmacol. 15:185T-191T. 11. Garrett, E. R., and S. M. Heman-Ackah Microbial kinetics and dependence of individual and combined antibiotic inhibitors of protein biosynthesis. Antmicrob. Agents Chemother. 4: Garrett, E. R., and G. H. Miller Kinetics and mechanisms of action of antibiotics on micro-organisms III. Inhibitory action of tetracycline and chloramphenicol on Escherichia coli established by total and viable counts. J. Pharm. Sci. 54: Garrett, E. R., G. H. Miller, and M. R. Brown Kinetics and mechanisms of antibiotics on microorganisms. V. Chloramphenicol and tetracycline affected Escherichia coli generation rates. J. Pharm. Sci. 55: Garrett, E. R., and C. M. Won Effect of novobiocin and its combination with tetracycline, chloramphenicol, erythromycin, and lincomycin on the microbial generation of Escherichia coli. Antimicrob. Agents Chemother. 4: Garrod, L. P., and F. O'Grady Antibiotic and chemotherapy, 3rd ed., p E. & S. Livingstone, London. 16. Heman-Ackah, S. M Microbial kinetics of drug ANTIMICROB. AGENTS CHEMOTHER. action against Gram-positive and Gram-negative organisms. I. Effect of lincomycin on Staphylococcus aureus and Escherichia coli. J. Pharm. Sci. 63: Hinshelwood, C Decline and death of bacterial populations. Nature (London) 167: Igarashi, K., and A. Kaji Relationship between sites 1, 2 and acceptor, donor sites for the binding of aminoacyl RNA to ribosomes. Eur. J. Biochem. 14: Izaki, K., K. Kiuchi, and K. Arima Specificity and mechanism of tetracycline resistance in a multiple drug resistant strain ofescherichia coli. J. Bacteriol. 91: Jordan, R. C., and S. E. Jacobs Studies in the dynamics of disinfection. J. Hyg. 43: Laskin, I Molecular aspects of tetracycline activity, p In Hamao Umezawa (ed.), Progress in antimicrobial and antineoplastic chemotherapy. Proceedings of the 6th International Congress on Chemotherapy, vol. II. University Park Press, Baltimore, Md. 22. Last, J. A Studies on the binding of tetracyclines to ribosomes. Biochim. Biophys. Acta 195: Maxwell, I. H Studies of the binding of tetracyclines-to ribosomes in vitro. Mol. Pharmacol. 4: Miller, G. H., S. A. Khalil, and A. N. Martin Structure-activity relationships of tetracyclines. I. Inhibition of cell division and protein and nucleic acid syntheses in Escherichia coli W. J. Pharm. Sci. 60: Reynard, A. M., L. F. Nellis, and M. E. Beck Uptake of 3H-tetracycline by resistant and sensitive Escherichia coli. Appl. Microbiol. 21: Rhan, The size of bacteria as the cause of the logarithmic order of death. J. Gen. Physiol. 13: Sarker, S., and R. E. Thach Inhibition of fornylmethionyl-transfer RNA binding to ribosomes by tetracyclines. Proc. Natl. Acad. Sci. 60: Sompolinsky, D., Y. Zaidenzaig, R. Zieglar-Schlomowitz, and N. J. Abramova Mechanism of tetracycline resistance in Staphylococcus aureus. J. Gen. Microbiol. 62: Weisblum, B Antibiotic inhibitors of the bacterial ribosome. Bacteriol. Rev. 32: Wilson, G. S., and A. A. Miles Principles of bacteriology and immunity, vol. 1, 5th ed., p Williams & Wilkins Co., Baltimore.