2 The Mode of Action of Novobiocin A. MORRIS*, B.Pharm., Ph.D., M.P.S. and A. D. RUSSELL, B.Pharm., Ph.D., M.P.S. Welsh School of Pharmacy, University of Wales Institute of Science and Technology, Card& Wales
DISCOVERY AND CHARACTERISTICS O F NOVOBIOCIN
MECHANISM O F ACTION O F NOVOBIOCIN Changes in bacterial morphology Effects on bacterial cell wall synthesis Effects on the synthesis and integrity of the bacterial cytoplasmic membrane Effects on bacterial protein synthesis Effects on bacterial nucleic acid syntheses: RNA synthesis DNA synthesis Induction of a magnesium deficiency Effects on enzyme systems and electron transport
41 41 43 46 41 47 47 48 49 54
ADSORPTION OF NOVOBIOCIN ON TO BACTERIA
RESISTANCE TO, AND CROSS-RESISTANCE WITH, NOVOBIOCIN
*Present address: Bacterial Chemotherapy Unit, Glaxo Research Laboratories Ltd.. Greenford, Middlesex, England
THE MODE OF ACTION OF NOVOBIOCIN
A study of the mode of action of an antimicrobial drug attempts to elucidate its biochemical effects on susceptible organisms. In addition, such a study provides an insight into hitherto unknown structures and processes present in micro-organisms and may offer clues to the chemist who wishes to design a drug with a specific effect. Success in elucidating the precise mechanism of action of a drug depends on the complete understanding of the sensitive cell and, although much is known about microbial structure and metabolism, the existing gaps in knowledge are still sufficient to hamper experimental design and the interpretation of experimental results. The mode of action of many antibacterial agents is still unknown. An attempt has been made in this review to clarify the diverwand sometimes conflicting reports concerning the proposed mechanism of action of novobiocin.
DISCOVER+ AND CHARACTERISTICS OF NOVOBIOCIN Novobiocin (I) is the official name given to an antibiotic discovered by no fewer than five independent groups of workers [l-51. It is a dibasic acid, and before its chemical structure was completely elucidated it was defined as
7-[4-carbamoyloxytetrahydro-3-hydroxy-5-methoxy-6,6-dimet hylpyran-2yloxy]-4-hydroxy-3-[4-hydroxy-3-(3-methyl-2-butenyl)benzamido]-8-methylcoumarin. In the light of present knowledge , a better name would be
41 A. MORRIS AND A. D. RUSSELL ~-[3-0-carbamoyl-5,5-dimethyl-4-O-methyl-a-~-lyxosyl]-4-hydroxy-3-[4hydroxy-3-(3-methylbut-2-enyl)benzamido]-8-methylcoumarin. Novobiocin is marketed under the following trade names: Albamycin, Biotexin, Cathocin, Cathomycin, Inamycin, Spheromycin, Vulcamycin and Vulkamycin. The antibiotic is soluble in aqueous solution above pH 7.5  and in polar organic solvents. It is stable in the dark, but is slightly lightsensitive . In practice, it is used either as the mono- or di-sodium salt or as the calcium salt. Novobiocin possesses a fairly wide range of activity and is active mainly against Gram-positive bacteria, particularly Staphylococcus aureus, having a minimum inhibitory concentration of between 0-1 and 5 p g / d for most strains of this organism. Other organisms that are susceptible include Neisseria, Haemophilus and Brucella species and certain strains of Proteus, but it is rarely used against these organisms in clinical practice. It is normally reserved for use against penicillin-resistant staphylococcal infections. For a more detailed account of the range of activity and the clinical applications of novobiocin, two earlier reviews should be consulted [8, 91. Both of these reviews also describe the effects of inoculum size, pH, sodium chloride, metal ions, and serum on the activity of novobiocin against various bacteria.
MECHANISM OF ACTION OF NOVOBIOCIN CHANGES IN BACTERIAL MORPHOLOGY
Novobiocin induces filamentation in Gram-negative rods [ 1&13] with subsequent vacuolation  and loss of intracellular materials . It is, however, debatable whether the induction of filamentous forms is characteristic of a particular biochemical effect. Some workers have advocated that filamentation indicates a specific inhibition of cell wall synthesis [15, 161, but a number of antibacterial agents that exert their effects elsewhere in the cell also induce filamentation, for example mitomycin C , acridines [18, 191, nalidixic acid , ultraviolet light  and m-cresol . It is probable, therefore, that filamentation induced by novobiocin is not wholly related to a specific effect of the antibiotic. Novobiocin also causes chaining in Streptococcusfaecium , although this effect is not produced in all cocci . The induction of spheroplasts in Escherichia coli by novobiocin has also been reported  and, in fact, the antibiotic has been recommended for preparing spheroplasts in Gramnegative bacteria . In contrast, however, it has been shown by various workers  that novobiocin does not induce spheroplasts in Serratia marcescens or in various strains of E. coli and  that it may even prevent spheroplast induction caused by benzylpenicillin in hypertonic medium (Figure 2.1).
THE MODE OF ACTION OF NOVOBIOCIN
3 L Time ( h o u r s )
Figure 2.1. Effect of novobiocin on the conversion of E.coli cells into spheroplasts by benq.1penicillin. Benzvlpenicillin concentration throughout, 250 units/ml. Novobiocin concentrations (pgiml): 0, ( L O ; 50. x-x ; 100. 0-0;500, A-A. (From Morris and Russell , by courtesy of Microhios).
A. MORRIS AND A. D. RUSSELL EFFECTS ON BACTERIAL CELL WALL SYNTHESIS
Early workers  found that, like benzylpenicillin, vancomycin, ristocetin and bacitracin, novobiocin caused an excessive accumulation of cell wall precursor, uridine diphosphate-N-acetylmuramic acid-L-alanine-D-glutamic acid-L-lysine-D-alanine-D-alanine(UDP-MurNAc-L-ala-D-glu-L-lys-D-alaD-ala) in Staph. aureus and it was thus considered that novobiocin was a specific inhibitor of peptidoglycan synthesis with an effect similar to that of penicillin. However, subsequent studies led to the withdrawal of this hypothesis , since novobiocin caused the accumulation of other precursor-type compounds and also strongly inhibited both nucleic acid and protein synthesis in this organism. Thus, accumulation of particular precursors does not necessarily reflect the site of action of an antibacterial agent . Other investigations have shown that novobiocin causes a non-specific inhibition of cell wall synthesis in Staph. aureus and Strept.faeciurn [23, 31, 321 and that in E. coli the novobiocin-induced inhibition of cell wall synthesis is secondary to the inhibition of nucleic acid syntheses . With cellular extracts of Staph. aureus and Micrococcus lysodeikticus, novobiocin inhibits the formation of the alternating N-acetylmuramic acidpentapeptide and N-acetylglucosamineresidues
during cell wall peptidoglycan synthesis, but only at concentrations many times those needed to inhibit growth of the particular organism [34, 351; for example with Staph. aureus, 0.03 pg/ml is required for 50 per cent growth inhibition and as much as 10 pg/ml for 50 per cent inhibition of peptidoglycan synthetase . Benzylpencillin also does not inhibit the peptidoglycan synthetase reaction, whereas ristocetin, vancomycin and bacitracin do at antibiotic concentrations closely related to those which are inhibitory to cell growth. Vancomycin and ristocetin allow the formation of the lipid intermediate, GlcNAc-MurNAc(pentapeptide)-P-P-phospholipid,but prevent its utilisation for peptidoglycan synthesis [36, 371. It may be concluded that novobiocin does not interfere with the formation or utilisation of the lipid intermediate involved in the above reaction [36-381. With other cell-free systems obtained from Bacillus licheniformis and Lactobacillus plantarum, novobiocin was found to inhibit both polyglycerophosphate and polyribitol phosphate syntheses [39, 401. However, the
44 THE MODE OF ACTION OF NOVOBIOCIN workers involved emphasised the need for caution in relating the primary mode of action to these inhibitions. Moreover, it has been observed  that B. megaterium is sensitive to novobiocin although this organism does not contain teichoic acid in its cell wall. In addition, another organism, M . lysodeikticus, which contains no teichoic acid in its cell wall, is also sensitive to novobiocin . Novobiocin has been found to inhibit the synthesis of teichuronic acid by cellular extracts of B. licheniformis , but as this occurs only at concentrations of antibiotic much greater than the minimal concentration needed to inhibit growth, it is difficult to propose that this is its primary effect. An indication of whether an antibiotic exerts a specific effect on cell wall synthesis may be obtained by elucidating whether the antibiotic induces lysis of growing cells or the formation of spheroplasts, protoplasts or L-forms [42-44]. Warren and Gray [45, 461 found that Staph. aureus cells grown in subinhibitory concentrations of nafcillin underwent lysis on subsequent incubation with lysozyme and trypsin, whereas cells grown in subinhibitory concentrations of novobiocin were not rendered susceptible to the enzyme combination. Novobiocin has also been shown not to induce lysis of Staph. aureus [131 or of E. coli . As already mentioned, although it was originally claimed that novobiocin induced the production of osmotically-sensitive forms, it is now clear that this is not so. Whereas it has been reported that the antibiotic induces the production of L-forms , there are a number of publications showing that this does not, in fact, occur [48-521. Roberts  for example, found that methicillin, ampicillin, cycloserine; cephalothin, ristocetin, bacitracin and vancomycin, but not novobiocin, induced L-forms in meningococci. The present evidence therefore indicates that novobiocin does not exert a specific effect on cell wall synthesis. Cell wall-deficient bacterial forms (protoplasts, spheroplasts and L-forms) are frequently used for determining whether a particular antibiotic exerts effects on the bacterial cell other than on the cell wall [53, 541, and this technique has been employed by many workers studying the effects of novobiocin. Thus, novobiocin is very active against penicillin-induced L-forms of E. coli , Proteus mirabilis  and Strept. faecalis [56, 571. Lysozymeinduced protoplasts of Strept. faecalis  and of B. megaterium , and L-forms of meningococci [51, 521 and of Staph. aureus  are as sensitive to novobiocin as are the corresponding parent forms. L-forms of Strept. pyogenes are lysed by the antibiotic , and similar observations have recently been made with penicillin-induced spheroplasts of E. coli and lysozyme-induced protoplasts of B. megaterium, both forms undergoing lysis in the presence of novobiocin although only when placed in a medium conducive to growth  (Figure 2.2). In addition, novobiocin has been found to interfere with the action of penicillins on Staph. aureus [62, 631. Thus, it seems highly unlikely that novobiocin exerts a specific effect on cellwall synthesis.
A. MORRIS A N D A. D. RUSSELL
0.5 0 0 In Y
. I -
- 0.3 73 0
-0.3 .-ul . I -
1 2 3 4 Time after addition (hours)
Figure 2.2. Effect of novobiocin on the growth (as represented by changes in optical density) o j (a) spheroplasts of E.coli, and (b)protoplasts of B.megaterium. Novobiocin concentrations (pgiml), in ( a ) :0,O-O; 100, 0-0; 500, A-A; in ( b ) :0 - 4 ; 0 (benzy!penicillin 1.5 units/ml), A-A; 2.5, x-x ; 5, 0-0; 25, 0-0. (From Morris and Russell , by courtesy of Biocheniical Pharnzacology.)
THE MODE OF ACTION OF NOVOBIOCIN
EFFECTS ON THE SYNTHESIS AND INTEGRITY OF THE BACTERIAL CYTOPLASMIC MEMBRANE
In an early report , it was stated that novobiocin inhibits the synthesis of the bacterial cytoplasmic membrane, as it decreased the crypticity of E. coli strain ML 35 and caused a loss of ribonucleic acid (RNA) into the surrounding medium. These effects were not induced in resting cells and indicated that the antibiotic interfered with the synthesis of new membrane material. Another early investigation had shown that novobiocin induced the loss of 260 nm absorbing material from washed suspensions of Staph. aureus [MI. More recently, it has been observed that novobiocin-induced membrane damage occurs in ML strains of E. coli but not in non-ML strains includTA- and W. Further investigations into this problem, ing KIz-3000,B,15 however, reveal that novobiocin also induces a loss of intracellular materials in non-ML strains , and as this was shown to occur in a hypertonic medium, it indicated a more specific effect of the antibiotic on the cytoplasmic membrane. In view of these findings and the earlier report by the same workers , it is probable that membrane damage induced by novobiocin in E. coli is not specific to any particular strain of this organism. A possible explanation for these conflicting observations has been suggested by Morris and Russell , who pointed out that, in their experiments , a minimal concentration (0.0025per cent w/v) of magnesium sulphate was present in the synthetic medium used because excess of MgZ ions markedly reduced the extent of leakage. Smith and Davis  did not investigate the effect of varying concentrations of magnesium sulphate on leakage. Several biochemical studies into the effects of novobiocin on membrane synthesis have been carried out. For example, the incorporation of radioactive glucose into the mannose fractions of the cytoplasmic membrane of M . lysodeikticus has been found to be inhibited by novobiocin [311, although similar inhibitions occur in other cellular fractions. With the use of protoplasts of B. meguterium, it has been shown  that the incorporation of radioactive amino acids and glycerol into both membrane and cytoplasmic fractions is inhibited by similar degrees. It was concluded that, although the membrane may be malfunctional, the cytoplasmic structure of novobiocintreated protoplasts can be similarly described. As already stated, lysis of L-forms of Strepr. pyogenes by novobiocin has been reported , indicating an effect of the antibiotic on the membrane. However, puromycin was also found in this investigation to induce lysis, and the results should therefore be interpreted with caution, since puromycin, which chemically closely resembles the end-group of aminoacyl-tRNA, interferes with protein synthesis by forming a peptide link with the terminal carboxyl group of the incomplete peptide chains . Thus, novobiocin exerts a fairly specific effect at the cytoplasmic mem+
A. MORRIS AND A. D. RUSSELL 47 brane of E. coli, and such an effect occurs in more strains than those suggested by Smith and Davis . Little is yet known about membrane biosynthesis , and it is not possible to propose that the effects of novobiocin so far noted constitute a primary lesion.
EFFECTS ON BACTERIAL PROTEIN SYNTHESIS
A number of workers have shown that novobiocin inhibits protein synthesis in bacteria. For example, it inhibits B-galactosidase synthesis in both Staph. aureus and E. cofi [48, 641. Although M-protein synthesis in a Group A streptococcus is not inhibited by novobiocin , the inhibition of protein synthesis in Strept. faecium has been attributed to the inhibition of tRNA synthesis . Novobiocin inhibits protein synthesis in Staph. aureus , but as the inhibition of other macro-molecules also occurs to a similar extent, this effect may not be the primary action of the antibiotic. The antibiotic also inhibits protein synthesis in E. cofi [33, 711 but this inhibition appears much later than the observed inhibition of deoxyribonucleic acid (DNA) synthesis. Thus, the inhibition of protein synthesis is probably not the primary effect of novobiocin on bacteria but is normally secondary to the inhibition of other metabolic processes. EFFECTS ON BACTERIAL NUCLEIC ACID SYNTHESES
R N A Synthesis
Brock and Brock  found that novobiocin induced the loss of RNA from E. coli, but they did not propose that the antibiotic exerted a specific effect on RNA synthesis. Later, Brock  attributed the strong inhibition of RNA synthesis in Strept. faecium to a novobiocin-induced magnesium deficiency. However, Smith and Davis [33, 711 showed that the inhibition of RNA synthesis in E. coli is secondary to the inhibition of DNA synthesis. The degradation and leakage of RNA is induced by novobiocin with ML, but not with non-ML strains of the organism . In contrast, however, Morris and Russell  observed that novobiocin induced the loss of RNA-like material from non-ML strains of E. coli (see also Figure 2 4 , and although RNA synthesis was strongly inhibited, the inhibition of DNA synthesis occurred to an even greater extent . It was concluded that the loss of RNAlike material into the surrounding medium was a result of a loss of membrane integrity. A recent report  states that the novobiocin-producing organism Streptomyces niveus is itself inhibited by the antibiotic and that this inhibition
THE MODE OF ACTION OF NOVOBIOCIN
is due to an inhibition of nucleic acid, and particularly RNA, synthesis. However, the majority of findings indicate that novobiocin exerts only a secondary effect on RNA synthesis.
Figure 2.3. Effect ofnovobiocin on leakage of260 nm-absorbing niaterialfrorii log phase cells of E.coli in synthetic rnediurn. Novobiocin concentrations (pg/ml): 0, M;5, 0-0 ; 10, V-V ; 50, x-x ; 100, 0-0; 500. A-A. (From Morris and Russell . by courtesy of Microbios.)
D N A Synthesis
The evidence that novobiocin exerts a specific and primary effect on DNA synthesis in E. coli was the subject of a recent report . In all strains of E. coli employed in this study, DNA synthesis was inhibited earlier and to a greater extent than were other macroaolecular syntheses. The antibiotic did not cause degradation of DNA; its action was reversible, for when the
A. MORRIS AND A. D. RUSSELL
novobiocin was removed from a culture of the organism, DNA synthesis was immediately resumed at a rate comparable to that of an untreated culture. A similar observation has been made by Morris and Russell . Smith and Davis  also reported that the T,,, value of bacterial DNA was not affected by novobiocin, indicating that the antibiotic does not form a stable complex with bacterial DNA. By means of partially purified E. coli DNA polymerase and native E. coli DNA, it was found that novobiocin inhibited polymerisation . It also inhibited the replication of RNA phage M5-2 and DNA phages T2 and T3. Earlier studies [48, 741 had shown that novobiocin induced a decrease in the DNA content of E. coli cells, but this was not regarded as being of primary importance. Wishnow, Strominger, Birge and Threnn  subsequently reported that novobiocin strongly inhibited the incorporation into Staph. aureus of radioactive lysine and of inorganic phosphate into nucleic acid, but the inhibition of inorganic phosphate also occurred in other macromolecular fractions of the cell, and the primary effect of the antibiotic was not defined. More recently [72, 751, novobiocin has been shown to exert an immediate effect on DNA synthesis in both E. coli and Staph. aureus, this inhibition occurring to a greater extent than that of RNA and protein syntheses (Figure 2.4). A low concentration of novobiocin (20 pg/ml) does not inhibit protein synthesis in E. coli in nutrient broth at 37°C [72, 751. However, Morris and Russell  considered that the inhibition of DNA synthesis was secondary to an effect on the cytoplasmic membrane; a similar proposal has been made regarding the mode of action of phenethyl alcohol . In view of the rapid effects of novobiocin on both spheroplasts and protoplasts  and the close association of the bacterial membrane and chromosome during growth [77, 781, such a hypothesis is feasible, although further work along this line is desirable. However, mitomycin C, which forms cross-links with DNA and inhibits DNA synthesis , also induces lysis of growing spheroplasts of E. coli , and this finding suggests that the inhibition of DNA synthesis precedes membrane damage. INDUCTION OF A MAGNESIUM DEFICIENCY
Early studies [72, 791 showed that metal ions, in particular those of magnesium, antagonised the action of novobiocin against Gram-negative organisms but not its action against Gram-positive bacteria. These findings have since been confirmed . Evidence was later provided that novobiocin forms a complex with magnesium ions  and it was proposed that an intracellular deficiency of these ions was thereby induced by the antibiotic. Many of the effects induced by novobiocin were then shown by Brock [23, 32, 801 to be similar to those induced by a magnesium deficiency. However, a deficiency
THE MODE OF ACTION OF NOVOBIOCIN
Time after addition (hours)
Figure 2.4. EJect of novobiocin on ( a )growth, as recorded by changes in opticul density, ( b )DNA synthesis, (c) RNA synthesis, and (d)protein synthesis in E.coli in nutrient broth at 37°C. Novobiocrn concentrations (pgiml): 0. C k O ; 20, m-m; 100, 0-0; 500, A- A.(From Morris and Russell , by courtesy of Microbios.)
A. MORRIS AND A. D. RUSSELL 51 of magnesium ions does not provide the complete explanation of the mode of action of novobiocin. Against Neurospora crassa, for example, novobiocin exerts less activity when the magnesium content of the medium is lowered [8 11 and the novobiocin-induced inhibition of the particulate cytochromelinked oxidase of Mycobacteriurn phlei is not alleviated by magnesium ions . DNA polymerisation in vitro is independent of the magnesium concentration, although the reaction is inhibited by novobiocin . Magnesium ions are essential for the formation of cell wall precursors and, as novobiocin causes the accumulation of these same precursors, it is difficult to envisage the induction of a magnesium deficiency to explain fully the effect of novobiocin . More recently, it has been shown  that magnesium ions antagonise the bactericidal effect of novobiocin against E. coli (Table 2.1) in both complex
Table 2.1 ANTAGONISM BY MAGNESIUM SULPHATE OF THE BACTERICIDAL EFFECT OF NOVOHIOCIN (500 pg/ml) ON LOG-PHASE CELLS OF E.coli IN NUTRIENT BROTH (OXOID) AT 37"c. FIGURES ARE VIABLE NUMBERS OF CELLS x 107/ml.(From Morris and Russell , by courtesy of Microbios). Concentration ( % w/v) of magnesium sulphate added to broth'
Time (h) after addition of novobiocin
0 0.5 1 3
0.025 14.0 7.1 3.1 0.6
14.0 8.1 3.8 0.8
14.0 8.8 4.7 1.9
14.0 9.3 8.7 4.1
0.25 14.2 13.2 11.9 9.2
14.2 13.3 12.4 10.7
'The broth itself contained traces of Mg' +,although the concentrationwas not determined.
and synthetic media, but do not against Staph. aureus. However, the inhibition of cell division induced by novobiocin is not antagonised by magnesium ions, which indicates that these ions do not interfere with the uptake of the antibiotic, as had been previously suggested . In addition, it has been found [66, 831 that magnesium ions reduce novobiocin-induced loss of intracellular materials (Figure 2.5), the metal ions in some way stabilising the membrane. Further evidence that novobiocin does not exert its effects on E. coli by inducing a magnesium ion deficiency was obtained when both DNA and RNA syntheses were found to be inhibited to a similar extent in media containing a trace and an excess of the metal ions . In addition to the report that novobiocin forms a complex with magnesium ions , there are two other accounts which show that such a complex is not formed [84,85]. In their study, Morris, Russell and Thomas  obtained ) difference spectra with a solution of novobiocin sodium (5 x l O P 3 ~and ) an equal ionic strength solution either magnesium chloride (5 x 1 0 - 3 ~or of sodium chloride and observed a small peak with the same ,A at 346 nm in each case. Potentiometric titration curves confirmed the lack of complex
52 THE MODE OF ACTION OF NOVOBIOCIN formation between novobiocin and MgZ+.Furthermore, resting protoplasts of B. megaterium are not lysed by novobiocin , even though they are dependent on magnesium ions for optimal stability (Figure 2.6). Thus, antagonism between novobiocin and magnesium ions may occur at the cytoplasmic membrane, which requires magnesium ions for stability [86,
Figure 2.5. Antagonistic effect of MgZCiomon leakage of 260 nm-absorbing niateriul induced by novobiocin (100 pg/ml) from E.coli growing in synthetic medium conruining the following concentrations (% w/v) of niagnesiwn sulphate: 0.0025, &-0; 0.01, x -x ; 0.05, 0-0;0.1, A-A. Control (niagnesiuni sulphate 0.0025% w/v, novobiocin absent), 0-0. (From Morris and Russell , by courtesy of Microbios.)
871 but there is no direct evidence to show that the induction of a deficiency of magnesium occurs in the sensitive cell. It is also difficult to explain why the effects of novobiocin on Gram-positive cells are not alleviated by magnesium ions. This phenomenon, however, may be attributable to a difference in the mode of action of the antibiotic against different organisms or to differences in the assimilation of exogenous metal ions by different organisms
ES Figure 2.6. Effect of Mgz+ions (added at J) on the stability, as recorded by changes in optical density. of protoplusts of B.megaterium. Concentrations ( % w/v) of mugnesiuni suiphate: 0, W;0+?02, x-x ;0905, A-A: 0.05, !I-U; 0.1, 0-0; 0.125, 7 - 7 . (From Morris and Russell 1751, by courtesy of Microbips.)
THE MODE OF ACTION OF NOVOBIOCIN
EFFECTS ON ENZYME SYSTEMS AND ELECTRON TRANSPORT
Both respiratory processes and oxidative phosphorylation are inhibited by novobiocin in rat liver homogenates  and in Mycobucterium phlei . The antibiotic also inhibits the activity of nitrate reductase in E. coli, succinic dehydrogenase and ethanol dehydrogenase in M. lysodeikticus, and adenosine triphosphatase (ATP-ase) activity of sonic extracts of Strept. fuecium [23, 32, 481. It also reduces the activity of ATP-ase of E. coli  and the enzymes necessary for polyglycero-phosphate and polyribitol phosphate [39, 401, but it does not inhibit peptidoglycan synthetase activity . Pyruvate oxidation by washed suspensions of Staph. uureus [64,91], oxygen uptake by M. lysodeikticus with either ethanol, succinate or glucose as substrate, and the production of lactic acid from glucose by Strept. fuecium  are all inhibited by novobiocin. These inhibitions do not, however, appear to be of primary importance and are probably secondary to novobiocin-induced membrane damage. STRUCTURE-ACTIVITY RELATIONSHIPS Novobiocin does not belong to any particular chemically defined group of antibiotics although there are a number of groupings in the molecule which may be related to specific effects. The antibiotic is a coumarin derivative (I) (see p. 40) and has been shown to inhibit oxidative phosphorylation in Mycobacterium phlei . A vitamin K compound is necessary for electron transport in this organism, and both novobiocin-induced inhibition of electron transport and inhibition of growth are reversed by vitamin K. The antibacterial activity of a number of synthetic 3-acylamino-4hydroxycoumarin derivatives  and of several other coumarins has been described [93,94]. However, as the carbamoyl group is of great importance to novobiocin activity, it may be that the activity of coumarin is not directly related to the activity of the antibiotic. Novobiocin also contains a phenolic grouping in its structure. Phenols interfere with membrane integrity [95, 961 and it is therefore tempting to speculate that novobiocin-induced membrane damage is attributable to the phenolic group . However, unlike phenols, novobiocin has no activity against resting cells, and whereas novobiocin induces lysis of spheroplasts , phenol itself does not . Noviose, the sugar moiety of novobiocin, also appears to be important for the activity of the antibiotic, as has been suggested for the sugar moiety of streptomycin . However, removal of the carbamoyl group on the 3-hydroxyl results in the formation of decarbamoylnovobiocin (descarbamylnovobiocin, 11, R as in I) and complete loss of activity. Similarly, when the carbamoyl group is on the 2-hydroxyl, the resulting compound. isonovo-
A. MORRIS A N D A. D. RUSSELL 55 biocin (111, R as in I), is inactive ; this indicates that both the presence and positioning of this grouping are of major importance. Novobiocin, but not
Me0 ( 1 1 ) Decar barnoylnovobiocin
050 N HZ
(111 1 lsonovobiocin
decarbamoylnovobiocin, has been stated to bind magnesium ions , but recent findings have cast doubt on the significance of this observation [84, 851. It is not possible, therefore, to attribute the major effects of novobiocin to any particular part of its unique chemical structure, the intact molecule being necessary for full activity. ADSORPTION OF NOVOBIOCIN ONTO BACTERIA It has been reported that novobiocin is strongly bound to bacterial cells , the binding being intermediate in strength between that of penicillin and polymyxin. This binding is pHdependent and occurs with both sensitive and resistant organisms [1001. With the use of tritium-labelleddihydronovobiocin , it has now been found that, at O'C, the antibiotic is bound rapidly to Strept. faecium and in amounts roughly proportional to the external concentrations. At 37'C, however, more antibiotic was adsorbed after short incubation periods than after long periods, and this anomaly may be the result of some metabolism of the antibiotic. Besides its ability to bind to bacterial cells, novobiocin also binds to proteins [101, 1021, cellulose filter paper [lo31 and membrane filters . It is possible therefore that binding to washed suspensions of bacteria occurs non-specifically ; this appears reasonable considering the inactivity of novobiocin against resting cells [13, 48, 831. RESISTANCE TO, AND CROSS-RESISTANCE WITH, NOVOBIOCIN There are many reports available dealing with the acquired resistance of bacteria to novobiocin [105-1081 and, in view of the rapidity with which resistance develops, the antibiotic is often administered in combination with other antibiotics. However, it has also been reported that an organism resistant to novobiocin is usually sensitive to other commonly-used anti-
56 THE MODE OF ACTION OF NOVOBIOCIN biotics, and the organisms resistant to particular antibiotics are sensitive to novobiocin [2, 108-1 111. These studies have involved the use of bacitracin, chloramphenicol, penicillin, tetracycline, polymyxin and streptomycin, and the results suggest that novobiocin exerts effects which are different from the other antibiotics used. Novick  has recently described the enhancement of penicillinase plasmid-negative segregants by novobiocin, the rare negatives originally present in the population being able to mutate more readily to slightly higher levels of resistance to novobiocin than the penicillinasepositive population.
CONCLUSIONS It is difficult, at present, to propose a mode of action for novobiocin. The antibiotic exerts a variety of effects on bacteria, and although a number of hypotheses has been proposed, complete evidence for each is still lacking. The effects on novobiocin on cell wall, RNA and protein synthesis and on enzyme systems can now be considered as secondary effects of the antibiotic, and the theory that novobiocin induces an intracellular deficiency of magnesium is far from convincing. Recent findings show that the antibiotic has an immediate and pronounced effect on DNA synthesis and this may be of primary importance. It has also been shown that novobiocin interferes at an early stage with membrane integrity and, in view of the close association between membrane and chromosome in the bacterial cell, these findings have to be carefully considered. REFERENCES 1. B. M. Frost, M. E. Salliant, L. McClelland, M. Solotorovsky and A. C. Cuckler, Antibiot. Annu., 1955/56,918 2. F. K . Lin and L. L. Coriell, Antibiot. Annu., 1955/56, 634 3. H. Welch and W. W. Wright, Antibiot. Chemotherapy, 1955,5, 670 4. G. Rolland, P. Sensi, G. A. DeFerrari, G. Maffii, M. T. Timbal and L. G. Silvestri, Farniaco Ed. Sci., 1956, 11, 549 5. M. Kuroya, K. Katagiri, K. Sat0 and M. Mayani,,J. Anfibiot. (Japan) 1958, A l l , 187 6. B. T. Golding and R. W. Rickards, Chem. Ind. (London), 1963, 1081
7. H. Hoeksema, M. E. Bergy, W. G. Jackson, J. W. Shell, J. W. Hinman, A. E. Fonken, G. A. Boyack, E. L. Caron, J. H. Ford, W. H. Devries and G. F. Crum, Antibiot. Chemotherapy, 1956, 6, 143 8. M. Finland and R. L. Nichols, Antibiot. Cheniotherapia, 1957,4, 209 9. P. E. Macey and D. F. Spooner, Experimental Chemotherapy (editors: R. J. Schnitzer and F. Hawking), Vol. 3. Academic Press, London and New York, 1964, p. 291 10. C. G. Smith, A. Dietz, W. T. Sokolski and G. M. Savage, Anfibiot. Chenlofherapy, 1956. 6, 135 1 1 . T. D. Brock, J . Bacteriol., 1956, 72, 320 12. A. Coppo, Antibiot. Chemotherapy, 1957, 7 , 297 13. A. Morris and A. D. Russell, Expeyientia, 1968, 24, 195
14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53.
A. MORRIS AND A. D. RUSSELL 57 A. Morris, Ph.D. Thesis, University of Wales, 1969 J. T. Park, Symp. Soc. Gen. Microbiol., 1958,8, 49 G. W. Bazill, Nature, 1967,216, 346 W. Szybalski and V. N. Iyer, Antibiotics: Mechanism of Action (editors: D. Gottlieb and P. D. Shaw), Vol. 1, Springer-Verlag, Berlin, 1967 J. Ciak and F. E. Hahn, Science, 156,655 J. H. S. Foster and A. D. Russell, Inhibition and Destruction of the Microbial Cell (editor: W. B. Hugo), Academic Press, London and New York, 1971 (in press) W. A. Goss, W. H. Dietz and T. M. Cook, J. Bacteriol., 1964, 88, 1112 J. P. Payne, P. E. Hartman, S. Mudd and A. W. Phillips, J. Bacteriol., 1956, 72,461 A. C. R. Dean and P. L. Rogers, Biochim. Biophys. Acta, 1967,148,774 T. D. Brock, Antibiotics: Mechanism of Actions (editors: D. Gottlieb and P. D. Shaw), Vol. 1, Springer-Verlag, Berlin, 1967, p. 651 F. E. Hahn cited by K. McQuillen, The Bacteria (editors: I. C. Gunsalus and R. Y. Stanier), Vol. 1, Academic Press, London and New York, 1960, p. 275 A. C. Baird-Parker and R. C. S. Woodruffe, Progress in Microbiological Techniques (editor: C. H. Collins), Butterworths, London, 1967, p. 89 R. M. Wishnow, J. L. Strominger, C. H. Birge and R. H. Threnn, J. Bacteriol., 1965, 89, 117 A. Morris and A. D. Russell, Biochem. Pharmacol., 1968.17, 1923 A. Morris and A: D. Russell, Microbios, 1970 (in press) J. L. Strominger and R. H. Threnn, Biochim. Biophys. Acta, 1959,33,280 B. D. Davis and D. S. Feingold, The Bacteria (editors: I. C. Gunsalus and R. Y.Stanier), Vol. 4, Academic Press, London and New York, 1962, p. 343 T. D. Brock, Antimicrob. Agents Annu., 1960, p. 297 T. D. Brock, Science, 1962, 136, 316 D. H. Smith and B. D. Davis, J. Bacteriol., 1967,93, 71 J. S. Anderson, M. Matsuhashi, M. A. Haskin and J. L. Strominger, Proc. Nat. Acad. Sci. U S . . 1965.53, 881 J. S. Anderson, P. M. Meadow, M. A. Haskin and J. L. Strominger, Arch. Biochem. Biophys., x966, 116,487 J. L. Strominger, K. Izaki, M. Matsuhashi and D. J. Tipper, Fed. Proc., 1967,26, 9 J. L. Strominger, K. Izaki, M. Matsuhashi and D. J. Tipper, Topics in Pharmaceutical Sciences, 1968, 1, 53 J. S. Anderson, M. Matsuhashi, M. A. Haskin and J. L. Strominger, J. Biol. Chem., 1967, 242,3180 M. M. Burger and L. Glaser, J. Biol. Chem., 1964,239, 3168 L. Glaser, J . Biol. Chem., 1964,239, 3178 R. C. Hughes, Biochem. J., 1966,101,692 J. R. Ward, S. Madoff and L. Dienes, Proc. Soc. Exp. Biol. Med., 1958,97, 132 R. E. 0. Williams, J . Gen. Microbiol., 1963, 33, 325 C. W. Molander, B. M. Kagan, H. J. Weinberger, E. M. Heimlich and R. J. Busser, J . Bacteriol., 1964, 88, 591 G. H. Warren and J. Gray, Proc. Soc. Exp. Biol. Med., 1963, 114,439 G. H. Warren and J. Gray, Proc. Soc. Exp. Biol. Med., 1965, 120, 504 H. Gooder, Microbial Protoplasts, SpheroplastsandL-forms(editor: L. B. Guze), Williams & Wilkins Co., Baltimore, 1968, p. 341 T. D. Brock and M. L. Brock, Arch. Biochem. Biophys., 1959,85, 176 E. L. Krawitt and J. R. Ward, Proc. SOC.Exp. Biol. Med., 1963,114,629 J. Rotta, W. W. Karakawa and R. M. Krause, J . Bacferiof., 1965.89, 1581 R. B. Roberts, Proc. Soc. Exp. Biol. Med., 1967,124,611 R. B. Roberts, Microbiol Protoplasts, Spheroplasts and L-forms (editor: L. B. Guze), Williams & Wilkins Co., Baltimore, 1968, p. 230 T. W. Chang and L. Weinstein, Nature, 1966,211, 763
THE MODE OF ACTION OF NOVOBIOCIN A. D. Russell, Lab. Pracr., 1968, 17, 804 U. Taubeneck, 2. Allg. Mikrobiol., 1962, 2, 132 J. 2. Montgomerie, G. M.Kalmenson and L. B. Guze, J . Lab. Ciin. Med., 1966,68, 543 J. 2. Montgomerie, G. M. Kalmenson and L. B. Guze, Microbiol Protoplasts, Spheroplasts and L-forms (editor: L. B. Guze), Williams & Wilkins Co., Baltimore, 1968, p. 306 58. G. D. Shockman and J. 0. Lampen, J . Bacteriol., 1962,84,508 59. R. Hancock and P. C. Fitz-James, J . Bacteriol., 1964,87, 1044 60. B. M. Kagan, Microbial Protoplasts, Spheroplasts and t-forms (editor: L. B. Guze),
54. 55. 56. 57.
61. 62. 63. 64. 65. 66. 67. 68. 69. 70. 71. 72. 73. 74. 75. 76. 77. 78. 79. 80. 81. 82. 83. 84. 85. 86. 87. 88. 89. 90. 91. 92. 93. 94. 95. 96. 97. 98. 99. 100.
Williams and Wilkins Co., Baltimore, 1968, p. 314 C. Panos, M. Cohen and G. Fagan, J . Gen. Microbiol., 1967,46,299 G. H. Warren and J. Gray, Proc. SOC.Exp. Biol. Med., 1967,126, 15 G. H. Warren and J. Gray, Proc. SOC.Exp. Biol. Med., 1968, 128, 776 K. E. Anderson and R. 0. Burns, Sci. Studies St. Bonaventure Unit, 1957, 19, 57 A. Morris and A. D. Russell, Microbios, 1970, 2, 253 A. Morris and A. D. Russell, Microbios, 1970, 2, 35 M. D. Yudkin, Biochem. J . , 1963,89,290 D. Nathan, Proc. Nut. Acad. Sci. U S . , 1964, 51, 585 M. R. J. Salton, Annu. Rev. Microbiol., 1967, 21, 417 T. D. Brock, J . Bacteriol., 1963,85, 527 D. H. Smith and B. D. Davis, Biochem. Biophys. Res. Commun., 1965,18,796 A. Morris and A. D. Russell, Microbios, 1970, In press M. A. Al-Nuri and N. S. Egorov, Microbioiogya, 1968, 37,413; Chem. Abstr., 1968,69, 50033h T. D. Brock, J . Bacteriol., 1956, 72, 320 A. Morris and A. D. Russell, Microbios, 1970, In press S. Silver and L. Wendt, J . Bacteriol., 1967, 93, 560 F. Jacob, A. Ryter and F. Cuzin, Proc. Roy. Soc., Ser. B, 1966, 164,267 A. Ryter, Bacteriol. Rev., 1968, 32, 39 E. D. Weinberg, Antibiot. Annu., 1956, 1056 T. D. Brock, J. Bacteriol., 1962, 84, 679 S. C. Kinsky, J . Bacteriol., 1961,82, 889 M. M. Weber and G. Rosso, Bacteriol. Proc., 1963,95 A. Morris and A. D. Russell, Microbios, 1969, 1, 335 P. J. Niebergall, D. A. Hussar, W. E. Cressman, E. T. Sugita and J. T. Doluisio, J . Pharm. Pharmacol., 1966, 18, 729 A. Morris, A. D. Russell and I. L. Thomas, Experientia, 1967, 23, 244 C. Weibull, J. Bacteriol., 1953, 66, 688 K. McQuillen, The Bacteria (editors: I. C. Gunsalus and R. Y.Stanier), Vol. 1, Academic Press, London and New York, 1960, p. 275 D. W. Tempest, J. W. Dicks and J. L. Meers, 2. Gen. Microbiol., 1967, 49, 139 J. Frei, N. Canal and E. Gori, Experientia, 1958, 14, 377 Kessler and Rickinberg,cited by T. D. Brock, Antibiotics: Mechanism of Action (editors: D. Gottlieb and P. D. Shaw), Vol. 1, Springer-Verlag, Berlin, 1967, p. 652 W. W. Umbreit, G. Microbiol., 1956, 2, 398 K. Okumura, Yakugaku, Zasshi, 1960,80, 525 R. Selleri, 0. Caldini and G. F. Ferretti, Boll. Chirn. Farm., 1965, 104, 248 K. Hodak, v. Jakesovaand V. Dadak, Cesk. Farm., 1967.16, 86 J. Judis, J. Pharm. Sci., 1962, 51, 261 A. D. Russell, Progr. Med. Chem., 1968,6, 135 A. Morris, Unpublishedobservations M. Kogut and J. W. Lightbown, Experimental Chemotherapy (editors: R. J. Schitzer and F. Hawking), Vol. 3, Academic Press, London and New York, p. 3 C . G. Smith, cited by T.D. Brock and M. L. Brock, Arch. Biochem. Biophys., 1959,85,176 C . G. Smith, cited by T.D. Brock, Antibiotics: Mechanism of Action (editors: D. Gottlieb
A. MORRIS AND A. D. RUSSELL
101. 102. 103. 104. 105. 106. 107. 108. 109. 110. 111. 112.
and P. D. Shaw), Vol. 1, Springer-Verlag, Berlin, 1967, p. 651 W. Ritzenfeld, Arzneim-Forsch., 1957,7,464 D. M. Tennent, R. C. Mason, W. K. Gunther, M. E. Valiant and M. Solotorovsky, Proc. Soc. Exp. Biol. Med., 1957, 94, 814 W. T. Sokolski, N. J. Eilers and J. W. Shell, Antibiot. Annu., 1956-57, 1031 J. W. Lightbown, Proc. 7th International Congress of Microbiological Standardisation, Edinburgh, 1962 W. J. Martin, F. R. Heilman, D. R. Nichols, W. E. Wellman and J. E. Geraci, Proc. Staff Meetings Mayo Clinic. 1955, 30,540 M. Finland, Antibiot. Annu,, 1955-56, 929 E. Jawetz, W. Bertie and M. Sonne, Antibiot. Med. Clin. Ther., 1957, 4, 40 W. F. Jones, R. L. Nichols and M. Finland, J. Lab. Clin. Med., 1956, 47, 483 L. P. Garrod and P. M. Waterworth, Brit. Med. J., 1956, 2, 61 M. Barber, A. Csillag and A. J. Medway, Brit. Med. J., 1958, 2, 1377 J. J. Vavra, J. Bacferiol., 1967,93, 801 R.P. Novick, Bacferiol. Rev., 1969, 33, 210