Mechanism of Action of Phenethyl Alcohol: Breakdown of the Cellular Permeability Barrier

Transcription

1 JOURNAL OF BACTERIOLOGY, Feb., 1967, p Vol. 93, No Amerian Soiety for Mirobiology Printed In U.S.A. Mehanism of Ation of Phenethyl Alohol: Breakdown of the Cellular Permeability Barrier SIMON SILVER' AND LOUIS WENDT Virus Laboratory and Department ofbiohemistry, University of California, Berkeley, California Reeived for publiation 1 November 1966 ABSTAr Phenethyl alohol (PEA) aused Esherihia oli to take up greatly inreased amounts of ariflavine, a ompound to whih healthy growing ells are impermeable. PEA also aused an inreased rate of efflux (leakage) of ellular potassium under onditions whih do not greatly alter the influx of potassium via the energy-dependent potassium pump. We therefore propose that the primary effet of PEA is a limited breakdown of the ell membrane. The inhibition of deoxyribonulei aid synthesis and other ellular funtions would then be seondary onsequenes of the alteration in the membrane struture. The use of phenethyl alohol (PEA) as a bateriostati agent was first studied by Lilley and Brewer (15); the idea gained impetus when Berrah and Konetzka (2) reported that PEA ats by seletively inhibiting deoxyribonulei aid (DNA) synthesis in Esherihia oli. Soon after, PEA was used in studies of two moleular geneti problems: the role of DNA synthesis in baterial onjugation (3, 11, 2), and the repliation and reombination of bateriophage DNA (6, 12). In addition to its use with miroorganisms, PEA has been shown to inhibit the growth of mammalian ells in tissue ulture (13) and the multipliation of mammalian viruses (21). However, there has been diffiulty in several laboratories (14, 18, 22) in finding onditions for the seletive inhibition of DNA synthesis while protein and ribonulei aid (RNA) synthesis ontinue, and several observations have appeared whih are inexpliable if PEA primarily inhibits DNA synthesis. These inlude reports of (i) the inhibition of the growth of an RNA phage by PEA (17), (ii) the inhibition of the germination of spores (14, 26) under onditions where DNA synthesis does not our even in the absene of PEA, and (iii) inhibition of messenger RNA synthesis and enzyme indution (18, 22) and sporulation (19, 26) with onentrations of PEA lower than those required to affet DNA synthesis. Treik and Konetzka (28) suggested that PEA may inhibit DNA synthesis in E. oli indiretly by altering the membrane site of the initiation of I Present address: Department of Biology, Washington University, St. Louis, Mo. DNA synthesis. Lester (14) showed that PEA alters the permeability of Neurospora rassa to a variety of amino aids. We examined the effets of PEA on ell permeability with E. oli and onluded that the primary ation of PEA is at the level of the ell membrane with resultant breakdown of the ellular permeability barriers. The inhibition of DNA synthesis ould be due to a oupling of the initiation of DNA repliation to the membrane as proposed by Treik and Konetzka (28), or it ould be a seondary onsequene of the initial hange in ell permeability. The membrane baflier of a ell ats in two ways: it keeps out many ompounds whih an only enter the ell via speifi metaboli pumps, and it keeps in moleules olleted by these pumps (as well as most intermediate metabolites), enabling the ell to aumulate needed materials against a onentration gradient. We studied two results of permeability breakdown. PEA aused a greatly inreased uptake of ariflavine (a drug normally exluded from the ells), and an inreased rate of efflux (leakiness) of ellular potassium (whih is aumulated against a 1-fold onentration gradient). MATERIALS AND METHODS E. oli B (the usual host for bateriophages T2 and T4) and K-12 (the usual strain in studies of baterial onjugation) were both used. Ariflavine uptake. Ariflavine uptake was measured aording to Silver (24; in preparation), who has shown that the uptake of dye is passive and a measure of permeability. If the ells are permeable, the dye enters and binds strongly to the ellular nulei aids. The present experiments provide insuffiient evidene 56

2 VOL. 93, 1967 MECHANISM OF ACTION OF PHENETHYL ALCOHOL 561 to prove that ariflavine uptake reflets altered permeability rather than surfae binding; however, we may assume this for our present purposes. Evidene in support of this assumption inludes the parallel between hanges deteted by ariflavine experiments and those deteted by inreased leakage of K2. Also, toluene and hloroform, agents whih are onsidered to destroy the ellular permeability barriers, ause similar and large hanges in ariflavine binding (Silver and Spielman, unpublished data). The 2% of the ariflavine absorbed by the impermeable ontrol ultures represents a noise level whih is affeted by baterial growth onditions. The ariflavine (1- methyl-2,8-diamino aridine) was a purified sample with only traes of proflavine, and was a gift of the Pharmaeutials Division of Imperial Chemial Industries, Ltd., Alderly Park, Cheshire, England. The ells were grown to about 5 X 18 per milliliter in the gluose-salts minimal medium M9 (1). Ariflavine was added at a onentration of.75,ug/ml, and samples were removed and entrifuged at room temperature. The amount of dye in the supernatant fluids and the resuspended pellets was measured with an Amino fluoromirophotometer (Amerian Instrument Co., Silver Spring, Md.) equipped with a primary filter (Kodak-Wratten 47B) and a seondary filter (a ombination of Kodak-Wratten 2A and 12 filters). The fluoresene of the supernatant fluids is expressed as perentage of the fluoresene of the.75 jug/ml of ariflavine in M9 medium. K42 efflux and influx. Potassium flux was measured as will be desribed in greater detail by Silver and Spielman (in preparation). The K42 was spetrographi grade from Iso/Serve, In., Cambridge, Mass. E. oli ells were grown for several generations at 37 C in tryptone broth (8 g of Difo tryptone, 5 g of NaCl per liter of water) to a onentration of about 5 X 18 ells per milliliter. For the efflux experiments,.2 to.5 u/ml of K4 was added 2 hr before harvesting the ells. The radioative ells were entrifuged at 2 C, washed one with broth, and resuspended at 5 X 18 ells per milliliter. The ells were distributed in a series of small flasks, a 1-ml sample of eah was filtered through membrane filters (type HA; Millipore Filter Corp., Bedford, Mass.), and then the drugs were added to eah flask. The flasks were plaed in a 29 C shaking water bath, and additional 1 -ml samples were removed and filtered. The filters were not washed beause washing did not affet appreiably the quantitative results. The filtrates were olleted in small polyvials, and both the filters and.5-ml samples of the filtrates were ounted in a Nulear-Chiago gas-flow ounter (Nulear-Chiago Corp., Des Plaines, Ill.). After orretions for geometry of the samples and deay of the K42, the results are given as the perentage of the total K42 in the filtrates or in the ells. For the influx experiments, the nonradioative ells were distributed into flasks; the drugs were added at zerotime, and K42 was added 1 min later. The flasks were plaed in the 29 C water bath; l-ml samples were removed from time to time, filtered, and washed twie with 5 ml eah of broth at room temperature (Silver and Spielman, in preparation). In the influx experiments only the filters were ounted. K4 was added in amounts between.5 X 1-4 and 1-4 M to tryptone broth, whih was the medium of hoie sine it ontains only 6 X 1-4 M potassium. In some experiments, deoxyribonulease was added to redue the visosity and speed filtration. Drugs. Phenethyl alohol was purhased from Eastman Organi Chemial Division of Eastman Kodak Co., Rohester, N.Y. Nalidixi aid was the gift of the Sterling-Winthrop Researh Institute, Rensselaer, N.Y. Phleomyin (pleomyin) was the gift of Bristol Laboratories, Syrause, N.Y. RESULTS Effet of PEA on permeability of bateria to ariflavine. Healthy, growing E. oli ells are generally impermeable to aridine dyes, but they an be made permeable in a variety of ways, inluding bateriophage infetion (24) and treatment with membrane-disrupting agents suh as hloroform or toluene (Silver and Spielman, in preparation). In these ases, the ariflavine enters the ells and is reversibly bound, apparently by the nulei aids. PEA aused E. oli ells to beome permeable to, and to aumulate, CL NAL or 9 PEA minutes at 25 FIG. 1. Effet of PEA on the ell permeability to ariflavine. PEA or NAL was added at time zero to S X 18 ells of Esherihia oli B per milliliter. Ariflavine was added at 5 min. The ordinate shows fluoresene of dye left in the supernatant fluids after entrifugation.

3 562 SILVER AND WENDT J. BACTERIOL. ariflavine (Fig. 1). This effet was readily seen at PEA onentrations as low as.1 % (v/v), and maximal permeability ourred with the addition of about.25% PEA (ompare Fig. 1 and 2). However, nalidixi aid (NAL), whih is being used in similar studies to those involving PEA (7, 9; Wendt and Wedel, in preparation), had no effet on permeability to ariflavine (Fig. 1). Similarly, pleomyin, an antibioti whih seletively inhibits DNA synthesis (27), did not affet ariflavine permeability at onentrations up to 1,ug/ml (data not shown). The effet of PEA on ariflavine permeability was more rapid than that of toluene (Fig. 2), possibly beause PEA is misible with the aqueous ulture medium and toluene is not. The PEA-indued permeability to ariflavine differs strikingly from the bateriophage-indued permeability (24) in that the PEA effets do not depend on ellular metabolism (unaffeted by hlorampheniol or yanide). The effet of PEA on permeability to ariflavine was readily reversible (Fig. 3). In this experiment, ells were pre-exposed to PEA and were then allowed a 1-min "reovery" period. The reovery to the previous impermeable state was omplete not only with pretreatment with CL ) U) 1- a) a F 5 4F 3[ I v_ PEA or Tioluene V * Bat.ontrol +.25% PEA o +.25% Toluene I'.25% PEA, but also with.5% PEA, a higher onentration than has been used in published reports on effets of PEA. With 1.% PEA, the permeability breakdown was irreversible, as was the toluene-indued permeability (data not shown). When the ells whih bound ariflavine in the presene of PEA were entrifuged, resuspended in fresh medium with neither PEA nor ariflavine, and then entrifuged again, the originally bound dye repartitioned between the ell pellet and the seond supematant fluid in the same proportions (about 75:25) as in the first entrifugation. This again showed that the ariflavine binding was not due to a ternary omplex of ariflavine, PEA, and ells. The reversal of PEA-indued permeability may not require energy (Table 1). The ells regained their normal impermeable state even when yanide was present during the reovery period. FIG. 3. Reversibility of PEA-produed permeability. A ulture of Esherihia oli B at 7 X 18 ells per milliliter was divided into two samples. The "ontrol" 2F sample reeived.75,ug/ml of ariflavine (at time zero 1o on the absissa) and was again divided into two parts. To one part,.25% PEA was added at 1 min. The "pretreated" ulture was exposed to.25% PEA for min at room temperature, entrifuged, and resuspended minutes at 25 in the absene of PEA. After an additional 1-min "reovery" period, during whih the ulture was divided FIG. 2. Kinetis of PEA- and toluene-produed into two parts and.25% PEA was added to one portt tion, ariflavine was added (at time zero on the ab- permeability to ariflavine. Ariflavine was added a time zero to 7 X 18 Esherihia oli B ells per mil- sissa). Ariflavine uptake was measured with the supernatantfluids after liliter. PEA or toluene was added at 1 min. entrifugation. C a a) C. a U) L- ) ) Do IC PEA PEA ( )I (A) E 7 9-_Controls E6 l JPretreated F A 'A el.c I AA minutes at 25 I_

4 VOL. 93, 1967 MECHANISM OF ACTION OF PHENETHYL ALCOHOL 563 TABLE 1. Reovery from PEA-indued permeability in the presene of yanide Treatment Per ent ariflavine in supernatant fluid a No addition PEA added None (ontrols) Pretreated and resuspended Pretreated and resuspended in yanide a Average of four samples; two were entrifuged at 45 min, and two at 6 min. The experiment was onduted as desribed in the legend to Fig. 3, exept that the ells were grown in.2% sodium suinate. Addition of 13 M sodium yanide was made during PEA pretreatment, and ells were resuspended in medium with previously added yanide. This experiment was done with suinate-grown ells to minimize glyolysis as a possible soure of energy. It was repeated several times, and the effet of yanide plus iodoaetate on gluosegrown ells, or yanide on broth-grown ells, was also measured. Although none of these experiments showed an energy effet on reovery from PEA-indued permeability, the possibility remained that the ells have a low level energy soure whih is not ompletely turned off by the metaboli inhibitors used. Effet of PEA on loss of intraellular potassium. PEA aused a rapid loss of intraellular potassium (Fig. 4). The potassium leakage was dependent on onentration of PEA as were the growth-inhibitory effets and ariflavine permeability. We observed no signifiant differenes between E. oli strains B and K-12 with regard to PEAindued permeability. NAL did not affet potassium permeability, at least for the first 3 min after its addition (Fig. 4). The potassium leakage was as rapid with PEA as with toluene, but not as omplete. With toluene, the ells lost over 9% of their potassium within 3 min (data not shown). The reason for this differene was that PEA, unlike toluene, did not stop the potassium pump (Fig. 5). The inhibition of potassium influx by PEA was transient, and, after about 5 min (Fig. 5 and 6), the net rate of uptake by PEA-treated ells was similar to that for untreated ells. The reason for the reovery of net uptake rate is not known, but reovery must result from either an inreased rate of influx or a dereased rate of efflux in the presene of PEA. This may reflet the ability of the ell's ation pumps to adjust the influx rate to 6 PEA or 25- C.) I 4- * +QI1% PEA 2 A~~~ +.2%PEA +.3%PEA y + 1gLg/ml NAL minutes at 29 FIG. 4. Effet of PEA on K42 efflux. Radioative (K42) Esherihia oli K-12 ells were entrifuged and resuspended in nonradioative broth. Samples were taken andfiltered. PEA or NAL was added at 2 min. maintain a onstant internal potassium onentration in spite of variations in the external onentration (5) and efflux (Silver and Spielman, in preparation). The alternative explanation that the PEA-indued efflux is transient is less likely, sine there was no apparent hange in efflux rate for the first 15 min after addition of PEA (Fig. 4 and 6) and the apparent essation of efflux in Fig. 4 and 6 is due to the establishment of equilibrium between efflux and influx (Silver and Spielman, in preparation). Figure 6 shows the results of an experiment designed to measure the reovery of the normal potassium efflux rate after treatment with PEA and then the removal of PEA by entrifugation. The reovery in this ase was only partial (.3% PEA used), but it should be noted that by the design of the experiment there was no reovery period (suh as in Fig. 3 and Table 2). The addition of PEA aused inreased (and idential) efflux rates from the pretreated ells and from those whih had not been previously exposed to PEA. In another experiment (Table 2), the ells were treated with.25% PEA for 25 min; the PEA was removed and the ells were "loaded'>

5 564 SILVER AND WENDT J. BACTERIOL. 4 -I minutes at 29 FIG. 5. Effet of PEA on the iqflux of KO. At time zero,.25% PEA or toluene was added to S X 18P E-sherihia oli B ells per milliliter. KU (.5,u/ml, 1-4m) was added at I min. Samples were filtered and washed as indiated. with K41 for 15 min. The potassium efflux from these pretreated ells showed a 9% reovery toward the ontrol rate. DISCUSSION These studies suggest that PEA auses a rapid and reversible breakdown in the permeability barriers of baterial ells. This effet ours over the same onentration range as the inhibitory effet of PEA on maromoleular synthesis. PEA is being ompared here with toluene, whih auses an equally rapid but irreversible permeability breakdown without ell lysis (1; Silver and Spielman, unpublished data). In fat, we think of PEA as a reversible toluene. It is reasonable to believe that the ellular permeability barrier is the primary site of ation of PEA, but primay annot be proven experimentally. The proposed mode of ation for PEA makes it easier to understand the very naffow onentration limits within whih PEA is reversibly inhibitory and not lethal. Toluene is ommonly used to disrupt ellular permeability barriers and to expose internal enzymes to external substrates, but little is known about the mehanism of ation of toluene (1). We have onfirmed and extended the findings of Jakson and DeMoss (1). The extent and kinetis of damage of E. oli ells by toluene depend on the relative volume of toluene added. With less than 1% (v/v) toluene, the ells do not lyse. There is a rapid breakdown of the ellular permeability barrier (Fig. 2), and small moleules (potassium and ariflavine) an pass through the membrane. With time there is further disruption, and larger and larger moleules, suh as proteins and nulei aids, an esape from the ells (1; Silver and Spielman, unpublished data). Eventually, after several hours, the ells lyse and release what remains of their ytoplasm (Silver and Spielman, unpublished data). The similarities and differenes between the effets of PEA and toluene an be onsidered in terms of their moleular strutures (Fig. 7). Both have hydrophobi rings, but only PEA has the additional hydrophili aloholi group. How an a nonspeifi attak on the ell membrane by PEA aount for the varied but seletive inhibitions of ellular proesses that have been reported? Two general types of explanation an be envisaged. The inhibition may be the result of a diret strutural oupling of the proess to the membrane. Baterial onjugation, sporulation, and the germination of spores all involve growth of, or alterations in, the ell membrane. Similarly, DNA synthesis appears to take plae on the membrane (for a review of the evidene, see 25). On the other hand, the inhibition may be more indiret and result, by leakage, from the ellular loss of neessary small moleules. A possible basis for seletive but indiret inhibition of DNA synthesis is provided by the observations of Ennis and o-workers (4, 16). By use of E. oli mutants genetially defetive in their ability to onentrate potassium, they showed that ellular DNA and protein synthesis is muh more dependent on the high intraellular potassium onentration than is the synthesis of RNA. The reovery from the PEA-indued permeability does not appear to require energy metabolism, indiating that repair synthesis is not neessary. Thus, the reovery from PEA like the initial ation of PEA may be a physial rather than a physiologial proess. This physial effet of a small moleule (PEA) on the properties of a large maromoleule (the ell membrane) an be thought of as an example of "membrane allostery." The absorption of PEA by the membrane alters the onformation of the membrane, resulting in a breakdown of its strutural integrity. When the PEA is removed, the membrane returns

6 VOL. 93, 1967 U) C) : 4. C) MECHANISM OF ACTION OF PHENETHYL ALCOHOL I I I I (A) Influx 9. 4/ 75 PEA 6. K42 o // _ i If I.~- I I minutes at 28 C FiG. 6. Reversibility of PEA-produed permeability to potassium. K" (.5 A/mi, 1-' ) was added to two samples ofa ulture ofesherihia oli B at 4 X 18 ells per milliliter. Samples were removed,filtered, and washed as indiated. At 2.5 min,.3% PEA was added to one ulture. At 25 min, both samples were entrifuged at 5, rev/min for 1 mi at 2 C. The ells were resuspended in equal volumes offresh broth and eah sample was divided in half. Samples were removed andfiltered as usual. PEA (.25%) was added to one portion eah of the pretreated and ontrol ells 4 min after the original addition ofk2. TABLE 2. Reovery from PEA-indued permeability to K4a Culture $ (min-) Control..2 Control plus PEA..87 Pretreated and "reovered"...27 Pretreated, "reovered," plus PEA...87 a A ulture of Esherihia oli B was exposed to.25% PEA for 25 min at 25 C, entrifuged, and "loaded" with K42 for 15 min. After an additional entrifugation to remove extraellular K4,.25% PEA was added to a sample of this "pretreated and reovered" ulture and to a parallel ontrol ulture. The k values were from the exponential part of the efflux urves and the equation: perentage of K4" remaining in ell = e-t. The PEAindued efflux shows a 4A-fold inrease in rate (.87/.2), and the frational reovery is 9% [( ). ( ) X 11. to its original onfiguration. Suh a reversible transition is readily imaginable within the framework of urrent ideas on membrane struture (8). We suggest that previous studies in whih PEA was used should be re-evaluated to onsider the la * jpea I I I I (B) Efflux \\i \4bOO C * +E pretreated A- - +PEA -I I I I I CH2CH2H Phenethyl alohol /CH3 Toluene FIG. 7. Moleular strutures of PEA and toluene. possibility that the primary site of ation of PEA is the ell membrane. Possible effets on the membrane of other agents whih are purported to at solely on DNA synthesis should be investigated. For example, phenethyl alohol protets against thymineless death (23), and this annot be due to inhibition of DNA synthesis. Thymine starvation of a thymine-requiring mutant also results in inreased permeability to ariflavine (Freifelder and Silver, unpublished data). It is also possible that PEA might be a useful reagent for studies of other membrane and permeability phenomena, suh as the struture and funtioning of hloroplasts and mitohrondri. _.ws C) 3 Z C-._ 2 < J a.

7 566 SILVER AND WENDT J. BACTERIOL. ACKNOWLEDGMENTS This investigation was supported by Publi Health Servie grant GM from the National Institute of General Medial Sienes and grant Al 6272 from the National Institute of Allergy and Infetious Diseases. We are indebted to P. M. Spielman for his assistane. LITERATURE CrrED 1. ADAMS, M. H Bateriophages. Intersiene Publishers, In., New York. 2. BERRAH, G., AND W. A. KoNErzKA Seletive and reversible inhibition of the synthesis of baterial deoxyribonulei aid by phenethyl alohol. J. Bateriol. 83: BOUCK, N., AND E. A. ADELBERG The relationship between DNA synthesis and onjugation in Esherihia oli. Biohem. Biophys. Res. Commun. 11: COHEN, P. S., AND H. L. ENNIs The requirement for potassium for bateriophage T4 protein and deoxyribonulei aid synthesis. Virology 27: EPSTEIN, W., AND S. G. SCHULTZ Cation transport in Esherihia oli. V. Regulation of ation ontent. J. Gen. Physiol. 49: FoLsoME, C. E Inhibition of reombination and heterozygosis in phenyl ethyl alohol treated phage T4-E. oli B omplexes. Biohem. Biophys. Res. Commun. 11: Goss, W. A., W. H. DEITZ, AN T. M. COOK Mehanism of ation of nalidixi aid on Esherihia oli. II. Inhibition of deoxyribonulei aid synthesis. J. Bateriol. 89: GREEN, D. E., AND J. F. PERDUE Membranes as expressions of repeating units. Pro. Natl. Aad. Si. U.S. 55: HOLLOM, S., AND R. H. PRITCHARD Effet of inhibition of DNA synthesis on mating in Esherihia oli K12. Genet. Res. Cambridge 6: JACKSON, R. W., AND J. A. DEMoss Effets of toluene on Esherihia oli. J. Bateriol. 9: JACOB, F., S. BRENNER, AND F. CUZIN On the regulation of DNA repliation in bateria. Cold Spring Harbor Symp. Quant. Biol. 28: KONETZKA, W. A., AND G. BERRAH Inhibition of repliation of bateriophage T2 by phenethyl alohol. Biohem. Biophys. Res. Commun. 8: LEACH, F. R., N. H. BEST, E. M. DAVIS, D. C. SANDERS, AND D. M. GIMLIN Effet of phenethyl alohol on ell ulture growth. I. Charaterization of the effet. Exptl. Cell Res. 36: LESTER, G Inhibition of growth, synthesis, and permeability in Neurospora rassa by phenethyl alohol. J. Bateriol. 9: LILLEY, B. D., AND J. H. BREWER The seletive antibaterial ation of phenylethyl alohol. J. Am. Pharm. Asso., Si. Ed. 42: LUBIN, M., AND H. L. ENNIS On the role of intraellular potassium in protein synthesis. Biohim. Biophys. Ata 8: NONOYAMA, M., AND Y. IKEDA Inhibition of RNA phage growth by phenethyl alohol. Biohem. Biophys. Res. Commun. 15: PREVOST, C., AND V. MOSES Ation of phenethyl alohol on the synthesis of maromoleules in Esherihia oli. J. Bateriol. 91: REMSEN, C. C., D. G. LUNDGREN, AND R. A. SLEPECKY Inhibition of the development of the spore septum and membranes in Baillus ereus by 3-phenethyl alohol. J. Bateriol. 91: ROESER, J., AND W. A. KONETZKA Chromosome transfer and the DNA repliation yle in Esherihia oli. Biohem. Biophys. Res. Commun. 16: ROIZMAN, B Reversible inhibition of herpes simplex multipliation in HEp-2 ells with phenethyl alohol. Virology 19: ROSENKRANZ, H. S., H. S. CARR, AND H. M. ROSE Phenethyl alohol. I. Effet on maromoleular synthesis of Esherihia oli. J. Bateriol. 89: ROSENKRANZ, H. S., H. S. CARR, AND H. M. ROSE Phenethyl alohol. II. Effet on thyminerequiring Esherihia oli. J. Bateriol. 89: SILVER, S Ariflavine resistane: a bateriophage mutation affeting the uptake of dye by the infeted baterial ells. Pro. Natl. Aad. Si. U.S. 53: SILVER, S Moleular genetis of bateria and bateriophages. Progr. Biophys. Mol. Biol. 16: SLEPECKY, R. A Inhibition of sporulation and germination of Baillus megaterium by phenethyl alohol. Biohem. Biophys. Res. Commun. 12: TANAKA, N., H. YAMAGUCHI, AND H. UMEZAWA Mehanism of ation of phleomyin, a tumor-inhibitory antibioti. Biohem. Biophys. Res. Commun. 1: TREICK, R. W., AND W. A. KONETZKA Physiologial state of Esherihia oli and the inhibition of deoxyribonulei aid synthesis by phenethyl alohol. J. Bateriol. 88: