JOURNAL OF BAcTEioLOY, Oct. 1969, p. 310-318 Vol. 100, No. 1Copyright © 1969 American Society for Microbiology Printed In U.S.A.

Mechanism of Action of the AntifungalAntibiotic PyrrolnitrinR. K. TRIPATHI AND DAVID GOTTLIEB

Department of Plant Pathology, University of Illinois, Urbana, Illinois 61801

Received for publication 22 July 1969

Pyrrolnitrin at 10 Ag/ml inhibited the growth of Saccharomyces cerevisiae,Penicillium atrovenetwn, and P. oxalicwn. The primary site of action of pyrrol-nitrin on S. cerevisiae was the terminal electron transport system between succinateor reduced nicotinamide adenine dinucleotide (NADH) and coenzyme Q. Atgrowth inhibitory concentrations, pyrrolnitrin inhibited endogenous and exogenousrespiration immediately after its addition to the system. In mitochondrial prepara-tions, the antibiotic inhibited succinate oxidase, NADH oxidase, succinate-cyto-chrome c reductase, NADH-cytochrome c reductase, and succinate-coenzyme Q6reductase. In addition, pyrrolnitrin inhibited the antimycin-insensitive reduction ofdichlorophenolindophenol and of the tetrazolium dye 2,2'-di-p-nitrophenyl-(3,3'-dimethoxy-4,4'-bi-phenylene)5,5'-diphenylditetrazolium. The reduction of anothertetrazolium dye, 2-p-iodophenyl-3-p-nitrophenyl-5-phenyltetrazolium chloride,that was antimycin-sensitive, was also inhibited by pyrrolnitrin. The antibiotic hadno effect on the activity of cytochrome oxidase, and it did not appear to bind withflavine adenine dinucleotide, the coenzyme of succinic dehydrogenase. In wholecells of S. cerevisiae, pyrrolnitrin inhibited the incorporation of "4C-glucose intonucleic acids and proteins. It also inhibited the incorporation of "4C-uracil, 'H-thymidine, and "C-amino acids into ribonucleic acid, deoxyribonucleic acid, andprotein, respectively. The in vitro protein synthesis in Rhizoctonia solani and Esch-erichia coli was not affected by pyrrolnitrin. Pyrrolnitrin also inhibited the uptakeof radioactive tracers, but there was no general damage to the cell membranes thatwould result in an increased leakage of cell metabolites. Apparently, pyrrolnitrininhibits fungal growth by inhibiting the respiratory electron transport system.

Pyrrolnitrin has been isolated from Pseu-domonas pyrrocinia (1, 11) and P. aureofaciens(14). The antibiotic is 3-chloro-4-(2-nitro-3-chlorophenyl)-pyrrole (Fig. 1). Its physical andchemical properties have been described byArima et al. (2). Pyrrolnitrin is an antifungalantibiotic and is most active against dermato-phytic fungi, especially the species of Tricho-phyton (2). It is synthesized from tryptophan, andthe synthesis is probably initiated by a chloro-peroxidase enzyme system acting on tryptophan(14). The present report deals with the mechanismof action of pyrrolnitrin.

MATERIALS AND METHODSPyrrolnitrin was obtained from Eli Lilly & Co.,

Indianapolis, Ind. The solutions of the antibioticwere prepared in 95% ethyl alcohol. Inorganic andorganic chemicals were purchased from either Mal-linckrodt Chemical Works, St. Louis, Mo., or FisherScientific Co., Pittsburgh, Pa., and biochemicalswere purchased from Sigma Chemical Co., St. Louis,

Mo. The chemicals for determination of radioactivitywere obtained from Packard Instrument Co., Inc.,Downers Grove, Ill., and radioisotopes were obtainedfrom New England Nuclear Corp., Boston, Mass.

S. cerevisiae, a strain of brewing yeast, was kindlysupplied by F. M. Clark, Department of Microbiology,University of Illinois. Penicillium atrovenetum andP. oxalicum were obtained from CommonwealthMycological Institute, Kew, England. The yeastculture was maintained in Difco Sabouraud DextroseAgar slants. Liquid cultures were grown in flaskscontaining glucose-yeast extract medium (GYE)consisting of 2% glucose and 1% yeast extract inglass-distilled water. The inhibition of growth of S.cerevisiae by pyrrolnitrin was studied by growing theyeast in tubes containing glucose-yeast nitrogen basemedium (0.2% glucose and 6.7% yeast nitrogen basein glass-distilled water) with or without pyrrolnitrin.Growth was measured turbidimetrically at 650 nmin a Coleman Junior Spectrophotometer. The yieldof P. atrovenetum and P. oxalicum was measured ona dry weight basis after growing the culture for 48hr on a reciprocal shaker. The mycelium was collected

310

on April 30, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

ACTION OF PYRROLNITRIN

by filtration on Whatman no. 1 filter paper and driedin an oven at 80 C to a constant dry weight.

Respiratory studies. Yeast cells were grown inGYE medium to a density equivalent to 6.0 mg (dryweight) per ml. They were collected by centrifugationat 12,000 X g and washed twice with deionizeddistilled water and once with 0.2 M sodium phosphatebuffer (pH 7.4). The washed cells were then suspendedin the same buffer and either were used as wholecells or were disintegrated in a French pressure cellat 15,000 psi to make the cell-free extracts. StandardWarburg respirometry was used (25).

Mitochondria from the yeast cell-free extractswere isolated by the method of Utter et al. (26).Beef heart mitochondria were isolated from a freshbeef heart by the method of Crane et al. (6). Theprotein content in these mitochondrial preparationswas determined by the biuret method of Gornallet al. (7).The effect of pyrrolnitrin on the production of

'4CO2 from 14C-glucose was studied in the intra-scintillation vial reaction tubes of Slater et al. (24).Enzyme assays. Succinate oxidase and cytochrome

oxidase were assayed manometrically by measuringthe oxygen uptake with sodium succinate and ascorbicacid and cytochrome c as substrates, respectively.Activities of other enzymes were assayed on a Beck-man DU spectrophotometer attached to a Gilfordmultiple sample absorbance recorder (model 2000).Absorption cells (1 ml) with a 1-cm light path wereused. Reduced nicotinamide adenine dinucleotide(NADH) oxidase was assayed by measuring thedecrease in absorbancy at 340 nm with NADH as asubstrate. Succinate and NADH-cytochrome creductases were assayed by measuring the increasein the absorbancy at 550 nm resulting from the reduc-tion of cytochrome c. An extinction coefficient of19.2 X 103 liters X mol-1 X cm-' was used to cal-culate the amount of reduced cytochrome c (12).Succinate-dichlorophenolindophenol (DPIP) reduc-tase was assayed by the decrease in absorbancy at600 nm resulting from the formation of the leucodye (reduced DPIP). The reduction of tetrazoliumdyes (formazan formation) was measured by anincrease in absorbancy at 530 nm for 2,2'-di-p-nitrophenyl-(3, 3'-dimethoxy-4,4'-bi-phenylene) 5,5'-diphenylditetrazolium (NBT) and at 570 nm for2-p-iodophenyl-3-p -nitrophenyl - 5 - phenyltetrazoliumchloride (INT). Succinate-coenzyme Q6 reductaseactivity was measured by the method of Ramasarmaand Lester (21). The reaction mixture for each of theenzyme systems is described below.

The possible binding of pyrrolnitrin with flavineadenine dinucleotide (FAD) was measured by thechanges in the absorption spectra of FAD and pyr-rolnitrin in the mixture of the two compounds.The effect of pyrrolnitrin on the permeability of

yeast cells was determined by two methods, i.e..the uptake of radioactive tracers and the leakage ofmetabolites from the cells. For leakage experiments,1 g (wet weight) of late exponential-phase cells wassuspended in 25 ml of 0.1 M sodium acetate buffer(pH 6.5) in 125-ml Erlenmeyer flasks. After addingappropriate volumes of pyrrolnitrin solutions or

uJclN

Pyrrolnitrih:[3-chloro- 4 - (2 nitro-31-chlorophenyl ) -Pyrrole]

FIG. 1. Structure ofpyrrolnitrin.

alcohol, the flasks were incubated on a reciprocalshaker for a total of 8 hr at 26 C. Portions (3 ml)were removed after 30 min and after 1, 3, 6, and 8 hr,and the cells were separated by centrifugation. Fromthe supernatant solutions, 0.5-ml samples were usedto determine the contents of ninhydrin-positive ma-terials (17), inorganic phosphate (4), reducing sugars(18), and nucleotides. The nucleotides were measuredby the absorption at 260 nm.The effect of pyrrolnitrin on the general cell me-

tabolism was studied with "C-glucose (UL). A 1-,ucamount of 14C-glucose (specific activity, 21 mc/mmole) was added to 25 ml of GYE medium in eachof fifteen 125-ml Erlenmeyer flasks. Pyrrolnitrin oralcohol was added to appropriate flasks, and theflasks were inoculated with mid-exponential-phaseyeast cells equivalent to 15 mg (dry weight). Theflasks were incubated at 26 C for 1 hr on a reciprocalshaker. The 14C02 given off by the cells was absorbedin hyamine hydroxide as described by Gottlieb andTripathi (10). This absorbed 14CO2 was counted intoluene scintillation fluid [2,5-diphenyloxazole(PPO), 5 g; 1 ,4-bis-2-(5-phenyloxazolyl)-benzene(POPOP), 0.3 g; toluene, 1 liter]. Water-misciblesamples were counted in dioxane scintillation fluid(PPO, 15 g; POPOP, 0.75 g; ethyl cellosolve, 250 ml;naphthalene, 75 g; dioxane, 1,250 ml). All measure-ments of radioactivity were done in a Packard Tri-Carb liquid scintillation spectrometer. Quench cor-rections were made by using 14C-toluene as an internalstandard.The amount of "C-glucose taken up by the cells

during incubation was calculated by determining theradioactivity of a portion of the incubation mixturebefore and after incubation. The cells were removedfrom the medium by centrifugation and were washedthree times with distilled water. The cells were ex-tracted twice with 5 ml of 80% ethyl alcohol to removesmall soluble molecules, e.g., sugars. The suspensionwas centrifuged at 1,200 X g, and the supernatantsolutions from ethyl alcohol extractions were com-bined. In this solution, the presence of amino acidsand reducing sugars was detected by the ninhydrintest (17) and the anthrone test (18), respectively.The residue from the ethyl alcohol treatment wasextracted first with 5 ml of acetone-ether (1:1) for3 hr and then with 5 ml of ether for 1 hr to remove thelipids. Residual ether from the cell residue was re-

VOL. 100, 1969 311

cl

N02

on April 30, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

TRIPATHI AND GOTTLIEB

moved by passing nitrogen gas over it. This residuewas extracted with 5 ml of ice-cold trichloroaceticacid for 2 hr at 4 C to remove nucleotides, smallpolysaccharides, etc. The residue obtained aftercentrifuging this suspension was then extracted with10 ml of 10% trichloroacetic acid for 1 hr in an ovenat 90 C to remove the nucleic acids. Samples from thesupernatant solution obtained were used to measureribonucleic acid (RNA; 3), deoxyribonucleic acid(DNA; 23), and absorbancyat 260 nm. After removingthe nucleic acids from the cells, protein was extractedwith 5 ml of 0.2 N NaOH at 90 C for 1 hr (10). Sam-ples were removed for ninhydrin (17) and biuretassays (7) to estimate the protein content in thisfraction. Samples (1 ml) from each of the abovefractions were used to determine the radioactivity.The inhibition of incorporation of 'IC-glucose intovarious fractions was then calculated.The effect of pyrrolnitrin on the incorporation of

14C-amino acids into protein in yeast cells was stud-ied by using algal protein hydrolysate, uniformlylabeled with a specific activity of 15 mc/mmole.Erlenmeyer flasks (500 ml) containing 100 ml ofGYE medium were inoculated with yeast equivalentto 2.5 mg (dry weight). After incubating these flasksfor 12 hr, alcohol or pyrrolnitrin solutions wereaseptically added. After 15 min of incubation toexpose the cells to the antibiotic, 4 ,c of 'IC-aminoacid mixture per ml was added to each flask, and thecontents were mixed by manual shaking. Sampleswere removed to determine the radioactivity in themedium. After an additional 2 hr of incubation, thecells were harvested and washed with distilled waterand centrifugation. Portions of the supernatant solu-tion were removed to determine the radioactivitytaken up by the cells. The cells were then extractedin given order with 10% ice-cold trichloroacetic acid,ethyl alcohol-ether (3:1), ether, and 10% hot tri-chloroacetic acid (90 C for 1 hr). The protein wasthen extrcted in 0.2 N NaOH by the method ofGottlieb et al. (9). The protein content in the super-natant solution was determined by the method ofLowry et al. (15) by using bovine serum albumin asa standard.The synthesis ofRNA was studied by incorporation

of 14C-uracil (30 mc/mmole) into RNA fraction. Thesame general methods were followed as for proteinsynthesis except that 125-ml flasks with 25 ml ofmedium were inoculated with cells equivalent to0.5 g (dry weight), and the 10% hot trichloroaceticacid and protein extractions were omitted. RNAwas extracted by a slightly modified method ofSchmidt and Thanhauser (22) as given below. Afterthe cold trichloroacetic acid extraction, the cellresidue was hydrolyzed with 0.5 N NaOH at 37 Cfor 12 hr. The RNA content in the supernatant solu-tion was measured by the orcinol method (3), andthe incorporation of 14C-uracil into RNA was cal-culated as counts per minute per milligram of RNA.The DNA synthesis in yeast cells was studied by

the incorporation of 'H-thymidine (methyl labeled,2 c/mmole) into 10% hot trichloroacetic acid extract(nucleic acids). A 2-pc amount of 'H-thymidine was

added in the medium, and the same general methodswere followed as for protein synthesis up to hottrichloroacetic acid extraction. The DNA contentin the trichloroacetic acid extract was measured bydiphenylamine test (23).

Polyuridylic acid (poly U)-directed in vitro proteinsynthesis in Rhizoctonia solani was studied by themethod of Obrig, Cerna, and Gottlieb (19). The invitro protein synthesis with Escherichia coli ribosomeswas studied by the method of Clark et al. (5).

RESULISPyrrolnitrin inhibition of the total growth of

S. cerevisiae increased with increasing concen-trations of antibiotic until, at 5 ,ug/ml, the in-hibition was 95% (Fig. 2). Cultures of the yeasttreated with 10 ,ug of pyrrolnitrin per ml did notgrow even after 1 week of incubation. The growthof P. atrovenetum and P. oxalicum was com-

pletely inhibited at 10 and 40 ,g/ml, respec-

tively. At 10 ug/ml, the yield of P. oxalicum wasreduced to 50%.

Pyrrolnitrin inhibited both endogenous andexogenous respiration in S. cerevisiae at verylow concentrations. The inhibition of respirationincreased with increasing concentrations ofpyrrolnitrin and was almost complete at 15,ug/ml (Table 1). Pyrrolnitrin also inhibited theproduction of '4CO2 from 14C-glucose, and theCO2 production decreased with increasingpyrrolnitrin concentrations (Fig. 3). Respirationof cell-free extracts of the yeast was also in-hibited by pyrrolnitrin. In these extracts thesuccinate oxidase activity more than doubled byexogenous cytochrome c (Table 2). The oxidationof succinate, with or without exogenous cyto-chrome c, was progressively inhibited with

c

0

i-CC

G)

U-

75

50

25

0

U '1 6 6 1

i

I0 2 4 6 8 1C

Concentration of PyrroIitrin, p.g/mlFio. 2. Inhibition of growth of S. cerevisiae by

pyrrolnitrin. The growth was measured turbidimetricallyat 650 nm after 24 hr ofincubation.

312 J. BAcrERioL.

on April 30, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

ACTION OF PYRROLNITRIN

TABLE 1. Inhibition of respiration of S. cerevisiaeby pyrrolnitrina

Amt ofpyrrolnitrin

Ag/mI025

1015202550

Exogenousrespiration

02 uptakeper mg(dr

weight)per hr

diters1007939185S44

Inhibitio

021618295959696

Endogenousrespiration

Os uptakeper mg

n (dry Inhibitionweirghtper hr

pliters63563623510O

I

011436493100100100

a Warburg flasks contained sodium phosphatebuffer (pH 7.4), 50 pmoles; glucose, 200 pmoles(omitted from flasks used for endogenous respira-tion); magnesium chloride, 35 umoles; alcohol oralcoholic solutions of pyrrolnitrin, 0.2 ml; anddistilled water to make the volume 2.6 ml. A cellsuspension of 0.4 ml (1.5 mg, dry weight) wasadded in the side arm of each flask.

increasing pyrrolnitrin concentrations until about90% of the succinate oxidation was inhibited at25 ,ug/ml.

Respiration of mitochondria was also pre-vented by pyrrolnitrin. In yeast mitochondria,the antibiotic inhibited the activities of succinateand NADH oxidase 90% at 25 and 10 ,ug/ml,respectively. This inhibition took place during10 min of incubation. The activities of theseoxidases were inhibited by antimycin A andsodium azide, indicating a normal terminalelectron transport pathway (Table 3). TheNADHoxidase was also inhibited 50% by 9 mm amytal.The NADH- and succinate-cytochrome c

reductases were inhibited by pyrrolnitrin bothin beef heart (Fig. 4) and yeast mitochondria(Table 4) immediately after adding the anti-biotic. In yeast mitochondria, pyrrolnitrin almostcompletely inhibited the activities of these en-zymes at 10 ug/ml and inhibited them about 60%at 5 jug/ml. The immediate inhibition of cyto-chrome c reductase at low concentrations ofpyrrolnitrin again suggested that electron trans-port is the primary site of action and that thesite of action is before cytochrome c.

In our studies, yeast mitochondria lost mostof their activity within 6- hr of isolation. There-fore, to further delineate the site of action ofpyrrolnitrin, beef heart mitochondria, whichretain the activity for months, were used. In

these mitochondria, cytochrome oxidase ac-tivity was not inhibited, even at 100 ,ug of anti-biotic per ml, whereas sodium azide (10 mM)completely inhibited oxygen consumption. Arti-ficial electron acceptors and antimycin A were

1i1 1 1 10 10 20 30 40 50

Concentration of Pyrrolnitrin, /ug/mlFio. 3. Inhibition by pyrrolnitrin of CO2 production

from i4C-glucose by S. cerevisiae. The reaction mixturecontained sodium phosphate buffer (pH 7.4), 50 umoles;magnesium chloride, 20,moles; 14C-UL-glucose, 20,umoles; and alcohol or pyrrolnitrin solution. The finalvolume of the mixture was 1.5 ml.

TABLE 2. Inhibition of succinate oxidation of cell-free extracts of S. cerevisiae by pyrrolnitrina

Amt ofpyrrolnitrin

pg/mi025

1015202550

Without addedcytochrome c

Amt of 02per mg ofproteinper hr

$diters312819129622

Inhibition

011396171819292

With cytochrome c

Amt of 0Oper mg ofproteinper hr

pliters7156382513855

Inhibition

021476581899393

a Contents in the Warburg flasks were the sameas described for Table 2, except that 0.4 ml of cell-free extract was added in the side arm and 2 mg ofcytochrome c was added when desired.

VOL. 100, 1969 313

on April 30, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

TRIPATHI AND GOTITLIEB

TABLE 3. Inhibition ofsuccinate oxidase and NADHoxidase activities of yeast mitochondria by

pyrrolnitrina

Succinate oxidase NADH oxidase

Inhibitor Amt of 02per mg of Inhibition AA/minb Inhibitionproteinper hr

juliters % %Pyrrolnitrin

(pg/ml)0 99 0 0.92 05 73 38 0.56 4010 38 60 0.12 8715 27 72 0.02 9825 9 92 0.01 9950 1 100 0.02 98

Antimycin A 0 100 0 100(20 pg/ml)

Sodium azide 2 99 0 100(8 mM)

Amytal (8 Not Not 0.47 49mM) used used

a For succinate oxidase, the Warburg flasks con-tained sodium phosphate buffer (pH 7.4), 200pAmoles; sodium succinate, 100,umoles; cytochromec, 2 mg; and mitochondrial preparation equivalentto 3.0 mg of protein. The control flasks contained95% alcohol corresponding to the volume of thepyrrolnitrin solution.

b For NADH oxidase, the reaction mixture con-tained sodium phosphate buffer, 50 pmoles; mito-chondrial preparation equivalent to 300 p&g ofprotein; NADH, 0.1 pmole; and water to make thevolume 1 ml. The reaction was started by addingNADH solution. The oxidation of NADH wasmeasured by the decrease in absorbancy at 340 nm.

used to determine whether pyrrolnitrin inhibitsthe electron transport prior to the antimycin-sensitive site. The reduction of DPIP was insen-sitive to antimnycin A. In contrast, this reductionwas progressively inhibited by increasing pyr-rolnitrin concentrations (Fig. 4). About 50% ofDPIP reduction was inhibited at 5 ,ug of pyr-rolnitrin per ml, and about 80% was inhibitedat 20 ,ug of pyrrolnitrin per ml. Beef heart mito-chondria reduced the tetrazolium dyes NBTand INT. Antimycin A, at 25 ,g/ml, had noeffect on this succinate-NBT reductase (Fig. 4)but inhibited succinate-INT reductase by 75%,indicating that NBT was reduced before andINT after the antimycin-sensitive site. Pyrrol-nitrin inhibited the reduction of NBT in boththe presence or absence of antimycin A. TheNBT-reductase activity was inhibited 60, 80,and 90% at 25, 50, and 100 pg of pyrrolnitrinper ml, respectively. At the same concentrations,

pyrrolnitrin inhibited the reduction of INT 27,38, and 61%, respectively.The results with artificial electron acceptors

suggested that the site of pyrrolnitrin action wasbefore the antimycin-sensitive site of the mito-chondrial respiratory chain. To determinewhether this site was before coenzyme Q, theeffect of the antibiotic on succinate-coenzymeQ6 reductase was studied. This coenzyme Qreductase activity was inhibited 50% by 10 mmthenoyltrifluoroacetone, the inhibitor of succinicdehydrogenase, and 90% by 50 Mig of pyrrolnitrinper ml. The inhibition of coenzyme Q6 reductaseincreased with increasing pyrrolnitrin concen-trations (Table 5). The inhibition of the electron

5 10 15 20 25 30 35 40 1900PYRROLNITRIN,,t&g/ml

FIG. 4. Inhibition ofNADH-cytochrome c reductase(I), succinate-cytochrome c reductase (II), succinate-DPIP reductase (III), and succinate-NBT reductase(IV) of beef heart mitochondria by pyrrolnitrin. Thefollowing reaction mixtures were usedfor these enzymeassays. (I) NADH-cytochrome c reductase: phosphatebuffer (pH 7.4), 50 pumoles; sodium azide, 5 pumoles;cytochrome c, 0.7 mg; mitochondria equivalent to 100pg ofprotein; NADH, 0.2 umole; and water to makethe volume 1.0 ml. (II) Succinate-cytochrome c reduc-tase: same as in NADH-cytochrome c reductase exceptthat 20 Amoles of sodium succinate was used instead ofNADH, and 300 pAg of mitochondrial protein. (III)Succinate-DPIP reductase: phosphate buffer (pH 7.4),50umoles; ethylenediaminetetraacetic acid, 1 ,umole;sodium succinate, 30 umoles; DPIP, 0.05 pmole; mito-chondria equivalent to 300 pug ofprotein; and water tomake the volume 1.0 ml. (IV) Succinate-NBT reductase:phosphate buffer (pH 7.4), 50 pmoles; sucrose, 25pmoles; sodium succinate, 30 pmoles; antimycin A,when added, 10 pg; mitochondria equivalent to 600 pugof protein; NBT, 1.0 pmole; and water to make thevolume 1.0 ml. The reduction of NBT was measuredspectrophotometrically at 530 nm.

314 J. BAerLqtioL.

on April 30, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

ACTION OF PYRROLNITRIN

TABLE 4. Inhibition ofNADH- and succinate-cyto-chrome c reductase of yeast mitochondria by

pyrrolnitrin

NADH-cytochrome c Succinate-cytochromeAmt of reductase' c reductaseb

pyrrolnitrinAA/minc Inhibition AA/min. Inhibition

pg/ml% %0 0.50 0 0.48 02 0.41 18 0.32 335 0.21 58 0.18 607 0.10 80 0.12 7510 0.01 98 0.01 9815 0.0 100 0.01 9820 0.0 100 0.0 100

aThe reaction mixture for succinate-cyto-chrome c reductase contained sodium phosphatebuffer (pH 7.4), 50 umoles; sodium azide, 5 ,moles;cytochrome c, 0.7 mg; mitochondrial preparationequivalent to 300 ug of protein; and water to makethe volume 0.95 ml. The reaction was initiated byadding 0.05 ml of sodium succinate (20 1umoles).bThe conditions for the assay of NADH-cyto-

chrome c reductase were the same except that 0.2pmole of NADH was added in place of succinate,and the amount of mitochondrial preparation wasreduced to correspond to 100 &g of protein.

c Absorbancy was measured at 550 nm.

transfer from succinate to coenzyme Q couldbe due to the interference of pyrrolnitrin withsuccinic dehydrogenase which has FAD as itscoenzyme. The ability of pyrrolnitrin to bindFAD was therefore studied by measuring theabsorption spectrum of both FAD and pyr-rolnitrin singly and in a mixture of the two com-pounds. The criteria for binding of pyrrolnitrinto FAD, or vice-versa, were any changes in theabsorption spectrum of either FAD or pyr-rolnitrin. FAD had peaks at 375 and 445 nm.Pyrrolnitrin had a single absorption peak at 250nm. In the mixture of two compounds, theseabsorption spectra were not altered.

Pyrrolnitrin also inhibited the synthesis ofvarious macromolecules. The antibiotic had noinhibitory effect on the incorporation of 14C-glucose into the 80% ethyl alcohol extract (aminoacids, soluble sugars, etc.) or the 5% cold tri-chloroacetic acid extract (nucleotides, smallpolysaccharides, etc.). The synthesis of nucleicacids and protein from "4C-glucose were inhibitedby pyrrolnitrin (Table 6). Studies with 8H-thymidine, 14C-uracil, and "4C-amino acids con-firmed that the antibiotic inhibited the synthesesof DNA, RNA, and protein. For example, at50 ,ug/ml, pyrrolnitrin inhibited the incorporationof 8H-thymidine into DNA by 76%; the incor-

poration of "4C-uracil into RNA and that of"4C-amino acids into protein were both inhibitedby 100% (Table 7).

Despite inhibition by pyrrolnitrin of proteinsynthesis in whole cells, the antibiotic had noeffect on in vitro protein synthesis in R. solanior in E. coli. In R. solani, the poly U-directedincorporation of phenylalanine was inhibitedby ribonuclease and puromycin, a characteristicof the typical in vitro protein synthesis system.

TABLE 5. Inhibition of succinate-coenzyme QG re-ductase of beef heart mitochondria by pyrrol-

nitrina

Inhibitor Absorbancy Inhibition

pg/mI %Pyrrolnitrin

0 0.401 05 0.369 810 0.320 2015 0.312 2220 0.244 3925 0.196 5150 0.040 9075 0.039 90100 0.039 90

Thenoyltrifluoroacetone 0.195 49(10 mM)a The assay mixture contained the following, in

glass-stoppered tubes: sodium phosphate buffer(pH 7.4), 100 pmoles; sodium azide, 5 Mmoles;sodium succinate, 50 Amoles; coenzyme Q, 0.4 mgas an alcoholic solution; ethyl alcohol or pyrrol-nitrin solution; 0.25 M sucrose to a total of 3.0 ml.Absorbancy was measured at 519 nm.

TABLE 6. Effect ofpyrrolnitrin on the metabolism of"4C-glucose by S. cerevisiae"

Per cent inhibition of

Amt of Hot tri-pyrrolnitrin chlor-

Uptake of Production oacetic Protein1C-glucose of 14CO2 acid-

solublefractionb

8g/ml0 0 0 0 010 7 12 13 225 14 25 25 550 27 56 75 55

100 75 86 95 96

a Exponential-phase cells were incubated inGYE medium containing "4C-glucose. The dis-appearance of the radioactivity from the mediumwas a measure of uptake of "4C-glucose. Themacromolecules were obtained as described.

I Nucleic acid.

VOL. 100, 1969 315

on April 30, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

TRIPATHI AND GOTFLIEB

But pyrrolnitrin, even at 40 pg/ml (20 pg perassay), did not inhibit the phenylalanine incor-poration. Instead, the antibiotic stimulated thisincorporation about 80% (Table 8). In vitroprotein synthesis in an E. coli system directed byturnip yellow mosaic virus genome was alsoinsensitive to pyrrolnitrin. Even at 125 pg/ml (the

TABLE 7. Effect ofpyrrolnitrin on the synthesis ofRNA, DNA, and protein by S. cerevisiaes

Per cent inhibition of

tf pDNA RNA Protein

rolnitrinUptake Incor- U take Incor- Uptake Incor-of 'II- poration Jf 14C. poration of 4C. porationthymi- into 1 into amino intodine DNA uracil RNA acids protein

0 0 0 0 0 0 05 28 29 17 10 23 3110 37 43 21 23 41 5620 44 57 48 62 67 9150 64 76 89 100 92 100

a Exponential-phase, washed cells were incu-bated for 2 hr in GYE medium containing the ap-propriate tracer. The uptake of the tracer wasmeasured by its disappearance from the medium.The cells were fractionated as described.

TABLE 8. In vitro protein synthesis in Rhizoctoniasolani in the presence of pyrrolnitrin

Assay condition Counts per minper alsayl

Complete systemO min . . . 6860min.2,590plus pyrrolnitrin (20 Ag).4,417plus puromycin (50 pg) 558plus ribonuclease (30 jig) 478minus ribosomes.389minus supernatant fraction 257minus poly U ..157

* All of the data are from the 60-min incubation,unless otherwise indicated. The complete assaysystem, in 0.5 ml, contained Tris(hydroxymethyl)-aminomethane-hydrochloride buffer (pH 7.8), 50pmoles; magnesium acetate, 10 ;moles; NH4C1, 25pmoles; 2-mercaptoethanol, 7.5 ;moles; reducedglutathione, 0.5 ;pmole; ATP, 1.5 ;moles; guano-sine triphosphate, 0.1 pmole; phosphoenolpyru-vate, 2.5 ;moles; pyruvate kinase, 10 Ag; yeasttransfer RNA, 2001g; poly U, 60 pg; 14C-L-phenyl-alanine, 0.3 pc; 19 other 12C-amino acids, 0.005pmole each; 105,000 X g supernatant fractionequivalent to 0.44 mg of protein; ribosomesequivalent to 0.29 mg of RNA; pyrrolnitrin, whenadded, 20 ;g.

minimal growth inhibitory concentration forE. coli was 100 pg/mi), pyrrolnitrin had noeffect on this protein synthesis (Table 9). How-ever, no stimulation was observed in the E. colisystem.The antibiotic inhibited the uptake of glucose,

uracil, thymidine, and amino acids (Table 6 and7) by S. cereviseae. This effect was not due to abreakdown of general cell permeability, sincethere was no leakage of phosphate, ninhydrin-positive materials, reducing sugars, nucleotides,etc.

DISCUSSIONSince growth of all organisms is the result

of complex, integrated, and interdependentprocesses, antibiotics affecting any one cellularprocess would ultimately impair most of theother cellular functions. The important questionin the study of the mode of action of an antibioticis the primary site of action. The reaction whichis inhibited at the lowest concentration andshortest time is usually accepted as the primarysite of action. For pyrrolnitrin, these criteria weresatisfied by the inhibition of respiration, sincevery low concentrations completely inhibitedthe respiration in whole cells, cell-free extracts,and mitochondrial preparations. Whole cellrespiration, as well as mitochondrial respiratoryenzymes, were rapidly inhibited, within 5 to 10min. The inhibition of syntheses of RNA, DNA,and proteins, on the other hand, required higherconcentrations of pyrrolnitrin and much longertimes, more than 1 hr.The locus of inhibition of cellular respiration

was the mitochondrial electron transport system,since pyrrolnitrin inhibited the activities ofsuccinate and NADH oxidase of isolated mito-chondria. In this electron transport system, thesite of inhibition by pyrrolnitrin was beforecytochrome c because pyrrolnitrin inhibitedtheNADH- and succinate-cytochrome c reductasewithout any inhibitory effect on cytochrome c

TABLE 9. In vitro protein synthesis of Escherichiacoli system in the presence of pyrrolnitrin

Assay condition Counts per minper assay

Complete system0 min ........................... 69760 min .......................... 140,692minus viral genome' .. . 3,714plus pyrrolnitrin (25 pg/ml) ....... 145,222plus pyrrolnitrin (125 pg/ml) .1.5.3.153,214

Background....................... 46

a Turnip yellow mosaic virus.

316 J. BAcTERioL.

on April 30, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

ACTION OF PYRROLNITRIN

oxidase. Since pyrrolnitrin inhibited the anti-mycin-insensitive reduction of DPIP and NBT,this inhibition must have been prior to the anti-mycin-sensitive site in the electron transportsystem. The inhibition by the antibiotic of INTreduction, which was antimycin-sensitive, wouldbe expected since the INT accepted electronsafter the antimycin-sensitive site. Lester andSmith (13) also found that the reduction of NBTin beef heart mitochondria is prior to and INTafter the antimycin-sensitive site. According tothese authors, the site of NBT reduction is ator before the site of coenzyme Q (13). Sincepyrrolnitrin inhibited the NBT reduction, thisinhibition should be at or before coenzyme Q inthe respiratory chain. The inhibition of succinate-coenzyme QB reductase by pyrrolnitrin suggeststhat the antibiotic inhibited the electron transferbetween succinate and coenzyme Q (Fig. 5).

This inhibition of respiratory electron transportwould also block adenosine triphosphate (ATP;energy) formation (16). Since energy is requiredfor the active uptake of various metabolitesand for the biosynthesis of macromolecules(16), pyrrolnitrin should ultimately affect theseenergy-dependent processes. The data fit thisconcept which explains the fact that pyrrolnitrininhibits the synthesis of RNA, DNA, and proteinand that the antibiotic also reduces the uptakeof amino acids, uracil, and thymidine. The effectof pyrrolnitrin on the biosynthesis of macro-molecules could be either on the incorporationof the precursor into the macromolecules or onthe inhibition of uptake of precursors. Probablyit is uptake that is prevented because pyrrolnitrinhad no effect on the in vitro protein-synthesizingsystems of R. solani and E. coli. Based on thesefacts, we conclude that the primary site of actionof pyrrolnitrin is on the respiratory electrontransport system.

After submission of this paper, an article onthe mechanism of action of pyrrolnitrin by Noseand Arima was published (19). Their data gen-erally agree with ours in showing that pyrrol-nitrin inhibits growth, synthesis of protein, RNA,DNA, and uptake of metabolites. In our experi-ments on S. cerevisiae, pyrrolnitrin did notstimulate leakage of 260-nm absorbing materials,sugars, and phosphate, whereas, in Nose andArima's work on Candida utilis, the antibioticcaused a leakage of 260-nm absorbing materials.Besides the different organisms used in the twostudies, the age of the cells used in these studiesalso differed. We used late exponential-phasecells, whereas Nose and Arima used early expo-nential-phase cells. The results on oxygen con-sumption also differed. Pyrrolnitrin, at 50 Ag/ml,almost completely inhibited the oxygen uptake

NADH, -- FD PN Co Q AfS>i< >_Cyt c,-Cytc-.Cyta

Succinate_-FS Cyt b ICyt a3

I

FIG. 5. Components and their sequence in therespiratory electron transport system of mitochondria(12) and the most likely site ofpyrrolnitrin action (PN).Abbreviations: FD, NADH-dehydrogenaseflavoprotein;FS, succinate dehydrogenase flavoprotein; Co Q, co-enzyme Q; Cyt, cytochrome; ASS, antimycin-sensitivesite.

by S. cerevisiae in our experiments, whereasNose and Arima reported that in C. utilis theinhibition was transitory. The oxygen uptake inthis yeast was completely inhibited for 10 minand then the yeast began to consume oxygen butnever at the rate of the control. At 20 and 30min, the inhibition was still about 90 and 50%,respectively. Nose and Arima attribute theantibiotic action of pyrrolnitrin to the permea-bility because of the deleterious effect of theantibiotic on bacterial protoplasts. On the otherhand, our data with S. cerevisiae whole cells,cell-free extracts, and mitochondria, as well asbeef heart mitochondria, all indicate a directeffect on aerobic respiration.

ACKNOWLEDGMENT

We thank Paul D. Shaw and Tom Obrig, Department ofPlant Pathology, and John Clark, Jr., Department of Chemistry,University of Illinois, for valuable assistance.

LITERATURE CITED

1. Arima, K., H. Imanaka, M. Kousaka, A. Fukuda, and G-Tamura. 1964. Pyrrolnitrin, a new antibiotic substanceproduced by Pseudomonas. Agr. Biol. Chem. 28:575-776.

2. Arima, K., H. Imanaka, M. Kousaka, A. Fukuda, and G.Tamura. 1965. Studies on pyrrolnitrin. I. Isolation andproperties of pyrrolnitrin. J. Antibiotics (Tokyo) Ser. A18:201-204.

3. Ashwell, G. 1957. Colorimetric analysis of sugars, p. 87-88.In S. P. Colowick and N. Kaplan (ed.), Methods in en-zymology, vol. 3. Academic Press Inc., New York.

4. Chen, P. S., T. Y. Toribara, and H. Warner. 1956. Micro-determination of phosphorus. Anal. Chem. 28:1756-1758.

5. Clark, J. M., Jr., A. Y. Chang, S. Spiegelman, and M. E.Reichman. 1965. The In vitro translation of monocistronicmessage. Proc. Nat. Acad. Sci. U.S.A. 54:1193-1197.

6. Crane, F. L., J. L. Glenn, and D. E. Green. 1956. Studieson the electron transfer system. IV. The electron transferparticle. Biochim. Biophys. Acta 22:475-487.

7. Gornall, A. G., C. J. Bardawill, and M. A. David. 1949.Determination of serum proteins by means of the biuretreaction. J. Biol. Chem. 177:751-766.

8. Gottlieb, D. 1968. Antibiotics and cell metabolism. HindustanAntibiot. Bull. 10:123-134.

9. Gottlieb, D., H. E. Carter, J. H. Sloneker, L. Chi Wu, andE. Gaudy. 1961. Mechanism of inhibition of fungi byfilipin. Phytopathology 51:321-330.

317VOL. 100, 1969

on April 30, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

TRIPATHI AND GOTITLIEB

10. Gottlieb, D., and R. K. Tripathi. 1968. The physiology ofswelling phase of spore germination in Penicillium atro-venetum. Mycologia 60:571-590.

11. Imanaka, H., M. Kousaka, G. Tamura, and K. Arima. 1965.Studies on pyrroinitrin. II. Taxonomic studies on pyrrol-nitrin producing strain. J. Antibiotics (Tokyo) Ser. A18:205-206.

12. Inoue, Y., and D. Gottlieb. 1967. Mechanism of action offlavensomycin on Penicillium oxalicun. AntimicrobialAgents and Chemotherapy-1966, p. 470-479.

13. Lester, R. L., and A. L. Smith. 1961. Studies on the electrontransport system. XXVIII. The mode of reduction oftetrazolium salts by beef heart mitochondria. Role ofcoenzyme Q and other lipids. Biochim. Biophys. Acta47:475-96.

14. Lively, D. H., M. Gorman, M. E. Haney, and J. A. Mabe.1967. Metabolism of tryptophans by Pseudomonas aurco-faciens. I. Biosynthesis of pyrrolnitrin. AntimicrobialAgents and Chemotherapy-1966, p. 462-469.

15. Lowry, 0. H., N. J. Rosebrough, A. L. Farr, and R. J.Randall. 1951. Protein measurement with the Folin phenolreagent. J. Biol. Chem. 193:265-275.

16. Mahler, H. R., and E. H. Cordes. 1968. Basic biologicalchemistry. Harper & Row, New York.

17. Moore, S., and W. H. Stein. 1948. Photometric ninhydrinmethod for use in chromatography of amino acids. J.Biol. Chem. 176:367-388.

18. Morris, D. L. 1948. Quantitative determination of carbo

hydrates with Dreywoods anthrone reagent. Science107:254-255.

19. Nose, M., and K. Arima. 1969. On the mode of action of anew antifungal antibiotic, pyrrolnitrin. J. Antibiotic(Tokyo) Ser. A 22:135-143.

20. Obrig, T. G., J. Cerna, and D. Gottlieb. 1969. Characteristicsof in vitro protein synthesis systems from Rhizoctoniasolani and Sckerotium bataticola. Phytopathology 59:187-221.

21. Ramasarma, T., and R. L. Lester. 1960. Studies on the elec-tron transport system. XXIV. The reduction and oxidationof exogenous coenzyme Q. J. Biol. Chem. 235:3309-3314.

22. Schmidt, G., and S. J. Thannhauser. 1945. A method for thedetermination of desoxyribonucleic acid, ribonucleic acidand phosphoproteins in animal tissues. J. Biol. Chem.161:83-89.

23. Schneider, W. C. 1957. Determination of nucleic acids intissues by pentose analysis, p. 680-681. In S. P. Colowickand N. 0. Kaplan (ed.), Methods in enzymology, vol. 3.680-81. Academic Press Inc., New York.

24. Slater, G. G., E. Geller, and A. Yuwiler. 1964. Intrascintilla-tion vial reaction tube. Anal. Chem. 36:1888.

25. Umbreit, W. W., R. H. Burris, and J. F. Stauffer. 1964.Manometric techniques. Burgess Publishing Co, Min-neapolis.

26. Utter, M. F., D. B. Keech, and P. M. Nossal. 1958. Oxidativephosphorylation by subcellular particles from yeast. Bio-chem. J. 68:431-440.

318 J. BACTERIOL.

on April 30, 2020 by guest

http://jb.asm.org/

Dow

nloaded from