5-Fluoropyrimidine-Resistant Mutants of Pneumococcus · 5-Fluoropyrimidine-Resistant Mutantsof Pneumococcus BARRYBEANAND ALEXANDERTOMASZ TheRockefeller University, New York, NewYork

JOURNAL OF BACTERIOLOGY, Mar. 1973, p. 1348-1355Copyright 0 1973 American Society for Microbiology

Vol. 113, No. 3Printed in U.S.A.

5-Fluoropyrimidine-Resistant Mutants ofPneumococcus

BARRY BEAN AND ALEXANDER TOMASZThe Rockefeller University, New York, New York 10021

Received for publication 3 October 1972

Three classes of 5-fluoropyrimidine-resistant mutants of Diplococcuspneumoniae have been characterized. The mutant strain upp is resistant to highconcentrations of the fluoropyrimidine bases fluorouracil (FU) and fluorocyto-sine (FC); strain upp has a defective uridine monophosphate pyrophosphoryl-ase. The mutant strain udk is resistant -to inhibition by fluorouridine (FUR)and exhibits defective uridine kinase activity. The mutant strain fun is resistantto inhibition by the nucleosides fluorodeoxyuridine, fluorodeoxycytidine, andFUR, but shows normal activity for all pyrimidine pathway enzymes tested.This strain may be defective in the activity of a transport system that governsthe cellular uptake of pyrimidine ribo- and deoxyribonucleosides. Biochemicalstudies on wild-type and fluoropyrimidine-resistant pneumococci are discussedwith respect to the transport and early metabolism of preformed pyrimidineprecursors by this organism.

In a previous communication, we reportedthat pneumococci respond in an unusual man-ner to treatment with various 5-fluoropyrimi-dines: fluorouracil (FU), fluorouridine (FUR),and fluorodeoxyuridine (FUdR), each causingdifferent patterns of inhibition (2). It wassuggested that the biochemical basis of thisobservation is the relative metabolic stability ofthe N-glycosidic bond of pyrimidine nucleo-sides, which, in turn, is due to the absence, orvery low activity, of enzymes that might cleavethis bond (nucleoside phosphorylase, hydro-lase, and N-transdeoxyribosidase).

In order to help elucidate the pathways forthe utilization of preformed pyrimidine com-pounds in this organism, we have isolated andcharacterized a series of fluoropyrimidine-resistant (FPr) mutants of this species.

MATERIALS AND METHODSBacterial strains and culture methods. The

wild-type and Ua (pyrimidine-requiring) strains ofDiplococcus pneumoniae R36A, clone R6, used inthese experiments and the methods used in theirpropagation have been previously described (2).Cells were cultured in the previously describedchemically defined basal medium (2) or, for some ofthe experiments reported here, in a casein-hydroly-sate medium (12) modified by the omission of uridineand adenosine and buffered with 0.05 M potassiumphosphate buffer at pH 8.0. In the experiments pre-sented here, there were no important differences be-tween these two media.

Genetic methods. The three new mutants de-scribed were selected as single-colony isolates arisingin a single step as spontaneous mutants of parentalstrain R6 carrying the b and d markers for sul-fonamide resistance. For the selection of each, mediawere supplemented as follows: FUr, 100 lAg of FUper ml and 20 Mg each of uridine (UR) and adenosineper ml. FURr, 10 lsg ofFUR per ml, and FUNr, 50 ,g ofFUdR, 100 ug of thymidine (TdR), and 20 lAg ofadenosine per ml.

Competent recipient R6 or Ua cells were preparedas previously described (11, 12). Transforming deox-yribonucleate was prepared, and genetic transforma-tion was carried out by the methods of Hotchkiss (6,7). Transformants to drug resistance were scored inthe presence of the following supplements: (i) dmarker for sulfonamide resistance, 40 Ag of sulfanila-mide per ml; (ii) FUr, 10 Mg of FU plus 0.1 MAg of URper ml; (iii) FURr, 5 gg of FUR per ml; and (iv)FUNr, 0.1 Mg of FUdR or 1 Mg of FUR per ml.

Studies on the utilization of radioactiveprecursors. Labeling conditions, filtration methodsfor the determination of intracellular isotope distri-bution, methods for compositional analysis of thesoluble pool by trichloracetic acid extraction andthin-layer chromatography (TLC) previously havebeen described (2).

In vitro enzyme studies. The preparation of crudeenzyme extracts, the assay of nucleoside phosphoryl-ase (EC 2.4.2), nucleoside hydrolase (EC 3.2.2.3),and nucleoside trans-N-deoxyribosidase (EC 2.4.2.6)activities were carried out as described previously(2).The complete uridine kinase (EC 2.7.1.48) assay

mixture contained, for each 0.1 ml of the final348

on January 20, 2021 by guesthttp://jb.asm

.org/D

ownloaded from

PNEUMOCOCCUS MUTANT RESISTANCE

volume: MgCl2, 1 umol; tris(hydroxymethylXamino-methane (Tris) buffer, pH 8.0, 10 Mmoles; adenosinetriphosphate, 1 Mmol; UR-2- 14C, 0.05 gCi/0.8nmoles; and 10 Mliters of enzyme extract (containing20 to 50 ug of protein). After incubation at 37 C, for30 or 60 min, samples were applied directly to TLCsheets for analysis or diluted into 5 ml of water (pH7.0), or both, for analysis by nucleotide adsorption todiethylaminoethyl-cellulose disks by a gravity flowmethod based on the procedure of Furlong (5). Underthese conditions, 7 to 15% of the uridine availablewas phosphorylated to a mixture of uridine mono-

phosphate (UMP), uridine diphosphate (UDP), anduridine triphosphate (UTP), indicating the presenceof active uridylate kinase in the pneumococcal ex-

tracts.By using the same assay system, but substituting

UMP-2-14C (0.02 gCi/0.11 Amol) for uridine, theuridine nucleotide kinases (EC 2.7.4.4 and EC2.7.4.6) could be assayed independently of uridinekinase. With UMP as substrate, a small amount ofnucleotidase activity for dephosphorylation of UMPto UR could also be detected in all pneumococcalstrains used. By omitting adenosine triphosphatefrom the assay mixture, the nucleotidase (EC 3.1.3.5)could be assayed without competition by any kinaseactivity.Attempts to assay for UdR/TdR kinase (EC

2.7.1.21) by substitution of these substrates in theuridine kinase assay system described above and inthe assay system of Okazaki and Kornberg (10) were

unsuccessful. Many additional variables were intro-duced into both assay systems, but it was notpossible to detect UdR or TdR kinase activities inpneumococcal extracts.

Thymidylate synthetase (EC 2.1.1.6) was assayedby the method of Wahba and Friedkin (13).

Attempts were made to assay possible nucleosidephosphotransferase (EC 2.7.1) activity by using a

procedure similar to that of Brawerman and Chargaff(4). The assay mixture contained, in a total volumeof 250 Mliters: UR-2-'4C, 0.25 uCi/1.0 Ag, or UdR-2-14C, 0.25 MCi/2.2 Mg; potassium acetate buffer, pH5.0, 50 MM, or Tris-maleate buffer, pH 7.9, 50 MM;phosphate donor to be tested, 1 MM; and 20 ,liters ofenzyme extract. Mixtures were incubated for 15 h at37 C and sampled for analysis by TLC. Neither inthis assay system nor in the UdR kinase assay systemwere any of the common nucleoside mono-, di-, or

triphosphates, sugar phosphates, or phosphoenol-pyruvate active as phosphotransferase donors.The UMP pyrophosphorylase (UMP:pyrophos-

phate phosphoribosyltransferase; EC 2.4.2.9) assaymixture contained in a final volume of 50 Mliters:MgCl2, 0.5 Mmol; Tris-hydrochloride buffer, pH 8.0,5 MM; 5-phosphoribosyl-1-pyrophosphate, 0.5 MM;uracil-2- 4C, 0.2 MCi/.007 MM; and enzyme extractcontaining 10 to 50 Mg of protein. After addition ofthe enzyme, assays were incubated at 37 C for 30min and chilled, and 6- to 10-Mliter samples were

applied directly to TLC sheets for chromatographicanalysis.

RESULTS5-Fluoropyrimidine resistant (FPr) mu-

tants. Some of the physiological and geneticproperties of the Fpr mutants are listed inTables 1 and 2, respectively. Cultures of thefluoropyrimidine-resistant mutants grew withnormal growth rates and to normal stationary-phase cell concentrations both in the presenceand absence of the appropriate drug. There wasno evidence for new nutritional requirementsaccompanying acquisition of fluoropyrimidineresistance.The data in Table 1 indicate that the FUr,

FURr, and FUNr mutant strains fall into threedistinct classes. The FUr mutant was cross-resistant to fluorocytosine (FC). The FUNrstrain was resistant to both FUR and FUdR,whereas the FURr strain showed no cross-resistance to FUdR.

In the transformation experiment listed inTable 2, genetically competent strain R6 wasused as recipient for deoxyribonucleic acid(DNA) purified from each of the three Fprstrains. Since each of the mutant strains alsocarries the d marker for sulfonamide resistance,transformation for this marker was also scoredas an internal control. The transformed cloneswere homogeneous for their resistance proper-ties. The results presented in Table 2 on theapparent spontaneous mutation rate to drugresistance and transformation frequencies forthe individual fluoropyrimidine resistancemarkers indicate that each of these traits is theresult of a single mutation.Biochemical characterization of Fpr mu-

tants. Table 3 reports some representativedata for the incorporation of natural pyrimidineprecursors and the fluoropyrimidine analogs bycultures of wild-type pneumococci and the Fpr

TABLE 1. Fluoropyrimidine concentrations requiredfor inhibition of wild-type and fluoropyrimidine

resistance mutantsa

Inhibitory concn (gg/ml)Strain

FU FUR FUdR FCdR FC

Wild type 0.005 0.1 0.05 7 5Fur >1,000l 0.1 0.05 7 > 400kFURr 0.005 100 0.05 7 15FUNr 0.005 5 5 40 5

a Scored as minimal concentration necessary toprevent colony growth from low inocula in the basalmedium anti-R system after 37 C incubation for 2days.

"These results were scored under the same condi-tions, except that media were supplemented with0.1 gg of UR per ml, which was found advantageousfor optimal expression of resistance by the FUrstrain.

VOL. 113, 1973 1349

on January 20, 2021 by guesthttp://jb.asm

.org/D

ownloaded from

BEAN AND TOMASZ

mutants. Each of the mutant strains showed acharacteristic defect in the utilization of cer-tain precursors.The FUr mutant had a defect in the utiliza-

tion of uracil and fluorouracil, the FURr mu-

tant showed abnormal incorporation of bothuridine and fluorouridine, and the FUNr strainshowed a generalized poor uptake for all thepyrimidine ribo- and deoxyribonucleosidestested, including UR, FUR, UdR, FUdR, deox-ycytidine (CdR), and TdR.Tables 4 and 5 present data on the nature of

the soluble-pool compounds derived from FUand FUR in the pertinent Fpr mutant versusthe sensitive parental strain. The data showthat the FUr and FURr mutants are eachdeficient in the net synthesis of phosphorylatedderivatives (including nucleic acids) of FU andFUR, respectively. Table 6 reports the resultsof in vitro assays for the enzymes UMP pyro-phosphorylase and uridine kinase in extractsprepared from the wild-type and Fpr mutantcells. The FUr mutant has greatly decreasedspecific activity for the enzyme UMP-pyro-phosphorylase, whereas the FURr mutant isdefective for uridine kinase activity. When the

appropriate fluoropyrimidine substrates weresubstituted for the natural ones, the samerelationships between these activities were ob-served.

Similar experiments with the FUNr mutantyielded more complex results. By using lowconcentrations of exogenous nucleosides onlyvery little, if any, radioactivity could be recov-ered in the soluble pool (Table 7). With highernucleoside concentrations, total cellular in-corporation was still low; however, under theseconditions the normal phosphorylated compo-nents were represented in the cellular solublepool (Table 5). These results suggest that theFUNr mutant may have a defect in the concen-tration or retention of pyrimidine nucleosides,or both. Further support for this suggestion isshown in Table 8, which summarizes the re-sults of experiments on the concentrative up-take of various pyrimidine precursors by thewild-type and FUNr strains. The data showthat wild-type pneumococci can build up sub-stantial concentrations of some of these precur-sors in unchanged chemical form in the cellularsoluble pool. In contrast, the FUNr strain wasunable to concentrate or retain any of the

TABLE 2. Frequencies for transformation and spontaneous mutation to fluoropyrimidine resistancea

Concn of drugDonor Recipient Fpr SAr FPr/SAr Mutation to FPr used in mutant

selection (jg/ml)

Fur R6 4.3 x 10-' 1.4 x 10-2 0.31 10- 6 100FURr R6 4.0 x 10-3 1.0 x 10-2 0.40 10-6 10FUNr R6 1.6 x 10-2 2.4 x 10-2 0.67 3 x 10' 50

a Phenotypes are denoted as follows: FPr, fluoropyrimidine-resistant; FUr, fluorouracil-resistant; FURr,fluorouridine-resistant; FUNr, fluoropyrimidine-nucleoside-resistant; and SA , sulfanilamide-resistant.

TABLE 3. Utilization of pyrimidine and fluoropyrimidine precursors by FPR mutantsa

Precursor concn FU (U) FUR UR FUdR UdR CdR TdR

(Ag/ml)10.0 0.75 0.34 1.0 0.08 0.11 5.0

Wild type 100.0 100.0 100.0 100.0 100.0 100.0 100.0FUr 11.0 100.0 101.0 96.0 88.0FURr 109.0 13.0 0.6 80.0FUNr 107.0 12.0 3.0 4.0 9.0 3.0 20.0

aPrecursors labeled with radiocarbon in the 2-position of the pyrimidine ring were added to matchedmid-logarithmic-phase cultures of wild-type and Fpr mutant pneumococci of identical cell concentrations.Data given are for samples taken after 60 min of incubation in the presence of the precursor and are presentedas the relative amount of radioactivity incorporated by the wild-type (taken as 100) and mutant strains. ForFU, FUR, and UR, data are based on incorporation to nucleic acid (trichloroacetic acid-precipitable)material; for FUdR, UdR, CdR, and TdR, data are based on whole-cell incorporation. Specific activities of theprecursors, and the absolute amount of trichloroacetic acid-precipitable or whole-cell radioactivity for wildtype samples, in counts per minute per milliliter of culture were as follows: FU-2- 4C, 1.3 MCi/10 Ag, 17,000counts/min; FUR-2- "C, 0.05 uCi/0.75 Mg, 1,900 counts/min; UR-2-'4C, 0.08 MCi/0.34 Mg, 81,300 counts/min;FUdR-6-_H, 2 MCi/ Mg, 24,600 counts/min; UdR-6-'H, 2.5 ACi/0.08 Mg, 37,600 counts/min; "C-CdR (U), 0.1MCi/0.11 Mg, 13,000 counts/min; TdR-2-'4C, 0.6 MCi/5 gg, 4,800 counts/min.

1350 J. BACTrERIOL.

on January 20, 2021 by guesthttp://jb.asm

.org/D

ownloaded from

PNEUMOCOCCUS MUTANT RESISTANCE

TABLE 4. Soluble-pool compounds derived from FU-2- 4Ca

Counts per min per ml Soluble-pool contents (%)

DeterminationNucleic acid Soluble pool Free FU and monophosphate diphosphate triphosphatenucleoside region region region

Wild type 17,120 81,480 41.6 21.8 22.4 14.2FUr 6,485 26,715 90.5 3.2 4.7 1.6

a Matched cultures of wild type and FU' were supplemented with FU-2-14C to give a final concentration of1.26 MCi per 10 ,g per milliliter at a titer of 5 x 107 viable units/ml. After 60 min of incubation, cultures weresampled for determination of isotope distribution, and trichloroacetic acid extracts were prepared. Data givenfor soluble-pool contents were determined by thin-layer chromatography of the extracts on silica gel sheets insolvent system 1 (2).

TABLE 5. Soluble-pool compounds derived from FUR-2- 4Ca

Counts per min per ml Soluble-pool contents (%)

DeterminationNucleic acid Soluble pool FUR monophosphate diphosphate triphosphateregion region region

Wild type 5,525 42,400 83.2 5.1 9.4 2.3FUR' 1,730 30,500 95.4 1.6 2.3 0.7FUN' 2,460 3,400 55.5 15.9 23.2 5.4

a Matched cultures of wild type, FUR', and FUNr were supplemented with FUR-2-'4C to give a finalconcentration of 0.7 MCi per 10 Ag per ml at a titer of 5 x 107 viable units/ml. After 60 min of incubation,cultures were sampled for determination of isotope distribution, and trichloroacetic acid extracts wereprepared. Data given for soluble-pool contents were determined by thin-layer chromatography of extracts onsilica gel sheets in solvent system 1 (2).

TABLE 6. In vitro activities for uridinemonophosphate (UMP)-pyrophosphorylase anduridine (UR) kinase in wild-type and FPr mutant

pneumococcia

Determination UMP-pyrophos- UR kinasephorylase

Wild type 100.0 100.0FU' <2.0 100.0FUR' 100.0 < 1.0FUN' 100.0 100.0

a Relative enzyme activities are reported as thepercentage of the wild-type value (100%). The spe-cific activities found for extracts of the wild-typecells, in nanomoles of substrate utilized per hour permilligram of protein, were: for UMP-pyrophosphory-lase, 270 + 10%, and for UR kinase, 1.0 + 10%.

nucleosides tested, or both, but retained nor-mal amounts of the unchanged free pyrimidinebase, FU.

Several pyrimidine pathway enzymes wereassayed directly as described in Materials andMethods by using extracts of both wild-typeand FUNr cells. All of the following activitieswere present at normal levels in the FUNrextracts: (i) thymidylate synthetase, dUMP -deoxythymidine monophosphate (dTMP); (ii)nucleotidase activities for dTMP - TdR and

dUMP - UdR; (iii) kinase activities for UR ,

UMP, UMP _ UDP -_ UTP, dTMP _ deoxy-thymidine diphosphate -_ deoxythymidine tri-phosphate and deoxycytidine monophosphate_ deoxycytidine diphosphate -_ deoxycytidinetriphosphate; and (iv) UMP-pyrophosphoryl-ase.

Exhaustive attempts were made to detectenzymatic phosphorylation of UdR and TdR inboth wild-type and mutant extracts, but theymet with no success. Although UdR-TdR ki-nase is known to be an elusive and labileenzyme in other bacterial systems, numerousattempts to monitor its activity in pneumococ-cal extracts seemed advisable because (i) fromin vivo experiments it is clear that such anactivity must be present in pneumococci, (3;unpublished observations) and (ii) this activitywas considered as a possibility for the enzymeaffected by the FUNr mutation.UdR and TdR kinase activities were assayed

as described in Materials and Methods. Con-trol experiments performed with extracts ofEscherichia coli W6 demonstrated that thematerials and techniques used were quite ade-quate for routine assay of these kinase activi-ties. For assays with pneumococcal extracts,many reasonable variables were introducedincluding: (i) conditions of cell growth and

VOL. 113, 1973 1351

on January 20, 2021 by guesthttp://jb.asm

.org/D

ownloaded from

BEAN AND TOMASZ

TABLE 7. Incorporation of nucleosides by wild-typeversus FUNr a

Total incorpo- Recovered inPrecursor Strain rated (counts soluble pool

per mmn per ml) per ml)

UdR Wild type 39,100 9,340FUNr 19,100 0

FUdR Wild type 24,600 15,930FUNr 870 0

TdR Wild type 8,690 1,780FUNr 4,480 180

CdR Wild type 12,990 5,530FUNr 390 0

FUR Wild type 9,020 7,820FUNr 580 334

a Precursor concentrations and incubation periodswere as follows: UdR-6-3H, 2.5 MCi per 0.08,ug per ml,30 min; FUdR-6-3H, 2 MACi per yg per ml, 60 min;TdR-2-14C, .65 MCi per 5 ug per ml; 120 min; CdR-'4C(U), 0.1 gCi per 0.11 gg per ml; 60 min; andFUR-2-14C, 0.05 MACi per 0.75 Ag per ml; 30 min.

enzyme extraction; (ii) type, concentration,and pH of the buffer; (iii) substrate and metalion composition; (iv) inclusion of protectivereagents including thiol compounds and pro-teins; (v) time and temperature of incubation;(vi) addition of possible effectors or alternatephosphate donors, or both, including variousnucleoside monophosphates (both as activatorsand as donors for possible phosphotransferaseactivity), diphosphates, and triphosphates (in-cluding those known to be activators for theEscherichia coli enzyme) and (vii) coupling ofUdR kinase activity to that of thymidylatesynthetase.With all of these attempts, it has been

impossible to measure UdR or TdR kinaseactivity in either the wild-type or mutantpneumococci.Biochemical characterization of FUr

transformants. Competent wild-type cellswere treated with DNA prepared from FUrmutant cells, and four FU-resistant clones fromthe transformation tube were isolated andcultured in nonselective media for many gener-ations. Upon being challenged again with FU,all exhibited high-level resistance typical of thedonor. One of these transformant clones wasused for further, more detailed, characteriza-tion. This strain was designated R6FUr.Table 9 summarizes the properties of recipi-

ent, donor, and transformant strains underconsideration here. From the table it is appar-ent that the properties of the transformant aresimilar to those of the donor FUr strain with

respect to utilization of nature and fluorinatedpyrimidines. In all other respects tested, thecharacteristics of the transformant are those ofstrain R6. Figure 1 shows the cellular utiliza-tion of FU, U, and UR by strain R6FUr versusits genetic parent strain R6 in detail.

Introduction of FU resistance bytransformation into a pyrimidine-requiringauxotroph. Competent Ua cells were testedwith DNA purified from the FUr strain. Thedonor cells do not require pyrimidine supple-ments and carry the wild-type allele for the

TABLE 8. Uptake of pyrimidines and pyrimidinenucleosides by wild-type and FUN'r pneumococcia

Ratio of intracellularto extracellular

Precursor Concn/ml concentrations ofunchanged precursor

Wild-type FUN'

FUR-2-14C 0.72 uCi/10 ,g 28.8 1.5FUdR-2-14C 0.06 MCi/1 Mg 68b 0.32UdR-2-14C 0.01 ,Ci/.08 Mg 24.0 < 1.5c"4C-UdR (U) 0.059 MCi/2.5 tsg 9.314C-CdR (U) 0.1,uCi/.11 Mg <0.5TdR-2-14C 0.65 MCi/5 Mg < 1.5c <0.15cFU-2-14C 1.26MgCi/10g 1.6b 1.6b

(5.3)d

a Radioactive precursors were added to matched,growing pneumococcal cultures in basal medium atthe concentrations specified, at initial cell concentra-tions of 2.5 to 5 x 107 viable units/ml. At the end ofthe incubation period (1 h for all precursors, with theexception of 2 h for TdR-2-14C), cultures were sam-pled for determination of intracellular isotope dis-tribution, and the radioactive composition of theculture medium and of the trichloroacetic acid-extractable soluble pool was determined by either orboth one- or two-dimensional thin-layer chromatog-raphy, autoradiography, and scintillation counting.Calculation of intracellular concentration was basedon the estimation of the packed cell volume (equal to1/1000th of the total culture volume at these cellconcentrations). Data is reported as the ratio of theconcentrations of unchanged precursor present in theintracellular soluble pool to that present in theculture medium at the end of the incubation period(i.e., a ratio of greater than 1.0 indicates concentra-tive uptake of the precursor).

I Under these conditions, the fluoropyrimidinetreatment is inhibitory to culture growth.

c In these cases thin-layer chromatography analy-sis of the soluble-pool extract was not performed. Thedata given are maximal values, based on calculationtaking the entire soluble-pool content as unchangedprecursor.

dThis value was determined by using the FUrmutant, where FU is an essentially nonmetabolizableprecursor.

1352 J.- BACTERIOL.

on January 20, 2021 by guesthttp://jb.asm

.org/D

ownloaded from

PNEUMOCOCCUS MUTANT RESISTANCE

TABLE 9. Introduction of FU resistance by genetictransformationa

Recipiet Donor Trans-Determination Rcipient Dono formantR6 upp ~R6upp

FU S R RFUR, FUdR, FCdR S S SU reversal of FUR + _

inhibitionFU incorporation + _U incorporation + _UR incorporation + + +

a S, Sensitive; R, resistant; +, normal; -, deficient.

Ua marker; for the purposes of this presenta-tion, the genotype corresponding to the FUrphenotype will be denoted Ua+upp. In order todetect transformation to pyrimidine independ-ence as well as to FU-resistance, samples of theexperimental tubes were scored under severalconditions, including both the presence andabsence of UR as a pyrimidine source and thepresence and absence of various concentrationsof FU. Analysis of this selective scoring datagave the following transformation frequencies:fIlTa = 4 x 10;f,p= 7 x 10-; f,., pp1X 10-5. Eleven clones derived from the DNA-treated culture were isolated after selection forFU-resistance (in the presence of 50 ,g of UR,20 yg of adenosine, and 1 ug of FU per ml).These 11 clones were cultured under nonse-

lective conditions (in medium containing 50 ,gof UR per ml) for many generations and werefurther characterized for their nutritional anddrug-resistance properties. Of the 11 clonespurified, all were resistant to high levels of FU,eight showed a nutritional requirement foruridine (which could not be satisfied by uracil),and three showed no pyrimidine requirement.

It is interesting to note that, in the geneticbackground of the Ua strain, transformation ofthe upp marker both from sensitivity to resist-ance (as above) and from resistance to sensitiv-ity can be readily detected by employing sim-ple selection methods. For the cross-Uaupp(recipient) x Uaupp+ (DNA), transformants ofthe type USupp+ can easily be distinguishedfrom cells of the recipient genotype on the basisof their newly acquired ability to satisfy thepyrimidine requirement with uracil.

DISCUSSIONThe existence of separate classes of noncross-

resistant mutations against the individual fluo-ropyrimidines is consistent with the previousobservation (2) that the different fluoropyrimi-dines each exhibit a different spectrum of

inhibitory effects, because these drugs are sub-ject to distinct patterns of utilization in pneu-mococci.Both mutation rates and frequencies of

transformation (Table 2) are consistent withthe interpretation that each of the FPr mutantscarries a different single-marker mutation. Inview of these findings, the physiological andbiochemical properties of each mutant strainmust be explained as the result of a singlegenetic alteration from the genotype of theparental strain.

All the physiological properties of the FUrand FURr pneumococci seem to be conse-quences of the mutationally decreased activityof UMP pyrophosphorylase and uridine kinase,respectively. Accordingly, the genotypes uppand udk have been assigned to the phenotypic-ally FUr and FURr strains, respectively.

cpm /mI

FU-2-'4CWild type

Tronsformant

20,000

U-2-'44C B

10,000 /

30 6020,000

UR-6-3H10,000 -

30 60Minutes

FIG. 1. Cellular uptake of FU, U, and UR by thewild-type (R6) and an FLUr transformant (R6 upp).At 0 min cultures were supplemented with radioac-tive precursors at the following final concentrations:(A) 0.1 gCi per 10 pg of FU-2-'4C per ml; (B) 0.4 uCiper 10 pg U-2- 'IC per ml; and (C) 2.5 pCi per 20 pg ofUR-6-9H per ml. Crosses (solid lines) and circles(broken lines) show whole-cell incorporation valuesin counts per minute per milliliter of culture for thewild type and transformant, respectively.

1353VOL. 113, 1973

on January 20, 2021 by guesthttp://jb.asm

.org/D

ownloaded from

BEAN AND TOMASZ

Both in vivo and in vitro studies with mutantand wild-type pneumococci on the utilizationof uracil, uridine, and the corresponding fluori-nated compounds show that the compoundsare metabolized by initial pathways shown inFig. 2. In contrast to other species of microor-ganisms (9), in pneumococcus there is no directinterconversion of uracil and uridine by way ofnucleoside phosphorylase or hydrolase. UMPnucleotidase is readily detectable both in vitroand in vivo; when labeled with radioactive FU,substantial amounts of the precursor are recov-erable from the pneumococcal soluble pool inthe form of FUR (2). No evidence has beenfound for operation of alternate pathways forthe metabolism of uracil or uridine in eitherhealthy or drug-inhibited cultures, ie., vianucleoside-N-transdeoxyribosidase or by ox-idative or reductive catabolism of the pyrimi-dine ring. Preliminary experiments suggestthat both UMP pyrophosphorylase and uridinekinase are subject to feedback inhibition bynucleoside triphosphates, as is the case in otherbacteria (1, 8).The nature of the mutational defect of the

FUNr strain, on the other hand, is less obvious.This strain shows normal growth in the pres-ence of 100 times the FUdR concentration re-quired to inhibit the wild-type, and is cross-resistant tQ moderate levels of FUR and FCdR(at levels approximately 50 and 5 timesthose required to inhibit the wild-type strain,respectively). FUNr shows normal sensitivityto FU and FC. The cross-resistance to FURshown by this strain is unilateral; that is, noreciprocal cross-resistance to FUdR is shown bythe FURr mutant.

Relative to its parental strain, the FUNrmutant exhibits decreased ability to utilizeseveral different pyrimidine deoxyribonucleo-sides and ribonucleosides, including FUdR,UdR, TdR, CdR, FUR, and UR, but showsnormal capacity for incorporation of FU. Dur-ing the utilization of these nucleosides, andparticularly when they are supplied at lowconcentrations (less than 1 /Ag/ml of culture),only tiny amounts of the precursor and itsderivatives can be recovered from the cellular

U P de novo pathway

O/ATP ATPUMP '- UDP- UTP

ATPF/

URFIG. 2. Early steps in the metabolism of uracil

and uridine.

soluble pool (Table 7). Thus, this mutantstrain appears to have lost the capacity toconcentrate or retain, or both, supplied pyrimi-dine nucleosides.

Further support for this suggestion comesfrom experiments in which the same precursorswere supplied at higher concentrations (Table5). Under these conditions, the precursors doenter the mutant cells in concentrations suffi-cient to permit their recovery from the cellularsoluble pool; and in this case the expectedphosphorylated derivatives are present. In vitroassays for uridine kinase, the uridine nucleo-tide kinases, UMP nucleotidase, thymidylatesynthetase, and other enzymes revealed thepresence of normal levels of activity for allenzymes tested. Therefore, the FUNr strainshows direct crypticity for at least one enzyme,uridine kinase; i.e., in spite of the fact that theutilization of uridine is defective in this mu-tant, the intact cells do contain the normalactive enzyme systems for its utilization.

Considered together, these several observa-tions suggest the the FUNr mutant strain isdefective for an enzymatic step concerned withthe uptake or retention, or both, of pyrimidinenucleosides by pneumococcus. Until more pre-cise identification of the enzymatic defect canbe made, the genotype fun has been assigned tothis strain. If this interpretation is correct, thenin normal pneumococci, there is at least onecommon element involved in the active trans-port of a number of different ribo- and deoxy-ribonucleosides (including UR, CdR, UdR,TdR, and perhaps other nucleosides). Theuptake of uracil and FU, on the other hand, isapparently mediated by a different system.Although uptake of TdR is not obviously

concentrative (Table 8), its utilization bythe FUNr mutant is defective (Table 7), sug-gesting that the uptake of TdR is mediated bythe same transport system that is responsiblefor the concentrative accumulation found forother nucleoside precursors.

In a previous publication it was proposedthat one of the important inhibitory roles ofFUdR in pneumococci may result from itsinhibition of the transport or phosphorylationof TdR (2), and results presented here suggestthat the permeation of both of these com-pounds is mediated by the same transportsystem.

LITERATURE CITED1. Anderson, E. P., and R. W. Brockman. 1964. Feedback

inhibition of uridine kinase by cytidine triphosphateand uridine triphosphate. Biochim. Biophys. Acta91:380-386.

1354 J. BACTRIOL.

on January 20, 2021 by guesthttp://jb.asm

.org/D

ownloaded from

PNEUMOCOCCUS MUTANT RESISTANCE

2. Bean, B., and A. Tomasz. 1971. Inhibitory effects andmetabolism of 5-fluoropyrimidine derivatives in Pneu-mococcus. J. Bacteriol. 106:412-420.

3. Bean, B., and A. Tomasz. 1973. Selective utilization ofpyrimidine deoxyribonucleosides for deoxyribonucleicacid synthesis in pneumococcus. J. Bacteriol. 113:1356-1362.

4. Brawerman, G., and E. Chargaff. 1954. On the synthesisof nucleotides by nucleoside phosphotransferases. Bio-chim. Biophys. Acta 15:549-559.

5. Furlong, N. B. 1963. A rapid assay for nucleotide kinasesusing C14 or H3 labeled nucleotides. Anal. Biochem.5:515-522.

6. Hotchkiss, R. D. 1957. Isolation of sodium deoxyribonu-cleate in biologically active form from bacteria, p.

692-696. In S. P. Colowick and N. 0. Kaplan (ed.),Methods in enzymology, vol. 3. Academic Press Inc.,New York.

7. Hotchkiss, R. D. 1957. Methods for characterization ofnucleic acids, p. 708-715. In S. P. Colowick and N. 0.

Kaplan (ed.), Methods in enzymology, vol. 3. Aca-

demic Press Inc., New York.8. Molloy, A., and L. R. Finch. 1969. Uridine-5'-monophos-

phate pyrophosphorylase activity from Escherichiacoli. Fed. Eur. Biochem. Soc. Lett. 5:211-213.

9. O'Donovan, G. A., and J. Neuhard. 1970. Pyrimidinemetabolism in microorganisms. Bacteriol. Rev.34:278-343.

10. Okazaki, R., and A. Kornberg. 1964. Deoxythymidinekinase of Escherichia coli. I. Purification and some

properties of the enzyme. J. Biol. Chem. 239:269-274.11. Tomasz, A. 1966. Model for the mechanism controlling

the expression of the competent state in pneumococ-cus cultures. J. Bacteriol. 91:1050-1061.

12. Tomasz, A., and R. D. Hotchkiss. Regulation of thetransformability of pneumococcal cultures by mac-

romolecular cell products. Proc. Nat. Acad. Sci.U.S.A. 51:480-487.

13. Wahba, A. J., and M. Friedkin. 1962. The enzymaticsynthesis of thymidylate. I. Early steps in the purifi-cation of theymidylate synthetase of Escherichia coli.J. Biol. Chem. 237:3794-3801.

VOL. 113, 1973 1355

on January 20, 2021 by guesthttp://jb.asm

.org/D

ownloaded from