sign in sign up
<<
Home , Adenylosuccinate lyase, GMP reductase, Guanosine phosphorylase, Inosine

JOURNAL OF BACTERIOLOGY, JUlY 1987, P. 2977-2983 Vol. 169, No. 7 0021-9193/87/072977-07$02.00/0 Copyright ) 1987, American Society for Microbiology Genetic and Physiological Characterization of Mutants Resistant to Analogs HANS H. SAXILD AND PER NYGAARD* Division, University Institute ofBiological Chemistry B, DK-1307 Copenhagen K, Denmark Received 14 November 1986/Accepted 1 April 1987

BaciUus subtilis mutants defective in purine have been isolated by selecting for resistance to purine analogs. Mutants resistant to 2-fluoroadenine were found to be defective in phosphoribosylbransferase (apt) activity and slightly impaired in adenine uptake. By making use of apt mutants and mutants defective in activity, it was shown that adenine deamination is an essential step in the conversion of both adenine and adenosine to . Mutants resistant to 8-azaguanine, pbuG mutants, appeared to be defective in and guanine transport and normal in hypoxanthine-guanine phosphoribosyltransferase activity. Purine auxotrophic pbuG mutants grew in a concentration-dependent way on hypoxanthine, while normal growth was observed on inosine as the purine source. Inosine was taken up by a different transport system and utilized after conversion to hypoxanthine. Two mutants resistant to 8-azaxanthine were isolated: one was defective in phosphoribosyltransferase (xpt) activity and xanthine transport, and another had reduced GMP synthetase activity. The results obtained with the various mutants provide evidence for the existence of specific purine base transport systems. The genetic lesions causing the mutant phenotypes, apt, pbuG, and xpt, have been located on the B. subtilis linkage map at 243, 55, and 198 degrees, respectively.

The sporeforming gram-positive bacterium Bacillus sub- sporulation, a shrinkage of the intracellular pools of GDP tilis is able to synthesize purine nucleotides from smaller and GTP occurs under different experimental conditions (8, metabolites via the purine de novo pathway and to reutilize 18, 26). The purpose of the present study was to gain insight and via the purine salvage path- into purine base and utilization. We report the ways (24) (Fig. 1). The preformed purine compounds may be isolation and the physiological and genetic characterization formed endogenously from the turnover of nucleic acids, of mutants of B. subtilis resistant to either 2-fluoroadenine, from nucleic acids taken up, or as degradation products in 8-azaguanine, or 8-azaxanthine. The results obtained have the surroundings from decaying cells. B. subtilis possesses contributed to our knowledge ofthe purine salvage pathways both extracellular and intracellular nucleotidases (6, 24). The and transport systems of B. subtilis. utilization of nucleosides as carbon or nitrogen sources is of minor importance in B. subtilis (9). The purine salvage pathways are well documented in many microorganisms (24) MATERIALS AND METHODS and have been investigated in some detail in B. subtilis (7). The B. used in this While the purine de novo pathway appears to be almost Organisms and media. strains of subtilis identical in the various organisms studied, the salvage path- study are listed in Table 1. The minimal medium used was ways seem far more diverse, as defined by the biochemical Spizizen salts (29) supplemented with 5 F.M MnSO4, 1 ,ug of reactions involved in purine utilization (24). Even among the thiamine hydrochloride per ml, 0.2% L-glutamic acid, and various bacilli there are major differences in how exogenous 0.4% D-glucose. Minimal agar plates contained Spizizen purine compounds are metabolized (11, 24). By comparison, salts, 5 ,uM MnSO4, 1 ,ug of thiamine hydrochloride per ml, less is known about the transport of purine bases and 0.4% glycerol, and 1.6% agar. MG1 medium was Spizizen nucleosides in microorganisms (3, 21). One major obstacle salts, S ,uM MnSO4, 1 p.g of thiamine hydrochloride per ml, has been the problem of separating transport from subse- 0.5% D-glucose, 0.2% yeast extract, and an quent intracellular conversion. mixture known to enhance competence in B. subtilis (30): L-histidine, L-tryptophan, L-argiline, L-valine, L-, L- Attempts to elucidate purine transport systems in B. threonine, L-glycine, L-asparagine, and L-methionine (all at subtilis have revealed the presence of several systems (3, 50 ,ug/ml). MG2 medium was MG1 supplemented with 2 mM 21). Useful tools for the identification of purine salvage MgCl2 and 0.5 mM CaCl2. As complex media, L-broth and pathways and purine transport systems are the use of brain-heart infusion broth (Difco Laboratories, Detroit, mutants defective in specific steps of the pathways or Mich.) were used. Nutrients required by auxotrophic strains transport systems and of analogs which specifically inhibit were added at a final concentration of 50 ,ug/ml for amino . Purine analogs which are not toxic until acids, 1 p.g/ml for vitamins, and 30 ,ug/ml for punne and they have been converted to nucleotides have been used compounds unless stated otherwise. extensively in microorganisms to select purine salvage and Growth conditions. Cells were cultured overnight at 30°C transport mutants (21, 24). Purine metabolism in B. subtilis is and diluted in the same medium to exponential growth at of particular interest because purine compounds appear to 37°C. Growth was followed by measuring the optical density play a key role in the initiation of sporulation (26, 28). Freese (OD) at 436 nm in an Eppendorf photometer. An optical and colleagues have shown that during the initiation of density of 1 (1-cm light path) corresponds to 0.2 mg (dry weight) of bacteria per ml. Resistance to 2-fluoroadenine (2 * Corresponding author. ,uM), 8-azaguanine (0.5 mM), 8-azaxanthine (2 mM), and 2977 2978 SAXILD AND NYGAARD J. BACTERIOL.

ATP---- AICAR-----PRPP GTP

Histidine 0 ADP FAICAW~I guaC r.npuuLJI

I purE purA grgguaA I AMP *- sAMP *- IMP XMP _-. GMP

ujaF Ixpt guaF pupA tapt adeC to guaP guaP Ado - Ade - Hyp * Ino Xan Gua .- Guo <-tt Interior t 4 4- 4 4 I j(3) (1) Exterior IHpuG 1(2) (4) pbuG 1(2) Ado Ade Hyp Ino Xan Gua Guo FIG. 1. Proposed pathways for purine salvage and transport in B. subtilis. The individual reactions are identified by symbols. Symbols in italics indicate mapped mutations. Numbers indicate transport systems identified by Beaman et al. (3). AICAR, Aminoimidazolecarboxamide ; FAICAR, formamidoimidazolecarboxamide ribonucleotide; purA, synthe- tase; purE, adenylosuccinate ; guaA, IMP dehydrogenase; guaB, GMP synthetase; guaC, GMP reductase; apt, adenine phosphoribosyltransferase; xpt, xanthine phosphoribosyltransferase; guaF, guanine-hypoxanthine phosphoribosyltransferase; adeC, adenine deaminase; pupA, adenosine phosphorylase; guaP, (inosine) phosphorylase; pbuG, purine base uptake (guanine/hypoxanthine). Numbers: 1, Adenine transport system; 2, guanosine (inosine) transport system; 3, adenosine transport system; 4, xanthine transport system. Other abbreviations: see Table 2, footnote b, and Table 3, footnote a.

4-azaleucine (0.2 mM) was scored on minimal agar plates fresh warmed medium, and suspended in 15 ml of warm containing the appropriate analog. The tsi and dnaD markers medium. After 5 min of incubation, 1 ml of culture was were scored on L-broth plates at 48 and 42°C, respectively. transferred to a tube containing the appropriate "C-labeled The tre marker was scored on minimal plates containing purine compound at a final concentration of 1 ,uM (50 0.4% trehalose as the carbon source. Ci/mol). At 15, 30, 45, and 60 s, 200-,ul samples were Chemicals and isotopes. Fine chemicals were purchased withdrawn, filtered through a membrane filter, and washed from Sigma Chemical Co., St. Louis, Mo. 14C-labeled com- with 2.5 ml of medium. The filter was dried, and radioactiv- pounds were purchased from Amersham International, ity was determined by liquid scintillation counting. For Amersham, England. 2-Fluoroadenine was kindly donated uptake measurements in purine auxotrophic strains, the cells by M. W. Taylor (Indiana University, Bloomington, Ind.). were suspended in minimal medium lacking purine, and Carrier-free [32P]phosphoric acid was from Fors0gsanlaeg, uptake was determined immediately thereafter. Ris0, Denmark. and PRPP pools. ATP, GTP, and Enzyme assays. Enzyme analyses were carried out at 37°C. PRPP pools were determined as described before (14) in cells Protein was determined by the method of Lowry et al. (20). cultured in low-phosphate medium (23) in the presence of Phosphoribosylpyrophosphate (PRPP) amidotransferase and 32P, (10 Ci/mol). Incorporation of [8-14C]adenine and [8- phosphoribosylglycinamide (GAR) synthetase activities "4C]adenosine into ATP and GTP was performed with cells were measured as described by Houlberg and Jensen (10). growing in minimal medium containing histidine (100 ,uglml). Activities of adenine, guanine, hypoxanthine, and xanthine At an OD436 of 0.5, the labeled compound was added to give phosphoribosyltransferases, , and pu- a final concentration of 50 ,uM (5 Ci/mol). At an OD436 of 1, rine nucleoside kinases were assayed as described before 25-,ul samples were withdrawn for pool analysis (15, 25). Purine nucleoside phosphorylase activity was de- (14). termined by the method of Jensen (13). AMP deaminase Transduction and phage production. The generalized trans- activity was determined as described by Murakami (22), and ducing phage AR9, which is related to the phage PBS1, was GMP synthetase activity as described by Lambden and used (19). Fifty microliters of producer lysate (propagated in Drabble (16). Adenase activity, the deamination of adenine Bacillus pumilus) and 50 ,ul of bacteria grown in L-broth to hypoxanthine, was measured in whole cells concentrated were added to 5 ml of brain-heart infusion broth and incu- to an OD436 of 3 to 5 in 0.1 M Tris hydrochloride (pH 7.6) bated for 24 h with shaking and then for 24 h without shaking containing 0.5 mM [8-14C]adenine (5 Ci/mol). Ten-microliter at 30°C. Chloroform was added, and the culture was left for samples were withdrawn at 4-min intervals, applied to 1 h at room temperature. The lysate was cleared by centrif- polyethyleneimine-impregnated cellulose plates on plastic ugation and treated with DNase (50 ,ug/ml) for 1 h at 37°C. sheets, and chromatographed in a solvent that allows the For transduction, the recipient strain was precultured in separation of hypoxanthine and adenine (25). After separa- MG1 medium plus requirements for 20 to 24 h at 30°C. Then, tion the chromatograms were examined under UV light, the 0.5 ml of culture was mixed with 200 ,ul of transducing lysate spots were cut out, and the radioactivity was determined by (usually 10 times diluted in L-broth). The cells were left for liquid scintillation counting (25). 30 min at room temperature without shaking and then Measurement of uptake of purine bases and nucleosides. washed twice with 0.5 ml of Spizizen minimal salts medium The rate of uptake of purine bases and nucleoside were before being plated on selective medium. determined at 37°C in exponentially growing cultures. At an Isolation of DNA and transformation. The donor strain was OD436 of 0.4, 15 ml of culture was collected on a membrane cultured overnight in L-broth, and DNA was isolated as filter (0.45-,um pore size; Sartorius), washed with 20 ml of described by Young and Wilson (31). Development of com- VOL. 169, 1987 METABOLISM OF PURINE BASES IN B. SUBTILIS 2979

TABLE 1. Strains of B. subtilis used Strain Relevant genotype' Original strain name and source or reference ED1 trpC2 168, C. Anagnostopoulos ED21 ilvAl sacA78 QB-16, F. Kunst ED71 aroI906 purB33 dal trpC2 QB-928, R. A. Dedonder ED72 tre-12 metCS glyB133 trpC2 QB-934, R. A. Dedonder ED77 hisAl thr-5 trpC2 QB-917, R. A. Dedonder ED90 ilvAl sacA78 (upp) J. Neuhard ED94 ilvAl sacA78 (upp pupA2) J. Neuhard ED95 ilvAl sacA78 (upp pupA3) J. Neuhard ED139 purB6 leuA8 metB5 Mu8u5u6, N. Sueoka ED141 purHi metC7 trpC2 61634, E. Freese ED145 tre-12 dal aroI906 trpC2 Tf. of ED71 by ED72,b Pur+ ED147 purE26 metB5 Mu5u26, M. Chilton ED154 purE26 tre-12 trpC2 Tf. of ED145 by ED147, Aro+ ED156 purB6 hisAl trpC2 Tf. of ED77 by ED139, Thr+ ED158 purHi hisAl trpC2 Tf. of ED77 by ED141, Thr+ ED163 tsi-23 leuA8 metB5 MB500, J. Marmur ED167 purCIlI metB5 Mu5ull, M. Chilton ED169 tre-12 trpC2 Tf. of ED154 by ED158, Pur+ ED171 ilvAl sacA78 pbuG4 ED21 spontaneous mutant resistant to 0.5 mM 8-azaguanine ED172 ilvAl sacA78 apt-i ED21 spontaneous mutant resistant to 10 FLM 2-fluoroadenine ED173 ilvAl sacA78 pbuG1 (upp) ED90 spontaneous mutant resistant to 0.5 mM 8-azaguanine ED174 ilvAl sacA78 pbuGl (upp) apt-5 ED173 spontaneous mutant resistant to 10 FM 2-fluoroadenine ED175 ilvAl sacA78 (upp) apt-2 ED90 spontaneous mutant resistant to 10 RM 2-fluoroadenine ED176 ilvAl sacA78 pbuG2 (upp pupA3) ED94 spontaneous mutant resistant to 0.5 mM 8-azaguanine ED177 ilvAl sacA78 apt-3 (upp pupA2) ED94 spontaneous mutant resistant to 100 ,uM 2-fluoroadenine ED178 ilvAl sacA78 pbuG3 (upp pupA3) ED95 spontaneous mutant resistant to 0.5 mM 8-azaguanine ED179 ilvAl sacA78 pbuG3 apt-6 (upp pupA3) ED178 spontaneous mutant resistant to 100 ,uM 2-fluoroadenine ED180 ilvAl sacA78 (upp pupA3) apt4 ED95 spontaneous mutant resistant to 100 ,uM 2-fluoroadenine ED181 purB6 pbuGI trpC2 Tf. of ED156 by ED173, His+ ED182 purB6 trpC2 Tf. of ED156 by ED173, His+ ED188 ilvC pheA trpC2 BR95, B. E. Reily ED189 argA2 azlB trpC2 CU456, S. A. Zahler ED191 purB6 apt-6 trpC2 Tf. of ED156 by ED179, His' ED192 ilvAl sacA78 guaBi ED21 spontaneous mutant resistant to 2 mM 8-azaxanthine ED193 ilvAl sacA78 pbuGi xpt (upp) ED173 spontaneous mutant resistant to 2 mM 8-azaxanthine ED194 purB6 xpt trpC2 Tf. of ED156 by ED193, His' ED197 pheA aziB trpC2 Tf. of ED189 by ED188, Arg+ ED203 leuA8 metB5 purAJ6 nic-38 BC50, J. C. Copeland ED204 ilvAl metB5 dnaD TsDNA-D23, D. Karamata ED206 leuA8 metB5 nic-38 Tf. of ED203 by ED188, Pur+ a Parentheses indicate unmapped mutations. b Indicates that the first strain was transformed with DNA from the second strain.

petence and transformation were done as described by agar plates (Table 1). To determine the phenotype, all the Boylan et al. (4) with the following modifications: Recipient mutants isolated were tested for impairment of purine utili-, cells were grown in MG1 medium plus requirements at 37°C. zation by measuring purine phosphoribosyltransferase activ- Ninety minutes after the end of exponential growth, the ities, purine uptake, and resistance to the other analogs culture was diluted 1:10 into MG2 medium. After 60 min in used. The mutants resistant to 2-fluoroadenine were also MG2 medium the cells had gained maximum competence. resistant to 2-fluoroadenosine (10 ,uM). However, this com- Then 1 ml of competent cells and 20 ,ul of DNA (5 to 10 ,ug) pound was only toxic in strains possessing adenosine were mixed and left for 35 min at 37°C with shaking. The phosphorylase activity. cells were then washed two times with 1 ml of Spizizen Four mutants were isolated which were resistant to 8- minimal salts, and proper dilutions were plated on selective azaguanine, ED171, ED173, ED176, and ED178. All showed medium. specific defects in hypoxanthine and guanine uptake, while hypoxanthine and guanine phosphoribosyltransferase activ- ities were normal. Data for one of the mutants, ED173, are RESULTS given in Table 2. To eliminate the possibility that the Isolation of mutants resistant to purine analogs. In a search inability to take up hypoxanthine and guanine was due to for purine analogs that were toxic and could therefore be alterations in the kinetic properties of hypoxanthine used to isolate mutants defective in purine salvage path- phosphoribosyltransferase, enzyme activity was determined ways, a few were of potential interest, namely 8-azaguanine, at various concentrations of hypoxanthine (10 to 100 ,uM) 2-fluoroadenine, 2-fluoroadenosine, 8-azaxanthine, and 6- and PRPP (0.1 to 1 mM) in cell extracts. The levels of thioguanine. Except for 6-thioguanine, we were able to hypoxanthine phosphoribosyltransferase activity in both isolate spontaneous mutants resistant to each analog. Dif- wild-type and pbuG cells were equivalent (data not shown). ferent mutants were isolated in various strains on minimal Six independent mutants resistant to 2-fluoroadenine were 2980 SAXILD AND NYGAARD J. BACTERIOL.

TABLE 2. Purine phosphoribosyltransferase activity and purine TABLE 4. Effect of hypoxanthine and inosine on purine de novo uptake rates in mutants resistant to purine analogs enzyme levels in pbuG and wild-type (pbuG+) cells Phosphoribosyltrans- Uptakeb (nmol/min per Enzyme activityb (nmol/min of protein) StrainRelevant ferase sp aCtb (nmol/ Strain Relevant Addition' per mg Strain genotype' nun per mg of protein) mg [dry wtJ) genotype PRPP GAR Ade Gua Xan Hyp Ade Gua Xan Hyp amidotransferase synthetase ED90 Wild type 28 69 34 36 3.3 1.6 2.2 3.0 ED90 Wild type None 18.3 9.0 ED173 pbuGI 30 70 37 38 2.5 0.1 2.2 0.3 Hypoxanthine 2.2 (8.3) 1.4 (6.4) ED172 apt-i 1 59 45 35 2.5 1.4 2.0 2.8 Inosine 3.3 (5.5) 4.6 (2.0) ED193 pbuGI xpt 29 63 1 35 3.2 0.1 <0.1 0.2 ED173 pbuGI None 19.2 8.6 ED181 pbuGI purB6 30 118 46 97 4.8 0.1 2.4 0.3 Hypoxanthine 6.0 (3.2) 4.4 (2.0) ED182 purB6 33 118 43 % 9.4 4.8 2.4 8.5 Inosine 2.3 (8.3) 5.5 (1.6) a pbuGI, Resistant to 8-azaguanine; apt-i, resistant to 2-fluoroadenine; xpt, a Purine compounds were present at a concentration of 200 FM. resistant to 8-azaxanthine. b Figures in parentheses indicate fold repression (data from two to four b Abbreviations: Ade, adenine; Gua, guanine; Xan, xanthine; Hyp, hypo- experiments). xanthine. DNA isolated from ED173 (pbuGl). The recombinant strain isolated, ED172, ED174, ED175, ED177, ED179, and ED181 (purB6 pbuGI), together with a control strain, ED182 ED180. All were defective in adenine phosphoribosyltrans- (purB6), obtained in the same selection, were assayed for ferase activity and showed normal adenine uptake. Data for purine uptake and phosphoribosyltransferase activities (Ta- one of the isolates, ED172, are given in Table 2. ble 2). The levels of both enzyme activities and uptake rates Two mutants resistant to 8-azaxanthine were isolated. differed somewhat from those of the original strains, but the One, ED193, was found to be defective in both xanthine pbuG phenotype was clear. Growth rate studies revealed uptake and phosphoribosyltransferase activity (Table 2). that the pbuG mutant ED181 grew at a reduced rate in low The other mutant, ED192, exhibited normal purine uptake concentrations of hypoxanthine, while the growth rate on and phosphoribosyltransferase activity (data not shown). inosine was the same at all concentrations tested, as ob- Physiological characteriaton of the pbuG mutants. Nucle- served for ED182 (Fig. 2). In a similar growth experiment, oside uptake was unaffected by the pbuG mutation, as PRPP and purine nucleoside triphosphate pools were mea- shown for pbuGl (Table 3). The results obtained indicate sured in ED181 and ED182 during growth on 30 ,uM hypo- separate transport systems for nucleosides and bases. We xanthine. When ED182 entered the purine starvation phase, also examined excretion of purine compounds in the pbuG a shrinkage of the pools of ATP and GTP and an increase in mutants by measuring their ability to cross-feed purine the PRPP pool were observed (Fig. 3). In contrast, ED181 auxotrophs and found no increase in purine excretion (data appeared to be permanently starved for , as judged not shown). from the increased level of PRPP and the low ATP and GTP Mutants resistant to 8-azaguanine have been isolated by pools. Normal pools were found in the pbuG mutants when others and shown to exhibit derepressed synthesis of some grown on 200 ,uM hypoxanthine (data not shown). of the purine de novo biosynthetic (12). Increased Physiological characterization of apt mutants. The apt levels of purine de novo biosynthetic enzymes were not mutation was introduced into ED156 (purB6) by transforma- observed in our pbuG mutant (Table 4). However, the tion with DNA from ED179 (apt-6). The desired genotype, synthesis of two de novo purine enzymes were repressed to purB6 apt-6 (ED191), lacked adenine phosphoribosyltrans- a lower extent in the mutant when hypoxanthine was present ferase activity. Adenine and hypoxanthine served equally in the growth medium. Addition of inosine, which is cleaved well as sole purine sources in ED191, the doubling time to hypoxanthine inside the cells, had the same effect on being 40 min for both compounds. This indicates that ade- purine enzyme synthesis in both the wild4type and mutant nine phosphoribosyltransferase activity is not essential for strains (Table 4). Both strains possessed inosine phosphor- adenine utilization. To evaluate the pathway of adenine ylase activity (Table 3) but no activity (less interconversion we determined the conversion of adenine in than 0.1 nmollmin per mg of protein). To analyze the wild-type cells and in apt and pupA mutants (Table 5). phenotype of the pbuG mutation in a purine auxotrophic background, strain ED156 (purB6) was transformed with 1.5[

TABLE 3. Uptake of purine nucleosides in pbuG and L. pupA mutants' 53 1.oF A Phosphory- lase activity Uptake (ntmol/nin per mg 0.5F Relevant Strain (nmol/min[dyw]per mg of [dry wti) genotype protein) Ado Ino Ado Guo Xao Ino 50 100 150 200 ED90 Wild type 30 52 2.3 1.5 0.2 1.0 Purine (^*A ED173 pbuGI 27 35 2.0 1.4 0.2 1.0 FIG. 2. Growth rates of ED182 (purB6) and ED181 (purB6 ED95 pupA3 1 49 0.2 ND" ND 1.0 pbuGl) on different concentrations of hypoxanthine and inosine. a Abbreviations: Ado, adenosine; Ino, inosine; Guo, guanosine; Xao, Symbols: O, ED182 supplemented with hypoxanthine; 0, ED181 . supplemented with hypoxanthine; A, ED181 and ED182 supple- bND, Not determined. mented with inosine. VOL. 169, 1987 METABOLISM OF PURINE BASES IN B. SUBTILIS 2981

purB6. p/>uGl 1.0 TABLE 6. Mapping of pbuG Recipient Donor Selection Recombia No. Gene order _--WI-o, 0.5

I a ED154 (purE ED173 pur+ pbuG+ tre+ 7 pbuG-purE- :&- 0.3°i tre) (pbuG) pbuG+ tre 2 tre li pbuG tre+ 58 0 pbuG tre 149 E ' ED181 (purB ED169 pur+ pbuG+ tre+ 170 pbuG-purB- 10.1 pbuG) (tre) pbuG+ tre 59 tre tre+ 0 1 2 3 pbuG 14 Time (h) pbuG tre 6 FIG. 3. Nucleotide and PRPP pools in ED182 (purB6) and ED181 ED181 (purB ED163 pur+ pbuG+ tsi+ 79 tsi-pbuG- (purB6 pbuGl) during growth on 30 ,uM hypoxanthine. Symbols: A, pbuG) (tst) pbuG+ tsi 103 purB tsi+ 17 ATP; l, GTP; 0, PRPP; 0, OD436- pbuG pbuG tsi 0 ED154 (purE ED156 pur+a purB+ tre+ 0 purE-purB- Neither adenosine phosphorylase nor adenine phos- tre) (purB) purB+ tre 22 tre phoribosyltransferase was required for adenine interconver- pirB tre+ 73 sion, and the cells possessed adenase activity. purB tre 245 For comparison, the utilization of adenosine was deter- a On hypoxanthine-containing plates, purE mutants require adenine for mined. The uptake of adenosine (Table 3) and the utilization growth and purB mutants will grow on hypoxanthine as the purine source. of adenosine for ATP synthesis (Table 5) depend on adeno- sine phosphorylase activity. B. subtilis has only low levels of adenosine kinase activity, 0.1 nmollmin per mg of protein, in wild-type cells (ED21). Strain ED192 grew normally on indicating that the conversion of adenosine to GTP in minimal medium, did not require guanine for growth, and wild-type and apt cells implies the intermediate formation of had normal pools of GTP and ATP (data not given). Trans- adenine. Furthermore, measurements of AMP deaminase formation studies indicated that the mutation was located in activity revealed levels below 3 pmol/min per mg of protein. the pur gene cluster, most likely in guaB, resulting in the The apparent adenosine deaminase activity measured (Table synthesis of an altered enzyme. 5) most likely can be ascribed to the combined action of Chromosomal location of the pbuG, apt and xpt . adenosine phosphorylase and adenase. The presence of Transducing lysates were made from ED173 (pbuGl), purine compounds in the culture medium resulted in a two- ED179 (apt-6), or ED193 (xpt). The nine strains of the to threefold increase in the level of inosine phosphorylase mapping kit (5) were transduced to wild type for each of the when adenosine or inosine was added, while the levels of selectable markers they contained. Approximately 100 re- adenase and adenosine phosphorylase were not affected. combinants from each such transduction were then purified, Physiologicil characterization of 8-azaxanthine-resistant and the possible cotransduction frequency with one of the mutants. DNA from strain ED193 (xpt) was used to trans- resistance markers was tested on selective plates containing form the purine-auxotrophic strain ED156 his' xpt, and the either 8-azaguanine, 2-fluoroadenine, or 8-azaxanthine. The pur transformants were tested for growth on 200 p.M pbuGI mutation was found to cotransduce with purB6, apt-6 xanthine as the sole purine source. The desired type, ED194 mutation with leuA, and xpt with trpC on the B. subtilis (purB6 xpt), could not grow on xanthine as the purine (27). To locate more precisely the pbuG, apt, source, lacked xanthine phosphoribosyltransferase activity, and xpt mutations on the chromosome, suitable strains were and required hypoxanthine or guanine for growth. The other constructed for three-factor crosses. The data of the detailed mutant resistant to 8-azaxanthine (ED192) exhibited wild- tnapping of pbuGI are presented in Table 6. By transducing type xanthine phosphoribosyltransferase activity and strain ED154 to purE+ with phages from ED173 (pbuGI), the xanthine transport function, but showed reduced GMP syn- gene order pbuG-purE-tre was suggested. To determine the thetase activity in cell extracts, less than 0.02 nmol/min per orientation of purB and pbuG relative to more distant mg of protein, compared with 12 nmol/min per mg of protein markers on either side of the purine cluster, strain ED181 (purB6 pbuGl) was transduced to pur+ with phages from ED163 (tsi) or ED169 (tre-12). The distribution of the dif- TABLE 5. Utilization of adenine and adenosine for ATP and ferent recombinant classes from these two crosses (Table 6) GTP synthesis in purine salvage mutants indicates the gene order tsi-pbuG-purB and pbuG-purB-tre. Nucleotide poolb (nmol/mg To determine the orientation of purE and purB relative to tre, ED154 Adenase activity' (purE26 tre-12) was transduced to pur+ with Strain genotypeleanotye (nmolUmin per mg [8-'C]ade- [8-4C]aden- phages from ED156 (purB6) (Table 6). This cross indicated [dry wt]) nine osine the gene order purE-purB-tre. A proposed map for the tsi-tre ATP GTP ATP GTP region is shown in Fig. 4. Both by transformation and by ED21 Wild type 4 9.2 1.7 8.6 1.6 transduction analysis, we found that pbuG2, -3, and 4 ED172 apt-i 5 (6) 8.8 1.8 7.1 1.6 mapped in the same region of the chromosome as pbuGl ED95 pupA3 5 (<0.1) 6.8 1.3 2.3 0.1 (data not shown). ED180 apt-4 pupA3 4 (<0.2) 6.1 1.1 0.9 0.1 For detailed mapping of the apt gene, two strains, ED197 (pheA aziB) and ED206 (leuA8 nic-38), aFigures in parentheses indicate adenosine phosphorylase-dependent were constructed and adenosine deaminase activity. transduced to phe+ and leu+, respectively, with phages from b For nucleotide pool size sleterminations, cells were grown in the presence ED179 (apt-6) (Table 7). The five other apt mutants isolated of [8-14CJadenine or [8-_4C]adenosine as described in Materials and Methods. carried mutations that were shown by transduction to map in 2982 SAXILD AND NYGAARD J. BACTERIOL.

ts/ pbuG purE purB tre TABLE 7. Mapping of apt and xpt Selec- Recombinant fI " . i . Recipiet Donor ClaSS t N. GeneGn Orderre 50 55 60 65 Recipient tiOn No. ED197 (pheA ED179 phe+ apt' azIB+ 5 pheA-apt-azlB azlB) (apt-6) apt' azlB 57 48 apt azIB+ 57 74 apt azlB 91 ED206 (leuA ED179 leu+ apt' nic+ 37 leuA-nic-apt 79 nic) (apt-6) apt' nic 64 apt nic+ 108 apt nic 1 leuA pheA nic apt az/B ED204 (metB ED193 met' xpt+ dnaD' 2 dnaD-xpt-metB 250I 24I3 dnaD) (xpt) xpt+ dnaD 24 25~0 245 240 235 xpt dnaD+ 138 31 xpt dnaD 45 48 ED1 (trpC) ED193 (xpt trp+ xpt+ ilvA+ 146 trpC-xpt-ilvA 30 - iIvA) xpt+ ilvA 2 70 xpt ilvA+ 80 xpt ilvA 104

trpC dnaD xpt netB ilvA I I z phoribosylation or that the xpt gene is located in an operon 205 200 195 together with a transport gene. 12 Recent studies of nucleoside transport-deficient manmma- 33 lian cell lines have provided evidence for a common nucle- 68 oside transport system which is also responsible for purine 45 base transport (1, 2). The results of the present study (Table FIG. 4. Genetic maps of the tsi-pbuG-tre, leuA-apt-azlB, and 5) and that of Endo et al. (7) indicate that the conversion of trpC-xpt-ilvA regions of the B. subtilis chromosome (27). Numbers adenine and adenosine to GTP requires adenase activity. In represent (100 - cotransduction)% for the markers shown and were agreement with this is the finding (P. Duckert, personal calculated from the data in Tables 6 and 7. Arrows point towards communication) that a purine-auxotrophic mutant defective unselected markers. in adenase activity fails to grow on adenine as the purine source. We have found no evidence for the involvement of the same region of the chromosome as apt-6 (data not adenosine or inosine (guanosine) phosphorylase in purine shown). A map of the leuA-az/B region is shown in Fig. 4. To base utilization, as observed for enterobacteria (24) and map the xpt gene relative to trpC2, strain ED204 (metBS Saccharomyces cerevisiae (17). The reported adenosine dnaD) was constructed. By transducing ED204 to met' with deaminase activity in B. subtilis (7) most likely can be phages from ED193 (xpt) the gene order dnaD-xpt-metB was ascribed to the combined action of adenosine phosphorylase found. Transduction of ED1 with phages from ED193 and and adenase. Two purine kinases have selecting for trp+ established the gene order trpC-xpt-i/vA been identified in B. subtilis; one enzyme phosphorylates (Table 7). A map of the trpC-i/vA region is shown in Fig. 4. and another both and . The latter enzyme also shows some activity towards adenosine (23). This activity may be responsible for DISCUSSION the slow incorporation of adenosine into ATP in the pupA3 New mutants resistant to various purine analogs have mutant (Table 5). In the incorporation experiment histidine been isolated and characterized. Except for one mutant, all was added to prevent the synthesis of anlinoimidazolecar- appeared to be altered in their capacity for purine salvage. boxamide ribotide, which is a byproduct of the Mutant strains (pbuG) were generated which were defective of histidine from ATP. This compound contains the radiola- in their hypoxanthine and guanine transport function but beling from [8-14C]adenine, and since it is an intermediate showed normal hypoxanthine-guanine phosphoribosyltrans- compound of the purihe biosynthetic pathway (Fig. 1), this ferase activity. Their characterization has provided the most labeling will appear in IMP and hence in ATP and GTP, definitive demonstration that the transport requires a genetic provided that the biosynthesis of histidine is not completely component and that purine phosphoribosylation is not re- shut down. B. subtilis does not seem to possess inosine and quired for the purine transport function (Table 2). The purine guanosine kinase activities (7). Based on our data and the nucleosides are taken up by separate transport systems and results of others (7), a scheme for purine base and ribonu- are phosphorolyzed by inosine (guanosine) or adenosine cleoside utilization in B. subtilis is proposed (Fig. 1). The phosphorylase (13). Adenine transport was only slightly genetic localization of pbuG, apt, and xpt genes on the reduced in apt mutants, while xanthine transport and chromosome and the identification of the biochemical le- xanthine phosphoribosyltransferase activity were defective sions in the mutants demonstrate for the first time the in the xpt mutant (Table 2). When present at high concen- gene-protein relationship of purine salvage in B. subtilis. trations in the medium or formed from the phosphorolysis of inosine, which was readily taken up by pbuG mutants (Table ACKNOWLEDGMENTS 3), hypoxanthine was efficiently utilized (Fig. 4). The finding that both transport and phosphoribosylation of xanthine We thank M. G. Sargent for introducing one of us (H.H.S.) to the were defective in the xpt mutant (Table 2) indicates either a techniques of transduction and transformation and B. Hove-Jensen strotg coupling between xanthine transport and phos- for critical comments. VOL. 16%, 1987 METABOLISM OF PURINE BASES IN B. SUBTILIS 2983

LITERATURE CITED chem. 98:254-263. 15. Jochimsen, B., P. Nygaard, and T. Vestergaard. 1975; Location 1. Aronow, B., P. Hollingsworth, J. Patrick, and B. Ullman. 1986. on the chromosome of of genes gbverning Incomplete nucleoside transpoft deficiency with increased hy- purine metabolism. Mol. Gen. Genet. 143:85-91. poxanthine transport capability in mutant T-lymphoblastoid 16. Lambden, P. R., and W. Drabble. 1973. The gua operon of cells. Mol. Cell. Biol. 6:1296-1303. Escherichia coli K-12: evidence for polarity from gulaB to guaA. 2. Aronow, B., and B. UlIman. 1986. Role of the nucleoside J. Bacteriol. 115:992-1002. transport function in the transport and salvage of purine 17. Laten, }1. M., P. J. Valentine, and C. A. Van Kast. 1986. nucleobases. J. IBiol. Chem. 261:2014-2019. Adenosine accumulation in Saccharomyces cerevisiae cultured 3. Beaman, T. C., A. D. Hitchins, K. Ochi, N. Vasantha, T. Endo, in medium containing low levels of adenine. J. Bacteriol. and E. Freese. 1983. Specificity and control of uptake of purines 166:763-768. and other compounds in Bacillus subtilis. J. Bacteriol. 18. Lopez, J. M., A. Dromerick, and E. Freese. 1981. Response of 156:1107-1117. guanosine 5'-triphosphate concentration to nutritional changes 4. Boylan, R. J., N. H. Mendelson, P. Brooks, and F. E. Young. and its significance for Bacillus subtilis sporulation. J. Bacteriol. 1972. Regulation of the bacterial cell wall: analysis of a mutant 146:1447-1449. of Bacillus subtilis defective in biosynthesis of teichoic acid. J. 19. Love, E., J. D'Ambrosio, and N. C. Brown. 1976. Mapping of the Bacteriol. 110:281-290. gene specifying DNA polymerase III of Bacillus 3ubtilis. Mol. 5. Dedonder, R. A., J.-A. Lepesant, J. Lepesant-Kejzlarova, A. Gen. Genet. 144:313-321. Billault, M. Steinmetz, and F. Kunst. 1977. Construction of a kit 20. Lowry, 0. H., N. J. Rosebrough, A. L. Farr, and R. J. Randall. ofreference strains for rapid genetincmapping in Bacillus subtilis 1951. Protein measurement with the Folin phenol reagent. J. 168. Appl. Environ. Microbiol. 33:989-993. lNiol. Chem. 193:265-275. 6. Dhariwal, K. R., N. Vasantha, and E. Freese. 1982. Partial 21. Munch-Petersen, A., and B. Mygind. 1983. Transport of nucleic nucleotide limitation induces phosphodiesterase I and 5'- acid precursors, p. 259-305. In A. Munch-Petersen (ed.), Me- nucleotidase in Bacillus subtilis. J. Bacteriol. 149:11461149. tabolism of nucleotides, nucleosides and nucleobases in micro- 7. Endo, T., B. Uratani, and E. Freese. 1983. Purine salvage organisms. Academic Press, Inc., New York. pathways of Bacillus subtilis and effect of guanine on growth of 22. Murakami, K. 1979. AMP deaminase from bakers yeast. Kinetic GMP reductase mutants. J. Bacteriol. 156:169-179. and molecular properties. J. Biochem. 86:1331-1336. 8. Freese, E., J. E. Heinze, and E. M. Galliers. 1979. Partial purine 23. M0olgaard, H. 1980. Deoxyadenosine/deoxycytidine kinase deprivation causes sporulation of Bacillus subtilis in the pres- from Bacillus subtilis. J. Biol. Chem. 255:8216-8220. ence of excess ammohia, glucose and phosphate. J. Gen. 24. Nygmard, P. 1983. Utilization of preformed purine bases and Microbiol. 115:193-205. nucleosides, p. 27-93. In A. Munch-Petersen (ed.), Metabolism 9. Haminer-Jespersen, K. 1983. Nucleoside catabolism, p. of nucleotides, nucleosides and nucleobases in microorganisms. 203-258. In A. Munch-Petersen (ed.), Metabolism of nucleo- Academic Press, Inc., New York. tides, nucleosides and nucleobases in microorganismns. Aca- 25. Nygaard, P. 1978. Adenosine deaminase.from Escherichia coli. demic Press, Inc., New York. Methods Enzymol. 51:508-512. 10. Houlberg, U., and K. F. Jensen. 1983. Role of hypoxanthine and 26. Ochi, K., J. Kandala, and E. Freese. 19g2. Evidence that guanine in the- regulation of Salmonella typhimurium pur gene Bacillus subtilis sporulation induced by the stringent response is expression. J, Bacteriol. 153:837-845. caused by a decrease in GTP and GDP. J. Bacteriol. 11. Ipata, P. L., F. Sgarreila, and M. G. Tozzi. 1985. Mechanisms of 151:704-711. exogenous purine nucleotide utilization in Bacillus cereus. Curr. 27. Piggotj P. J., and J. A. Hoch. 1985. Revised genetic linkage map Top. Cell. Reg. 26:419-432. of Bacilus subtilis. Microbiol. Rev. 49:158-179. 12. Ishii, K., and I. Shiio. 1972. Improved inosine production and 28. Sekar, V., S. P. Wilson, and J. H. Hageman. 1981. Induction of derepression of purine nucleotide biosynthetic enzymes in 8- Bacillus subtilis sporulation by nucleosides: inosine appears to azaguanine resistant mutants of Bacillus subtilis. Agric. Biol. be a sporogen. J. Bacteriol. 145:489-493. Chem. 36:1511-1522. 29. Spizizen, J. 1958. Transformation of biochemically deficient 13. Jensen, K. F. 1978. Two purine nucleoside in strains of Bacillus subtilis by deoxyribonucleate. Proc. Natl. Bacillus subtilis. Purification and some properties of the aden- Acad. Sci. USA 44:1072-1078. osine-specific phosphorylase. Biochim. Biophys. Acta 30. Wilson, G. A., and K. F. Bott. 1968. Nutritional factors influ- 525:346-356. encing the development of competence in the Bacillus subtilis 14. Jensen, K. F., U. Houlberg, and P. Nygaard. 1979. Thin-layer transformation system. J. Bacteriol. 95:1439-1449. chromatographic methods to isolate 32P-labeled 5- 31. Young, F. E., and G. A. Wilson. 1974. Bacillus subtilis, p. phosphoribosyl-1-pyrophosphate (PRPP): determination of cel- 69-114. In R. C. King (ed.), Handbook of genetics, vol. 1. lular pools and assay of PRPP synthetase activity. Anal. Bio- Plenum Publishing Corp., New York.