JOURNAL OF BACTERIOLOGY, Feb. 1973, p. 727-738Copyright i 1973 American Society for Microbiology

Vol. 113, No. 2Printed in U.S.A.

Isolation and Characterization of AcidPhosphatase Mutants in Saccharomyces

cerevisiaeAKIO TOH-E, YOSHINAMI UEDA, SEI-ICHIRO KAKIMOTO, AND YASUJI OSHIMA

Department of Fermentation Technology, Osaka University, Suita-shi, Osaka, Japan

Received for publication 31 October 1972

Saccharomyces cerevisiae strain H-42 seems to have two kinds of acidphosphatase: one which is constitutive and one which is repressible by inorganicphosphate. The constitutive enzyme was significantly unstable to heat inactiva-tion, and its Km of 9.1 x 10-4 M for p-nitrophenylphosphate was higherthan that of the repressible enzyme (2.4 x 10-4 M). The constitutive and therepressible acid phosphatases are specified by the phoC gene and by the phoB,phoD, or phoE gene, respectively. Results of tetrad analysis suggested that thephoC and phoE genes are linked to the lys2 locus on chromosome II. Since bothrepressible acid and alkaline phosphatases were affected simultaneously in thephoR, phoD, and phoS mutants, it was concluded that these enzymes were

under the same regulatory mechanism or that.they shared a common polypep-tide. The phoR mutant produced acid phosphatase constitutively, and the phoRmutant allele was recessive to its wild-type counterpart. The phoS mutantshowed a phenotype similar to that of a mutant defective in one of the phoB,phoD, or phoE genes. However, the results of genetic analysis of the phoSmutant clearly indicated that the phoS gene is not a structural gene for either ofthe repressible acid and alkaline phosphatases, but is a kind of regulatory gene.According to the proposed model, the phoS gene controls the expression of thephoR gene, and inorganic phosphate would act primarily as an inducer for theformation of the phoR product which represses phosphatase synthesis.

The existence of repressible phosphatases(EC 3.1.3.1 and EC 3.1.3.2) in Escherichia coli(10), Saccharomyces cerevisiae (16, 19), Neu-rospora crassa (13, 14), and Euglena gracilis (2)has been reported. Among these, the regulatorymechanism of alkaline phosphatase formationin E. coli was most extensively studied byGaren and his colleagues. They indicated thatthe enzyme formation was controlled by thestructural gene and two regulator genes, Rland R2, as well as by inorganic phosphate (6,7). However, the regulatory mechanism ofphosphatase formation has not been clearlyelucidated in eukaryotic microorganisms be-cause of the lack of an efficient method for theisolation of phosphatase mutants. Recently,Toh-e and Ishikawa (20), however, reported thegenetic analysis of acid and alkaline phospha-tase formation in N. crassa by developing anefficient procedure for the isolation of nucleasemutants, based on the ability to utilize nucleicacid as a sole source of phosphate, and for the

isolation of phosphatase mutants by a diazocoupling method which has been described byDorn (4). They suggested that one of the nu-cleases in the mycelial extract, i.e., nucleaseN, might be an inducing factor for the produc-tion of repressible phosphatases in N. crassa(20).Schurr and Yagil (17), using the diazo cou-

pling method, recently isolated and characteri-zed mutants of S. cerevisiae which lack therepressible acid phosphatase. They observedthat these mutants were also defective in therepressible alkaline phosphatase and that con-stitutive mutants could produce both enzymesin media containing enough phosphate for therepression of both enzymes in the wild-typestrain. They suggested that the formation ofboth enzymes might be controlled by a com-mon regulatory mechanism or that these twoenzymes might share a common polypeptide.We have isolated many strains of the S.

cerevisiae mutant for acid phosphatase by the727

on February 20, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

TOH-E ET AL.

diazo coupling method, and genetic and bio-chemical analyses of these mutants have beencarried out. In the present paper, we reportthat there seem to be two kinds of acid phos-phatases, constitutive and repressible, in thewild-type strain of S. cerevisiae, and we suggestthe possibility of a new regulatory gene whichmight control the expression of the regulatorgene for repressible acid phosphatase synthe-sis.

MATERIALS AND METHODSOrganisms. A heterothallic haploid strain, H-42

(a gal4), which was obtained from a diploid strain ofsake yeast, S. cerevisiae strain KM-46 (15), byrandom spore isolation, was used as the wild-typestrain for acid phosphatase formation. All phospha-tase mutants were derived from H-42 or its deriva-tives. Another haploid strain, T-1074-77D (a lys2gall), a mutant for phoC, was selected from ourbreeding stocks for yeast genetics. The gene symbolslys and gal indicate the requirement for L-lysine andthe inability to ferment galactose, respectively.

Media. The nutrient' medium contained 40 g ofglucose, 10 g of polypeptone, 5 g of yeast extract, 5 gof KH2PO4, and 2 g of MgSO, 7H,O per liter of tapwater. Synthetic minimal medium, in which 2 g ofrecrystallized asparagine per liter was the sole sourceof nitrogen, was prepared according to Burkholder(3). Phosphate-free medium was prepared by thesubstitution of 1.5 g of KCI per liter for 1.5 g ofKH2PO4 per liter in the minimal medium. The origi-nal minimal medium was used as the high phosphatemedium (high-Pi medium), and phosphate-free med-ium supplemented with 30 mg of KH2PO4 per literwas used as the low phosphate medium (low-Pimedium). If necessary, 0.2 mM L-lysine hydrochloridewas added to these synthetic media. For the prepa-ration of solid media, 20 g of agar per liter was added.The sporulation medium contained 5 g of anhydroussodium acetate and 15 g of agar per liter of tap water.

Isolation of mutants. All mutants were selectedby inspecting colonies in the petri dishes according tothe method described by Dorn (4). Cells to besubjected to mutagenesis were grown in nutrientmedium for 24 hr at 30 C without shaking. Then eachculture broth, containing about 107 cells per ml, wasdiluted 100-fold with sterilized water and a 10-mlportion was transferred to a sterilized petri dish 9 cmin diameter. Mutagenesis was performed by ultravio-let irradiation for 80 sec with an 18-w Toshibagermicidal lamp at a distance of 15 cm with stirringby a magnetic stirrer. Under these conditions, 1 to10% of the cells survived after being spread on anutrient plate. Each 0.1-ml sample was spread on aplate containing high-Pi or low-Pi medium, accord-ing to the mutants to be isolated. After 3 to 5 days ofincubation at 30 C, each colony which developed wasoverlaid with melted soft agar (50 C) containing astaining solution consisting of 5 mg of a-naphthyl-phosphate per ml, 50 mg of Fast blue salt B per ml,and 1% agar in 0.05 M acetate buffer (pH 4.0). In thepresence of acid phosphatase activity, the colonies

were stained to a dark red within 30 to 60 min at 30 C,whereas in its absence the colonies remained white.Since the cells were not killed by the staining,mutants could easily be isolated directly from thisplate; if necessary, each isolate was purified byrepeated spreading on a nutrient agar plate. For thediagnosis of phosphatase activity of the tetrad segre-gant, the same procedure was applied on the re-plicated plates. An adenine-requiring mutant such asadel was not useful for the present study, because itproduces colonies colored pink to dark red, and itseems to produce a substance which reacts with Fastblue salt B.Methods for genetic analysis. Diploid hybrids

were obtained by the mass mating method describedby Lindegren and Lindegren (11). For the sporulationof diploid hybrids, after cultivation for 1 or 2 days innutrient medium at 30 C, one loopful of cells wassmeared on sporulation medium and incubated foranother 2 days at 30 C. The cell walls of the asci weredigested by treatment with snail gut juice, andfour-spored asci were dissected with the aid of amicromanipulator.Preparation of cell extract. For the extraction of

acid phosphatase, the strains to be tested were grownin nutrient medium to full growth at 30 C withoutshaking. The cells were harvested, washed, andresuspended in the same volume of sterilized water; 1ml of cell suspension was inoculated in a 500-mlSakaguchi flask containing 100 ml of the indicatedmedium. Incubation was carried out at 30 C withshaking. Cells grown under the indicated conditionswere harvested and washed with 0.01 M acetate buffer(pH 4.0). All subsequent preparations were carriedout below 4 C. Cell concentration was measuredturbidimetrically at 660 nm. A 4-ml amount of heavycell suspension (optical density at 660 nm, approxi-mately 100) was shaken with 20 g of glass beads (0.45to 0.55 mm in diameter) for 15 sec by use of a Brauncell homogenizer (model MSK). To the mixture ofthe disrupted cell suspension and glass beads wasadded 4 ml of 0.01 M acetate buffer (pH 4.0), and theliquid layer was decanted and saved. This procedurewas repeated three more times. Insoluble materialwas removed from the pooled washings by centrifuga-tion at 17,000 x g for 20 min. The resultant superna-tant fluid was used as an enzyme solution, if nototherwise noted. For the extraction of both acid andalkaline phosphatases, 5 mm tris(hydroxymethyl)-aminomethane-maleate (Tris-maleate) buffer (pH6.5) was used instead of 0.01 M acetate buffer (pH4.0) in the above procedure. Protein concentrationwas determined according to Lowry et al. (12), withbovine serum albumin as the standard.Determination of phosphatase activity. Phos-

phatase activity was assayed according to themethod described by Torriani (21). The reaction mix-ture (2 ml) for the acid phosphatase assay consistedof 0.64 mg of p-nitrophenylphosphate per ml and en-zyme in 0.05 M acetate buffer (pH 4.0). The reactionwas carried out at 35 C for 10 min; the reaction wasstopped by the addition of the same volume of 10%trichloroacetic acid. Insoluble material was removedby centrifugation at 2,000 x g for 10 min. Then 2 mlof the supernatant fluid was added to 2 ml of a

728 J. BACTERIOL.

on February 20, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

ACID PHOSPHATASE MUTANTS IN S. CEREVISIAE

saturated solution of Na2CO,. When intact cells wereused as enzyme, cells were removed by centrifugationafter the addition of a saturated solution of Na2CO,.Liberated p-nitrophenol was determined spectro-photometrically at 420 nm. Alkaline phosphataseactivity was assayed by the same method describedabove, except that the reaction mixture contained0.05 M Veronal buffer (pH 9.0) and 5 mMMgSO4-7H20 instead of 0.05 M acetate buffer (pH4.0). One unit was defined as the amount of enzymewhich liberates 1 gmole of p-nitrophenol per minunder the conditions described above.

Chemicals. p-Nitrophenylphosphate was pur-chased from Wako Pure Chemical Industries, Ltd.,and Fast blue salt B was from Merck & Co., Inc.Snail gut juice (suc d'helix pomatia stabilisestandardise) was obtained from the Industrie Bi-ologique, Frangaise.

RESULTSAcid phosphatase species in S. cerevisiae.

Schmidt et al. reported that the acid phospha-tase in S. cerevisiae was derepressed in amedium in which the concentration of inor-ganic phosphate was kept low (16). The samerelationship between acid phosphatase forma-tion and the concentration of inorganic phos-phate in the medium was observed in ourwild-type strain, H-42. Strain H-42 was grownin a medium containing various amounts ofinorganic phosphate at 30 C with shaking, andthe time course of phosphatase activity wasdetermined during the cultivation. Results in-dicated that the acid phosphatase activityincreased as the concentration of inorganicphosphate was reduced, as shown in Fig. 1.Since both acid phosphatase formation (Fig.la) and cellular growth (Fig. lb) in the mediumcontaining 30 mg of KH2PO4 per liter weresatisfactory, this medium was used as thelow-Pi medium for giving a derepressed condi-tion for phosphatase formation. In the low-Pimedium, the highest level of acid phosphataseactivity was constantly maintained during 15to 24 hr of cultivation. Strain H-42 also pro-duced a small but significant amount of acidphosphatase in high-Pi medium (standardminimal medium containing 1.5 g of KHXP04per liter), and production was more efficient inthe nutrient medium.

Several enzymatic characteristics of theseacid phosphatase activities present in cellsgrown on nutrient medium or low-Pi mediumwere compared with crude extracts preparedfrom cells of the wild-type strain, H-42, grownin these media. Each extract was dialyzedagainst 0.01 M acetate buffer (pH 4.0) at 0 Covernight and used as the enzyme solution. Itwas evident that the acid phosphatase pro-duced in the low-Pi medium was markedlyresistant to heat inactivation at 55 C compared

'-- 0 .5 00u

w

a.

0.2 50

000000

E

coCD

o

0 10 20

H 0 U R S

0 1 0 20

H OU R S

FIG. 1. Effect of the concentration of inorganicphosphate on the formation of acid phosphatase. The

wild-type strain, H-42, was grown in various mediacontaining different amounts of inorganic phosphateat 30 C with shaking. Acid phosphatase activity was

determined with a suspension of intact cells as theenzyme source. Specific activity was expressed as

units per unit of optical density at 660 nm. (a) Timecourse of enzyme formation. (b) Time course ofgrowth. Symbols: 0, 1,500 mg; A, 30 mg; A, 15 mg;0, 3 mg; *, 0 mg ofKHPO4per liter in Burkholder'ssynthetic minimal medium; 0, nutrient medium.

with that produced in the nutrient medium(Fig. 2). A kinetic study was also performedwith p-nitrophenylphosphate as substrate. The

a

729VOL. 113, 1973

on February 20, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

TOH-E ET AL.

Km values were calculated from the Line-weaver-Burk plots shown in Fig. 3. It was foundthat the Km of the acid phosphatase producedin the high-Pi medium was 9.1 x 10- I hwhereas the acid phosphatase produced inlow-Pi medium had a lower Km of 2.4 x 10- 4 M.The same Km values for the respective acidphosphatases were also obtained when a sus-pension of intact cells was used as the enzymesource. These results tend to support the ideathat acid phosphatases produced under condi-tions of limited or of sufficient amounts of in-organic phosphate in the medium were differ-ent molecular species and corresponded to therepressible and constitutive acid phospha-tases, respectively.

It is evident that almost all of the activity ofthe acid phosphatase in the extract prepared

100 K I

1 0I-

4

zz

w

cx:

0.10 10 20 30

MINUTES AT 55"CFIG. 2. Effect of heat treatment on acid phospha-

tase activity. The wild-type strain, H-42, was grownto the early stationary phase in nutrient medium (0)and low-Pi medium (0) at 30 C with shaking.Extracts were prepared from these cells in 0.01 Macetate buffer (pH 4.0) and dialyzed against thesame buffer at 0 C overnight. An extract containingabout I unit of acid phosphatase per ml was prein-cubated at 55 C for various periods and then rapidlycooled in an ice bath, and the remaining activity wasdetermined. Activity without preincubation wasscored as 100% and the experimental values wereexpressed as percentages of this value.

0 1 2 3 4

FIG. 3. Lineweaver-Burk plots of acid phospha-tase with p-nitrophenylphosphate as substrate. Ex-tracts were prepared from cells of the wild-typestrain, H-42, grown to the early stationary phase innutrient medium (a) and low-Pi medium (0), andfrom cells of the mutant, 08-M3, grown in low-Pimedium (A). The velocity was expressed as thechange in absorbance measured at 420 nm per 10 minof incubation.

from H-42 cells grown in low-Pi medium wasdue to the activity of the repressible acidphosphatase, because the Km of the acid phos-phatase produced in low-Pi medium by 08-M3,which were derived from H-42 by mutation anddoes not show acid phosphatase activity in cellsgrown in nutrient medium, showed the samevalue as that from H-42 grown in low-Pimedium. In addition, no significant differencewas observed between these two enzymes withrespect to the other enzymatic characteristicstested, such as optimal pH (3.5 4.0) and theeffect of metal ions and ethylenediaminetetra-acetate on their activity.

Isolation of mutants lacking constitutiveacid phosphatase. Freshly cultivated cells ofthe strain H-42 were irradiated with ultravioletlight, plated on high-Pi medium, and in-cubated for 3 to 5 days at 30 C. The colonieswhich developed on the surface of each platewere stained by overlayering with the soft agarof the diazo coupling reagent. Colonies whichwere not stained were picked up. To test theformation of the constitutive acid phosphatase,the wild-type strain H-42 and mutant strains019-M20, 019-M32, and a tetrad segregantfrom hybrid P-22 (i.e., P-22-14D, which has thesame genotype for inability to produce con-stitutive acid phosphatase as 08-M3) weregrown in the nutrient and low-Pi media. Atappropriate intervals during the cultivation,the acid phosphatase activity of each strainwas determined with a cell suspension as anenzyme source. These mutant clones showed noconstitutive acid phosphatase activity in the

730 J.- BACTERIOL.

on February 20, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

ACID PHOSPHATASE MUTANTS IN S. CEREVISIAE

nutrient medium (Fig. 4), but they retained therepressible acid phosphatase activity. Themaximal specific activities of the repressibleacid phosphatase were, however, slightly differ-ent among these mutants. Another strainwhich does not have constitutive acid phospha-tase, T-1074-77D (a Iys2 gall), was selectedfrom our stock cultures for genetic study.Two strains, 08-M3 and T-1074-77D, were

found to lack the constitutive acid phosphataseand were subjected to genetic analysis. Eachstrain was crossed to the wild-type strain (H-42or its derivative) which had the constitutiveacid-phosphatase. The mutant present in08-M3, designated phoC-l, had segregated 2:2for constitutive versus absence of the constitu-tive acid phosphatase in all 76 tetrads tested.The diploid heterozygous for phoC-1 producedthe constitutive acid phosphatase. The allelephoC-1 is, hence, recessive. Segregations forthe mutation in T-1074-77D leading to theabsence of the constitutive acid phosphatase,designated as phoC-2, were essentially 2:2 for

0.075

0.0 50

0.0 2 5

0.000

0.5 0 0

Uw

a.

0.25 0

0.1 250

0 0000 10 20

H O U R S

FIG. 4. Comparison of the formation of acid phos-phatase in the wild-type strain, H-42, and mutantstrains. Each strain was grown in nutrient medium(a) and low-Pi medium (b). Acid phosphatase activ-ity was determined with a suspension of intact cellsas the enzyme source. Specific activity was expressedas units per unit of optical density at 660 nm.Symbols: 0, H-42; A, P-22-14D; *, 019-M20; and0, 019-M32. P-22-14D has the phoC-1 genotypederived from the mutant 08-M3.

the presence or absence of the constitutive acidphosphatase on high-Pi medium in 29 of 34 ascitested. However, one or two acid phosphatase-forming clones in each of five asci showed anambiguous color development. The ambiguouscolor development might be attributable to acertain difference in genetic background be-tween the parental strains and might not bedue to an abnormal meiotic event, because itwas, observed that the lys2 marker showed2+:2- segregation in each ascus. When asegregant from this cross bearing phoC-2 wasbackcrossed to H-42 (PHOC), the phoC-2 alleleagain segregated 2:2 in a total of 16 asci.Diploids heterozygous for phoC-2 also makeacid phosphatase constitutively. Thus, the al-lele phoC-2 is recessive.To check the allelism of phoC-l and phoC-2,

a cross was made to produce a phoC-l/phoC-2diploid. It did not produce acid phosphataseconstitutively. No segregant of this diploid ableto make acid phosphatase constitutively hasoccurred in the 11 asci so far examined. Thus,two mutations, phoC-l and phoC-2, have oc-curred in the same cistron. Additional mutantalleles at this locus have been obtained andcharacterized as above.Tetrad distribution from the diploids heter-

ozygous for lys2 and phoC-2 indicated thatthese two genes are linked. Pooled data on thenumbers of parental ditype, nonparental di-type, and tetratype asci were 47, 0, and 9,respectively. These values indicate a map dis-tance of 11 strains (18) between lys2 andphoC-2 on chromosome II (9).Mutants lacking the activity of the

repressible acid phosphatase. From the aphoC-1 and a phoC-l strains, many mutantslacking repressible acid phosphatase activitywere isolated on low-Pi medium. Several mu-tants were crossed with one of the originalphoC-1 strains according to their mating types.Each hybrid was sporulated and dissected, andit was observed that every ascus showed2+:2- segregation with respect to the forma-tion of the repressible acid phosphatase. Fromthe segregants of each hybrid, both a and amating type clones which lacked the repressi-ble acid phosphatase activity were collected.By use of the complementation test, thesemutants were divided into four groups. Thisexperiment also indicated that each mutantgene is recessive to the corresponding wild-typegene, because the diploid hybrid between dif-ferent classes of mutations was able to producethe repressible acid phosphatase. The numberof mutants so far obtained is listed in Table 1,according to the results of the complementa-

VOL. 113, 1973 731

on February 20, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

J. BACTERIOL.

TABLE 1. Classification of mutants for theproduction of repressible acid phosphatase

Original No. of Total no.Genotype phoC-1 mutants of mutants

strain isolated obtained

phoB 08-M3 1 4P-22-14D 3

phoD 08-M3 1 6P-22-14D 5

phoE 08-M3 1 61P-22-14D 60

phoSa 08-M3 1 16P-22-14D 15

a The phoS mutants showed the same phenotypeas other mutants lacking repressible acid phospha-tase activity, but their genetic behavior was signifi-cantly different; phoS presumably works as a kind ofregulatory gene, as discussed later in the text.

tion tests. These gene loci, controlling theformation of the repressible acid phosphatase,were designated as phoB, phoD, phoE, andphoS. Mutants belonging to phoE were mostfrequently isolated. Some of these mutantstrains were tested for the formation of acidand alkaline phosphatases in cells cultivated inhigh-Pi and low-Pi media. The results shown inTable 2 clearly indicate that the acid phospha-tase activity of all mutants was decreasedseveral-fold below that of the parental strain.The activity of the alkaline phosphatase, how-ever, did not show a significant difference be-tween mutant and parental strains, except forthe phoD and phoS mutants, in which theactivities of both acid and alkaline phospha-tases in low-Pi medium were significantly de-creased simultaneously. Although mutantsclassified as phoS showed a phenotype similarto that of the phoB, phoD, and phoE mutants,it was found that the phoS gene had a certainregulatory function in phosphatase formation,as will be described later. It is noteworthy thatwhen one of the phoE mutants, a phoC-l phoE,was crossed with the a PHOC PHOE strainand the resulting diploid hybrid was subjectedto tetrad analysis, all asci (20 asci) showed theparental ditype tetrad, with respect to thetraits controlled by the phoC and phoE genes.This result indicates that the phoE locus islinked to the phoC locus on chromosome II.The phoB, phoD, and phoS genes segregatedindependently of one another, and also in-dependently of phoC, gall, lys2, and matingtype.To confirm that the constitutive acid phos-

phatase lacking in the phoC mutants could notbe derepressed in low-Pi medium, a cross was

made between the a PHOC PHOD (H-42) andthe a phoC phoD (025-M57) strains. The re-sulting diploid hybrid, P-61, was sporulated,and four-spored asci were dissected. Acid phos-phatase formation was tested in the cells ofeach tetrad clone of the tetratype ascus cul-tivated in nutrient and low-Pi media. Since theamount of the acid phosphatase activity due tothe production of the repressible enzyme is fargreater than that from the constitutive enzymein low-Pi medium, it was easy to classify eachsegregant, by use of the staining method, todetermine whether the activity of acid phos-phatase was due to the PHOC or PHODgenotype. The results of a typical experimentare shown in Fig. 5. It is clear that the PHOCallele is indispensable for the production ofacid phosphatase in nutrient medium and thatthe PHOD allele is required for production ofthe repressible acid phosphatase. The constitu-tive acid phosphatase controlled by the phoCgene could not be derepressed even in low-Pimedium. These results, in addition to thearguments described earlier, may allow theconclusion that the activities of constitutiveand repressible acid phosphatases are at-tributable to different enzyme species.

TABLE 2. Comparison of the formation of repressibleacid and alkaline phosphatases among mutant

strains

Enzyme activitiesa

Acid AlkalineGenotype Strain" phosphatase phosphatase

High- Low- High- Low-Pi Pi Pi Pib

+ (phoC-1)c P-22-14D 0.009 2.41 0.084 0.316phoB 016-M3 0.004 0.010 0.076 0.208

021-M4 0.006 0.330 0.075 0.278phoD 016-M21 0.005 0.011 0.089 0.088

025-M57 0.006 0.020 0.067 0.077026-M68 0.010 0.018 0.069 0.076

phoE 017-M6 0.003 0.160 0.072 0.200021-Ml0 0.004 0.331 0.075 0.280022-Mi 0.004 0.279 0.076 0.283

phoS 022-M2 0.008 0.008 0.078 0.090022-Mll 0.009 0.007 0.089 0.095

phoR P-32-2B 0.268 2.78 0.133 0.450P-49-1C 0.169 3.45 0.264 0.435

a Specific activities of acid and alkaline phospha-tase are expressed as units per milligram of protein.

b Each strain was grown in high-Pi or low-Pimedium at 30 C for 19 to 22 hr with shaking. Extractswere prepared with 5 mm Tris-maleate buffer (pH6.5).

cThe original phoC-1 strain used for isolation ofmutants lacking repressible acid phosphatase.

732 TOH E ET AL.

on February 20, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

ACID PHOSPHATASE MUTANTS IN S. CEREVISIAE

0-200

0.1 00

I-

C.)

4

L)

w

a.

0*000

0.5 000

0.2 5 0

0.0 000 10 20

H 0 UR S

FIG. 5. Time course of the formation of acid phos-phatase in tetrad clones having the PHOC PHOD,PHOC phoD, phoC PHOD, and phoC phoD geno-types. These tetrad clones were obtained from a

diploid hybrid (P-61) of a phoC phoD x a PHOCPHOD combination. Each clone was grown in nutri-ent medium (a) or low-Pi medium (b), and theformation of acid phosphatase was determined atappropriate intervals. Acid phosphatase activity was

determined with a suspension of intact cells as theenzyme source. Specific activity is expressed as

units per unit of optical density at 660 nm. Symbols:0, P-61-1A (phoC PHOD); 0, P-61-IB (PHOCphoD); *, P-61-1C (phoC phoD); and 0, P-61-1D(PHOC PHOD).

Isolation of constitutive mutants from a

strain having the phoC-l genotype. Manymutants which showed the constitutive pheno-type with regard to acid phosphatase formationwere induced from a strain of the phoC-1genotype, such as 08-M3 (a) or P-22-14D (a),by ultraviolet irradiation. These constitutivemutants could have occurred as a result of oneof four possible mechanisms, as summarized inTable 3. The first possibility is that the con-

stitutive production of acid phosphatase isattributable to a true reversion or anothermutation which has occurred in the phoC locusand works as an intragenic suppressor. Thesecond is that an extragenic suppressor muta-tion effective for the original phoC-1 allele has

occurred. The third possible mechanism is thatthe constitutive production of the repressibleacid phosphatase might have resulted from amutation which had occurred in a regulatorgene. The fourth possibility is that of a muta-tion which exerts a function similar to that ofthe Oc mutation in a bacterial system. Thesemutations could be distinguished by examiningthe acid phosphatase formation in the diploidhybrid obtained by a cross between such aconstitutive mutant and the standard phoC-1and PHOC strains and by tetrad analysis ofthese hybrids. In the case of the first possibili-ty, the constitutive character of the mutant isdominant and the diploid hybrid made be-tween this mutant and the PHOC strain shouldshow essentially 4+:0- segregation in eachascus with respect to the constitutive formationof acid phosphatase. If a mutation caused anextragenic suppressor for the phoC-1 allele andif this suppressor was dominant, a diploid withthe phoC-1 strain should produce some amountof acid phosphatase constitutively. The diploidhybrid obtained by a cross between this type ofmutant and the PHOC strain should show4+:0-, 3+:1-, and 2+:2- segregation withrespect to the constitutive formation of acidphosphatase. If a mutation occurred as a resultof the third possibility, the constitutiveness isrecessive and the diploid hybrid obtained by across between the phoC-1 strain should notproduce the acid phosphatase on nutrient orhigh-Pi media. The diploid hybrid between thistype of mutant and the PHOC strain shouldshow 4+:0-, 3+: 1-, and 2+: 2- segregationfor the formation of acid phosphatase inhigh-Pi medium. If a mutation occurred as aresult of the fourth possibility, the diploidhybrid should produce the acid phosphataseconstitutively and the constitutive trait of thismutant would be dominant. The diploid hybridobtained by a cross between this mutant andthe PHOC strain should segregate a clonewhich is unable to produce acid phosphataseconstitutively. Although the genetic character-istics of the mutants of this class are quitesimilar to those of the mutants of the secondclass, it may be possible to distinguish them byexamining the enzymatic characteristics of theacid phosphatase produced in high-Pi mediumand by genetic analysis of hybrids obtained by across between such a mutant and the phoCphoB, phoC phoD, or phoC phoE strains.

Since the constitutiveness is recessive only inthe third class of mutant, constitutive mutantswere divided into two groups, dominant andrecessive, by examining the phenotype of thediploid hybrids obtained by crossing each mu-

VOL. 113, 1973 733

on February 20, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

TABLE 3. Representation of the hypothesis for establishing the genetic nature of constitutive mutation fromthe phoC-1 mutant strain

Phenotype of No.Possible genotype diploid hybrid Possible segre- of mu-

Hypothesis of constitutive in high-Pi gation pattern in tantsrevertant mediuma high-Pi medium iso-

lated

1. Reversion or intragenic sup- PHOC x PHOC: +t 4+:0- 6pressor at the phoC locus or x phoC-1 : + 2+ :2-

phoC-1 phoC-SUPc2. Extragenic suppressor to the phoC_1 SUp x PHOC + 4+ :0-, 3+:l1-, 2+ :2- 0

phoC-1 allele phoC_I : +d 2+:2 -3. Mutation in the regulator gene x phoC : +d 2 :2-

for the repressible enzyme phoC-1 phoR x PHOC + 4+:0-, 3+: 1-, 2+:2- 194. Mutation at the operator for x phoC-1: - 2+ :2-

the repressible enzyme phoC-1 phoOc x PHOC + 4+ :0-, 3+:1-, 2+ :2- 1x phoC-1: + 2+:2-

aEach mutant was crossed with the standard PHOC or phoC-1 strain. Each hybrid obtained was tested forthe ability to produce acid phosphatase constitutively and was subjected to tetrad analysis after sporulation.

Phenotype on high-Pi medium by the staining method; + and - indicate the ability and inability toproduce acid phosphatase in this medium, respectively.

c phoC-SUP is an intragenic suppressor for the phoC-1 allele.d A constitutive phenotype is expected if the suppressor is dominant.

tant with the phoC-1 strain. Dominant con-stitutive mutants were then classified accord-ing to the above criteria. Among several con-stitutive mutants, 012-M15 (a), 027-M66 (a),and 015-M2 (a) were used in further studies as

representative; the first two were dominantand the last one was recessive. These mutantswere crossed with the original phoC-1 strains,and the resulting diploid hybrids were sporu-lated and dissected. All asci showed 2+:2-segregation with respect to the formation ofacid phosphatase in high-Pi medium. From theabove crosses involving 012-M15 and 015-M2,constitutive segregants were isolated; they werecrossed to the wild-type strain, H-42 (PHOC),and the resulting diploids were subjected totetrad analysis. Mutant 027-M66 was crosseddirectly to H-42. Essentially all asci (33 of 34asci) showed 4+:0- segregation with respectto the constitutive formation of the acid phos-phatase in the cross involving the 012-M15derivative. Thus, constitutivity in 012-M15 isdue to reversion or intragenic suppression atphoC. In the case of the cross involving the015-M2 derivative, 4+:0-, 3+:1-, and2+:2- segregations appeared with respect tothe formation of acid phosphatase on high-Pimedium, as shown in Table 4. Since constitu-tivity in 015-M2 is recessive and since therepressible segregants were segregated in thecross to the wild-type (PHOC) strain, we havetentatively concluded that constitutivity in015-M2 is due to mutation in a regulator gene,

designated phoR. The data in Table 4 alsoindicated that constitutivity in 027-M66 is dueeither to extragenic suppression for phoC-1 orto 0-type mutation for the repressible acidphosphatase production, since the repressibleclones were segregated from the constitutivehybrid and constitutivity was dominant. Infurther testing, 027-M66 was crossed with016-M3 (a phoC-1 phoB), and the diploidhybrid was sporulated and dissected. If theconstitutivity in 027-M66 is attributable to thesecond possibility illustrated in Table 3,2+:2- segregation will be expected in eachascus on high-Pi medium. The diploid hybrid,however, showed a 2+: 2- (1 ascus) and a1 +: 3- (12 asci) segregation with respect to theformation of acid phosphatase in high-Pi medi-

TABLE 4. Tetrad segregations in crosses betweenPHOC and the constitutive mutants from the phoC-1

strain

Combination of hybrid Segregation in asciComblnalonofhybrd( +: - )

Mutanta PHOC 4:0 3:1 2:2 1:3 0:4

012-M15* H-42 34 1 0 0 0015-M2* H-42 2 18 5 0 0027-M66 H-42 3 12 0 0 0

a The constitutive mutants from the phoC-1 strain.Strains marked by an asterisk were replaced by tblgirderivative in the practical crosses.

b The formation of acid phosphatase was tested onhigh-Pi medium by the staining method.

734 TOH-E ET AL. J. BACTERIOL.

on February 20, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

ACID PHOSPHATASE MUTANTS IN S. CEREVISIAE

um; all asci showed a 2+:2- segregation inlow-Pi medium. This result suggests that themutation in 027-M66 belongs to the fourthclass of the mutation described in Table 3 andthat the phoB locus was not linked to thesupposed phoO locus. Further characterizationof this mutant is now underway. The constitu-tive mutants so far obtained were classifiedaccording to these criteria, and the number ofmutants belonging to each class is listed in thelast column of Table 3.

It was found that the formation of both acidand alkaline phosphatases was significantlyaffected by the mutation classified as phoDand phoS, as shown in Table 2. This observa-tion suggests that these repressible phospha-tases may be under the same controllingmechanism or that both enzymes share acommon factor. Therefore, the question ofwhether the repressible acid and alkaline phos-phatases are also controlled by the phoR genewas investigated. Two independently isolatedphoR mutants were grown in high-Pi andlow-Pi media, and the activity of both acid andalkaline phosphatases was assayed. The resultspresented in Table 2 clearly indicate that thephoR mutants produce a significant amount ofalkaline phosphatase as well as acid phospha-tase in high-Pi medium. Furthermore, fullderepression was observed only when the cellswere cultured on low-Pi medium. The phoRmutants showed a higher activity for bothenzymes in low-Pi medium than did the wild-type (phoC-1) strain.Role of the phoS gene. If a gene or a gene

system which controls the synthesis of anactive repressor is contained in the regulatorysystem of the acid phosphatase formation,some of the mutants which produce the activerepressor constitutively will show the samephenotype as a mutant with respect to thestructural gene(s). Furthermore, a double mu-tant for such a gene and for phoR shouldproduce the acid phosphatase constitutively.This possibility was examined by crossing themutant lacking the repressible acid phospha-tase with the phoR strain. Tetrad segregantsfrom each cross were tested for acid phospha-tase formation on a plate containing high-Pi orlow-Pi medium by the staining method. Twotypes of patterns in tetrad segregations fromthese crosses were observed (Table 5). One ofthe segregation patterns is characterized by a2+:2- segregation of the acid phosphataseformation in low-Pi medium; some of theseacid phosphatase producers are constitutive.This type of segregation would be expected ifthe phoR phoB, phoR phoD, and phoR phoE

genotypes could not produce the acid phospha-tase in the presence or absence of inorganicphosphate in the medium. Another type ofsegregation was observed in the hybrid, P-54,from the cross between the phoR and phoSmutants and is characterized by more than twospore cultures which produce acid phosphataseon low-Pi medium in each ascus; two of themwere always constitutive, as shown in Table 5.This type of segregation was most adequatelyexplained by the presence of another type ofregulatory gene, phoS. To confirm the abovehypothesis, one of the tetratype asci, consistingof P-54-5A, -5B, -5C, and -5D spore cultures,was selected for further study, and the geno-type of each strain was determined (Table 6).(Since all of the strains discussed below con-tained phoC-1, this genetic symbol has beenomitted in the following descriptions.) In thistetratype ascus, four genotypes should be pos-sible, namely, phoR PHOS, PHOR phoS, phoRphoS, and PHOR PHOS; the first two areparental types and the last two are recombi-nants. Judging from the phenotypes of thesegregants, P-54-5A, -5B, and -5D showed theparental phenotypes, and among them it canbe easily supposed that the genotype ofP-54-5B should be PHOR phoS because it doesnot produce acid phosphatase on either high-Pior low-Pi medium. Another clone, P-54-5C,behaved like the wild-type clone and should berecombinant having the PHOR PHOS geno-type. Therefore, either the P-54-5A or the -5Dclone has the genotype corresponding to phoRPHOS and the other has the genotype phoRphoS. Since the phoR and phoS alleles arerecessive to the corresponding wild-type allele,as mentioned above, the phoR PHOS straincan be distinguished from the phoR phoSstrain by examining the phenotype of diploidsobtained by crossing each of these strains to thePHOR phoS strains; the phoR PHOS/PHORphoS diploid should produce acid phosphatasein low-Pi medium but not in high-Pi medium.On the other hand, the phoR phoS/PHORphoS diploid should not produce the enzyme ineither low- or high-Pi media. These possibili-ties were tested by backcrosses of the tetradclones, P-54-5A or -5D, to the original phoSstrain, 016-M6. The ability of these diploidhybrids to produce the acid phosphatase wasexamined on high-Pi and low-Pi media by thestaining method. Results shown in Table 6clearly indicate that the diploid obtained by across between P-54-5A and 016-M6 could notproduce the acid phosphatase either on high- orlow-Pi media. On the other hand, the diploidobtained by crossing P-54-5D to 016-M6 could

VOL. 113, 1973 735

on February 20, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

TABLE 5. Tetrad segregations in crosses between thephoCphoR strain and thephoC phoB, phoC phoD, phoCphoE, or phoC phoS straina

Segregation in ascib

Hynborid Combination On low-Pi + + + + + + + - + +-- + +-- + +_Onhigh-Pi -+ - -++-- ++ - +-

P-44 P-32-2Bc x 016-M3 (phoC 0 0 2 19 2phoB)

P-45 P-32-2B x 016-M21 (phoC 0 0 4 6 9phoD)

P-75 P-32-2B x 017-M6 (phoC 0 0 3 8 3phoE)

P-54 P-32-2B x 016-M6 (phoC 10 39 17 0 0phoS)

a Four mutant strains, 016-M3, 016-M21, 017-M6, and 016-M6, were used as the representatives of thephoB, phoD, phoE, and phoS mutants, respectively.

b Each tetrad segregant was tested for its ability to produce acid phosphatase in high-Pi and low-Pi mediaby the staining method.

c P-32-2B has the phoC phoR genotype.

TABLE 6. Determination of the genotype of each tetrad clone in one of the tetratype asci from diploid P-54

Acid phosphatase formation

Segregant Mating In haploid segreganta In diploid hybridb Expectedtype~~ ~ ~ ~ ______genotype

High-Pi Low-Pi High-Pi Low-Pi

P-54-5A a +c (0.16)d + (0.41) _ _ phoR phoSP-54-5B a - (0.00) - (0.00) NTe NT PHORphoSP-54-5C a - (0.00) + (0.31) NT NT PHORPHOSP-54-5D a + (0.17) + (0.43) + phoR PHOS

a Phenotype of each tetrad segregant in high-Pi or low-Pi medium.bP-54-5A and -5D were constitutive and they were backcrossed with 016-M6 (a phoC PHOR phoS).c Phenotype determined by the staining method.d Specific activity of acid phosphatase (units per unit of optical density at 660 nm) determined with a

suspension of intact cells as the enzyme source. Each clone was grown in high-Pi or low-Pi medium for 20 hr at30 C with shaking.

e Not tested.

produce the enzyme, but only in low-Pi me-dium. Therefore, the genotypes of P-54-5A and-5D were labeled phoR phoS and phoR PHOS,respectively. It should be emphasized that thephoR phoS genotype results in a constitutiveformation of the acid phosphatase. To make aquantitative comparison of the formation of theacid phosphatase among these strains, eachstrain was cultivated in low- or high-Pi me-dium at 30 C for 20 hr with shaking. The activ-ity of the acid phosphatase was assayed with asuspension of intact cells as the enzyme source;the specific activity was expressed as units perunit of optical density at 660 nm. The resultspresented in Table 6 indicate that the enzymeformation of each strain reflects the character-istics shown by the staining method. In addi-tion, both of the constitutive strains showed ahigher acid phosphatase activity in low-Pimedium than in high-Pi medium.

DISCUSSIONIt has been reported in other organisms, such

as E. coli (8) and N. crassa (20), that derepres-sion of phosphatase has resulted from the denovo synthesis of the enzyme. It seems reasona-ble to assume that that is also the case in S.cerevisiae. The haploid strain H-42 producedacid phosphatases in both the repressed andderepressed conditions. These phosphatasesshowed a difference in their Km values forp-nitrophenylphosphate and in their heat sta-bility. It was suggested by these observationsthat these phosphatases were different species.This suggestion was also supported by geneticstudies of certain mutants lacking one of theseacid phosphatases. The genetic locus control-ling the formation of the constitutive acidphosphatase was designated as phoC, and itwas found that the phoC locus is linked withthe Iys2 locus on chromosome II (9).

736 TOH-E ET AL. J. BACTERIOL.

on February 20, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

ACID PHOSPHATASE MUTANTS IN S. CEREVISIAE

The mutants lacking repressible acid phos-phatase were divided into four classes; muta-tions at two of the loci, phoD and phoS, led toinability to derepress alkaline as well as acidphosphatases. The mutations at the phoB andphoE loci, however, did not have any signifi-cant effect on the activity of the alkaline phos-phatase. The phoA - mutants described bySchurr and Yagil (17) were also defective inboth acid and alkaline phosphatases. Theseauthors suggested the existence of a commonregulatory mechanism or a common polypep-tide to explain the pleiotropic effect of thesemutations. Since mutations at the phoD andphoS loci showed a similar pleiotropic effect onthese enzymes, they might correspond to thephoA - mutants. However, as will be discussedlater, the phoS gene is not a structural gene forthe repressible acid phosphatase. It is possibleto speculate that the phoD gene might act inthe manner suggested by Schurr and Yagil (17)or might have the positive function of derepres-sing these phosphatases. To date, however, it isstill unclear which mutation is in the structuralgene(s).

Genetic analyses of the constitutive mutantsderived from the phoC mutant showed that theconstitutive synthesis of acid phosphatase wasthe result of several distinct genetic events,such as the reversion to PHOC, mutationsoccurring in the regulator gene (phoR), andmutations occurring in an unknown gene otherthan phoC and phoR. Since the phoR mutantallele was recessive to the corresponding wild-type allele, and the acid phosphatase producedby the phoR mutant grown in high-Pi mediumwas similar to the repressible acid phosphatasewith respect to heat stability and the Km valuefor p-nitrophenylphosphate (unpublished da-ta), the simplest interpretation of the functionof the phoR gene is that it specifies the repres-sor molecule for the synthesis of the repressibleacid phosphatase. It was also found that thephoR mutation exerts a pleiotropic effect onboth acid and alkaline phosphatases, as de-scribed by Schurr and Yagil (17).

In bacteria, the site on which repressor acts iswell characterized as an operator locus (1); it islocalized at one end of the structural gene andbinds the repressor molecule reversibly. Such acontrolling site may exist in eukaryotic cells,although its presence has not been confirmedas yet, except for a report on the c-GAL4complex in a galactose-utilizing system in S.cerevisiae by Douglas and Hawthorne (5). Inthis respect, it would be interesting to studythe nature of a mutant which preserves thephoC PHOR genotype and shows constitutive

synthesis of the acid phosphatase. We suc-ceeded in isolating such a mutant from thephoC-1 strain; its characterization is now un-derway.The observation that the phoC phoR phoS

strain produced acid phosphatase in both high-and low-Pi media indicates that the phoS geneis not a structural gene but, rather, a kind ofregulatory gene. This result also eliminates thepossibility that the product of the phoS genemight positively control the synthesis of therepressible acid phosphatase directly, because,if this were the case, the phoC phoR phoSstrain would not produce acid phosphatase inhigh- and low-Pi media. The assumption thatthe phoS gene has a regulatory function isconsistent with the fact that the phoS mutantcould not derepress both acid and alkalinephosphatases simultaneously. The phenotypeof the phoC phoR phoS strain suggests that thephoS gene controls the synthesis of repressiblephosphatases via the expression of the phoRgene. At present, the simplest interpretation ofthe data from tetrad analysis of the diploidP-54 and the phenotype of the phoS mutant isthat: (i) the phoS mutant has a normal struc-tural gene(s) for both acid and alkaline phos-phatases, and is defective in regulatory geneother than phoR, (ii) the phoS gene controlsthe expression of the phoR gene via a cer-tain cytoplasmic substance somehow in-volved with inorganic phosphate, and (iii) thePHOR product represses the synthesis of therepressible acid and alkaline phosphatases re-gardless of the presence or absence of inorganicphosphate. According to this model, the pheno-type of the phoS and the phoR phoS mutantscan be explained as follows: (i) the phoSmutant produces the repressor for the struc-tural gene(s) constitutively and cannot dere-press these phosphatases, and (ii) although thephoR phoS double mutant produces the repres-sor for the structural gene(s) constitutively, therepressible phosphatases can be produced be-cause of an inactive repressor resulting from amutation occurring in the phoR gene.Assuming that the phoS mutation results in

a constitutive synthesis of the phoR productand that inorganic phosphate only affects theexpression of the phoR gene, the extent of theacid phosphatase produced by the phoR phoSdouble mutant in low-Pi medium should be thesame as that produced by the phoR mutant inhigh-Pi medium. However, as shown in Table6, the phoR phoS double mutant produced alarger amount of the acid phosphatase inlow-Pi medium than the phoR mutant did inhigh-Pi medium. This fact indicates that inor-

VOL. 113, 1973 737

on February 20, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

TOH-E ET AL. J. BACTERIOL.

ganic phosphate also affects the formation ofacid phosphatase in ways other than the ex-pression of the phoR gene.

LITERATURE CITED1. Beckwith, J. R., and D. Zipser (ed.). 1970. The lactose

operon. Cold Spring Harbor Laboratory, Cold SpringHarbor, N.Y.

2. Blum, J. J. 1956. Observation of the acid phosphatasesof Euglena gracilis. J. Cell Biol. 24:223-234.

3. Burkholder, P. R. 1943. Vitamin deficiencies in yeasts.Amer. J. Bot. 30:206-211.

4. Dorn, G. 1965. Genetic analysis of the phosphatases inAspergillus nidulans. Genet. Res. 6:13-26.

5. Douglas, H. C., and D. C. Hawthorne. 1966. Regulationof genes controlling synthesis of the galactose pathwayenzymes in yeast. Genetics 54:911-916.

6. Garen, A., and H. Echols. 1962. Properties of tworegulating genes for alkaline phosphatase. J. Bacteri-ol. 83:297-300.

7. Garen, A., and H. Echols. 1962. Genetic control ofinduction of alkaline phosphatase synthesis in E. coli.Proc. Nat. Acad. Sci. U.S.A. 48:1398-1402.

8. Garen, A., and C. Levinthal. 1960. A fine-structuregenetic and chemical study of the enzyme alkalinephosphatase of E. coli. I. Purification and characteri-zation of alkaline phosphatase. Biochim. Biophys.Acta 38:470-483.

9. Hawthorne, D. C., and R. K. Mortimer. 1968. Geneticmapping of nonsense suppressors in yeast. Genetics60:735-742.

10. Horiuchi, T., S. Horiuchi, and D. Mizuno. 1959. Apossible negative feedback phenomenon controllingformation of alkaline phosphomono-esterase in Esch-erichia coli. Nature (London) 183:1529-1530.

11. Lindegren, C. C., and G. Lindegren. 1943. A new methodfor hybridizing yeast. Proc. Nat. Acad. Sci. U.S.A.29:306-308.

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

13. Nyc, J. F. 1967. A repressible acid phosphatase inNeurospora crassa. Biochem. Biophys. Res. Commun.27:183-188.

14. Nyc, J. F., R. J. Kadner, and B. J. Crocken. 1966. Arepressible alkaline phosphatase in Neurospora crassa.J. Biol. Chem. 241:1468-1472.

15. Oda, M., and K. Wakabayashi. 1954. Yeasts isolatedfrom the sake-moromi (mash of sake) fermented bykimoto. J. Ferment. Technol. 32:289-294.

16. Schmidt, G., G. Bartsch, M. C. Lamont, T. Herman,and M. Liss. 1963. Acid phosphatase of baker's yeast:an enzyme of the external cell surface. Biochemistry2:126-131.

17. Schurr, A., and E. Yagil. 1971. Regulation and charac-terization of acid and alkaline phosphatase in yeast. J.Gen. Microbiol. 65:291-303.

18. Shult, E. E., and C. C. Lindegren. 1956. Mappingmethods in tetrad analysis. I. Provisional arrangementand ordering of loci preliminary to map constructionby analysis of tetrad distribution. Genetica28:165-176.

19. Suomalainen, H., M. Linko, and E. Oura. 1960. Changesin the phosphatase activity of baker's yeast during thegrowth phase and location of the phosphatases in theyeast cell. Biochim. Biophys. Acta 37:482-490.

20. Toh-e, A., and T. Ishikawa. 1971. Genetic control ofsynthesis of repressible phosphatases in Neurosporacrassa. Genetics 69:339-351.

21. Torriani, A. 1960. Influence of inorganic phosphate inthe formation of phosphatases by Escherichia coli.Biochim. Biophys. Acta 38:460-479.

738

on February 20, 2020 by guest

http://jb.asm.org/

Dow

nloaded from