The Bacillus subtilis phoAIV gene: effects of in vitro inactivation on total alkaline phosphatase production

Gene, 96 (1990) 95-100 Elsevier

95

GENE 03781

The Bacillus subtilis phoAIV gene: effects of in vitro inactivation on total alkaline phosphatase production (Insertional mutagenesis; pho regulon; recombinant DNA; phosphate starvation)

Niekolas V. Kapp, Charles W. Edwards, Ruth S. Chesnut and F. Marion Hulett Laboratory for Molecular Biology, Department of Biological Sciences, University of Illinois at Chicago, Chicago, IL 60680 (U.S.A.), Tel. (312)996-5460 Received by J.A. Hoch: 31 May 1990 Revised: 13 July 1990 Accepted: 7 August 1990

SUMMARY

A degenerative oligodeoxyribonucleot:~deprobe deduced from the first 19 amino acids of the mature alkaline phosphatase IV (APase IV) protein was used to clone a DNA fragment internah to the coding region of the phoAIV gene of Bacillussubtilis. An insertional mutation was constructed in the phoAIV locus using the integrative plasmid, pJM 103, containing the cloned DNA fragment. The straiv with the interrupted phoAIV gene showed no detectable APase IV product on Western-blot analysis. The impa~t of t~,e phoAIV interruption on total APase produ,~tion in B. subtilis 168 was analyzed under both phosphate starvation aad sporulation culturing co,9ditions. The mutation in phoAIV reduced total APase-specific activity by 75% in phosphate-starved cells, and resulted in the elimination of a salt-extractable membrane APase, as well as the secreted APase IV. Analysis of this membrane APase indicated that it is a phoAIV gene product which is localized within the membrane fraction of the lysed cell :~mdnot secreted. There was no effect on the production of sporulation APase. The phoAIV::pJMl03 insertion was mapped and determined to be located at approx. 73 ° on the B. subtilis 360 ° chromosome.

INTRODUCTION

The production of APase in B. subtilis is induced when cell growth is limited by inorganic phosphate concentrations of 0.1 mM or less. This APase synthesized in undifferentiated vegetative cells is referred to as vegetative APase

Correspondence to: Dr. F.M. Hulett, Laboratory for Molecular Biology (MC/067), University of Illinois at Chicago, P.O. Box 4348, Chicago, IL 60680 (U.S.A.) Tel. (312)996-2280; Fax (312)413-2691. Abbreviations: aa, amino acid(s); APase IV, alkaline phosphatase IV;

APIV, totally degenerative oligo encoding the N terminus of APase IV; bp, base pair(s); Cm, chloramphenicol; kF, ,,:'_'!obase(s)or 1000 bp; nt, nucleotide(s); oligo, oligodeoxyribonucleotide; PAGE, polyacrylamidegel electrophoresis; phoAIV,gene encoding APase IV; Pi, inorganic phosphate; SDS, sodium dodecyl sulfate; wt, wild type; tsp, transcription start point(s); XP, 5-bromo-4-chloro-3-indoxyl-phosphate-p-toluidine; ::, novel joint (f,,,~;.'.,1.i._sf'rtion). 0378-1119/90/$03.50 O If90 Elsevier Science Publishers B.V. (BiomedicalDivision)

(Hulett and Jensen, 1988; Piggot and Coote, 1976) and is distinguished from sporulation APase which is produced during sporulation even in the presence of high Pi concentrations. Various ioci involved in regulation of vegetative and/or sporulation APase have been identified. Vegetative APase in B. subtilis is controlled by the pho regulon, including the regulatory factors phoP and phog (Miki et al., 1965). The genes encoding PhoP and PhoR have been cloned and sequenced ($eki et al., 1987; :~S8). By analogy to other two-component regu'atory systems, PhoR is tile sensor that responds to low Pi concentrations, and PhoP is the activator that stimulates transcription of APase and other pho regulon genes (Nixon et al., 198g: Gross et al., 1989). Mutations in two other regulatory loci, phoT and phoS (Piggo~ and Taylor, 1977; Piggot et al., 1981), allow pro~.tuction of vegetative APase in the presence of Pi concentrations that would repress APase synthesis in wt cells. Sporuiation

96 APase is affected by mutations in several sporulation loci: spollA, spolIE, spollF and spoHG (Figgot and Coote, 1976). While vegetative APase productiot_~ is controlled by the pho regulon and sporulation AP~,se production is regulated by genes that regulate sporulation, there is some overlap between the two regulatory systems. Hulett and Jensen (1988) have shown that spoOH, which encodes a sigma factor required for initiation of sporulation, and spoOA, an early sporulation regulatory gene, are required for vegetative APase production. Recently, we reported the isolation of two APase enzymes, designated APase III and APase IV, from the culture medium of phosphate-starved vegetative cells (Hulett et al., 1990). The APase III protein i.s a 90-kDa dimer. APase IV appears to be unique among APases in that i~ elutes from a sieving column as an active 45-kDa monomer (Hulett et al., 1990). Both B. subtilis APases share homology with E. coil APase encoded by the phoA gene. We have shown that the two vegetative B. subtilis APase proteins are encoded by different genes. The gene for APase IH has been cloned, a mutant has been constructed and the impact ofinactivation of the APase III gene on total APase production in vegetative cells during phosphate starvation and in sporulating cells has been described (Bookstein et al., 1990). The mutation in APase III reduced the total vegetative APase specific activity by approximately 40% and sporulation APasc, specific activity by approx. 45 ~ . The APase III mutation caused no decrease in sporulation in Schaeffer's sporulation medium. The aims of the present study were to clone a D N A fragment itaternal to the coding region of the APase IV-encoding gene (phoAIV), to construct a phoAIV structural gene mut~ ~t and determi~e which APases are encoded by this gent

RESULTS AND DISCUSSION (a) Clot'ing o f an internal f r a g m e n t o f the phoAl~" gene

Th~s laboratory has previously described the isolation of the APase IV prot.~in from a phosphate-starved vegetative culture of B. subgias (Hule~t et al., 19q0). The first 19 aa of the APase !V protein were sequenced and an oligo (APIV) was constructed that hybridized to two different restriction fragments in most digests orB. subtilis cl:romosornal D N A (Hulett et al., 1990). In digests with EcoRI, HindlIl or PstI, APIV hybridized to the same two fragments that hybridized to a degenerative oligo based on the N-terminal sequence of APase III (Hulett et al., 1990). However, APIV hybridized to a 300-bp Haelll fragment not detected by the APase III oligo (data not shown). APIV was used to screen a B. subtilis HaelIl chromosomal D N A sub-library in pJM103. Colony hybridization at greater than 90~o stringency identitied six clones out of 3000. The clones were

identical when sequenced, and their nt sequence predicted an identical match to the first 22 aa of the purified mature APase IV protein (Fig. 1). The sequence of this HaeIII fragment does not contain a tsp or sites for translational starts, and has three codons 5' to the sequence of the mature protein. The aa encoded by these three codons are A-S-A. These three aa immediately precede the mature

GCC AGe GCC AAA AAA CAA GAC AAA GeT GAG ATe AGA AAT Ala Set A1a LYs Lvs ,Gln ASP Lvs Ala GIu Ile Ara ASh

39 13

GTC ATT GTG ATG ATA GGC GAC GGC ATG GGG ACG CCT TAC _Val Ile Val Met Ile GIv ASP Glv Met GIv Thr Pl-q Tyr

7B 26

ATA AGA GCC TAC CGT TCC ATG AAA AAT AAC GGT GAC ACA Ile Arg Ala Tyr Arg Set Met Lys Ash ASh Gly Asp Thr

117 39

CCG AAT AAC CCG AAG TTA ACA GAA TTT GAC eGG AAC CTG Pro Asn Asn Pro Lys Leu Thr "i., Phe Asp Arg ASh Leu

156 52

ACA GGC ATG ATG ATG ACG CAT CCG GAP GAC CCT GAC TAT Thr Gly Met Met Met Thr His Pro Asp Asp Pro ASp Tyr

195 65

AAT ATT ACA GAT TeA GCA GCA Gee GGA ACA GCA TTA GCG Asn ~le Thr Asp Set Ala Ala Ala G1¥ Thr Ala Leu Ala

234 78

ACA GGC GTT AAG ACA TAT AAC AAT GCA ATT GGC GTC GAT Thr Gly Val Lys Thr Tyr Ash Ash Ala Ile GI¥ Val Asp

273 91

AAA ~\C GGA AAA AAA GTG AAA TCT GTA CTT GAA GAG GCC Lys Ash G1¥ Lys Lys Val Lys Set Val Leu Glu Glu Ala

312 104

Fig. 1. Nucleotide sequence of the 312-bpHaelll fragment. A fragment internal to the phoAIV gcnewas clonedby colonyhybridizationusingthe total degenerative oligoAPIV (Hulett et al., 1990) as a probe. A library was constructed by ligating Sinai-cut plasmid pJMl03 (Ferrari and Hoch, 1989)to B. subtilis 168 HaeIII.cut chromosoma! DNA fragments 200-400 bp in length. This library was transformed into E. coli DHS~. Colonies were transferred to Amersham Hybond-N nylon membranes and screened according to Lampeet ai. (1988), as modifiedby Bookstein et ai. (1990). Twelve 85-mm filter disks were prehybridized in 10 ml of buffer, 107cpm of the riP-end-labeledoligo were added directly to the prehybridizationbuffer, and the colonyblots were hybridizedovernight a' 37°C. At lea~t three washes were done at 45°C for 30 m~neach. After initial autoradiography to allow alignment of filters with colonies, the filters were rewashed at + 5°C increments to idendfy only the most stringent hybrids. Conditions of stringencywere initiallydetermined by the formula Tm = 81.5°C + 16.6 x logM + 0.41(%G + C) - 500/n (Meinkoth and Wahl, 1984). Tm represents the melting temperature of duplexed DNA, M is the ionic strength of the buffer in tool/liter, and n refers to the shortest possiblelengthofthe duplex,in this case the 57-mer. The valueused for %G + C was (43%) was based on the averag~content of ~acillus DNA, since the G + C contents of the oli~qosin the mixture varied considerably.The maximumrate of hybridizationwas considered to be 25°C below T m. These conditionsapplied to assumptionsof 100% homology. To determine iowor stringencies,the method of Lathe (1985) was used. The change in Tm was estimated to equal -(100.h)t, where h is the 70 homologybetween sequencesand t i~ the change in temperature per% nonhomology.The value for t of 1.2°C per% homologywas that estimated empiricallyby Lathe (1985). By these calculations,53°C was considered to represent 90% stringency,and 66°C was considered to lepresent 100% stringency.Six positive clones were sequenced directly using Sequenase (U.S. Biochemical)according to the manufacturer's instructions and found *o be identical.The first line givesthe nt sequence of the entire HaeIII fragment.The second line gives the predicted translation, in frame with the known aa sequence of the mature APase IV protein. The N-terminal aa sequenceof the mature protein as determined by Hulett et al. (1990) is underlined. The deduced aa sequence begins with A-S-A, which may be the end of the signal sequence. The partial nt sequence and ~74erminaJ:~a sequence have been deposited with the GenBank under accession number M37165.

97 protein sequence of APace III (Bookstein et al., 1990). Because A P a s c IV !s a secreted protein (Hulett et at., 1990), these three codons may be part of a signal sequence. The HaellI fragment is not large enough to encode the complete A P a s e IV protein and the sequence does not contain a stop codon. Thus the rlaelll fragment is internal to the phoAIV structural gene and contains neither the promoter nor the 3' end of the A P a s e IV structural gene.

(b) Construction of a phoAIV mutant To facilitate further study of A P a s e IV, we constructed a strain containing a selectable m a r k e r in the phoAIV gene using plasmid insertional mutagenesis. When a fragment internal to the A P a s e IV-encoding region is successfully integrated by Campbell-type recombination into the B. subtilis chromosome, the result is the generation of two incomplete copies of the phoAIV transcriptional unit ~s illustrated in Fig. 2. Competent cells of B. subtilis were transformed with. plasmid pCE413 containing the internal HaeIII fragment of phoAIV in vector p J M 103 and selected for C m resistaace, integration of plasmid pCE413 into the B. subtilis c h r o m o s o m e was confirmed by Southern-blot analysis ( d a t a not shown).

(c) Western-blot an~.lysis of phoAIV mutants Both the wt and the phoAIV mutant were grown under the two conditions that induce production of B. subtilis APases: low Pi defined medium to induce the vegetative APases, and Schaeffer's sporulation medium to induce sporulation APases. Crude cell lysates were subjected to S D S - P A G E , blotted to lmmobilon membrane (Millipore), and incubated with polyclonal antibody raised against purified A P a s e III or A P a s e IV protein. The conditioi~s for transfer and for immunostaining were as described by Hulett et at. (1990). Fig. 3a and b shows Western. blots of cell lysates taken from vegetative growth. Analysis of these d a t a indicate that the phoAIV mutants do not produce A P a s e IV protein (Fig. 3a), but that the phoAIV mutation does not affect production of A P a s e III (Fig. 3b). Neither the wt nor the phoAIV mutants produce A P a s e IV in sporulating cells (Fig. 3c). Further, we were not able to isolate A P a s e IV protein from wt or phoAIV mutant cells grown in Schaeffer's sporulation medium (unpublished data). These data indicate that A P a s e IV induction is unique to vegetative, phosphate starved cells and does not occur in sporulation. A P a s e III is produced in both Pi starved vege-

1234567 CmR

A

~

.911

Ap R

B Bg IIf/~coRr~ ( ] I i

II

i

C

HaelII

Haelll

I ",," \,"

I

3'.']

Chromosomal

APaseIV-codingregion BglIII

HaeIII HdeIIIAp , , -

R

"

R Cm HaeIII n~.rTr. HaeIII D~:~ . -..-

HindIII

Fig. 2. Insertional mutagenesis of the phoAIV gene. The 312-bp Haelll DNA fragment was cloned into the Sma I site of the integration vector pJMl03 (Ferrari and Hoch, 1989), producing plasmid pCE413 (line 1). The HaeIII fragment was determined by sequencing to contain DNA from the internal portion of the phoAIV gene. When pCE413 was transform_ed (Dubrau and Davidoff-Abelson, 1971) into B. subtilis 1A!84 (pigY trpC2) (BSGSC) it integrated into the chromosome via homologous recombination. The recipient of the integrated plasmid contained two copies of the original HaeIII fragment, denoted by hatched boxes on line 3. The integration contained the 5'-end of the phoAIV gene followed by pJMl03 DNA (arc), the repeated HaelII fragment, and the Y-end of the APase IV-coding region. The result of the integration is two copies of the ItaeIII internal fragment, one adjacent to the 5' region and the other adjacent to the 3' region. In each case, DNA necessm:yfor production ofthe APase IV protein is missing, so that the result ofthe integration is a nonfunctional phoAIV gene that is tagged by a CmR marker.

e

't9

Fig. 3. Western-blot analysis of APase production during vegetative or sporulative growth. Vegetative samples were taken at the time points shown in the growth curve in Fig. 4. Lanes 1, 3 and 5 contain cell lysates ofB. subtilis168 taken at 4, 8, and 12 h ofvegetative growth, respectively. Lanes 2, 4 and 6 contain cell lysates of the phoAIV mutant t~ken at 4, 8, and 12 h, respcctively. The arrowheads indicate the position to which the 43-kDa protein marker ovalbumin migrated. Equal amounts of lysate were loaded on blots A avd B. Lane 7 is a positive control for blotting efficiency and antibody cross-reactivity. Lane 7 contains a previously analyzed B. subtilis 168 lysate taken at 12 h of growth in Prhmiting defined medium known to contain the ~,Pase IIl and APase IV proteins. Blot A was incubated with antibody raised against the purified APase IV protein. Blot B was probed with antibody raised again:st purified APase Ill protein. The phoAIV mutant cells do not produce detectable APase IV protein during vegetative growth. In blot C, lysates from cells grown in Schaeffer's sporulation medium (Schaeffer et al., 1965) were incubated with APase IV antibody. Lanes 1, 3, and 5 contain B. subtilis 168 lysates; lanes 2, 4, and 6 contain samples from the phoAIV mutant. Lanes I and 2 contain lysates from samples taken aRer 4 h of growth or t - 2; lanes 3 and 4 at 8 h or t + 2; and lanes 5 and 6 at 12 h or t + 6. Lane 7 contains the control lysate from Pi-starved vegetative B. subtilis 168. The occurrence of a band in this control lane and not in the other six lanes shows that APase IV would have been detected on the blot if it were present.

98 tative cells and sporulating cells (Bookstein et al, 1990). The difference in expression of these very similar proteins may imply an underlying difference in function that has not yet been identifie0. (d) Analysis of APase activity in a~ pt~oAIV m u t a n t The effect of the phoAIV mutation on total A:"ase activity was determined in assays using the general phosphatase substrate XP. When APase activity was determined by colony color on plates containing XP, the phoAIV mutant was indistinguishable from wt cells. These results occurred in plates of Schaeffer's sporulation medium (Schaeffer et al., 1965), a low P~ complex medium of 1.0% peptone (Seki et al., 1987), and the low P~ defined medium developed in this laboratory (Hulett et al., 1990). These data are consistent with previous reports on the inability to isolate phoAIV structural gene mutants in B. subtilis by assaying colony color in a plate screen (Le H6garat and Anagnostopoulos, 1973; Glenn, 1975; Grant, 1974). and can be explained by the fact that there are at least two APases in B. subtilis that can hydrolyze XP. The phoAIV mutation affects total APase activity in liquid culture. Fig. 4 shows the result of assays done to determine the effects of the phoAIV mutation on total

0.#0

i

w , i . ; , l o.4o

1.0

,_.~

0 O0

l I

/o %,

o n

o.2o

APase specific activity of cells grown in low-phosphate defned medium. Total APase produced in the phoAIV mutant was 25 % of that produced by the parental strain. Bookstein et al. (1990) reported that phoAIll mutants produced 60-70% of the APase activity seen in the va strain. Thus the phoAIll and the phoAIV genes are responsible for nearly 100% of the total APase induced by phosphate starvation in vegetative B. subtilis cells. (e) APase IV is both a secreted and a cell-bound e n z y m e

APases III and IV are the two major secreted APases (Hulett et al., 1990) under phosphate starvation conditions. Of the total vegetative APase-specific activity 20-40% is cell-associated, when the cells are separated from the supernatant fraction by a 100000 x g centfifugation for I h (Hulett and Jensen, 1988). Most of the cell-associated APase can be solubilized from the lysed cell pellet with 1 M Mg 2 +. Fig. 5 shows a Coomassie blue-stained gel of the proteins present in the APase activity peak eluted from a Mono S ion-exchange column dunng purification of saltextractable cell-associated APases. The arrow identifies a major protein band which is the APase IV protein based on the following criteria. At the N-terminus 10 aa were sequenced and determined to be identical to that of the secreted APase IV protein. Antibody to secreted APase IV cross-reacted strongly with the cell-associated APase compared to APase Ill or B. licheniformis APase antibody. Mr analysis, as judged by Superose 12 gel filtration, indicated that the Mg 2+-extracted (Fig. 5) APase was a monomer. The secreted APase IV was determined to be a monomer by the same technique (Hulett et al., 1990). Finally, this protein cannot be isolated from the cells of the phoAIV mutant. APase species from B. subtilis have previously been

o.15 ~ ; o

[]

/ o.1~,,~'~$1~ 0 2 4

o.~o

2

tea

A

B

o os ~ z ~ -~o.oo 6 8 10 12 (hours)

Fig. 4. Production of total APase by the phoAIV mutant in Pclimiting defined medium. The closed symbols denote growth as determined by absorbance ot S40 nm The open symbols are APase specific activity. Units of activity were calculated as /,~mol of p-nitr~phenylphosphate (PNPP) (Sigma)hydrolyzed/minat pH 9.5 in 1 M CAPS (Sigma)buffer with 0.1 mM CoC12/70mM MgSO4/10-5 mM ZnCI2. Circles represent wt B. subtilis and squares represent the phoA1Vmutant. Cultures were inoculated to an initial density ofA54o = 0.2, and grew exponentiallyfor 5 h before enteringstationaryphase because o f P i starvation. Both wt and mutant cells began to produce APase activity as the cultures entered stationary growth. The phoAIV interrupted strain produced 25% as much APase activity as d;d wt cells, which suggests that phoAIV is responsible for 75% of total APase activity.

68 b. 43~.

',i".~i"~ ~.........

.9-

Fig. 5. Identificationof cell-associated APase IV by isolation and protein sequencing. 20-40% of the total vegetative APase specific activity is cell..associated, most ofwhich can be solubilizedfrom the lysedcell pellet fraction with 1 M Mg2+. The major protein of this Mg2+-extracted APase fractionated identicallyto the purified secreted APase IV, eluting from a Mono S ion-exchange column with 0.64 M Mg2+ (Hulett et al., 1990). Proteins from the peak activity sample were separated by 0.1% SDS-7.5 % PAGE and transferred to lmmobilon membrane. The protein in lane B, indicated by the arrow, was identified by APase antibody and sequenced. Lane A contains Mr markers. Lane B contains the proteins from the peak activity fraction. APase IV is identified by the leftward arrow.

99 TABLE I Mapping of the phoAIV locus Recipient strain a

Secondary marker a

Number linked per total number tested b

Cotransduction frequency ( ~ ) c

1A5 IA603 1A630

glyB zce-g3::Tn917 thiA-82::Tn917

!40/200 83/400 81/300

70 21 27

(g) Conclusions (1) The phoAIV gene encodes an APase that produces

a The strains used for mapping were IA5 (glyB 133, metC3, tre-'12, and trpC2) (Dedonder et al., 1977), IA603 {(SP c2) thiA84::Tn917 trpC2} and IA630 {(SP c2) trpC2 zce-82::Tn917 } (Vaaldeyar and Zahler, 1986). All mapping strains were supplied by the Bacillus Genetic Stock Center (Ohio ~itate University). b A P~SI transducing phage lysate of the phoAIV insertional interruptiort strains was used to transduce the recipient mapping strains to Cm resistance. All available transduetants were then tested for loss ofth¢ second~Lry marker, indicating linkage. c Cotransduction frequency is the percentage of the phoAIV gene and the secondary marker cotransduced into the total population of PBS 1 transductants tested.

solubilized from the cytoplasmic membrane with 1 M Mg 2+ and purified to homogeneity (Glenn and Mandelstam, 1971; Le Htgarat and Anagnostopoulos, 1973; Takeda and Tsugita, 196'1). Electron microscopic histochemistry localized the active enzyme on the inner leaflet of the membrane (Ghosh et al., 1971). The relationship of APase IV to the previously reported APase(s) is not cleat', nor can it be established that the APase isolated by the three groups is the same. What is clear is that APase IV can be localized in two cell fractions; cell associated (presumably membrane) or secreted. Further studies should d e t e ~ n e if posttranslational modification is involved in dictating the final destination of the APase IV protein or if the cell-bound species is a precursor to the secreted species.

(f) Mapping of the phoAIVgene The mapping position of the pJM 103 integration into the

p/wAIV locus was determined by phage PBSl-mediated transduction. Table I shows the strains that were used for 75* glyB t

zce I 85~

~

phoAIV l

mapping and the frequency of cotransduction between the Cm R marker and the mapping locus used. The transduction frequencies indicate that the phoAIV locus is between glyB and thiA at approx. 73 ° on the B. :'ubtilis chromosome (Fig. 6).

thiA l

70g 27g

21g Fig. 6. PBSI transduction linkage map of phoAIE The arrows point from the selected to the unselected markers. Distances are marked as the percentage of cotransfer of the two markers. The glyB marker has been placed at 75 ° on the B. subtilis chromosome map by Piggot and Hoch (1985). The linkage between the zce locus and glyB is that published by Vandeyar and Zahler (1986).

75% of the vegetative APase specific activity measured under our assay conditions. Our laboratory has identified four APase proteins which contribute to total vegetative APase expression (Hulett et al., 1990). Mutations in APase Ill reduced total specific activity about 40~o (Bookstein et al., 1990). Taken with the data presented here on APase IV, we can conc!ude that APase III and APase IV must account for the major portion of the total vegetative APase activity induced during phosphate starvation of B. subtilis. We also propose that the promoter strength of the two genes may be quite similar under pilosphate starvation since the specific activity of APase IV is twice that of APase III (Hulett et al., 1990). (2) APase IV is not made during sporulation. It has been possible to gonetically separate the induction of phosphate starvation APase(s) synthesis from the induction of sporulation APase(s) (Piggot and Coote, 1976). Our working hypothesis was that multiple structural genes were responsible for the two types of APase and that they differed more in regulation and information for protein localization than in the information for the primary aa sequence. The question of how many APases are induced under each growth condition is not completely answered, but our recent data suggest at least two or more respond to either stimulus (Hulett et al., 1990; Bookstein et al., 1990). The first APase studied, APase III (Bookstein et al., 1990), was synthesized under both phosphate starvation and sporulation conditions. The data presented here prove that not all APase genes are induced under both conditions; phoAIF !s strictly a phosphate starvation inducible gene. Whether there are APase genes that are expressed only under sporulation conditions has not been determined. (3) The product of the phoAiV gene accounted for a portion of the total APase activity in two cellular locations. Before it was determined that Bacillus had multiple APase geaes it was difficult to envision a protein localization mechanism to account for the reported distribution of APases. With the construction of strains retaining a single functional APase gene it should be possible to determine if a precursor relationship exists between cell-bound and secreted APase proteins. (4) The phoAIV gene maps at 73 ° on the B. subtilis chromosome. The phoAlll gene mapped at approx. 50 ° (Bookstein et al., 1990). Therefore, the APase genes are not clustered on the chromosome; both map across the chro-

100 m o s o m e f r o m pho regulon regulatory genes, p h o P a n d phoR.

ACKNOWLEDGEMENTS This w o r k was s u p p o r t e d by Public H e a l t h Service grant G M 33471 from the N a t i o n a l Institutes o f Health.

REFERENCES Bookstein, C., Edwards, C.E., Kapp, N.V. and Hulett, F.M.: The Bacillus subtilis 168 alkahne phosphatase III gene: the impact of a phoAlll mutation on total alka ine phosphatase synthesis. J. Bacteriol. 172 (1990) 3730-3737. Dedonder, R.A., Lepe=':at, J.-A., Lepesant-Kejzlarova, J., Billault, A, Steinmetz, M. and Kunst, F.: Construction of a kit of reference strains for rapid genetic mapping in Bacillus subtilis 168, Appl. Environ. MicrobLl. 33 (1977) 31-40. Dubnau, D. and DavidoffAbelson R.: Fate of transforming DNA following uptake by competent Bacillus subtilis. J. Mol, Biol. 56 (1971) 209-221. Ferrari, E. and Hoch, J.: Genetics. In Harwood, C.R. (Ed.), Bacillus. Plenum, New York, 1989, pp. 37-72. Ghosh, B., Wouters, J. and .'ampen, C: Distribution of the sites of alkaline phosphatas¢(s) activity in vegetative cells of Bacillus subtilis. J. Baeteriol. 108 (1971) 928-937. Glenn, A-R.: Alkaline phosphatase mutants of Bacillus subtilis. Aust. J. Biol. Sci. 28 (1975) 323-330. Glenn, A. and Mandelstam, J.: Sporulation in Bacillus subtilis 168: comparison of alkaline phosphatase from sporulating and vegetative cells. J. Biochem. 123 (i971) 129-138. Grant, W.D.: Sporul'.'tion in Bacillus subtilis 168. Control of synthesis of alkaline phosphatase. J. Get:. Microbiol. 82 (1974) 263-269. Gross, R., Arieo, B. and Rappuoh, R.: Families of bacterial signaltransducing proteins. Mol. Microbic,I. 3 (1989) 1661-1667. Hulett, F.M. and Jensen, K.: Critical roles ofspoOA and spoOR in vegetative alkaline phosphatase production in Bacillus subtilis. J. Bacteriol. 170 (1988) 3765-3768. Hulett, F~M., Bookstein, C. and Jensen, K.: Evidence for two structural

genes for alkaline l;hosphatase in Bacillus subtilis. J. Bacteriol. 172 (1990) 735-740. Lampe, M., Binnie, C., Schmidt, R. and Losick, R.: Cloned gene encoding the delta subunit of Bacillus subtilis RNA polymerase. Gene 67 (1988) 13-19. Lathe, R.: Synthetic oligonucleotide probes deduced from amino acid sequence data: theoretical and practical considerations. J. Mol. Biol. 183 (1985) 1-12. Le H6garat, J.C. and Anagnostopoulos, C.: Purification, suhunit structure and properties of two repressible phosphohydrolases of Bacillus subtilis. Eur. L Biochem. 39 (1973) 525-539. Meinkoth, L and Wahl, G.: Hybridization of nucleic acids immobilized on solid supports. Anal. Biochem. 138 0984) 267-284. Miki, T., Minimi, Z. and Ikeda, Y.: The genetics of alkaline phosphatase formation in Bacillus subtilis. Genetics 52 (1965) 1093-1100. Nixon, B.T., Ronson, C.W. and Ausubel, F.M.: Two component regulatory systems responsive to environmental stimuli share strong conserved domains with the nitrogen assimilation regulatory genes ntrB and ntrC. Proc. Natl. Acad. Sci. USA 83 (1986) 7850-7854, Piggot, PJ. and Coote, J.G.: Genetic aspects of bacterial endospore formation. Bacteriol. Rev. 40 (1976) 908-962. Piggot, P.J. and Taylor, S.Y.: New types of mutation affecting formation of alkaline phosphatase by Bacillus subtilis in sporulation conditions. J. Gen. Microbiol. 102 (1977) 69-80. Piggot, P.L, Moir, A. and Smith, D.A.: Advances in the genetics of Bacillus subtilis differentiation. In Levinson, H.S., Sonenshein, A.L. and Tipper, DJ. (Eds.), Sporulation and Germination. American Society of Microbiology, Washington, DC, 1981, pp. 29-39. Piggot, PJ. and ~och, J.H.: Revised genetic linkage map of Bacillus subtilis. Microbiol. Rev. 49 (1985) 158-179. Schaeffer, P., Millet, J. and Aubert J.: Catabolic repression of bacterial sporulation. Proc. Natl. Acad. Sci. USA 54 (1965) 704-711. Seki, T., Yoshikawa, H., Takahashi, H. and Saito, H.' Cloning and nucleotide sequence of phoP, the regulatory gene fer alkaline phosphatase and phosphodiesterase in Bacillus subtilis. L Bacteriol. 169 (1987) 29!3-2916. Seki, T., Yoshikawa, H., Takahashi, H. and Saito, H.: Nucleotide sequence of the Bacillus subtilis phoR gene. J. Bacteriol. 170 (1988) 5935-5938. Takeda, K. and Tsugita, A.: Phosphodiesterase of Bacillus subtilis, If. Crystallization and properties of alkaline phosphatase. J. Biochem. (1967) 231-241. Vandeyar, M.A. and Zahler, S.A.: Chromosomal insertions of Tn917 in Bacillus subtilis. J. Bacteriol. 167 (1986) 530-534.