Proc. NatL Acad. Sci.

USA

Vol. 80, pp. 658-662, February 1983

Biochemistry

Early sporulation gene spoOF: Nucleotide sequence and analysis of

gene product
(recombinant DNA/in vitro transcription/maxicell)

HIDENORI SHIMOTSU*, Fujio KAWAMURA*, YASUO KOBAYASHIt, AND HIUGA SAITO*

*Institute of Applied Microbiology, University of Tokyo, Bunkyo-ku, Tokyo 113, Japan; and tDepartment of Applied Biochemistry, Hiroshima University,
Fukuyama 720, Japan

Communicated by Jesse C. Rabinowitz, October 8, 1982

tire spoOF gene did not restore the ability of mutant JH649 to
sporulate. This result indicated that the sporulation was inhibited, for an unknown reason, by the multi-copy plasmid (8).
Rhaese et al. (10) proposed that the spoOF gene encodes a
synthetase for a highly phosphorylated nucleotide adenosine
5',3'(2')bistriphosphate. However, the synthetase protein has
not been isolated, and the true function of the spoOF gene in
the early stage of the sporulation has not been discovered. In
order to clarify the initiation process of sporulation, it is necessary to study the cloned early sporulation gene at the molecular level. In the present communication we report the location
of the spoOF gene in a 2.2-kb EcoRI fragment, its direction of
transcription, and its sequence.

ABSTRACT We have determined the sequence of a 1,162base-pair DNA fragment containing a spoOF gene which is required for an early stage of sporulation in Bacillus subtilis. The
sequence has only one long open reading frame consisting of 173
codons, which has been confirmed to be the spoOF cistron by DNAmediated transformation and in vitro transcription. In UV-irradiated "maxicells" containing pUBSF13, the plasmid that carries
cloned spoOF gene, we have observed the synthesis of a 20-kilodalton polypeptide that is absent from cells carrying a vector plasmid pUBl10. The molecular weight of this protein is in agreement
with the calculated molecular weight of the spoOF gene product
(Mr, 19,065). The putative promoter sequences of spoOF gene were
5' T-A-T-A-A-T 3' at -10 and 5' T-T-G-A-T-T 3' at -35. An octamer sequence, 5' A-A-A-G-G-A-G-G 3', situated 8 base pairs
prior to the initiation codon was found to be perfectly complementary with the 3' end of 16S ribosomal RNA. This result offers
additional evidence for the proposal by Rabinowitz's group that
an extensive mRNA-rRNA interaction is a requirement for efficient translation by B. subtilis ribosomes.

MATERIALS AND METHODS

Bacterial Strains, Bacteriophage, and Plasmids. B. subtilis
JH649 (trpC2, pheAl, spoOF221), which was supplied by J.
Hoch (Scripps Clinic and Research Foundation, La Jolla, CA),
was used as a host strain for a temperate phage 4AO5spoOF. The
recombinant phage 4105spoOF which carries the spoOF gene
was constructed by prophage transformation method as described (7, 8, 11). B. subtilis UOT0277 (hisAl, metB5, recE4,
nonBi) was used as a host for plasmids. The recombinant plasmid pUBSF13 was constructed by inserting the spoOF-containing EcoRI fragment (OF fragment) of 41O5spoOF into pUB11O
(8). Deletion derivatives pUBSFAB, pUBSFAH, and pUBSFAS
were constructed in vitro by religation after the cleavage of
pUBSF13 with Bcl I, HindIII, and Sac I, respectively.
Restriction Enzyme Analysis. The recombinant phage
4lO5spoOF was used as a source of DNA for restriction analysis
and DNA sequence determination. After EcoRI cleavage, fragments were separated by electrophoresis on 0.8% agarose gel
and a 2.2-kb EcoRI fragment which carried the spoOF gene was
eluted according to the procedure of Yang et al. (12). The eluted DNA was further purified by DEAE-cellulose chromatography according to the procedure of Smith (13). The restriction
map was constructed by digesting the OF fragment with various
restriction endonucleases. Restriction enzymes were purchased
either from Bethesda Research Laboratories or Takara Shuzo
(Kyoto, Japan) and were used according to the manufacturers'
instructions with minor modifications.
Purification of RNA Polymerase. Preparation of RNA polymerase from cells of B. subtilis 168 (trpC2), harvested in early
exponential growth phase, was carried out as described (14)
except that: (i) the precipitate from the first ammonium sulfate
fraction was dialyzed against buffer A [ 10 mM Tris-HCl, pH 8. 0/
10 mM MgCl2/0.1 mM EDTA/0.1 mM dithiothreitol/10%
(vol/vol) glycerol] and then applied to a DEAE-Sephadex A-50
column; (ii) DEAE-Sephadex-purified fractions of RNA poly-

One of the primitive forms of differentiation is represented by

the formation of spores in certain bacteria. The formation of a
spore follows a sequence of morphological and biochemical
events. The process of sporulation in Bacillus subtilis has been
investigated in detail (1, 2). A large number of sporulation-defective mutants have been isolated and their mutations have
been mapped (3). These mutants have been classified into five
types (spoO, spoIl, spolI, spoIV, and spoV) according to the
stage at which sporulation of the mutant is blocked. Among
these mutants, the spoO type is particularly interesting because
initiation of the sporulation is dependent upon the products of
the spoO genes. At least nine loci of spoO genes have been
mapped (4) and a mutation in any one of these genes blocks
sporulation before the morphological change characteristic of
early spore development. Despite their crucial role in development, however, little is known about the spoO genes and how
they control the initial events of spore formation.
We recently developed a method, called "prophage transformation," for constructing specialized transducing phages of
p1l (5) and 0105(6). To study the initiation of sporulation, we
have used these phages to clone the early sporulation gene
spoOF. The activity to transform spoOF was found in a 2.2-kilobase-pair (kb) EcoRI fragment of the transducing phage DNAs
(7, 8). These transducing phages were designated pllspoOF and
4OSspoOF; upon lysogenization they could restore the ability
of mutant JH649 (spoOF221) to sporulate. In contrast to the
phage vectors, which exist as a single prophage genome, plasmid vectors such as pUB110 are known to reside as multiple
copies (9). The multicopy plasmid pUBSF13 harboring the enThe publication costs of this article were defrayed in part by page charge
payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U. S. C. 1734 solely to indicate this fact.

Abbreviations: bp, base pair(s); kb, kilobase pair(s).

658

Proc. NatL Acad. Sci. USA 80 (1983)

Biochemistry: Shimotsu et al.

merase were made 70% with ammonium sulfate, and the precipitate was dissolved in and dialyzed against buffer C [10 mM
Tris-HCl, pH 8.0/0.1 mM dithiothreitol/10% (vol/vol) glycerol] and the solution was applied to a DNA-cellulose column.
Samples.(25 Al) of the DNA-cellulose fractions were subjected
to electrophoresis through a NaDodSO4/polyacrylamide slab
gel and the fractions containing-core subunits and a a 55 subunit
were used for in vitro transcription studies.
RNA Synthesis. The RNA polymerase was incubated at. 370C
for 7 min with 0.5 Ag of DNA template in a 40-.lI reaction mixture cobtaining 50 mM Tris-HCl(pH 7.8), 0.1 mM EDTA, 0.1
mM dithiothreitol, 10 mM MgCl2, 0.2 mM spermidine, bovine
serum albumin at 100 pug/ml, ATP, GTP, CTP, and UTP at 0.1
mM each, and 0.6 AuM [a-32P]ATP (10 p.Ci; 1 Ci = 3.7 X 1010
Bq). The reaction was stopped by mixing with an equal amount
of phenol and the aqueous-layer was passed through a Sephadex
G-50 column (1 X 10 cm). Radioactive, RNA was subjected to
electrophoresis through polyacrylamide slab gels containing 7
M urea.
DNA Sequence Analysis. DNA fragments were labeled at
the 5' ends with'T4 polynucleotide kinase (Bethesda Research
Laboratories) and [_y-32P]ATP (Amersham). DNA sequence was
determined according to the procedures of Maxam and Gilbert
(15). The products of the "G","G+A","T+C","C", and "A+C"
reactions were electrophoresed on 0.4-mm-thick 8% or 20%
polyacrylamide gels.
Preparation.and Labeling of Maxicells. B. subtilis strain
UOT0277 (hisAl, metB5, recE4, nonBi) was used as a host
throughout the experiments. Vegetative cells containing plasmids were grown in Sterlini and Mandelstam's resuspension
medium (16) supplemented with 0.2% glucose and kanamycin
at 5 pg/ml. Cells were harvested by centrifugation 1 hr after
the end of exponential growth and suspended in S ml of the fresh
resuspension medium. The cell suspension was transferred to
,a 9-cm-diameter glass Petri dish and irradiated for 1 min at a
distance of 25 cm from a 15-W germicidal lamp. After 3 hr of
incubation, the cells were spun down and suspended in fresh
resuspension medium. The cell suspension was incubated for
5 min at 370C. Then 2.5 ACi of ['4C]leucine (340 mCi/mmol)
was added and incubation was continued for 20 min. The incorporation of radioactive amino acid was terminated by adding
nonradioactive leucine. The procedure of cell lysis was performed according to Hirochika et al. (17).

RESULTS
Location of spoOF Gene Within the Cloned Fragment. Previous studies have shown that the recombinant phage 0145spoOF

EHoRI

Hind0 ]T

pUBSF1 3
pUBS FAB

EcoRI

contains a B. subtilis chromosomal DNA insert of 2.2 kb which

carries the early sporulation gene spo0F(8). This insert was
cleaved from the phage DNA with EcoPJ and purified with an
agarose gel after electrophoresis. The restriction sites of
HindIII, Bcl I, and Sac I of the OF fragment were determined
as shown in Fig. 1. Recombinant plasmid pUBSF13 was constructed by inserting the OF fragment at the single EcoRI site
ofthe vector plasimd pUBilO. To determine the location of the
spoOF gene within the OF fragment, deletion derivatives of
pUBSF13 were constructed by site~specific cleavage and ligation (Fig. 1). When the 0.3,kb. Bcl I fragment was removed from
pUBSF13, the remaining 'DNA in the deletion plasmid pUBSFAB
lost the activity to transform spoOF221 to Spo'. Another plasmid, pUBSFAS which lacked the 0.7-kb Sac I fragment, also
lost the transforming activity. In contrast, deletion of the 0.3kb HindIII fragment in pUBSFAH did not affect the spoOF221
transforming activity. These results indicated that at least a part
of the spoOF gene was-located between the two Bcl I sites ofthe

OF fragment (Fig. 1).

Location of the Promoters on the 'OF Fragment. The initiation sites of in vitro transcription on the OF fragment were
mapped by analysis of RNA transcripts with various DNA templates (Fig. 2). Transcription of the 2.2-kb OF fragment yielded
two RNA species approximately 1,000 and 850 bases. When the
DNA fragment lacking the internal Bcl I fragment (ABcl I fragment) was used as template, only a single transcript of about
1,000 bases was detected. The largest HindIII fragment
(HindIII fragment A) gave a: transcript of 380 bases instead of
the 1,000-base transcript. This 380-base RNA was thought to
be a "run-off" product starting about 380 bp from left end of
HindIII fragment A. These results suggested that the 2.2-kb OF
fragment harbored two promoters where two transcripts were
initiated in opposite directions. The 850-base RNA was thought
to be the transcript of spoOF gene, because a part. of the spoOF
gene was involved in the internal Bcl I fragment.
Nucleotide Sequence Analysis. Our sequence analysis strategy is outlined in Fig. 3. To investigate the coding potential of
the DNA analyzed, a computer program was used to analyze
the relative positions of initiation and termination codons in all
six reading frames. The nucleotide sequence contains only one
open reading frame long enough to code for a.protein. The reading frame consists of 173 codons as shown in Fig. 4. When the
0.3-kb Bd I fragment was removed from the OF fragment, the
activity to transform spoOF221 to Spo' was lost. The internal
region lying between two Bcl I sites, which harbored the mutation site of spoOF221, was involved in the open reading frame
(Fig. 3).

EccoRI EcoRI EcoRI

I
mI- Sac
Hind
_
a_

659

BoUI Bcl

01 05spoOF

Sac
ab

T F
ACTIVITY

pUBS FAS
pUBS FAH

+4

FIG. 1. Restriction endonuclease maps of cloned OF fragment and its deletion derivatives. (Upper) The thick line on the 4105spoOF indicates
the cloned EcoRJ fiagment; the restriction map is shown below it. (Lower) Transforming activity (TF) of pUBSF13 and in vitro deletion plasmids.
The lines show spoOF EcoPJ inserts of the plasmids. flNA-mediated transformation was carried out as described (18). The activity of these DNAs
in transforming spoOF221 to Spo' is shown by +.

Biochemistry: Shimotsu. et al.

660

A
kb
1.0
O.(35

..

similar to, the corresponding -35 (5' T-T-G-A-C-A 3') and

-10 (5' T-A-T-A-A-T 3') consensus sequences of Escherichia
coli promoters (20). In the 5' flanking region' of the spoOF gene,
such sequences occur in the position predicted from the in vitro
transcription experiments (Fig. 4).
An inverted repeat sequence located in the 3' flanking region

are

1 2 3

Proc. Natl. Acad. Sci. USA 80 (1983)

1.I OF frecinnent

M-

may be' a

terminator because such a sequence can be found in

the transcriptional, terminator regions in E. coli (20). This presumptive terminator in the 3' flanking region of spoOF gene is
only 680 bases away from the putative promoter, whereas the
850-base RNA was considered to be the spoOF transcript made
in vitro. One possible explanation of this discrepancy could be
that the spoOF transcript made in vitro was not properly terminated and resulted in the "run-off" product, due to the salt
concentrations used in the in vitro reactions. The distance between the putative promoter and the 3' end of 0F'fragment was
estimated to be about 850 bases which corresponded to the
length of "run-off" transcript (Fig. 2).
It has been reported that the 3' end of 16S rRNA forms a basepaired complex with a sequence preceding the initiation codon
(21). In the spoOF gene, a 5' A-A-A-G-G-A-G-G 3' sequence
situated 8 bp prior to the initiation codon (Fig. 4) is perfectly
complementary with an octanucleotide sequence on 3' end of
16S rRNA (discussed below). The distance between the ribosomal binding site and the initiation codon is close to the average
distance found in other ribosome binding sites (21).
Analysis ofthe Gene Products. Expression of the spoOF gene
in pUBSF13 was detected by the maxicell method (22). Protein
synthesis in maxicells was carried out in. the presence of
[14C]leucine. The protein synthesized with cell lysates was analyzed by NaDodSO4/polyacrylamide gel electrophoresis. The
results, of these experiments suggested in vivo synthesis of a.
20-kilodalton protein by pUBSF13 harboring the OF fragment
(Fig. 5). A derivative plasmid pUBSFAB which lacked a part

1.0 kb

0.815 kb

2. OF-BeZ I

fragm1,enit

1.0

3.

kb

F-iSnd HI

4-O
A

ra

gm~ent

PP

0.85 kb

0.38 kb

FIG. 2. In vitro RNA synthesis with cleaved DNA templates. RNA

labeled with [a-32P]ATP and various templates. (A) Lanes: 1, OF
fragment (2.2 kb); 2, EcoRI fragment from pUBSFAB which lost the
internal Bel I fragment; 3, Hindu1 fragment A (1.5 kb) from the OF
fragment. E. coli rRNAs and tRNAs were used as size markers. (B)
DNA templates used for in vitro transcription experiments. Run-off
transcripts are indicated by dashed lines. p, Putative promoters.
was

Regulatory Region of the Transcription and Translation.

The principal form of the complete RNA polymerase in B. subtilis contains a or subunit of 55,000 daltons known as a 55. It
has been shown by Losick and Pero (19) that promoters. whose
recognition is mediated by or 55 exhibit a characteristic hexanucleotide sequence centered at position 35 and lying 10 bp
upstream from the start point of transcription. These hexamers
OF f ragment
HindM HindII

EcoRI

HinfISac I Bcl I
i

~
4444-

Z X |

1l ~1 1

BcZ I

Sa

AZu I EcoRI
I

~
~
4U 4X -|
1-z
-ray l l

4-

4--

.~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~r
I
tI.,

2
3

100

200

300 bp

s
_OF_

4_

FIG. 3. Restriction endonuclease maps and sequence determination experiments. (Upper) Detailed physical map in the Hinfi-AluI region of
the OF fragment. Each arrow represents- a single Maxam-Gilbert sequence determination experiment. (Lower) Coding potential of this DNA in
all six reading frames: m, open reading frame region; *, noncoding region. The coding frame for the spoOF gene was determined to be frame 2, as
described in the text.

Biochemistry: Shimotsu et al.

Proc. NatL Acad. Sci. USA 80 (1983)

661

80
GTCAGTTAGA CTTCAGGGGC AGATATTTTT TGACGGCGTC TCTGATTTCG TCGATGTCAA ACGGCTTGGC AAAGTGCGTC
Sacl

160

AGAGCGCCCA ATTCCTTCGA TTCCTGGATC ATGTCGAGCT CTCCGTATGC CGTCATGATA ATGACCCGGA TGTTTTCGTC

240
AATGACCTTC ATCCGTTTTA AGATTTCGAT TCCGTCCATG CCGGGAATTT TCATGTCCAA CAGCACAAGG TCGGGCCGTT
320
Pst I
CTTTTGTCAC AATGTCAAGC GCCTGCAGGC GTTCGCAGCC TGAAACGTCT GGTAGCCTTC.TTTATTGAAC ACTTCATTTA
BcI
400
GCAAAATACG AATGCCGTAT TGATCATCAA CGATTAAAAT TTTTTCATTC ATCATTTTAC ACCCCAATAT TATGATTTTC

GTCAAAAGTA AGCAGTATTG TATGTATTCT G

TGAT

T CCTATTTCCT TTAA

AA~GTCTACTT

480
TACGACATTT

555
ATG TTG AAA ATA TTC
Met L.eu Lys Ile Phe
621
ACG ACG CAG TTA ACA GGT ATT TTT TCC CGC ATT CAG GAT AAG GAA TCT GAC GCG ATT GAA GAT GGG
Thr Thr Gln Leu Thr Gly 1le Phe Ser.Arg Ile Gln Asp Lys Glu Ser Asp Ala Ile Glu Asp Gly
TCTGAGCATT TTCTCTTTTG TTGTATACTG ATATTGTACG TTATIAGrGTTCACTT

687

BcI I.

GCG CGG CTG CTT GCT CAA GCG GTG ATC AGC GGG CAT TCC ATT TAT TTA TAC GGA GCG AAT GAG CTT
Ala Arg Leu Leu Ala Gln Ala Val Ile Ser Gly His Ser Ile Tyr Leu Tyr Gly Ala Asn Glu Leu
753
CAG GGC GTC TTT TAT GAG GCC ACC GAA AGC AAA GAA CCC TTC CCA TCT GTC AAA GCC TTT CCA GAA
Gln Gly Val Phe Tyr Glu Ala Thr Glu Ser Lys Glu Pro Phe Pro Ser Val Lys Ala Phe Pro Glu
819
AAC GCT GAG GAA GTG ACA GAA AGC GAC AGG GTG CTG ATG TTT TGC TCA GGG ACG GGC ACA GCC GAA
Asn Ala Glu Glu ValThr Glu Ser Asp Arg Val Leu Met Phe Cys Ser Gly Thr Gly ThrAla Glu
885
Sac
GAA CAG GAG CTC GCA AAA GAG CTT TAT GAA AAA GGT GCG GGA GTC GTA TGC GTA TCG CCC GCA GCC
Glu Gln Glu Leu Ala Lys Glu Leu Tyr Glu Lys Gly Ala Gly Val Val Cys Val Ser Pro Ala Ala
951
AAA GAC AGT GCG GGA ATA GAA CAG TAT TGT GAT GTG CAT.ATT GAT TCT AAA TTA AAA ATG CCG CTT
Lys Asp Ser Ala Gly Ile Glu Gln Tyr Cys Asp Val His Ile Asp Ser Lys Leu Lys Met Pro Leu

1017
GTT CCC GAT GAA GAC GGC ACC CGT TAC GGG TTT CCC TCT TTA ATG.ACA GCA CTG TAT GTC TAT CAC
Val Pro Asp Glu Asp Gly Thr Arg Tyr Gly Phe Pro Ser Leu Met Tbr Ala Leu Tyr Val Tyr His

i1082

GCT TTA TCG TTT ACA CTA AAA GAA ATT CTG CAA GAG TAT GCA TAA
Ala Leu Ser Phe Thr Leu Lys Glu Ile.Leu Gln Glu Tyr Ala *

TATCTTATTG TACATGCTGG

1162
AACTTGCCGG AAACAAATAA AAAAGACTTG CCCGCTTTTG ACAAACGGCA AGTCTTTTTT ATTACTTCTG ATTTGCAGCT
FIG. 4. Nucleotide sequence of the spoOF gene and the adjacent region. Only the sequence confirmed by two independent experiments is shown.
The inferred amino acid sequence of spoOF is also shown. The presumptive "-35 sequence", Pribnow sequence, and Shine-Dalgarno ribosome
binding sequence are enclosed in boxes. The putative terminator and 5' T-A-T-T-G-T 3' sequence discussed in the text are underlined.

of the structural gene of spoOF lost the ability to synthesize the

protein. The molecular weight ofthis protein was in agreement
with the value calculated on the basis of the predicted amino
acid composition, 19,065. Another protein, of molecular weight
about 29,000, found in the same experiment was thought to be
-the product of a 1,000-base transcript of OF fragment because
this protein was synthesized by pUSFAB but not by pUBllO

(Fig. 5).
DISCUSSION
Reports of nucleotide sequences of B. subtilis chromosomal
genes have been scarce. We have determined the nucleotide
sequence of the spoOF gene region. The spoOF gene is one of
the-spoO type genes which are known to control the initial events
of spore formation. The nucleotide sequence was found to contain only one long open reading frame consisting of 173 codons,
and the presumptive molecular weight of the gene product was
calculated to be 19,065. Expression of the spoOF gene of
pUBSF13 was examined by the maxicell method, and a 20-kilo-

dalton protein-was detected by NaDodSO4/polyacrylamide gel

electrophoresis (Fig. 5). Direct correlation of-the spoOF gene
with the 20-kilodalton protein was proved because pUBSFAB
lacking a part of the spoOF cistron did not cause the protein
synthesis. Many papers have dealt with sporulation genes but
its gene product has not been identified at the molecular level.
The maxicell method used for identifying the spoOF gene
product originally was developed with E. coli CSR603 (recAl,
uvrA6, phr-1) by Sancar et al. (22). We adapted this method for
B. subtilis and obtained successful results with strain UOT0277
(recE4, hisAl, metB5, nonBi). This method was found to be
useful for identifying proteins encoded by plasmid DNA in B.
subtilis.
o55 is a major component of RNA polymerase in the actively
growing cells of B. subtilis, and it has been reported that o, 55
RNA polymerase recognizes the promoter sequence similar to
the E. coli consensus sequence (24). Recently, it has been shown
that minor transcriptional determinants o- 37, cr 29, and o- 28
also exist in growing and sporulating cells (25-28).

662

Biochemistry: Shimotsu et atS

Proc. Natl. Acad. Sci.. USA 80 (1983)

owitz's group (31, 32, 35) that an extensive mRNA-rRNA interaction is a requirement for efficient translation by B. subtilis
ribosomes.

1 2 3

'

69

46

x 30

._1

3. 18.4
e.

We are grateful to Dr. Sueharu Horinouchi for many stimulating

discussions. We are also grateful to Mr. H. Yoshikawa and H. Osada
for instruction in DNA sequence determination and fluorography. We
are indebted to Mr. H. Kojima of our laboratory for use of the computer
facilities. This research was supported in part by a Grant-in-Aid for
Scientific Research, by a Grant-in-Aid for Special Project Research, and
by a Grant-in-Aid for Encouragement of Young Scientists from the
Ministry of Education, Science and Culture, Japan.

29

-o
-20
-

1. Mandelstam, J. (1969) Symp. Soc, Gen. Microbiot 19, 377-404:

2. Hoch, J. A. (1976) Adv. Genet. 18, 69-98.
3. Piggot, P. J. & Coote, J. G. (1976) Bacteriot Rev. 40, 908-962.
4. Henner, D. J. & Hoch, J. A. (1982) in The Molecular Biology of
the Bacilli, Bacillus subtilis, ed. Dubnau, D. A. (Academic, New
York), Vol. 1, pp. 1-33.
5. Kawamura,. F., Saito, H. & Ikeda, Y. (1979) Gene 5, 87-91.
6. Iijima, T., Kawamura, F., Saito, H. & Ikeda, Y. (1980) Gene 9,

12.3

FIG. 5. Polypeptides synthesized by plasmids pUB110, pUBSF13,

and pUBSFAB. Maxicells carrying plasmids were incubated with
[14C]leucine. Lysates of these maxicells were electrophoresed on 15%
NaDodSO4/polyacrylamide gel (23). Lanes: 1, pUB110; 2, pUBSF13;
3, pUBSFAB. [14C]Methyl-labeled cytochrome c (12,300 daltons), lactoglobulin A (18,367 daltons), carbonic anhydrase (30,000 daltons),
ovalbumin (46,000 daltons), and bovine serum albumin (69,000 daltons) were used as molecular weight standards. Minor bands (other
than the 29- and. 20-kilodalton ones) were found in all lanes of the
original photograph.

Losick and Pero (19) suggested that. the divergent promoter

recognized by individual ar subunits in RNA
polymerases .of B. subtilis. A characteristic heptamer sequence
5' T-A-T-T-G-T-T 3' was found in the promoter region of spoVC
gene (29) which was known to be transcribed in vitro by r 37
RNA polymerase. Moran et al. reported that the two promoters
for the 0.4-kb (spoVG) gene, which is catalyzed in vitro by a
37 RNA polymerase, retain the similar sequences 5' A-A-T-TG-A-T 3' and 5' T-A-A-T-G-C-T 3' about 10 bp upstream from
the two start points of in vitro transcription (30). The promoter
region of spoOF gene also contains the 5' T-A-T-T-G-T-A 3'
heptamer (Fig. 4), although it is not yet known whether the
spoOF gene is transcribed in vitro by o- 37 RNA polymerase or

sequences were

not.

Another interesting finding in spoOF was concerned with

translation. It has been shown that the sequences of the ribosome binding sites for eight genes of Gram-positive, sources are
capable of forming stable complexes with the 3' end of B. subtilis 16S rRNA (31). McLaughlin et al. (32) have suggested that
mRNAs from Gram-positive organisms have stronger ShineDalgarno complementality with 16S rRNA than do those of E.
coli.
In the spoOF gene, a 5' A-A-A-G-G-A-G-G 3' sequence is
situated prior to the initiation codon. This polypurine region
can pair with eight nucleotides on the 3' end of B. subtilis 16S
rRNA (C. Woese, cited in ref. 31) as follows:
AAAGGAGG 3'
mRNA 5'...
rRNA' 3' (HO)UCUUUCCUCC 5'.
Based on the rules of Tinoco et al. (33) and Borer et al. (34), the
calculated AG for this interaction is -18.3 kcal/mol (1 cal =
4.184 J). Therefore, this sequence seems capable of forming a
strikingly stable complex with the 3' end of 16S rRNA. These
results offer an. additional evidence to the proposal -by Rabin..

115-126.
7. Kawamura, F., Saito, H., Hirochika, H. & Kobayashi, Y. (1980)
J. Gen. Appl. Microbiol 26, 345-355.
8. Kawamura, F., Shimotsu, H., Saito, H., Hirochika, H. & Kobayashi, Y. (1981)' in Sporulation and Germination,.eds. Levinson,
H. S., Sonenshein, A. L. & Tipper, D. J. (American Society for
Microbiology, Washington, DC), pp. 109-113.
9. Shivakumar, A. G. & Dubnau, D. (1978) Plasmid 1, 405-416.
10. Rhaese, H. J., Hoch, J. A. & Groscurth, R. (1977) Proc. Natl.
Acad. Sci. USA 74, 1125-1129.
11. Hirochika, H., Kobayashi, Y., Kawamura, F. & Saito, H. (1982)
J. Gen. Appl Microbiol 28, 225-229.
12. Yang, R. C. A., Lis, J. & Wu, R. (1979) Methods Enzymol 68,
176-182.
13. Smith, H. 0. (1980) Methods Enzymol 65, 371-380.
14. Kawamura, F. & Ito, J. (1977) J1 Virot 23, 562-577.
15. Maxam, A. M. & Gilbert, W. (1980) Methods Enzymol 65, 499560.
16. Sterlini, J. M. & Mandelstam, J. (1969) Biochem.J. 113, 29-37.
17. Hirochika, H., Kobayashi, Y., Kawamura, F. & Saito, H. (1981)
J. Bacteriol. 146, 494-506.
18. Shibata, T. & Saito, H. (1973) Mutat. Res. 20, 159-173.
19. Losick, R. & Pero, J. (1981) Cell 25, 582-584.
20. Rosenberg, M. & Court, D. (1979) Annu. Rev. Genet. 13, 319353.
21. Gold, L., Pribnow, D., Schneider, T., Shinedling, S., Singer, B.
S. & Stormo, G. (1981) Annu. Rev. Microbiol 35, 365-403.
22. Sancar, A., Hack, A. M. & Rupp, W. D. (1979)J. Bacteriol 137,
692-693.
23. Laemmli, U. K. & Favre, M. (1973) J. Mol Biol 80, 575-599.
24. Lee, G., Talkington, C. & Pero, J. (1980) Mol Gen. Genet. 180,
57-65.
25. Haldenwang, W. G. & Losick, R. (1980) Proc. Natl Acad. Sci.
USA 77, 7000-7004.
26. Haldenwang, W. G. & Losick, R. (1979) Nature (London) 282,
256-260.
27. Haldenwang, W. G., Lang, N. & Losick, R. (1981) Cell 23, 615624.
28. Wiggs, J. L., Gilman, M. Z. & Chamberlin, M. J. (1981) Proc.
Nati Acad. Sci. USA 78, 2762-2766.
29. Moran, C. P., Jr., Lang, N. & Losick, R. (1981) Nucleic Acids
Res. 9, 5979-5990.
30. Moran, C. P., Jr., Lang, N., Banner, C. D. B., Haldenwang, W.
G. & Losick, R. (1981) Cell 25, 783-791.
31. McLaughlin, J. R., Murray, C. L. & Rabinowitz, J. C. (1981)J.
Biol Chem. 256, 11283-11291.
32. McLaughlin, J. R.,.Murray, C. L. & Rabinowitz, J. C. (1981)
Proc. NatW Acad. Sci. USA 78, 4912-4916.
33. Tinoco, I., Jr., Borer, P. N., Dengler, B., Levine, M. D., Uhlenbeck, 0. C., Crothers, D. M. & Gralla, J. (1973) Nature (London) New Biol. 246, 40-41.
34. Borer, P. N., Dengler, B. & Tinoco, I., Jr. (1974)J. Mol Biol. 86,
843-853.
35. Murray, C. L. & Rabinowitz, J. C. (1982) J. Biol. Chem. 257,
1053-1062.