Download Effects of Transition Mutations in the Regulatory

Journal of General Microbiology (1989, 131, 959-962.
Printed in Great Britain
959
Effects of Transition Mutations in the Regulatory Locus spoZZA on the
Incidence of Sporulation in Bacillus subtilis
By M. D . Y U D K I N , * K . A. J A R V I S , S. E. R A V E N A N D P. FORT7
Microbiology Unit, Department of Biochemistry, University of Oxford, South Parks Road,
Oxford OX1 3QU, UK
(Received I7 October I984 ;revised 16 November 1984)
We have determined the changes in DNA sequence corresponding to three mutations in the
promoter-proximal open reading frame of spoIIA, a locus that regulates sporulation in Bacillus
subtilis. All three mutations prevent the synthesis of two sporulation-associated enzymes, but
they differ in their effects on spore incidence. We now find that mutation spo-42, which allows
spores to be produced at a low incidence, is a transition that changes Gly95 to Asp in the protein
encoded by the open reading frame. Mutation spo-69, which blocks sporulation entirely, consists
of two transitions: these change Gly62 to Asp and Ala116 to Thr. Mutation sas-I, which
partially suppresses spo-69, is also a transition : this changes residue 62 (which had become Asp
as a result of the spo-69 mutation) to Asn.
INTRODUCTION
When Bacillus subtilis cells are transferred from a growth medium to a medium with a poor
carbon source, they begin a process of differentiation that leads, after about 8 h at 37 "C, to the
formation of heat-resistant spores (for a review, see Young & Mandelstam, 1979). During this
time many specific genes, which were not transcribed during vegetative growth, become
expressed in a regulated sequence (Linn & Losick, 1976; Ollington & Losick, 1981; Yudkin et
al., 1982).One of the means by which the regulation of gene expression is thought to be achieved
involves changes in the specificity of RNA polymerase as development of the spore proceeds
(Losick, 1982). However, our understanding of the details of the control systems in sporulation is
very fragmentary.
Several dozen genetic loci are necessary for sporulation, as can be inferred from the effect of
mutations that block the sequence (reviewed by Piggot & Coote, 1976). We have been studying
one of these loci, spoIIA, which we have reason to believe is important in controlling gene
expression in sporulation. Different closely-linked mutations in spoIIA have different
phenotypic effects in cells placed under conditions that induce sporulation in the wild-type :
some abolish and others permit the synthesis of two enzymes, alkaline phosphatase and
DNAase, that are associated with stage I1 of sporulation (Errington & Mandelstam, 1983); some
prevent sporulation altogether and others allow a few spores to be made (Piggot, 1973); and one
mutation causes the progress of sporulation to be slowed (Yudkin & Turley, 1980).
The nucleotide sequence of the spoIIA region has recently been determined in this laboratory
(Fort & Piggot, 1984). It contains three open reading frames (ORFs), the promoter-proximal of
which (ORF 1) is expected to encode a protein of 117 amino-acid residues with an isoelectric
point of about 5.8.
We have now begun a study of the DNA sequences of different spoIIA mutants, and in this
paper we report the identification of the nucleotide changes corresponding to three mutations in
t Present address : Laboratoire de Biologie Molkculaire, Universitk de Sciences et Techniques du Languedoc,
Place E. Bataillou, 34060 Montpellier Cedex, France.
Abbreviation : ORF,open reading frame.
0001-2273 0 1985 SGM
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Mon, 19 Jun 2017 00:06:58
960
M. D . YUDKIN AND OTHERS
ORF 1. Each of these mutations prevents the synthesis of the sporulation-associated alkaline
phosphatase and DNAase, but the three differ in the extent to which they affect sporulation.
of the wild-type incidence (Yudkin
Mutation spo-42 allows spores to be produced at about
& Turley, 198l), spo-69 entirely abolishes sporulation (Piggot, 1973) and sas-I partially reverts
spo-69 (Yudkin & Turley, 1980).
METHODS
Plasmids. Plasmid pHM2 (Liu et al., 1982) carries a part of the spoIIA region including the whole of ORF 1.
When this plasmid is inserted into a strain that bears a mutation in the corresponding region of spoIIA, the
mutation is readily transferred to the plasmid (Chak et al., 1982). Dr P. J. Piggot (NIMR, Mill Hill, London, UK)
kindly gave us plasmids prepared in this way, that carried either spo-69 or spo-42 or sas-1. We transformed each
plasmid into Escherichia coli strain DH 1 (Hanahan, 1983),with selection for ampicillin resistance. Plasmids were
prepared from strain DH1 by the method of Guerry et al. (1973) after amplification with spectinomycin (Klein
et al., 1980), and purified on a gradient of CsCl.
We proved that each plasmid preparation was free of spo+ D N A by showing that it was unable to transform a B.
subtilis strain carrying the corresponding spo mutation to Spo+. We confirmed the identity of the spo mutation in
each preparation by transforming a spo+ strain to chloramphenicol resistance, recovering the spo mutation from
the plasmid on to the chromosome, allowing loss of the plasmid, and characterizing the resulting spo strain by
appropriate physiological and genetic tests including back-crosses.
Each plasmid was restricted, and the 0-67kbp EcoRI-BglII fragment that carries ORF 1 of spoIIA (Fort &
Piggot, 1984) was purified on low-melting-point agarose and cloned into M13mp9 (Messing & Vieira, 1982) that
had been cut with EcoRI and BamHI. In some experiments the EcoRI-BglII fragments were further digested with
A h 1 and cloned into the SmaI site of M13mp9.
Chemicals andenzymes. The materials needed for sequencing by the method of Sanger et al. (1977) were obtained
from Amersham, and the procedures were as described by the supplier. Restriction enzymes were obtained from
BRL, and used according to the directions of the supplier. Low-melting-point agarose was also from BRL.
RESULTS
Sequencingprocedure.All the mutations that we wished to study are in ORF 1 of spoIIA, which
is entirely contained within a 0-67kbp fragment bounded by EcoRI and BgZII sites (see Fig. 1).
By cloning this fragment from plasmids to which the mutations had been transferred,
sequencing from the EcoRI site, and comparing the results with the wild-type sequence, we were
able to identify the mutations. They all lay between two AluI sites, of which one is near the
middle of ORF 1 and the other is early in ORF 2 (Fig. 1). We could thus confirm the
identification of the mutations by subcloning the A h 1 fragment in each case in the two possible
orientations and then sequencing both DNA strands. We have thus identified each mutation by
three separate cloning and sequencing procedures.
Mutation spo-42. This mutation is a transition at position 670 from the nucleotide pair G - Cto
A-T (Fig. 2). This transition changes the codon GGT (Gly) to GAT (Asp), at residue 95 in the
ORF 1 protein.
Mutation spo-69. To our surprise, spo-69 turned out to be a double mutation, containing two
G - Cto A.T transitions. The first, at position 571, changes the codon for residue 62 from GGC
(Gly) to GAC (Asp). The second, at position 732, changes the codon for residue 116 from GCA
(Ala) to ACA (Thr).
Mutation sas-I. This mutation, which is also a transition from G.C to A.T, is at position 570,
within the same codon as the first of the spo-69 changes. Thus in the sas-I mutant the codon for
residue 62 reads AAC (Asn), as compared with GGC in the wild-type and GAC in the spo-69
mutant.
DISCUSSION
Although we have discovered two mutations, separated by 160 base pairs, in spo-69, we
believe that it is the first of these, at position 571, that is responsible for the Spo- phenotype. Our
reasons are these. First, we would expect a change from glycine to aspartic acid, which would
introduce an additional negative charge, to have a more dramatic effect on the ORF 1 protein
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Mon, 19 Jun 2017 00:06:58
Molecular identijication of spo mutations
961
U RIF ~
c
ORF 2
Fig. 1 . Part of the chromosome of B. subtilis in the spoIZA region. The numbering refers to the
nucleotide sequence determined by Fort & Piggot (1984).
Fig. 2. (a)Part of the DNA sequence of the spoZZA region of the wild-type. The sequence shown is part
of an AIuI fragment reading from ORF 1 towards ORF 2 (see Fig. 1). (b)The corresponding sequence of
the spo-42 mutant. The arrow points to the adenylic acid residue that replaces the wild-type guanylic
acid at position 670.
than a change from alanine to threonine. Secondly, we have tried, by using DNA from a spo69 Iys+ strain to transform a spo+ 1’s strain and selecting for Lys+, to obtain recombinants that
carry only the mutation at position 732; but even though we can calculate that several of these
should have arisen we have been unable to find any recombinants with a distinctive Spo
phenotype (M. D. Yudkin, unpublished work). Thirdly, genetic mapping unequivocally places
spo-69 on the lys-distal side of spo-42 (Yudkin & Turley, 1981),which suggests that what is being
mapped in these experiments is the mutation at position 571. We therefore conclude that it is
this mutation that is responsible for the Spo- phenotype; presumably the additional change at
position 732 occurred in the original N-methyl-N’-nitro-N-nitrosoguanidinemutagenesis that
produced spo-69, or, possibly, by subsequent genetic drift in the inactivated gene.
The fact that mutations spo-42 and spo-69 both lead to the replacement of glycine residues by
aspartic acid residues is presumably due to chance, but it allows us to focus attention on the way
in which the position of the amino-acid change affects the phenotype. Strains carrying spo-42
produce spores at about
of the wild-type incidence; in other words the mutation is leaky
(Yudkin & Turley, 1981). Strains carrying spo-69, by contrast, produce no detectable spores
whatever (Yudkin & Turley, 1981), so we may conclude that residue 62 is more critical to the
function of the ORF 1 protein than is residue 95.
The importance of amino-acid residue 62 is further emphasized by our finding that mutation
sas-2, which was selected as leading to partial reversion of spo-69, results in a change in this
of the
residue from aspartic acid to asparagine. Strains carrying sas-I make spores at about
wild-type incidence (Yudkin & Turley, 1980). Thus, although the protein has its full function
with glycine at position 62, a change to asparagine allows partial function; on the other hand,
aspartic acid at the same position totally abolishes activity, no doubt because of its charge.
Since sas-Z was selected from a mutant carrying spo-69, which has a mutation at nucleotide
732, one might ask why the sas-2 sequence is wild-type at this position. In order to answer this
question, we have to trace the history of sas-2. From a strain carrying spo-69, Yudkin & Turley
(1980) selected a partial revertant which proved still to harbour spo-69. This revertant was
presumed by Yudkin & Turley (1980) to be a double mutant, spo-69sas-I; in the light of the
present results, which show that spo-69 is itself a double mutant, we now believe that the
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Mon, 19 Jun 2017 00:06:58
962
M . D . YUDKIN AND OTHERS
revertant actually contained three mutations, at positions 570, 571 and 732. When DNA from
this revertant was used to transform a spo+ strain, recombinants were obtained at a frequency of
about 0.2%, which were presumed to carry sas-1 alone. It is this genotype that we have now
sequenced: since it contains transitions at positions 570 and 571 but not at position 732 we
conclude that the last of these mutations was crossed out in the transformation. Thus the
designation sas-I of Yudkin & Turley (1980) actually refers to transitions in two adjacent base
pairs.
Six different mutations in ORF 1 have been isolated on the basis of their phenotypic effects on
sporulation and have been genetically mapped in this laboratory (Yudkin & Turley, 1981;
Errington & Mandelstam, 1983). Of these six, spo-69 is the closest to that region of ORF 1 that
determines the N-terminus of the protein. The fact that we have now found that spo-69 alters
position 62 of a 117-residueprotein suggests that mutations in the N-terminal half of the protein
may be without effect on sporulation and may therefore have escaped detection.
These studies have allowed us to make a start in relating structure to function in a protein that
is involved in the regulation of spore development. We are now investigating other mutants,
including some that prevent sporulation but allow the synthesis of the sporulation-associated
alkaline phosphatase and DNAase to continue.
We are grateful to Professor J. Mandelstam for his valuable comments on the manuscript. P. Fort was supported
by Elf Aquitaine (UK).
REFERENCES
CHAK, K.F., LENCASTRE,
H. DE, LIU,H.-M. & PIGGOT,
P. J. (1982). Facile in vivo transfer of mutations
between the Bacillus subtilis chromosome and a
plasmid harbouring homologous DNA. Journal of’
General Microbiology 128, 28 13-28 16.
ERRINGTON,
J. & MANDELSTAM,
J. (1983). Variety of
sporulation phenotypes resulting from mutations in a
single regulatory locus, spoIIA, in Bacillus subtilis.
Journal oj’General Microbiology 129, 209 1-2 101.
FORT,P. & PIGGOT,P. J. (1984). Nucleotide sequence
of sporulation locus spoIIA in Bacillus subtilis.
Journal of General Microbiology 130, 2147-21 53.
GUERRY,P., LE BLANC,D. J. & FALKOW,S. (1973).
General method for the isolation of plasmid deoxyribonucleic acid. Journal o j Bacteriology 116, 10641066.
HANAHAN,
D. (1983). Studies on transformation of
Escherichia coli with plasmids. Journal of’ Molecular
Biology 166, 557-580.
KLEIN,R. D., SELSING,
E. & WELLS,R. D. (1980). A
rapid microscale technique for isolation of recombinant plasmid DNA suitable for restriction enzyme
analysis. Plasmid 3, 88-91.
LINN,T. & LOSICK,R. (1976). The program of protein
synthesis during sporulation in Bacillus subtilis. Cell
8, 103-114.
LIU, H.-M., CHAK,K. F. & PIGGOT,P. J. (1982).
Isolation and characterization of a recombinant
plasmid carrying a functional part of the Bacillus
subtilis spoIIA locus. Journal ojGeneral Microbiology
128, 2805-28 12.
LOSICK,R. (1982). Sporulation genes and their regulation. In The Molecular Biology ojthe Bacilli, pp. 179201. Edited by D. A. Dubnau. New York: Academic
Press.
MESSING,J. & VIEIRA,J. (1982). A new pair of
M13 vectors for selecting either DNA strand of
double-digest restriction fragments. Gene 19, 269276.
OLLINGTON,
J. F. & LOSICK,R. (1981). A cloned gene
that. is turned on at an intermediate stage of spore
formation in Bacillus subtilis. Journal ojBacteriology
147, 443-45 1.
PIGGOT, P. J. (1973). Mapping of asporogenous
mutations of Bacillus subtilis: a minimum estimate of
the number of sporulation operons. Journal of
Bacteriology 114, 1241-1 253.
PIGGOT,P. J. & COOTE,J. G. (1976). Genetic aspects
of bacterial endospore formation. Bacteriological
Reviews 40,908-962.
A. R. (1977).
SANGER,
F., NICKLEN,S. & COULSON,
DNA sequencing with chain terminating inhibitors.
Proceedings ojthe National Academy ojSciences ojthe
United States o j America 74, 5463-5468.
YOUNG, M. & MANDELSTAM,
J. (1979). Early events
during bacterial endospore formation. Advances in
Microbial Physiology 20, 103- 162.
YUDKIN,M. D. & TURLEY,L. (1980). Suppression of
asporogeny in Bacillus subtilis. Allele-specific suppression of a mutation in the spoIIA locus. Journal of’
General Microbiology 121, 69-78.
YUDKIN,
M.D. & TURLEY,
L. (1981). Mapping of six
mutations in the spoIIA locus of Bacillus subtilis and
studies of their response to a nonsense suppressor.
Journal o j General Microbiology 124, 255-261.
YUDKIN,M. D., BOSCHWITZ,H., LORCH, Y. &
KEYNAN,A. (1982). Changes in the pattern of
protein synthesis during the first three hours of
sporulation in Bacillus subtilis. Journal of General
Microbiology 128, 2 165-2 177.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Mon, 19 Jun 2017 00:06:58