Download Acquisition of Thymidylate Synthetase Activity by a Thymine

J. gen. Virol. (1969), 4, 489-5o4
Printed in Great Britain
489
Acquisition of Thymidylate
Synthetase Activity by a Thymine-requiring Mutant of Bacillus
subtilis following Infection by the Temperate Phage ¢3
By R. G. T U C K E R
Microbiology Unit, Department of Biochemistry, University of Oxford
(Accepted 29 October 1968)
SUMMARY
Infection of a thymineless strainof Bacillus subtiliswith citherthe temperate phage 43 or its clear plaque mutant ~3c enabled D N A to be formed in
the absence of added thymine. Growth of phage ~3c under these conditions
was associated with an enhanced rate of D N A synthesis comparcd with
uninfected cultures growing in the presence of thymine; when phage 43
became a prophage the lysogcnized cellssynthesized D N A at the same rate
as the uninfected bacteria with thyminc. The thymine independence of
phage-infected bacteria was due to the acquisition of the cnzyme thymidylate synthctase absent in the uninfected mutant bacteria.Phage ~3c growth
caused a rapid risein the specificactivityof the enzyme aftera short lag,and
by the time the cellslysed there was approximately ten times more activity
than in uninfected wild-type B. subtilis.In thymineless B. subtilislysogenizcd
with phage 43 the amount of thymidylate synthetase was the same as in
wild-type B. subti[is.Phage D N A was ablc to transform thymine-requiring
B. subtilisto thymine independence, showing that it contained the gene for
thymidylate synthctasc.Plaquc-forming particlescould not be separated by
density gradient centrffugationfrom those possessing the thymidylate synthetase gcne. This result, together with the failure to gct 'thyminelcss'
mutants of the phage to regain their abilityto promote thymine synthesis,
suggested that phage 43 was a converting phage. However, the rcsults of
transformation implied that the phage thymidylate synthctasc gene was
closely related to that of the recipientcells,and may have originated from
the bacterium.
INTRODUCTION
The acquisition of enzymic activity as a result of phage infection is a well-recognized
phenomenon and is often due to the transfer with the phage DNA of the genetic
information for the synthesis of the enzyme. The need for enzyme modification is most
evident where the phage contains components that are absent from the uninfected host
bacterium (Flaks & Cohen, 1959; Kahan, I963; Roscoe & Tucker, I966) but the reason
for other changes is not always clear. Some are presumably the result of subtle metabolic requirements for phage growth, but others seem to be fortuitous, having no role
in phage development, though possibly of importance in the broad context of genetic
exchange in bacteria (Sneath, I962). Transduction provides an example of the fortuitous acquisition of new enzymes following phage infection. Generalized transduction is widespread among the different genera of bacteria but far fewer examples of
specialized transduction are known; there is no reason to suppose that the process is
uncommon, merely that by its nature it is more difficult to detect. Similarly, phage
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 14:30:36
49 °
R.G. TUCKER
conversion is found in many different bacterial groups, and in the modified somatic
antigens of Group e Salmonellae the enzymic changes brought about by phages have
been examined in detail (Robbins et al. I965). As with specialized transduction, there
is no systematic way of detecting phage conversion, and this probably accounts for
the paucity of examples other than the easily recognizable changes due to altered
surface properties of cells. I report here the acquisition of the enzyme thymidylate
synthetase by bacteria infected with the temperate subtilis phage 43, and discuss the
possible relation this phenomenon bears to specialized transduction and conversion.
METHODS
Bacteria. Bacillus subtilis NCTC 3610 was used as the host for phage growth and as
the indicator for the phage assay; B. subtilis L 4 was a derivative of B. subtilis 36Io
lysogenized with phage ~3; B. subtilis 168 was the indole-rcquiring, transformable
strain described by Spizizen (I958); B. subtilis 168 thy-I, a thymine-requiring mutant
of B. subtilis 168 (Farmer & Rothman, 1965) was kindly supplied by Dr J. Farmer, and
an independently isolated thymineless mutant, B. subtilis 168 thy-2, was kindly supplied by Dr C. Anagnostopoulos. Stocks of bacteria were maintained on slopes of
nutrient agar (Oxoid Ltd, London, S.E. t) which for strains requiring thymine was
supplemented with 5 × lO-4 M-thymine. Liquid cultures for most of the experiments
were grown in 200 ml. volumes in I 1. conical flasks and aerated at 37 ° on a gyrotary
shaker. The number of bacteria in a culture was estimated routinely by measuring the
turbidity in an EEL (Evans Electroselenium Ltd, Halstead, Essex) photoelectric colorimeter with a neutral density filter, referring to a standard curve relating turbidity to
viable cell count.
Media. Nutrient broth for the production of phage lysates and the potato extract
for the growth of spores were described by Roscoe & Tucker 0966). Minimal medium
was the minimal salts solution of Davis & Mingioli (195o) used at twice the concentration and supplemented with ferric citrate and glucose before use (Haslam, Roscoe
& Tucker, I967). This medium was satisfactory for B. subtilis 36Io but for B. subtilis
I68 and its derivatives the addition of o.o 4 % acid hydrolysed casein (Oxoid) together
with any specific growth requirements was necessary for speedy growth. Solid, defined
media consisted of minimal salts solution containing 1"5 % (w/v) Oxoid no. 3 agar
which, after autoclaving, was supplemented with 5 × lO-3 M-glucose, 6.6 × Io -5 M-Lasparagine and 6.6 × lO-5 M-DL-glutamic acid. Further additions were made as follows:
5 × io -5 M-L-tryptophan (ST plates); 5 × lO-4 M-thymine (SP plates); 5 × IO-5 M-Ltryptophan plus 5 × lO-4 M-thymine (STP plates).
Phage assay. Standard plating procedures were used (Adams, I959). Agar medium
for the bottom layer of the plates contained per litre: nutrient broth CM1 granules
(Oxoid), 13 g. ; NaCl, 3 g. ; glucose, I g. ; Oxoid no. 3 agar, I5 g. For the overlay the
same medium was used but with agar (7 g./1.) Phage suspensions were diluted in
nutrient broth and added to tubes of 3 ml. melted overlay medium containing lO8
spores of B. subtilis 361o. Plates were incubated overnight at 25 °.
Phage lysates. Preparations of phage 43 were generally obtained as follows. An
overnight culture of B. subtilis L 4 in broth was added to I 1. fresh broth and aerated in
a rotating 5 1. flask (Mitchell, 1949) at 37 ° until the cell density reached Io 8 bacteria/ml.
Four ml. of 2 vol. H20~ were added per 1. of culture and aeration was continued for
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 14:30:36
Phage-induced thymidylate synthetase
49I
about an hour when the opacity of the culture had fallen maximally. Crude lysates of
phage 43 contained about lO2° p.f.u./ml., though sometimes titres were much lower.
In the later stages of the work more reproducible titres were obtained by adding
2 x lO-2 M-3-amino-I,2,4,-triazole, a catalase inhibitor (Margoliash, Novogrodsky &
Schejter, I96O), to the cultures induced by hydrogen peroxide.
Phage ~3c was a clear-plaque mutant of phage 43, picked from a phage 43 assay
plate and taken through three successive single-plaque isolations. Lysates of the phage
were prepared by infecting cultures of Bacillus subtilis 36IO, freshly grown to IOa cells
ml. in aerated broth, with approximately fivefold excess of phage ~3c. Aeration was
continued until lysis was maximal. Crude lysates contained about lOl° p.f.u./ml.
Cells and debris were removed from lysates by centrifugation at 2ooo g for 3o min.
and the clear supernatant fluid was transferred to a beaker to which solid ammonium
sulphate (4oo g./l.) was added with stirring. After standing overnight at 4 ° the precipitated material was collected by centrifuging, and resuspended in a buffered salts
solution (Hayes, 1957). The suspension was clarified by centrifugation at low speed,
then the phage was sedimented in an ultracentrifuge at 55,0oo g for 6o rain. The pellets
were suspended in buffered salts and treated with lysozyme (Armour Laboratories
Ltd, Hampden Park, Eastbourne, Sussex), IOO #g./ml. ; ribonuclease (Armour) IO #g.[
ml., and DNase (Calbiochem Ltd, London, W. I), IO #g./ml. at 37 ° for 30 min. The
suspension was clarified by centrifuging at IO,OOOg for 15 min., then the particles were
sedimented at 5o,0oo g for 6o min. and finally suspended in buffered salts at a concentration of approximately IOTM p.f.u./ml.
Preparation of DNA. A culture of B. subtilis, grown with aeration for I6 hr in
I 1. nutrient broth, was centrifuged at 2ooo g for 2o rain. and washed once with
40o ml. buffered citrate saline (o.14 M-NaC1; o.or M-sodium citrate; o'ot M-phosphate
buffer, pH 7"o). The bacteria were suspended in 30 ml. buffered citrate saline and
lysozyme added to a concentration of I mg./ml. After 2o min. at 37 °, sodium dodecylsulphate was added to a concentration of 0.2 % to complete the lysis and the mixture
diluted with 30 ml. ice-cold buffered citrate saline. An equal volume of phenol saturated with lO-7 M-phosphate buffer, pH 7"0, was now added and the mixture shaken
gently for 2o min. The aqueous and phenol phases were separated by centrifugation
and after re-extracting the aqueous layer with fresh phenol the nucleic acid was
precipitated by adding two volumes ethanol. The precipitate was dissolved in a solution containing o. 15 M-NaCI and o.o 15 M-sodium citrate (SSC) and treated with RNase
(50/zg./ml. of a solution in o.14 M-NaC1 which bad been heated at 80 ° for IO min. to
destroy any contaminating DNase) for 45 min. at 37 °. After another precipitation with
ethanol the nucleic acid was dissolved in SSC and dialysed overnight against the same
solution. For the purposes o f base analysis, traces of R N A that this material contained
were removed by treatment with charcoal (Zamenhof & Chargaff, 195 I). Phage nucleic
acid was prepared by the method of Mandell & Hershey (I96O), followed by overnight
dialysis against SSC.
Estimation of DNA in bacterial cultures. Ten ml. volumes of culture were pipetted
into I ml. 50 % trichloracetic acid in a centrifuge tube cooled in ice. After storage
overnight in ice the tubes were centrifuged and the supernatant fluid removed by
decantation and thorough draining. The precipitated material was resuspended in
1.5 ml. 5 % trichloracetic acid and heated in a water bath at 9°o for 30 min. The tubes
were cooled and centrifuged and 0"5 ml. volumes of the supernatant fluid used for the
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 14:30:36
492
R.G. TUCKER
determination of purine-bound deoxyribose by the diphenylamine reaction (Burton
I956). 2-Deoxyribose (Calbiochem) was used to make a standard curve.
Transformation. The media and procedures were those of Anagnostopoulos &
Spizizen (I96i) as modified for B. subtilis i68 thy-i by Farmer & Rothman (t965).
D N A was added to competent cultures after 60 min. incubation in the secondary
growth medium and exposure was terminated by adding DNase (50 #g./ml.) Io or
I5 min. later. The culture was incubated for a further 45 rain. before being plated on
suitable selective solid media.
Preparation of cell-free enzyme extracts. B. subtilis I68 thy-I was grown with aeration overnight in nutrient broth containing 5 × Io -4 M-thymine. The bacteria were
diluted to about 5 × Io7 cells/ml, in 200 ml. volumes of minimal medium supplemented
with acid hydrolysed casein (o'04 %, w/v), L-tryptophan (I'25 x lO-4 M) and thymine
(Io -4 M), and aerated at 37 ° till the cell count was 3 × Ioa cells/ml. The cultures were
cooled by pouring them into half their volume of frozen minimal salts solution and the
cells harvested by centrifugation. The bacteria were now resuspended in the original
culture volume of minimal medium with acid hydrolysed casein and L-tryptophan as
before, but lacking thymine, and distributed in IOO ml. volumes in I 1. conical flasks.
The bacterial suspensions were equilibrated at 37 ° for I5 rain., with aeration, phage
was added and the aeration continued. Individual flasks were removed at intervals and
their contents rapidly cooled by tipping into 50 ml. frozen minimal medium. The cells
were sedimented by centrifugation, resuspended in 5 ml. buffer A (Haslam et al. I967)
and disrupted at 25 kcyc./sec, for 4 min. in a Mullard Ultrasonic Generator Type
E 759oB. Unbroken cells and debris were removed by a preliminary centrifugation at
Io,ooo g for 60 rain. Extracts of uninfected bacteria were obtained in a similar manner.
The protein content of the extracts was estimated by the method of Lowry et al. (I95 I),
using crystalline bovine serum albumin (Armour) for the standard curve.
Assay ofthymidylate synthetase. The bacterial extracts, prepared as described above,
were used without further purification. Thymidylate synthetase was estimated by a
spectrophotometric method (Haslam et al. I967) based on that of Wahba & Friedkin
(1962).
Antiserum. Purified, concentrated phage suspensions were injected subcutaneously
and intravenously into rabbits on three alternate days, and a week later the intravenous series of injections was repeated. After a further week blood was collected and
tested for antiphage activity.
RESULTS
Properties of phage (~3
Phage ~3 was isolated from garden compost by conventional enrichment techniques
using Bacillus subtilis 36IO as the host bacterium. Electron microscopy showed tadpoleshaped particles with heads which were hexagonal in outline and 57 nm. in diameter,
and slender tails 270 nm. long. Under the plating conditions used the phage formed
turbid plaques approximately I mm. in diameter, with an occasional clear plaque on
crowded plates. Bacterial growth picked from the centre of a turbid plaque with a
sterile needle was streaked on to nutrient agar and several isolated colonies appearing
after incubation were purified by streaking on three successive occasions on fresh
nutrient agar plates. All the colonies obtained in this way produced peripheral zones
of lysis when replica-plated on to a lawn of B. subtilis 36Io, and one of them, designated strain L4, was selected as stock for further study.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 14:30:36
Phage-induced thymidylate synthetase
493
These initial results could be explained if phage ~b3 either formed some carrier-type
association with the host or if there was a lysogenic relation between them. Evidence
for the latter was obtained from a number of experiments. Thus when isolated colonies
of Bacillus subtilis L4 were replica-plated on a lawn of B. subtilis 36IO virtually all of
the transferred regions were surrounded by a zone of lysis after overnight incubation.
This behaviour persisted after numerous single colony isolations of B. subtilis L 4 and
was unaffected either by prior growth of the bacteria in the presence of antiphage
serum or by two cycles of growth in the presence of Io -4 M-proflavine. No difference
between the rates of growth of B. subtilis 36Io and B. subtilis L4 in broth or minimal
medium could be detected. Culture supernatants of the latter organism, however,
always contained free phage 43 particles, and the numbers could be greatly increased
by a variety of treatments, such as u.v. irradiation, exposure to mitomycin C, addition
of hydrogen peroxide, known to induce phage growth in lysogenic bacteria. For routine
purposes lysates of phage ~b3 were obtained by hydrogen peroxide treatment of broth
cultures of B. subtilis L 4.
The rates of growth of strains 36Io, I68 and I68 thy-I of Bacillus subtilis in minimal
medium, supplemented where necessary with tryptophan and thymine, were little
affected by the addition of phage ~b3, even with a I5-fold excess of phage to cells. In
contrast, infection of the same organisms with clear-plaque mutant phage ~b3cresulted
in distinct lysis. The growth of cultures of B. subtilis L 4 was unchecked either by phage
43 or by phage ff3c, and neither phage produced plaques on indicator lawns of this
bacterium.
Formic acid hydrolysis of the nucleic acids extracted from phages ~b3 and ~3c
followed by paper chromatography (Wyatt, I95 I) showed that they contained all the
normal DNA bases but that the guanine plus cytosine (GC) content of 36 ~o was significantly less than that of the DNAs of B. subtilis 36Io or B. subtilis L4, both of which
had a GC content of 43 %- A limited comparison was made between phage ~b3 and
two subtilis phages that might have been related to it: phage SP3, which has the same
GC content as phage ~b3 (Marmur & Cordes, I963) but is not stably lysogenic, and
phage ~kIo5, a temperate phage isolated by Dr B. Reilly. Neither of these phages was
inactivated by antiserum against phage ~b3, and both formed plaques on B. subtilis L4,
which was immune to phage ~b3. Phage ~b3 therefore was distinct from these other
subtilis phages.
DNA synthesis in bacteria infected with phage (~3
As with many phages, the growth of phage ~3c in host bacteria, e.g. Bacillus subtilis
36Io or B. subtilis I68, was accompanied by an enhanced rate of DNA synthesis until
the culture lysed. Experiments were also done to determine the effect of phage infection on the DNA metabolism of thymineless mutants of B. subtilis. A culture of
B. subtilis I68 thy-I was grown to a concentration of 3 x io 8 ceUs/ml, in 4oo ml. minimal medium supplemented with acid-hydrolysed casein (0"o4 % w/v), t.-tryptophan
(I'5 x Io -4 M) and thymine (IO -4 M). The organisms were chilled rapidly by pouring
them on to frozen double-strength minimal salts solution (Davis & Mingioli, I95O),
then centrifuged and washed with fresh salts solution. The bacteria were taken up in
the original culture volume of minimal medium, supplemented with acid-hydrolysed
casein and tryptophan as before but without added thymine, and the suspension divided
into four IOO ml. volumes in I 1. conical flasks. Thymine (I0 -4 M) was added to two of
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 14:30:36
494
R.G. TUCKER
the flasks and phage 43c was added to one of these to give approximately tenfold
excess of phage to cells. Phage 43c was also added to one of the two remaining flasks
lacking thymine. All the cultures were incubated with aeration at 37° and Io ml.
samples were withdrawn at intervals for the estimation of DNA. In the presence of
thymine the results were no different from those with wild type bacteria. However,
there was also appreciable DNA synthesis by the culture infected with phage 43c in
the absence of thymine (Fig. r a), suggesting the ability of this phage, like certain
coliphages (Barner & Cohen, I959), to promote thymine biosynthesis in thyminerequiring bacteria. Similar burst sizes were obtained in one-step growth experiments in
which B. subtilis I68 thy-I was infected in minimal medium with or without thymine,
showing that the DNA synthesized in the absence of thymine was fully functional.
The clear-plaque mutant phage 43 used in the previous experiments lysed the bacteria
it infected, so it was of interest to determine what effect the temperate phage 43 would
have on the DNA synthesis of thymineless bacteria. A culture of B. subtilis I68 thy-I
was grown, washed and resuspended as above, then infected with a tenfold excess of
phage 43. The infected culture synthesized DNA in the absence of added thymine and
it continued to do so well beyond the time at which cultures infected with the clear
plaque mutant phage 43 lysed (Fig. I b). Thus it may be concluded that the phageinduced synthesis of thymine can be obtained even when the phage enters the prophage
state.
Acquisition of thymine independence by phage 43-infected B. subtilis
Since Bacillus subtilis I68 thy-I infected by phage 43 can form DNA in the absence
of exogenous thymine it seemed likely that lysogeny would confer on the bacteria the
ability to grow without thymine. A culture ofB. subtilis I68 thy-I, grown to a concentration of 3"5 x ioa cells/ml, in minimal medium with added acid-hydrolysed casein
(o'o4 % w/v), L-tryptophan 0.25 × IO-4 M) and thymine 0 o -4 M), was infected with a
suspension of phage 43 at a cell:phage ratio of Io: i and incubated at 37°. Samples of
the infected culture were removed at intervals and after dilution were spread in duplicate on three sets of minimal agar medium supplemented with tryptophan plus thymine
(STP plates), with thymine (SP plates) and with tryptophan (ST plates). The first set
permitted the growth of all the bacteria: the other two the growth only of those able to
synthesize tryptophan and thymine respectively. All the plates were incubated overnight at 37°. Infection of B. subtilis I68 thy-I with phage 43 resulted in the rapid
acquisition by the majority of the cells of the ability to grow on the medium lacking
added thymine, but there was no corresponding growth on the plates that lacked
tryptophan (Fig. 2).
Two consistent features of this type of experiment were the slight initial decrease in
numbers of colonies appearing on STP plates and a later decrease, about 5o min. after
infection, in colony counts both on ST and STP plates. The reason for these apparently
temporary losses of viability are unknown: possibly they were the result of physiological changes caused by plating at these particular times, since the balance between
lysogenic and lytic response is often susceptible to environmental changes (Luria et al.
1958). The relation between phage:cell ratio and the number of organisms gaining the
ability to synthesize thymine and thereby grow on ST plates is shown in Fig. 3. The loss
of viability occurring when the bacteria were plated immediately after infection with
phage 43 was a complicating factor, and an exposure time of 2o rain. before plating
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 14:30:36
Phage-induced thymidylate synthetase
0.7 !
495
(a)
~, 0.6
0.5
0
,.o
0
,i
0.4
I
0.3
0
J~
<
Z
O~
0
'"
0.1
20
40
60
80
20
40
60
Minutes after infection
80
I00
120
t40
160
Fig. I. Effect of phage infection on DNA synthesis by thymine-requiringBacillus subtilis 168
thy-I. (a) DNA content of control cultures, without ( O - - O ) and with ( 0 - - 0 ) thymine,
and of phage ¢3c-infected cultures, without ( A - - A ) and with ( & - - A ) thymine. (b) DNA
content of control cultures, without ( O - - O ) and with ( 0 - - 0 ) thymine, and of phage
¢3-infected cultures, with ( A - - A ) and without ( & - - A ) thymine.
I
I
20
40
I
I
I
I
60
80
1 O0
120
600
~
500
400
300
o
~ 2oo!
100
140
Minutes after infection
Fig. 2. Acquisition of thymine independence after infection with phage ¢3.
STP plates (O--O), ST plates ( Q - - O ) SP plates ( A - - A ) .
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 14:30:36
496
R.G. TUCKER
was therefore chosen, since it combined good recovery with least growth of the culture.
With a tenfold multiplicity of phage to cells, 8o % of the bacteria became capable of
growing in the absence of thymine within 2o rain.
Growth of phage ff3-infected Bacillus subtilis 168 thy-I without thymine is a stable
property. A colony picked at random from those growing on an ST plate in the
previous experiment was transferred to nutrient broth containing 5 x io - t M-thymine
plus antiphage serum and incubated overnight at 37 °. The culture was then diluted and
suitable volumes were spread on to STP plates to permit the growth of thyminedependent as well as thymine-independent organisms. The colonies appearing after
I
I
I
1
I
0
O
8
.8> 7
I::I
0
¢,1
0
6
5
i
,,I
Log. phage:coil ratio
Fig. 3. Relation between phage :cell ratio and growth without exogenous thymine. STP
plates (O--©), ST plates (Q--O). Colonies were counted after overnight incubation at 37°.
incubation were replica-plated on to ST plates and also on to a lawn of B. subtilis 361o
on nutrient agar to detect phage. In all, 45o colonies were examined: all grew in the
absence of thymine and all were lysogenic.
To determine the proportion of phages able to confer thymine independence on
Bacillus subtilis I68 thy-I, stock suspensions of phage 43 were diluted and plated
against B. subtilis 168 thy-I as indicator using nutrient agar medium supplemented with
5 x Io -4 M-thymine. Individual plaques were picked off with a sterile needle and transferred to marked positions on fresh nutrient agar containing thymine. After overnight
incubation at 37 ° patches of bacterial growth appeared where the plaques had been
transferred, consisting of B. subtilis 168 thy-1 lysogenized with phage 43, and these
were replica-plated on to ST plates. Any replicated patches that failed to grow on
further incubation were presumed to contain 'thymineless' phage, but this was
tested by spreading the bacteria from these patches on to ST and STP plates, and also
spotting on to a lawn of B. subtilis 36IO to confirm the presence of phage. Out of a
total of 1364 plaques examined in this way, 46 (3"4 %) consisted of phage unable to
confer thymine independence on B. subtilis 168 thy-i. These 'thymineless' derivatives
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 14:30:36
Phage-induced thymidylate synthetase
497
of phage 43, designated phage ~b3T-, were indistinguishable from the present phage
apart from their failure to convey thymine independence. Moreover, crowded assay
plates contained clear-plaque mutants (phage ~b3cT-) which lysed growing cultures of
B. subtilis and gave similar burst sizes to those of phage $3c in one-step growth
experiments.
Thymidylate synthetase activity following ~b3 infection
The growth and formation of DNA by Bacillus subtilis 168 thy-I in the absence of
thymine after phage infection implied that phages 43 and $3c could correct the biochemical defect of this bacterium, namely the absence of the enzyme thymidylate
synthetase (Farmer & Rothman, 1965). To confirm this, B. subtilis 168 thy-I was
grown in minimal medium containing acid-hydrolysed casein (0"04 %), L-tryptophan
(i'25 x lO-4 M) and thymine (lO -4 M) to a concentration of 3 x io s cells/ml. The bacteria
were harvested, resuspended in the same medium without added thymine and infected
|
300
!
I
(a)
I
I
I
~0
O
I
(b)
2
~ 200
E
0
Q.
100
ff
j
10
20
30
20
40
60
80
100
120
Minutes after infection
Fig. 4- Specific activity of thymidylate synthetase in extracts of Bacillus subtilis 168 thy-I
followingphage infection. (a) Cultures infectedwith phage ~b3c,(O--O) with L-tryptophan,
(A--&), without L-tryptophan, (A--A), with L-tryptophan and 50 #g./rnl. chloramphenicol; (O--O), uninfected culture with L-tryptophan. (b) Cultures infected with phage ~3:
(O--O), with L-tryptophan; (Q--O), uninfected control culture with L-tryptophan.
with a tenfold excess of phage ~b3c. Samples were removed at intervals and cell-free
extracts of the bacteria were assayed for thymidylate synthetase activity by the spectrephotometric method (Fig. 4a). After a lag of at least 8 min. the enzyme was detectable
and its specific activity increased rapidly to 30o m/,mole H4PtG oxidized/hr/mg, protein by the time the cells had started to lyse. The validity of the spectrophotometric
method for assaying thymidylate synthetase activity under these conditions was confirmed by showing that extracts from infected cells had acquired the ability to incor32
J. Virol. 4
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 14:30:36
498
R.G. TUCKER
porate label from [14C]HCHO into thymidylic acid. No increase in enzyme activity
was obtained if chloramphenicol (5o/zg./ml.) was added to the medium with the phage
or if tryptophan, one of the growth requirements of the bacterium, was omitted from
the medium used for phage infection, both results showing that de novo protein synthesis was needed for the appearance of the enzyme. Similar experiments were made
with the temperate phage 43 to determine whether it, too, could generate thymidylate
synthetase activity in B. subtilis I68 thy-I. The bacteria were grown and transferred to
medium lacking thymine as before. A tenfold excess of phage to cells was added,
giving approximately 80 % lysogenization, and samples were taken at intervals to
determine their enzyme content. The specific activity of thymidylate synthetase in the
300
T
T
I
I
i
1
I
10
10
o
2
?
8~
x
7i
0
r~
~E 200
80
~
X
6-
7. I
O
"6
6~J
O
=
N 10o
N
O
3-
: 4
2-
-- 3
1 -
.I.
I ~
10
20
Minutes after infection
30
50
60
70
Drop number
80
Fig. 6
Fig. 5
Fig. 5. Specific activity of thymidylate synthetase in cell extracts of Bacillus subtilis 168
following phage infection. (©--©), culture infected with phage ~b3c; (A--A), culture
infected with phage ~b3cT-; (@--@), uninfected control culture.
Fig. 6. Density gradient centrifugation of phage ~3. (@--@), plaque-forming particles
against Bacillus subtilis 36Io; (©--©), particles carrying the thymidylate synthetase gene.
extracts increased rapidly from zero at the time of infection to 60 m#mole H4PtG
oxidized/hr/mg, protein zo min. later. Thereafter the specific activity declined, its
value remaining at about 35 m/~mole H4PtG oxidized/hr/mg, protein until the end of
the experiment (Fig. 4b).
Infection of wild type Bacillus subtilis 36[o or B. subtilis r68 with phage ~3c also
caused an increase in the thymidylate synthetase activity of cell-free extracts. In a
typical experiment the specific activity of the enzyme increased from an initial value of
36 to 4o0 m#mole H4PtG oxidized/hr/mg, protein after zo min. The dramatic increase
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 14:30:36
2
Phage-induced thymidylate synthetase
499
in enzyme activity after phage infection, both of the wild type and of thymineless
B. subtilis, probably represented an uncontrolled expression of the thymidylate synthetase gene rather than a requirement for thymine that the existing host mechanism
was unable to provide. For example, when the 'thymineless' phage ¢3cT- was used to
infect B. subtilis I68 the thymidylate synthetase activity remained at about 36 m/zmoles H4PtG oxidized/hr/mg, protein but the yield of phage was the same as with
phage ¢3c which did increase enzyme activity (Fig. 5).
The origin of the thymidylate synthetase gene of phage ¢3
The appearance of the enzyme thymidylate synthetase after infection of thymineless
Bacillus subtilis with phage ¢3 could be readily explained if the DNA injected by the
phage contained the genetic information necessary for the formation of this enzyme.
Phage ¢3 must therefore have been capable either of lysogenic conversion or of transduction of a portion of the bacterial chromosome. The possibility that the phage was a
generalized transducing phage was ruled out by the experiment showing that approximately 97 ~/o of the plaque-forming particles were able to confer thymine independence
on B. subtilis 168 thy-I, and also more directly by the failure to transduce a variety of
amino-acid auxotrophs of B. subtilis 36IO. If phage ¢3 is a specialized transducing
phage it should have been possible, by analogy with phage A, for' thymineless' derivatives of the phage to acquire the thymidylate synthetase gene following growth in wild
type bacteria. Attempts to demonstrate transduction of thymine biosynthesis using
lysates of B. subtilis 361o lysogenized with phage ¢3T- which were obtained either by
hydrogen peroxide treatment or by induction with u.v. irradiation were, however,
unsuccessful. Further experiments were done to determine whether it was possible to
separate plaque-forming particles from transducing particles by centrifugation in a
density gradient (Weigle, Meselson & Paigen, 1959). A suspension of phage ¢3 diluted
to 5 x io 10p.f.u./ml, in 1.5 X IO-~ M-tris+HC1, pH 8, was mixed with solid CsCI to
give a solution of density 1"511 g./cm8. The phage was centrifuged at 27,0oo rev./min.
for 16 hr at 2o ° in the SW 39 rotor of a Spinco Model L ultracentrifuge. The centrifuge
tubes were then punctured at the bottom and single drops collected in 2 ml. o.14 MNaC1. Fractions were analysed for plaque-forming particles by mixing with spores of
B. subtilis 36IO and plating on nutrient agar, and also for particles conferring thymine
independence by mixing with a culture of B. subtilis 168 thy-1 and plating on ST medium after 15 min. incubation at 37°. Coincident peaks of activity were obtained (Fig. 6)
showing that there was no difference in density between the plaque-forming particles
of phage ¢3 and those carrying the thymidylate synthetase gene. In a further experiment B. subtilis 168 thy-I lysogenized with the 'thymineless' phage ¢3T- was used as
the indicator for phage carrying the thymidylate synthetase gene in order to supply
functions that might be absent from any defective transducing particles. Again, there
was no density difference between the two phage activities, and it must be presumed
that the thymidylate synthetase gene is not carried by defective particles.
Synthesis of thymine after the uptake of phage nucleic acid
Since competent cultures of Bacillus subtilis 168 can incorporate homologous DNA
it should be possible to determine whether the thymidylate synthetase gene of phage
¢3 is specific for the phage or whether the gene originally came from the bacterium.
Cultures of B. subtilis 168 thy-I were prepared for transformation by conventional
32-2
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 14:30:36
500
R.G. TUCKER
procedures (Anagnostopoulos & Spizizen, I96I; Farmer & Rothman, I965). After
60 rain. incubation at 37 ° in the secondary growth medium containing Io -~ Mthymidine, D N A from phage ~b3 was added, followed I5 min. later by DNase (50/~g./
ml.). Incubation was continued for another 45 min. before the cultures were diluted
and spread on plates of ST and STP medium. Phage ~b3c D N A was able to transform
B. subtilis I68 thy-I to thymine independence, and the frequency of transformation
increased linearly with the concentration of D N A up to a plateau (Fig. 7). In control
experiments with phage D N A used over the same range of concentration and for the
same duration there was neither transformation of B. subtilis 168 thy-t to tryptophan
5
!
I
!
I
I
S
I
I
I
I
///y
o
o
2
,I
I
I
I
~
~.
~
Log. DNA (/~g.]ml.)
I
o
1
I
~
I
I
I
~
~
~
o
Log. DNA (#g./rnl.)
Fig. 7
Fig. 8
Fig. 7. Transforming activity of phage ~b3cDNA.
Fig. 8. Comparison of transforming activity of bacterial and phage DNA. Phage ~b3cDNA
( 0 - - 0 ) and DNA from Bacillus subtilis 36Io (O--O) was added to competent cultures of
B. subtilis i68 thy-1.
independence nor transfection and phage reproduction. Identical transformation activity was shown by D N A obtained from phage ~b3c and phage ~b3. However, at concentrations below saturation, phage D N A was more active in transforming than D N A
from wild type B. subtilis 361o; similar numbers of thymine-independent transformants being given by 25 to 5o times less phage D N A than bacterial. Since the bacterial
D N A used in these experiments had been subjected to more vigorous treatment during
extraction than the phage DNA it was probably more fragmented and so less active in
transformation (Litt et al. 1958). To find out whether this accounted for the greater
relative activity of the phage D N A similar comparisons were made with phage D N A
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 14:30:36
Phage-induced thymidylate synthetase
5oi
and bacterial DNA isolated by the gentler procedure of Berns & Thomas (I965).
Again, the phage DNA was at least ten times more active than the bacterial DNA in the
concentration range below o'5 #g. DNA/ml. (Fig. 8).
DISCUSSION
The association between phage 43 and its host Bacillus subtilis is considered to be of
the lysogenic type rather than a pseudo-lysogenic carrier state. The criteria for this
view include the failure of the infected cells to lose their phage-producing potential
after many single-colony isolations or after growth in phage ¢3 antiserum, but more
convincing evidence comes from the phage-induced acquisition by thymineless mutants
of B. subtilis of the power to synthesize DNA and grow without the addition of thymine to the medium. Thymine independence is a stable property of these bacteria and
is accompanied by the ability to produce phage even after many single colony isolations. A chromosomal location for prophage 43 has yet to be demonstrated but if
there is integration into the bacterial chromosome as there is for phage A (Campbell,
I962) then some similarity in the sequence of bases along the bacterial and phage
chromosomes would be expected. Preliminary experiments using the DNA-agar technique (Cowie & McCarthy, I963) have shown approximately 30 ~/ohomology, giving at
least qualitative evidence for similarity which might permit genetic interaction in spite
of the difference in the gross DNA base composition of the phage and bacterium.
Examination of extracts ofthymineless bacteria showed that they gained thymidylate
synthetase activity as a result of phage 43 infection and that this could account for
their subsequently being able to dispense with exogenous thymine. The vegetative
growth of phage 43 is accompanied by an enormous increase in the specific activity of
the enzyme so that by the time the cells start to lyse it is about ten times the value
normally found in Bacillus subtilis I68. On the other hand, the enzyme activity after
infection of B. subtilis I68 thy-r with the temperate phage ~b3 is regulated to about the
level found in the parent organism B. subtilis 168. The loss of specific activity of thymidylate synthetase following its initial increase after phage 43 infection is puzzling; it
could be due to direct inhibition of the enzyme (Haslam et al. I967) or to dilution of
enzyme, formed in an initially unregulated manner, by cell growth. Another possibility is that the phage grows vegetatively in a small number of infected cells and that
enzyme activity is lost when these cells lyse. While the synthesis of thymine accompanying phage 43 growth in thymine-requiring cells might be explained in a variety of
ways, the simplest hypothesis is that the phage contains the thymidylate synthetase
gene, especially since the enzyme activity of wild-type B. subtilis also increases after
infection. Support for this is provided by the experiments showing that DNA from
phages if3 and ¢3c can transform B. subtilis I68 thy-I to thymine independence. Since
the DNA content of each phage is less than that of the bacterium the phage DNA would
be expected to be more active in transformation than bacterial DNA. In practice, IO to
5o times as many transformants were obtained with phage DNA compared with the
same amount of bacterial DNA, but no great significance should be attached to the
absolute size of this difference because the methods of DNA extraction were not the
same and varying degrees of fragmentation of DNA could affect the efficiency of
transformation. Also, the efficiency of integration of a given gene seems to depend
upon the nature of genes in the vicinity of the particular marker (Schaeffer, I958), and
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 14:30:36
502
R.G. TUCKER
these are presumably different in the two cases. A few transformation experiments were
made with an independently isolated thymine-requiring strain, B. subtilis t68 thy-2.
Identical results were obtained showing that the phenomenon of transformation by
phage if3 DNA is not a peculiarity of B. subtilis I68 thy-I.
It has been assumed throughout that thymidylate synthetase is the only enzyme
responsible for forming thymine after phage infection. From genetic experiments,
Anagnostopoulos & Schneider-Champagne 0966) and Wilson, Farmer & Rothman
0966) concluded that there is another, as yet unknown, enzymic mechanism for synthesizing thymidylic acid. No attempt was made to determine whether phage 43 infection leads to the appearance of this second mechanism too. Further study of bacteria
infected with phage 43 may shed light on the way the two mechanisms for forming
thymidylic acid are interrelated, and the way their control is coordinated with that of
other enzymes forming DNA components.
Of the two likely methods of carriage of the thymidylate synthetase gene by phage
43, the balance of evidence favours phage conversion rather than specialized transduction. The phage as first isolated possessed the gene and all attempts to get' thymineless' mutants of the phage to regain it failed, though a very low frequency of gene
pick-up cannot at the moment be excluded. Defective phage particles certainly play no
part in the process, but it is doubtful whether a clear distinction can be made between
conversion and specialized transduction of the type reported by Matsushiro, Sato &
Kida (I964) for derivatives of coliphage ~8o, where the phage chromosome may acquire
host genes without replacement of its own genetic material. Also, Young, Fukazawa &
Hartman (I964) showed that phage P22 can lose its converting properties without
becoming defective, as if the gene responsible for conversion were not essential for
phage development. Such seems to be the situation with the 'thymineless' mutants of
phage 43. If the conclusion that phage ~3 is a converting phage is correct, the discovery
that phage DNA can transform thymine-requiring strains of B. subtilis has added
significance. Marmur, Seaman & Levine 0963) found that there were strict requirements for genetic homology in the B. subtilis transforming system, so the integration of
DNA from the phage could mean that a fragment of DNA including the gene for
thymidylate synthetase was derived from the chromosome of the wild-type bacterium
at some stage in the history of the phage. It may be that some other converting phages
have gained their characteristic properties in this way, and should be regarded as
extreme forms of specialized transducing phages (Luria, Adams & Ting, I96o; Matsushiro et al. t964). In the particular instance of phage ~b3 a comparison of the properties of the purified enzymes from wild-type B. subtilis and from thymineless bacteria
infected with phage ~3 would tell whether the thymidylate synthetase genes are as
closely related as the transformation experiments suggest.
I am grateful to Professor J. Mandelstam for his interest and helpful advice, and to
Mrs J. Newell for expert technical assistance. I also wish to thank Dr D. Kay for
electron microscopy, and Drs C. Anagnostopoulos, J.L. Farmer, B. Reilly and
W. R. Romig for kindly supplying bacteria and phages.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 14:30:36
Phage-induced thyrnidylate synthetase
503
REFERENCES
ADAMS, M. H. (I959). Bacteriophages. New York: Interscience.
ANACNOSTOVOOLOS,C. & SCn~aDrR-CrtAMPAO~rr, A.-M. (I 966). D6terminisme g6n6tique de l'exigence
en thymine chez certains mutants de Bacillus subtilis. C. r. hebd. S~anc. Acad. Sci., Paris 262,
I3II.
ANA~NOSTOVOULOS,C. & SV~Iz~N, J. (I96I). Requirements for transformation in Bacillus subtilis.
J. Bact. 8x, 741.
BAR~rER, H. D. & CoheN, S. S. (I959). Virus-induced acquisition of metabolic tunction. IV. Thymidylate synthetase in thymine-requiring Escherichia coli infected by T2 and T5 bacteriophages.
J. biol. Chem. 234, 2987.
BERNS, K. I. & TnOUAS, C. A. (I965). Isolation of high molecular weight D N A from ttemophilus
influenzae. J. molec. Biol. xx, 476.
BURTON, K. (I956). A study of the conditions and mechanism of the diphenylamine reaction for the
colorimetric estimation of deoxyribonucleic acid. Biochem. J. 62, 315.
CA~WBELL,A. M. (1962). Episomes. Adv. Genet. xx, IOI.
Cow~, D. B. & McCARThY, B. J. (I963). Homology between bacteriophage A D N A and E. coli DNA.
Proc. natn. Acad. Sci. U.S.A. 50, 537.
DAws, B. D. & MrNGIOLI, E. S. (195o). Mutants of Escheriehia coli requiring methionine or vitamin
BI~. J. Bact. 60, 17.
FAR,,mR, J. L. & ROVHMAN,F. (I965). Transformable thymine-requiring mutant of Bacillus subtilis.
J. Bact. 89, 262.
FLAICS, J. G. & Cormlq, S. S. (I959). Virus-induced acquisition of metabolic function. I. Enzymatic
formation of 5-hydroxymethyldeoxycytidylate. J. biol. Chem. 234, I5oI.
HASLAM, E . A . , ROSCOE, D. H. & TUCKER, R. G. (I967). Inhibition of thymidylate synthetase in
bacteriophage-infected Bacillus subtilis. Biochim. biophys. Acta x34, 312.
HAYES, W. (I957). The kinetics of the mating process in E. coil J. gen. MicrobioL x6, 97.
KAt-IA~q,F. M. (I963). Novel enzymes formed by Bacillus subitilis infected with bacteriophage. Fedn.
Proc. Fedn. Am. Socs exp. Biol. 22, 4o6.
Ln"r, M., MAR~,roR, J., EPHRUSSI-TAVLOR,H. & Doa~', P. (I958). The dependence of pneumocoecal
transformation on the molecular weight of deoxyribose nucleic acid. Proc. hath. Acad. Sci. U.S.A.
44, I44.
LowRY, O. H., ROSEBROUCU,N. J., FAP.X, A. L. & RANDALL,R. J. (I95I). Protein measurement with
the Folin reagent. J. biol. Chem. x93, 265.
Ltr~A, S. E., ADAMS, J. N. & T~G, R. C. 096o). Transduction of lactose-utilizing ability among
strains of E. coli and S. dysenteriae and the properties of the transducing phage particles. Virology
x2, 348.
LtrgiA, S. E., FRASER, D. K., ADAMS, J. N. & BtrgROUS, J. W. (I958). Lysogenization, transduction
and genetic recombination in bacteria. CoM Spring Harb. Symp. quant. Biol. z3, 7I.
Mm,rDELL, J . D . & HEgsrmY, A . D . (196o). A fractionating column for analysis of nucleic acid.
Analyt. Biochem. x, 66.
MARGOLIASH,E., NOVOGRODSKY,A. &; SCHEJTER,A. (1960). Irreversible reactions of 3-amino-1 : z : 4triazole and related inhibitors with the protein of catalase. Biochem. J. 74, 339.
MARMUR, J. & CORD~S, S. (I963). Studies on the complementary strains of bacteriophage DNA. In
Informational Macromolecules, p. 86, Ed. by H. J. Vogel, V. Bryson and J. O. Lampen. New
York: Academic Press Inc.
MAgMtm, J., SEAMAN,E. & LEvn~rE,J. (I963). Interspecific transformation in Bacillus. J. Bact. 85, 46L
MATSUSrImO,A., SATO, K. & KIDA, S. (I964). Characteristics of the transducing elements of bacteriophage ~8o. Virology 23, 299.
MITCr~LL, P. (I949). A new technique for stirred, aerated cultures. Nature, Lend. x64, 846.
ROBmNS, P. W., K~LL~R, I. M., WgtG~T, A. W. & Be~NST~rN, R. (I965). Enzymatic and kinetic studies
on the mechanism of O antigen conversion by bacteriophage e~. J. biol. Chem. 24o, 384.
RoscoE, D. H. & TUCKER, R. G. (I966). The biosynthesis of 5-hydroxymethyldeoxyuridylic acid in
bacteriophage-infected Bacillus subtilis. Virology 29, I57.
S c t a ~ r a ~ , P. (I958). Interspecific reactions in bacterial transformation. Syrup. Soc. exp. Biol. xz, 60.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 14:30:36
504
R.G. TUCKER
SNEATI-I,P. H. A. (I962). Sex factors as episomes. Brit. med. Bull. x8, 4I.
SPIZlZEN, J. (I958). Transformation of biochemicaUy deficient strains of Bacillus subtilis by deoxyribonucleate. Proc. natn. Acad. Sci. U.S.A. 44, IO72.
WAI-mA,A. J. & FRmDKIN,M. (1962). The enzymatic synthesis of thymidylate. I. Early steps in the
purification of thymidylate synthetase of Escherichia coli. d. biol. Chem. 237, 3794WEIGLE, J'., MESELSON,M. & PAIGEN,K. (I959). Density alterations in the bacteriophage lambda.
J. molec. Biol. I, 379.
WILSON, M. C., FARMER,J. L. & ROTI-IMAN,F. (1966). Thymidylate synthesis and aminopterin resistance in Bacillus subtilis. J. Bact. 92, 186.
WYATT,G. R. (I95I). The purine and pyrimidinecomposition of deoxypentose nucleic acids. Biochem.
J. 48, 584.
YOUNG, B. G., FUKAZAWA,Y. & HARTMAN,P. E. (I964). A P22 bacteriophage mutant defective in
antigen conversion. Virology 23, 279.
ZAMENHOF,S. • CI-IARGAFF,E. (195I). Separation of deoxypentose and pentose nucleic acids. Nature,
Lend. x68, 6o4.
(Received 23 S e p t e m b e r 1968)
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 14:30:36