J. gen. ViroL (I979), 43, 57-73
57
Printed in Great Britain
The Strategy of Infection as a
Criterion for Phylogenetic Relationships of Non-Coli Phages
Morphologically Similar to Phage T7
By K. H. K O R S T E N , C. T O M K I E W I C Z AND R. H A U S M A N N
Institut fiir Biologie III der Universitdt, Schiinzlestrasse I,
D-78 Freiburg, West Germany
(Accepted 26 September I978)
SUMMARY
Five phages which are morphologically similar to coliphage T7 but attack other
host bacteria have been compared to T7 and to its relative, T3, by the following
criteria: (a) cross-reactivity with antisera against T7 and T3, (b) DNA base sequence
homologies, as determined by the Cot technique, (c) synthesis of two phagecoded enzymes: RNA polymerase and SAMase, (d) patterns of phage-directed
protein synthesis, as determined by SDS-polyacrylamide gel electrophoresis followed
by autoradiography, (e) SDS-polyacrylamide gel electrophoresis of phage coat
subunits. As judged by all these criteria, Pseudomonas phage PX 3 is not related to
T7; thus, morphological similarity was attributed to convergent evolution. The
other phages, i.e. Serratia phage IV, Pseudomonas phage gh-I, Citrobacter phage
ViIII and Klebsiella phage No. I I, were considered to be related to T7 on the
basis of similarities in the patterns of phage-coded proteins and because, early
after infection, these phages induced, as T7 does, an RNA polymerase which
specifically transcribes the DNA of the homologous phage. Phages IV and No. t I
also induced the early synthesis of SAMase (previously only known to occur upon
T3 infection). With the exception of phage IV, however, DNA base sequence homologies with T 7 or T3 seem to be poor or non-existent. The tested phages, again with
the exception of phage IV, did not react with antiserum against T3 or T7.
It is concluded that a particular pattern of phage-directed protein synthesis (as
characterized by polyacrylamide gel electrophoresis and enzyme tests) may provide
evidence for phylogenetic relationships between phages, even in cases where other
criteria, such as genetic recombination, serological cross-reaction, and DNA base
sequence homologies, fail to indicate relatedness.
INTRODUCTION
Nothing is known about the phylogenetic origins of bacteriophages. However, during the
last three decades, analysis of a large number of phages has made it clear that newly isolated
phages can often be assigned to a given natural group of previously known ones, on the
basis of such characteristics as morphological similarity, serological relatedness, ability to
recombine genetically, or sequence homologies of the corresponding nucleic acids. Thus,
with regard to Escherichia coli phages with double-stranded DNA, we distinguish such
groups as the T-even-related, the ,~-related or lambdoids, the T5-related, and, of special
concern to us, the T7-related [or, as proposed by Summers (I972), the haptoid] phages.
oo22-I317/79/oooo-3412 $02.00 ~ I979 S G M
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 28 Jul 2017 15:11:58
58
K.H. KORSTEN, C. TOMKIEWICZ AND R. HAUSMANN
Each group of phages is characterized by a peculiar 'strategy of infection' (Subak-Sharpe,
I971 ), a term which encompasses the sum of co-ordinated genetic and biochemical events
which are elicited by the invading virus genome and which culminate in the liberation of
infective progeny particles.
Thus, the strategy of infection of T7-related phages could be summarized as follows.
Upon invasion of a host cell by the phage genome, the host RNA polymerase recognizes
promoters located near the conventional left end of the phage DNA and transcribes a
segment of about 20 % of that molecule. At that point the polymerase is thrown off the
template at a transcription termination signal. The segment of the phage genome transcribed
by the host RNA polymerase, i.e. the early region, encompasses only a few genes (five, in
the case of T7), among which gene 1 is of paramount importance. This gene codes for a
polypeptide, with a mol. wt. of about 1oooo0, which shows rifampicin-resistant RNA
polymerase activity. This phage-coded RNA polymerase recognizes only specific promoters located on the phage DNA, and absent on the host DNA (Chamberlin et al. 197o).
Since the expression of the early region of the phage genome also results in the inactivation
of the host RNA polymerase (for review see Hausmann, I976), it is thus assured that all
transcription is diverted towards expression of the late portion of the phage genome. The
crucial early feature of the T7-specific strategy of infection is thus the synthesis of a phagecoded, rifampicin-resistant RNA polymerase.
In this paper we will attempt to show that by characterizing a particular strategy of
infection one may be able to produce circumstancial evidence for or against a phylogenetic
relationship among two phages, even in cases where the classical criteria, namely morphological similarity, serological cross-reaction, genetic recombination and DNA base sequence
homologies, give inconclusive results. As a model system we chose a series of five bacteriophages which, based on morphological criteria, have to be assigned to the group C phages,
described by Bradley (i967). This group encompasses phages whose particles display a
polyhedral head and a tail which is much shorter than the head diam. The prototype of this
group is phage T7, whose particles have a head diam. of 6o nm and a tail length of about
I5 nm. The phages chosen for this comparative study were the following: Serratia phage IV;
Pseudomonas phage gh- 1 ; Pseudomonas phage PX 3; Citrobacter phage VilII; Klebsiella phage
No. I I (see Table I). The first two were included in this series because there was further
evidence for a phylogenetic relatedness with T 7; Serratia phage IV had been shown by Adams
& Wade (1954) to cross-react with antisera against T7 or against T3, which is a close relative
of T 7. And Pseudomonas phage gh-i encodes a specific RNA polymerase which transcribes
only the DNA of this phage (Towle et al. 1975). The other three phages were chosen solely
on the basis of their morphological similarity to T 7. The fact that T7 only grows on certain
strains of Escherichia coli and Shigella, while none of the other phages does, was an indication that their phylogenetic relationship to T 7, if any, would be remote in comparison to
the group of about a dozen coliphages which are related to T7 (Hausmann, 1976). The
absence of a known common host precluded mixed infection experiments for monitoring
genetic recombination among the tested phages. Thus, our comparative studies encompassed,
as classical criteria for relatedness, the investigation of serological properties and of
DNA-DNA hybridization patterns. On the other hand, information on the strategies of
infection was sought through enzyme tests for a phage-induced, rifampicin-resistant RNA
polymerase and for SAMase (an S-adenosylmethionine-cleaving enzyme characteristic for
phage T3), as well as by comparisons of the temporal expression of the banding patterns
of phage-coded proteins, as revealed by the method of sodium dodecyl sulphate (SDS)
polyacrylamide gel electrophoresis and autoradiography (Studier, 1973). Our results suggest
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 28 Jul 2017 15:11:58
Non-coli phages related to T 7
59
Table I. Strains of phages and corresponding host bacteria
Phage
Host
Source, references for description
IV
Escheriehia coli B
Escheriehia eoli B
Serratia mareescens4
gh-i
Pseudomonas putida A.3. t 2
PX3
ViIII
Pseudomonas aeruginosa z
Citrobacter spp. Ci23
No. i i
Klebsiella spp. 390 (03: KI I)
T7
T3
R. S. Edgar, Caltech; Hausmann & Gomez (1967)
R. S. Edgar, Caltech; Hausmann & Gomez 0967)
American Type Culture Collection (Cat. No. 11634-B and
1I634, respectively); Wassermann & Seligmann (2953);
Adams & Wade (t954)
J. A. Boezi, Michigan State University; Lee & Boezi (I966);
Towle et al. 0975)
R. H. Olsen, University of Michigan; Olsen et aL 0968)
S. Stirm, University of Giessen (W. Germany); Kwiatkowski
et al. (I975)
S. Stirm, University of Giessen (W. Germany); Bessler et aL
(I973); Rudolph et al. 0975)
that, by applying these methods routinely in the course of the characterization of new
phages, many more distant phylogenetic relationships between phages will be discovered
than has been possible up to now and thus throw more light on the problem of phage
evolution.
METHODS
Phage and bacteria. The phage and the corresponding host bacteria used in this work are
listed in Table I. For storage, bacterial cultures were grown in NB medium, mixed with an
equal vol. of glycerin and stored in liquid nitrogen. Phage stocks for our work were derived
from a single plaque isolated upon arrival of the requested samples. Lysates made from
these isolates were concentrated and purified as described by Y a m a m o t o et aL (197o). For
general phage work, standard procedures were used (Adams, 1959)- For plating Serratia
phage IV, not more than IOv host cells were poured on a plate, otherwise no plaques would
develop. Pseudomonas putida was grown at 32 °C, other bacteria at 37 °C. All infection
experiments were at 37 °C.
Media. NB medium contained, per litre, IOg nutrient broth (Difco) and to g NaC1; this
medium was used for growing all host bacteria except Pseudomonas aeruginosa, which was
grown in TGE-medium, containing, per litre, 5 g Bacto Tryptone (Difco), 2"5 g Yeast
Extract (Difco) and I g glucose. For incorporation studies with aSS-methionine, the cells
were grown in a synthetic medium (S-medium), which contained, per litre, IO g di-sodium
hydrogen phosphate, 4"5 g potassium di-hydrogen phosphate, 1 g ammonium sulphate,
o. 5 g sodium citrate, o.2 g magnesium chloride, 2 g glucose and 4o mg of each amino acid
required for protein synthesis, except methionine, which was not added.
Phage inactivation by antiserum. Rabbit antisera against T7 or T3 were obtained by
weekly subcutaneous injections, for 4 weeks, of lO11 purified phage particles each time; no
adjuvant was used. For assaying the activity of an antiserum it was diluted ioo-fold into a
phage suspension in S-medium, with a titre of lO~ particles/ml, and incubated at 37 °C for
io to 3o rain. Every 2 min a sample was withdrawn and titrated in order to determine the
K value of the original serum, according to the equation P t / P o = e -r×~/D (Adams, 1959).
Po and Pt represent, respectively, the phage titre at the times of serum action zero and t, in
rain; D is the serum dilution factor; and K is the activity constant for the inactivation
reaction (which closely followed first order kinetics until about 95 % of the homologous
phage were inactivated). K values obtained ranged from 200 to 4oo. For assaying hetero-
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 28 Jul 2017 15:11:58
60
K . H . K O R S T E N , C. T O M K I E W I C Z
AND R. HAUSMANN
logous activities, antisera were diluted into suspension of the various phages to be tested
(IO 7 particles/ml) and samples were withdrawn for titration, as described for the homologous
reaction. Heterologous K values (Khot) were then expressed as a fraction of the corresponding homologous values (Khom).
E n z y m e assays. Phage-induced, rifampicin-resistant RNA polymerase activities were
monitored as described by Chamberlin et al. (I97O) except that I~C-UTP (500 ct/min/nmol)
was used to label the RNA product. S-adenosylmethionine-cleaving activity (SAMase
activity) was monitored as described by Gefter et al. (~966). For both types of assays,
extracts of phage-infected cells were used as enzyme source (Beier & Hausmann, I974); for
RNA polymerase assays, 20/zg of cell-free extract were added to the assay mixture, IO #g
for SAMase assays.
D N A - D N A hybridization. In order to evaluate the relative degrees of genetic homology
between the DNA of T7 or T 3 and the DNAs of the other phages, the Cot technique of
Britten & Kohne (I968) was used. The Cot method is based on the fact that, under appropriate conditions, the extent of re-association observed in a given preparation of denatured
D N A depends on the product (Cot) of the DNA concentration (in tool of nucleotides per
litre) and the time of incubation (in seconds). The reassociation kinetics of a denatured
D N A may conveniently be expressed on a semi-logarithmic plot, where the fraction of
re-associated DNA fragments is given on an arithmetic scale on the ordinate and the product of DNA concentration and time of incubation (Cot) on a logarithmic scale on the
abscissa. This form of representation yields sigmoidal curves for the renaturation kinetics
and allows comparisons over a very wide range of renaturation rates. It also allows an easy
graphical determination o f the Cot values which correspond to 50% renatured DNA
fragments (C0t~ values). Based on the measurements of Cot½ values, our experimental procedure was similar in principle to that of Gelb et al. (I97I) for detecting SV4o sequences in
virus-transformed cells. First, the renaturation kinetics of a dilute solution (o'05 #g/ml) of
all-labelled T 7 DNA (or T3 DNA) was followed. In a second step, a Io- or 2o-fold excess
of non-radioactive T7 DNA (or T 3 DNA, respectively) was added to another sample of
the dilute solution of labelled T 7 DNA (or T 3 DNA) and again the renaturation kinetics
followed. It was thus shown, that the renaturation of the labelled D N A was speeded up :
the apparent Cot½values for the renaturation of labelled DNAs were lowered in proportion
to the excess of homologous unlabelled DNA added (Fig. 2 and 3). In a third step, unlabelled
phage DNAs with only partial sequence homologies were used in renaturation experiments.
The capacity of this heterologous DNA, relative to the capacity of homologous DNA, to
decrease the apparent Cot½ value of labelled T7 DNA (or T3 DNA) was considered to be a
function of the degree of genetic homology between the two DNA species (see also Discussion). In detail, the procedure was as follows: DNA of all phages was obtained by phenol
extraction of purified phage suspensions. 3H-thymidine-labelled T7 and T3 DNA were
prepared from phages grown in aH-thymidine-labelled Escherichia coli B cells resuspended
in non-radioactive NB medium. The host cells were labelled as described elsewhere (Issinger
& Hausmann, 1973). The DNAs were sheared to fragments with a tool. wt. of about 2 × io 5
(i.e. to less than i / i o o genome length) by means of ultrasonic treatment (Branson Sonifier,
microtip, intensity setting no. 5). The shearing was done in 2 ml portions o f a DNA solution
o f 50/zg/ml, in dilute saline citrate buffer (o-I % sodium chloride; o.o2 % sodium citrate).
The mol. wt. of the sheared samples were checked by zone sedimentation in neutral sucrose
gradients, according to Burgei & Hershey (I963). The samples of sheared phage D N A were
diluted in salmon sperm DNA (used as a carrier), sheared under identical conditions.
Appropriate mixtures of sheared DNA samples (see Results) were then denatured (too °C,
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 28 Jul 2017 15:11:58
Non-coli phages related to T7
6I
2o min) and incubated at 68 °C. At different times of incubation, samples (0.2 ml) were
placed on water jacketed hydroxyapatite columns at 6o °C (total vol. of column: 2 ml;
binding capacity about I"3 mg DNA) and washed with 3 ml of o'I4 M-phosphate buffer in
order to elute single stranded segments. A second washing, with 3 ml of o'4 M-phosphate
buffer, yielded a fraction with double-stranded fragments. The fractions were precipitated
with 5 % trichloroacetic acid and their radioactivity determined by scintillation counting.
The percentage value of renatured D N A corresponding to each incubation time was then
plotted against Cot [i.e. the product of D N A concentration (tool of nucleotides/1) and time
(s) allowed for renaturation at 68 °C].
Characterization of phage-coded proteins by SDS polyacrylamide slab gel electrophoresis
and autoradiography. The experimental layout for polyacrylamide slab gel electrophoresis
in the presence of SDS was similar to that described by Studier (I973). Host ceils were
grown in S medium to a concentration of 3 × IO8 cells/ml and were irradiated with u.v. light
in IO ml portions in uncovered Petri dishes (Sylvania germicidal lamp G8T5, I2 cm
distance for z min with continuous agitation). After 5 min of incubation at 37 °C, phage
(IO particles/cell) and 3~S-methionine (4o #Ci/ml, final concentration) was added to a 3 ml
sample of irradiated ceils, and incubation continued. At intervals, o.2 ml samples
were poured into tubes containing 25 #g of chloramphenicol and kept at o °C. The infected
cells were harvested by low speed centrifugation, resuspended in t5 #1 sample buffer
(o.I M-tris, pH 6.8; o.8 i-mercapoethanol, 2 % SDS, ~o% glycerin), and heated at IOO °C
for 2 min. Samples of IO #1 were applied to the gel pockets of an SDS polyacrylamide
gradient slab gel (IO to r8 % polyacrylamide), as described by Studier (r973), and subject to
electrophoresis under a constant voltage of 15o V for 3 h. The gels were stained, washed
and dried (Studier, I973). The dried gels were autoradiographed for 7 days, by use of
Osray T 4 X-ray film. For visualizing the bands of phage coat proteins (Fig. 8), purified
phage suspensions 0 o 12 particles/ml) were dialysed against sample buffer and heated at
IO0 °C for 2 min. A Io/zl portion was then subjected to gel electrophoresis as described.
The gel was kept in a staining solution (o'1% Coomassie brilliant blue R25o, 4o % methanol,
I o % acetic acid) for I2 h and placed in a destaining solution (5% methanol, 7"5 % acetic
acid) for rz h at 50 °C.
RESULTS
Serological cross-reactivity of anti-T7 and anti-T3 sera
Inactivation experiments with anti-T7 and anti-T3 sera were performed at two different
serum concentrations. In one series of experiments, Khor~values of 0"7 were chosen. This
relatively low serum activity allowed us to measure the inactivation rates which characterize the initially exponential inactivation kinetics. As shown in Fig. I, Serratia phage IV
was the only non-coli phage to react measurably with antisera against T7 or T 3. In these
experiments, the heterologous inactivation constant of phage IV with anti-T7 serum was
about o'o5 (Khet ~ O'O7 Kho~). Phage IV reacted with anti-T 3 serum with a K value of
about o'25 (Khot ~ o'36 Kho~), i.e. at a rate about five times higher than that corresponding
to T7 antiserum. The heterologous K values of anti-T3 serum with regard to T7 was only
about one third of the value corresponding to phage IV. Thus, from these experiments it
seems that Serratia phage IV is more closely related to T3 than T7 is, although phage IV
and T3 share no known common host. (Additional evidence for a close relationship between
phage IV and T3 will be discussed later.) The phages which were not inactivated at antiseum
concentrations o f K ~ o'7 were also shown not to react at a 2o-fold higher serum concentration. Under these conditions, heterologous reaction rates of less than one thousandth of
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 28 Jul 2017 15:11:58
62
K.H. KORSTEN, C. T O M K I E W I C Z AND R. HAUSMANN
l (a) I
100 ~
I
I
I
(b)
-
I
I
:~-~L--d~--- d~--0--
0.1~-
\
I
I
10
20
t
I
I
t
30
10
20
Time of serum action (mint
I
30
Fig. I. Action of anti-T7 (a) and anti-T3 serum (b) on heterologous phages. Diluted samples of
rabbit antiserum against T7 or T3 were mixed with phage suspensions ( ~ io 7 particles per ml) and
incubated at 37 °C. At timeintervalsthe reaction mixtures were assayed for surviving p.f.u. The serum
dilution factor was chosen to give a K value of about 0'7 when the serum was assayed with the
homologous phage. • - - • , T7; I1--111,T3; • - - A, Serratia phage IV; II-- - I , Pseudomonas phage
gh-1; G - - - C ) . Pseudomonas phage PX3; A - - - A , Citrobacter phage ViIII; C]----D, Klebsiella
phage No. It.
the homologous ones could easily have been measured. Thus, it can be stated that phages
gh-I, PX3, Villi and No. 1I are not serologically related to T3 and T7 by the criterion of
cross-reactivity with the tested sera.
Re-annealing kinetics of the DNAs of T 7 and T3 with the DNAs of
phages IV, gh-I, PX3, Villi and No. I I
As stated in more detail in the Methods, possible base sequence homologies between T7
or T3 and the other phages investigated here were sought by the Cot technique of Britten
& Kohne (I968). As seen in Fig. 2 and 3, the addition of a Io- or 2o-fold excess of nonradioactive, homologous DNA reduced the apparent Cot½ values, of the labelled DNAs of
T3 or T7 by the expected factor of zo or 20. On the other hand, the addition of a 2o-fold
excess of DNA from Citrobaeter phage ViIII or from Pseudomonas phage PX 3 did not
interfere with the re-association kinetics of labelled T3 or T7 DNA. Thus, it seems that no
base sequence homology exists between these phages and phage ViIII or phage PX3.
A 2o-fold excess of DNA of phage gh-l had no appreciable effect on the re-association
kinetics of 3H-DNA of T7 but reduced the apparent Cot½ of 3H-DNA of T 3 by about 5 % of
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 28 Jul 2017 15:11:58
63
Non-coli phages related to T7
I
I
I
I
I
I
l
i
9O
-...
-
-
,.
- ===2" . . . . . . . . .
-o.
50
~
\,.
\.
c).,
t
mx \ ' ~
~,
rl x
_
_
m
t0 L_
I
I
I
10-4
I
10-2
10-3
3H -
DNA Co t
Fig. 2. Reassociation kinetics of all-labelled T7 D N A in the presence of non-radioactive D N A s of
heterologous phages. Mixtures of sheared, 3H-labelled T7 D N A and sheared non-radioactive D N A s
(tool. wt. about z x io 5) of T7, or heterologous phages, were denatured at IOO °C, in the presence of
sheared salmon sperm carrier D N A and incubated at 68 °C. At various times of incubation, samples
were placed on hydroxyapatite columns at 6o °C. The fractions of single-stranded D N A were
eluted with o'z4 M-phosphate buffer; then, double-stranded D N A was eluted with o'4 M-phosphate
buffer. Cot represents the product of the time of incubation at 68 °C (in seconds) and the concentration of ZH-DNA of phage T7 (in mol of nucleotides per litre). Incubation mixtures contained,
per m h o ' o 5 # g aH-DNA of T7; 2"5 mg salmon sperm D N A ; one of the following additions:
© - - - © , no addition; I-I---17, I-o#g T7 D N A ; A - - - A , o ' 5 # g T7 D N A ; O - - O , o ' 5 # g T3
D N A ; D----D, z'o #g D N A of Klebsiella phage No. IZ; • - - • ,
o'5 #g D N A of Serratia phage
IV; m - - I , z'o/~g D N A of Pseudomonas phage gh-I ; A - - A , I / t g D N A of Citrobacter phage ViIII.
4------I,-
. . . . .
~
-L
I
I
I
I
b.
a
:~ so
\
I0
I
I
10 -4
I
I
10-3
[
"13.. \x..
I"--~--==~=
10-2
3 H - DNA Co t
Fig. 3. Reassociation kinetics of 3H-labelled T3 D N A in the presence of non-radioactive D N A s of
heterologous phages. The experimental layout was as described in the legend of Fig. z, except that
3H-labelled T3 D N A was substituted for 3H-labelled T7 DNA. Incubation mixtures contained, per
m h o'o25 #g 3H-DNA of T3; 2"5 mg salmon sperm D N A ; one of the following additions: ( 3 - - - ©
no addition; IS]----[2, o'5 #g T3 D N A ; A - - - ~ , 0"25 #g T3 D N A ; Q - - O, o'5/zg T7 D N A ; • - - • ,
o'5 #g D N A of Serratia phage IV; D--if] o'5 #g D N A of Klebsiella phage No. [ z ; IN--m, o'5 #g
D N A of Pseudomonas phage gh-I ; A - - / k , 0"5 #g D N A of Citrobaeter phage ViIII.
3
vza 43
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 28 Jul 2017 15:11:58
K. H. KORSTEN, C. T O M K I E W I C Z AND R. HAUSMANN
64
!
)
100
i
i
(a)
!
~,,,
i
i
i
f
)
)
(b)
a
50
,
E
l
-
g IO0
oe~
<
z
r
j)
,
-
I
I
) I
()
(d)
50
r
l
j
I
5
10
15
5
l0
15
Time after infection (min)
Fig. 4. Time course of synthesis of phage-directed, rifampicin-resistant RNA polymerase. Growing
host bacteria were infected with phage (Io particles per cell) and incubated at 37 °C. At different
times, samples of the infected cells were centrifuged. The pellets were resuspended in buffer and the
cells disrupted by ultrasound. After removing particulate material by centrifugation, the supernatant extracts were used as a source of enzyme in RNA polymerase tests in the presence of
rifampicin (2o #g/ml). DNA from the homologous phage was used as template DNA. (a) Eseherichia eoli B, infected with phage T7; (b) E. eoli B, infected with phage T3; (e) Serratia mareeseens 4,
infected with phage IV; (d) Pseudomonas putida A. 3. I 2, infected with phage gh- z ; (e) Citrobaeter sp.
Ci23, infected with phage ViIII; (f) Klebsiella spp. 39o, infected with phage No. xz.
5
10
15
the value expected on the assumption o f complete homology. D N A from Klebsiella phage
No. I I shifted the Cot½ values o f labelled T3 or T7 D N A s by about 8 % o f the values expected for fully homologous D N A . With Serratia phage IV D N A the apparent Cot½ o f
labelled T3 D N A was shifted by only 2 o % o f the value expected if the same a m o u n t o f
homologous D N A had been used. The addition of phage IV D N A had no appreciable
effect on the renaturation behaviour o f labelled T7 D N A . F o r comparison, in the case of
3 H - D N A o f T7 with excess o f T3 D N A (Fig. 2), as well as in the case o f 3 H - D N A o f T 3
with excess of T7 D N A (Fig. 3), we found that the shift in apparent Cot½ values of labelled
D N A s was about 3 o % o f the shift observed u p o n addition o f a corresponding excess o f
homologous D N A .
Rifampicin-resistant R N A polymerase activities after infection by phages IV,
gh-I, Villi, and No. II
As mentioned in the Introduction, a characteristic feature o f the strategy o f infection of
T7-related phages is the early synthesis of a phage-coded polypeptide (of tool. wt. a b o u t
1ooooo) which shows specific R N A polymerase activity depending on the presence o f the
D N A o f the h o m o l o g o u s phage as template. Thus, the search for such an R N A polymerase
activity was o f p a r a m o u n t importance for further characterizing the phages under investigation. F o r this, growing host cell cultures were infected with phage and further incubated at
37 °C. At various times after infection, samples o f these cultures were withdrawn in order
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 28 Jul 2017 15:11:58
Non-coli phages related to T7
100 ]
(a)
65
(br
(c)
~E
50
i
5
i
10
15
I
I
5
10
15
Time after infection (min)
I
5
I
10
I
15
Fig. 5. Time course of synthesis of phage-directed S-adenosylmethionine-cleaving enzyme. The
same cell-free extracts of phage-infected host cells, as described in Fig. 4, were used for assaying
SAMase activity. (a) Escherichia eoli B, infected with phage T3 ; (b) Serratia marcescens 4, infected
with phage IV; (e) Klebsiella spp. 39o, infected with phage No. I I.
to prepare cell-free extracts of the infected cells. These extracts were used as enzyme source
in RNA polymerase assays in the presence of rifampicin. The DNA used as template was
from the same phage which had been used to infect the host cells whose extracts were
assayed. The results (Fig. 4) m a d e it clear that, within 2 to 3 rain after infection, a sharp
rise in rifampicin-resistant RNA polymerase activity occurred in Pseudomonas putida cells
infected with gh-I, Serratia marcescens cells infected with phage IV, Citrobacter cells
infected with phage ViIII, and Klebsiella cells infected with phage No. I I. In all cases, the
kinetics of RNA polymerase synthesis was virtually identical to that of T7. No new RNA
polymerase activity was found to appear after infection of Pseudomonas aeruginosa with
PX 3, even when assays were made in the absence of rifampicin. In a second series of
experiments, heterologous phage DNAs were used as templates in RNA polymerase assays
with samples of the extracts which gave the highest activity in the homologous assays
(usually the 5 rain sample). None of the phage RNA polymerases tested showed any
affinity for the D N A of the heterologous phages, except T3 RNA polymerase, which transcribed T7 DNA, and T 7 RNA polymerase, which transcribed T3 DNA. In these two cases,
we confirmed the previously reported levels of heterologous activity of IO% and 5o%
respectively (Dunn et al. I97I ; Hausmann & Tomkiewicz, x976).
Phage-coded SAMase activity after infection by Serratia phage IV and
Klebsiella phage No. I I
Phage T3 codes for an S-adenosylmethionine (SAM)-cleaving enzyme, SAMase, the
product of gene 0"3. This phage gene is the first one to be expressed after infection (Hausmann & H~trle, ~97~) and acts to overcome the SAM-dependent restriction mechanism of
the host (Studier & Movva, ~976). SAMase synthesis is a specific feature of T3 infection
and had not been known to occur during the infection cycle of any other phage. Enzyme
assays done with extracts of phage-infected cells as putative enzyme sources revealed, however, that Serratia phage IV and Klebsiella phage No. I I upon infection induce SAMase
activity according to the same pattern observed during T3 infection, i.e. a peak of enzyme
activity was reached at only 3 to 4 rain after infection (Fig. 5)- The other phages under
investigation showed no detectable SAMase production.
3-2
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 28 Jul 2017 15:11:58
66
K . H . KORSTEN, C. T O M K I E W I C Z AND R. HAUSMANN
(a)
(b)
~
~16
~
mind ~
qmlmP 1
....
~
~
_
,m~
~
12; 15; 5
19; 8; 17
4
0"71.
--
--
~
w_~.
.... ~
lo
~, 6
5 .....
3.5
0.3
Uninf.
2
4
6
8
10
12 15 20
25
2
4
6
8
I0 12 15 20 25rnin
Fig. 6. Time course of synthesis of phage-coded proteins after infection of (a) Pseudomonasaeruginosa 2 by phage PX3, (b) Eseheriehia eoli B by phage T7. Host bacteria were u.v. irradiated in order
to inhibit host protein synthesis. Phage (lo particles per cell) and 35S-methionine were then added and
the cultures incubated at 37 °C. At the times indicated, samples were withdrawn. The cells were
disrupted and proteins separated electrophoretically. (I5o V, 3 h), in the presence of SDS, on a
Io to x8% polyacrylamide gradient slab gel. Dried gels were autoradiographed for selective
visualization of the bands of phage-coded proteins. Numbers at right indicate corresponding T7
genes. [Our T7 reference type carries a deletion in the o'7 gene region (Hausmann, 1976); therefore
no protein band shows at the corresponding position.] The origin is at the top. Uninf., uninfected,
irradiated Pseudomonas aeruginosa cells (a) were incubated for 25 min in order to evaluate residual
protein synthesis by the host. (Similar controls with other uninfected hosts showed no significant
residual protein synthesis.)
Time course of synthesis of intraeellular phage-eoded proteins, as revealed by
S D S polyacrylamide gel electrophoresis and autoradiography
Phage T7 shows a very characteristic pattern o f synthesis o f phage-coded proteins (Studier,
I973; Fig. 6b). Shortly after infection, five early proteins are synthesized in the following
order (for review, see Hausmann, i976): gene o'3 product (mol. wt. ~ Ioooo), gene o'7
product (tool. wt. ~ 4oooo), gene ir R N A polymerase (mol. wt. ~ IOOOO), a short peptide
coded by gene z.z, and a ligase, the product o f gene I ' 3 (tool. wt. ~ 4oooo). After the appearance o f these early proteins, the synthesis of the late proteins starts at about 5 min after
infection (37 °C). There are about 25 such late proteins which can be distinguished autoradiographically as individual bands after polyacrylamide gel electrophoresis o f appropriately labelled extracts o f infected cells (Studier, I973; Fig. 6b). A comparison o f the gel
electrophoretic-autoradiographic patterns of phage-coded proteins obtained after infection
o f our host strains by the corresponding phage revealed striking c o m m o n features with the
T7-specific pattern in the case o f Serratia phage IV, Pseudomonas phage gh-I, Citrobacter
phage ViIII, and Klebsiella phage No. I I (Fig. 7). All these phages directed the synthesis,
early in the course o f infection, o f a protein with a mol. wt. o f about 1ooooo. This protein
we tentatively consider as h o m o l o g o u s to the gene I product o f phage T 7. Similarly, the
early protein o f m o l , wt. in the range o f 4oooo, whose appearance precedes that o f the I o o o o o
tool. wt. protein would be h o m o l o g o u s to the gene o'7 product of phage T7. In the cases o f
Serratia phage IV and Klebsiella phage No. I I, the first phage-coded protein to appear had
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 28 Jul 2017 15:11:58
Non-coli phages related to T7
67
(b)
(a) '
.
...
,....,
.
.
.,.
.
,.,,
.
~
~
(16)
(16)
. . . ~
. . . . .
(1)
,..
,..
.,,.
..,,
..,.
(1)
~
~
~
~
(0.7)
(0.7)
~
-
2
~
:
4
~
:
6
~
:
8
-
~
~
10
~
12
~
!
....
.
~
6
8
10 12
(0-3)
~
15 20
25 min
2
4
15 20
25min
(d)
(c)
(16)
(16)
~
~
.......
.-,,,. ,,,..- . , , ~ - , - - ~
~
~ !
,,,~
W
(1)
...,
(0.7)
-,
,..,
,,-
~,,
,.,
m N ~ u m l m W
.,,,
(1)
(0.7)
; i
2
4
6
8
10 12
15
20
25rain
2
~
*,, - - ~
~
4
6
12
8
10
~ ~ ~0.3)
15 20
25rain
Fig. 7. Time course of synthesis of phage-coded proteins after infection of (a) Serratia mareeseens
by phage IV, (b) .Pseudomonas putida A.3.I2 by phage gh-I, (e) Citrobacter spp. Ci23 by phage
ViIII, (d) Klebsiella spp. 39o by phage No. I i. See legend to Fig. 6 for experimental procedures.
a mol. wt. around IOOOO;based on these findings, and considering that these two phages
were found to code for a SAMase, these proteins are here tentatively considered homologous
to the 0"3 gene products of T3 and T7. The topmost band (Fig. 7), corresponding to a late
protein of mol. wt. of about I5OOOO,was tentatively considered homologous to the product
of T 7 gene z6, a coat subunit (see also Fig. 8). Phage PX3, which showed no phage-induced
RNA polymerase activity, is also the only phage whose autoradiographic protein banding
pattern displayed no obvious similarities with that of T7 (Fig. 6a).
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 28 Jul 2017 15:11:58
68
K. H. K O R S T E N , C. T O M K I E W I C Z
A N D R. H A U S M A N N
--,~4
~
~'~
12
15
8
N
10
14
Fig. 8. Banding patterns of phage coat proteins. A sample (c. lo 12 particles) of a purified suspension
of particles of each phage was heated in the presence of SDS and then subjected to electrophoresis
0 5 o V, 3 h) in an SDS polyacrylamide gradient (Io to ~8 %) slab gel. The protein bands were
stained with Coomassie blue. Numbers at the right indicate correspondiag genes of T7.
SDS polyacrylamide gel-electrophoretic banding pattern of particle
coat proteins
The five phage types compared here to T3 and T7 were chosen because of their morphological similarity to T 7 under the electron microscope: they have hexagonal heads of about
60 nm in diam. and a stubby tail about I5 to 20 nm long. From these morphological similarities it does not necessarily follow, however, that the patterns of protein subunits composing the phage coats are also similar. Convergent evolution of phylogenetically totally
unrelated phage lines could have resulted in superficial morphological resemblance of coats
built up of totally different subunits. With this possibility in mind, we separated electrophoretically the coat protein subunits of the phages investigated here (on an SDS polyacrylamide slab gel). As shown in Fig. 8, each phage displayed an individual, unique
banding pattern of its coat protein subunits. Nevertheless, the overall similarity of pattern
of phages T7, T3, IV, gh-I, ViIII and No. Ix is in clear contrast to the quite different
pattern of PX 3 coat subunits.
DISCUSSION
We think that our findings clearly show that the close morphological similarity between
coliphage T 7 and Pseudomonas phage PX3 may be best interpreted as an example of convergent evolution and not as a result of a phylogenetic relationship between these two
phages. This conclusion is based on the lack of similarity in the subunit composition of the
phage coats and the banding pattern of phage-coded proteins, as well as the absence in
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 28 Jul 2017 15:11:58
Non-coli phages related to Y 7
69
PX 3 of the most striking feature of the T7-specific strategy of infection: the early synthesis
of a single peptide enzyme with a mol. wt. of about IOOOOO,endowed with an RNA polymerase activity restricted to the transcription of the DNA of the homologous phage. On the
other hand, this, in our opinion, most crucial criterion for phylogenetic relatedness of T7 is
fulfilled in the other four non-coli phages investigated here. However, except for the known
cross-reactivity in the case of T3 and T7 (Dunn et al. 1971; Hausmann & Tomkiewicz,
I976), we could detect no phage RNA polymerase activity in any heterologous combination
(RNA polymerase of one phage with template DNA from another). The common ancestor
of these phages which is proposed here obviously existed sufficiently long ago to allow
divergent evolution to erase all functional stereospecific complementarity between late
promoters on the phage DNAs and the non-homologous phage RNA polymerases. The
alternative explanation of a convergent evolution of unrelated RNA polymerases seems
extremely unlikely in view of present ideas on the evolution of protein superfamilies
(Dayhof, I976).
Lack of functional stereospecific complementarity was also patent in all serological
cross-reaction experiments, except in the case of Serratia phage IV, which reacted with a
relatively high Khet with anti-T3 serum (Kh~ ~ o'36 K~o,~)and with lesser intensity with
anti-T7 serum (K~ot ~ o'o7 Khom).This suggests a relatively close relationship of this phage
to T 3. This impression is corroborated by the fact that phage IV, like T3, codes for an early
protein with SAMase activity and that denatured phage IV DNA formed heteroduplexes
with denatured phage T 3 DNA at a relatively high rate, as discussed in the next paragraph,
in which the reannealing patterns with heterologous DNAs are considered.
In order to evaluate the relative degree of phylogenetic divergence between the DNAs of
two related phages we have measured the rates of reannealing of the heterologous DNAs
(leading to heteroduplex formation) and compared these rates to the rate of reannealing of
DNA from T3 or from T7 (homologous DNA) treated in the same conditions. For this
purpose we have the Cot technique of Britten & Kohne (1968), as described in the Methods.
Until recently evaluations of base sequence homologies between different DNAs were done
mainly by determining saturation levels upon hybridization with corresponding RNA or by
measuring the melting temperature (Tin) of corresponding DNA heteroduplexes. However,
saturation levels do not allow the detection of base substitutions and small deletions or
additions distributed randomly through the genome. But this is exactly the situation found
if one compares T 3 with T7, as has been shown in analyses of phage-coded proteins separated by SDS-polyacrylamide gel-electrophoresis (Hyman et al. 1974), and by electron
microscopy of T3 DNA-T7 DNA heteroduplexes (Davis & Hyman, 1971). Therefore,
measurements of saturation levels were not considered appropriate for our purpose. However, one could argue that, instead of determining the relative rates of heteroduplex formation, it would be preferable to measure the differences between Tm values of homoduplexes
and heteroduplexes (ATm), since the work of other investigators would then be available for
comparison. If data on ATm values gave precise information on the degree of base sequence
divergence between two DNAs, such a comparison would indeed be of great value. One
first problem in evaluating base sequence homologies byATm determinations is that genome
regions of low or no homology will not undergo heteroduplex formation in the first place.
In such cases, the average value of sequence homology will be systematically overestimated.
In addition to this, as was recently pointed out by Britten & Davidson 0976): 'The melting
temperature reduction is a good measure of the sequence divergence, but the conversion
factor is not accurately known. For this discussion we assume that a reduction of 1 °C in
Tm corresponds to about 1% nucleotide substitution in the two ancestral species lines since
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 28 Jul 2017 15:11:58
70
K.H. KORSTEN, C. TOMKIEWICZ AND R. HAUSMANN
their divergence from a common ancestor.' Their assumption was based on studies with
synthetic polynucleotides in which only single base substitutions were present [and which
gave results varying by a factor of more than 3 (King & Wilson, I975)]; the precise effects
of random small deletions or insertions have never been subjected to a rigorous study.
Since, as mentioned above, many such small deletions (or insertions) have been detected
by comparing T3 with T7, and since most spontaneous mutations seem not to be simple
base substitutions (e.g. Drake, I97o), it seems to us a fair statement that measurements of
ATm would not yield more precise information on the actual extent of base sequence
homologies between the investigated phages than measurements of the rates of heteroduplex
formation, especially if an appropriate comparison with data obtained by still another
technique is possible. In the case of T 3 and T7, our data can indeed be compared with those
of the electron microscopic study of partially annealed T3-T 7 heteroduplexes, by Davis &
Hyman 0971). These authors examined the extent and position of duplex regions formed
along T3-T7 hybrid D N A molecules annealed under conditions of different stringency
brought about by varying formamide concentrations. They found continuously varying
degrees of homology (ranging from less than 40 ~o to over 9o %) along the two compared
genomes. From their data we have calculated a weighted average of about 6o ~o genetic
homology between the DNAs of T 3 and T7. On the other hand, in the present study we
found that an excess of unlabelled T3 DNA shifted the apparent Cot value of labelled T 7
DNA by 3o ~o of the value found by the use of the same amount of excess homologous D N A
(Fig. 2) and the corresponding value for the reciprocal experiment (excess non-radioactive
T7 DNA plus labelled T3 DNA) was 32 %. Thus, in our experimental conditions, in the
case of the heterologous system of T3 and T7, a shift in apparent Cot½ value of about 30 ~'o
of the shift in the homologous systems would correspond to about 60 % of base sequence
homology, as defined by Davis & Hyman (I97I). We would like to emphasize, however,
that these authors based their calculations of sequence homologies on the uncertain correspondence factor between ATm and percentage of base substitutions. And, as stated
above, this uncertainty is enhanced by the fact that a sizable but unknown fraction of
mutational events which accumulated in T3 and T 7 since their divergence is not due to
single base substitutions. However, in spite of all these uncertainties, we may state, with
regard to the non-coli phage DNAs tested, that none had the ability to shift the apparent
Cot½ values of either T3 or T 7 D N A to the same extent as the DNAs of these two phages in
heterologous combination. Thus, by this criterium, none of these phages is as closely
related to T 3 or T7 as these phages are to themselves; in some cases (see Fig. 2 and 3),
where no effect on apparent Cot½values of T 3 DNA and T 7 D N A could be measured by
adding heterologous DNA, base sequence homologies, if present, seem to be relatively
small.
In addition to that, we think that our data on the rates of heteroduplex formation allow
some tentative inferences on the nature of the genetic heterologies which characterize the
phages we compared to T3 and T7. In our study two extreme kinds of sequence divergence
can be anticipated: 0 ) base substitutions, small deletions and insertions distributed
randomly through large sections of the genomes [the results of Davis & Hyman 0 9 7 0 and
of Hyman et al. 0974) suggest this to be the case for T3 and TT]; (2) major changes such
as block substitutions or deletions (large relative to the average fragment size, which in
our case was about o.oI of the genome size), leading to no relatedness. The effects of changes
of type I on the rate of reannealing are not known precisely (a statement which also applies
to their effects on ATm), but, nevertheless, one may predict that in experiments such as
these discussed here, they will lead to a parallel shift of the C0t-curves of labelled hetero-
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 28 Jul 2017 15:11:58
~
Non-coli phages related to T7
0-9
e-
q..
0-5
z
7I
d
0-1
I
I
0-01
0-I
l
1
l0
100
Cot
Fig. 9. Computer-simulated time courses of annealing reactions of a labelled D N A preparation in
the presence of an excess of unlabelled DNA. The two DNAs share some regions of total base
sequence homology and some of total non-homology in the following proportions: (a) 75 % homology and 25 % non-homology; (b) 5o% and 5o%, respectively; (e) 15 % and 75 %, respectively; (d)
I OO% non-homology; (e) I OO% homology. Homologous and heterologous regions are very large
relative to the average fragment size of the D N A preparations. The equations for the main graph
are as follows:
C
I
( d ) C--~= I +KC0t;
i.e. an ideal second order reaction where Co, the initial concentration of homologous DNA, and the
constant K, are considered equal to unit; C is the fraction of labelled D N A remaining single
stranded at the time t;
(a) C -
0"75
0"25
I+IOt +~;I+t
(c) C ----- o'2_____~5+ 0"75.
~+~ot
x+t'
0"5
(b) C =
~+Iot
(e) C =
I
r+~ot"
0"5
+ - -I ;+ t
As can be seen from these equations, the excess of unlabelled DNA was io-fold, i.e. corresponded
to the conditions of most of our experiments. In the inset, unlabelled D N A was considered to be
present in Ioo-fold excess in otherwise identical conditions.
logous DNA. In contrast, the effects of changes of type 2 are not immediately predictable. At
first glance one might think that biphasic reannealing curves will result, in such a way that
a plateau will be found at the percentage level of single strands corresponding to the
fraction of the tested D N A which has no counterpart in the related D N A added in excess
(this fraction would anneal as if no related D N A was present). In order to examine this
question more precisely, we performed computer simulations of reannealing experiments
where one D N A probe was in Io-fold or 2o-fold excess over the other, i.e. the conditions in
which our experiments were done. We tested three different assumptions: (a) both D N A s
share a 75 % segment of full homology while 25 % of the genome have no homology at all;
(b) as in (a) but the corresponding proportions are 5o% and 5o % ; (c) as in (a) but the
corresponding proportions are z 5 % and 75 %- In no case did a plateau develop (a small
tendency to plateau formation could only be observed in simulation experiments with over
a Ioo-fold excess of one D N A species). However, the slopes of the simulated C0t-curves were
not shifted in a parallel way, but the shifts were more accentuated in the upper part; this
resulted in what we would like to call a diagonal shift (Fig. 9). Since in our experiments all
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 28 Jul 2017 15:11:58
72
K . H . K O R S T E N , C. T O M K I E W I C Z
A N D R. H A U S M A N N
shifts were clearly parallel, we favour the idea that the genetic divergence among the investigated phages is mainly of type I (i.e. consists mainly of base substitutions and small
deletions or insertions). In extreme cases such differences would prevent the detection of
any base sequence homologies by means of reanneating experiments. An example of this
situation might be Citrobacter phage ViIII, on the one hand, and T 3 or T7 on the other. In
this case, we have based our hypothesis of phylogenetic relatedness to T7 solely on the
apparent similarities in the strategies of infection of these phages, as revealed by the presence
of a phage-specific RNA polymerase and the gel electrophoretic patterns of phage-coded
proteins.
In order to characterize further the group of T7-related phages, many other phage types
will have to be investigated. One interesting question, for instance, refers to the possible
phylogenic relatedness between T 7 and such morphologically similar structures as bluegreen algae virus LPP-I (Adolph & Haselkorn, I972). In addition, in order to allow comparative genetics with T7, hosts for amber mutants of some of these phages should be
searched for. Such studies will substantially contribute to the long range goal of establishing
a detailed pedigree of what seems to constitute the large and diverse group of T7-1ike phages.
As a result of this, the problems of phage systematics and taxonomy (Bradley, T967;
Wildy, 197x ; Fenner, 1976) might also be more easily resolved.
We would like to thank Monika Messerschmidt for efficient technical assistance. We are
also grateful to Professor H.-Ch. Spatz for programming the computer. This work was
supported by the Deutsche Forschungsgemeinschaft (grant Ha 7o9).
REFERENCES
ADAMS, M. H. (1959). In Bacteriophages. New York: Interscience.
ADAMS, M. H. & WADE, E. 0954). Classification of bacterial viruses: the relationship of two serratia phages
to coil-dysentery phages T3, T7 a n d D44. Journal of Bacteriology 68, 320-325.
ADOLPH', K. W. & HASELKORN, R. (I972). C o m p a r i s o n of the structures of blue-green algal viruses L P P - I H
a n d LPP-2 a n d b a c t e r i o p h a g e T7. Virology 47, 7 o i - 7 I o .
BEIER, H'. & H'AUSMANN,R. (1974). T3 x T7 phage crosses leading to recombinant R N A polymerases. Nature,
London 25I, 538-54o.
BESSLER, W., FREUND-M()LBERT, E., KNUFERMANN, H., RUDOLPH, C., THUROW, H. & STIRM, S. 0973). A bacteriophage-indueed depolymerase active on Klebsiella K I I capsular polysaccharide. Virology 56, 134-I5 I.
BRADLEY, D. E. (I967). Ultrastructure of bacteriophages and bacteriocins. Bacteriological Reviews 3x, 23o314.
BRITTEN, R. J. & DAVIDSON, E. H. 0 9 7 6 ) . D N A sequence arrangement and preliminary evidence on its evolution. Federation Proceedings 35, 2151-2157.
BRITTEN, R. J. & KOHNE, D. E. (1968). Repeated sequences in D N A . Science, New York x6x, 529-54o.
aURGI, E. & HERSHEY, g . D . (I963). Sedimentation rates as a m e a s u r e o f molecular weight of D N A . Biophysical Journal 3, 3o9-32 I.
CHAMBERLIN, M., McGRATH, J. & WASKELL, L. 097O). N e w R N A polymerase from Escherichia cull infected
with bacteriophage T7. Nature, London 228, 227-231.
DAYHOFF, M. O. 0976). T h e origin a n d evolution o f protein superfamilies. Federation Proceedings 35, 21322138.
DAVIS, R. W. & HYMAN, R.W. (I971) A s t u d y in evolution: the D N A b a s e s e q u e n c e h o m o l o g y between
coliphages T7 a n d T3. Journal of Molecular Biology 6~, 287-3oi.
DRAKE, J. W. (I970). In The Molecular Basis of Mutation, pp. 178. San Franeisco. H o l d e n - D a y .
DUNN, J. J., BAUTZ, E. A. & BAUTZ, E. K. F. (I971). Different template specificities of p h a g e T3 a n d T7 R N A
polymerases. Nature New Biology 23o, 84-96.
FENNER, F. (1976). T h e classification a n d n o m e n c l a t u r e o f viruses. S u m m a r y o f results of meetings of the
international committee on t a x o n o m y of viruses in M a d r i d , September I975. Virology 7x, 371-378.
GEETER, M., HAUSMANN, R., GOLD, M. & HLtRWITZ, J. (1966). T h e e n z y m a t i c m e t h y l a t i o n of ribonucleic acid
and deoxyribonucleic acid. X. Bacteriophage T3-induced S - a d e n o s y l m e t h i o n i n e cleavage. Journal of
Biological Chemistry 24x, I995-2oo5.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 28 Jul 2017 15:11:58
Non-coli phages related to T7
73
GELB, L. D., KOHNE,D. E. & MARTIN, M. A. (I97I). Quantitation of simian virus 4o sequences in African green
monkey, mouse and virus-transformed cell genomes. Journal of Molecular Biology 57, 129-I45.
~AUSMANN, R. (1976). Bacteriophage T7 genetics. Current Topics in Microbiology and Immunology 75,
77-11o,
HAUSMANN,R. & GOMEZ, B. 0967)- Amber mutants of bacteriophages T3 and T7 defective in phage-directed
deoxyribonucleic acid synthesis. Journal of Virology x, 779-792.
nAUSMANN, R. & H~RLE, E. (I970. Expression of the genomes of the related bacteriophages T3 and T7. In
Proceedings of the First European Biophysics Congress, vol. L PP. 467-488. Edited by E. ]3roda, A. Locker
& H. Springer-Lederer. Vienna: Wiener medizinische Akademie.
HAUSMANN,R. & TOMKIEWlCZ, C. (I976). Genetic analysis of template specificity of RNA polymerases (gene
1 products) coded by phage T3 × T7 recombinants within gene 1. In RNA Polymerase, pp. 731-743.
Edited by R. Losick & M. Chamberlin. New York: Cold Spring Harbor Laboratory.
HYMAN, R. W., BRIJNOVSKIS, I. & SUMMERS,W. C. (I974). A biochemical comparison of the related bacteriophages T7, OI, OII, W3I, H and T3. Virology 57, 189-2o6.
ISStN~ER, O. 6. & HAUSMANN,R. (I973). Synthesis of bacteriophage-coded gene products during infection of
Escherichia coli with amber mutants of T3 and T7 defective in gene 1. Journal of Virology xx, 465-472.
KIN~, M. C. & WILSON, A. C. 0975). Evolution at two levels in humans and chimpanzees. Science, New York
x88, IO7-116.
KWlATKOWSKI, B., BEIL~ARZ,H. & s'rmM, s. (I975). Disruption of Vi bacteriophage III and localization of its
deacetylase activity. Journal of General Virology ~-9, 267-280.
LEE, L. V. & ~OEZI, J. A. (1966). Characterization of bacteriophage gh-1 for Pseudomonas putida. Journal of
Bacteriology 92, 1821-1827.
OLSEN, R. It., METCALF, E. S. & TODD, J. K. (1968). Characteristics of bacteriophages attacking psychrophilic
and mesophilic pseudomonads. Journal of Virology 2, 357-364.
RUDOLPh, C., EREUND-M/)LBERT,E. & STIRM, S. (I975). Fragments of Klebsiella bacteriophage No. 11. Virology
64, 236-246.
STUDIER, F. W. (1973)- Analysis of bacteriophage T7 early RNAs and proteins on slab gels. Journal of
Molecular Biology 79, 237-248.
STUDIER, E. W. & MOVVA,Y. R. 0976). The SAMase gene of bacteriophage T3 is responsible for overcoming
host restriction. Journal of Virology x9, I36-145.
SUBAK-SHARPE,J. It. (197I). In Strategy of the Viral Genome, pp. 383-394. Edited by G. E. W. Wolstenholme
& M. O'Connor. A Ciba Foundation Symposium. London: Churchill Livingstone.
SUMMERS,W. C. (1972). Regulation of RNA metabolism of T7 and related phages. Annual Review of Genetics
6, I9i-2o2.
TOWt~E, H. ¢., JOLLY, J. E. & BOEZ~, J. A. (I975). Purification and characterization of bacteriophage gh-iinduced deoxyribonucleic acid-dependent ribonucleic acid polymerase from Pseudomonas putida. The
Journal of Biological Chemistry 250, I723-I733.
WASSERMANN,M. M. & SELIGMANN,E. 0953). Serratia marcescens bacteriophages. Journal of Bacteriology 66,
1 I9-12o.
WILDY, l'. (I971). Classification and Nomenclature of Viruses. First Report of the International Committee
on Nomenclature of Viruses. In Monographs in Virology, No. 5, Basel: Karger.
YAMAMOTO, K. R., ALBERTS, B. M., BENZINGER, R., LAWHORNE, L. & TREmER, G. (1970). Rapid bacteriophage
sedimentation in the presence of polyethylene glycol and its application to large-scale virus purification.
Virology 4 o, 734-744.
(Received 4 July I978)
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 28 Jul 2017 15:11:58