J. gen. ViroL (I978), 38, 263-272
263
Printed in Great Britain
A Generalized Transducing Salmonella Phage ES18 Can Recombine
with a Serologically Unrelated Phage Fels 1
By N O B U T O
YAMAMOTO
Fels Research Institute and Department of Microbiology,
Temple University School of Medicine, Philadelphia, Pennsylvania I914o, U.S.A.
(Accepted I5 August I977)
SUMMARY
A long-tailed and generalized transducing Salmonella phage ES 18 can recombine
with a serologically unrelated phage Feb ~. When ESI8 phage stocks grown on
Feb ~ lysogens were plated on Feb • lysogens resistant to ESI8, plaques of a new
hybrid phage PI8I were found at a frequency of about IO-s. Serological analysis of
phage PI8I showed that it carries the protein coat of Fels z and has no serological cross-reactivity with Es~g. Phage PIgI expresses the clear plaque
morphology similar to that of ESI8. ESI8 was also found to carry a homology
with P22 in at least the c regions. Thus, the c markers ( c + , ct, c2 and c3) of P22
can be transferred to Esrg. When these ESI8 strains were used for isolation of
P18I phage, the clear plaque morphology of PI8I mimicked that of the P22c
marker in the ESI8 strain which gave rise to the PI8I strain. Salmonella typhimurium strain QI, lysogenic for PI8r, is immune to ESI8 but not to Fels f. Furthermore, ESI8 lysogens are immune to PI8I. Therefore it is concluded that the hybrid
phage PI8I conserves the protein coat of Fels I and carries at least the c regions
of ESI8.
When ESI8 was grown on Fels z lysogens of recombination-deficient (reeA)
mutants, PI8I hybrid particles were found in the ESI8 stocks at a frequency of
about i o-11. This frequency is about IOOO-fold lower than that found in ES 18 stocks
grown on the wild type hosts lysogenic for Fels z. Therefore, it is concluded that the
bacterial recombination mechanism plays an important role in the formation of
PI81 by recombination between the prophage Fels r and the superinfecting phage
ESI8.
Although phage ESI8 belongs to B group phages, it possesses a generalized
transducing capability. We have studied a possible origin of ESI8.
INTRODUCTION
We previously found that the generalized transducing Salmonella phage P22 recombines
with the prophages Feb I and Fels 2 of Salmonella typhimurium LT-2 (abbreviated St) to
yield hybrid phage P221 at a frequency of about 1o -l° (Yamamoto, 1967) and hybrid phage
F22 at a frequency of about IO-11 (Yamamoto, I969a), respectively. P22 can also recombine
with an E. coli phage A (Gemski, Baron & Yamamoto, i972). Several other generalized
transducing phages (Salmonella typing phages ~IO, 22 and 23 obtained from Dr E. S.
Anderson and phage 427) which are antigenically indistinguishable from P22 are also able
to recombine with the prophage Fels I, resulting in formation of hybrid phages similar to
Downloaded from www.microbiologyresearch.org by
IP: 78.47.19.138
On: Sat, 01 Oct 2016 03:40:37
264
N. YAMAMOTO
Table I. Host range and antigenicity of ESI8 and its relatives
Phage* (source)
Isolated from (source)
ES I 8
Typing phage ~I8
(Dr B. A. D. Stocker)
ESI8hl
(Dr B. A. D. Stocker)
Antigenicity
by
Host range +
neutraliza~ - ~. . . . .
~ tion tests
(phage
Q Q/ee Q/22/18 type)
+
+
+
+
-
-
-
-
ESI8
ESI8
ESI9
Typing phage :~I9 stockt
(Dr E. S. Anderson)
+
+
--
ESI8
ESI9A
Typing phage ~19 stockt
(Dr E. S. Anderson)
+
-
-
P22
q~I9
Typing standard bacterial strain no. I9
(Dr E. S. Anderson)
+
+
-
ESI8
~519A
Typing standard bacterial strain no. I9
(Dr E. S. Anderson)
+
-
-
P22
* ESI9 and ESI9A were isolated from an old stock of the typing phage :~I9 which was grown on the typing
standard bacterial strain no. I9. ~ 9 and ~I9A are temperate phages which we isolated from the typing
standard bacterial strain no. I9.
t An old phage stock (prepared more than I8 years ago by B. R. Callow) of typing phage gI9.
z~ + , Indicates sensitive and plaque formation; - , indicates resistant and no plaques.
P22I phage. However, other Salmonella phages which are not capable of generalized
transduction, did not produce P22 I-type phage (Yamamoto, 1969 b). Thus, our accumulated
observations led us to postulate that although it is not known whether generalized transduction is an absolute prerequisite for recombination between unrelated bacteriophages, at
least in the case where one partner is a generalized transducing phage, this type of recombination accelerates the evolution of bacteriophage.
The above transducing phages are all A group phages by Boyd's classification (Boyd &
Bidwell, I957). However, Kuo & Stocker 097o) found that a long-tailed B group Salmonella
phage ESI8 is a generalized transducing phage and is serologicaUy and morphologically
unrelated to P22. Our preliminary study showed that ESI8 is also able to recombine with
Fels z to yield a new hybrid phage species designated PI8I (Yamamoto, 197o). In this
communication we report characterization of the recombination system and discuss a general
mechanism of recombination between unrelated phages.
METHODS
Bacteria. Salmonella typhimurium strain LT-2 (abbreviated St) which is known to carry
prophages Fels I and Fels 2, S. typhimurium strain Qz (Boyd & Bidwell, I957), and their
following mutants were used. SDI 4 (Spicer & Datta, 1959), kindly supplied by Dr B. A. D.
Stocker, Stanford University, is a derivative of St cured of phage Fels 2. We found that
although this cured strain is free of prophage Feb 2, it still carries prophage Fels ~. P22resistant mutants used were St/22 and Q/22. ESI8-resistant mutants of these strains,
St/22/x8 and Q/22/x8 were also isolated. Moreover, St recA mutants (recA leu I97 and
recA his ~t22-0 (Yamamoto, I969b) and Qz recA35 were used. Dr E. S. Anderson, Central
Enteric Reference Laboratory, London, kindly sent us Salmonella typhimurium typing
bacterial standard strain no. I9 (Callow, I959).
Downloaded from www.microbiologyresearch.org by
IP: 78.47.19.138
On: Sat, 01 Oct 2016 03:40:37
Hybrid between serologically unrelated phages
265
Phages. A long-tailed and generalized transducing Salmonella phage ESI 8 and its host
range mutant ESI8hI were kindly supplied by Dr B. A. D. Stocker. These ESI8 phages and
their relatives are listed in Table I. The temperate phages Fels z and Fels ~ of St were
described previously (Yamamoto, I967). ESI8 is morphologically and serologically distinguishable from P22 (Kuo & Stocker, I97o). ESI8 forms clear medium-sized plaques
(about i.o mm in diameter) with slightly turbid centres on SDz 4, SDz4[e2, Qz and Q[22
whereas it forms tiny plaques on St and St[22 with low plating efficiencies fluctuating
between 1o-a and 1o-5. This partial resistance of St and St/ee to ESI8 is conferred by the
Fels e prophage (Kuo & Stocker, I97O). The plating efficiencies of ES18 on Fels e lysogens
are similar to those of P22 mutants sensitive to restriction by the Fels e prophage (Fukuda
& Yamamoto, i972 a). This restriction sensitive (Rs) gene of P22 is located between the erf
and c3 genes of P22 (Fukuda & Yamamoto, 197zb). ESI8hI forms large clear plaques
(about I "5 mm in diameter) on Fels e lysogens, St, St/e2 and Q(Fels 2), as well as non-Fels 2
lysogens, SDI4, SDz4/e2 , Qz and Qee. In addition, an old stock of typing phage :~i 9
previously grown on its propagating host, the typing bacterial standard strain no. I9
(Callow, I959) , was supplied about I8 years ago by Dr E. S. Anderson of Central Enteric
Reference Laboratory, Central Public Health Laboratory, Colindale, London, N.W. 9.
Antiserum. Anti-Pzz and anti-P22i sera were previously prepared by Yamamoto &
Anderson (~96 0 and anti-Fels z serum was previously prepared by Yamamoto (r967).
Since anti-Paz I neutralizes Fels z as well as P22 I, and Fels z is antigenically indistinguishable
from P221 (Yamamoto, 1967), anti-Pz21 was used instead ofanti-Fels z for some experiments.
Anti-ES 18 was kindly supplied by Dr B. A. D. Stocker.
Media. Nutrient broth consisting of 8 g Difco nutrient broth and 5 g sodium chloride
per litre of distilled water was used for making phage lysate and bacterial aeration culture.
For phage plating, an agar base containing 23 g Difco nutrient agar and 5 g sodium chloride
per litre, and soft nutrient overlay agar containing 7"5 g Difco bacto-agar, 5 g sodium
chloride and 8 g Difco nutrient broth per litre were used. Phosphate-buffered saline contained M/t 5 phosphate in o'I M-NaC1 at pH 7"o and was used for dilution a n d assay of
phages.
Preparation of ESz8 stocks. Confluent lysis plates were prepared by plating about
4 × Io ~ or more p.f.u, of ES18 particles on various derivatives of St and Qz on freshly
prepared nutrient agar plates. After overnight incubation at 37 °C the soft agar of the confluent lysis plate was scraped, suspended in 2"5 ml of the buffered saline and incubated for
I h at room temperature. This was centrifuged for I5 min at 3ooo rev/min. The supernatant
contained at least Io 1° p.f.u. ESI8 particles.
Classification of phages by heat resistance test. Boyd 095 O) divided Salmonella typhimurium phages into two groups A and B by heat stability. The phages of group A show
a degree of resistance to heat which considerably exceeds that of group B phages. We
defined newly isolated phages by the criterion that exposure of phages in nutrient broth for
5 min at 75 °C causes more than 99 % loss of plaque forming activity of B group phages but
no change in activity of A group phages.
RESULTS
Plaque morphology, host range and restriction of ES18
The S. typhimurium typing phage ~I8 (designated ESI8 by Dr B. A. D. Stocker) forms
clear medium-sized plaques (I.o mm) on the supposedly non-lysogenic S. typhimurium
strain Qz of Boyd (Boyd & Bidwell, I957). A doubly lysogenic S. typhimurium strain LT-2
Downloaded from www.microbiologyresearch.org by
IP: 78.47.19.138
On: Sat, 01 Oct 2016 03:40:37
266
N. YAMAMOTO
T a b l e 2. Host range of ESI8, Fels I and P181 phages*
Bacterial strain
ESI8
ESI8hl
Fels I
PI8I
Q
i
i
i
i
Q/22
Q/22/z8
i
i
i
i
< io
9
< IO
0
i
i
Q(Fels x)
I
I
< IO
6
Q/22(Fels I )
Q/22/I8(Fels
I
I
< Io
6
I
SDz4
SDI4/22
SDI4/22/I8
St
St/22
St/22/I8
1)
I
< IO -9
< IO -9
< IO
6
I
I
I
< 10-9
I
I
< I0 -9
< IO-6
< tO-~
< I0 -6
I
I
I
IO
3.~
I
< IO -6
I*
IO 3?
I
< I0 -6
IS
< 10 -6
I$
< IO -9
< I0 -9
* Determined by plating phage stocks, previously grown on non-lysogen Qz, on various bacterial strains.
Numeral ' I ' indicates e.o.p, of about I with small errors, e.o.p, ranging from o'3 to 1.2. In tests where e.o.p.
was < IO-9 or < IO-6, phage of titre /> io 9 or >~IO6 p.f.u, produced no plaques.
~E.o.p. of Jo 3 is due to restriction conferred by prophage Fels 2. Thus it seems likely that ESI8 carries
the restriction sensitive (Rs) genotype (see text).
~c Occasionally we found Pt81 strains expressing Rs phenotype, suggesting that some PI8I strains carry
the Rs genotype of ESI8.
( a b b r e v i a t e d St) is p a r t i a l l y resistant to E S I 8 whereas its cured derivative SDz4, which is
free o f p r o p h a g e Fels 2 b u t still carries the p r o p h a g e Fels z, is fully sensitive to E S I 8 for its
lytic r e p l i c a t i o n ( K u o & Stocker, I97o).
A s shown in T a b l e 2, E S I 8 f o r m s distinct plaques o n strains singly lysogenic for Fels I,
SDx4, SDx4/22, Q(Fels z) a n d Q[22(Fels z), with p l a q u e m o r p h o l o g y similar to t h a t on
Qx a n d with p l a t i n g efficiency o f a b o u t I. W h e n E S I 8 is p l a t e d on St a n d St/22 which c a r r y
Fels x a n d Fels 2 p r o p h a g e s after I5 min p r e i n c u b a t i o n , p i n p o i n t plaques are f o u n d at
p l a t i n g efficiencies o f a b o u t I o -3. W i t h o u t the p r e i n c u b a t i o n the p l a t i n g efficiency decreases
to Io -5, p r o b a b l y b e c a u s e smaller plaques m a y n o t be detected. I n a g r e e m e n t with the w o r k
o f K u o & Stocker (197o), we also c o n c l u d e d t h a t Fels 2 p r o p h a g e confers o n the h o s t
b a c t e r i a p a r t i a l resistance to E S I 8 whereas Fels z p r o p h a g e shows no effect on E S I 8
replication. However, the b e h a v i o u r o f E S I 8 on Fels 2 lysogens is similar to t h a t o f
restriction sensitive P22 m u t a n t s , PzzRs, which are sensitive to restriction conferred b y
Fels 2 p r o p h a g e . I n d e e d we f o u n d t h a t the p a r t i a l resistance o f Fels 2 lysogens to E S I 8 is
due to restriction b y p r o p h a g e Fels 2 ( F u k u d a & Y a m a m o t o , I972a).
A h o s t range m u t a n t o f ESI8, E S I 8 h I , forms clear plaques o n Fels 2 lysogens, St a n d
St/22 at a p l a t i n g efficiency o f a b o u t I. The p l a q u e s o f ES18hI o n the Fels 2 lysogens are
similar to those on non-Fels 2 lysogens a n d are slightly larger t h a n those o f E S I 8 o n
non-Fels 2 lysogens. Thus E S I 8 h I is insensitive to restriction b y p r o p h a g e Fels 2.
Isolation o f a new phage species
W h e n E S I 8 p h a g e stocks c o n t a i n i n g a b o u t i o 1° E S I 8 p.f.u, particles previously g r o w n on
S D x 4 or SDz4/22 were p l a t e d o n an E S I 8 resistant h o s t SDz4122[I8, a b o u t 2oo small
plaques (o'7 m m in d i a m e t e r ) were found. These p l a q u e formers are designated P18I phage.
A m o n g Iz5 E S I 8 p h a g e stocks tested, a b o u t 96 ~ o f the stocks c o n t a i n e d 5 to 3ooo P I 8 I
p.f.u, particles. Thus, we r o u g h l y e s t i m a t e d t h a t the frequency o f P I 8 I in E S I 8 p h a g e stocks
was a b o u t i o -8. Similar results were also f o u n d with 45 E S I 8 h I p h a g e stocks g r o w n on
S D z 4 a n d SDz4/22. W h e n 77 E S I 8 h I p h a g e stocks g r o w n on St a n d St/22 were p l a t e d on
Downloaded from www.microbiologyresearch.org by
IP: 78.47.19.138
On: Sat, 01 Oct 2016 03:40:37
Hybrid between serologically unrelated phages
267
Table 3. Immunity relationships between Salmonella phages*
Lysogen
Q(ESz8I)
Q(ESz8hz)
Q(PI8I)J;
Q(Fels l )
ESI8
< IO - °
< I0 -°
< IO 9
I
ESI8hI
< I0 9
< IO -9
< IO -0
I
PI8I'~
< IO - 6
< I 0 -6
< IO -0
I
Fels I
I
I
I
< IO - s
* Phage stocks grown on non-lysogen Qz were assayed on lysogens for immunity relationships as expressed
b y e . o . p . Q(Fels z) is f u l l y s e n s i t i v e t o t h e s e p h a g e s a s s h o w n b y a n e . o . p , o f a b o u t I. P I 8 I s t r a i n s d e r i v e d
from ESI8 and from ESI8hI showed the same immunity response used as a plating phage'~ and as a
prophage:~.
St[ee[z8, P i 8 i phage was also found at frequencies of about Io -8. However, ESI8 and
ESI8hr phage stocks grown on non-lysogens Qir or Qee contained no PI8t phage.
As shown in Table 2, PI8I phage forms clear plaques on ES18-resistant hosts, SDz4/2e[I8
and St[e2118. Though PI8I plaques are smaller than those of ESI8, the clearness of P~8I
plaques is similar to that of the ESI8 strain used, implying that PI8I phage has inherited
the c allele of ESI8. However, since ESI8 and ESi8hi are clear plaque forming phages, we
were unable to study their wild type ( e + ) and other c markers.
Immunity relationships between ESI8 and PI 8I
Derivatives of S. typhimurium strain QI, lysogenized with either ESI8, PI81 or Fels I
were scored for their pattern of immunity to these phages (Table 3). P I 8I forms plaques on
Fels I lysogens but not on ESI8 lysogens. Moreover P18I lysogens are immune to ESI8.
Therefore it should be concluded that ES18 and PI8I are co-immune. This is consistent
with the view that PI8I carries the c region of ESI8. However, PI8I lysogens are sensitive
to Fels x simply because the phage P I 81 carries the c region of E S 18 in place of the c region
of Fels I.
Homology between ES18 and P22
Kuo & Stocker (I97O) reported that ES~8 and P22 are both generalized transducing
phages and share the same prophage insertion site though they are morphologically and
serologically unrelated. Because of these surprising similarities, we conducted genetic
recombination studies to demonstrate a possible homology between these phages. When
S. typhimurium QI was mixedly infected with P22c+ and ESI8, E S I 8 e + recombinants
(carrying the c + of P22) were found at a frequency of o'o3 %. Despite their serological
unrelatedness this recombination frequency is very high, suggesting that ES 18 and P22 have
a homology containing at least the e region. With the ES 18c + recombinants the formation
of Q1 lysogens was more efficient (about 9o %) than that with ESI8 (less than 5 %) by the
method of Kuo & Stocker (I97o). When Q1 lysogenic for E S I 8 e + was superinfected with
P22cl, P22c2 or P22c 3 and the resulting lysates were assayed on Q]e2, ES18 recombinants
carrying these P22c markers (ESI8eI, ESI8c2 and ESI8e3) were found at a frequency of
about o.1%.
The ES 18 recombinants carrying the P22c markers were used for isolation of phage P 18 I.
Although PI 81 plaques appear small, it was evident that the e markers of PI 81 phage mimic
those of the P22c markers in the ESI8 recombinant used. For example, when E S I 8 c + is
used PI81 phage exhibits turbid plaques. In contrast, ESI8cI, ESi8c2 and ESI8c3 yield
PI8I phages that express the corresponding degree of clearness typical of P22c markers in
the ESI8 recombinant used.
Downloaded from www.microbiologyresearch.org by
IP: 78.47.19.138
On: Sat, 01 Oct 2016 03:40:37
268
N. YAMAMOTO
Table 4. Serological cross-reactivity between Salmonellaphages as expressedby rate constants*
of neutralization of plaque forming activity
Rabbit
antiserum
ESI8
Anti-ESI8
Anti-P22
14o
.
Anti-Fels
I
-
2
.
Anti-P22I
Anti-Feb
ESI8hI
PI8I
145
.
.
-.
-~"
.
I28
I04
I22
lO2
P22
I
.
.
I 15 °
.
-
--
Eels
.
.
.
- -
-.
P221
Eels
g
.
-I3 t
I IO
---85o
* V a l u e s i n d i c a t e r a t e c o n s t a n t s , k , m i n 1.
"~ - , N o e f f e c t o f s e r u m o n p h a g e v i a b i l i t y .
Antigenic analysis of PI8I and ESI8
Since P I 8 I is distinguishable from ESI8 and ESI8hI by host range, we were able to
obtain pure P I 8 I clones free of ESI8 and Fels I by several cloning cycles on Q[22[I8.
Several P I 8 I clones derived from independent ESI8 and ESI8hI stocks were purified and
tested to determine whether P I 8 I is antigenically related to ESI8.
As shown in Table 4, rabbit antiserum against ES 18 inactivated ES 18 with a rate constant,
k, of I4o min -1 but showed no effect on P I 8 I infectivity. In addition, antiserum against
P22 inactivated P22 with a rate constant of II5O rain -1 but both P I 8 I and ESI8 were
unaffected by the anti-Pzz serum.
Since all of the ES 18 and ES 18h I propagating hosts carry Fels I prophage and the isolation
principle and procedure of P181 are similar to those of the hybrid phage P22I between P22
and Fels I ( Y a m a m o t o & Anderson, I96I; Y a m a m o t o , I967), it was desirable to study
possible antigenic relationships between P I 8 I and Fels z. Anti-Fels z, which inactivates
Fels I at a rate constant of izz min -~, inactivated P I 8 I with about the same rate constant
I28 min -1. We previously reported that the antigenicity of Fels z is identical to that of
P22I (Yamamoto, 1967). Thus anti-Pz2I serum was also used for analysing the antigenicity
of P I 8 I . As seen in Table 4, anti-Pz2I, which inactivated P22I with a rate constant of
i i o min -1, inactivated both Fels z and P I 8 I with about the same rate constant, IOZ to
to4 min -1. Furthermore, it should be noted here that anti-serum against Fels 2, which is
another prophage of St and absent in SDz 4, inactivated Fels 2 with a rate constant of
85o min -1 but did not inactivate any of the above phages. These data indicated that the tail
antigens responsible for adsorption of PI 81 are serologically identical to those of Fels i. Due
to its antigenic structure, capacity to plate on ESI8 resistant derivatives of Fels I lysogens
and inheritance of the c marker of ESI8, we concluded that P I 8 I is a hybrid between ESI8
and Feb L
Demonstration of recombination between ESI8 and the prophage Fels I
Although we concluded that Pr8i is a hybrid between ES~8 and Fels I, several attempts
to isolate P I 8 I by mixed infections of a non-lysogenic S. typhimurium strain Q I with ESI8
and Fels I were unsuccessful. This observation suggests that the prophage form of Fels I
may be required for recombination between ESI8 and Fels I genomes. We thus prepared
various Fels z lysogens of Qz derivatives, Q(Fels x), Q/22(Fels I) and Q/22[i8(Fels I).
When ESI8 stocks and ESIShI stocks grown on Q(Fels I) and Q[22(Fels I) were plated
on Q[22]z8(Fels z), P I 8 I plaques were readily found at a frequency of about lO-8.
Downloaded from www.microbiologyresearch.org by
IP: 78.47.19.138
On: Sat, 01 Oct 2016 03:40:37
Hybrid between serologically unrelated phages
269
Contribution of the host cell functions to recombination between ESI8 and the
prophage Fels I
Requirement of the prophage form ofFels r for recombination between ESI8 and Fels z
genomes may suggest a possible contribution of the bacterial functions to this type of
recombination. Accordingly, St recA leu I97, St recA his :~2z-I and Qz recA35 (Fels I) were
used for preparation of ESI8hI phage stocks. Among 5o ESI8hI stocks (containing about
2 x Io 1° p.f.u.) tested, only four stocks contain a few (2 to 7) P~8I p.f.u, particles. Thus the
frequency of PI8I formation in recA hosts is roughly xo-11. This frequency is about IOOOfold lower than that of PI8I formation in the wild type hosts, suggesting that the host cell
recombination function plays an important role for recombination between ESr8 and the
prophage Fels z.
The origin of ESI 8
Since ES 18 is a clear plaque forming phage, it was desirable to obtain the wild type parent
phage of ESI8 for genetic study. Callow 0959) reported that ESi8 was derived from a temperate phage carried by the typing standard bacterial strain no. I9. Although we did not
carry S. typhimurium typing standard bacterial strains, an old Callow's phage stock of typing
phage :~I9 grown on its propagating host, typing standard bacterial strain no. ~9, was
available in our laboratory. Thus, we attempted to isolate the parent phage of ESI8 from
the :~I9 typing phage stock. Using S. typhimurium strains Qz and Q/e2, we isolated two types
of phages designated ESI9 and ESr9A. ESI9 is a clear plaque forming B group phage and
serologically indistinguishable from ES 18 whereas ES 19A is a turbid plaque forming A group
phage and serologically indistinguishable from P22. Later, Dr E. S. Anderson of Central
Enteric Reference Laboratory, London, kindly sent us the typing standard bacterial strain
no. I9. We then identified two types of turbid plaque forming temperate phages designated
~I 9 and ~br9A, from a nutrient broth culture of the standard bacterial strain no. I9. As
shown in Tables I and 5, ~I9A is an A group phage and is identical to ESI9A in antigenicity
and some genetic traits. Thus ~bI9A is the parent phage of ESI9 A. In contrast ~bI9 is a bidtur
plaque forming B group phage and serologically indistinguishable from ESI 8. Therefore, it
is concluded that ~I9 is the wild type parent strain of ESI8 and that ESI8 and ESr9 are c
mutants of ~br9.
As expected from their serological relatedness, electron microscopic analyses (data not
shown) showed that ~b~9 and ESI9 are morphologically identical to ESI8 which carries
a flexible long tail while ~5x9A and ES 19A are morphologically indistinguishable from phage
P22 which has a short stubby tail with a hexagonal base plate attached to six spikes.
As shown in Table 5, the above newly isolated four phages, ~bI9, ESI9, ~519Aand ESI9A,
are all generalized transducing phages and can recombine with the prophage Fels z to yield
hybrid phages with the corresponding c markers of the phages which give rise to them. In
addition, like ES 18, they are all sensitive to restriction by Fels e prophage. This phenotype is
analogous to that of the P22 restriction sensitive gene, PzeRs, which is located between the
c3 and erfgenes of P22 (Fukuda & Yamamoto, r972b).
Although ~bI9 is serologically and morphologically distinguishable from A group phages
such as ~I9A and P22, we found that ~bI9 can exchange the c marker with ~bI9A and P22
at frequencies of about 0"03 ~o- Since ~519 and ~bIgA share some common genetic traits
(including the c markers, restriction sensitivity and generalized transducing activity), it may
be suggested that ~bI9 is a hybrid between ~i9 A and an unidentified B group prophage in
the typing standard bacterial strain no. I9 or its ancestor bacterium.
Downloaded from www.microbiologyresearch.org by
IP: 78.47.19.138
On: Sat, 01 Oct 2016 03:40:37
Downloaded from www.microbiologyresearch.org by
IP: 78.47.19.138
On: Sat, 01 Oct 2016 03:40:37
Turbid
~b19
St
+
Rs
+
Rs
Rs
Rs
Rs
+
Rs
r
+
+
+
+
+
+
+
Generalized
transduction
-
--
-
Rs
-
Rs
+
Rs
St / 22
traits
and its relatives
+
+
+
+
+
+
+
+
-
_
-
+
--
+
+
+
+
+
+
+
+
+
+
+
-
_
-
+
-
+
+
+
Rs
+
Rs
Rs
Rs
Rs
+
Rs
(P22I)
+
(P22l like)
+
+
(P181)
+
(P22I like)
+
(PI81)
+
(P180
(PI8I)
+
+
Ability to
recombine with
Host range on lysogens*
F e b z prophage
~
~
(resulting hybrid
SDz4
SDz4/2.~
Q ( F e l s x) Q e e ( F e l s z) Q ( F e l s 2)
type)
of ESl8
* R x , Sensitive to restriction conferred by F e l s 2 prophage (see text). The m a x i m u m e.o.p, is about ~o-3. + , Indicates sensitivity and
plaque formation; - , indicates resistance and no plaques.
Various c mutants (ci, c2 and c3) and wild type c + were tested.
c or c +
c or c +
Turbid
ES~9A
P22 +
P22Rs'~"
Clear
ES 19
Turbid
Clear
ES 18h x
19A
Clear
E118
Phage strain
Plaque
morphology
(c marker)
T a b l e 5. G e n e t i c
to
©
©
O
Hybrid between serologically unrelated phages
27I
DISCUSSION
The present study suggests that recombination between unrelated bacteriophages requires
at least the following three indispensable conditions. (~) The host bacteria have to be wild
type recA hosts (recA +); (2) one of the parental phage has to be a prophage of the recA +
host; and (3) the other parental phage should be a generalized transducing phage.
Generalized transducing phage protein coat can pack DNA, presumably by the headful
mechanism (Thomas, ~967), from a terminal of the randomly fragmented bacterial chromosome as well as the concatenated phage genomes. It seems likely that the fragmentation of
the bacterial chromosome possibly by endonuclease is initiated by infection with the
generalized transducing phage. If a random cut is introduced in the prophage genome, the
host cell recombination function may join the superinfecting transducing phage genome
or its segments with a segment of the prophage at a terminal of the fragmented chromosome.
If these requisites for this type of recombination are satisfied, hybrids between various
unrelated phages may be created. For example, we also found that ESI8hI which is able to
grow in Fels 2 lysogens can recombine with the prophage Fels 2 to yield a new hybrid phage
species carrying the Fels e protein coat and at least the c regions ofESi8h~. This is a situation
analogous to formation of the hybrid phage between P22 and the prophage Fels 2
(Yamamoto, I969a).
In our unpublished work, we have studied effects of the bacterial repair and recombination
functions on the formation mechanism of P22I (the hybrid between the unrelated phages
P22 and Fels I). P22 phage stocks grown on Fels I lysogens of wild type strains [i.e. St,
S D x 4 and Q(Fels O] and their u.v. excision mutants (i.e. UVrB) contain P22I phage
particles at frequencies of about I0 -l° whereas P22 stocks grown on Fels I lysogens of recA
mutants contain P22I phage particles with frequencies lower than Io -12. These results
support the aforementioned hypothesis that the bacterial recombination function, recA,
plays an essential role in recombination between unrelated phage and prophage. However,
the u.v. excision function is not involved in this type of recombination. Furthermore, when
P22 phage mutants lacking the essential recombination function (err) (Yamagami &
Yamamoto, I970) or integration function (int) (Smith & Levine, I967) were used, we found
that all of these P22 mutant stocks grown on wild type bacterial strains lysogenic for Fels I
contain P22I particles at frequencies similar to that with the wild type P22 phage. This
observation suggests that the phage recombination functions of the superinfecting phage do
not contribute to recombination between the superinfecting phage and its unrelated
prophage.
This work was supported in part by U.S. Public Health Service Grants AI-o6429 and
CA-I2zz 7 from the National Institutes of Health.
REFERENCES
8OYD, J. S. K. (I950). T h e symbiotic bacteriophages of Salmonella typhimurium. Journal of Pathology and
Bacteriology 72, 5oi-517.
BOYD, J. s. K. & BIDWELL, D. E. (I957). T h e type A phages o f Salmonella typhimurium: identification by a
standardized c r o s s - i m m u n i t y test. Journal of General Microbiology x6, 217-228.
CALLOW, B. R. (I959). A new phage-typing s c h e m e for Salmonella typhimurium. Journal of Hygiene 57,
346-359.
FUKUDA, S. & YAMAMOTO,N. (I972a). A restriction sensitive m u t a n t of Salmonella p h a g e P22. Bacteriological
Proceedings, p. 237.
FUKUDA, S. & YAMAMOTO, N. (I972b). F o r m a t i o n o f various g e n o m e lengths o f hybrids between the serologically a n d morphologically unrelated bacteriophage species. Virology 5o, 727-732.
Downloaded from www.microbiologyresearch.org by
IP: 78.47.19.138
On: Sat, 01 Oct 2016 03:40:37
272
N. Y A M A M O T O
GEMSIO, P., BARON,L. S. & YAMAMOTO,N. (I972). Formation of hybrids between coliphage y and Salmonella
phage P22 with a Salmonella typhimurium hybrid sensitive to these phages. Proceedings of the National
Academy of Sciences o f the United States of America 69, 3110-3II 4.
KUO, T. & STOCKER,B. A. D. (I970). ESI 8, a general transducing phage for smooth and nonsmooth Salmonella
typhimurium. Virology 42, 621-632.
SMITH, H. O. & LEVlNE,M. (1967)- A phage P22 gene controlling integration of prophage. Virology 3x, 2o7-2I 6.
Sl'ICER, C. C. & DATTA, N. (I959). Reversion of transduced antigenic characters in Salmonella typhimurium.
Journal of General Microbiology 2o, I36-143 .
THOMAS, C. A. (1967). The rule of the ring. Journal of Cellular Physiology 70, suppl. I, 13-I4.
YAMAGAMI,H. & YAMAMOTO,N. (1970). Contribution of the bacterial recombination function to replication
of bacteriophage P22. Journal of Molecular Biology 53, 281-285.
YAMAMOTO,N. (I967). The origin of bacteriophage P22I. Virology 33, 545-547.
YAMAMOTO, N. (1969a). Genetic evolution of bacteriophage. I. Hybrids between unrelated bacteriophages
P22 and Fels 2. Proceedings of the National Academy of Sciences of the United States of America 62,
63-69.
YAMAMOTO, N. 0969b). Damage, repair and recombination. 1. Contribution of the host to recombination
between the superinfecting phage and the prophage. Virology 38, 447-456.
YAMAMOTO,N. (1970). Generalized transducing phage ES~8 recombines with other phage species. Bacteriological Proceedings, p. 2o2.
YAMAMOTO,N. & ANDERSON,T. r. (1961). Genomic masking and recombination between serologically unrelated
phages P22 and P22I. Viroiogy I4, 43o-439 .
(Received 18 M a y t977)
Downloaded from www.microbiologyresearch.org by
IP: 78.47.19.138
On: Sat, 01 Oct 2016 03:40:37