957
J. gen. Virol. (1987), 68, 957-963. Printed in Great Britain
Key words:phage T5/macromolecular assembly~head-tail linker protein
The Head-Tail Linker Protein of Bacteriophage T5: Genetic and
Immunological Studies
By D E S M O N D
K A Y , * A N N E. W A K E F I E L D t
AND W I L L I A M
S. J A M E S
Sir William Dunn School o f Pathology, University o f Oxford,
South Parks Road, Oxford OX1 3RE, U.K.
(Accepted 7 January 1987)
SUMMARY
We investigated the properties of amber mutants of coliphage T5 defective in a late
stage of phage assembly. The non-functional heads synthesized by mutants am25 or
am 158 were unable to combine with functional tails to produce viable phage particles.
These mutants were shown to lack a single protein, designated the head-tail linker
protein (HTLp), which was identified by polyacrylamide gel electrophoresis and
Western blot analysis and had an Mr of 18000. An extract containing the HTLp, when
added to the HTLp-deficient heads, restored their ability to combine with functional
tails.
INTRODUCTION
The bacteriophage T5 is a coliphage with a particle tool. wt. of 109 x 106 (Dubin et al., 1970).
The icosahedral head contains a single molecule of d s D N A 121.3 kbp in length with an 8%
terminal repetition (Rhoades, 1982). T5 possesses a non-contractile tail, 200 nm long, to which
three L-shaped tail fibres (LTF) and one straight tail fibre (STF) are attached at the distal end
(Saigo, 1978). The major structural proteins of the head and tail have been identified and their
relative molecular masses determined by S D S - P A G E . A total of six head proteins and nine tail
proteins have been described (Zweig & Cummings, 1973; Heller, 1984).
Complete phage heads filled with D N A and functional tails are assembled by independent
pathways (Lunt & Kay, 1968). These functional phage substructures are able to complement
each other in vitro to form viable phage particles (Zweig & Cummings, 1973). Two amber
mutants, both in the same complementation group, have been identified which produce heads
filled with D N A and functional tails, but the heads are unable to combine with functional tails
to produce viable phage particles in an in vitro complementation test (Kay & Wakefield, 1986).
We have proposed that this is due to the absence of a head-tail linker protein (HTLp), and in
this paper we describe the identification of this protein.
METHODS
Strains of bacteria and bacteriophage. Escherichia coli F (Lanni, 1958) was the non-permissive strain and E. coli
HR34 (Lanni & Lanni, 1966)was the permissive strain for growth of amber mutants. The T5 amber mutants used
in the present work were characterized by Kay & Wakefield (1986): am25 and am158, complementation group 11,
producers of functional tails and non-functional heads in E. coli F; am123, complementation group 20, and am l 7,
complementation group 8, producers of functional heads but no tails; am28, complementation group 13, producer
of functional tails but no heads.
Media and buffers. These were as previously described (Kay & Wakefield, 1986). For infections with T5, media
were made 1 mM with CaC12. All incubations were carried out at 34 °C.
In vitro complementation. This was done as previously described (Kay & Wakefield, 1986).
Large-scalepreparation and purification of phage and amber mutant products. Four 500 ml nutrient broth cultures
in 51flasks were aerated by rotation, grownto2 × 108cells/mlandinfectedatanm.o.i.of4witheitherwild-type
t Present address: Department of Paediatrics, University of Oxford, John Radcliffe Hospital, Oxford
OX3 9DU, U.K.
0000-7476 © 1987 SGM
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 29 Jul 2017 00:53:21
958
D. KAY, A. E. WAKEFIELD AND W. S. JAMES
T5 (T5am÷) or an amber mutant (am17, am25, am123 or am 158). After incubation overnight in the case of T5am÷
or 100 min in the case of the amber mutants, the cultures (except T5am÷) were treated with chloroform (0-5 ml/l)
and stored at 4 °C overnight. After sedimentation of cell debris at 3000 r.p.m, for 15 min at 4 °C, the supernatant
was treated with NaC1 (27 g/l) and polyethylene glycol (PEG) 6000 (100 g/l) and stored at 4 °C overnight. The
precipitate was collected by sedimentation at 10000 r.p.m, for 15 min at 4°C, resuspended in T5 buffer and freed
from PEG 6000 by sedimentation at 10000 r.p.m, followed by washing the sediment with T5 buffer and pooling
the phage-containing supernatants. The T5am÷ or am mutant product was sedimented at 30000 r.p.m, for 60 min
at 4 °C, resuspended in 20 ml T5 buffer. CsCI (19.5g) was then dissolved in this suspension. After centrifugation at
40000 r.p.m, for 18h at 20 °C the visible band was removed with a hypodermic needle, diluted with T5 buffer and
sedimented at 30000 r.p.m, for 60 min at 4 °C. The pellet was resuspended in T5 buffer (8 ml) and fractionated on
sucrose density gradients (15~ to 30~, w/v) at 14200 r.p.m, for 60 min at 20 °C in a Beckman SW28 rotor. The
position of the phage was recorded using an Isco u.v. analyser and the peak fractions were pooled and stored with
10-2 M-sodiumazide. The content of phage or heads was standardized by their u.v. absorption at 260 nm.
Preparation of HTL protein-containing extract. Four 500 ml cultures of E. coli F were grown in nutrient broth to
1-5 x 108 cells/ml and infected with T5am÷ phage at an m.o.i, of 5. Incubation was stopped after 110 min (before
lysis)and the cellswere harvested by centrifugation at 3000 r.p.m, for 15 min at 4 °C. The cells were resuspended in
15 ml ofT5 buffer with 10-: M-sodiumazide, centrifuged at 30000 r.p.m, for 60 min at 4 °C and the supernatant
was subjected to u.v. light 340 mm below a 15 W germicidal lamp for 5 min.
Antisera. Rabbits were injected intramuscularly three times with 0.5 ml of a mixture of either T5am+ or am 17p
with Freund's adjuvant at 20 day intervals and bled after a further 20 days. Both the phage and the aml7p used
were peak fractions from sucrose density gradients with titre or phage equivalent concentrations of 7 x 10~~/ml.
Adsorbed antiserum was prepared by mixing anti-am17p antiserum with an equal volume of sucrose gradientfractionated am25p, incubating for 60 min at 34 °C and removing remaining am25p and antigen-antibody
complexes by centrifugation (30000 r.p.m., 60 min). The process was repeated three times.
Sucrose density centrifugation. This was as described previously (Kay & Wakefield, 1986).
SDS-PAGE. This was performed by the method of Laemmli & Favre (1973).
Westernblotting. This was performed as described by Burnette (1981). The first antibody was phage-specific(see
text) and the second antibody was radioiodinated horse anti-rabbit immunoglobulin.
RESULTS
In vitro complementation analysis o f am25 product
The amber mutants, am25 and am158, have been assigned to complementation group 11 by
the in vivo complementation test (Kay & Wakefield,. 1986). During infection of the nonpermissive host, these mutants produced heads filled with D N A , as detected by sucrose density
gradient analysis. In vitro complementation analysis of the products of infection under nonpermissive conditions revealed that both am25 and am158 synthesized functional tails but that
the head-like particles which had been synthesized were unable to yield infectious particles
when treated with functional tails (Kay & Wakefield, 1986). This observation was investigated
further and the results are shown in Fig. 1. The products of infection of the non-permissive host
were prepared, a m l 7 being chosen as the functional head producer and am28 as the functional
tail producer. These structures were used for the in vitro complementation experiments and the
products of the complementation were analysed on sucrose density gradients. Fig. 1 (a) shows
the position of the T5am ÷ particles in the sucrose density gradient and Fig. 1 (b) shows the
absence of any peak in the am28 product, am28p (the suffix 'p' after the m u t a n t denotes the
product of infection under non-permissive conditions). Fig. 1 (c) and 1 (d) show the peaks
obtained with a m l 7 p and am25p, respectively, indicating that both functional heads, aml7p,
and inactive heads, am25p, run to the same position in the gradient which is lower than that of
complete T5am* particles. Incubation of a mixture of aml 7p and am28p resulted in a reduction
of the peak corresponding to the free heads and the appearance of a peak corresponding to that
of T5am ÷ (Fig. I e). Analysis of the samples for plaque-forming ability showed that the peak of
infectivity corresponded to the latter peak. In contrast, when am25p was incubated with am28p,
no peak corresponding to complete phage appeared (Fig. 1 f). Therefore, the heads produced by
am25, though indistinguishable by sucrose gradient analysis from those produced by aml 7, are,
in contrast, unable to combine with functional tails to produce complete phage particles. W h e n
am25p and a m l 7 p were mixed, complete phage particles were produced (Fig. 1g). This shows
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 29 Jul 2017 00:53:21
Head-tail linker protein of T5
(c)
0-08 (a)
959
(e)
4.0
g-,
r
0.06
1
3.0 •.~~ -×
09~
E
0.04
2-0 -=
r
1-0
0-02
,.Q
e~
0
(d)
'~ 0-08 (b)
t-
09
-
~)
I
0.06
i
0.04
0-02 _
0
Direction of sedimentation
Fig. 1. Analysis of T5 phage wild-type, mutant products and the results of in vitro complementation on
sucrose density gradients. Cultures of E. coli F, grown to a density of 2.5 × 108 ceUs/ml in nutrient broth
were infected with phage at an m.o.i, of 3 to 5. After incubation at 34 °C with shaking for 100 min,
chloroform (0.5 ml/1) was added and the cultures were stored overnight at 4 °C Bacterial debris was
removed by centrifugation. Phage or mutant products were incubated alone or in mixtures for 60 min at
34 °C, concentrated 20-fold by centrifugation at 30000 r.p.m, for 60 min, applied to gradients (4.5 ml) of
15 to 3070 sucrose (w/v) in 50 mM-Tris-HC1 pH 7.5 and centrifuged at 18000 r.p.m, for 25 min at 20 °C
in a Beckman SW50.1 rotor. The gradients were analysed by upward displacement through an Isco
ultraviolet fractionator. (a) T5am ÷, (b) am28p, (c) aml7p, (d) am25p, (e) aml7p + am28p (peak of
infectivity defined by open circles), ( f ) am25p + am28p and (g) aml7p + am25p.
that am25 synthesizes functional tails which, although unable to combine with its own nonfunctional heads, can combine with the functional heads of aml7.
SDS-PAGE analysis of non-functional heads
PAGE was used to detect any differences between the protein composition of the functional
and non-functional heads. In these experiments, am123 was used as the functional headproducer. T5am ÷, am25p, am123p and am158p particles were analysed by SDS-PAGE (see Fig.
2a). In order to reveal differences between the minor proteins, some of the major bands were
heavily overloaded. The protein bands of the wild-type, numbered according to Zweig &
Cummings (1973) and Heller (1984) comprise bands 1 and 2, the tail fibre proteins (Mr 125000
and 108 000); bands 3 and 4, tail proteins (Mr 103 000 and 75 000); band 5, the receptor-binding
protein (Mr 67000); band 6, the major tail protein (Mr 58000); band 7, a head protein (Mr
44000); band 8, the major head protein (Mr 32000); band 9, a tail protein (Mr 22000); bands 10
and 11, head proteins (Mr 19 000 and 18 000); and bands 12 and 13, tail proteins (Mr 17 000 and
15000).
Neither am25p, am158p nor am123p contained any of the structural proteins of the phage tail
(Fig. 2a), but the non-functional heads, am25p and am158p also lacked the band corresponding
to the head protein 11 which was present in the functional heads, aml23p. By comparison with
the mobility of molecular weight markers, this protein appears to have an Mr of 18 000. Since it is
absent from heads which are unable to link with functional tails, we propose that it constitutes
the HTLp of phage T5.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 29 Jul 2017 00:53:21
960
D. K A Y , A. E. W A K E F I E L D A N D W . S. JAMES
(a)
(b)
1
3
4
1
2
3
4
(c)
1
2-3/4-5-6--
8--
9-10-._
11-12-13--
Fig. 2. Analysis ofT5 wild-type and mutant products by SDS-PAGE and immunoblotting. Samples of
lane t, T5-am +; lane 2, am25p; lane 3, am123pand lane 4 am158p, purified on sucrose density gradients
(see legend to Fig. l) were analysed on 10~ discontinuous SDS-polyacrylamide gels. (a) Coomassie
Brilliant Blue-stained gel, (b) Coomassie Brilliant Blue-stained gel at 1/10 loading of (a). (c) Immunoblot
of (b) using am25p-adsorbed rabbit anti-am123p serum as first antibody and radioiodinated horse antirabbit second antibody.
T a b l e 1. Effect of adsorbed and non-adsorbed serum on wild-type T5 and in vitro
complementation between amber mutants
Phage or
mutant product
T5am +
T5am ÷
T5am ÷
am123p
am28p
am123p
am123p
am123p
Pre-incubation
Whole serum
Adsorbed serum
Whole serum
Adsorbed serum
Second
mutant product
am28p
am28p
am28p
Titre
(p.f.u./ml)
3.2 x 10lo
6-4 × 107
3.1 x 10 l0
5-9 x 107
2.9 x 108
1-6 x 1010
5.6 x 108
5.0 × 109
The effect of anti-HTLp antibody on in vitro complementation
A serum was raised against purified T5 heads (see Methods). In an a t t e m p t to p r e p a r e a
monospecific a n t i b o d y against the H F L p , the whole a n t i s e r u m was adsorbed three times w i t h
the H T L p - d e f i c i e n t heads, am25p. T h e whole s e r u m and the a d s o r b e d serum were tested for the
ability to neutralize the infectivity o f wild-type T5 (Table 1). W h e r e a s the c o m p l e t e a n t i s e r u m
reduced the titre o f T 5 a m + by 98 %, the adsorbed serum had no significant effect. P r e - i n c u b a t i o n
with adsorbed serum o f the functional heads, am123p, reduced the efficiency w i t h w h i c h they
f o r m e d viable p h a g e w h e n i n c u b a t e d w i t h excess functional tails, a l t h o u g h u n a d s o r b e d s e r u m
was still m o r e p o t e n t t h a n a d s o r b e d s e r u m (Table 1).
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 29 Jul 2017 00:53:21
Head-tail linker protein of T5
961
Table 2. Conversion of am25p heads to functional form
Sample number
1
2
3
4
5
6
7
8
9
10
Components
am25p alone
am28p alone
Extract alone
amt23p alone
am28p + am25p
am28p + am123p
Extract + 28p
Extract + am123p
Extract + am25p
Extract + am28p + am25p
Phage titres (p.f.u./ml)
2.8 x 105
2.0 x 103
1.2 X 10 3
7-2 × 107
2.0 x 105
5.8 x 101°
3.2 x 103
4-0 × 10t°
7.2 x 107
1.2 x 10s
Western blot analysis of HTLp-deficient, non-functional heads
We probed a nitrocellulose blot of wild-type and mutant phage proteins with adsorbed and
unadsorbed serum (see Methods) to test the specificity of our adsorbed serum and confirm that
the putative HTLp was indeed absent from defective heads and corresponded to band 11. The
quantity of sample used was such that band 11 was not detected by Coomassie Brilliant Blue
staining (Fig. 2b). The Western blot of these samples, using adsorbed serum as the first
antibody, clearly displayed band 11 in the wild-type and am123p samples, but not in those of the
inactive heads, am25p and am158p (Fig. 2c). The adsorbed serum however, revealed two other
minor head proteins, corresponding to band 7 Mr 43000 and band 10 Mr 19000, in all four
samples. Although the role of these proteins is unknown, they may be internal proteins and thus
unavailable for removal of the corresponding antibodies during adsorption.
Conversion of non-functional to functional heads using a source of HTLp
We considered that cells infected with wild-type phage might contain HTLp in a free form
just before lysis. Accordingly, an extract of such cells was treated with u.v. to inactivate the
phage (see Methods) and was added to non-functional heads, am25p, for 60 min at 34 °C. The
results (Table 2) show that although addition of excess functional tails (am28p) to the nonfunctional heads (am25p) made little difference to the titre of infectious phage (sample 5), the
addition of u.v.-irradiated extract of wild-type phage-infected cells to the non-functional heads
increased the titre by more than 100-fold (sample 9). This was not as large an increase in titre as
that obtained by mixing functional heads with functional tails (sample 6) or functional heads
with extracts of wild-type-infected cells (sample 8). This presumably reflects either a molar ratio
of free HTLp to other components of less than one or the inefficiency of reassociation of H T L p
with pre-formed heads. The extract of wild-type-infected cells did not complement am28p
(sample 7) which confirms that it contained no active heads but a large quantity of functional
tails in addition to HTLp. It would seem, therefore, that HTLp can be added to otherwise
complete non-functional heads to restore their ability to combine with functional tails.
DISCUSSION
All the double-stranded DNA-containing bacteriophages which have been examined in detail
possess heads of regular or prolate icosahedral shape. One of the vertices of the head carries the
complex tail structure. The icosahedral head capsid has a fivefold symmetry at the vertex while
the tubular tail has a sixfold rotational symmetry. This difference in symmetry was at one time
viewed as an architectural paradox (Moody, 1965) but is now partially explained by the structure
of the proteins found at the vertex. In the case of coliphages lambda, T4, P22, T3 and T7 and the
Bacillus subtilis phage ¢29, the vertex is formed by an oligomer composed of one or, in some
cases, several of the minor structural components of the head. In the case of lambda, T4 and ¢29,
this oligomer has a 12-fold rotational symmetry (Kochan et al., 1984; Driedonks et al., 1981;
Carrascosa et al., 1985; for review, see Bazinet & King, 1985). This protein structure has also
been implicated in the initiation of prohead assembly (Murialdo & Becker, 1978) and in the
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 29 Jul 2017 00:53:21
962
D. KAY, A. E. WAKEFIELD AND W. S. JAMES
packaging and release of the phage DNA (Earnshaw & Casjens, 1980; Bazinet & King, 1985)
and has therefore variously been named the initiator, the connector and the DNA-translocating
vertex.
In the case of phage T5, less is known about the functions of the structural proteins. Zweig &
Cummings (1973) and Heller (1984) found at least 15 structural proteins by SDS-PAGE of
which the functions of five are known. Protein band 1 corresponds to the LTF (Saigo, 1978),
band 2 to the STF (Heller & Schwarz, 1985), band 5 to the receptor-binding protein (Heller &
Schwarz, 1985), band 6 to the major tail protein and band 8 to the major head protein (Zweig &
Cummings, 1973). Our findings show that the HTLp (band 11) is essential for the attachment of
the formed tail to the filled head to give active phage.
Our studies show that whereas the product of infection of non-permissive strains by a headproducing, tail-deficient amber mutant such as am123 can combine with functional tails
produced by a second amber mutant, am28, the corresponding product of group 11 mutants,
am25 and am158, although indistinguishable from functional heads on sucrose gradients and by
electron microscopy, cannot complement with am28 tails.
Two possible explanations were considered: first, am25 heads may be deficient in some factor
essential for the attachment of tails; second, such attachment may occur but the resulting
particles are blocked in some function necessary for infection. We showed that whereas
functional heads (am123p) and functional tails (am28p) could combine to yield particles which
banded on velocity sedimentation gradients at a position typical of complete wild-type phage
particles, non-functional heads (am25p) could not do so. This result eliminates the second
possibility and confirms the first.
We have named the protein involved in this effect HTLp (head-tail linker protein). Its
structure must differ from the vertex proteins of lambda (monomer Mr 62 000) and T4 (monomer
Mr 65000) since it has a monomer of Mr 18000. Zweig & Cummings (1973) estimated that it
amounted to 1 ~ of the phage protein, which suggests a copy number of four to five per virion.
Another important difference between T5 HTLp and the vertex proteins reviewed above is that
it does not appear necessary for DNA packaging, since the non-functional, HTLp-deficient
heads contain the full complement of DNA as determined by restriction analysis (data not
shown). One would therefore expect that one of the other T5 head proteins has the function of
DNA translocation, and this may be an internal head protein such as that of band 7, Mr 43 000,
but it cannot be the protein of band 10 since that one is not essential (Saigo, 1978).
Since the yields of the products am28p, am25p and am123p from the non-permissive host
appeared to be similar to that of wild-type phage, as judged by the heights of u.v.-absorbing
peaks on the preparative gradients and the titres of active phage obtained from successful in vitro
complementation, we conclude that the products of the genes defined by the corresponding
mutations are not necessary for cellular lysis.
Since we observed that extracts of T5-infected cells prepared late in the latent period can
bring about the ability of non-functional heads to combine with tails and produce active phage,
we conclude, first, that an excess of HTLp over DNA-filled heads is produced during normal
infection, and second, that the sequence of events during T5 maturation includes the addition of
HTLp to the filled heads, followed by the attachment of tails to generate the complete phage.
Since HTLp-deficient heads are as stable as HTLp-containing heads (data not shown), we
conclude that HTLp is not necessary for the integrity of the filled head.
We should like to thank Mrs K. W. Brownsill for technical assistance.
REFERENCES
BAZtNET, ¢. ~ KIN~, ~. 0985). The DNA translocating vertex of dsDNA bacteriophage. Annual Review of
Microbiology 39, 109-129.
BURNETr~, W. N. (1981). "Western blotting": electrophoretic transfer of proteins from sodium dodecyl sulfatepolyacrylamide gels to unmodified nitrocellulose and radiographic detection with antibody and
radioiodinated Protein A. Analytical Biochemistry ll2, 195-203.
CARRASCOSA,L L., CARAZO,J. M., IBANEZ,C. & SANTISTEBAN,A. 0985). Structure of phage 429 connector protein
assembled in vitro. Journal of Molecular Biology 15,4, 311-324.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 29 Jul 2017 00:53:21
Head-tail linker protein of T5
963
DRIEDONKS, R. A., ENGEL, A. B., TEN HEGGELER, B. & VAN DRIEL, R. (1981). G ene 20 product of bacteriophage T4. Its
purification and structure. Journal of Molecular Biology 152, 641-662.
DUmN, S. a., BENADEK,G. a., BANCROFT, F. C. & FREIFELDER, D. (1970). Molecular weights of coliphages and
coliphage DNA. II. Measurements of diffusion coefficients using optical mixing spectroscopy, and
measurements of sedimentation coefficients. Journal of Molecular Biology 54, 547-556.
EAKNSHAW,W. C. & CASJENS,S. R. (1980). DNA packaging by the double-stranded DNA bacteriophages. Cell 21,
319-331.
HELLER,K. J. (1984). Identification of the phage gene for host receptor specificity by analyzing hybrid phages of T5
and BF23. Virology 139, 11-21.
HELLER, K. J. a SCHWARZ,H. (1985). Irreversible binding to the receptor of bacteriophages T5 and BF23 does not
occur with the tip of the tail. Journal of Bacteriology 162, 621-625.
KAY,D. & WAKEFIELD,A. E. (1986). The characterization of a set of amber mutants of bacteriophage T5. Journalof
General Virology 67, 1663-1672.
KOCHAN, J., CARP,ASCOSA, J. L. & MURIALDO,H. (1984). Bacteriophage lambda preconnectors. Purification and
structure. Journal of Molecular Biology 174, 433-447.
LAEMMLI,V. K. a rxvrtE, M. (1973). Maturation of the head of bacteriophage T4. I. DNA packaging events. Journal
of Molecular Biology 80, 575-599.
LANNI, Y. T. (1958). Lysis inhibition with a mutant of bacteriophage T5. Virology 5, 481-501.
LANNI,V. & LANNI,V. T. (1966). Genetic supressors of bacteriophage T5 amber mutants. Journal of Bacteriology 92,
521-523.
LLrNT,M. R. & KAY,D. (1968). The conversion of viral heads to T5 phage in vivo. Journal of General Virology 3, 459463.
MOODy, M. F. (1965). The shape of the T-even bacteriophage head. Virology 26, 567-576.
MURIALDO, H. & BECKER, A. (1978). Head morphogenesis of complex double-stranded deoxyribonucleic acid
bacteriophages. Microbiological Reviews 42, 529-576.
RHOADES, M. (1982). New physical map of bacteriophage T5 DNA. Journal of Virology 43, 566-573.
SAIGO, K. (1978). Isolation of high-density mutants and identification of nonessential structural proteins in
bacteriophage T5 : dispensability of L-shaped tail fibres and a secondary major head protein. Virology 85,
422-433.
ZWEIG, M. & CUMMINGS,D. J. (1973). Structural proteins of bacteriophage T5. Virology 51, 443-453.
(Received 26 September 1986)
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 29 Jul 2017 00:53:21