Induction of Shiga toxin prophage by abiotic environmental stress in food

45 Pages·2017·1.36 MB·English

AEM Accepted Manuscript Posted Online 4 August 2017
Appl. Environ. Microbiol. doi:10.1128/AEM.01378-17
Copyright © 2017 American Society for Microbiology. All Rights Reserved.
1 Induction of Shiga toxin prophage by abiotic environmental stress in food
2 Yuan Fang, Ryan G. Mercer, Lynn M. McMullen and Michael. G. Gänzle
3 University of Alberta, Dept. of Agricultural, Food and Nutritional Science, Edmonton,
D
4 AB, Canada. o
w
n
5 Running title: Induction of shiga-toxin prophage in food lo
a
d
e
6 Corresponding author footnote: d

f
r
o
m
7 Michael Gänzle,

h
t
t
p
8 University of Alberta, Dept. of Agricultural, Food, and Nutritional Science :
/
/
a
e
m
9 4-10 Ag/For,
.
a
s
m
10 Edmonton, AB, Canada, T6G 2P5
.
o
r
g
11 phone, + 1 780 492 0774; e-mail, [email protected] /
o
n

N
12
o
v
e
m
b
e
r

1
9
,

2
0
1
8

b
y

g
u
e
s
t
1
13 Abstract
14 The prophage-encoded Shiga toxin is a major virulence factor in Stx producing
15 Escherichia coli (STEC). Toxin and phage production are linked, and occur after
D
16 induction of the RecA-dependent SOS response. However, food-related stress and Stx o
w
n
17 prophage induction have not been studied on the single-cell level. This study investigated lo
a
d
e
18 the effects of abiotic environmental stress on stx expression by single-cell quantification d

f
r
o
m
19 of gene expression in STEC O104:H4 Δstx2:gfp:ampr. In addition, the effect of stress on

h
t
t
p
20 production of phage particles was determined. The lethality of stressors, including heat, :
/
/
a
e
m
21 HCl, lactic acid, hydrogen peroxide and high hydrostatic pressure, was selected to reduce
.
a
s
m
22 cell counts by 1-2 log(CFU/mL). The integrity of the bacterial membrane after exposure
.
o
r
g
23 to stress was measured by propidium iodide (PI). The fluorescent signals of GFP and PI /
o
n

N
24 were quantified by flow cytometry. The mechanism of prophage induction by stress was
o
v
e
m
25 evaluated by relative gene expression of recA and cell morphology. Acid (pH <3.5) and
b
e
r

26 H O (2.5 mM) induced the expression of stx2 in about 18 % and 3 % of the population, 1
2 2 9
,

2
27 respectively. The mechanism of prophage induction by acid differs from induction by 0
1
8

b
28 H2O2. H2O2 induction but not acid induction corresponded to production of infectious y

g
u
e
29 phage particles, upregulation of recA upregulation, and cell filamentation. Pressure (200
s
t
30 MPa) or heat did not induce the Stx2-prophage. Overall, the quantification method
31 developed in this study allowed investigation of prophage induction and physiological
32 properties on the single cell level. H O and acids mediate different pathway to induce
2 2
2
33 Stx2-prophage.
34 Importance: Induction of the Stx-prophage in STEC results in production of phage
35 particles and Stx and thus relates to virulence as well as the transduction of virulence
D
36 genes. This study developed a method for a detection of the induction of Stx-prophages at o
w
n
37 the single cell level; membrane permeability and an indication of SOS response to lo
a
d
e
38 environmental stress were additionally assessed. H O and mitomycin C induced d
2 2
f
r
o
m
39 expression of the prophage and activated a SOS response. In contrast, HCl and lactic acid

h
t
t
p
40 induced the Stx-prophage but not the SOS response. The lifestyle of STEC exposes the :
/
/
a
e
m
41 organism to intestinal and extra-intestinal environments that impose oxidative and acid
.
a
s
m
42 stress. A more thorough understanding of the influence of food-processing related
.
o
r
g
43 stressors on Stx-prophage expression thus facilitates control of STEC in food systems by /
o
n

N
44 minimizing prophage induction during food production and storage.
o
v
e
m
45 Keywords: Singe cell detection, E. coli O104: H4, prophage induction, environmental
b
e
r

46 stress, RecA-independent, SOS response, membrane permeability 1
9
,

2
47 0
1
8

b
y

g
u
e
s
t
3
48 Introduction
49 Shiga toxin producing Escherichia coli (STEC) cause the life-threatening hemolytic
50 uremic syndrome (HUS), which is associated with acute renal failure and significant
D
51 mortality (1, 2). Shiga toxins (Stx) include Stx1 and Stx2; Stx2 is the more potent of the o
w
n
52 two toxins causing renal failure (2). Shiga toxins are N-glycosidases that remove one lo
a
d
e
53 adenine from 28S rRNA, block binding of amino-acyl-tRNA to the ribosome, and thus d

f
r
o
m
54 inhibit protein synthesis, inducing apoptosis in human renal cell and tissue (3, 4). The

h
t
t
p
55 genes encoding Stx (stx) are located on a lambdoid prophage (5). The production of Stx is :
/
/
a
e
m
56 dependent on the expression of the Stx-encoding prophage. Induction to the lytic cycle
.
a
s
m
57 lyses the host cell, releasing toxins and phage particles (5, 6). Stx-phages can create novel
.
o
r
g
58 STEC by lysogenic infection. A STEC outbreak with more than 4000 cases and 50 deaths /
o
n

N
59 in Germany 2011 provides a prominent example for the evolution of novel pathotypes by
o
v
e
m
60 transduction of stx genes (7, 8). The outbreak strain is a novel pathotype with the serotype
b
e
r

61 O104:H4, combining the virulence factors of STEC and enteroaggregative E. coli (EAEC) 1
9
,

2
62 (7). Stx2-phages can also establish lysogeny in non-STEC including enteropathogenic, 0
1
8

b
63 enteroinvasive and enterotoxigenic E. coli and commensal E. coli strains in vitro (9-11). y

g
u
e
64 This demonstrates that Stx2-phages have a broad host range among strains of E. coli.
s
t
65 Production of Stx2-phages and Stx2 are controlled by the phage late gene promoter. The
66 lysogenic state is maintained by the phage repressor CI (12, 13). Expression of stx2 and
67 prophage induction results from proteolytic cleavage of the CI repressor. Many agents
4
68 that induce prophages also induce the SOS response (14, 15). Induction of SOS response
69 is due to the cleavage of the LexA repressor by the RecA nucleoprotein, which is formed
70 by RecA and single stranded DNA (16). Cleavage of LexA derepresses over 30 genes for
D
71 DNA repair and inhibition of cell division. CI repressors are also destroyed by RecA o
w
n
72 activation, leading prophage induction (17). lo
a
d
e
73 Environmental stress encountered by E. coli in the human intestine or in food d

f
r
o
m
74 processing also induces Stx-prophages (18-20). Human neutrophils in the intestine

h
t
t
p
75 release antibacterial molecules including H O which stimulate Stx and phage production :
2 2, /
/
a
e
m
76 (19). The Stx-phage released in the intestine may transduce to commensal strains and
.
a
s
m
77 thus amplify toxin production and exacerbate disease symptoms (10). Hydrostatic
.
o
r
g
78 pressure, a physical method of food preservation, also induced Stx prophages (18, 21). /
o
n

N
79 Prophage induction in any of the ecological niches populated by STEC drives horizontal
o
v
e
m
80 gene transfer by release of infectious phage particles that allows transduction of
b
e
r

81 Shiga-toxin negative phages, and thus disseminates genes coding for the Shiga-toxin in 1
9
,

2
82 intestinal and food systems. This study aimed to evaluate the effect of different stress on 0
1
8

b
83 Stx2-prophage induction and SOS response in an outbreak strain that caused significant y

g
u
e
84 mortality and morbidity (7, 8). To quantify stx2 expression, stx2 was replaced with the
s
t
85 gene (gfp) that encodes for green fluorescence protein in STEC O104:H4. Flow
86 cytometry was used to measure the fluorescence signals to determine the stress response
87 of the E. coli culture at the single cell level.
5
88 RESULTS
89 Generation and validation of E. coli O104:H4 Δstx2:gfp:ampr. To quantify the
90 expression of the stx2 during stress, stx2 was replaced by gfp as fluorescent reporter gene;
D
91 ampr was used as a selection marker to screen mutants. PCR amplification and o
w
n
92 sequencing confirmed that the stx2 was replaced with gfp and ampr in E. coli O104:H4 lo
a
d
e
93 Δstx2:gfp:ampr. The transcription of gfp and stx2 in response to stress was determined by d

f
r
o
m
94 reverse transcription (RT) qPCR. Mitomycin C increased expression of stx2 and gfp

h
t
t
p
95 about 5 fold (Fig. S1). The expression of stx2 in the wild type strain relative to gfp :
/
/
a
e
96 expression in the mutant strain at stress and control conditions was 0.76±0.034 and 0.68 m
.
a
s
m
97 ±0.26, respectively. Expression of stx2 and gfp at induced and un-induced conditions
.
o
r
g
98 was not different (P>0.05), demonstrating that the relative expression of gfp was /
o
n

N
99 equivalent to the expression of stx2 in response to mitomycin C.
o
v
e
m
100 Quantification of prophage induction by mitomycin C. Mitomycin C causes
b
e
r

101 cross-links of double stranded DNA, activates a SOS response, and induces cell 1
9
,

2
102 filamentation and phage production (22, 23). Exposure of E. coli O104:H4 0
1
8

b
103 Δstx2:gfp:ampr to mitomycin C for 2 and 3 h induced gfp expression in 17 and 56 % of y

g
u
e
104 cells of the population, respectively (Fig. 1). Of the cells that expressed GFP after 3 h of
s
t
105 treatment with mitomycin C, 19.4 % were also filamented, confirming that mitomycin C
106 induced a SOS response.
107 Stress resistance of E. coli O104:H4 Δstx2:gfp:ampr. The effect of environmental stress
6
108 on expression of stx2 and induction of the Stx2-prophage was evaluated with stressors
109 related to food preservation, which included heat, oxidative and acid stress, and pressure.
110 To compare prophage expression in response to different stressors, each stress was
D
111 applied to reduce bacterial cell counts by 1-2 log (CFU/mL) (Fig. 2). o
w
n
112 Effect of heat and hydrogen peroxide on membrane permeability and GFP lo
a
d
e
113 expression in E. coli O104:H4 Δstx2:gfp:ampr. Prophage expression and loss of d

f
r
o
m
114 membrane permeability were simultaneously assessed at the single cell level by flow

h
t
t
p
115 cytometric determination of GFP and PI fluorescence, respectively. The change of the :
/
/
a
e
m
116 GFP-expressing and PI permeant proportions of the population were measured over 3 h
.
a
s
m
117 of exposure to stress (Fig. 3), corresponding to the time required for full GFP induction
.
o
r
g
118 with mitomycin C. In control cultures, less than 3 % of the cells expressed GFP, or were /
o
n

N
119 stained with PI (Fig. 3A). Exposure to 50 °C for 3 h increased membrane permeability in
o
v
e
m
120 64 % of the cells but did not induce GFP (Fig. 3B). Exposure to 2.5 mM H O for 3 h
2 2 b
e
r

121 increased membrane permeability in less than 10 % of cells and induced GFP in 3 % of 1
9
,

2
122 the cells (Fig. 3C). Of note, treatment of E. coli O104:H4 Δstx2:gfp:ampr with H O for 3 0
2 2 1
8

b
123 h reduced viable cell counts by more than 6 log (CFU/mL) without increasing membrane y

g
u
e
124 permeability.
s
t
125 Effect of acids on membrane permeability and GFP expression in E. coli O104:H4
126 Δstx2:gfp:ampr. The effect of HCl and lactic acid on GFP expression and membrane
127 permeability in E. coli O104:H4 Δstx2:gfp:ampr is shown in Fig. 4. HCl induced GFP in a
7
128 higher proportion of cells (12 %) when compared to lactic acid, which induced GFP in 6 %
129 of cells. During treatment with HCl, a GFP positive and PI negative population appeared
130 after 20 – 40 min of treatment but decreased from 12 to 0.5 % during subsequent
D
131 incubation. Correspondingly, a GFP and PI positive population increased from 2 to 17 % o
w
n
132 between 40 and 180 min of incubation (Fig. 4A). A similar trend was observed for lactic lo
a
d
e
133 acid treatment, where the GFP positive and PI negative population dropped to less than 1 % d

f
r
o
m
134 after 40 min treatment, while the GFP and PI positive population increased to 12.7 % (Fig.

h
t
t
p
135 4B). These data indicate that GFP was expressed first but GFP expressing cells lost :
/
/
a
e
m
136 membrane integrity and likely viability during subsequent incubation at acid conditions.
.
a
s
m
137 Effect of pressure on membrane permeability and GFP expression in E. coli
.
o
r
g
138 O104:H4 Δstx2:gfp:ampr. The effect of pressure on GFP expression and membrane /
o
n

N
139 permeability in E. coli O104:H4 Δstx2:gfp:ampr is shown in Fig. 5. Control cultures
o
v
e
m
140 were heat-sealed in sample tubing and stored at ambient pressure and temperature; here,
b
e
r

141 the membrane integrity was compromised in around 10 % of the population even without 1
9
,

2
142 pressure treatment (Fig. 5A). After treatment with 200 MPa for 7 min, 2 % of the 0
1
8

b
143 population was positive for GFP and PI. During recovery at ambient pressure and 37 °C y

g
u
e
144 for 3 h, the proportion of cells with damaged membranes decreased owing to membrane
s
t
145 repair, or to growth of surviving cells, and GFP positive cells were not detected.
8
146 Effect of acids and H O on phage production in wild type E. coli O104:H4 and E.
2 2
147 coli O104:H4 Δstx2:gfp:ampr. To determine whether induction of gfp and tsx2 also leads
148 to release of infectious phage particles, spot on lawn assays were carried out after
D
149 induction with diverse stressors (Table 3). Filtrates isolated from control cultures and o
w
n
150 acid-stressed cultures of the mutant and wild type strains did not contain infectious phage lo
a
d
e
151 particles. In contrast, phages isolated from wild type and mutant strain after mitomycin C d

f
r
o
m
152 and H O treatments were lysed E. coli DH5α, indicating the presence of infectious
2 2
h
t
t
p
153 phages. :
/
/
a
e
m
.
154 Cell morphology of stress treated cells. To determine whether GFP expression in E. a
s
m
.
155 coli O104:H4Δstx2:gfp:amp is linked to a SOS response, the morphology of untreated o
r
g
/

o
156 and stressed cells were compared by microscopic observation. The SOS response inhibits n

N
o
157 cell division and causes filamentation (22, 23). No morphological changes were observed v
e
m
b
158 after treatment with HCl (pH 2.5), lactic acid (pH 3.5) and HHP (200 MPa) (Fig. 6). e
r

1
9
159 However, after the treatment with mitomycin C and H O , some of the cells formed ,
2 2
2
0
1
160 filaments (Fig. 6E & F). 8

b
y

g
u
161 Effect of environmental stress on recA expression. To further confirm the link between e
s
t
162 GFP expression and the SOS response in response to stress, we also evaluated the
163 expression of recA (Fig.7). Mitomycin C and H O induced expression of recA; however,
2 2
164 acid treatment did not affect recA expression. These results indicate that mitomycin C
9
165 and H O activated the SOS response, which corresponds to the microscopic observation
2 2
166 of cell filamentation as indicator of an induction of the SOS response (Fig 6 and 7).
167 However, stressors inducing recA did not overlap with conditions inducing GFP.
D
o
w
168 DISCUSSION
n
lo
a
169 The induction of the Stx prophage regulates toxin and phage production, which are d
e
d

f
170 integral to virulence of STEC (5). We developed a method for simultaneous detection of r
o
m

h
171 the expression of Stx2 prophages, changes of membrane permeability, and the induction t
t
p
:
/
/
172 of the SOS response. This novel tool was used to assess the role of environmental and a
e
m
.
173 food-related stressors on the induction of Stx2-prophage and the SOS response. a
s
m
.
174 Classical techniques for the evaluation prophage induction employ plaque assays (21, 24), o
r
g
/

o
175 protein assays (6, 10) and quantification of gene expression (25); however, these methods n

N
o
176 inform only on prophage induction at a population level. Using a fluorescent reporter v
e
m
b
177 gene to replace stx2 was done in previous studies to investigate the induction of Stx2 e
r

1
9
178 prophages (6, 26). We generated a single-cell detection method using flow cytometry to ,

2
0
1
179 assess the induction of Stx2-prophage and the SOS response by quantification of GFP 8

b
y

180 and determination of cell morphology (27). We also combined the fluorescence probe, PI, g
u
e
s
181 with flow cytometry to measure membrane permeability. Treatment with H O reduced t
2 2
182 cell counts without disrupting membrane permeability. The link of prophage induction to
183 the SOS response was further confirmed by quantification of recA expression. Mitomycin
10