Subscriber access provided by UNIV OF ALABAMA BIRMINGHAM
Remediation and Control Technologies
Bacteriophages To Sensitize A Pathogenic New Delhi Metallo #-lactamase-positive Escherichia coli To Solar Disinfection Nada Al-Jassim, David Mantilla-Calderon, Giantommaso Scarascia, and Peiying Hong Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.8b04501 • Publication Date (Web): 20 Nov 2018 Downloaded from http://pubs.acs.org on November 21, 2018
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 30
Environmental Science & Technology
Inactivation curve with bacteriophages
Log10 E. coli
Log10 E. coli
Inactivation curve without bacteriophages
time
time Pathogenic & antibiotic- resistant E. coli
Pathogenic & antibioticresistant E. coli + Bacteriophages
ACS Paragon Plus Environment
Environmental Science & Technology
Log10 reduction in CFU numbers Log10(C/C0)
1
0
50
Page 2 of 30
Fluence (J/cm2) 150 200
100
250
300
350
1.00
0
0.00
-1
-1.00 R² = 0.9733
-2
-2.00
-3
-3.00
R² = 0.9745
-4
-4.00
-5
-5.00
-6
-6.00
0
2
4
6 Time (h)
Figure 1.
ACS Paragon Plus Environment
8
10
12
Page 3 of 30
Environmental Science & Technology
1 0
Log10 reduction in PFU numbers Log10(C/C0)
-1 -2 -3 -4 R² = 0.9509 -5 -6
0
2
4
6
8
Time (h) Dark control
Figure 2.
Irradiated
ACS Paragon Plus Environment
10
12
Environmental Science & Technology
Page 4 of 30
1
Bacteriophages to Sensitize a Pathogenic New Delhi Metallo β-lactamase-positive
2
Escherichia coli to Solar Disinfection
3
Nada Al-Jassim 1, David Mantilla-Calderon 1, Giantommaso Scarascia 1, Pei-Ying Hong * 1
4 5 6 7
1. King Abdullah University of Science and Technology (KAUST), Water Desalination and Reuse Center (WDRC), Biological and Environmental Sciences & Engineering Division (BESE), Thuwal, 23955-6900, Saudi Arabia
8 9 10 11 12 13 14 15 16 17 18 19 20 21 22
* Corresponding author: Pei-Ying Hong Email:
[email protected] Phone: +966-12-8082218
23 24 25
Keywords: Ultraviolet disinfection; decay kinetics; antibiotic-resistant bacteria; transcriptomics
26 27 28 29
Running title: Bacteriophage-supplemented disinfection of E. coli
1 ACS Paragon Plus Environment
Page 5 of 30
Environmental Science & Technology
30
Abstract
31
Bacteriophages active against a New Delhi metallo beta lactamase (NDM)-positive E. coli PI-7
32
were isolated from municipal wastewater and tested for their lytic effect against the bacterial
33
host. Bacteriophages were highly specific to E. coli PI-7 when tested for host-range. After
34
determining host-specificity, bacteriophages were tested for their ability to sensitize E. coli PI-
35
7 to solar irradiation. Solar irradiation coupled with bacteriophages successfully reduced the
36
length of the lag-phase for E. coli PI-7 from 4.11 ± 0.78 h to 1.82 ± 0.60 h in buffer solution.
37
The reduction of lag-phase length was also observed in filtered wastewater effluent and
38
chlorinated effluent. Previously, we found through gene expression analysis that cell wall,
39
oxidative stress and DNA repair functions played a large role in protecting E. coli PI-7 against
40
solar damage. Here, gene expression analysis of bacteriophage-supplemented solar-irradiated
41
E. coli PI-7 revealed downregulation of cell wall functions. Downregulation of functions
42
implicated in scavenging and detoxifying reactive oxygen species, as well as DNA repair genes
43
was also observed in bacteriophage-supplemented solar-irradiated E. coli PI-7. Moreover, solar
44
irradiation activates recA which can induce lytic activity of bacteriophages. Overall, the
45
combined treatment led to gene responses that appeared to make E. coli PI-7 more susceptible
46
to solar disinfection and bacteriophage infection. Our findings suggest that bacteriophages
47
show good potential to be used as a biocontrol tool to complement solar irradiation in
48
mitigating the persistence of antibiotic-resistant bacteria in reuse waters.
49
2 ACS Paragon Plus Environment
Environmental Science & Technology
50
Page 6 of 30
1. Introduction
51
Antibiotic-resistant bacteria (ARB) are among emerging microbial contaminants of
52
increasing concern, and are challenging our current antimicrobial strategies. In fact, several
53
reports have raised concerns that the development of new drugs alone will not adequately
54
address the threat posed by the spread of antibiotic-resistant pathogens1-3. A major hotspot
55
implicated in the creation of superbugs, and the dissemination of these emerging contaminants
56
is wastewater and its treatment facilities as abundance of nutrients, high bacterial numbers and
57
sub-lethal levels of antibiotics are considered favorable for both the survival ARB and the
58
horizontal transfer of bacterial resistance genes4, 5.
59
In particular, an earlier study has isolated an E. coli strain PI-7 from local wastewater
60
influent in Jeddah, Saudi Arabia, and found it to be pathogenic, highly antibiotic-resistant and,
61
more alarmingly, carrying the New Delhi metallo-β-lactamase (NDM) antibiotic resistance
62
gene, blaNDM6. Other studies have also independently reported the presence of blaNDM genes
63
and NDM-harboring bacteria in urban wastewater6-8, and in the final chlorinated secondary
64
effluent despite having received treatment through a municipal wastewater treatment process7.
65
The blaNDM gene codes for a carbapenemase, a class of enzymes that confer resistance to
66
carbapenems and other β-lactam antibiotics that are normally reserved as a last line of defense
67
against gram-negative bacterial infections. Moreover, NDM-harboring bacteria typically carry
68
resistance determinants against multiple other antibiotic classes, leaving very few treatment
69
options available 8-11. These findings highlight the risks of disseminating ARB and their genes
70
through wastewater reuse12, 13, with potential implications to public health. This is particularly
71
relevant in places like Saudi Arabia, where pressure on water resources is high, and treated
72
wastewater is being considered to alleviate considerable portions of this water-scarcity pressure
73
14.
74
Solar irradiation can serve as a biocidal barrier to mitigate this problem. Many studies
75
found rapid inactivation of fecal indicator organisms within a few hours of exposure to natural
76
sunlight15-19. However, it was found that the NDM-positive E. coli PI-7 underwent a
77
significantly longer lag-phase prior to decay, and exhibited a longer inactivation half-life
78
compared to another non-pathogenic and less antibiotic-resistant E. coli DSM1103 upon solar
79
irradiation20. This suggests that a higher solar fluence or longer exposure time would be
80
required to achieve the same log reduction for the NDM-positive E. coli compared to a non-
81
resistant strain. Furthermore, E. coli PI-7 was observed to upregulate a larger suite of genes
82
related to cell repair, including upregulation of cell wall synthesis and extracellular polymeric 3 ACS Paragon Plus Environment
Page 7 of 30
Environmental Science & Technology
83
substance genes that may have served as a physical barrier to solar damage, upregulation of
84
oxidative stress response and ROS (reactive oxygen species) scavenger genes which may have
85
protected the cell from indirect mechanisms of solar damage, as well as upregulation of DNA
86
synthesis and repair genes that may have enabled longer survival. Moreover, the same study
87
found upregulation of genes involved in persister cell formation, along with genes related to
88
antibiotic resistance and virulence functions. These observations suggest that pathogenic
89
antibiotic-resistant bacteria like E. coli PI-7 remain viable for longer period compared to an
90
antibiotic-susceptible strain despite solar irradiation. There exists a need for an additional
91
mitigation strategy besides solar irradiation to enhance biocidal effects against the E. coli PI-
92
7.
93
A complementary treatment method to explore would be the use of lytic
94
bacteriophages. Lytic bacteriophages are viruses that infect the bacterium host, replicate within
95
and kill bacteria by lysing the host. In wastewater related applications, bacteriophages have
96
been applied to control biofilm formation on filtration membranes and to restore flux to fouled
97
membranes1, 21. Bacteriophage therapy has also been demonstrated to solve the problem of
98
sludge bulking caused by filamentous bacteria22-24. However, those studies did not demonstrate
99
the complementary use of bacteriophages along with solar irradiation to inactivate ARB like
100
E. coli PI-7.
101
It is hypothesized that bacteriophages can aid in achieving further reduction of E. coli
102
PI-7 numbers during disinfection by solar irradiation. This is inferred from the earlier study
103
where downregulation of phage-shock functions as well as upregulation of phage integrase and
104
DNA transfer genes were observed in E. coli PI-7 in response to solar irradiation20. The aim of
105
this study is to investigate the use of bacteriophages as a biocidal tool in combination with solar
106
irradiation for inactivating NDM-carrying E. coli PI-7. In this study, seven bacteriophages
107
targeting E. coli PI-7 were isolated from wastewater. They were then tested for host-specificity
108
and stability in controlling E. coli PI-7’s growth under various conditions. Three bacteriophage
109
candidates were used in combination with solar irradiation to test the possibility of hastening
110
E. coli PI-7’s inactivation. The response was assessed in terms cell survival via quantification
111
of culturable cell counts, and gene expression changes to reveal survival and damage responses
112
at the molecular level.
113 114 115 4 ACS Paragon Plus Environment
Environmental Science & Technology
116
2. Materials and Methods
117
2.1. Bacteriophage isolation
Page 8 of 30
118
Bacteriophages specific to a NDM-positive pathogenic E. coli strain PI-76 were isolated
119
from wastewater as follows. 50 ml of wastewater influent was centrifuged at 10,000 x g for 15
120
min, and the supernatant was filtered through a 0.2 µm filter. Then 50 ml of 2X LB broth was
121
added to the filtrate, as well as 50 ml of overnight PI-7 culture with 8 µg/ml meropenem. The
122
mixture was incubated at 37 °C and 200 rpm for 24 h. Chloroform was added in to a final
123
concentration of 1% and the mix was incubated for 1 h to lyse bacterial cells. Next, the mixture
124
was centrifuged at 10,000 x g for 30 min at 4 °C, and the supernatant was filtered through a
125
0.2 µm filter. The phage lysate filtrate was diluted in SM buffer (100 mM NaCl, 8 mM
126
MgSO4•7H2O, 50 mM Tris buffer at pH 7.5). To isolate individual phages, 10 µl of phage
127
lysate and 100 µL of overnight PI-7 culture were mixed together in soft LB agar supplemented
128
with 2 mM CaCl2, then poured over CaCl2-supplemented LB (1.5 % w/v) agar. After 24 h
129
incubation at 37 °C, isolated plaques were individually picked, re-suspended in SM buffer and
130
filtered through a 0.2 µm filter. Each plaque then went through seven rounds of plating and re-
131
isolation to isolate a total of seven individual bacteriophages.
132
2.2. High-titer bacteriophage stock preparation
133
Bacteriophage stocks containing high phage titer were prepared by first mixing 50 µl
134
of original phage stock with 100 µl of fresh E. coli PI-7 in 3 ml of soft LB agar (0.3%)
135
supplemented with 2 mM CaCl2. The mixture was plated onto 1.5% w/v LB agar plates
136
supplemented with 2 mM CaCl2, and incubated overnight at 37 °C. Then, 3 ml of SM buffer
137
was added onto each plate, and a sterile spreader was used to scrape off the top agar layer
138
containing bacteriophages and bacterial cells. This mixture was then centrifuged at 10,000 g
139
for 15 min, and the supernatant was filtered through 0.2 µm filters (GE Healthcare Life
140
Sciences, Little Chalfont, Buckinghamshire, UK). Afterwards, polyethylene glycol 8000 was
141
added to the filtrate to form 15% w/v, and the mixture was incubated overnight with stirring at
142
4 °C. Subsequently, the mixture was centrifuged at 10,000 g for 20 mins, and the cell pellet
143
containing the bacteriophage particles was resuspended in 1 ml of 1X PBS. Stocks were
144
enumerated for titer concentration by plating serial dilutions of the stock with fresh E. coli PI-7
145
using the soft agar method described above, and quantifying PFUs (or plaque-forming units)
146
after overnight incubation. Bacteriophages were characterized for their morphology and
147
genome sizes as described in Supplementary Information 1. 5 ACS Paragon Plus Environment
Page 9 of 30
148
Environmental Science & Technology
2.3. Bacteriophage lytic activity
149
To check the lytic activity of the bacteriophage isolates against different bacteria,
150
cultures of various bacteria from clinical and environmental origins (Table 1) were prepared
151
overnight. 100 µl of bacteria with were plated in soft agar, mixed with the bacteriophages to
152
perform a plaque assay, or followed by spotting of serial dilutions of different bacteriophages
153
for spot test evaluation25.
154
Based on the lytic activity tests, bacteriophages were specific against E. coli PI-7 and
155
further determination was done to monitor the impact on growth curve of E. coli PI-7 in
156
presence of bacteriophages. A fresh culture of E. coli PI-7 was prepared overnight, then 100 µl
157
was inoculated into 10 ml of 1:1 SM buffer mixed with LB broth in sterile transparent glass
158
tubes. The OD600 of each tube was periodically measured to monitor E. coli PI-7’s growth in
159
presence or absence of bacteriophages at different temperatures (40, 37, 32 and 20 °C) and pH
160
(6, 7 and 8).
161
2.4. Solar inactivation and decay kinetics
162
Solar inactivation trials were conducted using an Atlas Suntest ® XLS+ photosimulator
163
(Chicago, IL) equipped with a xenon arc lamp as described in our previous work20. The solar
164
simulator’s irradiance is described in Supplementary Information 2 and shown in Figure S1.
165
Briefly, prior to the experiments, overnight cultures of E. coli PI-7 were thoroughly washed in
166
1 mM NaHCO3, then re-suspended in 80 ml of the final matrix of either 1X PBS, filtered
167
tertiary-treated wastewater effluent (collected after treatment in an anoxic-oxic activated
168
sludge tank followed by microfiltration) or filtered chlorinated effluent collected after
169
microfiltration. All wastewater samples were filtered through 0.2 µm polycarbonate membrane
170
filters (GE Healthcare Life Sciences, Little Chalfont, Buckinghamshire, UK) and stored at 4°C
171
in sterile glass bottles within 4 h of collection. No additional step was taken to quench residual
172
chlorine in chlorinated effluent samples prior to inactivation experiment as no residual chlorine
173
were detected by DPD method (Hach, Loveland, CO). Starting bacterial numbers for solar
174
inactivation experiments were always approximately 1-9x108 CFU/ml. For bacteriophage-
175
supplemented experiments (SI+phage), 1 ml of high titer phage solution was added to the
176
microcosms to a final PFU concentration similar to the bacterial concentration (i.e., MOI = 1).
177
Microcosms were then either covered with foil (dark controls) or glass filters permitting
178
wavelength 280 nm and above (Newport Corporation, Irvine, CA). A set of non-bacteriophage-
179
supplemented experiments (SI-only), both dark and irradiated were always included in parallel 6 ACS Paragon Plus Environment
Environmental Science & Technology
Page 10 of 30
180
to bacteriophage-supplemented solar irradiation (SI+phage) trials, but SI-only microcosms
181
were inoculated with sterile buffer solution instead of phage solution. After inoculation with
182
bacteria and bacteriophages where applicable, the microcosms were placed inside the Atlas
183
Suntest and exposed to solar irradiation. SI+phage and SI-only inactivation kinetics data
184
presented in this study was obtained from four biological replicates.
185
A set of solar inactivation experiments with seven bacteriophages individually
186
inoculated into buffer solution in separate microcosms were also performed without replication
187
in order to denote the stability of these bacteriophages upon exposure to sunlight. This
188
observation, along with that obtained from lytic activity tests and growth curve impediment
189
experiments (described in section 2.3), were used to select for the most suitable bacteriophage
190
isolates to form a bacteriophage mixture. Upon selection of the bacteriophage mix (i.e.,
191
bacteriophage P1, P4 and P5a), solar inactivation was performed on this mixture inoculated in
192
sterile buffer solution and in presence of E. coli PI-7. Three biological replicates were
193
performed for this experiment to determine the bacteriophage mixture’s decay kinetics.
194
The number of colony forming units (CFU/ml) and/or plaque forming unites (PFU/ml)
195
in the irradiated and dark experimental microcosms was quantified at the starting point and
196
over irradiation time. Inactivation results were converted to log10 and natural log (ln) curves of
197
concentration values normalized as CFUt/CFUt=0 for E. coli PI-7 or PFU/PFUt=0 for
198
bacteriophages, and plotted against time to obtain the inactivation curves. The decay rate
199
constants (k) of dark control and irradiated sample inactivation curves were calculated as the
200
negative slopes of natural log inactivation curves, with the following caveat: in cases where
201
there was a detectable lag-phase, the curve was considered bi-modal and the lag-phase data
202
points were excluded from the slope calculation. Further detail on lag-phase can be found in
203
Supplementary Information 3 and Figure S2. The decay rate constants were mathematically
204
corrected for attenuation of light penetration caused by light screening (due to variable light
205
absorbance) but not light scattering (due to turbidity) in the samples. The corrected decay rate
206
constants were then used in statistical comparisons or half-life calculations. Because
207
differences in light absorbance and screening can also occur due to sample to sample variation
208
or different water matrices (Figure S3), this adjustment was hence performed for each
209
experimental microcosm by measuring each individual sample’s absorbance at the start of the
210
experiment as described previously,
211
inactivation by full spectrum irradiation (280-700 nm) as is shown in this study, correction
212
factor for the decay rate constants was based on the concept that different wavelengths differ
26, 27
and shown in Supplementary Information 2. For
7 ACS Paragon Plus Environment
Page 11 of 30
Environmental Science & Technology
213
in their inactivation efficiencies. UV portions of the spectrum are generally the most important
214
wavelengths that are most lethal to the microorganism compared to the visible range
215
wavelengths. Mathematical calculations of correction factors therefore also weigh in the
216
difference in contribution of UV and visible range to the decay27. The half-lives for each
217
experiment, or the durations needed to reduce the bacterial or bacteriophage concentration by
218
half (after and excluding lag-phase), were calculated using the first-order decay kinetics
219
equation:
220
ln(CFUt/CFU0) = -k*t
221
where k* is the decay rate constant, corrected for light screening, of the natural log inactivation
222
curve, and t is time.
223
Statistical analyses was performed using Minitab version 1.4.0. One-way ANOVA was
224
performed to compare the lengths of lag-phases, the corrected decay rate constants and the
225
half-life values of the different experimental conditions and isolates. Linear regression analysis
226
was performed to compare the slope of each experiments inactivation curve to 0 (α = 0.05) to
227
determine if significant change/decay is observed (Supplementary Information 3). Errors
228
presented at various points in the text and tables, and as error bars in some charts, are standard
229
deviation from the mean values reported. No error was propagated on C0 in (C/C0)-based plots.
230
This is because each sample had its own C0. After samples were enumerated for C, Log10(C/C0)
231
or Ln(C/C0) values were generated for each microcosm. Afterwards the average Log10(C/C0)
232
or Ln(C/C0) – including all technical and biological replicates - for one experimental condition
233
was calculated, and then the standard deviation error for that average was calculated. Lag-
234
phase length and half-life were not analyzed for dark control samples due to the lack of
235
significant decay in these samples, as determined by single linear regression analysis.
236
2.5. RNA-seq sampling, extraction and sequencing
237
RNA-seq was performed to provide comparative analysis of gene expression between
238
samples with and without bacteriophage addition upon solar irradiation. Samples were taken
239
at various time points (at 0, 1, 2, 4, 6, 8 and 10 h) to capture the gene expression at different
240
phases of the bacterial CFU inactivation curves. For bacteriophage-supplemented experiments
241
(SI+phage), three solar inactivation experiments were conducted for RNA sampling to obtain
242
three biological replicates, each with two dark and two irradiated technical replicate
243
microcosms at every sampling point. For non-bacteriophage-supplemented (SI-only)
8 ACS Paragon Plus Environment
Environmental Science & Technology
Page 12 of 30
244
experiments, two solar inactivation experiments for RNA sampling were performed, also with
245
two dark and two irradiated technical replicate microcosms at every sampling point.
246
RNA preservation, extraction and sequencing were performed as described
247
previously20. Briefly, sample volumes varying between 1 to 4 mL were taken from the
248
experimental microcosms, and an equal volume of RNAProtect reagent (Qiagen, Valencia,
249
CA) was added in. Samples were stored at 4 °C for less than a week until RNA extraction.
250
Samples were centrifuged at 5,000 g for 10 mins, and the pellets used for total RNA extraction
251
using the RNeasy Mini kit (Qiagen, Hilden, Germany) with the on-column DNase treatment.
252
The RNA was eluted with RNase-free water and RNA quality was determined using 2200
253
Tapestation Bioanalyzer (Agilent Technologies, Santa Clara, CA). All samples that passed the
254
quality control were submitted to KAUST Genomics Core lab for RNA-Seq on Illumina HiSeq
255
platform. Prior to sequencing, the bacterial ribosomal RNA was removed from 1 µg of total
256
RNA using the Ribo Zero rRNA removal kit (Illumina, San Diego, CA, USA). The enriched
257
mRNA fraction was converted to RNA-seq libraries compatible for runs on Illumina HiSeq
258
platforms based on protocols described by manufacturer. Fastq files were generated and
259
demultiplexed by KAUST Bioinformatics core lab. All fastq files associated with E. coli PI-7
260
with and without bacteriophages are available on the ENA SRA public depository under
261
accession numbers PRJEB25911 and 25812.
262 263
2.6. RNA-seq data analysis
264
The RNA-seq data was analyzed using CLC Genomics Workbench version 8.0.1 from CLC
265
Bio (Cambridge, MA). For the reference genome sequence, and to assign genes to specific
266
categories, genomic DNA sequences annotated using RAST were used for E. coli PI-7 6. RNA-
267
seq reads were mapped onto the reference sequences using CLC. Reads were only assembled
268
if the fraction of reads that aligned to the reference genome was greater than 0.9 and if the read
269
matched other regions of the reference genome at fewer than 10 nucleotide positions.
270
Summarized RNA-seq mapping results are shown in Tables S1 and S2 and Supplementary
271
Information 4.
272
CLC was used to generate normalized mean expression values by applying a scaling
273
correction to samples to facilitate cross-sample comparisons within and across biological
274
batches. The gene expression results were then analyzed for statistical significance, separately
275
for each of the biological experimental replicate batches. The statistical comparison was based 9 ACS Paragon Plus Environment
Page 13 of 30
Environmental Science & Technology
276
on the proportion-based test described by Baggerly et al. 28. Mean gene expression values as
277
normalized RPKM (reads per kilobase million) in irradiated samples were compared with the
278
mean normalized RPKM values of their respective dark controls at each time point for each
279
experimental batch. Then, for each condition, the fold-change in expression level was taken
280
into consideration if it was statistically significant (p-value