Suppression of Enteric Bacteria by Bacteriophages: Importance ... - CDN

Subscriber access provided by ORTA DOGU TEKNIK UNIVERSITESI KUTUPHANESI

Article

Suppression of Enteric Bacteria by Bacteriophages: Importance of Phage Polyvalence in the Presence of Soil Bacteria Pingfeng Yu, Jacques Mathieu, Yu Yang, and Pedro J. J. Alvarez Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.7b00529 • Publication Date (Web): 17 Apr 2017 Downloaded from http://pubs.acs.org on April 18, 2017

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 free 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 accessible to all readers and 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.

Environmental Science & Technology 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 31

Environmental Science & Technology

Suppression of Enteric Bacteria by Bacteriophages: Importance of Phage Polyvalence in the Presence of Soil Bacteria

Pingfeng Yu, Jacques Mathieu, Yu Yang and Pedro J.J. Alvarez* Department of Civil and Environmental Engineering, Rice University, Houston, TX 77005

RUNNING TITLE: Growth Suppression by polyvalent phages and competitive exclusion *

Corresponding Author: Phone: 713-348-5903. Fax: 713-348-5203. E-mail: alvarez@ rice.edu.

ACS Paragon Plus Environment


Page 2 of 31

1

ABSTRACT

2

Bacteriophages are widely recognized for their importance in microbial ecology and bacterial

3

control. However, little is known about how phage polyvalence (i.e., broad host range) affects

4

bacterial suppression and interspecies competition in environments harboring enteric pathogens

5

and soil bacteria. Here we compare the efficacy of polyvalent phage PEf1 versus coliphage T4 in

6

suppressing a model enteric bacterium (E. coli K-12) in mixtures with soil bacteria

7

(Pseudomonas putida F1 and Bacillus subtilis 168). Although T4 was more effective than PEf1

8

in infecting E. coli K-12 in pure cultures, PEf1 was 20-fold more effective in suppressing E. coli

9

under simulated multispecies biofilm conditions because polyvalence enhanced PEf1

10

propagation in P. putida. In contrast, soil bacteria do not propagate coliphages and hindered T4

11

diffusion through the biofilm. Similar tests were also conducted under planktonic conditions to

12

discern how interspecies competition contributes to E. coli suppression without the confounding

13

effects of restricted phage diffusion. Significant synergistic suppression was observed by the

14

combined effects of phages plus competing bacteria. T4 was slightly more effective in

15

suppressing E. coli in these planktonic mixed cultures, even though PEf1 reached higher

16

concentrations by reproducing also in P. putida (7.2±0.4 vs. 6.0±1.0 log10PFU/mL). Apparently,

17

enhanced suppression by higher PEf1 propagation was offset by P. putida lysis, which decreased

18

stress from interspecies competition relative to incubations with T4. In similar planktonic tests

19

with more competing soil bacteria species, P. putida lysis was less critical in mitigating

20

interspecies competition and PEf1 eliminated E. coli faster than T4 (36 vs. 42 h). Overall, this

21

study shows that polyvalent phages can propagate in soil bacteria and significantly enhance

22

suppression of co-occurring enteric species.

23

1 ACS Paragon Plus Environment

Page 3 of 31

24 25


INTRODUCTION Bacteriophages, which are viruses that infect and replicate within bacteria, are the most

26

abundant and diverse living entities in the biosphere.1 Estimates place their numbers in the range

27

of 1031, with a common environmental virus-to-prokaryote ratio of about 10:1.2, 3 As antibacterial

28

agents and important vectors for horizontal gene transfer4, phages are significant drivers of

29

bacterial evolution and community structure. Consequently, phages can also significantly

30

influence biogeochemical cycles and energy flows in various ecosystems.5 Given their ecological

31

importance and potential applications for controlling problematic bacteria6, there is a need to

32

advance fundamental understanding of phage-host interactions in complex microbial

33

communities. This includes discerning how phage host range affects their propagation dynamics

34

and microbial community structure in both planktonic (i.e., suspended) and biofilm systems.

35

Phages alter their hosts’ metabolic repertoire, fitness and competitive capability, or directly

36

eliminate the most susceptible bacteria.7, 8 Interspecific bacterial competition in the presence of

37

phages has been studied theoretically8, 9 and empirically10, 11. Depending on the nature of phage-

38

host associations and environmental conditions, phages can facilitate competitive exclusion9, 12

39

or enhance the bacterial hosts’ coexistence11, 13. However, most these studies have only

40

considered phages that typically infect only one species and cannot reproduce by using hosts

41

from other genera in the microbial community. No previous publications have considered

42

how (broad host range) polyvalent phages (i.e., phages capable of intergeneric infection14)

43

affect bacterial suppression and interspecific competition. This is a critical knowledge gap

44

given the recent recognition of the prevalence of polyvalent phages in the environment15-17 and

45

their potential to exhibit more complex propagation dynamics. Specifically, whereas polyvalent

46

phages use a wider variety of bacteria to reproduce, they generally experience lower efficiency



47

of infection (i.e., the ratio between the infectious titer for a given host and the maximum

48

observed titer18) and slower growth17, 19, which may offset their ability to influence microbial

49

community structure.

Page 4 of 31

50

This study addresses how phage host range affects interspecific bacterial competition and

51

enteric bacteria suppression. We compare the effect of coliphage T4 versus polyvalent phage

52

PEf1 on three competing bacteria: (1) Escherichia coli, representing an enteric bacterium

53

commonly associated with fecal pollution and (for some E. coli strains) infectious disease20, (2)

54

Pseudomonas putida (alternative host to PEf1 but not to T4) and (3) non-host Bacillus subtilis,

55

which cannot be infected by these phages. The latter two represent indigenous soil bacteria

56

commonly associated with biodegradation of pollutants.21 E. coli suppression is quantified under

57

both planktonic conditions and model biofilms, which offer potential resistance to phage

58

diffusion and to propagation of phage-resistant mutants. We discern the contributions to E. coli

59

suppression from phages versus competition by soil bacteria (demonstrating synergism), and

60

highlight the importance of polyvalence in enhancing phage propagation in alternative hosts (e.g.,

61

soil bacteria) and significantly boosting suppression of potential enteric pathogens.

62

MATERIALS AND METHODS

63

Bacteria, bacteriophage and culture conditions. The bacterial strains in this study included

64

E. coli K-12 (ATCC 700926), P. putida F1 (ATCC 700007), and B. subtilis 168 (ATCC 23857).

65

Culturing was conducted in M63 glucose medium at 30°C [12 g of KH2PO4, 28 g of K2HPO4, 8

66

g of (NH4)2SO4 per liter water supplemented with 1 mM of MgSO4, 0.2% glucose and 0.5%

67

Casamino Acids].22 Coliphage T4 (Carolina Biological 12-4330), which was reported to infect

68

only E. coli and closely related Shigella species23, infected E. coli K-12 but none of other tested

69

bacterial strains. The polyvalent phage PEf1, which can infect both E. coli and P. putida (Fig.


Page 5 of 31


70

S1) but not B. subtilis, was previously isolated by a sequential multi-host approach.19 Bacteria in

71

exponential phase were used for phage treatment, phage enumeration and phage characterization.

72

Double-layer plaque assays24 used tryptone agar plates containing 0.7% agar for the soft agar and

73

1.1% for the base layer. Phage titration was performed using E. coli K-12 as the host and

74

expressed as plaque forming units (PFU) per ml. The samples were centrifuged to obtain

75

supernatant for extracellular phage titration. Phages were enumerated after direct dilution using

76

the double-layer plaque assay. Quantitative phage abundance analysis was performed by

77

observation of virus-like-particles (VLP) via fluorescence microscopy after staining with SYBR

78

Green I according to the standard protocol.25

79

Bacteriophage growth parameters characterization. The fraction of free phages was

80

measured as function of time in adsorption tests, and phage adsorption rate constants were

81

determined based on the first-order kinetics: ݇‫(= ܤ‬ln(ܲ0) − ln (ܲ‫))ݐ‬/‫ݐ‬, where k is the adsorption

82

rate constant (mL/min), B is bacterial density (CFU/mL), P0 is number of free phages at time 0

83

(PFU/mL), and Pt is number of free phages at time t (PFU/mL).26 One-step growth curve

84

experiments were conducted to measure the burst size and latent time of each phage.27 Efficiency

85

of plating (EOP) was quantified by calculating the ratio of phage plaque titers obtained with

86

a given host to the phage plaque titers obtained with a reference strain18 (i.e., E. coli K-12 in

87

this case). All parameters were determined with bacteria in exponential phase in M63 medium at

88

30 °C (Table S1).

89

Bacterial challenge test in planktonic state. To compare the inhibitory efficiency of phage

90

T4 and PEf1 in planktonic states, 10 ml 105 CFU/ml E. coli K-12 were subjected to the following

91

treatments: (i) 106 PFU/ml phage T4, (ii) 106 PFU/ml phage PEf1, (iii) soil bacteria (1 P. putida

92

and 1 B. subtilis per E. coli), (iv) soil bacteria phage mixture I (1 P. putida, 1 B. subtilis and 10



Page 6 of 31

93

T4 per E. coli), and (v) soil bacteria phage mixture II (1 P. putida, 1 B. subtilis and 10 PEf1 per

94

E. coli). During the 72-h experiment, to avoid nutrient depletion and the accumulation of phage

95

particles (which confounds analysis), samples were collected at 12h intervals and 0.1 ml culture

96

were transferred into 9.9 ml fresh medium. The samples were centrifuged to obtain supernatant

97

for extracellular phage titration and cell pellets for bacterial DNA extraction. Since cultures were

98

diluted 100-fold each step, bacteria were considered completely eliminated when densities were

99

below 100 CFU/ml.

100

To avoid interference by dead cells, ethidium bromide monoazide (EMA) was used to bind

101

the DNA from dead cells 20. Briefly, cell suspensions were stained with 20 µg/ml EMA (Sigma-

102

Aldrich) in the dark for 5 min, placed on ice for 1 min, and exposed to bright visible light for 10

103

min prior to DNA extraction. Bacterial genomic DNA was then extracted using an UltraClean

104

Microbial DNA Isolation Kit (MoBio, Carlsbad, CA) according to the manufacturer’s

105

instructions.

106

Challenge tests with bacteria embedded in soft agar. The two-species mixture (105 CFU

107

E. coli K-12 and B. subtilis 168, or 105 CFU E. coli K-12 and P. putida F1) and three-species

108

mixture (105 CFU E. coli K-12, B. subtilis 168 and P. putida F1) were inoculated with coliphage

109

T4 or polyvalent phage PEf1 (105 PFU) in the soft agar of double-layer plates. The plates were

110

incubated for 12h at 30°C and then three disk-shaped samples (2cm in diameter) were randomly

111

taken from the soft agar of each plate. The samples were physically ground and bacterial DNA

112

was extracted using a PowerSoil DNA Isolation Kit (MoBio, Carlsbad, CA). Phage λ DNA was

113

used as an internal standard to calculate the recovery rate of all DNA extractions 28.

114 115

Isolation and characterization of BIMs. Bacteriophage-insensitive mutants (BIMs) resistant to phage PEf1 and T4 were isolated as follows 29. The soft agar of double-layer plates


Page 7 of 31


116

was seeded with 107 CFU hosts and 109 PFU phage (multiplicity of infection, MOI = 100) and

117

then incubated for 24 h at 30°C. Surviving colonies were picked from the confluent lysis assay

118

plate and streaked onto a nutrient agar. Bacterial contamination was excluded by using selective

119

agar for plating (EMB agar for E. coli and Pseudomonas selective agar for P. putida), and

120

verifying purity by colony PCR with specific biomarkers (E. coli-specific universal stress protein

121

gene A30 and P. putida-specific toluene dioxygenase gene todC131). The ratio between the

122

number of surviving colonies and the number of initially inoculated bacteria (107 CFU) was

123

calculated as the BIM frequency.

124

BIMs were streaked once to avoid re-sensitizing and remove potentially co-evolving

125

phages.32 Spot tests were conducted to confirm the phage resistance of new isolated BIMs.18 To

126

determine potential fitness costs associated with phage resistance development, growth kinetics

127

and biofilm formation were measured. Bacterial specific growth rates were determined based on

128

OD600 measurements during the 12-h exponential growth phase (all tested bacteria reached

129

stationary phase after 12 h). The biofilm formed on the surface of 96-well tissue culture plates

130

after a 24h incubation was measured according to a microtiter dish biofilm formation

131

assay.33

132

Inhibition of E. coli K-12 in newly-established biofilm. E. coli K-12, P. putida F1 and B.

133

subtilis 168 (105 CFU/ml for each bacterium) were incubated in glass vials (KimbleTM 20 mL

134

Disposable Scintillation Vials) containing 5 ml of medium and 1.0 g of quartz sand (Fig. S3).

135

The microcosms were shaken at 240 rpm and 30 °C using a Multi-ThermTM shaker (Benchmark

136

Scientific Inc. Edison, NJ). After 24 h incubation, the bulk solution was replaced with 5 ml fresh

137

medium inoculated with bacteria (105 CFU/ml for each strain) and phages (T4 or PEf1 at 106

138

PFU/ml) to adjust the initial MOI in the bulk solution to the same value used in the prior liquid



139

culture experiments in order to facilitate comparison. The time point when phages were added

140

was considered as time zero. No phages were added in the control group. During the 5 day

141

experiment, 4.9 ml of liquid culture was replaced with an equal volume of fresh medium at 24 h

142

intervals.

Page 8 of 31

143

The bulk solution was collected and centrifuged to obtain supernatant for phage titration and

144

cell pellets for bacteria enumeration. The sand was gently rinsed three times with 0.2 ml of PBS.

145

Sand sampled were suspended in 0.2 ml of PBS supplemented with 0.05% (v/v) Tween-20, and

146

gently sonicated at 40 kHz for 20 min in a 4°C bath sonicator (Branson, Danbury, CT) to disrupt

147

the biofilm matrix and disperse the bacterial cells and entrapped phages.34 The 0.2 ml suspension

148

was centrifuged to obtain the supernatant for phage titration and the cell pellet for bacteria

149

quantification.

150

Primer design and real-time quantitative PCR analysis. The E. coli-specific universal

151

stress protein gene A (uspA) and toluene dioxygenase gene todC1 were chosen as specific

152

biomarkers to assess the abundance of E.coli K-12 and P. putida F1, respectively. Both genes

153

were quantified by qPCR using SYBR Green. The bacterial 16S rRNA gene, quantified by

154

Taqman qPCR, was used for the total bacteria enumeration 35. The extracted genomic DNA from

155

E.coli K-12 and P. putida F1 with different bacterial densities were used to establish the standard

156

curve between threshold cycle (Ct) value and log10 CFU for each bacteria individually. Detailed

157

information of SYBR Green qPCR and Taqman qPCR on primers and probes, reaction reagents

158

and temperature programs is included in the supplementary material (Table S2).

159

Scanning electron microscopy of biofilm on sand surface. The washed sand samples were

160

fixed in 4% glutaraldehyde 0.1M phosphate buffer solution (PBS, pH=7.4) overnight at 4°C. The

161

samples were dehydrated in a series of ethanol-water solutions (50, 60, 70, 80, 90, 95, 100%,


Page 9 of 31


162

vol/vol) for 30 min at 4°C. The sand was placed on the carbon tape pasted stub and then sputter-

163

coated with ~7 nm gold film under vacuum using Denton Desk V Sputter System. The

164

microstructure of biofilm on the sand surface was observed using JEOL 6500 scanning electron

165

microscopy. Duplicates were performed for samples from both control and phage treatment

166

microcosms and more than 10 representative images were captured for each sample.

167

Statistical analysis. Student's t-test (two-tailed) was used to determine the significance of the

168

differences between treatments. Differences were considered to be significant at the 95% level (p

169

< 0.05). Synergy quantification of combination treatments was performed using the Bliss

170

independence model 36. Combination indices (CI) were calculated as follows:

171

CI = (ECompetitors + EPhages - ECompetitors × EPhages)/ ECompetitors+Phages

Eqn. (1)

172

where ECompetitors, EPhages and ECompetitors+Phages represent suppression factors for E. coli caused by

173

interspecific competition, phage infection and combination of interspecific competition and

174

phage infection, respectively 37. The Bliss model was selected because the modes of action for

175

bacterial inhibition by interspecies competition and phage infection are independent. Each

176

suppression factor was calculated as the ratio of E. coli density in the presence of a specific

177

stressor (phages or competitors) to the density of unexposed E. coli.

178

RESULTS

179

Coliphage T4 suppressed E. coli monocultures more effectively than polyvalent phage PEf1.

180

E. coli growth in suspended pure culture was inhibited to a greater extent by T4 than by PEf1

181

(Fig. 1A). Relative to uninfected controls, E. coli viability (i.e., the ratio between E. coli

182

abundance in the phage-treated group to that in the phage-free control) after 12-h incubation in

183

the presence of T4 at MOI = 1 (i.e., one phage initially present per E. coli cell) was 12±4%,

184

which is significantly lower (p < 0.05) than 28±4% in the presence of PEf1 at the same MOI.



Page 10 of 31

185

Similar results were observed for biofilm formation, T4 was more effective than PEf1 (10±3%

186

versus 18±4% E. coli viability after 24 h exposure, n = 3, p < 0.05) at inhibiting E. coli growth in

187

microtiter dish biofilm formation assays (Fig. 1B). These results are consistent with the higher

188

adsorption rate constant (2.3×10-9 vs. 0.6×10-9 ml/min), shorter latent time (35 vs. 50 min), and

189

larger burst size (117±8 vs. 99±5 PFU/cell) for T4 than PEf1 (Table S1), which results in faster

190

propagation and higher probability of infecting E. coli in monoculture.

191

E. coli resistance to T4 results in higher fitness costs than resistance to PEf1.

192

Development of resistance is a common bacterial response to phage infection, but a given

193

host may experience different susceptibility and fitness costs for different phages. We assessed

194

the BIM frequency of E. coli exposed to T4 or PEf1 (MOI = 100) and compared the resulting

195

fitness costs associated with phage resistance. The BIM frequency of E. coli with T4 was

196

6.2±1.1×10-7, which is significantly lower p < 0.05) than that of E. coli toward PEf1

197

(3.8±0.5×10-6). Resistance to T4 decreased the (exponential phase) specific growth rate of E. coli

198

by 28.8±3.6% (from 0.52±0.03 to 0.37±0.02 h-1), compared to a 19.3±1.6% decrease (from

199

0.52±0.03 to 0.42±0.02 h-1) associated with resistance to PEf1 (Fig. 2A). Similarly, the 24 h

200

biofilm formation of E. coli resistant to T4 decreased by 22.4±2.1% (from 0.69±0.04 to

201

0.53±0.04 OD595), while that of E. coli resistant to PEf1 decreased by only 14.3±2.9% (from

202

0.69±0.04 to 0.60±0.05 OD595) (Fig. 2B). The higher fitness cost associated with resistance to T4

203

than to PEf1 is consistent with the more effective suppression of E. coli pure cultures by T4.

204

Polyvalent phage PEf1 proliferates to a greater extent and is more effective than T4 at

205

suppressing E. coli in mixed cultures under simulated biofilm and sandy slurry conditions.

206 207

Although PEf1 was less effective than T4 in suppressing E. coli in mono-culture (Fig. 1), it was significantly more effective than T4 under more realistic mixed cultures in both sandy slurry


Page 11 of 31


208

microcosms (Table 1, Fig. 3 and Fig. 4A) and in soft agar microtiter formation assays (Table S3

209

and Fig. 4B). In these mixed cultures, polyvalence enhanced PEf1 propagation in other hosts (i.e.,

210

P. putida F1 serving as alternative host for PEf119) and reached significantly higher

211

concentrations than T4 in the bulk solution (7.88±0.16 vs 6.83±0.17 log10PFU/ml, p < 0.05),

212

which enhanced E. coli suppression.

213

In sandy slurry microcosms, PEf1 could reproduce in both E. coli and P. putida and reached

214

higher (10-fold) concentrations than T4, which could reproduce only in E. coli (Fig. 3 and Fig.

215

S5). Higher PEf1 concentrations in mixed cultures facilitated E. coli infection and suppression

216

(Fig. 3 and Fig. 4A). E. coli concentrations in the bulk liquid stabilized after 3 days at 4.7±0.1

217

log10CFU/ml in the presence of PEf1, which is 1.3 orders of magnitude lower than in

218

microcosms with T4 (Fig. 3). PEf1 also suppressed E. coli to a greater extent and for a longer

219

duration than T4 in the mixed-culture biofilm attached on the sand surface (Fig. 4A). Prior to

220

phage amendment, the initial densities of E. coli in newly formed mixed-species biofilm were

221

5.64±0.15 log10 CFU/mg sand (Fig. 4A). After PEf1 amendment, the attached 5-day E. coli

222

density decreased by 93±2% to 4.51±0.21 log10 CFU/mg sand, while that in microcosms

223

amended with T4 it increased by 44±26% to 5.80±0.23 log10 CFU/mg sand (Fig. 4A). Compared

224

to controls without phage, PEf1 reduced attached E. coli viability by 2.4 orders of magnitude,

225

while T4 achieved only 1.1 order of magnitude suppression. Furthermore, scanning electron

226

microscopy (SEM) of the sand biofilm revealed significantly greater disruption in microcosms

227

treated with PEf1 than those treated with T4 (Fig. 5).

228

Additional experiments were conducted with bacteria embedded in soft agar plates to assess

229

the effect of phages under biofilm-like conditions without confounding effects such as phage

230

reproduction in suspended hosts or recolonization from the bulk solution. Whereas soft agar tests



Page 12 of 31

231

do not represent the complexity of biofilms commonly encountered in natural and engineered

232

systems, they offer a simple system to investigate factors that influence bacterial attachment and

233

biofilm formation in reductionist experiments that exclude such confounding factors.38, 39 When

234

E. coli and B. subtilis were co-cultured without phages, the E. coli density reached 8.57±0.17

235

log10 CFU/cm2 after 12 h (Fig. 4B). E. coli densities decreased dramatically when the bacteria

236

were challenged with either PEf1 (6.60±0.18 log10 CFU/cm2) or T4 (6.77±0.18 log10 CFU/cm2)

237

(Fig. 4B, p

Suppression of Enteric Bacteria by Bacteriophages: Importance ... - CDN

Recommend Documents