Microbial Dynamics in Multi-Chamber Ozone ... - ACS Publications

Subscriber access provided by READING UNIV

Article

Microbial Dynamics in Multi-Chamber Ozone Contactors at a Full-Scale Drinking Water Treatment Plant Nadine Kotlarz, Nicole Rockey, Terese M Olson, SarahJane Haig, Larry Sanford, John J. LiPuma, and Lutgarde Raskin Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.7b04212 • Publication Date (Web): 04 Jan 2018 Downloaded from http://pubs.acs.org on January 4, 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 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 33

Environmental Science & Technology

Microbial Dynamics in Multi-Chamber Ozone Contactors at a Full-Scale Drinking Water Treatment Plant

Nadine Kotlarz†, Nicole Rockey†, Terese M. Olson†, Sarah-Jane Haig†, Larry Sanfordǂ, John J. LiPuma§, Lutgarde Raskin*,†

†

Department of Civil and Environmental Engineering, University of Michigan, Ann Arbor, Michigan

48109, United States ǂ

Drinking Water Treatment Utility, Ann Arbor, Michigan 48103, United States

§

Department of Pediatrics and Communicable Diseases, University of Michigan Medical School, Ann

Arbor, Michigan 48109, United States

*Corresponding author: Lutgarde Raskin Department of Civil and Environmental Engineering University of Michigan 1351 Beal Ave., 107 EWRE Bldg. Ann Arbor, MI 48109-2125 Phone: 734-647-6920 Email: [email protected]

ACS Paragon Plus Environment


Page 2 of 33

1

ABSTRACT

2

Concentrations of viable microbial cells were monitored using culture-based and culture-independent

3

methods across multi-chamber ozone contactors in a full-scale drinking water treatment plant. Membrane-

4

intact and culturable cell concentrations in ozone contactor effluents ranged from 1,200 to 3,750 cells/mL

5

and from 200 to 3,850 colony forming units/mL, respectively. Viable cell concentrations decreased

6

significantly in the first ozone contact chamber, but rose, even as ozone exposure increased, in subsequent

7

chambers. Our results implicate microbial detachment from biofilms on contactor surfaces, and from

8

biomass present within lime softening sediments in a hydraulic dead zone, as a possible reason for

9

increasing cell concentrations in water samples from sequential ozone chambers. Biofilm community

10

structures on baffle walls upstream and downstream from the dead zone were significantly different from

11

each other (p = 0.017). The biofilms downstream of the dead zone contained a significantly (p = 0.036)

12

higher relative abundance of bacteria of the genera Mycobacterium and Legionella than the upstream

13

biofilms. These results have important implications as the effluent from ozone contactors is often treated

14

further in biologically active filters and bacteria in those effluents continuously seed filter microbial

15

communities.

Page 2 of 27 ACS Paragon Plus Environment

Page 3 of 33


16

INTRODUCTION

17

Ozone, a strong oxidant, is widely used for water treatment as it inactivates microbial pathogens,1

18

decomposes natural organic matter (NOM), removes taste-, odor- and color-causing compounds,2,3

19

oxidizes inorganic contaminants,4,5 and improves clarification and filtration by inducing formation of

20

precipitates or through reducing the stabilizing effects of NOM6. Wastewater treatment, drinking water

21

treatment, and water reuse plants have increasingly been including ozonation in their treatment trains to

22

achieve contaminant removal.7-12 Particularly, ozone’s use has increased in some European countries for

23

tertiary treatment of micropollutants in wastewater,13 in North America for drinking water treatment

24

where ozone application followed by biofiltration has increased 10-fold from 1993 to 2013,14 and in

25

several water scarce regions of the world where ozone combined with biofiltration has been explored as

26

an alternative to reverse osmosis for direct or indirect potable reuse (e.g., in Virginia,15 Australia,7

27

California16).

28

Ozone inactivation kinetics can be difficult to resolve in lab-scale experiments because ozone’s

29

fast reactivity results in non-detectable microbial levels after short ozonation times.17 Nevertheless, viable

30

microbial cells have been reported to remain present after ozonation of raw water18 or partially treated

31

water19-21 in full-scale and pilot-scale water treatment systems. A few reasons may explain the presence of

32

viable microorganisms in ozone contactor effluents. First, it is not surprising that different types of

33

microbes exhibit differential ozone resistance since the effectiveness of a disinfectant varies based on

34

variations in cellular and physiological features such as cell envelope composition and growth rate,22

35

growth state such as planktonic or biofilm growth,23,24 and nutrient availability25. In particular, some

36

opportunistic bacterial pathogens exhibit higher ozone resistance than other bacteria.22,26 Second,

37

disinfection in full-scale systems is often less effective than predicted by lab-scale inactivation

38

experiments because of matrix effects, such as the protection of microorganisms from disinfectant

39

exposure when present in flocs21 and lower than intended disinfection concentrations due to oxidant

40

demand exerted by NOM27 or inorganic compounds in untreated or partially treated water.



Page 4 of 33

41

Reduced disinfection efficiency is relevant for drinking water treatment plants (DWTPs) that

42

employ vertically baffled, multi-chamber ozone contactors. Nonideal flow conditions such as short-

43

circuiting and internal recirculation in such contactors can lead to insufficient mixing and reduced ozone

44

exposure.28 A combination of nonideal hydrodynamic conditions and the production of readily

45

biodegradable substrates through oxidation of NOM may increase the potential for biofilm formation on

46

contactor surfaces. Biofilm formation would present problems for disinfection since biofilm extracellular

47

polymeric substances (EPS) may shield microorganisms from ozone exposure and cells grown in biofilms

48

exhibit reduced sensitivity to disinfection.23,24 Surveys of microbial communities in full-scale DWTPs

49

19,21,29,30

50

in contactor influents and effluents only. Research is needed to elucidate factors that explain the presence

51

of viable microorganisms in ozone contactor effluents.

have treated ozone contactors as “black boxes,” focusing on the planktonic microbial community

52

In this study, viable microbial cell concentrations were monitored at five locations across multi-

53

chamber ozone contactors at a full-scale DWTP to gain insight into processes that shape the microbial

54

quality (i.e., microbial concentrations and community composition) of ozonated water. Furthermore, we

55

evaluated the bacterial community structures in sludge and in biofilms on baffle walls to generate

56

hypotheses for how nonideal flow conditions influence the microbial community in ozone contactor

57

effluent.


Page 5 of 33


58

MATERIALS AND METHODS

59

Drinking Water Treatment Plant. The maximum capacity of the DWTP in Ann Arbor, MI is 50 million

60

gallons per day (MGD) (1.9 × 105 m3/day) and the average daily water production is 15 MGD (5.7 × 104

61

m3/day). Pinto et al. previously provided a description of the Ann Arbor DWTP.30 Briefly, raw water is

62

obtained from the Huron River (80-85%) and from groundwater wells (15-20%) and is treated using lime

63

softening, coagulation, flocculation, sedimentation, ozonation, filtration, and chloramination. Before

64

ozonation, CO2 contactors reduce the pH of the water to approximately 7.8 (mean pH was 7.8 ± 0.4 for

65

December 2015 through December 2016). Ozonated water is applied to biologically active, dual media

66

filters containing granular activated carbon and sand.

67

Ozone Contactors. The Ann Arbor DWTP has four parallel ozone contactors each containing seven

68

chambers separated by vertical baffle walls and inlet and outlet basins (Figure 1). Two of the four

69

contactors are typically operated to meet water demand. Onsite ozone generators are fed oxygen and

70

produce a gas stream with approximately 7-10% ozone. Ozone is dispersed by ceramic fine bubble

71

diffusers into chambers 2 and 3 of each contactor. Specifically, 70% of the ozone generated is dispersed

72

into chamber 2 and 30% is dispersed into chamber 3. Water flows by gravity counter-currently to the

73

ozone in chamber 2 and co-currently in chamber 3. A sufficient ozone dose is added to maintain 0.1 mg/L

74

ozone residual in the effluent of chamber 2 to achieve 1-log virus inactivation credit under the United

75

States Environmental Protection Agency (U.S. EPA) Surface Water Treatment Rule. The effluent of the

76

seventh chamber in each contactor flows over a weir to dissipate dissolved gases, and water from the two

77

contactors in operation is mixed in an outlet basin.

78

In each contactor, five sample lines extend from the contactor chambers to an ozone analyzer

79

(Figure 1). Samples collected from sample line 1 represent clarified water after pH adjustment but before

80

ozone exposure (CT = 0). Samples from sample lines 2, 3, and 4 represent water after contact with ozone

81

gas in chambers 2 and 3, and water after exposure to any residual ozone in chamber 4 and the outlet basin,

82

respectively. The residence time for a complete contactor is approximately 10 min (average was 9.8 ± 1.3

83

min across sampling events for heterotrophic plate count and flow cytometry analyses). The length of Page 5 of 27 ACS Paragon Plus Environment


Page 6 of 33

84

chamber 3 (i.e., the distance between consecutive baffle walls) is twice the length of the other chambers.

85

While uniform chamber sizes are most common in ozone contactors, some contactor designs provide

86

larger chambers for ozone dissolution. The co-current flow in chamber 3 is hydraulically less optimal

87

relative to the counter-current flow in chamber 2.31 Longer chamber length provides additional contact

88

time in chamber 3.

89

Sample Collection. Water samples were collected from three contactors over the course of a year (Table

90

S1). Before sample collection, each sample line was flushed until meter readings of ozone concentration

91

were stable (approximately 3 min). 500 mL water samples were collected into sterile Nalgene

92

polypropylene bottles containing sodium thiosulfate (5 × stoichiometric requirement (SR)17,32 to quench

93

0.5 mg/L ozone) from each sample line. Separate 500 mL samples were collected into sterile, carbon-free

94

amber bottles containing sodium thiosulfate for assimilable organic carbon (AOC) analyses by Eurofins

95

Eaton Analytical (described below).

96

Ozone Exposure (CT) Calculation Method. Ozone concentration and temperature were monitored in

97

each sample line by online Ozone 499A OZ Sensors. The ozone exposure or CT (concentration × time)

98

for each chamber was calculated as the average ozone concentration (C) in the chamber multiplied by the

99

T10. The U.S EPA Surface Water Treatment Rule designates T10 as the time it takes for 10 percent of a

100

tracer to exit a disinfection contactor. For reaction chambers 4-7 (no ozone addition), a first order ozone

101

decay model was used to determine the average ozone concentration. Residual ozone concentrations at

102

sample line 5 (in the outlet basin) were typically below the quantification limit (0.01 mg/L). The CT

103

calculation method is described in detail in the Supporting Information (Table S2).

104

Heterotrophic Plate Counts (HPC). HPC were conducted using the pour plate method according to

105

Standard Methods 9215 B.33 Samples were diluted (1:100, 1:10; 1:5, or 1:2 depending on the sample)

106

with sterile phosphate buffered saline (PBS, pH = 7.2 ± 0.2) solution and 1 mL diluted sample was mixed

107

with liquefied R2A agar in triplicate and plated (n = 6 plates per sample). Colony forming units (CFUs)

108

were counted using a Quebec Darkfield Colony Counter (Reichert, Inc., Depew, NY) after seven days of

109

incubation at 28 °C. Ratios of culturability were calculated for water from sample lines 1 and 5, for each Page 6 of 27 ACS Paragon Plus Environment

Page 7 of 33


110

of the two days when HPC and flow cytometry methods (described below) were both used. The values

111

reported in the manuscript are the mean ± range for sample lines 1 or 5 across the two days.

112

Taxonomic Classification of Isolates. For one of the sampling days, 190 isolates were cut from R2A

113

plates for taxonomic analysis of culturable populations. More information is provided in the Supporting

114

Information (Figure S5).

115

Flow Cytometry Analysis. Sample aliquots (1 mL) were stained with propidium iodide (PI,

116

ThermoFisher Scientific, Waltham, MA) and SYBR Green I (SGI, ThermoFisher Scientific) at final

117

concentrations of 3 µM PI and 1X SGI. After staining, samples were incubated at 37°C for 15 min to

118

improve cell staining.34,35 CountBrightTM absolute counting beads (Life Technologies, Carlsbad, CA) were

119

added to each sample per the manufacturer’s instructions to determine flow cytometer process volume.

120

All samples were stained and analyzed in duplicate and stored on ice until analysis with an LSR Fortessa

121

cell analyzer (BD Biosciences, San Jose, CA). Events were triggered off fluorescence in the SGI or PI

122

channels above a threshold of 200. For each sample, forward and side-scatter as well as fluorescence in

123

the 610/10, 530/30, and 780/60 nm wavelengths were measured for detection of PI positive, SGI positive,

124

and bead events, respectively. At least 10,000 events were collected for each sample, and no

125

compensation was applied to sample measurement. Data analysis was carried out using FlowJo software

126

(FlowJo, Ashland, OR). Electronic gating of events in a density plot of SGI and PI fluorescence channels

127

was used to determine the number of events positive for SGI and negative for PI (membrane-intact cells).

128

Organic Carbon Analysis. Each quenched sample was filtered through sterile 0.45 µm nylon syringe

129

filters (ThermoFisher Scientific) previously washed with 40 mL MilliQ water and 10 mL of sample, and

130

the filtrate was frozen for dissolved organic carbon (DOC) analysis with a Total Organic Carbon

131

Analyzer TOC-V CSH (Shimadzu, Columbia, MD).

132

Samples were submitted to Eurofins Eaton Analytical (South Bend, IN) for AOC analysis

133

according to Standard Methods 9217B.33 At Eurofins Eaton Analytical, water samples were aliquoted into

134

vials (40 mL each) and pasteurized (70-80 °C for 30 min). Spirillum NOX and Pseudomonas fluorescens

135

P-17 strains, which have different substrate specificities, were inoculated into each vial at an initial Page 7 of 27 ACS Paragon Plus Environment


Page 8 of 33

136

concentration of 500 CFU/mL and then incubated at 25 °C. Growth in each vial was determined on R2A

137

agar plates in duplicate on days 5, 6 and 7 (n = 6 plates per sample). Average CFU was converted into

138

AOC concentration based on the growth yields of the NOX and P-17 strains for acetate (4.1 × 106 and 1.2

139

× 107 CFU/µg acetate-C for P17 and NOX, respectively). AOC determined by P-17 growth and AOC

140

determined by NOX growth were added and expressed as the total sample AOC.

141

Contactor Shutdown Experiment. Water flow and ozone supply for one of the contactors were turned

142

off. Samples were collected from chambers 2 and 3 before shutdown, and 2 h and 9.45 h after shutdown

143

for HPC and flow cytometry analyses as described above. The ozone contactor was kept dark and

144

relatively full during shutdown. Immediately after shutdown, water from chamber 2 was collected into a

145

sterile glass bottle protected from light and kept at room temperature to serve as an experimental control,

146

representing the bulk water in the contactor but without exposure to contactor surfaces. After shutdown,

147

the contactor could be considered a batch system, resulting in a low rate of microbial cell diffusion into

148

the bulk water, motivating the selected sampling time points on the order of hours (2 h and 9.45 h).

149

Residual ozone concentrations were monitored in chamber 2 by online Ozone 499A OZ Sensors.


Page 9 of 33


150

Ozone Contactor Biofilm and Sludge Samples. After the shutdown experiment, the contactor was

151

drained and, for safety purposes, ozone was allowed to degas from the concrete walls for two weeks

152

before we entered the contactor. Biofilm was swabbed from specific areas (18.8 cm2 each) on baffle

153

walls, at about 1.5 m from the floor, from chambers 2-5. Duplicate biofilm samples, collected for 3 out of

154

5 wall locations sampled, were swabbed next to each other. Biofilm from one diffuser in chamber 3 (n =

155

1) was collected. Each swab was placed in sterile PBS. Swabs were vortexed for 2 min and sonicated for

156

2 min in an ultrasonicator (ThermoFisher Scientific) to release biomass. The biomass was filtered onto

157

0.2 µm polycarbonate filters (Millipore) and filters were stored at -80 °C. Sludge from a hydraulic dead

158

zone in chamber 3 (n = 1), the chamber where sludge accumulation was the greatest, was collected into a

159

sterile Falcon tube (Corning, Corning, NY) and stored on ice for up to 1 h before storage at -80 °C. A

160

hydraulic dead zone is a zone of insufficient mixing resulting from internal recirculation inside the

161

chamber.36

162

Total DNA was extracted from biofilm and sludge samples using a modified recipe of the

163

Universal Extraction Lysis Buffer and a phenol-chloroform-isoamyl extraction protocol37,38 to improve

164

recovery of difficult-to-lyse bacteria. The controls were processed in parallel with the samples. PCR,

165

sample multiplexing, and Illumina MiSeq sequencing of partial 16S rRNA genes were performed by the

166

Microbial Systems Laboratory at the University of Michigan Medical School (Ann Arbor, Michigan).

167

Barcoded dual-index primers specific to the V4 region of the 16S rRNA gene39,40 were used to amplify

168

the DNA. PCR reactions included 5 µL of 4 µM forward and reverse primers, 0.15 µL of AccuPrime Taq

169

DNA High Fidelity Polymerase (Life Technologies), 2 µL of 10x AccuPrime PCR Buffer II

170

(ThermoFisher Scientific), 1 µL of DNA template, BSA (0.5 mg/mL final concentration), and nuclease-

171

free water. PCR conditions consisted of 2 min at 95 °C followed by 30 cycles of 95 °C for 20 s, 55 °C for

172

15 s, 72 °C for 5 min, followed by 72 °C for 10 min. Sequencing was done on the Illumina MiSeq

173

platform using a MiSeq Reagent Kit V2 500 cycles according to the manufacturer’s instructions with

174

some modifications.39



Page 10 of 33

175

16S rRNA gene sequences from all samples were curated using mothur (version 1.39.0).41

176

Sequences that were within two nucleotides of each other were merged using the pre.cluster command.

177

After preclustering, 439 sequences in the control samples (putative contaminants) were removed from the

178

6,429 unique sequences in the dataset using the remove.seqs command (6.8% sequences removed).

179

Curated sequences (131,264 total sequences) were clustered into OTUs using a 97% similarity cutoff with

180

the opticlust algorithm.42 Subsampling was performed based on the sample with the lowest number of

181

sequences, resulting in 4,387 paired-end reads per sample.39 Sequences were classified using a naïve

182

Bayesian classifier43 trained against a 16S rRNA gene training set provided by the RDP.44 An RDP

183

consensus taxonomy was generated for each OTU.

184

Thermal Gravimetric Analysis (TGA) of Sludge. TGA analysis was performed to estimate the

185

composition of contactor sludge. A sample of the sludge was weighed, dried in an oven at 103 °C for 1 h

186

and weighed again to determine water loss. The sludge was subject to TGA in an oxygen atmosphere. The

187

temperature was ramped up from ambient to 800 °C at a rate of 10 °C min-1 in a TA Q500 (TA

188

Instruments, New Castle, DE). Continuous online records of weight loss and temperature were used to

189

plot a derivative TGA curve.

190

Statistical Analyses. Results are reported as the mean ± standard deviation unless otherwise noted.

191

Statistical significance was based upon the probability of p < 0.05. A Wilcoxon test was used for

192

comparison of cell concentrations. An Analysis of Molecular Variance (AMOVA) was used to test for

193

significant separation of samples on the NMDS plot. A Pearson correlation coefficient was used for the

194

correlation analysis with individual OTUs and shifts in samples along axes in the NMDS plot.


Page 11 of 33


195

RESULTS

196

Because the Ann Arbor DWTP did not operate the same ozone contactors throughout the sampling

197

campaign, water samples were collected from three contactors on different days (Table S1). Plant

198

operators change ozone feed gas concentrations and influent flow rates through the contactors daily and

199

this contributed to the variability in ozone CT profiles across the sampling days (Figure 1). Therefore,

200

viable cell concentrations are plotted against sample line numbers (1-5) instead of ozone CT. The average

201

cumulative ozone CT was 0.96 ± 0.32 mg-min/L across eight sampling days (Figure 1).

202

Cell Concentrations Across Ozone Contactors. Culturing and culture-independent flow cytometry were

203

used to monitor microbial communities during ozone disinfection in multi-chamber contactors. Figure 2

204

shows viable cell concentrations measured by culturing and culture-independent flow cytometry. HPC

205

based inactivation monitoring indicated 1.88 ± 0.47 log reduction from 1,110 ± 420 CFU/mL before

206

ozonation to 17 ± 19 CFU/mL in chamber 2, the first chamber where ozone contact occurs (ozone

207

exposure of 0.26 ± 0.06 mg-min/L; Figure 1) (Figure 2a). As expected, greater numbers of intact cells

208

were measured using flow cytometry than with HPC. However, a similar log reduction of 1.86 ± 0.19 was

209

observed from 106,000 ± 31,000 intact cells/mL before ozonation to 1,610 ± 730 intact cells/mL in

210

chamber 2 (Figure 2b). The levels of culturable and membrane-intact cells increased in the third chamber,

211

even as ozone exposure increased by 0.63 ± 0.17 mg-min/L (Figure 1). Specifically, in chamber 3, there

212

was a significant (p < 0.0001) increase in the level of culturable cells to 750 ± 350 CFU/mL, and the level

213

of intact cells increased to 2,700 ± 1,100 cells/mL (p = 0.05). Intact cell concentrations increased further

214

in chamber 4 (p = 0.05), but culturable cell concentrations did not increase significantly. In spite of

215

increasing intact cell concentrations in chambers 3 and 4 on most days (Figure S1b), ozonation

216

significantly reduced intact cell concentrations overall (1.62 log reduction, p = 0.0001), with

217

concentrations ranging from 1,200 to 3,750 intact cells/mL in water from the outlet basin (sample line 5).

218

The levels of culturable bacteria in the outlet basin ranged over an order of magnitude, from 200 to 3,850

219

CFU/mL. In comparison with cell concentrations before ozonation (sample line 1), there was a slight

220

reduction in culturable cells after ozonation (sample line 5) on three sampling days, but a significant (p < Page 11 of 27 ACS Paragon Plus Environment


Page 12 of 33

221

0.0001) increase in CFU concentrations after ozonation in one of the contactors sampled on three

222

different days during one week (Figure S1a). The culturability (i.e., the ratio of HPC to membrane-intact

223

cells) of water samples before ozonation was 0.9 ± 0.4% (mean ± range) but increased 12-fold to 10.4 ±

224

7.8% after ozonation (Figure S2).

225

Microbial Growth Substrates. Dissolved organic carbon (DOC) levels did not change significantly

226

during ozonation (p = 0.08) (Figure S3). The average concentration of DOC across the five sampling lines

227

was 4,000 ± 300 µg/L.. However, AOC levels increased significantly (p < 0.0001) after ozone exposure

228

began in chamber 2. The average AOC level was 36 ± 0.3 µg/L (mean ± range across two sampling

229

events) before ozonation and increased to 115 ± 12 µg/L (mean ± range) after ozonation. The increase in

230

AOC was caused by an increase in growth substrates specific for the Spirillum NOX strain, which

231

primarily uses carboxylic acids as growth substrates45.

232

Cell Concentrations after Contactor Shutdown. A “no flow” (batch) experiment was conducted in a

233

contactor to test whether walls and accumulated solids were the source of increasing cell concentrations

234

in the bulk water. Cell concentrations were monitored in stagnant water in chambers 2 and 3 after influent

235

water and ozone gas supply were shut off. Chambers 2 and 3 were chosen since chamber 2 is the contact

236

chamber where the greatest reduction in viable cell levels occurred and chamber 3 is where cell

237

concentrations increased (Figure 2). Immediately after contactor shutdown, water was collected from

238

chamber 2 and incubated at room temperature in the dark to serve as a control with the same planktonic

239

cell concentration and growth substrate level as the stagnant water in chamber 2 at the start of the

240

experiment. An increase in cell concentrations in the control would be explained by planktonic cell

241

growth or recovery, not detachment of cells from biofilm or sediments.

242

Residual ozone concentrations dropped to below quantification (0.01 mg/L) after 12 min in

243

chamber 2 (Figure S4). The initial CFU and intact cell concentrations in chamber 2 were 10 ± 4 CFU/mL

244

and 1,660 ± 110 intact cells/mL, respectively. The CFU concentration in chamber 2 increased

245

substantially to 800 ± 210 CFU/mL 2 h after shutdown, and then increased further to 4,800 ± 1,300

246

CFU/mL 9.5 h after shutdown (Figure 3a). The concentration of intact cells in chamber 2 increased Page 12 of 27 ACS Paragon Plus Environment

Page 13 of 33


247

similarly (Figure 3b). The initial CFU and intact cell concentrations in chamber 3 were 600 ± 50 CFU/mL

248

and 1,400 ± 20 intact cells/mL, respectively. The CFU concentration in chamber 3 decreased to 160 ± 20

249

CFU/mL 2 h after shutdown, but then increased to 6,010 ± 1,800 CFU/mL 9.5 h after shutdown (Figure

250

3a). The concentration of intact cells in chamber 3 increased consistently at 2 and 9.5 h after shutdown

251

(Figure 3b). In contrast, the concentration of culturable cells in the control decreased and was below

252

detection after 2 h and remained below detection during the 9.5 h experiment. The levels of membrane-

253

intact cells in the control decreased slightly to 820 ± 30 intact cells/mL after 2 h, and then increased to

254

3,890 ± 4,400 intact cells/mL after 9.5 h (Figure 3b).

255

Taxonomic Classification of Isolates. The colonies on R2A plates generally became smaller and more

256

uniform (mostly white colonies) in water sampled from sequential chambers (Figure S5). Table 1 shows

257

taxonomic classification at the genus level of 190 isolates based on 16S rRNA gene sequencing.

258

Pseudomonas, Blastomonas and Limnobacter spp. predominated among the strains cultured from water

259

before ozone exposure (CT = 0 mg-min/L). The relative abundance of Pseudomonas strains decreased

260

and Blastomonas strains were not detected after ozone exposure began (CT = 0.30 mg-min/L), whereas

261

the relative abundance of Limnobacter strains increased substantially to greater than 80% after ozone

262

exposure began. Other bacteria cultured from water in the outlet basin (sample line 5) included

263

Hydrogenophaga strains and unclassified Comamonadaceae spp. Isolates obtained from water before

264

ozonation belonged to Alphaproteobacteria, Betaproteobacteria, and Gammaproteobacteria, but strains

265

isolated after ozonation began were mostly Betaproteobacteria and, by the end of ozonation, all cultured

266

populations were Betaproteobacteria.

267

Biofilm and Sludge Samples. Partial 16S rRNA gene sequencing of biofilms on walls in chambers 2

268

through 5 indicated that some taxa cultured from water samples collected during ozonation were also

269

detected in the biofilm samples. For example, an OTU classified as Limnobacter (OTU0004) was present

270

in biofilm sampled from the wall in chamber 2 (1.4% relative abundance), and its relative abundance was

271

substantially higher in biofilm on the left wall of chamber 3 (29%) (Table S5). This OTU was also present



Page 14 of 33

272

in biofilm on a diffuser in chamber 3 (2.1% relative abundance). However, Limnobacter spp. were not

273

detected in biofilm samples on the right wall in Chamber 3, or in biofilm samples from chambers 4 and 5.

274

Sludge had accumulated near the right wall in chamber 3 and TGA of a sludge sample (Figure

275

S6) showed significant weight loss occurred between 542 and 830 °C, corresponding to the decarbonation

276

of calcium carbonate46. This finding suggests sludge accumulation was due to incomplete removal of

277

calcium carbonate during lime softening. Smaller weight losses during TGA occurring at 502 °C and

278

between 183 and 407 °C may have corresponded to dihydroxylation of calcium hydroxide and

279

volatilization of organic content, respectively.47,48

280

DNA yields from baffle wall biofilm samples after the dead zone were significantly higher than

281

those before the dead zone (p = 0.036) (Table S3). Furthermore, non-metric multidimensional scaling of

282

ϴyc distance ordinations indicated that the biofilm community structures of samples collected before and

283

after the dead zone clustered separately (Figure 4), and the separation was statistically significant (p =

284

0.017). An OTU (OTU0001) classified as Gp4 in Acidobacteria became significantly more abundant in

285

biofilm samples after the dead zone (p = 0.036). The relative abundance of this OTU was significantly

286

correlated with a shift in the biofilm community structure along NMDS axis 1 (p = 0.002), shifting from

287

the community structure before the dead zone toward the community structure after the dead zone (Figure

288

4). In addition, the relative abundance of OTUs classifying as Mycobacterium and Legionella, two genera

289

that contain species of public health importance as opportunistic pathogens, increased significantly in

290

biofilm samples after the dead zone (p = 0.036 and p = 0.036 for Mycobacterium and Legionella,

291

respectively) (Figure 5). The bacterial community in the sludge that accumulated near the right baffle wall

292

in chamber 3 contained sequences classified as Gp4 OTU0001 (1.5% relative abundance) as well as

293

Mycobacterium and Legionella (0.41% and 2.4% relative abundance, respectively) (Figure 6). The

294

Shannon diversity indices of biofilms from baffle walls varied between 3.1 and 4.5 (mean of 4.02 ± 0.45,

295

Table S4), whereas the Shannon diversity indices of biofilm from a diffuser and from the sludge in


Page 15 of 33


296

chamber 3 were 5.2 and 5.0, respectively. Shannon evenness ranged from 0.58 to 0.86 across all samples,

297

with the sludge bacterial community being the most even.

298

DISCUSSION

299

Viable cell concentrations. The overall level of inactivation across each contactor, as determined by

300

culture-independent flow cytometry, was higher than inactivation measured by culture-based methods.

301

Across the contactors, 1.62 log reduction of membrane-intact cells was observed on average (Figure 2b),

302

which is consistent with 1-2 log reductions of membrane-intact cells, as measured by flow cytometry,

303

reported for ozonation (0.33-0.35 mg O3/mg C) of partially treated raw waters in two other full-scale

304

DWTPs49. However, on days when HPC levels were reduced across the contactors, the reductions

305

observed (0.32 ± 0.05 log for April 22, Oct 4, Oct 19, 2016) were lower than those reported for a pilot-

306

scale ozone contactor column (1.5 log)50 and a baffled ozone reactor treating secondary wastewater

307

effluent (1.4-1.6 log)21. The greater culturability of the bacteria in water samples after ozonation (Figure

308

S2) may have contributed to the modest reductions in HPC levels observed. The potential for changes in

309

culturability to impact estimates of microbial reductions is one reason why flow cytometry was used in

310

addition to HPC to enumerate viable cells. HPC levels in contactor effluents were sometimes higher than

311

influent levels (May 5, 9, 13, 2016; Figure S2). At a DWTP in the Netherlands, levels of heterotrophic

312

bacteria isolated on conventional plate count agar increased after ozonation, although the increase was not

313

observed when R2A agar was used.49 A major limitation of culture-based studies is that only a small

314

proportion of viable cells are amenable to cultivation. Therefore, HPC may not be as responsive to

315

changes in bacterial abundance as culture-independent methods such as flow cytometry.51-53 The observed

316

increase in HPC and intact cell concentrations in chamber 3 and subsequent chambers indicates that

317

microbial growth occurred or that viable cells were introduced to the water through another mechanism.

318

Most of the overall reduction of viable cells occurred in the first ozone contact chamber (chamber

319

2), where an approximate 2-log reduction of HPC and intact cell concentrations was observed. An



Page 16 of 33

320

investigation of bacterial concentrations at the same plant conducted three months after ozonation was

321

implemented reported 2.6-log reduction of HPC determined by culturing on R2A agar (5.43 × 103

322

CFU/mL in contactor influent to 13 CFU/mL in effluent).19 Results of this earlier study taken together

323

with our findings suggest the reductions originally achieved over the full contactor length were

324

constrained to the first ozone contact chamber. While the first contact chamber provides a disinfection

325

barrier to microorganisms present in the partially treated raw water given the significant reduction in

326

viable cells (Figure 2a and 2b) observed after chamber 2, our data suggest that characteristics of chambers

327

3-7 shape the microbial quality of the water exiting the contactors.

328

Biofilm Detachment. The increase in viable cell concentrations after the first contact chamber cannot be

329

attributed to planktonic cell growth considering the short residence time in the contactors (approximately

330

10 min). The doubling rates required to produce such an increase would be much shorter than the fastest

331

doubling rates reported for microorganisms (e.g., 9.8 min for Vibrio natriegens54). Within 2 h of shutting

332

off water flow and ozone gas supply, HPC and intact cell concentrations increased in chambers 2 and 3,

333

but not in a control (Figure 3), suggesting that the interaction of bulk water with contactor surfaces or

334

accumulated solids was necessary to produce the increasing cell concentrations observed.

335

The contactors are made from poured concrete. Carbonation, i.e., the replacement of calcium

336

hydroxide with calcium carbonate, reduces the pH of concrete over time55 thereby resulting in more

337

favorable conditions for biofilm formation. Passive biofilm detachment from contactor surfaces may

338

occur continuously by erosion56 or through discrete detachment events (i.e., sloughing),56,57 and may be

339

due to shear created by fluid moving past the biofilm58 and/or an increase in detachment rate caused by

340

higher growth rates due to greater AOC availability59. Had shear-based detachment been important,

341

higher effluent cell concentrations would have been expected with increased flow through a contactor.

342

However, the times when effluent CFU concentrations were greatest (May 5, 9, 13, 2016; Figure S2) did

343

not correlate with days when the plant was operating at higher flowrates (Table S1). Instead, microbial

344

levels in bulk water increased quickly under stagnant conditions (Figure 3). AOC in contactor influent


Page 17 of 33


345

and AOC generated in chamber 2 (Figure S3) provided a continuous supply of biofilm growth substrates.

346

The results support a scenario whereby AOC generated by ozonation provides growth substrate for

347

biofilm formation on contactor surfaces. Subsequently, biofilm detaches, thereby increasing planktonic

348

cell concentrations in contactor effluents.

349

Biofilm community structure. Results from the contactor shutdown experiment, and other evidence

350

discussed above, suggest that biofilm detachment may be responsible for the presence of microorganisms

351

detected in water samples. This underlines the importance of understanding the factors that shape the

352

microbial community structure of biofilms on contactor surfaces to control microbial quality of the

353

effluent. Biofilm microbes may have cellular and physiological features that facilitate their resistance to

354

ozone exposure. Additionally, biofilm EPS and carbonate on contactor surfaces may serve to protect

355

microorganisms from ozone and ozone decomposition products, respectively, as carbonate is an effective

356

hydroxyl radical scavenger60.

357

Recent research has focused on the potential for enrichment of disinfectant resistant,

358

opportunistic pathogenic bacteria through disinfection processes in water treatment.61,62 We detected

359

Mycobacterium and Legionella OTUs in biofilm samples from contactor walls in chamber 2, where the

360

highest ozone residuals were applied. Their resistance to ozone would be consistent with lab-scale

361

inactivation experiments demonstrating that certain Legionella and Mycobacterium species (e.g.,

362

Legionella pneumophila63 and Mycobacterium avium22) are more resistant to ozone exposure than E. coli.

363

The finding that ozonation of a wastewater treatment plant effluent selected for bacteria with high

364

genomic guanine-cytosine (GC) content is relevant for ozone selection of Mycobacterium species, which

365

have GC rich genomes.64 While we did not culture Mycobacterium spp. from the ozonated water samples,

366

in an earlier study performed shortly after the plant began ozonating, Mycobacterium was the most

367

common group and comprised 31% of bacterial isolates identified after ozonation.19 These results raise

368

the possibility that the presence of Legionella and Mycobacterium spp. in biofilm on contactor walls was

369

related to their resistance to ozone and not solely their protection from ozone exposure.



Page 18 of 33

370

Smaller colony sizes after ozone contact in chamber 2 (Figure S5) would be consistent with the

371

view that ozone exposure selected for disinfectant-resistant microorganisms as bacteria with lower

372

specific growth rates often exhibit higher resistance to stress,22,65,66 although this is not always the case23.

373

However, the ozone resistance of the cultured microorganisms was not tested in this study. Ozone

374

inactivation experiments are needed to confirm ozone resistance of microorganisms in contactor effluents.

375

Furthermore, given the range of viable cell concentrations we observed in contactor effluents,

376

fluorescence activated cell sorting of viable populations71 combined with amplicon sequencing could be

377

used for identification of viable cells in ozonated water. While the observed enrichment for Limnobacter

378

spp. in this study was consistent with their detection in biofilms from water distribution networks67,68 and

379

premise plumbing,69 their resistance to ozone is unknown since ozone inactivation kinetics for

380

Limnobacter spp. have not been reported. However, in a recent study, the relative abundance of

381

Limnobacter increased significantly in a surface water disinfected with photocatalytic ozonation.70 We

382

had expected that Limnobacter would be present in all biofilm samples based on its predominance among

383

isolates from contactor water samples (Table 1), but the community structure of biofilm samples on

384

contactor walls past a dead zone in chamber 3 was substantially different from those in chamber 2

385

(discussed below). The water samples for culturing were collected from a contactor in operation whereas

386

biofilm samples were collected from a contactor that had been taken out of operation. This contactor was

387

chosen for shutdown because ozone demand was unusually high. Upon inspection, we observed an

388

accumulation of sludge in the contactor, which may have caused the high ozone demand. It is likely that

389

this sludge had a greater impact on the biofilm community structure of the baffle walls in this contactor

390

than in the contactor from which water samples had been collected for culturing.

391

Dead zone effects. The length of chamber 3 is double the length of chamber 2 to provide for additional

392

disinfectant contact time. As the water velocity through chamber 3 decreases, more sediments are

393

deposited in chamber 3. This greater length is also expected to increase internal recirculation inside the

394

chamber and result in zones of insufficient mixing (dead zones).28 The accumulation of sludge observed

395

near the right baffle wall in chamber 3 was in a location predicted to be a dead zone by modeling of flow Page 18 of 27 ACS Paragon Plus Environment

Page 19 of 33


396

through multi-chamber contactors.28 This sludge accumulation, and our data suggesting the presence of

397

biomass in the sludge, may be one way viable cells were introduced to the water in chamber 3.

398

Furthermore, the significantly different biofilm community structure past the dead zone raises the

399

possibility that microorganisms in sludge in the dead zone seeded biofilms downstream. However, it is

400

possible there was a greater impact of sludge on the biofilm community structure of baffle walls in this

401

contactor than in the contactor from which water samples for culturing were collected. Although lime

402

softening precipitates are expected to have a high pH unfavorable for microbial life, the Shannon diversity

403

of the bacterial community in a sludge sample from one dead zone was relatively high (5.0 ± 0.01, Table

404

S4), possibly due to the rapid decomposition of ozone when exposed to the high surface pH of the lime

405

softening precipitates, thus providing protection of microorganisms from ozone exposure. This notion

406

raises the possibility that sludge in dead zones harbor microorganisms that present a public health risk.

407

The presence of Mycobacterium and Legionella OTUs in sludge, and an increase in their relative

408

abundance in biofilms past the dead zone, suggest that adverse microbial effects of dead zones are an

409

argument for improving hydraulic efficiency in multi-chamber contactors by decreasing chamber length,

410

in spite of greater energy requirements to drive flow over and under more baffle walls36. Therefore,

411

chamber length should be an important feature for contactor design. Additionally, use of materials that are

412

less reactive with ozone and do not quench hydroxyl radicals (e.g., acidic materials) in contactor

413

construction may minimizes biofilm formation on contactor surfaces. As retrofitting ozone contactors is

414

not an option for many existing DWTPs, other options need to be considered to improve ozone

415

disinfection efficiency. Strategies that focus on increasing ozone dose to avoid biofilm formation need to

416

consider undesirable ozonation byproducts such as bromate72 and nitrosamines73. Better pretreatment

417

before ozonation and more frequent contactor cleanings are obvious strategies to reduce the effects of

418

sediments entering the ozone contactor in the present study. Our findings from one contactor need to be

419

confirmed with sampling campaigns in other full-scale, multi-chamber ozone contactors.



Page 20 of 33

420

While previous research has shed light on the importance of biofilms in determining microbial

421

water quality in distribution systems,74 our results provide a new appreciation for the potential role of

422

biofilm detachment in determining multi-chamber ozone contactor effluent water quality. Our findings

423

suggest implications for post-ozonation processes as ozone contactors may serve as an important seed

424

source of microorganisms to downstream biofilters. More research is needed to understand the interplay

425

between ozonation and biofiltration, two unit processes that may represent critical control points for

426

drinking water microbial quality30,75,76.


Page 21 of 33


427

ACKNOWLEDGEMENTS

428

We thank James Yonts, Ryan Justin, Jeff Keown, Matt Brown, Mike Switzenberg, Aaron Buza, Duane

429

Weible, Wendy Schultz and Brian Steglitz for assistance with sampling and data collection; Michael

430

Pihalja, Aaron Robida, Krista Wigginton, and Frederik Hammes for helpful discussions; and Johannes

431

Schwank, Xiaoyin Chen and Yun Shen for assistance with TGA analysis. We also thank anonymous

432

reviewers for their valuable feedback. Partial funding for this work was provided by the Cystic Fibrosis

433

Foundation and the University of Michigan MCubed Program. NK was supported by Dow Sustainability,

434

Rackham Merit and Integrated Training in Microbial Systems (ITiMS) Fellowships from the University

435

of Michigan (UM). The ITiMS program “Instructional Program Unifying Population and Laboratory

436

Based Sciences” at UM was funded by the Burroughs Wellcome Fund. NR was partially supported by a

437

U.S. National Science Foundation Graduate Research Fellowship. SJH was supported by Alfred P. Sloan

438

Foundation, Microbiology of the Built Environment (G- 2014-13739) and UM Dow Sustainability

439

Fellowships.



Page 22 of 33

440

SUPPORTING INFORMATION

441

Supporting Information for this study is provided and includes an overview of ozone contactors sampled,

442

description of the method used to determine ozone exposure, results of HPC analyses for each sampling

443

day, HPC and flow cytometry measurements on two days, results of organic carbon analyses, residual

444

ozone measurements after contactor shutdown, results of thermal gravimetric analysis of sludge, DNA

445

extraction yields and diversity estimates for biofilm and sludge samples, and taxonomy and relative

446

abundance of the top 10 most abundant OTUs in biofilm and sludge samples.

447


Page 23 of 33


448 449

References

450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494

1. 2. 3.

4. 5.

6. 7.

8.

9.

10.

11.

12.

13.

14. 15.

16.

Hunt, N. K. Mariñas, B. J. Inactivation of Escherichia coli with ozone: chemical and inactivation kinetics. Water Res. 1999, 33 (11), 2633-41. Nerenberg, R.; Rittmann, B. E., Soucie, W. J. Ozone/biofiltration for removing MIB and geosmin. J Am Water Works Assoc. 2000, 92 (12), 85. Saroj, D. P.; Kumar, A.; Bose, P.; Tare, V., Dhopavkar, Y. Mineralization of some natural refractory organic compounds by biodegradation and ozonation. Water Res. 2005, 39 (9), 192133. Zhang, Y. Pagilla, K. Gas-Phase Ozone Oxidation of Hydrogen Sulfide for Odor Treatment in Water Reclamation Plants. Ozone Sci. Eng. 2013, 35 (5), 390-8. Sung, B.; Chu, K.; Yun, S.; Ahn, J.; Lee, J.; Yoo, S.; Lee, J., Ko, K. Removal of iron and manganese ions from abandoned neutral or alkaline mine drainage via ozone oxidation and micro-sand filtration: a pilot-scale operation. Desalin Water Treat. 2015, 53 (9), 2354-62. Jekel, M. R. Effects and mechanisms involved in preoxidation and particle separation processes. Water Sci Technol. 1998, 37 (10), 1-7. Blackbeard, J.; Lloyd, J.; Magyar, M.; Mieog, J.; Linden, K. G., Lester, Y. Demonstrating organic contaminant removal in an ozone-based water reuse process at full scale. Environ Sci Wat Res. 2016, 2 (1), 213-22. Gerrity, D.; Gamage, S.; Holady, J. C.; Mawhinney, D. B.; Quiñones, O.; Trenholm, R. A., Snyder, S. A. Pilot-scale evaluation of ozone and biological activated carbon for trace organic contaminant mitigation and disinfection. Water Res. 2011, 45 (5), 2155-65. Hollender, J.; Zimmermann, S. G.; Koepke, S.; Krauss, M.; McArdell, C. S.; Ort, C.; Singer, H.; von Gunten, U., Siegrist, H. Elimination of organic micropollutants in a municipal wastewater treatment plant upgraded with a full-scale post-ozonation followed by sand filtration. Environ. Sci. Technol. 2009, 43 (20), 7862-9. Huber, M. M.; GÖbel, A.; Joss, A.; Hermann, N.; LÖffler, D.; McArdell, C. S.; Ried, A.; Siegrist, H.; Ternes, T. A., von Gunten, U. Oxidation of pharmaceuticals during ozonation of municipal wastewater effluents: a pilot study. Environ. Sci. Technol. 2005, 39 (11), 4290-9. Reungoat, J.; Escher, B.; Macova, M.; Argaud, F.; Gernjak, W., Keller, J. Ozonation and biological activated carbon filtration of wastewater treatment plant effluents. Water Res. 2012, 46 (3), 863-72. Rosal, R.; Rodríguez, A.; Perdigón-Melón, J. A.; Petre, A.; García-Calvo, E.; Gómez, M. J.; Agüera, A., Fernández-Alba, A. R. Occurrence of emerging pollutants in urban wastewater and their removal through biological treatment followed by ozonation. Water Res. 2010, 44 (2), 57888. Eggen, R. I.; Hollender, J.; Joss, A.; Schärer, M., Stamm, C. Reducing the discharge of micropollutants in the aquatic environment: the benefits of upgrading wastewater treatment plants. Environ. Sci. Technol. 2014, 48 (14), 7683-9. J, B.; G, U.; J, C., T, B., Water Research Foundation North American Biofiltration Knowledge Base Project #4459. 2016. Schimmoller, L.; Bott, C.; Mitchell, J.; Bernas, J.; Henifin, T.; Snyder, E., Holloway, D. Comparing Advanced Treatment Approaches for a Large-Scale Potable Reuse Project in Southeastern Virginia. Proc Water Env Fed. 2016, 2016 (12), 4792-806. Gerrity, D. Snyder, S. Review of ozone for water reuse applications: toxicity, regulations, and trace organic contaminant oxidation. Ozone Sci. Eng. 2011, 33 (4), 253-66.



495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545

17.

18.

19. 20. 21.

22.

23. 24.

25. 26.

27. 28. 29. 30. 31. 32. 33. 34.

35. 36. 37.

Page 24 of 33

Ramseier, M. K.; von Gunten, U.; Freihofer, P., Hammes, F. Kinetics of membrane damage to high (HNA) and low (LNA) nucleic acid bacterial clusters in drinking water by ozone, chlorine, chlorine dioxide, monochloramine, ferrate (VI), and permanganate. Water Res. 2011, 45 (3), 1490-500. Hammes, F.; Berney, M.; Wang, Y. Y.; Vital, M.; Koster, O., Egli, T. Flow-cytometric total bacterial cell counts as a descriptive microbiological parameter for drinking water treatment processes. Water Res. 2008, 42 (1-2), 269-77. Lee, J. Deininger, R. A. Survival of bacteria after ozonation. Ozone Sci. Eng. 2000, 22 (1), 65-75. Vital, M.; Hammes, F., Egli, T. Competition of Escherichia coli O157 with a drinking water bacterial community at low nutrient concentrations. Water Res. 2012, 46 (19), 6279-90. Czekalski, N.; Imminger, S.; Salhi, E.; Veljkovic, M.; Kleffel, K.; Drissner, D.; Hammes, F.; Bürgmann, H., von Gunten, U. Inactivation of Antibiotic Resistant Bacteria and Resistance Genes by Ozone: From Laboratory Experiments to Full-Scale Wastewater Treatment. Environ. Sci. Technol. 2016, 50 (21), 11862-71. Taylor, R. H.; Falkinham, J. O.; Norton, C. D., LeChevallier, M. W. Chlorine, chloramine, chlorine dioxide, and ozone susceptibility of Mycobacterium avium. Appl. Environ. Microbiol. 2000, 66 (4), 1702-5. Berry, D.; Xi, C., Raskin, L. Effect of Growth Conditions on Inactivation of Escherichia coli with Monochloramine. Environ. Sci. Technol. 2009, 43 (3), 884-9. Steed, K. A. Falkinham, J. O. Effect of growth in biofilms on chlorine susceptibility of Mycobacterium avium and Mycobacterium intracellulare. Appl. Environ. Microbiol. 2006, 72 (6), 4007-11. Du, Z.; Nandakumar, R.; Nickerson, K. W., Li, X. Proteomic adaptations to starvation prepare Escherichia coli for disinfection tolerance. Water Res. 2015, 69, 110-9. Lezcano, I.; Pérez Rey, R.; Baluja, C., Sánchez, E. Ozone Inactivation of Pseudomonas aeruginosa, Escherichia coli, Shigella sonnei and Salmonella typhimurium in Water. Ozone Sci. Eng. 1999, 21 (3), 293. Broadwater, W. T.; Hoehn, R., King, P. Sensitivity of three selected bacterial species to ozone. Appl. Microbiol. 1973, 26 (3), 391-3. Kim, D.; Kim, D.-I.; Kim, J.-H., Stoesser, T. Large eddy simulation of flow and tracer transport in multichamber ozone contactors. J. Environ. Eng. 2009, 136 (1), 22-31. Vital, M.; Stucki, D.; Egli, T., Hammes, F. Evaluating the Growth Potential of Pathogenic Bacteria in Water. Appl. Environ. Microbiol. 2010, 76 (19), 6477-84. Pinto, A. J.; Xi, C. W., Raskin, L. Bacterial Community Structure in the Drinking Water Microbiome Is Governed by Filtration Processes. Environ. Sci. Technol. 2012, 46 (16), 8851-9. Langlais, B.; Reckhow, D. A., Brink, D. R., Ozone in water treatment: application and engineering. 1991: American Water Works Research Foundation, CRC press. Hunt, N. K. Mariñas, B. J. Kinetics of Escherichia coli inactivation with ozone. Water Res. 1997, 31 (6), 1355-62. APHA AWWA. WEF, 2005. Standard Methods for the Examination of Water and Wastewater. 2005, 21, 258-9. Barbesti, S.; Citterio, S.; Labra, M.; Baroni, M. D.; Neri, M. G., Sgorbati, S. Two and three‐ color fluorescence flow cytometric analysis of immunoidentified viable bacteria. Cytometry. 2000, 40 (3), 214-8. Van Nevel, S.; Koetzsch, S.; Weilenmann, H.-U.; Boon, N., Hammes, F. Routine bacterial analysis with automated flow cytometry. J. Microbiol. Methods. 2013, 94 (2), 73-6. Zhang, J.; Tejada-Martínez, A., Zhang, Q. Hydraulic efficiency in RANS of the flow in multichambered contactors. J. Hydraul. Eng. 2013, 139 (11), 1150-7. Hill, V. R.; Narayanan, J.; Gallen, R. R.; Ferdinand, K. L.; Cromeans, T., Vinjé, J. Development of a Nucleic Acid Extraction Procedure for Simultaneous Recovery of DNA and RNA from Diverse Microbes in Water. Pathogens. 2015, 4 (2), 335-54. Page 24 of 27 ACS Paragon Plus Environment

Page 25 of 33

546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595


38.

39.

40.

41.

42. 43.

44.

45. 46. 47.

48.

49.

50. 51.

52.

53. 54. 55.

Haig, S.-J.; Kotlarz, N.; LiPuma, J. J., Raskin, L. A high-throughput approach for identification of nontuberculous mycobacteria in drinking water reveals relationship between water age and Mycobacterium avium. Submitted. mBio. Kozich, J. J.; Westcott, S. L.; Baxter, N. T.; Highlander, S. K., Schloss, P. D. Development of a dual-index sequencing strategy and curation pipeline for analyzing amplicon sequence data on the MiSeq Illumina sequencing platform. Appl. Environ. Microbiol. 2013, 79 (17), 5112-20. Caporaso, J. G.; Lauber, C. L.; Walters, W. A.; Berg-Lyons, D.; Lozupone, C. A.; Turnbaugh, P. J.; Fierer, N., Knight, R. Global patterns of 16S rRNA diversity at a depth of millions of sequences per sample. Proc. Natl. Acad. Sci. 2011, 108 (Supplement 1), 4516-22. Schloss, P. D.; Westcott, S. L.; Ryabin, T.; Hall, J. R.; Hartmann, M.; Hollister, E. B.; Lesniewski, R. A.; Oakley, B. B.; Parks, D. H.; Robinson, C. J.; Sahl, J. W.; Stres, B.; Thallinger, G. G.; Van Horn, D. J., Weber, C. F. Introducing mothur: Open-Source, Platform-Independent, Community-Supported Software for Describing and Comparing Microbial Communities. Appl. Environ. Microbiol. 2009, 75 (23), 7537-41. Westcott, S. L. Schloss, P. D. OptiClust, an Improved Method for Assigning Amplicon-Based Sequence Data to Operational Taxonomic Units. mSphere. 2017, 2 (2), e00073-17. Wang, Q.; Garrity, G. M.; Tiedje, J. M., Cole, J. R. Naive Bayesian classifier for rapid assignment of rRNA sequences into the new bacterial taxonomy. Appl. Environ. Microbiol. 2007, 73 (16), 5261-7. Cole, J. R.; Wang, Q.; Cardenas, E.; Fish, J.; Chai, B.; Farris, R. J.; Kulam-Syed-Mohideen, A.; McGarrell, D. M.; Marsh, T., Garrity, G. M. The Ribosomal Database Project: improved alignments and new tools for rRNA analysis. Nucleic Acids Res. 2009, 37 (suppl 1), D141-D5. Van der Kooij, D. Hijnen, W. Substrate utilization by an oxalate-consuming Spirillum species in relation to its growth in ozonated water. Appl. Environ. Microbiol. 1984, 47 (3), 551-9. Alarcon-Ruiz, L.; Platret, G.; Massieu, E., Ehrlacher, A. The use of thermal analysis in assessing the effect of temperature on a cement paste. Cem. Concr. Res. 2005, 35 (3), 609-13. Ben‐Dor, E. Banin, A. Determination of organic matter content in arid‐zone soils using a simple “loss‐on‐ignition” method. Communications in Soil Science & Plant Analysis. 1989, 20 (15-16), 1675-95. Frost, R.; Hales, M., Martens, W. Thermogravimetric analysis of selected group (II) carbonateminerals—Implication for the geosequestration of greenhouse gases. Journal of Thermal Analysis and Calorimetry. 2008, 95 (3), 999-1005. Vital, M.; Dignum, M.; Magic-Knezev, A.; Ross, P.; Rietveld, L., Hammes, F. Flow cytometry and adenosine tri-phosphate analysis: alternative possibilities to evaluate major bacteriological changes in drinking water treatment and distribution systems. Water Res. 2012, 46 (15), 4665-76. Norton, C. D. LeChevallier, M. W. A pilot study of bacteriological population changes through potable water treatment and distribution. Appl. Environ. Microbiol. 2000, 66 (1), 268-76. Van Nevel, S.; Koetzsch, S.; Proctor, C. R.; Besmer, M. D.; Prest, E. I.; Vrouwenvelder, J. S.; Knezev, A.; Boon, N., Hammes, F. Flow cytometric bacterial cell counts challenge conventional heterotrophic plate counts for routine microbiological drinking water monitoring. Water Res. 2017, 113, 191-206. Berney, M.; Vital, M.; Hülshoff, I.; Weilenmann, H.-U.; Egli, T., Hammes, F. Rapid, cultivationindependent assessment of microbial viability in drinking water. Water Res. 2008, 42 (14), 40108. Ward, D. M.; Weller, R., Bateson, M. M. 16S rRNA sequences reveal numerous uncultured microorganisms in a natural community. Nature. 1990, 345 (6270), 63-5. Eagon, R. G. Pseudomonas natriegens, a marine bacterium with a generation time of less than 10 minutes. J. Bacteriol. 1962, 83 (4), 736-7. Pade, C. Guimaraes, M. The CO2 uptake of concrete in a 100 year perspective. Cem. Concr. Res. 2007, 37 (9), 1348-56.



596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645

56. 57. 58. 59.

60. 61.

62.

63. 64. 65. 66.

67.

68. 69.

70.

71. 72. 73.

74.

75.

Page 26 of 33

Rittmann, B., Detachment from biofilms. In: Characklis WG, Wilderer PA, editors. Structure and function of biofilms. 1989, New York: John Wiley & Sons. p. 49-58. Bester, E.; Wolfaardt, G. M.; Aznaveh, N. B., Greener, J. Biofilms’ role in planktonic cell proliferation. Int. J. Mol. Sci. 2013, 14 (11), 21965-82. Picioreanu, C.; Van Loosdrecht, M. C., Heijnen, J. J. Two-dimensional model of biofilm detachment caused by internal stress from liquid flow. Biotechnol. Bioeng. 2001, 72 (2), 205-18. Bester, E.; Kroukamp, O.; Wolfaardt, G. M.; Boonzaaier, L., Liss, S. N. Metabolic differentiation in biofilms as indicated by carbon dioxide production rates. Appl. Environ. Microbiol. 2010, 76 (4), 1189-97. Hoigne, J. Bader, H. Ozonation of water: selectivity and rate of oxidation of solutes. Ozone Sci. Eng. 1979, 1 (1), 73-85. Alexander, J.; Knopp, G.; Dötsch, A.; Wieland, A., Schwartz, T. Ozone treatment of conditioned wastewater selects antibiotic resistance genes, opportunistic bacteria, and induce strong population shifts. Sci. Total Environ. 2016, 559, 103-12. Chiao, T.; Clancy, T. M.; Pinto, A. J.; Xi, C., Raskin, L. Differential Resistance of Drinking Water Bacterial Populations to Monochloramine Disinfection. Environ. Sci. Technol. 2014, 48 (7), 4038-47. Botzenhart, K.; Tarcson, G., Ostruschka, M. Inactivation of bacteria and coliphages by ozone and chlorine dioxide in a continuous flow reactor. Water Sci Technol. 1993, 27 (3-4), 363-70. Wayne, L. G. Gross, W. M. Base composition of deoxyribonucleic acid isolated from mycobacteria. J. Bacteriol. 1968, 96 (6), 1915-9. Mah, T.-F. C. O'Toole, G. A. Mechanisms of biofilm resistance to antimicrobial agents. Trends Microbiol. 2001, 9 (1), 34-9. Berney, M.; Weilenmann, H.-U.; Ihssen, J.; Bassin, C., Egli, T. Specific growth rate determines the sensitivity of Escherichia coli to thermal, UVA, and solar disinfection. Appl. Environ. Microbiol. 2006, 72 (4), 2586-93. Liu, R.; Zhu, J.; Yu, Z.; Joshi, D.; Zhang, H.; Lin, W., Yang, M. Molecular analysis of long-term biofilm formation on PVC and cast iron surfaces in drinking water distribution system. J. Environ. Sci. (China). 2014, 26 (4), 865-74. Wang, H.; Hu, C.; Hu, X.; Yang, M., Qu, J. Effects of disinfectant and biofilm on the corrosion of cast iron pipes in a reclaimed water distribution system. Water Res. 2012, 46 (4), 1070-8. Feazel, L. M.; Baumgartner, L. K.; Peterson, K. L.; Frank, D. N.; Harris, J. K., Pace, N. R. Opportunistic pathogens enriched in showerhead biofilms. Proc. Natl. Acad. Sci. U. S. A. 2009, 106 (38), 16393-8. Becerra-Castro, C.; Macedo, G.; Silva, A. M.; Manaia, C. M., Nunes, O. C. Proteobacteria become predominant during regrowth after water disinfection. Sci. Total Environ. 2016, 573, 313-23. Müller, S. Nebe-von-Caron, G. Functional single-cell analyses: flow cytometry and cell sorting of microbial populations and communities. FEMS Microbiol. Rev. 2010, 34 (4), 554-87. von Gunten, U.; Bruchet, A., Costentin, E. Bromate formation in advanced oxidation processes. J Am Water Works Assoc. 1996, 88 (6), 53. Gerrity, D.; Pisarenko, A. N.; Marti, E.; Trenholm, R. A.; Gerringer, F.; Reungoat, J., Dickenson, E. Nitrosamines in pilot-scale and full-scale wastewater treatment plants with ozonation. Water Res. 2015, 72, 251-61. Liu, S.; Gunawan, C.; Barraud, N.; Rice, S. A.; Harry, E. J., Amal, R. Understanding, monitoring, and controlling biofilm growth in drinking water distribution systems. Environ. Sci. Technol. 2016, 50 (17), 8954-76. Lautenschlager, K.; Hwang, C.; Liu, W.-T.; Boon, N.; Köster, O.; Vrouwenvelder, H.; Egli, T., Hammes, F. A microbiology-based multi-parametric approach towards assessing biological stability in drinking water distribution networks. Water Res. 2013, 47 (9), 3015-25.


Page 27 of 33

646 647


76.

Lin, W.; Yu, Z.; Zhang, H., Thompson, I. P. Diversity and dynamics of microbial communities at each step of treatment plant for potable water generation. Water Res. 2014, 52, 218-30.



Page 28 of 33

Cumulative 75% + 1.5*IQR 25-75% 25% - 1.5*IQR Median Outliers

5 (Outlet basin)

Inlet Inlet basin

1

2

3

4

5

6

7

Outlet basin

Outlet

Ozone gas Sample Line

Figure 1. Box and whisker plot for ozone CT (average concentration × contact time) for each chamber across contactors sampled on eight days. Cumulative ozone exposure is represented by orange triangles (mean values measured on eight days; error bars represent standard deviations). Water from an inlet basin flows through seven chambers and then into an outlet basin. Ozone gas is added into chambers 2 and 3. The cumulative residence time in a contactor is approximately 10 min. The width of chamber 3 is larger than the widths of the other chambers to provide additional disinfectant contact time.


Page 29 of 33


75% + 1.5*IQR

a

25-75% 25% - 1.5*IQR Median Outliers

(200 - 3,850 CFU/mL)

b

(1,200 - 3,750 cells/mL)

Figure 2. Box and whisker plots representing viable cell concentrations of lime-softened water (sample line 1) and in different chambers during ozone exposure monitored by (a) HPC on six days, and (b) flow cytometry on four days. HPC data are reported as the average of duplicate dilutions plated in triplicate (n = 6 counts per sample except for chamber 1 and chamber 4 samples collected on May 13, 2016, which had only 5 and 4 counts, respectively). The numbers in parentheses indicate the range of cell concentrations at sample line 5.



a

Page 30 of 33

b Water in chamber 2

Water in chamber 2

Water in chamber 3

Water in chamber 3

Chamber 2 water in control

Chamber 2 water in control

Time after shutdown (h)

Time after shutdown (h)

Figure 3. Average concentrations of viable cells monitored by (a) HPC and (b) flow cytometry before, 2 h after, and 9.5 h after contactor shutdown in water sampled from chambers 2 and 3, and water sampled from chamber 2 stored separately as an experimental control. Error bars represent standard deviation of duplicate dilutions plated in at least triplicate (except the control at 9.45 h in which duplicate plates were counted) or range of duplicate flow cytometry measurements.


Page 31 of 33


Sludge

OTU0010 OTU0005 Diffuser OTU0004 2R 2R 3L

3R 3R

OTU0016

4R 4R 5R OTU0013 OTU0001 OTU00012 OTU00014

Figure 4. Non-metric multidimensional scaling (NMDS) plot of ϴyc distance ordinations showing differences in community structure for total bacteria in biofilm and sludge samples. Biofilm samples were collected from chambers 2-5 on right or left baffle walls (indicated by R or L). There was significant (p = 0.017) separation between bacterial communities in biofilm on baffle walls located before (green points) and after (blue points) a dead zone. Of the top 20 most abundant OTUs across the samples, the OTUs with relative abundances that increased or decreased significantly (p < 0.01) along NMDS axis 1 are shown in red. OTU00010, OTU0005, OTU0004, OTU00016, OTU00013, OTU0001, OTU0012, and OTU0014 were classified as Brevundimonas, Methylophilus, Limnobacter, Sphingomonas, Sphingopyxis, Gp4, Hyphomicrobiaceae, and Sphingomonadaceae, respectively.



Page 32 of 33

` Legionella spp. Mycobacterium spp.

2R

2R

1

3L Diffuser Sludge

2

3

3R

4

3R

5

4R

6

4R

5R

7

Inlet Inlet basin

2R

3L

3R

4R

5R

Diffuser

Outlet basin

Outlet

Sludge Sample Line

Figure 5. Relative abundance of 20 Legionella OTUs (red) and 4 Mycobacterium OTUs (blue) in biofilm samples collected from chambers 2-5 on right or left baffle walls (indicated by R or L), a diffuser (Diffuser), and sludge (Sludge) from one contactor. Chamber numbers 1-7 are shown at the top of the contactor schematic.


160,000

80,000 Page 33 Environmental of 33 Science & Technology Membrane intact cells/mL

40,000 20,000 10,000 5,000 2,500 1,000

ACS Paragon Plus Environment Ozone gas

Microbial Dynamics in Multi-Chamber Ozone ... - ACS Publications

Recommend Documents