Quantification and identification of particle-associated bacteria in unchlorinated drinking water from three treatment plants by cultivation-independent methods

Accepted Manuscript Quantification and identification of particle-associated bacteria in unchlorinated drinking water from three treatment plants by cultivation-independent methods G. Liu, F.Q. Ling, A. Magic-Knezev, W.T. Liu, J.Q.J.C. Verberk, J.C. Van Dijk PII:

S0043-1354(13)00306-0

DOI:

10.1016/j.watres.2013.03.058

Reference:

WR 9880

To appear in:

Water Research

Received Date: 16 November 2012 Revised Date:

2 March 2013

Accepted Date: 31 March 2013

Please cite this article as: Liu, G., Ling, F.Q., Magic-Knezev, A., Liu, W.T., Verberk, J.Q.J.C., Van Dijk, J.C., Quantification and identification of particle-associated bacteria in unchlorinated drinking water from three treatment plants by cultivation-independent methods, Water Research (2013), doi: 10.1016/ j.watres.2013.03.058. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

Quantification and identification of particle-associated bacteria in unchlorinated drinking

2

water from three treatment plants by cultivation-independent methods

3

G. Liu1 *, F. Q. Ling2, A. Magic-Knezev3, W. T. Liu2, J.Q.J.C.Verberk1, J.C. Van Dijk1

4

1. Section Sanitary Engineering, Department of Water Management, Faculty of Civil

5

Engineering and Geosciences, Delft University of Technology, PO BOX 5048, 2600 GA Delft,

6

the Netherlands E-mail: [email protected]

7

2. Department of Civil and Environmental Engineering, University of Illinois Urbana-

8

Champaign, 205 N. Mathews Ave., Urbana, Illinois 61801, U.S.A.

9

3. Het Water Laboratorium, PO BOX 734, 2003 RS Haarlem, the Netherlands

SC

M AN U

10 11

TE D

12

Corresponding author:

14

Gang Liu

15

Email: [email protected]

16

Tel: 0031 15 2785457\ 0031 6 41866671

AC C

18

EP

13

17

RI PT

1

1

ACCEPTED MANUSCRIPT

Abstract

20

Water quality regulations commonly place quantitative limits on the number of organisms

21

(e.g., heterotrophic plate count and coliforms) without considering the presence of multiple

22

cells per particle, which is only counted as one regardless how many cells attached. Therefore,

23

it is important to quantify particle-associated bacteria (PAB), especially cells per particle. In

24

addition, PAB may house (opportunistic) pathogens and have higher resistance to disinfection

25

than planktonic bacteria. It is essential to know bacterial distribution on particles. However,

26

limited information is available on quantification and identification of PAB in drinking water.

27

In the present study, PAB were sampled from the unchlorinated drinking water at three

28

treatment plants in the Netherlands, each with different particle compositions. Adenosine

29

triphosphate (ATP) and total cell counts (TCC) with flow cytometry were used to quantify the

30

PAB, and high-throughput pyrosequencing was used to identify them. The number and

31

activity of PAB ranged from 1.0-3.5×103 cells ml-1 and 0.04-0.154 ng l-1 ATP. There were

32

between 25 and 50 cells found to be attached on a single particle. ATP per cell in PAB was

33

higher than in planktonic bacteria. Among the identified sequences, Proteobacteria were

34

found to be the most dominant phylum at all locations, followed by OP3 candidate division

35

and Nitrospirae. Sequences related to anoxic bacteria from the OP3 candidate division and

36

other anaerobic bacteria were detected. Genera of bacteria were found appear to be consistent

37

with the major element composition of the associated particles. The presence of multiple cells

38

per particle challenges the use of quantitative methods such as HPC and Coliforms that are

39

used in the current drinking water quality regulations. The detection of anoxic and anaerobic

40

bacteria suggests the ecological importance of PAB in drinking water distribution systems.

41

Keywords: particle associated bacteria (PAB), drinking water, adenosine triphosphate (ATP),

42

flow cytometry, pyrosequencing

AC C

EP

TE D

M AN U

SC

RI PT

19

2

ACCEPTED MANUSCRIPT

43

1. Introduction When distributing drinking water, the regrowth of bacteria and other organisms may occur

45

and lead to water quality deterioration (Ridgway and Olson, 1982; Van Der Kooij, 2000).

46

Depending on the source water and water treatment, more or less planktonic bacteria (PB), as

47

well as particle-associated bacteria (PAB) and biodegradable compounds, are present in the

48

treated water. They enter the drinking water distribution system (DWDS) and may serve as

49

“seeds” for regrowth. The generation of PAB during drinking water treatment is caused by the

50

action of particles as the site for attachment and growth of bacteria (Gregory, 2005;

51

Winkelmann and Harder, 2009). It has been reported that PAB represent a small number

52

compared to the PB population in treated water (Brazos and O'Connor, 1996). Nevertheless,

53

the PAB that may pass through or be generated during treatment have been considered an

54

important source of bacteria entering the drinking water distribution systems both for bacterial

55

regrowth (Camper et al., 1986) and bacteria in accumulated loose deposits (Gauthier et al.,

56

1999; Vreeburg et al., 2008; Liu et al., 2013). PAB have been detected in 41.4% of the

57

samples of granular activated carbon filtered water at water treatment plants (Camper et al.,

58

1986), and in 17% of samples collected from fire hydrants in drinking water distribution

59

systems and water well outlets (Ridgway and Olson, 1981).

60

Since the attachment and growth of bacteria can lead to biofilm formation on particles

61

(Winkelmann and Harder, 2009), a major concern regarding PAB is their resistance to

62

disinfection (Brazos and O'Connor, 1996; Dietrich et al., 2009; Hess-Erga et al., 2008;

63

Hoadley and Gould, 1977; Lin et al., 2010; Wojcicka et al., 2008). PAB have been proven to

64

be more resistant to disinfection by chlorine (Ridgway and Olson, 1982), ozone (Hess-Erga et

65

al., 2008) and ultraviolet (UV) (Mamane and Linden, 2006; Wu et al., 2005) than PB are. As a

66

result, the regrowth or survival of pathogens in drinking water distribution systems may be

67

enhanced (Herson et al., 1987). Considering the disinfection resistance of PB and PAB, the

AC C

EP

TE D

M AN U

SC

RI PT

44

3

ACCEPTED MANUSCRIPT

nutrient limitation approach (Van der Kooij, 1992) to produce biologically stable drinking

69

water is likely to control the regrowth of both PB, PAB and bacteria in biofilms attached to

70

pipe walls.

71

Another concern regarding PAB is the potential underestimation of bacterial numbers because

72

no matter how many bacteria have been attached to one particle, they will be counted as one

73

by traditional culture methods (Camper et al., 1986). Water quality regulations commonly

74

place quantitative limits on the number of organisms (e.g., heterotrophic plate count and

75

coliforms) and particle densities (e.g., turbidity), resulting in a substantial underestimation of

76

the PAB bacteria present (Dietrich et al., 2007). In addition, PAB may house (opportunistic)

77

pathogens and the dose of microbes may differ significantly if PAB rather than PB are

78

ingested, thereby increasing the potential risk to customers. For instance, Herson et al. (1991)

79

found that a large number of coliforms added to particle-containing drinking water could not

80

be reflected by plate counting because they accumulated as PAB.

81

All the above-mentioned studies have improved knowledge of the importance of PAB in

82

drinking water. However, limited studies on PAB in drinking water have been conducted,

83

most of which applied cultivation-dependent methods (Camper et al., 1985; Ridgway and

84

Olson, 1982; Wu et al., 2005) or microscopic observations (Brazos and O'Connor, 1996).

85

Considerable bias and underestimation may be introduced by applying these methods.

86

Consequently, PAB in drinking water have been poorly documented. Cultivation-independent

87

techniques for bacterial quantification and identification offer new posibilities to reevaluate

88

PAB in drinking water. The main goals of this study were to determine the presence of PAB

89

in treated water from Dutch drinking water treatment plants by cultivation-independent

90

methods, (i.e., use total cell count (TCC) with flow cytometry to quantify attached bacteria

91

and adenosine triphosphate (ATP) to quantify activity), and use high-throughput

92

pyrosequencing to identify the PAB. This study was undertaken to understand what PAB

AC C

EP

TE D

M AN U

SC

RI PT

68

4

ACCEPTED MANUSCRIPT

levels in drinking water are, what the fraction of PAB in the total bacteria levels is; how many

94

bacteria are associated with a single particle; and what the PAB community is, and if the PAB

95

community has a relation to the characteristics of the particles from different water treatment

96

plants.

97

2. Materials and Methods

98

2.1 Description of water treatment plants

RI PT

93

Three drinking water treatment plants with different particle compositions in their treated

100

water were selected: a treatment plant using artificial recharge and recovery (ARR) with river

101

water as source water (TP1), and two groundwater treatment plants (TP2, TP3). TP1 takes

102

source water from the Meuse River. The source water, after pre-treatment, is transported over

103

30 km to a dune area of natural lakes, where it recharges the groundwater. After an average

104

residence time of 2 months, the water is abstracted from the dunes. Abstracted ARR water is

105

post-treated by softening, powdered activated carbon, aeration, rapid sand filtration, and slow

106

sand filtration before being pumped into the distribution system.

107

At TP2 anoxic groundwater is treated by aeration and rapid sand filtration, and afterwards fed

108

to the distribution system. At TP3, after abstraction, the groundwater is treated by aeration,

109

filtration, softening, carry-over filtration, activated carbon filtration and UV disinfection. The

110

treated groundwater contains somewhat higher levels of iron, manganese and ammonia

111

concentrations than at TP1. The concentrations of these elements are also different between

112

TP2 and TP3 due to the different treatments applied at the two treatment plants. The quality of

113

treated water is summarized in Table S1 in the supplementary data.

114

AC C

EP

TE D

M AN U

SC

99

2.2 Sampling

115

The sampling spots are located at the treatment plants just before the water enters the

116

distribution system. PAB were collected with a specially designed multiple-particle filtration

5

ACCEPTED MANUSCRIPT

system (MuPFiS, Figure 1). Each line of MuPFiS consists of 47mm Swinnex filter holder

118

followed by a flow meter. Multiple samples can be collected at the same time, and with the

119

recorded water volume, the concentration of quantified PAB can be calculated. Particles were

120

pre-concentrated by filtering approximately 200 liters of water through 1.2µm pore-size glass

121

fiber filters (Whatman, 1822-047). Different pore size filters (1-10µm) have been used to

122

collect PAB in water systems (Power and Nagy 1999; Riemann and Winding 2001; Zhang,

123

Liu et al. 2007; Parveen, Reveilliez et al. 2011). In drinking water environment, previous

124

researches done by scanning electron microscope (Ridgway and Olson 1982) and differential

125

size filtration experiments (Azam and Hodson 1977) have demonstrated that single bacteria

126

are usually less than 1.0µm in diameter. Although bigger particles or bacteria may exist in

127

source water, the multiple treatment processes applied can remove them efficiently (Brazos

128

and O’Connor 1996). Brazos and O’Connor (1996) did not collect PAB successfully from

129

drinking water by using 3µm filters. In this study, a pore size of 1.2µm was selected. On one

130

hand, it is bigger than reported diameter of single bacteria and will not lead to losing too

131

much small bio-aggregates; on the other hand, 1.2µm filters are commonly used for

132

suspended solids sampling which makes it possible to interpret and combine the biological

133

analysis with physiochemical analysis.

134

Three samples were taken by running the MuPFiS on the same day of the week in three

135

different weeks for PAB quantification. On each sampling day, water samples were taken

136

before filtration at the MuPFiS-connected points. For pyrosequencing analysis, triplicate

137

samples were taken by one run of MuPFiS (finished within one day for all locations). A

138

particle counter (Met one, 32 channels, 1-100µm) was run parallel to MuPFiS. Every filtration

139

run was standardized to 3 hours. A particle counter was run at each sampling point for weeks

140

to monitor the particle load in the treated water.

141

AC C

EP

TE D

M AN U

SC

RI PT

117

2.3 Physiochemical characteristics of collected particles 6

ACCEPTED MANUSCRIPT

The physiochemical characteristics of collected particles were studied by scanning electron

143

microscopy (SEM) coupled with energy dispersive spectroscopy (EDS). SEM-EDS was used

144

to obtain high resolution images and the elemental composition. The JEOL JSM-840A

145

scanning electron microscope is configured with secondary and backscattered electron

146

detectors as well as an energy dispersive X-ray spectrometer. The working distance was 7-39

147

mm for high–resolution imaging and 39 mm for EDS analysis and element mapping

148

(Echeverría et al., 2009).

SC

149

RI PT

142

2.4 PAB quantification

The filter with pre-concentrated PAB was submerged upside down in 5 ml autoclaved tap

151

water with glass beads immediately after filtration. Samples were kept in a cool box and

152

transported to the laboratory within two hours. Bacteria were detached from the particles by

153

low energy ultrasonic treatment for 3 mins (Branson ultrasonic water bath, 43 kHz) as

154

described by (Magic-Knezev and van der Kooij, 2004). Obtained suspensions were used for

155

further analyses. Cultivation-independent methods, Adenosine triphosphate (ATP) and total

156

cell counts (TCC) by flow cytometry were applied to quantify collected PAB in the

157

suspension. ATP was measured as previously described (Magic-Knezev and van der Kooij,

158

2004); TCC was measured by C6 flow cytometer (BD Accuri C6, United States); damaged

159

cells and intact cells were distinguished, as described by Hammes et al. (2008). Both ATP and

160

TCC analyses were done at Het Water Laboratorium in Haarlem, the Netherlands.

161

The ATP and TCC values obtained from PAB are defined as associated ATP (A-ATP) and

162

associated TCC (A-TCC). Based on ATP and TCC results, ATP per cell was calculated for

163

water samples and PAB samples according to the equation used by Berney et al.(2008), that is,

164

dividing the cell ATP by the intact cell number. In this study, ATP was measured as total ATP.

165

As for PAB, no free ATP was determined because the PAB were collected by filtration. For

166

PB, it has been demonstrated that the unchlorinated drinking water samples in the Netherlands

AC C

EP

TE D

M AN U

150

7

ACCEPTED MANUSCRIPT

167

do not contain significant amounts of free ATP (Van der Kooij, 1992; Van der Wielen and

168

Van der Kooij, 2010).

ATP per cell (10 −16 g cell −1 ) =

ATP concentration ( ng l −1 ) intact cell concentration (cells l −1 )

(1)

RI PT

169

170

To quantify bacteria attached to particles, the average number of cells per particle was

171

calculated from A-TCC and particle count measurements as follows:

Cells per particle ( cells particle −1 ) =

A − TCC ( cells ml −1 ) particle count ( particles ml −1 )

SC

172

(2)

Considering PAB as a form of biofilm on a sphere, A-ATP and particle counts were used to

174

calculate the ATP of PAB (pg per cm2) using Equation (3):

175

ATP ( pg cm−2 ) =

M AN U

173

A − ATP N ×S

(3)

Where A-ATP represents measured PAB ATP results; N the number of particles, and S the

177

surface area per particle. An average diameter of 1.5µm was used based on the particle size

178

distribution in tested water, where most particles had a diameter between 1.2µm and 2µm

179

(results not shown).

EP

180

TE D

176

2.4 454 pyrosequencing

Suspended PAB samples were further processed to study their bacterial diversity. DNA was

182

extracted from the suspension using a chemical and enzymatic DNA extraction protocol as

183

described by Hong et al. (2010) and was amplified with bacterium-specific forward primer

184

27F and reverse primer 534R (Hong et al., 2010). DNA extraction was done at KWR Water

185

Cycle Research Institute. The 454 pyrosequencing was carried out with a 454 Life Sciences

186

GS FLX series genome sequencer (Roche, Switzerland). The sequences were trimmed

187

(resulting in an average sequence length of 230 bp). Merged alignments of the sequences

AC C

181

8

ACCEPTED MANUSCRIPT

aligned via the infernal aligner from the Ribosomal Database Project (RDP) pyrosequencing

189

pipeline (http://pyro.cme.msu.edu/) and the NAST alignment tool from Greengenes were

190

obtained via software developed by the biotechnology center at the University of Illinois (UI)

191

(http://acai.igb.uiuc.edu/bio/merge-nast-infernal.html). The RDP Classifier was used for

192

taxonomical assignments of the aligned 454 pyrosequences at the 95% confidence level. Both

193

PCR amplification and pyrosequencing were performed at the UI biotechnology center. The

194

total PAB communities from the three treatment plants were analyzed for the number of

195

operational taxonomic units (OTUs), and species richness by using the DOTUR program

196

(Hong et al., 2010).

M AN U

SC

RI PT

188

197

3. Results and Discussion

198

3.1 Physiochemical characterization of PAB

Although it was not quantifiable, SEM pictures confirmed the finding of PAB and multiple

200

cells attached on a single particle (Figures 2A, 2B and 2C). Different particle sizes and

201

morphologies were observed at the three treatment plants. EDS elemental analysis showed

202

that particles mainly consisted of carbon (C), oxygen (O), silicon (Si), sodium (Na), calcium

203

(Ca), iron (Fe), and manganese (Mn) (Figure 2D). The results complied with reported findings

204

on particles in drinking water distribution systems (Gauthier et al., 2001; Matsui et al., 2007;

205

Verberk et al., 2006; Vreeburg et al., 2008).

206

Compared to TP2 and TP3, TP1 particles were smaller and contained mainly C, O and Si.

207

This may be due to the fact that TP1 used ARR surface water as source water and larger

208

particles were removed by subsequent filtration steps applied at the plant. At TP2, high

209

percentages of Ca, Fe and Mn were found. The results were as expected, since the

210

groundwater is treated with conventional treatments of aeration and rapid sand filtration. The

211

Ca, Fe, and Mn concentrations in the treated water were also higher than that of TP1 and TP3

212

(Table S1, supplementary data). At TP3, although the groundwater source is the same as at

AC C

EP

TE D

199

9

ACCEPTED MANUSCRIPT

TP2, different particles were collected. Ca, Fe and Mn were not found, instead the highest

214

concentration of carbon was found. The difference between TP2 and TP3 can be related to the

215

more advanced treatments applied in TP3, such as the softening, multiple filtration steps and

216

additional activated carbon filtration (ACF). The highest carbon concentration may be due to

217

the last treatment step (ACF) which may release carbon fines from the activated carbon media

218

into the treated water. 3.2 Quantification of PAB

SC

219

RI PT

213

3.2.1 A-TCC and A-ATP

221

ATP and TCC were analyzed for both water and PAB samples (Figure 3 and Figure 4). The

222

TCC of the produced water from the three pumping stations ranged from 0.45 × 105 cells ml-1

223

to 1.65 × 105 cells ml-1, while ATP ranged from 1 ng l-1 to 6 ng l-1. Both TCC and ATP results

224

are typical values for public drinking water without disinfectant residuals (Hammes et al.,

225

2008; Van der Wielen and Van der Kooij, 2010). For PAB, A-TCC results ranged from 1.0 ×

226

103 cells ml-1 to 3.5 × 103 cells ml-1, A-ATP ranged from 0.04 ng l-1 to 0.154 ng l-1. A-TCC

227

and A-ATP accounted for less than 2% of that detected in the bulk water phase. One potential

228

risk of PAB over PB, as mentioned in the introduction, is that it can be protected from

229

disinfection. Therefore, although it forms less than 2% of PB, it may serve as seed feed to

230

drinking water distribution system that can form biofilm on pipe wall and grow during

231

distribution in bulk water and sediments. In the distribution system where disinfectant

232

residuals applied, growth of PB can be controlled by the residuals, whereas, persistence,

233

growth and accumulation of PAB will not be limited.

234

On the other hand, the amount of PAB highly depends on the particle load (Liu et al., 2013).

235

In the present study, the selected treatment plants produced high quality and low particle load

236

drinking water (Table S1). The monitoring of particle counts in drinking water distribution

237

system of TP1 found that the particle count increased from 20 # ml-1 at treatment plant to

AC C

EP

TE D

M AN U

220

10

ACCEPTED MANUSCRIPT

1,500-2,000 # ml-1 at customers’ taps. The monitoring of PB showed stable results, as the

239

water is produced with high biological stability (AOC=5.8 µg cl-1, Table S1). Particles in

240

distribution network may come from treatment plant such as filter material of sand and carbon,

241

or generated during drinking water distribution such as corrosion. The physiochemical

242

properties of different particles will lead to different communities, as corroborated by the

243

present study (will be further discussed in the following section). Furthermore, PAB may also

244

be accumulated in distribution systems as loose deposits and be released to bulk water during

245

hydraulic peaks (Gauthier et al., 1999; Vreeburg et al., 2008; Liu et al., 2013), while, PB is

246

rapid washed out of distribution system (Boe-Hansen et al., 2002).

247

The lowest A-ATP and A-TCC were found at TP1, which is reasonable since, as mentioned

248

above, TP1 is using ARR and has extended post-treatment, including slow sand filtration. The

249

highest A-ATP was found at TP2, where only a conventional groundwater treatment is used.

250

Results also demonstrated that, for water samples, intact cells account for 95%, 92% and 85%

251

for TP1, TP2 and TP3, respectively. The percentage decreased to 75%, 85% and 72% for

252

PAB samples. The lower intact cell ratio of PAB can be explained by the attachment and

253

accumulation of damaged cells to extracellular polymeric substances (EPS) that surround

254

living cells (Liu et al., 2004).

255

Though A-ATP and A-TCC results were relatively low compared to the bulk water, sufficient

256

amounts of PAB were sampled successfully for further analysis. In a previous study, using

257

heterotrophic plate counts (HPC), a value of 0.07 to 0.15 CFU ml-1 cultivable bacteria were

258

detected, and the value seemed to depend on the particle size (Lin et al., 2010). However,

259

direct comparison of results obtained between two methods is difficult because the number of

260

cultivable bacteria represents only a small percentage of the total number of bacteria

261

(Hammes et al., 2008; Staley and Konopka, 1985).

AC C

EP

TE D

M AN U

SC

RI PT

238

11

ACCEPTED MANUSCRIPT

The application of direct microscopy cell (DC) counting found few PAB in the treated water

263

(Brazos and O'Connor, 1996). In their research, PAB was quantified by comparing DCs in

264

water before and after filtering through 3µm filters. The failure of quantifying PAB may be

265

related to quantifying the difference in bacterial numbers before and after filtration instead of

266

direct analysis of PAB retained on the filter. As mentioned, PAB only accounted for less than

267

2% of biomass in the water sample and the accuracy of most bacteria quantification methods

268

make it difficult to detect such small differences.

269

Although ATP has been widely used in drinking water research to quantify bacterial activity

270

in bulk water and biofilm, it has not been used previously to quantify PAB in water systems.

271

The ATP per surface area of particles was calculated using Equation (3). In the present study,

272

9300 to 49,700 pg cm-2 ATP were derived for the PAB samples. The ATP values of pipe wall

273

biofilm were reported to be from 100 to 4000 pg cm-2 (Lehtola et al., 2004; Lehtola et al.,

274

2006; Yu et al., 2010). It is difficult to compare these results since they were obtained from

275

different systems. It should be noted that unlike pipe wall biofilms, which are colonized on

276

the pipe wall surface, PAB are suspended in the bulk water during drinking water distribution

277

that can reach customers’ taps and consumed by customers. The higher ATP per surface area

278

and better mobility of PAB suggest that PAB, depending on the microbial community, may

279

have a higher potential health risk to customers than pipe wall biofilm. Therefore, PAB

280

require more research attention than they have had thus far.

281

3.2.2 Multiple cells per particle

282

Cells per particle were calculated using A-TCC (Figure 3) and particle counting results (Table

283

1) by Equation (2). On average, 25-50 cells per particle were found (Table 1). The values are

284

higher than the previously reported number counted by microscopy counting, which is 1.7-17

285

cells per particle (Brazos and O'Connor, 1996; Ridgway and Olson, 1981). It has been

AC C

EP

TE D

M AN U

SC

RI PT

262

12

ACCEPTED MANUSCRIPT

suggested that particles with five or more attached cells should be considered important

287

(Brazos and O'Connor, 1996).

288

It should be noted that, although TP3 had a somewhat lower number of cells per particle, the

289

total amount of PAB in TP3 was higher due to a higher particle load. TP1 was just the

290

opposite of TP3. TP2, on the other hand, was characterized by high particles and high cell

291

counts per particle. It was reported that more particles could offer more surface area for

292

bacterial cells to attach and form biofilms (Gregory, 2005), and particles with biofilms could

293

promote particle aggregation (Paris et al., 2009). Thus, it is likely that high particle load in

294

combination with high cell numbers per particle could result in a higher level of deposition

295

and accumulation of particulate matter in the drinking water distribution system. Particle

296

generating processes during water distribution such as corrosion, flocculation, aggregation

297

and biofilm detachment may increase the number of PAB at the tap.

298

The water quality regulations commonly place quantitative limits on the number of cultivable

299

organisms (e.g., coliforms) and particle densities (e.g., turbidity). However, these regulations

300

do not take into account whether bacterial cells are present in the water as PB or PAB

301

(Dietrich et al., 2007). No matter how many bacteria are associated with a particle, the cells

302

will likely result in only a single colony if measured by routine HPC method (Camper et al.,

303

1986) and will be counted as single cell when subjected to flow cytometry without further

304

processing. Microscopic methods may be able to determine PAB, but are cumbersome and

305

subjective. As a result, the presence of PAB in treated water will result in the underestimation

306

of both specific and total bacterial cell counts in most methods, especially for the high particle

307

load drinking water systems. This will give direct influence to bacteria injected to consumers.

308

Using total coliforms as an example, it has been reported that adding coliforms in drinking

309

water containing particles could not be correctly counted in the water sample, due to the

310

association of the added coliforms with suspended particles (Herson et al., 1991). The further

AC C

EP

TE D

M AN U

SC

RI PT

286

13

ACCEPTED MANUSCRIPT

attachment of cells to particles during water distribution will lead to increase of particle size,

312

which will change the surface and settling properties, disinfectant resistance and micro-

313

environments (oxygen diffusion) inside the particle.

314

3.2.3 ATP content in planktonic bacteria and attached bacteria

315

The average ATP per cell was calculated based on obtained total ATP and intact cell count

316

results for both water and PAB samples according to Formula 1. It was found that the ATP

317

per cell ranged from 0.21 × 10-16 g cell-1 to 0.33 × 10-16 g cell-1 for PB and from 0.38 × 10-16 g

318

cell-1 to 0.54 × 10-16 g cell-1 for PAB (Table 1). The average ATP per cell values of PAB is

319

2.5 times higher than that of PB, suggesting that PAB could have a higher metabolic activity

320

(Magic-Knezev and van der Kooij, 2004). This may be due to the fact that nutrients

321

originating from treatment plant as particles can be used firstly and directly by PAB. The

322

results agreed with findings obtained in other water environments (DeBruyn and Sayler, 2009;

323

Lemarchand et al., 2006) (Table 2). The results obtained in the present study are similar to

324

that from drinking water biofilters in the Netherlands (Magic-Knezev and van der Kooij, 2004)

325

and tap water (Berney et al., 2008), but lower than that from different aquatic samples

326

(Hammes et al., 2010). This may be associated with the low nutrients concentration in treated

327

drinking water.

SC

M AN U

TE D

EP

3.3 Identification of PAB

AC C

328

RI PT

311

329

3.3.1 PAB diversity

330

A total of 2,988, 8,921 and 2,818 bacterial 16S rRNA gene sequences were obtained from

331

TP1, TP2 and TP3, respectively. The sequences (2,818-8,921 sequences) were significantly

332

larger than that of conventional cloning and sequencing methods (generally around 200

333

sequences, Kwon et al., 2011). The increased sequences made it possible to detect more

334

microorganisms (i.e., 207-495 operational taxonomic units (OTUs) at a 3% cutoff). These

14

ACCEPTED MANUSCRIPT

values are higher than those reported studies on samples from freshwater based on clone

336

libraries (less than 100 OTUs, Eichler et al., 2006). Compared to available pyrosequencing

337

studies, the number of OTUs observed in the present study were higher than those from

338

biofilm in water meters sampled from drinking water distribution systems (133 and 208 OTUs)

339

(Hong et al., 2010). The number of OTUs was lower than from membrane filtration systems

340

for drinking water production (1133-1731 OTUs) (Kwon et al., 2011). The Chao1 index

341

estimated 1470, 653, and 448 OTUs at a 3% cutoff for the samples from TP1, TP2, and TP3,

342

respectively. The highest bacterial diversity for PAB was found at TP1. Other nonparametric

343

diversity indices such as the Shannon index and Evenness gave similar results (Table 3).

344

3.3.2 PAB community composition

345

Figure 5 indicates the major phyla (>3% in total sequences) found in different treatment

346

plants. The results suggest that the three treatment plants have slightly different bacterial

347

community compositions. Proteobacteria were observed to dominate in all three locations,

348

ranging from 37% to 68%, and were represented by Alphaproteobacteria, Betaproteobacteria,

349

Deltaproteobacteria, and Gammaproteobacteria.

350

Proteobacteria) and Gammaproteobacteria (23% to 34%) dominated in all three plants. At

351

TP1, Deltaproteobacteria were found to account for 22% of Proteobacteria, whereas at TP2

352

and TP3 more Alphaproteobacteria (16 - 33%) were found.

353

Among the remaining phyla, OP3 candidate phylum (OP3) and Nitrospirae were also

354

commonly found at the three treatment plants. A high percentage of OP3 was found at TP1

355

(18.7%) and TP2 (13.6%). At TP3, only Proteobacteria had an abundance > 10%. Members

356

within the Planctomycetes, Cyanobacteria, Euryarchaeota and Acidobacteria were found at

357

one or more treatment plants with a percentage > 3%. Members of the Actinobacteria,

358

Bacteroidetes, Crenarchaeota, Chloroflexi, Gemmatimonadetes, GN02, NC10 and SBR1093

Betaproteobacteria (25% to 36% of

AC C

EP

TE D

M AN U

SC

RI PT

335

15

ACCEPTED MANUSCRIPT

were also found, but had an abundance < 3%. It should be noted that significant fractions of

360

sequences were assigned as unclassified, ranging from 6.8% (TP3) to 22.2% (TP1).

361

Little is known about the diversity of the PAB community in drinking water distribution

362

systems. The typical drinking water bacteria that have been reported correspond only to PB

363

and

364

Gammaproteobacteria are repeatedly found in drinking water environment (Eichler et al.,

365

2006; Hong et al., 2010; Kalmbach et al., 1997; Magic-Knezev et al., 2009; Mathieu et al.,

366

2009). Compared to Alphaproteobacteria, Betaproteobacteria and Gammaproteobacteria,

367

species from Deltaproteobacteria, were not detected or reported. Results from this

368

investigation confirm that, similar to the bulk water and biofilm studies by (Kormas et al.,

369

2010; Tokajian et al., 2005), Proteobacteria are of central importance in the drinking water

370

environment.

371

This is also the case for PAB. The subclasses of Proteobacteria have been reported to have a

372

different resistance to disinfectants (Mathieu et al., 2009). Disinfectants can promote or

373

suppress the proliferation of certain subclasses of Proteobacteria; for instance, increased

374

chlorine in distributed water results in an increased percentage of Gammaproteobacteria and

375

a decrease in Alphaproteobacteria (Mathieu et al., 2009). Hence, the presence of all four

376

classes of Proteobacteria in the present study may reflect the fact that no chemical

377

disinfectant is applied in the Netherlands.

378

It was noted that, compared to previous water and pipe wall biofilm studies in drinking water

379

and PAB studies in other water systems, a high percentage of bacteria belonging to candidate

380

division OP3 has been detected. OP3 was originally defined based on a single 16S rRNA gene

381

sequence obtained from obsidian pool sediment in Yellowstone National Park (Hugenholtz et

382

al., 1998). Bacteria belonging to OP3 were found to thrive in anoxic environments, such as in

383

marine sediment, fresh water lakes and aquifers (Glöckner et al., 2010; Kolinko et al., 2011).

wall

biofilm

bacteria.

Alphaproteobacteria,

Betaproteobacteria

and

AC C

EP

TE D

M AN U

SC

pipe

RI PT

359

16

ACCEPTED MANUSCRIPT

Regarding the habits and characteristics of OP3 bacteria, Glöckner’s group noted that OP3

385

bacteria were frequently detected in anoxic environments that are defined by redox cycling of

386

iron, manganese and other metals, and/or sulphur as major drivers of microbial activity.

387

Groundwater originates from anoxic aquifers with high concentrations of iron and manganese.

388

This high concentration may be associated with the high prevalence of OP3, compared to

389

what is found in marine and fresh water systems.

390

This same anoxic condition is present in the ARR dune area. The contact time of more than

391

two months with dune soils may allow soil bacteria to be introduced into the water phase.

392

Considering the aerobic environment of drinking water distribution systems, it is reasonable

393

to expect less/no presence of OP3 in the community study of distribution system water and

394

the biofilm. In the present study, the percentage of OP3 becomes significant (Figure 5) by

395

sampling particulate matter from the water phase indicating that the OP3 found in our treated

396

water originated from source water and passed through the treatment systems. Another

397

possibility is that in the aerobic drinking water environment, anoxic micro-environments exist

398

in or around the PAB structure. This hypothesis was supported by the detection of specific

399

anaerobic microorganisms, such as bacteria belonging to Rhodospirillales and Chromatiales,

400

and archaea belonging to Methanosarcinales. However, the particle size examined in the

401

present study is far too small to maintain a multiple micro-environment structure.

402

Evaluation at the genus level showed that 66 genera (genera accounting for more than 1%)

403

were found, and the identified genera are listed in Table 4. There were 53, 57, and 57 genera

404

found in TP1, TP2 and TP3, respectively. In this study, most of the genera were found in all

405

the treated waters. Unidentified genera belonging to OP3 GIF10 and Comamonadeceae were

406

found at all three treatment plants in a percentage higher than 3%. At TP2, Legionella,

407

Nitrospira, Gallionella, Nitrosomonas, Crenothrix and Thermodesulfavibrionaceae LCP-6

408

were also found in (relatively) high percentages (> 3%). Rhodospirillaceae, Leptolyngbya, a

AC C

EP

TE D

M AN U

SC

RI PT

384

17

ACCEPTED MANUSCRIPT

genus belonging to unclassified Comamonadeceae and unclassified OP3 GIF10 were

410

observed at TP3.

411

As shown in the Figure 2d, iron, manganese, ammonia, and sulfate were of significant

412

concentrations in the raw water of the three treatment plants, and in the treated water at TP2.

413

Therefore, not surprisingly, bacterial groups related to Fe and Mn (Gallionella and

414

Crenothrix), sulfate (unidentified OP3 GIF10, unidentified Syntrophobacteraceae) and

415

ammonia (Nitrospira and Nitrosomonas) cycles were found at high percentages. Within the

416

two treatment plants using groundwater as source water (TP2 and TP3), a higher percentage

417

of these bacteria was found at TP2 than at TP3. This might be explained by more intensive

418

treatment and the use of subsurface aeration for iron, manganese and ammonia removal at

419

TP3 (de Vet et al., 2009). For TP1, with a higher percentage of genera belonging to

420

unclassified phyla, an even higher diversity than listed in the results can be expected.

SC

M AN U

4. Conclusion

TE D

421

RI PT

409

The results obtained by cultivation-independent methods allow us to evaluate PAB in treated

423

water at water treatment plants and to conclude that:

424

• Regardless of the sources of water and treatment processes applied in the three treatment

425

plants, treated drinking water contains PABs. Levels of A-TCC were 1.0-3.5×103 cells and A-

426

ATP 0.04-0.154 ng l-1 ATP. This represented less than 2% of the TCC and ATP in the treated

427

water.

428

• ATP per cell of PAB is higher than that of PB in bulk water. On average, 25-50 cells were

429

found attached to a single particle. The presence of multiple cells per particle challenges the

430

use of quantitative methods such as HPC and coliforms that are present in the current drinking

431

water quality regulations.

AC C

EP

422

18

ACCEPTED MANUSCRIPT

• Members of the Proteobacteria phylum dominated in all sampled PAB communities,

433

followed by OP3 and Nitrospirae, findings that are similar to previous drinking water and

434

drinking water biofilm studies. This study is the first report of OP3 being a main part of the

435

population in drinking water and PAB in water systems. Nitrospira was the main population

436

of treated groundwater.

437

• Genera of bacteria were found in the PAB communities that appear to be consistent with the

438

particle characteristics: bacterial groups related to Fe and Mn (e.g., Gallionella and

439

Crenothrix), sulfate (e.g., unidentified OP3 GIF10, unidentified Syntrophobacteraceae) and

440

ammonia (e.g., Nitrospira and Nitrosomonas), present in groundwater and water from ARR.

441

• Although this study demonstrated complex microbial communities of PAB in treated

442

drinking water, it has not classified PAB to the species level. It is noted that the genus

443

Legionella, order Thiotrichales, and family Burkholderiaceae, all of which contain pathogenic

444

or opportunistic pathogenic species, were found through this study. To evaluate the presence

445

of (opportunistic) pathogenic bacteria, more specific studies are necessary.

446

• Since the results were observed in three treatment plants with different source water and

447

treatment processes, the study suggests that similar results would be obtained at other

448

treatment plants. However, for the PAB in distribution systems, depending on the particle

449

load, particle characteristics, pipe material and hydraulic conditions, differences can be

450

expected.

451

The data and approaches presented in this study can be useful to elucidate the complexity and

452

dynamics of PAB and PB in drinking water, both for treatment plants and distribution systems.

453

However, the mechanism behind particle-bacteria interacting phenomena is still unclear.

454

Further research is needed to study the behavior and significance of PAB during water

455

distribution and the interaction between particle properties (size, number, elemental

456

composition) and PAB, water quality changes and PAB changes in drinking water distribution

AC C

EP

TE D

M AN U

SC

RI PT

432

19

ACCEPTED MANUSCRIPT

systems. This will, in turn, improve the understanding of the potential risks associated with

458

the PAB groups present in drinking water production and distribution networks.

459

Acknowledgements

460

The authors would like to acknowledge the support from the Chinese Scholarship Council

461

(2008612022). The authors thank Oasen, Dunea and Vitens water companies for their

462

cooperation in this study. Thanks are also due to Maarten Lut, Ed van der Mark, and Geo

463

Bakker for their assistance in the study. The authors thank Gertjan Medema for his critical

464

reading of the manuscript and Yu Tao for his valuable comments.

AC C

EP

TE D

M AN U

SC

RI PT

457

20

ACCEPTED MANUSCRIPT

References

466

Azam, F. and R. E. Hodson (1977). "Size Distribution and Activity of Marine

467

Microheterotrophs." Limnology and Oceanography 22(3): 492-501.

468

Berney, M., Vital, M., Hülshoff, I., Weilenmann, H.U., Egli, T. and Hammes, F. (2008) Rapid,

469

cultivation-independent assessment of microbial viability in drinking water. Water Research

470

42(14), 4010-4018.

471

Brazos, B.J. and O'Connor, J.T. (1996) Seasonal effects on generation of particle-associated

472

bacteria during distribution. Journal of Environmental Engineering 122(12), 1050-1057.

473

Camper, A.K., LeChevallier, M.W., Broadaway, S.C. and McFeters, G.A. (1985) Evaluation

474

of procedures to desorb bacteria from granular activated carbon. Journal of Microbiological

475

Methods 3(3-4), 187-198.

476

Camper, A.K., LeChevallier, M.W., Broadaway, S.C. and McFeters, G.A. (1986) Bacteria

477

associated with granular activated carbon particles in drinking water. Applied and

478

Environmental Microbiology 52(3), 434-438.

479

de Vet, W.W.J.M., Dinkla, I.J.T., Muyzer, G., Rietveld, L.C. and van Loosdrecht, M.C.M.

480

(2009) Molecular characterization of microbial populations in groundwater sources and sand

481

filters for drinking water production. Water Research 43(1), 182-194.

482

DeBruyn, J.M. and Sayler, G.S. (2009) Microbial community structure and biodegradation

483

activity of particle-associated bacteria in a coal tar contaminated creek. Environmental

484

Science and Technology 43(9), 3047-3053.

485

Dietrich, J.P., Darby, J.L. and Loge, F.J. (2009) Potential health risks associated with particles

486

in reclaimed wastewater. Journal of Environmental Engineering 135(5), 285-290.

AC C

EP

TE D

M AN U

SC

RI PT

465

21

ACCEPTED MANUSCRIPT

Dietrich, J.P., Loge, F.J., Ginn, T.R. and Başaǧaoǧlu, H. (2007) Inactivation of particle-

488

associated microorganisms in wastewater disinfection: Modeling of ozone and chlorine

489

reactive diffusive transport in polydispersed suspensions. Water Research 41(10), 2189-2201.

490

Echeverría, F., Castaño, J.G., Arroyave, C., Peñuela, G., Ramírez, A. and Morató, J. (2009)

491

Characterization of deposits formed in a water distribution system. Caracterización de

492

depósitos formados en un sistema de distribución de agua potable 17(2), 275-281.

493

Eichler, S., Christen, R., Höltje, C., Westphal, P., Bötel, J., Brettar, I., Mehling, A. and Höfle,

494

M.G. (2006) Composition and dynamics of bacterial communities of a drinking water supply

495

system as assessed by RNA- and DNA-based 16S rRNA gene fingerprinting. Applied and

496

Environmental Microbiology 72(3), 1858-1872.

497

Eydal, H.S.C. and Pedersen, K. (2007) Use of an ATP assay to determine viable microbial

498

biomass in Fennoscandian Shield groundwater from depths of 3-1000 m. Journal of

499

Microbiological Methods 70(2), 363-373.

500

Gauthier, V., Barbeau, B., Millette, R., Block, J.C. and Prévost, M. (2001) Suspended

501

particles in the drinking water of two distribution systems, pp. 237-245.

502

Gauthier, V., Gérard, B., Portal, J.M., Block, J.C. and Gatel, D. (1999) Organic matter as

503

loose deposits in a drinking water distribution system. Water Research 33(4), 1014-1026.

504

Glöckner, J., Kube, M., Shrestha, P.M., Weber, M., Glöckner, F.O., Reinhardt, R. and Liesack,

505

W. (2010) Phylogenetic diversity and metagenomics of candidate division OP3.

506

Environmental Microbiology 12(5), 1218-1229.

507

Gregory, J. (2005) Particles in water: properties and processes, CRC Press.

508

Hammes, F., Berney, M., Wang, Y., Vital, M., Köster, O. and Egli, T. (2008) Flow-

509

cytometric total bacterial cell counts as a descriptive microbiological parameter for drinking

510

water treatment processes. Water Research 42(1-2), 269-277.

AC C

EP

TE D

M AN U

SC

RI PT

487

22

ACCEPTED MANUSCRIPT

Hammes, F., Goldschmidt, F., Vital, M., Wang, Y. and Egli, T. (2010) Measurement and

512

interpretation of microbial adenosine tri-phosphate (ATP) in aquatic environments. Water

513

Research 44(13), 3915-3923.

514

Herson, D.S., Marshall, D.R., Baker, K.H. and Victoreen, H.T. (1991) Association of

515

microorganisms with surfaces in distribution systems. Journal / American Water Works

516

Association 83(7), 103-106.

517

Herson, D.S., McGonigle, B., Payer, M.A. and Baker, K.H. (1987) Attachment as a factor in

518

the protection of Enterobacter cloacae from chlorination. Applied and Environmental

519

Microbiology 53(5), 1178-1180.

520

Hess-Erga, O.K., Attramadal, K.J.K. and Vadstein, O. (2008) Biotic and abiotic particles

521

protect marine heterotrophic bacteria during UV and ozone disinfection. Aquatic Biology 4(2),

522

147-154.

523

Hoadley, A.W. and Gould, J.P. (1977) Disinfection. Journal of the Water Pollution Control

524

Federation 49(6), 1067-1074.

525

Hong, P.Y., Hwang, C., Ling, F., Andersen, G.L., LeChevallier, M.W. and Liu, W.T. (2010)

526

Pyrosequencing analysis of bacterial biofilm communities in water meters of a drinking water

527

distribution system. Applied and Environmental Microbiology 76(16), 5631-5635.

528

Hugenholtz, P., Pitulle, C., Hershberger, K.L. and Pace, N.R. (1998) Novel division level

529

bacterial diversity in a Yellowstone hot spring. Journal of Bacteriology 180(2), 366-376.

530

Jensen, B.K. (1989) ATP-related specific heterotrophic activity in petroleum contaminated

531

and uncontaminated groundwaters. Canadian Journal of Microbiology 35(8), 814-818.

532

Kalmbach, S., Manz, W. and Szewyk, U. (1997) Dynamics of biofilm formation in drinking

533

water: Phylogenetic affiliation and metabolic potential of single cells assessed by formazan

534

reduction and in situ hybridization. FEMS Microbiology Ecology 22(4), 265-279.

AC C

EP

TE D

M AN U

SC

RI PT

511

23

ACCEPTED MANUSCRIPT

Kolinko, S., Jogler, C., Katzmann, E., Wanner, G., Peplies, J. and Schüler, D. (2011) Single-

536

cell analysis reveals a novel uncultivated magnetotactic bacterium within the candidate

537

division OP3. Environmental Microbiology.

538

Kormas, K.A., Neofitou, C., Pachiadaki, M. and Koufostathi, E. (2010) Changes of the

539

bacterial assemblages throughout an urban drinking water distribution system. Environmental

540

Monitoring and Assessment 165(1-4), 27-38.

541

Kwon, S., Moon, E., Kim, T.S., Hong, S. and Park, H.D. (2011) Pyrosequencing

542

demonstrated complex microbial communities in a membrane filtration system for a drinking

543

water treatment plant. Microbes and Environments 26(2), 149-155.

544

Lehtola, M.J., Juhna, T., Miettinen, I.T., Vartiainen, T. and Martikainen, P.J. (2004)

545

Formation of biofilms in drinking water distribution networks, a case study in two cities in

546

Finland and Latvia. Journal of Industrial Microbiology and Biotechnology 31(11), 489-494.

547

Lehtola, M.J., Laxander, M., Miettinen, I.T., Hirvonen, A., Vartiainen, T. and Martikainen,

548

P.J. (2006) The effects of changing water flow velocity on the formation of biofilms and

549

water quality in pilot distribution system consisting of copper or polyethylene pipes. Water

550

Research 40(11), 2151-2160.

551

Lemarchand, C., Jardillier, L., Carrias, J.F., Richardot, M., Debroas, D., Sime-Ngando, T. and

552

Amblard, C. (2006) Community composition and activity of prokaryotes associated to detrital

553

particles in two contrasting lake ecosystems. FEMS Microbiology Ecology 57(3), 442-451.

554

Lin, T., Chen, W. and Wang, L. (2010) Particle properties in granular activated carbon filter

555

during drinking water treatment. Journal of Environmental Sciences 22(5), 681-688.

556

Liu, Y.Q., Liu, Y. and Tay, J.H. (2004) The effects of extracellular polymeric substances on

557

the formation and stability of biogranules. Applied Microbiology and Biotechnology 65(2),

558

143-148.

AC C

EP

TE D

M AN U

SC

RI PT

535

24

ACCEPTED MANUSCRIPT

Liu, G., Lut, M., Verberk, J.Q.J.C., Van Dijk, J.C. (2013) A comparison of additional

560

treatment processes to limit particle accumulation and microbial growth during drinking water

561

distribution. Water Research, DOI: 10.1016/j.watres.2013.02.035

562

Magic-Knezev, A. and van der Kooij, D. (2004) Optimisation and significance of ATP

563

analysis for measuring active biomass in granular activated carbon filters used in water

564

treatment. Water Research 38(18), 3971-3979.

565

Magic-Knezev, A., Wullings, B. and Van Der Kooij, D. (2009) Polaromonas and

566

Hydrogenophaga species are the predominant bacteria cultured from granular activated

567

carbon filters in water treatment. Journal of Applied Microbiology 107(5), 1457-1467.

568

Mamane, H. and Linden, K.G. (2006) Impact of particle aggregated microbes on UV

569

disinfection. I: Evaluation of spore-clay aggregates and suspended spores. Journal of

570

Environmental Engineering 132(6), 596-606.

571

Mathieu, L., Bouteleux, C., Fass, S., Angel, E. and Block, J.C. (2009) Reversible shift in the

572

α-, β- and γ-proteobacteria populations of drinking water biofilms during discontinuous

573

chlorination. Water Research 43(14), 3375-3386.

574

Matsui, Y., Yamagishi, T., Terada, Y., Matsushita, T. and Inoue, T. (2007) Suspended

575

particles and their characteristics in water mains: Developments of sampling methods. Journal

576

of Water Supply: Research and Technology - AQUA 56(1), 13-24.

577

Paris, T., Skali-Lami, S. and Block, J.C. (2009) Probing young drinking water biofilms with

578

hard and soft particles. Water Research 43(1), 117-126.

579

Parveen, B., J. P. Reveilliez, et al. (2011). "Diversity and dynamics of free-living and particle-

580

associated Betaproteobacteria and Actinobacteria in relation to phytoplankton and

581

zooplankton communities." FEMS Microbiology Ecology 77(3): 461-476.

AC C

EP

TE D

M AN U

SC

RI PT

559

25

ACCEPTED MANUSCRIPT

Power, K. N. and L. A. Nagy (1999). "Relationship between bacterial regrowth and some

583

physical and chemical parameters within Sydney's drinking water distribution system." Water

584

Research 33(3): 741-750.

585

Ridgway, H.F. and Olson, B.H. (1981) Scanning electron microscope evidence for bacterial

586

colonization of a drinking-water distribution system. Applied and Environmental

587

Microbiology 41(1), 274-287.

588

Ridgway, H.F. and Olson, B.H. (1982) Chlorine resistance patterns of bacteria from two

589

drinking water distribution systems. Applied and Environmental Microbiology 44(4), 972-987.

590

Riemann, L. and A. Winding (2001). "Community dynamics of free-living and particle-

591

associated bacterial assemblages during a freshwater phytoplankton bloom." Microbial

592

Ecology 42(3): 274-285.

593

Staley, J.T. and Konopka, A. (1985) Measurement of in situ activities of nonphotosynthetic

594

microorganisms in aquatic and terrestrial habitats. Annual Review of Microbiology 39, 321-

595

346.

596

Tokajian, S.T., Hashwa, F.A., Hancock, I.C. and Zalloua, P.A. (2005) Phylogenetic

597

assessment of heterotrophic bacteria from a water distribution system using 16S rDNA

598

sequencing. Canadian Journal of Microbiology 51(4), 325-335.

599

Van der Kooij, D. (1992) Assimilable organic carbon as an indicator of bacterial regrowth.

600

Journal / American Water Works Association 84(2), 57-65.

601

Van Der Kooij, D. (2000) Biological stability: A multidimensional quality aspect of treated

602

water. Water, Air, and Soil Pollution 123(1-4), 25-34.

603

Van der Wielen, P.W.J.J. and Van der Kooij, D. (2010) Effect of water composition, distance

604

and season on the adenosine triphosphate concentration in unchlorinated drinking water in the

605

Netherlands. Water Research 44(17), 4860-4867.

AC C

EP

TE D

M AN U

SC

RI PT

582

26

ACCEPTED MANUSCRIPT

Velten, S., Hammes, F., Boller, M. and Egli, T. (2007) Rapid and direct estimation of active

607

biomass on granular activated carbon through adenosine tri-phosphate (ATP) determination.

608

Water Research 41(9), 1973-1983.

609

Verberk, J.Q.J.C., Hamilton, L.A., O'Halloran, K.J., Van Der Horst, W. and Vreeburg, J.

610

(2006) Analysis of particle numbers, size and composition in drinking water transportation

611

pipelines: Results of online measurements. Amy, G., Annachhatre, A., Arvin, E., Chen, J. and

612

Cho, J. (eds), pp. 35-43.

613

Vreeburg, J.H.G., Schippers, D., Verberk, J.Q.J.C. and van Dijk, J.C. (2008) Impact of

614

particles on sediment accumulation in a drinking water distribution system. Water Research

615

42(16), 4233-4242.

616

Winkelmann, N. and Harder, J. (2009) An improved isolation method for attached-living

617

Planctomycetes of the genus Rhodopirellula. Journal of Microbiological Methods 77(3), 276-

618

284.

619

Wojcicka, L., Baxter, C. and Hofmann, R. (2008) Impact of particulate matter on distribution

620

system disinfection efficacy. Water Quality Research Journal of Canada 43(1), 55-62.

621

Wu, Y., Clevenger, T. and Deng, B. (2005) Impacts of goethite particles on UV disinfection

622

of drinking water. Applied and Environmental Microbiology 71(7), 4140-4143.

623

Yu, J., Kim, D. and Lee, T. (2010) Microbial diversity in biofilms on water distribution pipes

624

of different materials, pp. 163-171.

625

Zhang, R., B. Liu, et al. (2007). "Particle-attached and free-living bacterial communities in a

626

contrasting marine environment: Victoria Harbor, Hong Kong." FEMS Microbiology Ecology

627

61(3): 496-508.

AC C

EP

TE D

M AN U

SC

RI PT

606

628

27

ACCEPTED MANUSCRIPT

Table 1 Quantifiable characteristics of PAB from three treatment plants

Parameters

TP1

TP2

TP3

Particle counts (# ml-1) (on line)

20(±3)

70(±5)

120(±9)

A-TCC/particle (average cells particle-1)

50 (±14)

49 (±4)

25 (±3)

0.21

0.29

0.33

(±0.05)

(±0.04)

Average ATP/intact cell (10-16 g cell-1) (n=3)

SC

(n=3)

RI PT

629

M AN U

(±0.05) Average A-ATP/intact cell (10-16 g cell-1)

0.53

0.50

0.51

(n=3)

(±0.01)

(±0.01)

(±0.01)

AC C

EP

TE D

630

28

ACCEPTED MANUSCRIPT

Table 2 ATP per cell from reported literature

631

literature

Groundwater

0.2-4.0

(Jensen, 1989)

Groundwater

2.2-5.2

(Eydal and Pedersen, 2007)

Drinking water biofilter

0.21-3.8

(Magic-Knezev and van der Kooij,

RI PT

ATP per cell (10-16 g cell-1)

0.65-2.28

(Velten et al., 2007)

Tap water

0.31-0.55

(Berney et al., 2008)

Average aquatic

0.89

(Hammes et al., 2010)

M AN U

Drinking water biofilter

AC C

EP

TE D

samples

632

SC

2004)

29

ACCEPTED MANUSCRIPT

Table 3 PAB bacterial diversity in the three treatment plants (average ± std., n=3)

Sample

TP1

Number of

Observed

Chao 1

Shannon-Wiener

Evenness

sequences analysed

OTUs (97%)

(97%)

Index (H)

(E)

2988

495 ± 12

1470 ±

8.1 ± 0.10

0.89 ± 0.01

4.9 ± 0.31

0.64 ± 0.02

103 8921

207 ± 24

653 ± 148

2818

239 ± 7

448 ± 83

6.3 ± 0.19

M AN U

TP3

SC

TP2

RI PT

633

AC C

EP

TE D

634

0.80 ± 0.02

30

ACCEPTED MANUSCRIPT

Table 4 The genera identified in PAB of different treatment plants

Percentage of sequences (%) TP2

TP3

Legionella

0.34

3.03

1.0

Nitrospira

0.11

3.10

0.42

Gallionella

0.34

Planctomyces

1.00

SC

RI PT

TP1

0.29

2.10

0.48

M AN U

4.47

CandidatusOdyssella

0.04

1.06

0.46

Caulobacter

0.17

0.22

0.11

0.16

0.20

0.31

Aquabacterium

0.27

0.71

0.07

Comamonadeceae (unclassified)

3.70

3.02

10.85

12.09

10.28

3.38

Caldilinea

0

0.02

0.08

Nitrosomonas

0.19

5.17

0.22

Crenothrix

0.13

4.49

0.30

Thermodesulfovibrionaceae LCP-6

1.90

3.50

3.71

EP

TE D

Hyphomicrobium

Unclassified OP3 GIF10

AC C

635

31

ACCEPTED MANUSCRIPT

0.31

0

13.79

Rhodopirellula

0

0.07

0

Leptolyngbya

0

0

3.39

Unclassified Alphaproteobacteria

1.77

6.58

6.08

Unclassified phylum bacteria

22.25

12.79

6.84

Other

55.23

SC 39.19

48.22

M AN U

636

RI PT

Rhodospirillaceae (un-identified)

AC C

EP

TE D

637

32

ACCEPTED MANUSCRIPT

638

Figure 1 Schematic drawing of multiple particle filtration system (MuPFiS) and parallel

639

running particle counter used for PAB sampling

640

Figure 2 SEM pictures (a: TP1, b: TP2 and c: TP3) and elemental composition of PAB (d)

RI PT

641 642

Figure 3 Average TCC and A-TCC results at each treatment plant (n=3)

SC

643 644

Figure 4 Average ATP and A-ATP results of at each treatment plant (n=3)

M AN U

645 646

Figure 5 Taxonomic assignment of 16s rRNA gene sequences retrieved from PAB samples

648

classified by phylum. (a: TP1; b: TP2; c: TP3)

AC C

EP

TE D

647

33

ACCEPTED MANUSCRIPT Highlights

• The number and activity of PAB ranged from 1.0-3.5×103 cells/ml and 0.04-0.154ng/l ATP. • There were 25-50 cells found to be attached on a single particle. • Anoxic and anaerobic bacteria were detected; OP3 was a main part of the PAB population.

RI PT

• Bacterial genera associated with particles are consistent with the particle characteristics. • PAB are important to bacteria enumeration and ecology of water distribution systems.

AC C

EP

TE D

M AN U

SC

 

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

Supplementary data Table S1 water quality information of selected treatment plants (N.A.: not available)

TP2

Surface water

Groundwater

Extracted ARR

Treated

Raw

N.A.

12.9

N.A.

Temperature

pH

N.A.

Turbidity (NTU)

N.A.

AOC

N.A.

l-1) ATP (ng l-1)

Groundwater

Raw

12

Treated

N.A.

12

8.5

N.A.

8.3

N.A.

8.2

<0.03

N.A.

0.22

N.A.

<0.1

5.8

N.A.

12.

N.A.

5.9

TE D

(µg c

Treated

M AN U

(°C)

TP3

RI PT

TP1

SC

Parameters

N.A.

1

N.A.

4.5

N.A.

4.5

N.A.

0.45

N.A.

1.65

N.A.

1.62

0.60

0.0052

0.55

0.0052

0.67

<0.03

Fe (mg l-1)

0.33

<0.01

0.95

0.012

0.92

<0.01

Mn (mg l-1)

0.09

<0.01

0.79

0.023

0.085

<0.01

SO4 (mg l-1)

N.A.

50.8

N.A.

34

N.A.

31

EP

TCC (cells ml-1)

AC C

NH4 (mg l-1)

ACCEPTED MANUSCRIPT

AOC of TP1 was averaged of measurements in the year 2010 (n=4). AOC of TP2 and TP3 was measured in 2010. The result of parameters were averaged results of samples taken

TE D

M AN U

SC

RI PT

together with PAB sampling (n=3).

AC C

EP

Figure S1. The release of ATP during LES treatment, each LES treatment last for 2 minutes and triplicate measurements were conducted.

ACCEPTED MANUSCRIPT

Quantification and identification of particle-associated bacteria in unchlorinated drinking water from three treatment plants by cultivation-independent methods

RI PT

G. Liu1 *, F. Q. Ling2, A. Magic-Knezev3, W. T. Liu2, J.Q.J.C.Verberk1, J.C. Van Dijk1 1. Section Sanitary Engineering, Department of Water Management, Faculty of Civil Engineering and Geosciences, Delft University of Technology, PO BOX 5048, 2600 GA Delft, the Netherlands E-mail: [email protected]

SC

2. Department of Civil and Environmental Engineering, University of Illinois Urbana-

M AN U

Champaign, 205 N. Mathews Ave., Urbana, Illinois 61801, U.S.A.

AC C

EP

TE D

3. Het Water Laboratorium, PO BOX 734, 2003 RS Haarlem, the Netherlands

1

ACCEPTED MANUSCRIPT

References Azam, F. and R. E. Hodson (1977). "Size Distribution and Activity of Marine Microheterotrophs." Limnology and Oceanography 22(3): 492-501.

RI PT

Berney, M., Vital, M., Hülshoff, I., Weilenmann, H.U., Egli, T. and Hammes, F. (2008) Rapid, cultivation-independent assessment of microbial viability in drinking water. Water Research 42(14), 4010-4018.

SC

Boe-Hansen, R., Albrechtsen, H.J., Arvin, E. and Jørgensen, C. (2002) Bulk water phase and biofilm growth in drinking water at low nutrient conditions. Water Research 36(18), 4477-

M AN U

4486.

Brazos, B.J. and O'Connor, J.T. (1996) Seasonal effects on generation of particle-associated bacteria during distribution. Journal of Environmental Engineering 122(12), 1050-1057. Camper, A.K., LeChevallier, M.W., Broadaway, S.C. and McFeters, G.A. (1985) Evaluation

TE D

of procedures to desorb bacteria from granular activated carbon. Journal of Microbiological Methods 3(3-4), 187-198.

Camper, A.K., LeChevallier, M.W., Broadaway, S.C. and McFeters, G.A. (1986) Bacteria

EP

associated with granular activated carbon particles in drinking water. Applied and Environmental Microbiology 52(3), 434-438.

AC C

de Vet, W.W.J.M., Dinkla, I.J.T., Muyzer, G., Rietveld, L.C. and van Loosdrecht, M.C.M. (2009) Molecular characterization of microbial populations in groundwater sources and sand filters for drinking water production. Water Research 43(1), 182-194. DeBruyn, J.M. and Sayler, G.S. (2009) Microbial community structure and biodegradation activity of particle-associated bacteria in a coal tar contaminated creek. Environmental Science and Technology 43(9), 3047-3053.

2

ACCEPTED MANUSCRIPT

Dietrich, J.P., Darby, J.L. and Loge, F.J. (2009) Potential health risks associated with particles in reclaimed wastewater. Journal of Environmental Engineering 135(5), 285-290. Dietrich, J.P., Loge, F.J., Ginn, T.R. and Başaǧaoǧlu, H. (2007) Inactivation of particle-

RI PT

associated microorganisms in wastewater disinfection: Modeling of ozone and chlorine reactive diffusive transport in polydispersed suspensions. Water Research 41(10), 2189-2201. Echeverría, F., Castaño, J.G., Arroyave, C., Peñuela, G., Ramírez, A. and Morató, J. (2009)

SC

Characterization of deposits formed in a water distribution system. Caracterización de depósitos formados en un sistema de distribución de agua potable 17(2), 275-281.

M AN U

Eichler, S., Christen, R., Höltje, C., Westphal, P., Bötel, J., Brettar, I., Mehling, A. and Höfle, M.G. (2006) Composition and dynamics of bacterial communities of a drinking water supply system as assessed by RNA- and DNA-based 16S rRNA gene fingerprinting. Applied and Environmental Microbiology 72(3), 1858-1872.

TE D

Eydal, H.S.C. and Pedersen, K. (2007) Use of an ATP assay to determine viable microbial biomass in Fennoscandian Shield groundwater from depths of 3-1000 m. Journal of Microbiological Methods 70(2), 363-373.

EP

Gauthier, V., Barbeau, B., Millette, R., Block, J.C. and Prévost, M. (2001) Suspended particles in the drinking water of two distribution systems, pp. 237-245.

AC C

Gauthier, V., Gérard, B., Portal, J.M., Block, J.C. and Gatel, D. (1999) Organic matter as loose deposits in a drinking water distribution system. Water Research 33(4), 1014-1026. Glöckner, J., Kube, M., Shrestha, P.M., Weber, M., Glöckner, F.O., Reinhardt, R. and Liesack, W. (2010) Phylogenetic diversity and metagenomics of candidate division OP3. Environmental Microbiology 12(5), 1218-1229. Gregory, J. (2005) Particles in water: properties and processes, CRC Press.

3

ACCEPTED MANUSCRIPT

Hammes, F., Berney, M., Wang, Y., Vital, M., Köster, O. and Egli, T. (2008) Flowcytometric total bacterial cell counts as a descriptive microbiological parameter for drinking water treatment processes. Water Research 42(1-2), 269-277.

RI PT

Hammes, F., Goldschmidt, F., Vital, M., Wang, Y. and Egli, T. (2010) Measurement and interpretation of microbial adenosine tri-phosphate (ATP) in aquatic environments. Water Research 44(13), 3915-3923.

SC

Herson, D.S., Marshall, D.R., Baker, K.H. and Victoreen, H.T. (1991) Association of microorganisms with surfaces in distribution systems. Journal / American Water Works

M AN U

Association 83(7), 103-106.

Herson, D.S., McGonigle, B., Payer, M.A. and Baker, K.H. (1987) Attachment as a factor in the protection of Enterobacter cloacae from chlorination. Applied and Environmental Microbiology 53(5), 1178-1180.

TE D

Hess-Erga, O.K., Attramadal, K.J.K. and Vadstein, O. (2008) Biotic and abiotic particles protect marine heterotrophic bacteria during UV and ozone disinfection. Aquatic Biology 4(2), 147-154.

EP

Hoadley, A.W. and Gould, J.P. (1977) Disinfection. Journal of the Water Pollution Control Federation 49(6), 1067-1074.

AC C

Hong, P.Y., Hwang, C., Ling, F., Andersen, G.L., LeChevallier, M.W. and Liu, W.T. (2010) Pyrosequencing analysis of bacterial biofilm communities in water meters of a drinking water distribution system. Applied and Environmental Microbiology 76(16), 5631-5635. Hugenholtz, P., Pitulle, C., Hershberger, K.L. and Pace, N.R. (1998) Novel division level bacterial diversity in a Yellowstone hot spring. Journal of Bacteriology 180(2), 366-376. Jensen, B.K. (1989) ATP-related specific heterotrophic activity in petroleum contaminated and uncontaminated groundwaters. Canadian Journal of Microbiology 35(8), 814-818. 4

ACCEPTED MANUSCRIPT

Kalmbach, S., Manz, W. and Szewyk, U. (1997) Dynamics of biofilm formation in drinking water: Phylogenetic affiliation and metabolic potential of single cells assessed by formazan reduction and in situ hybridization. FEMS Microbiology Ecology 22(4), 265-279.

RI PT

Kolinko, S., Jogler, C., Katzmann, E., Wanner, G., Peplies, J. and Schüler, D. (2011) Singlecell analysis reveals a novel uncultivated magnetotactic bacterium within the candidate division OP3. Environmental Microbiology.

SC

Kormas, K.A., Neofitou, C., Pachiadaki, M. and Koufostathi, E. (2010) Changes of the bacterial assemblages throughout an urban drinking water distribution system. Environmental

M AN U

Monitoring and Assessment 165(1-4), 27-38.

Kwon, S., Moon, E., Kim, T.S., Hong, S. and Park, H.D. (2011) Pyrosequencing demonstrated complex microbial communities in a membrane filtration system for a drinking water treatment plant. Microbes and Environments 26(2), 149-155.

TE D

Lehtola, M.J., Juhna, T., Miettinen, I.T., Vartiainen, T. and Martikainen, P.J. (2004) Formation of biofilms in drinking water distribution networks, a case study in two cities in Finland and Latvia. Journal of Industrial Microbiology and Biotechnology 31(11), 489-494.

EP

Lehtola, M.J., Laxander, M., Miettinen, I.T., Hirvonen, A., Vartiainen, T. and Martikainen, P.J. (2006) The effects of changing water flow velocity on the formation of biofilms and

AC C

water quality in pilot distribution system consisting of copper or polyethylene pipes. Water Research 40(11), 2151-2160. Lemarchand, C., Jardillier, L., Carrias, J.F., Richardot, M., Debroas, D., Sime-Ngando, T. and Amblard, C. (2006) Community composition and activity of prokaryotes associated to detrital particles in two contrasting lake ecosystems. FEMS Microbiology Ecology 57(3), 442-451. Lin, T., Chen, W. and Wang, L. (2010) Particle properties in granular activated carbon filter during drinking water treatment. Journal of Environmental Sciences 22(5), 681-688. 5

ACCEPTED MANUSCRIPT

Liu, Y.Q., Liu, Y. and Tay, J.H. (2004) The effects of extracellular polymeric substances on the formation and stability of biogranules. Applied Microbiology and Biotechnology 65(2), 143-148.

RI PT

Liu, G., Lut, M., Verberk, J.Q.J.C., Van Dijk, J.C. (2013) A comparison of additional treatment processes to limit particle accumulation and microbial growth during drinking water distribution. Water Research, DOI: 10.1016/j.watres.2013.02.035

SC

Magic-Knezev, A. and van der Kooij, D. (2004) Optimisation and significance of ATP analysis for measuring active biomass in granular activated carbon filters used in water

M AN U

treatment. Water Research 38(18), 3971-3979.

Magic-Knezev, A., Wullings, B. and Van Der Kooij, D. (2009) Polaromonas and Hydrogenophaga species are the predominant bacteria cultured from granular activated carbon filters in water treatment. Journal of Applied Microbiology 107(5), 1457-1467.

TE D

Mamane, H. and Linden, K.G. (2006) Impact of particle aggregated microbes on UV disinfection. I: Evaluation of spore-clay aggregates and suspended spores. Journal of Environmental Engineering 132(6), 596-606.

EP

Mathieu, L., Bouteleux, C., Fass, S., Angel, E. and Block, J.C. (2009) Reversible shift in the α-, β- and γ-proteobacteria populations of drinking water biofilms during discontinuous

AC C

chlorination. Water Research 43(14), 3375-3386. Matsui, Y., Yamagishi, T., Terada, Y., Matsushita, T. and Inoue, T. (2007) Suspended particles and their characteristics in water mains: Developments of sampling methods. Journal of Water Supply: Research and Technology - AQUA 56(1), 13-24. Paris, T., Skali-Lami, S. and Block, J.C. (2009) Probing young drinking water biofilms with hard and soft particles. Water Research 43(1), 117-126.

6

ACCEPTED MANUSCRIPT

Parveen, B., J. P. Reveilliez, et al. (2011). "Diversity and dynamics of free-living and particleassociated Betaproteobacteria and Actinobacteria in relation to phytoplankton and zooplankton communities." FEMS Microbiology Ecology 77(3): 461-476.

RI PT

Power, K. N. and L. A. Nagy (1999). "Relationship between bacterial regrowth and some physical and chemical parameters within Sydney's drinking water distribution system." Water Research 33(3): 741-750.

SC

Ridgway, H.F. and Olson, B.H. (1981) Scanning electron microscope evidence for bacterial colonization of a drinking-water distribution system. Applied and Environmental

M AN U

Microbiology 41(1), 274-287.

Ridgway, H.F. and Olson, B.H. (1982) Chlorine resistance patterns of bacteria from two drinking water distribution systems. Applied and Environmental Microbiology 44(4), 972-987. Riemann, L. and A. Winding (2001). "Community dynamics of free-living and particle-

Ecology 42(3): 274-285.

TE D

associated bacterial assemblages during a freshwater phytoplankton bloom." Microbial

Staley, J.T. and Konopka, A. (1985) Measurement of in situ activities of nonphotosynthetic

346.

EP

microorganisms in aquatic and terrestrial habitats. Annual Review of Microbiology 39, 321-

AC C

Tokajian, S.T., Hashwa, F.A., Hancock, I.C. and Zalloua, P.A. (2005) Phylogenetic assessment of heterotrophic bacteria from a water distribution system using 16S rDNA sequencing. Canadian Journal of Microbiology 51(4), 325-335. Van der Kooij, D. (1992) Assimilable organic carbon as an indicator of bacterial regrowth. Journal / American Water Works Association 84(2), 57-65. Van Der Kooij, D. (2000) Biological stability: A multidimensional quality aspect of treated water. Water, Air, and Soil Pollution 123(1-4), 25-34. 7

ACCEPTED MANUSCRIPT

Van der Wielen, P.W.J.J. and Van der Kooij, D. (2010) Effect of water composition, distance and season on the adenosine triphosphate concentration in unchlorinated drinking water in the Netherlands. Water Research 44(17), 4860-4867.

RI PT

Velten, S., Hammes, F., Boller, M. and Egli, T. (2007) Rapid and direct estimation of active biomass on granular activated carbon through adenosine tri-phosphate (ATP) determination. Water Research 41(9), 1973-1983.

SC

Verberk, J.Q.J.C., Hamilton, L.A., O'Halloran, K.J., Van Der Horst, W. and Vreeburg, J. (2006) Analysis of particle numbers, size and composition in drinking water transportation

M AN U

pipelines: Results of online measurements. Amy, G., Annachhatre, A., Arvin, E., Chen, J. and Cho, J. (eds), pp. 35-43.

Vreeburg, J.H.G., Schippers, D., Verberk, J.Q.J.C. and van Dijk, J.C. (2008) Impact of particles on sediment accumulation in a drinking water distribution system. Water Research

TE D

42(16), 4233-4242.

Winkelmann, N. and Harder, J. (2009) An improved isolation method for attached-living Planctomycetes of the genus Rhodopirellula. Journal of Microbiological Methods 77(3), 276-

EP

284.

Wojcicka, L., Baxter, C. and Hofmann, R. (2008) Impact of particulate matter on distribution

AC C

system disinfection efficacy. Water Quality Research Journal of Canada 43(1), 55-62. Wu, Y., Clevenger, T. and Deng, B. (2005) Impacts of goethite particles on UV disinfection of drinking water. Applied and Environmental Microbiology 71(7), 4140-4143. Yu, J., Kim, D. and Lee, T. (2010) Microbial diversity in biofilms on water distribution pipes of different materials, pp. 163-171.

8

ACCEPTED MANUSCRIPT

Zhang, R., B. Liu, et al. (2007). "Particle-attached and free-living bacterial communities in a contrasting marine environment: Victoria Harbor, Hong Kong." FEMS Microbiology Ecology

AC C

EP

TE D

M AN U

SC

RI PT

61(3): 496-508.

9