Dr-HYDER 3

See discussions, stats, and author profiles for this publication at: https://www.researchgate.
net/publication/332987859
Continuous ozonation of urban wastewater: Removal of antibiotics,

antibiotic-resistant Escherichia coli and antibiotic resistance genes and
phytotoxicity
Article in Water Research · May 2019

DOI: 10.1016/j.watres.2019.05.025
CITATIONS READS
200 1,254
10 authors, including:
Iakovos C.Iakovides Ana Rita Lado Ribeiro

University of Cyprus University of Porto
9 PUBLICATIONS 626 CITATIONS 101 PUBLICATIONS 6,578 CITATIONS
SEE PROFILE SEE PROFILE
Telma Fernandes
Universidade NOVA de Lisboa
15 PUBLICATIONS 806 CITATIONS
SEE PROFILE
Some of the authors of this publication are also working on these related projects:
COST Action CA18225 - WaterTOP - Taste and Odor in early diagnosis of source and drinking Water Problems View project
NanoBiofilms View project
All content following this page was uploaded by Iakovos C.Iakovides on 31 May 2019.
The user has requested enhancement of the downloaded file.

Water Research 159 (2019) 333e347
Contents lists available at ScienceDirect
Water Research
journal homepage: www.elsevier.com/locate/watres
Continuous ozonation of urban wastewater: Removal of antibiotics,

antibiotic-resistant Escherichia coli and antibiotic resistance genes and
phytotoxicity
I.C. Iakovides a, b, I. Michael-Kordatou b, N.F.F. Moreira c, d, A.R. Ribeiro d, T. Fernandes e,
M.F.R. Pereira d, O.C. Nunes c, C.M. Manaia e, A.M.T. Silva d, **, D. Fatta-Kassinos a, b, *
a
Department of Civil and Environmental Engineering, School of Engineering, University of Cyprus, P.O. Box 20537, 1678, Nicosia, Cyprus
b
Nireas-International Water Research Centre, University of Cyprus, P.O. Box 20537, 1678, Nicosia, Cyprus
c
LEPABE-Laboratory for Process Engineering Environment, Biotechnology and Energy, Faculdade de Engenharia, Universidade do Porto, Rua Dr. Roberto
Frias, 4200-465, Porto, Portugal
d
Laboratory of Separation and Reaction Engineering-Laboratory of Catalysis and Materials (LSRE-LCM), Faculdade de Engenharia, Universidade do Porto,
Rua Dr. Roberto Frias, 4200-465, Porto, Portugal
e
Universidade Catolica Portuguesa, CBQF - Centro de Biotecnologia e Química Fina e Laborato rio Associado, Escola Superior de Biotecnologia, Rua
Arquiteto Loba~o Vital, 172, 4200-374, Porto, Portugal
a r t i c l e i n f o a b s t r a c t
Article history: This work evaluated the removal of a mixture of eight antibiotics (i.e. ampicillin (AMP), azithromycin
Received 1 February 2019 (AZM), erythromycin (ERY), clarithromycin (CLA), ofloxacin (OFL), sulfamethoxazole (SMX), trimetho-
Received in revised form prim (TMP) and tetracycline (TC)) from urban wastewater, by ozonation operated in continuous mode at
6 May 2019
different hydraulic retention times (HRTs) (i.e. 10, 20, 40 and 60 min) and specific ozone doses (i.e. 0.125,
Accepted 7 May 2019
Available online 10 May 2019
0.25, 0.50 and 0.75 gO3 gDOC 1). As expected, the efficiency of ozonation was highly ozone dose- and
contact time-dependent. The removal of the parent compounds of the selected antibiotics to levels below
their detection limits was achieved with HRT of 40 min and specific ozone dose of 0.125 gO3 gDOC 1. The
Keywords:
Antibiotics
effect of ozonation was also investigated at a microbiological and genomic level, by studying the effi-
Antibiotic resistance ciency of the process with respect to the inactivation of Escherichia coli and antibiotic-resistant E. coli, as
Phytotoxicity well as to the reduction of the abundance of selected antibiotic resistance genes (ARGs). The inactivation
Ozonation of total cultivable E. coli was achieved under the experimental conditions of HRT 40 min and 0.25 gO3
Continuous mode gDOC1, at which all antibiotic compounds were already degraded. The regrowth examinations revealed
that higher ozone concentrations were required for the permanent inactivation of E. coli below the Limit
of Quantification (<LOQ ¼ 0.01 CFU mL 1). Also, the abundance of the examined ARGs (intl1, aadA1,
dfrA1, qacED1 and sul1) was found to decrease with increasing HRT and ozone dose. Despite the fact that
the mildest operating parameters were able to eliminate the parent compounds of the tested antibiotics
in wastewater effluents, it was clearly demonstrated in this study that higher ozone doses were required
in order to confer permanent damage and/or death and prevent potential post-treatment re-growth of
both total bacteria and ARB, and to reduce the abundance of ARGs below the LOQ. Interestingly, the
mineralization of wastewater, in terms of Dissolved Organic Carbon (DOC) removal, was found to be
significantly low even when the higher ozone doses were applied, leading to an increased phytotoxicity
towards various plant species. The findings of this study clearly underline the importance of properly
optimising the ozonation process (e.g. specific ozone dose and contact time) taking into consideration
both the bacterial species and associated ARGs, as well as the wastewater physicochemical properties
(e.g. DOC), in order to mitigate the spread of ARB&ARGs, as well as to reduce the potential phytotoxicity.
© 2019 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND
license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
* Corresponding author. Department of Civil and Environmental Engineering, School of Engineering, University of Cyprus, P.O. Box 20537, 1678, Nicosia, Cyprus.
** Corresponding author.
E-mail addresses: [email protected] (A.M.T. Silva), [email protected] (D. Fatta-Kassinos).
https://doi.org/10.1016/j.watres.2019.05.025
0043-1354/© 2019 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
334 I.C. Iakovides et al. / Water Research 159 (2019) 333e347
1. Introduction radicals (HO) under alkaline pH conditions, which result from

ozone decay in water and wastewater. Ozonation has been shown
Nowadays, the ecosystems and the human health are threat- to have a high potential for the oxidation of a variety of pharma-
ened by the environmental occurrence of both chemical and bio- ceutical compounds, including antibiotics, and the inactivation of
logical contaminants of emerging concern (CECs), such as total pathogenic bacteria in drinking water and wastewater
antibiotics and acquired antibiotic resistance genes (ARGs), mobi- (Antoniou et al., 2013; Von Gunten, 2018). The few studies pub-
lized by antibiotic-resistant bacteria (ARB). ARB&ARGs are consid- lished so far suggest that ozonation is effective to inactivate ARB
ered as a serious public health problem by various international and remove ARGs and also to eliminate the potential bacterial
organizations and the European Union (EU), because of their regrowth, under proper optimized conditions, specific ozone dose
widespread in the environment, food chain, and even drinking and exposure time (Michael-Kordatou et al., 2018). However,
water, among others (Rizzo et al., 2013; Berendonk et al., 2015). further studies are necessary to fill in the gaps of knowledge in
Recently, the environmental dimension of antimicrobial resistance relation to the effectiveness of ozonation to the inactivation of ARB
has been identified as one of the six emerging issues of environ- and the reduction of the abundance of ARGs.
mental concern according to the United Nations Environment Within this context, the aim of the present work was to inves-
(UNEP, 2017; APHA, 2005). In fact, ARB can often do work syner- tigate and optimize the efficiency of ozonation operated in
gistically by interacting with each other in a variety of mechanisms continuous mode, for the abatement of a diverse array of antibiotic-
that enhance remarkably their collective capability to transfer ARGs related chemical and microbiological microcontaminants from
leading to antibiotic resistance, described two decades ago by secondary treated wastewater effluents. The objectives of this
Salyers and Amabile-Cuevas (1997) as an ‘easy-to-get, hard-to-lose’ study were to assess: (i) the removal of a mixture of eight antibi-
phenomenon. In the face of regional and global antibiotic resistance otics belonging to different classes and exhibiting different physi-
challenges, the European Commission stands at the forefront for cochemical properties, being the macrolides included in the EU
intensifying its efforts to shape a relevant legislative instrument watch list of substances for Union-wide monitoring (EU, 2017b, EU,
with regard to the minimum quality criteria for the assessment of 2018), namely azithromycin (AZM), erythromycin (ERY), clari-
reclaimed wastewater intended for agricultural irrigation and thromycin (CLA), as well as ampicillin (AMP), ofloxacin (OFL), sul-
aquifer recharge, including antibiotic resistance and related risks famethoxazole (SMX), trimethoprim (TMP), and tetracycline (TC);
(EU, 2017a). (ii) the inactivation of wastewater autochthonous total and
In the last decade, urban wastewater treatment plants (UWTPs) antibiotic-resistant Escherichia coli and to further evaluate their
have been considered as critical hotspots for the propagation of regrowth potential after treatment; (iii) the reduction of the
antibiotic resistance in the environment, with the consequences abundance of selected genes originally present in wastewater,
being potentially compounded by the reuse of treated wastewater. specifically the bacterial biomarker 16S rRNA, the class 1 integron
The fact that UWTPs are not specifically designed for the elimina- integrase intI1 and ARGs (i.e., qacDE1, sul1, aadA1 and dfrA1); (iv)
tion of antibiotic compounds (Michael et al., 2013), accompanied by the extent of mineralization of the treated wastewater; and (v) the
an increase in the prevalence of ARB and their associated ARGs, has evolution of phytotoxicity against selected plant species. According
triggered an imperative need to improve the efficiency of the to the authors’ knowledge this work is among the first studies
existing treatment processes (Pruden, 2013; Karkman et al., 2017; revealing comprehensive data regarding the oxidation of a mixture
Michael-Kordatou et al., 2018; Manaia et al., 2018). In this sense, of antibiotics (rather than one compound) at environmentally low
various advanced chemical oxidation processes (AOPs) have been concentrations (rather than high concentrations) during the
extensively investigated to efficiently remove antibiotic residues application of ozonation in continuous mode (rather in batch
(Michael et al., 2013), inactivate ARB and remove ARGs from mode) simulating thus a more realistic scenario from the UWTP
wastewater effluents at bench-, pilot-, and full-scale applications full-scale perspective, combined with the assessment of the process
(Zhuang et al., 2015; Alexander et al., 2016; Czekalski et al., 2016; in removing ARB&ARGs (rather than evaluating only the disinfec-
Ferro et al., 2016; Lee and Von Gunten, 2016; Sousa et al., 2017; tion efficiency in relation to the inactivation of various bacterial
Karaolia et al., 2018). In a recent review, Michael-Kordatou et al. groups) and phytotoxicity. The results obtained by existing studies
(2018) investigated the fate of ARB&ARGs during AOPs, the key performed on the ozonation of wastewater (Table S1) may be
operating conditions affecting their efficiency to inactivate ARB and sometimes contradictory, while in many cases the lack and/or
remove ARGs, and the main oxidative damage pathways involved in heterogeneity of the scientific data (e.g. different target micro-
these processes, indicating that besides the operating conditions, contaminants, different experimental configurations, etc), as well
the variable behavior observed by the various examined genetic as the different methodological approaches applied in the various
constituents of the microbial community, may be directed by the studies, make difficult the accurate evaluation of the efficiency of
process distinct oxidative damage mechanisms in place during the the ozonation. The innovation of this study lies in the fact that it
application of each treatment technology. evaluates the efficiency of ozonation operated in continuous mode in
A great deal of interest has been focused on the application of relation to the simultaneous removal of both chemical (i.e. antibiotics)
ozone-based systems, which have been demonstrated to be effec- and biological (i.e. ARB&ARGs) microcontaminants, as well as phyto-
tive in achieving significant abatement of various micro- toxicity, which is another important parameter when wastewater
contaminants, while also providing sufficient disinfection of reuse in crop irrigation is considered. Also, this is one of the very first
wastewater at different ozone doses and contact times (Von attempts made to investigate if the variable region of class1 integrons
Gunten, 2003, 2018; Ikehata et al., 2006). In fact, the knowledge might undergo excision after exposure to ozone (herein indicated by a
on the efficiency of ozonation to eliminate ARB&ARGs is limited reduction in aadA1 and dfrA1).
since most studies performed so far only focused on the effect of
ozone on cultivable bacterial populations, with the effect of the 2. Material and methods
process on ARB or ARGs being still unclear (Michael-Kordatou et al.,
2018). Ozone is a powerful oxidant and has been widely used 2.1. Chemicals and reagents
during the last years for the treatment of wastewater. It reacts
selectively with organic compounds via a direct reaction with The antibiotic reference standards for liquid chromatography
ozone molecules, or through an indirect pathway with hydroxyl (>98% purity) for AMP (CAS number 69-53-4), AZM (CAS number
I.C. Iakovides et al. / Water Research 159 (2019) 333e347 335
117772-70-0), ERY (CAS number 114-07-8), CLA (CAS number below the method's detection limit (MDL): AMP MDL,
81103-11-9), OFL (CAS number 82419-36-1), SMX (CAS number AZM ¼ 184.5e358.8 ng L1, CLA ¼ 433.2e474.8 ng L1, ERY MDL,
723-46-6), TMP (CAS number 738-70-5), and TC (CAS number 60- OFL MDL-6.12 ng L1, SMX MDL, TC MDL, and
54-8), as well as the surrogate standards azithromycin-d3 and TMP ¼ 129.5e334.1 ng L1. The MDL values for each compound are
ofloxacin-d3 were purchased from Sigma Aldrich (Steinhein, Ger- provided in Table 1.
many). Methanol (MS grade) and hydrogen peroxide (H2O2, 30% w/
w) were acquired from VWR International (Fontenay-sous-Bois, 2.3. Ozonation set-up and procedure
France) and Merck (Darmstadt, Germany), respectively. Ethanol
(HPLC grade) was purchased from Fisher Scientific (Leicestershire, Ozonation experiments were performed in a continuous mode,
UK). Formic acid (99%), potassium iodide, sodium dihydrogen through a flow-through reactor where untreated wastewater ef-
phosphate and phosphoric acid (85%) were provided by Fluka fluents were pumped on a continuous basis, and treated samples
(Steinhein, Germany). Catalase from bovine liver powder were exiting continuously. A cylindrical borosilicate bubble column
(2,000e5,000 units/mg protein), supplied by Fluka, was used to reactor (Fig. 1a; 3.0 cm internal diameter (I.D). 70 cm height,
quench the reactions in the treated samples prior to chromato- useful volume of approximately 310 mL) packed with ca. 400 glass
graphic and toxicity/microbiological analysis. Also, sodium sulphite rings, was used in order to promote the contact between the gas
(Sigma Aldrich) solution was used in the treated samples to remove phase and wastewater, increasing the mass transfer of ozone. The
excess oxidants for the DOC determination. Potassium indigo tri- use of borosilicate in the reactor column ensured that no adsorp-
sulfonate (Sigma Aldrich), sodium dihydrogen phosphate (Fluka) tion/desorption of the examined compounds on the reactor walls
and phosphoric acid (85%, Fluka) were used for the preparation of occurred. This material has been widely used in ozonation reactors
the indigo solution. Ultrapure water (resistivity > 18.2 MU cm at (Mecha et al., 2017; Orhon et al., 2017; Rozas et al., 2017). Also, the

25 C) was supplied by a Milli-Q water system (Millipore). Oasis® borosilicate bubble column provides the possibility of visual con-
HLB (Hydrophilic-Lipophilic Balanced) cartridges (150 mg, 6 mL) tact in the reactor to avoid any malfunctions during the perfor-
used for sample pre-concentration, were supplied by Waters mance of the experiments. Regardless of the type of experiment
(Milford, MA, USA). Stock solutions of each individual compound performed, the reactor was always filled with ultrapure water and
(approximately 5,000 mg L1) were prepared in methanol and the experiments commenced (t0 ¼ 0 min) when the wastewater to
working solutions were prepared by diluting these solutions in be treated was pumped into the reactor through the peristaltic
ultrapure water. The hydrolytic stability of each stock solution was pump (Fig. 1b), being diluted until achieving steady-state condi-
routinely evaluated through chromatographic analysis after storage tions. Preliminary experiments, through conductivity measure-
in a refrigerator for 2 months and it was found to be stable. ments with a tracer solution of NaCl (concentration of
2,000 mg L1), were performed to determine the time required to
2.2. UWTP samples achieve the steady state for each HRT (Fig. S1). For the examined
HRTs (i.e. 10, 20, 40 and 60 min), the inlet liquid flow rate varied
All experiments were performed using wastewater effluent between 40 and 6 mL min1. According to Fig. 1, the wastewater
samples collected between October 2017 and January 2018, from effluents were continuously entering the reactor by the bottom of
the secondary clarifier of an UWTP located in Northern Portugal. the column (Fig. 1c) and leaving it from its top (Fig. 1d). The effi-
This UWTP, serving a population equivalent of 80,000, has an ciency of the process was estimated after achieving the steady state
average flow of 18,000 m3 day1 and employs conventional acti- for each selected HRT. A BMT 802X ozone generator (Fig. 1e) was
vated sludge (CAS) as secondary treatment. Grab samples were used to produce ozone from pure oxygen. The desired ozone con-
collected in sterile glass bottles and processed immediately after centration was produced through the adjustment of the O2 gas flow
being transferred to the laboratory. The qualitative parameters (pH, rate, which varied from 20 to 160 mL min1 (adjusted by a mass
conductivity, DOC, chemical oxygen demand (COD), and total sus- flow controller (Fig. 1f) and the electric intensity of the ozone
pended solids (TSS)) of the secondary-treated wastewater samples generator). The ozone concentration was normalised to the DOC
used during the experimental period were determined via routine values of each sample at each treatment time point. The examined
analysis (Table S2). The experiments were performed with specific ozone doses were 0.125, 0.25, 0.50 and 0.75 gO3 gDOC 1.
secondary-treated effluents, as collected or spiked with a mixture The ozone concentration in the liquid phase was measured by the
of the target antibiotics (AMP, AZM, CLA, ERY, OFL, SMX, TMP and Indigo colorimetric method (Bader and Hoigne , 1981). The ozone
TC) with an initial concentration of 100 mg L1, without prior pH escaping the reactor in the gas phase was carefully removed by gas
adjustment (the inherent wastewater pH values ranged between washing bottles filled with a KI solution (Fig. 1g). Two sets of ex-
7.2 and 7.8) and at the inherent temperature values (22e25 C). It is periments were performed using: (1) wastewater effluents as
noted that the inherent concentrations of the antibiotics in the collected and (2) wastewater effluents spiked with the mixture of
wastewater samples were, for half of the examined antibiotics, the selected antibiotics. Moreover, experiments with the addition
Table 1
Occurrence of antibiotics (expressed as a concentration range) in secondary-treated wastewater effluent samples taken from the secondary clarifier of the
UWWTP. The concentration range refers to the samples taken between October 2017 and January 2018.
Compounds Method detection limit (MDL) (ng L1) Concentration range (ng L1)
Ampicillin (AMP) 0.97 <MDL

Azithromycin (AZM) 0.15 184.5e358.8
Clarithromycin (CLA) 0.01 433.2e474.8
Erythromycin (ERY) 0.74 <MDL
Ofloxacin (OFL) 1.01 <MDL - 6.12
Sulfamethoxazole (SMX) 0.18 <MDL
Tetracycline (TC) 1.12 <MDL
Trimethoprim (TMP) 0.41 129.5e334.1
MDL ¼ Method Detection Limit.

Fig. 1. Experimental set-up of the ozonation reactor operating in continuous mode: (a) bubble column reactor filled with glass rings for gas-liquid-solid contact, (b) peristaltic
pump, (c) inlet solution (secondary-treated effluents), (d) outlet solution, (e) ozone generator, (f) mass flow controller and (g) ozone gas destroyer.
of H2O2 at different concentrations (i.e. 0, 0.04, 0.08 and 0.12 mM) autosampler (SIL-30AC), an oven (CTO-20AC), a degasser (DGU-20A
were also performed. The selected concentrations were based on 5R) and a system controller (CBM-20A), coupled to a triple quad-
previous studies (Valero et al., 2015), in an attempt to optimize the rupole mass spectrometer detector (Ultra-Fast Mass Spectrometry
concentration of H2O2 able to enhance the degradation efficiency of series LCMS-8040) with an ESI source operating in both positive
pollutants by catalytic decomposition of O3, generating HO, and negative ionization modes. An analytical method described
avoiding the use of an excessive amount which could favour the elsewhere was used (Barbosa et al., 2018) for the detection and
reaction with HO to form a weaker hydroperoxyl radical (HO2) quantification of the target antibiotics. A Kinetex™ 1.7 mm XB-C18
(Ribeiro et al., 2019). 100 Å column (100 2.1 mm i.d.) supplied by Phenomenex, Inc.
Prior to chromatographic analysis, the obtained samples were (California, USA) was employed. The mobile phase consisted of
transferred immediately to glass vials with a stoichiometric excess ultrapure water and methanol, both acidified with 0.1% formic acid,
of catalase solution in order to remove any residual oxidants. All operating at gradient mode, with a flow rate of 0.25 mL min1.
samples were subsequently filtered through 1.2-mm glass-fiber fil- Column oven and autosampler temperatures were set at 30 C and
ters (47 mm GF/C, Whatman™; Maidstone, United Kingdom). The 4 C, respectively. The volume of injection was 5 mL. The optimized
non-spiked wastewater samples were pre-concentrated by solid ESI parameters were the following: nebulizing gas flow,
phase extraction (SPE) using Oasis® HLB cartridges, as described 3.0 L min1; drying gas flow, 15 L min1, capillary voltage, 4.5 kV;
elsewhere (Ribeiro et al., 2015). Each experiment was performed in desolvation temperature, 400 C; and source temperature, 250 C.
triplicate and average values are quoted as results. The error bars Argon at 230 kPa was the collision induced dissociation gas (CID).
depicted in the figures represent the relative standard deviation The two selected reaction-monitoring (SRM) transitions
(RSD) of three independent measurements derived from three (Table S3) between the precursor ion and the two most abundant
separate experimental runs. RSD values were always below 10%. fragment ions were employed for quantification (SRM1) and
identity confirmation (SRM1/SRM2 ratio). All MS data was pro-
cessed using the software package LC Solution Version 5.41SP1.
2.4. Analytical methods A Shimadzu TOC-5000A instrument (Shimadzu Scientific In-
struments, Japan) was employed in order to monitor the DOC in the
For Ultra-High Performance Liquid Chromatography with tan- samples before and after treatment and therefore, to assess the extent
dem Mass Spectrometry (UHPLCeMS/MS) analysis, a Shimadzu of the mineralization. The pH and conductivity were measured with a
Corporation apparatus (Tokyo, Japan) was used, consisting of a pH meter pHenomenal® pH 1100 L (VWR, Germany).
UHPLC equipment (Nexera) with two pumps (LC-30AD), an
2.5. Enumeration of total and antibiotic-resistant Escherichia coli ARGs are provided in Section 3.3). Three sets of experimental
conditions were selected to assess the effect of the experimental
Bacteria enumeration was performed using the membrane parameters (i.e. HRT and specific ozone dose) on the reduction of
filtration method, as described elsewhere (Novo and Manaia, 2010; the abundance of the selected genes: (i) 0.25 g O3 gDOC 1 and HRT
Michael et al., 2012). The chromogenic agar of tryptone bile X- of 10 min, which can be considered as a realistic scenario in a full-
glucuronide (t-BX) (Sigma Aldrich, Steinhein, Germany) was used scale UWTP, with many studies in the literature reporting these
for the detection and enumeration of Escherichia coli. For the conditions; (ii) 0.75 gO3 gDOC 1 and HRT of 40 min, which were
enumeration of ARB, the medium was spiked with TMP or SMX, at the optimum conditions determined in this study for the degra-
concentrations of 16 or 516 mg L 1, respectively. These concen- dation of the target antibiotics and the bacterial inactivation; and
trations were chosen based on the minimum inhibitory concen- (iii) 0.25 gO3 gDOC 1 and HRT of 40 min, in order to examine the
tration (MIC) of each of the examined antibiotics, as determined in influence of lower ozone dose along with higher contact time.
the Clinical and Laboratory Standards Institute (CLSI) (Wayne, The total genomic DNA extraction was performed using the
2007). Serial dilutions of each sample were prepared in a saline PowerWater® DNA Isolation Kit (MOBIO Laboratories Inc.)
solution (NaCl 0.85%) and filtered through mixed cellulose ester following filtration of the samples (volume 900e1000 mL) through
membranes (0.22 mm pore size, Millipore), which were placed onto polycarbonate filter membranes (45 mm diameter, 0.22 mm pore
the culture medium and incubated for 24 h at 44 ± 1 C for E. coli. size, Millipore). DNA quantification was assessed using a Qubit 3.0
Bacteria counts, expressed as Colony Forming Units per mL (CFU/ Fluorometer (ThermoFIsher Scientific). Quantitative polymerase
mL), were enumerated on the respective culture media, with the chain reaction (qPCR, StepOne TM Real-Time PCR System, Life-
Limit of Quantification (LOQ) of E. coli being equal to 1 CFU Technologies, Carlsbad, CA, USA) using SYBR Green chemistry was
100 mL1, since 100 mL was the maximum volume filtered. The employed to measure the abundance (per mL of sample) of aadA1,
prevalence of antibiotic resistant E. coli was assumed as the ratio of dfrA1, qacED1 and sul1, intI1 and 16S rRNA. Calibration curves based
CFU mL1 observed on the medium containing the antibiotic and on adequate ten-fold dilution series of the standards for the
that CFU mL1 on the medium without antibiotic, as described respective gene were run along with the test samples. Primers in-
elsewhere (Novo and Manaia, 2010; Karaolia et al., 2017). Addi- formation and qPCR conditions are listed in Table 2.
tionally, the regrowth potential of the examined bacterial groups
was investigated in ozonated samples in which the examined 2.7. Phytotoxicity assays
bacterial groups were at levels below LOQ. Regrowth was examined
after 24, 48 or 72 h of storage in the dark at both 24 ± 1 C and The phytotoxicity assessment of samples collected before and
44 ± 1 C. This investigation took place in order to examine after the ozonation treatment (at the same experimental conditions
whether the reduction of the examined bacterial groups denoted mentioned in Section 2.6) was carried out in triplicate using the
their permanent inactivation, or if it was a transient effect of injury Phytotestkit microbiotest (MicroBioTests Inc.). The selection of this
and temporary inactivation, with regrowth after some time for phytotoxicity assay among various toxicity tests, aimed at
repair/reactivation. The examination of the ozonation regrowth providing evidence on the suitability of the ozonation process as a
capacity aimed at giving insights on the potential bacterial tertiary treatment for wastewater effluents intended for reuse in
regrowth after treatment, in order to have an indication of what agricultural irrigation. The following plants were used in this assay,
could happen in terms of faecal contamination, if the treated due to their rapid germination and growth of roots and shoots, and
wastewater was discharged in a natural receiving environment or their sensitivity to low concentrations of phytotoxic substances:
stored to be used for reuse purposes. monocotyl Sorgho (Sorghum saccharatum), dicotyl garden cress
(Lepidium sativum), and dicotyl mustard (Sinapis alba). A control
2.6. Quantification of selected ARGs test was performed using tap water. The data interpretation is
extensively described elsewhere (Michael et al., 2012).
Total genomic DNA was used to monitor selected genes and
ARGs in the treated samples obtained from different experimental 3. Results and discussion
runs performed at different operating conditions. The genes
examined to this study were the four ARGs of aadA1, dfrA1, qacED1 3.1. Degradation of antibiotics
and sul1, the class 1 integron e integrase (intI1) and the house-
keeping gene of 16S rRNA (the reasons of selecting these genes/ The efficiency of the ozonation process, operated in continuous
Table 2
List of primers and conditions used to detect the target genes.
Target Primers Primers sequence Conditions Primers reference

gene
16S rRNA 1114F CGGCAACGAGCGCAACCC 95 C for 10 min (1 cycle); 95 C for 15 s, 55 C for 20 s and 72 C for 10 s (Denman and Mcsweeney,
1275R CCATTGTAGCACGTGTGTAGCC (35 cycles) 2006)
aadA1 aadA-01F GTTGTGCACGACGACATCATT 95 C 2 min (1 cycle), 95 C 15 s - 60 C 1 min (40 cycles) Zhu et al. (2013)
aadA-01R GGCTCGAAGATACCTGCAAGAA
dfrA1 dfA1F GGAATGGCCCTGATATTCCA 95 C 10 min (1 cycle), 95 C 15 s - 60 C 1 min (40 cycles) Zhu et al. (2013)
dfA1R AGTCTTGCGTCCAACCAACAG
qacED1 qacED1- CCCCTTCCGCCGTTGT
95 C 5 min (1 cycle), 95 C 10 s - 60 C 30 s (35 cycles) Zhu et al. (2013)
02F
qacED1- CGACCAGACTGCATAAGCAACA
02R
sul1 sul1-FW CGCACCGGAAACATCGCTGCAC 95 C 5 min (1 cycle), 95 C 15 s - 60 C 30 s (35 cycles) Pei et al. (2006)
sul1-RV TGAAGTTCCGCCGCAAGGCTCG
intI1 intI1-LC1 GCCTTGATGTTACCCGAGAG 95 C 10 min (1 cycle), 95 C 15 s - 60 C 1 min (40 cycles) Barraud et al. (2010)
intI1-LC5 GATCGGTCGAATGCGTGT
mode, in degrading the target antibiotics was assessed under 0.125 gO3 gDOC 1 (Fig. S2). However, considering that the appli-
various HRTs (i.e. 10, 20, 40 and 60 min) and specific ozone con- cation of ozonation at a full-scale UWTP for 40 min is not cost-
centrations (i.e. 0.125, 0.25, 0.50 and 0.75 gO3 gDOC1). These effective and that no significant differences were found between
conditions (i.e. HRTs and ozone doses) were selected based on the the different applied ozone concentrations for the HRTs of 10 and
findings of previous studies dealing with the application of ozon- 20 min, a new set of experiments was carried out to determine the
ation for the removal of various organic microcontaminants optimum experimental conditions with respect to the antibiotics’
(Reungoat et al., 2012; Prieto-Rodríguez et al., 2013; Margot et al., degradation at the lowest HRTs of 10 and 20 min, by examining the
2013; Marce et al., 2016; Sousa et al., 2017), and taking into ac- contribution of the addition of hydrogen peroxide (H2O2) to the
count the full-scale applicability of the process in relation to the overall process efficiency.
operating cost. The latter was taken into consideration by applying The degradation of the target antibiotics was investigated by
the lowest ozone dose able to effectively eliminate the examined comparing single ozonation (absence of H2O2) and ozonation
compounds under continuous mode operation. The results assisted by the addition of H2O2 at different concentrations (i.e.
revealed that the spiked antibiotics, in mixture, were efficiently 0.04, 0.08 and 0.12 mM), which were selected according to the
degraded (>99%) by ozonation applying both HRTs of 40 and findings reported in previous studies (Cho and Yoon, 2006; Lanao
60 min, at all examined ozone concentrations (Figs. S2c and d). For et al., 2008; Valero et al., 2015). The addition of H2O2 to the ozon-
the lower HRTs of 10 min and 20 min with an ozone dose of 0.125 ation process leads to the so-called peroxone process (O3/H2O2),
gO3 gDOC 1 (Figs. S2a and b), only SMX, OFL, and TC were which is expected to promote higher removals of organic com-
degraded rapidly below the LOD, while AMP, AZM, CLA, ERY and pounds than single ozonation, due to the enhanced ozone decay
TMP were degraded up to 84, 73, 98, 70 and 96%, respectively. and production rate of HO (Von Gunten, 2003; Von Sonntag and
When gradually increasing the ozone dose, the elimination of the Von Gunten, 2012). Fig. 2 shows the effect of the addition of H2O2
parent compounds of all antibiotics was shown to slightly increase on the degradation of each antibiotic by ozonation for the HRT of
for both HRTs of 10 and 20 min (10 min). 10 min, as a function of the applied ozone concentration.
This behaviour was expected even at low ozone doses, since this It can be observed in Fig. 2bed that the increase of the oxidant
oxidant is able to either selectively attack double bonds and mol- concentration had a marginal influence on the degradation of the
ecules containing electron rich moieties present in the different examined antibiotics. The results obtained in this study revealed
classes of the examined antibiotic compounds (e.g., macrolides, that the removal of the antibiotics which were not completely
sulfonamides and fluoroquinolones) or develop a radical oxidation degraded by single ozonation (AMP, AZM, and ERY), was similar
pathway due to its self-decomposition at the natural pH of the with that observed when applying the O3/H2O2 process for all the
wastewater (pH 7.3e7.8). Considering that the experiments were tested concentrations of H2O2. AMP, AZM and ERY were not
conducted at the inherent pH of the wastewater effluents, it can be completely removed after 10 min of treatment under all the tested
assumed that the elimination of the parent compounds was pre- ozone doses and H2O2 concentrations, and their removal percent-
dominantly driven by HO, rather than the direct reaction with ages by single (O3) and combined (O3/H2O2) treatments was
molecular O3 (Michael-Kordatou et al., 2017). The results presented differed only by 2e3%. This may be caused by the rapid consump-
herein are consistent with those reported in other relevant scien- tion of ozone by dissolved effluent organic matter (dEfOM), when
tific studies using different reactor configurations and applying using low specific ozone doses (gO3 gDOC1 < 0.5). This reaction
different ozone concentrations (Reungoat et al., 2012; Lee et al., outcompetes the relatively slow reaction of ozone with H2O2 (Von
2013; Margot et al., 2013; Prieto-Rodríguez et al., 2013). For Sonntag and Von Gunten, 2012; Lee et al., 2014; Miklos et al., 2018).
instance, Lee et al. (2013) reported a degradation higher than 90% Similarly, Acero and Von Gunten (2001) also reported the effect of
for TMP and SMX in secondary-treated effluents treated by a the wastewater qualitative parameters on the efficiency of ozona-
bench-scale semi-continuous ozonation. Another similar bench- tion and O3/H2O2.
scale study using ozonation at semi-operation mode demon- Several studies have reported similar findings, indicating that
strated that the elimination of the parent compounds of sulfon- the enrichment of the ozonation reaction with H2O2 slightly
amide and macrolide antibiotics spiked in ultrapure water and in affected the degradation of microcontaminants. For instance, Lee
pharmaceutical wastewater effluents was achieved in less than et al. (2013) reported negligible variations when adding H2O2 in
20 min (Lin et al., 2009). In a pilot-scale ozonation study (0.32e1.23 wastewater (the H2O2/O3 molar ratio and ozone doses ranged be-
gO3 gDOC1; HRTs up to 20 min), Margot et al. (2013) reported the tween 0.5-1.0 and 0.25e1.5 gO3 gDOC1, respectively), when using
removal of 40 microcontaminants (including AZM, CLA, TMP, and a lab-scale ozone reactor operating in batch mode to examine the
SMX) over 80% from secondary-treated wastewater effluents. A degradation of 16 microcontaminants, including TMP and SMX, in
removal of 90% of microcontaminants (including CLA, ERY, OFL, secondary-treated effluents. Similar results were obtained in the
SMX and TMP) from secondary-treated effluents was reported after study of Lee et al. (2014) who reported that the removal of the
a treatment time of 20 min using pilot-scale ozonation operating in tested microcontaminants during O3/H2O2 process increased only
a semi-continuous mode (0.19 gO3 gDOC1) (Prieto-Rodríguez slightly (<10%) by the addition of H2O2 (molar ratio H2O2/O3: 0,
et al., 2013). Similar results were reported by Reungoat et al. 0.25, and 0.5). In a pilot-scale ozonation (0.64e1.08 gO3 gDOC1;
(2010), in a study dealing with full-scale ozonation (5 mgO3 L1 HRTs 12e23 min; addition of H2O2 up to 2.5 mg L 1), Kovalova et al.
and HRT 15 min) of secondary-treated wastewater. Among the 25 (2013) reported a negligible improvement of ±10% for micro-
selected microcontaminants quantified in the samples, 9 showed a contaminants elimination by O3/H2O2 in comparison to single
reduction of more than 90% (including ERY, TMP, and SMX) and 13 ozonation.
others were reduced by more than 70%. Therefore, these results showed that the addition of H2O2, which
Overall, the degradation of the parent compounds of the anti- would increase the operating cost of the process, did not enhance
biotics was improved with increasing the HRT and ozone dose, as remarkably the degradation of the examined compounds. However,
expected. According to the aforementioned results obtained from it is important to mention that in the experiments performed with
the spiked experiments, the HRT of 40 min could be considered as non-spiked wastewater effluents (inherent antibiotics’ concentra-
optimum, since it allowed to achieve a rapid and complete elimi- tions), even the mildest experimental conditions tested with single
nation of the parent compounds of all the examined antibiotics at ozonation (HRT of 10 min and 0.125 gO3 gDOC 1) were found to be
mg L1 level, even when using the lowest applied ozone dose of able to provide a removal of the examined microcontaminants
Fig. 2. Degradation of the target antibiotics by continuous mode ozonation with an HRT of 10 min, varying the O3 doses (0.125, 0.25, 0.5 and 0.75 gO3 gDOC1) and the H2O2
concentration: (a) 0 mM, (b) 0.04 mM, (c) 0.08 mM, and (d) 0.12 mM of H2O2). Experimental conditions: [A0] ¼ 100 mg L1; matrix: secondary treated effluents; pH 7.3e7.8;
T ¼ 24 ± 1 οC.
(Table 1) to concentrations below their limit of detection. In order impacted environments, together with their persistence in the
to investigate whether the mildest ozonation conditions were environment, as well as genome plasticity, make them important
adequate towards the inactivation of ARB and removal of ARGs, a tracers to assess the antibiotic resistance status of environmental
new series of optimization experiments was performed with samples. Additionally, in previous studies including both Gram-
varying ozone doses and HRTs, as discussed in the following negative and Gram-positive faecal indicator bacteria (E. coli and
Sections. Enterococcus spp., respectively), noteworthy differences were not
observed regarding their inactivation. Comparing both groups,
Enterococcus spp. has usually lower abundance and suffers higher
3.2. Inactivation of total and antibiotic-resistant cultivable inactivation than E. coli (Moreira et al., 2016, 2018; Biancullo et al.,
Escherichia coli 2019). It is also noted that the selection of E. coli as the target
bacterium in this study was based on the fact that the latter is
The ozonation process operated in continuous mode was included in the Proposal for a “Regulation of the European Parlia-
investigated with regard to its efficiency to inactivate the total and ment and of the Council”, on minimum requirements for water
antibiotic-resistant cultivable E. coli. The selection of E. coli was reuse (EU, 2019). The selection of SMX and TMP to investigate the
based on its role as indicator of microbiological contamination. antibiotic resistance was based on their wide use and frequent
E. coli and faecal coliforms in general, are used worldwide as in- detection in wastewater and aquatic environments.
dicators of faecal contamination, in particular to assess the micro- The initial abundance of E. coli in the secondary-treated efflu-
biological water/wastewater quality and are included in the ents was up to 8 103 CFU mL1. The experiments were performed
wastewater legislation for reuse purposes (APHA 2005, ISO9308- in triplicate for all the examined HRTs and ozone concentrations.
1:2000). E. coli are excreted in faeces and consequently they The obtained results regarding the inactivation of total E. coli
reach UWTPs, where they can survive and thrive throughout the (Fig. 3a), indicated by absence of colonies in 100 mL of treated
CAS treatment (Forster et al., 2002; Wellington et al., 2013; Karaolia wastewater (<LOD ¼ 0.01 CFU mL1), were found using an HRT of
et al., 2017). Therefore, the ubiquity of these bacteria in human-
of total heterotrophs, enterococci and coliforms (including of

antibiotic resistant E. coli strain A2FCC14) was higher than 99%,
within 30 min of ozonation (Sousa et al., 2017).
The capacity of ozonation operated in continuous mode, was
also investigated in terms of its efficiency to inactivate antibiotic-
resistant E. coli. The obtained results (Fig. S3) demonstrated the
inactivation, below LOD, of E. coli harbouring resistance to TMP and
SMX, by increasing HRTs and applied ozone doses. The inactivation
of the examined ARB occurred for HRTs of 10 and 20 min and for
ozone doses of 0.5 and 0.25 gO3 gDOC1, respectively. The apparent
observation that cultivable TMP- and SMX-resistant E. coli was lost
earlier than the total cultivable E. coli is probably due to the lower
abundance of the first, which had already reached the LOQ value.
Indeed, those results were expected due to the lower number of
bacteria exhibiting resistance to TMP and SMX. It should be noted
however that oxidizing agents (i.e. O3 and HO) might have led to
the induction of a ‘Viable But Not Cultivable state’ (VBNC) in the
bacteria, explaining that non-cultivable bacteria were detected in
the presence of antibiotics. Michael-Kordatou et al. (2017) observed
that total E. coli and ERY-resistant E. coli were inactivated to values
below LOQ within 15 min, using an ozone dose of 0.3 mgO3 L 1
during batch-mode ozonation. Oh et al. (2016) investigated ozon-
ation at batch mode with an ozone dose of 3 mgO3 L 1, for the
potential inactivation of E. coli containing the 64,508 bp nucleotide
sequence of the Inc-P-1beta antibiotic resistance plasmid pB10,
which has multiple resistance to a number of antibiotics. It was
shown that after 15 min of contact time, the total and resistant
E. coli reduced by more than 90%. Alexander et al. (2016) investi-
gated the abundance of ARB in wastewater effluents treated by
ozonation (0.9 gO3 gDOC1 and HRT of 18 ± 2 min), with Entero-
coccus spp. exhibiting the highest susceptibility to ozone (98%
reduction), and P. aeruginosa the highest tolerance. Moreira et al.
(2016) reported the inactivation to vales below LOQ of total het-
erotrophs and Enterococcus spp., as well as of bacteria harbouring
resistance to ciprofloxacin, gentamicin and meropenem, after
26 min of treatment with ozone dose of 50 g (Nm3)1 in a lab-scale
continuous mode ozonation system.
The results obtained for the HRTs of 10 and 20 min, revealed the
sensitivity of the indigenous E. coli (total cultivable and antibiotic-
resistant E. coli) towards the oxidizing agents (O3 and HO). This
could be attributed to the physical vulnerability and the damage of
the cell wall and membrane, which may then lead to the cell lysis
Fig. 3. Inactivation profile of E. coli (expressed as E. coli CFU mL1) by continuous mode
and consequent leakage of cell components to the external envi-
ozonation (a) varying the O3 doses (0.125, 0.25, 0.50 and 0.75 gO3 gDOC1) and HRTs
(10, 20, 40, 60 min); and (b) using an HRT of 10 min and varying the O3 doses and H2O2 ronment (Zuma et al., 2009; Dodd, 2012; Von Sonntag and Von
concentrations (0 mM, 0.04 mM, 0.08 mM, 0.12 mM of H2O2). Experimental conditions: Gunten, 2012; Pak et al., 2016).
[A0] ¼ 100 mg L1; matrix: secondary treated effluents; pH 7.3e7.8; T ¼ 24 ± 1 οC. The regrowth potential of E. coli was investigated for the two
mildest operating conditions, which led to their inactivation to
levels below the LOQ just after treatment (i.e., according to Fig. 3a:
60 min for all the examined ozone concentrations. Similarly, an HRT
HRT of 20 min and 0.75 gO3 g DOC 1; HRT of 40 min and 0.25 gO3 g
of 40 min was found to be adequate to inactivate E. coli to values
DOC 1). The idea behind this approach was to understand the al-
below LOD for almost all tested ozone concentrations, except 0.125
terations on the faecal microbiota occurring in treated wastewater
gO3 g DOC 1. For the HRT of 20 min, the higher ozone concentra-
stored for reuse purposes. Incubation periods of 24, 48, and 72 h
tion tested (0.75 gO3 gDOC1) was the only one able to inactivate
and temperatures set at 24 ± 1 C and 44 ± 1 C were employed. The
E. coli to values below the LOD, whereas inactivation was not
selection of the incubation temperatures was made according to
attained when using an HRT of 10 min. One study dealing with
the environmental conditions (environmental temperature:
similar experimental conditions, reported a 1-log reduction of
22e25 C), and the maximal temperature tolerated by faecal co-
E. coli in a pilot-scale ozonation reactor treating secondary-treated
liforms, namely E. coli (44e45 C). The environmental incubation
effluents with an ozone dose up to 0.73 gO3 gDOC 1 and a contact
temperature constitutes a more realistic approach concerning the
time of 20 min (Lüddeke et al., 2015). Also, Ostoich et al. (2013)
regrowth potential of the treated samples, whereas incubation at
reported the complete inactivation (<LOD ¼ 0.01 CFU mL1), of
44 C provided an indication on the potential of regrowth under a
faecal coliforms by ozonation in a UWTP of Italy, using ozone doses
stressful temperature value. Fig. 4 shows that ozonation operated
varying from 10 to 15 mgO3 L1 and a contact time of 30 min.
under low ozone dose and high HRT was efficient to avoid the
Another study focusing on the inactivation of cultivable bacterial
regrowth at 24 ± 1 C, whereas the experiments using higher ozone
populations after batch-mode ozonation under different contact
dose and lower HRT led to the regrowth of E. coli at both temper-
times and ozone doses up to 50 g (Nm3)1, showed that the removal
atures, suggesting that shorter contact time is a key factor
Fig. 4. (a) Regrowth profile of E. coli (expressed as CFU mL1) after continuous mode ozonation of wastewater samples (a) using a HRT of 20 min and an O3 dose of 0.75 gO3 gDOC1
and (b) using an HRT of 40 min and an O3 dose of 0.25 gO3 gDOC 1. Experimental conditions: [A0] ¼ 100 mg L1; matrix: secondary treated effluents; pH 7.3e7.8; T ¼ 24 ± 1 οC.
determining the observed regrowth. The results of this study injured, and may need time to repair the damages. Depending on
indicated that low contact times at high ozone doses were not the extent of cell damage, the time needed to recover may vary
sufficient to induce non-cultivability in the E. coli wastewater among surviving cells. Equally, it is expected that they will not grow
population. As consequence, the microorganisms affected by ozone at the same rate. According to the scientific literature (Lüddeke
and HO maintained viability, and therefore exhibited regrowth et al., 2015; Pak et al., 2016; Marce et al., 2017; J€
ager et al., 2018),
after a certain time period, when the stress was relieved. It is ozonation time, the applied ozone concentration, the physical
important to mention here that a high number of cells present in characteristics of the examined organisms and the wastewater
wastewater may be aggregated in flocs of different sizes. Hence, composition, are the major parameters affecting the bacterial
inner bacterial cells may be protected from the deleterious effects regrowth after ozonation.
of ozone and radicals. Therefore, after treatment some cells can be Further experiments in the presence of ozone and H2O2 (0.04,
Fig. 5. Examined genes, expressed as log10 mean values, for the examined experimental conditions of HRT of 10 min with 0.25 gO3 gDOC1, HRT of 40 min with 0.25 gO3 gDOC1 and
HRT of 40 min with 0.75 gO3 gDOC1 and their mean values after 72 h of incubation in the dark at room temperature (22e25 C). Mean values are represented in log10 copies
number per 1 mL of sample. Experimental conditions: matrix: secondary treated effluents; pH 7.3e7.8.
0.08 and 0.12 mM), did not reveal significant differences in terms of been suggested as proxies for acquired antibiotic resistance in the
E. coli inactivation (Fig. 3b), in comparison to the experiments environment (Gillings et al., 2015). These genetic elements have
performed in the absence of H2O2 at the same experimental con- conserved regions, namely intI1, sul1 and qacDE1, and a variable
ditions (HRT of 10 min, ozone doses: 0.125, 0.25, 0.5 and 0.75 gO3 region which often includes genes such as aadA1 or dfrA1 (Ferreira
gDOC1). Specifically, a slight reduction (<5%) was observed as the Da Silva et al., 2007, Moura et al., 2012).This study aimed at
H2O2 concentration increased. However, this reduction falls within assessing the reduction of the abundance of intI1 and the conserved
the statistical error of the measurements. Similar results were ob- genes of the class 1 integrons and at investigating if ozonation led
tained for the inactivation of TMP- and SMX-resistant E. coli to gene excision, herein indicated by a reduction in aadA1 and dfrA1.
(Fig. S4). It is important to note that the results obtained in the Genes quantification was performed to understand how the pro-
presence of H2O2 for the inactivation of total and resistant bacteria duced oxidizing agents (i.e. O3 and HO) interact with the integrons
were in agreement with those observed in the case of the degra- and if the ozonation treatment is capable of efficiently dis-
dation of antibiotics, i.e. the process efficiency was not affected by integrating this mobile genetic element (Zhang et al., 2009).
the addition of H2O2. The comparison between secondary-treated effluent and ozo-
Hübner et al. (2012) investigated the bacteria inactivation in nated samples showed that the reduction of the abundance of
secondary-treated effluents by ozonation with the addition of H2O2 ARGs, bacterial load and intI1 were HRT- and dose-dependent. An
and reported no differences after the addition of H2O2. Similar re- HRT of 40 min and 0.75 gO3 gDOC1 were found to be the optimum
sults were also found by Valero et al. (2015), during the inactivation operating conditions for the reduction of the genes abundance
of Enterococcus spp. through ozonation in the presence and absence (Fig. 5). The log reduction varied between 2 and 3, 3e4 and 4e6, for
of H2O2. These findings can be attributed to the strong competition an HRT of 10 min and 0.25 gO3 gDOC 1, HRT of 40 min and 0.25 gO3
of H2O2 with the highly reactive moieties present in the dEfOM gDOC1 and HRT of 40 min with 0.75 gO3 gDOC1, respectively. The
occurring in the secondary-treated effluents, for consumption of results revealed that when using an HRT of 40 min, even when
ozone molecules (Pocostales et al., 2010; Miklos et al., 2018). applying a low ozone dose, the reduction of the abundance of the
examined genes was similar to that achieved when using the
higher ozone dose, suggesting that an appropriate HRT would lead
3.3. Removal of selected genes and ARGs
to the desired disintegration levels of the examined ARGs, even
with low ozone doses. Ozone disinfection, with an appropriate HRT
Continuous ozonation was evaluated with regard to its effi-
and ozone dose, is apparently enough to compromise DNA integrity
ciency to remove selected genes and ARGs. Three experimental
into a condition in which it could not act as PCR template. These
conditions as mentioned previously (Section 2.7) were selected to
results are in agreement with the fact that once the bacteria cell
assess the influence of the experimental parameters (i.e. HRT and
surface is compromised by oxidizing agents, and the cell interior is
ozone dose) towards the selected genes. The housekeeping 16S
exposed to the external environment, O3 and HO may interfere
rRNA gene was used as a biomarker for bacteria, aiming at assessing
with DNA (Von Sonntag and Von Gunten, 2012; Dodd, 2012; Pak
the reduction in the load of total bacteria. Class 1 integrons have
Fig. 6. Examined genes, expressed as the relative abundance of the log10 mean values, for the examined experimental conditions of HRT of 10 min with 0.25 gO3 gDOC1, HRT of
40 min with 0.25 gO3 gDOC1 and HRT of 40 min with 0.75 gO3 gDOC1 and their mean values after 72 h of incubation in the dark at room temperature (22e25 C). Mean values are
represented in log10 copies number per 1 mL of sample. Experimental conditions: matrix: secondary treated effluents; pH 7.3e7.8.
et al., 2016). the complexity of the microbiota in the wastewater may be the
One important observation made in this study was that the major factors explaining the variations of the efficiency of ozona-
resistant E. coli decreased throughout ozonation, whereas the tion treatment (Zhuang et al., 2015; Sousa et al., 2017). The fluc-
reduction of the abundance of the selected genes observed after tuation of the wastewater qualitative characteristics can influence
treatment was apparently transient. The reactivation of the the reaction mechanism of ozone affecting thus the overall disin-
respective bacterial hosts may explain that in the samples treated fection efficiency towards the living cells in UWTPs. Although not
with 0.25 gO3 gDOC1 and an HRT of 10 min, the examined genes covered in this study, according to Pak et al. (2016) damage of the
were detected after 72 h at almost the same levels as the initial surface of the cells by ozonation may lead to the release of DNA. If
values (1-log cycle below the initial), under dark and ambient relatively intact, ARGs containing free fragments may be trans-
temperature (22e25 C) (Fig. 5). Additionally, from Fig. 6, which ferred to other microorganisms, through horizontal mechanisms,
presents the relative abundance of the examined genes, it can be and contribute to the spread of antibiotic resistance. All those fac-
observed that despite the detected reduction of the abundance of tors seemed to have influence on the ozonation efficiency and more
the examined ARGs and integrase genes after the treatment, an detailed studies are needed to identify the proper conditions
increase in the prevalence of these genetic determinants was required to achieve the optimum performance of the ozonation
detected 72 h after the treatment. These observations were ex- treatment in terms of genes and ARGs inactivation.
pected and are in accordance with the results presented in Section
3.2, where no total inactivation of the examined bacteria was ob-
3.4. Mineralization and phytotoxicity assessment
tained for the same samples. Having in mind that the 16S rRNA
gene is a biomarker of the load of total bacteria, its increase in the
Fig. 7a shows the results for the mineralization observed in the
treated samples under all operating conditions tested implies that
treated effluents, in terms of DOC removal. It must be pointed out
other bacterial groups thriving in wastewater (e.g., Aeromonads,
that the contribution of the antibiotics to the DOC of the spiked
Pseudomonads), surviving ozonation, have contributed to the in-
wastewater effluent, was negligible since the DOC of the samples
crease of the abundance of the examined genes. A recovery on the
examined genes was also observed for the more severe experi-
mental conditions of HRT and ozone dose. These results could be
explained through the assumption made by Alexander et al. (2016),
who suggested that bacteria may develop different levels of
robustness or resistance against oxidative stress. Additionally, the
high organic content of the wastewater (Table S2) may react with
the majority of the produced oxidizing species, resulting in a very
small fraction able to cause lethal damages to the live cells existing
in the medium.
Alexander et al. (2016) investigated the efficiency of ozonation
treatment to inactivate selected ARGs in a pilot-scale system
(18 ± 2 min of HRT and ozone dose up to 0.9 ± 0.1 gO3 gDOC1) and
the obtained results showed variations of the examined genes
removed through ozonation, with ARGs of vanA, blaVIM, ampC and
ermB being reduced by 49.9%, 18.7%, 69.8% and 99.3%, respectively.
Zhuang et al. (2015) reported a slight removal of the examined
genes at low ozone doses using a bench-scale system with ozone
dose varying between 27 and 128 mg L1 a fact that may be
attributed to the rapid consumption of ozone by the organic matter
present in wastewater (e.g. COD: 13e29 mgO2 L1). Also, the
inactivation of the selected genes observed by Zhuang et al. (2015)
was found to be enhanced by increasing ozone doses, and this is in
line with the results presented in this study. Continuous mode
ozonation experiments performed in synthetic and real wastewater
samples with constant ozone dose for different contact times (e.g.
15, 30 and 60 min) indicated that increasing contact time led to the
inactivation of the selected genes (e.g. 16S rRNA gene, intI1, blaTEM,
qnrS and sul1) up to 99% (in relation to the initial load before
treatment) after 30 min of exposure (Sousa et al., 2017). In addition,
Sousa et al. (2017) showed that the regrowth after 72 h of incuba-
tion occurred for samples treated for 15 and 30 min, whereas no
reactivation was observed after 60 min, suggesting that this period
of contact time was adequate to kill the majority of the cells.
However, in the real wastewater samples, which are characterized
by higher microbial diversity and chemical complexity, compared
to synthetic wastewater, the inactivation efficiency and the
regrowth potential in the examined samples were lower and more
intense, respectively. This may be ascribed to the fact that the
structural organization of the bacterial cells is different, with real Fig. 7. DOC removal percentages by continuous mode ozonation: (a) varying the O3
doses (0.125, 0.25, 0.5 and 0.75 gO3 gDOC1) and HRTs (10, 20, 40, 60 min); and (b)
wastewater containing a high number of aggregated cells forming using an HRT of 10 min and varying the O3 doses and H2O2 concentrations (0, 0.04 mM,
flocs of different dimensions. 0.08 mM, 0.12 mM of H2O2). Experimental conditions: [A0] ¼ 100 mg L1; matrix: sec-
It is already known that the suspended solids, the dEfOM and ondary treated effluents; pH 7.3e7.8; T ¼ 24 ± 1 οC.
varied between 18.2 and 21.2 mg L1 (Table S2), whereas the required mild ozone doses, while under these conditions the
theoretical DOC corresponding to the 8 spiked antibiotics was reduction of the organic matter (in terms of COD and DOC) was low
0.45 mg L1 (<3%). A significant improvement was observed when (15e25%). dEfOM contains a variety of organic compounds that
increasing both the ozone dose and HRT. Specifically, when during their reaction with ozone and hydroxyl radicals, may lead to
applying an ozone dose of 0.25 gO3 gDOC 1 and HRT of 10 min, the formation of persistent oxidation products. These products may
DOC removal was found to be 10e12%, whereas an ozone dose of be not susceptible to ozone and can be even more toxic than the
0.75 gO3 gDOC 1 and HRT of 60 min led to DOC removal up to 25%. initial compounds.
Several studies are in line with the results presented herein and Phytotoxicity tests using three plants species (Lepidium sativum,
many of them attributed the low yield of mineralization to the Sinapis alba and Sorghum saccharatum) were considered suitable to
formation of recalcitrant organic intermediates deriving from the evaluate the toxicity of the treated wastewater before its use for
oxidation of dEfOM originally present in wastewater (Von Sonntag agriculture irrigation (Rizzo, 2011). To evaluate the phytotoxicity of
and Von Gunten, 2012; Margot et al., 2013; Prieto-Rodríguez et al., the treated samples and assess the influence of the ozone dose and
2013; Rodríguez-Chueca et al., 2015; Marce et al., 2016; Michael- HRT, three experimental conditions reported in Section 3.3 were
Kordatou et al., 2017). In a pilot-scale study dealing with ozona- selected. The phytotoxicity of the treated samples towards
tion of secondary-treated effluents, Liu et al. (2014) reported a DOC L. sativum, S. alba and S. saccharatum was expressed as percentage
removal up to 28%, after 30 min of contact time, using an ozone of inhibition in the seed germination (GI), shoot germination (SI)
concentration of 4 mg L1. Carbajo et al. (2015) observed 35% of and root growth (RI). The experiments were conducted using
DOC removal even using 130 mg L 1 of ozone and a contact time of wastewater effluent samples as collected (Fig. 8a and b) and the
20 min. No remarkable differences were found for DOC removal by same samples spiked with the target antibiotics (Fig. 8c and d), in
adding H2O2 (Fig. 7b), and this is in accordance to the results order to elucidate the role of the oxidation products formed during
observed for the antibiotics in Section 3.1, suggesting that H2O2 did ozonation in the phytotoxicity against the three plants species. The
not improve the removal of the oxidation products formed during raw wastewater samples did not cause any GI, whereas an inhibi-
ozonation. tion effect was observed in the ozonated samples, with
In this work, it was clearly shown that the degradation of the S. saccharatum being the only seed showing a GI in the three sets of
parent antibiotic compounds (down to their limit of detection) experiments (Fig. S5).
Fig. 8. Root growth inhibition (RI) and shoot growth inhibition (SI) of Lepidium sativum, Sinapis alba and Sorghum saccharatum, before and after continuous mode ozonation of
wastewater samples, as collected or spiked at 100 mg L1, using the selected experimental conditions: 0.25 gO3 gDOC 1 and HRT of 10 min; 0.25 gO3 gDOC 1 and HRT of 40 min; and
0.75 gO3 gDOC 1 and HRT of 40 min. Experimental conditions: Matrix: secondary treated effluents; pH 7.3e7.8; T ¼ 24 ± 1 οC.
The main finding of the phytotoxicity assessment (Fig. 8) was in relation to phytotoxicity. Also, more studies are needed to
that the higher ozone doses or contact times increased the phyto- establish conditions capable of limiting regrowth of ARB after
toxicity levels of the treated samples. The inhibition expressed as SI ozonation. These efforts will contribute to improving the quality of
and RI percentages in the spiked experiments were higher than the treated wastewater, while minimising the potential risks of
those observed in the non-spiked experiments. According to Fig. 8, spread of antibiotic resistance in the receiving environment.
the highest SI and RI were observed in the experiment applying
ozone dose of 0.75 gO3 gDOC 1 with HRT of 40 min. These results Acknowledgments
are in agreement with those obtained by Michael-Kordatou et al.
(2017), showing that ozonation resulted in the augmentation of This paper is part of a project that has received funding from the
the phytotoxicity (i.e. root and shoot inhibition) even after 15 min European Union's Horizon 2020 research and innovation pro-
of treatment, indicating that dEfOM oxidation products were gramme under the Marie Skłodowska-Curie grant agreement No
potentially more toxic than the original matrix. This behaviour 675530.
might be due to direct oxidation of various organic constituents This work was also partially supported by projects POCI-01-
present in the dEfOM of the wastewater samples or to their 0145-FEDER-006984 e Associate Laboratory LSRE-LCM, POCI-01-
oxidation via a hydroxyl radical-mediated mechanism. The high 0145-FEDER-006939 (Laboratory for Process Engineering, Envi-
values of DOC and COD of secondary treated effluent used for the ronment, Biotechnology and Energy e UID/EQU/00511/2013) and
implementation of the experiments are an indication of the exis- UID/Multi/50016/2013 - Associate Laboratory Centro de Bio-
tence of those organic constituents. It is well known that a number tecnologia e Química Fina, CBQF, funded by ERDF through
of oxidation products can be formed during ozonation from the COMPETE2020 - POCI and FCT. NFFM and ARR acknowledge PD/BD/
oxidation of both the examined antibiotics and the dEfOM present 114318/2016 and SFRH/BPD/101703/2014, respectively.
in the secondary-treated wastewater effluents (Von Sonntag and
Von Gunten, 2012; Lee and Von Gunten, 2016; Michael-Kordatou Appendix A. Supplementary data
et al., 2017). In fact, the low DOC removal could be correlated
with the existence of ozonation products in the treated samples, Supplementary data to this article can be found online at
which could contribute to the increased phytotoxicity of the sam- https://doi.org/10.1016/j.watres.2019.05.025.
ples. Also, results obtained from other studies using a variety of
toxicological tests indicated that ozonation may cause other bio-
References
logical effects, e.g. toxicity and genotoxicity (Stalter et al., 2010).
The above findings clearly indicate that the ozonation treatment Acero, J.L., Von Gunten, U., 2001. Characterization of oxidation processes: ozonation
is insufficient to completely mineralize the wastewater and poses and the Aop O3/H2O2. J. Am. Water Work. Assoc. 93, 90e100.
Alexander, J., Knopp, G., Do €tsch, A., Wieland, A., Schwartz, T., 2016. Ozone treatment
risks of producing oxidation products with potentially phytotoxic
of conditioned wastewater selects antibiotic resistance genes, opportunistic
effects, in particular when the most severe operating conditions are bacteria, and induce strong population shifts. Sci. Total Environ. 559, 103e112.
tested. Considering that nowadays ozonation finds an increasing Antoniou, M.G., Hey, G., Vega, S.R., Spiliotopoulou, A., Fick, J., Tysklind, M., La Cour
application in full-scale UWTPs, more research should be per- Jansen, J., Andersen, H.R., 2013. Required ozone doses for removing pharma-
ceuticals from wastewater effluents. Sci. Total Environ. 456, 42e49.
formed on this issue in order to identify the specific oxidation APHA, I, 2005. Water quality-Detection and enumeration of Escherichia coli and
products that may be responsible for the observed phytotoxic coliform bacteria-Part 1: Membrane filtration method. International Organi-
effects. zation for Standardization (ISO) Geneva, Switzerland.
Bader, H., Hoigne , J., 1981. Determination of ozone in water by the indigo method.
Water Res. 15, 449e456.
4. Conclusions Barbosa, M.O., Ribeiro, A.R., Ratola, N., Hain, E., Homem, V., Pereira, M.F.R., Blaney, L.,
Silva, A.M., 2018. Spatial and seasonal occurrence of micropollutants in four
Portuguese rivers and a case study for fluorescence excitation-emission
The findings of this study demonstrated that ozonation of
matrices. Sci. Total Environ. 644, 1128e1140.
wastewater operated at continuous mode constitutes a promising Barraud, O., Baclet, M.-C., Denis, F., Ploy, M.-C., 2010. Quantitative multiplex real-
treatment technology for the elimination of the selected time pcr for detecting class 1, 2 and 3 integrons. J. Antimicrob. Chemother.
65, 1642e1645.
A&ARB&ARGs. The results revealed that low ozone doses accom-
Berendonk, T.U., Manaia, C.M., Merlin, C., Fatta-Kassinos, D., Cytryn, E., Walsh, F.,
panied by prolonged treatment time was needed to eliminate the Bürgmann, H., Sørum, H., Norstro €m, M., Pons, M.-N., 2015. Tackling antibiotic
parent compounds of the examined antibiotics to concentration resistance: the environmental framework. Nat. Rev. Microbiol. 13, 310.
below their LOD in the spiked experiments, whereas the mildest Biancullo, F., Moreira, N.F., Ribeiro, A.R., Manaia, C.M., Faria, J.L., Nunes, O.C., Castro-
Silva, S.M., Silva, A.M., 2019. Heterogeneous photocatalysis using uva-leds for
conditions (i.e. lower HRT of 10 min for the same ozone dose of the removal of antibiotics and antibiotic resistant bacteria from urban waste-
0.125 gO3 gDOC 1) were sufficient to degrade the parent com- water treatment plant effluents. Chem. Eng. J. 367, 304e313.
pounds in the non-spiked experiments. No substantial minerali- n, P., García-Calvo, E., Fern
Carbajo, J.B., Petre, A.L., Rosal, R., Herrera, S., Leto andez-
Alba, A.R., Perdigo n-Melo n, J.A., 2015. Continuous ozonation treatment of
zation was observed at the end of the treatment time at all ofloxacin: transformation products, water matrix effect and aquatic toxicity.
experimental conditions examined, a fact that may explain the J. Hazard Mater. 292, 34e43.
enhanced phytotoxicity of the ozonated samples towards the tested Cho, M., Yoon, J., 2006. Enhanced bactericidal effect of O3/H2o2 followed by Cl2.
Ozone: Sci. Eng. 28, 335e340.
plant species. Ozonation operated at high HRT and ozone doses Czekalski, N., Imminger, S., Salhi, E., Veljkovic, M., Kleffel, K., Drissner, D.,
(HRT 40 min; 0.25 gO3 gDOC 1) was capable of inactivating total as Hammes, F., BüRgmann, H., Von Gunten, U., 2016. Inactivation of antibiotic
well as TMP- or SMX-resistant E. coli, with the simultaneous resistant bacteria and resistance genes by ozone: from laboratory experiments
to full-scale wastewater treatment. Environ. Sci. Technol. 50, 11862e11871.
reduction of the abundance of the examined genes. However, Denman, S.E., Mcsweeney, C.S., 2006. Development of a real-time pcr assay for
regrowth cannot be avoided after ozonation, as it was demon- monitoring anaerobic fungal and cellulolytic bacterial populations within the
strated that living cells might still exist in the treated samples even rumen. FEMS Microbiol. Ecol. 58, 572e582.
Dodd, M.C., 2012. Potential impacts of disinfection processes on elimination and
after prolonged treatment. Overall, it was shown that ozonation
deactivation of antibiotic resistance genes during water and wastewater
upon its proper optimisation constitutes an efficient treatment treatment. J. Environ. Monit. 14, 1754e1771.
option for the removal of antibiotic-related microcontaminants EU, 2018. Establishing A Watch List of Substances for Union-Wide Monitoring in the
from wastewater effluents. Nevertheless, it is suggested that after Field of Water Policy Pursuant to Directive 2008/105/Ec of the European
Parliament and of the Council and Repealing Commission Implementing De-
ozonation, post-treatment by biological filtration or activated car- cision (Eu) 2015/495.
bon adsorption, may overcome the main limitations of this process EU, 2019. Regulation of the European Parliament and of the Council on Minimum
Requirements for Water Reuse. Fatta-Kassinos, D., 2017. On the capacity of ozonation to remove antimicrobial
EU, A., 2017. Commission Implementing Decision Establishing A Watch List of compounds, resistant bacteria and toxicity from urban wastewater effluents.
Substances for Union-Wide Monitoring in the Field of Water Policy Pursuant to J. Hazard Mater. 323, 414e425.
Directive 2008/105. Ec Of The European Parliament And The Council. Michael-Kordatou, I., Karaolia, P., Fatta-Kassinos, D., 2018. The role of operating
EU, B., 2017. Minimum Quality Requirements for Water Reuse in Agricultural Irri- parameters and oxidative damage mechanisms of advanced chemical oxidation
gation and Aquifer Recharge - towards A Water Reuse Regulatory Instrument at processes in the combat against antibiotic-resistant bacteria and resistance
Eu Level Re e
dition. genes present in urban wastewater. Water Res. 129, 208e230.
Ferreira Da Silva, M., Vaz-Moreira, I., Gonzalez-Pajuelo, M., Nunes, O.C., Michael, I., Hapeshi, E., Michael, C., Varela, A., Kyriakou, S., Manaia, C., Fatta-
Manaia, C.M., 2007. Antimicrobial resistance patterns in enterobacteriaceae Kassinos, D., 2012. Solar photo-fenton process on the abatement of antibiotics
isolated from an urban wastewater treatment plant. FEMS Microbiol. Ecol. 60, at a pilot scale: degradation kinetics, ecotoxicity and phytotoxicity assessment
166e176. and removal of antibiotic resistant enterococci. Water Res. 46, 5621e5634.
Ferro, G., Guarino, F., Castiglione, S., Rizzo, L., 2016. Antibiotic resistance spread Michael, I., Rizzo, L., Mcardell, C., Manaia, C., Merlin, C., Schwartz, T., Dagot, C., Fatta-
potential in urban wastewater effluents disinfected by uv/H2o2 process. Sci. Kassinos, D., 2013. Urban wastewater treatment plants as hotspots for the
Total Environ. 560, 29e35. release of antibiotics in the environment: a review. Water Res. 47, 957e995.
Forster, S., Snape, J.R., Lappin-Scott, H.M., Porter, J., 2002. Simultaneous fluorescent Miklos, D.B., Remy, C., Jekel, M., Linden, K.G., Drewes, J.E., Hübner, U., 2018. Evalu-
Gram staining and activity assessment of activated sludge bacteria. Appl. En- ation of Advanced Oxidation Processes for Water and Wastewater TreatmenteA
viron. Microbiol. 68, 4772e4779. Critical Review. Water Research.
Gillings, M.R., Gaze, W.H., Pruden, A., Smalla, K., Tiedje, J.M., Zhu, Y.-G., 2015. Using Moreira, N.F., Narciso-Da-Rocha, C., Polo-Lo pez, M.I., Pastrana-Martínez, L.M.,
the class 1 integron-integrase gene as a proxy for anthropogenic pollution. Faria, J.L., Manaia, C.M., Fern n
andez-Iba ~ ez, P., Nunes, O.C., Silva, A.M., 2018. Solar
ISME J. 9, 1269. treatment (H2o2, Tio2-P25 and go-tio2 photocatalysis, photo-fenton) of organic
Hübner, U., Miehe, U., Jekel, M., 2012. Optimized removal of dissolved organic micropollutants, human pathogen indicators, antibiotic resistant bacteria and
carbon and trace organic contaminants during combined ozonation and arti- related genes in urban wastewater. Water Res. 135, 195e206.
ficial groundwater recharge. Water Res. 46, 6059e6068. Moreira, N.F., Sousa, J.M., Macedo, G., Ribeiro, A.R., Barreiros, L., Pedrosa, M.,
Ikehata, K., Jodeiri Naghashkar, N., Gamal El-Din, M., 2006. Degradation of aqueous Faria, J.L., Pereira, M.F.R., Castro-Silva, S., Segundo, M.A., 2016. Photocatalytic
pharmaceuticals by ozonation and advanced oxidation processes: a review. ozonation of urban wastewater and surface water using immobilized Tio2 with
Ozone: Sci. Eng. 28, 353e414. leds: micropollutants, antibiotic resistance genes and estrogenic activity. Water
€ger, T., Alexander, J., Kirchen, S., Do
Ja €tsch, A., Wieland, A., Hiller, C., Schwartz, T., Res. 94, 10e22.
2018. Live-dead discrimination analysis, Qpcr assessment for opportunistic Moura, A., Oliveira, C., Henriques, I., Smalla, K., Correia, A., 2012. Broad diversity of
pathogens, and population analysis at ozone wastewater treatment plants. conjugative plasmids in integron-carrying bacteria from wastewater environ-
Environ. Pollut. 232, 571e579. ments. FEMS Microbiol. Lett. 330, 157e164.
Karaolia, P., Michael-Kordatou, I., Hapeshi, E., Alexander, J., Schwartz, T., Fatta- Novo, A., Manaia, C.M., 2010. Factors influencing antibiotic resistance burden in
Kassinos, D., 2017. Investigation of the potential of a membrane bioreactor municipal wastewater treatment plants. Appl. Microbiol. Biotechnol. 87,
followed by solar fenton oxidation to remove antibiotic-related micro- 1157e1166.
contaminants. Chem. Eng. J. 310, 491e502. Oh, J., Medriano, C.A., Kim, S., 2016. The Effect of Tetracycline in the Antibiotic
Karaolia, P., Michael-Kordatou, I., Hapeshi, E., Drosou, C., Bertakis, Y., Christofilos, D., Resistance Gene Transfer before and after Ozone Disinfection. Desalination and
Armatas, G.S., Sygellou, L., Schwartz, T., Xekoukoulotakis, N.P., 2018. Removal of Water Treatment, 57, pp. 646e650.
antibiotics, antibiotic-resistant bacteria and their associated genes by Orhon, K.B., Orhon, A.K., Dilek, F.B., Yetis, U., 2017. Triclosan removal from surface
graphene-based Tio2 composite photocatalysts under solar radiation in urban water by ozonation-kinetics and by-products formation. J. Environ. Manag. 204,
wastewaters. Appl. Catal. B Environ. 224, 810e824. 327e336.
Karkman, A., Do, T.T., Walsh, F., Virta, M.P., 2017. Antibiotic-resistance genes in waste Ostoich, M., Serena, F., Falletti, L., Fantoni, A., 2013. Control of dangerous substances
water. Trends Microbiol. 26, 220e228. in discharges and microbiological abatement: european framework and a case
Kovalova, L., Siegrist, H., Von Gunten, U., Eugster, J., Hagenbuch, M., Wittmer, A., study of an ozone disinfection system. Water Sci. Technol. 67, 1238e1246.
Moser, R., Mcardell, C.S., 2013. Elimination of micropollutants during post- Pak, G., Salcedo, D.E., Lee, H., Oh, J., Maeng, S.K., Song, K.G., Hong, S.W., Kim, H.-C.,
treatment of hospital wastewater with powdered activated carbon, ozone, Chandran, K., Kim, S., 2016. Comparison of antibiotic resistance removal effi-
and uv. Environ. Sci. Technol. 47, 7899e7908. ciencies using ozone disinfection under different ph and suspended solids and
Lanao, M., Ormad, M., Ibarz, C., Miguel, N., Ovelleiro, J., 2008. Bactericidal effec- humic substance concentrations. Environ. Sci. Technol. 50, 7590e7600.
tiveness of O3, O3/H2o2 and O3/tio2 on Clostridium perfringens. Ozone: Sci. Pei, R., Kim, S.-C., Carlson, K.H., Pruden, A., 2006. Effect of river landscape on the
Eng. 30, 431e438. sediment concentrations of antibiotics and corresponding antibiotic resistance
Lee, Y., Gerrity, D., Lee, M., Bogeat, A.E., Salhi, E., Gamage, S., Trenholm, R.A., genes (Arg). Water Res. 40, 2427e2435.
Wert, E.C., Snyder, S.A., Von Gunten, U., 2013. Prediction of micropollutant Pocostales, J.P., Sein, M.M., Knolle, W., Von Sonntag, C., Schmidt, T.C., 2010. Degra-
elimination during ozonation of municipal wastewater effluents: use of kinetic dation of ozone-refractory organic phosphates in wastewater by ozone and
and water specific information. Environ. Sci. Technol. 47, 5872e5881. ozone/hydrogen peroxide (peroxone): the role of ozone consumption by dis-
Lee, Y., Kovalova, L., Mcardell, C.S., Von Gunten, U., 2014. Prediction of micro- solved organic matter. Environ. Sci. Technol. 44, 8248e8253.
pollutant elimination during ozonation of a hospital wastewater effluent. Water Prieto-Rodríguez, L., Oller, I., Klamerth, N., Agüera, A., Rodríguez, E., Malato, S., 2013.
Res. 64, 134e148. Application of solar Aops and ozonation for elimination of micropollutants in
Lee, Y., Von Gunten, U., 2016. Advances in predicting organic contaminant abate- municipal wastewater treatment plant effluents. Water Res. 47, 1521e1528.
ment during ozonation of municipal wastewater effluent: reaction kinetics, Pruden, A., 2013. Balancing Water Sustainability and Public Health Goals in the Face
transformation products, and changes of biological effects. Environ. Sci.: Water of Growing Concerns about Antibiotic Resistance. Acs Publications.
Res. Technol. 2, 421e442. Reungoat, J., Escher, B., Macova, M., Argaud, F., Gernjak, W., Keller, J., 2012. Ozon-
Lin, A.Y.-C., Lin, C.-F., Chiou, J.-M., Hong, P.A., 2009. O3 and O3/H2o2 treatment of ation and biological activated carbon filtration of wastewater treatment plant
sulfonamide and macrolide antibiotics in wastewater. J. Hazard Mater. 171, effluents. Water Res. 46, 863e872.
452e458. Reungoat, J., Macova, M., Escher, B., Carswell, S., Mueller, J., Keller, J., 2010. Removal
Liu, P., Zhang, H., Feng, Y., Yang, F., Zhang, J., 2014. Removal of trace antibiotics from of micropollutants and reduction of biological activity in a full scale reclamation
wastewater: a systematic study of nanofiltration combined with ozone-based plant using ozonation and activated carbon filtration. Water Res. 44, 625e637.
advanced oxidation processes. Chem. Eng. J. 240, 211e220. Ribeiro, A.R., Moreira, N.F., Puma, G.L., Silva, A.M., 2019. Impact of water matrix on
€ ffler, H., 2015. Removal of total
Lüddeke, F., Heß, S., Gallert, C., Winter, J., Güde, H., Lo the removal of micropollutants by advanced oxidation technologies. Chem. Eng.
and antibiotic resistant bacteria in advanced wastewater treatment by ozona- J. 363, 155e173.
tion in combination with different filtering techniques. Water Res. 69, 243e251. Ribeiro, A.R., Pedrosa, M., Moreira, N.F., Pereira, M.F., Silva, A.M., 2015. Environ-
Manaia, C.M., Rocha, J., Scaccia, N., Marano, R., Radu, E., Biancullo, F., Cerqueira, F., mental friendly method for urban wastewater monitoring of micropollutants
Fortunato, G., Iakovides, I.C., Zammit, I., 2018. Antibiotic resistance in waste- defined in the directive 2013/39/Eu and decision 2015/495/Eu. J. Chromatogr. A
water treatment plants: tackling the black box. Environ. Int. 115, 312e324. 1418, 140e149.
Marce, M., Domenjoud, B., Esplugas, S., Baig, S., 2016. Ozonation treatment of urban Rizzo, L., 2011. Bioassays as a tool for evaluating advanced oxidation processes in
primary and biotreated wastewaters: impacts and modeling. Chem. Eng. J. 283, water and wastewater treatment. Water Res. 45, 4311e4340.
768e777. Rizzo, L., Manaia, C., Merlin, C., Schwartz, T., Dagot, C., Ploy, M., Michael, I., Fatta-
Marce, M., Palacios, O., Bartolome , A., Caixach, J., Baig, S., Esplugas, S., 2017. Appli- Kassinos, D., 2013. Urban wastewater treatment plants as hotspots for antibiotic
cation of ozone on activated sludge: micropollutant removal and sludge quality. resistant bacteria and genes spread into the environment: a review. Sci. Total
Ozone: Sci. Eng. 39, 319e332. Environ. 447, 345e360.
Margot, J., Kienle, C., Magnet, A., Weil, M., Rossi, L., De Alencastro, L.F., Abegglen, C., Rodríguez-Chueca, J., Melero, M.P.O., Abad, R.M., Finol, J.E., Narvio n, J.L.O., 2015.
Thonney, D., Che vre, N., Sch€arer, M., 2013. Treatment of micropollutants in Inactivation of Escherichia coli in fresh water with advanced oxidation pro-
municipal wastewater: ozone or powdered activated carbon? Sci. Total Environ. cesses based on the combination of O 3, H 2 O 2, and tio 2. Kinetic modeling.
461, 480e498. Environ. Sci. Pollut. Res. 22, 10280e10290.
Mecha, A.C., Onyango, M.S., Ochieng, A., Momba, M.N., 2017. Ultraviolet and solar Rozas, O., Baeza, C., Nún ~ ez, K., Rossner, A., Urrutia, R., Mansilla, H.D., 2017. Organic
photocatalytic ozonation of municipal wastewater: catalyst reuse, Energy re- micropollutants (omps) oxidation by ozone: effect of activated carbon on
quirements and toxicity assessment. Chemosphere 186, 669e676. toxicity abatement. Sci. Total Environ. 590, 430e439.
Michael-Kordatou, I., Andreou, R., Iacovou, M., Frontistis, Z., Hapeshi, E., Michael, C., Salyers, A.A., Amabile-Cuevas, C.F., 1997. Why are antibiotic resistance genes so
resistant to elimination? Antimicrob. Agents Chemother. 41, 2321. Wayne, P., 2007. Clinical and Laboratory Standards Institute. Performance Standards
Sousa, J.M., Macedo, G., Pedrosa, M., Becerra-Castro, C., Castro-Silva, S., for Antimicrobial Susceptibility Testing,, vol. 17.
Pereira, M.F.R., Silva, A.M., Nunes, O.C., Manaia, C.M., 2017. Ozonation and uv254 Wellington, E.M., Boxall, A.B., Cross, P., Feil, E.J., Gaze, W.H., Hawkey, P.M., Johnson-
Nm radiation for the removal of microorganisms and antibiotic resistance genes Rollings, A.S., Jones, D.L., Lee, N.M., Otten, W., 2013. The role of the natural
from urban wastewater. J. Hazard Mater. 323, 434e441. environment in the emergence of antibiotic resistance in gram-negative bac-
Stalter, D., Magdeburg, A., Oehlmann, J., 2010. Comparative toxicity assessment of teria. Lancet Infect. Dis. 13, 155e165.
ozone and activated carbon treated sewage effluents using an in vivo test Zhang, X.-X., Zhang, T., Zhang, M., Fang, H.H., Cheng, S.-P., 2009. Characterization
battery. Water Res. 44, 2610e2620. and quantification of class 1 integrons and associated gene cassettes in sewage
Unep 2017. United Nation Environment (Unep), Frontiers, 2017. Emerging Issues of treatment plants. Appl. Microbiol. Biotechnol. 82, 1169e1177.
Environmental Concern. United Nations Environment Programme, Nairobi. Zhu, Y.-G., Johnson, T.A., Su, J.-Q., Qiao, M., Guo, G.-X., Stedtfeld, R.D.,
Valero, P., Mosteo, R., Ormad, M.P., L azaro, L., Ovelleiro, J.L., 2015. Inactivation of Hashsham, S.A., Tiedje, J.M., 2013. Diverse and abundant antibiotic resistance
Enterococcus sp. By conventional and advanced oxidation processes in syn- genes in Chinese swine farms. Proceedings Of The National Academy Of Sci-
thetic treated urban wastewater. Ozone: Sci. Eng. 37, 467e475. ences, p. 201222743.
Von Gunten, U., 2003. Ozonation of drinking water: Part I. Oxidation kinetics and Zhuang, Y., Ren, H., Geng, J., Zhang, Y., Zhang, Y., Ding, L., Xu, K., 2015. Inactivation of
product formation. Water Res. 37, 1443e1467. antibiotic resistance genes in municipal wastewater by chlorination, ultraviolet,
Von Gunten, U., 2018. Oxidation processes in water treatment: are we on track? and ozonation disinfection. Environ. Sci. Pollut. Res. 22, 7037e7044.
Environ. Sci. Technol. 52, 5062e5075. Zuma, F., Lin, J., Jonnalagadda, S.B., 2009. Ozone-initiated disinfection kinetics of
Von Sonntag, C., Von Gunten, U., 2012. Chemistry of Ozone in Water and Waste- Escherichia coli in water. J. Environ. Sci. Health A 44, 48e56.
water Treatment. Iwa Publishing.
View publication stats

Dr-HYDER 3

Uploaded by

Copyright:

Available Formats

Dr-HYDER 3

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Dr-HYDER 3

Uploaded by

Copyright:

Available Formats

See discussions, stats, and author profiles for this publication at: https://www.researchgate.

Continuous ozonation of urban wastewater: Removal of antibiotics,

Article in Water Research · May 2019

Iakovos C.Iakovides Ana Rita Lado Ribeiro

SEE PROFILE SEE PROFILE

NanoBiofilms View project

The user has requested enhancement of the downloaded file.

Contents lists available at ScienceDirect

Continuous ozonation of urban wastewater: Removal of antibiotics,

1. Introduction radicals (HO) under alkaline pH conditions, which result from

Ampicillin (AMP) 0.97 <MDL

MDL ¼ Method Detection Limit.

Target Primers Primers sequence Conditions Primers reference

of total heterotrophs, enterococci and coliforms (including of

View publication stats

You might also like