Water Research 126 (2017) 79e87

Contents lists available at ScienceDirect

Water Research
journal homepage: www.elsevier.com/locate/watres

Negative environmental impacts of antibiotic-contaminated effluents

from pharmaceutical industries


Ana Bielen a, 1, Ana Simatovi 
c b, 1, Josipa Kosi
c-Vuk
si
c c, Ivan Senta b, Marijan Ahel b,
Sanja Babi
c b, Tamara Jurina a, Juan Jose  Gonzalez Plaza b, Milena Milakovi c b,
*
Nikolina Udikovi c-Koli
c b ,

a
Faculty of Food Technology and Biotechnology, University of Zagreb, Pierottijeva 6, Zagreb, Croatia
b
RuCer Bo
skovic Institute, Bijeni
cka 54, Zagreb, Croatia
c
Andrija

Stampar Teaching Institute of Public Health, Mirogojska cesta 16, Zagreb, Croatia

a r t i c l e i n f o a b s t r a c t

Article history: Effluents from pharmaceutical industries are recognized as significant contributors to aquatic pollution
Received 21 July 2017 with antibiotics. Although such pollution has been mostly reported in Asia, knowledge on industrial
Received in revised form discharges in other regions of the world, including Europe, and on the effects associated with such ex-
7 September 2017
posures is still limited. Thus, we performed chemical, microbiological and ecotoxicological analyses of
Accepted 8 September 2017
Available online 9 September 2017
effluents from two Croatian pharmaceutical industries during four seasons. In treated effluents of the
company synthesizing macrolide antibiotic azithromycin (AZI), the total concentration of AZI and two
macrolide by-products from its synthesis was 1e3 orders of magnitude higher in winter and springtime
Keywords:
Antibiotic pollution
(up to 10.5 mg/L) than during the other two seasons (up to 638 mg/L). Accordingly, the highest total
Drug manufacturing concentrations (up to 30 mg/L) in the recipient river were measured in winter and spring. Effluents from
Macrolides second company formulating veterinary antibiotics contained fluoroquinolones, trimethoprim, sulfon-
Antibiotic resistance amides and tetracyclines ranging from low mg/L to approx. 200 mg/L. Low concentrations of these anti-
Ecotoxicity biotics, from below the limit of quantification to approx. few mg/L, have also been measured in the
recipient stream. High frequency of culturable bacteria resistant to AZI (up to 83%) or sulfamethazine (up
to 90%) and oxytetracycline (up to 50%) were also found in studied effluents. Finally, we demonstrated
that toxicity to algae and water fleas often exceeded the permitted values. Most highly contaminated
effluents induced multiple abnormalities in zebrafish embryos. In conclusion, using a wide array of
analyses we have demonstrated that discharges from pharmaceutical industries can pose a significant
ecological and public health concern due to their toxicity to aquatic organisms and risks for promoting
development and spread of antibiotic resistance.
© 2017 Elsevier Ltd. All rights reserved.

1. Introduction production facilities (Andersson and Hughes, 2014). A significant

amount of research in the area has focused on municipal effluents
Pharmaceuticals are considered emerging pollutants since they and data from municipal wastewater plants are relatively acces-
are continuously being released into the environment and affect sible. Complex cocktail of various pharmaceuticals is assumed to
the biota on multiple levels. Consequently, global concern about enter municipal sewage by human excretion and their concen-
their potential adverse impacts on both human health and the trations are generally in ng/L in treated sewage effluents (Kostich
environment is growing. The primary sources of pharmaceuticals et al., 2014; Luo et al., 2014). In contrast to this, predominant
in the environment are municipal wastewater treatment plants, compounds in pharmaceutical effluents depend on the production
agricultural settings, aquaculture, hospitals and pharmaceutical site and research on such effluents is more problematic due to
difficulties in accessing information. However, concentrations of
pharmaceuticals in industrial effluents can be several orders of
* Corresponding author.
E-mail address: [email protected] (N. Udikovi
c-Koli
c).
1
Equal contribution.

http://dx.doi.org/10.1016/j.watres.2017.09.019
0043-1354/© 2017 Elsevier Ltd. All rights reserved.
80 A. Bielen et al. / Water Research 126 (2017) 79e87

magnitude greater than the concentrations found in sewage ef- and vertebrates (zebrafish embryos).
fluents as reported recently (Cardoso et al., 2014; Larsson, 2014).
Besides pharmaceuticals, industrial effluents can also contain 2. Materials and methods
various intermediate products and by-products such as solvents,
acids and inorganic compounds (including heavy metals, nitrate 2.1. Sampling of the effluents
and phosphate). Moreover, these effluents can have high level of
chemical and biochemical oxygen demand and total suspended Effluents from two pharmaceutical industries situated in
solids (COD, BOD and TSS, respectively) (Gadipelly et al., 2014). northern Croatia were collected (Company 1 and Company 2) four
However, in contrast to some other physico-chemical parameters, times a year over one year monitoring period.
the concentration of pharmaceuticals and their metabolites is Company 1 has a long tradition in manufacturing the macrolide
currently not legally regulated, although some of them, such as antibiotic azithromycin via synthesis from another macrolide
macrolide antibiotics, have been included in watch lists of the antibiotic, erythromycin. The final effluent from this industry (after
regulation bodies (Cardoso et al., 2014; Decision, 2015; Rivera- treatment in industry's membrane bioreactor) is discharged into
Utrilla et al., 2013). the Sava river together with effluent from baker's yeast production.
Among pharmaceuticals, special attention is given to antibiotics Such mixed effluents were collected as grab samples at the
because environmental contamination with antibiotics facilitates discharge pipe located at the bank of the Sava river in winter
the development and spread of antibiotic resistance among bac- (February), spring (May), summer (August) and autumn
teria, including pathogens (Andersson and Hughes, 2014). High (November) 2016. In addition to effluent samples, surface river
emissions of antibiotics from manufacturing plants have been re- water was collected as grab sample from three locations along the
ported mostly in Asian countries, while much less information is river: at 3.5 km upstream and at 700 m and 4.5 km downstream of
available for other regions of the world, including Europe (Larsson, the discharge site.
2014). In addition, metals and other biocides present in industrial Company 2 is involved in manufacturing of various human and
effluents have the potential to co-select for antibiotic resistance animal health products, but is mostly known for formulation of
(Baquero et al., 2008; Roose-Amsaleg and Laverman, 2016). Besides veterinary antibiotics (sulfonamides, fluoroquinolones, tetracy-
risks for promoting resistance, antibiotic pollution can induce clines and b-lactams) into doses. The technological effluent pro-
direct biostatic or bio-toxic effects to diverse organisms and alter duced by this industry is discharged into the nearby small stream
species distribution, which is harmful for the ecosystem as a whole after being diluted with industrial sewage and passed through a
(Halling-Sørensen, 2000; Harada et al., 2008; Isidori et al., 2005; primary treatment (mechanical removing of larger, floating solids).
Wang et al., 2014a). The 24-h flow proportional samples were collected directly with an
Noteworthy, most studies, such as the ones mentioned above, automatic sampler that is connected to a compatible flow
predict the environmental effects of single chemicals. Such measuring device. Sampling was performed in winter (January),
research is certainly useful, but does not provide much information spring (April), summer (July) and autumn (October) 2016. Also,
about toxic effects of effluents as mixtures of pharmaceuticals that stream samples were collected as grab samples from three loca-
can interact synergistically, additively or antagonistically (Białk- tions along the stream: at 300 m upstream and at 2 m and 3 km
 ska et al., 2013; Caldwell et al., 2016; Flaherty and Dodson,
Bielin downstream of the discharge site.
2005; Spurgeon et al., 2010). Therefore, from the environmental Effluent and surface water samples were collected in sterile
risk assessment perspective, the whole effluent testing is of polyethylene wide mouth jars and stored at þ4  C or 20  C,
importance to account for unknown substances present in effluents depending on analysis performed. Parts of samples were sterile-
and combined cocktail' effects. Existing studies investigating the filtered through a 0.22 mm filter (Millipore, USA).
effects of pharmaceutical effluents on aquatic environment mostly
focus on only one or few endpoints, reporting for instance only 2.2. Chemical analyses
chemical analyses of the pharmaceuticals in the wastewaters (Fick
et al., 2009), only analyses of antimicrobial resistance (Guardabassi The samples were analyzed for a number of basic parameters
et al., 1998) or only toxicological data (Carlsson et al., 2009; Ti
sler within their holding time using internationally validated methods
and Zagorc-Kon
can, 1999). The work of Larsson et al. (2007) is (ISO standards) for examination of water and wastewater (for more
the most detailed to date, including both chemical analyses of details see Table S1).
pharmaceuticals in the effluent and toxicity bioassays using bac- Antibiotics (azithromycin, AZI; dehydrated erythromycin, ERY-
teria (Vibro fisheri), invertebrates (Daphnia magna) and plants H2O; enrofloxacin, ENR; oxytetracycline, OTC; sulfadiazine, SDZ;
(Lactuca sativa) as models. sulfamethazine, SMZ; trimethoprim, TMP; Sigma, Germany) and
Thus, there is a clear need for more research in these areas to intermediate from AZI production (N-Desmethyl Azithromycin, N-
better understand the exposure routes and the effects pharma- DMA) were quantified using the previously described method
ceutical pollution might have on human and ecological health. In based on liquid chromatography - tandem mass spectrometry
order to take a step toward this understanding, we have sampled (Senta et al., 2017, 2008). Briefly, samples were enriched using
final (treated or partially treated) effluents from two Croatian solid-phase extraction on Oasis HLB cartridges. Typical volumes
pharmaceutical industries that are involved in synthesis and were between 1 and 250 mL, depending on the sample type. It
formulation of antibiotics during four seasons in one year moni- should be noted that erythromycin (ERY) was not analyzed in the
toring period. We have analyzed the physico-chemical properties original form but as its dehydrated transformation product ERY-
and chemical composition of effluent samples, including quantifi- H2O.
cation of antibiotics and heavy metals. We also measured antibiotic Concentrations of heavy metals in effluent samples, arsenic (As),
concentrations in the recipient water bodies. Finally, we have cadmium (Cd), chromium (Cr), copper (Cu), nickel (Ni), lead (Pb)
assessed the frequency of culturable antibiotic-resistant bacteria in and zinc (Zn), were determined by inductively coupled plasma
effluents and the effect of effluents on various levels of biological mass spectrometry (ICP-MS) following the methodology described
organization: bacteria, freshwater algae, invertebrates (Daphnia) by Dautovi c et al. (2014).
 A. Bielen et al. / Water Research 126 (2017) 79e87 81

2.3. Bioassays 3. Results

2.3.1. Determination of the frequency of culturable antibiotic- 3.1. Physicochemical properties and chemical composition of
resistant bacteria pharmaceutical effluents, including heavy metal and antibiotic
Serial dilutions of effluents were prepared in physiological sa- content
line (0.9% NaCl) and plated on R2A agar (R2A broth [Lab M Limited,
UK] amended with 1.5% agar [BioLife Italiana, Italy]) supplemented Effluent quality parameters are given in Table 1, Table 2 and
with yeast extract (5 g/L) (BioLife Italiana, Italy) to enumerate total Table S2. Effluents from Company 1 had a high level of organic
bacteria. In parallel, serial dilutions of effluents were plated on R2A pollution, especially in winter and spring, as COD, BOD5 and TOC
agar with yeast extract and antibiotic AZI (15 mg/L) (Fluka, Ger- values were far higher than permitted values (Table 1). Concen-
many) for the analysis of Company 1 effluents, or antibiotic SMZ trations of nitrites were higher than maximal permitted value only
(350 mg/L) (Sigma, Germany) and OTC (25 mg/L) (Sigma, Germany) in the winter sampling (1.5 mg/L), while nitrates were elevated in
for the analysis of Company 2 effluents, to enumerate resistant summer sample. Total phosphorus concentrations were above the
bacteria. The culture medium was supplemented with cyclohexi- limit for winter, spring and summer effluents from Company 1.
mide (100 mg/L final concentration; Sigma, Germany) to inhibit Company 2 effluents had satisfactory quality regarding organic
growth of fungi. The number of colony forming units (CFU) was and nutrient pollution; only BOD5 value in the summer sample was
determined after incubation at 28  C for 5 days. The frequency of slightly above the limit (Table 2). However, BOD5/COD ratios were
antibiotic-resistant bacteria was calculated as the ratio of resistant lower than 0.4 for effluents of both Companies indicating low
bacteria CFU and total CFU. biodegradability (Chamarro et al., 2001).
The measured values of temperature, pH and sulfates of efflu-
2.3.2. Freshwater algal growth inhibition test ents of both companies were in compliance with the permissible
Effect of effluents on the growth of planktonic unicellular limits for wastewaters discharge in surface waters set by Croatian
freshwater green microalgae Pseudokirchneriella subcapitata was legislative (Official Gazette of the Republic of Croatia 80/13, 43/14,
measured using the standardized ISO 8692:2012 method (ISO, 27/15, 3/16; Table S2). Conductivity of effluents is not regulated by
2012). The algae were exposed to effluent dilution series, as well legislation but it can serve as a useful measure of dissolved inor-
as appropriate positive and negative control, and incubated for ganic materials. All effluents from Company 1 and Company 2 had
72 h. The toxicity of effluents was expressed as the lowest inef- high conductivity, 2e10 times higher than the values reported in
fective dilution (LID) at which growth inhibition of <5% was the literature for pharmaceutical effluents (Idris et al., 2013; James
observed. et al., 2014).
The results on heavy metal content in analyzed effluents are
2.3.3. Acute Daphnia magna toxicity test given in Table 3. Concentrations of Cr, Cu, Pb, Ni and Zn are regu-
Effect of effluents on motility of water flea Daphnia magna Straus lated by Croatian law and their concentrations in the effluents were
(Cladocera, Crustacea) was determined using the standardized ISO lower or approx. the same as maximal permitted values (Official
6341:2013 method (ISO, 2013). After 24 h exposure of water fleas to Gazette of the Republic of Croatia 80/13, 43/14, 27/15, 3/16). Only
effluent dilution series, as well as appropriate positive and negative concentration of zinc in the autumn sample from Company 2 was
control, the toxicity of effluents was expressed as LID for which four times higher than maximal permitted value.
10% of the fleas were immobile. In winter and spring, effluents from Company 1 had the highest
total concentration of AZI and two macrolide by-products from its
P
2.3.4. Mutagenicity to Salmonella synthesis, N-DMA and ERYeH2O ( ¼ 6.4 and 10.5 mg/L, respec-
Mutagenicity was tested on two most contaminated effluent tively) (Table 4). These compounds were also present in summer
samples from Company 1 (winter and spring samples) and on two and autumn effluents but in 1e3 orders of magnitude lower total
effluent samples from Company 2 that showed toxicity to algae concentration (up to 638 mg/L) than in winter and spring samples.
(summer and autumn samples). The test was performed in tripli- Surface water samples from the Sava river taken upstream from the
cates (with or without S9 fraction, Sigma, Germany) using the discharge mostly contained ng/L levels of the analyzed macrolides,
standard Ames bacterial test (Maron and Ames, 1983; Sauerborn which are much lower than in downstream locations. The con-
Klobu
car et al., 2013) with Salmonella typhimurium, strains TA98 centrations of macrolides in downstream river generally decreased
(his D3052, rfa, uvrB, pKM101) and TA100 (his G46, rfa, uvrB, as the compounds were carried downstream, with an exception in
pKM101). Revertants were scored after 48 h incubation at 37  C. autumn when higher concentration of macrolides was observed in
Distilled water was used as negative solvent control and positive the river at a greater distance from the discharge.
controls were 9-aminoacridine for TA98 strain without metabolic In effluents of Company 2 (Table 5), the concentrations of sul-
activation (50 mg per plate), sodium azide for TA100 strain without fonamides were measured in the range from 6.7 to 231.0 mg/L for
metabolic activation (1 mg per plate), and 2-aminoantracene for SMZ, from 3.0 to 20.0 mg/L for SDZ, from 7.4 to 29.0 mg/L for OTC,
TA98 and TA100 strains with metabolic activation (2 mg per plate). from 3.6 to 98 mg/L for ENR and at approx. few mg/L for TMP. Con-
centrations of these antibiotics did not generally exceed 10 mg/L in
2.3.5. Zebrafish embriotoxicity assay the receiving stream near the discharge and were further decreased
Embriotoxicity assay on zebrafish Danio rerio (wild type) was at 3 km downstream location ranging from below the limit of
performed on the same selected samples as mutagenicity according quantification to low mg/L. Nevertheless, these levels were gener-
to Babic et al. (2017). Each sample was tested in triplicates per 10 ally 4e34  higher than those detected at upstream location.
embryos (amounting to a total of 30 embryos per sample/dilution)
e undiluted, 2  and 4  diluted with artificial water (ISO, 1996). 3.2. Antibiotic-resistant bacteria as biological pollutants in studied
Toxicological endpoints, both lethal and sublethal, were recorded at effluents
24, 48 and 72 h post fertilization (hpf) (ISO, 2007). The mortality in
all negative control groups was less than 5%. Additionally, physio- Our culturing data showed that studied effluents contained high
logical endpoints, i.e. heart rate, pigmentation formation and numbers of total culturable bacteria (in the range of 105e107 CFU/
hatching rate, were evaluated at 48 and 72 hpf. mL) and high frequency of antibiotic-resistant bacteria (Fig. 1).
82 A. Bielen et al. / Water Research 126 (2017) 79e87

Table 1
Quality parameters of Company 1 effluents discharged into the Sava river and their permissible limits by Croatian legislative. Recorded values higher then permitted are
marked in bold.

Parameters Sampling campaigns Maximal permitted valuea

Winter Spring Summer Autumn

b
COD 1915 1166 398 596 125
mg O2/L
BOD5c 564 352 28 119 20
mg O2/L
BOD5/COD 0.29 0.30 0.07 0.20 e
ratio
TOCd 616.2 177.6 86.2 134.9 30
mg C/L
Total phosphorus mg P/L 6.77 2.5 4.41 4.2 1.5
Nitrite 1.5 <0.01 <0.01 <0.01 1
mg N/L
Nitrate 1.4 0.5 3.8 0.2 2
mg N/L
a
Maximal permissible values for pharmaceutical effluents discharged into surface waters (Official Gazette of the Republic of Croatia 80/13, 43/14, 27/15, 3/16).
b
Chemical oxygen demand.
c
Biochemical oxygen demand.
d
Total organic carbon.

Table 2
Quality parameters of Company 2 effluents discharged into nearby stream and their permissible limits by Croatian legislative. Recorded values higher then permitted are
marked in bold.
a
Parameters Sampling campaigns Maximal permitted value

Winter Spring Summer Autumn

CODb 37.0 54.0 99.0 27.0 125

mg O2/L
BOD5c 7.3 13.0 33.0 7.0 20
mg O2/L
BOD5/COD 0.20 0.24 0.33 0.26 e
ratio
TOCd 7.9 5.4 11.7 4.84 30
mg C/L
Total phosphorus mg P/L 0.62 0.42 0.008 0.17 1.5
Nitrites 0.07 0.09 <0.01 0.06 1
mg N/L
Nitrates 0.84 1.2 <0.05 0.9 2
mg N/L
a
Maximal permissible values for pharmaceutical effluents discharged into surface waters (Official Gazette of the Republic of Croatia 80/13, 43/14, 27/15, 3/16).
b
Chemical oxygen demand.
c
Biochemical oxygen demand.
d
Total organic carbon.

Table 3
The heavy metal ions concentration in the effluents from two Croatian pharmaceutical companies, Company 1 (a) and Company 2 (b), during four sampling campaigns and
their permissible limits by Croatian legislative. Recorded values higher then permitted are marked in bold.

Heavy metal ions (mg/L) Winter Spring Summer Autumn Maximal permitted valuea

a)
Arsenic (As) 0.73 0.59 0.41 0.67 e
Cadmium (Cd) 0.28 0.79 0.07 0.71 e
Chromium (Cr) 7.23 2.62 15.73 1.98 50
Copper (Cu) 26.92 6.19 18.36 10.36 100
Lead (Pb) 4.24 3.73 2.70 3.30 500
Nickel (Ni) 20.66 13.62 12.42 24.77 50
Zinc (Zn) 90.94 103.06 56.35 70.42 100
b)
Arsenic (As) 0.53 0.46 0.79 50.27 e
Cadmium (Cd) 0.02 0.02 0.05 0.12 e
Chromium (Cr) 0.51 0.16 1.13 0.38 50
Copper (Cu) 7.46 12.75 12.70 15.42 100
Lead (Pb) 2.33 1.55 9.3 1.97 500
Nickel (Ni) 1.25 0.81 2.12 1.31 50
Zinc (Zn) 33.88 28.72 47.64 417.67 100
a
Maximal permitted value for pharmaceutical effluents discharged into surface waters (Official Gazette of the Republic of Croatia 80/13, 43/14, 27/15, 3/16).
 A. Bielen et al. / Water Research 126 (2017) 79e87 83

Table 4
Concentration of three macrolide compounds, expressed in mg/L, in effluents of Company 1 and in the recipient Sava river.

Antibiotic Sampling campaign Effluent concentration Sava Upstream 3.5 km Sava Downstream 700 m Sava Downstream 4.5 km

Azithromycin (AZI) Winter 2137.00 0.13 3.40 0.39

Spring 3776.00 0.01 3.40 1.60
Summer 4.90 0.02 0.03 0.02
Autumn 218.00 0.06 0.01 2.80
N-Desmethyl Azithromycin (N-DMA) Winter 2341.00 0.37 13.00 1.50
Spring 5660.00 0.06 8.60 6.20
Summer 11.10 < LOD < LOD 0.02
Autumn 390.00 < LOD < LOD 0.16
Erythromycin-H2O (ERY-H2O) Winter 2009.00 3.40 4.20 22.00
Spring 1069.00 0.13 18.00 10.00
Summer 1.00 < LOD < LOD < LOD
Autumn 30.0 < LOD < LOD 0.02
P
Macrolides total, Winter 6487.00 3.90 20.60 23.89
Spring 10 505.10 0.20 30.00 17.80
Summer 17.00 0.02 0.03 0.04
Autumn 638.00 0.06 0.01 2.98

< LOD e below detection limit.

Table 5
Concentration of five antibiotics, expressed in mg/L, in effluents of Company 2 and in the recipient stream.

Antibiotic Sampling campaign Effluent concentration Stream Upstream 300 m Stream Downstream 2 m Stream Downstream 3 km

Sulfadiazine (SDZ) Winter 20.00 0.07 6.60 2.40

Spring 8.12 0.07 2.30 0.29
Summer 7.10 0.04 < LOD 0.78
Autumn 3.00 0.06 < LOD 0.70
Sulfamethazine (SMZ) Winter 10.00 0.10 5.30 1.10
Spring 44.70 0.79 12.00 1.70
Summer 231.00 1.00 2.80 11.00
Autumn 6.70 0.20 0.70 1.70
Trimethoprim (TMP) Winter 5.60 < LOD 0.11 1.10
Spring 1.57 < LOD 0.37 0.24
Summer 5.40 < LOD 0.11 0.42
Autumn 1.00 0.01 < LOD 0.12
Enrofloxacin (ENR) Winter 98.00 < LOD 10.60 1.00
Spring 6.66 < LOD 1.00 < LOD
Summer 4.30 < LOD 0.40 0.21
Autumn 3.60 < LOD 0.30 < LOD
Oxytetracycline (OTC) Winter 17.00 < LOD 10.00 < LOD
Spring 7.40 < LOD < LOD < LOD
Summer 29.00 < LOD 8.00 < LOD
Autumn 10.00 < LOD 1.00 0.43

< LOD e below detection limit.

Fig. 1. The frequency of culturable antibiotic-resistant bacteria present in pharmaceutical effluents. The percentage of antibiotic-resistant bacteria was calculated as the ratio of
resistant bacteria CFU and total CFU. The mean of calculated resistance frequencies is represented as column and the error bar indicates standard deviation (n ¼ 3). AZI - azi-
thromycin; SMZ - sulfamethazine; OTC e oxytetracycline.
84 A. Bielen et al. / Water Research 126 (2017) 79e87

Effluents of Company 1 contained up to 82.5% of AZI-resistant including heart and yolk edema (Fig. S2; f), scoliosis (Fig. S2; g, h, i)
bacteria, whereas Company 2 effluents contained up to 91.0% of and lack of pigmentation formation (Fig. S2; e). During exposure to
SMZ-resistant and up to 49.7% of OTC-resistant bacteria. Company 1 spring sample, embryos often exhibited delayed
development (e.g. reaching the 24 h stage at 72 hpf), yolk deformity
3.3. Effects on biota and pericardial edemas (Fig. S2; a, b, d, i). Observed developmental
delay was in correlation with reduced hatching rate, which was
We have tested the effects of our pharmaceutical effluents on most pronounced after exposure of embryos to 2  diluted spring
various levels of biological organization: (i) bacteria (potential sample. Among other explored endpoints, embryo pigmentation
mutagenic effect on Salmonella typhimurium tester strains), (ii) was significantly reduced upon exposure to 2 and 4  diluted
freshwater algae (growth inhibition of green alga Pseudokirchner- winter sample at 48 hpf (Table S3). Further, heart beat rate was also
iella subcapitata), (iii) invertebrates (induction of immobility in significantly reduced upon exposure to dilutions of winter sample
water flea Daphnia magna), and (iv) vertebrates (lethal and sub- at 48 and 72 hpf. In contrast to diluted winter sample, 2 and 4 
lethal toxicity to fish Danio rerio embryos). diluted spring sample significantly increased heart beat frequency
The Ames Salmonella assay (Fig. S1) showed no mutagenic effect at 72 hpf.
after exposure of TA98 and TA100 tester strains to Company 1 ef- The only observed abnormality during exposure to Company 2
fluents. However, incubation with Company 2 effluents caused summer and autumn effluents was pericardial and yolk sack
slight increase in the number of S. typhimurium TA98 revertants, edema. Additionally, those samples did not affect pigmentation
especially pronounced after metabolic activation (up to 1.3 fold formation, but significantly increased heart beat rate at 48 and 72
increase in comparison with negative control), indicating the hpf. Undiluted autumn sample caused the largest hatching rate
occurrence of frameshift mutations. reduction at 72 hpf (Table S3), which was also revealed through
Toxic effects of the effluents on P. subcapitata and D. magna are constant twitching of embryo body.
reported in Table 6. Company 1 effluents (especially winter and
spring samples) were very toxic to algae as LID values of all samples 4. Discussion
were higher than permitted. When comparing acute toxicity to
Daphnia, only winter and spring samples were toxic, with winter The present study evaluated chemical composition and eco-
sample being more toxic than the spring one (LID values 16 and 5, toxicity of final effluents from two antibiotic-manufacturing plants
respectively). Only summer and autumn effluent samples from in Croatia as well as antibiotic contamination of receiving surface
Company 2 were toxic to green algae (LID values 4 and 10, waters.
respectively). Our chemical analyses of effluents from Croatian Company 1
Embriotoxicity assay on zebrafish Danio rerio was performed on revealed occasionally (winter and spring season) high concentra-
winter and spring effluent samples from Company 1 and summer tions of AZI and two other macrolides associated with AZI pro-
and autumn samples from Company 2. All tested samples caused duction (N-DMA and ERY-H2O), in the order of few mg/L. It is
adverse effects on zebrafish embryos, which were manifested important to stress that both identified macrolide by-products have
through lethal and sub-lethal alterations (Fig. 2, Fig. S2, Table S3). antimicrobial activity (Shepard et al., 1990; Wang et al., 2014b). The
Most strikingly, all embryos exposed to undiluted effluents of observed mg/L-levels of AZI in Company 1 effluents are three or-
Company 1 were coagulated at 24 hpf (see mortality rate in Fig. 2). ders of magnitude greater than the levels generally found in treated
In comparison, undiluted effluents of Company 2 induced much hospital and sewage effluents (around or below 1 mg/L) (Michael
lower mortality rate (<10%). After we have diluted the samples we et al., 2013; Verlicchi et al., 2012). These findings together with
still observed some mortality after exposure of the embryos to high levels of organic, phosphorus and nitrate compounds detected
Company 1 effluents. In overall, mortality was more pronounced in the effluents are indicative of unsatisfactory wastewater treat-
after exposure to winter than to spring sample. Even when the ment of Company 1 and of unauthorized discharges. They also
samples were diluted 4 , effluents of Company 1 still caused up to indicate that these effluents are sources of macrolide pollution
50% mortality at 72 hpf, while no mortality was observed during observed in the recipient river downstream from the discharge
exposure to 2 or 4  diluted Company 2 effluents. point. Indeed, during the winter and springtime, all studied mac-
To get a more detailed insight into the toxic effects of effluents to rolides were found in the river at enhanced concentrations (up to
zebrafish embryos, type of observed abnormalities (Fig. S2) and approx. 30 mg/L total). The observation of a declining gradient of
some physiological endpoints (heart beat rate, pigmentation for- macrolide concentration from the discharge site might be due to
mation, hatching rate, Table S3) were followed upon exposure. further dilution of macrolides with the river water and/or some
Type of abnormalities depended on the sample. Exposure to elimination processes such as sorption and degradation. However,
Company 1 winter sample induced multiple malformations reverse situation observed occasionally indicates that the industrial

Table 6
Toxicity of effluents to freshwater organism. Toxicity to algae is expressed as the lowest ineffective dilution (LID) at which growth inhibition of <5% was observed. Toxicity to
Daphnia is expressed as the LID for which 10% of the fleas were immobile. Recorded values higher then permitteda are marked in bold.

Sampling campaign Alga Pseudokirchneriella subcapitata Water flea Daphnia magna

Company 1 Winter >32 16

Spring >32 5
Summer 12 2
Autumn 4 2
Company 2 Winter 1 1
Spring 3 1
Summer 4 2
Autumn 10 2
Maximal permissible valuea 3 2
a
Maximal permissible value for pharmaceutical effluents discharged into surface water (Official Gazette of the Republic of Croatia 80/13, 43/14, 27/15, 3/16).
 A. Bielen et al. / Water Research 126 (2017) 79e87 85

Fig. 2. Mortality of D. rerio embryos exposed to effluent samples, expressed as mean ± SD (n ¼ 30).

effluents are not the only source of macrolide pollution of studied resistance in the recipient Sava river. Indeed, AZI concentrations in
river. Discharge from the wastewater treatment plant of the City of the river at two downstream locations measured during winter and
Zapre
sic located between two studied downstream locations may springtime exceeded the corresponding PNEC values, suggesting
be responsible for this. the risk for selecting macrolide resistance in the environment.
In contrast to mg/L concentrations of macrolides in effluents Although much lower antibiotic concentrations were detected in
from AZI production, the concentrations of sulfonamides, effluents of antibiotic-formulating Company 2 and in the receiving
trimethoprim, enrofloxacin and oxytetracycline antibiotics in ef- stream near discharge, they were still above the corresponding
fluents from Company 2 were much lower and ranging from low PNECs and thus predicted to be selective for resistance (Bengtsson-
mg/L to approx. 200 mg/L. Even though, peak values were approx. Palme and Larsson, 2016). However, antibiotic concentrations
one order of magnitude higher than the levels observed in the measured in the stream 3 km downstream of discharge were
receiving studied stream and the levels that are regularly found in mostly lower than PNEC values.
treated sewage effluents (Michael et al., 2013). Noteworthy, con- Additionally, metals detected in effluents have the potential to
centrations of the same antibiotic in Company 2 effluents varied co-select for antibiotic resistance (Baker-Austin et al., 2006; Pal
significantly, sometimes even by approx. 30 times as observed for et al., 2015). Despite the fact that the concentrations of most eco-
sulfamethazine (between 7 and 231 mg/L). Such fluctuations in toxic heavy metals (As, Cd, Cr, Cu, Pb, Ni) were lower than the le-
antibiotic concentrations were observed for effluents of both gally set upper limit, they were still up to 85 times higher than
Companies and could be due to multiple reasons, including minimum concentrations that were reported to have co-selective
different production cycles and washing of reaction/formulating potential in surface waters (Seiler and Berendonk, 2012;
tanks (Larsson, 2014; Pruden et al., 2013). Stepanauskas et al., 2006). This suggests that co-selection could
The high public concern associated with release of antibiotics be at play, at least in effluents. However, in current lack of data on
from pharmaceutical production is promotion of antibiotic resis- metal concentrations in the receiving water bodies, we do not
tance, an event not restricted to the area or even country of know if metal concentrations in the receiving waters are still
emergence but a potential threat to human health worldwide. selective.
Indeed, enrichment of resistant bacteria or their resistance genes Effluents from both industries were found to contain high fre-
has already been shown in aquatic environment polluted with quency of antibiotic-resistant bacteria i.e. up to 91% of SMZ-
antibiotics from pharmaceutical manufacturing (Kristiansson et al., resistant bacteria and 50% of OTC-resistant bacteria in effluents
2011; Li et al., 2011, 2009, 2010). In addition to this, antibiotic from Company 2 and up to 83% of AZI-resistant bacteria in effluents
resistance can be promoted even at antibiotic levels far below the from Company 1. The frequencies of AZI-resistant bacteria in
inhibitory concentrations, similar to those found in some aquatic Company 1 effluents and OTC-resistant bacteria in Company 2 ef-
environments (Andersson and Hughes, 2014; Berendonk et al., fluents were found to be positively associated with the occurrence
2015; Gullberg et al., 2011). Despite this, there are still no regula- of AZI and OTC suggesting that increased antibiotic concentrations
tion regarding surface water levels for any antibiotic, although AZI in these effluents created a breeding ground for proliferation of
and ERY were recently added to the so-calledwatch list' within the resistant bacteria. In addition, the fact that effluents from Company
European Water Framework Directive due to their toxicity, persis- 2 are mixed with human sewage within the industry implies a risk
tence and bioaccumulation potential [Decision, 2015/495/EU of 20 that pathogens are exposed to antibiotics at concentrations
March 2015 (Barbosa et al., 2016)]. In the current lack of regulatory selecting for resistance. Therefore, discharge of such poorly treated
system, Bengtsson-Palme and Larsson (2016) have recently esti- effluents into the receiving stream offers a potential route for
mated Predicted No Effect Concentrations (PNECs) for resistance dissemination of resistant bacteria, including human pathogens.
selection for 111 antibiotics. AZI concentration in our effluents can This suggests that there is a need for better management of dis-
be 4 orders of magnitude higher then estimated PNEC (250 ng/L), charges from antibiotic manufacturing which should involve limits
suggesting that it might be sufficient to specifically promote on both antibiotic-resistant bacteria and antibiotic levels.
86 A. Bielen et al. / Water Research 126 (2017) 79e87

The direct effects of effluent exposure on biota was assessed and spring suggest that discharge of studied effluents can pose a
using eukaryotic algae, crustacean invertebrates and zebrafish great risk for environmental and human health. Studied effluents
embryo models. In overall, majority of effluent samples induced can be seen as a reservoirs from which resistance traits, in the
some level of toxic effects, but Company 1 effluents proved to be presence of residual antibiotics and other co-selective agents as
more toxic to all tested organisms. For example, clearly detectable well as nutrients, may be further spread, with the potential to ul-
effects on D. magna motility were found after exposure to Company timately reach human pathogens.
1 winter effluents diluted up to 16 . On the other hand, Larsson Most importantly, this study emphasizes the environmental risk
et al. (2007) demonstrated the same effects after exposure to of macrolides that are recognized as antibiotics with high toxicity,
higher dilutions of Patancheru industrial effluents contaminated persistence and bioaccumulation potential (Decision, 2015/495/EU
with fluoroquinolones, suggesting even higher toxicity of Patan- of 20 March 2015). We report high levels of macrolides in final,
cheru effluents on D. magna in comparison with our macrolide- treated effluents from azithromycin-production during winter and
contaminated effluents. Similarly, lower dilutions of our Company spring season. Even after dilution of the effluent with the receiving
1 winter and spring effluents (2 and 4 ) caused high zebrafish river water, the concentrations of macrolides are likely to be se-
embryo mortality, reduced the heart rate and embryo pigmenta- lective for antibiotic resistance and to have toxic effect on fresh-
tion, whereas Carlsson et al. (2009) observed the same effects at water green algae.
higher dilutions of Patancheru industrial effluents. In comparison Finally, our work prompts for further research and strategies for
to D. magna and zebrafish embryos, alga P. subcapitata was shown improved management of discharges from antibiotic production,
to be more sensitive to exposure to tested effluents because even including more efficient wastewater treatment technologies and
more than 32 times diluted winter and spring Company 1 samples defining emission limits for individual antibiotics, with emphasis
still caused growth inhibition. However, majority of the observed on the compounds that were shown to pose a greatest environ-
lethal and sub-lethal toxic effects can not be attributed to any single mental and health risk, such as macrolides.
antibiotic, since previously reported EC50 values for antibiotic
toxicity to algae, daphnids and fish (Table S4) are mostly higher Acknowledgements
than antibiotic concentrations we measured in the effluents
(Tables 3 and 4). The only exception is AZI: we have measured The authors are thankful to Nenad Muhin for the help during the
approx. 100  higher concentrations of AZI in winter and spring collection of samples and Lidija Buzuk for carrying out the Salmo-
Company 1 effluents than its EC50 value for algae (19 mg/L) (Harada nella genotoxicity assay. The research has been supported in part by
et al., 2008). Therefore, presence of macrolides could partly explain the Croatian Science Foundation under the project number UIP-
the toxic effects of Company 1 effluents to P. subcapitata. It is also 2014-09-9350 and in part by the Foundation of Croatian Academy
possible that some other undetected potent compound(s) in ef- of Sciences and Arts.
fluents are causative agent(s) of the observed effects, but a mixture
effect from multiple compounds is also possible. Indeed, it was Appendix A. Supplementary data
reported that synergism often occurs in antibiotic mixtures and
that certain combinations of antibiotics, such as mixtures of Supplementary data related to this article can be found at http://
different macrolides, present an additional ecological risk for dx.doi.org/10.1016/j.watres.2017.09.019.
aquatic ecosystems (Gonza lez-Pleiter et al., 2013; Yang et al., 2008).
Also, Zhang et al. (2012) have showed that antibiotics (OTC and CIP) References
make complexes with heavy metals (Cu, Zn and Cd) which are more
toxic to bacterium Vibrio fischeri and alga Scenedesmus obliquus Andersson, D.I., Hughes, D., 2014. Microbiological effects of sublethal levels of an-
than individual compounds. Such scenarios likely contributed to tibiotics. Nat. Rev. Microbiol. 12 (7), 465e478.
Babic, S., Bari
si
c, J., Vi
si
c, H., Klobu
car Sauerborn, R., Topi c Popovic, N., Strunjak-
the high toxicity of pharmaceutical effluents to various organisms


Perovi c, I., Co z-Rakovac, R., Klobu
car, G., 2017. Embryotoxic and genotoxic ef-
presented in this study and in the previous studies (Carlsson et al., fects of sewage effluents in zebrafish embryo using multiple endpoint testing.
2009; Larsson et al., 2007). Water Res. 115, 9e21.
Baker-Austin, C., Wright, M.S., Stepanauskas, R., McArthur, J., 2006. Co-selection of
Finally, the results from the Salmonella mutagenicity assay
antibiotic and metal resistance. Trends Microbiol. 14 (4), 176e182.
showed a mild mutagenicity of Company 2 effluents which sug- Baquero, F., Martínez, J.-L., Canto n, R., 2008. Antibiotics and antibiotic resistance in
gests that these effluents contained compounds with mutagenic water environments. Curr. Opin. Biotechnol. 19 (3), 260e265.
effects. This could be potentially ascribed to the presence of ENR in Barbosa, M.O., Moreira, N.F., Ribeiro, A.R., Pereira, M.F., Silva, A.M., 2016. Occurrence
and removal of organic micropollutants: an overview of the watch list of EU
those samples as fluoroquinolones were reported to induce geno- Decision 2015/495. Water Res. 94, 257e279.
toxic effects (Gocke, 1991; Hayasaki et al., 2006). However, other Bengtsson-Palme, J., Larsson, D.J., 2016. Concentrations of antibiotics predicted to
compounds or mixture of compounds cannot be excluded as well. select for resistant bacteria: proposed limits for environmental regulation.
Environ. Int. 86, 140e149.
Together, our data indicate that effluents from local antibiotic Berendonk, T.U., Manaia, C.M., Merlin, C., Fatta-Kassinos, D., Cytryn, E., Walsh, F.,
manufacturing plants can be contaminated to the point where they Bürgmann, H., Sørum, H., Norstro € m, M., Pons, M.-N., 2015. Tackling antibiotic
affect aquatic wild-life. Additional studies are required to elucidate resistance: the environmental framework. Nat. Rev. Microbiol. 13 (5), 310e317.
Białk-Bielin ska, A., Kumirska, J., Stepnowski, P., 2013. What do we know about the
adverse effects for human health which may rather be related to chronic and mixture toxicity of the residues of sulfonamides in the environ-
indirect risks coupled to potential promotion and spread of anti- ment? SMR 127, 79e77.
biotic resistance as a consequence of the environmental pollution Caldwell, D.J., Mertens, B., Kappler, K., Senac, T., Journel, R., Wilson, P.,
Meyerhoff, R.D., Parke, N.J., Mastrocco, F., Mattson, B., 2016. A risk-based
with antibiotics. approach to managing active pharmaceutical ingredients in manufacturing
effluent. Environ. Toxicol. Chem. 35 (4), 813e822.
5. Conclusions Cardoso, O., Porcher, J.-M., Sanchez, W., 2014. Factory-discharged pharmaceuticals
could be a relevant source of aquatic environment contamination: review of
evidence and need for knowledge. Chemosphere 115, 20e30.
This study demonstrates the multiple negative potential effects €
Carlsson, G., Orn, S., Larsson, D., 2009. Effluent from bulk drug production is toxic to
of antibiotic-contaminated pharmaceutical effluents for the aquatic vertebrates. Environ. Toxicol. Chem. 28 (12), 2656e2662.
receiving waters. Observed high concentrations of macrolides (up Chamarro, E., Marco, A., Esplugas, S., 2001. Use of Fenton reagent to improve organic
chemical biodegradability. Water Res. 35 (4), 1047e1051.
to few mg/L) along with high load of antibiotic-resistant bacteria Dautovi
Bare
si
c, J., Fiket, Z., c, J., Ahel, M., Mikac, N., 2014. Sources, distribution and
(105e107 CFU/mL) and high organic and nutrient loads in winter behavior of major and trace elements in a complex karst lake system. Aquat.
 A. Bielen et al. / Water Research 126 (2017) 79e87 87

Geochem. 20 (1), 19e38. Li, D., Yu, T., Zhang, Y., Yang, M., Li, Z., Liu, M., Qi, R., 2010. Antibiotic resistance
Decision, E., 2015. 495/2015, Commission Implementing Decision (EU) 2015/495 of characteristics of environmental bacteria from an oxytetracycline production
20 March 2015 establishing a watch list of substances for Union-wide moni- wastewater treatment plant and the receiving river. Appl. Environ. Microbiol. 76
toring in the field of water policy pursuant to Directive 2008/105/EC of the (11), 3444e3451.
European Parliament and of the Council. Offic. J. Eur. Union L 78, 40e42. Luo, Y., Guo, W., Ngo, H.H., Nghiem, L.D., Hai, F.I., Zhang, J., Liang, S., Wang, X.C.,
€derstro
Fick, J., So €m, H., Lindberg, R.H., Phan, C., Tysklind, M., Larsson, D., 2009. 2014. A review on the occurrence of micropollutants in the aquatic environ-
Contamination of surface, ground, and drinking water from pharmaceutical ment and their fate and removal during wastewater treatment. Sci. Total En-
production. Environ. Toxicol. Chem. 28 (12), 2522e2527. viron. 473, 619e641.
Flaherty, C.M., Dodson, S.I., 2005. Effects of pharmaceuticals on Daphnia survival, Maron, D.M., Ames, B.N., 1983. Revised methods for the Salmonella mutagenicity
growth, and reproduction. Chemosphere 61 (2), 200e207. test. Mutat. Res.Environ. Mutagen. Relat. Subj. 113 (3e4), 173e215.
Gadipelly, C., Pe rez-Gonza lez, A., Yadav, G.D., Ortiz, I., Iba n
~ ez, R., Rathod, V.K., Michael, I., Rizzo, L., McArdell, C., Manaia, C., Merlin, C., Schwartz, T., Dagot, C., Fatta-
Marathe, K.V., 2014. Pharmaceutical industry wastewater: review of the tech- Kassinos, D., 2013. Urban wastewater treatment plants as hotspots for the
nologies for water treatment and reuse. Ind. Eng. Chem. Res. 53 (29), release of antibiotics in the environment: a review. Water Res. 47 (3), 957e995.
11571e11592. Pal, C., Bengtsson-Palme, J., Kristiansson, E., Larsson, D.J., 2015. Co-occurrence of
Gocke, E., 1991. Mechanism of quinolone mutagenicity in bacteria. Mutat. Res.Fund. resistance genes to antibiotics, biocides and metals reveals novel insights into
Mol. Mech. Mutagen. 248 (1), 135e143. their co-selection potential. BMC Genom. 16 (1), 964.
Gonza lez-Pleiter, M., Gonzalo, S., Rodea-Palomares, I., Legane s, F., Rosal, R., Pruden, A., Larsson, D.J., Ame zquita, A., Collignon, P., Brandt, K.K., Graham, D.W.,
Boltes, K., Marco, E., Ferna ndez-Pin ~ as, F., 2013. Toxicity of five antibiotics and Lazorchak, J.M., Suzuki, S., Silley, P., Snape, J.R., 2013. Management options for
their mixtures towards photosynthetic aquatic organisms: implications for reducing the release of antibiotics and antibiotic resistance genes to the envi-
environmental risk assessment. Water Res. 47 (6), 2050e2064. ronment. Environ. Health Perspect. 121 (8), 878.
Guardabassi, L., Petersen, A., Olsen, J.E., Dalsgaard, A., 1998. Antibiotic resistance in Rivera-Utrilla, J., S anchez-Polo, M., Ferro-García, M.A.,  Prados-Joya, G., Ocampo-
Acinetobacter spp. isolated from sewers receiving waste effluent from a hospital Perez, R., 2013. Pharmaceuticals as emerging contaminants and their removal
and a pharmaceutical plant. Appl. Environ. Microbiol. 64 (9), 3499e3502. from water. A review. Chemosphere 93 (7), 1268e1287.
Gullberg, E., Cao, S., Berg, O.G., Ilb€ack, C., Sandegren, L., Hughes, D., Andersson, D.I., Roose-Amsaleg, C., Laverman, A.M., 2016. Do antibiotics have environmental side-
2011. Selection of resistant bacteria at very low antibiotic concentrations. PLoS effects? Impact of synthetic antibiotics on biogeochemical processes. Environ.
Pathog. 7 (7), e1002158. Sci. Pollut. Res. 23 (5), 4000e4012.
Halling-Sørensen, B., 2000. Algal toxicity of antibacterial agents used in intensive Sauerborn Klobu
car, R., Brozovi

c, A., Stambuk, A., 2013. Ecotoxicological assessment
farming. Chemosphere 40 (7), 731e739. of nitrofurantoin in fish cell lines, unicellular algae Desmodesmus subspicatus,
Harada, A., Komori, K., Nakada, N., Kitamura, K., Suzuki, Y., 2008. Biological effects of and bacterial strains of Salmonella typhimurium. Fresenius Environ. Bull. 22 (9
PPCPs on aquatic lives and evaluation of river waters affected by different A), 2669e2675.
wastewater treatment levels. Water Sci. Technol. 58 (8), 1541e1546. Seiler, C., Berendonk, T.U., 2012. Heavy metal driven co-selection of antibiotic
Hayasaki, Y., Itoh, S., Kato, M., Furuhama, K., 2006. Mutagenesis induced by 12 resistance in soil and water bodies impacted by agriculture and aquaculture.
quinolone antibacterial agents in Escherichia coli WP2uvrA/pKM101. Toxicol. Front. Microbiol. 3, 399.
Vitro 20 (3), 342e346. Senta, I., Krizman-Matasic, I., Terzic, S., Ahel, M., 2017. Comprehensive determina-
Idris, M., Kolo, B., Garba, S., Ismail, M., 2013. Physico-chemical analysis of phar- tion of macrolide antibiotics, their synthesis intermediates and transformation
maceutical effluent and surface water of river Gorax in Minna, Niger State, products in wastewater effluents and ambient waters by liquid chromatogra-
Nigeria. BEPLS 2 (3), 45e49. phyetandem mass spectrometry. J. Chromatogr. A 1509, 60e68.
Isidori, M., Lavorgna, M., Nardelli, A., Pascarella, L., Parrella, A., 2005. Toxic and Senta, I., Terzic, S., Ahel, M., 2008. Simultaneous determination of sulfonamides,
genotoxic evaluation of six antibiotics on non-target organisms. Sci. Total En- fluoroquinolones, macrolides and trimethoprim in wastewater and river water
viron. 346 (1), 87e98. by LC-tandem-MS. Chromatographia 68 (9e10), 747e758.
ISO 7346-1, 1996. Water quality - Determination of the acute lethal toxicity of Shepard, R., Fouda, H., Ferraina, R., Mullins, M., 1990. Disposition and metabolism of
substances to a freshwater fish [Brachydanio rerio Hamilton-Buchanan (Tele- azithromycin in rats, dogs and humans. Abstract. In: Proceedings of the 6th
ostei, Cyprinidae)]. International Congress of Infectious Diseases, p. 173 (Montreal, Canada).
ISO 15088, 2007. Water quality - Determination of the acute toxicity of waste water Spurgeon, D.J., Jones, O.A., Dorne, J.-L.C., Svendsen, C., Swain, S., Stürzenbaum, S.R.,
to zebrafish eggs (Danio rerio). 2010. Systems toxicology approaches for understanding the joint effects of
ISO 8692, 2012. Water quality - Fresh water algal growth inhibition test with uni- environmental chemical mixtures. Sci. Total Environ. 408 (18), 3725e3734.
cellular green algae. Stepanauskas, R., Glenn, T.C., Jagoe, C.H., Tuckfield, R.C., Lindell, A.H., King, C.J.,
ISO 6341, 2013. Water quality - Determination of the inhibition of the mobility of McArthur, J., 2006. Coselection for microbial resistance to metals and antibiotics
Daphnia magna Straus (Cladocera, Crustacea) - Acute toxicity test. in freshwater microcosms. Environ. Microbiol. 8 (9), 1510e1514.
James, O.O., Oluwaleye, S.E., Saka, K.L., Olabanji, D.J., 2014. Physico-chemical char- Ti
sler, T., Zagorc-Kon
can, J., 1999. Toxicity evaluation of wastewater from the
acteristics of pharmaceutical effluents from sango industrial area, Nigeria. Bull. pharmaceutical industry to aquatic organisms. Water Sci. Technol. 39 (10e11),
Env. Pharmacol. Life Sci. 3, 78e81. 71e76.
Kostich, M.S., Batt, A.L., Lazorchak, J.M., 2014. Concentrations of prioritized phar- Verlicchi, P., Al Aukidy, M., Galletti, A., Petrovic, M., Barcelo , D., 2012. Hospital
maceuticals in effluents from 50 large wastewater treatment plants in the US effluent: investigation of the concentrations and distribution of pharmaceuti-
and implications for risk estimation. Environ. Pollut. 184, 354e359. cals and environmental risk assessment. Sci. Total Environ. 430, 109e118.
Kristiansson, E., Fick, J., Janzon, A., Grabic, R., Rutgersson, C., Weijdegård, B., Wang, H., Che, B., Duan, A., Mao, J., Dahlgren, R.A., Zhang, M., Zhang, H., Zeng, A.,
€ derstro
So € m, H., Larsson, D.J., 2011. Pyrosequencing of antibiotic-contaminated Wang, X., 2014a. Toxicity evaluation of b-diketone antibiotics on the develop-
river sediments reveals high levels of resistance and gene transfer elements. ment of embryo-larval zebrafish (Danio rerio). Environ. Toxicol. 29 (10),
PloS One 6 (2), e17038. 1134e1146.
Larsson, D.J., 2014. Pollution from drug manufacturing: review and perspectives. Wang, S., Fan, C., Low, A., He, J., 2014b. Comparison of microbial communities in
Phil. Trans. Roy. Soc. B 369 (1656), 20130571. sequencing batch reactors (SBRs) exposed to trace erythromycin and erythro-
Larsson, D.J., de Pedro, C., Paxeus, N., 2007. Effluent from drug manufactures con- mycin-H2O. Appl. Microbiol. Biotechnol. 98 (6), 2667e2673.
tains extremely high levels of pharmaceuticals. J. Hazard. Mater. 148 (3), Yang, L.H., Ying, G.G., Su, H.C., Stauber, J.L., Adams, M.S., Binet, M.T., 2008. Growth-
751e755. inhibiting effects of 12 antibacterial agents and their mixtures on the fresh-
Li, D., Qi, R., Yang, M., Zhang, Y., Yu, T., 2011. Bacterial community characteristics water microalga Pseudokirchneriella subcapitata. Environ. Toxicol. Chem. 27 (5),
under long-term antibiotic selection pressures. Water Res. 45 (18), 6063e6073. 1201e1208.
Li, D., Yang, M., Hu, J., Zhang, J., Liu, R., Gu, X., Zhang, Y., Wang, Z., 2009. Antibiotic- Zhang, Y., Cai, X., Lang, X., Qiao, X., Li, X., Chen, J., 2012. Insights into aquatic tox-
resistance profile in environmental bacteria isolated from penicillin production icities of the antibiotics oxytetracycline and ciprofloxacin in the presence of
wastewater treatment plant and the receiving river. Environ. Microbiol. 11 (6), metal: complexation versus mixture. Environ. Pollut. 166, 48e56.
1506e1517.