Antibiotics are among the most commonly used group of pharmaceuticals in human medicine. They can therefore reach surface and groundwater bodies through different routes, such as wastewater treatment plant effluents, surface runoff, or infiltration of water used for agricultural purposes. It is well known that antibiotics pose a significant risk to environmental and human health, even at low concentrations. The aim of the present study was to evaluate the presence of aminoglycosides and phenicol antibiotics in municipal wastewaters, sea water and pharmaceutical effluents in Tunisia. All analysed water samples contained detectable levels of aminoglycoside and phenicol antibiotics. The highest concentrations in wastewater influents were observed for neomycin and kanamycin B (16.4ng mL−1 and 7.5 ng mL−1, respectively). Chloramphenicol was found in wastewater influents up to 3 ng mL−1. It was observed that the waste water treatment plants were not efficient in completely removing these antibiotics. Chloramphenicol and florfenicol were found in sea water samples near aquaculture sites at levels up to, respectively, 15.6 ng mL−1 and 18.4 ng mL−1. Also aminoglycoside antibiotics were found near aquaculture sites with the highest concentration of 3.4 ng mL−1 for streptomycin. In pharmaceutical effluents, only gentamycin was found at concentrations up to 19 ng mL−1 over a sampling period of four months.

During recent decades, research of environmental contaminants has expanded beyond classical contaminants like pesticides, biocides and dioxins, to personal care products and pharmaceuticals (Ternes et al. 2004). Antibiotic compounds are widely used in human and veterinary medicine as a preventive or curative treatment for bacterial infections or as growth promoters (Sapkota et al. 2008). After administration, these compounds are only partially metabolized and are usually excreted unchanged in the urine and faeces (Junker et al. 2006; Mitani & Kataoka 2006).

Wastewaters become one of the main routes for the introduction of antibiotics into the environment (Ericson 2002; Karthikeyan & Meyer 2006; Xu et al. 2007) since they are often not completely eliminated during waste water treatment (Fatta-Kassinos et al. 2011; Yan et al. 2013a, 2013b). Resistance to antibiotics constitutes a major threat to public health and ought to be recognized as such more widely than it is at present. Moreover, wildlife could be severely affected by the presence of antibiotics in the environment, even at trace levels. Therefore, the European Union (EU) recommends the prudent use of antimicrobial agents in medicine (Werner & Bronzwaer 2007; Torres et al. 2010).

Aminoglycoside antibiotics are widely used in hospitals for treatment of serious human infection by Gram-negative and Gram-positive bacteria (Löffler & Ternes 2003) and in veterinary medicine (Salisbury 1995). Gentamicin, spectinomycin and lincomycin are among the most commonly used aminoglycosides (Mitscher et al. 2008; Beale 2011; Vucicevic-Prceti et al. 2011). Chloramphenicol, thiamphenicol and florfenicol have been used in human and veterinary medicine since the 1950s. Although many countries have strictly banned the use of these antibiotics in food-producing animals (Gantverg et al. 2003; European Commission EEC 2377/90). Illegal use of chloramphenicol and thiamphenicol in livestock and aquaculture still exists in China due to their low prices and steady antibiosis effectiveness (Chen et al. 2009). However, few researches have examined the contamination of those compounds, especially in the aquatic environment (Yan et al. 2013b). In our study, we evaluate the presence of aminoglycosides and phenicol antibiotics in Tunisian waters and pharmaceutical effluents. Samples were analysed using ultra-performance liquid chromatography-tandem mass spectrometry (UPLC-MS/MS). To our best knowledge, this is the first study to determine those antibiotics in pharmaceutical effluents, waste waters and sea waters nearby aquaculture sites in North African countries.

Water sampling

All samplings were performed in 2013. Influent and effluent waste water samples were obtained from four full-scale wastewater treatment plants (WWTPs) in the region of Grand-Tunis, a big metropolitan area in Tunisia, covering Tunis-nord, Charguia, Chotrana and Ben Arous.

The pharmaceutical effluents were collected at two discharge sites in the environment in the north of Tunisia (Zghouen and Ariana). At each sampling site, one monthly sampling was performed for four successive months (February–May).

Sea water samples were taken at Tunis, Bizerte, Monastir, Sousse and Hergla. The five sampling sites were chosen because of their human activities (aquaculture and hotels).

All samples were collected in a 1L amber recipient. After filtration through a 1 μm membrane filter (Euro Scientific, Belgium), aliquots were taken and stored at ≤ −20°C until analysis.

Chemicals and reagents

All analytical standards, except chloramphenicol-d5, were purchased from Sigma-Aldrich (Bornem, Belgium). Chloramphenicol-d5 was purchased from Witega (Berlin, Germany). Separate stock solutions of each antibiotic were prepared in methanol (1 mg mL−1) and stored in amber glass vials at ≤ −20°C. ULC-grade solvents were purchased from Biosolve (Valkenswaard, The Netherlands). All other chemicals for extraction and purification were of high-performance liquid chromatography-grade quality (HPLC-grade). HPLC-grade water was produced by a Milli-Q® Gradient A10 water purification system (Millipore, Bedford, MA, USA) Oasis® HLB (600 mg, 6 mL) and Oasis® CBA (500 mg, 3 mL) solid phase extraction (SPE) cartridges were purchased from Waters (Milford, MA, USA).

Sample analysis

Water filtrates (100 mL) were spiked with the internal standard (chloramphenicol-d5 or tobramycin for phenicol or aminoglycosides respectively). The water samples were further purified by SPE using Oasis® HLB cartridges for phenicol antibiotics and Oasis® CBA cartridges for aminoglycoside antibiotics. After elution, evaporation to dryness and reconstitution, an 8 μL aliquot was injected on the analytical column. UPLC-MS/MS analysis was performed using a Waters Acquity UPLC® class system and an Acquity UPLC® BEH C18 column (100 × 2.1 mm) interfaced with a Xevo TQ-S tandem quadrupole mass spectrometer, all from Waters (Milford, MA, USA). The instrument was used in the positive electrospray ionization mode for the analysis of the aminoglycosides and in the negative mode for phenicols. The method was validated according to commission decision 2002/657/EC for linearity, specificity, sensitivity, accuracy and precision (European Commission 2002/657/EC).

Two distinct analytical protocols were applied to extract the phenicols and the aminoglycosides in waste and sea water. After extraction and purification, the samples were analysed with UPLC-MS/MS within 14 min. The methods were fully validated and the validation data (data not shown) were compliant with the acceptance criteria of the commission decision 2002/657/EC. The methods were able to detect and quantify all compounds below 0.5 ng mL−1.

Waste water samples were taken at influent and effluent from four sewage treatment plants in Grand Tunis, i.e. Tunis-nord, Charguia, Chotrana and Ben Arous. The results are presented in Table 1. Florfenicol and neomycin were detected in all waste water samples. The highest concentrations were observed for neomycin and kanamycin B (16.4 ng mL−1 and 7.5 ng mL−1, respectively). The most detected antibiotics were present in both treated and untreated municipal water samples. This might indicate that the wastewater treatment process in Tunisia, which consists of pretreatment, primary settling and biological treatment (activated sludge) (Saddoud et al. 2007), is not very efficient in completely removing aminoglycoside and phenicol antibiotics from wastewater.

Table 1

Antibiotic concentrations detected in wastewater samples (ng mL−1) using UPLC-MS/MS analysis

 Sampling sites
 Tunis-nord
Charguia
Chotrana
Ben Arous
CompoundIEIEIEIE
Chloramphenicol 3.3 1.1 nd nd 0.5 0.3 nd nd 
Thiamphenicol 1.2 nd nd nd nd nd nd nd 
Florfenicol 3.3 0.8 1.2 0.8 1.4 0.6 0.9  0.1 
Paromycin 4.2 1.3 nd nd nd nd 1.2 0.9 
Dihydrostreptomycin 0.9 0.4 nd nd nd nd nd nd 
Kanamycin B 7.5 5.4 1.2 0.7 1.0 0.7 0.5 nd 
Aparamycin 1.5 0.5 1.2 0.4 nd nd 0.7 nd 
Streptomycin nd nd 2.7 1.2 1.6 0.8 nd nd 
Amikacin 2.1 1.8 nd nd nd nd 2.3 1.0 
Sisomycin 6.7 3.9 nd nd 2.3 1.0 4.1 2.0 
Neomycin 12 4.6 16.4 11.2 6.2 2.1 1.8 0.4 
Gentamycin c1a 1.6 0.6 0.5 0.2 nd nd nd nd 
Gentamycin c1 0.8 0.3 0.8 0.4 nd nd 0.6 nd 
Gentamycin c2 1.0 0.3 nd nd nd nd nd nd 
 Sampling sites
 Tunis-nord
Charguia
Chotrana
Ben Arous
CompoundIEIEIEIE
Chloramphenicol 3.3 1.1 nd nd 0.5 0.3 nd nd 
Thiamphenicol 1.2 nd nd nd nd nd nd nd 
Florfenicol 3.3 0.8 1.2 0.8 1.4 0.6 0.9  0.1 
Paromycin 4.2 1.3 nd nd nd nd 1.2 0.9 
Dihydrostreptomycin 0.9 0.4 nd nd nd nd nd nd 
Kanamycin B 7.5 5.4 1.2 0.7 1.0 0.7 0.5 nd 
Aparamycin 1.5 0.5 1.2 0.4 nd nd 0.7 nd 
Streptomycin nd nd 2.7 1.2 1.6 0.8 nd nd 
Amikacin 2.1 1.8 nd nd nd nd 2.3 1.0 
Sisomycin 6.7 3.9 nd nd 2.3 1.0 4.1 2.0 
Neomycin 12 4.6 16.4 11.2 6.2 2.1 1.8 0.4 
Gentamycin c1a 1.6 0.6 0.5 0.2 nd nd nd nd 
Gentamycin c1 0.8 0.3 0.8 0.4 nd nd 0.6 nd 
Gentamycin c2 1.0 0.3 nd nd nd nd nd nd 

nd: not detected.

I: influent water. E: effluent water.

The phenicol antibiotics chloramphenicol and florfenicol were most frequently found in Hergla and Monastir sea waters (Table 2). The highest concentrations were detected in Hergla at 15.6 ng mL−1 and 18.4 ng mL−1 for chloramphenicol and florfenicol, respectively. Chloramphenicol is not authorized for use in food-producing animals in the EU (European Commission EEC 2377/90). The presence of this antibiotic in the vicinity of aquaculture sites of Hergla and Monastir might suggest that chloramphenicol is likely to enter the Tunisian food chain.

Table 2

Antibiotic concentrations in sea water samples (ng mL−1) using UPLC-MS/MS analysis

 Sampling sites
CompoundTunisBizerteMonastirSousseHergla
Chloramphenicol 0.4 nd 3.5 0.2 15.6 
Florfenicol nd nd 2.9 nd 18.4 
Paromycin nd nd nd nd nd 
Dihydrostreptomycin 0.7 0.3 0.9 0.3 1.9 
Kanamycin B 0.1 0.2 0.5 0.6 1.5 
Aparamycin 0.2 0.4 0.9 1.8 0.4 
Streptomycin nd 0.7 1.4 0.7 3.4 
Amikacin 0.1 0.3 0.1 0.8 1.2 
Sisomycin nd nd 0.4 nd 0.4 
Neomycin nd nd 0.1 nd 0.7 
Gentamycin c1a nd nd 0.8 0.2 1.4 
Gentamycin c1 nd nd nd nd nd 
Gentamycin c2 nd nd nd nd nd 
 Sampling sites
CompoundTunisBizerteMonastirSousseHergla
Chloramphenicol 0.4 nd 3.5 0.2 15.6 
Florfenicol nd nd 2.9 nd 18.4 
Paromycin nd nd nd nd nd 
Dihydrostreptomycin 0.7 0.3 0.9 0.3 1.9 
Kanamycin B 0.1 0.2 0.5 0.6 1.5 
Aparamycin 0.2 0.4 0.9 1.8 0.4 
Streptomycin nd 0.7 1.4 0.7 3.4 
Amikacin 0.1 0.3 0.1 0.8 1.2 
Sisomycin nd nd 0.4 nd 0.4 
Neomycin nd nd 0.1 nd 0.7 
Gentamycin c1a nd nd 0.8 0.2 1.4 
Gentamycin c1 nd nd nd nd nd 
Gentamycin c2 nd nd nd nd nd 

nd: not detected.

In comparison, chloramphenicol was quantified up to 68 ng L−1 in sewage treatment plant effluent in the study of Alexy & Kümmerer (2006). Lin et al. (2008) quantified the chloramphenicol in hospital effluent at 1 ng L−1 and the thiamphenicol at 4 ng L−1. In the study of Lopez-Serna et al. (2013), chloramphenicol was not found in the groundwater in Spain, however this antibiotic was detected in Victoria, Pearl River, Harbour and in sea water in China (Xu et al. 2007). Xu et al. (2007) reported chloramphenicol concentrations up to 266 ng L−1 in the Pearl River in China.

Aminoglycoside antibiotics were only detected at low levels in all sea water samples. On the other hand, almost every sampling point was contaminated with these antibiotics. The highest concentration was found for streptomycin in sea water of Hergla at 3.4 ng mL−1. Results of sea water samples are presented in Table 2.

Aminoglycosides are mostly non-metabolized after administration and excreted unchanged via the urine (Marzo & Dal Bo 1998). The analysis of hospital wastewater in Germany revealed that the concentration of gentamycin was between 0.4 and 7.6 μg mL−1 (Löffler & Ternes 2003). In our study, the presence of antibiotics in sea water can be explained by the discharges of municipal treated water and wastewaters from hotels in the sea around the populous region of Sousse and Tunis. Regarding sea waters, and as expected, concentrations found were lower than the ones in waste water samples. This could indicate that a dilution factor occurs once antibiotics enter the sea.

In pharmaceutical effluents, gentamycin c1a was detected at one sampling point (Zghouen) on each monthly taken sample. The average concentration found was 14.8 ng mL−1 (Table 3). It seems that this pharmaceutical company continuously releases these antibiotics into the environment.

Table 3

Antibiotic concentrations determined in a pharmaceutical effluent (ng mL−1) over a period of 4 months (in 2013) in Zghouen, North-Tunisia using UPLC-MS/MS analysis

CompoundFebruaryMarchAprilMay
Chloramphenicol nd nd nd nd 
Florfenicol nd nd nd nd 
Paromycin nd nd nd nd 
Dihydrostreptomycin nd nd nd nd 
Kanamycin B nd nd nd nd 
Aparamycin nd nd nd nd 
Streptomycin nd nd nd nd 
Amikacin nd nd nd nd 
Sisomycin nd nd nd nd 
Neomycin nd nd nd nd 
Gentamycin c1a 15.2 19.1 10.8 14.4 
Gentamycin c1 nd nd nd nd 
Gentamycin c2 nd nd nd nd 
CompoundFebruaryMarchAprilMay
Chloramphenicol nd nd nd nd 
Florfenicol nd nd nd nd 
Paromycin nd nd nd nd 
Dihydrostreptomycin nd nd nd nd 
Kanamycin B nd nd nd nd 
Aparamycin nd nd nd nd 
Streptomycin nd nd nd nd 
Amikacin nd nd nd nd 
Sisomycin nd nd nd nd 
Neomycin nd nd nd nd 
Gentamycin c1a 15.2 19.1 10.8 14.4 
Gentamycin c1 nd nd nd nd 
Gentamycin c2 nd nd nd nd 

nd: not detected.

The widespread occurrence of antibiotics in the environment may cause deleterious health effects. The transfer of resistant bacteria to humans via water and food or to other pathogenic bacteria is an important health issue (Fatta-Kassinos et al. 2011). In addition to the risk of development and spread of resistant bacteria, the antibiotics may be toxic for human health (anaemia and hypersensitivity) (Roybal 1998).

To the best of our knowledge, this is the first time that antibiotics were monitored in Tunisian waste and sea waters. All analysed water samples contained detectable levels of aminoglycoside and phenicol antibiotics. Influent water of Tunis-Nord WWTP was heavily contaminated and contained not less than 14 antibiotics at concentrations ranging from 0.9 to 12 ng mL−1. The highest concentrations in wastewater were observed for neomycin. It was further observed that the WWTPs were not efficient in removing the antibiotics.

Florfenicol and the EU-forbidden chloramphenicol antibiotic were found in sea water samples near aquaculture sites at levels up to 18.4 ng mL−1 and 15.6 ng mL−1, respectively. Also aminoglycoside antibiotics were found nearby aquaculture sites with the highest concentration of 3.4 ng mL−1 streptomycin. The measured antibiotics are quite ubiquitous pollutants in the Tunisian region. The concentrations found in the waste and sea water samples might pose environmental and health risks.

Alexy
 
R.
Kümmerer
 
K.
2006
Antibiotics for human use
. In:
Organic Pollutants in the Water Cycle: Properties, Occurrence, Analysis and Environmental Relevance of Polar Compounds
(
Reemtsma
 
T.
Jekel
 
M.
, eds).
Wiley-VCH
,
Weinheim
,
Germany
, pp.
65
85
.
Beale
 
J. H.
2011
Antibacterial antibiotics
. In:
Wilson and Gisvold's Textbook of Organic Medicinal and Pharmaceutical Chemistry
, 12th edn (
Beale
 
J. M.
Block
 
J. H.
, eds).
Lippincott Williams & Wilkins Publishers
,
Philadelphia
,
USA
, pp.
294
300
.
Ericson
 
B. E.
2002
Analyzing the ignored environmental contaminants
.
Environ. Sci. Technol.
36
,
140A
145A
.
European Commission (2002/657/EC)
of 12 August 2002, Implementing Council Directive 96/23/EC concerning the performance of analytical methods and the interpretation of results, Official Journal of the European Commission M221, 8–36
.
European Commission (2377/90)
of 26 June 1990, Laying down a Community procedure for the establishment of maximum residue limits of veterinary medicinal products in food stuffs of animal origin, Official Journal of the European Commission L244, 1–8
.
Mitscher
 
L. A.
Lemke
 
T. L.
Gentry
 
E. J.
2008
Antibiotics and antimicrobial agents
. In:
Foye's Principles of Medicinal Chemistry
, 6th edn (
Williams
 
D. A.
Lemke
 
T. L.
, eds).
Lippincott Williams & Wilkins Publishers
,
Philadelphia
,
USA
, pp.
1028
1083
.
Roybal
 
J. E.
1998
Chloramphenicol and related drugs
. In:
Analytical Procedures for Drug Residues in Food of Animal Origin
(
Turnipseed
 
S. B.
Long
 
A. R.
, eds).
Science Technology System
,
West Sacramento, CA
,
USA
, pp.
227
260
.
Saddoud
 
A.
Ellouze
 
M.
Dhouid
 
A.
Sayadi
 
S.
2007
Anaerobic membrane bioreactor treatment of domestic waste water in Tunisia
.
Desalination
207
,
205
215
.
Salisbury
 
C. D. C.
1995
Chemical analysis aminoglycoside antibiotics
. In:
Chemical Analysis for Antibiotics Used in Agriculture
(
Oka
 
H.
Nakazawa
 
H.
Harada
 
K. K. E.
MacNeil
 
J. D.
, eds).
AOAC International
,
Arlington, USA
.
Sapkota
 
A.
Sapkota
 
A. R.
Kucharski
 
M.
Burke
 
J.
McKenzie
 
S.
Walker
 
P.
Lawrence
 
R.
2008
Aquaculture practices and potential human health risks: current knowledge and future priorities
.
Environ. Int.
34
,
1215
1226
.
Ternes
 
T. A.
Joss
 
A.
Siegrist
 
H.
2004
Scrutinizing pharmaceuticals and personal care products in wastewater treatment
.
Environ. Sci. Technol.
38
,
392
399
.
Torres
 
C.
Moreno
 
M. A.
Zarazaga
 
M.
2010
Prudent use of antimicrobial agents: not just for humans
.
Enfer. Infecc. Microbiol. Clin.
28
,
669
671
.
Werner
 
G.
Bronzwaer
 
S.
2007
Ensuring prudent use of antimicrobials in human medicine in the European Union
.
Eurosurveillance
12
,
64
67
.
Yan
 
C.
Dinh
 
Q. T.
Chevreuil
 
M.
Garnier
 
J.
Roose-Amsaleg
 
C.
Labadie
 
P.
Laverman
 
A.
2013a
The effect of environmental and therapeutic concentrations of antibiotics on nitrate reduction rates in river sediment
.
Water Res.
47
,
3654
3662
.
Yan
 
C.
Yang
 
Y.
Zhou
 
J.
Liu
 
M.
Nie
 
M.
Shi
 
H.
Gu
 
L.
2013b
Antibiotics in the surface water of the Yangtze Estuary: occurrence, distribution and risk assessment
.
Environ. Pollut.
175
,
22
29
.