FUNDACIÓN MAPFRESeguridad y Medio Ambiente

Year 31 Nº 124 2011

Antibiotic-resistant bacteria in aquatic environmentsENVIRONMENT

Increasing bacteria resistance to antibiotics is now causing many health problems, raising the population’s morbidity and mortality rates. The best two ways of reducing resistance are the rational use of antibiotics and the elimination of wastewater-excreted resistant bacteria. The aims of this project are: 1- To ascertain the abundance of antibiotic-resistant faecal bacteria in aquatic environments receiving wastewater from an antibiotic producing factory and from a wastewater treatment plant. 2- Study the capacity of wastewater treatment plants to eliminate or boost antibiotic resistance. 3- Study the variation in antibiotic resistance throughout the year. The study shows that wastewater outflows significantly increase the amount of resistant bacteria. The outflows from wastewater treatment plants have higher percentages of antibiotic-resistant bacteria than the inlet waters. This study found no significant increase in resistance between conventional systems (activated sludge) and low-cost systems (constructed wetlands). Neither did wetland design differences make any significant difference in the amount of resistant bacteria in their outflows.

Área de Ecología, Facultad de CC Biológicas y Ambientales, Universidad de León, 24071 León.


Antibiotic resistance as a bacteria adaptive strategy

Antibiotic resistance is one of the burning issues in human health. The rapid and proven increase in infections caused by antibiotic-resistant pathogenic bacteria is producing, in evolutionary terms, a sort of «arms race» (Van Valen 1973), in which the development of new weapons by man (antibiotics) is falling behind the micro-organism’s capacity of obviating their effect (increase of resistance).

The cause of antibiotic resistance is natural, i.e., the bacteria themselves have developed survival mechanisms to deactivate the antibiotic substances either in their own defence or as a competitive strategy against other micro-organisms (Martínez, 2009). But this resistance, or the bacteria’s adaptive strategy, is being inadvertently boosted by human action with the use of antibiotics to treat human and animal diseases. In this case man’s selective mechanism has increased the likelihood of contact between antibiotics and bacteria, thereby provoking a rapid selection of the bacterial populations and favouring the most resistant ones (Alonso et al., 2001). Although antibiotic resistance may crop up by genetic mutation in the absence of any antibiotic (Henriques et al., 2006), it is the excessive use of antibiotics that has caused such an alarming increase in resistance, prompting the introduction of legislation to curb their free acquisition (Kümmerer, 2004).

This use of antibiotics and the consequent increase of resistant bacteria is producing a concomitant increase in the excretion into the environment not only of resistant bacteria but also of antibiotics, mainly through wastewater. The emission of bacteria into the aquatic environment also favours genetic exchange with previously non-resistant populations, thereby increasing the dispersion of this resistant capacity in the bacteria of the environment (Davison 1999).

Survival of antibiotic-resistant bacteria in aquatic environments

Resistant bacteria enter rivers in wastewater. One of the aspects of interest here is to find out the capacity of these bacteria to survive in natural conditions and also to exchange genetic material in said environments. Although the natural mortality of pathogenic bacteria or their indicators is very high in extraenteral environments, their great abundance (wastewater values of 105-1010 CFUs/ml ) (McFeters et al., 1990) and the environmental conditions (turbidity, temperature, etc) might keep these populations viable for quite some time (Davies-Colley et al., 1999).

This study sets out to assess the viability of these antibiotic-resistant bacteria along a river receiving two outflows of a different nature. Firstly, treated domestic wastewater and secondly wastewater from a factory producing penicillin and cephalosporin. The aim is to find out the rate at which the main groups of faecal bacteria (total coliforms, E. coli, enterobacteria and sulphite-reducing clostridium) disappear with distance, comparing this rate of disappearance against the rate of disappearance of antibiotic resistance in the surviving bacteria. Cross checks with different antibiotics will then tell us the variation in the multi-resistance, which factor betrays genetic exchange within the river’s populations of faecal bacteria themselves, both the bacteria suspended in the water and held in the sediment. Resistance studies will be accompanied by the chemical analysis of the antibiotics potentially present, both in the raw wastewater and in the river.

Elimination of antibiotic-resistant bacteria. Role of the wastewater treatment plants

One of the methods for destroying wastewater pathogenic bacteria, or their indicator groups, is the treatment thereof in wastewater treatment plants. The treatment plants present hostile conditions for the survival of said bacteria due to their ambient temperature, the physicochemical conditions of the reagents and the great abundance of predators (bacterivorous organisms and viruses). In general the treatment plants reduce the abundance of inlet water bacteria by 1-3 log units (Hirata et al., 2003). This reduction in bacteria abundance, however, is not accompanied by a reduction in resistant bacteria. On the contrary, conventional wastewater treatment plants (activated sludge) tend to increase the percentage of resistant bacteria in the outflow as a result of the great abundance of bacteria in the biological reagent and the time of contact between them; this increases the genetic exchange rate (conjugation, transformation and transduction) and, ipso facto, the percentage of antibiotic-resistant bacteria (Da Costa et al., 2006).

Due to their high cost, activated sludge systems are replaced in rural areas by low-cost systems such as percolating filters or extensive systems like sludge lagoons or constructed wetlands.

Another of this study’s objectives is to compare conventional activated sludge systems with low cost systems like constructed wetlands. These systems differ in terms of cell retention times, bacteria density within the reagents and exposure to environmental conditions (Ferrer et al., 2007).

This study sets out to ascertain whether the «natural» wastewater treatment systems have the same capacity of increasing resistance rates as conventional systems and what role is played in this by the various design elements (type of plant used, hydraulic flow employed). It has been proven that natural systems are more efficient than conventional systems in the elimination of pathogenic and faecal indicator bacteria (García and Bécares, 1997), but there is no information on the capacity of exchanging antibiotic-resistant genes within said processes.

Furthermore, although the capacity of conventional processes for eliminating pharmaceutical compounds is well known by now (e.g., Hijosa-Valsero et al., forthcoming), there are as yet no figures on the role of low cost systems in eliminating antibiotics.


Water- and sediment-sample taking from the River Bernesga

The study area lies to the south of the city of León, in the vicinity of the River Bernesga. This area contains not only the Wastewater Treatment Plant (WWTP) of León and its hinterland but also the outflow of an antibiotics factory (AB) producing cephalosporin and penicillin, which at the moment runs its own wastewater treatment plant.

A river sampling point was set up before and after each outflow (AB and WWTP) (fig. 1), plus two points downriver of the latter outflow, at different distances, for taking both water and sediment samples and measuring the various physicochemical parameters. Samples were also taken and readings made at the two outflows.

The water and sediment samples were taken on 24, 30 and 31 August 2010 (n=3), between 9.00 and 13.00 hours. No rainfall was recorded during the sample taking campaign or in the previous two weeks. The sampling sites are shown in figure 1. In the sites River 1 (point 1), River 2 (point 2), River 3 (point 3), River 4 (point 4), River 5 (point 5) and River 6 (point 6) samples were taken along a transect that crossed the river from bank to bank. In the case of the water samples, 1000 ml were collected at each transect point. In the case of the sediment samples, a 9-cm diameter, 50-cm long methacrylate cylinder was used for extracting the core sample. All the samples from the same site were then mixed to obtain an integrated sample, both of water and of sediment.

The temperature, pH, dissolved oxygen, redox potential and conductivity readings were taken in situ by means of electrode probes (WTW, Weilheim, Germany).

Analysis of antibiotics in water and sediment of the River Bernesga

The analyses were conducted in the Departamento de Hidrogeología y Química Analítica (Analytical Chemistry and Hydrogeology Department) of Universidad de Almería under the coordination of Dr. Roberto Romero González. The analysis method is briefly described below. One thousand millimetres of river water was filtered through a cellulose acetate membrane filter with a diameter of 47 mm and pore size of 0.45 μm . Formic acid was then added until building up to pH = 3. A solid phase extraction was then made with Oasis HLB 200 mg 6 ml cartridges previously conditioned with 5 ml of methanol and 5 ml of water at pH = 3 (acidified with formic acid). The water samples were passed through the cartridges at a rate of about 5 ml / min-1 after which the cartridges were left to dry off for one hour. The cartridges were then gravity eluted with 10 ml of methanol. This methanol was then evaporated off with a gentle current of nitrogen gas. The dry residue was recomposed in 1 ml of a (1:1) mixture comprising methanol and an aqueous solution of formic acid 0.01%. This sample was then analysed with an ultra high performance liquid chromatograph (UHPLC) coupled up to a triple quadrupole.

As for the sediment analysis, the sediment was first of all dried to 30º C for 24 h to eliminate all the water. Two grams of dry sediment was then mixed with 10 ml of acetonitrile, 1 ml of citric acid solution 1 M at pH 4 and 0.5 ml of a solution of Na2-EDTA 0.5 M. The mixture was homogenised for 20 minutes in a sonicator and then centrifuged at 3000 rpm for 10 minutes. The supernatant was skimmed off and kept, while the precipitate was thrown away. This liquid was then evaporated off with a gentle current of nitrogen gas. The dry residue was recomposed in 1 ml of a (1:1) mixture comprising methanol and an aqueous solution of formic acid 0.01%. This sample was then analysed with an ultra high performance liquid chromatograph (UHPLC) coupled up to a triple quadrupole.

The chemical oxygen demand (COD), biological oxygen demand (BOD5), total suspended solids (TSS) and volatile suspended solids (VSS) were determined, respectively, with the procedures 5220 C, 5210 B, 2450 D and 2540 E as laid down in Standard Methods (APHA, 2005).

Microbiological variables sampling

Both the water and the sediment samples were micro-biologically analysed on the same day they were taken. Faecal indicators were analysed, i.e,. E. coli, total coliforms (TC), Enterococci and sulphite-reducing clostridia . The cultivation mediums used were: Chomocult coliform agar (Merck 1.10426.0500) for E. coli and TC, a selective chromogenic medium that distinguishes these two groups on the same plate according to the colony colour; Enterococci were analysed by means of the SB agar (Membrane-filter enterococcus selective agar according to Slanetz and Barley, Merck 1.05262.0500), and sulphite-reducing clostridia by SPS agar (Perfringens selective Agar according to ANGELOTTI, Merck 1.10235.0500).

Antibiotic resistance analysis

Antibiotic resistance was studied by means of two different methods: plate dilution and agar diffusion.

Plate Dilution

Under this method a measurement was made of the resistance of each bacterial group (except for clostridia) to two antibiotics, amoxicillin (A) and cefalexin (C). The number of colonies growing on the plates with antibiotic were compared with the control plates without antibiotic to check the percentage of resistant bacteria. In the constructed wetlands only the resistance to amoxicillin was tested but two different concentrations were applied to the mixture, 5 and 50 μg/ml (TC and E. coli) or 1 and 10 μg/ml (Enterococci).

Agar Diffusion (antibiogram)

The antibiotic discs used were: two β-lactamides (penicillin, 10U and ampicillin, 10μg), two tetracyclines (doxicycline, 30μg and tetracycline, 30μg), two macrolides (erithromycin, 15μg and azithromycin, 15μg) and one aminoglycoside (streptomycin, 10μg), (BBL Sensi-Disc Antimicrobial Susceptibility Test Discs, BD).

Colonies showing a «resistant» or «intermediate» behaviour were classified as resistant, or otherwise «sensitive» (Constanzo, 2005).

Elimination of bacteria in natural wastewater treatment systems (constructed wetlands)

As well as the samples taken along the river, samples were also taken from the aquatic macrophyte treatment plant (constructed wetlands) on the site of León’s WWTP and consisting of eight 1-m2 tanks with different plants and a different design in each one (figures 3 and 4). Each tank simulates a different design and system, as indicated below:

  • H1 and H5: hydroponic cultures of lesser reedmace (Typha angustifolia) and reed (Phragmites australis), respectively.
  • H2: Free water surface flow (FWSF) system planted up with lesser reedmace.
  • H3: Free water subsurface flow (FWSSF) system planted up with lesser reedmace.
  • H4: Free water surface flow system with no planted crop (control of H3).
  • H6: Subsurface flow (SSF) system planted up with phragmites and with triple the load of the rest of the systems.
  • H7: Subsurface flow (SSF) system planted up with phragmites.
  • H8: Subsurface flow system not planted up with any crop (control of H7).

Figure 3. Diagram of experiment tanks simulating the different designs of constructed wetlands

The aim of the study is to compare the most commonly used constructed wetland designs to ascertain whether any of them might affect the elimination of antibiotics or the generation of antibiotic-resistant bacteria, and to compare both aspects with a conventional treatment system like activated sludge.


Conventional water quality variables

The COD, BOD5, TSS and VSS readings of the water samples are shown in figure 5. The values of these parameters are constant and similar in all sites, except for the outflow from the antibiotics factory. This outflow has relatively high values of COD, TSS and VSS (regulated by the Resolution of 20 April 2009 of Castilla y León). Despite this, the discharge of this wastewater did not significantly increase the COD, TSS and VSS readings in the river (point 2, figure 5). This is largely due to the dilution effect, given that the river flow-rate at point 1 (1.24-1.79 m3 s-1) is much higher than the flow-rate of the outflow from the antibiotics factory (0.09 m3 s-1). Moreover the outflow of León’s WWTP has a very low charge of COD, BOD5 and solids in suspension, coming as it does from a WWTP that has to meet the requirements of Directive 91/271/EC; despite its high flow, therefore (1.24 m3 s-1), it does not excessively affect the river, at least insofar as these contaminants are concerned (i.e., readings at point 4 show no great increase on point 3; see figure 5).

Figure 5. Mean concentrations of the conventional water quality parameters (COD, BOD5, TSS and VSS) in the River Bernesga. The error bars show the confidence interval of 95%.

Concentration of antibiotics in the river

Antibiotics in the river water

The only substances detected in the water of the River Bernesga were enrofloxacin, danofloxacin, difloxacin, sarafloxacin, flumequine (quinolones), oxitetracyclin, chlortetracycline (tetracyclines), oxacillin (®-lactam antibiotic, penicillin) and benzathine (a penicillin-associated substance). The most abundant substances were enrofloxacin and benzathine.

In general, both the antibiotics-factory outflow and the León WWTP outflow show a higher concentration of antibiotics than at the first point of the river, proving that these outflows are a source feeding antibiotics into the natural waters of the river. Nonetheless, enrofloxacin, flumequine and benzathine were all detected at the first sampling point, although far away from the first outflow that might have affected it.

An earlier study, Hijosa-Valsero et al. (forthcoming) measured the concentration of antibiotics at the inlet and outlet of the León WWTP in the winter. In this case the method used showed up the presence of antibiotics not found in this study.


Antibiotics in the river sediments

The readings of this study tally with those shown in the international literature on river sediments. In terms of quinolones a French study (Pouliquen et al., 2009) found flumequine levels of 0-700 ng/g and oxolinic acid levels of 0-20 ng/g in river sediments. As regards the tetracycline group, these antibiotics are very common in sediment due to their chemical nature. Oxitetracyclin has been frequently detected in sediment from diverse parts of the world in concentrations of 0-200 μg/kg in China (Liu et al., 2009; Yang et al., 2010), of 0-8,3 μg/kg in the USA (Arikan et al., 2008; Kim and Carlson, 2007) and 0-180 ng/g in France (Pouliquen et al., 2009). Studies conducted in US rivers (Arikan et al., 2008; Kim and Carlson, 2007) detected doxicycline concentrations of 0 to 28 μg/kg in the sediments. Trimethoprim has been detected in China in concentrations of 34.6 ng/g (Tang et al., 2009).

River bacteria communities and antibiotic resistance

Determination of the bacteria community

The bacteria counts in the control plates without antibiotics show that both the antibiotics-factory treatment plant and the WWTP of León input bacteria into the river; the amount of bacteria of all groups is much higher (difference of two or three log units) after the WWTP outflow (River 4, River 5 and River 6) (figure 6).

Figure 6. Log 10 of the colony forming units per millimetre at each sampling point of the river

These differences stand out more clearly if the bacteria abundance is expressed in relation to the river flow-rate and outflows (bacteria load). River flow-rates were measured at point 1 (1.2 m3 seg-1) and point 6 (3 m3 seg-1), the outflows of AB and the WWTP are known (0.09 and 1.2 m3 seg-1 respectively), and the rest of the river points can be calculated (points 2 and 3: 1.8 m3 seg-1. Points 4 and 5: 3 m3 seg-1). This shows that the bacteria levels at each point of the river are higher after the outflows by an average difference of 3 log units (figure 7).

Figure 7. Bacteria load (CFUs/day) at each sampling point

Sediment readings also showed a substantial increase in the total number of bacteria in all groups as outflows were phased in down the river (figure 8).

Figure 8. Log 10 of the Colony Forming Units per millimetre at each sediment sampling point

Plate dilution resistance study

The water samples collected during the sampling period show a disparate antibiotic resistance pattern. In the case of cefalexin, the bacteria group with the highest resistance percentage is the total coliforms, showing readings of up to 100% resistance after the antibiotics-factory outflow, although these percentages than fall off along the river (figure 9), unlike the situation with the total number of bacteria. In the case of amoxicillin, both the total coliforms and E. coli show fairly high resistance percentages, also falling off further down the river. No resistant Enterococci were found except in the WWTP outflow (figure 10).

Figure 9. Cefalexin resistance percentages in river water samples.

Figure 10. Amoxicillin resistance percentages in river water samples


Sediment patterns are similar to those of water: the highest cefalexin resistance occurs after the outflow from the antibiotics factory, falling off thereafter down the river, and the bacteria group showing the highest resistance percentage is total coliforms (figure 11). As for amoxicillin, both total coliforms and E. coli show high resistance, falling off in general further down the river. Amoxicillin-resistant Enterococci were not found in sediment either (figure 12).

Figure 11. Cefalexin resistance percentages in river sediment samples

Figure 12. Amoxicillin resistance percentages in river sediment samples

Table 1. Resistance patterns of isolated E. coli colonies in the river and outflows
Concentration Colonies of resistant E. coli at each sampling point (%)
Antibiotic Concentration
River 1
River 2
River 3
River 4
River 5
River 6
Ampicillin 10 75,0 40,0 83,3 94,7 81,8 82,8 75,9 65,5
Doxicycline 30 25,0 20,0 0,0 52,6 61,9 58,6 69,0 55,2
Tetracycline 30 35,0 20,0 16,7 63,2 59,1 75,9 58,6 65,5
Streptomycin 10 40,0 100,0 0,0 63,2 52,4 65,5 55,2 34,5
Erithromycin 15 100,0 100,0 100,0 100,0 100,0 100,0 100,0 100,0
Azithromycin 15 11,8 0,0 16,7 31,6 9,1 17,2 17,2 20,7
Penicillin 10 (U) 100,0 100,0 100,0 100,0 100,0 100,0 100,0 100,0
Table 2. Resistance patterns of isolated E. coli colonies in sediments
Concentration Colonies of resistant E. coli at each sampling point (%)
Antibiotic Concentration
ment 1
ment 2
ment 3
ment 4
ment 5
ment 6
Ampicillin 10 100,0 100,0 100,0 81,5 44,4 82,8
Doxicycline 30 80,0 100,0 46,7 66,7 55,6 55,2
Tetracycline 30 92,0 57,1 66,7 70,4 51,9 48,3
Streptomycin 10 40,0 28,6 46,7 51,9 40,7 34,5
Erithromycin 15 96,0 85,7 100,0 100,0 100,0 100,0
Azithromycin 15 16,0 0,0 7,1 33,3 14,8 3,4
Penicillin 10 (U) 100,0 100,0 100,0 100,0 100,0 100,0
Antibiogram study of multiple resistances

A total of 289 E. coli colonies were isolated from the 14 sampling points for river water, sediment and outflows, and an analysis was then made of their antibiotic resistance patterns. Table 1 shows the resistance patterns obtained for the river water samples and also the outflow from the antibiotics factory and the outflow of the WWTP of León. Table 2 shows the sediment results. These tables show that both erithromycin and penicillin have resistance rates of 100% in all points while the rate of azithromycin-resistant bacteria never tops 35% at any point. The percentage of resistant bacteria also falls just after the outflow from the antibiotics factory (River point 2), then increasing and holding steady along the rest of the river. In the case of the sediment the resistance rates are fairly similar along all river sampling points.

Multi-resistance patterns

Figure 13 shows the distribution of these percentages in the river sampling and outflow points. In all points 100% of the bacteria are resistant to 2 or more antibiotics while over 80% show a resistance to 3 or more antibiotics. This figure also shows how the percentage of bacteria resistant to 3 antibiotics plunges as from point 3, giving way to multi-resistant bacteria resistant to more than 4 antibiotics.

Figure 14 shows the same figures for sediment, with a similar pattern: at point 2 multi-resistant bacteria resistant to 4 antibiotics predominate, then falling away with a concomitant increase of bacteria resistant to 5 or more antibiotics.

Figure 13. Distribution of the multi-resistance percentages of the river sampling points (left) and of the outflows (right).

Figure 14. Distribution of the multi-resistance percentages of the sediment sampling points

These figures suggest that crossover resistance is commonly developed in the ecosystem, since all isolated colonies were resistant to at least 2 antibiotics and many of them to more than 3 and more than 4; a significant number of colonies were also resistant to all the antibiotics tested.

Elimination of antibiotic-resistant bacteria in constructed wetlands

Determination of the bacteria community

In the effluents of constructed wetlands the dominant bacteria group is the TCs, followed by E. coli; the Enterococci are the least abundant group (figure 15). As for the elimination of the total number of bacteria, all the wetlands and also the WWTP show statistically significant differences (Kruskal-Wallis, p<0,001) with respect to the inflow waters (Infl.) for all bacteria groups. Furthermore, wetlands 3, 4, 7 and 8 have a significantly better elimination performance for all bacteria groups (Kruskal-Wallis, p<0.05) than the WWTP. The H7 set-up shows the best elimination performance of all.

Figura 15. Log 10 of the forming units colonies per milliliter of effluents Constructed wetlands (H #), the effluent from the EDAR (EDAR) and the influent of both (Infl.).


Analysis of antibiotic resistance in constructed wetlands

Analysis of the resistance readings of the wetland outflows shows that they do not present significantly different resistance rates (Kruskal-Wallis, p>0.05) from the effluent or inflow waters of the WWTP . Total coliforms (figure 16) show very high resistance rates with mean values of around 100%; in some cases (H7) higher growth rates are even observed in the antibiotic plate than in the control. The readings also show that an increase in the concentration of the antibiotic translates into a lower percentage of resistant bacteria. In the case of E. coli (figure 17) similar patterns show up among the different systems, but with somewhat lower resistance rates, about 50%. In general, the percentage of resistant bacteria also falls with an increase in the concentration of the antibiotic, but less than so than with the TCs. Enterococci (figure 18) show a much higher sensitivity to the antibiotic, the maximum percentages being lower than 10%.

Figure 16. Percentages of total coliforms resistant to amoxicillin in the outflows of the constructed wetlands (H#), the outflow of the WWTP (WWTP) and the inflow waters of both (Infl.).

Figure 17. Percentages of E. coli resistant to amoxicillin in the outflows of the constructed wetlands (H#), the outflow of the WWTP (WWTP) and the inflow waters of both (Infl.).

Figure 18. Percentages of Enterococci resistant to amoxicillin in the outflows of the constructed wetlands (H#), the outflow of the WWTP (WWTP) and the inflow waters of both (Infl.).


Although the mortality of pathogenic bacteria or their indicators is very high in extraenteral environments, their great abundance (McFeters et al., 1990) and certain environmental conditions might keep these populations viable for quite some time (Davies-Colley et al., 1999). This factor is particularly important in the case of sediment, which acts as a reservoir of bacteria (Fernandes and Watanabe, 2008; Alm et al., 2003; Howel et al., 1995) due, firstly, to the availability of nutrients (Davies et al., 1995), secondly, to the additional protection that these provide against the light (Davies-Colley et al., 1999) and, thirdly, protozoon predation (Davies and Bavor, 2000). Several authors have found a high correlation between bacteria density in sediment and water of different environments (beaches and freshwater shores) (Fernandes and Watanabe, 2008; Alm et al., 2003; Junco et al., 2005), showing the existence of a continual flow of microorganisms, and ipso facto of their genetic material (including resistance genes) between the two. All these factors might be impinging on the high resistance readings shown by the sediment in points 2 and 3, then falling further along the river.

In general, treatment plants reduce the abundance of inflow water bacteria by between 1 and 3 log units (Hirata et al., 2003; Reinthaler et al., 2003). This reduction, however, is not necessarily accompanied by a reduction in the number of resistant bacteria; quite the contrary, the number of resistant bacteria increases (Da Costa et al., 2006). The bacteria quantified in this study are the faecal indicators and are therefore easily detected in WWTP outflows where contamination is predominantly faecal; this is not the case in industrial wastewater, such as the outflow from the antibiotics factory. In this outflow, therefore, even though a high number of faecal indicators is not detected, it is quite possible that resistance genes are still being input through other types of undetected or even uncultivatable bacteria (Roszak, 1987; Ash, 2002). Constructed wetlands (natural systems) have been proven to be more effective in eliminating faecal-indicator bacteria than conventional treatment plants (García and Bécares, 1997); this has been borne out by this study.

The resistance percentages obtained in this study tally with the resistance ranges found by other authors. Thus, the Enterococci show very low amoxicillin resistance rates (0% in most cases ), coinciding with the findings of Fernandes and Watanabe (2008). The resistance of E. coli and the TCs to this same antibiotic are also comparable to those found by other authors (Carrol et al., 2005; Fars et al., 2005; Lefkowitz and Durán, 2009), although the TC readings are slightly higher in this study (nearly 20%). TC resistance readings in constructed wetlands are higher in this study than elsewhere. A comparison of antibiogram results shows ranges of between 3 and 66% of ampicillin-resistant E. coli, 100% resistant to erithromycin, between 15 and 33% resistant to tetracycline, 30% resistant to streptomycin and the remaining 70% to penicillin (Reinthaler et al., 2003; Lefkowitz and Durán, 2009; Costanzo et al., 2005; Carrol et al., 2005; Schwartz et al., 2003), all chiming in with the findings of this study.

Multiresistance is another constantly studied factor. Chelosi et al. (2003) found that more than 56% of the Gram negative bacteria from cultivated marine sediment were resistant to 5 or more antibiotics. Lefkowitz and Durán (2009) measured the multi-resistance of E. coli in wastewater treatment plants, obtaining outflow readings of 60% of bacteria multi-resistant to 2 or more antibiotics and 25% to 4 or more. Other authors have studied the same factor (Tendencia and de la Peña, 2001; Pillai et al., 1997; Lin and Biyela, 2005; Toroglu, 2005; Chapin et al., 2005), and the findings of this study fall within the same ranges found therein.


To FUNDACIÓN MAPFRE, which subsidised this study. To Roberto Romero of Universidad de Almería for his collaboration in the chemical analysis of the antibiotics. The Spanish Ministry of Science and Innovation (Ministerio de Ciencia e Innovación) subsidised the construction of the pilot wetland plants (projects CTM2005-06457-C05-03 and CTM2008-06676-C05-03TECNO).


  1. Alm, E. W., Burke, J., Spain, A. (2003). Fecal indicator bacteria are abundant in wet sand at freshwater beaches. Water Research 37 (16), 3978-3982.
  2. Alonso, A., Sánchez P., Martínez, J.L. (2001). Environmental selection of antibiotic resistance genes. Environmental Microbiology 3, 1-9.
  3. Arikan, O.A., Rice, C., Codling, E. (2008). Occurrence of antibiotics and hormones in a major agricultural watershed. Desalination 226, 121–133.
  4. Ash, R. J., Mauck, B. and Morgan, M. (2002). Antibiotic resistance of gram-negative bacteria in rivers, United States. Emerging Infectious Diseases 8, 713-716.
  5. Bécares, E. 1995. Biología de una planta depuradora de fangos activados de doble etapa tratando aguas residuales de la industria farmacéutica. Tesis doctoral. Universidad de León.
  6. Carroll, S., Hargreaves, M. and Goonetilleke, A. (2005). Sourcing faecal pollution from onsite wastewater treatment systems in surface waters using antibiotic resistance analysis. Journal of Applied Microbiology 99 , 471-482.
  7. Chapin, A., Rule, A., Gibson, K., Buckley, T. and Schwab, K. (2005). Airborne multidrug-resistant bacteria isolated from concentrated swine feeding operation. Enviromental Health Perspectives 113 , 137-142.
  8. Chelossi, E., Vezzulli, L., Milano, A., Branzoni, M., Fabiano, M., Riccardi, G. and Banat, I. M. (2003). Antibiotic resistance of benthic bacteria in fish-farm and control sediments of the Western Mediterranean. Aquaculture 219 (1-4), 83-97.
  9. Constanzo, S. D., Murby, J. and Bates, J. (2005). Ecosystem response to antibiotics entering the aquatic environment. Marine Pollution Bulletin 51 , 218-223.
  10. Da Costa P.M.M., Vaz-Pires P.M., Bernardo F.M. (2006). Antimicrobial resistance in Enterococci spp. Isolated from wastewater isolated in inflow, effluent and sludge from municipal sewage water treatment plants. Water Research 40, 1735-1740.
  11. Davies, C. M., Bavour, H. J. (2000). The fate of storm water associated bacteria in constructed wetland and water pollution control pond systems. Journal of Applied Microbiology 89 (2), 349-460.
  12. Davies, C. M., Long, J. A. H., Donal, M. and Ashbolt, N. J. (1995). Survival of fecal microorganism in marine and freshwater sediments. Applied Environmental Microbiology 61 (5), 1888-1896.
  13. Davies-Colley, R.J., Donninson A.M., Speed D.J., Ross C.M., Nagels J.W. (1999). Inactivation of faecal indicators microorganisms in waste stabilisation ponds: Interactions of environmental factors with sunlight. Water Research 33, 1220-1230.
  14. Davison, J. (1999). Genetic exchange between bacteria in the environment. Plasmid 42, 73-91.
  15. Fars, S., Oufdou, K., Nejmeddine, A., Hassani, L., Melloul, A. A., Bousselhaj, K., Amahmid, O., Bouhoum, K., Lakmichi, H. and Mezrioui, N. (2005). Antibiotic resistance and survival of faecal coliforms in activated sludge system in a semi- arid region (Beni Mellal, Morocco). World Journal Of Microbiology & Biotechnology 21 (4), 493-500.
  16. Fernandes, A. J. and Watanabe J. M. (2008). Antimicrobial resistance and especies composition of Enterococci spp. isolated from waters and sands of marine recreational beaches in Southeastern Brazil. Water Research 42 , 2242-2250.
  17. Ferrer, C., Sangüesa, I., Bécares, E. and Llopis, J.A. 2007. La depuración en pequeños municipios de Castellón. FACSA. Diputación de Castellón. 194 pp.
  18. García M. and Bécares, E. (1997). Bacterial removal in three pilotscale wastewater treatment systems for rural areas. Water Science Technology 35 (11-12), 197-200
  19. Henriques, I., Moura, A., Alves, A., Saavedra, M.J. and Correia, A. (2006). Analysing diversity among betalactamase encoding genes in aquatic environments. FEMS MIcrobiol. Ecol. 56:418-429.
  20. Hijosa-Valsero, M., Matamoros, V., Sidrach-Cardona, R., Villacorta, M., Bécares, E. and Bayona J.M. (2010). Comprehensive assessment of the design configuration of constructed wetlands for the removal of pharmaceuticals and personal care products from urban wastewater. Water Res. 44:3669-3678
  21. Hijosa-Valsero, M., Sidrach-Cardona, R., Martín-Villacorta, J. and Bécares, E. (2010). Optimization of performance assessment and design characteristics in constructed wetlands for the removal of organic matter. Chemosphere. 81:651-657
  22. Hijosa-Valsero, M., Fink, G., Schlüsener, M.P., Sidrach-Cardona, R., Martín-Villacorta, J., Ternes, T. and Bécares, E. Removal of antibiotics from urban wastewater by means of constructed wetland. Water Res.
  23. Hirata, T., Kawamura, K., Yano, K., Kaneko, M., Moniwa, T., Tosa, K. and Taguchi, K. (1993). Removal efficiencies of microorganisms in wastewater treatment processes. Water Science Technology 28 (7), 55-61.
  24. Howell, J. M., Coyne, M. S., Cornelius, P. L. (1996). Effect of sediment particle size and temperature on fecal bacteria mortality rates and the fecal coliform/fecal streptococci ratio. Journal of Environmental Quality 25 (6), 1216-1220.
  25. Junco, M. T. T., Martin, M. G., Toledo, M. L. P., Gómez, P. L. and Barrasa, J. L. M. (2001). Identification and antibiotic resistance of faecal enterococci isolated from water samples. International Journal Of Hygiene And Environmental Health 203 (4), 363-368.
  26. Kim, S.C., Carlson, K. (2007). Quantification of human and veterinary antibiotics in water and sediment using SPE/LC/MS/MS. Analytical and Bioanalytical Chemistry 387, 1301–1315.
  27. Kümmerer, K. ( 2004). Resistance in the environment. J. Antimicrob. Chemother. 54:311-320.
  28. Lefkowitz, J. R., Durán, M., (2009). Changes in antibiotic resistance patterns of Escherichia coli during domestic wastewater treatment. Water Environment Research 81 (9), 878-885
  29. Lin, J., & Biyela, P. T. (2005). Convergent acquisition of antibiotic resistance determinants amongst the Enterobacteriaceae isolates of the Mhlathuze River, KwaZulu-Natal (RSA) . Water SA 31 , 257-260.
  30. Liu, H., Zhang, G., Liu, C.Q., Li, L. and Xiang, M. (2009) The occurrence of chloramphenicol and tetracyclines in municipal sewage and the Nanming River, Guiyang City, China. Journal of Environmental Monitoring 11, 1199–1205.
  31. Martínez, J.L. (2009). The role of natural environments in the evolution of resistance traits in pathogenic bacteria. Proc. R. Soc. B 276, 2521-2530.
  32. McFeters, G.A. (1990). Drinking water microbiology. Springer-Verlag, New York.185-203.
  33. Pillai, S. D., Widmer, K. W., Maciorowski, K. G. and Ricke, S. C. (1997). Antibiotic resistance profiles of Escherichia coli isolated from rural and urban environments. Journal of Environmental Science and Health Part A: Environmental Science and Engineering & Toxic and Hazardous Substance Control 32 , 1665-1675.
  34. Pouliquen, H., Delépée, R., Thorin, C., Haury, J., Larhantec-Verdier, M., Morvan, M.L. and Le Bris, H. (2009). Comparison of water, sediment, and plants for the monitoring of antibiotics: a case study on a river dedicated to fish farming. Environmental Toxicology and Chemistry 28, 496–502.
  35. Reinthaler, F. F., Posch, J., Feierl, G., Wust, G., Haas, D., Ruckenbauer, G., Mascher, F. and Marth, E. (2003). Antibiotic resistance of E. coli in sewage and sludge. Water Research 37 (8), 1685-1690
  36. Roszak, D. B., & Colwel, R. R. (1987). Survival strategies of bacteria in the natural environment. Microbiological Reviews 51 , 365-379.
  37. Schwartz, T., Kohnen, W., Jansen, B. and Obst, U. (2003). Detection of antibiotic-resistant bacteria and their resistance genes in wastewater, surface water, and drinking water biofilms. FEMS Microbiology Ecology 43 , 325-335.
  38. Tang, C.M., Huang, Q.X., Yu, Y.Y. and Peng, X.Z. (2009). Multiresidue determination of sulfonamides, macrolides, trimethoprim and chloramphenicol in sewage sludge and sediment using ultrasonic extraction coupled with solid phase extraction and liquid chromatography- tandem mass spectrometry. Chinese Journal of Analytical Chemistry, Volume 37, 1119–1124.
  39. Tendencia, E. A. and De la Peña, L. D. (2001). Antibiotic resistance of bacteria from shrimp ponds. Aquaculture 195 , 193-204.
  40. Toroglu, S., Dinçer, S. and Kormaz, H. (2005). Antibiotic resistance in gram negative bacteria isolated from Aksu River (Kahramanmaras) in Turkey. Annals of Microbiology 55 , 229-233.
  41. Van Valen L. (1973). A new evolutionary law. Evolutionary Theory 1:1-30.
  42. Yang, J.F., Ying, G.G., Zhao, J.L., Tao, R., Su, H.C. and Chen, F. (2010). Simultaneous determination of four classes of antibiotics in sediments of the Pearl Rivers using RRLC–MS/MS. Science of the Total Environment 408, 3424–3432.

go top

Antibiotic-resistant bacteria in aquatic environments

Figure 1. Distribution of river sampling points (cumulative distances in metres).

Figure 2. Sample taking. Sediment core

Figure 4. View of the experimental constructed wetlands system in summer