Environ Monit Assess (2015) 187:342 DOI 10.1007/s10661-015-4573-8
Densities and antimicrobial resistance of Escherichia coli isolated from marine waters and beach sands. Vanessa da Costa Andrade & Bruna Del Busso Zampieri & Eliete Rodrigues Ballesteros & Aline Bartelochi Pinto & Ana Julia Fernandes Cardoso de Oliveira
Received: 24 March 2015 / Accepted: 27 April 2015 # Springer International Publishing Switzerland 2015
Abstract Bacterial resistance is a rising problem all over the world. Many studies have showed that beach sands can contain higher concentration of microorganisms and represent a risk to public health. This paper aims to evaluate the densities and resistance to antimicrobials of Escherichia coli strains, isolated from seawater and samples. The hypothesis is that microorganisms show higher densities in contaminated beach sands and more antimicrobial resistance than the water column. Density, distribution, and antimicrobial resistance of bacteria E. coli were evaluate in seawater and sands from two recreational beaches with different levels of pollution. At the beach with higher degree of pollution (Gonzaguinha), water samples presented the highest densities of E. coli; however, higher frequency of resistant strains was observe in wet sand (71.9 %). Resistance to a larger number of antimicrobial groups was observe in water (betalactamics, aminoglycosides, macrolides, rifampicins, and tetracyclines) and sand (betagalactamics and aminoglycosids). In water samples, highest frequencies of resistance were obtain against ampicilin (22.5 %), streptomycin (15.0 %), and rifampicin (15.0 %), while in sand, the highest frequencies were observe in relation to ampicilin (36.25 %) and V. da Costa Andrade : B. Del Busso Zampieri : A. B. Pinto Campus de Rio Claro, Universidade Estadual Paulista, UNESP, Avenida 24A, 1515, 13506-900 Rio Claro, SP, Brazil E. R. Ballesteros : A. J. Fernandes Cardoso de Oliveira (*) Laboratory of Marine Microbiology, Campus do Litoral Paulista, Universidade Estadual Paulista, UNESP, Praça Infante Dom Henrique, s/n, 11330-900 São Vicente, SP, Brazil e-mail: [email protected]
streptomycin (23.52 %). At the less polluted beach, Ilha Porchat, highest densities of E. coli and higher frequency of resistance were obtain in wet and dry sand (53.7 and 53.8 %, respectively) compared to water (50 %). Antimicrobial resistance in strains isolated from water and sand only occurred against betalactamics (ampicilin and amoxicilin plus clavulanic acid). The frequency and variability of bacterial resistance to antimicrobials in marine recreational waters and sands were related to the degree of fecal contamination in this environment. These results show that water and sands from beaches with a high index of fecal contamination of human origin may be potential sources of contamination by pathogens and contribute to the dissemination of bacterial resistance. Keywords Escherichia coli . Antimicrobial resistance . Marine recreational beaches . Sand . Water
Introduction Bacterial resistance is a rising problem all over the world, generated by processes of selection that result from the massive use of antibiotics (Frieden et al. 1993). The sewage discharge of domestic effluents and from strongly selective sources such as hospitals, industries, aquiculture, and among others has led to an increase in the distribution and frequency of bacterial resistance genes in several environments, including fresh and marine waters (Schwartz et al. 2003).
342
Environ Monit Assess (2015) 187:342
Page 2 of 10
In coastal regions, the high development of tourism, especially because of recreational activities such as diving, swimming, aquatic sports, and fishery, generates several impacts of anthropogenic origin. One of these impacts is the high densities of pathogens that are found in water column and sediments, including beach sands. These pathogens come from the sewage discharge which may be partially treated or in natura and carry several microorganisms such as bacteria, viruses, and protozoans (WHO 1998). When these pathogens reach waters and sands, they expose beach goers to the possibility of acquiring diseases, although the incidence of these diseases depends upon a variety of factors, such as the level of pollution in the water, the nature and length of exposure, and the immunological condition of the beachgoer (Bartram and Rees 2000). Many studies have showed that beach sands can contain higher concentrations of microorganisms, including pathogenic bacteria (Mendes et al. 1993; Papadakis et al. 1997; Alm et al. 2003; Whitman and Nevers 2003; Mudryk 2005; Oliveira and Pinhata 2008; Whitman et al. 2014), than the water column, once bacteria find in this environment favorable conditions of nutrients, protection against sunlight and predation by protozoans (Davies et al. 1995; Davies-Colley et al. 1999; Davies and Bavor 2000; Sinton et al. 1994; Pianetti et al. 2004; Villar et al. 1999). Thus, beach sands contaminated by pathogenic microorganisms can represent a higher risk to the visitors’ health than the contact with the marine recreational waters. In addition, marine recreational waters and sands that receive domestic sewage can be potential sources of resistant bacteria, and these areas may contribute to the establishment of dissemination routes of microorganisms carrying antimicrobial resistance genes. This is because besides revealing intrinsic resistance to several antibiotics bacteria can also, through conjugation with bacteria from the same or different species, acquire resistance factors present in plasmid (Huycke et al. 1998; Kühn et al. 2000). The frequency and variability of resistant strains of bacteria to different antimicrobial have increased in the last time, becoming a great challenge for the treatment of infectious diseases (Arvanitidou et al. 2001). The presence in sands and recreational waters of bacterial strains, which show resistance to antibiotics, is relevant due to the relationship between waterborne disease and recreational activities. It is expected for sand beaches to show higher bacterial densities than the water
column, since it can act as a reservoir for pathogenic microorganisms and can collaborate to increase the dissemination of resistance genes, which represent a risk to public health specially because there are no parameters or monitoring programs for sands of recreational beaches on the current legislation. Therefore, the present study aimed to evaluate the densities and resistance to antimicrobials of Escherichia coli strains, which were isolated from seawater and sand samples from two recreational beaches under different levels of pollution. In addition, environment potentials as sources of dissemination of bacterial resistance were evaluate based on the correlation between quality of marine recreational waters, bacterial habitat, frequency, and type of resistance.
Materials and methods Study area São Vicente (23° 57′ S and 46° 23′ W) is located on the Santos Basin (BSantos Lowland^), south coast of São Paulo State (Fig. 1). The Santos Basin is so named due to its geography, since it borders the Serra do Mar (BSea Highlands^). It occupies an area of 2373 km2, with 161 km2 of beaches. Due to the heterogeneity of its beaches and the proximity to urban centers, such as São Paulo city, one of the most important economic activities in São Vicente is tourism. Five beaches belong to São Vicente: Itararé, Milionários, Ilha Porchat, Divisa, and Gonzaguinha beach. The last ones are the most affected by organic contamination, while Ilha Porchat shows the lowest levels of pollution (CETESB 2008). The worst bathing water quality at Gonzaguinha beach is associated with its geographical configuration and the high inflow of organic load brought by the effluents discharged at Mar Pequeno, São Vicente Channel and the adjacent hydrographical system, which ultimately reach this beach (CETESB 2006). Since Gonzaguinha Beach is located in a sheltered area, dispersal of the microorganisms is limited, allowing bacteria and other microorganisms to be accumulate in areas used by beach goers. On the other hand, Ilha Porchat Beach is located in a less sheltered area, where circulation processes and the constant renewal of the water facilitate dispersal of the pollutants. These processes
Environ Monit Assess (2015) 187:342
Page 3 of 10 342
Fig. 1 Map of Brazil showing the location of the coast of the São Paulo State (left above) and the study areas at São Vicente (right): Gonzaguinha and Ilha Prochat Beaches. (Source:https://maps.google.com.br/maps)
favor the dispersal of microorganism seaward and provide better bathing water quality. In São Vicente, there is a sewage treatment plant in operation, which is responsible for treating the total collected sewage (56.6 %). Part of the collected sewage is send to the submarine outfall of Santos city (CETESB 2006). After passing by a bars screen to remove solid residuals and after chlorination, it is release 4 km away from the coast, in the central part of Santos Bay. During summer, the treatment of domestic sewage is not sufficient, since population in these areas can dramatically increase, reaching twice the number of local residents (CETESB 2006). Thus, there is an increase of domestic sewage load, affecting the quality of marine recreational water in the region (Sato et al. 2005; CETESB 2006). Besides the increase of domestic sewage discharge in seawaters, the urban water drainage reaching beaches, litter brought by bathers and tides, as well as the presence of animals on the beach, can also be pointed out as noteworthy problems in the region that are intensified during the holiday season.
Sample collection Seawater and sand samples used in the microbiological analyses were collect during summer (2012), in five consecutive days at Ilha Porchat and Gonzaguinha beaches, in São Vicente. Water samples were collected in 6 points of Gonzaguinha and 4 points of Ilha Porchat at 1 m isobath with sterile containers for 5 consecutive days, totaling 20 samples of water from Ilha Porchat and 30 samples from Gonzaguinha. Following, samples were maintain under refrigeration and analyzed within 6 h. Concurrently, sand samples were collected on two distinct zones of the beaches: dry zone, the area not flooded by seawater and correspondent to the section that is used most frequently by beach users, and wet zone, the area that is influenced by the tides. The sampling points were distributed in 6 points in Gonzaguinha beach and 4 points in Ilha Porchat, for 5 consecutive days, totaling 100 samples. Approximately 200 g of sand were collected using a sterile spatula and then transported in
342
Environ Monit Assess (2015) 187:342
Page 4 of 10
sterile bags under refrigeration to the Marine Microbiology laboratory of the São Paulo State University. Beaches sampled in this study are evaluated by Beach Quality Monitoring Program, which is perform by CETESB. The program conforms with the Brazilian legislation on bathing water quality (CONAMA Resolution n° 274/2000), and complies the evaluation of recreational seawater quality from beaches of 15 cities in São Paulo State, with weekly sampling (on Sundays) during low tides at 1 m isobath. According to the Brazilian legislation, E. coli density levels higher than 800 UFC 100 mL−1, in more than 20 % of five consecutive samples in sea water, or values higher than 2000 UFC 100 mL−1 in one sample are used to classify beaches as unsuitable for swimming and bathing.
Enumeration of Escherichia coli Densities of E. coli in seawater and sand samples were determined using the membrane filter technique (APHA 2012). In the laboratory, sand samples were weighed, added to sterile distilled water (1:10), and agitated twice in a Kline agitator for 10 min to wash and extract the bacteria from the samples. Volumes of 50, 10, 5, and 1 mL from the supernatant of the sand wash were filter in 0.45-μm-pore size membrane. These membranes were deposited in plates containing modified m-Tec Agar culture media and incubated at 37 °C for 24/ 48 h. Following this period, colonies with purple coloration were counted as E. coli. From each site and substrate, ten colonies presumptive counted as E. coli were biochemically confirmed by Gram staining, Cytochrome Oxidase test, Indole production, Methyl Red, Voges–Proskauer and Simmons Citrate tests (IMViC tests). To analyze the bacterial densities in seawater, sample volumes of 100, 50, 25, 10, and 5 mL were filtered. Culture media, incubation, counting, and confirmation procedures were the same as for the sand samples. Mean density and standard deviation of bacteria densities in the water and sand samples were obtain from bacterial counts. Mean densities of bacteria E. coli in seawater were express in colony-forming unities (CFU) per 100 mL and on sand in CFU per 100 g. After confirmation, strains were randomly selected from the plates to perform the susceptibility test being 246 from Ilha Porchat (water=84, wet sand=80, and dry
sand=82) and 210 from Gonzaguinha (water=80, wet sand=64, and dry sand=66) totaling 456 strains tested. Statistical analysis Pearson’s correlation coefficient (at 5 % significance level) was apply to E. coli density values from seawater, dry sand, and wet sand. Antimicrobial susceptibility testing The microbial sensitivity tests were carry out by the agar disc diffusion method from Kirby-Bauer, using MüllerHinton Agar and following the National Committee for Clinical Laboratory Standards (NCCLS 2003). We tested 456 colonies, randomly selected from the isolates, for the following antimicrobials: amoxicillin+clavulanic acid 30 μg, ampicillin 10 μg, erythromycin 15 μg, streptomycin 10 μg, gentamicin 10 μg, rifampicin 5 μg, tetracycline 30 μg, vancomycin 30 μg, and penicillin 10 μg. Strains of E. coli ATCC 25922 and E. coli ATCC 35218 were used as control for the results following the criteria established by the National Committee for Clinical Laboratory Standards (NCCLS 2004)
Results Enumeration of Escherichia coli Water According to the Brazilian legislation, E. coli densities above 800 CFU 100 mL−1 in two or more samples during a period of 5 weeks or values above 2000 CFU 100 mL−1 at the preceding sampling characterize the beach as inappropriate for primary contact recreation. Density values of E. coli, obtained in the present study, for either Gonzaguinha Beach or Ilha Porchat Beach, indicated that the former remained inappropriate for bathing during most of the period studied, while the latter was consider one of the best beaches in São Vicente. These results corroborate with the data obtained by CETESB (2012). At Gonzaguinha Beach, mean density of E. coli in seawater was above 2000 CFU 100 mL−1 during two sampling days (02/21 and 02/23) and reached maximum value of 2640 CFU 100 mL−1 on 02/23 (Fig. 2). Density
Environ Monit Assess (2015) 187:342
Page 5 of 10 342
value below 2000 CFU 100 mL−1 was observed in the last day (460 CFU 100 mL−1) and on 02/22, the density was exactly 2000 CFU 100 mL−1. At Ilha Porchat Beach, however, densities of E. coli were below 2000 CFU 100 mL−1 during the entire study period (Fig. 2).
Antimicrobial susceptibility testing A total of 456 samples were tested for antimicrobial susceptibility of which 56.6 % demonstrated resistance to at least one antibiotic. Water
Sand
Among the colonies isolated of Gonzaguinha Beach waters, 67.5 % showed resistance to antimicrobials. Among these strains, the highest frequencies of resistance were found to ampicillin (22.5 %), streptomycin (15.0 %), and rifampicin (15.0 %) (Table 1). Moreover, strains resistant to erythromycin (7.5 %) and tetracycline (7.5 %) were also isolated in waters from this beach. Most of isolated strains in waters from Gonzaguinha Beach were resistant to one antibiotic (78.4 %). Double resistance was observed in 19.5 % of the strains and only 2,1 % demonstrated resistance to 3 antimicrobials. At Ilha Porchat Beach, 50.0 % of the isolated strains exhibited resistance to at least one antimicrobial, and the highest frequency of resistance was observed to ampicillin and amoxicillin + clavulanic acid (25.0 %). Moreover, 70 % of the isolated strains from water samples showed single resistance while 30 % of the strains were resistant to two antimicrobials. None of the colonies from this samples presented triple resistance to antimicrobials.
At Gonzaguinha Beach, all wet and dry sand samples showed lower densities of E. coli compared to seawater samples. In dry sand, densities varied between 200 and 1240 CFU 100 g−1; while in wet sand, the values ranged from 80 to 1340 CFU 100 g−1. At Ilha Porchat Beach, densities of E. coli were higher than in water for both wet and dry sand. In dry sand, values ranged between 200 and 1100 CFU 100 g−1, and in wet sand, between 200 and 940 CFU 100 g−1 (Fig. 3). On the first 3 days of sampling, wet sand showed higher densities of E. coli than dry sand at Gonzaguinha Beach, while densities in wet sand of Ilha Porchat Beach were lower than those in dry sand only on the first sampling day (02/21) (Fig. 3). Statistical analysis Based on the standard deviation of bacterial density at each sampling site, we confirmed that no significant differences occurred among replicates. Mean density of E. coli in seawater, wet sand, and dry sand, compared using a correlation analysis, indicated a positive correlation between bacterial density in seawater and wet sand for both beaches (Gonzaguinha; r = 0.6, Ilha Porchat; r=0.5). Nevertheless, no significant correlation was observed between bacterial densities in seawater and dry sand (Gonzaguinha; r=−0.9, Ilha Porchat; r=0).
The isolated colonies from Gonzaguinha Beach sands presented 58.46 % of strains resistant to antimicrobials. Between these strains, the highest frequencies of resistance (23.52 %) were related to streptomycin (16.15 %)
3000 2500
CFU 100ml -1
Fig. 2 Escherichia coli counts in water at Gonzaguinha and Ilha Porchat beaches during the study period in comparison to the Brazilian E. coli standards. Error bars indicate standard error of the mean. Lines indicate the limits established by Brazilian legislation
Sand
2000 Gonzaguinha Beach
1500
Ilha Porchat Beach
1000 500 0 20
21
22
23
Sampling days
24
342
Environ Monit Assess (2015) 187:342
Page 6 of 10
Fig. 3 Escherichia coli counts in dry and wet sands at Gonzaguinha and Ilha Porchat beaches during the study period. Error bars indicate standard error of the mean
and to ampicillin (20.77 %) (Table 1), with resistance to penicillin (3.08 %), gentamicin (7.69 %), and amoxicillin+clavulanic acid (10.77 %) been observed as well. At Ilha Porchat Beach, 53.09 % of the isolated strains from sand demonstrated resistance to antimicrobials, and the highest frequency of resistance occurred to ampicillin (32.72 %) and amoxicillin+clavulanic acid (20.37 %). On both Gonzaguinha and Ilha Porchat Beaches, resistance frequencies of the isolated strains were higher in wet sand (71.9 and 53.8 %, respectively) (Fig. 4) than in dry sand (45.5 and 52.5 %, respectively). Gonzaguinha Beach showed higher percentage of resistant strains in seawater (67.5 %) compared to the results from Ilha Porchat Beach (50 %), and sand samples showed similar results, 58.46 % at Gonzaguinha and 53.09 % at Ilha Porchat. In regard to the antibiotic groups, at Ilha Porchat Beach waters, strains resistant to three types of antibiotics from two different groups were isolated (Fig. 5)— betalactamics (50.0 %) and aminoglycosides (50.0 %), whereas at Gonzaguinha Beach, strains resistant to six antibiotics from five distinct groups—betalactamics (33.4 %), rifampicins and aminoglycosides (22.2 %), and macrolides and tetracyclines (11.1 %).
On sand samples from Ilha Porchat Beach, strains resistant to four types of antibiotics from the betalactamics group were isolated, while at Gonzaguinha Beach, strains resistant to two antibiotics belonging to betalactamics (66.7 %) and aminoglycoside (33.3 %) groups were observed.
Discussion Coastal cities attract many tourists because of the scenic beauty and recreational activities. Usually, people look for beaches for recreational practices, such as swimming, surfing, boat ride, and others. In this sense, it is prerogative that those beaches maintain monitoring programs to attest their sanitary quality. In Brazil, water quality monitoring program is perform by CETESB according to Conselho Nacional do Meio Ambiente (CONAMA) legislation. According to this legislation, for a beach to be considered suitable for swimming, E. coli densities must be up to 800 CFU 100 mL−1 in two or more samples during a period of 5 weeks or values below 2000 CFU 100 mL−1 at the preceding sampling (CONAMA 2000). However, there is no legislation regarding beach sands. This is a concern
Environ Monit Assess (2015) 187:342
Page 7 of 10 342
Table 1 Percentages of isolates exhibiting antimicrobial agent resistance, by location and type of sample. Antimicrobial agent
% Resistant species (Number) Ilha Porchat (n=246)
Gonzaguinha (n=210)
Water (n=84) Wet sand (n=80) Dry sand (n=82) Water (n=80) Wet sand (n=64) Dry sand (n=66) Rifampicin
0
0
0
15.0 (12)
0
0
Streptomycin
0
0
0
15.0 (12)
23.44 (15)
9.09 (6)
Ampicillin
25.0 (21)
36.25 (29)
29.27 (24)
22.5 (18)
23.44 (15)
18.18 (12)
Erytromyciin
0
0
0
7.5 (6)
0
0
Tetracycline
0
0
0
7.5 (6)
0
0
Gentamicin
0
0
0
0
6.25 (4)
9.09 (6)
Vancomycin
0
0
0
0
0
0
Amoxicillin+clavulanic acid 25.0 (21)
17.50 (14)
23.17 (19)
0
12.50 (8)
9.09 (6)
Penicillin
0
0
0
0
6.25 (4)
0
Amoxicillin
0
0
0
0
0
0
to public health, once that beachgoers spend more time in sand than in the seawater, and a widely variety of pathogens should be included for screening on different sand zones (Halliday and Gast 2011; Sabino et al. 2014). In the present study, Gonzaguinha beach showed higher densities of E. coli in both samples, sand and water, compared to Ilha Porchat’s samples. Indeed, according to CETESB’s report, Gonzaguinha beach have a sanitary quality lower than that of Ilha Porchat beach. It probably happens due to its location, since Gonzaguinha beach is located at the end of an estuary that receives a large amount of organic effluent and for its physiography (since Gonzaguinha beach is an enclosed bay). Such as the study of Koczura et al. (2012), were almost all strains samples were multiresistant, depending on the sampling site, most of samples in our study
were resistant and the higher percentage of resistant strains was observed in Gonzaguinha samples. Is well known that antibiotics have a direct biological influence on microorganisms and, in many cases, can act like persistent pollutants for its continuous release in the environment (Constanzo et al. 2005; Wellington et al. 2013), and more than 30 antibiotic substances have been
Fig. 4 Percentage of antibiotic-resistant Escherichia coli isolated from wet and dry sands in Gonzaguinha and Ilha Porchat beaches
Fig. 5 Distribution of resistance frequency by antibiotic groups in water and sands from Gonzaguinha and Ilha Porchat beaches
342
Page 8 of 10
found in sewage (Kamper 2008), including tetracyclines and betalactamics (Oberlé et al. 2012). Antibiotics are a major concern nowadays since they have a wide variety of use. For its use in public health, prevention and treatment in animals, plant infection, and others, there is a great release along with domestic effluents in the environment for about 30 to 90 % of an administered dose of most antibiotics to humans and animals are excreted in the urine as an active substance (Rang et al. 1999; Kumarasamy et al. 2010). The major concern as to the massive use of antibiotics relates to its potential to accelerate widespread bacteria resistance. Constanzo et al. (2005) found strains of E. coli, isolated from aquatic environment, resistant to a least two of six antibiotics used. The same authors also demonstrated that antibiotics were detected in effluent entering receiving waters and 500 m from the source showing its power of dispersion. Galvin et al. (2010) reported that antimicrobialresistant E. coli survived the wastewater treatment process of a modern secondary treatment facility in Ireland, showing the importance of this specie on the spread of antibiotic resistance genes. In Europe, findings from SENTRY study showed a significant increase in antibiotic resistance for E. coli (11.6 %) in 2008 compared with 2004–2006 (
Densities and antimicrobial resistance of Escherichia coli isolated from marine waters and beach sands. Vanessa da Costa Andrade & Bruna Del Busso Zampieri & Eliete Rodrigues Ballesteros & Aline Bartelochi Pinto & Ana Julia Fernandes Cardoso de Oliveira
Received: 24 March 2015 / Accepted: 27 April 2015 # Springer International Publishing Switzerland 2015
Abstract Bacterial resistance is a rising problem all over the world. Many studies have showed that beach sands can contain higher concentration of microorganisms and represent a risk to public health. This paper aims to evaluate the densities and resistance to antimicrobials of Escherichia coli strains, isolated from seawater and samples. The hypothesis is that microorganisms show higher densities in contaminated beach sands and more antimicrobial resistance than the water column. Density, distribution, and antimicrobial resistance of bacteria E. coli were evaluate in seawater and sands from two recreational beaches with different levels of pollution. At the beach with higher degree of pollution (Gonzaguinha), water samples presented the highest densities of E. coli; however, higher frequency of resistant strains was observe in wet sand (71.9 %). Resistance to a larger number of antimicrobial groups was observe in water (betalactamics, aminoglycosides, macrolides, rifampicins, and tetracyclines) and sand (betagalactamics and aminoglycosids). In water samples, highest frequencies of resistance were obtain against ampicilin (22.5 %), streptomycin (15.0 %), and rifampicin (15.0 %), while in sand, the highest frequencies were observe in relation to ampicilin (36.25 %) and V. da Costa Andrade : B. Del Busso Zampieri : A. B. Pinto Campus de Rio Claro, Universidade Estadual Paulista, UNESP, Avenida 24A, 1515, 13506-900 Rio Claro, SP, Brazil E. R. Ballesteros : A. J. Fernandes Cardoso de Oliveira (*) Laboratory of Marine Microbiology, Campus do Litoral Paulista, Universidade Estadual Paulista, UNESP, Praça Infante Dom Henrique, s/n, 11330-900 São Vicente, SP, Brazil e-mail: [email protected]
streptomycin (23.52 %). At the less polluted beach, Ilha Porchat, highest densities of E. coli and higher frequency of resistance were obtain in wet and dry sand (53.7 and 53.8 %, respectively) compared to water (50 %). Antimicrobial resistance in strains isolated from water and sand only occurred against betalactamics (ampicilin and amoxicilin plus clavulanic acid). The frequency and variability of bacterial resistance to antimicrobials in marine recreational waters and sands were related to the degree of fecal contamination in this environment. These results show that water and sands from beaches with a high index of fecal contamination of human origin may be potential sources of contamination by pathogens and contribute to the dissemination of bacterial resistance. Keywords Escherichia coli . Antimicrobial resistance . Marine recreational beaches . Sand . Water
Introduction Bacterial resistance is a rising problem all over the world, generated by processes of selection that result from the massive use of antibiotics (Frieden et al. 1993). The sewage discharge of domestic effluents and from strongly selective sources such as hospitals, industries, aquiculture, and among others has led to an increase in the distribution and frequency of bacterial resistance genes in several environments, including fresh and marine waters (Schwartz et al. 2003).
342
Environ Monit Assess (2015) 187:342
Page 2 of 10
In coastal regions, the high development of tourism, especially because of recreational activities such as diving, swimming, aquatic sports, and fishery, generates several impacts of anthropogenic origin. One of these impacts is the high densities of pathogens that are found in water column and sediments, including beach sands. These pathogens come from the sewage discharge which may be partially treated or in natura and carry several microorganisms such as bacteria, viruses, and protozoans (WHO 1998). When these pathogens reach waters and sands, they expose beach goers to the possibility of acquiring diseases, although the incidence of these diseases depends upon a variety of factors, such as the level of pollution in the water, the nature and length of exposure, and the immunological condition of the beachgoer (Bartram and Rees 2000). Many studies have showed that beach sands can contain higher concentrations of microorganisms, including pathogenic bacteria (Mendes et al. 1993; Papadakis et al. 1997; Alm et al. 2003; Whitman and Nevers 2003; Mudryk 2005; Oliveira and Pinhata 2008; Whitman et al. 2014), than the water column, once bacteria find in this environment favorable conditions of nutrients, protection against sunlight and predation by protozoans (Davies et al. 1995; Davies-Colley et al. 1999; Davies and Bavor 2000; Sinton et al. 1994; Pianetti et al. 2004; Villar et al. 1999). Thus, beach sands contaminated by pathogenic microorganisms can represent a higher risk to the visitors’ health than the contact with the marine recreational waters. In addition, marine recreational waters and sands that receive domestic sewage can be potential sources of resistant bacteria, and these areas may contribute to the establishment of dissemination routes of microorganisms carrying antimicrobial resistance genes. This is because besides revealing intrinsic resistance to several antibiotics bacteria can also, through conjugation with bacteria from the same or different species, acquire resistance factors present in plasmid (Huycke et al. 1998; Kühn et al. 2000). The frequency and variability of resistant strains of bacteria to different antimicrobial have increased in the last time, becoming a great challenge for the treatment of infectious diseases (Arvanitidou et al. 2001). The presence in sands and recreational waters of bacterial strains, which show resistance to antibiotics, is relevant due to the relationship between waterborne disease and recreational activities. It is expected for sand beaches to show higher bacterial densities than the water
column, since it can act as a reservoir for pathogenic microorganisms and can collaborate to increase the dissemination of resistance genes, which represent a risk to public health specially because there are no parameters or monitoring programs for sands of recreational beaches on the current legislation. Therefore, the present study aimed to evaluate the densities and resistance to antimicrobials of Escherichia coli strains, which were isolated from seawater and sand samples from two recreational beaches under different levels of pollution. In addition, environment potentials as sources of dissemination of bacterial resistance were evaluate based on the correlation between quality of marine recreational waters, bacterial habitat, frequency, and type of resistance.
Materials and methods Study area São Vicente (23° 57′ S and 46° 23′ W) is located on the Santos Basin (BSantos Lowland^), south coast of São Paulo State (Fig. 1). The Santos Basin is so named due to its geography, since it borders the Serra do Mar (BSea Highlands^). It occupies an area of 2373 km2, with 161 km2 of beaches. Due to the heterogeneity of its beaches and the proximity to urban centers, such as São Paulo city, one of the most important economic activities in São Vicente is tourism. Five beaches belong to São Vicente: Itararé, Milionários, Ilha Porchat, Divisa, and Gonzaguinha beach. The last ones are the most affected by organic contamination, while Ilha Porchat shows the lowest levels of pollution (CETESB 2008). The worst bathing water quality at Gonzaguinha beach is associated with its geographical configuration and the high inflow of organic load brought by the effluents discharged at Mar Pequeno, São Vicente Channel and the adjacent hydrographical system, which ultimately reach this beach (CETESB 2006). Since Gonzaguinha Beach is located in a sheltered area, dispersal of the microorganisms is limited, allowing bacteria and other microorganisms to be accumulate in areas used by beach goers. On the other hand, Ilha Porchat Beach is located in a less sheltered area, where circulation processes and the constant renewal of the water facilitate dispersal of the pollutants. These processes
Environ Monit Assess (2015) 187:342
Page 3 of 10 342
Fig. 1 Map of Brazil showing the location of the coast of the São Paulo State (left above) and the study areas at São Vicente (right): Gonzaguinha and Ilha Prochat Beaches. (Source:https://maps.google.com.br/maps)
favor the dispersal of microorganism seaward and provide better bathing water quality. In São Vicente, there is a sewage treatment plant in operation, which is responsible for treating the total collected sewage (56.6 %). Part of the collected sewage is send to the submarine outfall of Santos city (CETESB 2006). After passing by a bars screen to remove solid residuals and after chlorination, it is release 4 km away from the coast, in the central part of Santos Bay. During summer, the treatment of domestic sewage is not sufficient, since population in these areas can dramatically increase, reaching twice the number of local residents (CETESB 2006). Thus, there is an increase of domestic sewage load, affecting the quality of marine recreational water in the region (Sato et al. 2005; CETESB 2006). Besides the increase of domestic sewage discharge in seawaters, the urban water drainage reaching beaches, litter brought by bathers and tides, as well as the presence of animals on the beach, can also be pointed out as noteworthy problems in the region that are intensified during the holiday season.
Sample collection Seawater and sand samples used in the microbiological analyses were collect during summer (2012), in five consecutive days at Ilha Porchat and Gonzaguinha beaches, in São Vicente. Water samples were collected in 6 points of Gonzaguinha and 4 points of Ilha Porchat at 1 m isobath with sterile containers for 5 consecutive days, totaling 20 samples of water from Ilha Porchat and 30 samples from Gonzaguinha. Following, samples were maintain under refrigeration and analyzed within 6 h. Concurrently, sand samples were collected on two distinct zones of the beaches: dry zone, the area not flooded by seawater and correspondent to the section that is used most frequently by beach users, and wet zone, the area that is influenced by the tides. The sampling points were distributed in 6 points in Gonzaguinha beach and 4 points in Ilha Porchat, for 5 consecutive days, totaling 100 samples. Approximately 200 g of sand were collected using a sterile spatula and then transported in
342
Environ Monit Assess (2015) 187:342
Page 4 of 10
sterile bags under refrigeration to the Marine Microbiology laboratory of the São Paulo State University. Beaches sampled in this study are evaluated by Beach Quality Monitoring Program, which is perform by CETESB. The program conforms with the Brazilian legislation on bathing water quality (CONAMA Resolution n° 274/2000), and complies the evaluation of recreational seawater quality from beaches of 15 cities in São Paulo State, with weekly sampling (on Sundays) during low tides at 1 m isobath. According to the Brazilian legislation, E. coli density levels higher than 800 UFC 100 mL−1, in more than 20 % of five consecutive samples in sea water, or values higher than 2000 UFC 100 mL−1 in one sample are used to classify beaches as unsuitable for swimming and bathing.
Enumeration of Escherichia coli Densities of E. coli in seawater and sand samples were determined using the membrane filter technique (APHA 2012). In the laboratory, sand samples were weighed, added to sterile distilled water (1:10), and agitated twice in a Kline agitator for 10 min to wash and extract the bacteria from the samples. Volumes of 50, 10, 5, and 1 mL from the supernatant of the sand wash were filter in 0.45-μm-pore size membrane. These membranes were deposited in plates containing modified m-Tec Agar culture media and incubated at 37 °C for 24/ 48 h. Following this period, colonies with purple coloration were counted as E. coli. From each site and substrate, ten colonies presumptive counted as E. coli were biochemically confirmed by Gram staining, Cytochrome Oxidase test, Indole production, Methyl Red, Voges–Proskauer and Simmons Citrate tests (IMViC tests). To analyze the bacterial densities in seawater, sample volumes of 100, 50, 25, 10, and 5 mL were filtered. Culture media, incubation, counting, and confirmation procedures were the same as for the sand samples. Mean density and standard deviation of bacteria densities in the water and sand samples were obtain from bacterial counts. Mean densities of bacteria E. coli in seawater were express in colony-forming unities (CFU) per 100 mL and on sand in CFU per 100 g. After confirmation, strains were randomly selected from the plates to perform the susceptibility test being 246 from Ilha Porchat (water=84, wet sand=80, and dry
sand=82) and 210 from Gonzaguinha (water=80, wet sand=64, and dry sand=66) totaling 456 strains tested. Statistical analysis Pearson’s correlation coefficient (at 5 % significance level) was apply to E. coli density values from seawater, dry sand, and wet sand. Antimicrobial susceptibility testing The microbial sensitivity tests were carry out by the agar disc diffusion method from Kirby-Bauer, using MüllerHinton Agar and following the National Committee for Clinical Laboratory Standards (NCCLS 2003). We tested 456 colonies, randomly selected from the isolates, for the following antimicrobials: amoxicillin+clavulanic acid 30 μg, ampicillin 10 μg, erythromycin 15 μg, streptomycin 10 μg, gentamicin 10 μg, rifampicin 5 μg, tetracycline 30 μg, vancomycin 30 μg, and penicillin 10 μg. Strains of E. coli ATCC 25922 and E. coli ATCC 35218 were used as control for the results following the criteria established by the National Committee for Clinical Laboratory Standards (NCCLS 2004)
Results Enumeration of Escherichia coli Water According to the Brazilian legislation, E. coli densities above 800 CFU 100 mL−1 in two or more samples during a period of 5 weeks or values above 2000 CFU 100 mL−1 at the preceding sampling characterize the beach as inappropriate for primary contact recreation. Density values of E. coli, obtained in the present study, for either Gonzaguinha Beach or Ilha Porchat Beach, indicated that the former remained inappropriate for bathing during most of the period studied, while the latter was consider one of the best beaches in São Vicente. These results corroborate with the data obtained by CETESB (2012). At Gonzaguinha Beach, mean density of E. coli in seawater was above 2000 CFU 100 mL−1 during two sampling days (02/21 and 02/23) and reached maximum value of 2640 CFU 100 mL−1 on 02/23 (Fig. 2). Density
Environ Monit Assess (2015) 187:342
Page 5 of 10 342
value below 2000 CFU 100 mL−1 was observed in the last day (460 CFU 100 mL−1) and on 02/22, the density was exactly 2000 CFU 100 mL−1. At Ilha Porchat Beach, however, densities of E. coli were below 2000 CFU 100 mL−1 during the entire study period (Fig. 2).
Antimicrobial susceptibility testing A total of 456 samples were tested for antimicrobial susceptibility of which 56.6 % demonstrated resistance to at least one antibiotic. Water
Sand
Among the colonies isolated of Gonzaguinha Beach waters, 67.5 % showed resistance to antimicrobials. Among these strains, the highest frequencies of resistance were found to ampicillin (22.5 %), streptomycin (15.0 %), and rifampicin (15.0 %) (Table 1). Moreover, strains resistant to erythromycin (7.5 %) and tetracycline (7.5 %) were also isolated in waters from this beach. Most of isolated strains in waters from Gonzaguinha Beach were resistant to one antibiotic (78.4 %). Double resistance was observed in 19.5 % of the strains and only 2,1 % demonstrated resistance to 3 antimicrobials. At Ilha Porchat Beach, 50.0 % of the isolated strains exhibited resistance to at least one antimicrobial, and the highest frequency of resistance was observed to ampicillin and amoxicillin + clavulanic acid (25.0 %). Moreover, 70 % of the isolated strains from water samples showed single resistance while 30 % of the strains were resistant to two antimicrobials. None of the colonies from this samples presented triple resistance to antimicrobials.
At Gonzaguinha Beach, all wet and dry sand samples showed lower densities of E. coli compared to seawater samples. In dry sand, densities varied between 200 and 1240 CFU 100 g−1; while in wet sand, the values ranged from 80 to 1340 CFU 100 g−1. At Ilha Porchat Beach, densities of E. coli were higher than in water for both wet and dry sand. In dry sand, values ranged between 200 and 1100 CFU 100 g−1, and in wet sand, between 200 and 940 CFU 100 g−1 (Fig. 3). On the first 3 days of sampling, wet sand showed higher densities of E. coli than dry sand at Gonzaguinha Beach, while densities in wet sand of Ilha Porchat Beach were lower than those in dry sand only on the first sampling day (02/21) (Fig. 3). Statistical analysis Based on the standard deviation of bacterial density at each sampling site, we confirmed that no significant differences occurred among replicates. Mean density of E. coli in seawater, wet sand, and dry sand, compared using a correlation analysis, indicated a positive correlation between bacterial density in seawater and wet sand for both beaches (Gonzaguinha; r = 0.6, Ilha Porchat; r=0.5). Nevertheless, no significant correlation was observed between bacterial densities in seawater and dry sand (Gonzaguinha; r=−0.9, Ilha Porchat; r=0).
The isolated colonies from Gonzaguinha Beach sands presented 58.46 % of strains resistant to antimicrobials. Between these strains, the highest frequencies of resistance (23.52 %) were related to streptomycin (16.15 %)
3000 2500
CFU 100ml -1
Fig. 2 Escherichia coli counts in water at Gonzaguinha and Ilha Porchat beaches during the study period in comparison to the Brazilian E. coli standards. Error bars indicate standard error of the mean. Lines indicate the limits established by Brazilian legislation
Sand
2000 Gonzaguinha Beach
1500
Ilha Porchat Beach
1000 500 0 20
21
22
23
Sampling days
24
342
Environ Monit Assess (2015) 187:342
Page 6 of 10
Fig. 3 Escherichia coli counts in dry and wet sands at Gonzaguinha and Ilha Porchat beaches during the study period. Error bars indicate standard error of the mean
and to ampicillin (20.77 %) (Table 1), with resistance to penicillin (3.08 %), gentamicin (7.69 %), and amoxicillin+clavulanic acid (10.77 %) been observed as well. At Ilha Porchat Beach, 53.09 % of the isolated strains from sand demonstrated resistance to antimicrobials, and the highest frequency of resistance occurred to ampicillin (32.72 %) and amoxicillin+clavulanic acid (20.37 %). On both Gonzaguinha and Ilha Porchat Beaches, resistance frequencies of the isolated strains were higher in wet sand (71.9 and 53.8 %, respectively) (Fig. 4) than in dry sand (45.5 and 52.5 %, respectively). Gonzaguinha Beach showed higher percentage of resistant strains in seawater (67.5 %) compared to the results from Ilha Porchat Beach (50 %), and sand samples showed similar results, 58.46 % at Gonzaguinha and 53.09 % at Ilha Porchat. In regard to the antibiotic groups, at Ilha Porchat Beach waters, strains resistant to three types of antibiotics from two different groups were isolated (Fig. 5)— betalactamics (50.0 %) and aminoglycosides (50.0 %), whereas at Gonzaguinha Beach, strains resistant to six antibiotics from five distinct groups—betalactamics (33.4 %), rifampicins and aminoglycosides (22.2 %), and macrolides and tetracyclines (11.1 %).
On sand samples from Ilha Porchat Beach, strains resistant to four types of antibiotics from the betalactamics group were isolated, while at Gonzaguinha Beach, strains resistant to two antibiotics belonging to betalactamics (66.7 %) and aminoglycoside (33.3 %) groups were observed.
Discussion Coastal cities attract many tourists because of the scenic beauty and recreational activities. Usually, people look for beaches for recreational practices, such as swimming, surfing, boat ride, and others. In this sense, it is prerogative that those beaches maintain monitoring programs to attest their sanitary quality. In Brazil, water quality monitoring program is perform by CETESB according to Conselho Nacional do Meio Ambiente (CONAMA) legislation. According to this legislation, for a beach to be considered suitable for swimming, E. coli densities must be up to 800 CFU 100 mL−1 in two or more samples during a period of 5 weeks or values below 2000 CFU 100 mL−1 at the preceding sampling (CONAMA 2000). However, there is no legislation regarding beach sands. This is a concern
Environ Monit Assess (2015) 187:342
Page 7 of 10 342
Table 1 Percentages of isolates exhibiting antimicrobial agent resistance, by location and type of sample. Antimicrobial agent
% Resistant species (Number) Ilha Porchat (n=246)
Gonzaguinha (n=210)
Water (n=84) Wet sand (n=80) Dry sand (n=82) Water (n=80) Wet sand (n=64) Dry sand (n=66) Rifampicin
0
0
0
15.0 (12)
0
0
Streptomycin
0
0
0
15.0 (12)
23.44 (15)
9.09 (6)
Ampicillin
25.0 (21)
36.25 (29)
29.27 (24)
22.5 (18)
23.44 (15)
18.18 (12)
Erytromyciin
0
0
0
7.5 (6)
0
0
Tetracycline
0
0
0
7.5 (6)
0
0
Gentamicin
0
0
0
0
6.25 (4)
9.09 (6)
Vancomycin
0
0
0
0
0
0
Amoxicillin+clavulanic acid 25.0 (21)
17.50 (14)
23.17 (19)
0
12.50 (8)
9.09 (6)
Penicillin
0
0
0
0
6.25 (4)
0
Amoxicillin
0
0
0
0
0
0
to public health, once that beachgoers spend more time in sand than in the seawater, and a widely variety of pathogens should be included for screening on different sand zones (Halliday and Gast 2011; Sabino et al. 2014). In the present study, Gonzaguinha beach showed higher densities of E. coli in both samples, sand and water, compared to Ilha Porchat’s samples. Indeed, according to CETESB’s report, Gonzaguinha beach have a sanitary quality lower than that of Ilha Porchat beach. It probably happens due to its location, since Gonzaguinha beach is located at the end of an estuary that receives a large amount of organic effluent and for its physiography (since Gonzaguinha beach is an enclosed bay). Such as the study of Koczura et al. (2012), were almost all strains samples were multiresistant, depending on the sampling site, most of samples in our study
were resistant and the higher percentage of resistant strains was observed in Gonzaguinha samples. Is well known that antibiotics have a direct biological influence on microorganisms and, in many cases, can act like persistent pollutants for its continuous release in the environment (Constanzo et al. 2005; Wellington et al. 2013), and more than 30 antibiotic substances have been
Fig. 4 Percentage of antibiotic-resistant Escherichia coli isolated from wet and dry sands in Gonzaguinha and Ilha Porchat beaches
Fig. 5 Distribution of resistance frequency by antibiotic groups in water and sands from Gonzaguinha and Ilha Porchat beaches
342
Page 8 of 10
found in sewage (Kamper 2008), including tetracyclines and betalactamics (Oberlé et al. 2012). Antibiotics are a major concern nowadays since they have a wide variety of use. For its use in public health, prevention and treatment in animals, plant infection, and others, there is a great release along with domestic effluents in the environment for about 30 to 90 % of an administered dose of most antibiotics to humans and animals are excreted in the urine as an active substance (Rang et al. 1999; Kumarasamy et al. 2010). The major concern as to the massive use of antibiotics relates to its potential to accelerate widespread bacteria resistance. Constanzo et al. (2005) found strains of E. coli, isolated from aquatic environment, resistant to a least two of six antibiotics used. The same authors also demonstrated that antibiotics were detected in effluent entering receiving waters and 500 m from the source showing its power of dispersion. Galvin et al. (2010) reported that antimicrobialresistant E. coli survived the wastewater treatment process of a modern secondary treatment facility in Ireland, showing the importance of this specie on the spread of antibiotic resistance genes. In Europe, findings from SENTRY study showed a significant increase in antibiotic resistance for E. coli (11.6 %) in 2008 compared with 2004–2006 (
