Analysis of Antibiotic Resistance in Bacteria Isolated from the Surface Microlayer and Underlying Water of an Estuarine Environment

Abstract

We compared the prevalence of cultivable antibiotic-resistant bacteria and resistance genes in the surface microlayer (SML) and underlying waters (UW) of an estuary. Prevalence of resistant bacteria was determined in antibiotic-supplemented agar. Bacterial isolates from the UW (n=91) and SML (n=80), selected in media without antibiotic, were characterized concerning susceptibility against nine antibiotics. The presence of genes bla_TEM, bla_OXA-B, bla_SHV, bla_IMP, tet(A), tet(B), tet(E), tet(M), cat, sul1, sul2, sul3, aadA, IntI1, IntI2, and IntI3 was assessed by PCR. The variable regions of integrons were sequenced. Ampicillin- and streptomycin-resistant bacteria were significantly more prevalent in SML. Resistance levels among the bacterial collections were generally low, preventing detection of significant differences between SML and UW. The tet(E) gene was detected in two Aeromonas isolates and tet(M) was detected in a Pseudomonas isolate. Gene sul1 was amplified from three Aeromonas isolates. Prevalence of intI genes was 2.11%. Cassette arrays contained genes encoding resistance to aminoglycosides and chloramphenicol. A higher prevalence of antibiotic-resistant bacteria in the SML, although only detectable when bacteria were selected in antibiotic-supplemented agar, suggests that SML conditions select for antibiotic resistance. Results also showed that antibiotic resistance was uncommon among estuarine bacteria and the resistance mechanisms are probably predominantly intrinsic.

Introduction

Alarge fraction of the Earth's surface is covered by water. The interface between hydrosphere and atmosphere is represented by a thin gelatinous surface film designated as the sea-surface microlayer (SML), where many ecological processes take place.²⁷

The SML is distinct from underlying waters (UW),²⁷ since higher fluctuations in salinity and temperature occur and the exposure to ultraviolet radiation is also greater than in UW. Also, in the SML, higher amounts of organic compounds (proteins, lipids, and polysaccharides) and anthropogenic contaminants (hydrocarbons, metalloids, pesticides, and antibiotic residues) accumulate.^7,11,10 Moreover, even under greater selective pressures, the SML harbors a more abundant and active bacterial community, the bacterioneuston, comparing to UW.^2,16,37

The gelatinous nature of the SML combined with strong selective pressures and high bacterial densities stimulate the production of secondary metabolites.²⁹ In fact, for example, antibiotic producers have been frequently retrieved from the SML.¹⁴

The production of antimicrobials, together with persistence of anthropogenic-derived antibiotic residues in the SML³⁹ may select for and foment the dissemination of antibiotic resistance genes (ARGs). In addition, bacterial growth in biofilms, expected to occur in the SML, might also promote horizontal gene transfer and resistance to antimicrobial agents.^20,10,35 Thus, the SML may be a natural reservoir of ARGs and an incubator of new gene combinations.

However, most studies concerning antibiotic bacterial resistance in estuarine and coastal water systems focused on bulk water,^18,25,13 and consequently the relevance of the SML in developing, selecting, and spreading ARGs has been almost ignored.

Estuarine systems are strongly influenced by anthropogenic activities accumulating high levels of pollutants and often becoming eutrophicated,²³ and thus gathering ideal conditions for the establishment of a distinct bacterioneuston community.

In a previous study, we have focused our attention on the phylogenetic composition of the bacterioneuston and bacterioplankton communities from a highly polluted estuarine system (Ria de Aveiro, Portugal).² The aims of the present study are (1) to compare the prevalence of cultivable antibiotic-resistant bacteria in the SML and UW in Ria de Aveiro; (2) to compare antibiotic resistance profiles between isolates previously obtained from SML and UW samples; (3) to assess and compare the occurrence and diversity of ARGs and integrons in the same isolates.

Material and Methods

Heterotrophic plate counts on antibiotic-selective media

SML and UW samples were collected on three occasions (C1, C2, and C3) in 2008 at three sites named Cais do Chegado (CC), Costa Nova (CN), and Cais do Sporting (CS) within the estuary Ria de Aveiro (Supplementary Fig. S1; Supplementary Data are available online at www.liebertpub.com/mdr) as described by Azevedo et al.² Briefly, SML samples (250 ml) represented by a water layer of 60–100 μm were collected with an acrylic or a glass plate.² UW (1 L) were collected at ∼0.4-m depth in 2-L autoclaved bottles.

From each water sample, 100 μl were directly spread onto EA and GSP plates without added antibiotics. Also, 1 ml of each sample was serially diluted in 0.9% NaCl and 100 μl of each dilution was spread in the same media to obtain colony counts between 30 and 300 per plate. Total heterotrophic plate counts (HPC) were determined using these plates.

Resistant bacteria were cultivated by the same spreading procedure in the same culture media supplemented with ampicillin (50 μg/μl), tetracycline (20 μg/μl), and streptomycin (10 μg/μl). These concentrations are above the breakpoints defined by the Clinical and Laboratory Standards Institute (CLSI)⁸ for most phylogenetic groups and were chosen to select for highly resistant bacteria. Colony-forming units (CFU/ml) were counted in triplicate (three independent plates) after 4 days of incubation at 22°C in the dark.

Bacterial isolates

A total of 352 bacterial isolates were obtained from SML and UW within the estuary Ria de Aveiro in a previous work, corresponding to 10 to 20 isolates selected from each sample.² From those, 13 isolates representing 12 genera were impossible to recover in successive cultures and were excluded from the current study. Isolates were typed using REP-PCR with primers REP-1R and REP-2I as described previously (Versalovic et al.).³⁸ PCR reactions were carried out with Taq polymerase, nucleotides, and buffers from Promega. The PCR reaction mixtures (25 μl) contained 1×PCR buffer, 200 μM of each nucleotide, 3 mM MgCl₂, 5% dimethylsulfoxide, 7.5 pmol of each primer, 1 U of Taq polymerase, and 50–100 ng purified DNA. Amplification was carried out as follows: initial denaturation for 7 min at 95°C, followed by 30 cycles of denaturation for 1 min at 94°C, annealing for 1 min at 40°C and extension for 8 min at 65°C and by a final extension of 16 min at 65°C. The products were electrophoresed for 90 min under a constant voltage of 80 V on a 1.5% (w/v) agarose gel containing 0.5×TAE (20 mM Tris–acetate, pH 7.4, 10 mM sodium acetate, 0.5 mM Na₂EDTA) and DNA markers purchased from MBI Fermentas. The gel images were acquired using a Molecular Imager FX system (Bio-Rad Laboratories) and analyzed using the software package GelCompar 4.0 (Applied Maths). From 339 isolates, a total of 171 isolates displaying different REP profiles were identified based on the partial sequence of the 16S rRNA gene.² Isolates (80 isolates from SML and 91 from UW) affiliated with 34 genera within six classes (Table 1) and were included in the present study.

Table 1.

Summary of Bacteria Isolated from Surface Microlayer and Underlying Waters Samples

	Phylogenetic group		Number of isolates ^a
	Class	Genera	SML	UW
Gram-positive	Actinobacteria	Agrococcus	0	6
		Arthrobacter	1	1
		Brevibacterium	0	1
		Corynebacterium	1	0
		Kocuria	0	2
		Microbacterium	0	1
		Micrococcus	1	0
		Unclassified Micrococcaceae	2	5
	Bacilli	Aerococcus	1	0
		Bacillus	3	1
		Exiguobacterium	0	1
		Planococcus	2	0
		Staphylococcus	1	0
Gram-negative	Alfaproteobacteria	Brevundimonas	0	3
		Erythrobacter	1	1
		Agrobacterium	1	0
		Pseudorhodobacter	2	2
	Betaproteobacteria	Alcaligenes	2	1
		Comamonas	1	0
		Delftia	0	1
	Gamaproteobacteria	Aeromonas	10	7
		Shewanella	6	8
		Pseudoalteromonas	0	1
		Rheinheimera	2	4
		Erwinia	2	0
		Halomonas	2	0
		Marinomonas	1	2
		Acinetobacter	5	8
		Pseudomonas	10	7
		Psychrobacter	15	13
		Vibrio	1	7
		Lysobacter	0	1
		Stenotrophomonas	2	1
		Unclassified Enterobacteriaceae	5	6

Isolates selected as representative of different REP profiles.

SML, surface microlayer; UW, underlying waters.

DNA extraction

Isolates were cultured in Luria-Bertani agar plates (Merck) and were incubated at 30°C overnight. The total DNA was obtained by resuspending two isolated colonies in 100 μl of buffer B1 (50 mM Tris-Cl pH 8.0; 50 mM EDTA pH 8.0; 0.5% Tween 20; 0.5% Triton X-100) and 100 μl of master-mix (1 ml of buffer B1; 2 mg/ml of Lysozyme; 4.5 mg/ml of Proteinase K: 7.5 U of RNAse A). Cell suspension was mixed vigorously and incubated for 30 min at 37°C. Following, 70 μl of B2 (3 M Guanidine Hydrochloride; 20% Tween 20) were added and the mixture was incubated for 30 min at 50°C. Subsequently, the DNA Extraction Kit (#K0513–MBI; Fermentas) was used according to the manufacturer's instructions.

Antimicrobial susceptibility testing

Antimicrobial susceptibility was tested by the agar disk diffusion method as recommended by the CLSI.⁸ Isolates representing different REP profiles (n=171) were tested for susceptibility to 9 antimicrobial agents representing six classes of antibiotics: (1) β-lactams—ampicillin (10 μg), imipenem (10 μg), cephalothin (30 μg); (2) aminoglycoside—streptomycin (10 μg), gentamicin (10 μg); (3) phenicols—chloramphenicol (30 μg); (4) tetracyclines—tetracycline (30 μg); (5) the combination sulfamethoxazole/trimethoprim (25 μg) and (6) quinolones—nalidixic acid (30 μg). Disks were purchased from Oxoid. Antimicrobial activities were determined as clear zones of inhibition around the antibiotic disks, after incubation at 30°C for 24 hr. Isolates were classified as sensitive, intermediate, or resistant taking into account the CLSI guidelines.⁸ Whenever genus-specific guidelines were not available, guidelines for the Enterobacteriaceae were used for Gram-negative strains and guidelines for Staphylococcus spp. were used for Gram-positive strains. Escherichia coli ATCC 25922 was used as a quality control strain.

Antimicrobial resistance genes

All isolates were tested by PCR for the presence of genes conferring resistance to β-lactams, tetracyclines, chloramphenicol, aminoglycoside, and sulfonamides by using primers and PCR conditions previously reported^18,19 (see also Table 2). Isolates carrying resistance genes were included in each PCR run as positive controls. The positive controls for amplification of the sulfonamide resistance genes sul1, sul2, and sul3 were as described previously.^18,9 A PCR mixture with no DNA added was used as a negative control. Gel electrophoresis was performed on 1.5% agarose gel and stained with ethidium bromide.

Table 2.

Primers and PCR Conditions

Primer pair	Target	Sequence (5′–3′)	Annealing temperature (°C)	Amplicon size (bp)	Control strain
TEM_F	bla_TEM	AAAGATGCTGAAGATCA	44	425	Klebsiella pneumoniae 6T¹⁸
TEM_R		TTTGGTATGGCTTCATTC
SHV_F	bla_SHV	GCGAAAGCCAGCTGTCGGGC	62	304	K. pneumoniae 2s¹⁸
SHV_R		GATTGGCGGCGCTGTTATCGC
OXA-B_F	bla_OXA-B	CAAGCCAAAGGCACGATAGTTG	56	561	Aeromonas sp. G.N1.15¹⁸
OXA-B_R		CTCAACCCATCCTACCCACC
IMPF	bla_IMP	GAATAGAGTGGATTAATTCTC	55	232	K. pneumoniae KP99c196¹⁸
IMPR		GGTTTAAYAAAACAACCACC
TetAF	tetA	GCTACATCCTGCTTGCCTTC	53	211	Escherichia coli M.I10.34¹⁹
TetAR		GCATAGATCGCCGTGAAGAG
TetBF	tetB	TCATTGCCGATACCACCTCAG	53	391	E. coli M.N1.616¹⁹
TetBR		CCAACCATCATGCTATTCCATCC
TetEF	tetE	ATGAACCGCACTGTGATGATG	53	744	Aeromonas sp. G.I10.2¹⁹
TetER		ACCGACCATTACGCCATCC
TetMF	tetM	GTGGACAAAGGTACAACGAG	55	406	E. coli M.I10.34¹⁹
TetMR		CGGTAAAGTTCGTCACACAC
aadA1_F	aadA1	TATCAGAGGTAGTTGGCGTCAT	54	485	Aeromonas sp. G.N1.15¹⁸
aadA1_R		AATGAAACCTTAACGCTATGGAAC
catF	Cat	CCTGCCACTCATCGCAGT	55	623	Aeromonas sp. G.N1.15¹⁸
catR		CCACCGTTGATATATCCC
sulF1	sul1	CTGAACGATATCCAAGGATTYCC	50	239	E. coli M.I10.40¹⁸
sulR1		AAAAATCCCATCCCCGGRTC
sul2-F	sul2	GCGCTCAAGGCAGATGGCATT	69	293	E. coli M.I10.40¹⁸
sul2-R		GCGTTTGATACCGGCACCCGT
sul3F	sul3	AAGAAGCCCATACCCGGRTC	50	236	K. pneumoniae (intI3+)³⁰
sul3R		ATTAATGATATTCAAGGTTTYCC
intI1F	Class 1 integrase gene	CCTCCCGCACGATGATC	55	280	Salmonella enterica ser Typhimurium (intI1+)³⁰
intI1R		TCCACGCATCGTCAGGC
intI2F	Class 2 integrase gene	TTATTGCTGGGATTAGGC	52	233	E. coli (intI2+)³⁰
intI2R		ACGGCTACCCTCTGTTATC
intI3F	Class 3 integrase gene	AGTGGGTGGCGAATGAGTG	50	600	K. pneumoniae (intI3+)³⁰
intI3R		TGTTCTTGTATCGGCAGGTG
5′-CS	Class 1 integron variable region	GGCATCCAAGCAGCAAG	58.5	variable	Aeromonas sp. G.N1.15¹⁸
3′-CS		AAGCAGACTTGACCTGA

Detection and characterization of integrons

The presence of IntI1, IntI2, and IntI3 genes, encoding class 1, 2, and 3 integrases, was investigated by PCR as previously described and using the same positive controls³⁰ (see Table 2). All integrase-positive isolates were examined to determine the sizes of integrons variable regions. Primer set 5′CS/3′CS was used to amplify the class 1 integron gene cassette region according to what has been described elsewhere³⁰ (see Table 2). PCR products were purified with the Jetquick PCR Product Purification Spin Kit (Genomed) and used as a template in the sequencing reaction carried out by the company GATC (Germany). When bands with different sizes were amplified, PCR products were cloned using a TA cloning kit (Invitrogen) according to the manufacturer's instructions. Clones representative of different amplicons were sent for sequencing. Online similarity searches were performed using the BLAST software.¹

Statistical analysis

Data retrieved from HPC on antibiotic-selective media were analyzed using the t-test through PAST version 2.04 (http://palaeoelectronica.org/2001_2001/past/issue2001_2001.htm).¹⁵

Nucleotide sequence accession numbers

The nucleotide sequences were deposited in the GenBank nucleotide database under the following accession numbers: JX646699 to JX646702.

Results

Prevalence of antibiotic-resistant bacteria

The percentages of ampicillin-, streptomycin-, and tetracycline-resistant cultivable bacteria found in the two water layers are shown in Table 3. In general, resistance to streptomycin was the most prevalent (5.86% to 42.87% of total HPC), followed by ampicillin (2.29% to 12.37%), and tetracycline (0.87% to 7.21%). Levels of resistance were significantly higher for the SML samples (p-value<0.05) with only a few exceptions (resistance to tetracycline in the C2 and C3 campaigns). The prevalence of ampicillin-resistant bacteria in the SML was three to four times higher than in UW and streptomycin-resistant bacteria were two to three times more prevalent in the SML. For tetracycline, significant differences were only detected in the first campaign (five times more prevalent in the SML).

Table 3.

Prevalence of Antibiotic-Resistant Bacteria in Surface Microlayer and Underlying Waters

		Total of HPC	% of HPC resistant to:
Campaign	Sample	(CFU/ml)	Streptomycin	Tetracycline	Ampicillin
C1	SML	2.9×10⁴	34.40±2.24	7.21±0.33	12.37±1.23
	UW	6.2×10³	21.33±1.37	1.49±0.25	3.20±0.74
C2	SML	1.3×10⁴	17.60±2.11	0.87±0.60	10.11±1.04
	UW	1.8×10⁴	5.86±2.66	0.99±1.90	3.99±1.82
C3	SML	3.6×10⁴	42.87±6.67	0.91±0.33	7.73±0.45
	UW	1.9×10⁴	12.53±0.71	1.17±1.01	2.29±0.60

HPC, heterotrophic plate counts.

Antibiotic susceptibility testing

Susceptibility to all tested antibiotics was observed in 36% (61 out of 171) of the isolates. Predominant resistances were observed to cephalothin (46% of the isolates were resistant to this antibiotic), followed by ampicillin (40%), and streptomycin (34%). Isolates were most frequently susceptible to imipenem (only 6% of the isolates were resistant to this antibiotic), tetracycline (4%), and gentamicin (3%). Significant differences between isolates collected from SML and UW were not observed (Fig. 1).

FIG. 1.

Percentages of strains obtained from the surface microlayer (SML) and underlying waters (UW) that were resistant to each of the antibiotics tested. Antibiotic abbreviations: AMP, ampicillin; IMP, imipenen; TE, tetracycline; CN, gentamicin; NA, nalidixic acid; SXT, sulfamethoxazole/trimethoprim; C, chloramphenicol; S, streptomycin; KF, cephalothin.

Since antibiotic resistance profiles may be dependent on the phylogenetic groups, results were separately analyzed for the most commonly retrieved genera (Aeromonas, Acinetobacter, Pseudomonas, Psychrobacter, and Shewanella) as well as for the isolates classified as Enterobacteriaceae (Table 4). Most groups were highly resistant (resistance rates >30%) to ampicillin, cephalothin, and streptomycin. Lower levels of resistance to these antibiotics were only registered for Acinetobacter against streptomycin and for Shewanella against ampicillin. Additionally, Psychrobacter members were exceptionally sensitive (resistance rates <18%) to all antibiotics tested. All groups were highly susceptible (resistance rates <18%) to imipenem, gentamicin, and tetracycline. Resistance levels to chloramphenicol, nalidixic acid, and trimethoprim/sulfamethoxazole were generally low with a few exceptions: Acinetobacter and Pseudomonas were highly resistant to trimethoprim/sulfamethoxazole and Pseudomonas members were also highly resistant to chloramphenicol and nalidixic acid.

Table 4.

Antibiotic Resistance Rates of the Most Frequently Retrieved Genera and Enterobacteriaceae

	Resistance to antibiotics (%)
Phylogenetic group (number of isolates ^a )	Ampicillin	Cephalothin	Imipenem	Tetracycline	Chloramphenicol	Gentamicin	Nalidixic acid	Streptomycin	Trimethoprim/sulfamethoxazole	MDR ^b
Acinetobacter (n=13)	53.8	92.3	7.7	7.7	15.4	0.0	15.4	7.7	38.5	15.4
Aeromonas (n=17)	100	82.2	5.9	17.6	0.0	0.0	17.5	88.2	0.0	23.5
Enterobacteriaceae (n=13)	46.2	30.8	0.0	7.7	7.7	0.0	7.7	30.8	7.7	15.4
Pseudomonas (n=17)	82.4	88.2	0.0	5.9	47.1	0.0	47.1	41.2	47.1	52.9
Psychrobacter (n=28)	3.6	3.6	7.1	0.0	0.0	0.0	10.7	17.9	0.0	0.0
Shewanella (n=14)	21.4	100	7.1	0	0.0	7.1	14.3	42.9	7.1	14.3

n, number of isolates affiliated to each phylogenetic group.

MDR, percentage of multiresistant strains (resistant to antibiotics included in 3 or more classes).

The overall frequency of multiresistant isolates (resistant to three or more classes of antibiotics) was 14% (24 isolates out of a total of 171 isolates). In terms of multiresistance, results obtained for SML and UW isolates were also similar (Fig. 2). For multiresistant isolates, 16 antibiotic resistance patterns were observed (Table 5). The most common resistance profile included resistance to ampicillin, nalidixic acid, streptomycin, cephalothin, chloramphenicol, and trimethoprim/ sulfamethoxazole (two isolates from SML and two from UW). The comparison between phylogenetic groups revealed that multiresistance was more frequent among Pseudomonas strains (Table 4).

FIG. 2.

The percentage of multiple antibiotic-resistant bacteria from the SML (bacterioneuston) and UW (bacterioplankton).

Table 5.

Antibiotic Resistance Patterns of Surface Microlayer and Underlying Waters Isolates

	No. of multiresistant isolates (n=24)
Antibiotic resistance patterns	SML	UW
AMP, TE, C^b	0	1
AMP, NA, S^a	2	1
TE, C, KF^b	0	1
NA, S, KF^b	0	1
AMP, NA, SXT, S^b	0	1
AMP, NA, SXT, KF^c	1	0
AMP, NA, S, KF^a	1	1
AMP, C, S, KF^b	0	1
AMP, TE, NA, S, KF^b	0	3
AMP, TE, NA, SXT, KF^c	1	0
AMP, NA, SXT, C, KF^c	1	0
AMP, NA, C, S, KF^c	1	0
IMP, CN, NA, SXT, S, KF^b	0	1
AMP, IMP, CN, C, S, KF^c	1	0
AMP, IMP, SXT, C, S, KF^c	1	0
AMP, NA, SXT, C, S, KF^a	2	2

The most common pattern of multiple-drug resistance among both water collections. Antibiotic resistance patterns specific of ^bUW and ^cSML isolates.

AMP, ampicillin; KF, cephalothin; IMP, imipenem; TE, tetracycline; C, chloramphenicol; CN, gentamicin; NA, nalidixic acid; S, streptomycin; SXT, sulfamethoxazole/trimethoprim.

Screening for antibiotic resistance and integrase genes

The presence of genes conferring resistance to beta-lactams, tetracyclines, chloramphenicol, aminoglycoside, and sulfonamides was analyzed by PCR in all the 171 isolates (Table 2). Genes bla_TEM, bla_OXA-B, bla_SHV, bla_IMP, tet(A), tet(B), tet(E), tet(M), cat, sul1, sul2, sul3, and aadA were chosen as they have been the most frequently detected acquired genes both in clinical and environmental isolates. Surprisingly, despite the considerable high levels of resistance against beta-lactams and aminoglycosides, genes conferring resistance to these antibiotics were not detected. Genes conferring resistance to chloramphenicol were also not found. Among seven isolates displaying resistance to tetracycline, tet(E) was detected in two isolates belonging to genus Aeromonas, and tet(M) was detected in one Pseudomonas isolate. The gene sul1 was amplified from three isolates belonging to genus Aeromonas, and sul2 was detected in one Pseudomonas isolate.

The intI1 gene was present in 2.11% of the isolates, all affiliated with Aeromonas, while genes intI2 and intI3 were not detected in any isolate. The variable region of class 1 integrons was successfully amplified with the primers 5′CS/3′CS. Sequencing analysis revealed four different arrays: aadA1, aadA13, catB8 and catB8, aadA1. The last two arrays were amplified from the same isolate.

Discussion

We compared the persistence of antibiotic resistance among cultivable bacteria from the SML and UW in an estuarine system. We hypothesized that SML would be a reservoir of antibiotic resistance.

Specifically, we compared the prevalence of tetracycline-, ampicillin-, and streptomycin-resistant bacteria between both water layers. These antibiotics have been used for long not only in human medicine, but also in agriculture, aquaculture, and veterinary.⁶ Accordingly, resistance to these drugs is highly disseminated in several environments.^3,34 Results obtained during this study showed significant differences between SML and UW in terms of prevalence of resistance to all tested antibiotics. Differences may arise from the fact that SML accumulates pollutants, including antibiotics,^20,39 and antibiotic concentrations in the environment are the main factors contributing to the selection of resistant bacteria.²⁴ In addition to the selective pressure exerted by antibiotics themselves, other compounds, such as heavy metals, disinfectants, or pesticides also contribute to the maintenance of antibiotic resistance.²² SML has been described to accumulate nickel, copper, arsenic, chlorinated pesticides, and polychlorinated biphenyls.^11,42 On the other hand, producers of antimicrobial compounds have often been described among particle-attached bacteria rather than free-living bacteria.^14,40,41 Particle-attached bacteria may be enriched in SML due to accumulation of higher amounts of organic particles in this water layer.¹⁰

Besides determining the prevalence of antibiotic-resistant bacteria in both layers using antibiotic-supplemented agar, during this study, we also compared antibiotic resistance profiles of isolates from SML and UW. Isolates were obtained in nonselective agar without an antibiotic. No significant differences in terms of resistance profiles or prevalence of multiresistance were detected between SML and UW isolates. These apparently contradictory results are probably due to the fact that generally resistant bacteria were a minor component of the SML and UW bacterial communities preventing the detection of differences between culture collections. In fact, antibiotic resistance levels among our bacterial collections were generally low and multiresistance was infrequent. For example, members of the genus, Psychrobacter, were highly sensitive to all antibiotics tested. Psychrobacter was the most frequently retrieved genus representing almost 17% of the total number of isolates. Although other studies reported considerable higher resistance levels in estuarine bacteria,⁴³ those studies were either focused on resistant bacteria selected on antibiotic-supplemented agar¹² or on specific bacterial groups known for their high resistance levels to several antibiotics.¹⁸ When unbiased selections were conducted, results were similar to the ones obtained during this study.^32,33

Comparative information on antibiotic resistance between SML and UW is still scarce. Even so, a few studies reported differences between SML and UW in terms of prevalence of resistant bacteria.^20,22 Those studies were also based on counting antibiotic-resistant bacteria in antibiotic-supplemented agar.

In this study, colonies were randomly selected from Petri dishes, which may have excluded some of the less common members of the community. Even so, this procedure was adopted since colony selection based on morphological traits commonly results in overrepresentation of these less prevalent members. Also, the representativeness of our culture collection was improved by collecting a high number of samples and by using two different culture media.

We assessed the presence and diversity of acquired ARGs commonly detected both in clinical and environmental isolates. As previously described, antibiotic resistance phenotypes and genotypes may not match.¹⁸ Thus, we decided to screen for the presence of resistance genes in all the isolates independently of their resistance phenotypes. The presence of the ARGs in isolates with a susceptible phenotype is worth evaluating, since in combination with other resistance mechanisms or in a different genomic context these genes may confer or potentiate resistance.

Most of the times we failed to detect any of the inspected genes even in resistant isolates. For example, the presence of the common acquired bla genes encoding β-lactamases was not demonstrated in any of the isolates, even when an ampicillin resistance phenotype was detected. Again, our results may be due to the fact that an unbiased selection was conducted. This strategy resulted in a collection of isolates belonging to a wide range of genera, for some of which antibiotic resistance mechanisms had never been characterized. These mechanisms may be intrinsic rather than the acquired ones inspected during this study. For example, trends of antibiotic resistance were essentially the same for the genera most commonly retrieved in our study and included high levels of resistance to ampicillin, cephalothin, and streptomycin. Intrinsic antibiotic resistance mechanisms to beta-lactams have already been described, such as the production of chromosomal-encoded beta-lactamases in Aeromonas, Pseudomonas, Acinetobacter, Vibrio, Stenotrophomonas, or Shewanella.²⁸ Additionally, Acinetobacter, Pseudomonas, and Aeromonas are noted for their intrinsic resistance to antibiotics due to several other mechanisms, such as the production of amynoglicoside-modifying enzymes, the diminished expression of outer membrane proteins, and the overexpression of efflux pumps.^4,21 For these genera, the combination of several resistance mechanisms often occurs leading to the emergence of multiresistance strains.^4,12 In our study, multiresistance was particularly associated to Pseudomonas and Aeromonas.

Concerning tetracyclines, the genes tet(E) and tet(M) were detected in Aeromonas and Pseudomonas, respectively. tet(E) has previously been described and has the most frequent tetracycline resistance mechanism in Aeromonas¹⁹ and tet(M) is also common among Gram-negative isolates, including Pseudomonas.⁵

The prevalence of integrons was also assessed and class 1 integrons were detected in 2.11% of our isolates. This prevalence is comparable to what has been reported for estuarine bacteria from other geographical locations.³⁶ All class 1 integrons were detected in Aeromonas isolates (two from SML and one from UW samples), predominantly carrying the aadA-type cassettes (aadA1, aadA13), conferring resistance to streptomycin. These cassettes are reported as the most frequent gene cassettes in bacterial isolates³¹ and have been frequently detected in Aeromonas.^18,30,26

Results obtained during this study must be interpreted having in mind that only the culturable fraction of the bacterial community was considered. We are aware that only a minor fraction of the environmental bacterial community can be cultivated under laboratory conditions.² However, the large majority of the studies on antibiotic resistance conducted so far are based on the characterization of pure cultures. The main reasons are related to the current limitations of culture-independent methodologies.¹⁷ On one hand, fragments amplified from total DNA are not necessarily indicative of the presence of functional genes. On the other hand, the hosts of those genes usually cannot be identified by culture-independent methods, which present a severe limitation when studying antibiotic resistance.

In summary, the prevalence of antibiotic-resistant bacteria was different between the SML and UW only when selection was conducted using antibiotic-supplemented agar. Even so, these results suggest that SML conditions select for antibiotic resistance. The overall results also showed that, although antibiotic resistance occurs among estuarine bacteria, the resistance mechanisms are different from the most common acquired mechanisms, being probably predominantly intrinsic. Although we cannot exclude the contribution of transferable genetic elements, our results suggest that SML primarily selects for intrinsic antibiotic resistance.

Footnotes

Acknowledgments

This work was financed by the Fundacão para a Ciência e a Tecnologia (FCT) through grants SFRH/BPD/65820/2009 (CSO) and SFRH/BPD/63487/2009 (IH) and SFRH/BD/64057/2009 (JSNA). We are also grateful to A. Moura (Department of Biology, University of Aveiro) for reviewing English style and grammar.

Disclosure Statement

No competing financial interests exist.

References

Altschul

S.F.

, Madden

T.L.

, Schaffer

A.A.

, Zhang

et al. 1997. Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res, 25:3389–3402.

Azevedo

J.S.N.

, Ramos

, Araújo

, Oliveira

C.S.

et al. 2012. Spatial and temporal analysis of estuarine bacterioneuston and bacterioplankton using culture-dependent and culture-independent methodologies. Antonie van Leeuwenhoek, 101:819–835.

Barkovskii

A.L.

, Bridges

2011. Persistence and profiles of tetracycline resistance genes in swine farms and impact of operational practices on their occurrence in farms' vicinities. Water Air Soil Pollut., 223:49–62.

Bonomo

R.A.

, Szabo

2006. Mechanisms of multidrug resistance in Acinetobacter species and Pseudomonas aeruginosa. Clin. Infect. Dis., 43,Suppl:2S49–2S56.

Brown

M.G.

, Mitchell

E.H.

, Balkwill

D.L.

2008. Tet 42, a novel tetracycline resistance determinant isolated from deep terrestrial subsurface bacteria. Antimicrob. Agents Chemother., 52:4518–4521.

Chelossi

, Vezzulli

, Milano

, Branzoni

, Fabiano

, Riccardi

, Banat

I.M.

2003. Antibiotic resistance of benthic bacteria in fish-farm and control sediments of the Western Mediterranean. Aquaculture, 219:83–97.

Cincinelli

, Stortini

A.M.

, Checchini

, Martellini

et al. 2005. Enrichment of organic pollutants in the sea surface microlayer (SML) at Terra Nova Bay, Antarctica: influence of SML on superficial snow composition. J. Environ. Monit., 7:1305–1312.

[CLSI] Clinical, Laboratory Standards Institute. Performance standards for antimicrobial susceptibility testing; twentieth informational supplementDocument M100-S20Wayne, PA: CLSI, 2005.

Correia

, Boavida

, Grosso

et al. 2003. Molecular characterisation of a new class 3 integron in Klebsiella pneumoniae. Antimicrob. Agents. Chemother., 47:2838–2843.

10.

Cunliffe

, Murrell

J.C.

2009. The sea-surface microlayer is a gelatinous biofilm. ISME J., 3:1001–1003.

11.

Cuong

D.T.

, Karuppiah

, Obbard

J.P.

2008. Distribution of heavy metals in the dissolved and suspended phase of the sea-surface microlayer, seawater column and in sediments of Singapore's coastal environment. Environ. Monit. Assess., 138:255–272.

12.

Evangelista-Barreto

N.S.

, Carvalho

F.C.T.

, Vieira

R.H.S.F.

, Reis

C.M.F.

et al. 2010. Characterization of Aeromonas species isolated from an estuarine environment. Braz. J. Microbiol., 41:452–460.

13.

Figueira

, Serra

, Manaia

C.M.

2011. Differential patterns of antimicrobial resistance in population subsets of Escherichia coli isolated from waste- and surface waters. Sci. Total. Environ., 409:1017–1023.

14.

Hakvåg

, Fjærvik

, Josefsen

K.D.

, Ian

et al. 2008. Characterization of Streptomyces spp. Isolated from the Sea Surface Microlayer in the Trondheim Fjord, Norway. Mar. Drugs, 6:620–635.

15.

Hammer

, Harper

D.A.T.

, Ryan

P.D.

2001. PAST: paleontological statistics software package for education and data analysis. Palaeontologia Electronica, 4,1:9.

16.

Hardy

J.T.

1982. The sea surface microlayer: biology, chemistry and anthropogenic enrichment. Prog. Oceanogr., 11:307–328.

17.

Henriques

, Moura

, Alves

, Saavedra

M.J.

, Correia

2006. Analysing diversity among beta-lactamase encoding genes in aquatic environments. FEMS Microbiol. Ecol., 56:418–429.

18.

Henriques

, Fonseca

, Alves

, Saavedra

M.J.

et al. 2006. Occurrence and diversity of integrons and beta-lactamase genes among ampicillin-resistant isolates from estuarine waters. Res. Microbiol., 157:938–947.

19.

Henriques

, Fonseca

, Alves

, Saavedra

M.J.

et al. 2008. Tetracycline-resistance genes in Gram-negative isolates from estuarine waters. Lett. Appl. Microbiol., 47:526–533.

20.

Hermansson

, Jones

G.W.

, Kjelleberg

1987. Frequency of antibiotic and heavy metal resistance, pigmentation and plasmids in bacteria of the marine air-water interface. Appl. Environ. Microbiol., 53:2338–2342.

21.

Janda

J.M.

, Abbott

S.L.

2010. The genus Aeromonas: taxonomy, pathogenicity, and infection. Clin. Microbiol. Rev., 23:35–73.

22.

Jones

J.G.

, Gardner

, Simon

B.M.

, Pickup

R.W.

1991. Heterotrophic bacteria of the freshwater neuston and their ability to act as plasmid recipients under nutrient deprived conditions. Microb. Ecol., 22:15–25.

23.

Keddy

P.A.

2000. Wetland Ecology: Principles and Conservation. Cambridge University Press, Cambridge, pp., 3–77.

24.

Kummerer

2004. Resistance in the environment. J. Antimicrob. Chemother., 54:311–320.

25.

Kümmerer

2009. Antibiotics in the aquatic environment–a review–part I. Chemosphere, 75:417–434.

26.

Laroche

, Pawlak

, Berthe

, Skurnik

, Petit

2009. Occurrence of antibiotic resistance and class 1, 2 and 3 integrons in Escherichia coli isolated from a densely populated estuary (Seine, France) FEMS Microbiol. Ecol., 68:118–130.

27.

Liss

P.S.

, Duce

R.A.

1997. The Sea Surface and Global Change. Cambridge University Press: Cambridge.

28.

Livermore

D.M.

1995. Beta-lactamases in laboratory and clinical resistance. Clin. Microbiol. Rev., 8:557–584.

29.

Maki

J.S.

2002. Neuston microbiology: life at the air–water interface. Bitton

The Encyclopedia of Environmental Microbiology. John Wiley & Sons: New York, 2133–2144.

30.

Moura

, Henriques

, Ribeiro

, Correia

2007. Prevalence and characterization of integrons from bacteria isolated from a slaughterhouse wastewater treatment plant. J. Antimicrob. Chemother., 60:1243–1250.

31.

Moura

, Soares

, Pereira

, Leitao

et al. 2009. INTEGRALL: a database and search engine for integrons, integrases and gene cassettes. Bioinformatics, 25:1096–1098.

32.

Mudryk

2002. Antibiotic resistance among bacteria inhabiting surface and subsurface water layers in estuarine lake Gardno. Pol. J. Environ. Stud., 11:401–406.

33.

Mudryk

, Skorczewski

2009. Frequency of antibiotic resistance in bacteria inhabiting water of downtown pond. Balt. Coast. Zone, 13a:135–146.

34.

Olaniran

A.O.

, Naicker

, Pillay

2009. Antibiotic resistance profiles of Escherichia coli isolates from river sources in Durban, South Africa. World J. Microbiol. Biotechnol., 25:1743–1749.

35.

Oliveira

C.S.

, Lázaro

, Azevedo

J.S.N.

, Henriques

et al. 2012. New molecular variants of epsilon and beta IncP-1 plasmids are present in estuarine waters. Plasmid, 67:252–258.

36.

Rosser

S.J.

, Young

H.-K.

1999. Identification and characterization of class 1 integrons in bacteria from an aquatic environment. J. Antimicrob. Chemother., 44:11–18.

37.

Sieburth

J.M.

, Willis

, Johnson

K.M.

, Burney

C.M.

et al. 1976. Dissolved organic matter and heterotrophic microneuston in the surface microlayers of the North Atlantic. Science, 194:1415–1418.

38.

Versalovic

, Koeuth

, Lupski

J.R.

1991. Distribution of repetitive DNA sequences in eubacteria and application to fingerprinting of bacterial genomes. Nucleic Acids Res., 19:6823–6831.

39.

Walczak

, Donderski

2004. Antibiotic sensitivity of neustonic bacteria in lake Jeziorak Mały. Pol. J. Environ. Stud., 13:429–434.

40.

Wilson

G.S.

, Raftos

D.A.

, Corrigan

S.L.

, Nair

S.V.

2010. Diversity and antimicrobial activities of surface-attached marine bacteria from Sydney Harbour, Australia. Microbiol. Res., 165:300–315.

41.

Wilson

G.S.

, Raftos

D.A.

, Nair

S.V.

2011. Antimicrobial activity of surface attached marine bacteria in biofilms. Microbiol. Res., 166:437–448.

42.

Wurl

, Obbard

J.P.

2005. Chlorinated pesticides and PCBs in the sea-surface microlayer and seawater samples of Singapore. Mar. Pollut. Bull., 50:1233–1243.

43.

Zheng

, Qiu

, Chen

, Yu

et al. 2011. Antibiotics pollution in Jiulong River estuary: Source, distribution and bacterial resistance. Chemosphere, 84:1677–1685.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.05 MB