Abundance and characteristics of the recreational water quality ...

Abundance and characteristics of the recreational waterquality indicator bacteria Escherichia coli and enterococciin gull faeces

L.R. Fogarty1, S.K. Haack1, M.J. Wolcott2 and R.L. Whitman3

1US Geological Survey, Lansing, MI, 2US Geological Survey, Madison, WI, and 3US Geological Survey, Porter, IN, USA

2002/326: received 3 September 2002, revised 8 January 2003 and accepted 23 January 2003

ABSTRACT

L.R . FOGARTY, S.K . HAACK, M.J . WOLCOTT AND R.L . WHITMAN. 2003.

Aims: To evaluate the numbers and selected phenotypic and genotypic characteristics of the faecal indicator

bacteria Escherichia coli and enterococci in gull faeces at representative Great Lakes swimming beaches in the United

States.

Methods and Results: E. coli and enterococci were enumerated in gull faeces by membrane filtration. E. coli

genotypes (rep-PCR genomic profiles) and E. coli (Vitek� GNI+) and enterococci (API� rapid ID 32 Strep and

resistance to streptomycin, gentamicin, vancomycin, tetracycline and ampicillin) phenotypes were determined for

isolates obtained from gull faeces both early and late in the swimming season. Identical E. coli genotypes were

obtained only from single gull faecal samples but most faecal samples yielded more than one genotype (median of

eight genotypes for samples with 10 isolates). E. coli isolates from the same site that clustered at ‡85% similarity

were from the same sampling date and shared phenotypic characteristics, and at this similarity level there was

population overlap between the two geographically isolated beach sites. Enterococcus API� profiles varied with

sampling date. Gull enterococci displayed wide variation in antibiotic resistance patterns, and high-level resistance

to some antibiotics.

Conclusions: Gull faeces could be a major contributor of E. coli (105–109 CFU g)1) and enterococci (104–

108 CFU g)1) to Great Lakes recreational waters. E. coli and enterococci in gull faeces are highly variable with

respect to their genotypic and phenotypic characteristics and may exhibit temporal or geographic trends in these

features.

Significance and Impact of the Study: The high degree of variation in genotypic or phenotypic characteristics

of E. coli or enterococci populations within gull hosts will require extensive sampling for adequate characterization,

and will influence methods that use these characteristics to determine faecal contamination sources for recreational

waters.

Keywords: enterococci, Escherichia coli, faecal contamination, gulls, water quality.

INTRODUCTION

Recreational waters are susceptible to a variety of sources of

microbiological pollution (USEPA 1986; Levesque et al.

2000; Rose et al. 2001). In 1986, the United States

Environmental Protection Agency (USEPA) published

numerical standards for Escherichia coli and enterococci for

fresh US recreational waters (USEPA 1986). In October

2000, the US Congress required states with coastal (marine

or Great Lakes) recreational waters to adopt (by April 2004)

the USEPA criteria, and to establish monitoring and

public notification programmes. As the responsible agencies

develop beach-monitoring programmes in response to the

new legislation, new and more detailed informationCorrespondence to: L.R. Fogarty, USGS-WRD, 6520 Mercantile Way Suite 5,

Lansing, MI 48911, USA (e-mail: [email protected]).

ª 2003 The Society for Applied Microbiology

Journal of Applied Microbiology 2003, 94, 865–878

concerning sources of E. coli and enterococci will be

required to manage Great Lakes recreational waters.

Because faecal contamination can come from many different

human (septic systems or sewers) and animal sources

(animal pasture runoff, waterfowl, wildlife or domestic

animals), identifying sources of faecal contamination can be

an aid to improving management of recreational waters.

Recently, various methods have been proposed to identify

sources of faecal contamination by classifying faecal bacteria

[faecal coliforms, faecal streptococci (more recently, entero-

cocci) or E. coli] from known sources based on phenotypic or

genotypic characteristics and using these characteristics to

classify faecal bacteria of unknown source that were isolated

from the environment. Phenotypic source-determination

methods have included multiple antibiotic resistance (MAR)

profiles (Krumperman 1983; Kaspar et al. 1990; Wiggins

1996; Parveen et al. 1997; Hagedorn et al. 1999; Harwood

et al. 2000), O-serotyping and fatty acid methyl ester

analysis (Parveen et al. 2001). Genotypic source-determin-

ation methods have included ribotyping (RT; Parveen et al.

1999; Carson et al. 2001), pulsed-field gel electrophoresis

(Parveen et al. 2001) and rep-PCR profiles (Dombek et al.

2000).

Very seldom has the range of phenotypic or genotypic

characteristics of faecal indicator bacteria within host

populations been considered in source-determination

studies. Even though early studies indicated that individ-

ual animals host a variety of phenotypes and genotypes of

E. coli including resident strains and continuous immi-

grants from the environment (Selander et al. 1987), most

source-determination studies have used samples taken on

multiple dates or from multiple locations. For example,

Wiggins et al. (1999) studied antibiotic resistance profiles

of faecal streptococci isolated from humans, cattle, poultry

and wild animals over a 4-year period, and classified the

isolates with respect to source, using discriminant analysis.

The average rate of correct classification (ARCC; number

of correctly classified isolates divided by the total number

studied) was 64–78%. It was hypothesized in this study

that the relatively low ARCC might have been because of

changes within the source-specific populations from which

the samples were collected. Other studies applying

discriminant analysis to different source-tracking methods

have reported ARCCs of: RT, 82% (Parveen et al. 1999)

or 74–96% depending on number of sources analysed

(Carson et al. 2001); rep-PCR, 87–93% (Dombek et al.

2000); MAR of faecal streptococci, 87% (Hagedorn et al.

1999); MAR of faecal streptococci or faecal coliforms,

62–64% (Harwood et al. 2000). Additionally, most source-

determination studies have shown varying success of

classification depending on source type. For example,

Dombek et al. (2000) used rep-PCR DNA fingerprints as

a method of source determination for E. coli isolated from

faecal samples. In that study, 100% of the E. coli isolates

from chickens and cows were classified correctly; however,

only 80–89% of the E. coli isolates from waterfowl (ducks

and geese) were classified correctly. Evaluation of variab-

ility in the physiological and genomic characteristics of

faecal bacteria populations within hosts may help to

explain the percentage of incorrect classification of source

samples and refine the usefulness of the proposed source-

determination methods.

Gulls (Larus sp.) have not been addressed in any

source-determination studies to date, despite their poten-

tial or documented significance as a major source of faecal

contamination to reservoirs and recreational waters and at

bathing beaches (Jones et al. 1978; Levesque et al. 1993,

2000; Hatch 1996; Alderisio and Deluca 1999; Jones and

Obiri-Danso 1999; Obiri-Danso and Jones 2000). Gull

faecal material is considered a threat to human health

(Hatch 1996; Levesque et al. 2000). Studies have docu-

mented the presence in gull faeces of human bacterial

pathogens such as Salmonella spp., Aeromonas spp.,

Campylobacter spp, and E. coli serotype O157 (Jones et al.

1978; Hatch 1996; Wallace et al. 1997; Levesque et al.

2000; Obiri-Danso and Jones 2000). No studies have

specifically reported the numbers of USEPA-recommen-

ded recreational water faecal indicator bacteria E. coli and

enterococci in gull faeces, although faecal coliforms

(Levesque et al. 1993, 2000; Alderisio and Deluca 1999)

and faecal streptococci (Jones and Obiri-Danso 1999) have

been reported. Information on the abundance and phen-

otypic and genotypic characteristics of these indicator

bacteria in gull faeces will be useful to managers of

recreational waters in the US Great Lakes and similar

environmental settings.

The objective of this study was to evaluate, over the

typical recreational swimming season, numbers and selected

phenotypic and genotypic characteristics of the indicator

bacteria E. coli and enterococci in gull faeces. We quantified

E. coli and enterococci in gull faeces collected at two Lake

Michigan, USA, beaches between May and October 2000.

We used E. coli rep-PCR genomic profiles and E. coli and

enterococci phenotypic tests to characterize E. coli and

Enterococcus populations in gull faecal material. This paper

presents the results of this study.

MATERIALS AND METHODS

Gull faecal sample collection

Gull faecal samples were collected at Lake Michigan

beaches in Chicago, IL, USA (CHI) and Traverse City,

MI, USA (TC) between May and October 2000. These

beaches lie on the opposite shores of Lake Michigan

(Figure 1). Sample nomenclature is presented in Table 1.

866 L.R. FOGARTY ET AL.

ª 2003 The Society for Applied Microbiology, Journal of Applied Microbiology, 94, 865–878

Faecal samples were collected just after defecation. CHI

gull faecal samples were collected by rolling a sterile swab

in the centre of gull faecal droppings. TC samples were

collected with a sterile spatula. Care was taken during

sampling to be sure no surrounding beach sediment was

collected. Samples were placed in a sterile tube, stored on

ice, and processed in the laboratory 24–48 h after sample

collection. Faecal material weight was determined for CHI-

A21 and all TC samples but not for CHI-JU or CHI-A1

samples. All samples were suspended in a known volume

of phosphate buffered saline, diluted in series and filtered

by membrane filtration method for isolation of E. coli and

enterococci.

Bacteria isolation and identification

Escherichia coli and enterococci were isolated from all

samples using membrane filtration (American Public Health

Association 1998; USEPA 2000). A sterile buffered saline

control and a series of dilutions were passed through

individual sterile 0Æ45 lm pore size, gridded cellulose nitrate

membrane filters (Advantec MFS, Inc., Pleasanton, CA,

USA). Dilution tubes were thoroughly mixed before

filtration. Total coliform bacteria were identified on

mENDO agar LES medium (DIFCO Laboratories, Detroit,

MI, USA). For E. coli identification, membranes with

15–50 well-separated coliform colonies were transferred to

300 km

lllionois

Michigan

United States

Canada

Chicago

TraverseCity

0

85˚ 45′

45˚ 00′

Lake

Mic

higa

n

Fig. 1 Study location

E. COLI AND ENTEROCOCCI IN GULL FAECES 867


Nutrient Agar containing 4-methylumbelliferyl-b-DD-glu-

curonide (Na-MUG agar; DIFCO). Colonies that fluoresced

blue under UV light were identified presumptively as E. coli.

Selected presumptive E. coli were transferred to LB medium

and confirmed by three different physiological tests: negat-

ive indophenol oxidase production (BBL� oxidase reagent

droppers, Becton Dickinson Microbiology Systems, Coc-

keysville, MD, USA), positive b-DD-galactosidase reaction

(Taxo� ONPG discs, Becton Dickinson Microbiology

Systems), and positive indole production (BBLTM Dry-

SlideTM Indole, Becton Dickinson Microbiology Systems).

CHI isolates subsequently were verified as E. coli based on

reactions from the Vitek� GNI+ system (bioMerieux,

Hazelwood, MO, USA).

Enterococcus isolates were identified using membrane

filtration on mEI agar as described by USEPA (USEPA

2000). Enterococci with representative morphologies on

mEI agar were isolated and confirmed by growth on brain

heart infusion agar with 6Æ5% NaCl at 35�C, esculin

hydrolosis on bile esculin agar and negative catalase activity

(USEPA 2000). Enterococcus isolates were further charac-

terized using multiple physiologic assays (API� rapid ID 32

Strep, bioMerieux) as well as colony colour and haemolysis

on Columbia sheep blood agar (BBL Becton Dickinson).

Rep-PCR genomic profiles of Escherichia coliisolates

Rep-PCR procedures were revised slightly from those

described in Rademaker and de Bruijn (1997). Primers used

were REP 1R and REP 2I (Versalovic et al. 1991) obtained

from Genosys Biotechnologies (The Woodlands, TX, USA)

and diluted in TE (10 mmol l)1 Tris, pH 8Æ0, 1 mmol l)1

EDTA). The rep-PCR reaction components consisted

Table 1 Numbers of Escherichia coli, E. coli genotypes and enterococci in gull faecal samples

E. coli Enterococci

Seagull faecal sample Collection date CFU No. of isolates No. of genotypes* CFU No. of isolates

CHI-Ju-A 26 June 2000 – 3 2 – 5

CHI-Ju-B 26 June 2000 – 3 3 – 5

CHI-Ju-C 26 June 2000 – 8 7 – 5

CHI-Ju-D 26 June 2000 – 6 6 – –

CHI-Ju-E 26 June 2000 – – – – 5

CHI-Ju-F 26 June 2000 – – – – 5

CHI-A1-B5 1 August 2000 – 2 2 – –

CHI-A1-B6 1 August 2000 – 3 3 – –

CHI-A1-B7 1 August 2000 – 2 2 – –

CHI-A1-B8 1 August 2000 – 4 4 – –

CHI-A1-SB1 1 August 2000 – 1 1 – –

CHI-A1-SB2 1 August 2000 – 3 3 – –

CHI-A1-SB3 1 August 2000 – 2 2 – –

CHI-A1-SB4 1 August 2000 – 2 2 – –

CHI-A21-A 21 August 2000 1Æ9 · 109 10 9 4Æ0 · 106 5

CHI-A21-B 21 August 2000 2Æ3 · 107 10 7 2Æ8 · 105 6

CHI-A21-C 21 August 2000 1Æ9 · 107 5 5 6Æ5 · 107 5

CHI-A21-D 21 August 2000 5Æ0 · 106 9 3 2Æ0 · 104 10

TC-My-1 24 May 2000 5Æ7 · 106 2 2 – –

TC-My-2 25 May 2000 1Æ6 · 106 10 8 2Æ1 · 106 –

TC-A-A 9 August 2000 <1 · 105� – – 4Æ0 · 106 –

TC-A-B 9 August 2000 1Æ8 · 107 10 8 1Æ3 · 108 5

TC-A-C 9 August 2000 <1 · 106� – – <1 · 106� –

TC-A-D 9 August 2000 1Æ4 · 107 – – 2Æ4 · 108 5

TC-Oc-A 14 October 2000 6Æ9 · 107 4 3 2Æ3 · 107 1

TC-Oc-B 14 October 2000 1Æ0 · 106 3 2 1Æ0 · 106 1

Total 102 84 63

*Different genotypes defined as isolates having rep-PCR profiles of <100% similarity.

�Number less than the lowest dilution tested, as indicated.

CFU, colony-forming unit; CHI, Chicago; TC, Traverse City.



of a final concentration of: 1 · PCR reaction buffer

(100 mmol l)1 Tris–HCl pH 8Æ5, 500 mmol l)1 KCl; Gibco

BRL, Gaithersburg, NY, USA), 3Æ3 mmol l)1 MgCl2,

125 lmol l)1 of each dNTP (Pharmacia, Piscataway, NJ,

USA), 0Æ01 lg ll)1 BSA (Boehringer Mannheim, Indianap-

olis, IN, USA), 10% DMSO, 2 lmol l)1 of each primer, 2U

Taq DNA Polymerase (Gibco BRL), 1 ll of a 1 : 10 diluted

E. coli culture (18–24 h culture in LB broth), and sterile

tissue culture water to bring the volume up to 25 ll. To

confirm purity, cultures used for the PCR were streaked

onto EMB (DIFCO) and TSA with 5% sheep blood (BBL

Becton Dickinson). DNA amplification was carried out in a

Perkin Elmer 2400 Gene Amp PCR system (Perkin Elmer-

Cetus, Norwalk, CN, USA) with the following conditions:

95�C for 7 min; 34 cycles of: 94�C for 3 s, 92�C for 30 s,

40�C for 1 min, 65�C for 8 min; a final elongation of 16 min

at 65�C; and a final hold at 4�C. PCR products (7 ll of each)

were electrophoresed on a 2% agarose gel for 100 min at

75 V in a Wide Mini-Sub Cell GT system (Bio-Rad

Laboratories, Hercules, CA, USA) and visualized with

ethidium bromide staining. On each gel, a laboratory strain

of E. coli (ATCC 25922) was included as a positive control

and standard for comparisons. Banding patterns of scanned

images were compared using BioNumerics version 2Æ5(Applied Maths, Kortrijk, Belgium), with a resolution of

600 dpi. Similarities between banding patterns were estab-

lished using unweighted pair-group method using arith-

metic averages (UPGMA) clustering, based on the Dice

correlation coefficient with 1Æ0% optimization, 2% position

tolerance and 2% minimum height.

Enterococci biochemical tests

Enterococcus faecalis ATCC 19433 was used as control for

each series of tests. Results were interpreted according to

API� rapid ID 32 Strep manufacturer’s instructions. API�

rapid ID 32 Strep tests only indicate eight Enterococcus

species: Ent. avium, Ent. casseliflavus, Ent. durans, Ent.

faecalis, Ent. faecium 1 or 2, Ent. gallinarum, Ent. hirae and

Ent. saccharolyticus. For the purpose of discussion, species

names were assigned to isolates based on the API� test

results, although it is broadly recognized that physiologic

tests are inadequate to definitively identify enterococci from

environmental sources (Muller et al. 2001; Svec et al. 2002).

To establish similarities between isolates, the API� test

responses for the Ent. faecalis ATCC control, all gull

Enterococcus isolates and the eight Enterococcus species

reported in the API� manual were converted to binary data

and cluster analysis of the binary response profiles was

conducted (agglomerative clustering, Ward method) using

S-Plus 2000 (MathSoft Inc., Seattle, WA, USA). Resistance

of enterococci to the antibiotics streptomycin, gentamicin,

tetracycline, vancomycin and ampicillin was determined

using the Etest� (AB Biodisk, Piscataway, NJ, USA). These

five antibiotics are commonly used to treat enterococcal

infections in humans, and levels of resistance to each

antibiotic were defined using standard criteria (National

Committee for Clinical Laboratory Standards 2002).

RESULTS

Abundance of Escherichia coli and enterococciin gull faeces

Escherichia coli concentrations ranged from <1Æ0 · 105–

109 g)1 of faeces, and enterococci ranged from 104 to 108 g)1

(Table 1). The mean number (± standard deviation) of E. coli

for Chicago faecal samples was 4Æ9 · 108 ± 9Æ4 · 108 g)1

and for Traverse City faecal samples it was 1Æ4 · 107 ±

2Æ3 · 107 g)1. For Chicago faecal samples, enterococci

numbers ranged from 104 to 107 CFU g)1 (mean: 1Æ7 ·107 ± 3Æ2 · 107 g)1) and for Traverse City samples entero-

cocci numbers ranged from 105 to 108 (mean: 5Æ7 · 107 ±

9Æ3 · 107 g)1).

Genotypic and phenotypic characteristics ofEscherichia coli from gull faeces

Chicago isolates. Escherichia coli isolates from Chicago

seagull samples were characterized by both rep-PCR

genomic profiles and Vitek� phenotype. Clustering of these

isolates based on their rep-PCR profiles, and the relation of

these profiles to Vitek� phenotype, is depicted in Figure 2.

Isolates having <100% similarity of banding patterns

(Figure 2) were defined as different genotypes. In every

case, CHI isolates with identical genotypes were obtained

only from single seagull faecal samples (Figure 2). Typically,

isolates with identical genotypes also had identical or very

similar phenotypes. There was no unique phenotype for

each rep-PCR banding pattern at the 100% similarity level,

primarily because the isolates exhibited variability in only 12

phenotype characters, resulting in less phenotypic variation

than genotypic variation. At approximately 85% similarity

of banding patterns, clusters tended to be characterized by

isolates from a single sampling date (although not necessarily

the same sample). Clusters at approximately 85% banding

pattern similarity also tended to share phenotypic features,

and tended to be distinct in phenotype from adjacent

clusters at lower levels of similarity. There was some

association between phenotype and sampling date. The only

isolates positive for all phenotypic tests were from the A21

sampling event (faecal samples A21 B and C). Failure to

utilize raffinose was more common in June than August

samples. At >70% similarity, seven clusters (A–G in

Figure 2) with broadly similar banding pattern and phen-

otypic features could be identified.



Sam

ple

95 100

90858075706560REP-PCR, Kb

0·50

0·60

0·70

0·80

0·90

1·00

1·20

1·40

1·60

1·80

2·00

2·50

3·00

3·50

5·00

8·00

VITEK

OF

GG

CLA

CM

LTM

AN

XY

LR

AF

SO

RS

UC

AD

OC

OU

ON

PR

HA

AR

AG

LULY

SN

C

CHICHICHI

CHICHICHICHICHICHICHI

CHICHICHICHI

CHICHICHI

CHICHICHICHI

CHICHICHI

CHICHICHICHICHICHICHICHI

CHICHICHI

CHICHICHICHI

CHICHICHI

CHICHICHICHI

CHICHICHI

CHICHICHICHI

CHICHICHICHICHICHICHICHICHICHICHI

CHICHICHI

CHICHICHICHI

CHICHI

A1A21A21

A21A21A21A21A21A21A1

A1A21JUA21

JUA1A1

A1A1A21A21

A21A1A1

JUJUA1JUA21JUJUJU

A1A1A21

A1A21A21A21

A21A21A21

A21A21A21A1

A21JUJU

JUJUJUJU

A21JUJUJUJUJUA1A21A21A21A21

A21A21A1

JUA21A1A21

A1A1

B8CA

AAAABCSB2

B8ABB

DSB3SB1

SB2SB2BB

CB8SB4

BCB8DCAAA

B7B5A

B7DDD

DDD

DDDB6

BCC

CCCD

ACCBDDSB3AABB

BBB6

DCB5B

B6SB4

186

8452511

2732

632

3243

541

11342321

129

2312

1049

8562

654

7867

132452231097

8104

1411

12B

Similarity (%)

Dat

e

Isol

ate

A

B

C

D

E

F

G



Traverse City isolates. Traverse City isolates were char-

acterized only by rep-PCR genomic profiles. When rep-PCR

profiles for TC isolates were analysed with those for the

CHI isolates (Figure 3), seven of 29 TC isolates fell into the

existing CHI clusters A–D, some as much as 90% similar to

CHI isolates, and relations among Chicago isolates in these

clusters remained essentially unchanged (Figure 3). The

remaining 22 TC isolates formed new, TC-only, groups at

£70% similarity with CHI isolates in clusters D–G. As for

the CHI isolates, identical TC genotypes were only

identified in single gull faecal samples (Figure 3). There

was no case in which the same genotype was found in both

CHI and TC gull faecal samples.

Number of genotypes per sample. No prior information

was available on gull E. coli intraspecies population genomic

variability; therefore, we collected multiple isolates from all

but one gull faecal sample (Table 1). The number of

different E. coli genotypes was determined for each gull

faecal sample (Table 1). Every faecal sample with more than

one isolate yielded multiple E. coli genotypes (Table 1).

There was a tendency for the number of genotypes to

increase with increasing number of isolates per sample, with

a median of eight genotypes for the four samples with 10

isolates (Figure 4).

Phenotypic characteristics of enterococcifrom gull faeces

API test responses. Enterococcus phenotypes differed

between sites and on different sampling dates within sites.

The results of cluster analysis of API� Rapid ID 32 Strep

test responses for 51 CHI enterococci and 12 TC enterococci

are shown in Figure 5. Cluster A is composed of enterococci

exhibiting similar API� test responses to those of Ent.faecalis or Ent. avium. This cluster was dominated by 23 of

the 25 June isolates from CHI gull faecal samples. In

addition, TC isolates from October fell in this cluster.

Cluster B was composed of enterococci having similar test

responses to those of Ent. gallinarum, Ent. durans, Ent. hirae

and Ent. faecium. The species Ent. durans, Ent. hirae and

Ent. faecium form a highly related phylogenetic group, and

are difficult to distinguish based on biochemical tests

(Devriese et al. 2002). The two remaining isolates from

June CHI samples fell in Cluster B, along with the majority

of August TC isolates. Finally, Cluster C was dominated by

22 of the 26 August CHI isolates and included two TC

August isolates. The majority of the isolates in Cluster C

could not be identified by the API� Rapid ID 32 Strep test.

Group C1 was composed of enterococci uniformly positive

for mannitol, sorbitol and ribose acidification, positive for

arginine dihydrolase and alkaline phosphatase, and variably

positive for LL-arabinose acidification. The majority of Group

C2 isolates exhibited yellow pigmentation. API� Rapid ID

32 Strep only identifies the yellow-pigmented Ent. casse-

liflavus. Only the two TC August isolates exhibited the

correct biochemical test responses to be classified as Ent.

casseliflavus by the API� tests. The remaining isolates in

Group C2 were all isolates from CHI August gull faecal

samples.

Antibiotic resistance. Enterococci from CHI and TC

exhibited a variety of resistance patterns to the five tested

antibiotics (Table 2) with no obvious pattern with respect to

sampling date or location. The highest level of streptomy-

cin tested was 256 lg ml)1, and several isolates were fully

resistant at this level. It is possible that some of these

isolates may be resistant at levels >1000 lg ml)1, which

would indicate acquired resistance. Several isolates were

resistant to tetracycline at the highest tested concentration

(256 lg ml)1). One isolate was resistant to the highest

concentration of gentamicin tested, and at a level indicating

acquired resistance. No isolate was resistant to ampicillin.

No isolate exhibited resistance to vancomycin; however,

susceptibility to this antibiotic is defined as £4 lg ml)1. The

three isolates exhibiting an inhibitory concentration of

6 lg ml)1 were all yellow-pigmented isolates (two identified

as Ent. casseliflavus) and intrinsic, intermediate levels of

vancomycin resistance are typical of some yellow-pigmented

enterococci, including Ent. flavescens and Ent. casseliflavus.

DISCUSSION

The high concentrations of both the E. coli and the

enterococci associated with gull faeces suggest that gulls

Fig. 2 Dendrogram (UPGMA clustering based on Dice correlation

coefficient) of 73 Escherichia coli rep-PCR profiles for isolates obtained

from gull faeces collected on beaches in Chicago, IL. Phenotypes based

on Vitek� GNI+ response are shown but were not used in the

clustering; j ¼ a positive response, u ¼ a negative response. [DP-3

(DP-300 fermentation), OFG (glucose, oxidative utilization), GC

(growth control), ACE (acetamide utilization), ESC (esculin hydroly-

sis), PLI (Plantindican reaction), URE (urea utilization), CIT (citrate

utilization), MAL (malonate utilization), TDA (tryptophan deami-

nase), PXB (Polymixin B growth), LAC (lactose oxidation), MLT

(maltose oxidation), MAN (mannitol oxidation), XYL (xylose oxida-

tion), RAF (raffinose utilization), SOR (sorbitol utilization), SUC

(sucrose utilization), INO (inositol utilization), ADO (adonitol util-

ization), COU (p-coumaric fermentation), H2S (hydrogen sulphide

production), ONP (ortho-nitrophenol galactopyranoside hydrolysis),

RHA (rhamnose utilization), ARA (LL-arabinose utilization), GLU

(glucose fermentation), ARG (argenine dihydrolation), LYS (lysine

decarboxylation), NC (decarboxylation control), ORN (ornithine

decarboxylation)]. Isolate designations: CHI – Chicago; JU – June, A1

– early August, A21 – late August

b



TCTCTC

AAA

BBB

FJG

Loca

tion

TCTCTC

AAA

BBB

FJG

100

908070605040302010

CHICHICHICHITCCHICHICHICHICHICHICHICHICHICHICHICHICHITCCHITCCHICHICHICHICHITCCHICHICHICHITCCHICHICHICHICHICHICHICHICHITCCHICHICHICHICHICHICHICHICHITCCHICHITCTCTCTCTCTCTCCHICHICHICHICHICHITCTCTCTCCHICHICHICHICHICHI

TCTCTC

AAA

CHITCTCTCCHICHICHITCTCCHICHICHITCCHICHICHICHICHITCTCTCTCCHICHITC

A21A21A21A21OCA21A21A21A1A1A21A1A21JUA21JUJUA1MYA1MYA1A1A1A1A1MYA21A21A21JUMYJUA21JUJUJUA1A1A21A1MYA21A21A21A21A21A21A21A21A21AA1A21OCOCAOCMYMYMYJUJUJUJUJUJUAAAAA21A21A21A21A1JUA21AAAA21JUJUMYMYJUA1A1MYJUJUA1A21A21AOCOCOCA1A21MY

AAAAAABCSB2B8AB8CBBDCB82SB32SB2SB2SB1B8SB42BBCB1DCAAAB7B5AB71DDDDDDDDDBB6BAABA222CCCCCDBBBBBBBBB6DCBBBACC22BSB4B62DDSB3AABBBBB5B2

6845B2511271832613A3J3

241K4351C423211292B3121049856K26ECHALHG547867BDCA97810414FJG132FC42B1E522310ECBD11D

Similarity (%)

Dat

e

Sam

ple

Isol

ate

A

B

C

D

G

E

E

F



may be a significant source of these indicator bacteria

recommended by the USEPA for monitoring recreational

waters (USEPA 2000). Other authors have noted the large

numbers of faecal bacteria associated with the faeces of gulls

and other shore birds and have documented the impact gull

faeces may have on bacterial contamination of reservoirs,

beach sediments and coastal waters (Jones et al. 1978;

Levesque et al. 1993, 2000; Hatch 1996; Alderisio and

Deluca 1999; Jones and Obiri-Danso 1999; Obiri-Danso and

Jones 2000). However, no prior studies have specifically

addressed the number of E. coli or enterococci in gull faeces.

Levesque et al. (1993, 2000) reported that 95–99% of faecal

coliforms in gull faeces were E. coli. E. coli concentrations

per gram of gull faeces in our study (<1Æ0 · 105 to 1Æ9 · 109,

Table 1) were similar to faecal coliform concentrations

found in other studies (1Æ1 · 106 to 1Æ1 · 1010; Levesque

et al. 1993; Alderisio and Deluca 1999; Jones and Obiri-

Danso 1999). A similar range in the concentration of

enterococci per gram of faeces (2Æ0 · 104 to 2Æ4 · 108) was

also observed in our study. Gould and Fletcher (1978)

determined that the average wet weight of faeces excreted by

different gull species ranged from 11Æ2 to 24Æ9 g day)1. This

would result in an average daily load of E. coli and entero-

cocci from one gull on the Chicago beach up to 1Æ2 · 1010

and 4Æ2 · 108, respectively (3Æ5 · 108 and 1Æ4 · 109,

respectively, for Traverse City).

We observed no temporal or geographic trend in E. coli or

enterococci concentration in gull faeces. Likewise, Levesque

et al. (2000) demonstrated little difference in the numbers of

faecal coliforms in gull faecal material with respect to gull

age group, colony or sampling date. In addition, a study by

Alderisio and Deluca (1999) indicated a fairly stable

concentration of faecal coliforms (107–108 g)1) over four

seasons across two sampling years.

Within the Great Lakes region, populations of the ring-

billed gull (Larus delawarensis) have been increasing in

recent years, especially in Illinois (Sauer et al. 2002). In

recent years, more than 7000 breeding pairs of ring-billed

gulls were counted at six sites along the Chicago shoreline,

and as many as 17 700 breeding pairs at nearby locations

outside the Chicago metropolitan area (F. Cuthbert, Uni-

versity of Minnesota, Minneapolis, MN, USA, personal

communication). As many as 13 000 breeding pairs have

been counted at an island located west of Grand Traverse

Bay. There are over 250 000 breeding pairs of ring-billed

gulls in the Great Lakes region, accompanied each year by

non-breeding immatures up to 2 years of age. Other water

birds found along the Chicago and Traverse City shorelines

include mallard ducks and Canada geese, for which the

population numbers have also been increasing in the Great

Lakes region (Sauer et al. 2002). However, Canada geese

were rarely seen on the beaches during the swimming

season, and mallard ducks were less numerous than gulls at

both beaches.

As noted by others for both Europe and the US (Hatch

1996; Jones and Obiri-Danso 1999), large and increasing

populations of ring-billed gulls may bring increased risk of

human exposure to endemic bacterial pathogens (Campylo-

bacter spp.) as well as those acquired through feeding at

landfills, animal pastures and sewage disposal sites (Salmon-

ella, other enteric bacteria). The increasing populations of

ring-billed gulls in the Great Lakes, combined with the large

numbers of faecal bacteria they carry, may constitute a major

non-point source of water pollution. With this concern in

mind, means to discriminate gull faecal pollution from other

potential sources would be especially valuable to beach

managers and water pollution control authorities.

Our results indicate a high degree of intra-species

population variation for E. coli (defined by rep-PCR profiles

and Vitek� biotype) in gull faeces taken from Lake

Michigan beaches. Rep-PCR has been shown to be a

powerful method for identifying intra- and inter-species

genotypic relations, and rep-PCR profiles have been shown

to be correlated with intra-specific phenotypic characteris-

tics in Xanthomonas, Pseudomonas and Ochrobactrum spp.

(Lemanceau et al. 1995; Lebuhn et al. 2000; Rademaker

et al. 2000). Johnson and O’Bryan (2000) showed that E. coli

rep-PCR profiles were related to E. coli clusters based on

multiple locus enzyme electrophoresis (MLEE). In our

study, E. coli isolates with identical genotypes had very

similar or identical Vitek� phenotypes. Although our sample

set was small, we found a median of eight genotypes in four

Fig. 3 Dendrogram (UPGMA cluster analyses based on Dice corre-

lation coefficient) of Escherichia coli rep-PCR profiles for 73 isolates

obtained from gull faeces collected on beaches in Chicago, IL (CHI)

and 29 isolates from gull faeces collected at Traverse City, MI (TC)

beaches. Isolate designations as for Figure 2, except TC – Traverse

City; A – August; OC – October

y = 3·1861Ln(x) – 0·3422

R2 = 0·7852

0123456789

10

0 2 4 6 8 10 12

Number of isolates

Num

ber

of g

enot

ypes

Fig. 4 Relation between number of Escherichia coli rep-PCR geno-

types and number of isolates obtained per gull faecal sample

b



casseliflavus

durans

A

CHI A21 C1

CHI A21 C5

Distance

TC A D4 casseliflavusTC A D3 casseliflavus

CHI A21 B2BCHI A21 B4

CHI A21 C4CHI A21 A6CHI A21 A2

CHI A21 A5CHI A21 A3

CHI A21 D1CHI A21 D4CHI A21 B5CHI A21 B6

CHI A21 A1

CHI A21 D2CHI A21 D3

CHI A21 D5



CHI A21 D10hiraegallinarum

TC A D5 hirae

TC A B2TC A D2 durans

TC A D1hirae

TC A B1 duransTC A B5TC A B4 durans

TC A B3 durans faecium 1faecium 2CHI JU C3 faecium 1CHI JU C1gallinarumCHI JU C8 faecalis

CHI A21 B1CHI JU E1 faecalisCHI JU C2

CHI A21 C3saccharolyticus

CHI JU E4 faecalisCHI JU E3 faecalis

CHI JU E2 faecalis

CHI A21 C2CHI JU C6

CHI JU F6 faecalis

CHI JU B3 faecalis

CHI JU B5CHI JU F2 faecalis

CHI JU E5 faecalis

CHI JU F5

CHI JU F3

CHI JU B2 faecalis

CHI JU A5 faecalis

CHI JU A1 aviumCHI JU B1 avium

avium

CHI JU B4CHI JU F4CHI JU A4

CHI A21 B2A

CHI JU A3TC OC B faecalis

TC OC A faecalisCHI JU A2

B

C

0·00·51·01·5



faecal samples for which we collected 10 isolates, all but one

faecal sample with multiple isolates yielded more than one

genotype, and we found no instance where identical

genotypes occurred in more than one seagull faecal sample.

Our data suggest that large numbers of E. coli isolates from

gull faecal samples, and more than one isolate per sample,

would be required to fully characterize intra-specific pop-

ulation diversity, and to adequately characterize the popu-

lation for source-determination studies.

We observed structure in the rep-PCR profile clusters at

‡85% and around 70% similarity. CHI isolates that clustered

at ‡85% similarity were typically from the same sampling

date, shared identical or highly similar phenotypes, and

retained their close association even after TC isolates were

added to the data set. Five TC isolates were 85–90% similar

to CHI isolates. The close association of some CHI and TC

isolates in the same clusters suggests some population overlap

at the two geographically distinct sites. At >70% similarity of

banding patterns, CHI and TC isolates exhibited broadly

similar banding patterns, and for CHI isolates, such clusters

were accompanied by some unique phenotypic characteris-

tics (Figure 2). Previous studies (Lemanceau et al. 1995;

Johnson and O’Bryan 2000; Lebuhn et al. 2000; Rademaker

et al. 2000) have shown that large and consistent variations in

banding pattern have intra-specific genotypic and phenotypic

significance. Much more data on other phenotypic and

genotypic features of our isolates would be required to

establish the population significance of these broad clusters.

Nevertheless, our results suggest that several intra-species

groups of E. coli occur within gull faeces at two geograph-

ically separated Lake Michigan beaches and that the

proportion of isolates in these groups varies temporally.

Fig. 5 Dendrogram (agglomerative clustering, Ward method) of

binary API� Rapid ID 32 Strep biochemical response profiles of

enterococci isolates obtained from gull faeces collected on beaches in

Chicago, IL, and Traverse City, MI, compared to the standard

responses (shown in italics) of Ent. avium, casseliflavus, durans, faecalis,

faecium 1, faecium 2, gallinarum, hirae, and saccharolyticus. Isolate

designations: CHI, Chicago; TC, Traverse City; JU, June; A21 or AG,

August; OC, October

b

Table 2 Antibiotic resistance patterns of

enterococci from gull faecesAntibiotic resistance level (lg ml)1)

Isolate Cluster* Date Strep, >1000� Gent, >500� Van, 8–16� Tet, ‡16§ Amp, ‡16§

CHI JU B3 A1 JU >256 64 2 >256 1

CHI JU E5 A1 JU >256 64 2 >256 1

CHI JU F6 A1 JU >256 16 4 0Æ5 0Æ5CHI JU A2 A2 JU 128 16 1 0Æ125 0Æ19

CHI JU F4 A2 JU >256 24 4 1 0Æ5TC OC A A2 OC >256 64 3 64 0Æ75

TC OC B A2 OC >256 24 3 48 0Æ25

CHI JU B1 A3 JU 64 4 1Æ5 0Æ38 0Æ38

CHI JU C8 A4 JU >256 12 1 0Æ25 0Æ5CHI JU E4 A4 JU 48 4 1 0Æ38 0Æ25

CHI JU C3 B JU 32 12 0Æ75 48 1Æ5TC AG B2 B A >256 1024 3 >256 3

TC AG B3 B A >256 48 0Æ38 >256 2

TC AG D1 B A 96 24 2 0Æ75 0Æ38

CHI A21 B6 C1 A >256 16 2 0Æ38 0Æ5CHI A21 D2 C1 A 128 12 2 0Æ38 0Æ5CHI A21 D4 C1 A >256 24 2 128 0Æ5CHI A21 D9 C1 A >256 48 2 >256 0Æ75

CHI A21 A2 C2 A 16 8 6 0Æ75 0Æ5CHI A21 B4 C2 A 32 12 3 0Æ25 2

CHI A21 C5 C2 A 24 6 4 0Æ25 0Æ38

TC AG D3 C2 A 32 8 6 1 1

TC AG D4 C2 A >256 24 6 96 3

*Cluster as shown in Figure 5.

�Concentration above which isolate should be tested for high-level resistance.

�Range for intermediate resistance requiring further testing.

§Defined resistance level.

Strep, streptomycin; Gent, gentamicin; Van, vancomycin; Tet, tetracycline; Amp, ampicillin.



Genomic characteristics of E. coli populations have also been

observed to vary temporally in human waste and faeces from

feral mice (Gordon 1997; Gordon et al. 2002). This variation

would be consistent with the concepts of E. coli resident and

immigrant strains suggested by earlier studies (Selander

et al. 1987).

Enterococci biochemical profiles support the concept of

bacterial population dynamics in gull faeces suggested by

the E. coli results. Most (23 of 25) June CHI Enterococcus

isolates had biochemical profiles identified as similar to

Ent. faecalis. In contrast, August CHI Enterococcus isolates

had different API� biochemical profiles than those of June

CHI isolates, and where species could be assigned, August

Enterococcus species were different from those obtained in

June. API biochemical tests alone cannot be used to assign

environmental enterococci isolates to species (Muller et al.

2001; Svec et al. 2002). The API tests have been developed

primarily for characterization of clinical isolates of entero-

cocci, and environmental isolates exhibit test responses

inconsistent with those established for clinical isolates of

the same species (Muller et al. 2001). Nevertheless, when

recently developed genotyping methods for enterococci have

been applied, biochemical test responses were consistent

within source-specific genotypes (ecovars; Svec et al. 2002)

and were useful in distinguishing closely related phylogenet-

ic groups of enterococci from less-related groups (Devriese

et al. 2002). In our study, antibiotic resistance patterns did

not closely parallel groups defined by biochemical test

response. Likewise, Muller et al. (2001) found little corre-

lation between antibiotic resistance patterns and species or

genotypes of enterococci isolated from forage grass. Never-

theless, some isolates in our study exhibited high-level

resistance to medically significant antibiotics.

The causes and ecological significance of population

variation in E. coli and enterococci in gull faeces remain to

be determined. Our results suggest the existence of ecovars

of both E. coli and enterococci in gull faeces, which might be

related to feeding ecology, age structure or colony charac-

teristics not determined in this study. In particular, gull diets

may be extremely variable. They are opportunistic feeders,

feeding on the nearest food supply (fish, worms, insects,

trash, etc.; Weseloh and Blokpoel 1979; Drury 1980; Hatch

1996) near the lakeshore, at landfills, in pastures or in city

parking lots. Early studies suggested that hosts are subject to

continuous immigration of E. coli strains from the environ-

ment, with food being a major source (Selander et al. 1987).

The abundance of Salmonella in gulls may be affected by

season and by age-specific differences in feeding ecology

(Hatch 1996). Studies performed on other animals have

shown that the intestinal microflora can be affected by small

changes in diet (Selander et al. 1987; Netherwood et al.1999; Souza et al. 1999; Leser et al. 2000). However, pigeons

have a very characteristic and host-specific enterococcal flora

(Baele et al. 2002). Further studies of E. coli and enterococci

population dynamics in gull faeces might lead to improved

understanding of gut ecology, and provide insight into the

significance of ecovars of these common bacterial genera.

The variation in faecal indicator bacteria populations seen

in this study is also significant in the context of current

efforts to determine the sources of faecal bacterial pollution

to ambient waters. First, our results suggest that the large

degree of variation in population characteristics of both E. coli

and enterococci in gull faeces will require extensive sampling

for adequate characterization. Second, our results may help

to better understand the variable success in correct classifi-

cation (ARCC) of bacteria with respect to source in other

studies (Hagedorn et al. 1999; Parveen et al. 1999; Dombek

et al. 2000; Carson et al. 2001). The variability in ARCC has

been similar regardless of method used. Therefore, it is likely

due to factors that are not method-related. Such factors may

include features of gut/faecal microbiology that exhibit

temporal, geographic or ecological variability. Intra-specific

variation in E. coli or Enterococcus genomic structure within a

given animal population may affect both the ARCC and the

reliability of various genomic typing procedures to correctly

classify these bacteria from various animal sources. Results of

this study suggest that variation in physiological and genomic

characteristics of E. coli and enterococci occurs at many

levels: within faecal samples, between faecal samples collec-

ted on the same date and between samples collected on

different dates. These variations are all important consider-

ations when building libraries to be used in faecal contam-

ination source determination.

ACKNOWLEDGEMENTS

We thank all those who helped in sample collection and

analyses including Joel Underwood from the US Geological

Survey, Lansing, MI; Maria Goodrich, from the US

Geological Survey, Porter, IN; and Brenda Berlowski and

Heather Gutzman from the US Geological Survey National

Wildlife Disease Center, Madison, WI. We would also like

to thank the City of Chicago for administrative, field and

technical support. This project was funded in part by the

City of Chicago, with contributions from the US Geological

Survey and the Michigan Great Lakes Protection Fund.

REFERENCES

Alderisio, K.A. and Deluca, N. (1999) Seasonal enumeration of fecal

coliform bacteria from the feces of ring-billed gulls (Larus

delawarensis) and Canada geese (Branta canadensis). Applied and

Environmental Microbiology 65, 5628–5630.

American Public Health Association (1998) Standard Methods for the

Examination of Water and Wastewater, 20th edn. Washington DC:

American Public Health Association.



Baele, M., Devriese, L.A., Butaye, P. and Haesebrouck, F. (2002)

Composition of enterococcal and streptococcal flora from pigeon

intestines. Journal of Applied Microbiology 92, 348–351.

Carson, C.A., Shear, B.L., Ellersieck, M.R. and Asfaw, A. (2001)

Identification of fecal Escherichia coli from humans and animals

by ribotyping. Applied and Environmental Microbiology 67, 1503–

1507.

Devriese, L.A., Vancanneyt, M., Descheemaker, P., Baele, M.,

Landuyt, H.W.V., Gordts, B., Butaye, P., Swings, J. and

Haesebrouck, F. (2002) Differentiation and identification of Entero-

coccus durans, E. hirae, and E. villorum. Journal of Applied Micro-

biology 92, 821–827.

Dombek, P.E., Johnson, L.K., Zimmerley, S.T. and Sadowsky, M.J.

(2000) Use of repetitive DNA sequences and the PCR to differen-

tiate Escherichia coli isolates from human and animal sources. Applied

and Environmental Microbiology 66, 2572–2577.

Drury, W.H. (1980) Herring Gull, Hinterland Who’s Who, no. CW69-

4/10-2002E/. Ottawa, Canada: Canadian Wildlife Service, Envi-

ronment Canada.

Gordon, D.M. (1997) The genetic structure of Escherichia coli

populations in feral house mice. Microbiology 143, 2039–2046.

Gordon, D.M., Bauer, S. and Johnson, J.R. (2002) The genetic

structure of Escherichia coli populations in primary and secondary

habitats. Microbiology 148, 1513–1522.

Gould, D.J. and Fletcher, M.R. (1978) Gull droppings and their effects

on water quality. Water Research 12, 665–672.

Hagedorn, C., Robinson, S.L., Filtz, J.R., Grubbs, S.M., Angier, T.A.

and Reneau, R.B. Jr. (1999) Determining sources of fecal pollution

in a rural Virginia watershed with antibiotic resistance patterns in

fecal streptococci. Applied and Environmental Microbiology 65, 5522–

5531.

Harwood, V.J., Whitlock, J. and Withington, V. (2000) Classification of

antibiotic resistance patterns of fecal indicator bacteria by discrimi-

nant analysis: use in predicting the source of fecal contamination in

subtropical waters. Applied and Environmental Microbiology 66, 3698–

3704.

Hatch, J.J. (1996) Threat to public health from gulls. International

Journal of Environmental Health Research 6, 5–16.

Johnson, J.R. and O’Bryan, T.T. (2000) Improved repetitive-element

PCR fingerprinting for resolving pathogenic and nonpathogenic

phylogenetic groups within Escherichia coli. Clinical and Diagnostic

Laboratory Immunology 7, 265–273.

Jones, F., Smith, P. and Watson, D.C. (1978) Pollution of water supply

catchment by breeding gulls and the potential environmental health

implications. Journal of the Institution of Water Engineers and

Scientists 32, 469–482.

Jones, K. and Obiri-Danso, K. (1999) Non-compliance of beaches with

the EU directives of bathing water quality: evidence of non-point

sources of pollution in Morecambe Bay. Journal of Applied

Microbiology Symposium Supplement 85, 101S–107S.

Kaspar, C.W., Burgess, J.L., Knight, I.T. and Colwell, R.R. (1990)

Antibiotic resistance indexing of Escherichia coli to identify sources

of fecal contamination in water. Canadian Journal of Microbiology 36,

891–894.

Krumperman, P.H. (1983) Multiple antibiotic resistance indexing of

Escherichia coli to identify high-risk sources of fecal contamination of

foods. Applied and Environmental Microbiology 46, 165–170.

Lebuhn, M., Achouak, W., Scholter, M., Berge, O., Meier, H.,

Barakat, M., Hartmann, A. and Heulin, T. (2000) Taxonomic

characterization of Ochrobactrum sp. isolates from soil samples and

wheat roots, and description of Ochrobactrum tritici sp. nov. and

Ochrobactrum grignonense sp. nov. International Journal of Systematic

and Evolutionary Microbiology 50, 2207–2223.

Lemanceau, P., Corberand, T., Gardan, L., Latour, X., Laguerre,

Boeufgras, J. and Alabouvette, C. (1995) Effect of two plant species,

flax (Linum usitatissinum L.) and tomato (Lycopersicon esculentum

Mill.), on the diversity of soilborne populations of fluorescent pseu-

domonads. Applied and Environmental Microbiology 61, 1004–1012.

Leser, T.D., Lindecrona, R.H., Jensen, T.K., Jensen, B.B. and Møller,

K. (2000) Changes in bacterial community structure in the colon of

pigs fed different experimental diets and after infection with

Brachyspira hyodysenteriae. Applied and Environmental Microbiology

66, 3290–3296.

Levesque, B., Brousseau, P., Simard, P., Dewailly, E., Meisels, M. and

Joly, J. (1993) Impact of the ring-billed gull (Larus delawarensis) on

the microbiological quality of recreational water. Applied and


Levesque, B., Brousseau, P., Bernier, F., Dewailly, E. and Joly, J.

(2000) Study of the bacterial content of ring-billed gull droppings in

relation to recreational water quality. Water Research 34, 1089–1096.

Muller, T., Ulrich, A., Ott, E.-M. and Muller, M. (2001) Identification

of plant-associated enterococci. Journal of Applied Microbiology 91,

268–278.

National Committee for Clinical Laboratory Standards (2002) Per-

formance Standards for Antimicrobial Susceptibility Testing. 12th

Informational Supplement, National Committee for Clinical Labor-

atory Standards Document M100-S12, Wayne, Pennsylvania.

Netherwood, T., Gilbert, H.J., Parker, D.S. and O’Donnell, A.G.

(1999) Probiotics shown to change bacterial community structure in

the avian gastrointestinal tract. Applied and Environmental Microbio-

logy 65, 5134–5138.

Obiri-Danso, K. and Jones, K. (2000) Intertidal sediments as reservoirs

for hippurate negative camplybacters, salmonellae and faecal indi-

cators in three EU recognized bathing waters in North West

England. Water Research 34, 519–527.

Parveen, S., Murphree, R.L., Edmiston, L., Kaspar, C.W., Portier

K.M. and Tamplin, M.L. (1997) Association of multiple-antibiotic-

resistance profiles with point and nonpoint sources of Escherichia coli

in Apalachicola Bay. Applied and Environmental Microbiology 63,

2607–2612.

Parveen, S., Portier, K.M., Robinson, K., Edmiston, L. and Tamplin,

M.L. (1999) Discriminant analysis of ribotype profiles of Escherichia

coli for differentiating human and nonhuman sources of fecal

pollution. Applied and Environmental Microbiology 65, 3142–3147.

Parveen, S., Hodge, N.C., Stall, R.E., Farrah, S.R. and Tamplin, M.L.

(2001) Phenotypic and genomic characterization of human and

nonhuman Escherichia coli. Water Research 35, 379–386.

Rademaker, J.L.W. and de Bruijn, F.J. (1997) Characterization and

classification of microbes by rep-PCR genomic fingerprinting and

computer-assisted pattern analysis. In DNA Markers: Protocols,

Applications and Overviews ed. Caetano-Anolles, G. and Gresshoff,

P.M. pp. 151–171. New York: John Wiley.

Rademaker, J.L.W., Hoste, B., Louws Frank, J., Kersters, K., Swings,

J., Vauterin, L., Vauterin, P. and de Bruijn Frans, J. (2000)



Comparison of AFLP and rep-PCR genomic fingerprinting with

DNA–DNA homology studies: Xanthomonas as a model system.

International Journal of Systematic and Evolutionary Microbiology 50,

665–677.

Rose, J.B., Atlas, R.M., Gerba, C.P., Gilchrist, M.R., LeChevallier,

M.W., Sobsey, M.D. and Yates, M.V. (2001) Microbial Pollutants in

our Nation’s Water: Environmental and Public Health Issues. Wash-

ington DC: American Society for Microbiology.

Sauer, J.R., Hines, J.E. and Fallon, J. (2002) The North American

breeding bird survey, results and analysis 1966–2001. Version 2002.1.

USGS Patuxent Wildlife Research Center, Laurel, MD.

Selander, R.K., Caugant, D.A. and Whittam, T.S. (1987) Genetic

structure and variation in natural populations of Escherichia coli. In

Escherichia coli and Salmonella typhimurium, Cellular and Molecular

Biology, Vol. 2 ed. Neidhardt, F.C. pp. 1625–1645. Washington DC:

American Society for Microbiology.

Souza, V., Rocha, M., Valera, A. and Eguiarte, L.E. (1999) Genetic

structure of natural populations of Escherichia coli in wild hosts on

different continents. Applied and Environmental Microbiology 65,

3373–3385.

Svec, P., Devriese, L.A., Sedlacek, I., Baele, M., Vancanneyt, M.,

Haesebrouck, F., Swings, J. and Doskar, J. (2002) Characterization

of yellow-pigmented and motile enterococci isolated from intestines

of the garden snail Helix aspera. Journal of Applied Microbiology 92,

951–957.

USEPA (1986) Ambient water quality criteria for bacteria-1986.

EPA440/5-84-002. Washington DC: United States Environmental

Protection Agency.

USEPA (2000) Improved enumeration methods for the recreational

water quality indicators: enterococci and Escherichia coli. EPA/821/

R-97/004. Washington DC: United States Environmental Protec-

tion Agency.

Versalovic, J., Koeuth, T. and Lupski, J.R. (1991) Distribution of

repetitive DNA sequences in eubacteria and application to finger-

printing of bacterial genomes. Nucleic Acids Research 19, 6823–6831.

Wallace, J.S., Cheasty, T. and Jones, K. (1997) Isolation of Vero

cytotoxin-producing Escherichia coli O157 from wild birds. Journal of

Applied Microbiology 82, 399–404.

Weseloh, D.V. and Blokpoel, H. (1979). Ottawa, Canada: Canadian

Wildlife Service, Environment Canada.

Wiggins, B.A. (1996) Discriminant analysis of antibiotic resistance

patterns in fecal streptococci, a method to differentiate human and

animal sources of fecal pollution in natural waters. Applied and


Wiggins, B.A., Andrews, R.W., Conway, R.A., Corr, C.L., Dobratz,

E.J., Dougherty, D.P., Eppard, J.R., Knupp, S.R., Limjoco, M.C.,

Mettenburg, J.M., Rinehardt, J.M., Sonsino, J., Torrijos, R.L. and

Zimmerman, M.E. (1999) Use of antibiotic resistance analysis to

identify nonpoint sources of fecal pollution. Applied and Environ-

mental Microbiology 65, 3483–3486.



Abundance and characteristics of the recreational water quality ...

Documents

Transcript of Abundance and characteristics of the recreational water quality ...