This article was downloaded by: [University of Leeds]On: 26 April 2013, At: 02:48Publisher: Taylor & FrancisInforma Ltd Registered in England and Wales Registered Number: 1072954Registered office: Mortimer House, 37-41 Mortimer Street, London W1T3JH, UK

International Journal ofEnvironmental HealthResearchPublication details, including instructions forauthors and subscription information:http://www.tandfonline.com/loi/cije20

Antibiotic resistance in soiland water environmentsNwadiuto Esiobu a , Lisa Armenta a & Joseph Ike aa Biology Department, College of Liberal Arts,Florida Atlantic University, 2912 College Avenue,Davie, FL, 33314, USAVersion of record first published: 21 Jul 2010.

To cite this article: Nwadiuto Esiobu , Lisa Armenta & Joseph Ike (2002): Antibioticresistance in soil and water environments, International Journal of EnvironmentalHealth Research, 12:2, 133-144

To link to this article: http://dx.doi.org/10.1080/09603120220129292

PLEASE SCROLL DOWN FOR ARTICLE

Full terms and conditions of use: http://www.tandfonline.com/page/terms-and-conditions

This article may be used for research, teaching, and private studypurposes. Any substantial or systematic reproduction, redistribution,reselling, loan, sub-licensing, systematic supply, or distribution in any formto anyone is expressly forbidden.

The publisher does not give any warranty express or implied or make anyrepresentation that the contents will be complete or accurate or up todate. The accuracy of any instructions, formulae, and drug doses shouldbe independently verified with primary sources. The publisher shall notbe liable for any loss, actions, claims, proceedings, demand, or costs or

damages whatsoever or howsoever caused arising directly or indirectly inconnection with or arising out of the use of this material.

Dow

nloa

ded

by [

Uni

vers

ity o

f L

eeds

] at

02:

48 2

6 A

pril

2013

Correspondence: Nwadiuto Esiobu, Ph.D., Biology Department, College of Liberal Arts, Florida AtlanticUniversity, 2912 College Avenue, Davie, FL 33314, USA. Tel.: + 1-954-236-1128; Fax:+ 1-954-236-1099; E-mail: nesiobu@fau.edu

ISSN 0960-3123 printed/ISSN 1369-1619 online/02/020133-12 © 2002 Taylor & Francis LtdDOI: 10.1080/09603120220129292

International Journal ofEnvironmental Health Research 12, 133–144 (2002)

Antibiotic resistance in soil and waterenvironmentsNWADIUTO ESIOBU, LISA ARMENTA and JOSEPH IKE

Biology Department, College of Liberal Arts, Florida Atlantic University, 2912 College Avenue, Davie FL 33314,USA

Seven locations were screened for antibiotic-resistant bacteria using a modified agar dilution technique.Isolates resistant to high levels of antibiotics were screened for r plasmids. Low-level resistance(25 mg ml–1) was widespread for ampicillin, penicillin, tetracycline, vancomycin and streptomycin but notfor kanamycin. Resistant populations dropped sharply at high antibiotic levels, suggesting that intrinsicnon-emergent mechanisms were responsible for the multiple drug resistance exhibited at low doses. Dairyfarm manure contained significantly (P < 0.01) more (%) resistant bacteria than the other sites. Bacteriaisolated from a dairy water canal, a lake by a hospital and a residential garden (fertilized by farm manure)displayed resistance frequencies of 77, 75 and 70%, respectively. Incidence of tetracycline resistance wasmost prevalent at 47–89% of total bacteria. Out of 200 representative isolates analyzed, Pseudomonas,Enterococcus-like bacteria, Enterobacter and Burkholderia species constituted the dominant reservoirs ofresistance at high drug levels (50–170 mg ml–1). Plasmids were detected in only 29% (58) of these bacteriawith tetracycline resistance accounting for 65% of the plasmid pool. Overall, resistance trends correlatedto the abundance and type of bacterial species present in the habitat. Environmental reservoirs of resistanceinclude opportunistic pathogens and constitute some public health concern.

Keywords: Antibiotic resistance; environmental reservoirs of resistance; bacteria; r plasmids; impact ofnon-therapeutic antibiotic usage.

Introduction

The growing challenge of the emergence and spread of antibiotic resistance has been the subjectof several investigations and reviews (Linton 1986; Bates et al. 1994; Jacobs-Reitsma et al.1994; Piddock 1996; McDonald et al. 1997; Anonymous 1998; Levy 1998; Schnabel and Jones1999). Antibiotics have been actively employed in human and veterinary medicine for thetreatment of infectious diseases and as feed additives for about 50 years. Antibiotic productionsoared from 2 million pounds in 1954 to over 50 million pounds presently in the United Statesalone (Levy 1998). It is estimated that more than 70% of the annual antibiotic out-put is fed tochicken, pigs and cows for non-therapeutic purposes and used in agriculture to protect plantsand fruits (Anonymous 2001).

Dow

nloa

ded

by [

Uni

vers

ity o

f L

eeds

] at

02:

48 2

6 A

pril

2013

134 Esiobu et al.

Research has consistently linked the use of antibiotics in food production to increasedincidence of drug-resistant pathogens in animals and man (Bates et al. 1994; Piddock 1995;Blanco et al. 1997; McDonald et al. 1977; Al-Mustafa and Al-Ghamdi 2000) and animal gutflora is seen as a major reservoir of the R plasmid (Linton and Hinton 1984). However, asystematic evaluation of the impact of non-discriminate use of antibiotics on environmentalmicroorganisms is in its infancy. Earlier studies on antibiotic resistance in aquatic habitat forinstance, concentrated on its incidence in bacteria of fecal origin or those resistant to heavymetals (Niemi et al. 1983; Calomiris et al. 1984). Walter and Vennes (1985) among others haveshown that waste effluents from hospitals contain higher levels of antibiotic-resistant entericbacteria than waste effluents from other sources, but the effect of such releases on indigenousbacteria is poorly documented. The discharge of wastewater from a pharmaceutical plant hasbeen associated with increased prevalence of both single and multiple resistance amongAcinetobacter species (Guardabassi et al. 1998). In contrast, Jones et al. (1986) found that theincidence of resistance in indigenous bacteria was by far greater than that of introduced speciesand remained high even in undisturbed habitats. Schnabel and Jones (1999) correlated increasedincidence of tetracycline resistance in a Michigan apple orchard with the use of antibiotics in thefield.

Treated wastewater, garbage and manures may contain resistant bacteria even when levels ofroutinely used indicator microbes are in compliance. Therefore certain wastewater andbiological fertilizers (example: composts and manures) may require scrutiny for resistantbacteria before disposal into the environment. Such regulations need scientific support data onwhether or not human activities and other anthropogenic factors affect resistance frequencies inindigenous bacteria. It is also pertinent to test the implication of intrinsic and/or introducedresistance on the overall pattern of antibiotic resistance in a given environment.

The questions that this study attempts to address include: (1) Which bacterial speciesconstitute the principal reservoirs of resistance in the environment? (2) What proportion of theseresistant population carry plasmids? (3) What is the relative resistance patterns for variousantibiotics? And, finally, (4) What are the public health implications? Seven antibiotics,penicillin G (PEN), ampicillin (AMP), vancomycin (VAN), chlorotetracycline (CHTET) &oxytetracycline (OXTET), kanamycin (KAN) & streptomycin (STR), were chosen for thisinvestigation. Vancomycin is a highly valued antibiotic used in treating serious infections suchas sepsis caused by drug-resistant pathogens and to which there are few alternates (FDA 2000).Kanamycin was chosen because it is widely used in antibiotic sensitivity testing onenvironmental samples (Jones et al. 1986). Other antibiotics employed in the study arecommonly used in local medical and veterinary practice. Finally, the mechanisms of resistanceto these antibiotics are very different and cover the major mechanisms, such as alteration oftarget site, modification or detoxification, extrusion from cell and enzyme substitution.

Materials and methods

Sampling and sites

Table 1 shows the seven sites chosen to reflect varying environments. Composite samples werecollected from the top 10 cm at each location in duplicates during four sampling periods (twoin the Fall and two in early Spring). Fifty grams of each soil sample were analyzed. Soilsampling methods conformed to standards described by van Elsas and Smalla (1997). For thewater samples, the canals or lakes were vigorously stirred prior to aseptic collection of 500 mlof samples from 20 cm below water surface.

Dow

nloa

ded

by [

Uni

vers

ity o

f L

eeds

] at

02:

48 2

6 A

pril

2013

Antibiotic resistance in the environment 135

Antibiotics

Penicillin G (PEN), ampicillin (AMP), vancomycin (VAN), chlorotetracycline (CHTET) andoxytetracycline (OXTET), kanamycin (KAN) & streptomycin (STR) (Sigma, St. Louis, MO,USA) were employed in the study. For preliminary screening, all antibiotics were used at aconcentration of 15 mg ml–1 and later increased to 25 mg ml–1. Higher doses of vancomycin,kanamycin and tetracycline were used to further delineate the pattern of resistance to theseantibiotics.

Protocol

A modified agar dilution screening method was used to analyze all samples within 20 h ofcollection. Serial 10-fold dilutions of each sample were prepared in physiological saline andaliquots were pour plated on Mueller Hinton Agar (Difco) modified by incorporating theappropriate antibiotic at molten state. Identical sample aliquots were plated on regular MullerHinton Agar (no antibiotics) to estimate total bacterial density. After 24 – 48 h of incubation at30°C, colony-forming units (CFU) on the plates were estimated using a colony counter.Resistant colonies were replica plated on other antibiotic media to test for multiple resistance.Visible tangible growth on the plates was scored as positive for resistance, whereas no growthto trace colony development was recorded as susceptible.

Next, using the same set of samples in a similar protocol as that described above, threeantibiotics: kanamycin (35 and 100 mg ml–1), tetracycline (35 and 170 mg ml–1) and vancomycin(50 mg ml–1) were employed at clinically significant concentrations. The drugs were incorpo-rated in to various selective and differential media (Pseudomonas agar, Eosine methylene blueagar (EMB) and Phenyl ethyl alcohol agar (PEA)). An estimate of the total culturable resistantEnterobacteriaceae and Gram-negatives (EMB), total resistant Gram-positives (PEA) and totalresistant pseudomonads were obtained by plate counts and expressed as colony-forming units(CFU) per 100 ml water or CFU per g solid, after incubation at 37°C. A presumptive detectionand enumeration of enterococci involved replica plating the colonies on antibiotic-modified

Table 1. Description of samples and sampling sites

Sample code Description Remark

DFS Dairy farm soil Soil mixed with fresh cow manure

DFAM Dairy farm animal manure Composting animal manure

DFWC Water canal by dairy farm Receives farm run-off

RGS Residential garden soil Far from hospitals, no history ofuse of antibiotics

CLH City lake by hospital Recycles irrigation water;receives run-off from environment

PPWC Public park water canal Collects run-off from recreationalpark

RLW Residential lake water Man-made lake. No nearbyhospital or farm.

Dow

nloa

ded

by [

Uni

vers

ity o

f L

eeds

] at

02:

48 2

6 A

pril

2013

136 Esiobu et al.

PEA (a medium used to isolate Gram-positive organisms) on to bile esculin azide agar by pointinoculation (24 h at 40°C) (Anonymous 1999). The EPA-approved mEI agar would have beena good alternative for direct enumeration of enterococci; however, this study aimed atdocumenting total resistant Gram-positive bacteria as well.

Identification of species composition and reservoirs of resistance at the sites

Based on source, colony color and morphology, at least five representatives of culturally distinctbacteria from antibiotic-free and antibiotic-amended media were purified for identification ateach sampling time (500 total). After preliminary characterization by Gram staining,microscopy, catalase and oxidase tests, the isolates were grouped into 20 operational taxonomicunits (OTUs). Then 10 representatives from each group (a total of 200 bacterial isolates) wereidentified by the following taxonomic id kits/data bases: Biolog GN and GP plates (Biolog Inc.,CA, USA) for Pseudomonas, Burkholderia, Salmonella and Enterobacter species and several‘no id’ Gram-positive and -negative isolates; Enterotubes and Oxiferm tubes (BBL) forEnterobacter, Kingella, Pseudomonas, and Sphingobacterium species. Protocol was as permanufacturer’s instructions. Gram-positive cocci in short chains which formed black colonieson bile esculin azide agar at 41°C (positive esculin hydrolysis) were designated enterococci.Confirmatory tests used here may result in the inclusion of other non-Enterococcus fecalstreptococci in the resistant group designated ‘enterococci’. Motility and pigmentation testswere employed to attempt separating the enterococci into the van C (intrinsic, non transmissibleresistance group), and the acquired resistance types (Anonymous 1999).

Plasmid isolation and purification from resistant bacteria

Selected isolates (200 strains in all) from the OTUs were cultured over-night in LB brothamended with the appropriate antibiotic. Plasmids were purified by means of Promega WizardPlus SV Minipreps DNA purification system according to the manufacturer’s instructions andanalyzed by 0.8% agarose gel electrophoresis in TBE at 6 V cm–1. They were visualized afterstaining with ethidium bromide.

Correlation analysis and test for significant differences between resistant bacteria in variouslocations was achieved by the MicroSoft Excel Statistical Analysis ToolPak.

Results and discussions

The task of evaluating the impact of human activities and anthropogenic forces on antibioticresistance frequencies in the environment is inherently complex and difficult at this time. First,there is no baseline data of resistance profiles or patterns in various habitats before the use ofantibiotics and the so-called undisturbed control areas may not be absolutely pristine due toflood, earthquakes and other natural events that mix flora in the environment. Next, there is nostandardized assay for assessing susceptibility or resistance of bacteria from the environment. Inaddition, the fact that medium composition and sub-culturing alter resistance in environmentalisolates (Jones et al. 1986b) makes comparisons with previous data difficult. The resultsreported herein are valid for the purpose of comparing various habitats and give a realisticevaluation of total resistance (especially of numerically dominant and culturable bacteria), sincethe samples were plated directly on to antibiotic-amended Mueller Hinton agar without theattenuation encountered by sub-culturing. Incorporation of antibiotics into selective-differentialmedia as used in some segments of this study could exert a synergistic negative effect ongrowth, resulting in an underestimation of resistance in the isolated bacterial groups. Some

Dow

nloa

ded

by [

Uni

vers

ity o

f L

eeds

] at

02:

48 2

6 A

pril

2013

Antibiotic resistance in the environment 137

researchers prefer to isolate a specific opportunistic pathogen prior to evaluating its resistancepatterns. The goal in this study was to document the resistance profiles of bacteria found in theenvironment. Data presented here are useful in understanding the proportion of intrinsic versusactual resistance and reveals major bacterial genera, which serve as reservoirs of resistance.

Preliminary screening tests (15 mg ml–1 of antibiotics) yielded too many resistant colonies tocount for all antibiotics except KAN. Subsequent screening experiments therefore employedantibiotics at a dose comparable to the content of the sensitivity discs used in Jones et al.(1986).

In Fig. 1, the total bacterial counts and the distribution of resistance to low level antibiotics(25 mg ml–1) in the environment are depicted. Resistance to all antibiotics was widespread inthese environments regardless of the prevailing human activity, except for kanamycin.Chlortetracycline-resistant populations, for instance, ranged from 1.68́ 10–4 CFU 100 ml–1 inthe residential lake to 9.28́ 10–6 CFU g–1 of dairy farm soil. The proportion of total bacteriafound to be resistant to at least one antibiotic was 1.5, 8.1, 54.2, 70.3, 75.1, 77 and 95% forresidential lake, dairy farm soil, recreation park canal water, residential garden soil, city lake byhospital, dairy farm water canal and dairy farm manure, respectively. Analysis of varianceindicated that the dairy farm soil (mixed with fresh dung), the residential garden and lake by thehospital harbored significantly higher (P < 0.01) numbers of drug-resistant bacteria than otherhabitats. There was also a positive correlation between total bacterial counts and the prevalenceof resistant populations. The percent resistant bacteria for each site was, however, significantlyhigher in the farm manure (95%) than the rest of the sites (P < 0.01), suggesting that, while totalbacterial numbers declined during composting (1.4́ 10–8 CFU g–1 fresh dung and soil; 4.0́ 10–6

Fig. 1. Distribution of bacterial resistance to various antibiotics (25 mg ml–1) in soil and waterenvironments. DFS, dairy farm soil; DFAM, dairy farm animal manure; DFWC, water canal by dairy farm;RGS, residential garden soil; CLH, city lake by hospital; PPWC, public park water canal; RLW, residentiallake water.

Dow

nloa

ded

by [

Uni

vers

ity o

f L

eeds

] at

02:

48 2

6 A

pril

2013

138 Esiobu et al.

g–1 composted manure), antibiotic-resistant strains were not equally affected. The residentialgarden yielded surprisingly high numbers of antibiotic-resistant bacteria. Further inquiriesrevealed that farm manure obtained from a nearby dairy farm, was used to fertilize the garden.The garden soil could have derived its drug-resistant population intrinsically by harboringantibiotic producers (Jones et al. 1986) or from the allochthonous species in the organic manure.It is also possible that the residual-resistant bacteria in the farm manure were selected or favoredin the garden soil by additives such as pesticides and herbicides. Associations of antibiotic-resistance genes with those used to resist other toxic substances have been demonstrated (Raniand Mahadevan 1992; Wireman et al. 1997). Though the above scenario is plausible, it can beconfirmed by tracing the species composition in both locations and via molecularepidemiology.

Tetracycline resistance was the most prevalent at all sites, and resistance frequencies rangedfrom 47 to 89% of total bacteria. Baynes et al. (1983) and Blanco et al. (1997) reported similarresults for bacteria in the environment but Jones et al. (1986) did not find tetracycline resistanceto be prevalent in their studies. Differences in species composition of a habitat appear to dictatethe pattern and nature of resistance observed (Jones et al. 1986b). According to the NationalCommittee for Clinical Laboratory Standards (1993) resistance to ampicillin, penicillin andvancomycin (for otherwise susceptible bacteria), is defined as when more than 16 mg ml–1 of theantibiotic is required to inhibit bacteria. Resistance patterns were similar for the b-lactams (PEN& AMP); even though incidence of penicillin resistance was consistently higher than that ofAMP resistance in all habitats. With the exception of the dairy farm soil, residential garden soiland lake by the hospital, all other samples recorded low resistance frequencies (15–21%) toampicillin and moderate levels to penicillin G (31–38%). Bacteria resistant to the aminoglyco-side KAN, ranged from 38 to 59% of total bacteria and were only found in the dairy farm soiland hospital lake.

Table 2 displays the relative abundance of selected bacterial groups in the various study sites.Overall, high Pseudomonas and coliform counts correlated with high frequency of low levelresistance as well as multiple drug resistance. Some Gram-negative bacteria are naturallyresistant to low levels of several anti-microbial agents due of their impermeable outer membraneand efficient efflux proteins (Quintilliani and Courvalin 1995). The incidence of multipleantibiotic resistance in indigenous bacteria from various sites is shown in Table 3. Multipleresistance to low level antibiotics was most pronounced in dairy farm soil isolates and followedthe same pattern as the single resistance in the habitats. Jones et al. (1986) found that site speciescomposition was an important determinant of this trend. They reported that 82% ofPseudomonas isolates were multiply resistant to antibiotics. Data in Tables 2 and 3 show thathabitats intrinsically rich or allochthonously enriched with Pseudomonas sp. expressed highlevels of single and multiple drug resistance. Pseudomonas may be an important reservoir andshuttle of resistance between locations. The high incidence of A*T resistance (73%) can beexplained by the reported linkage and co-selection of AMP and TET resistance (Quintilliani andCourvalin 1995).

When the incidence of resistance at selected sites was determined using high levels ofvancomycin (VAN), kanamycin (KAN) and tetracycline (TET), incorporated into several media,the results showed a very clear site-specific trend (Table 4). Significantly higher incidences ofresistance to VAN, KAN, and TET were observed in the dairy farm soil and residential gardenfertilized with organic manure (P < 0.05) than in the other sites. To conclude that this dataclearly demonstrates that anthropogenic factors enhance the frequency and spread ofenvironmental drug resistance is perhaps premature, since there was no baseline data before the

Dow

nloa

ded

by [

Uni

vers

ity o

f L

eeds

] at

02:

48 2

6 A

pril

2013

Antibiotic resistance in the environment 139

Tabl

e 2.

Rel

ativ

e ba

cter

ial

com

posi

tion

of t

he s

ites

stu

died

Sam

plin

g si

te*L

og (

CF

U/g

or

100

ml)

*

Tota

l co

unts

Gra

m-n

egat

ive *

*G

ram

-pos

itiv

e **

Pse

udom

onad

sE

. co

liE

nter

obac

ter

Ent

eroc

occi

†

Dai

ry f

arm

soi

l8.

15a

8.02

a7.

54a

5.18

a2.

47a

4.8a

3.08

a

Dai

ry f

arm

cow

man

ure

6.6b

5.62

c6.

55b

2.88

c2.

5a3.

45b

1.22

c

Dai

ry c

anal

wat

er5.

48c

5.31

c4.

99c

2.2c

1.75

b2.

01c

1.1c

Res

iden

tial

gar

den

soil

6.3b

5.9c

6.08

b3.

92b

2.4a

3.55

b2.

35b

Lak

e by

hos

pita

l5.

9c5.

77c

5.32

c3.

5b2.

35a

2.88

c1.

99b

Pub

lic

park

can

al w

ater

4.7d

4.6d

4.05

d1.

8d0.

8d1.

27d

2.01

b

Res

iden

tial

est

ate

lake

6.51

b6.

48b

5.32

c2.

1c2.

02b

3.5b

1.04

c

*C

olon

y fo

rmin

g un

its p

er g

ram

soi

l or

per

100

ml

wat

er.

Col

umn

mea

ns w

ith t

he s

ame

lette

r ar

e no

t si

gnif

ican

tly d

iffe

rent

**To

tal

num

ber

of c

ultu

rabl

e G

ram

-neg

ativ

e an

d G

ram

-pos

itive

bac

teri

a w

ere

estim

ated

by

plat

e co

unts

of

eosi

ne m

ethy

lene

blu

e an

d ph

enyl

eth

yl a

lcoh

ol a

gar,

resp

ectiv

ely,

and

con

firm

ed b

y G

ram

-sta

inin

g.†

Thi

s gr

oup

may

inc

lude

oth

er n

on-e

nter

ococ

cus

feca

l st

rept

ococ

ci g

iven

the

non

-str

inge

nt c

onfi

rmat

ory

test

s us

ed.

Dow

nloa

ded

by [

Uni

vers

ity o

f L

eeds

] at

02:

48 2

6 A

pril

2013

140 Esiobu et al.

farm was established. Also crowded plates (as was the case for the Enterobacteriaceae/Gram-negative plates) tend to disallow accurate assessment of resistance. Nevertheless, results inTables 3 and 4 are strong indicators of the probable impact of human dairy farming for example,on the incidence and distribution of drug resistance in the environment. The high incidenceobserved in the residential garden soil was surprising but could be explained by theallochthonous introduction of resistant strains from animal manure or by intrinsic resistance ofisolates.

To investigate the major reservoirs of resistance in the sites studied, a total of 200 resistantisolates that represented operational taxonomic units (OTUs) defined in the Materials andmethods were characterized. Among the identified isolates, Pseudomonas, Gram-positiveesculin-hydrolyzing cocci including enterrococci, Enterobacter and Burkholderia species werethe dominant resistant groups in the habitats studied (Fig. 2). Whereas, most Enterobacterstrains were identified with high probability as E. agglomerans, the specific epithet for the othergenera varied or could not be identified. It was not surprising to encounter these autochthonousand ubiquitous bacteria as prominent reservoirs of resistance. They possess several mechanismsfor adaptation and survival, which sometimes confer resistance traits (Jones et al. 1986b;Quintilliani and Courvalin 1995; McDonald et al. 1997; Zhao et al. 1998). Dominant resistantgenera varied from niche to niche (Table 4). Similarly, types and levels of antibiotics resisted area function of bacterial community composition and their genetic disposition.

The public health implications from data presented here are important. Firstly, intrinsic lowlevel resistance appears to be widespread and non-transmissible. It correlates with the presenceof autochthonous bacteria whose structure and/or metabolism permit growth in the presence ofantibiotics. However, the worrisome group is the high drug level antibiotic-resistant bacteria thatsurvive well in the environment and which can cause serious opportunistic and nosoccomialdiseases. Pseudomonas sp. and Burkholderia are notorious in this regard. Though a detailed

Table 3. Incidence of multiple antibiotic resistance in bacteria from various environmental locations

Antibiotics* Proportion of resistant bacteria exhibiting multiple resistance (%)

DFS DFAM DFWC RGS CLH PPWC RLW

A*V*T*K 10 0 0 0 6 0 0A*V*T 15 5 0.5 14 0.5 5 0A*V*K 20 12 0 0 10 0 0A*T*K 10 0 0 0 18 0 0V*T*K 20 0 0 0 10 0 0A*V 30 8 8 21 5 3 0.2A*K 20 0 0 0 15 0 0A*T 73 55 12 65 25 18 18V*K 20 0 0 0 11 0 0T*K 27 0 0 0 20 0 0V*T 19 20 12 18 10 15 1

* A ampicillin; V, vancomycin; T, oxytetracycline; K, kanamycin.All antibiotics were used at a concentration of 25 mg/ml.Total resistant bacteria (log CFU per g or per ml) were 6.9, 6.5, 5.3, 6.1, 5.7, 4.4 and 4.6 for DFS, DFAM, DFWC,RGS, CLH, PPCW and RLW, respectively.

Dow

nloa

ded

by [

Uni

vers

ity o

f L

eeds

] at

02:

48 2

6 A

pril

2013

Antibiotic resistance in the environment 141

Tabl

e 4.

Hig

h le

vel

anti

biot

ic r

esis

tanc

e in

bac

teri

a fr

om s

oil

and

wat

er

Sam

plin

g si

teB

acte

rial

gro

ups

Tota

l re

sist

ant

bact

eria

(C

FU

per

g so

il o

r pe

rm

l w

ater

) *

Vanc

omyc

in50

mg/

ml

Kan

amyc

in

35m

g/m

l10

0m

g/m

l

Tetr

acyc

line

35m

g/m

l17

0m

g/m

l

Pse

udom

onad

s16

5,00

014

2,40

012

8,00

099

,200

82,4

00D

airy

far

m s

oil

Ent

eroc

occi

& o

ther

gra

n-po

siti

ve12

,000

120,

000

100,

000

21,0

0019

,000

Ent

erob

acte

riac

ea &

gra

m-n

egat

ive

>20

0,00

0>

200,

000

>20

0,00

0>

200,

000

>20

0,00

0P

seud

omon

ads

800

100

––

–

Dai

ry c

anal

wat

erE

nter

ococ

ci &

oth

er g

ran-

posi

tive

100

––

––

Ent

erob

acte

riac

ea &

gra

m-n

egat

ive

5,90

030

030

0–

–P

seud

omon

ads

3,50

050

060

0–

–

Res

iden

tial

gar

den

soil

Ent

eroc

occi

& o

ther

gra

n-po

siti

ve1,

300

3,50

07,

800

––

Ent

erob

acte

riac

ea &

gra

m-n

egat

ive

70,0

00 >

200,

000

>20

0,00

0 >

200,

000

>20

0,00

0P

seud

omon

ads

5545

0–

––

Lak

e be

side

hos

pita

lE

nter

ococ

ci &

oth

er g

ran-

posi

tive

120

300

––

–E

nter

obac

teri

acea

& g

ram

-neg

ativ

e10

010

0–

––

*R

esis

tant

pop

ulat

ions

rep

orte

d as

‘>

200,

000’

wer

e to

o m

any

to c

ount

(T

MC

) .

Dow

nloa

ded

by [

Uni

vers

ity o

f L

eeds

] at

02:

48 2

6 A

pril

2013

142 Esiobu et al.

speciation of the resistant Gram-positive cocci was not done, the level of vancomycin resistance(50 mg ml–1) stresses the need for protecting the environment from antibiotics. Secondly, therelatively high proportion of resistant population in composted manure, despite a significantdrop in total bacteria, spells caution on how indicator parameters are set for treated wastes.

The need for a close surveillance of bacterial discharge and antibiotic usage is buttressed bythe data in Fig. 3, which shows that the bulk of plasmid-bearing bacteria encountered were

Fig. 2. Important bacterial groups that serve as reservoirs of drug resistance in the environment.

Fig. 3. Distribution of plasmids among antibiotic resistant bacteria isolated from soil and water.

Dow

nloa

ded

by [

Uni

vers

ity o

f L

eeds

] at

02:

48 2

6 A

pril

2013

Antibiotic resistance in the environment 143

opportunistic pathogens. Only 29% (58) of all isolates (n = 200) harbored plasmids. Very largeplasmids and low copy number plasmids may not have been visualized by the methodsemployed. Notwithstanding, the consequences of a potential genetic exchange between theseand commensals in animals and man as well as the environment are severe. The probability ofsuch gene transfers in nature has been documented by Marcinek et al. (1998). Plasmids weremost prevalent in tetracycline-resistant strains at 65%. Schnabel and Jones (1999) found Tet Bgenes (in Enterobacter agglomerans), and Tet A, C or G (in Pseudomonas spp.) borne on largeplasmids in all but 16% of TET-resistant isolates.

It is increasingly evident (Schnabel and Jones 1999) that non-therapeutic uses of antibioticsand human activities do impact the antibiotic resistance profile of a given habitat. Humanactivities can alter the species composition of a niche (through selection or allochthonous in-put), which in turn determines resistance trends. A strict regulation of non-medical use ofantibiotics; as exemplified in the recent move by the US Food and Drug Administration (FDA)to withdraw fluoroquinolone antibiotics from poultry farming (Anonymous 2001) should beupheld. Regulations guiding the discharge of treated effluent and wastes containing bacteriashould be expanded to test for resistance genes and strains. Molecular epidemiology may beuseful in tracking resistance and ascertaining the potential health risk posed by various uses anddischarge of antibiotics in the environment.

Acknowledgement

This project was supported by funds from the College of Liberal Arts, Florida AtlanticUniversity. We thank Mr. Chris Howell for skillful technical assistance.

References

Al-Mustafa, Z.H. and Al-Ghamdi, M.S. (2000) Use of norfloxacin in poultry production in the easternprovince of Saudi Arabia and its possible impact on public health. Int. J. Environ. and HealthResearch 10, 291– 9.

Anonymous. Food and Drug Administration. (1998) Evaluation of the human health impact of themicrobial effects of new antimicrobial animal drugs intended for use in food-producing animals.Fed. Reg. 63, 64094.

Anonymous (1999) Vancomycin-resistant enterococci (VRE) and the clinical laboratory. Centers forDisease Control and Prevention. http://www.cdc.gov/ncidod/hip/Lab/Factsheet/vre.htm

Anonymous (2001) Report Says agricultural use of antibiotics and impact on resistance are high. ASMNews 67 (March), 3pp.

Bates, J., Jordens, J.Z. and Griffiths, D.T. (1994) Farm animals as a putative reservoir for vancomycin-resistant enterococcal infection in man. J. Antimicrob. Chemother. 34, 507–16.

Bayne, S., Blankson, M. and Thirkell, D. (1983) Enumeration and speciation of group D streptococci fromabove and below a sewer outfall, their susceptibilities to six antibiotics and a comparison withclinical isolates. Antonie van Leeuwenhoek 49, 299–10.

Blanco, J.E., Blanco, M., Mora, A. and Blanco, J. (1997) Prevalence of bacterial resistance to quinolonesand other antimicrobials among avian Escherichia coli strains isolated from septicemic andhealthy chickens in Spain. J. Clin Microbiol 35, 2184– 5.

Calomiris, J.J., Armstrong, J.L. and Seidler, R.J. (1984) Association of metal tolerance with multipleantibiotic resistance of bacteria isolated from drinking water. Appl. Environ. Microbiol. 47,1238– 42.

Dow

nloa

ded

by [

Uni

vers

ity o

f L

eeds

] at

02:

48 2

6 A

pril

2013

144 Esiobu et al.

Food and Drug Administration (FDA). (2000) A proposed Framework for evaluating and assuring thehuman safety of the microbial effects of antimicrobial new animal drugs intended for use in food-producing animals.

Guardabassi, L., Petersen, A., Olsen, J.E. and Dalsgaard, A. (1998) Antibiotic resistance in Acinetobacterspp. isolated from sewers receiving waste effluent from a hospital and a pharmaceutical plant.Appl. Environ. Microbiol. 64, 3499–502.

Jacobs-Reitsma, W., Kan, C. and Bolder, N. (1994) The induction of quinolone resistance inCampylobacter in broilers by quinolone treatment. Lett. Appl. Microbiol. 19, 228– 31.

Jones, J.G., Gardener, S., Simon, B.M. and Pickup, R.W. (1986a) Antibiotic resistant bacteria inWindermere and two remote upland tarns in the English Lake district. J. Applied Bacteriol. 60,443– 53.

Jones, J.G., Gardener, S., Simon, B.M. and Pickup, R.W. (1986b) Factors affecting the measurement ofantibiotic resistance in bacteria isolated from lake water. J. Applied Bacteriol. 60, 455– 62.

Levy, S.B. (1998) Multi-drug resistance: a sign of the times. New Engl. J. Med. 338, 1376– 8.Linton, A.H. (1986) Flow of resistance genes in the environment and from animals to man. J. Antimicrob.

Chemother. 18 (Suppl. C), 189– 97.Linton, A.H. and Hinton, M.H. (1984) The ecology of antibiotic-resistant bacteria in animals and their

environment. In Antimicrobials and Agriculture (M. Woodbine, ed.), pp. 533–49. London:Butterworths.

Marcinek, H., Wirth, R. and Musholl-Silberhorn, A. (1998) Enterococcus faecalis gene transfer undernatural conditions in municipal sewage water treatment plants. Appl. Environ. Microbiol. 64,626– 32.

McDonald, C.L., Kuehnert, M.J., Tenover, F.C. and Jarvis, W.R. (1997) Vancomycin-resistant enterococcioutside the health-care setting: Prevalence, sources, and public health implications. Emerg InfectDis 3, 311–7.

NCCLS. (1993) Methods for dilution, antimicrobial susceptibility tests for bacteria that grow aerobically.Approved Standard. M7-A3. NCCLS, Villinova, Pa.

Niemi, M., Sibakov, M and Niemala, S. (1983) Antibiotic resistance among different species of faecalcoliforms isolated from water samples. Appl. Environ. Microbiol. 45, 73–7.

Piddock, L.J.V. (1995) Quinolone resistance and Campylobacter spp. J. Antimicrob. Chemother. 36,891– 8.

Piddock, L. (1996) Does the use of antimicrobial agent in veterinary medicine and animal husbandry selectantibiotic resistant bacteria that infect man and compromise antimicrobial therapy? J. Antimicrob.Agents Chemother. 38, 1–3.

Quintiliani, R. Jr. and Courvalin, P. (1995) Mechanisms of resistance to antimicrobial agents. In Manualof Clinical Microbiology (P. Murray, E.J. Barron, M. Pfaller, F. Tenover and R.H. Yolken, eds), pp.1308–20. Washington: ASM Press.

Rani, D.B.R. and Mahadevan, A. (1992) Plasmid mediated metal and antibiotic resistance in marinePseudomonas. BioMetals 5, 73–80.

Schnabel, E.L. and Jones, A.L. (1999) Distribution of tetracycline resistance genes and transposons amongphylloplane bacteria in Michigan apple orchards. Appl. Environ. Microbiol. 65, 4898–907.

Van Elsas, JD. and Smalla, K. (1997) Methods for sampling soil microbes. In Manual of EnvironmentalMicrobiology (C.J. Hyrst, G.R. Knudsen, M.J. Mclnerney, L.D. Stetzenbach and M.V. Walter eds),pp. 383–90. Washington: ASM Press.

Wireman, J., Liebert, C.A., Smith, T. and Summers, A.O. (1997) Association of mercury resistance withantibiotic resistance in the gram negative fecal bacteria of primates. Appl. Environ. Microbiol. 63,4494– 503.

Zhao, Q., Lee, X.Z., Srikumar, R. and Poole, K. (1998) Contribution of outer membrane efflux proteinoprM to antibiotic resistance in Pseudomonas aeruginosa independent of MexAB. Antimicrob.Agents Chemother. 41, 1682– 8.

Dow

nloa

ded

by [

Uni

vers

ity o

f L

eeds

] at

02:

48 2

6 A

pril

2013