International Journal of Antimicrobial Agents 24 (2004) 439–443

Characterisation of viridans group streptococci with different levels ofTet(M)-mediated tetracycline resistance

Paul Stapletona,∗, Victoria Adamsa, Rachel Pikeb, Victoria Lucasb, Graham Robertsb,Peter Mullanyb, Robin Rowburya, Michael Wilsonb, Hilary Richardsa

a Department of Biology, University College London, Gower Street, London, UKb Department of Microbiology, Eastman Dental Institute, University College London, Grays Inn Road, London, UK

Received 8 May 2004; accepted 8 June 2004

Abstract

Streptococcus oralis264-3,Streptococcus mitis254-1 andS. mitis264-1, isolated from the oral cavities of two children were each foundto carry thetet(M) gene but exhibited different degrees of reduced susceptibility to tetracycline (tetracycline MICs of 2, 8 and 64 mg/L,r .E arentals ).D is(S cture fore id sequenceso©

K

1

fptrsTptt

(W

ry

heseacy-ough

ancee at-

t-tself,ey we) re-

andrent

0d

espectively). The aim of this study was to determine the molecular basis for the different levels of tetracycline resistance (TcR) observedscherichia coliHB101 carrying the clonedtet(M) genes exhibited similar levels of tetracycline susceptibility to those observed in the ptreptococcal strains (MICs of 1, 16, and 64 mg/L fortet(M) genes fromS. oralis264-3,S. mitis254-1 andS. mitis264-1, respectivelyNA sequencing revealed thatS. oralis264-3 had atet(M) gene highly homologous totet(M) carried by Tn916 fromEnterococcus faecal

99.6% identity), while the intermediate- and high-level TcR strains hadtet(M) sequences that resembled thetet(M) gene of Tn5251fromtreptococcus pneumoniae(99.3% and 99.4% identity, respectively). No differences were observed in the upstream attenuator struach of the strains and differences in reduced tetracycline susceptibilities could be attributed to changes in the deduced amino acf the Tet(M) proteins.2004 Elsevier B.V. and the International Society of Chemotherapy. All rights reserved.

eywords:Viridans group streptococci; Tetracycline resistance; Heterogeneous resistance;tet(M)

. Introduction

Tetracycline resistance in oral bacteria is commonly con-erred by thetet(M) gene[1,2] which encodes a ribosomalrotection protein (RPP) that confers tetracycline resistance

hrough promoting the release of tetracycline from the 30Sibosomal subunit in a GTP-dependent manner[3–5]. Expres-ion oftet(M) is regulated by transcriptional attenuation[6].he majority oftet(M) genes are located on conjugative trans-osons, such as Tn916and Tn5253, that are responsible for

heir widespread distribution, not only amongst different bac-erial species of the oral cavity, but the larger bacterial com-

∗ Corresponding author. Present address: Department of PharmaceuticsMicrobiology), School of Pharmacy, 29-39 Brunswick Square, London

C1N 1AX, UK. Tel.: +44 20 7753 5848; fax: +44 20 7753 5942.E-mail address:paul.stapleton@ulsop.ac.uk (P. Stapleton).

munity as a whole[7]. Recombination between evolutionadivergenttet(M) genes has led to the formation oftet(M)genes with mosaic structures[8]. Consequentlytet(M) genesexhibit nucleotide variations at defined segments within tgenes[9]. The effect of these mosaic structures on tetrcline resistance has not been formally addressed althit is known that streptococci carrying differenttet(M) genesubtypes can confer different levels of tetracycline resist[10]. Since differences in tetracycline resistance could btributed to isolates carrying more than one copy of thetet(M)gene, or due to changes intet(M) promoter sequence, atenuator structure, or due to changes in the structure iideally the influence of differenttet(M) subtypes should bexamined in the same genetic background. In this studexamined three streptococcal isolates that exhibited; (aduced tetracycline susceptibility; (b) intermediate-; (c)high-level tetracycline resistance and show that the diffe

924-8579/$ – see front matter © 2004 Elsevier B.V. and the International Society of Chemotherapy. All rights reserved.oi:10.1016/j.ijantimicag.2004.06.003

440 P. Stapleton et al. / International Journal of Antimicrobial Agents 24 (2004) 439–443

levels of reduced tetracycline susceptibilities were due tochanges in the deduced amino acid sequences of the Tet(M)proteins.

2. Materials and methods

2.1. Bacterial strains

Viridans group streptococci from the saliva and plaquesamples of children (age range: 5 to 15 years), were iso-lated on Mueller–Hinton agar containing HgCl2 (40�M)[11]. Isolates were identified to species level with the APIidentification system (bioM̀erieux UK Limited, Basingstoke,United Kingdom) and by carbohydrate substrate analysis[12].

2.2. Susceptibility testing

MICs of tetracycline, ampicillin and erythromycin (allpurchased from Sigma–Aldrich, Poole, United Kingdom)were determined by agar dilution technique on isosensitestagar (Oxoid Ltd., Basingstoke, UK) with 5% horse blood forstreptococcal isolates or without blood forEscherichia coliHB101 carrying thetet(M) gene. An inoculum of about 104

o after2

2

iveD edK es( idi-s outa lec-u in-lb lledd ionm d byA y-m 0P -m uenceUt[G ic-tu dt es mP tc 4a

rpsL20] by a heat-shock procedure[14]. The orientationof the insert in pMOSBlue was confirmed by restrictionendonuclease digestion withHindIII and only clones withthe same insert orientation were used. Primers used tosequence thetet(M) gene sub-cloned into pMOSBluewere: Univm13 (5′-GTTTTCCCAGTCACGACGTTGTA-3′), and T7P (5′-TAATACGACTCACTATAGGG-3′)binding to sequences within the vector and; tetM-F3[5′-TCTGTATGCTTTGTATGCCTATGG-3′], tetM-F4[5′-TGTGACGAACTTTACCGAATCTGA-3′], tetM-F5[5′-TCCCTCTGCTGCAAACGACTG-3′], tetM-F6 [5′-GTATGGTTGGAATGTGACGGACTG-3′], and tetM-R3[5′-GTTGTACCTTTGTCCACGCTTCCT-3′] binding tosequences within the sub-cloned gene. Independentlyamplified PCR products were sequenced directly to confirmthe nucleotide changes found in the cloned genes. DNAsequencing was performed with the aid of an AppliedBiosystems automated fluorescent sequencer (model 373A).The nucleotide sequences of thetet(M) genes forS. oralis264-3, S. mitis 254-1, andS. mitis 264-1 have the ac-cession numbers, AJ580976, AJ580977, and AJ580978,respectively.

3. Results

sis-t ciw sus-c d

TA ls

I

SSSSSSSSSSSSSSSSSSSSSS

(+) tet(M) present, (−) tet(M) absent.

rganisms per spot was used and the MICs were read4 h incubation at 37◦C.

.3. DNA techniques

DNA extraction was performed with a Gram-positNA isolation kit (Flowgen, Ashby de la Zouch, Unitingdom). Southern blotting to Nylon N+ membran

Amersham Pharmacia Biotech) and DNA–DNA hybration with digoxygenin-labelled probes was carriedccording to the manufacturer’s instructions (Roche Molar Biochemicals, Lewes, United Kingdom). Digoxygen

abelledtet(M) probe was prepared from Tn916DNA by PCRy the incorporation of alkali-stable digoxygenin-labeUTP (Roche Molecular Biochemicals) in the amplificatixture. PCR amplification was performed as describeusubel et al.[13] with Dynazyme EXT thermostable polerase (Flowgen) and an annealing temperature of 5◦C.rimers used for amplification of thetet(M) gene and prooter regions (2841 bp) were based on GenBank seq09422 for Tn916 from Enterococcus faecalisDS16:

etM-F1 (5′-ACGACGTTCTTCAAGCTCTATCCT-3′)positions 11492-11515] and tetM-R1 (5′-CATTTTATCG-CTCTGCGTCTTT-3′) [positions 14310–14332]. Restr

ion endonuclease digestion withHindIII and DraI wassed to distinguish betweentet(M) and the closely relateet(O) and tet(S) genes. The amplifiedtet(M) genes werub-cloned into theEcoRV site of pMOSBlue (Amershaharmacia Biotech) and introduced into competenE.oli HB101 [F− �(gtp-proA)62 recA13 leuB6 supE4ra-14 galK2 lacY1� (mcrC-mrr) mtl-1 proA2 xyl-5

During a study examining mercury and antibiotic reance in oral bacteria[11], 22 viridans group streptococere isolated that exhibited varying degrees of reducedeptibilities to tetracycline (Table 1). Three isolates exhibite

able 1ntimicrobial resistance profiles and the presence of thetet(M) gene for oratreptococcal isolates with reduced susceptibility to tetracycline

solate MIC (mg/L) tet(M)a

Tetracycline Ampicillin Erythromycin

. mitis152-2 2 0.12 0.03 −

. parasanguis181-2 2 1 0.016 −

. parasanguis179-1 2 0.5 1 −

. oralis264-3 2 0.03 <0.008 +

. oralis283-4 2 0.06 2 +

. oralis204-4 4 0.12 0.016 +

. mitis254-1 8 0.016 0.06 +

. mitis160-1 16 0.12 0.03 −

. oralis277-2 16 0.12 1 −

. mitis166-4 32 0.016 0.03 +

. mitis173-1 32 0.016 0.016 +

. parasanguis181-1 32 1 0.03 −

. oralis279-1 32 <0.008 <0.008 −

. oralis262-3 32 0.06 0.5 +

. oralis271-2 32 0.016 0.016 +

. oralis260-1 64 <0.008 0.03 +

. mitis264-1 64 1 <0.008 +

. oralis271-1 64 0.25 0.03 +

. oralis318-1 64 0.016 0.016 +

. oralis254-2 64 0.12 1 −

. oralis262-1 64 0.016 8 +

. oralis283-4 64 0.06 2 +a

P. Stapleton et al. / International Journal of Antimicrobial Agents 24 (2004) 439–443 441

reduced susceptibility to ampicillin (MIC of 1 mg/L), andsix isolates were resistant to erythromycin (MIC≥ 1 mg/L).Fourteen of the twenty-two isolates were found to carry thetet(M) gene by PCR and by DNA–DNA hybridisation witha tet(M) probe (Table 1). Threetet(M)-positive isolates,S.oralis 264-3,S. mitis254-1 andS. mitis264-1 with vary-ing degrees of reduced tetracycline susceptibilities (tetracy-cline MICs of 2, 8 and 64 mg/L, respectively) were chosento be investigated further. Two of the isolates,S. mitis264-1 andS. oralis264-3 were isolated from the oral cavity ofthe same child. Thetet(M) gene was amplified from the iso-lates by PCR with a proof-reading DNA polymerase and theresulting products were sub-cloned into pMOSBlue.E. coliHB101 carrying the sub-clonedtet(M) genes exhibited sim-ilar levels of reduced tetracycline susceptibility observed inthe parental streptococcal strains (tetracycline MICs of 1, 1,16, and 64 mg/L for HB101, 264-3, 254-1 and 264-1, respec-tively) (Table 2).

DNA sequencing of thetet(M) genes revealed thatS.oralis264-3 had atet(M) gene highly homologous to that carriedby Tn916 fromE. faecalisDS16 (99.6% identity) (Table 2),while isolates,S. mitis264-1 andS. mitis254-1 hadtet(M)sequences that resembled thetet(M) gene of Tn5251 fromStreptococcus pneumoniae(99.3% and 99.4% identity, re-spectively) (Table 2). No differences were observed in theu ft line-r in the−T inoaQ acids line-r 9Q,K

4

d viri-d latesw tiono ity oft ed tot , or toc f dif-f ra-c lyb kets[ sus-cta ikef ce. Ta

ble

2Te

trac

yclin

eM

ICs

for

stre

ptoc

occa

liso

late

san

dE.coli

HB

101

carr

yingtet(M

)ge

nes

codi

ngfo

rTe

t(M

)pr

otei

nsw

ithdi

ffere

ntde

duce

dam

ino

acid

sequ

ence

s

Hos

tstr

ain

orge

netic

elem

ent

Tetr

acyc

line

MIC

(mg/

L)fo

r:N

ucle

otid

epo

sitio

nsfo

rno

n-co

nser

vativ

ese

quen

cedi

ffere

nces

amon

gst

tet(M

)ge

nesc

Hos

taR

ecip

ientb

490

627

756

857

951

980

1348

1352

1372

1892

S.oralis

254-

32

1A

(Met 1

64)

G(G

ln20

9)T

(Asn

252)

A(L

ys28

6)T

(His

317)

G(A

rg32

7)T

(Ser

450)

C(P

ro45

1)A

(Met

458)

T(V

al63

1)S.m

itis2

54-1

816

A(M

et 164

)T

(His

209)

T(A

sn25

2)A

(Lys

286)

G(G

ln31

7)T

(Leu

327)

G(A

la45

0)A

(Gln

451)

G(V

al45

8)T

(Val

631)

S.m

itis2

64-1

6464

G(V

al 164

)G

(Gln

209)

A(L

ys25

2)G

(Arg

286)

G(G

ln31

7)T

(Leu

327)

G(A

la45

0)A

(Gln

451)

G(V

al45

8)C

(Ala

631)

tet(M

)on

Tn916

N/d

dN

/dd

G(V

al16

4)G

(Gln

209)

T(A

sn25

2)A

(Lys

286)

A(G

ln31

7)T

(Leu

327)

T(S

er45

0)C

(Pro

451)

A(M

et45

8)T

(Val

631)

tet(M

)on

Tn5251

N/d

dN

/dd

A(M

et16

4)T

(His

209)

A(L

ys25

2)A

(Lys

286)

G(G

ln31

7)T

(Leu

327)

T(S

er45

0)C

(Pro

451)

A(M

et45

8)T

(Val

631)

aS

trep

toco

ccal

isol

ate.

bE.coli

HB

101.

cN

ucle

otid

epo

sitio

nsar

eba

sed

onth

ead

enos

ine

ofth

ein

itiat

ion

codo

n(A

TG

)be

ing

assi

gned

posi

tion

1.T

heam

ino

acid

and

posi

tion

for

the

corr

espo

ndin

gnu

cleo

tide

are

give

nin

brac

kets

.d

N/d

,not

dete

rmin

ed.

pstream attenuator structure of thetet(M) gene for each ohe isolates, but the intermediate- and high-level tetracycesistant strains each had a single nucleotide change35 region of thetet(M) promoter (TTTACA to TTGACA).et(M) from isolateS. oralis264-3 had three deduced amcid differences compared with Tet(M) from Tn916: V164M,307H and L327R. Comparison of the deduced aminoequences of the intermediate- and high-level tetracycesistant isolates revealed five differences: M164V, H20252N, K286R and V631A.

. Discussion

In common with other reports[2,15] thetet(M) resistanceeterminant was frequently encountered amongst theans group streptococci isolated in this study. The isoere not uniformly resistant to tetracycline despite detecf the same resistance determinant amongst the major

he isolates. Such differences could have been attributhe presence of more than one resistance determinanthanges affecting gene expression, or the presence oerent Tet(M) variants with different abilities to confer tetycline resistance. Differenttet(M) variants have previouseen identified in bacteria isolated from periodontal poc

9] but whether these variants had different tetracyclineeptibilities was not reported. Thetet(M) variants identified inhis study, resemble thetet(M) subtypes, M1 (S. oralis264-3)nd M2 (S. mitis254-1 andS. mitis264-1), reported by Olst al. [9]. However, unlike the M1 subtype, thetet(M) gene

rom S. oralis264-3 did not confer tetracycline resistan

442 P. Stapleton et al. / International Journal of Antimicrobial Agents 24 (2004) 439–443

Comparison of the nucleotide sequences revealed that thetet(M) gene fromS. oralis264-3 differed from M1 at twonucleotide positions (951, A to T and 980, T to G) that resultin changes to the deduced amino acid sequence of the Tet(M)protein (Q317H and L327R). Consequently, either one ofthese amino acid substitutions or both of them are likely tobe responsible for the reduced ability of the Tet(M) proteinto confer tetracycline resistance.

The existence oftet(M)-positive isolates which remainsusceptible to tetracycline has been previously noted amongstStreptococcus pneumoniaeclones[10,16]. MIC testing with-out prior induction with tetracycline has been proposed as anexplanation for the existence oftet(M)-positive tetracycline-sensitive clones[10]. In this study, we have sequenced the re-gion upstream from thetet(M) gene, including the promoterand the attenuator structure which is responsible for inducibletetracycline resistance, and found no nucleotide differenceswhen compared with the upstreamtet(M) gene sequence fromTn916. Although the ability to induce tetracycline resistancewas not investigated in this study, indirect evidence from thesequence of the upstream region of thetet(M) gene, com-bined with the lack of ability of thetet(M) gene to confertetracycline resistance when introduced intoE. coli,suggestschanges in the deduced amino acid sequence of the Tet(M) isthe primary factor responsible for tetracycline susceptibilityi

lybg p-s d,a eom thes ft s int ore,t ucedw toE acids typeso

ionp andT ch asO nss F-Gi acids d pu-ta ffectt st n ofTt thea R).O Leu

residue at position 327 shows some conservation: the Leu-327 residue is found in Tet(M), Tet(O), Tet(S), Tet(Q) andEF-G. Since this amino acid residue is located within the pu-tative ribosomal-binding domain of Tet(M), the substitutionof Arg for Leu-327 may affect binding of Tet(M) to the ri-bosome and consequently explain the significantly reducedtetracycline resistance observed inS. oralis264-3.

S. mitis254-1 andS. mitis264-1 had five deduced aminoacid differences between them. All the amino acid substitu-tions, with the exception of V631A, are found in other RPPsand it is not possible to deduce which amino acid substi-tutions confer the level of tetracycline resistance observed.However, it appears that different mosaic gene structures cangive rise to moderate (found inS. mitis254-1) and high-level(found inS. mitis264-1) tetracycline resistance. The signif-icance of these differenttet(M) gene variants remains to bedetermined.

Tetracycline MICs for viridans group streptococci havebeen reported to exhibit a broad range with tetracycline MICsvarying in increments throughout the range[20-22]. Fromthis study we have shown that for isolates carrying thetet(M)gene, changes in the deduced amino acid sequence of Tet(M)is the main reason for this variation. However, further workinvolving site-directed mutagenesis combined with detailedsubstrate profiling is required to confirm the contributionsm

A

ncil(

R

with

cy-other

mes

er-the

biol

ausBO

ion.

ppli-nce.

heid

n this case.The region upstream from thetet(M) gene has previous

een reported to be highly conserved even amongsttet(M)enes from diverse hosts[17]. In this study, the regions utream from thetet(M) gene were also highly conservelthough a single nucleotide change in the−35 sequencf the tet(M) promoter was found in the sequences fromS.itis 254-1 andS. mitis264-1. Since both isolates had

ame nucleotide change, alteration to the−35 sequence ohe tet(M) gene was not responsible for the differencehe levels of tetracycline resistance observed. Furthermhe different levels of tetracycline resistance were reprodhen the correspondingtet(M) genes were introduced in. coli suggesting that changes in the deduced aminoequences of Tet(M) were responsible for the phenobserved.

Tet(M) belongs to a wider family of ribosomal protectroteins (RPP) that includes the closely related Tet(O)et(S) determinants and more divergent determinants sutr(A) fromStreptomyces[7]. Ribosomal protection protei

hare homology with elongation factors, EF-Tu and Envolved in protein synthesis and alignments of aminoequences of RPPs and elongation factors have identifieative functional regions in Tet(M) and Tet(O)[18,19]. Alter-tions to certain conserved amino acids are known to a

he function of the RRP and elongation factors[18]: changeo Asn-128 located in the putative GTP-binding domaiet(O) results in a decrease in tetracycline resistance[19]. Inhis study,S. oralis264-3 had three deduced alterations inmino acid sequence of Tet(M) (V164M, Q317H, L327f these positions in the amino acid sequence, only the

ade by each of the deduced amino acid changes.

cknowledgements

This work was funded by the Medical Research Cougrants G9810729 and G9810341).

eferences

[1] Olsvik B, Olsen I, Tenover FC. Detection oftet(M) and tet(O) usingthe polymerase chain reaction in bacteria isolated from patientsperiodontal disease. Oral Microbiol Immunol 1995;5:299–303.

[2] Villedieu A, Diaz-Torres ML, Hunt N, et al. Prevalence of tetracline resistance genes in oral bacteria. Antimicrob Agents Chem2003;47:878–82.

[3] Burdett V. Tet(M)-promoted release of tetracycline from ribosois GTP dependent. J Bacteriol 1996;178:3246–51.

[4] Connell SR, Trieber CA, Stelz U, Einfeldt E, Taylor DE, Nihaus KH. The tetracycline resistance protein Tet(O) perturbsconformation of the ribosomal decoding center. Mol Micro2002;45:1463–72.

[5] Connell SR, Trieber CA, Dinos GP, Einfeldt E, Taylor DE, NierhKH. Mechanism of Tet(O)-mediated tetracycline resistance. EMJ 2003;22:945–53.

[6] Su YA, He P, Clewell DB. Characterization of thetet(M) determi-nant of Tn916: Evidence for regulation by transcription attenuatAntimicrob Agents Chemother 1992;36:769–78.

[7] Chopra I, Roberts M. Tetracycline antibiotics: mode of action, acations, molecular biology, and epidemiology of bacterial resistaMicrobiol Mol Biol Rev 2001;65:232–60.

[8] Oggioni MR, Dowson CG, Smith JM, Provvedi R, Pozzi G. Ttetracycline resistance genetet(M) exhibits mosaic structure. Plasm1996;35:156–63.

P. Stapleton et al. / International Journal of Antimicrobial Agents 24 (2004) 439–443 443

[9] Olsvik B, Tenover FC, Olsen I, Rasheed JK. Three subtypes of thetet(M) gene identified in bacterial isolates from periodontal pockets.Oral Microbiol Immunol 1996;5:299–303.

[10] Doherty N, Trzcinski K, Pickerill P, Zawadzki P, Dowson CG. Ge-netic diversity of thetet(M) gene in tetracycline-resistant clonal lin-eages ofStreptococcus pneumoniae. Antimicrob Agents Chemother2000;44:2979–84.

[11] Pike R, Lucas V, Stapleton PM, et al. Prevalence and antibioticresistance profile of mercury-resistant oral bacteria from childrenwith and without mercury amalgam fillings. J Antimicrob Chemother2002;5:777–83.

[12] Beighton D, Hardie JM, Whiley RA. A scheme for the identificationof viridans streptococci. J Med Microbiol 1991;35:367–72.

[13] Ausubel FM, Brent R, Kingston RE, et al. Current protocols inmolecular biology. New York: John Wiley and Sons; 1989.

[14] Sambrook J, Fritsch EF, Maniatis T. Molecular cloning: a laboratorymanual. 2nd ed Cold Spring Harbor: Cold Spring Harbor LaboratoryPress; 1989.

[15] Lacroix JM, Walker CB. Detection and incidence of the tetracyclineresistance determinanttet(M) in the microflora associated with adultperiodontitis. J Periodontol 1995;66:102–8.

[16] Corso A, Severina EP, Petruk VF, Mauriz YR, Tomas A. Molecu-lar characterisation of penicillin-resistant Streptococcus pneumoniae

isolates causing respiratory disease in the United States. MicrobialDrug Resist 1998;4:325–37.

[17] Roberts M. Tetracycline resistance determinants: mechanism of ac-tion, regulation of expression, genetic mobility, and distribution.FEMS Microbiol Rev 1996;19:1–24.

[18] Taylor DE, Chau A. Tetracycline resistance mediated by ribosomalprotection. Antimicrob Agents Chemother 1996;40:1–5.

[19] Grewal J, Manavathu EK, Taylor DE. Effect of mutational alterationof Asn-128 in the putative GTP-binding domain of tetracycline re-sistance determinant Tet(O) fromCampylobacter jejuni. AntimicrobAgents Chemother 1993;37:2645–9.

[20] Doern GV, Ferraro MJ, Brueggemann AB, Ruoff KL. Emergenceof high rates of antimicrobial resistance among viridans groupstreptococci in the United States. Antimicrob Agents Chemother1996;40:891–4.

[21] Teng L-J, Hsueh P-R, Chen Y-C, Ho S-W, Luh K-T. Antimicrobialsusceptibility of viridans group streptococci in Taiwan with an em-phasis on the high rates of resistance to penicillin and macrolides inStreptococcus oralis. J Antimicrob Chemother 1998;41:621–7.

[22] Wisplinghoff H, Reinert RR, Cornely O, Shifert H. Molecular rela-tionships and antimicrobial susceptibilities of viridans group strep-tococci isolated from blood of neutropenic cancer patients. J ClinMicrobiol 1999;37:1876–80.