THE NEW ENGLAND JOURNAL OK MEDICINE 
July 21, 1008 
Table 1. Properties of the Bacterial Strains Studied.* 
Rm.evant Charm hrisiks 
StRAlN 
ano Plasmid CoNir.NTt 
SoCRLL OR Rl.n RFNCt 
E. faecium DM4147 
pi PM 1 6 Tra‘ Vin 
Pc letM 
Wild strain 
E. faecium BM4147-1 
Pc leiM 
Curing of BM4147 
E. faecium BM4148 
pi PH 1 6 Tra" Vm 
Pc letM LY 
Spontaneous mutant of BM4147 
E. faecium BM4I52 
plPK 1 7 Tra" Vm ermAM auciE 
plP8 1 8 
Pc te/M 
Wild strain 
E. faeciu n B.M4152-1 
plP8 1 8 
Pc leiM 
Curing of BM4152 
E. faecium BM4153 
p!P8 1 7 Tra* Vm ermAM andE 
pll’8 18 
Pc letM LY 
Spontaneous mutant of BM4152 
E. faecium BM4153-1 
pi P8 1 8 
Pc leiM LY 
Curing of BM4153 
S. sanguis Challis 
Pakula et al. 12 
5. sanguis BM4163 
p!P8!6 Tra" Vm 
Transformation 
pIP8 16 X Challis 
S. sanguis BM4164 
pIP8 1 7 Tra" Vm ermASf andE 
Transformation 
plP817 X Challis 
*Tra” denotes oon-self-lransferable, Vui vancomycin resistance. Pc penicillin resistance, and LY LY146032 resistance. 
tThe designation of the plasmid and its phenotypic characters are aligned. 
staining, the absence of catalase, their inability to produce gas, the 
presence of Lancefield antigen group D, growth on 40 percent bile, 
in 6.5 percent sodium chloride, in 0. 1 percent methylene blue, at pH 
9.6. Species identification 13,14 was based on the absence of a reduc- 
tion in potassium tellurite and tests for acid production from 50 
carbohydrates in API 50 CH galleries (API-System). Strain 
BM4147 caused fermentation of xylose and rhamnose, whereas 
BM4152 did not. The patterns of the penicillin-binding protein 
were identical to those of E. faecium reference strain D359, which is 
resistant to penicillins. 15 E.faecalis strainsJH2-2 (resistant to fusidic 
acid and rifampin 16 ) and BM4110 (resistant to streptomycin 17 ), 
Staphylococcus aureus RN450, and E. faecium strains BM4105 and 
BM4106 (susceptible to glycopeptides) were from our laboratory 
collection. 
Mediums 
Brain-heart infusion broth and agar (Difco) were used for the 
cultures. Sensitivity tests were done on Mueller-Hinton agar (Diag- 
nostics Pasteur) supplemented with 5 percent horse blood. All incu- 
bations were at 37°C. 
Genetic-Analysis Techniques 
ct 011 x 111 rations of the antibiotics were deter- 
mined in cells with and without induced re- 
sistance. 
Inactivation of vancomycin and leico- 
planin was assessed with the Guts test 7 *'; Ba- 
cillus subtilis A TCC6633 was used as the indi- 
cator organism. Inactivation of antibiotics 
in liquid medium by resting cells was per- 
formed as described elsewhere.^ 1 Nitroccfin 
was used to test for the presence of a penicil- 
linase. 22 The aminoglycoside-modifying 
enzymes were assayed in supernatants 
(100,000x^) after ultrasonic disintegration, 
as described elsewhere. 75 *’ 4 
DNA-DNA Hybridization 
Purification of plasmid and M13 deriva- 
tive rcpiicativc-form phage DNA was done 
as previously described. 251 '' The DNA re- 
striction fragments were separated by elec- 
trophoresis in horizontal slab gels (20 by 20 
by 0.7 cm) containing 0.8 percent agarose 
Type II or VII (Sigma Chemical). Frag- 
ments were extracted from low-tcmpcrature- 
gelling agarose Type VII as previously de- 
scribed. 27 
Plasmid p I P8 1 6 DNA and the purified re- 
striction DNA fragments used as probes 
were labeled with [a 37 P]dcoxyadcnosine triphosphate by nick 
translation. 7 ” DNA fragments cloned in bacteriophage N113 deriva- 
tives were hybridized with the 15-bp distal primer and labeled by 
means of DNA synthesis in the presence of dcoxyguanosine triphos- 
phate, dcoxycytidine triphosphate, deoxythymidine triphosphate, 
[a 37 P]deoxyadenosine triphosphate, and DNA polymerase I 
(Klenow fragment). 28 Dot blot hybridization was performed in 50 
percent formamide at 42°C for 24 hours and was followed by three 
washings in 2 X SSCP(lX SSCP = 0. 12 M hydrochloric acid, 0.015 
M sodium citrate, 0.013 M potassium phosphate, and 0.001 M 
EDTA, pH 7.2) at room temperature for 15 minutes and two 
washings in 0.2 X SSCP at 65°C for 1 hour. DNA restriction frag- 
ments fractionated by agarose-gel electrophoresis were transferred 
to nitrocellulose sheets (Sartorius), as described by Southern, 29 and 
hybridized 30 with the probes. The aadE probe was obtained by 
cloning a 470-bp Hpall fragment of the streptococcal plasmid 
pJHl 31 into the Accl site of M13mp9. 26 The ermAM and tet.W probes 
have previously been described. 32,33 
Restriction endonucleases were obtained from Boehringer 
GmbH and were used according to the manufacturer’s recom- 
mendations. Lysozyme was obtained from Sigma Chemical, and 
RNase A (bovine pancreas) and proteinase K from Calbiochcm- 
Behring. 
Mating on filters, transformation, and curing of antibiotic- 
resistance traits with novobiocin at 42°C were performed as de- 
scribed elsewhere. 1217,18 The antibiotic concentrations for the selec- 
tion of transformants were 4 mg of erythromycin per liter, 100 mg 
of streptomycin per liter, and 2 mg of vancomycin per liter. For the 
selection of resistant mutants, plates containing twofold dilutions of 
LY 146032, from 0.125 to 64 mg per liter, were inoculated by flood- 
ing with 10 fi to 10 10 bacteria per plate. Clones that grew at antibiotic 
levels higher than the minimal inhibitory concentrations after 48 
hours at 37'C were streaked and selected for further study. 
Resistance Studies 
To assess the resistance to antibiotics, the disk-agar diffusion test 
was used. The method of Steers ct al. 19 was employed to determine 
the minimal inhibitory concentrations of the antibiotics. To induce 
resistance to vancomycin and teicopianin, resistant cells were grown 
in the presence of subinhibitory concentrations of glycopep- 
lides. The concentrations of vancomycin and teicopianin were 5 
and 50 mg per liter for enterococcal strains and 0.5 mg per liter for 
transformants. After incubation overnight, the minimal inhibitory 
Results 
Characteristics of the Antibiotic Resistance of E. faecium 
BM4147 and GM4152 
Strains BM4147 and BM4152 were highly resistant 
to vancomycin and teicopianin, but remained sus- 
ceptible to LY146032 (Table 2). This resistance was 
inducible by subinhibitory concentrations of glyco- 
peptidcs (data not shown). No inactivation of the 
antibiotics could be detected with the microbiologic 
techniques used. These strains were also resistant to 
penicillins, kanamycin, tobramycin, amikacin, sisomi- 
cin, netilmicin, tetracycline, and minocycline (Table 
3). Resistance to penicillins could be attributed to the 
pattern of lower-affinity penicillin-binding proteins 
that is characteristic of E. faecium; production of beta- 
lactamase was not detected in assessments with Nitro- 
[ 410 ] 
Recombinant DNA Research, Volume 13 
