J Periodontol • February 2014

Antimicrobial Resistance and Prevalence

of Resistance Genes of Obligate
Anaerobes Isolated From Periodontal
Abscesses
Yi Xie,* Jiazhen Chen,† Junlin He,* Xinyu Miao,† Meng Xu,* Xingwen Wu,* Beiyun Xu,* Liying Yu,*
and Wenhong Zhang†

Background: This study attempts to determine the anti-

microbial resistance profiles of obligate anaerobic bacteria
that were isolated from a periodontal abscess and to evalu-
ate the prevalence of resistance genes in these bacteria.
Methods: Forty-one periodontal abscess samples were
cultivated on selective and non-selective culture media to
isolate the oral anaerobes. Their antibiotic susceptibilities

P
eriodontal abscess has been de-
to clindamycin, doxycycline, amoxicillin, imipenem, cefra- fined as a suppurative lesion that is
dine, cefixime, roxithromycin, and metronidazole were de- associated with periodontal break-
termined using the agar dilution method, and polymerase down and localized pus in the gingival
chain reaction assays were performed to detect the pres- wall of the periodontal pocket.1 The vast
ence of the ermF, tetQ, nim, and cfxA drug resistance majority of dental abscesses with end-
genes. odontic or periodontal pocket origins
Results: A total of 60 different bacterial colonies was are polymicrobial anaerobic infections.2
isolated and identified. All of the isolates were sensitive These infections are predominantly at-
to imipenem. Of the strains, 6.7%, 13.3%, 16.7%, and 25% tributable to strict anaerobes, such as
were resistant to doxycycline, metronidazole, cefixime, and Gram-negative anaerobic bacteria, and
amoxicillin, respectively. The resistance rate for both clinda- partially facultative anaerobes, such as
mycin and roxithromycin was 31.7%. Approximately 60.7% the Streptococcus milleri group.3 How-
of the strains had the ermF gene, and 53.3% of the amoxicil- ever, in dental abscess, anaerobes out-
lin-resistant strains were found to have the cfxA gene. Two number aerobic and facultative bacteria
nim genes that were found in eight metronidazole-resistant in ratios ranging from 10:1 to 10,000:1,
strains were identified as nimB. with anaerobic Gram-negative bacilli
Conclusions: In the present study, the Prevotella species predominating.3 In periodontal abscesses,
are the most frequently isolated obligate anaerobes from Bacteroides forsythus, Fusobacterium
periodontal abscesses. The current results show their alarm- spp., Prevotella intermedia, Prevotella ni-
ingly high resistance rate against clindamycin and roxithro- grescens, and Porphyromonas gingivalis
mycin; thus, the use of these antibiotics is unacceptable for are the most prevalent species,4,5 and
the empirical therapy of periodontal abscesses. A brief these bacteria are more commonly iso-
prevalence of four resistance genes in the anaerobic bacte- lated from periodontal abscesses than
ria that were isolated was also demonstrated. J Periodontol endodontic abscesses. However, previous
2014;85:327-334. studies using swabs of purulent material
have demonstrated poor recovery of
KEY WORDS
strict anaerobes, which implicated strep-
Anti-infective agents; microbiology; periodontics. tococci or staphylococci as the causative
microorganisms of suppurative dental
* Department of Stomatology, Hua Shan Hospital, Fudan University, Yangpu, Shanghai, infection. This difference in recovery was
China.
† Department of Infectious Disease, Hua Shan Hospital, Fudan University. attributable to poor sampling techniques
and inadequate culture methods in these
investigations.6,7

doi: 10.1902/jop.2013.130081

327
Drug Resistance of Obligate Anaerobes of Periodontal Abscess
Antibiotics are often prescribed to limit the
spread of the infection after the surgical drainage of
pus from periodontal abscesses. However, reports
have indicated that abscess formation may be at-
tributable to the overgrowth of resistant pathogens.8
Therefore, it is critical to know the susceptibility
profiles of clinically relevant oral pathogens (which
may vary considerably between patients in different
countries) for effective antibiotic treatment.
Amoxicillin, metronidazole, roxithromycin, and
clindamycin are commonly used to treat oral in-
fections. Resistance against metronidazole was long
considered to be rare in anaerobes. However, recent
studies have shown that this resistance is no longer
uncommon.9 Metronidazole resistance in Bacter-
oides spp. appears to be encoded by the nimA
through nimG genes.9-11 The use of tetracycline has
decreased because of the high resistance rates
observed in various microorganisms. The resistance
in Gram-negative anaerobic bacteria, especially
in the genus Prevotella, was primarily attributable to
the presence of the tetQ gene,12,13 which encodes
the ribosomal protection protein.14 In oral anaer-
obes, resistance to clindamycin–erythromycin has
significantly increased during the past 20 years15
and is usually encoded by the ermF gene (which
has the widest host range and was found in 10 of
the examined anaerobic genera).16 Amoxicillin
continues to exhibit a high level of activity against
the majority of oral anaerobes.17 However, cfxA
(class A/group 2e) has been characterized recently
in and cloned from oral amoxicillin-resistant Pre-
votella species.18
In this study the aim is to investigate the an-
timicrobial resistance profiles of clinically cultured
anaerobes from periodontal abscesses and their
resistance to clindamycin, doxycycline, amoxicillin,
imipenem, cefradine, roxithromycin, cefixime, and
metronidazole, which are commonly used in the
treatment of periodontitis. The prevalence of four
relevant resistance genes is also studied.
MATERIALS AND METHODS
Participants
Forty-one patients (21 males and 20 females; aged
45 – 10 years [mean age: 47 years]) with peri-
odontal abscesses were recruited at the Department
of Stomatology of Hua Shan Hospital, Shanghai,
China, for this study from August 2011 to August
2012. Informed written consent was obtained from
each patient, and the study was approved by the
Ethics Committee of Hua Shan Hospital. The ex-
clusion criteria were: 1) pregnancy; 2) consumption
of systemic antimicrobials or anti-inflammatory
drugs in the previous 6 months; and 3) periodontal
therapy during the previous 6 months.
328
Volume 85 • Number 2
Microbial Samples
The abscesses were drained after decontamination
of the mucosa. A sterile 10-mL inoculating loop‡ was
inserted into the deep area of the fistula for 20
seconds. The loop was then immediately inoculated
on two prereduced culture media using quadrate
section streak methods. Anaerobe basal agar§ me-
dium was used as a non-selective culture medium
and was supplemented with 5% sterile defibrinated
sheep blood.i The anaerobe basal sheep blood agar
was supplemented with kanamycin¶ (100 mg/mL),
and vancomycin# (75 mg/mL) was used as a selec-
tive culture medium for the isolation of Gram-neg-
ative anaerobic bacteria. Both the selective and non-
selective culture media were immediately incubated
in transparent bags** at 37C and were studied after
2 to 4 days of growth. Each colony (usually two to six
colonies) with a distinct morphology (including the
colony size, pigmentation, transparency, viscidity,
edge situation, apophysis situation, and hemolytic
features) was selected from the primary culture for
two or three repeated oxygen tolerance tests. The
isolates that only grew anaerobically were studied
further. If several colonies from one specimen were
identified as the same species, only one colony was
used for additional study.
Strain Species Identification
Single anaerobic colonies were inoculated onto an-
other anaerobe basal sheep blood agar plate for
proliferation. Certain bacterium colonies from the
same plate for the identification sequencing of 16S
rRNA were sampled. Approximately 20 to 50 mg of
bacteria was harvested, suspended in a Tris–EDTA
buffer, and pelleted for DNA extraction. The genomic
DNA was extracted using a DNA extraction kit††
according to the instructions of the manufacturer.
Approximately 1,400-bp 16S rDNA fragments were
amplified with the universal ribosomal 16S
primers 8FLP (59-AGAGTTTGATCCTGGCTCAG-
39) and 1492RPL (59-GGTTACTTGTTACGACTT-39),
as described previously.19 The amplification products
were detected using electrophoresis in agarose gels
(1%) and sequenced using a DNA analyzer.‡‡ A se-
quence database of 16S ribosomal RNA sequences
(bacteria and archaea) from GenBank (using the
BLASTN program through the National Center for
Biotechnology Information sever) was used to com-
pare 1,400-bp 16S rDNA sequences.20 Identifica-
tion at the species level was defined as a 16S rDNA
‡ 10-mL Inoculating Loops, JET BIOFIL, Canada.
§ Anaerobe basal agar, Oxoid Limited, Basingstoke, Hampshire, U.K.
i Defibrinated sheep blood, Oxoid Limited.
¶ Kanamycin, Sigma-Aldrich, St. Louis, MO.
# Vancomycin, Sigma-Aldrich.
** GENbag, bioMérieux, Marcy l’Etoile, France.
†† DNA extraction kit, TianGen Biotech, Shanghai, China.
‡‡ 3730xl DNA Analyzer, Applied Biosystems, Foster City, CA.
J Periodontol • February 2014
Table 1.
Oligonucleotide Sequences and PCR Conditions Used to Detect Target Resistance Genes
Resistance
Genes Oligonucleotide Sequence 59-39
cfxA/cfxA2 CGT AGT TTT GAG TAT AGC TTT
GAT GTT GCC TAT ATA TGT C
tetQ TTA TAC TTC CTC Cgg CAT C
ATC ggT TCg AgA ATg TCCA
nim ATG TTC AGA GAA ATG GGG CGT
AAG
GCT TCC TTG CCT GTC ATG TGC TC
ermF CGG GTC AGC ACT TTA CTA TTG
GGA CCT ACC TCA TAG AGA AG
bp = base pair.
sequence similarity of >99% to that of the prototype
strain sequence in GenBank; identification at the
genus level was defined as a 16S rDNA sequence
similarity of >97% to that of the prototype strain
sequence in GenBank.21
Antimicrobial Susceptibility Testing
Antibiotic susceptibilities to clindamycin,§§ doxy-
cycline,ii amoxicillin,¶¶ cefradine,## cefixime,***
roxithromycin,††† metronidazole,‡‡‡ and imipenem§§§
were determined using the agar dilution method22
with Brucella agariii plates that were supplemented
with 5 mg/mL hemin,¶¶¶ 1 mg/mL vitamin k1,### and
5% defibrinated sheep blood. Two-day-old bacteria
(104 to 105 colony forming units) were inoculated
onto the plates using multipoint inoculators.****
The minimal inhibitory concentration (MIC) was
defined as the lowest concentration of the antibiotic
that repressed visible growth. Bacteroides fragilis
ATCC 25285 served as the control strain. The
breakpoints were used in accordance with the rec-
ommendations of the Clinical and Laboratory
Standards Institute (CLSI):23 1) clindamycin (‡8
mg/mL); 2) imipenem (‡16 mg/mL); 3) metronida-
zole (‡32 mg/mL); 4) doxycycline (‡8 mg/mL); and
5) amoxicillin (‡2 mg/mL). When CLSI antimicrobial
breakpoints were not established, the breakpoints
of the European Committee on Antimicrobial
Susceptibility Testing (EUCAST)24 were followed.
Because the breakpoints of roxithromycin and ce-
fixime have not yet been determined by CLSI and
Xie, Chen, He, et al.
Amplicon
Amplification Cycles Size (bp) Reference
94C · 30 seconds, 35 cycles of 94C · 30 966 This study
seconds, 52C · 30 seconds, 72C · 75
seconds, and 72C · 5 minutes
94C · 30 seconds, 35 cycles of 94C · 30 904 This study
seconds, 52C · 60 seconds, 72C ·
1.15 minutes, and 72C · 5 minutes
94C · 30 seconds, 35 cycles of 94C · 30 458 9
seconds, 55C · 60 seconds, 72C · 10
minutes, and 72C · 5 minutes
94C · 30 seconds, 35 cycles of 94C · 30 487 15
seconds, 50C · 30 seconds, 72C · 1
minutes, and 72C · 5 minutes
EUCAST, the breakpoints of ‡32 and ‡8 mg/mL for
roxithromycin and cefixime, respectively, were used
in accordance with previous studies.25,26
Detection of Drug Resistance Genes
The ermF, tetQ, nim, and cfxA genes were amplified
using polymerase chain reaction (PCR) assays under
the conditions shown in Table 1 and with a total
volume of 50 mL that was composed of 25 mL pre-
mix,†††† 20 mM of each primer, and 5 to 100 ng
bacteria genomic DNA. An ermF amplification was
performed in all strains because of its high per-
centage of resistance to clindamycin, whereas tetQ,
nim, and cfxA amplification were only performed in
the resistant strains because of their relatively low
percentage of resistance to metronidazole, doxycy-
cline, and amoxicillin. The amplification products
were sequenced using 3,730 and compared using
the BLASTN program on the Nucleotide collection
(nr/nt) database in GenBank.
§§ Clindamycin, National Institutes for Food and Drug Control, Beijing,
China.
ii
¶¶
Doxycycline, National Institutes for Food and Drug Control.
Amoxicillin, National Institutes for Food and Drug Control.
## Cefradine, National Institutes for Food and Drug Control.
*** Cefixime, National Institutes for Food and Drug Control.
††† Roxithromycin, National Institutes for Food and Drug Control.
‡‡‡ Metronidazole, National Institutes for Food and Drug Control.
§§§ Imipenem, Hangzhou MSD Pharmaceutical, Hangzhou, China.
iii Brucella agar, BD Difco, San Jose, CA.
¶¶¶ Hemin, Sigma-Aldrich.
### Vitamin k1, Sigma-Aldrich.
**** Multipoint inoculators, Denley Instruments, Billinghurst, Sussex, U.K.
†††† Premix Ex Taq, Takara, Kyoto, Japan.
329
Drug Resistance of Obligate Anaerobes of Periodontal Abscess Volume 85 • Number 2

Table 2. The susceptibility assays were

repeated twice for all of the isolates.
Identification of Anaerobic Bacteria Isolated From
As shown in Table 3, the most ef-
Periodontal Abscesses by 16S rRNA fective antibiotic was imipenem, to
which all were sensitive. Low-to-in-
Genus of Strains/Total termediate resistance was found for
Microorganisms Strains Strains (percentage) other antimicrobial agents. In total,
Prevotella genus 42 42 of 60 (70%) four of 60 (6.7%), eight of 60
Black-pigmented Prevotella 29 29 of 60 (48.3%) (13.3%), 10 of 60 (16.7%), and 15 of
P. intermedia 12 — 60 (25%) of the strains were resistant
P. melaninogenica 7 — to doxycycline, metronidazole, ce-
P. nigrescens 3 — fixime, and amoxicillin, respec-
P. denticola 7 - tively. The least effective antibiotics
Non-black-pigmented Prevotella 13 13 of 60 (21.7%) in the present study are clindamy-
P dentalis 4 — cin and roxithromycin, which both
P. oris 3 —
had a 19 of 60 (31.7%) resistance
P. salivae 1 —
rate. Ten strains were resistant
Candidatus Prevotella conceptionensis 1 —
Prevotella sp. oral taxon 317 to both clindamycin and roxi-
Prevotella spp. 2 — thromycin. The resistances to
P. buccae 2 — amoxicillin and clindamycin were
high in the Prevotella species,
Veillonella genus 11 11 of 60 (18.3%) reaching 30.9% and 38.1%, re-
V. parvula/V. dispar 10 —
spectively, whereas the other
V. vegosae 1 —
anaerobic species had lower re-
Fusobacterium genus 2 2 of 60 (3.3%) sistances to these antibiotics (al-
F. canifelinum 1 — though not statistically significant
F. nucleatum 1 — because of the limited amount of
Actinomyces genus 1 1 of 60 (1.67%) strains) (Table 3).
A. odontolyticus 1 — The ermF gene was detected in
17 of the 60 strains, and 16 of these
Bacteroides 1 1 of 60 (1.67%) strains were phenotypically resistant
B. heparinolyticus 1 — to clindamycin and/or roxithromycin,
Shuttleworthia 1 1 of 60 (1.67%) whereas only one strain was sen-
S. satelle 1 — sitive to clindamycin and interme-
diately resistant to roxithromycin
Pentostrentococcus 1 1 of 60 (1.67%)
(MIC = 16 mg/mL). As shown in
P. stomatis 1 —
Table 4, in the amoxicillin-, metro-
Propionibacterium 1 1 of 60 (1.67%) nidazole-, and doxycycline- re-
P. acnes 1 — sistant strains, cfxA, nim, and tetQ
— = The proportion of species was not listed in this table. were detected in 53.3% (eight of
15), 25% (two of eight), and 50%
RESULTS (two of four) of the strains, respectively. Both of
A total of 60 different bacterial colonies were iso- the amplified nim genes had nucleotides that were
lated from the periodontal abscesses of 41 patients. 100% identical to nimB (National Center for Bio-
Identifications of these colonies are shown in Table technology Information gene identification no.
2; 42 isolates (70%) belonged to the Prevotella ge- 9496386); thus, they were identified as such.
nus, and 29 of 60 (48.3%) were black-pigmented
Prevotella, whereas 13 of 60 (21.7%) were non- DISCUSSION
black-pigmented Prevotella. Of the 60 strains, 11 In agreement with previous studies,1,3 Gram-nega-
(18.3%) belonged to the Veillonella genus. Other tive anaerobic rods are the most frequent isolates
genera, including Actinomyces, Peptostreptococcus, in periodontal abscesses, and the Prevotella spe-
Bacteroides, Shuttleworthia, and Propionibacterium, cies are those most frequently isolated in this
also had minor occurrences. Only three rods, Ac- research, suggesting their predominance in peri-
tinomyces odontolyticus, Shuttleworthia satelle, odontal abscess infection. It has been reported
and Propionibacterium acnes, were Gram positive, that non-black-pigmented Prevotella species (such
whereas the others were Gram negative. as P. oralis, P. buccae, and P. oris) are nearly as

330
J Periodontol • February 2014 Xie, Chen, He, et al.

Table 3.
MIC (mg/mL) Range and MIC50 and MIC90 Values of Prevotella and Non-Prevotella
Isolated From Periodontal Abscesses for Eight Antibiotics Tested in China

Resistant Strains/Total
Antibiotics Range (mg/mL) MIC50 MIC90 (resistance percentage)

Total (n = 60)
Cefradine <0.06 to >128 0.5 16 NA
Cefixime <0.06 to 32 0.125 8 10 of 60 (16.7%)
Amoxicillin <0.06 to >128 <0.06 32 15 of 60 (25%)
Clindamycin <0.06 to >128 <0.06 >128 19 of 60 (31.7%)
Metronidazole <0.06 to >128 0.5 >128 8 of 60 (13.3%)
Roxithromycin <0.06 to >128 2 >128 19 of 60 (31.7%)
Doxycycline <0.06 to >8 <0.06 1 4 of 60 (6.7%)
Imipenem <0.06 to 0.5 <0.06 <0.06 0 of 60 (0%)
Prevotella (n = 42)
Cefradine <0.06 to >128 0.5 32 NA
Cefixime <0.06 to 32 0.125 4 4 of 42 (9.5%)
Amoxicillin <0.06 to >128 <0.06 16 13 of 42 (30.9%)
Clindamycin <0.06 to >128 <0.06 >128 16 of 42 (38.1%)
Metronidazole <0.06 to >128 0.5 8 4 of 42 (9.5%)
Roxithromycin <0.06 to >128 2 64 11 of 42 (26.2%)
Doxycycline <0.06 to >8 <0.06 0.5 2 of 42 (4.8%)
Imipenem <0.06 to 0.5 <0.06 <0.06 0 (0%)
Non-Prevotella anaerobes (n = 18)
Cefradine <0.06 to 8 0.5 8 NA
Cefixime <0.06 to 16 1 16 6 of 18 (33.3%)
Amoxicillin <0.06 to 64 <0.06 8 2 of 18 (11.1%)
Clindamycin <0.06 to >128 <0.06 >128 3 of 18 (16.7%)
Metronidazole <0.06 to >128 0.5 128 3 of 18 (16.7%)
Roxithromycin <0.06 to >128 4 >128 6 of 18 (33.3%)
Doxycycline <0.06 to >8 <0.06 8 2 of 18 (11.1%)
Imipenem <0.06 to 0.5 <0.06 <0.06 0 of 60 (0%)
The following breakpoints were used in accordance with CLSI (2012):23 1) clindamycin (‡8 mg/mL); 2) imipenem (‡16 mg/mL); 3) metronidazole (‡32 mg/
24
mL); and 4) doxycycline (‡8 mg/mL). The following breakpoint was used according to EUCAST (2012): amoxicillin (‡2 mg/mL). The following breakpoints
were used according to references 25 and 26: 1) roxithromycin (‡32 mg/mL); and 2) cefixime (‡8 mg/mL).
MIC50 or MIC90 = MICs for 50% or 90% of the organisms, respectively; NA = Because no breakpoint was available for cefradine, resistance percentage was not
evaluated.

common as the black-pigmented species.1 How-
ever, in the research of the present authors, black-
pigmented Prevotella (48.3%) had a higher ratio
than non-black-pigmented Prevotella (21.7%), and
this finding may be because of the higher virulence
of the black-pigmented Prevotella to cause in-
fections.27 However, some of the non-Prevotella
genera, such as Shuttleworthia and Propioni-
bacterium, that were isolated in this research have
rarely been reported in periodontitis,28,29 and more
information is needed to understand their clinical
role in periodontal abscesses.
In research by the present authors, clindamycin
was the least effective antibiotic against Prevotella,
which had 38.1% (16 of 42) resistant strains, and
this finding was in accordance with a previous
study30 in which 31% of the Prevotella species was
resistant to clindamycin. This alarming resistance
rate renders it unacceptable for the empirical ther-
apy of severe infections that are predominantly
caused by Prevotella species. The detection rate for
the ermF gene was 60.7% (13 of 19) in clindamycin-
and/or roxithromycin-resistant Prevotella species,
which was slightly higher compared to the findings
of Chung et al.,31 who found a 55.6% (20 of 36)
detection rate in clinical anaerobic bacteria. Other
erm genes (ermB, ermC, ermQ, ermT, and erm35)
have also been reported to encode macrolides and
lincomycin resistance in anaerobes,15 which may
explain the resistance of the non-tested ermF-neg-
ative strains in this study.
Tetracycline is one of the most widely used an-
tibiotics worldwide, but its usage has decreased
because of increasing bacterial resistance to this
drug. Doxycycline as a semisynthetic tetracycline
has been shown to significantly reduce the anaerobic

331
Drug Resistance of Obligate Anaerobes of Periodontal Abscess
Table 4.
Distribution of Resistance Genes of Resistant Strains Isolated From Periodontal Abscess
Total Positive Resistance Gene Strains/
Resistance Genes Phenotypic Resistant Strains (percentage)
nim 2 of 8 (25%)
cfxA 8 of 15 (53.3%)
ermF 17 of 28 (60.7%)
tetQ 2 of 4 (50%)
— = no phenotypic resistance strains in the genera.
population in plaque without increasing the pop-
ulation of resistant bacteria or allowing the bacteria
to acquire antibiotic resistance.32 In the present
research, the rate of resistance to doxycycline is as
low as 6.7%, and some articles reported doxycy-
cline-resistant anaerobic bacteria that were isolated
from periodontitis. Doxycycline could be recom-
mended as one of the first-line antimicrobial drugs
against periodontal disease because of its high ac-
tivity and low resistance. Tetracycline resistance has
always been associated with the presence of the tetQ
gene, which has been found in 12 anaerobic gen-
era.33,34 However, only two of four doxycycline-
resistant strains were detected to have tetQ, and
other genes that promoted resistance are not iden-
tified in this study.
It is inspiring that all of the strains in the present
study are susceptive to imipenem. However, imi-
penem has an extremely broad spectrum of activity
and is not recommended to use empirically in treating
light oral infections to prevent uncontrollable re-
sistance.
An amoxicillin–metronidazole combination has
been used to treat periodontitis for the long term. In
the present study, 30.9% (13 of 42) of the Prevotella
strains are resistant to amoxicillin; Kuriyama et al.35
showed that amoxicillin resistance occurred in 34%
of Prevotella isolates and that all of these resistant
strains were found to produce b-lactamase. Reports
indicated that the resistance rate for amoxicillin
ranged from 9% to 54% in common isolates from
acute dental abscesses.36 Amoxicillin is advocated
as a suitable first-line agent in these situations,
although the increased resistance of Prevotella
strains to penicillins remains a matter of con-
cern.37 Prevotella species are known to produce
b-lactamases that are encoded by cfxA,38 which
may explain their resistance to amoxicillin. In the
present research, 61.5% (eight of 13) of amoxicillin-
resistant Prevotella strains were found to harbor
332
Volume 85 • Number 2
Positive Resistance Gene Strains/Phenotypic Resistant
Strains in the Genera (percentage)
Prevotella Veillonella Other Bacteria
2 of 2 (100%) — 0 of 6 (0%)
8 of 13 (62%) 0 of 2 (0%) —
13 of 19 (68%) 3 of 7 (42%) 1 of 2 (50%)
2 of 4 (50%) — —
cfxA, although their ability to produce b-lactamases
remains untested.
Metronidazole is commonly used for the treatment
of infections that are caused by anaerobic organ-
isms. In the present research, it was found that eight
of the 60 anaerobic strains were resistant to met-
ronidazole and that four of these strains belonged to
Prevotella. Some studies have examined the distri-
bution of nim genes in Prevotella spp. Lubbe et al.39
detected one nimA-positive strain among seven
metronidazole-resistant Prevotella species, whereas
Katsandri et al.40 found one nimC-positive strain and
two nimE-positive isolates among 57 metronidazole-
resistant Prevotella species. In the present research,
two Prevotella species strains (P. dentalis and P.
denticola) were detected to harbor nimB. Because
five nim-negative strains with an MIC >128 mg/mL
were recovered, other potential resistance mecha-
nisms may exist for these isolates, such as decreased
pyruvate:ferredoxin oxidoreductase activity, over-
expression of efflux pumps, or alterations of the
rhamnose catabolism pathway.9,41
There are some limitations to the present study.
First, the anaerobic organisms should have been
incubated for 7 to 14 days. However, after 4 days of
incubation, the fast-growing colonies had increased
in size and would have grown over some of the strict
anaerobes. Therefore, the inoculations were limited
to 4 days of incubation, and the development of the
slow-growing anaerobes was likely inhibited. This
may explain the high proportion of fast-growing
Prevotella and the loss of slow-growing anaerobes.
Second, a limited amount of isolates from the primary
plates based on typical morphology was chosen; thus,
some anaerobes with similar phenotypes may have
been overlooked by the present study. Last, only the
antimicrobial resistance genes that had been fre-
quently reported in Prevotella were detected; all other
resistance genes that could encode for other anaer-
obes were not studied.
J Periodontol • February 2014
CONCLUSIONS
The use of macrolides and lincomycins should be
monitored when treating periodontal abscesses be-
cause of resistance of these anaerobic bacteria to
these antibiotics. Their reduced susceptibility to
penicillins and metronidazole should also be noted
and discussed. Some studies have addressed anti-
microbial susceptibility profiles or detected resistance
genes in the anaerobes that are isolated from peri-
odontal abscesses. Careful monitoring of antimicro-
bial resistance and the detection of these resistance
genes may verify the presence and spread of certain
anaerobic strains, and these data could likely aid
clinical practitioners in the empirical use of antibiotics
and could help identify the origin of the resistance.
ACKNOWLEDGMENTS
This study was supported by the Postgraduate Inno-
vation Fund projects of Fudan University. Xie and Dr.
Chen contributed equally to this work. The authors
report no conflicts of interest related to this study.
REFERENCES
1. Herrera D, Roldán S, Sanz M. The periodontal abscess:
A review. J Clin Periodontol 2000;27:377-386.
2. Finegold SM. Anaerobic Gram-Negative Bacilli. In:
Baron S, ed. Medical Microbiology, 4th ed. Galveston,
TX: University of Texas Medical Branch at Galveston;
1996:131-142
3. Robertson D, Smith AJ. The microbiology of the acute
dental abscess. J Med Microbiol 2009;58:155-162.
4. Peterson LJ. Principles of management and prevention
of odontogenic infections. In: Contemporary Oral and
Maxillofacial Surgery. St. Louis, MO: Mosby; 1998:
392-417.
5. Jaramillo A, Arce RM, Herrera D, Betancourth M,
Botero JE, Contreras A. Clinical and microbiological
characterization of periodontal abscesses. J Clin Peri-
odontol 2005;32:1213-1218.
6. Lewis MA, MacFarlane TW, McGowan DA. A microbi-
ological and clinical review of the acute dentoalveolar
abscess. Br J Oral Maxillofac Surg 1990;28:359-366.
7. Olsen I, Socransky SS. Comparison of three anaerobic
culture techniques and media for viable recovery of
subgingival plaque bacteria. Scand J Dent Res 1981;
89:165-174.
8. Dahlén G. Microbiology and treatment of dental ab-
scesses and periodontal-endodontic lesions. Periodon-
tol 2000 2002;28:206-239.
9. Alauzet C, Mory F, Teyssier C, et al. Metronidazole
resistance in Prevotella spp. and description of a new
nim gene in Prevotella baroniae. Antimicrob Agents
Chemother 2010;54:60-64.
10. Haggoud A, Reysset G, Azeddoug H, Sebald M. Nucle-
otide sequence analysis of two 5-nitroimidazole re-
sistance determinants from Bacteroides strains and of
a new insertion sequence upstream of the two genes.
Antimicrob Agents Chemother 1994;38:1047-1051.
11. Reysset G. Genetics of 5-nitroimidazole resistance in
Bacteroides species. Anaerobe 1996;2:59-69.
12. Lacroix JM, Walker CB. Detection and prevalence of
the tetracycline resistance determinant Tet Q in the
Xie, Chen, He, et al.
microbiota associated with adult periodontitis. Oral
Microbiol Immunol 1996;11:282-288.
13. Roberts MC. Tetracycline resistance determinants:
Mechanisms of action, regulation of expression, ge-
netic mobility, and distribution. FEMS Microbiol Rev
1996;19:1-24.
14. Roberts MC, Sutcliffe J, Courvalin P, Jensen LB,
Rood J, Seppala H. Nomenclature for macrolide and
macrolide-lincosamide-streptogramin B resistance
determinants. Antimicrob Agents Chemother 1999;43:
2823-2830.
15. Roberts MC. Acquired tetracycline and/or macrolide-
lincosamides-streptogramin resistance in anaerobes.
Anaerobe 2003;9:63-69.
16. Kuriyama T, Williams DW, Yanagisawa M, et al. Anti-
microbial susceptibility of 800 anaerobic isolates from
patients with dentoalveolar infection to 13 oral antibi-
otics. Oral Microbiol Immunol 2007;22:285-288.
17. Madinier I, Fosse T, Giudicelli J, Labia R. Cloning and
biochemical characterization of a class A beta-lacta-
mase from Prevotella intermedia. Antimicrob Agents
Chemother 2001;45:2386-2389.
18. Handal T, Olsen I, Walker CB, Caugant DA. Detection
and characterization of beta-lactamase genes in sub-
gingival bacteria from patients with refractory peri-
odontitis. FEMS Microbiol Lett 2005;242:319-324.
19. Smart CD, Schneider B, Blomquist CL, et al. Phyto-
plasma-specific PCR primers based on sequences of
the 16S-23S rRNA spacer region. Appl Environ Micro-
biol 1996;62:2988-2993.
20. Altschul SF, Madden TL, Schäffer AA, et al. Gapped
BLAST and PSI-BLAST: A new generation of protein
database search programs. Nucleic Acids Res 1997;
25:3389-3402.
21. Drancourt M, Bollet C, Carlioz A, Martelin R, Gayral JP,
Raoult D. 16S ribosomal DNA sequence analysis of
a large collection of environmental and clinical un-
identifiable bacterial isolates. J Clin Microbiol 2000;38:
3623-3630.
22. Brown WJ, Waatti PE. Susceptibility testing of clinically
isolated anaerobic bacteria by an agar dilution tech-
nique. Antimicrob Agents Chemother 1980;17:629-635.
23. Clinical and Laboratory Standards Institute. Performance
Standards for Antimicrobial Susceptibility Testing;
Twenty-Second Informational Supplement, M100-S22.
Wayne, PA: Clinical and Laboratory Standards Institute;
2012:1-184.
24. European Committee on Antimicrobial Susceptibility
Testing. Breakpoint tables for interpretation of MICs
and zone diameters, v.2.0. Basel, Switzerland: Euro-
pean Committee on Antimicrobial Susceptibility Test-
ing; 2012:58-63
25. Bryskier A. Roxithromycin: Review of its antimicrobial
activity. J Antimicrob Chemother 1998;41(Suppl. B):
1-21.
26. Fuchs PC, Barry AL, Jones RN. Cefixime disk suscep-
tibility test criteria. J Clin Microbiol 1986;24:647-649.
27. Finegold SM, Strong CA, McTeague M, Marina M. The
importance of black-pigmented gram-negative anaer-
obes in human infections. FEMS Immunol Med Micro-
biol 1993;6:77-82.
28. Downes J, Munson MA, Radford DR, Spratt DA, Wade
WG. Shuttleworthia satelles gen. nov., sp. nov., iso-
lated from the human oral cavity. Int J Syst Evol
Microbiol 2002;52:1469-1475.
29. Bulkacz J, Erbland JF. Exocellular phospholipase C
activity from a Propionibacterium acnes strain isolated
333
Drug Resistance of Obligate Anaerobes of Periodontal Abscess
from a periodontal pocket. J Periodontol 1997;68:369-
374.
30. Teng LJ, Hsueh PR, Tsai JC, Liaw SJ, Ho SW, Luh KT.
High incidence of cefoxitin and clindamycin resistance
among anaerobes in Taiwan. Antimicrob Agents Che-
mother 2002;46:2908-2913.
31. Chung WO, Werckenthin C, Schwarz S, Roberts MC.
Host range of the ermF rRNA methylase gene in
bacteria of human and animal origin. J Antimicrob
Chemother 1999;43:5-14.
32. Walker CB, Godowski KC, Borden L, et al. The effects of
sustained release doxycycline on the anaerobic flora
and antibiotic-resistant patterns in subgingival plaque
and saliva. J Periodontol 2000;71:768-774.
33. Connell SR, Tracz DM, Nierhaus KH, Taylor DE.
Ribosomal protection proteins and their mechanism
of tetracycline resistance. Antimicrob Agents Chemo-
ther 2003;47:3675-3681.
34. Roberts MC. Update on acquired tetracycline resis-
tance genes. FEMS Microbiol Lett 2005;245:195-203.
35. Kuriyama T, Absi EG, Williams DW, Lewis MA. An
outcome audit of the treatment of acute dentoalveolar
infection: Impact of penicillin resistance. Br Dent J
2005;198:759-763, 754, 778.
36. Baumgartner JC, Xia T. Antibiotic susceptibility of
bacteria associated with endodontic abscesses. J En-
dod 2003;29:44-47.
37. Rogers MB, Parker AC, Smith CJ. Cloning and charac-
terization of the endogenous cephalosporinase gene,
334
Volume 85 • Number 2
cepA, from Bacteroides fragilis reveals a new subgroup
of Ambler class A beta-lactamases. Antimicrob Agents
Chemother 1993;37:2391-2400.
38. Iwahara K, Kuriyama T, Shimura S, et al. Detection of
cfxA and cfxA2, the beta-lactamase genes of Prevo-
tella spp., in clinical samples from dentoalveolar
infection by real-time PCR. J Clin Microbiol 2006;
44:172-176.
39. Lubbe MM, Stanley K, Chalkley LJ. Prevalence of nim
genes in anaerobic/facultative anaerobic bacteria iso-
lated in South Africa. FEMS Microbiol Lett 1999;172:
79-83.
40. Katsandri A, Avlamis A, Pantazatou A, Houhoula DP,
Papaparaskevas J. Dissemination of nim-class genes,
encoding nitroimidazole resistance, among different
species of Gram-negative anaerobic bacteria isolated
in Athens, Greece. J Antimicrob Chemother 2006;58:
705-706.
41. Diniz CG, Farias LM, Carvalho MA, Rocha ER, Smith
CJ. Differential gene expression in a Bacteroides fragilis
metronidazole-resistant mutant. J Antimicrob Chemo-
ther 2004;54:100-108.
Correspondence: Prof. Liying Yu, Department of Stomatol-
ogy of Hua Shan Hospital, Fudan University, 200040
Shanghai, China. E-mail: wuyu1984@hotmail.com.
Submitted February 5, 2013; accepted for publication
April 20, 2013.