Clinical microbiology
In vitro
effect of antibiotics on bio
fi
lm formation by
Bacteroides fragilis
group strains isolated from intestinal microbiota of dogs and their
antimicrobial susceptibility
Janice Oliveira Silva
a,*, Ana Catarina Martins Reis
a, Carlos Quesada-Gómez
b,
Adriana Queiroz Pinheiro
c, Rosemary Souza Freire
d, Reinaldo Barreto Oriá
d,
Cibele Barreto Mano de Carvalho
aaCentro de Biomedicina, Programa de Pós Graduação em Microbiologia Medica, Laboratório de Bacteriologia, Universidade Federal do Ceará, Fortaleza,
CE, Brazil
bLaboratorio de Investigación en Bacteriología Anaerobia, Facultad de Microbiología and Centro de Investigación en Enfermedades Tropicales, Universidad de Costa Rica, San José, Costa Rica
cFaculdade de Veterinária, Universidade Estadual do Ceará, Fortaleza, CE, Brazil
dNúcleo De Estudos Em Microscopia E Processamento De Imagem (NEMPI), Departamento de Morfologia, Faculdade de Medicina,
Universidade Federal do Ceará, Fortaleza, CE, Brazil
a r t i c l e
i n f o
Article history:
Received 26 October 2013 Received in revised form 17 April 2014
Accepted 18 April 2014 Available online 4 May 2014
Keywords: Bacteroides fragilis Biofilm
Antimicrobials resistance Intestinal microbiota
a b s t r a c t
The Bacteroides fragilis group strains colonize the intestinal tract of dogs as commensal bacteria. Nevertheless, they can be opportunistic pathogens responsible for significant morbidity and mortality rates in dogs, like in oral infections, abscesses and wound infections. The purpose of this study was to evaluate antimicrobial susceptibility inB. fragilisstrains isolated from dogs intestinal microbiota and to evaluate the effect of subinhibitory concentrations of some antimicrobials on biofilm formation. A total of 30B. fragilisgroup strains were tested for susceptibility to ten antimicrobial agents by broth micro-dilution method. ThirteenB. fragilisstrains were tested for biofilm formation and the biofilm producer strains were chosen to evaluate the effect of subinhibitory concentrations of six antimicrobials on biofilm formation. The isolates were susceptible to amoxicillin-clavulanic acid, metronidazole, imipenem and chloramphenicol. Tetracycline and clindamycin were active against 50% and 33% of the strains, respec-tively. When biofilm-forming strains were grown in the presence of sub-MICs of imipenem and metronidazole, the inhibition of biofilm formation was observed. In contrast, enrofloxacin at ½ MIC caused a significant increase in biofilm formation in two of four strains examined. In conclusion, the
B. fragilisgroup strains isolated were susceptible to most of the antimicrobials tested and the sub-MIC concentrations of imipenem, metronidazole and clindamycin were able to inhibit the biofilm formation. Ó2014 Elsevier Ltd. All rights reserved.
1. Introduction
Bacteroides fragilis group and Parabacteroides microorganisms are anaerobic, bile-resistant, non-spore-forming, gram-negative rods[1]. They are normally commensal bacteria of the gutflora of dogs. Nevertheless, these microorganisms can also be responsible for infections with significant morbidity and mortality rates in dogs, like oral infections, abscesses and wound infections[2e4].
Numerous factors contribute to the ability of B. fragilis to persist as commensal in the gut, such as the capacity to use a
wide range of dietary polysaccharides, high bile tolerance, capsule formation and the presence of variable surface antigens that allow the bacteria to evade the host’s immune responses. The capacity for adhesion and biofilm formation is also important factors[5].
The ability to form biofilm is an attribute of a majority of the microorganisms. A biofilm is a structured consortium of bacteria embedded in a self-produced polymer matrix consisting of poly-saccharide, protein and DNA. The biofilm enables the bacteria to survive in hostile environments and increase antibiotic resistance due to restricted penetration of antimicrobials, the heterogeneous metabolic activity of microorganisms contained in biofilm and differences in gene expression patterns compared with planktonic cells[6].
*Corresponding author. Centro de Biomedicina, Programa de Pós Graduação em Microbiologia Medica, Laboratório de Bacteriologia, Universidade Federal do Ceará, Rua Cel. Nunes de Melo 1315, Rodolfo Teófilo, 60430-270 Fortaleza, CE, Brazil.
E-mail address:janice.o.silva@gmail.com(J.O. Silva).
Contents lists available atScienceDirect
Anaerobe
j o u r n a l h o m e p a g e : w w w . e l s e v i e r . c o m / l o c a t e / a n a e r o b e
Recent studies have demonstrated thatBacteroidesstrains from human gastrointestinal microbiota can form biofilmin vitro. Biofilm is responsible for a wide variety of infections in veterinary medicine such as pneumonia, liver abscesses, bacterial gastroenteritis, wound infections and mastitis[7e10]. The purpose of this study
was to evaluate the antimicrobial susceptibility ofB. fragilisgroup andParabacteroidesisolates and their ability to form biofilm in the presence of sub-MICs of some antimicrobials.
2. Material and methods
2.1. Strains and samples
From January to June 2011, 30 non-duplicated microorganisms of theBacteroidesand Parabacteroidesgenera were isolated from the intestinal tract of 50 healthy dogs. The animals used were in medical appointments (mainly vaccine) from the Veterinary Hos-pital Unit of the State University of Ceará. The hosHos-pital’s research ethics committee (N10610110-2/57) approved the study. Animals
that had undergone antimicrobial chemotherapy during the last 30 days were not included.
Rectal swabs with the feces were collected for each animal. The swabs were inoculated in semi-solid pre-reduced Cary & Blair medium (HiMediaÒ
) and sent to the Bacteriology Laboratory of Federal University of Ceará (UFC). The samples were plated on BacteroidesBile Esculin agar (BBE, HiMediaÒ
) supplemented with gentamycin (100
m
g/mL), under anaerobic conditions (90% N2and10% CO2) in a jar with AnaerobacÒsystem (ProbacÒ).
2.2. Identification of isolates
Bacterial strains isolated on BBE were examined for oxygen tolerance and bacterial morphology by Gram staining. The identi-fication was determined by the fatty acid profile using gas chro-matography system (6850CG, Agilent TechnologiesÒ
) and the MIDI-SherlockÒ
software, according to manufacturer’s instructions (Agilent TechnologiesÒ
).
2.3. Determination of minimum inhibitory concentration (MIC) to antimicrobials
The MIC was determined by broth micro-dilution method, ac-cording to the CLSI guidelines[11]. The antimicrobial drugs evalu-ated were: penicillin, amoxicillin-clavulanic acid, cefoxitin, imipenem, clindamycin, ciprofloxacin, enrofloxacin, tetracycline, chloramphenicol and metronidazole (SigmaÒ
). To evaluate the susceptibility we used break points according to CLSI, TheB. fragilis ATCCÒ
25285 reference strain was included as control. All tests were performed twice.
2.4. Biofilm formation assays
Biofilm formation assays were performed as described previ-ously using 96-wellflat-bottom plate[10]. ThirteenB. fragilis iso-lates were evaluated. Briefly, after a 48 h incubation period of bacterial cultures at 37C under anaerobic conditions, the content
of each well was removed and the wells were carefully washed 3 times with 200
m
L of phosphate-buffered saline (PBS). The plates were dried at 60C for 60 min and stained with 100m
L of 0.01% w/vcrystal violet solution. Crystal violet was removed after 20 min and the dye bound to the adherent cells was solubilized with 95% ethanol. The optical density (OD540 nm) was determined using an automated microtitre plate reader (Multiscan FC, Thermo ScientificÒ
).
The strains were classified according to their adherence ability into the following categories: non-adherent, weakly adherent, moderately adherent, and strongly adherent, using the classifi ca-tion described by Sproule-Willoughby[12]. The experiments were conducted in triplicate and repeatedfive different times.
2.5. Influence of antimicrobials on biofilm formation
The strongest 4 biofilm-producing strains were chosen to eval-uate the effect of sub-MIC concentrations of some antimicrobials on biofilm formation. The antimicrobials tested were cefoxitin, clin-damycin, chloramphenicol, enrofloxacin, imipenem and metroni-dazole (SigmaÒ
). The concentrations tested were ½MIC and ¼MIC of each antibiotic after 48 h of incubation (under anaerobic con-ditions at 37 C). The strains were tested using a 96-wellfl
at-bottom plate and Brucella broth (BDÒ
Company) with the same method described for the biofilm formation. The tests, performed in triplicate, were repeated four different times [10]. The four B. fragilisstrains in Brucella broth without antibiotics were used as control of biofilm formation. For each strain tested we calculated the average of the four values obtained from OD540in the presence
and absence of antimicrobials.
The results of biofilm formation were compared individually using the test one-way ANOVA followed by Bonferroni multiple comparison using GraphPad PrismÒ
version 5.00 for Windows. (P value<0.05 was considered significant).
Furthermore, the biofilm formation after 48 h of incubation was evaluated using a confocal laser scanning microscopy (CLSM). The same strongest 4 biofilm-producerB. fragilis strains used in the experiment described above were chosen for evaluation of the sub-inhibitory concentrations (½MIC) of two antibiotics (clindamycin and imipenem). These antimicrobials were chosen due the large resistance of theB. fragilisstrains and to be used a lot in veterinary medicine.
For the CLSM assays, each well of a 12-well plastic tissue culture plate, with a 13-mm diameter glass coverslip placed on the bottom, wasfilled with 200
m
L of a 24 h-culture (0.5 McF) of each strain and 1.8 mL of Brucella broth and incubated under anaerobic conditions for 48 h at 37C. The biofilms grown on the coverslips werefixedwith 3.7% paraformaldehyde at room temperature for 30 min and were stained with the LIVE/DEADÒ
BacLightÔBacterial Viability
Kits (InvitrogenÒ
). Fluorescence from biofilms was documented using a CLSM (OlympusÒ
, using program FV10-ASW and version 01.07).
3. Results
The following species were identified of the 30 isolates from the fecal samples: B. fragilis (50%), Parabacteroides distasonis (14%), Bacteroides vulgatus,Bacteroides thetaiotaomicron,Bacteroides ova-tus(10%, each one), Parabacteroides merdaeandBacteroides egger-thii(n¼3%, each one).
The isolates showed lower susceptibility to penicillin, tetracy-cline and clindamycin (0%, 50% and 33%, respectively). The strains were uniformly susceptible to amoxicillin-clavulanic acid, cefoxitin, chloramphenicol, imipenem and metronidazole (Table 1).
The MIC90values for enrofloxacin and ciprofloxacin were 2 and
32
m
g/mL, respectively. Furthermore, the clindamycin showed the highest MIC50and MIC90values from all isolates.TheB. fragilisstrains were investigated due to their ability to adhere in vitro to plastic tissue culture plates to form biofilm. Among B. fragilis isolates studied by the first method, 8 (62%) strains were capable of producing biofilm.
metronidazole had a higher reduction in the formation of biofilms of isolates.
The reduction percentages were determined, using the control (strains without antimicrobials) like standard for biofilm growth. Variable reduction in biofilm formation for the four strains was seen, with cefoxitin, clindamycin and chloramphenicol. The reduction in biofilm production of all four strains was uniformly and significantly higher (P<0.05) after growth with ½ MIC and ¼
MIC of imipenem and metronidazole. Enrofloxacin at ½ MIC caused a significant increase in biofilm formation in two of four strains examined (Table 2). The data presented inTable 2also indicate that ½ MIC was the most active antibiotic concentration in biofilm reduction.
Control strains 29C2 and 44C7 produced the highest biofilm density as seen under confocal microscopy (Fig. 1). Imipenem was the drug that showed the greatest ability of biofilm inhibition in the four strains, with significant reduction of cell viability. On the other hand, clindamycin showed less action of biofilm inhibition in three strains. For the strain 33C2, it has been observed in confocal mi-croscopy that clindamycin killed the bacteria but the bacteria remained adhered to the surface.
4. Discussion
Infections caused by anaerobic bacteria in animals are similar to those seen in humans and include abscesses, osteomyelitis, myositis, aspiration pneumonia and peritonitis. TheB. fragilisgroup strains are the anaerobic non-sporulated microorganisms most commonly associated with those infections[3].B. fragiliswas the species most frequently isolated from dog’s intestinal tract fol-lowed byB. vulgatus, P. distasonis,B. thetaiotaomicron, B. ovatus, P. merdaeandB. eggerthii.
The antimicrobial susceptibility tests detailed in this study revealed, as in other reports, that penicillin is not active against Bacteroidesand Parabacteroides. In recent decades, an important increase in resistance to cephamycins has been reported for the B. fragilis group isolated from humans. These results were also found by our study, with resistance of 7%. In contrast, the rate of resistance to amoxicillineclavulanic acid, metronidazole and
imi-penem is still low, as in human infections[13,14].
Clindamycin was long consideredfirst choice in the treatment choice of anaerobic infections. However, its resistance has signifi -cantly increased over the past two decades[14]. This was evident in this study, because only 33% of the isolates were susceptible to clindamycin. These results are even more important considering that the isolates were isolated from the normal gut microbiota of animals.
Ciprofloxacin and enrofloxacin are antimicrobials commonly used in veterinary medicine but no criteria for breakpoint suscep-tibilities have been established for these quinolones[15]. Studies testing ciprofloxacin against anaerobic bacteria isolated from human infections have found MIC50and MIC90values of 8
m
g/mL and 32m
g/ mL respectively[16]. The results obtained in our study with dogs strains are similar. On the other hand, the MIC50and MIC90values ofenrofloxacin in this study were lower than that those described elsewhere (MIC50and MIC90of 2
m
g/mL and 8m
g/mL)[2].The high antimicrobial susceptibility observed may be due to low prescription of antibiotics at the Veterinary Hospital Unit of
Table 1
The Minimal Inhibitory concentration (MIC) values for antimicrobial ofBacteroides andParabacteroidesstrains isolated from intestinal microbiota of dogs.
Antimicrobial MIC (mg/mL) Susceptibility
Range 50 90
Penicillin 1e128 32 64 0%
Amoxicillin-clavulanic acid 0.25/0.125e8/4 0.5/0.25 2/1 97%
Cefoxitin 1e64 8 16 93%
Imipenem 0.125e4 0.5 2 100% Clindamycin 0.5e128 16 128 33% Chloramphenicol 0.25e8 2 4 100% Metronidazole 0.125e8 1 2 100% Tetracyclin 0.064e32 8 32 50%
Enrofloxacin: Range MIC: 0.125e4; MIC50: 1; MIC90: 2mg/mL. Ciprofloxacin: Range MIC: 8e32; MIC50: 8; MIC90: 32.mg/mL.
Table 2
Reduction of biofilm formation by the effect of sub-inhibitory concentrations of six different antimicrobials inB. fragilisisolates (n¼4), after 48 h of incubation.
Drug Strains MICmg/mL OD 540 Control OD 540 ½MIC %Rd OD 540 ¼ MIC %Rd
Cefoxitin 29C2 4 0.77bb 0.5727b 25.62% 0.7303 5.15%
33C2 4 0.373aa 0.1567b 57.98% 0.2488 33.29%
42C3 8 0.453aa 0.269b 40.57% 0.2932 35.23%
44C7 8 0.872aa 0.3067b 64.80% 0.3486 60.00%
Clindamycin 29C2 16 0.77aa 0.1272b 83.48% 0.3079 60.00%
33C2 16 0.373bb 0.3292b 11.74% 0.3256 12.70%
42C3 8 0.453ab 0.201a 55.59% 0.3684 18.62%
44C7 4 0.872ab 0.4257b 51.18% 0.4755 45.50%
Chloramphenicol 29C2 8 0,77aa 0.104b 86.49% 0.1098 85.74%
33C2 2 0.373ab 0.248b 33.51% 0.3229 13.43%
42C3 1 0.453aa 0.2632b 41.90% 0.284 37.26%
44C7 1 0.872aa 0.2851b 67.30% 0.4833 44.60%
Enrofloxacin 29C2 0.125 0.77bb 0.5707b 25.88% 0.5661 26.48%
33C2 1 0.373bb 0.5098b 36.67% 0.4552 22.03%
42C3 0.25 0.453bb 0.6632b 46.49% 0.4995 13.34%
44C7 0.25 0.872ab 0.4059b 53.45% 0.5536 36.50%
Imipenem 29C2 0.5 0.77aa 0.09624b 87.50% 0.0957 87.57%
33C2 0.125 0.373aa 0.1619b 56.59% 0.1427 61.74%
42C3 1 0.453aa 0.1373b 69.67% 0.1538 63.61%
44C7 0.5 0.872aa 0.1654b 81.00% 0.21 75.90%
Metronidazole 29C2 1 0.77aa 0.1794b 76.70% 0.2159 71.96%
33C2 1 0.373ab 0.1713b 54.07% 0.1933 48.17%
42C3 0.5 0.453aa 0.2307b 49.00% 0.2419 46.56%
44C7 2 0.872ab 0.5763b 33.90% 0.5744 34.10%
%Rd¼Reduction percentage of biofilm formation (compared to control).
aa¼significantly reduced comparing control with ½MIC and ¼MIC, respectively.
a
¼Significantly reduced (comparing ½MIC and ¼MIC).
b
State University of Ceará. However, clindamycin and tetracycline resistance rates were high. Tetracycline is not prescribed for ani-mals with less than 1 year of age, which was the age of the aniani-mals examined. In light to the high clindamycin and tetracycline resis-tance rates found in this study, it appears necessary to determine whether there are environmental selection pressures, other than antibiotic use, that contribute to the spread and maintenance of resistance genes and that explain the high level of resistance in areas where antibiotics appear not to be present[17].
As B. fragilis is considered the most virulent species in the
B. fragilis group and its capacity to form biofilms has been
demonstrated, it was chosen in this study for the biofilm formation inhibition assays using sub-inhibitory antibiotic concentrations
[12,18,10]. The concentrations of ½MIC and ¼ MIC were used. Mi-croorganisms often grow in the presence of sub-MICs, which although not able to inactivate microorganisms are potentially capable of altering the chemical and physical cell-surface charac-teristics and consequently the functionality and expression of some virulence properties such as adhesion, biofilm formation, hydro-phobicity and motility[19].
Most studies on the effects of sub-MICs of antibiotics have focused largely on Escherichia coli and Pseudomonas aeruginosa
[20,19]and to the best of our knowledge, there is little information in the literature concerningB. fragilis. Imipenem showed a very good effect on biofilm reduction and clindamycin showed approximately 50% inhibition of biofilm formation. In contrast, enrofloxacin ½MIC was able to significantly induce the production of biofilm in two strains. More studies must be done for confirm this result.
The results of this study must be taken into account when choosing the antibiotics for treatment of chronic infections in dogs. Moreover, these results clearly indicate that all three antibiotics tested are able to interfere with biofilm formation by theB. fragilis strains tested.
Acknowledgments
The authors are grateful to the Brazilian agency CNPq for financial support and Vicerrectoría de Investigación at Universidad de Costa Rica for partialfinancial support. We would like to thank the veterinarians of the Veterinary Clinic of UECE-Brazil, José Olavo Moraes (UFC-Brazil), Pablo Vargas and Diana López-Ureña (UCR-Costa Rica) for their technical assistance.
References
[1] Hirsch DC, MacLachlan NJ, Walker RL. Veterinary microbiology. 2nd ed. Iowa State University Press; 2004. p. 536.
[2] Silley P, Stephan B, Greife HA, Pridmore A. Comparative activity of prado-floxacin against anaerobic bacteria isolated from dogs and cats. J Antimicrob Chemother 2007;60:999e1003.
[3] Wagner KA, Hartmann FA, Trepanier LA. Bacterial culture results from liver, gallbladder, or bile in 248 dogs and cats evaluated for hepatobiliary disease: 1998e2003. J Vet Intern Med 2007;21:417e24.
[4] Ledbetter EC, Scarlett JM. Isolation of obligate anaerobic bacteria from ulcer-ative keratitis in domestic animals. Vet Ophthalmol 2008;11:114e22. [5] Wexler HM. Bacteroides: the good, the bad, and the nitty-gritty. Clin Microbiol
Rev 2007;20:593e621.
[6] Jacques M, Aragon V, Tremblay YDN. Biofilm formation in bacterial pathogens of veterinary importance. Anim Health Res Rev 2010;11:97e121.
[7] Pumbwe L, Skilbeck CA, Nakano V, Avila-Campos MJ, Piazza RM, Wexler HM. Bile salts enhance bacterial co-aggregation, bacterial-intestinal epithelial cell adhesion, biofilm formation and antimicrobial resistance of Bacteroides fra-gilis. Microb Pathog 2007;43:78e87.
[8] Pumbwe L, Skilbeck CA, Wexler HM. Presence of quorum-sensing systems associated with multidrug resistance and biofilm formation in Bacteroides fragilis. Microb Ecol 2008;56:412e9.
[9] Clutterbuck AL, Woods EJ, Knottenbelt DC, Clegg PD, Cochrane CA, Percival SL. Biofilms and their relevance to veterinary medicine. Vet Microbiol 2007;121: 1e17.
[10] Donelli G, Vuotto C, Cardines R, Mastrantonio P. Biofilm-growing intestinal anaerobic bacteria. FEMS Immunol Med Microbiol 2012;10:1574e695. [11] CLSI Clinical and Laboratory Standards Institute. Performance standards for
antimicrobial susceptibility testing of anaerobic bacteria: Informational Sup-plement M11S1. 940 West Valley Road, Wayne, P.A., U.S.A: Clinical and Lab-oratory Standards Institute; 2011.
[12] Sproule-Willoughby KM, Stanton MM, Rioux KP, McKay DM, Buret AG, Ceri H. In vitro anaerobic biofilms of human colonic microbiota. J Microbiol Methods 2010;83:296e301.
[13] Hawser SP, Hackel M, Hoban DJ. Antibiotic susceptibility profiles of European Bacteroides fragilis with reduced carbapenem susceptibility. J Antimicrob Chemother 2010;65:803e4.
[14] Snydman DR, Jacobus NV, McDermott LA, Golan Y, Hecht DW, Goldstein EJ, et al. Lessons learned from the anaerobe survey: historical perspective and review of the most recent data (2005e2007). Clin Infect Dis 2010;50:26e33. [15] CLSI Clinical and Laboratory Standards Institute. Methods for antimicrobial susceptibility testing of anaerobic bacteria. 7th ed. Wayne, P. A., U. S.A: CLSI; 2007. Approved Standard M11-A7.
[16] Goldstein EJ, Citron DM. Comparative activity of ciprofloxacin, oflox-acin,sparfloxacin, temafloxacin, CI-960, CI-990, and WIN 57273 against anaerobic bacteria. Antimicrobial Agents Chemother 1992;36:1158e62.
[17] Salyers AA, Amabile-Cuevas CF. Why are antibiotic resistance genes so resistant to elimination? Antimicrobial Agents Chemother 1997;41:2321e 5.
[18] Guaglianone E, Cardines R, Vuotto C, Di Rosa R, Babini V, Mastrantonio P, et al. Microbial biofilms associated with biliary stent clogging. FEMS Immunol Med Microbiol 2010;59:410e20.
[19] Majtán J, Majtánová L, Xu M, Majtán V.In vitroeffect of subinhibitory con-centrations of antibiotics on biofilm formation by clinical strains ofSalmonella entericserovarTyphimuriumisolated in Slovakia. J Appl Microbiol 2008;104: 1294e301.
