SEARCH

SEARCH BY CITATION

Keywords:

  • antibiotic compatibility;
  • Japanese probiotics

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Results
  6. Discussion
  7. Acknowledgement
  8. References

Aim:  To determine the antimicrobial resistance of the Japanese probiotics available in the market without a pharmacist’s supervision.

Methods and Results:  A total of 43 isolates were obtained from 40 samples of probiotics (30 dairy products and 10 products in tablet form). Isolates were identified using 16S rRNA gene sequencing and tested for their susceptibility to 14 antimicrobials. They were screened using PCR for some antibiotic resistance genes. Inactivation of cefepime, clarithromycin and vancomycin by different inocula of 11 strains was evaluated using the antibiotic inactivation bioassay. None of the dairy probiotics showed a level of constitutive resistance or carried inducible resistance genes, making them suitable to be administrated with macrolides. Among the probiotics in tablet form only Enterococcus faecium strains carrying the msrC gene showed an MIC90 of 4 μg ml−1. Extended-spectrum β-lactams, tetracyclines and ampicillin exhibited powerful germicidal activity against the vast majority of the probiotic strains.

Conclusions:  There is a limited choice of the Japanese probiotics that can be administered with clinically used antibiotics.

Significance and Impact of the Study:  Japanese probiotics are widely distributed all over the world. Through the findings of our study, we have attempted to provide guidance for clinicians interested in using the Japanese probiotics in combination with antibiotics.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Results
  6. Discussion
  7. Acknowledgement
  8. References

Whereas the discovery and use of antimicrobial agents has been one of medicine’s greatest achievements, frequent and lengthy use of antibiotics usually results in alteration in the complexity of the intestinal commensal flora. This enhances the ability of otherwise silent intestinal microbes to cause disease. How antibiotic-mediated elimination of commensal bacteria promotes infection by other bacteria is still a fertile area for speculation, with few defined mechanisms (Pamer 2007; Brandl et al. 2008). Moreover, preventive therapy for diseases and side effects emerging during treatment with antibiotics, such as antibiotic-associated diarrhoea (AAD), Clostridium difficile disease and infection with antibiotic-resistant bacteria (Pamer 2007; Brandl et al. 2008), is still not available.

Medical professionals in this era have found that prescribing probiotics with antibiotics is an effective approach in counteracting the adverse effects of the antibiotics. In addition, some viable probiotics have shown strong germicidal activity against certain pathogens, such as Helicobacter pylori (Zou et al. 2009). As a consequence, they are coprescribed with antibiotics. This raises an important question: What are the rationales for the choice of probiotics to be coadministered with antibiotics? We believe that a deeper understanding of their antibiotic resistance mechanisms will lead to the development of more effective coadministration protocols. The conclusion that probiotics are unlikely to offer a protective effect in vivo, when they are themselves susceptible to the same antibiotics considered to be at high risk of causing AAD (Gould and Short 2008) may underestimate some valuable probiotic strains. A bacterial strain carrying an inducible antibiotic resistance gene may appear susceptible to some antibiotics in in vitro tests, but have the potential to develop resistance to these antibiotics upon in vivo selection by the appropriate antibiotic (Schmitz et al. 2002).

On the other hand, should maintaining the potency of the antibiotic in the probiotic–antibiotic combination therapy be considered a safety criterion in this therapeutic approach? It has been known that the relative potency of vancomycin declines with the density of bacteria exposed (Udekwu et al. 2009). Reduction in the effective concentration of the antibiotic (potency) in the medium may be because a number of factors that include antibiotic-inactivating or denaturing enzymes, binding of the antibiotic to cell structures and the DNA or RNA of killed as well as viable bacteria (Hunt et al. 1987; Davies 1994; Udekwu et al. 2009). This may lead to the hypothesis that probiotics that are usually ingested in large numbers, during probiotic–antibiotic combination therapy, may interact antagonistically with antibiotics. Antibiotic inactivation bioassay is a simple and reliable technique for studying this theory.

In recent years, Japan has undergone regulatory reform. In addition to the large variety of dairy probiotics available in the Japanese market, some of the over-the-counter drugs, including probiotics with well established safety records, have been reclassified into the category of ‘quasi drugs’ or as functional foods to reduce health care costs (Amagase 2008). These products are no longer limited to being sold at drug stores and can now be sold more widely in the marketplace. It is unclear whether the probiotics available in the Japanese market are compatible for use in probiotic–antibiotic combination therapy. Phenotypic and molecular assessment of their antimicrobial resistance is needed to elucidate this issue. Therefore, the objective of this study was to determine the spectrum of antibiotic resistance of the probiotics available in the Japanese market and to verify their genetic mechanisms of resistance.

Material and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Results
  6. Discussion
  7. Acknowledgement
  8. References

Collection of samples and isolation of bacterial strains

A total of 40 samples of probiotic products, comprising 30 dairy products and 10 products in tablet form produced by different companies, were collected from Japanese supermarkets. Bacterial isolation was performed as follows. Briefly, after vigorous shaking of the product (probiotic or probio-yogurt) in its closed package, 1 ml was mixed with 9 ml buffered peptone water (Difco, Becton Dickinson and Co., Sparks, MD, USA) and incubated for 3 h at 37°C. From the incubated dairy products, loopfuls were streaked on to De Man Rogosa and Sharpe Agar medium (Merck, Darmstadt, Germany) supplemented with 0·3 g l−1 cysteine hydrochloride (Sigma-Aldrich, Tokyo, Japan) for the isolation of Lactobacillus and Bifidobacterium strains and on to M17 medium (Merck) for isolation of Streptococcus strains.

In the case of products in tablet form, one tablet was dissolved in 9 ml buffered peptone water and incubated for 3 h at 37°C and then, according to the strain names declared on the product labels, loopfuls were streaked on different media. In addition to the two previously mentioned media, Kanamycin esculin azide agar (Merck) was used for the isolation of Enterococcus spp. All plates with different media were incubated under standard conditions.

DNA extraction

Total genomic DNA was extracted using the DNeasy Tissue kit (Qiagen, Tokyo, Japan) following the manufacturer’s protocol for Gram-positive bacteria.

Identification of isolates

The 16S rRNA gene was amplified by PCR using primers as described by Frank et al. (2008). Another set of primers consisting of rrs (forward) (Kariyama et al. 2000) and 1492r (reverse) (Frank et al. 2008) were used to amplify 16S rRNA gene in DNA templates that could not be amplified by the first set of primers. For species identification of enterococcal isolates, a pair of degenerate primers was used to characterize a 438-bp-long DNA fragment internal to the sodA gene encoding the manganese-dependent superoxide dismutase (Poyart et al. 2000). The PCR products were purified using a QIAquick PCR purification kit (Qiagen) according to the manufacturer’s instructions and subsequently commercially sequenced using an ABI 3730 XL automated sequencer (Applied Biosystems, Foster City, CA, USA).

Antimicrobial susceptibility tests

Characterized strains were tested for their susceptibility to 14 antimicrobial agents by the disc diffusion method. They were as follows (μg): ampicillin (10), cefepime (30), cefotaxime (30), cefpirome (30), ceftriaxone (30), ciprofloxacin (5), clarithromycin (15), erythromycin (15), gentamicin (10), kanamycin (30), metronidazole (50), streptomycin (10), tetracycline (30) and vancomycin (30). They were tested according to the guidelines of the Clinical and Laboratory Standards Institute (CLSI) [formerly, the National Committee for Clinical Laboratory Standards (NCCLS)] (CLSI 2005) but with the following modifications: plates of lactic acid bacteria (LAB) susceptibility test medium supplemented with cysteine (LSM + cysteine) [i.e. 90% Iso-Sensitest broth (Oxoid, Wesel, Germany), 10% MRS broth (Merck) and 15 g l−1 agar supplemented with 0·3 g l−1 l-cysteine–HCl (Sigma)] were used for all LAB and bifidobacteria (Klare et al. 2005). Streptococcus thermophilus sensitive medium (90% Iso-Sensitest broth, 10% M17 broth, 0·5% lactose and 15 g l−1 agar) was used for Strep. thermophilus strains (Tosi et al. 2007). Muller Hinton agar (Oxoid) was used for non-LAB. The antimicrobial discs (Table 1) were purchased from Nissui Pharmaceutical Co. (Tokyo, Japan), with the exception of metronidazole, which was purchased from Oxoid. The minimal inhibitory concentrations (MICs), expressed in μg ml−1, of the same antibiotics purchased from different companies (Bristol-Myers Squibb, Tokyo Japan; Nacalai Tesque, Kyoto, Japan; Sigma-Aldrich; Wako Chemical Co., Osaka, Japan) were determined using the broth microdilution method, using the same media without agar, according to the guidelines of CLSI (CLSI 2005).

Table 1.   Susceptibility of probiotic bacteria to antimicrobial agents using the disc diffusion method
Species (number of strains)Inhibition zone diameter range (mm)
CLR 15 μgERY 15 μgVAN 30 μgTET 30 μgSTR 10 μgGEN 10 μgKAN 30 μg
  1. AMP, ampicillin; CFR, cefpirome; CIP, ciprofloxacin; CLR, clarithromycin; CRO, ceftriaxone; CTX, cefotaxime; ERY, erythromycin; FEP, cefepime; GEN, gentamicin; KAN, kanamycin; MTR, metronidazole; STR, streptomycin; TET, tetracycline; VAN, vancomycin.

  2. *Pin point colonies often observed within inhibition zone.

  3. †Haze of growth observed within inhibition zone.

Lactobacillus casei (12)29–3525–30630–338–1019–2414–16
Lactobacillus acidophilus (6)31–3730–3329–3030–3312–2614–256–20
Lactobacillus gasseri (3)31–3529–3022–2327–2814–1510–156
Lactobacillus delbrueckii ssp. bulgaricus (6)30–3728–3823–2529–3313–2211–2614–28
Lactobacillus plantarum (2)22–2519–21615613–1611–13
Lactobacillus helveticus (1)34*30*2933262325
Lactobacillus brevis (1)2926615112014
Streptococcus thermophilus (3)28–3029–3022–2528–3211–1214–1515–17
Bifidobacterium animalis ssp. lactis (3)30–4030–3225–2720–2266–96
Bifidobacterium longum (2)29–3031–322530–35666
Enterococcus faecium (4)14–1613–1421–2222–266–1213–146
 CIP 5 μgMTR 50 μgCTX 30 μgCRO 30 μgFEP 30 μgCFR 30 μgAMP 10 μg
Lact. casei (12)14–17625–2915–196–1522–2723–26
Lact. acidophilus (6)6629–3325–2725–3030–3432–36
Lact. gasseri (3)6629–3123–2627–3029–3727–30
Lact. delbrueckii ssp. bulgaricus (6)11–13625–3728–3615–3725–3532–38
Lact. plantarum (2)6630–3229–3130–3230–3320–25
Lact. helveticus (1)664035283542
Lact. brevis (1)661717141624
Strep. thermophilus (3)19–20630–3230–3330–3134–3634–35
Bif. animalis ssp. lactis (3)11–1239–41*30–3230–3636–3834–3630–38
Bif. longum (2)1737–3932–3529–3331–3635–3732–35
Ent. faecium (4)12–16624–26†20–23†25–29†24–28†27–29

Antibiotic inactivation bioassays

Inactivation of clarithromycin, cefepime and vancomycin by one strain representing each probiotic species (total 11 strains) was tested as follows: three tubes of LSM broth supplemented with cysteine and containing either 10 μg ml−1 of clarithromycin or cefepime or 500 μg ml−1 vancomycin were inoculated with exponential cultures of these strains until the concentrations were equal to 0·5 and 1 McFarland standard (for Enterococcus faecium strains, Muller–Hinton broth was used instead of LSM broth). Then, they were incubated for 12 h at 37°C. After centrifugation, sample aliquots of these cultures were filtered (0·22 μm) to remove the cells after 6 and 12 h. Thirty microlitres of the supernatants was deposited on sterile discs over Muller–Hinton agar plates, previously seeded with Kocuria rhizophila ATCC 9341 as an indicator organism for clarithromycin and cefepime, and with a vancomycin sensitive Bacillus subtilis laboratory strain as indicator for vancomycin bioassay. The plates were incubated for 18 h and the zone sizes around the discs, which indicate the antibiotic remaining in the culture medium, were measured. The zone sizes were compared with those produced when the antibiotic was incubated with medium alone.

The broth microdilution assay for the detection of effective residual concentration of antibiotics in the culture medium, which is thought to be more sensitive than the disc method, was carried out as described by Udekwu et al. (2009) to confirm the negative results. Briefly, probiotic cultures were challenged with each antibiotic and sample aliquots of these cultures were filtered as mentioned earlier. Then, overnight cultures of K. rhizophila ATCC 9341 (in case of clarithromycin and cefepime) or B. subtilis laboratory strain (in case of vancomycin) were diluted to c. 5 × 105 CFU ml−1, and the MICs of the filtered media were estimated by broth microdilution assay modified from the CLSI protocol.

Detection of antibiotic resistance genes

The presence of antibiotic resistance genes was investigated in all strains isolated in this work. PCR amplifications of genes associated with resistance to macrolides (ermA, ermB, ermC, msrA/B, ereA, ereB, mphA and mefA/E genes) (Sutcliffe et al. 1996), aminoglycosides (aph[3′], ant[6], aadE and aacA-aphD) (Van de Klundert and Vliegenthart 1993; Clark et al. 1999; Leelaporn et al. 2008), vancomycin (vanA, vanB, vanC1, and vanC2 or vanC3) (Bell et al. 1998), tetracycline (the ribosomal protection proteins degenerate primer pairs DI/DII (Clermont et al. 1997), Ribo-2-FW/Ribo-2-RV and tet(W) (Aminov et al. 2001) were carried out. Amplicons were purified and sequenced as previously described.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Results
  6. Discussion
  7. Acknowledgement
  8. References

Identification of isolates

16S rRNA sequence analyses using primers described by Frank et al. (2008) allowed species identification of nearly all strains. The alternative set of primers was needed for the identification of Lactobacillus delbrueckii ssp. bulgaricus, Lactobacillus helveticus and Lactobacillus acidophilus strains that did not have amplicons suitable for sequencing with the first set. From the dairy probiotics, the lactobacilli were Lactobacillus casei (n = 11), Lact. acidophilus (n = 5), Lactobacillus gasseri (n = 2), Lactobacillus plantarum (n = 2), Lact. delbrueckii ssp. bulgaricus (n = 6), Lactobacillus brevis (n = 1) and Lact. helveticus (n = 1). The streptococci were Strep. thermophilus (n = 3). The bifidobacteria were Bifidobacterium animalis ssp. lactis (n = 3) and Bifidobacterium longum (n = 2). From probiotics in tablet form, one strain each of Lact. casei, Lact. acidophilus and Lact. gasseri were identified from three different products. In addition, there were four Ent. faecium strains. In the case of dairy products, identified species were always in agreement with the strain names declared on the product label (when written), whereas in the case of products in tablet form, Ent. faecium strains were mislabelled as ‘Enterococcus faecalis’ in all the four products containing this species.

Antimicrobial susceptibility tests

The results of the antimicrobial disc diffusion susceptibility tests on 43 strains for 14 antibiotics are summarized in Table 1. Their MICs are listed in terms of MIC50 [MIC (μg ml−1) that inhibited 50% of the number of isolates tested], MIC90 [MIC (μg ml−1) that inhibited 90% of the number of isolates tested] and the MIC range (Table 2). Because no cut-off values have been officially defined for LAB or bifidobacterial species, the breakpoints established by the FEEDAP Panel of the European Food Safety Authority (EC 2008) were employed as a reference for most of the antibiotics tested. Using these cut-offs, none of the dairy probiotics showed a level of constitutive resistance for macrolides (MIC breakpoint 4 μg ml−1). Among the probiotics in tablet form, only Ent. faecium strains showed the highest natural resistance to erythromycin (MIC90 of 4 μg ml−1). High resistance levels to vancomycin were observed in Lact. casei, Lact. brevis and Lact. plantarum strains (MIC ≥128 μg ml−1). Indeed, heterofermentative lactobacilli are naturally resistant to glycopeptides and their MICs for vancomycin are not reliable for the differentiation of intrinsic and acquired resistance mechanisms (EC 2005). On the other hand, the MICs of vancomycin shown by obligate homofermentative lactobacilli including Lact. gasseri, Lact. delbrueckii ssp. bulgaricus and Lact. helveticus were lower than the MIC breakpoint of 4 μg ml−1 for vancomycin. Tetracycline and ampicillin resistant strains appeared to be rare in the Japanese market. Only Bif. animalis ssp. lactis (n = 3), Lact. brevis (n = 1) and Lact. plantarum (n = 2) showed moderate resistance levels to tetracycline (MIC range 8–16 μg ml−1). In contrast, both naturally resistant aminoglycosides and metronidazole strains were abundant. It should be noted that there was a tendency towards lower MICs for gentamicin when compared with kanamycin and streptomycin. This finding was inconsistent with the results obtained by Danielsen and Wind (2003). Pinpoint colonies were observed within the inhibition zone of metronidazole using the disc diffusion test for all bifidobacterial strains.

Table 2.   Susceptibility of probiotic bacteria to antimicrobial agents using broth microdilution method
AntibioticsSpecies (number of strains)MIC (μg ml−1)
MIC rangeMIC50MIC90
  1. MIC, minimal inhibitory concentrations.

  2. MIC50 and MIC90, MICs (μg ml−1) that inhibited 50 and 90% of the number of isolates tested, respectively.

  3. *All enterococcal strains were able to give a haze of growth in presence of cefotaxime, ceftriaxone and cefpirome till higher concentrations (cefotaxime and ceftriaxone MIC ≥128 and cefpirome MIC at 128).

ClarithromycinLactobacillus casei (12)≤0·125–0·25≤0·125≤0·125
Lactobacillus acidophilus (6)≤0·125≤0·125≤0·125
Lactobacillus gasseri (3)≤0·125≤0·125≤0·125
Lactobacillus delbrueckii ssp. bulgaricus (6)≤0·125≤0·125≤0·125
Lactobacillus plantarum (2)0·250·250
Lactobacillus helveticus (1)≤0·125
Lactobacillus brevis (1)0·25
Streptococcus thermophilus (3)≤0·125≤0·125≤0·125
Bifidobacterium animalis ssp. lactis (3)≤0·125≤0·125≤0·125
Bifidobacterium longum (2)≤0·125≤0·125
Enterococcus faecium (4)2–444
ErythromycinLact. casei (12)≤0·125–0·25≤0·125≤0·125
Lact. acidophilus (6)≤0·125≤0·125≤0·125
Lact. gasseri (3)≤0·125≤0·125≤0·125
Lact. delbrueckii ssp. bulgaricus (6)≤0·125≤0·125≤0·125
Lact. plantarum (2)0·25–0·50·25–0·5
Lact. helveticus (1)≤0·125
Lact. brevis (1)0·25
Strep. thermophilus (3)≤0·125≤0·125≤0·125
Bif. animalis ssp. lactis (3)≤0·125≤0·125≤0·125
Bif. longum (2)≤0·125≤0·125
Ent. faecium (4)444
VancomycinLact. casei (12)≥128≥128≥128
Lact. acidophilus (6)0·50·50·5
Lact. gasseri (3)111
Lact. delbrueckii ssp. bulgaricus (6)0·50·50·5
Lact. plantarum (2)≥128
Lact. helveticus (1)0·25
Lact. brevis (1)≥128
Strep. thermophilus (3)≤0·125–0·250·250·25
Bif. animalis ssp. lactis (3)0·50·50·5
Bif. longum (2)0·250·25
Ent. faecium (4)111
TetracyclineLact. casei (12)111
Lact. acidophilus (6)1–211
Lact. gasseri (3)1–222
Lact. delbrueckii ssp. bulgaricus (6)0·5–111
Lact. plantarum (2)88
Lact. helveticus (1)1
Lact. brevis (1)16
Strep. thermophilus (3)0·25–111
Bif. animalis ssp. lactis (3)888
Bif. longum (2)≤0·125–0·5≤0·125
Ent. faecium (4)0·5–20·50·5
StreptomycinLact. casei (12)8–1688
Lact. acidophilus (6)0·5–844
Lact. gasseri (3)4–3244
Lact. delbrueckii ssp. bulgaricus (6)0·5–888
Lact. plantarum (2)16–3216–32
Lact. helveticus (1)1
Lact. brevis (1)8
Strep. thermophilus (3)888
Bif. animalis ssp. lactis (3)323232
Bif. longum (2)1616
Ent. faecium (4)8–323232
GentamicinLact. casei (12)1–222
Lact. acidophilus (6)0·25–444
Lact. gasseri (3)4–844
Lact. delbrueckii ssp. bulgaricus (6)0·25–824
Lact. plantarum (2)2–82–8
Lact. helveticus (1)0·5
Lact. brevis (1)1
Strep. thermophilus (3)444
Bif. animalis ssp. lactis (3)8–161616
Bif. longum (2)1616
Ent. faecium (4)888
KanamycinLact. casei (12)161616
Lact. acidophilus (6)4–321616
Lact. gasseri (3)32–643232
Lact. delbrueckii ssp. bulgaricus (6)1–161616
Lact. plantarum (2)1616
Lact. helveticus (1)2
Lact. brevis (1)16
Strep. thermophilus (3)161616
Bif. animalis ssp. lactis (3)646464
Bif. longum (2)3232
Ent. faecium (4)128–≥128128128
CiprofloxacinLact. casei (12)1–222
Lact. acidophilus (6)8–323232
Lact. gasseri (3)646464
Lact. delbrueckii ssp. bulgaricus (6)444
Lact. plantarum (2)8–168
Lact. helveticus (1)64
Lact. brevis (1)32
Strep. thermophilus (3)1–211
Bif. animalis ssp. lactis (3)444
Bif. longum (2)22
Ent. faecium (4)2–444
MetronidazoleLact. casei (12)≥128≥128≥128
Lact. acidophilus (6)64–≥128≥128≥128
Lact. gasseri (3)≥128≥128≥128
Lact. delbrueckii ssp. bulgaricus (6)≥128≥128≥128
Lact. plantarum (2)≥128≥128≥128
Lact. helveticus (1)≥128
Lact. brevis (1)≥128
Strep. thermophilus (3)≥128≥128≥128
Bif. animalis ssp. lactis (3)444
Bif. longum (2)11
Ent. faecium (4)≥128≥128≥128
CefotaximeLact. casei (12)2–444
Lact. acidophilus (6)0·25–0·50·50·5
Lact. gasseri (3)0·5–10·51
Lact. delbrueckii ssp. bulgaricus (6)≤0·125–111
Lact. plantarum (2)0·250·25
Lact. helveticus (1)0·5
Lact. brevis (1)16
Strep. thermophilus (3)≤0·125≤0·125≤0·125
Bif. animalis ssp. lactis (3)0·50·50·5
Bif. longum (2)0·25–0·50·25–0·5
Ent. faecium (4)32*32*32*
CeftriaxoneLact. casei (12)8–161616
Lact. acidophilus (6)0·5–111
Lact. gasseri (3)0·25–0·50·5
Lact. delbrueckii ssp. bulgaricus (6)≤0·125–111
Lact. plantarum (2)≤0·125–0·25≤0·125–0·25
Lact. helveticus (1)1
Lact. brevis (1)16
Strep. thermophilus (3)≤0·125≤0·125≤0·125
Bif. animalis ssp. lactis (3)0·5–10·50·5
Bif. longum (2)1–21–2
Ent. faecium (4)16–32*32*32*
CefepimeLact. casei (12)16–321616
Lact. acidophilus (6)0·5–10·50·5
Lact. gasseri (3)≤0·125–0·250·250
Lact. delbrueckii ssp. bulgaricus (6)≤0·125–811
Lact. plantarum (2)≤0·125≤0·125
Lact. helveticus (1)2
Lact. brevis (1)16
Strep. thermophilus (3)≤0·125≤0·125≤0·125
Bif. animalis ssp. lactis (3)≤0·125≤0·125≤0·125
Bif. longum (2)≤0·125≤0·125
Ent. faecium (4)≥128≥128≥128
CefpiromeLact. casei (12)2–422
Lact. acidophilus (6)≤0·125–0·0·250·250·25
Lact. gasseri (3)≤0·125–0·250·250·25
Lact. delbrueckii ssp. bulgaricus (6)≤0·125–0·50·50·5
Lact. plantarum (2)≤0·125≤0·125
Lact. helveticus (1)1
Lact. brevis (1)8
Strep. thermophilus (3)≤0·125≤0·125≤0·125
Bif. animalis ssp. lactis (3)≤0·125≤0·125≤0·125
Bif. longum (2)≤0·125≤0·125
Ent. faecium (4)2–4*4*4*
AmpicillinLact. casei (12)0·5–111
Lact. acidophilus (6)0·250·250·25
Lact. gasseri (3)0·250·250·25
Lact. delbrueckii ssp. bulgaricus (6)≤0·125≤0·125≤0·125
Lact. plantarum (2)≤0·125–0·25≤0·125–0·25
Lact. helveticus (1)0·25
Lact. brevis (1)1
Strep. thermophilus (3)≤0·125≤0·125≤0·125
Bif. animalis ssp. lactis (3)≤0·125≤0·125≤0·125
Bif. longum (2)≤0·125–0·25≤0·125–0·25
Ent. faecium (4)0·5–10·51

Extended-spectrum β-lactams including cefpirome, cefepime, cefotaxime and ceftriaxone showed high germicidal activity against the vast majority of tested strains. Lactobacillus casei strains exhibited a different behaviour for this group. They showed high levels of resistance to cefepime and ceftriaxone but appeared almost susceptible to cefpirome and cefotaxime. On the other hand, all enterococcal strains showed zones of partial inhibition in the disc diffusion test for this group. Also, in the MIC test, all enterococcal strains were able to give a haze of growth in the presence of these antibiotics until higher concentrations were reached (cefotaxime and ceftriaxone MIC ≥128 μg ml−1 and cefpirome MIC 128 μg ml−1).

Ciprofloxacin exhibited different patterns of response in the tested strains. A wide range of MICs was observed in Lactobacillus species. However, according to the criteria suggested by Danielsen and Wind (2003) for lactobacilli (MIC breakpoint of 32 μg ml−1), nearly all strains, with the exception of Lact. gasseri and Lact. helveticus, appeared to possess intrinsic mechanisms of resistance.

Antibiotic inactivation bioassays

No decline in the effective concentrations of vancomycin, clarithromycin and cefepime in the filtrates was observed, after the incubation of heavy inoculums of the 11 tested strains with these three antibiotics.

Antibiotic resistance genes

Even though strains of the same species identified in this study appeared phenotypically resistant to antibiotics, they did not carry antibiotic resistance genes. This may be attributed to intrinsic mechanisms of resistance or the presence of antibiotic resistance genes that were not screened in this study. Positive amplicons were obtained using msrA/B primers from all enterococcal strains. Sequence analysis revealed the msrC gene. All Bif. animalis ssp. lactis with an MIC90 of 8 μg ml−1 for tetracycline were found to have positive amplicons using degenerate primers that identify all known ribosome-protection-type Tcr genes. All had the tet(W) gene as revealed from the sequencing of the positive amplicons using tet(W) specific primers.

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Results
  6. Discussion
  7. Acknowledgement
  8. References

There has been a significant worldwide rise in the sales and consumption of probiotic products being taken together with antibiotics. It is important that such probiotic products are correctly prescribed with antibiotics. In other words, antibiotic and probiotic combination therapy can only be considered to be compatible if the potency of the antibiotic and the viability of the probiotic are maintained. This study is the first to provide collective data on antibiotic compatibility of the Japanese probiotics. Through the findings of our study, we have attempted to provide guidance for clinicians interested in using the Japanese probiotics in combination with antibiotics and to improve the knowledge of the general public with regard to the safety aspects of the ingestion of probiotics during antibiotic therapy.

Probiotics are defined as ‘live micro-organisms administered in adequate amounts which confer a beneficial health effect on the host’ (FAO/WHO 2001). Starting from this vague definition and in the absence of approved standards for the phenotypic or genotypic evaluation of the antibiotic resistance of probiotics, moving towards a therapeutic approach with probiotics is complicated. This definition is based on the beneficial effects of viable bacteria and should not be confused with the probiotic effects achieved by nonviable bacteria or even by bacterial DNA (Szajewska et al. 2006). We believe that two important concepts should be considered when choosing this therapeutic approach. First, viable probiotics should reach their target site in the gastrointestinal tract in large enough numbers to achieve their effects. Therefore, they should have high natural constitutive resistance levels or nontransferable inducible antibiotic resistance genes enabling them to survive in the actual concentration of antibiotics that can be present in the human gastrointestinal tract. Second, when susceptible bacteria are exposed to bactericidal concentrations of an antibiotic, most of the population is killed within 1–2 h and some bacteria survive even in the presence of active antibiotic (Wiuff and Andersson 2007). Thus, in randomized double-blind placebo-controlled trials, the beneficial effects of probiotics isolated after the end of therapy should not always be attributed to viable normal probiotics.

Probiotic strains have been tested for their clinical efficacy against H. pylori infection (Zou et al. 2009). This infection is one of the most important established risk factors for the development of gastric cancer, with more than 100 000 new cases each year in Japan (Ferlay et al. 2002). The Lact. gasseri, Lact. acidophilus, Lact. casei and Bifidobacterium lactis Bb12 strains were found to amplify the eradication of this pathogen (Zou et al. 2009). However, physicians usually prescribe probiotics in combination with antibiotic therapy to achieve complete eradication. Helicobacter pylori eradication regimen, proton pump inhibitor + amoxicillin (AMPC) + clarithromycin (CAM), or PPI/AC is widely used all over the world and is the only currently approved H. pylori eradication regimen for insurance coverage in Japan. Surprisingly, according to our findings, none of the dairy probiotic strains investigated could be appropriately combined with this regime, as they were highly susceptible to clarithromycin (MIC90 of ≤0·125 μg ml−1). Indeed, the gastric concentration of clarithromycin (μg ml−1) was recorded to be 13 and 43 after 6 h of low and high dose regimes, respectively (Nakamura et al. 2003). Thus, it is not expected that these probiotics will remain viable if coadministered with clarithromycin. Additionally, these probiotics did not carry any inducible antibiotic resistance genes that could have enabled them to survive in vivo. Moreover, the same mentioned species isolated from probiotic products in tablet form were also very susceptible to clarithromycin. It should be noted that the MIC90 for macrolides obtained in this study for Lactobacillus species was lower than the MIC90 obtained in some previous studies (Danielsen and Wind 2003). However, it was inconsistent with the erythromycin MIC90 for the large number of Lactobacillus species tested in a European project intended to propose tentative epidemiological cut-off values for recognizing intrinsic and acquired antimicrobial resistances (Klare et al. 2007). On the other hand, the expected concentration of metronidazole, the second line of antimicrobial treatment, in the gastric juice was around 20 μg ml−1 at 1 h after administration (Calafatti 2000). This was higher than the MIC of Bif. animalis ssp. lactis strains. The appearance of pinpoint colonies around the metronidazole discs (50 μg) using the disc diffusion test indicated that the emergence of mutants is more likely to occur in vivo. However, the number of these mutants is not expected to reach the proposed number of probiotics that can deliver beneficial effects.

Cephalosporins are rapidly losing their usefulness as frontline antimicrobial agents, because of their potential to cause Cl. difficile-associated diarrhoea (Billyard 2007). The mode of excretion of these antibiotics results in an MIC concentration in the lower part of the intestinal tract that is reported to be higher than the MIC that can be tolerated by all of the intestinal microflora, leading to their partial elimination (Brismar et al. 1990). Using the currently proposed MIC breakpoint for ciprofloxacin (Danielsen and Wind 2003), our Lactobacillus strains appeared to be susceptible. Therefore, their survival during therapy is a cause for suspicion. Our in vitro results were inconsistent with the outcomes of previous clinical trials using similar Lactobacillus strains (Sullivan et al. 2003). Sullivan et al. (2003) could detect only low number of probiotic bacteria in stools of few treated patients after 10 days of quinolone therapy. This indicated the emergence of a few phenotypically resistant probiotics that could only be detected after this period.

In addition, the vast majority of the strains detected in this study were clearly susceptible to tetracycline and extended-spectrum β-lactams, making the choice of probiotics that remain viable with these antibiotics limited. A Bif. animalis ssp. lactis strain carrying the tet(W) gene, which had a similar MIC to the strains identified in this study, has been shown to survive during tetracycline therapy (Saarela et al. 2007). Its survival depended on the evolution of a genetic mechanism of resistance, rather than its natural ability to withstand the intestinal concentration of tetracycline. This strain had tetracycline MICs of >256, 256 and 32 μg ml−1 during therapy. More importantly, it did not revert to the susceptible state after growing in antibiotic-free media. This means that a genetic mechanism of resistance, rather than phenotypic resistance, was responsible for the acquired phenotype (Balaban et al. 2004).

Interestingly, all enterococcal strains identified in this study showed two unique characters. First, they carried msrC gene that is thought to confer low-level protection against the 14-membered-ring macrolides and type B streptogramins. This is the first report of msrC gene in Ent. faecium strains used as probiotics. It has been reported that msrC is prevalent, but not intrinsic, in Ent. faecium isolates and encodes an efflux pump that is likely to be of an inducible nature (Singh et al. 2001; Werner et al. 2001). From a clinical perspective, induction of this gene during macrolide therapy may enable these strains to survive in vivo. The second unique character of the isolated enterococcal strains appeared in antibiotic susceptibility tests of the extended-spectrum β-lactams (cefpirome, cefepime and cefotaxime). No clear-cut results were detected in both disc diffusion and MIC tests. This phenomenon has not been recorded in the guidelines for the susceptibility testing of Enterococcus spp. (CLSI 2005), but a similar phenomenon has been reported for anaerobes (National Committee for Clinical Laboratory Standards, NCCLS) (CLSI 2001). In such cases, MIC was determined as the lowest concentration of antimicrobial agent that resulted in a significant drop-off in the amount of growth. It should be noted that in the evaluation of the antibiotic resistance levels of probiotics, we should consider the survival of the proposed number of probiotics that are expected to give beneficial effects. Further clinical studies are still needed to clarify this point in these probiotic strains.

Taking into consideration the high cell density of probiotics in tablet form and in dairy products on the one hand and the traditional habitat of eating large amounts of probiotics in Asian countries on the other, we tried in this study to investigate the impact of 11 strains from different species with different cell components and genomes on the effective concentrations of three different classes of antibiotics. It was found that an increase in inoculum size of LAB resulted in elevation of their MICs to some antibiotics, including β-lactam (ampicillin) and macrolide (erythromycin) antibiotics (Egervärn et al. 2007). On the other hand, vancomycin was proved to be inactivated by heavy inoculums of vancomycin susceptible Staphylococcus aureus strains (Udekwu et al. 2009). It is unclear whether the heavy inoculum of probiotic bacteria has the ability to inactivate certain classes of antibiotics. Therefore, the interactions of cefepime, clarithromycin and vancomycin, antibiotics widely used in clinic, with probiotic bacteria were investigated in this study using antibiotic inactivation bioassay. The effective residual concentrations of the three antibiotics in the filtrate of culture media were not affected by heavy inoculums of the probiotic strains. Interestingly, LAB including certain species belonging to the genus Lactobacillus synthesize low affinity peptidoglycan precursors ending exclusively in D-Lac that lack the ability to bind with vancomycin (Handwerger et al. 1994; Deghorain et al. 2007). In agreement with this view, the heavy inoculums of resistant strains in this study showed no effect on the potency of vancomycin. Our in vitro finding clearly showed that bacterial components of probiotics had no antagonistic effect on the potency of antibiotics tested. Further studies are still needed to investigate whether the high cell density of probiotics adversely affects the rate and extent of antibiotic-mediated killing (the efficacy of antibiotic) during probiotic–antibiotic combination therapy.

With regard to a general concern about the safety of probiotics, such as potential transferability of resistance determinants, the vast majority of Japanese probiotics, with their low natural resistance to antibiotics tested, appear risk-free. The previous published data (Portillo et al. 2000; Flórez et al. 2006) concerning the antibiotic resistance genes identified in this study revealed that they were always chromosomally encoded. Moreover, given that transferable resistance genes might be present in all strains of a given species, further studies are also still needed to elucidate the mechanisms of resistance of some strains exhibiting high MICs.

In conclusion, according to the molecular and phenotypic characteristics of the Japanese probiotics detected in this study, probiotic–antibiotic combination therapy in many cases may not achieve the same positive outcomes as those obtained in clinical trials using viable probiotics alone. This is because many probiotic strains are susceptible to antibiotics used in clinic. In view of the similarity between our in vitro susceptibility profile of Japanese probiotics and the viability of probiotics demonstrated in previous clinical studies, clinicians are advised to use only those probiotic strains shown to be viable in the presence of antibiotics.

Acknowledgement

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Results
  6. Discussion
  7. Acknowledgement
  8. References

The authors gratefully acknowledge funding from a Grant-in-Aid for Scientific Research to T.S. from the Ministry of Education, Culture, Sports, Science and Technology of Japan.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Results
  6. Discussion
  7. Acknowledgement
  8. References
  • Amagase, H. (2008) Current marketplace for probiotics: a Japanese perspective. Clin Infect Dis 46(Suppl.), S73S75.
  • Aminov, R.I., Garrigues-Jeanjean, N. and Mackie, R.I. (2001) Molecular ecology of tetracycline resistance: development and validation of primers for detection of tetracycline resistance genes encoding ribosomal protection proteins. Appl Environ Microbiol 67, 2232.
  • Balaban, N.Q., Merrin, J., Chait, R., Kowalik, L. and Leibler, S. (2004) Bacterial persistence as a phenotypic switch. Science 305, 16221625.
  • Bell, J.M., Paton, J.C. and Turnidge, J. (1998) Emergence of vancomycin-resistant enterococci in Australia: phenotype and genotype characteristics of isolates. J Clin Microbiol 36, 21872190.
  • Billyard, T. (2007) Probiotics and diarrhoea. No high risk antibiotics? BMJ 335, 171.
  • Brandl, K., Plitas, G., Mihu, C.N., Ubeda, C., Jia, T., Fleisher, M., Schnabl, B., DeMatteo, R.P. et al. (2008) Vancomycin-resistant enterococci exploit antibiotic-induced innate immune deficits. Nature 455, 804807.
  • Brismar, B., Edlund, C., Malmborg, A.S. and Nord, C.E. (1990) Ciprofloxacin concentrations and impact of the colon microflora in patients undergoing colorectal surgery. Antimicrob Agents Chemother 34, 481483.
  • Calafatti, S.A. (2000) Transfer of metronidazole to gastric juice: impact of Helicobacter pylori infection and omeprazole. Scand J Gastroenterol 35, 699704.
  • Clark, N.C., Olsvik, O., Swenson, J.M., Spiegel, C.A. and Tenover, F.C. (1999) Detection of a streptomycin/spectinomycin adenylyltransferase gene (aadA) in Enterococcus faecalis. Antimicrob Agents Chemother 43, 157160.
  • Clermont, D., Chesneau, O., De Cespédès, G. and Horaud, T. (1997) New tetracycline resistance determinants coding for ribosomal protection in streptococci and nucleotide sequence of tet(T) isolated from Streptococcus pyogenes A498. Antimicrob Agents Chemother 41, 112116.
  • Clinical and Laboratory Standards Institute (2001) National Committee for Clinical Laboratory Standards, Approved Standard M11-A5. Methods for Antimicrobial Susceptibility Testing of Anaerobic Bacteria, 5th edn. Wayne, PA: CLSI.
  • Clinical and Laboratory Standards Institute (2005) National Committee for Clinical Laboratory Standards, Approved Standard M100-S15. Performance Standards for Antimicrobial Susceptibility Testing, 15th edn. Wayne, PA: CLSI.
  • Danielsen, M. and Wind, A. (2003) Susceptibility of Lactobacillus spp. to antimicrobial agents. Int J Food Microbiol 82, 111.
  • Davies, J.E. (1994) Inactivation of antibiotics and the dissemination of resistance genes. Science 264, 375382.
  • Deghorain, M., Goffin, P., Fontaine, L., Mainardi, J.L., Daniel, R., Errington, J., Hallet, B. and Hols, P. (2007) Selectivity for D-lactate incorporation into the peptidoglycan precursors of Lactobacillus plantarum: role of Aad, a VanX-like D-alanyl-D-alanine dipeptidase. J Bacteriol 189, 43324337.
  • Egervärn, M., Lindmark, H., Roos, S., Huys, G. and Lindgren, S. (2007) Effects of inoculum size and incubation time on broth microdilution susceptibility testing of lactic acid bacteria. Antimicrob Agents Chemother 51, 394396.
  • European Commission (2008) Opinion of the FEEDAP Panel on the updating of the criteria used in the assessment of bacteria for resistance to antibiotics of human or veterinary importance. EFSA J 732, 915.
  • FAO/WHO (2001) Joint FAO/WHO Expert Consultation on Evaluation of Health and Nutritional Properties of Probiotics in Food Including Powder Milk with Live Lactic Acid Bacteria. Cordoba, Argentina: FAO/WHO.
  • Ferlay, J., Bray, F., Pisani, P. and Parkin, D.M. GLOBOCAN (2002) Cancer Incidence, Mortality and Prevalence Worldwide. IARC CancerBase. No. 5, version 2.0. Lyon, France: IARC Press, 2004.
  • Flórez, A.B., Ammor, M.S., Álvarez-Martín, P., Margolles, A. and Mayo, B. (2006) Molecular analysis of tet(W) gene-mediated tetracycline resistance in dominant intestinal Bifidobacterium species from healthy humans. Appl Environ Microbiol 72, 73777379.
  • Frank, J.A., Reich, C.I., Sharma, S., Weisbaum, J.S., Wilson, B.A. and Olsen, G.J. (2008) Critical evaluation of two primers commonly used for amplification of bacterial 16S rRNA genes. Appl Environ Microbiol 74, 24612470.
  • Gould, K. and Short, G. (2008) Probiotics and antibiotic-associated diarrhoea – a logical flaw? J Antimicrob Chemother 61, 761.
  • Handwerger, S., Pucci, M.J., Volk, K.J., Liu, J.P. and Lee, M.S. (1994) Vancomycin-resistant Leuconostoc mesenteroides and Lactobacillus casei synthesize cytoplasmic peptidoglycan precursors that terminate in lactate. J Bacteriol 176, 260264.
  • Hunt, G.H., Oberholzer, V.G. and Price, E.H. (1987) In-vitro antibiotic inactivation by mammalian cell and killed bacterial preparations. J Antimicrob Chemother 19, 637646.
  • Kariyama, R., Mitsuhata, R., Chow, J.W., Clewell, D.B. and Kumon, H. (2000) Simple and reliable multiplex PCR assay for surveillance isolates of vancomycin-resistant enterococci. J Clin Microbiol 38, 30923095.
  • Klare, I., Konstabel, C., Müller-Bertling, S., Reissbrodt, R., Huys, G., Vancanneyt, M., Swings, J., Goossens, H. et al. (2005) Evaluation of new broth media for microdilution antibiotic susceptibility testing of lactobacilli, lactococci, pediococci, and bifidobacteria. Appl Environ Microbiol 71, 89828986.
  • Klare, I., Konstabel, C., Werner, G., Huys, G., Vankerckhoven, V., Kahlmeter, G., Hildebrandt, B., Müller-Bertling, S. et al. (2007) Antimicrobial susceptibilities of Lactobacillus, Pediococcus and Lactococcus human isolates and cultures intended for probiotic or nutritional use. J Antimicrob Chemother 59, 900912.
  • Leelaporn, A., Yodkamol, K., Waywa, D. and Pattanachaiwit, S.A. (2008) Novel structure of Tn4001-truncated element, type V, in clinical enterococcal isolates and multiplex PCR for detecting aminoglycoside resistance genes. Int J Antimicrob Agents 31, 250254.
  • Nakamura, M., Spiller, R.C., Barrett, D.A., Wibawa, J.I., Kumagai, N., Tsuchimoto, K. and Tanaka, T. (2003) Gastric juice, gastric tissue and blood antibiotic concentrations following omeprazole, amoxicillin and clarithromycin triple therapy. Helicobacter 8, 294299.
  • Pamer, E.G. (2007) Immune responses to commensal and environmental microbes. Nat Immunol 8, 11731178.
  • Portillo, A., Ruiz-Larrea, F., Zarazaga, M., Alonso, A., Martinez, J.L. and Torres, C. (2000) Macrolide resistance genes in Enterococcus spp. Antimicrob Agents Chemother 44, 967971.
  • Poyart, C., Quesnes, G. and Trieu-Cuot, P. (2000) Sequencing the gene encoding manganese-dependent superoxide dismutase for rapid species identification of enterococci. J Clin Microbiol 38, 415418.
  • Saarela, M., Maukonen, J., Von Wright, A., Vilpponen-Salmela, T., Patterson, A.J., Scott, K.P., Hämynen, H. and Mättö, J. (2007) Tetracycline susceptibility of the ingested Lactobacillus acidophilus LaCH-5 and Bifidobacterium animalis subsp. lactis Bb-12 strains during antibiotic/probiotic intervention. Int J Antimicrob Agents 29, 271280.
  • Schmitz, F.J., Petridou, J., Jagusch, H., Astfalk, N., Scheuring, S. and Schwarz, S. (2002) Molecular characterization of ketolide-resistant erm(A)-carrying Staphylococcus aureus isolates selected in vitro by telithromycin, ABT-773, quinupristin and clindamycin. J Antimicrob Chemother 49, 611617.
  • Singh, K.V., Malathum, K. and Murray, B.E. (2001) Disruption of an Enterococcus faecium species-specific gene, a homologue of acquired macrolide resistance genes of staphylococci, is associated with an increase in macrolide susceptibility. Antimicrob Agents Chemother 45, 263266.
  • Sullivan, A., Barkholt, L. and Nord, C.E. (2003) Lactobacillus acidophilus, Bifidobacterium lactis and Lactobacillus F19 prevent antibiotic-associated ecological disturbances of Bacteroides fragilis in the intestine. J Antimicrob Chemother 52, 308311.
  • Sutcliffe, J., Grebe, T., Tait-Kamradt, A. and Wondrack, L. (1996) Detection of erythromycin-resistant determinants by PCR. Antimicrob Agents Chemother 40, 25622566.
  • Szajewska, H., Setty, M., Mrukowicz, J. and Guandalini, S. (2006) Probiotics in gastrointestinal diseases in children: hard and not-so-hard evidence of efficacy. J Pediatr Gastroenterol Nutr 42, 454475.
  • Tosi, L., Berruti, G., Danielsen, M., Wind, A., Huys, G. and Morelli, L. (2007) Susceptibility of Streptococcus thermophilus resistance of to antibiotics. Antonie Van Leeuwenhoek 92, 2128.
  • Udekwu, K.I., Parrish, N., Ankomah, P., Baquero, F. and Levin, B.R. (2009) Functional relationship between bacterial cell density and the efficacy of antibiotics. J Antimicrob Chemother 63, 745757.
  • Van de Klundert, J.A.M. and Vliegenthart, J.S. (1993) PCR detection of genes coding for aminoglycoside-modifying enzymes. In Diagnostic Molecular Microbiology. Principles and Applications ed. Persings, D.H., Smith, T.F., Tenover, F.C. and White, T.J. pp. 547552. Washington, DC: American Society of Microbiology.
  • Werner, G., Hildebrandt, B. and Witte, W. (2001) The newly described msrC gene is not equally distributed among all isolates of Enterococcus faecium. Antimicrob Agents Chemother 45, 36723673.
  • Wiuff, C. and Andersson, D.I. (2007) Antibiotic treatment in vitro of phenotypically tolerant bacterial populations. J Antimicrob Chemother 59, 254263.
  • Zou, J., Dong, J. and Yu, X. (2009) Meta-analysis: Lactobacillus containing quadruple therapy versus standard triple first-line therapy for Helicobacter pylori eradication. Helicobacter 14, 97107.