Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Carbohydrate stress-related response in Bifidobacterium pseudolongum subsp. globosum

Carbohydrate stress-related response in Bifidobacterium pseudolongum subsp. globosum Ann Microbiol (2012) 62:1751–1756 DOI 10.1007/s13213-012-0432-9 ORIGINAL ARTICLE Carbohydrate stress-related response in Bifidobacterium pseudolongum subsp. globosum Stefano Tacconi & Barbara Sgorbati & Monica Modesto & Bruno Biavati & Lorenzo Nissen & Paola Mattarelli Received: 23 September 2011 /Accepted: 7 February 2012 /Published online: 1 March 2012 Springer-Verlag and the University of Milan 2012 . . . Abstract Bifidobacteria are indigenous components of hu- Keywords Bifidobacterium Plasmid Probiotic man and animal gastrointestinal microbiota, and their Environmental stress Carbohydrate starvation health-promoting benefits have long been recognized. Of the 36 currently described species of the Bifidobacterium genus, 8 contain plasmids, most of which are cryptic. It is Introduction possible that plasmid presence is related very closely to environmental change, so in conditions of stress this pres- Normal gut microbiota are involved in host physiology and ence could be specifically controlled. For plasmid-positive play a key role in animal health and disease. Bifidobacteria Bifidobacterium pseudolongum subsp. globosum RU809/1, are indigenous components of human and animal gastroin- the influence of the type and concentration of the carbohy- testinal microbiota and their health-promoting benefits have drate source is evident in the dramatic pVS809 curing effect long been recognized, although the molecular mechanisms when growth is conducted in the presence of 0.15% (w/v) underlying their beneficial effects are unclear (Lee and glucose, lactose, maltose, melibiose, raffinose or starch. The O’Sullivan 2010). Besides their benefits, such as immunity effect is linked to carbohydrate starvation, not to carbohy- system stimulation (Abt and Artis 2009), protection from drate abundance, and is independent of biomass growth. pathogen infection (Macfarlane and Cummings 2002; Plasmid curing was achieved after one or two consecutive Shirasawa et al. 2010), energy production from the host’s transfers, also in cells grown on medium containing 0.15% undigested carbohydrates, and vitamin production (Saulnier et arabinose, fructose, galactose and sucrose, but not mannose, al. 2009;Wongetal. 2006), there is a growing commercial ribose or xylose. Knowing plasmid behavior in stressful interest in the utilization of bifidobacteria as probiotics. More- conditions, like carbon source availability, has allowed an over, bifidobacteria were also identified recently as potential early insight into carbohydrate starvation as a curing agent expression vectors for proteins of pharmacological and food- for bifidobacteria. Furthermore, knowledge of plasmid be- related interest (Hidaka et al. 2007; Prather et al. 2003; havior in stressful conditions could be important not only in Sangrador-Vegas et al. 2007; Shareck et al. 2004). To facilitate genetics and ecology but also in food-grade and pharma- genetic manipulation of bifidobacteria, and the development ceutical applications for the development of cloning and and improvement of their cloning vectors and gene transfer expression vector systems for bifidobacteria. systems, a basic knowledge of plasmid genetics and physiol- ogy is very important. Bifidobacteria are subject to stressful conditions in both biotechnological processes and nature, and their ability to : : : : : respond quickly to stress is essential for survival. For bacterial S. Tacconi B. Sgorbati M. Modesto B. Biavati L. Nissen P. Mattarelli (*) cells living in a natural environment, one of the biggest prob- Department of Agroenvironmental Sciences and Technologies, lems is nutritional limitation—cells are starved most of the Bologna University, time (Matin 1991); the opposite occurs in the laboratory: Via Fanin 42, bacteria are usually cultivated under optimal or over-optimal 40127 Bologna, Italy growth conditions. Thus, in conditions of starvation or e-mail: paola.mattarelli@unibo.it 1752 Ann Microbiol (2012) 62:1751–1756 abundance, bacterial cultures undergo numerous metabolic normal tryptone-phytone yeast (nTPY) (pH 6.8) containing changes to facilitate survival, and the dynamism lies in the 1.5% (w/v) of glucose (Biavati and Mattarelli 2006). An terms of cell number and cell type. Also chromosomal and overnight culture at 4% (v/v) was used as inoculum. extrachromosomal DNA replication undergoes specific con- trols. Therefore, the investigation of plasmid maintenance in Growth in different glucose concentrations bacterial adaptation to stress is important. Plasmids have been detected in only 8 of the 36 known TPY medium supplemented with different concentrations Bifidobacterium species: in B. bifidum, B. breve, B. catenu- [0.1, 0.15, 0.2, 0.4, 0.6, 0.8, 1.0, 1.25, 1.5, 3.6, 5.4, 7.2, latum, B. pseudocatenulatum and B. longum subsp. longum 9.0, 10.8, 12.6, 14.4, 16.2, 18.0, 27.0% (w/v)] of glucose of the 13 human species, in B. pseudolongum subsp. globo- was used. After filtration, the glucose solutions were added sum species of the 17 animal species and in B. asteroides to the cooled sterile growth medium without sugar. All and B. indicum of the 6 insect species (Lee and O’Sullivan strains were grown in 500 mL medium at 37°C on each 2010; Sgorbati et al. 1982). It can be expected that species glucose concentration. The medium was inoculated as soon with plasmids that occur naturally in nature are more able to as possible after autoclaving with cells derived from an maintain them in a stable condition than other species. overnight 20 mL TPY culture centrifuged and resuspended Among the known species, B. pseudolongum subsp. glo- in glucose-free TPY to avoid inoculum interference; bosum, which is typical of animal habitats, has the most 0.5 mL/L NaHCO (10% w/v) filter sterilized solution was cosmopolitan distribution. It colonizes the intestine of cattle, added to the flask and incubation was conducted under air. pig, dog, rat and chicken (Biavati and Mattarelli 2006), and The experiments were conducted in three replications. was isolated recently from wild animals like the wallaby and springbok (Endo et al. 2010); furthermore, it has been sporad- Growth in different carbohydrates ically isolated from human feces (Mattarelli et al. 1993). A study by Turroni et al. (2010) divided the Bifidobacterium TPY medium was supplemented with different carbohy- genus into five phylogenetic groups and, on analyzing human drates at concentrations of 1.5% and 0.15% (w/v) in place microbiota, it was found that the 16S rRNA sequences of of glucose. Carbohydrates used were: arabinose, fructose, about 50% of unculturable bifidobacteria have high homology galactose, glucose, lactose, maltose, mannose, melibiose, with the "pseudolongum group". This species is typical of the raffinose, ribose, starch, sucrose, and xylose. Media were animal habitat, and this underlines its potential importance. autoclaved separately from the carbohydrates, which were Bifidobacterial plasmids are generally considered cryptic, added to the cooled broth after filtration. All strains were except for the plasmid from B. bifidum NCFB 1454, which grown in 500 mL medium inoculated on the appropriate was proposed as being involved in bifidocin B production, carbohydrate concentrations as described above. The experi- and the plasmid pVS809 from B. pseudolongum subsp. ments were conducted in three replications. globosum, which seems to be involved in leucine auxotro- phy (Mattarelli and Biavati 1999). Dry cell matter concentration Plasmid pVS809 occurs naturally in extrachromosomal DNA in B. pseudolongum subsp. globosum RU809/1, and, Aliquots (10 mL) of cell culture were centrifuged at 5,000 g on the basis of mapping restriction analysis, has been esti- for 10 min. The supernatant was decanted and the cell mated to be 28 Kb in size (Mattarelli et al. 1994). pellets washed with an equal volume of sterile water. The The present work investigates how carbon type and the washed cells were dried in a dry oven at 105°C for 24 h and concentration of carbon source availability—both important then weighed. stressful conditions in culture media—influence pVS809 stability and dry matter concentration in B. pseudolongum Plasmid analyses subsp. globosum RU 809/1. The aim of the study was to obtain significant information to increase our still incom- Plasmid DNA was isolated by a method described previously plete knowledge of plasmid physiology in bifidobacteria. (Sgorbati et al. 1982), and was resolved on 0.8% agarose gel at 80 V for 4 h. Semiquantitative videodensitometric analysis of the gel was performed using the Gel Doc™ XR + System Materials and methods (Bio-Rad, Hercules, CA). RU 809/1 was subcultured three times in TPY containing Bacterial strain, media, and growth different glucose concentrations, or different carbon sources at different concentrations, and each subculture was ana- The strain Bifidobacterium pseudolongum subsp. globosum lyzed for plasmid presence. Plasmid-negative or low plas- RU 809/1 was grown routinely anaerobically at 37°C in mid quantity cultures were retested and plated on 1.8% agar Ann Microbiol (2012) 62:1751–1756 1753 0.8 TPY containing the standard 1.5% glucose: 70 colonies 0.7 were isolated from each subculture in the appropriate dilu- tions. The cultures derived from each of the picked colonies 0.6 were analyzed for plasmid content. 0.5 0.4 Chromosomal insertion of plasmid DNA 0.3 The possibility of plasmid presence in a chromosomally 0.2 integrated form was tested using a pVS809 DNA probe 0.1 against RU809/1 chromosomal DNA from 30 plasmid- negative clones isolated from each of three consecutive subcultures at 0.15% (w/v) glucose concentration. The de- tailed procedure for plasmid chromosomal insertion has Fig. 2 Dry matter concentrations (mg/mL) of RU809/1 grown on been described previously (Mattarelli et al. 1994). 1.5% (w/v) of different carbohydrate as carbon source Statistical analysis Effect of different carbohydrates on RU809/1 performance All experiments were performed in triplicate. All quantita- Each of the 13 carbohydrates tested (arabinose, fructose, tive data, unless otherwise stated, are presented as means, galactose, glucose, lactose, maltose, mannose, melibiose, raf- with error represented by standard deviation. finose, ribose, sucrose, starch, xylose) was utilized as carbon source at a normal concentration of 1.5% in TPY medium. The dry matter concentration was similar in all the carbohy- Results drates tested, with a little higher growth in glucose (Fig. 2). When the carbohydrates were added at 0.15% w/v in TPY, the Effect of glucose on RU809/1 performance dry matter decreased one- or two-fold with respect to nTPY. The three tested subcultures show similar dry matter biomass, RU 809/1 was cultivated in a range of glucose concentra- with the exception of mannose, ribose and xylose that showed tions varying from 0.1 to 27% (w/v). All the glucose con- less dry matter, i.e., low growth rate (Fig. 3). centrations permitted growth with the exception of 27% (w/ v). The biomass value based on dry matter calculation is Morphological observations of cells grown at different sugar almost steady in the glucose concentration range of 0.4– concentrations 14.4% (w/v), while it diminishes at the highest and lowest glucose concentrations (Fig. 1). The maximum dry matter The RU809/1 cells grown at the extreme glucose concen- concentration (0.93±0.013 mg/ml) was obtained at a 3.6% trations (below or equal to 0.4% and above or equal to (w/v) glucose concentration. The dry matter of subcultures 16.2%) show a reduction in size with respect to that at at the same glucose concentration maintains the same values 1.5% (w/v), the nTPY glucose concentration (Fig. 4). This (data not shown). observation is in agreement with the literature, highlighting that bacterial cells can utilize different strategies to conserve energy during stress conditions, reducing not only their 0.9 growth rate but also their size (Gilbert et al. 1990; Novitsky 0.8 and Morita 1976). 0.7 0.6 Effect of different glucose concentrations on the loss 0.5 of plasmid in RU809/1 0.4 0.3 RU809/1 cells grown in nTPY possess one large plas- 0.2 mid of 28,000 bp (Mattarelli et al. 1994): this plasmid 0.1 band was considered as a positive control in the fol- 0 2 4 6 8 101214161820 lowing experiments. %(w/v) glucose concentration The cells were subcultured once at glucose concentra- tions ranging from 0.1 to 18.0% (w/v), and the plasmid was Fig. 1 Dry matter concentrations of RU809/1 grown at different percentage (w/v) of glucose as carbon source present at concentrations equal to, or above, 0.4%. Cells mg/ml dry matter mg/ml dry matter 1754 Ann Microbiol (2012) 62:1751–1756 Fig. 3 Dry matter concentra- 0.60 tions (mg/mL) of RU809/1 grown on 0.15% (w/v) of differ- ent carbohydrates as carbon 0.50 source determined at first , second □ and third ■ consecutive 0.40 subcultures 0.30 0.20 0.10 0.00 grown at 0.2% (w/v) glucose showed reduced plasmid in- (w/v) (Table 1). Like cells grown in glucose, cells grown tensity, dropping to about 40% (densitometric analysis), and in lactose, maltose, melibiose, raffinose and starch lost the no plasmid band was observed at 0.15% and 0.1% (w/v) plasmid at the first subculture containing these carbohy- glucose concentrations. Plasmid-free and plasmid-positive drates (Fig. 5). Instead, in the three consecutive subcultures cultures were subcultured three times and the previous examined, mannose and ribose had no effect, the plasmid results were confirmed. being maintained without modification. However, after one Three consecutive subcultures derived from 0.2 and consecutive subculture in arabinose, fructose and sucrose, 0.15% (w/v) glucose cultures were produced. For each and after two in galactose, we again observed plasmid loss. subculture, 70 clones were obtained. Each of these clones Cells grown on medium containing xylose showed, in all was tested for the presence of plasmid: 35±3 plasmid pos- three subcultures, a diminished plasmid band, revealed by itive clones were obtained from 0.2% (w/v) glucose con- densitometric analysis (about 35% with respect to normal centration and 0 plasmid positive clones were obtained from density of plasmid positive cells), but not completely lost 0.15% glucose concentration. This result is in agreement pVS809 (Table 1). Given the result shown above for glucose with densitometric data. at 0.2%, it is conceivable that, also for the different carbo- hydrate sources, the plasmid decrease is not due to a copy Effect of different carbon source concentrations on plasmid number decrease but to the presence of mixed plasmid positive and negative populations. No direct relationship loss in RU809/1 was observed between dry matter concentration and plasmid The different carbon sources utilized at 1.5% (w/v) did not maintenance. For example, the cells grown in 0.15% (w/v) mannose, ribose or xylose, which had the lowest dry matter influence the pVS809 presence, but there were some changes in the plasmid content in cells grown at 0.15% concentrations, maintained the plasmid in all three subcultures: Fig. 4 Phase-contrast photomicrographs showing cellular morphology of RU 809/1 grown at a 0.15%, b 1.5% and c 18% (w/v) of glucose concentrations. Bar 10 μm mg/ml dry matter Ann Microbiol (2012) 62:1751–1756 1755 Table 1 Plasmid presence in three consecutive subcultures of RU809/ Discussion 1 grown at 0.15% (w/v) concentration of different carbohydrates. + Plasmid positive cells, +/− plasmid band intensity 60–65% of control The basic biology of plasmids in the Bifidobacterium species value, − plasmid-negative cells is unexpectedly poorly known. The biological significance of Carbon source Plasmid presence these plasmids is unclear as most of the open reading frames, apart from those predicted to specify replication functions, 1st subculture 2nd subculture 3rd subculture code for proteins of unknown function (Cronin et al. 2007). Arabinose +/− –– In general, naturally occurring plasmids are firmly maintained Fructose + –– (O’Riordan and Fitzgerald 1999), but no studies have been carried out on how stressing environmental conditions Galactose +/− +/− – affects plasmid stability in bifidobacteria, with the excep- Glucose –– – tion of plasmid curing experiments with chemical agents Lactose –– – (Mattarelli et al. 1994). Maltose –– – The survey of glucose concentrations, which ranged from Mannose + + + 18 to 0.1% (w/v), showed that plasmid stability became Melibiose –– – affected at 0.2% glucose concentration, with a loss of about Raffinose –– – 60%, while at concentrations of 0.15% and 0.1%, pVS809 Ribose + + + was lost completely. The partial plasmid loss was related to Sucrose + –– the presence of mixed plasmid-positive and -negative clones Starch –– – in the population, as revealed by plasmid analysis of clones Xylose +/− +/− +/− derived from different subcultures, rather than a lowering of plasmid copy number. A direct correlation was found be- tween low plasmid band intensity and the number of mannose and ribose at 100% while in xylose, the plasmid is plasmid-positive clones present, and the presence of chro- maintained with a value of 65%. mosomally integrated plasmid in plasmid cured clones was not detectable. Therefore, there is a strong relationship be- Plasmid-chromosome integration tween plasmid presence and glucose concentration, the in- tensity of this relationship being such that glucose starvation The results of the hybridization of the pVS809 probe to the could be comparable to the 50 μg/ml ethidium bromide chromosomal DNA in 30 plasmid negative clones, 10 from curing effect, with 100% plasmid-free clones, and with a each of the three consecutive subcultures at 0.15% (w/v) curing effect higher even than that of 75 μg/ml acridine glucose concentration, indicate that no plasmid integrated orange, with its 58% plasmid-free clones (Mattarelli et al. 1994). Moreover, the loss of pVS809 in the presence of with the chromosome. 0.15% (w/v) concentration of 10 of the 13 tested carbohy- drates is shown after one, two or three consecutive subcul- tures. These results confirm previous observations that the type and concentration of nutrients available in the growth medium can have a profound effect on cellular metabolism and DNA. Thus, for microbes growing under specific envi- ronmental conditions, there could be a marked effect on the plasmid molecules replicating autonomously (independently of chromosomes) as non-essential genetic elements in cells (Kim et al. 2001;O’Kennedy et al. 2003). For example, plasmid stability studies on Streptomyces arenae showed that any variation in copy number after a change in culture conditions caused instability of the antibiotic resistant plas- mid (Braxenthaler et al. 1991), as was also observed in Lactobacillus plantarum caTC2 growth and maintenance in the presence of certain carbohydrates (Sinha 1992). As well as plasmid yield, the type and concentration of Fig. 5 Agarose gel electrophoresis of pVS809 isolated from cells the basic ingredients used in the culture medium determines derived from the first subcultures at 0.15% (w/v) of the following the amount of biomass produced. The decrease in dry matter carbohydrates: 1 maltose, 2 lactose, 3 sucrose, 4 galactose, 5 ribose, 6 starch, 7 arabinose, 8 glucose, 9 raffinose, 10 fructose could be due to high or low glucose concentrations in the 1756 Ann Microbiol (2012) 62:1751–1756 dormancy, and stringent response. Antimicrob Agents Chemother growth medium, putting an extra metabolic burden on the 34:1865–1868 cells and retarding their proliferation rate, resulting in low Hidaka A, Hamayi Y, Sasaki T, Taniguchi S, Fujimori M (2007) dry matter concentration. At very low (below 0.2% w/v) and Exogeneous cytosine deaminase gene expression in Bifidobacte- very high (above 12.6% w/v) glucose concentrations, the rium breve I-53-8w for tumor-targeting enzyme/prodrug therapy. Biosci Biotechnol Biochem 71:2921–2926 biomass concentration produced is one-third and one-half Kim WS, Park JH, Ren J, Su P, Dunn NW (2001) Survival that of nTPY, respectively. Also, cell morphology changes at response and rearrangement of plasmid DNA of Lactococcus such extreme glucose concentrations, the cell size being lactis during long- term starvation. Appl Environ Microbiol smaller than cells grown at a biomass level similar to cells 67:4594–4602 Lee JH, O’Sullivan DJ (2010) Genomic insights into bifidobacteria. grown in nTPY. The decrease in biomass concentration is Microbiol Mol Biol Rev 74:378–416 related to plasmid loss only in starvation conditions (0.2%, Macfarlane GT, Cummings JH (2002) Probiotics, infection and immu- w/v and below), not at high glucose concentrations (12.6% nity. Curr Opin Infect Dis 15:501–506 and above). Matin A (1991) The molecular basis of carbon-starvation-induced general resistance in Escherichia coli. Mol Microbiol 5:3–10 This paper is the first to report the marked effect of Mattarelli P, Biavati B (1999) L-leucine auxotrophy in Bifidobacterium stressful environmental conditions, particularly carbohy- globosum. Microbiologica 22:73–76 drate starvation, on bifidobacterial plasmid stability. These Mattarelli P, Biavati B, Crociani F, Scardovi V, Prati G (1993) Bifido- results suggest a simple alternative technique for plasmid bacterial cell-wall proteins (Bifops) in Bifidobacterium globosum. Res Microbiol 144:581–590 curing in bifidobacteria, culturing them in low concentra- Mattarelli P, Biavati B, Alessandrini A, Crociani F, Scardovi V (1994) tions of specific carbon sources. The mechanism of why Characterization of the plasmid pVS809 from Bifidobacterium some carbohydrates cause plasmid loss while others have no globosum. Microbiologica 17:327–331 effect is not understood. In recent years plasmid DNA has Novitsky JA, Morita RY (1976) Morphological characterization of small cells resulting from nutrient starvation of a psychrophilic been the subject of appreciable interest due to its attractive marine vibrio. Appl Environ Microbiol 32:617–622 potential application in gene therapy, DNA vaccines, and O’Kennedy R, Ward J, Keshavarz-Moore E (2003) Effect of fermen- food-grade applications, so its maintenance in a bacterial tation strategy on the characteristics of plasmid DNA production. host is very important. Furthermore, plasmid behavior in Biotechnol Appl Biochem 37:83–90 O’Riordan K, Fitzgerald GF (1999) Molecular characterisation of a stressed conditions could be relevant not only to genetics 5.75-kb cryptic plasmid from Bifidobacterium breve NCFB 2258 and ecology but also to the development of cloning and and determination of mode of replication. FEMS Microbiol Lett expression vector systems for bifidobacteria. 174:285–294 Future studies will address in depth the mechanisms in- Prather KJ, Sagar S, Murphy J, Chartrain M (2003) Industrial scale production of plasmid DNA for vaccine and gene therapy: plas- volved in carbohydrate-mediated plasmid curing—an effect mid design, production and purification. Enzyme Microb Technol comparable to the action of antibiotics or other curing agents. 33:865–883 Sangrador-Vegas A, Stanton C, van Sinderen D, Fitzgerald GF, Ross RP (2007) Characterization of plasmid pASV479 from Bifidobac- terium pseudolongum subsp. globosum and its use for expression References vector construction. Plasmid 58:140–147 Saulnier DM, Kolida S, Gibson GR (2009) Microbiology of the human Abt MC, Artis D (2009) The intestinal microbiota in health and intestinal tract and approaches for its dietary modulation. Curr disease: the influence of microbial products on immune cell Pharm Des 15:1403–1414 homeostasis. Curr Opin Gastroenterol 25:496–502 Sgorbati B, Scardovi V, LeBlanc DJ (1982) Plasmids in the genus Biavati B, Mattarelli P (2006) The family Bifidobacteriaceae.In: Bifidobacterium. J Gen Microbiol 128:2121–2131 Dworkin M, Falkow S, Rosenberg E, Schleifer K-H, Stackebrandt Shareck J, Choi Y, Lee B, Miguez CB (2004) Cloning vectors based on E (eds) The Prokaryotes: Archaea. Bacteria: Firmicutes, Actino- cryptic plasmids isolated from lactic acid bacteria: their character- mycetes, vol 3, 2nd edn. Springer, New York, pp 322–382 istics and potential applications in biotechnology. Crit Rev Bio- Braxenthaler M, Poetsch B, Fröhlicha K-U, Meckea D (1991) pSAR1, technol 24:155–154 a natural plasmid from Streptomyces arenae, shows rapid increase Shirasawa Y, Shibahara-Sone H, Iino T, Ishikawa F (2010) Bifidobac- and decrease of copy numbers on changes of growth media. terium bifidum BF-1 suppresses Helicobacter pylori-induced FEMS Microbiol Lett 83:311–316 genes in human epithelial cells. J Dairy Sci 93:4526–4534 Cronin M, Knobel M, O’Connell-Motherway M, Fitzgerald GF, van Sinha RP (1992) Plasmid instability in Lactobacillus plantarum strain Sinderen D (2007) Molecular dissection of a bifidobacterial rep- caTC2. Curr Microbiol 25:219–223 licon. Appl Environ Microbiol 73:7858–7866 Turroni F, van Sinderen D, Ventura M (2010) Genomics and ecological Endo A, Futagawa-Endo Y, Dicks LM (2010) Diversity of Lactobacil- overview of the genus Bifidobacterium. Int J Food Microbiol lus and Bifidobacterium in feces of herbivores, omnivores and 149:37–44 carnivores. Anaerobe 16:590–596 Wong JM, de Souza R, Kendall CW, Emam A, Jenkins DJ (2006) Gilbert P, Collier PJ, Brown MRV (1990) Influence of growth rate Colonic health: fermentation and short chain fatty acids. J Clin on susceptibility to antimicrobial agents: biofilms, cell cycle, Gastroenterol 40:235–243 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Annals of Microbiology Springer Journals

Carbohydrate stress-related response in Bifidobacterium pseudolongum subsp. globosum

Loading next page...
 
/lp/springer-journals/carbohydrate-stress-related-response-in-bifidobacterium-pseudolongum-0fp8Ia3A06

References (26)

Publisher
Springer Journals
Copyright
Copyright © 2012 by Springer-Verlag and the University of Milan
Subject
Life Sciences; Microbiology; Microbial Genetics and Genomics; Microbial Ecology; Mycology; Medical Microbiology; Applied Microbiology
ISSN
1590-4261
eISSN
1869-2044
DOI
10.1007/s13213-012-0432-9
Publisher site
See Article on Publisher Site

Abstract

Ann Microbiol (2012) 62:1751–1756 DOI 10.1007/s13213-012-0432-9 ORIGINAL ARTICLE Carbohydrate stress-related response in Bifidobacterium pseudolongum subsp. globosum Stefano Tacconi & Barbara Sgorbati & Monica Modesto & Bruno Biavati & Lorenzo Nissen & Paola Mattarelli Received: 23 September 2011 /Accepted: 7 February 2012 /Published online: 1 March 2012 Springer-Verlag and the University of Milan 2012 . . . Abstract Bifidobacteria are indigenous components of hu- Keywords Bifidobacterium Plasmid Probiotic man and animal gastrointestinal microbiota, and their Environmental stress Carbohydrate starvation health-promoting benefits have long been recognized. Of the 36 currently described species of the Bifidobacterium genus, 8 contain plasmids, most of which are cryptic. It is Introduction possible that plasmid presence is related very closely to environmental change, so in conditions of stress this pres- Normal gut microbiota are involved in host physiology and ence could be specifically controlled. For plasmid-positive play a key role in animal health and disease. Bifidobacteria Bifidobacterium pseudolongum subsp. globosum RU809/1, are indigenous components of human and animal gastroin- the influence of the type and concentration of the carbohy- testinal microbiota and their health-promoting benefits have drate source is evident in the dramatic pVS809 curing effect long been recognized, although the molecular mechanisms when growth is conducted in the presence of 0.15% (w/v) underlying their beneficial effects are unclear (Lee and glucose, lactose, maltose, melibiose, raffinose or starch. The O’Sullivan 2010). Besides their benefits, such as immunity effect is linked to carbohydrate starvation, not to carbohy- system stimulation (Abt and Artis 2009), protection from drate abundance, and is independent of biomass growth. pathogen infection (Macfarlane and Cummings 2002; Plasmid curing was achieved after one or two consecutive Shirasawa et al. 2010), energy production from the host’s transfers, also in cells grown on medium containing 0.15% undigested carbohydrates, and vitamin production (Saulnier et arabinose, fructose, galactose and sucrose, but not mannose, al. 2009;Wongetal. 2006), there is a growing commercial ribose or xylose. Knowing plasmid behavior in stressful interest in the utilization of bifidobacteria as probiotics. More- conditions, like carbon source availability, has allowed an over, bifidobacteria were also identified recently as potential early insight into carbohydrate starvation as a curing agent expression vectors for proteins of pharmacological and food- for bifidobacteria. Furthermore, knowledge of plasmid be- related interest (Hidaka et al. 2007; Prather et al. 2003; havior in stressful conditions could be important not only in Sangrador-Vegas et al. 2007; Shareck et al. 2004). To facilitate genetics and ecology but also in food-grade and pharma- genetic manipulation of bifidobacteria, and the development ceutical applications for the development of cloning and and improvement of their cloning vectors and gene transfer expression vector systems for bifidobacteria. systems, a basic knowledge of plasmid genetics and physiol- ogy is very important. Bifidobacteria are subject to stressful conditions in both biotechnological processes and nature, and their ability to : : : : : respond quickly to stress is essential for survival. For bacterial S. Tacconi B. Sgorbati M. Modesto B. Biavati L. Nissen P. Mattarelli (*) cells living in a natural environment, one of the biggest prob- Department of Agroenvironmental Sciences and Technologies, lems is nutritional limitation—cells are starved most of the Bologna University, time (Matin 1991); the opposite occurs in the laboratory: Via Fanin 42, bacteria are usually cultivated under optimal or over-optimal 40127 Bologna, Italy growth conditions. Thus, in conditions of starvation or e-mail: paola.mattarelli@unibo.it 1752 Ann Microbiol (2012) 62:1751–1756 abundance, bacterial cultures undergo numerous metabolic normal tryptone-phytone yeast (nTPY) (pH 6.8) containing changes to facilitate survival, and the dynamism lies in the 1.5% (w/v) of glucose (Biavati and Mattarelli 2006). An terms of cell number and cell type. Also chromosomal and overnight culture at 4% (v/v) was used as inoculum. extrachromosomal DNA replication undergoes specific con- trols. Therefore, the investigation of plasmid maintenance in Growth in different glucose concentrations bacterial adaptation to stress is important. Plasmids have been detected in only 8 of the 36 known TPY medium supplemented with different concentrations Bifidobacterium species: in B. bifidum, B. breve, B. catenu- [0.1, 0.15, 0.2, 0.4, 0.6, 0.8, 1.0, 1.25, 1.5, 3.6, 5.4, 7.2, latum, B. pseudocatenulatum and B. longum subsp. longum 9.0, 10.8, 12.6, 14.4, 16.2, 18.0, 27.0% (w/v)] of glucose of the 13 human species, in B. pseudolongum subsp. globo- was used. After filtration, the glucose solutions were added sum species of the 17 animal species and in B. asteroides to the cooled sterile growth medium without sugar. All and B. indicum of the 6 insect species (Lee and O’Sullivan strains were grown in 500 mL medium at 37°C on each 2010; Sgorbati et al. 1982). It can be expected that species glucose concentration. The medium was inoculated as soon with plasmids that occur naturally in nature are more able to as possible after autoclaving with cells derived from an maintain them in a stable condition than other species. overnight 20 mL TPY culture centrifuged and resuspended Among the known species, B. pseudolongum subsp. glo- in glucose-free TPY to avoid inoculum interference; bosum, which is typical of animal habitats, has the most 0.5 mL/L NaHCO (10% w/v) filter sterilized solution was cosmopolitan distribution. It colonizes the intestine of cattle, added to the flask and incubation was conducted under air. pig, dog, rat and chicken (Biavati and Mattarelli 2006), and The experiments were conducted in three replications. was isolated recently from wild animals like the wallaby and springbok (Endo et al. 2010); furthermore, it has been sporad- Growth in different carbohydrates ically isolated from human feces (Mattarelli et al. 1993). A study by Turroni et al. (2010) divided the Bifidobacterium TPY medium was supplemented with different carbohy- genus into five phylogenetic groups and, on analyzing human drates at concentrations of 1.5% and 0.15% (w/v) in place microbiota, it was found that the 16S rRNA sequences of of glucose. Carbohydrates used were: arabinose, fructose, about 50% of unculturable bifidobacteria have high homology galactose, glucose, lactose, maltose, mannose, melibiose, with the "pseudolongum group". This species is typical of the raffinose, ribose, starch, sucrose, and xylose. Media were animal habitat, and this underlines its potential importance. autoclaved separately from the carbohydrates, which were Bifidobacterial plasmids are generally considered cryptic, added to the cooled broth after filtration. All strains were except for the plasmid from B. bifidum NCFB 1454, which grown in 500 mL medium inoculated on the appropriate was proposed as being involved in bifidocin B production, carbohydrate concentrations as described above. The experi- and the plasmid pVS809 from B. pseudolongum subsp. ments were conducted in three replications. globosum, which seems to be involved in leucine auxotro- phy (Mattarelli and Biavati 1999). Dry cell matter concentration Plasmid pVS809 occurs naturally in extrachromosomal DNA in B. pseudolongum subsp. globosum RU809/1, and, Aliquots (10 mL) of cell culture were centrifuged at 5,000 g on the basis of mapping restriction analysis, has been esti- for 10 min. The supernatant was decanted and the cell mated to be 28 Kb in size (Mattarelli et al. 1994). pellets washed with an equal volume of sterile water. The The present work investigates how carbon type and the washed cells were dried in a dry oven at 105°C for 24 h and concentration of carbon source availability—both important then weighed. stressful conditions in culture media—influence pVS809 stability and dry matter concentration in B. pseudolongum Plasmid analyses subsp. globosum RU 809/1. The aim of the study was to obtain significant information to increase our still incom- Plasmid DNA was isolated by a method described previously plete knowledge of plasmid physiology in bifidobacteria. (Sgorbati et al. 1982), and was resolved on 0.8% agarose gel at 80 V for 4 h. Semiquantitative videodensitometric analysis of the gel was performed using the Gel Doc™ XR + System Materials and methods (Bio-Rad, Hercules, CA). RU 809/1 was subcultured three times in TPY containing Bacterial strain, media, and growth different glucose concentrations, or different carbon sources at different concentrations, and each subculture was ana- The strain Bifidobacterium pseudolongum subsp. globosum lyzed for plasmid presence. Plasmid-negative or low plas- RU 809/1 was grown routinely anaerobically at 37°C in mid quantity cultures were retested and plated on 1.8% agar Ann Microbiol (2012) 62:1751–1756 1753 0.8 TPY containing the standard 1.5% glucose: 70 colonies 0.7 were isolated from each subculture in the appropriate dilu- tions. The cultures derived from each of the picked colonies 0.6 were analyzed for plasmid content. 0.5 0.4 Chromosomal insertion of plasmid DNA 0.3 The possibility of plasmid presence in a chromosomally 0.2 integrated form was tested using a pVS809 DNA probe 0.1 against RU809/1 chromosomal DNA from 30 plasmid- negative clones isolated from each of three consecutive subcultures at 0.15% (w/v) glucose concentration. The de- tailed procedure for plasmid chromosomal insertion has Fig. 2 Dry matter concentrations (mg/mL) of RU809/1 grown on been described previously (Mattarelli et al. 1994). 1.5% (w/v) of different carbohydrate as carbon source Statistical analysis Effect of different carbohydrates on RU809/1 performance All experiments were performed in triplicate. All quantita- Each of the 13 carbohydrates tested (arabinose, fructose, tive data, unless otherwise stated, are presented as means, galactose, glucose, lactose, maltose, mannose, melibiose, raf- with error represented by standard deviation. finose, ribose, sucrose, starch, xylose) was utilized as carbon source at a normal concentration of 1.5% in TPY medium. The dry matter concentration was similar in all the carbohy- Results drates tested, with a little higher growth in glucose (Fig. 2). When the carbohydrates were added at 0.15% w/v in TPY, the Effect of glucose on RU809/1 performance dry matter decreased one- or two-fold with respect to nTPY. The three tested subcultures show similar dry matter biomass, RU 809/1 was cultivated in a range of glucose concentra- with the exception of mannose, ribose and xylose that showed tions varying from 0.1 to 27% (w/v). All the glucose con- less dry matter, i.e., low growth rate (Fig. 3). centrations permitted growth with the exception of 27% (w/ v). The biomass value based on dry matter calculation is Morphological observations of cells grown at different sugar almost steady in the glucose concentration range of 0.4– concentrations 14.4% (w/v), while it diminishes at the highest and lowest glucose concentrations (Fig. 1). The maximum dry matter The RU809/1 cells grown at the extreme glucose concen- concentration (0.93±0.013 mg/ml) was obtained at a 3.6% trations (below or equal to 0.4% and above or equal to (w/v) glucose concentration. The dry matter of subcultures 16.2%) show a reduction in size with respect to that at at the same glucose concentration maintains the same values 1.5% (w/v), the nTPY glucose concentration (Fig. 4). This (data not shown). observation is in agreement with the literature, highlighting that bacterial cells can utilize different strategies to conserve energy during stress conditions, reducing not only their 0.9 growth rate but also their size (Gilbert et al. 1990; Novitsky 0.8 and Morita 1976). 0.7 0.6 Effect of different glucose concentrations on the loss 0.5 of plasmid in RU809/1 0.4 0.3 RU809/1 cells grown in nTPY possess one large plas- 0.2 mid of 28,000 bp (Mattarelli et al. 1994): this plasmid 0.1 band was considered as a positive control in the fol- 0 2 4 6 8 101214161820 lowing experiments. %(w/v) glucose concentration The cells were subcultured once at glucose concentra- tions ranging from 0.1 to 18.0% (w/v), and the plasmid was Fig. 1 Dry matter concentrations of RU809/1 grown at different percentage (w/v) of glucose as carbon source present at concentrations equal to, or above, 0.4%. Cells mg/ml dry matter mg/ml dry matter 1754 Ann Microbiol (2012) 62:1751–1756 Fig. 3 Dry matter concentra- 0.60 tions (mg/mL) of RU809/1 grown on 0.15% (w/v) of differ- ent carbohydrates as carbon 0.50 source determined at first , second □ and third ■ consecutive 0.40 subcultures 0.30 0.20 0.10 0.00 grown at 0.2% (w/v) glucose showed reduced plasmid in- (w/v) (Table 1). Like cells grown in glucose, cells grown tensity, dropping to about 40% (densitometric analysis), and in lactose, maltose, melibiose, raffinose and starch lost the no plasmid band was observed at 0.15% and 0.1% (w/v) plasmid at the first subculture containing these carbohy- glucose concentrations. Plasmid-free and plasmid-positive drates (Fig. 5). Instead, in the three consecutive subcultures cultures were subcultured three times and the previous examined, mannose and ribose had no effect, the plasmid results were confirmed. being maintained without modification. However, after one Three consecutive subcultures derived from 0.2 and consecutive subculture in arabinose, fructose and sucrose, 0.15% (w/v) glucose cultures were produced. For each and after two in galactose, we again observed plasmid loss. subculture, 70 clones were obtained. Each of these clones Cells grown on medium containing xylose showed, in all was tested for the presence of plasmid: 35±3 plasmid pos- three subcultures, a diminished plasmid band, revealed by itive clones were obtained from 0.2% (w/v) glucose con- densitometric analysis (about 35% with respect to normal centration and 0 plasmid positive clones were obtained from density of plasmid positive cells), but not completely lost 0.15% glucose concentration. This result is in agreement pVS809 (Table 1). Given the result shown above for glucose with densitometric data. at 0.2%, it is conceivable that, also for the different carbo- hydrate sources, the plasmid decrease is not due to a copy Effect of different carbon source concentrations on plasmid number decrease but to the presence of mixed plasmid positive and negative populations. No direct relationship loss in RU809/1 was observed between dry matter concentration and plasmid The different carbon sources utilized at 1.5% (w/v) did not maintenance. For example, the cells grown in 0.15% (w/v) mannose, ribose or xylose, which had the lowest dry matter influence the pVS809 presence, but there were some changes in the plasmid content in cells grown at 0.15% concentrations, maintained the plasmid in all three subcultures: Fig. 4 Phase-contrast photomicrographs showing cellular morphology of RU 809/1 grown at a 0.15%, b 1.5% and c 18% (w/v) of glucose concentrations. Bar 10 μm mg/ml dry matter Ann Microbiol (2012) 62:1751–1756 1755 Table 1 Plasmid presence in three consecutive subcultures of RU809/ Discussion 1 grown at 0.15% (w/v) concentration of different carbohydrates. + Plasmid positive cells, +/− plasmid band intensity 60–65% of control The basic biology of plasmids in the Bifidobacterium species value, − plasmid-negative cells is unexpectedly poorly known. The biological significance of Carbon source Plasmid presence these plasmids is unclear as most of the open reading frames, apart from those predicted to specify replication functions, 1st subculture 2nd subculture 3rd subculture code for proteins of unknown function (Cronin et al. 2007). Arabinose +/− –– In general, naturally occurring plasmids are firmly maintained Fructose + –– (O’Riordan and Fitzgerald 1999), but no studies have been carried out on how stressing environmental conditions Galactose +/− +/− – affects plasmid stability in bifidobacteria, with the excep- Glucose –– – tion of plasmid curing experiments with chemical agents Lactose –– – (Mattarelli et al. 1994). Maltose –– – The survey of glucose concentrations, which ranged from Mannose + + + 18 to 0.1% (w/v), showed that plasmid stability became Melibiose –– – affected at 0.2% glucose concentration, with a loss of about Raffinose –– – 60%, while at concentrations of 0.15% and 0.1%, pVS809 Ribose + + + was lost completely. The partial plasmid loss was related to Sucrose + –– the presence of mixed plasmid-positive and -negative clones Starch –– – in the population, as revealed by plasmid analysis of clones Xylose +/− +/− +/− derived from different subcultures, rather than a lowering of plasmid copy number. A direct correlation was found be- tween low plasmid band intensity and the number of mannose and ribose at 100% while in xylose, the plasmid is plasmid-positive clones present, and the presence of chro- maintained with a value of 65%. mosomally integrated plasmid in plasmid cured clones was not detectable. Therefore, there is a strong relationship be- Plasmid-chromosome integration tween plasmid presence and glucose concentration, the in- tensity of this relationship being such that glucose starvation The results of the hybridization of the pVS809 probe to the could be comparable to the 50 μg/ml ethidium bromide chromosomal DNA in 30 plasmid negative clones, 10 from curing effect, with 100% plasmid-free clones, and with a each of the three consecutive subcultures at 0.15% (w/v) curing effect higher even than that of 75 μg/ml acridine glucose concentration, indicate that no plasmid integrated orange, with its 58% plasmid-free clones (Mattarelli et al. 1994). Moreover, the loss of pVS809 in the presence of with the chromosome. 0.15% (w/v) concentration of 10 of the 13 tested carbohy- drates is shown after one, two or three consecutive subcul- tures. These results confirm previous observations that the type and concentration of nutrients available in the growth medium can have a profound effect on cellular metabolism and DNA. Thus, for microbes growing under specific envi- ronmental conditions, there could be a marked effect on the plasmid molecules replicating autonomously (independently of chromosomes) as non-essential genetic elements in cells (Kim et al. 2001;O’Kennedy et al. 2003). For example, plasmid stability studies on Streptomyces arenae showed that any variation in copy number after a change in culture conditions caused instability of the antibiotic resistant plas- mid (Braxenthaler et al. 1991), as was also observed in Lactobacillus plantarum caTC2 growth and maintenance in the presence of certain carbohydrates (Sinha 1992). As well as plasmid yield, the type and concentration of Fig. 5 Agarose gel electrophoresis of pVS809 isolated from cells the basic ingredients used in the culture medium determines derived from the first subcultures at 0.15% (w/v) of the following the amount of biomass produced. The decrease in dry matter carbohydrates: 1 maltose, 2 lactose, 3 sucrose, 4 galactose, 5 ribose, 6 starch, 7 arabinose, 8 glucose, 9 raffinose, 10 fructose could be due to high or low glucose concentrations in the 1756 Ann Microbiol (2012) 62:1751–1756 dormancy, and stringent response. Antimicrob Agents Chemother growth medium, putting an extra metabolic burden on the 34:1865–1868 cells and retarding their proliferation rate, resulting in low Hidaka A, Hamayi Y, Sasaki T, Taniguchi S, Fujimori M (2007) dry matter concentration. At very low (below 0.2% w/v) and Exogeneous cytosine deaminase gene expression in Bifidobacte- very high (above 12.6% w/v) glucose concentrations, the rium breve I-53-8w for tumor-targeting enzyme/prodrug therapy. Biosci Biotechnol Biochem 71:2921–2926 biomass concentration produced is one-third and one-half Kim WS, Park JH, Ren J, Su P, Dunn NW (2001) Survival that of nTPY, respectively. Also, cell morphology changes at response and rearrangement of plasmid DNA of Lactococcus such extreme glucose concentrations, the cell size being lactis during long- term starvation. Appl Environ Microbiol smaller than cells grown at a biomass level similar to cells 67:4594–4602 Lee JH, O’Sullivan DJ (2010) Genomic insights into bifidobacteria. grown in nTPY. The decrease in biomass concentration is Microbiol Mol Biol Rev 74:378–416 related to plasmid loss only in starvation conditions (0.2%, Macfarlane GT, Cummings JH (2002) Probiotics, infection and immu- w/v and below), not at high glucose concentrations (12.6% nity. Curr Opin Infect Dis 15:501–506 and above). Matin A (1991) The molecular basis of carbon-starvation-induced general resistance in Escherichia coli. Mol Microbiol 5:3–10 This paper is the first to report the marked effect of Mattarelli P, Biavati B (1999) L-leucine auxotrophy in Bifidobacterium stressful environmental conditions, particularly carbohy- globosum. Microbiologica 22:73–76 drate starvation, on bifidobacterial plasmid stability. These Mattarelli P, Biavati B, Crociani F, Scardovi V, Prati G (1993) Bifido- results suggest a simple alternative technique for plasmid bacterial cell-wall proteins (Bifops) in Bifidobacterium globosum. Res Microbiol 144:581–590 curing in bifidobacteria, culturing them in low concentra- Mattarelli P, Biavati B, Alessandrini A, Crociani F, Scardovi V (1994) tions of specific carbon sources. The mechanism of why Characterization of the plasmid pVS809 from Bifidobacterium some carbohydrates cause plasmid loss while others have no globosum. Microbiologica 17:327–331 effect is not understood. In recent years plasmid DNA has Novitsky JA, Morita RY (1976) Morphological characterization of small cells resulting from nutrient starvation of a psychrophilic been the subject of appreciable interest due to its attractive marine vibrio. Appl Environ Microbiol 32:617–622 potential application in gene therapy, DNA vaccines, and O’Kennedy R, Ward J, Keshavarz-Moore E (2003) Effect of fermen- food-grade applications, so its maintenance in a bacterial tation strategy on the characteristics of plasmid DNA production. host is very important. Furthermore, plasmid behavior in Biotechnol Appl Biochem 37:83–90 O’Riordan K, Fitzgerald GF (1999) Molecular characterisation of a stressed conditions could be relevant not only to genetics 5.75-kb cryptic plasmid from Bifidobacterium breve NCFB 2258 and ecology but also to the development of cloning and and determination of mode of replication. FEMS Microbiol Lett expression vector systems for bifidobacteria. 174:285–294 Future studies will address in depth the mechanisms in- Prather KJ, Sagar S, Murphy J, Chartrain M (2003) Industrial scale production of plasmid DNA for vaccine and gene therapy: plas- volved in carbohydrate-mediated plasmid curing—an effect mid design, production and purification. Enzyme Microb Technol comparable to the action of antibiotics or other curing agents. 33:865–883 Sangrador-Vegas A, Stanton C, van Sinderen D, Fitzgerald GF, Ross RP (2007) Characterization of plasmid pASV479 from Bifidobac- terium pseudolongum subsp. globosum and its use for expression References vector construction. Plasmid 58:140–147 Saulnier DM, Kolida S, Gibson GR (2009) Microbiology of the human Abt MC, Artis D (2009) The intestinal microbiota in health and intestinal tract and approaches for its dietary modulation. Curr disease: the influence of microbial products on immune cell Pharm Des 15:1403–1414 homeostasis. Curr Opin Gastroenterol 25:496–502 Sgorbati B, Scardovi V, LeBlanc DJ (1982) Plasmids in the genus Biavati B, Mattarelli P (2006) The family Bifidobacteriaceae.In: Bifidobacterium. J Gen Microbiol 128:2121–2131 Dworkin M, Falkow S, Rosenberg E, Schleifer K-H, Stackebrandt Shareck J, Choi Y, Lee B, Miguez CB (2004) Cloning vectors based on E (eds) The Prokaryotes: Archaea. Bacteria: Firmicutes, Actino- cryptic plasmids isolated from lactic acid bacteria: their character- mycetes, vol 3, 2nd edn. Springer, New York, pp 322–382 istics and potential applications in biotechnology. Crit Rev Bio- Braxenthaler M, Poetsch B, Fröhlicha K-U, Meckea D (1991) pSAR1, technol 24:155–154 a natural plasmid from Streptomyces arenae, shows rapid increase Shirasawa Y, Shibahara-Sone H, Iino T, Ishikawa F (2010) Bifidobac- and decrease of copy numbers on changes of growth media. terium bifidum BF-1 suppresses Helicobacter pylori-induced FEMS Microbiol Lett 83:311–316 genes in human epithelial cells. J Dairy Sci 93:4526–4534 Cronin M, Knobel M, O’Connell-Motherway M, Fitzgerald GF, van Sinha RP (1992) Plasmid instability in Lactobacillus plantarum strain Sinderen D (2007) Molecular dissection of a bifidobacterial rep- caTC2. Curr Microbiol 25:219–223 licon. Appl Environ Microbiol 73:7858–7866 Turroni F, van Sinderen D, Ventura M (2010) Genomics and ecological Endo A, Futagawa-Endo Y, Dicks LM (2010) Diversity of Lactobacil- overview of the genus Bifidobacterium. Int J Food Microbiol lus and Bifidobacterium in feces of herbivores, omnivores and 149:37–44 carnivores. Anaerobe 16:590–596 Wong JM, de Souza R, Kendall CW, Emam A, Jenkins DJ (2006) Gilbert P, Collier PJ, Brown MRV (1990) Influence of growth rate Colonic health: fermentation and short chain fatty acids. J Clin on susceptibility to antimicrobial agents: biofilms, cell cycle, Gastroenterol 40:235–243

Journal

Annals of MicrobiologySpringer Journals

Published: Mar 1, 2012

There are no references for this article.